Energy Savings: Performance Contracts Offer Benefits, but	 
Vigilance Is Needed to Protect Government Interests (22-JUN-05,  
GAO-05-340).							 
                                                                 
The federal government is the nation's largest energy consumer,  
spending, by latest accounting, $3.7 billion on energy for its	 
500,000 facilities. Upfront funding for energy-efficiency	 
improvements has been difficult to obtain because of budget	 
constraints and competing agency missions. The Congress in 1986  
authorized agencies to use Energy Savings Performance Contracts  
(ESPCs) to privately finance these improvements. The law requires
that annual payments for ESPCs not exceed the annual savings	 
generated by the improvements. GAO was asked to identify (1) the 
extent to which agencies used ESPCs; (2) what energy savings,	 
financial savings, and other benefits agencies expect to achieve;
(3) the extent to which actual financial savings cover costs; and
(4) what areas, if any, require steps to protect the government's
financial interests in using ESPCs.				 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-340 					        
    ACCNO:   A27178						        
  TITLE:     Energy Savings: Performance Contracts Offer Benefits, but
Vigilance Is Needed to Protect Government Interests		 
     DATE:   06/22/2005 
  SUBJECT:   Agency missions					 
	     Energy						 
	     Energy costs					 
	     Financial analysis 				 
	     Financial management				 
	     Government contracts				 
	     Energy savings performance contracts		 
	     Performance-based contracting			 
	     Savings estimates					 
	     DOE Federal Energy Management 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-05-340

                 United States Government Accountability Office

GAO Report to Congressional Requesters

June 2005

ENERGY SAVINGS

    Performance Contracts Offer Benefits, but Vigilance Is Needed to Protect
                              Government Interests

                                       a

GAO-05-340

[IMG]

June 2005

ENERGY SAVINGS

Performance Contracts Offer Benefits, but Vigilance Is Needed to Protect
Government Interests

What GAO Found

Although comprehensive data on federal agencies' use of ESPCs are not
available, in fiscal years 1999 through 2003, we found that 20 federal
agencies undertook 254 ESPCs to finance investments in energy-saving
improvements for 5 to 25 years. Through the ESPCs, federal agencies plan
to make annual payments amounting to at least $2.5 billion spread over the
lifetime of the contracts.

Agencies expect to achieve benefits that include energy savings worth at
least $2.5 billion over the life of the contracts, as well as other
benefits that cannot be easily quantified, such as improved reliability of
the newer equipment over the aging equipment it replaced, environmental
improvements, and additional energy and financial savings once the
contracts have been paid for. While these benefits could be achieved using
upfront funds and with lower financing costs, agencies stated that they
generally have not received sufficient funds upfront for doing so and see
ESPCs as a necessary supplement to upfront funding in order to achieve the
benefits cited. Agencies believe that ESPCs also provide unique benefits
such as a partial shift of risk from agencies to private energy services
companies and a more integrated approach to providing efficiency measures.

Agencies structure ESPCs so that financial savings cover costs and they
reported that many do. However, GAO could not verify that conclusion using
the data on ESPCs, and GAO work and agency audits disclosed ESPCs in which
unfavorable contract terms, missing documentation, and other problems
caused GAO to question how consistently savings cover costs. Furthermore,
differing interpretations of the law establishing ESPCs about what
components of costs must be paid for from the savings generated by the
project or may be paid for using other funding sources have contributed to
uncertainties about whether savings are appropriately covering costs.

GAO identified concerns in the areas of expertise and related information
and competition that are fundamental to ensuring that savings cover costs
and to protecting the government's financial interests in using ESPCs.
According to agency officials, they often lacked the technical and
contracting expertise and information (such as interest rates and markups)
to negotiate ESPCs and to monitor contract performance in the long term.
The officials also think there may be insufficient competition among
finance and energy services companies and that this could lead to higher
costs for ESPCs.

                 United States Government Accountability Office

Contents

  Letter

Results in Brief
Background
Many Agencies Used ESPCs, Although the Extent of Use Varied
Agencies Expect ESPC-Financed Projects to Result in Energy

Savings As Well As Other Benefits Agencies Believe Financial Savings Cover
Costs, but Whether Savings Actually Do So Is Uncertain

Agencies Are Concerned About Officials' Lack of Necessary Expertise and
Information and About Competitiveness of the Super ESPCs

Conclusions Matter for Congressional Consideration Recommendations for
Executive Action Agency Comments

                                                                     1 4 8 12

19

27

35 47 49 50 51

Appendixes                                                              
                Appendix I:       Objectives, Scope and Methodology        53 
               Appendix II:    Comments from the Department of Defense     56 
              Appendix III:     Comments from the Department of Energy     59 
              Appendix IV:     Comments from the Department of Veterans    65 
                                               Affairs                     
                Appendix V:       Comments from the General Services       68 
                                            Administration                 
              Appendix VI:      GAO Contact and Staff Acknowledgments      71 

Tables   Table 1: Number and Cost of ESPC Projects Undertaken in Fiscal 
                               Years 1999 through 2003                     14 
               Table 2: Energy and Financial Savings for ESPC Projects     
                    Undertaken in Fiscal Years 1999 through 2003           20 
                  Table 3: Steps Contracting Centers Are Taking to Address 
                                                                  Concerns 
                         About Expertise, Information, and Competitiveness 46 
Figures Figure 1: Percentage of ESPC Financed Projects, by Contract        
           Value, Undertaken in Fiscal Years 1999 through 2003             15
              Figure 2: Agencies' Use of Contracting Centers for ESPCs     
                    Undertaken in Fiscal Years 1999 through 2003           18 

Contents

Abbreviations

DOD Department of Defense
DOE Department of Energy
DOJ Department of Justice
ESPC Energy Savings Performance Contracts
FEMP Federal Energy Management Program
GSA General Services Administration
MMBTU million British thermal units
VA Department of Veterans Affairs

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

June 22, 2005

The Honorable Tom Davis Chairman The Honorable Henry A. Waxman Ranking
Minority Member Committee on Government Reform House of Representatives

The federal government is the single largest energy consumer in the
nation, spending about $3.7 billion in fiscal year 20021 on energy for its
approximately 500,000 facilities in the United States. The Energy Policy
Act of 1992 and several subsequent executive orders require federal
agencies to reduce their consumption of energy in federal facilities. Most
notably, Executive Order 13123, issued in 1999, requires agencies, by
2010, to reduce energy consumption by 35 percent from a 1985 baseline.
Additional provisions in the act and various executive orders have added
other goals, such as conserving water and using renewable fuels.

Whether to pay for energy-efficiency improvements that reduce energy
consumption through up-front appropriations or through private financing
is a matter of concern to many. Agencies and members of the Congress have
long recognized that upfront funding for energy-efficiency improvements
has often been difficult to obtain because of budget constraints and
competing agency mission priorities. In 1986, the Congress provided
agencies with an alternative mechanism for obtaining energyefficiency
improvements when it authorized agencies to use Energy Savings Performance
Contracts (ESPCs), a type of share-in-savings contract, to privately
finance the improvements. This step reflected the trend in the federal
government toward increased reliance on performancebased contracting to
improve the services agencies receive from contractors. In
performance-based contracting, the agency specifies the result it desires
and leaves it to the contractor to decide how best to achieve the desired
result. Through share-in-savings contracting, one performance-based
technique, the agency compensates a contractor from the financial benefits
derived as a result of contract performance. Under an

1Fiscal year 2002 is the latest year for which the Department of Energy
has reported data on federal energy use, see U.S. Department of Energy,
Annual Report to Congress on Federal Energy Management and Conservation
Programs Fiscal Year 2002 (Washington, D.C.: Sept. 29, 2004).

ESPC, agencies enter into a long-term contract (up to 25 years) with a
private energy services company in which the company installs
energyefficiency improvements financed from private funds. The agency then
repays the company until the improvements have been paid for. The law
requires that annual payments for the ESPCs not exceed the value of the
annual utility savings generated by the installed energy-efficiency
improvements. As part of an ESPC, the agency and the energy services
company estimate the annual energy and financial savings and develop a
plan to monitor and verify that the expected savings actually occur. ESPCs
are designed to shift performance risk associated with energy-efficiency
improvements from the agency to the company. This shift is to be made by
conditioning annual payments to the company on verification that the
expected financial savings have been realized. These savings are to be
calculated as the difference between the baseline cost for energy
consumption that would have been incurred without the ESPC and the cost of
the energy consumption with the ESPC's energy-efficiency improvements in
place.

Agencies tended to use ESPCs sparingly during the late 1980s and early
1990s, largely because negotiating an ESPC can be a highly technical and
time-consuming process. Agencies began using ESPCs more during the late
1990s. Agencies' energy reduction goals became more ambitious as a result
of Executive Order 13123, and the new goals required agencies to allocate
additional funds to install energy-efficiency improvements. To help
simplify and shorten the ESPC negotiation process, the Department of
Energy's (DOE) Federal Energy Management Program (FEMP) negotiated "super"
ESPCs with energy services companies that FEMP prequalified, via a
competitive process, to provide services under the contracts. FEMP's super
ESPCs are umbrella contracts that federal agencies may use to purchase
energy equipment and services. The Department of the Air Force and the
U.S. Army Corps of Engineers negotiated similar "super" ESPCs. Agencies
have the option to use the super ESPCs to take advantage of some
prenegotiated terms and conditions. In this regard, agencies can implement
delivery orders more quickly using the super ESPCs because the competitive
selection process has already been completed and key terms of the contract
negotiated. Alternatively, the agencies may still enter into "stand-alone"
ESPCs with energy services companies using a separate competitive
selection process.

The use of ESPCs in recent years has raised questions about how these
contracts should be reflected in the federal budget. At present, they are
not reflected-"scored"--upfront in the budget when the contract is signed,
and

budget agencies disagree about whether they should be.2 The Congressional
Budget Office believes that the obligation to make payments for the
energy-efficiency improvements and the financing costs is incurred when
the government signs the ESPC and that scoring the full cost is consistent
with governmentwide accounting principles that the budget reflect this
commitment as a new obligation at the time of signing. The Office of
Management and Budget, on the other hand, includes the costs of ESPCs in
the budget on an annual basis as they are incurred. The scoring treatment
is based on the contingent nature of the contract-payments are contingent
on achieving expected financial savings and, therefore, the government is
not fully committed to the entire long-term cost of the ESPC at the time
it is signed. Agencies have statutory authority to enter into a multiyear
contract even if funds are available only to pay for the first year of the
contract. Although authorization for ESPCs lapsed on October 1, 2003, it
was renewed on October 28, 2004, through fiscal year 2006, and retroactive
authorization was provided for any ESPCs signed between the time the
authority expired and was reinstated.

In this context, you asked us to determine, for contracts agencies
undertook in fiscal years 1999 through 2003, (1) the extent to which
agencies used ESPCs; (2) what energy savings, financial savings, and other
benefits agencies expect to achieve; (3) the extent to which actual
financial savings from ESPCs cover costs; and (4) what areas, if any,
require steps to protect the government's financial interests in using
ESPCs.

To answer these questions, we first obtained basic contract data from the
databases of the four federal contracting centers that assist agencies
with ESPCs-the Air Force Civil Engineer Support Agency, the U.S. Army
Corps of Engineers' Huntsville Center, FEMP, and the Naval Facilities
Engineering Service Center, which reflect the majority of all federal
ESPCs undertaken during fiscal years 1999 through 2003. We did not
completely assess these data for reliability; however, we reviewed the
steps each agency took to ensure the data were reliable and determined
that these steps were sufficient for our reporting purposes. We also
obtained more detailed contract data for the same period from the seven
federal agencies having the most facility floor space and highest energy
use and, therefore, the most potential to use ESPCs. These agencies were
DOE; the Departments of the Air Force, the Army, the Navy (including the
Marine Corps), Justice,

2GAO, Capital Financing: Partnerships and Energy Savings Performance
Contracts Raise Budgeting and Monitoring Concerns, GAO-05-55 (Washington,
D.C.: Dec. 16, 2004).

and Veterans Affairs; and the General Services Administration. We did not
perform formal benefit/cost analyses of individual ESPC projects or of
ESPCs as a whole because of data limitations. Consequently, to assess the
costs and benefits of ESPCs, we supplemented the limited data analysis we
were able to conduct with agencies' assessments of their own ESPCs and the
additional information we obtained from agency files and through more than
60 interviews with officials from the agencies, energy services companies,
and financiers. We also reviewed relevant regulations, policies, and
agency procedures. For more information regarding the scope and method we
followed, see appendix I. We conducted our work from January 2004 through
May 2005 in accordance with generally accepted government auditing
standards.

Results in Brief	In fiscal years 1999 through 2003, 20 federal agencies
undertook a total of 254 ESPCs to finance investments in energy-efficiency
improvements for up to 25 years. However, we could not determine the full
extent of ESPC use because there is no comprehensive database on federal
agencies' use of ESPCs. Although DOE is required to report to the Congress
some governmentwide annual data on the new ESPCs that agencies undertake
each year, DOE's data are not comprehensive or cumulative. The 20 agencies
for which we do have data have committed the federal government to annual
payments totaling about $2.5 billion over the terms of these contracts,
conditional on either the savings guaranteed in the contracts being
verified or as stipulated in the contracts.3 The energyefficiency
improvements have been or are in the process of being installed at
locations across the nation and cover many types of equipment including
lighting, boilers, geothermal heat pumps, and energy management systems.
The extent of ESPC use has varied across agencies. For example, the
Department of Defense (DOD) agencies undertook about 153 ESPCs to finance
about $1.8 billion in costs at about 100 military installations, while the
Department of Justice undertook only 2 ESPCs to finance about $43 million
in energy-efficiency improvements. Department of Defense officials

3By law, payment to an energy services company must reflect the savings
guarantee. Because energy services companies are accountable for
guaranteeing the performance of the equipment installed, if savings are
reduced due to equipment performance, the company must correct any related
problems. In some instances the contract may stipulate an amount of
savings that will be achieved. In the event that this stipulation
overstates actual savings, the agency must still make payments based on
the amount of savings stipulated. However, if stipulation understates
savings, the agency obtains the additional savings at no additional cost.

told us they relied on ESPCs to augment the upfront funding they receive
to purchase such improvements and achieve their energy efficiency goals.
Justice officials told us they undertook two projects to help meet similar
goals.

Agencies expect to achieve energy savings worth at least $2.5 billion over
the life of their ESPCs, as well other benefits that we could not attach a
dollar value to, including improved ability to accomplish their missions
by replacing aging infrastructure and environmental benefits from using
newer and cleaner technologies. Agencies also generally expect benefits to
continue after the contracts end because the improvements financed by the
ESPCs should operate and continue to save energy beyond the point at which
they have been paid for. While these benefits could be achieved using
upfront funding with associated financial cost savings to the government,
agencies told us they generally have not received appropriations for these
types of investments in sufficient amounts to achieve their energy savings
goals and maintain their energy infrastructure in a timely manner.
Therefore, they stated that meeting their energy savings and other goals
often depends on using ESPCs to supplement the upfront funding. In
addition, agencies and industry experts told us that ESPCs provide
benefits that are not typically obtained when agencies use upfront funding
to purchase the investments. For example, ESPCs shift some of the risk
from the government to the energy services companies by making payments
conditional on verification of expected performance, which in turn yields
energy savings. Such performance clauses are not generally included when
agencies purchase improvements using upfront funds, though it might be
possible to do so. Agency officials also said using ESPCs enabled them to
develop an integrated approach to energy management in their buildings by
ensuring, for example, that new and existing equipment work together
efficiently. In contrast, they said that obtaining up-front funding is
uncertain and episodic, making it difficult to ensure that improvements
work effectively together and with existing equipment.

Agencies believe that ESPCs' financial savings generally cover the costs,
and they provided examples of when this has occurred; however, the
available data are not conclusive and our work, agency audits of ESPCs,
and agencies' different interpretations about the components of costs that
must be covered by savings under the ESPC legislation raise questions
about how consistently savings actually cover costs. The ESPC legislation
requires agencies to design their ESPCs so that the upfront estimates of
savings exceed the costs. In addition, payments on the contracts are
conditioned on the savings guaranteed in the contracts being verified.

Although the agencies in our review told us about projects for which
savings covered costs and provided data on verified savings for most of
their projects, the data were not sufficient for us to conclude whether
project savings have covered costs. Furthermore, we found instances that
caused us to question whether savings consistently cover costs. For
example, a 2002 Army audit of a 1999 project covering five locations found
that the project's guaranteed savings were based on faulty assumptions,
potentially leading to payments of about $96 million that may not be
covered by savings if corrections are not made and if the contract is not
renegotiated. Finally, the agencies have adopted different interpretations
of which costs must be covered by savings under the ESPC authorizing
legislation. In practice, it remains uncertain whether contract payments
may be made only from utility savings resulting from the ESPC or from
funds already earmarked for equipment replacement and other sources to
reduce the length of the contract and finance charges. As a result,
agencies expressed the need for legislative clarification in this area.

During our review, the expertise and information needs of the agencies and
competitiveness issues related to the contracts emerged as concerns for
the protection of the government's financial interests in using ESPCs.
First, according to a number of the agency officials we interviewed, they
often lacked the necessary technical and contracting expertise and related
information to effectively develop and negotiate the terms of ESPCs and to
monitor contract performance once the energy-efficiency improvements were
operating. Even when the officials obtained assistance from the Department
of Defense's and FEMP's contracting centers, which the officials generally
believed to be helpful, they told us they sometimes could have benefited
from additional help with some aspects of developing the contracts, such
as evaluating the proposed financing, and monitoring savings during the
term of the contract. However, for various reasons, such as resource
constraints, they did not always get that assistance. As a result, they
sometimes relied on the energy services companies for help in these areas,
thereby calling into question whether they negotiated the best contracts
and ensured that the savings guaranteed by the contracts were realized.
The officials lacked necessary expertise largely because they were
inexperienced with ESPCs. They lacked necessary information because
information on ESPCs negotiated in the past is generally neither collected
and disseminated above the individual project level; nor is it required to
be. In addition to their concerns about expertise and information, agency
officials believe they may be paying too much for financing and other
terms in the contracts, in part, because there may not be enough
competition among the companies that finance ESPCs and among the energy
services

companies. One reason for lack of competition among financiers may be the
limited number of companies involved in financing ESPCs. Another reason
may be the risk associated with financing ESPCs because of the performance
requirements-risk that tends to limit the number of financiers interested
in participating. Regarding insufficient competition among energy services
companies, most officials believe that the super ESPCs, including their
lists of prequalified companies, are outdated and the contracts should be
put out for recompetition more frequently. The individual agencies and the
contracting centers have taken a number of steps to address concerns about
expertise, information, and competition. For example, in 2000, DOE began
requiring that each of its departmental projects be approved by a team of
experts in headquarters, and each of the contracting centers has developed
guidance for verifying actual savings. In addition, the agencies have
begun to address some of these concerns more collectively through an
interagency steering committee. We did not attempt to assess the
effectiveness of the agencies' efforts.

To strengthen the ESPC process, we are recommending that the Congress
consider clarifying the components of costs that must be covered by
savings in the statute relevant to ESPCs. We are also making
recommendations concerning the use of data, expertise, audits, and
competition to the heads of the agencies that use ESPCs; to the
Secretaries of Defense and Energy because the contracting centers answer
to them; and to the Secretary of Energy because of that agency's ESPC
oversight and reporting responsibilities.

In commenting on a draft of this report, the Departments of Defense (for
the Departments of the Air Force, the Army, and the Navy), Energy,
Justice, and Veterans Affairs, and the General Services Administration,
all stated their concurrence with our findings, conclusions, and
recommendations and provided technical and clarifying comments, which we
have incorporated, as appropriate.

Background	Federal agency use of ESPCs was authorized by the Congress to
provide an alternative to direct appropriations for funding
energy-efficiency improvements in federal facilities.4 Many agencies were
hard-pressed to pay for planned maintenance and repairs in their
facilities, let alone make more significant building improvements. As a
result of this situation, many federal facilities were in a state of
deterioration with agencies estimating restoration and repair needs in the
tens of billions of dollars. Although energy-efficiency improvements were
likely to save money over the life of the investments and replace aging
infrastructure, budgetary constraints prevented agencies many times from
receiving appropriations for such investments. Under the ESPC legislation,
agencies could take advantage of private-sector expertise, often lacking
at the agencies, with little or no upfront cost to the government. Under
these contracts, private-sector firms are supposed to bear the risk of
equipment performance in return for a share of the savings. This
arrangement permitted agencies to meet mission requirements and upgrade
their energy efficiency to reduce energy usage at the same time, while
recognizing only the first year's cost upfront in the budget. The Congress
authorized agencies to retain some or all of any annual savings available
after required contractual payments to the energy services companies have
been made.5

ESPC Process	To begin an ESPC project, agency officials work on their own
or with the assistance of one of the federal contracting centers at the
U.S. Air Force, the U.S. Army Corps of Engineers' Huntsville Center, the
Navy, or FEMP, to

4ESPCs were first introduced under the Comprehensive Omnibus Budget
Reconciliation Act of 1985, Pub. L. 99-272, which amended the National
Energy Conservation Policy Act. Agencies' authority to use ESPCs was
further extended under the Energy Policy Act of 1992, Pub. L. No. 102-486,
to authorize agencies to use energy savings performance contracts as a
tool for implementing energy-efficiency improvements. Prior to the Energy
Policy Act, the Federal Energy Management Improvement Act of 1988 mandated
a 10 percent reduction in energy used per square foot in federal buildings
between 1985 and 1995. Executive Order 12759 issued April 17, 1991,
extended these reduction requirements to the year 2000, requiring a 20
percent reduction from 1985 levels. These requirements were incorporated
into the Energy Policy Act of 1992 (42 U.S.C. S: 8253 (a) (1)). Executive
Order 12902 issued March 8, 1994, increased the reduction to 30 percent
per gross square foot by 2005 compared to 1985 to the extent that the
improvements are cost effective, and Executive Order 13123, issued June 3,
1999, extended this further to 35 percent by 2010.

5Currently DOD and GSA may retain and use 100 percent of all savings
without further appropriation. Other agencies can retain 50 percent of
savings and must return the other 50 percent to the Treasury.

choose an energy services company for the project and to identify the
energy-efficiency improvements the company will finance for the agency.6
Usually, multiple companies submit initial proposals that include
information on their qualifications and preliminary cost and savings
projections for the project. During this phase, all costs are borne by the
companies.

To continue developing the project, the agency chooses one company and
agrees to pay for a detailed energy survey. According to contracting
center officials, this survey typically takes up to 1 year and includes
such items as an assessment of baseline energy use and cost, projections
of energy use and savings once the improvements have been put in place,
maintenance schedules, and prices. Improvements must be "life-cycle cost
effective," that is, the benefits must meet or exceed total costs over the
contract. Determining life-cycle cost effectiveness is an agency
responsibility, but the agency can request this service from the company,
generally for a separate fee. A final proposal that includes the detailed
survey becomes the basis for comment and negotiation between the agency
and/or contracting center and the company. Included in these negotiations
are such contract terms as the "markups" added to the direct cost of each
improvement to cover the energy services company's indirect costs and
profit associated with its implementation,7 operations and maintenance
arrangements, guaranteed savings amounts, financing, and methods to verify
that savings are achieved.

Once the agency and energy services company have reached final agreement
on contract terms, the company designs and installs the energyefficiency
improvements and tests the improvements' operating performance. Agency
officials review test results and have the company make any necessary
corrections. To install, test, and accept the improvements typically takes
up to 2 years to complete. Upon accepting

6ESPC contracting assistance from FEMP and the U.S. Army Corps of
Engineers is available to all agencies, although the contracting
assistance from the Air Force is available only to Air Force installations
or military tenants located there. Currently, the Navy and Marine Corps
use the FEMP super ESPCs for their ESPC projects, and because the Navy has
centralized technical and contracting support staff who are familiar with
ESPCs, the Navy and Marine Corps use the Navy technical and contracting
staff to provide most of the support for Navy and Marine Corps ESPCs.

7Markups are expressed as a percentage of the cost of a particular
energy-efficiency improvement.

the project, the agency starts payments to the company, which must be
supported by regular measurement and verification reviews.

Although agencies may develop an ESPC themselves, doing so can be a
complicated process; consequently, most agencies seek assistance from one
of the contracting centers at DOD or FEMP. To streamline the procurement
process, three of these contracting centers-Air Force, U.S. Army Corps of
Engineers' Huntsville Center, and FEMP-have awarded super ESPCs, from
which multiple projects can be developed, to prequalified energy services
companies in different regions of the country.8 The super ESPC awards to
selected energy services companies complied with Federal Acquisition
Regulation rules and requirements for competition. With these
multiple-award contracts in place, agencies can implement ESPCs in a
fraction of the time it would take to undertake an ESPC alone because the
competitive process to select qualified companies has been completed and
key terms of the contract broadly negotiated, such as setting maximum
markups the companies may charge. In addition to managing the super ESPCs,
the contracting centers support agencies in negotiating aspects of
specific projects for a separate fee. For example, FEMP provides
facilitation services, where a third party assists the agency and energy
services company in agreeing on terms such as markup rates, financing
options, and the appropriateness of plans to measure and verify savings
for proposed improvements. In addition, FEMP issues guidelines, offers
training, and provides other support to agencies using the FEMP super
ESPC.

ESPC Savings Are Intended to Cover Contract Costs

Under an ESPC, company-incurred costs are paid from savings resulting from
improvements during the life of the contract. These savings include such
things as reductions in energy costs, operation and maintenance costs, and
repair and replacement costs directly related to the new efficiency
improvements. In addition to direct costs for the improvements, other
costs that savings should cover include financing charges, monitoring
services, and company-provided maintenance. Savings to an agency must
exceed payments to the energy services company. By law, aggregate annual
payments by an agency to both utilities and energy services companies
under an ESPC may not exceed the amount that the agency would have

8DOE has certified as prequalified energy services companies in six
regions of the country; the Air Force has prequalified companies in six
regions; and the U.S. Army Corps of Engineers has two major contracts: a
46-state and a 4-state contract.

paid for utilities without the ESPC. To ensure that energy savings cover
the contract costs, companies are required to guarantee the performance of
the new equipment and assume the risk for its operation and maintenance
during the contract, even though the agency may perform the maintenance.
Agencies still assume some risks, for example, for changes in utility
rates and in hours of operation, over which the energy services company
has no control.

To measure and verify that the guaranteed savings are achieved, an agency
compares baseline energy usage and costs prior to the ESPC with
consumption and costs after the improvements have been installed.
Typically, the company develops a baseline during its detailed survey,
while the agency is responsible for ensuring that the baseline has been
properly defined. The company then estimates the energy that will be saved
by installing the improvements and calculates the financial savings
expected in the future. At least annually, and sometimes more often, the
company provides measurement and verification inspections and reports to
the agency to substantiate the expected savings.

Several measurement and verification protocols are available to determine
energy savings. For example, under FEMP guidelines, four options are
discussed that range in complexity and costs. The simplest, and perhaps
least expensive, option is to measure the capacity or efficiency of the
new equipment and "stipulate" hours of operation, expected energy
consumption, and other factors rather than specifically measure them. Such
stipulation is often used for simpler improvements, such as lighting. A
more costly option might include constant monitoring of energy usage
through metering or computer simulation models of whole building energy
consumption. These methods may involve metering performance and operating
factors before and after the installation of the improvements. When
choosing among the alternatives, agencies balance the need for accuracy of
their estimates with the costs of verifying those estimates. As part of
its guidance, FEMP includes a matrix that describes a number of factors
and associated risks involving financial, operational, and performance
issues. When guaranteed savings are not achieved directly

due to the performance of the equipment, the agency may withhold payment
from the energy services company until the conditions are corrected.9

Prior GAO Work Compared the Financing Costs of ESPCs with Upfront Funding

As we reported in December 2004, while ESPCs provide an alternative
financing mechanism for agencies' energy-efficiency improvements, for the
cases we examined, such funding was more expensive than using timely
upfront appropriations. This is because the federal government is able to
obtain capital at a lower financing rate than private companies can. In
this regard, our earlier work examining six projects found that financing
these projects with ESPCs cost 8 to 56 percent more than had the projects
been funded at the same time with upfront funds.10 The report noted that
other factors, such as required measurement and verification of savings,
may also affect the cost of projects financed with ESPCs. Agency officials
commenting on this work agreed that timely upfront appropriations would be
less costly than privately financing energy-efficiency improvements, if
such appropriations were available, but stated that any delays in funding
would result in a subsequent loss of energy and cost savings and these
losses over time could offset the lower financing costs of the upfront
funding. We did not analyze the likelihood nor the costs of such delays.

                               Many Agencies Used
                              ESPCs, Although the
                              Extent of Use Varied

During fiscal years 1999 through 2003, numerous agencies undertook ESPCs
to finance energy-efficiency improvements, committing the federal
government to annual payments totaling about $2.5 billion over the terms
of these contracts. The use of ESPCs has been geographically widespread,
with many types of equipment installed, and the extent of use has varied
across the agencies.

During our review, we found that there is no source of comprehensive data
on federal agencies' use of ESPCs, either in DOE, the contracting centers,
or the agencies. DOE is required to collect data on the numbers, costs,
and expected energy and financial savings for the new ESPCs that agencies

9By law, payment to an energy services company must reflect the savings
guarantee. Since energy services companies are accountable for
guaranteeing the performance of the equipment installed, if savings are
reduced due to equipment performance, the company must correct any related
problems.

10GAO-05-55.

undertake each year and report these data annually to the Congress. The
data in DOE's reports, however, were not adequate for our review for
several reasons: they did not include some critical elements, such as
actual energy savings; they were not cumulative from year to year; and
they did not include ESPCs begun in fiscal year 2003 because DOE has not
yet issued the report for that year. Similarly, the DOD and FEMP
contracting centers' data were not comprehensive enough for our purposes.
The centers' data were limited to those contracts for which they provided
assistance; like DOE's reports, they did not include certain critical
elements; and, with the exception of Navy's, did not incorporate
information on modifications or progress on the contracts past the point
at which the centers' assistance to the agency was completed-usually only
up to 1 year after the contract was signed. Furthermore, most agencies do
not have a comprehensive, centralized electronic or paper system for
tracking their ESPCs and keep some contract data only in project files at
the facilities where the contracts are being implemented.

Consequently, to examine ESPC use across the federal government, we
obtained data from the four contracting centers and from the seven
agencies included in our review. We combined the data from all the
agencies into a consistent format, deleted duplicate records, and
performed basic tests to ascertain the reliability of the data. Although
the data for some projects were incomplete, the overall results of our
analyses appear to be consistent with information published from other
sources. The results of our analyses follow.

Twenty Agencies Used ESPCs

During fiscal years 1999 through 2003, 20 agencies undertook 254 ESPC
projects to finance investments in energy-efficiency improvements. The
ESPCs commit the federal government to annual payments totaling about $2.5
billion over the terms of these contracts, conditional on either the
savings guaranteed in the contracts being verified or as stipulated in the
contracts. Because energy services companies are accountable for
guaranteeing the performance of the equipment installed, if savings are
reduced due to equipment performance, the company must correct any related
problems. In some instances, the contract may stipulate an amount of
savings that will be achieved. In the event that this stipulation
overstates actual savings, the agency must still make payments based on
the amount of savings stipulated. However, if stipulation understates
savings, the agency obtains the additional savings at no additional cost.
Table 1 shows the numbers and costs of ESPCs the 20 agencies undertook, as
well as the percentage of total ESPCs attributable to each agency.

Table 1: Number and Cost of ESPC Projects Undertaken in Fiscal Years 1999
                                  through 2003

                                                                     Agency's 
                                                           percentage of $2.5 
                     Percentage of  Number of Cost to be     billion in total 
                                              paid                       cost 
          Number of total number of projects        over      to be paid over 
                                      with      contract 
Agency  projects        projects cost data       term        contract term 

                             Department of Defense

                Air Force                63    24.8       63   760,012,668    
                  Army                   47    18.5       46   324,374,960    
      Navy, including Marine Corps       40    15.7       35   653,376,185    
           Other DOD agencies            3      1.2        3    21,040,420    
      Subtotal for Defense agencies     153    60.2      147  1,758,804,233   
     General Services Administration     30    11.8       30   222,500,840    
     Department of Veterans Affairs      24     9.4       17   146,818,918    
          Department of Energy           10     3.9       10    38,076,458    
      Department of Transportation       8      3.1        8    56,516,373    
         Department of Interior          5      2.0        5    26,787,215    
           Department of Labor           4      1.6        4    11,543,796    
     National Aeronautics and Space                          
             Administration              4      1.6        4    54,300,894    
     Department of Health and Human                          
                Services                 3      1.2        3    20,004,872    
      National Archives and Records                          
             Administration              3      1.2        3    14,762,964    
        Department of Agriculture        3      1.2        3    37,046,526    
          Department of Justice          2      0.8        2    42,984,767    

         Department of Commerce          1    0.4       1   8,689,639     0.4 
     Environmental Protection Agency     1    0.4       1   8,687,513     0.4 
    Kennedy Center for the Performing                                    
                  Arts                   1    0.4       0      NA a      
         National Gallery of Art         1    0.4       1   5,108,785     0.2 
           Department of State           1    0.4       1   12,847,527    0.5 
                  Total                 254         240 a 2,465,481,320  

Source: GAO's analysis of ESPC data reported by the four ESPC contracting
centers and seven individual agencies included in GAO's review.

aOf the 254 projects agencies reported undertaking, the agencies reported
cost data for 240. We did not receive cost data for 1 Army project, 5
Marine Corps projects, 7 Veterans Affairs projects, or the Kennedy
Center's project. Furthermore, the agencies reported estimated savings for
only 237 of the 240 with cost data. To allow a fair comparison of costs
(shown in table 1) to savings (shown in table 2), we calculated total cost
for only the 237 projects with both cost and savings data. As a
consequence, we have understated total cost by the costs of the 3 projects
for which we did not receive savings data and by the costs of the
additional 14 projects for which we received neither cost nor savings
data.

The size of ESPC projects varied greatly over the 5-year period, ranging
from $241,943 to $137,515,074. About 72 percent of the projects in this
time period are valued at $10 million or less, as shown in figure 1. The
contract length of all ESPC projects ranges from 5 to 25 years, with an
average of 15.8 years.

Figure 1: Percentage of ESPC Financed Projects, by Contract Value,
Undertaken in Fiscal Years 1999 through 2003

<$5 million

>$5 million to $10 million

>$10 million to $15 million

>$15 million to $20 million

>$20 million >$10 million

<$10 million

Source: GAO's analysis of ESPC data reported by the four ESPC contracting
centers and seven individual agencies included in GAO's review.

Using the ESPCs, agencies financed energy-efficiency improvements that
have been or are in the process of being installed at locations in 49
states and on U.S. military installations in Guam, Cuba, Italy, Germany,
and Korea. Numerous types of energy-efficiency improvements were financed,
including replacement of boiler and chiller plants for heating and
cooling, energy management control systems, geothermal heat pumps, and
lighting. In the largest ESPC project during the 5-year period, the Marine
Corps committed to spend almost $138 million at a facility in California
to install a cogeneration plant, solar hot water and photovoltaic systems,
heating, ventilating, and air conditioning at various sites, and waste
water pump

upgrades. This ESPC project, awarded in July 2002, has a contract term of
18 years.

Extent of ESPC Use Varied Across Agencies

The extent to which agencies have used ESPC financed projects has varied,
as shown in table 1. DOD agencies have used the contracts the most,
undertaking about 153 ESPCs to finance about $1.8 billion in costs at
about 100 military installations during the 5-year period. DOD officials
told us they relied heavily on ESPCs to achieve energy infrastructure
improvements, in part because of difficulties they encountered in
obtaining adequate upfront funding for energy projects that were not
categorized as being mission-critical. They noted that these improvements
also helped the agencies meet other national energy goals as well.

After DOD, the General Services Administration (GSA) and Veterans Affairs
(VA) used ESPCs the most during the 5-year period, undertaking 30 and 24
projects, respectively. Together these agencies account for about 21
percent of projects. Both GSA and VA officials told us that adequate
upfront funding for their energy projects has been difficult to obtain in
recent years. At the same time, they have faced increasing backlogs of
these projects in their capital management plans. Consequently, the
agencies have moved toward using more ESPCs to meet mandated energy
reduction goals and to make badly needed upgrades to aging and inefficient
equipment.

DOE's departmental ESPC projects represent about 4 percent of the total
projects undertaken over the period, valued at about $38 million. DOE
officials told us that the agency has mainly used ESPCs since 1999 to
supplement limitations in upfront funding for energy-efficiency projects.
After GSA and VA, among civilian agencies, DOE has a high percentage of
federal facility square footage; however, the agency has not been among
the largest users of ESPCs for two reasons. First, the agency has found it
relatively easy to meet its mandated energy reduction goals because it has
in recent years closed a number of its facilities, such as those producing
nuclear weapons, that were no longer needed. Furthermore, many DOE
facilities have negotiated low utility rates or are in regions of the
country where utility rates are relatively low. This makes developing an
ESPC for which savings will cover costs difficult, because the low utility
rates hold down the amounts that can be saved with the energy-efficiency
improvements. As a result, DOE's major goal in using ESPCs, we were told,
has been for energy infrastructure improvement.

Of the seven agencies in our review, the Department of Justice (Justice)
used ESPCs the least, undertaking only two ESPCs totaling about $43
million in costs. According to Justice officials, because many of their
facilities are prisons, security concerns can make undertaking
energyefficiency projects on existing buildings difficult. Nonetheless,
the agency undertook two ESPC projects in 2003, one each under the Bureau
of Prisons and the Federal Bureau of Investigation. According to the
officials, the agency undertook the ESPCs because it was concerned about
meeting the mandated energy reductions, and upfront funding for
energy-efficiency projects was decreasing. In addition, for one of the
projects, the agency saw a chance to use an ESPC to accomplish
environmental goals established by Executive Order 13123, such as making
more use of renewable energy. In that case, the agency undertook a project
at a California prison site. After the California energy crises in 2000
and 2001, the agency sought to decrease its dependence on the electricity
grid, so the project included installation of renewable energy sources,
including a wind turbine and photovoltaic panel, which furthered the
agency's energy security interests as well as helping it meet its energy
reduction and environmental goals.

Finally, five agencies-the Departments of Commerce and State, the
Environmental Protection Agency, the John F. Kennedy Center for the
Performing Arts, and the National Gallery of Art-that we did not contact
for additional information for our review each undertook one project
during the 5-year period. We did not receive cost data for the Kennedy
Center. The other four totaled about $35 million in costs.

Agencies Increasingly Used FEMP's Services

Figure 2 shows agency use of the contracting centers at the Air Force, the
U.S. Army Corps of Engineers' Huntsville Center, the Navy, and FEMP for
fiscal years 1999 through 2003. With the exception of 2002, the data show
that, over the period, agencies increasingly used FEMP's contracting
center more relative to the other agencies' centers. Although there was an
average of 51 ESPC financed projects undertaken each year, there was a 54
percent increase in projects awarded from 2002 (37 projects) to 2003 (57
projects). According to agency officials, this increase was largely
because agencies put significant effort into awarding ESPC financed
projects, anticipating the sunset of the legislation on October 1, 2003.
This was particularly true for ESPCs done through FEMP's contracting
center. As discussed previously, on October 28, 2004, ESPC authority was
renewed through fiscal year 2006.

Figure 2: Agencies' Use of Contracting Centers for ESPCs Undertaken in
Fiscal Years 1999 through 2003

Number of ESPCs 50

40

30

20

10

0 1999 2000 2001 2002 2003 Fiscal year

Air Force Army Corps of Engineers FEMP Navy

Source: GAO's analysis of ESPC data reported by the four ESPC contracting
centers and seven individual agencies included in GAO's review.

  Agencies Expect ESPC-Financed Projects to Result in Energy Savings As Well As
  Other Benefits

ESPCs awarded by federal agencies to finance energy-efficiency
improvements are expected to achieve energy savings worth at least $2.5
billion during the life of their contracts. Agencies estimate that they
are annually reducing energy use by at least 9 million MMBTUs.11 Some
savings are also expected to continue after the ESPCs end. Agencies
receive other benefits through ESPCs as well, such as environmental
improvements and better mission capability resulting from replacing aging
infrastructure with more reliable equipment. Although these benefits could
be achieved through up-front appropriations at a lower cost, this funding
has often not been available on a timely basis. Furthermore, ESPCs provide
additional benefits not typically associated with investments purchased
through upfront appropriations, such as shifting some of the performance
risk of the equipment to the energy services companies and allowing
agencies to more easily combine multiple energy-efficiency improvements
into an integrated package.

ESPC-Financed Projects Have Reduced Energy Use and Agencies Expect to
Achieve Energy Savings Worth At Least $2.5 Billion

Over the life of the ESPC financed projects included in our review,
agencies expect to achieve energy savings worth at least $2.5 billion and
amounting to over 9 million MMBTUs, as shown in table 2. These estimated
savings are likely to be understated because the agencies did not report
financial savings for 17 projects and energy savings for 45 projects. The
military services account for about 64 percent of the financial savings
and about 71 percent of energy savings for the ESPCs awarded during the 5
years. Savings at some specific locations are expected to be substantial.
For example, reported data show that total estimated savings at each of
three military installations will exceed $100 million, ranging from $117
to $138 million for a total of $378 million. The ESPC at Elmendorf Air
Force Base in Alaska is expected to reduce the base's energy consumption
by more than 1 million MMBTUs per year, which are valued at $123 million
for the 22-year contract term. According to the base energy manager, this
is the largest ESPC ever awarded by the Air Force.

11MMBTU stands for million British thermal units and is a standard unit
used to measure energy usage. The estimated 9.1 million MMBTUs in energy
savings from ESPCs is equal to the annual energy needed for about 98,000
households, at an average of about 92 MMBTUs per household per year.

  Table 2: Energy and Financial Savings for ESPC Projects Undertaken in Fiscal
                            Years 1999 through 2003

Agency

                               Number of projects with financial savings data

                                                  Number of 
             Estimated                        projects with                   
                       Percentage of              Estimated     Percentage of
            cumulative       financial     estimated annual         estimated 
             financial                               energy 
     savings over life savings for all energy savings          energy savings 
     of                                savings in           
              contract       contracts          data MMBTUs for all contracts 

                             Department of Defense

                       Page 20 GAO-05-340 Energy Savings
 Air                                                                                                                 Other                              Subtotal                                                       General                                                                       Veterans                                  Department                                                                                          Department                            Department                             National                                                                                   and                               National                                                                                                                      Department                                                                            NA                        Protection                              Kennedy          the        NA       NA    
Force 63 750,533,703 30.0 60 3,448,867  Army 44 334,403,496 26.7 34 383,674  Navy 35 667,164,060 13.4 38 1,866,509    DOD    3 21,089,559 0.8 3 89,065    for          agencies 145 $1,773,190,818 70.9 135 5,788,115  Services       Administration 30 233,000,518 9.3 30 697,413  Department       Affairs  16 154,879,631 6.2 13 1,887,625  of Energy  10 38,099,795 1.5 10 271,403  Department       Transportation 8 57,161,461 2.3 3 49,233      of     5 26,572,468 1.1 5 75,292   of Labor  4 11,602,330 0.5 2 20,489  Aeronautics        and        Administration 4 54,567,011 2.2 3 167,833  Department        Human   3 20,033,135 0.8 1 20,144  Archives       Records       Administration 3 13,636,305 0.5 1 4,962  Department       Agriculture 3 39,267,423 1.6 2 32,329  of Justice 2 43,008,699 1.7 2 26,994 0.3 Department       Commerce 1 8,689,649 0.3 0a a  -- Environmental         Agency   1 8,966,682 0.4 1 24,900 0.3 Center        Performing 0a  a -- 0a a  --
                                                                                                                    agencies                            Defense                                                                                                                         of                                                                                                  of                                                      Interior                                                                                     Space                                                  of Health        Services                              and                                                                     of                                                                                               of                                                                                                           for            Arts                   

(Continued From Previous Page)

           Number of                                                 
            projects                            Number of            
                with       Estimated Percentage  projects  Estimated Percentage 
                                     of              with                    of 
           financial cumulative       financial estimated annual      estimated 
                     financial                            energy     
             savings savings over    savings    energy    savings in     energy 
                     life of         for all    savings                 savings 
  Agency        data        contract  contracts      data     MMBTUs    for all 
                                                                      contracts 
National                                                             
Gallery of         1       5,184,179        0.2         1     22,796 
Art                                                                  
Department         1      12,847,609        0.5        0a       NA a         -- 
 of State                                                            
  Total          237 $2,500,707,713b                  209 9,089,527c 

Source: GAO's analysis of ESPC data reported by the four ESPC contracting
centers and seven individual agencies included in GAO's review.

aWe did not receive financial savings data for the Kennedy Center's
project. We did not receive energy savings data for the projects of the
Departments of Commerce or State, or for the Kennedy Center's project.

bAgencies reported financial savings data for 237 of the 254 projects;
consequently, the total financial savings reported here understates total
savings by the unknown amount of the savings of the 17 projects for which
we did not receive savings data.

cAgencies reported estimated energy savings to date for only 209 of the
254 projects, understating estimated savings achieved to date.

The installation of energy efficient equipment has already resulted in
some energy savings and is expected to result in further savings, lower
utility bills, and reduced operations and maintenance expenses. Over the
5-year period, the agencies estimate they reduced their energy use by at
least 9 million MMBTUs annually.12 According to agency officials, these
reductions have assisted, and will continue to assist, agencies in meeting
their mandated goals for reducing BTUs of energy used. For example,
agencies reported that they exceeded by 4 percent their goal for fiscal
year 2000-a 20 percent reduction in BTUs of energy consumed relative to
their fiscal year 1985 usage. Agencies report their progress in meeting
the goals by each agency as a whole and do not indicate the portion that
could be attributed to the agency's ESPCs. However, officials we
interviewed representing most of the agencies believe they would not have
met the 2000 goal without the contracts. Furthermore, they expect their
ability to meet the remaining goals-30 percent reduction by fiscal year
2005 and 35 percent by fiscal year 2010-depends largely on being able to
use ESPCs to finance energy efficiency improvements. DOD officials told us
that in recent years ESPCs have accounted for over half of DOD agencies'
annual energy savings. Furthermore, they believe that DOD will have
significant difficulty in achieving the 2005 energy reduction goal because
a number of ESPC projects planned for fiscal years 2004 and early 2005
were not

12The agencies reported estimated, rather than actual, BTUs saved.

undertaken because authority for ESPCs was suspended during that time. DOE
is an exception-according to DOE officials, the agency has already met its
goals for 2005 and 2010, largely because it has closed facilities that
produced nuclear weapons, thereby significantly reducing the energy
consumed by the agency.

Agencies may also benefit from substantial energy and financial savings
once the contracts are paid for. Energy and related financial savings
should continue beyond a project's payback period through annual energy
saving, as well as through reduced operations and maintenance costs.
Currently, financial savings retained by agencies are small because most
agencies use their savings to pay off their contracts with the energy
services companies as quickly as possible, thereby reducing debt more
rapidly and saving interest costs to the government. For example, GSA,
which currently pays energy services companies 98 percent of the agency's
annual financial savings from ESPCs, estimates that it will save about $16
million annually from its 30 projects after it has repaid the companies.
Similarly, data provided by the Air Force and the Navy show expected
annual financial savings for those agencies of almost $45 and $40 million,
respectively, once the contracts are paid for, and Army and Marine Corps
projects also expect to garner financial savings past the contract terms.
In another instance, officials at Fort Bragg told us that they would
continue to obtain lower utility rates, which were negotiated as part of
the ESPC by the energy services company, even after the contract period.

                ESPC Financed Projects Offer Additional Benefits

In addition to energy savings and lower overall utility costs,
ESPC-financed projects, like projects funded with upfront appropriations,
can provide agencies with environmental benefits through installation of
newer, cleaner technologies. The ESPC financed projects in our review, we
were told, are assisting the agencies in eliminating environmental
hazards, reducing outdoor air pollution, and improving indoor air quality.
The project at Elmendorf Air Force Base allowed the Air Force to replace
old steam plants insulated with asbestos, a known environmental hazard. In
another instance, in the ESPC at Portsmouth Naval Shipyard, in Maine, the
Navy installed a cogeneration unit for generating power. As a result, the
shipyard eliminated its reliance on bunker fuel oil and is producing
significantly fewer greenhouse gas emissions.

ESPC-financed projects also allow agencies to replace aging infrastructure
without having to obtain upfront appropriations. Officials at six of the
seven agencies in our review noted the importance of using ESPCs to

replace aging infrastructure. The upgrades, the officials told us,
improved the agencies' abilities to carry out their primary missions and
provide a more comfortable work environment for employees. At Elmendorf
Air Force Base, for example, the energy manager told us the base was able
to replace a 50-year-old cogeneration power plant with a new, much more
efficient decentralized natural gas system. Navy officials told us they
faced a similar situation with a power plant built in 1945, which was
failing at their Portsmouth facility. The backlog of maintenance work on
the power plant was continuing to increase. Due to the geographic location
in Maine, with severe winter weather and the continual repairs needed on
the old power plant, an upgrade was essential to support the nuclear
submarines at the shipyard. The officials noted each day's loss of power
cost the shipyard $1.5 million. By using an ESPC to replace the power
plant, the base was able to eliminate eight full-time staff positions
(saving about $448,000 annually) because the new power plant is easier to
operate and does not require frequent emergency maintenance, as the old
one did.

Upfront Funds Could Provide These Benefits But Are Often Not Available on
a Timely Basis

Although the benefits from ESPC financed projects discussed above could be
achieved using upfront funding, agencies have found that sufficient
amounts of such funding were generally not available-making it necessary
for the agencies to use ESPCs to supplement the upfront funding they
receive in order to obtain these benefits. A study by Oak Ridge National
Laboratory that compared ESPCs with upfront funded projects concluded that
when sufficient upfront funds are not available, the most expensive choice
may be to do nothing, allowing inefficient equipment to remain in service
and wasting funds on unnecessary energy use and emergency repairs and
replacement. Officials at six of the seven agencies we reviewed-the Air
Force, the Army, GSA, Justice, the Navy, and VA-told us that, in spite of
attempts to obtain upfront appropriations for energy projects, adequate
amounts of such funds were generally not available.13 For example:

o 	GSA officials said the agency received no funds for any
energyefficiency work included in their capital management plans for
fiscal years 2002 and 2003, although they requested $32 million and $8
million, respectively. As a result, they used other financing options,
such as ESPCs.

13Attempts to obtain appropriations included requesting funds in the
President's budget, from the Office of Management and Budget, or
internally within the agency.

o 	Army officials at Aberdeen Proving Ground noted that failing heating
and air conditioning systems in the base's family housing had become a
fire hazard and were too expensive to maintain. These officials said they
repeatedly attempted to obtain upfront appropriations for the upgrades
but, being unsuccessful, negotiated an ESPC.

o 	Navy officials told us their planned investments for energy-efficiency
projects range from $100 million to $150 million annually in order to meet
their BTU reduction goals. However, because the Congress will only provide
$50 million for all of DOD, and the Navy only gets about $15 million of
that amount-or none, as in fiscal year 2000-the Navy questions the
usefulness of requesting the funds while foregoing making
energy-efficiency improvements.

Furthermore, officials at both the VA and the Navy told us that even when
they can obtain upfront funds, the project typically takes 4 to 5 years to
obtain approval and be completed, compared with about 2 years for an ESPC.
Navy officials pointed out that up-front-funded projects take longer
because projects must be submitted 2 years in advance of the budget year;
in addition, they said that most projects are not fully funded and have to
be resubmitted in subsequent years. According to these and other agency
officials, their agencies were achieving savings through lower utility
bills and reduced operation and maintenance costs during the extra years
that equipment installed under ESPCs was operational. DOE's Oak Ridge
National Laboratory reported in March 2003 that, on average, upfront
funded projects that were approved took 63 months to award, design, and
construct, compared with 27 months for ESPCs.

In a recent report, GAO performed a case study analysis of six ESPC
projects and compared the actual costs of financing the energy-efficiency
improvements incurred in the ESPCs with an estimate of what the financial
costs would have been had the improvements been paid for through timely
upfront appropriations.14 We found that the financial cost to the
government of private financing was significantly higher than the
financial costs of upfront appropriations and also that monitoring and
verification costs-included with ESPCs but typically not included in
projects paid for with up-front appropriations-also added to the cost
difference between private versus upfront financing. Specifically, our
case studies found that

14GAO, Capital Financing: Partnerships and Energy Savings Performance
Contracts Raise Budgeting and Monitoring Concerns, GAO-05-55 (Washington,
D.C.: Dec. 16, 2004).

ESPC financed projects increased the government's cost of acquiring the
energy-efficiency improvements by 8 to 56 percent compared to timely,
full, upfront appropriations. Our analysis assumed that the energy savings
and other benefits associated with the energy-efficiency improvements were
independent of how they were financed.

While our earlier work found higher financing costs associated with the
use of ESPCs, a recent study of ESPCs, undertaken by the Lawrence Berkeley
National Laboratory, analyzed both the costs and government benefits of
109 ESPCs and compared the net benefits of these projects with the net
benefits under several alternative scenarios involving direct, upfront
appropriations.15 The study assumed that the performance of the equipment
installed was dependent to varying degrees on which financing method was
used. Specifically, they evaluated scenarios in which energy savings from
equipment installed using upfront appropriations decay over time (1 or 2
percent per year) because projects funded up-front typically do not
include the same level of monitoring and verification to ensure sustained
performance of the equipment. The study concluded that "delays of more
than one year in obtaining congressional appropriations result in reduced
net benefits relative to ESPC-financed projects." Although we did not
independently verify all of the study's assumptions, data, and results, we
did review several studies of energy audits that the Lawrence Berkeley
authors used to support their assumption regarding savings decay to verify
their assumption that energy systems' savings decay in the absence of
proper monitoring and verification. In discussions with experts on the
performance of energy equipment, we were told that many of the
energyefficiency improvements require careful monitoring and verification
to ensure that they perform up to their specifications and that, without
such monitoring and verification, energy savings would indeed decay over
time, in some cases very quickly; however, we found that agencies often
lack sufficient expertise in monitoring and verifying performance of
energy equipment on their own. Thus, although we could not conclude on the
actual extent of savings decay for upfront-funded projects, there is
evidence that savings decay occurs. While it is likely that agencies could
purchase monitoring and verification services from the private sector in
the case of equipment paid for with up-front appropriations, they have
typically not done so in the past and the additional cost of doing so is
unknown. We

15Ernest Orlando Lawrence Berkeley National Laboratory, Public and
Institutional Markets for ESCO Services: Comparing Programs, Practices and
Performance, LBNL-55002 (University of Calif. Berkeley, California; March
2005).

cannot conclude definitively the extent to which decreased savings decay
and other benefits from ESPC-financed projects may offset the significant
savings achieved from using upfront funding that we found previously in
six case studies.

Some ESPC Benefits Not Readily Available With Upfront Funding

ESPCs also provide two benefits not typically associated with investments
purchased through upfront appropriations: (1) some performance risk is
shifted from the government to the energy services companies and (2)
agencies find it easier to combine multiple energy-efficiency improvements
into an integrated package. First, as noted by agency officials and
industry experts, because ESPCs require energy services companies to
guarantee equipment performance over the lifetime of the contract, which
in turn yields energy savings, agencies benefit as these risks are shifted
from the agencies to the companies. As part of these guarantees, energy
services companies are ultimately responsible for insuring that adequate
operations and maintenance are conducted and for any repairing and
replacing equipment if it fails. These requirements reduce the risks from
possible faulty engineering, poor equipment installation, or equipment
failure. For projects funded with upfront appropriations, energy services
companies are generally only responsible for equipment risks during the
warranty period, which typically is shorter than an ESPC's contract
guarantee. While it may be possible to supplement upfront-funded projects
with additional warranty or performance coverage, agency officials told us
that this would add costs and typically is not done. According to FEMP
ESPC program managers, ESPCs create an incentive for energy services
companies to develop highly efficient improvements and maintain the
equipment so that it is in peak operating condition. This incentive occurs
because the companies' compensation is directly linked to the savings
achieved through their work. Officials from both the Navy and the Army
told us that because the value of energy savings must cover the annual
payments to the energy services company, the company bears the risk when
it encounters problems. For any problems related to the performance of the
equipment that are defined as company risks and that were not explicitly
determined to be an agency risk, the agency can withhold future payments
from the energy services company until the problem has been corrected.
Officials at Fort Bragg told us that they withheld payment from a
contractor for a short period until an equipment problem was fixed on
their ESPC. In many cases, the agency, rather than the energy services
company, performs the operations and maintenance. An official from the DOE
departmental energy management program, however, noted that it is not
altogether clear when a piece of equipment fails, whether payment to the
energy services

company can be stopped directly or whether a review of maintenance
records, for example must be performed to determine if the agency or the
company is responsible for the failure. Typically, when problems occur for
equipment purchased with up-front funds, if the warranty period is over,
the agency is responsible for fixing or replacing the equipment at its own
expense.

Second, with ESPC-financed projects agencies find it easier to bundle a
number of energy-efficiency improvements so they can function as an
integrated system. In this way, one energy services company is responsible
for the guaranteed performance of all the equipment. Agency officials told
us that, due to tight budgets, upfront funding is limited even when it is
available and the agency can typically install only a few of the necessary
energy-efficiency improvements. They said it may be years before the
agency receives authority to fund additional projects and, due to the
competition requirements of federal procurement practices, it is quite
possible a different energy services company would be selected to install
them. Besides potential problems of integrating the controls for system
components installed by two different companies, some savings that would
have been obtained if all energy-efficiency improvements had been
installed without delay at one time are lost. Energy savings can be
achieved more quickly through an integrated approach than implementing
efficiency improvements on a piecemeal basis. The lack of a performance
guarantee over the life of the equipment purchased with up-front funding
and the uncertain, episodic nature of upfront funding can make those
projects more risky and less capable of generating an integrated approach
to energy management for new and existing equipment.

  Agencies Believe Financial Savings Cover Costs, but Whether Savings Actually
  Do So Is Uncertain

Agencies generally believe that ESPCs' financial savings cover the costs
because they design their contracts to cover costs and because they must
obtain verification reports from the energy services companies that
confirm this point or take steps to correct shortfalls in savings. They
cited examples of projects that realized savings in excess of costs and
provided data on verified savings for most of their projects. However, the
data provided were insufficient to conclude whether savings covered costs
of the projects in our review. Furthermore, our work, agency audits of
ESPCs, and agencies' differing interpretations about the components of
costs that must be covered by savings caused us to question whether
savings consistently cover costs. FEMP officials recognize the difficulty
in ensuring that actual savings cover costs and have formed a special
working group to address uncertainties regarding savings.

Agencies Generally Believe That Savings Cover Costs Because They Are
Designed to Do So and Because Companies Must Verify Savings, but
Uncertainties Arise from Limitations of Available Data

In response to statutory requirements, agencies design ESPCs so that
savings are sufficient to cover costs. In addition, the agencies refrain
from committing themselves to ESPCs when they determine beforehand that
savings will be inadequate or when the payback will exceed their preferred
time frames for the contracts. For example, a DOE official cited several
departmental projects that advanced to the final proposal stage but that
the agency dropped because the economics for the projects were either poor
or the agency did not agree with the savings projections. For one project,
the low utility rate (which reduced the amount of savings that could be
accrued) and the high cost of performing the work in an area with access
controlled for security reasons forced the project's abandonment. In
another case, the agency did not agree with the company's projected
savings and believed that very little savings would be achieved. FEMP
officials noted a requirement for performing a life-cycle cost analysis of
individual energy-efficiency improvements, which are then bundled to
ensure that the project's overall savings cover costs.

Another reason for agencies' general confidence regarding savings is that
energy services companies are required to submit annual measurement and
verification reports confirming the savings and, in case of a shortfall,
take corrective steps to recoup the savings. These annual reports provide
the specific figures on which agencies base their payments to the energy
services companies. In some cases, the reports are updated quarterly to
give the officials monitoring the project more current data on the
performance of equipment, enabling them to spot shortfalls in savings and
have the energy services company correct them quickly. In addition, agency
officials cited projects that realized savings in excess of costs. For
example, the ESPC at Fairchild Air Force Base in Washington State has
garnered about $180,000 more per year than it cost. The extra savings have
resulted from the equipment operating more efficiently than estimated and
actual utility costs that were higher than estimated in the contract.

We asked the seven agencies in our review to provide data on verified
savings for each of their projects. In many cases, the projects have not
entered their performance periods, so verified savings data are not yet
available. To approximate the number of projects that should have verified
savings available, we looked at the 111 projects (about 44 percent of the
projects) that had been under way for 3 years or more and could reasonably
be expected to have at least 1 year of verified savings to

report.16 In this regard, the seven agencies reported verified savings for
most of the 111 projects, but they did not provide cost data that could be
compared with the annual verified savings. We did not take steps to obtain
the data, which are contained in files at projects located across the
country. Thus, we could not conclude from the data provided to us that
verified savings were, in fact, covering the costs of these projects.
Furthermore, while federal officials are expected to accompany energy
services company officials when the data are being gathered for the
reports to provide an extra level of confidence in the data's validity,
FEMP officials cautioned that this added check may not be happening as
often as it should.

An additional limitation of the data is that the measurement and
verification process relies not only on actual measurements but on
estimates as well. As will be discussed more fully later, estimates may be
used extensively in this process, introducing the possibility of incorrect
assumptions and errors in the calculations. Moreover, the process
evaluates not only the performance of the equipment, but additional
factors such as the cost of energy that affect actual savings.

Agency Officials and Audits Cite Projects for Which Savings May Not Cover
Costs

Agencies cited specific projects in which the savings have not covered
costs. According to a DOE departmental official, savings for 4 of its 10
projects have fallen short of costs because of unexpected problems. DOE's
analysis has shown that, in three of the four instances where savings are
inadequate, the shortfall has resulted from unpredictable mission changes
in the use of the facilities. For example, in one of these cases, the
discovery of beryllium contamination forced the closure of some of the
buildings involved in the contract. Reductions in electricity consumption
accounted for the fourth case. In this instance, in 7 out of 12 months
each year, DOE is not meeting the minimum required demand cost that was
projected and has to pay for the electrical demand it does not use. As a
result, for 7 months of the year, the new equipment associated with the
project is not providing any electrical demand savings, so the overall
cost savings of the equipment is less than expected. In general, according
to the DOE official, it is extremely difficult to accurately predict all
the variables that affect energy savings over the 10 to 15 year ESPC
contract term, so agencies have to bear some of the risk of inaccurate
assumptions at the outset.

16We were told that, on average, project construction/installation takes
up to 2 years to complete and be accepted by the agency, after which the
performance period begins.

While most agencies have not audited their ESPCs, the Army and Air Force
audits of ESPCs have found several instances in which savings may not have
covered costs. For example, a 2002 Army audit of a 1999 project covering
five locations found the Army could pay about $96 million that may not be
covered by savings over the 18-year life of the project because savings
that the Army agreed to were overestimated. First, the report found the
baselines were incorrect because the contractor inflated labor costs for
the operation and maintenance baseline by $66 million over the life of the
project. Second, it found the contractor also overstated the baselines for
electrical consumption and water conservation by more than $30 million
over the life of the project. This inflation of both baselines occurred,
according to the report, because the agency relied heavily on the
contractor to prepare them. Other major contributing factors, the report
stated, appeared to be insufficient time to review contract proposals and
a desire to award the contract and pay the contractor prematurely. As a
consequence, the report concluded, the agency could pay for nonexistent
savings over the term of the contract. As another example of questionable
estimates, the contractor for an Air Force ESPC increased the original
consumption baseline by over 11,000 kilowatts with no indication that Air
Force officials questioned this adjustment.

Poor documentation adds to the problem of ensuring that savings cover
costs. For example, energy services companies at 8 locations reviewed in a
2003 Air Force audit reported savings of $6.7 million associated with $78
million in ESPC investments, but civil engineering officials could not
provide support that they reviewed or validated these numbers. The
auditors projected the results for the sample of 8 locations to the 36
included in their review and concluded that the lack of documentation made
it impossible to assess the savings that the agency will receive for about
$600 million in costs for energy efficient equipment. As noted in the
report, this condition occurred because Air Force guidelines did not
specifically require maintenance of baseline supporting documentation, a
methodology for savings computation, or validation of cost savings. In
response to recommendations in the report, Air Force officials stated that
they were taking steps to correct these problems. However, we could not
determine the status of the payments for either the Army or the Air Force
projects in these audits because the audit documentation did not indicate
whether payments were made despite the potential savings shortfalls.

Difficulties in Calculating Savings, Ensuring Adequate Equipment
Maintenance, and Verifying Savings Also Contribute to Uncertainty About
Savings Covering Costs

Accurately calculating financial savings is fundamentally difficult for
agency officials to do. Major components of financial savings-baseline
energy consumption, the consumption once the energy efficiency
improvements have been installed, and the cost of energy associated with
both the baseline and the later consumption-are partly stipulated, or
estimated, rather than actually measured. In this regard, striking the
"right" balance between stipulation (which is less costly but also less
accurate) and measurement (which is more costly but also more accurate) is
a challenge for agencies. To the extent that stipulation is used in lieu
of actual measurement, according to DOE officials, savings calculations
may be based on inadequate data or incorrect assumptions, which contribute
to uncertainties about the actual savings.

Agency officials commented on how difficult it is to identify the
consumption and cost of energy, which forms the basic equation
(consumption times cost) that establishes the energy-related baseline and
the future financial savings. For example, contractual arrangements with
regard to consumption can affect the savings. In the case of "take-or-pay"
contracts, agencies may have to pay for a certain amount of projected
minimum demand even if they do not actually use it. As noted earlier, this
situation has occurred in one of DOE's ESPCs and has reduced its savings
to the point where savings will not cover payments under the contract. The
cost of energy, as shown in utility rates, can also be difficult to
determine. Given their potential complexity, it is easy for energy
services companies or federal officials to provide incorrect utility
rates, which in turn will have important consequences for the level of
savings. Rates need not only to be determined accurately to establish the
baseline but also projected as accurately as possible into the future to
determine eventual savings. These rates are projected 10 years into the
future in ESPC contracts, according to agency officials, but the actual
rates can change at any point during the contract period. Anomalies due to
weather, fluctuations in energy prices, or other influences can affect the
rates. In general, if utility rates go down or increase more slowly than
projected, then the actual savings will not materialize. In essence, these
rates are stipulated, and the agency bears the risk.

Ensuring that the equipment installed under ESPCs has been adequately
operated and maintained is essential for agencies and can affect whether
savings cover costs. According to an expert who has worked on Army and Air
Force ESPCs, calculations of guaranteed savings assume a high level of
operation and maintenance activities, but rapid loss of energy efficiency
if equipment output is not maintained can jeopardize savings. He said that

typically a 10 to 20 percent degradation in savings occurs annually on a
given ESPC in the event of improper operation and maintenance. He cited
the virtual ruin of a chiller (for air conditioning) in only 3 years as a
result of improper maintenance.

Similarly, measurement and verification are critical in the longer term
for achieving guaranteed savings. In determining these savings, however,
energy services companies blend the use of measurement with stipulations
in their reporting process. The expert who has worked on Army and Air
Force ESPCs noted that, despite the importance of using measurement in
addition to stipulation, there are numerous barriers to performing actual
measurements. These barriers include a lack of appropriate metered
equipment and reluctance by energy services companies to perform
measurement and verification because it might work against their
interests. An Air Force official observed that in his experience energy
services companies prefer stipulation and have limited the number of
actual measurement for projects as much as possible. A report for FEMP
examining seven ESPCs noted the reliance on stipulation in the projects'
measurement and verification plans. The primary reason for using this
method was its low cost; however, the report concluded that the large use
of stipulated savings left the agencies at risk of unrealized savings. To
help agencies use stipulation correctly, in 2000, FEMP issued supplemental
guidance on measurement and verification that specifies that some
stipulation may be used in lieu of measurement when there is a reasonable
degree of certainty about the stipulated values, their contribution to
overall uncertainty is small, and they are based on reliable and
documented sources of information.

A Special Working Group Has Been Formed to Address These Uncertainties

DOD and FEMP recently established a special working group to address the
uncertainties about actual savings. The Energy Savings Discrepancy
Resolution Working Group, formed in late 2004, is developing approaches to
compare projected and actual savings and to explain any deviations.
Because it has just commenced these studies, the group has obtained
preliminary results regarding only one project. The group found that the
projected savings for this project were diminished by consolidations of
agency missions, expanded construction, and new demands for energy that
had nothing to do with the ESPC. Officials said they chose this project
because it came with a well-developed baseline, which is often not
available for careful evaluations of this sort.

Agencies' Differing Interpretations of the Components of Costs That Must
Be Covered by Savings under the ESPC Legislation Add to Uncertainty That
Savings Cover Costs

The statute governing ESPCs provides that "aggregate annual payments"
under an ESPC may not exceed the amount the agency would have paid for
energy without such a contract.17 However, agencies differ in their
interpretation of this statute. In practice, it remains uncertain whether
contract payments may be made only from utility savings resulting from the
ESPC or from funds already earmarked for equipment replacement and other
sources to reduce the length of the contract and finance charges. Within
DOE, for example, disagreement about the interpretation of the statute is
shown by a FEMP guide on the one hand and an opinion provided by DOE's
Office of General Counsel on the other.

According to a DOE departmental official, the main source of guidance for
agencies regarding lump-sum payments is FEMP's "Practical Guide to Savings
and Payments in Super ESPC Delivery Orders," issued in January 2003.
Section 3.6 explains that agencies may use existing funds that would
otherwise be used for operation and maintenance and repair and replacement
projects (1) to increase ESPC project investment and include a more
comprehensive set of energy-efficiency improvements than would be possible
otherwise, or (2) to lower the financed amount and shorten the term,
thereby reducing interest costs over the term. The section adds that
one-time energy-related cost savings are often applied as a
preperformance-period payment to the energy services company. However,
such payments may also be scheduled as payments during the contract
performance period.

Similarly, section 4.4.1 of the FEMP guide states that if appropriated
funds are available for general maintenance, operation, repair, and
replacement of energy-consuming systems (as opposed to being earmarked for
a specific project via a capital line item), they may be used for payments
to the energy services company. Adding that one-time savings and payments
from general operation and maintenance and repair and replacement accounts
merit further clarification, the discussion notes that the intent of the
ESPC statute is to permit the use of funds available in general operation
and maintenance and repair and replacement accounts that could be used for
energy-related purposes for preperformance-period ESPC payments. It also
notes that one-time payments scheduled during the performance period may
not exceed the amount planned and budgeted in the general

1742 U.S.C. S: 8287 (a)(2)(B).

operation and maintenance and repair and replacement accounts for the
avoided project.

Despite the FEMP guide's attempt to clarify allowable sources of funding
for ESPC projects, some uncertainties remain. Even within DOE, for
example, the General Counsel's office expressed an opinion at variance
with the FEMP guide. A memo from the General Counsel's office to the
assistant secretary for Energy Efficiency and Renewable Energy in August
2000 stated that, in the case of buyouts and buydowns in super ESPC
projects, energy cost savings must exceed payments in each of the contract
years. The memo added that, because ESPCs are performance contracts,
payment is conditional upon the realization of energy cost savings. The
memo stated that buy downs are in effect prepayments which, in any
contract year, may not exceed guaranteed and verified energy costs savings
for that year. The memo concluded that prepayments have the effect of
paying a contractor before the savings have occurred and under this
analysis such prepayments are prohibited.18

GSA's policy regarding buydowns is drawn primarily from the FEMP guide. In
GSA, the motivation for using the funds allowed by this guide is the low
utility rates in some of its regions. These low utility rates reduce the
savings accrued by a proposed project, necessitating a longer contract
term so that sufficient savings can be generated for covering costs.
According to GSA officials, the agency has used upfront buydowns
frequently, which has enabled GSA to reduce the cost and length of its
contracts. They noted that even a small buy down has a large impact over
the typical length of such contracts.

GSA officials told us that the lack of clarity regarding financial terms
in earlier FEMP guidance led to GSA being unable to buy down ESPCs in some
cases. One of GSA's main complaints in this regard stemmed from
inconsistencies across its regions about what funding sources could be
applied to buy downs. Following comments to FEMP and FEMP's revision of
the guidance, GSA officials noted that there have been no complaints since
October 2002. Asked if there are any remaining improvements needed for the
sake of clarity, GSA officials told us that there is still some

18In commenting on a draft of this report, DOE disagreed with our
assessment that the opinion of the General Counsel's office and FEMP
guidance are at variance. Nevertheless, DOE plans to address this point
through guidance and in an upcoming report on ESPCs to the Congress.
Consequently, we did not change the report text.

uncertainty about how much can be financed and how much can be bought down
on any given ESPC project.

The Navy has no written policy on the use of buydowns and defers to
contracting officers to determine when additional payments can be made.
Because of the lack of clarity in this area, the Energy Programs division
director at the Naval Facilities Engineering Service Center has asked for
written guidance from the Navy but has not received it. The director told
us that contracting officers evaluate the legislation and the terms of the
contract and apply them to individual contracts and situations. He said
that there have been three different situations in which the Navy has used
buydowns. First, before or during construction, the Navy has identified
avoided costs for equipment whose purchase is already included in the
budget but that will not be needed as a result of an ESPC. Funds
associated with these avoided costs can be used to reduce the amount of
money owed in the contract because the Navy views these avoided costs as
resulting directly from the ESPC. Second, during the actual performance
period of the contract, the Navy has used other utilities budget monies
from its working capital fund and mission funding to reduce the amount of
money owed. However, it has stopped this practice because GAO raised
concerns about the money not being linked with savings from the ESPC.
Third, in cases of terminating specific energy efficiency improvements or
terminating a number of years from a contract, the Navy has used funds
from its utilities budget. The division director stated that greater
clarity regarding the use of funds to make additional performance period
payments from the utilities budget, but not directly associated with the
ESPC, would be helpful because these payments can reduce long-term
financing costs and save money for the government.

  Agencies Are Concerned About Officials' Lack of Necessary Expertise and
  Information and About Competitiveness of the Super ESPCs

Agencies expressed concerns about the expertise and information needs of
the agencies and insufficient competition among financiers and energy
services companies, all of which can affect agencies' ability to protect
the government's financial interests in using ESPCs. Regarding expertise
and information, agency officials many times lacked technical and
contracting expertise and information on past contracts needed to
effectively evaluate the ESPC proposals and monitor the contracts for
savings. As a result, they often relied on the energy services companies,
calling into question the quality of the deals the officials struck and
their certainty that guaranteed savings were realized. Expertise was
lacking mainly because of inexperience with ESPCs, and information was
lacking mainly because agencies are not required to collect and
disseminate it. Regarding

insufficient competition, agencies believe there may not be enough
competition among finance companies and energy services companies. As a
result, agencies may be paying too much for financing and other terms of
the contracts and may be getting poor services after the contracts have
been signed. In recognition of these shortcomings, the agencies are taking
a number of corrective steps on an ad hoc basis and have developed an
interagency steering committee to address some of them collectively. We
did not assess the effectiveness of the agencies' efforts.

Agencies Often Lacked the Expertise and Information Needed to Effectively
Develop and Monitor ESPCs

Those project officials we interviewed who were able to marshal the
expertise and information they needed believe that having adequate
expertise and information are critical to the success of the ESPC. For
example, officials at the Portsmouth Naval Shipyard, which undertook a $43
million ESPC in 1999 to upgrade its power plant system, relied on the U.S.
Army Corps of Engineers' Huntsville Center and the Navy contracting
centers for technical and contracting support and on a consultant for
engineering support and analysis of utilities forecasting. According to
the Navy official who developed and oversees the Portsmouth project, the
expertise provided by the three sources was essential to the success of
the project. In particular, the consultant's analysis of electricity rate
projections, made possible because of the consultant's knowledge of
utility markets in New England, allowed the Portsmouth officials to
question the energy services company's rate projections and negotiate more
favorable rates for the ESPC.

As previously discussed, developing and monitoring an ESPC are difficult,
requiring both technical and contracting expertise. In particular, for the
development phase of ESPCs, we learned that agencies frequently had
difficulty with technical responsibilities such as accurately calculating
energy-use baselines and forecasting utility rates. For example, the Air
Force and Army audits of ESPCs noted a number of instances in which
baselines were incorrectly established, and numerous officials told us how
difficult it is to accurately establish these baselines. Along those
lines, the manager of DOE's departmental energy management program told us
that officials at the project level do not always have the necessary
expertise to forecast utility rates and, given the complexity of
forecasting these rates, particularly over the long terms typical of
ESPCs, it is easy for the officials to agree to incorrect estimates. ESPC
experts at DOE's Oak Ridge National Laboratory agreed, saying it may be
unrealistic to expect a government contracting officer to be able to
effectively negotiate some contract terms

such as utility rates because they are technically difficult to understand
and forecast.

Regarding monitoring ESPCs once the energy-efficiency improvements are in
place and operating, the measurement and verification reports the energy
services companies submit to substantiate savings pose a challenge for
agencies because of their technical nature. A number of the officials we
interviewed told us that the level of expertise at the project level is
often inadequate to perform a thorough evaluation of the measurement and
verification reports. The manager of DOE's departmental energy management
program noted that, in the past, DOE has not reviewed measurement and
verification reports. The challenge of effectively reviewing these
reports, however, has led DOE to consider requiring that DOE headquarters
become involved in measurement and verification evaluations. In addition,
according to an expert in measurement and verification for the Air Force,
lack of technical knowledge is the primary cause for agencies' failure to
conduct appropriate measurement and verification oversight. In this
regard, a lack of basic adherence to measurement and verification plans
has also been observed. The project manager of the Air Force audit noted
that, among the eight bases included in his review, only one had properly
followed its plan.

Another area requiring technical expertise involves a careful balance
between stipulation and measurement and striking this balance has been
difficult for agencies. According to DOE officials, key guidelines for
measurement and verification do not define the best method for each
energy-efficiency improvement that balances the trade-offs between cost
and accuracy. Consequently, the "right" amount of measurement and
verification for many improvements remains uncertain and requires
expertise to determine in each case. Agency officials have generally
agreed that measurement and verification, at least in the first years of
using the super-ESPC contracts, tended to rely more heavily on stipulation
than on actual measurements for determining long-term savings. An Air
Force official told us that, in his view, the heavy reliance on
stipulation during the earlier years of the program worked to his agency's
disadvantage with regard to savings. In more recent contracts, however, he
believes that a better balance between stipulation and measurement has
been reached because there has been a greater reliance on expertise in
this area.

In some cases, we were told, the officials may have the technical, but not
contracting, expertise they need. Managers of the VA's ESPC program are
confident that the agency's project-level officials have enough
engineering know-how to understand the technology and construction process
involved with ESPCs; however, the Managers are concerned that project
level officials do not understand the financing, markups, or other aspects
of the business end of ESPCs well enough, giving the energy services
companies an advantage over the agency officials, who, in turn, may not be
able to make the best business decisions for government. For example,
according to the Manager of DOE's departmental energy management program,
in his experience, markups and financing rates often go unchallenged by
projectlevel staff, even though they are negotiable, because the project
officials do not have the expertise to challenge them. Furthermore,
officials who oversee GSA's energy program told us that GSA energy
managers have had to negotiate with energy services companies on markups
and financing terms, even though they were not adequately trained in that
contracting technique.

Related to expertise, ESPC project-level officials also may not have the
information at their disposal that would help them develop the best
possible contracts and effectively oversee contract implementation. A
number of officials we interviewed said they had neither benchmarking data
on prices and other contract terms agreed to for other ESPCs, nor
knowledge of "lessons learned" on other contracts, making it difficult for
the officials to evaluate project proposals and to negotiate effectively.
Of the seven agencies included in our review, DOE, GSA, and the Navy
compile and maintain some data on their ESPCs in one location. Although
individual project files contain some data that could be used for
benchmarking prices and terms, agencies are not required to compile and
disseminate such information across their ESPCs, and the other four
agencies told us they do not. Similarly, as discussed previously, although
agencies are required to monitor the performance of energy services
companies on individual projects to determine whether expected savings are
being realized, they are not required to keep track of that information at
the agency level. As a result, the agencies may not have historical
information on contract performance to use in choosing energy services
companies and developing terms of the contracts, such as the measurement
and verification plans.

Officials responsible for ESPCs do not always have the expertise and
related information they need for a number of reasons. Many of the
projectlevel officials are inexperienced with ESPCs. In that regard,
several of the

military project officials we interviewed said that their current
experience is their first encounter with an ESPC, and the limited training
they received did not adequately prepare them. Furthermore, DOD officials
told us that because military staff are frequently reassigned after a few
years, it is not likely that one person will be on site throughout the
entire ESPC contract, and the officials expect their replacements to be
similarly inexperienced with the contracts. Further exacerbating the
problem, we were told that many of the military and civilian officials
charged with developing and overseeing ESPCs only work on the contracts
part-time so the efforts they can devote to the process are limited.

Most agencies do not require their officials to use the contracting
centers in DOD and FEMP when developing ESPC projects; nonetheless, most
of them do. In the case of interviews with officials from 27 projects in
which the officials discussed their use of contracting centers or other
sources of expertise in detail, officials from 26 of these projects said
they found the expertise helpful. However, for 13 of the 26 projects that
got assistance, officials cited areas of the ESPC development process in
particular for which they could have used more expertise. For example, GSA
and VA officials told us that their FEMP project facilitators, which cost
the agency $30,000 for each project, did not perform some functions that
the agencies thought would have been beneficial, such as preparing
estimates of project costs or advocating for the agencies during contract
negotiations. Some project officials that used the contracting centers
found the centers to be inadequately funded. For example, one Air Force
project official told us that the Air Force's center provided the project
with excellent support, but could not visit the project site due to
resource constraints. Similarly, another official told us he did not
consider using the U.S. Army Corps of Engineers' Huntsville Center because
he thought, based on his previous experience with the center, that it was
understaffed and would not be able to devote enough effort to the project.

As a result, we were told that agencies often rely on the energy services
companies to provide much of the needed expertise to develop and monitor
the ESPC projects, potentially raising a conflict of interest.19 One
company representative told us that agency officials are typically not
familiar with the energy savings potential of the new equipment being

19As we pointed out in our earlier report, once agencies decide to use an
ESPC and select an energy services company to work with, they must ensure
that the government's interests are protected from the potential conflicts
that may arise.

proposed for installation, for example, and another representative said
that agencies need more ESPC expertise. A number of agency officials
agreed that they rely on the energy services companies because they lack
certain expertise themselves. For example, an Air Force official told us
that project officials on remote air bases tend to have less-experienced
staff and rely on the energy services companies for essential ESPC
activities such as performing life-cycle cost analyses.

Agencies Expressed Concerns About Competition Among Finance Companies and
Energy Services Companies

Some agency officials we spoke with expressed concerns that there may not
be enough competition among finance companies and that this could lead to
higher than necessary financing costs for ESPCs. Some agency officials
told us they believe the financing rates for ESPCs are high compared with
rates to finance energy-efficiency improvements by other means. For
example, according to VA ESPC program managers, the rates VA has
negotiated to purchase energy-related equipment via another financing
mechanism-enhanced-use leases-are generally lower than its ESPC rates.20
For the 241 ESPC delivery orders for which we received financing data,
financing rates ranged from 5 to 13 percent, with an average across all
projects of almost 8 percent. According to an ESPC expert at DOE's Oak
Ridge National Laboratory, improving the financing of ESPC projects is one
of the most important ways to achieve a better deal for the government.

Agency officials stated that there may be too few companies available to
finance ESPCs. For example, the head of the Navy's ESPC program told us
that there have been difficulties in finding investors for its ESPCs and
needs more investors in the program. VA officials responsible for
overseeing the agency's ESPCs echoed this concern. They believe there are
only three or four "boutique" companies that specialize in financing
ESPCs, and that the absence of more financing companies drives up the
financing rates. They cited findings by a consultant the agency hired to
review ESPCs that reported that the lack of competition among energy
financiers creates higher rates, and the officials believe that injecting
more competition into the process may result in better rates. The head of
FEMP's Super ESPC Program estimates that there are eight financiers that
have provided bids for financing ESPCs.

20Enhanced-use leases allow authorized agencies to enter into long-term
real property leases. Like ESPCs, they allow agencies to use private funds
to finance improvements to real property, including energy-efficiency
improvements.

Agency officials also said they have seen little evidence that the energy
services companies are seeking out the most favorable financing rates.
Historically, energy services companies were not required to provide
documentation of having sought favorable rates. According to a contracting
officer who reviews the Army ESPCs, the agency has sometimes obtained
better rates when it required at least three quotes from financiers.
According to the Air Force and Navy officials responsible for reviewing
ESPC proposals, some proposals did not contain sufficient information to
adequately determine if the financing costs were reasonable. The Deputy
Manager of DOE's departmental energy management program told us that
including documentation of competition among financiers in the ESPC
proposals is needed to provide better assurances that the government is
getting the best financing rates. In his experience, an energy services
company often wants to use a single financier for all of its ESPCs, so he
believes little or no competition for financing exists for those
contracts. The energy services company representatives rebut this
contention, saying they consistently seek the most favorable financing for
ESPCs. They told us that lower financing costs allow more of the project's
savings to be spent on energy-efficiency improvements, from which the
companies profit, rather than on finance costs.

Other agency officials and representatives of finance companies and an
energy services company have offered other explanations for why finance
rates for ESPCs are as high as they are. For example, according to FEMP
and GSA ESPC program managers, as well as representatives of the three
financing companies in our review, agency officials generally do not
understand that certain characteristics of ESPCs increase the risk of
financing those contracts and may drive the rates up. Chronic late
payments by agencies are one such characteristic. Others include the
possibility that the agency will withhold its payments if the savings
guaranteed in the contract are not realized, the additional uncertainty
about contract performance due to the long contract terms typical of an
ESPC, and the possibility that the agency will make unscheduled payments
that will reduce the financier's return on the contract. According to
GSA's ESPC Program Manager, these risk factors limit the number of
companies willing to finance ESPCs, and the complexity of the contracts
drives financing rates higher.

We were unable in the scope of this work to determine the extent, if any,
to which a lack of competition, rather than other factors, has caused
finance rates for ESPCs to be higher than for other methods of financing
energy-

efficiency improvements. However, due to the large number of questions
raised by agencies, we believe this question should be explored in more
depth.

Some agency officials also expressed concern that there may not be enough
competition among energy services companies. In general, they told us
there may be too few companies on the lists and those companies may be
charging prices that are too high and providing inadequate services.
Regarding the number of companies available, some officials told us that
often only the large companies on the lists are willing to undertake
ESPCs, effectively limiting agencies to three or four companies to choose
from. FEMP ESPC program managers affirmed that it may be only the largest
companies that can afford the extended negotiation and contract
implementation periods of ESPCs before getting paid for their services.
Further, GSA ESPC managers told us they have received complaints from
energy services companies that would like to take on smaller ESPCs, but
believe they are disadvantaged in obtaining that business because they are
not on the lists and have not been given a sufficient chance to compete
for that status. In that regard, officials from some agencies told us that
the companies approved for the lists often will not bid on projects unless
they are worth at least $1 to $2 million. As a result, the agencies must
forego undertaking the smaller projects or combine multiple locations into
a project to meet the threshold. According to DOE and GSA officials, it is
more difficult to manage projects with multiple locations. In addition,
according to the head of FEMP's Super ESPC program, multiple energy
services companies that did not compete in the original super ESPC
competitions have communicated their desire to participate in a
recompetition and to be added to lists of prequalified energy services
companies.

Some agency officials linked a perceived lack of competition among energy
services companies with high markups and prices for other components of
ESPCs and poor services-especially after the contract is signed. Regarding
markups, energy services companies charge a percentage of the cost of each
energy-efficiency improvement to cover company costs for, among other
things, overhead, sales, markup on subcontractor-supplied materials and
labor, and profit. Both the Army and FEMP super ESPCs contain
pre-negotiated markup maximums that are intended to cap the amount of
markup that the energy services company can add to the basic price of each
energy-efficiency improvement covered by the contracts. FEMP's markup
maximums typically range from 26 to 31 percent-but may be as low as 5
percent and as high as 100 percent-depending on the

energy-efficiency improvement on which they are based and the region of
the country the improvement is implemented. The markup maximums the U.S.
Army Corps of Engineers' Huntsville Center provided to us range from 15 to
30 percent. A number of agency officials told us that, as a practical
matter, the energy services companies resist agencies' efforts to
negotiate markups that are lower than the caps. According to an Air Force
contracting center official, the Air Force super ESPCs do not contain
prenegotiated markup maximums for energy-efficiency improvements and the
negotiators that use the Air Force super ESPCs typically obtain more
favorable markups than those who use the Army Corps of Engineers'
Huntsville Center's or FEMP's super ESPCs. To test this assertion, we
examined data on markups in ongoing ESPCs that agencies reported to us.
The reported markups ranged from 10 to 40 percent for projects under
FEMP's super ESPC, from 13 to 32 percent for the U.S. Army Corps of
Engineers' Huntsville Center's, and from 9 to 29 percent for the Air
Force's. However, because the agencies did not report markups for all of
the projects in our review and because data did not tie markups to
individual energy-efficiency improvements, we could not determine whether
the projects using the Air Force super ESPC actually resulted in more
favorable markups.

With regard to prices of some components of ESPCs, a number of agency
officials we interviewed expressed concern about their ability to
negotiate reasonable prices in their ESPCs. DOD agencies are required to
give all energy services companies prequalified for a super ESPC an
opportunity to participate in a limited competition at the initial
proposal stage of a project. The competition is limited because,
ostensibly, the companies have already passed government scrutiny in order
to be included on the super contract. Although civilian agencies do not
have the same requirement, they may choose to conduct a limited
competition, and most did for the projects in our review. For the limited
competition, the agencies provide the companies with such information as
current utility rates and the types of improvements the agency is
considering that the companies can use to develop their initial proposals.
The initial proposals contain preliminary cost estimates and other
information the agencies use to narrow the field to the single company it
will do business with on the project. Prices are not discussed with any
specificity until after the selected company has prepared its formal
project proposal, even though the formal proposal can take more than 6
months to complete and review. By that time, we were told, the agency may
feel pressure to continue with the company, possibly accepting prices that
are too high because it is too costly to start over with another company.
Lacking the ability to force the energy companies to

compete more rigorously on prices, ESPCs may cost more than they should.

Finally, some officials complained about the unsatisfactory services
provided by the energy services companies. For example, one Air Force
energy manager told us that the quality of work by the energy services
company declined substantially after the delivery order was awarded.
According to this official, the energy services company lacked the
internal capability to properly do the work yet resisted hiring additional
staff. In addition, the company did not use the subcontractor identified
in the project proposal; as a result, the agency could not determine if
the costs claimed by the company were valid. Some other problems cited by
officials included inflated costs and over-billing for equipment and
labor, insufficient and/or redundant design work, substitution of cheaper
materials, untimely responses, and disruptive staffing changes.

None of the companies on the super ESPC lists have had to re-compete for
their positions on the lists since they won them 6 to 9 years ago, and the
recompetitions planned for them will not occur for another 1 to 2 years.
The companies on those lists have not changed unless they merged with
others, went out of business, or chose to be taken off the list. While
there are no requirements for how frequently the super ESPCs must be put
out for competition, GSA's practice regarding its contracts for the
Federal Supply Schedule, which are multiyear contracts similar to the
super ESPCs, is to renegotiate the contracts every 5 years to help ensure
the contracts remain competitive. According to the head of FEMP's Super
ESPC Program, DOE policy calls for re-competing contracts such as the
super ESPCs every 5 years.

Our own analysis of agency data on ESPC use indicates that ESPC contracts
appear to be highly concentrated among a relative few companies in some
regions. We calculated the Herfindahl-Hirschman Index (HHI)-an index used
by the Federal Trade Commission and the Department of Justice to evaluate
mergers-for each of the six regions defined by the FEMP super ESPCs. In
four of the six FEMP regions, the HHI was above the level at which
industries are typically considered to be moderately to severely
concentrated. While such measures do not by themselves indicate a lack of
competition, they do suggest that a more complete evaluation of the
competitiveness of the ESPC contracts is warranted.

Individual Agencies and the Contracting Centers Are Taking Steps to
Address Concerns about Expertise, Information, and Competition

Individual agencies have taken steps to address concerns about expertise
and related information and competition. Among other steps, to bring
expertise and information from previous ESPCs to bear on new ones being
undertaken by their agencies, DOE and the Navy each require ESPC proposals
be reviewed by experts either in-house or at FEMP and do not allow the
projects to proceed into implementation without approval of these experts.
DOE and GSA compiled lists of lessons learned and have shared them among
project officials within their agencies. In addition, the Air Force, the
Army, the Navy, DOE, GSA, and VA each have begun requesting evidence of
competition for financing rates before they will agree to an ESPC for
their agencies. Furthermore, rather than relying exclusively on the super
ESPC contracts, officials from VA are pursuing alternatives to introduce
price competition into the process.

The contracting centers have also taken steps to bolster the expertise and
information available to their officials and to address the
competitiveness problems with the super ESPCs. Most notably, all the
centers have, among other things, issued guidance to help agencies with
developing and monitoring their ESPCs and begun requiring that project
proposals contain documentation of multiple financing bids. Furthermore,
the centers are working to have newly competed super ESPCs available to
agencies between fiscal years 2007 and 2008. The U.S. Army Corps of
Engineers' Huntsville Center plans to have its new super ESPCs in place by
the beginning of fiscal year 2008 and has begun that process. According to
a U.S. Army Corps of Engineers' Huntsville Center contracting office
official, the center has not recompeted its super ESPCs because the
current contracts do not expire until the end of 2007 and developing the
revisions to the contracts has proven to be a slow process that requires
coordinated input from multiple ESPC experts and contracting centers. FEMP
also plans to recompete its super ESPCs. It plans to begin the process in
2006 and have the new contracts in place sometime in 2007. According to
the FEMP Super ESPC Program manager, FEMP has not recompeted its super
ESPCs to date primarily because FEMP has focused its efforts on helping
agencies undertake successful ESPCs and developing guidance for the
agencies to use. The Air Force does not presently plan to recompete its
contracts but will reconsider that decision over the next 2 years.
According to managers of Air Force's contracting center, Air Force ESPC
projects are increasingly using FEMP's super ESPCs because doing so
provides the project-level officials with more contract and energy
services company options. Consequently, rather than re-competing the Air
Force super ESPCs, the managers may begin to phase out their agency's use
of the Air

Force super ESPCs as they increasingly use FEMP's with Air Force-specific
clauses added.

Although most of the steps agencies and contracting centers have taken to
address expertise, information, and competition needs have been ad hoc,
they have recently begun to address them more collectively via an
interagency steering committee and its working groups. The purposes of the
steering committee include sharing experiences and lessons learned among
the federal agencies that use ESPCs the most, identifying process and
procedural improvements, and developing best practices. The steering
committee plans to develop performance metrics by which its efforts can be
evaluated by June 2005. In addition, the steering committee and its
working groups have accomplished some of their objectives to date. For
example, the working group on measurement and verification issued a
template standardizing the measurement and verification process. Each of
the contracting centers used some of the group's recommendations when it
developed new measurement and verification guidance. See table 3 for a
more complete list of steps the contracting centers have taken.

We believe that many of the steps the individual agencies and contracting
centers have taken to address expertise, information, and competition
issues promise to help improve those areas. However, because of their
adhoc nature and because many are relatively new and untested, we did not
attempt to assess their effectiveness.

    Table 3: Steps Contracting Centers Are Taking to Address Concerns About
                  Expertise, Information, and Competitiveness

Contracting
centers Expertise and data Competition

FEMP	Expertise: (1) Developed ESPC delivery order guidelines in October
1999 and updated in February 2004, (2) Developed guidance on measurement
and verification in September 2000, (3) Developed a guide regarding ESPC
savings and payments in January 2003, (4) Developed ESPC-specific training
on process, tools, and best practices in 2004, (5) Facilitates ongoing
working groups involved in ESPC issues including measurement and
verification, performance period administration, and energy savings
discrepancy resolution, (6) Offers Project Facilitators Financing: (1)
Facilitates the financing working group of the ESPC Steering Committee,
(2) Implementing financing reforms in super ESPC modification as of late
2004, (3) Educating agencies and fiance companies more on ESPC financing

Contract terms: (1) Issued modifications to super ESPC language and terms
in late-2004 with changes effective in 2005; (2) Plans to update markup
maximums during re-competition of super ESPCs by 2007; (3) Will begin the
super ESPC re-competition in late 2006 to be completed by 2007; (4) Plans
to recompete every 5 yrs

(Continued From Previous Page)

Contracting
centers Expertise and data Competition

Data: Collects data on basic contract terms at time of signing Pricing
competition: Facilitates the Price but does not collect data on contract
modifications past the first Reasonableness working group of the ESPC year
or monitoring data showing energy savings Steering Committee - tasked to
provide tools and

guidance in negotiating competitive pricing

Navy	Expertise: (1) Provides centralized engineering and contracting
expertise, (2) Developed process and contract guidance in October 2003,
(3) Developed a price reasonableness guide, (4) Shares lessons learned
throughout agency, (5) Requires inhouse or FEMP approval of new ESPCs
Contract Terms: (1) Helping sites implement additional contract language
to better protect the agency's interests

Data: Maintains a central ESPC database	Financing: Requesting
documentation of multiple financier bids

Air Force	Expertise: (1) Developed M&V templates for each technology and
other guidance beginning in 2001, (2) Provides expertise and support to
project officials (3) Created a Resource Efficiency Manager position to
address expertise needs.

Data: Centrally collects basic contract data and modification information,
but does not collect monitoring or savings data Financing: Requesting
documentation of multiple financier bids

Contract terms: (1) Modifying M&V language in the super ESPCs, (2)
Contract agency helping implement additional contract language to other
super ESPC templates; (3) 2005-2006 evaluating either re-competing Air
Force's super ESPCs or using FEMP's with Air Force-specific clauses added
Army Corps of Expertise: (1) Developed ESPC training document, (2)
Contract terms: (1) Currently revising proposal Engineers' Participates in
FEMP working group, (3) Uses a template to requirements and other language
via bi-lateral Huntsville assist agency officials negotiating with
contractors negotiations with energy services companies; (2) Center
revising additional language and terms by 3rd quarter

Data: Collects data on some contract terms, and may not have FY2005, to be
included in planned recompetition for

all modifications on record FY2007

            Source: GAO analysis of contracting center information.

Conclusions	ESPCs provide a valuable and practical tool that federal
agencies use to meet energy reduction, environmental, infrastructure, and
other goals. Clearly, agencies that have used ESPCs to install more
efficient, energysaving equipment have reduced their energy consumption
and associated environmental impacts. Further, by using private financing,
agencies have also been able to more quickly and consistently replace an
aging and energy-wasting infrastructure-an infrastructure that the
agencies have identified in their capital management plans as being in
need of billons of dollars of repair and restoration.

While using ESPC-financed projects has permitted agencies to reduce energy
consumption and achieve other goals, the extent to which savings cover
costs as required by legislation remains uncertain. The complexity of
ESPCs accounts for much of this uncertainty. ESPCs are complicated

because of the wide array of technical, financial, legal, and
energy-related issues that must be resolved both in the short and
long-term. Because of this complexity and the cost of more extensive
reliance on actual measurements, agencies have tended in the past to defer
to the expertise of energy services companies and the use of stipulation
in lieu of measurements. In doing so, they may have paid contractors for
energy savings that did not occur or may have negotiated contracts that
are more expensive than necessary. Limited agency audits and our
interviews have disclosed indications of these problems in dozens of
projects. Since most agencies have not audited their use of ESPCs and
broad performance information and documentation are unavailable, we could
not determine how widespread these problems are. Without comprehensive
information on actual performance of the contracts once they have begun to
unfold, however, the agencies' task of overseeing the contracts becomes
difficult. In turn, the lack of comprehensive information on ESPC
performance makes it more difficult for the Congress to determine the
level of support it should lend to agency use of the financing mechanism.
Finally, because DOE reports to the Congress about agencies' progress
toward achieving energy goals, the lack of comprehensive data on the
results of ESPCs also reduces congressional awareness in this area. In a
more general context, additional information would be useful in comparing
the costs and benefits of ESPCs relative to alternative financing
mechanisms. This information could include, among other things, the
effects of deterioration of energy efficiency savings in the absence of
measurement and verification and delays in obtaining up-front
appropriations relative to obtaining funds through ESPCs.

In response to these problems, agencies have begun to recognize the
importance of developing and using their own expertise more effectively,
but this has occurred only recently and, at this point, they have not
ensured that it is brought to bear during negotiations and in the longer
term. The ability to correct these problems requires the availability of
high-quality information and the expertise to use it effectively during
negotiations and throughout the life of these long-term contracts. In
developing and using appropriate expertise and information, agencies can
also begin to assemble better information about governmentwide experiences
with ESPCs, including ways of improving such areas as measurement and
verification. They can also draw conclusions regarding the effectiveness
of agencies' working relationships with individual energy services
companies, which could provide another valuable tool for agencies to
consider. Finally, as ESPC use continues, sharing best practices or
lessons learned in all of these areas would go a long way toward making
ESPCs as cost effective as

possible while also helping to ensure that the federal government's
financial interests are protected. Absent further efforts to rely on
appropriate expertise and improve the quality of information, agencies
will continue to be at a disadvantage in negotiating effective ESPCs and
less likely to achieve long-term energy and financial savings.

Agencies have expressed concerns about the adequacy of competition among
financiers and energy services companies in developing ESPCs and
consequently their ability protect their interests. Agency officials and
others expressed concerns that financing costs may be too high because
there may be too few companies that finance ESPCs and energy services
companies may not seek the most favorable financing. Other problems such
as the length of time between competitions for the approved list of energy
services companies and the lack of price competition inherent in using the
super ESPCs also reinforce these concerns. Agency officials have taken
some steps to address these concerns, but the question of sufficient
competition points toward the need for further measures such as requiring
greater competition among financial service companies to potentially
reduce interest rates and putting the super ESPCs out for competition more
frequently.

Differing agency interpretations of the law establishing ESPCs have
contributed to agency uncertainties about the use of funding sources other
than savings for reducing investments in ESPCs through upfront payments.
Within DOE, inconsistencies and uncertainties about interpretation of the
statute are apparent. In practice, some agencies believe that contract
payments may be made only from utility savings resulting from the ESPC
while other agencies make a lump-sum payment on the contract-from funds
already earmarked for equipment replacement or from other sources-to
reduce the length of the contract and finance charges. In our view, these
inconsistencies reflect a lack of clarity about the use of down payments
in general and what does-or does not-constitute a legitimate source of
funds for such down payments if they are allowed.

Matter for To ensure that agencies use ESPCs as the Congress intends, we

recommend that the Congress consider revising the relevant statute
toCongressional more clearly define the components of costs that must be
covered byConsideration savings. In particular, the Congress could clarify
whether agencies may

make lump sum payments using funds other than their current year utility

savings.

  Recommendations for Executive Action

To better ensure that federal agencies undertake only those ESPCs having
the greatest likelihood that savings will cover costs and that the
agencies negotiate the best possible contract terms and monitor the
contracts properly, we are making recommendations to the heads of those
agencies included in our review, namely the Secretaries of Defense,
Energy, and Veterans Affairs, the Attorney General of the Department of
Justice, and the Administrator of the General Services Administration. Our
recommendations focus on the areas of information, expertise, and audits:

o 	Collect and use ESPC-related data more effectively by (1) compiling
information on key contract terms-such as interest rates and markups for
energy-efficiency equipment-for each ESPC and as a key part of best
practices make information accessible to agency officials in negotiating
subsequent ESPCs and (2) tracking actual costs, verified savings, and any
changes to ESPC projects that may affect these costs and savings.

o 	Ensure that the agency officials responsible for ESPC decision-making
use appropriate expertise when they undertake an ESPC. If the officials do
not have sufficient expertise themselves, they should be required to
obtain it from such independent sources as a centralized pool within the
agency; the contracting centers of the Air Force, the U.S. Army Corps of
Engineers, the Navy, and FEMP; or from private parties. The costs of
acquiring this expertise should be considered in deciding whether to use
an ESPC.

o 	Require, as appropriate and in line with available resources, that
inspectors general or other audit offices conduct audits of ESPC projects
to ensure the projects are achieving their expected results.

Because the contracting centers can play an important role in helping the
agencies develop and monitor their ESPCs, we recommend that the
secretaries of Defense and Energy require the contracting centers to

o 	work with the agencies that use them to ensure that the contracting
centers have the information and expertise needed to effectively develop
and monitor their ESPCs; and

o 	continue and expand their ongoing efforts regarding competition,
including taking steps such as re-competing the super ESPCs as soon as
possible and then more regularly.

Finally, to strengthen the information available to the Congress for
assessing the progress and effectiveness of ESPCs, we recommend that the
Secretary of Energy collect more extensive information on agencies' ESPCs,
including such critical elements as cumulative verified savings and costs,
and include that information in its annual report to the Congress. As a
part of this effort, we also recommend that the Secretary compare projects
funded by ESPCs with projects funded by upfront appropriations to
determine their relative costs and benefits. Specifically, the Secretary
should determine, among other things, the effects of deterioration of
energy efficiency savings in the absence of measurement and verification
and delays in obtaining upfront appropriations relative to obtaining funds
through ESPCs.

Agency Comments	We provided the Departments of Defense, Energy, Justice,
and Veterans Affairs, and the General Services Administration, with a
draft of this report for their review and comment. DOD, DOE, VA, and GSA
provided written comments, which are presented in appendixes II through V.
The Department of Justice responded by email on June 2, 2005. All of the
agencies generally concurred with the findings, conclusions, and
recommendations and stated their intention of implementing the
recommendations. The agencies also submitted technical and clarifying
comments, which we have incorporated as appropriate.

In addition, DOE expressed concerns in two areas. First, regarding our
discussion about confusion over the allowable sources of funding for
ESPCs, DOE expressed the view that its General Counsel's office's opinion
regarding prepayments was not at variance with FEMP guidance as we
reported. Nevertheless, the agency noted that it will take steps to ensure
that FEMP guidance is consistent on this point to avoid future confusion.
Furthermore, DOE supports our recommendation that the Congress more
clearly define the components of ESPC costs that must be covered by
savings and the agency stated that it will address the issue in a report
to the Congress on ESPCs that is currently in the review and approval
process within the agency. We have added language to the report noting
DOE's disagreement with our discussion of this issue. Second, DOE
expressed concern that FEMP does not have authority to do more to
facilitate oversight of ESPCs, as we recommended. While we recognize DOE's
concern with taking on additional oversight responsibilities, we note
that, in commenting on our draft report, all of the agencies stated their
intention to work cooperatively with DOE and the other agencies to
implement our recommendations. In recommending that DOE facilitate
oversight of

ESPCs, we intended that the agency take such actions as collecting data on
verified savings and costs and reporting such information to the Congress,
as well as to the agencies themselves, to aid the Congress and the
agencies in their ESPC oversight actions. We believe that it is
appropriate at this point for DOE and the other agencies to continue to
use a cooperative approach, such as through the Federal ESPC Steering
Committee, to develop and implement consistent and best practices for
ESPCs.

As agreed with your office, unless you publicly announce the contents of
this report earlier, we plan no further distribution until 30 days from
the date of this letter. At that time, we will send copies of this report
to the appropriate congressional committees; the Secretaries of Defense,
Energy, and Veterans Affairs; the Attorney General; the Administrator,
GSA; and other interested parties. We will also 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.

If you or your staff have any questions, please call 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 contributors to this report are listed in appendix VI.

Jim Wells Director, Natural Resources

and Environment

Appendix I

                       Objectives, Scope and Methodology

Objectives	We were asked to determine (1) the extent to which federal
agencies used ESPCs; (2) what energy savings, financial savings, and other
benefits agencies expect to achieve; (3) the extent to which actual
financial savings from ESPCs cover costs; and (4) what areas, if any,
require steps to protect the government's financial interests in using
ESPCs.

Scope and Methodology	To satisfy these objectives, we included in our
review any ESPCs that agencies undertook in fiscal years 1999 through
2003. We did not perform formal cost benefit analyses of individual ESPC
projects or of ESPCs as a whole because of data limitations.

To address the data analysis component of these objectives, we first
obtained basic contract data from the four federal contracting centers
that assist agencies with ESPCs-the Air Force Civil Engineer Support
Agency, the U.S. Army Corps of Engineers' Huntsville Center, the Naval
Facilities Engineering Service Center, and (FEMP), which reflect the
majority of all ESPCs undertaken during fiscal years 1999 through 2003. We
did not completely assess these data for reliability; however, we reviewed
the steps that the contracting centers take to ensure data reliability and
determined that these steps were sufficient for our reporting purposes. We
obtained more detailed contract data for the same period from the seven
federal agencies in our review having the most facility floor space and
the highest energy use and therefore the highest potential to use ESPCs.
These agencies were the Departments of the Air Force, the Army, the Navy
(including the Marine Corps), Energy, Justice, Veterans Affairs, and the
General Services Administration.

Before analyzing the contract data, we combined the data from the
contracting centers and the agencies into a single data set. Because some
agency contract data could also be included in the contracting center
data, we identified the projects that appeared to have duplicate records.
We asked each agency to confirm those records that were duplicates and,
using our best judgment, retained those records with the most complete
information.

To address the objectives overall, we interviewed and obtained
documentation from a wide range of stakeholders. From the seven agencies
and four contracting centers, we talked with officials at headquarters, in
regions, and at specific project sites. We also discussed the issues with
officials from the Congressional Research Service; Oak Ridge

Appendix I Objectives, Scope and Methodology

National Laboratory; Lawrence Berkeley National Laboratory; the Defense
Energy Support Service Center; and the states of Maryland and Louisiana,
both of which use ESPCs extensively. In addition, we talked with officials
from the energy services and financial services sectors and an academic
expert knowledgeable about ESPCs. We also reviewed relevant legislation,
regulations, policies, and agency procedures.

We also reviewed studies by the Oak Ridge National Laboratory and the
Lawrence Berkeley National Laboratory that analyzed the costs and benefits
of ESPCs and compared net benefits of using ESPCs to finance energy
savings improvements with the net benefits of using direct appropriations.
To evaluate these studies, we interviewed some of the authors and reviewed
other studies and reports that the authors had referred to and which
supported some of the assumptions they used to model the net benefits. In
particular, we asked the authors of the Lawrence Berkeley National
Laboratory study about their support for the assumption that energy
savings decay over time in the absence of monitoring and verification.
They referred us to a body of literature on energy
commissioning-essentially energy audits of buildings-in which there is
evidence of energy savings decay. We reviewed several studies from this
literature and concluded that there was sufficient evidence for savings
decay to warrant inclusion of the Lawrence Berkeley National Laboratory
study results, with the proper caveat that we could not definitively
conclude on the extent of savings decay or on the extent to which
decreased savings decay and other benefits from ESPC-funded projects may
offset the significant savings achieved from using upfront funding that we
found previously in six case studies.

To better understand specific benefits, problems and suggested
improvements for ESPCs as well as evaluate whether savings were covering
costs, we interviewed and obtained appropriate documentation from
officials who either negotiated and/or currently manage specific ESPC
projects at 15 geographically dispersed locations. We judgmentally
selected these officials from lists of knowledgeable officials provided to
us by the ESPC program managers of each agency. We also discussed these
issues in general for an additional 10 projects with the officials who
manage DOE's departmental ESPCs. Furthermore, we reviewed 13 audit reports
conducted by the Army Audit Agency or Air Force Audit Agency and discussed
the results with auditors involved in the reviews; reviewed a report by a
consultant hired by VA to assess that agency's use of ESPCs; reviewed
relevant GAO reports and consulted with subject matter experts

Appendix I Objectives, Scope and Methodology

at GAO; and reviewed other reports and information on ESPCs identified by
searching the literature.

We conducted our work from January 2004 through May 2005 in accordance
with generally accepted government auditing standards.

                                  Appendix II

                    Comments from the Department of Defense

Appendix II
Comments from the Department of Defense

Appendix II
Comments from the Department of Defense

                                  Appendix III

                     Comments from the Department of Energy

Appendix III
Comments from the Department of Energy

Appendix III
Comments from the Department of Energy

Appendix III
Comments from the Department of Energy

Appendix III
Comments from the Department of Energy

Appendix III
Comments from the Department of Energy

Appendix IV

Comments from the Department of Veterans Affairs

Appendix IV
Comments from the Department of Veterans
Affairs

Appendix IV
Comments from the Department of Veterans
Affairs

Appendix V

Comments from the General Services Administration

Appendix V
Comments from the General Services
Administration

Appendix V
Comments from the General Services
Administration

Appendix VI

                     GAO Contact and Staff Acknowledgments

                     GAO Contact Jim Wells, (202) 512-3841

Staff 	In addition to the individual named above, Dan Haas, Dennis
Carroll, Randy Jones, Hugh Paquette, Frank Rusco, Karla Springer, Barbara
Timmerman,

Acknowledgments	and Jena Whitley made key contributions to this report.
Chris Bonham, Carol Henn, Cynthia Norris, and Jena Sinkfield also
contributed.

  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. ***