This is the accessible text file for GAO report number GAO-05-55
entitled 'Capital Financing: Partnerships and Energy Savings
Performance Contracts Raise Budgeting and Monitoring Concerns' which
was released on January 18, 2005.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
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. 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.
Report to the Chairman, Committee on the Budget, U.S. Senate:
December 2004:
CAPITAL FINANCING:
Partnerships and Energy Savings Performance Contracts Raise Budgeting
and Monitoring Concerns:
GAO-05-55:
GAO Highlights:
Highlights of GAO-05-55, a report to the Chairman, Committee on the
Budget, U.S. Senate:
Why GAO Did This Study:
ESPCs finance energy-saving capital improvements, such as lighting
retrofits for federal facilities, without the government incurring the
full cost up front. Partnerships tap the capital and expertise of the
private sector to develop real property. This report describes (1) what
specific attributes of ESPCs and partnerships contributed to budget
scoring decisions, (2) the costs of financing through ESPCs compared to
the costs of financing via timely, full, and up-front appropriations,
and (3) how ESPCs and partnerships are monitored. Using case studies,
GAO reviewed GSA and Navy ESPCs and DOE and VA partnerships.
What GAO Found:
Energy savings performance contracts (ESPC) and public/private
partnership arrangements we examined were authorized by Congress and
did not require reporting of the full, long-term costs up front in the
budget. ESPCs are financed over time through annual cost savings from
energy conservation measures (ECM) and only their initial-year costs
must be recognized up front. OMB policy determined how agencies
obligated ESPCs in their budgets. With partnerships, agencies sometimes
used short-term leases to acquire assets constructed for the
government’s long-term use and benefit. As a result, budgetary
decisions may favor alternatively financed assets. However, spreading
costs over time enabled agencies to acquire capital that might not have
been obtainable if full, up-front appropriations were required.
A number of factors may cause third-party financing to be more
expensive than timely, full, and up-front appropriations. For example,
a higher rate of interest is incurred by using ESPCs and partnerships
than if the same capital is acquired through timely, full, and up-front
appropriations. For our six ESPC case studies, the government’s costs
of acquiring assets increased 8 to 56 percent by using ESPCs rather
than timely, full, and up-front appropriations. However, officials
noted that there are opportunity costs, such as foregone energy and
maintenance savings, associated with delayed appropriations, but there
are insufficient data to measure this effect. For ESPC and partnership
case studies, agency officials said they did not specifically consider
or request full up-front appropriations because they did not believe
funds would be available in a timely manner and because alternative
mechanisms were authorized. An evaluation of funding alternatives on a
present value basis could have helped agencies determine the most
appropriate way of funding capital projects.
Implementation and monitoring of ESPCs is a relatively uniform process.
Since partnerships take a variety of forms, their implementation and
monitoring is more complex. Although third-party financing can make it
easier for agencies to manage within a given amount of budget
authority, it also increases the need for effective implementation and
monitoring by agencies to ensure the government’s interests are
protected.
What GAO Recommends:
GAO recommends that OMB require and suggests Congress consider
requiring agencies that use ESPCs to present an annual analysis
comparing the total contract cycle costs of ESPCs entered into during
the fiscal year with estimated up-front funding costs for the same
ECMs. GAO also recommends (1) OMB work with scorekeepers to develop a
scorekeeping rule to ensure that the budget reflects the government’s
full commitment for partnerships and (2) agencies perform business case
analyses and ensure that the full range of funding alternatives,
including useful segments, are analyzed when making capital financing
decisions. Case study agencies had mixed comments on this report.
www.gao.gov/cgi-bin/getrpt?GAO-05-55.
To view the full product, including the scope and methodology, click on
the link above. For more information, contact Susan Irving at (202)
512-9142 or Irvings@gao.gov.
[End of section]
Contents:
Letter:
Results in Brief:
Background:
Various Features Enabled Agencies to Delay Recognition in the Budget:
Higher Interest Rates and Other Factors May Increase the Cost of Third-
Party Financing Compared to Timely, Full, and Up-Front Appropriations:
Different Financing Alternatives Present Different Implementation and
Monitoring Challenges:
Conclusions:
Matter for Congressional Consideration:
Recommendations for Executive Action:
Agency Comments and Our Response:
Appendixes:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: ESPC Case Studies:
Navy Region Southwest, California:
Patuxent River Naval Air Station, Maryland:
Naval Submarine Base, Bangor, Washington:
GSA Gulfport Federal Courthouse, Mississippi:
GSA North Carolina Bundled Sites:
GSA Atlanta Bundled Sites, Georgia:
Appendix III: Public/Private Partnership Case Studies:
DOE Oak Ridge National Laboratory, Tennessee:
VA Atlanta Regional Office Collocation, Georgia:
VA Medical Campus at Mountain Home, Tennessee:
VA Vancouver Single Room Occupancy, Washington:
VA North Chicago Energy Center, Illinois:
Appendix IV: Comments from the Department of Defense:
GAO's Comments:
Appendix V: Comments from the Department of Energy:
GAO's Comments:
Appendix VI: Comments from the General Services Administration:
GAO's Comments:
Appendix VII: Comments from Veterans Affairs:
GAO's Comments:
Appendix VIII: GAO Contact and Staff Acknowledgments:
GAO Contact:
Acknowledgments:
Tables Tables:
Table 1: Cost Analysis of Six ESPCs:
Table 2: Case Studies Included in This Review:
Table 3: Cost Analysis of Navy Region Southwest ESPC:
Table 4: Cost Analysis of Patuxent River Naval Air Station ESPC:
Table 5: Cost Analysis of Naval Submarine Base, Bangor ESPC:
Table 6: Cost Analysis of Gulfport Federal Courthouse ESPC:
Table 7: Cost Analysis of North Carolina Bundled Sites' ESPC:
Table 8: Cost Analysis of Atlanta Bundled Sites' ESPC:
Figures:
Figure 1: Definition of an Operating Lease:
Figure 2: Criteria for Assessing Private Sector Risk:
Figure 3: Elements of VA's EU Lease Authority:
Figure 4: Third-Party Financing Delays Recognition of Full Costs to
Taxpayers:
Figure 5: Buildings on the ORNL Reservation:
Figure 6: Contract Cycle Costs, Up-Front Payments, and Savings of Six
ESPCs:
Figure 7: Relationship Between ORNL Parties:
Figure 8: ESPCs Reallocate the Federal Government's Payments for Energy
and Energy-Related Operations & Maintenance Expenses (E+O&M):
Figure 9: Number of FEMP Super ESPC Awards by Agency, Fiscal Years
1998-2003:
Figure 10: Number of Navy ESPC Awards by Contract Vehicle, Fiscal Years
1998-2003:
Figure 11: Covered Parking Photovoltaic System at Navy Region
Southwest, California:
Figure 12: Partnerships and Financing of ORNL's Revitalization:
Letter December 16, 2004:
The Honorable Don Nickles:
Chairman:
Committee on the Budget:
United States Senate:
Dear Mr. Chairman:
As you know, one of the major recommendations of the 1967 Commission on
Budget Concepts dealt with coverage of the budget. According to the
Commission, the "budget should…be comprehensive of the full range of
Federal activities. Borderline agencies and transactions should be
included in the budget unless there are exceptionally persuasive
reasons for exclusion." With specific regard to capital investments,
the Commission recommended strongly against a capital budget that would
spread outlays over an asset's life, noting that it would likely
"distort decisions about the allocation of resources." We have long
supported an inclusive budget that discloses up-front the full
commitments of the government.[Footnote 1] However, in an era of
limited resources and growing mission demands, Congress has authorized
agencies to use approaches other than full, up-front funding to finance
capital acquisitions such as land improvement projects, buildings, and
equipment.[Footnote 2] Accordingly, these alternative financing
mechanisms[Footnote 3] have been used by agencies to acquire assets by
spreading the cost over a number of years in their budgets. Thus, the
full cost of an asset is not presented or recognized[Footnote 4] in the
budget at the time the decision is made to acquire it.[Footnote 5] As a
result, resource allocation decisions made through the budgeting
process may not consider the full financial commitment the U.S.
government is making and, consequently, assets financed through
alternative approaches may be preferred over other equally worthy
projects that are competing for full funding.
From an agency's perspective, the ability to record the acquisition
costs of a capital asset over the life of that asset can be very
attractive because the capital asset could be obtained without first
having to secure sufficient appropriations to cover the full cost of
that asset. From the agency's standpoint, absorbing the entire cost of
these relatively high-priced assets in a single year's budget may seem
prohibitively expensive, particularly in light of the long-term
benefits of the assets. Accordingly, alternative financing mechanisms
are frequently desirable to agencies because they make it easier for
them to quickly meet mission capital demands within a given amount of
budget authority. From a governmentwide budget perspective, however,
the situation can be different. The costs associated with these
financing approaches may be greater than would be the case with timely,
full, and up-front budget authority due, in part, to higher interest
costs.[Footnote 6] This is of particular concern at a time of rising
budget deficits and concern about underrecognition of long-term costs
and commitments. Moreover, when capital costs are not fully recognized
up-front, before funds are committed, important information about the
full budgetary effects may not be considered as trade-offs are made
among competing priorities. For the purchase of any given capital
equipment, agencies receive the same program benefits regardless of the
financing approach used.
Agencies have estimated restoration and repair needed to address the
alarming state of deterioration of many federal assets to be in the
tens of billions of dollars.[Footnote 7] Given this estimate agencies
have relied heavily on costly leasing instead of ownership to meet new
needs. Since the budget scorekeeping rules[Footnote 8] were
established, decision makers have struggled to address agencies'
tendencies to choose operating leases instead of ownership. One option
we have suggested be considered would be to recognize that many
operating leases[Footnote 9] are used for long-term needs and should be
treated on the same basis as purchases. This would entail scoring up
front the payments covering the same time period used to analyze
ownership options. We have suggested this scoring for those leases that
are perceived by all sides as long-term federal commitments so that all
options are treated equally.[Footnote 10] Although this could be
viable, there would be implementation challenges if this were pursued,
including the need to evaluate the validity of agencies' requirements
based on their long-term plans. Finding a solution for this problem has
been difficult.[Footnote 11] While leasing to meet long-term needs
almost always results in excessive long-term costs to taxpayers and
does not necessarily reflect the best approach to capital asset
management, it also provides the government opportunities to spend more
on other mission objectives. This same problem arises for any asset
that is acquired to meet long-term needs.
In August 2003, based on your request, we issued a report that
identified and briefly described 10 different capital financing
approaches used by 1 or more of 13 federal agencies.[Footnote 12]
Subsequently, as requested, we have analyzed in greater detail two of
the identified approaches: Energy Savings Performance Contracts
(ESPC)[Footnote 13] and public/private partnerships
(partnerships).[Footnote 14] In particular, we determined (1) what
specific attributes of ESPCs and partnerships contributed to budget
scoring decisions, (2) the costs of financing through ESPCs and
partnerships compared to the costs of financing via timely, full, and
up-front appropriations, and (3) how ESPCs and partnerships are
implemented and monitored.
To obtain the detail necessary to respond to this request, we used a
case study approach, which does not allow us to generalize our findings
across the government. In order to understand the budgetary treatment
and oversight of ESPCs and partnerships, we reviewed relevant
legislation, ESPC files, partnership agreements, and relevant guidance
issued by agencies and the Office of Management and Budget (OMB). We
selected case studies from agencies that had awarded a large dollar
volume of ESPCs awarded under the Department of Energy's (DOE) Federal
Energy Management Program's (FEMP) umbrella contract, had broad
partnership authority, or were discussed in our prior report.[Footnote
15] We also interviewed staff within the General Services
Administration (GSA), the Department of Defense (DOD), the Department
of Veterans Affairs (VA), the Department of Energy (DOE), the
Congressional Budget Office (CBO), and OMB to understand the features
of ESPCs and partnerships, and how the arrangements were scored. In
addition, we spoke with representatives of Energy Service Companies
(ESCO) and UT-Battelle, agency contractors involved in our case
studies. In total, we analyzed 11 case studies--6 ESPCs and 5
partnerships--across 4 agencies. Because of our focus on budget
scoring, our analysis was confined to the government's acquisition cost
and was not a cost-benefit analysis.[Footnote 16] To analyze ESPC
costs, we reviewed the delivery orders of each of our six ESPC case
studies. Although we were able to analyze ESPC costs and savings, we
were unable to perform a similar analysis of the partnerships we
reviewed because we were unable to evaluate claims that other factors,
such as lower labor costs and fewer bureaucratic requirements available
to private partners, may have reduced costs. All of the partnership
case studies we reviewed were executed before OMB's 2003 changes to its
instructions on the budgetary treatment of lease-purchases and leases
of capital assets. According to OMB staff, some of these partnerships
may have been scored differently under the revised instructions. Our
work was done in accordance with generally accepted government auditing
standards, from September 2003 through November 2004, in Washington,
D.C., Atlanta, Ga., Oak Ridge, Tenn., and Port Hueneme, Calif. A
complete description of our objectives, scope, and methodology can be
found in appendix I. Appendix II provides a summary of our ESPC case
studies and appendix III summarizes our partnership case studies.
Written comments from DOD, DOE, GSA, and VA are reproduced and
addressed in appendixes IV through VII. OMB provided oral comments. We
have incorporated these comments as appropriate throughout. Key
contributors to this report are listed in appendix VIII.
Results in Brief:
For all of the case studies we reviewed, Congress had enacted
legislation that authorized agencies to enter into ESPCs or
partnerships. Accordingly, many of the ESPC and partnership
arrangements we examined were structured to include specific attributes
that did not require agencies to reflect the full, up-front costs in
the budget even though they have features indicative of long-term
commitments. For example, agencies had statutory authority to purchase
new equipment through ESPCs over a 25-year period without an
appropriation for the full amount of the purchase price[Footnote 17]
and OMB has directed that ESPCs should be obligated on an annual basis.
With respect to several of the partnerships we examined, scoring
decisions were driven by the transfer of government land from federal
agencies to third parties. Both VA and DOE used existing authorities to
transfer land to nonfederal entities.[Footnote 18] In some cases, the
agencies then leased back, in short-term increments, assets constructed
on the land to ensure that annual lease payments rather than the full,
up-front costs of the assets were scored. Regardless of how these
transactions were structured, they had features that indicate a long-
term commitment by the government. For example, agencies will retain
control of capital assets acquired through ESPCs. Some of the
partnerships we examined were completely invisible in the budget
because they involved noncash consideration. Because the budget does
not reflect up-front the full costs of ESPCs and partnerships, decision
makers may not be able to weigh the full costs of capital acquisitions
against their potential benefits nor consider the full financial
commitment that the government is undertaking. This can make
comparisons to other proposed acquisitions difficult and can lead to a
situation in which budget decisions may favor alternatively financed
capital over programs that include their full costs up-front in the
budget.
Officials from each of our case study agencies agreed that timely,
full, and up-front appropriations were the least-cost alternative for
financing capital acquisitions. A number of factors may cause these
alternative financing approaches to be more expensive than timely,
full, and up-front appropriations. For example, case study agencies
incurred a higher rate of interest by using ESPCs and partnerships than
if they had obtained that same capital through timely, full, and up-
front appropriations because of their reliance on private financing as
opposed to Department of the Treasury financing. Also, for our ESPC
case studies, the government likely incurred additional costs for the
measurement and verification (M&V)[Footnote 19] of equipment
performance. For our six ESPC case studies, the government's costs of
acquiring energy conservation measures (ECM), such as lighting
retrofits and ventilation systems, increased by 8 to 56 percent by
using ESPCs rather than timely, full, and up-front appropriations. None
of the partnership case studies lent themselves to this type of cost
analysis for various reasons. Some of the partnerships did not involve
cash consideration. For others, while the government incurs a higher
interest rate as a result of the partnership, it is uncertain whether
the project as a whole is more or less expensive because the extent to
which other factors cited by agencies--such as lower labor costs and
fewer bureaucratic requirements--could make partnership financing less
expensive.
Additionally, some agency officials said that ESPCs and partnerships
can be cost effective because they allow agencies to acquire capital if
appropriations are not immediately available and reduce the
government's financial risk if the agency no longer needs the asset.
Although for both ESPCs and partnerships, agency officials agreed they
could acquire capital less expensively through timely, full, and up-
front appropriations, they did not specifically request full, up-front
appropriations to finance the capital projects we reviewed. Frequently
they said this was because they did not believe funds would be
available in a timely manner, that there are costs such as higher
utility bills associated with delayed appropriations, and that they had
statutory authority to use the alternative financing mechanisms.
FEMP has issued uniform guidelines for implementing and monitoring
ESPCs. Under this uniform process, agencies rely heavily on the ESCOs
to recommend potential ECMs, install the equipment, and then verify
that the improvements yield intended results. In contrast, partnerships
take a variety of forms and thus uniform implementation and monitoring
of these arrangements is difficult. In general, however, partnership
arrangements entail a government agency engaging a third party to,
among other things, renovate, construct, operate, or maintain a public
facility. Such relationships increase the need for effective oversight
to ensure the government's interests are protected. Although we did not
find any instances of fraud, waste, or abuse, the structure of ESPCs is
such that they may be compromised by potential conflicts of interest of
contractors that determine what equipment is needed and then monitor
the performance of the equipment that they recommend, install, and
guarantee. Partnerships also require monitoring because of the
complicated relationships involved. For example, at DOE's Oak Ridge
National Laboratory (ORNL), officers of ORNL's management and
operations (M&O) contractor--UT-Battelle, LLC, (1) recommended the
transfer of land free of charge to another organization--UT-Battelle
Development Corporation (UTBDC) and (2) served as officers of UTBDC,
which received the land. In addition, two of the five partnerships we
reviewed prepared no business case analysis to ensure the government's
interests were protected. Following leading capital planning
practices,[Footnote 20] including an evaluation of alternatives to
satisfy capital needs, could help agencies determine whether third-
party financing is the most appropriate way of acquiring capital. Three
of the six ESPC case studies paid a significant portion of the total
contract cycle costs[Footnote 21] in the first year of the contract.
While these large buy-downs of principal allowed agencies to lessen
their interest costs, they could also imply opportunities exist to
acquire ECMs in smaller, useful segments[Footnote 22]--when technically
feasible--with timely, full, and up-front appropriations for each of
these segments instead of through ESPCs. None of the case study
agencies considered acquiring the assets we reviewed in useful
segments.
Given the recent extension of ESPC authority until the end of fiscal
year 2006 and the competing pressures Congress faces to support energy-
saving investments while at the same time seeking to ensure budgetary
transparency of full program costs, we recommend that OMB require and
that Congress consider requiring agencies that use ESPCs to present to
Congress an annual analysis comparing the total contract cycle costs of
ESPCs entered into during the fiscal year with estimated up-front
funding costs for the same ECMs. Congress could use this information in
evaluating whether to further extend ESPC authority beyond its current
expiration date.
We also are making a recommendation to the OMB Director to develop a
scorekeeping rule to ensure the budget reflects the full commitment of
the government for partnerships, considering the substance of all
underlying agreements. Finally, we are recommending to the heads of
case study agencies--GSA, Energy, Navy, and VA--that they ensure that
business case analyses are performed and that the full range of funding
alternatives are analyzed.
In a draft of this report, we had a recommendation that the Director of
OMB work with scorekeepers to develop a rule that would ensure that the
full commitments of ESPCs are reflected in the budget. Several agencies
did not agree with this recommendation, citing concerns that such a
rule would likely discourage or prevent agencies from entering into
ESPCs. In light of Congress' recent expression of its current
priorities by extending ESPC authority through fiscal year 2006, we
dropped this recommendation with respect to ESPCs and included instead
the recommendation to OMB and the matter for congressional
consideration to require agencies to annually compare total contract
cycle costs of ESPCs, with estimated up-front costs for the same ECMs.
We obtained comments from OMB and our case study agencies--DOD, DOE,
VA, and GSA. OMB agreed in concept with our first recommendation that
OMB work with scorekeepers to develop a rule for partnerships that
would ensure the budget reflects the full commitment of the government,
considering the substance of all underlying agreements. DOE and VA
disagreed with this recommendation based on concerns that such a rule
would effectively make alternative financing unavailable to federal
agencies. While it is not our intent to discourage or eliminate
partnerships with the private sector, recognizing the full commitment
up-front in the budget enhances transparency and enables decision
makers to make appropriate resource allocation choices among competing
demands that all have their full costs recorded in the budget. GSA did
not address this recommendation in its comments. DOE, GSA, and VA
agreed at least in part with our final recommendation, that case study
agencies should perform business case analyses to ensure the full range
of funding alternatives are analyzed when making capital financing
decisions. DOD disagreed with this recommendation and OMB did not
address it in its comments. Business case analyses are well accepted as
a leading practice among public and private entities and OMB requires
all executive branch agencies to prepare such analyses for major
investments as part of their budget submissions to OMB. Therefore, we
believe our recommendation is appropriate.
Written comments from DOD, DOE, GSA, and VA are included and addressed
in appendixes IV through VII. Representatives from OMB provided oral
comments. We have incorporated changes as a result of these comments
throughout, as appropriate.
Background:
The extent to which capital costs are reflected in the budget depends
on how they are "scored." CBO and OMB separately "score" or track
budget authority, receipts, outlays, and the surplus or deficit
estimated to result as legislation is considered and enacted. CBO
develops estimates of the budgetary impact of bills reported by the
different congressional committees. For the many individual
transactions done under existing authorities (thus not requiring annual
legislation), CBO's estimates play no role in determining how much
budget authority must be obligated.[Footnote 23] However, OMB
interprets the scorekeeping guidelines to determine the costs that
should be recognized and recorded as an obligation at the time the
agency signs a contract or enters into a lease. It is not always
obvious whether a transaction or activity should be included in the
budget. Where there is a question, OMB normally follows the
recommendation of the 1967 President's Commission on Budget Concepts to
be comprehensive of the full range of federal agencies, programs, and
activities. However, under some circumstances, it may choose not to
record obligations and outlays up front.
Budget scorekeeping rules and OMB instructions on the budgetary
treatment of lease-purchases and leases of capital assets are published
in OMB Circular A-11. Revised in 2003 to address lease-backs from
partnerships, among other things, these instructions consider both the
government's legal obligation and how risk is shared between the
government and the contractor for three types of leases: capital
leases, lease-purchases, and operating leases. The instructions state
that when agencies enter into a capital lease contract or lease-
purchase,[Footnote 24] budget authority is scored in the year in which
the authority is first made available in the amount of the net present
value of the government's total estimated legal obligations over the
life of the contract. Alternatively, for operating leases that include
a cancellation clause, agencies only need budget authority sufficient
to cover the first year's lease payments, plus cancellation costs.
Figure 1 summarizes the criteria and other guidelines for defining an
operating lease.
Figure 1: Definition of an Operating Lease:
An operating lease meets all the following criteria: [A]
* Ownership of the asset remains with the lessor during the term of the
lease and is not transferred to the government at or shortly after the
end of the lease term;
* The lease does not contain a bargain-price purchase option;
* The lease term does not exceed 75 percent of the estimated economic
life of the asset.[B]
* The present value of the minimum lease payments over the life of the
lease does not exceed 90 percent of the fair market value of the asset
at the beginning of the lease term;
* The asset is a general-purpose asset rather than being for a special
purpose of the government and is not built to the unique specification
of the government as lessee.[C];
* There is a private sector market for the asset.
Source: OMB Circular A-11, Appendix B.
[A] According to OMB's scoring instructions, if the government ground-
leases property to a nonfederal party and subsequently leases back the
improvements, the lease will not be considered a lease-back from a
public/private partnership, as long as the lessor is a totally
nonfederal entity. Such lease-backs may be treated as operating leases
if they meet the criteria for an operating lease.
[B] Scoring instructions state that if the lease agreement contains an
option to renew that can be exercised without additional legislation,
it will be presumed that the option will be exercised.
[C] Scoring instructions state that if the project is constructed or
located on government land, it will be presumed to be for a special
purpose of the government.
[End of figure]
While we have previously reported that up-front funding permits
disclosure of the full costs to which the government is being
committed, OMB's budget scorekeeping instructions allow costly
operating leases to appear cheaper in the short term and have
encouraged an overreliance on them for satisfying long-term
needs.[Footnote 25]
Partnerships must conform with OMB's scorekeeping instructions. The
instructions for partnerships consider the degree of private
participation in the partnership to determine its scoring. Private
participation is judged by the level of substantial private
participation and private sector risk as evidenced by the absence of
substantial government risk. Substantial private participation means
(1) the nonfederal partner has a majority ownership share of the
partnership and its revenues, (2) the nonfederal partner has
contributed at least 20 percent of the total value of the assets owned
by the partnership, and (3) the government has not provided indirect
guarantees of the project, such as a rental agreement or a requirement
to pay higher rent if it reduces its use of space. If government risk
is considered high and private participation not deemed substantial,
the partnership would be considered governmental for budget purposes
and its transactions would be scored. Figure 2 presents OMB's
illustrative criteria for assessing private sector risk.
Figure 2: Criteria for Assessing Private Sector Risk:
The following types of illustrative criteria indicate ways in which a
project contains more private sector risk.
* There is no provision of government financing and no explicit
government guarantee of third-party financing;
* Risks incident to ownership of the asset (e.g., financial
responsibility for destruction or loss of the asset) remain with the
lessor unless the government was at fault for such losses;
* The asset is a general purpose asset rather than being for a special
purpose of the government and is not built to the unique specification
of the government as lessee;
* There is a private sector market for the asset;
* The project is not constructed on government land.
Source: OMB Circular A-11, Appendix B.
[End of figure]
Using ESPCs and partnerships, agencies have been allowed to spread the
costs of capital assets over several years. Agencies sometimes used
these financing mechanisms when they believed that timely, full, and
up-front appropriations would not be made available to support capital
needs. Moreover, they believe these alternative mechanisms enabled them
to avoid costs, such as higher utility bills associated with waiting
for appropriations. Nevertheless, several of the agencies we spoke with
agree that they could acquire capital less expensively through timely,
full, and up-front appropriations.
Energy Savings Performance Contracts:
ESPCs finance energy-saving capital improvements[Footnote 26] such as
lighting retrofits and ventilation systems for federal facilities
without the government recording the full cost up-front.[Footnote 27]
According to DOE, ESPCs have been used as an alternative financing
mechanism to finance over a billion dollars of energy system upgrades
and installations. Federal agencies' use of ESPCs was authorized and
encouraged by both Congress[Footnote 28] and the executive branch. In
Executive Order 13123, dated June 3, 1999, the executive branch defined
requirements for agencies to meet specific energy reduction goals and
supported the use of ESPCs to achieve them. The Energy Policy Act of
1992 and Executive Order 13123 require federal agencies to reduce their
consumption of energy in federal buildings. The act set a goal for the
agencies of lowering their consumption per gross square foot by 20
percent below fiscal year 1985 baseline consumption levels by fiscal
year 2000.[Footnote 29] Executive Order 13123 requires a 30 percent
reduction from 1985 levels by the year 2005 and a 35 percent reduction
by 2010.
Under the Energy Policy Act of 1992, federal agencies had the authority
to enter into ESPCs for as long as 25 years with qualified ESCOs that
purchase and install new energy systems in federal buildings. The ESCO
assumes much of the up-front capital costs and, in return, receives a
portion (nearly all) of the annual energy savings attributable to the
improvements until the principal and interest have been repaid. The
ESCOs guarantee the performance of the equipment installed, within
certain parameters, for the term of the ESPC. The agency makes the
payment to the ESCO from funds that the agency would otherwise have
used to pay the higher utility costs (which are lower because of the
ECM installed by the ESCO). Consequently, agencies argue that they will
not need an increase in future appropriations relative to the current
amount of appropriations in order to pay the ESCOs.[Footnote 30]
Agencies other than DOD and GSA[Footnote 31] must transfer 50 percent
of the energy savings realized from energy savings performance
contracts (after paying the negotiated amount to the contractor) to the
Treasury. The remaining 50 percent saved may be retained and is
available for additional energy and water conservation projects until
expended.[Footnote 32]
According to FEMP, 18 federal agencies and departments have implemented
ESPC projects worth $1.7 billion. Without ESPCs, agencies would have to
reassess their budget plans to accommodate investments in ECMs and/or
Congress would be asked to appropriate funds today to finance
investments to meet currently required energy consumption goals.
Proponents of ESPCs argue that, as an alternative financing method,
these contracts help agencies overcome the problem of insufficient
funding to meet federal energy reduction goals. Without ESPCs, agencies
would need to adjust their program plans within expected appropriation
levels to make energy efficiency improvements or possibly do nothing if
the funds are unavailable in a given year. In this regard, proponents
note that delays of even 1 year can result in greater utility,
maintenance, and other costs. Moreover, by using ESPCs, agencies did
not have to make some difficult trade-offs between purchasing ECMs and
other claims on resources.
Critics of ESPCs, however, point out that for any given ECM the direct
purchase of more efficient energy systems would allow all future
savings to accrue to the government, rather than paying out a
percentage of the savings to private contractors. In addition, the
government incurs certain costs in using an ESPC, such as the M&V fees
paid to the contractor, that it would not necessarily incur if the
energy improvements were financed up front with federal appropriations.
The ESCO's also pay a higher cost of capital than the federal
government. As a result, over the long term, financing ECMs through
ESPCs is likely to be more expensive than acquiring them through
timely, full, and up-front appropriations. Finally, dependence on the
annual budget cycle is the process by which decision makers weigh
competing federal priorities. Permitting ESPCs to be recorded in the
budget at less than their full cost up front affects this process,
possibly resulting in lower priorities receiving funding ahead of
higher priorities. Not addressing some difficult resource allocation
decisions is seen as an advantage to agencies. However, long-standing
budget concepts hold that a budget should be a forum for resource
allocation decisions and that all competing claims should be compared
on a consistent basis.
CBO and OMB disagree over the appropriate budget scoring for ESPCs. CBO
recognizes that the law enables agencies to use ESPCs to pay for ECMs
over a period of up to 25 years without an appropriation for the full
amount of the purchase price. However, the law does not prohibit
scoring the full cost of the contract up front. In CBO's view, the
obligation to make payments for the energy efficient equipment and the
financing costs is incurred when the government signs the ESPC.
Further, CBO believes it is consistent with governmentwide accounting
principles that the budget reflect this commitment as new obligations
at the time that an ESPC is signed. Accordingly, CBO scored recent ESPC
legislation[Footnote 33] such that the total (long-term) commitments to
an ESCO would be counted in the budget at the time the ESPC delivery
order is signed. In contrast, OMB recognizes obligations, budget
authority, and outlays for ESPCs on a year-to-year basis. According to
OMB staff, this decision was based on the savings component of ESPCs
and agencies' statutory authority to enter into a multiyear contract
even if funds are available only to pay for the first year of the
contract.
Public/Private Partnerships:
Partnerships are designed to tap the capital and expertise of the
private sector to improve or redevelop federal real property
assets.[Footnote 34] Partnerships are sometimes used when excess
capacity, such as unused land, exists within a federal asset. Ideally,
the partnerships are designed such that each participant makes
complementary contributions that offer benefits to all parties. From a
budget scoring perspective, recording an agency's full commitments up
front in the budget can be difficult because the precise level of an
agency's financial commitment and control in the partnership may be
unclear.
Congress has enacted a variety of laws that provide agencies with
authority to enter into partnerships with private firms. For instance,
VA possesses enhanced use (EU) lease authority to outlease federal real
property to private firms (see fig. 3). Alternatively, GSA possesses
limited authority for specific situations. For example, GSA entered
into a partnership with the Georgia Cooperative Services for the Blind
to operate a food court within the Atlanta Federal Center, using
authority provided by the Randolph Shepard Act.[Footnote 35] GSA has
also used authorities granted in other legislation, such as the Public
Buildings Cooperative Use Act of 1976, as amended,[Footnote 36] and the
National Historic Preservation Act, as amended,[Footnote 37] to work
with nonfederal partners. Despite the significance of GSA's role in
federal property management, its limited authority to enter into
partnerships has prevented it from taking a substantive role in
partnership activities.
Figure 3: Elements of VA's EU Lease Authority:
The Secretary of VA has unique statutory authority (38 U.S.C.§§ 8161
- 8169) to enter into long-term agreements called "enhanced use"
leases. The basic elements of this authority are:
* The leases can be for up to 75 years in instances of new or
substantial construction;
* The real property must be controlled by VA;
* The lease allows for non-VA uses or activities on VA property;
* The overall lease must enhance a VA mission or program;
* In return for the lease, VA may obtain monetary consideration,
services, or facilities or other benefits from the operation of the
non-VA uses so long as the consideration is determined by the Secretary
as being "fair consideration;" and;
* Upon expiration of the enhanced use lease, all improvements become
the property of VA.
Source: Department of Veterans Affairs.
[End of figure]
Proponents of partnerships argue that the approach provides a
realistic, less costly alternative to leasing when planning and
budgeting for real property needs. Proponents also note that federal
partners benefit from improved, modernized, or new facilities plus a
minority share of the income stream generated by the partnership or use
of the asset at a lower cost than a commercial lease.
Critics of partnerships caution that these ventures are not the least
expensive means of meeting capital needs, although they may appear to
be in the short term. They remind policymakers that up-front funding
with appropriated funds is the least expensive way to obtain assets and
results in the inclusion of the government's long-term commitments in
the budget.
Our prior work has shown that, as part of a capital review and approval
process, leading organizations develop decision packages, such as
business case analyses, to justify capital project requests.[Footnote
38] A business case analysis is a planning and decision support tool
used to ensure that (1) the objectives for a proposed facility-related
investment are clearly defined, (2) a broad range of alternatives for
meeting the objectives are developed, (3) the alternatives are
evaluated to determine how well the objectives will be met, and (4)
trade-offs are explicit. The overriding purpose of a business case
analysis is to make transparent to the various decision making and
operating groups all of the objectives to be met by the investment, the
underlying assumptions, and the attendant costs and potential
consequences of alternative questions. Business case analyses are
supported by detailed economic and financial analyses such as cost-
benefit, return on investment, life-cycle cost, and comparative
alternative analyses, and recommend the most cost-effective option.
Various Features Enabled Agencies to Delay Recognition in the Budget:
ESPCs and partnership arrangements were authorized by Congress. With
these arrangements, both the government and third parties share the
risk of a long-term financial commitment. However, agencies were not
required to reflect the full cost of these commitments up front in the
budget when the commitments were being made. For example, ESPCs
finance energy-efficient equipment over time by using savings in
agencies' utility bills to repay ESCOs for the up-front equipment and
installation costs. Because the ESCOs are repaid over time, the full
up-front costs of ECMs are not reflected in the budget. (Fig. 8 on
page 55 illustrates how ESPCs affect agencies' cash flows before,
during, and after the contract term.) With respect to most of the
partnerships we reviewed, scoring decisions were driven by the transfer
of government land from agencies to third parties. Case study agencies
sometimes leased assets in short-term increments that third parties
constructed on the transferred land specifically for the government's
use and benefit. As a result, the full cost of the assets were not
required to be reflected in the budget.[Footnote 39] Given that the
federal budget is primarily measured on a cash and obligations basis,
some of the partnerships we examined were completely invisible in the
budget because they involved noncash transactions. The financial
commitment of the government is illustrated in figure 4--although costs
through third-party financing that appear in the budget may be
initially lower, the government is committed to years of future
payments.
Figure 4: Third-Party Financing Delays Recognition of Full Costs to
Taxpayers:
[See PDF for image]
[End of figure]
ESPC Commitments Are Not Fully Recognized Up Front in the Budget:
ESPCs represent long-term commitments of the government. Agencies
generally retain control of capital assets acquired through ESPCs for
their entire life cycle, and frequently contractors transfer title of
the assets to the government after the assets are installed and
accepted by the government. Moreover, the term of ESPC delivery orders
spans as long as 25 years. Finally, agencies' termination liability for
ESPCs typically corresponds to their outstanding principal balance.
Although these arrangements represent long-term commitments, funds for
ESPCs are obligated on an annual basis. Therefore, the budget does not
recognize the government's long-term commitment up front, when
decisions are made. This policy was formalized in a 1998 OMB
memorandum[Footnote 40] that stated ESPC obligations, budget authority,
and outlays would be recognized on an annual basis. The memorandum did
not discuss OMB's rationale for scoring ESPCs in this manner. According
to OMB staff, this memorandum reflected OMB's early examination of the
issues. Specifically, the policy was based on the savings component of
the contracts and the statutory authority to enter into a multiyear
contract even if funds are available only to pay for the first year of
the contract.
Long-term Partnership Arrangements Were Not Fully Recognized Up Front
in the Budget:
Capital assets acquired through the partnership arrangements we
reviewed were structured such that third parties have an ongoing, long-
term relationship with the government. However, OMB's budget scoring
instructions required that only the short-term costs associated with
assets acquired through case study partnerships be scored in the
budget. As shown previously in figure 1, the definition of an operating
lease (which permits obligations to be scored annually) specifies that
the lease term may not exceed 75 percent of the estimated economic life
of the asset. Assets we reviewed were constructed in areas where case
study agencies have long maintained a presence and have a continuing
mission. It seems unlikely that the agencies will vacate or abandon
these assets before the end of their economically useful lives. To the
extent agencies continue to occupy leased spaces, the 75 percent
criteria for an operating lease may be exceeded.
For example, through a series of transactions, DOE entered into a
partnership with the nonprofit UT-Battelle Development Corporation
(UTBDC) to revitalize its Oak Ridge National Laboratory (ORNL) in the
state of Tennessee.[Footnote 41] ORNL was first established in 1943 and
is DOE's largest science and energy laboratory. Many of the buildings
on the ORNL reservation have become obsolete, dilapidated, and
expensive to maintain. Accordingly, UTBDC arranged for the sale of
bonds through Keenan Development Associates of Tennessee to finance and
construct three general-use office buildings at ORNL (specifically to
support DOE research). UT-Battelle, LLC, on behalf of DOE, leased the
buildings through a series of leases extending to 25 years and
involving UTBDC and Keenan. (See app. III for a full explanation of
this complicated arrangement.) DOE's ORNL Project Manager told us that,
even if ORNL's mission was downsized, it was unlikely that DOE would
terminate any of the leases of the three new, state-of-the art
buildings to reoccupy the now empty, dilapidated buildings. Figure 5
shows one vacant office building and an artist's rendition of the
revitalized ORNL reservation, including the three privately financed
buildings.
Figure 5: Buildings on the ORNL Reservation:
[See PDF for image]
[A] Vacant office building at ORNL.
[B] Artist's rendition of revitalized reservation.
[End of figure]
Evidence of ORNL's long-term commitment is further bolstered by
Standard and Poor's A+ rating of the ORNL bond issuance. The rating
report stated that a strong lease revenue stream from DOE, for a period
of up to 25 years, would be pledged as security for the payment of the
bonds. Moreover, the unique mission of ORNL makes it unlikely that DOE
will move its operations from the Oak Ridge site. DOE officials stated
that the likelihood that ORNL will close during the term of the bonds
is very low, so DOE is unlikely to terminate any part of the leases.
According to DOE officials, DOE transferred the government-owned land
on which the buildings are located to UTBDC via quitclaim deed[Footnote
42] so that appropriations for the full, up-front costs of the three
buildings were unnecessary. DOE later approved the facility subleases
between UTBDC and UT-Battelle, LLC. DOE then obtained the use of the
newly constructed buildings from UT-Battelle, LLC, reimbursing UT-
Battelle, LLC, for the sublease rents. DOE officials told us the deal
was deliberately structured with a quitclaim deed to ensure that the
arrangement was scored as an operating lease rather than a capital
lease.[Footnote 43] Because OMB allowed DOE to record this arrangement
as an operating lease, DOE needed to obligate only the annual cost of
the lease payments, rather than the full cost of the
construction.[Footnote 44]
In another case study, VA entered into an enhanced use (EU) lease for
up to 35 years with the Dekalb County Development Authority (the
Authority)[Footnote 45] to finance and construct VA's Atlanta Regional
Office (VARO) building and parking area.[Footnote 46] Dekalb County
issued 35-year revenue bonds to finance the project. VA officials said
construction of the new building allowed the agency to collocate its
regional office with an existing VA medical center and provide enhanced
"one-stop" service for veterans. Although VA leases the facilities from
the Authority in 2-year increments, there is no current plan to vacate
the property in the near future and the leases automatically renew
unless VA takes positive action to terminate.[Footnote 47]
Additionally, VA agreed not to replace the regional office building
financed by the Authority with another regional administration or
headquarters building in Georgia using its EU leasing authority during
the term of the bonds. VA may enter into another EU lease in the
Atlanta region so long as the new building does not disrupt VA's
occupancy within the collocated office.
VA did not need to obligate the full up-front cost of the regional
office building and parking area because it used its EU lease authority
to outlease the government-owned land to the Authority on which the
Authority could build. VA then leased the newly constructed building
and parking area from the Authority's developer through 2-year
operating leases, which automatically renew for up to nine consecutive
terms unless VA takes positive action to terminate the automatic
renewal clause. Accordingly, OMB only required the annual lease
payments and any termination costs[Footnote 48] to be reflected in the
budget.
Some Partnerships May Not Be Included in the Budget Because They
Involve Noncash Consideration:
In some cases, partnerships are arranged for reasons other than an
agency's belief that appropriations are not available. For example, VA
entered into an EU lease with the City of Vancouver's Housing Authority
to construct an estimated $4 million homeless shelter on VA
property.[Footnote 49] In exchange, veterans receive priority placement
in 50 percent of the shelter units; VA receives no cash consideration.
Because VA can discharge patients into the homeless shelter rather than
extending inpatient care in VA medical facilities, VA estimated that it
avoids costs of roughly $1.8 million annually. Additionally, VA
anticipated that the homeless shelter would provide veterans with
greater outpatient services and improve the availability of affordable
housing for single homeless individuals. Because the partnership
involves no cash consideration, it is not reflected in the budget.
Higher Interest Rates and Other Factors May Increase the Cost of Third-
Party Financing Compared to Timely, Full, and Up-Front Appropriations:
A number of factors may cause third-party financing to be more
expensive than timely, full, and up-front appropriations. For example,
case study agencies incurred a higher rate of interest by using ESPCs
and partnerships than if they had obtained that same capital through
timely, full, and up-front appropriations.[Footnote 50] Also, for
ESPCs, officials told us that the government likely incurred additional
costs for the measurement and verification (M&V) of equipment
performance. In our six ESPC case studies, use of ESPCs increased the
government's costs of acquiring ECMs by 8 to 56 percent compared to the
use of timely, full, and up-front appropriations. None of the
partnership case studies lent themselves to this type of cost analysis
because comparable data were not available. Some of the partnerships
did not involve cash consideration. For others, although the government
incurred higher interest costs compared to up-front funding, we did not
evaluate claims that other factors such as lower labor costs and fewer
bureaucratic requirements might lower costs because data were not
readily available. Thus, we were unable to judge whether partnerships
could be less expensive overall. For both ESPCs and partnerships,
agency officials said they did not request full, up-front
appropriations to finance the specific capital projects we reviewed.
Frequently, they said this was because they did not believe funds would
be available in a timely manner and they had statutory authority to use
the alternative mechanisms. However, there are insufficient data to
measure this effect.
Acquiring Capital through ESPCs Is More Expensive Than Acquiring the
Same Capital through Timely, Full, and Up-Front Appropriations in Our
Case Studies:
Since the federal government's cost of capital is lower than that of
the private sector, alternative financing mechanisms may be more
expensive than timely, full, and up-front appropriations. Accordingly,
all case study agencies could have acquired the same ECMs less
expensively through timely, full, and up-front appropriations than
through ESPCs.
In addition to a higher cost of capital, agencies also likely incur
additional M&V costs when they finance ECMs through ESPCs rather than
timely, full, and up-front appropriations. Agencies contract for M&V of
energy savings financed through ESPCs because they are required to show
that annual savings generated by ECMs meet or exceed annual contractor
payments. M&V of savings also acts as insurance; if actual savings fall
below those guaranteed by the contractor, the contractor may be
obligated to take corrective actions at its own expense. Officials we
spoke with said they believed that M&V resulted in higher sustained
savings but is an expense that would not be incurred if the ECMs were
acquired through timely, full, and up-front appropriations.
Representatives from the ESCOs said that their private sector clients
do not always purchase M&V, and, if they do, it is for a shorter period
than contracts secured by the federal government.
Table 1 presents cost comparisons using the installation and
construction price of ECMs (based on delivery order files) as a proxy
estimate for timely, full, and up-front appropriations costs.[Footnote
51] It shows that ECMs obtained through our six ESPC case studies might
be roughly 8 to 56 percent more expensive than they could have been for
the same ECMs had they been obtained through timely, full, and up-front
appropriations. The percentage difference between financing through
ESPCs and estimated timely, full, and up-front appropriations is shown
in the far right column. The difference in:
costs between the two financing mechanisms is a function of (1) the
higher cost of capital incurred through ESPC financing and (2) the M&V
costs incurred through ESPC financing.[Footnote 52]
Table 1: Cost Analysis of Six ESPCs:
Dollars in millions, present value.
Navy Region South West (10 years);
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $13.66;
Cost of ECMs financed through ESPCs[B]: $14.69;
Percentage increase due to financing: 8%.
Patuxent River Naval Air Station (20 years);
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $4.33;
Cost of ECMs financed through ESPCs[B]: $5.77;
Percentage increase due to financing: 33%.
Naval Submarine Base Bangor (9 years);
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $4.33;
Cost of ECMs financed through ESPCs[B]: $5.34;
Percentage increase due to financing: 23%.
GSA Gulfport Federal Courthouse (17 years);
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $1.60;
Cost of ECMs financed through ESPCs[B]: $2.50;
Percentage increase due to financing: 56%.
GSA North Carolina bundled sites (19 years);
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $1.39;
Cost of ECMs financed through ESPCs[B]: $1.93;
Percentage increase due to financing: 39%.
GSA Atlanta bundled sites (20 years);
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $6.15;
Cost of ECMs financed through ESPCs[B]: $7.78;
Percentage increase due to financing: 27%.
Source: GAO analysis of ESPC case study delivery order files.
Note: Analysis based on delivery order files as signed upon final award
of contracts. In some cases, the government later modified these
delivery orders to add more ECMs.
[A] This column represents the present value of the installation and
construction price for the ECMs. It does not include operations and
maintenance (O&M) expenses, since these costs typically are
appropriated annually. The price could be viewed as a proxy for the
amount Congress would have had to appropriate had the ECMs been
financed through timely, full, and up-front appropriations rather than
an ESPC. However, because it is difficult to predict what the true cost
of the asset would have been had it been financed differently, this is
not a precise measure.
[B] This column represents the present value of the installation and
construction price for the ECMs under an ESPC. In addition, it includes
the interest and M&V costs that must be paid under an ESPC. It does not
include O&M expenses, since these costs typically are appropriated
annually.
[End of table]
The performance of ECMs installed through the use of ESPCs are
guaranteed to reduce energy use during the term of the contract so that
payments to the contractor can be made from the savings from lower
utility bills. ESPCs contain assumptions for such things as hours of
operation and ECM efficiency which, taken together, determine estimated
savings. However, if the assumptions are incorrect and estimated
savings are not achieved, the agency is still required by contract to
pay the ESCO the agreed-upon savings specified in the ESPC. According
to agency officials, ECMs may continue to accrue savings beyond the
contract cycle as they continue to operate more efficiently than the
equipment they replace. The additional savings along with the savings
realized during the contract cycle may cover the entire cost of the
equipment. (See app. II for additional detail on verification of ESPC
savings.)
As shown in figure 6, for all six ESPC case studies, contract cycle
energy cost savings specified by the contractor did not fully cover
total contract cycle costs (including O&M expenses) because agencies
made up-front payments. All six of the case study ESPCs used a
combination of funds from their existing budgets and third-party
financing via an ESPC to implement packages of ECMs. The up-front
payments from their existing budgets covered the difference between
total contract cycle costs and savings. Accordingly, the agencies
reduced the amount they had to finance through ESPCs, thereby reducing
their interest payments. In the case of the ESPCs for the Bangor Naval
Submarine Base and Navy Region Southwest, some of these up-front
payments came out of a special appropriation provided to address energy
supply shortages in the West. With respect to the ESPCs at North
Carolina, Atlanta, and Patuxent River, funds used to pay down the
principal on the ESPC had previously been appropriated to renovate,
renew, or repair old energy consuming systems. Since implementation of
the ESPC made these activities unnecessary and may be providing
benefits other than costs savings, such as maintaining an acceptable
level of service, the funds from the avoided costs were put toward the
ESPC. According to guidance issued by DOE's Federal Energy Management
Program (FEMP), agencies are permitted to use funds generated by these
types of avoided costs to pay for ESPCs. For the six ESPC case studies,
up-front payments ranged between 2 and 45 percent of total contract
cycle costs.[Footnote 53]
Figure 6: Contract Cycle Costs, Up-Front Payments, and Savings of Six
ESPCs:
[See PDF for image]
Note: Savings amounts are specified in the delivery orders. Our
analysis was based on final award of delivery order files. In some
cases, the government later modified these delivery orders to add more
ECMs.
[End of figure]
Full Cost of Partnerships Is Unclear:
None of our five partnership case studies lent themselves to a
budgetary cost analysis because comparable data were not available. Two
did not involve cash transfers of any kind. In the case of the other
three, it was unclear how much the projects would have cost had they
used timely, full, and up-front appropriations rather than partnership
financing. In these cases, agencies sometimes incurred higher interest
costs by using partnership financing rather than timely, full, and up-
front appropriations. For example, DOE benefited from the use of a
roughly $70 million private bond offering to finance the revitalization
of three buildings at ORNL. The financing structure[Footnote 54] used
over the course of 25 years means DOE actually will obligate funds
equaling approximately $96 million, present value (PV). Similarly, VA
will obligate funds equaling approximately $43 million (PV) over 35
years to pay off the approximately $33 million in bonds the Authority
issued to finance the construction of an office building and parking
area in Dekalb County, Ga.
However, officials said that other factors associated with
partnerships, such as lower labor costs and fewer bureaucratic
requirements, could make partnership financing overall less expensive
than financing through timely, full, and up-front appropriations. For
example, officials with DOE's M&O contractor said that, in the case of
the ORNL revitalization project, greater efficiencies existed in
private sector construction since the government would have had to
enter into more costly union labor agreements had it financed the
project through timely, full, and up-front appropriations. Similarly, a
VA official said that using an EU lease to construct an energy center
would be less expensive than financing the asset through timely, full,
and up-front appropriations because federal labor agreements and
acquisition regulations created inefficiencies in federal
construction. As part of this engagement, we did not analyze these
claims to determine whether the efficiencies associated with private
sector construction would offset the higher interest costs of
partnership financing.
Uncertainty and Timing of Appropriations Affect Cost Effectiveness:
Some agency officials said that ESPCs and partnerships are cost
effective because they allow agencies to acquire capital more quickly
than through appropriations. They noted that it is uncertain when or
whether timely, full, and up-front appropriations will be made
available. Officials from Navy, DOE, and GSA expressed their belief
that funds obtained through third parties would be available much more
quickly than through appropriations. Consequently, officials said that
agencies could accrue more savings and avoid more costs using ESPCs and
partnerships than they would have if they had waited for
appropriations. For example, Navy officials said that, had they not
used ESPCs, Naval installations would have had to pay higher utility
bills while waiting for appropriations to finance ECMs. Similarly, at
ORNL, one DOE official pointed out that although the idea for three
privately financed buildings was conceived at the same time as the
highest priority federally funded building, the three privately
financed buildings were completed and occupied at least a year in
advance of the one funded through appropriations.[Footnote 55] Other
DOE officials said that the costs of maintaining obsolete, dilapidated
buildings at ORNL while waiting for appropriations would have added to
the cost of waiting for full, up-front appropriations.[Footnote 56]
Thus, according to these officials, capital obtained through ESPCs and
partnerships may be less expensive relative to full, up-front
appropriations than it seems.
Since the agencies did not request additional appropriations or adjust
their plans to accommodate needed capital investments, it cannot be
known whether agencies were correct in assuming that timely
appropriations would not be available. Agencies are responsible for
establishing funding priorities to achieve their missions, including
capital needs and mandated energy savings. Capital plans supported by
strong analysis could help them in setting priorities for funding
requests.
Agencies Did Not Request Full, Up-Front Appropriations before Entering
into ESPCs or Partnerships:
Given the federal government's ability to obtain capital at lower
interest rates than private companies, officials from each of our case
study agencies agreed that timely, full, and up-front appropriations
were the least-cost alternative for financing capital acquisitions.
However, officials also stated that they did not request additional
appropriations for the case studies we reviewed because they were
authorized to use the alternative financing mechanism. Further, they
did not believe appropriations would have been available in a timely
manner. For example, DOE officials said that they had not requested
full, up-front appropriations for certain aspects of the ORNL
revitalization project because, in the past, they had tried and failed
to obtain funding for similar projects. The Director of DOE's Office of
Science stated that it was particularly difficult to obtain funding for
general use office buildings compared to buildings specifically
designed for scientific research. However, the poor condition of these
general use buildings negatively affected DOE's ability to recruit and
retain high-quality scientists. Because the agencies never requested
appropriations for these specific projects, it is impossible to know
whether their assumptions were correct.
GSA and Navy officials also said recent declines in up-front
appropriations for ECMs affected their decision to use ESPCs. For
example, according to GSA officials, GSA's budget authority for energy
efficiency projects declined from $20 million in fiscal year 1999 to
$4.2 million in fiscal year 2004, and it received no funds in fiscal
years 2002 and 2003. They also pointed to GSA's $6 billion backlog of
identified repair and alteration needs. According to Navy officials,
appropriations for its Energy Conservation Improvement Program dropped
from $21.7 million in fiscal year 1999 to zero dollars in fiscal year
2000. Although funding has increased in recent years, it still remains
well below 1999 levels. According to the Director of the Navy's Energy
Programs Division, the department receives less than 10 percent of the
estimated $140 million needed each year to meet energy savings goals.
Navy officials said that other priorities in the Navy's budget had
taken precedence over energy reduction projects. According to FEMP,
ESPC projects worth $1.7 billion have been implemented by 18 federal
agencies and departments. Without ESPCs, agencies would have to
reassess their budget plans to accommodate investments in ECMs and/or
Congress would be asked to appropriate funds today to meet currently
required energy consumption standards.
Officials also pointed out that agencies had been granted statutory
authority to use ESPCs and partnerships. For example, the Energy Policy
Act of 1992 authorized agencies to fund ECMs through ESPCs.
Additionally, OMB issued two memorandums encouraging agencies to use
ESPCs to achieve long-term energy savings. In addition, the Atomic
Energy Act granted DOE authority to give away land for mission purposes
and enabled it to finance improvements on that land through private
sector financing. Similarly, VA officials said that the agency's EU
lease authority specifically enabled it to enter into partnerships with
nonfederal sector entities to finance capital.
Different Financing Alternatives Present Different Implementation and
Monitoring Challenges:
Third-party financing can make it easier for agencies to manage in the
short term within a given amount of budget authority but may have
additional long-term costs. With ESPCs, case study agencies relied
heavily on ESCOs to recommend potential ECMs, install the equipment,
and then verify that the recommended improvements yield intended
results. Partnership arrangements generally entail a government agency
engaging another party to, among other things, renovate, construct,
operate, or maintain a public facility. Such relationships increase the
need for effective implementation and monitoring by agencies to ensure
the government's interests are protected. For example, reliance on
outside parties can leave the government open to problems resulting
from conflicts of interest and presents monitoring challenges. The
ESPCs and one of the partnerships we reviewed highlighted these
vulnerabilities. An evaluation of funding alternatives was not always
done to determine the most appropriate way of funding capital projects.
Finally, VA has used partnership financing to engage in an activity
that is not related to VA's mission and which it ordinarily would not
fund through full, up-front appropriations.
Third-Party Financing Increases the Risk of Conflicts of Interest and
Presents Monitoring Challenges:
As the government teams with outside parties to acquire capital, it
must ensure that its welfare is protected from conflicts of interest.
ESPCs introduce concerns over conflicts of interest due to the heavy
reliance upon ESCOs. Partnership arrangements can also create
management challenges as outside participants gain influence over
projects. Active participation and scrutiny by agencies can help ensure
the government's interests are not compromised.
ESPC Process Creates Potential for Conflicts of Interest:
Once agencies decide to use an ESPC and select an ESCO to work with,
they must ensure that the government's interests are protected from the
potential conflicts of interest that may arise from the ESCO's
comprehensive role in recommending what ECMs are needed and then in
monitoring and verifying the performance of the equipment that they
recommended, installed, and guaranteed.[Footnote 57] For example, an
ESCO prepares a Detailed Energy Survey (DES), which is an investment-
grade audit that determines what ECMs will be installed as part of the
ESPC. This serves as the basis for the project's estimated savings,
M&V, and O&M, and is used to develop the final energy project proposal.
After the ESCO installs the ECMs, it measures and verifies that the
contractually guaranteed savings it estimated are being
achieved.[Footnote 58] To ensure the government's interests were
protected, staff at both GSA and the Navy reviewed documentation and
participated throughout the ESPC process. Given its decentralized
process for managing ESPCs, GSA uses a FEMP project facilitator for
technical assistance on its ESPC delivery orders. To the extent desired
by federal agencies using its Super ESPC, FEMP provides assistance
through training, project development tools, and technical
support.[Footnote 59] According to FEMP and Navy officials, the Navy's
centralized process for managing ESPCs enables it to maintain
sufficient in-house technical expertise and the Navy does not typically
employ FEMP's assistance. However, the Navy frequently uses FEMP's free
services, such as reviewing proposals, and has purchased FEMP's support
in special circumstances.
Although both GSA and the Navy took an active role in negotiating the
case study ESPCs to protect the government's interests, the process by
which these contracts are structured can still introduce problems
resulting from ESCO's conflicts of interest. For example, case study
agencies relied heavily upon the ESCOs to estimate facilities' energy
use after ECM installation compared to what baseline energy use would
have been if ECMs had not been installed. Projected energy savings are
calculated by subtracting estimated energy use after ECMs have been
installed from baseline energy use. According to FEMP guidance, these
calculations should be examined in detail because they are the basis
for determining whether the contractually guaranteed savings are
achieved. Nonetheless, a number of Army Audit Agency reports[Footnote
60] issued over the last several years stated that energy savings
baselines established by the ESCOs were faulty, resulting in
overpayments to the ESCO. For example, some baselines used incorrect
assumptions such as overstated operating hours. Representatives from
two ESCOs noted that the greater the experience of the government team,
the greater the intensity of the negotiations.[Footnote 61] Therefore,
FEMP assistance is particularly important for agencies with relatively
little ESPC experience. Given agencies' reliance on the ESCOs in the
ESPC process, agencies must be diligent to ensure that the government's
best interests are protected. Employing best practices in using ESPCs
also may provide opportunities to better ensure the government receives
the best value for its investment.[Footnote 62]
Monitoring Challenges Existed for Complicated ORNL Partnership:
The partnership between DOE, its management and operations (M&O)
contractor UT-Battelle, LLC, and UTBDC, created monitoring challenges.
Although we found no evidence of fraud, waste, or abuse, these
challenges were created when officers of the M&O contractor recommended
that DOE transfer land, without charge, to UTBDC.[Footnote 63] The same
officers of the M&O contractor that recommended this course of action
to DOE also served as officers of UTBDC, the organization that received
the land. UT-Battelle, LLC, contracted with UTBDC, which arranged for
the private financing to construct three general-use office buildings.
Accordingly, because the M&O contractor, not DOE, was directly involved
in the contract, DOE was presented with monitoring challenges. DOE
counsel both at Oak Ridge and headquarters told us that DOE's risk was
minimal and that monitoring of the partnership was not necessary. At
Oak Ridge, counsel told us that so long as the end product was what
they wanted, DOE did not have much of a role. At headquarters, counsel
told us that DOE does not provide oversight or micromanage how M&O
contractors work with subcontractors. Further, we were told that DOE
does not question M&O contractors' practices because DOE officials
believe these contractors to be trustworthy. Nonetheless, the primary
purpose of the partnership was to obtain facilities for DOE's use and
ultimately the revenue stream supporting the financing will be paid
through DOE appropriations.[Footnote 64] Thus, we believe greater
monitoring and oversight was warranted to ensure that the contractor
operates in the government's best interest.
Figure 7: Relationship Between ORNL Parties:
[See PDF for image]
[End of figure]
Business Case Analysis Needed to Ensure Third-Party Financing Is in the
Government's Best Interest:
Our prior work has shown that, as part of a capital review and approval
process, leading organizations develop decision packages, such as
business case analyses, to justify capital project requests.[Footnote
65] These packages are supported by detailed economic and financial
analyses such as cost-benefit, return on investment, life-cycle costs,
and comparative alternative analyses, and recommend the most cost-
effective option. Both OMB and our guidance stress that, when a
performance gap between needed and current capabilities has been
identified, it is important that organizations carefully consider how
best to bridge the gap by identifying and evaluating a full range of
alternatives to construct or purchase a new capital asset. This type of
analysis was not always performed for the case studies we reviewed. For
example, large buy-downs of ESPC principal raised questions about the
need for ESPC financing. A business case analysis might have
demonstrated that sufficient funds were available to purchase ECMs in
smaller, useable segments, when technically feasible. In addition, not
all partnerships included a business case analysis to determine whether
third-party financing was the most cost-effective alternative.
Large Buy-Downs of Principal Raise Questions About the Need for ESPC
Financing:
One key attraction for using ESPCs is that they enable agencies to
acquire ECMs even if funds are available only to pay for the first year
of the contract. However, three of the six case studies we reviewed
obligated and paid a significant portion of the total cost of the ECMs
in the first year of the contract.[Footnote 66] These large buy-downs
of principal avoided repair, replacement, and renovation costs as a
result of implementing the ESPC. They also imply opportunities exist to
acquire ECMs in smaller, useful segments, when technically
feasible[Footnote 67] with timely, full, and up-front appropriations
instead of through ESPCs. For example, agencies could individually
acquire an ECM such as an air chiller without bundling it into an ESPC.
Navy and GSA officials indicated they typically did not consider
financing ECMs through useful segments or through full and up-front
appropriations. They also told us they did not did not perform a
business case analysis before deciding to use ESPCs because of the
administrative cost of such an analysis since they believed there was
no other viable option. Officials explained that their agencies did not
request full, up-front appropriations since appropriations might not
have been made available in a timely manner and the use of ESPCs had
been authorized. Analyzing the full range of funding alternatives would
help agencies determine if acquiring ECMs in these useful segments
would be a more cost-effective alternative.
As previously discussed, FEMP guidance permits agencies to apply
avoided repair or replacement expenses for large equipment as one-time
cost savings to buy-down principal. For three of our case studies that
followed this guidance, these one-time savings were approximately 7
percent, 38 percent, and 39 percent of contract cycle costs. According
to GSA and Navy officials, these funds had already been appropriated
for the repair or replacement of old equipment. The ESPC made this
repair or replacement unnecessary and thus freed-up funds for other
uses, such as buying down the ESPC principal. Although paying down
principal up-front has the benefit of reducing financing costs to the
government, the availability of these funds highlights the need for an
analysis of the feasibility of purchasing ECMs in useful segments
rather than through an ESPC.
Agency officials expressed concern that acquiring ECMs in smaller,
useful segments would mean that some energy inefficient equipment would
be kept in use longer than if acquired in bulk through the ESPCs. In
addition, they noted that buying the ECMs in smaller quantities might
cause the government to lose economies of scale achieved through the
larger contracts. This concern was echoed by an ESCO representative,
who pointed out that fixed costs for smaller contracts would be similar
to those of larger projects. Nonetheless, given the higher financing
and likely additional M&V expenses, agencies should formally assess the
costs, on a present value basis, to determine the most cost-effective
alternative.
Agencies Did Not Always Assess the Cost-Effectiveness of Partnership
Arrangements:
Three of the five partnership arrangements we reviewed were undertaken
to obtain project financing.[Footnote 68] Leading capital practices for
capital decision-making would call for business case analyses in such
cases, which include analyzing the full range of funding alternatives.
However, for two of these three partnerships, agencies did not assess
the full range of alternatives to determine how best to fund the
project. For example, DOE officials could not provide evidence that the
department had prepared or reviewed a business case analysis for the
financing arrangement at ORNL. While ORNL's M&O contractor, UT-
Battelle, LLC told DOE employees that private financing of three
general-use office buildings was in the government's best interest, in
our opinion the data provided to DOE were summarized at such a high
level that DOE could not have done a comparative analysis of financing
alternatives. Although UT-Battelle, LLC's detailed analysis was readily
available to the department, UT-Battelle, LLC officials told us that
data were not requested from nor provided to DOE. This analysis
contained much greater detail, including an analysis of the cost of
maintaining old, dilapidated buildings but did not analyze timely up-
front appropriations as an alternative. DOE staff informed us that they
had a good relationship with UT-Battelle, LLC and had no reason to
doubt the summary analysis provided. However, a memorandum issued by
DOE's Assistant General Counsel for General Law regarding the
applicability of OMB Circular A-11 to the Oak Ridge National Laboratory
land transfer proposal said that the department's policy would require
DOE to do a comparative cost-analysis between using appropriated funds
to build the facilities now and the cost of funding UT-Battelle, LLC's
sublease payments.
VA did not compare the cost of financing the regional office building
through up-front appropriations to the cost of financing it through
Georgia's DeKalb County Development Authority. To avoid steep rent
increases in GSA leased space and to collocate its regional office with
an existing medical center, VA formed a partnership with the Authority
to finance the construction of its new regional office building on VA-
owned land. Although VA prepared a business case analysis comparing
leasing from GSA to financing construction through the Authority, VA
told us it did not compare the cost of financing through the Authority
versus up-front appropriations.
One Partnership Led to Involvement in a Nonmission-Related Activity:
VA has used partnership financing to engage in an activity that is not
related to its mission and in which it ordinarily would not fund
through full, up-front appropriations. In one instance VA used its EU
lease authority to construct a power plant for its North Chicago
campus. VA officials said that it is doubtful that VA would ever
construct and operate a power plant on its own since (1) power
generation is not a core activity within VA's mission and (2) VA does
not possess the necessary expertise. However, according to VA
officials, the Navy, the only provider of steam in the area, had been
charging VA rates above those charged by other suppliers to consumers
in neighboring areas of Chicago and this EU lease enabled VA to reduce
its energy costs.
Conclusions:
ESPCs and partnerships that we reviewed were authorized by Congress.
The financing approaches used in many of the case studies were
structured to include features for which OMB did not require up-front
budget recognition even though they established long-term commitments
of the government. One or more of the following features were used by
case study agencies: (1) the transfer of government land to third
parties, (2) use of a third-party rather than the U.S. Treasury to
finance assets over time, (3) use of short-term leases for potentially
long-term needs, (4) noncash transactions, (5) contractually guaranteed
savings, and (6) statutory authority to enter into ESPCs without funds
to obligate the full contract price. In the case studies we reviewed,
the capital assets acquired offered benefits to the government such as
energy conservation and a collocated VA regional office and medical
center. It is not the purpose of this report to second guess the
benefits of the assets. This report focuses only on the budgetary
process of justifying the means of acquiring and financing assets. Even
though project financing may be obtained more quickly by using
alternative financing mechanisms, these mechanisms do not disclose the
federal government's measure of long-term obligations in the budget. As
a result, when resource allocation decisions are made, costs are not
shown on comparable bases. This can favor capital programs financed
through these mechanisms over other programs (including capital) that
include their full costs up front in the budget.
In our work on capital planning, we noted that leading practices
include analyzing alternative approaches to financing capital by using
methods, such as net present value analysis, to analyze relevant
alternatives to address capital needs. OMB's capital planning guidance
also suggests that agencies need to select the alternative with the
most cost-effective results over the long term, based on a present
value analysis. This analysis would include all relevant federal
financing costs associated with the alternatives and any potential
savings that can be attributed to the various alternatives.
Long-standing federal budget concepts and our own work reinforce the
principle that full accountability for budgetary decisions is best
guaranteed by recognizing the full costs of federal initiatives at the
time when the decision is made to commit federal resources. One way to
ensure that costs of assets used for long-term commitments are
appropriately considered in the budget would be to score up front the
expected payments over the same period of time used to analyze
ownership options. This would require going beyond the strict terms of
a proposed transaction and scoring based on the substance of the deal.
Although ensuring the validity of agencies' long-term plans may pose
implementation challenges, such as the need to validate agencies' long-
term capital requirements, such scoring could result in a better
reflection of the government's full commitment.
In addition to potentially affecting budget decisions, our case studies
showed that funding capital projects through alternative financing
mechanisms may be more expensive to the government than funding through
appropriations because the private sector's cost of capital is
generally higher than the federal government's. Other factors such as
additional M&V and lower labor costs also may affect the cost of
alternative financing. Using a proxy of their cost to the government,
ECMs obtained through ESPCs we reviewed at the Navy and GSA cost
between 8 and 56 percent more than the same ECMs funded through up-
front appropriations. Also, agencies did not specifically analyze and
compare all alternatives, nor did they investigate the feasibility of
purchasing ECMs in useful segments. Given the federal government's
ability to obtain capital at lower interest rates than private
companies, officials at case study agencies agreed that funding through
timely, full, and up-front appropriations is less expensive than third-
party financing. However, with respect to partnerships, other factors
such as lower labor and fewer bureaucratic requirements may offset
higher financing costs. Therefore, it is uncertain whether using
partnerships is more or less expensive than using up-front financing.
Agencies did not adjust their capital plans to accommodate needed
capital investments nor request appropriations to finance capital
projects because they did not believe sufficient appropriations would
be available in a timely manner. Instead, they used the authorities
provided to them to finance projects over time, through third parties.
By incurring potentially higher costs in the future to avoid making
difficult trade-offs today, agencies merely defer the trade-offs to a
later date and a subsequent Congress. These trade-offs would lead
Congress to either increase appropriations to maintain the current
level of investment or fund fewer projects.
Agencies are faced with requirements for energy savings and need
appropriations to implement energy conservation measures. At the same
time, Congress is faced with allocating scarce resources for many needs
across the government. Recently, Congress expressed its current
priorities for energy saving projects by extending ESPC authority until
October 1, 2006, to permit the financing of such projects through
private companies over time. As shown in our report, this favorable
budget treatment comes at a cost--a cost that Congress needs to monitor
as these contracts are used during the next 2 years.
Implementation and monitoring of ESPCs is a relatively uniform process.
Since partnerships take a variety of forms, their implementation and
monitoring is more complex. While third-party financing can make it
easier for agencies to quickly finance projects within a given amount
of budget authority, it also presents monitoring challenges. In a
federal setting, even the appearance of a problem such as a conflict of
interest is of concern because it can erode the public's confidence in
the government and ultimately degrade an agency's ability to carry out
its mission. The use of third-party participants increases the
importance of ensuring that the government's interests are protected
and the performance of these third-party participants should be
carefully monitored and verified.
Matter for Congressional Consideration:
Given the competing pressures faced by Congress to support energy
saving investments while at the same time seeking to ensure budgetary
transparency of full program costs, Congress should consider requiring
agencies that use ESPCs to present Congress with an annual analysis
comparing the total contract cycle costs of ESPCs entered into during
the fiscal year with estimated up-front funding costs for the same
ECMs. Congress could use this information in evaluating whether to
further extend ESPC authority beyond its current expiration date.
Recommendations for Executive Action:
First, we recommend that the Director of OMB instruct agencies that use
ESPCs to report to OMB and to their committees of jurisdiction an
annual analysis comparing the total contract cycle costs of ESPCs
entered into during the fiscal year with estimated up-front funding
costs for the same ECMs. Congress could use this information in
evaluating whether to further extend ESPC authority beyond its current
expiration date.
Second, we recommend that the Director of OMB work with the
scorekeepers to develop a scorekeeping rule for the acquisition of
capital assets to ensure that the budget reflects the full commitment
of the government for partnerships, considering the substance of all
underlying agreements, when third-party financing is employed.
Finally, we recommend the Secretaries of Energy, VA, and the Navy and
the GSA Administrator perform business case analyses and ensure that
the full range of funding alternatives, including the technical
feasibility of useful segments, are analyzed when making capital
financing decisions.
Agency Comments and Our Response:
In a draft of this report, we had a recommendation that the Director of
OMB work with scorekeepers to develop a rule that would ensure that the
full commitments of ESPCs are reflected in the budget. Several agencies
did not agree with this recommendation, citing concerns that such a
rule would likely discourage or prevent agencies from entering into
ESPCs. In light of Congress' recent expression of its current
priorities by extending ESPC authority through fiscal year 2006, we
dropped this recommendation with respect to ESPCs and included instead
the first recommendation to OMB and the matter for congressional
consideration described above.
We obtained comments from OMB and our case study agencies--DOD, DOE,
VA, and GSA. OMB agreed in concept with our second recommendation that
OMB work with scorekeepers to develop a rule for partnerships that
would ensure the budget reflects the full commitment of the government,
considering the substance of all underlying agreements. DOE and VA
disagreed with this recommendation based on concerns that such a rule
would effectively make alternative financing unavailable to federal
agencies. While it is not our intent to discourage or eliminate
partnerships with the private sector, recognizing the full commitment
up-front in the budget enhances transparency and enables decision
makers to make appropriate resource allocation choices among competing
demands that all have their full costs recorded in the budget. GSA did
not address this recommendation in its comments. DOE, GSA, and VA
agreed at least in part with our final recommendation, that case study
agencies should perform business case analyses to ensure the full range
of funding alternatives are analyzed when making capital financing
decisions. DOD disagreed with this recommendation and OMB did not
address it in its comments. Business case analyses are well accepted as
a leading practice among public and private entities and OMB requires
all executive branch agencies to prepare such analyses for major
investments as part of their budget submissions to OMB. Therefore, we
believe our recommendation is appropriate.
Representatives from OMB, including staff from the Office of General
Counsel, provided oral comments on our draft. These representatives
stated that OMB "meets regularly with Congressional scorekeepers to
review the scorekeeping rules and updates A-11 guidance in order to
accurately reflect the types of transactions and obligations the
government is entering into. OMB staff stated that they would take into
consideration the findings of the report and the agencies' comments on
the report, including whether contractually guaranteed savings are
equitably considered and given due credit when evaluating ESPCs. OMB
staff also noted that recent updates in 2003 to scorekeeping guidance
related to lease-backs from public/private partnerships may address
some of the concerns GAO noted in the draft." OMB staff said that the
new scorekeeping guidance attempts to ensure that the substance of the
entire transaction is scored. We have clarified in the report that
OMB's instructions were revised in 2003 and the possible effect on how
case studies were scored.
With respect to ESPCs, OMB representatives said there is no current
plan to revisit the 1998 decision to obligate funds on an annual basis.
They said ESPCs are treated differently from partnerships in part
because of the savings component and in part because they believe doing
so would negate the statutory authority provision permitting agencies
to enter into a multiyear contract even if funds are available only to
pay for the first year of the contract. We recognize the statutory
authority enabling agencies to enter into ESPC multiyear contracts
without funds available for the full contract price but note that the
budget does not reflect full ESPC commitments as new obligations at the
time that ESPCs are signed. In light of these circumstances and
Congress' recent action to extend ESPC authority through fiscal year
2006, we removed our scorekeeping recommendation with respect to ESPCs
and are now suggesting that Congress consider requiring agencies that
use ESPCs to present Congress with an annual analysis comparing the
total contract cycle costs of ESPCs entered into during the fiscal year
with estimated up-front funding costs for the same ECMs. Congress could
use this information in evaluating whether to further extend ESPC
authority beyond its current expiration date.
DOD partially concurred with our recommendation (now deleted) about
developing a new scorekeeping rule for ESPCs. In his letter, the
Principal Assistant Deputy Under Secretary of Defense (Installations
and Environment) said DOD would concur in full if the recommendation
was modified to properly consider guaranteed savings. As we note in our
report, recognizing the full commitment up-front in the budget when the
commitment is made enables decision makers to make more informed
resource allocation choices among competing demands from equally worthy
projects that all have their full costs recorded in the budget. DOD
stated that it did not concur with our recommendation that business
case analyses should be performed when making capital financing
decisions because such an analysis would only increase administrative
costs to the department in the absence of a viable option to directly
finance energy conservation projects. Business case analyses are well
accepted as a leading practice among public and private entities and
OMB requires all executive agencies to prepare a business case analysis
for major investments as part of their budget submissions. Only by
doing a business case analysis can the government ensure that it
selects the best alternative and that taxpayers' interests are
protected. Therefore, we believe our recommendation is valid. DOD also
provided technical comments on the draft. DOD's complete comments and
our responses are contained in appendix IV.
The Acting Under Secretary for Energy, Science and Environment agreed
with our recommendation that agencies should perform business case
analyses and we commend DOE for drafting a policy that will require a
business case analysis for public/private partnerships. However, DOE
strongly disagreed with our recommendation on scoring ESPCs (now
deleted) primarily because it believes it would negate the
congressional objective of promoting energy conservation through the
use of ESPCs. It is not the intent of this report to discourage or to
eliminate energy conservation efforts. We do not believe that up-front
funding would necessarily lead to reduced support as long as energy
conservation was viewed as a priority within the appropriations
process. However, recognizing the full commitment up front in the
budget when the commitment is made enhances transparency and enables
decision makers to make more informed resource allocation choices among
competing demands that all have their full costs recorded in the
budget. We believe our recommendation that OMB require and our
suggestion that Congress consider requiring agencies to provide an
annual analysis comparing total contract cycle costs of ESPCs with
estimated up-front funding costs for the same ECMs would be an
appropriate balance between budget transparency and energy savings at
this time. DOE's complete comments and our responses are contained in
appendix V.
Because GSA does not normally engage in public/private partnerships,
its comments were confined to ESPCs. The GSA Administrator noted that
GSA's policy is to perform business case analyses, but that such
analyses do not always consider the full range of funding alternatives
for ECMs. An official from GSA's Atlanta region noted that, given GSA's
$6 billion alterations and repair backlog, other financing alternatives
may not be viable. GSA also said it has decided to revise its energy
conservation project evaluation process to include consideration of
useful segments. We commend GSA's decision to do so. Finally, GSA
pointed out that, aside from ancillary up-front costs that must be
incurred to carry out the project, ESCOs guarantee that project savings
will be met or exceeded during the contract term and that GSA enforces
these guarantees. Because these up-front payments are part of the costs
in the delivery orders we reviewed, we included them in our analysis of
total contract cycle costs--the payments ranged between 2 and 45
percent of the total contract cycle costs. For each of the three GSA
ESPCs we reviewed, total contract cycle costs exceeded contract cycle
savings. GSA's written comments and our complete response are contained
in appendix VI.
VA disagreed with our report's conclusions and recommendation to OMB.
Although our report only looked at VA partnerships, VA chose to comment
both on partnerships and ESPCs. In the Secretary's comments, he noted
that implementation of the recommendation to develop a new scorekeeping
rule would limit, discourage, and possibly eliminate the enhanced-use
lease program, thus resulting in a loss of benefits and services to
veterans. Again, it is not the intent of this report to discourage or
to eliminate energy conservation efforts or partnerships with the
private sector. However, from a budgetary standpoint, recognizing the
full commitment up front in the budget when the commitment is made
enables decision makers to make more informed resource allocation
choices among competing demands. With respect to our second
recommendation regarding the need for business case analyses, VA noted
that it had a process for this to occur for capital investments above a
threshold amount. VA's complete comments and our responses are
contained in appendix VII.
As agreed with your office, unless you release this report earlier, we
will not distribute it until 30 days from the date of the letter. At
that time, we will send copies of this report to the Ranking Minority
Member of the Senate Committee on the Budget and the Chairman and
Ranking Minority Member of the House Committee on the Budget. We will
also send copies to the Chairmen and Ranking Minority Members of the
House and Senate Appropriations Committees, House and Senate Veterans
Committees, and House and Senate Energy Committees. In addition, we are
sending copies to the Secretaries of Defense, Energy, and Veterans
Affairs as well as the Administrator of the General Services
Administration and the Director of the Office of Management and Budget.
Copies will also be made available to others upon request. In addition,
the report is available at no charge on GAO's Web site at
[Hyperlink, http://www.gao.gov].
This report was prepared under the direction of Susan J. Irving,
Director, Federal Budget Analysis, Strategic Issues, who can be reached
at (202) 512-9142 or [Hyperlink, irvings@gao.gov] and Mark L.
Goldstein, Director, Physical Infrastructure Issues, who can be
reached at (202) 512-6670, [Hyperlink, goldsteinm@gao.gov]. Questions
may also be directed to Christine Bonham, Assistant Director, Strategic
Issues, at (202) 512-9576 or [Hyperlink, bonhamc@gao.gov]. Other key
contributors to this report are listed in appendix VIII.
Sincerely yours,
Signed by:
Susan J. Irving:
Director, Federal Budget Analysis:
Strategic Issues:
Signed by:
Mark L. Goldstein:
Director, Physical Infrastructure Issues:
[End of section]
Appendixes:
Appendix I: Objectives, Scope, and Methodology:
The objectives of this study were to determine (1) what specific
attributes of energy savings performance contracts (ESPC) and public/
private partnerships (partnerships) contributed to budget scoring
decisions, (2) the costs of financing through ESPCs and partnerships
compared to the costs of financing via timely, full, and up-front
appropriations, and (3) how ESPCs and partnerships are implemented and
monitored. To obtain the detail necessary to respond to this request,
we used a case study approach. Accordingly, our findings cannot be used
to generalize across the government. We selected case study agencies
based on our August 2003 report on alternative approaches to finance
capital.[Footnote 69] In total, we analyzed 11 case studies--6 ESPCs
and 5 partnerships--across 4 agencies.
For ESPCs, we selected case studies at the General Services
Administration (GSA) and the Department of the Navy (Navy). We chose
these two agencies because they had awarded the largest dollar volume
of delivery orders under the Department of Energy's (DOE) super ESPC
program.[Footnote 70] In addition, our discussions with DOE's Federal
Energy Management Program (FEMP) officials, who administer the Super
ESPC program, indicated that the differences in the way GSA and the
Navy administer ESPCs (decentralized versus centralized, respectively)
might provide some interesting insights. Finally, the Committee's
request specifically asked us to include a military department in our
review of ESPCs.
For partnerships, we selected case studies at the Departments of
Veterans Affairs (VA) and Energy. We chose VA because of its broad
authority to enter into enhanced use (EU) lease partnerships and the
significant number of EU leases that have been awarded. We selected one
case from DOE based on the preliminary work on the Oak Ridge National
Laboratory (ORNL) partnership done for our August 2003 report. Also, it
was our understanding that this type of transaction might be replicated
at other DOE facilities.
The initial selection of case studies within each agency was based on
(1) project costs, (2) availability/location of data, and (3) time
frame of the project. These criteria narrowed the number of available
case studies and we judgmentally selected cases from this pool. We
chose three ESPC projects each from GSA and the Navy. Also, we chose to
review one partnership case study from DOE and four from VA. Table 2
lists the case studies reviewed at each agency.
Table 2: Case Studies Included in This Review:
Agency: Navy;
ESPCs:
* Navy Region Southwest, Calif;
* Patuxent River Naval Station, Md;
* Naval Submarine Base, Bangor, Wash;
Partnerships: None.
Agency: GSA;
ESPCs:
* Gulfport Federal Courthouse, Miss;
* North Carolina Bundled Sites;
* Atlanta Bundled Sites, Ga;
Partnerships: None.
Agency: VA;
ESPCs: None;
Partnerships:
* Atlanta Regional Office Collocation, Ga;
* Vancouver Single Room Occupancy, Wash;
* Medical Campus at Mountain Home, Tenn;
* North Chicago, Energy Center, Ill.
Agency: DOE;
ESPCs: None;
Partnerships:
* Oak Ridge National Laboratory, Tenn.
Source: GAO.
[End of table]
Case studies were selected based on their cost and data availability.
Data for the selected cases were in Washington, D.C., for VA cases;
Atlanta, Georgia, for GSA cases; Port Hueneme, California, for Navy
cases; and Oak Ridge, Tennessee, for the DOE case study. We selected
ESPCs for which contracts/delivery orders were awarded no later than
fiscal year 2001 so that the respective agencies would have had
opportunities to analyze whether cost savings realized to date
approximate expected savings.
To understand the features of the selected ESPCs and partnerships, we
reviewed laws authorizing the agencies to enter into ESPCs and
partnerships, relevant GAO products, and ESPC files and partnership
agreements. We interviewed officials and/or staff from the Office of
Management and Budget (OMB) and agencies involved in the development of
the selected ESPCs and partnerships. We interviewed officials from FEMP
in order to understand the general features of ESPCs. In addition, we
met with officers of UT-Battelle, LLC--ORNL's management and operations
(M&O) contractor--to gain a better appreciation of the DOE partnership.
To gain an understanding of how the selected ESPCs and partnerships
were scored and the reasoning behind the scoring, we reviewed relevant
portions of OMB's Circular A-11, analyzed the terms and conditions of
the selected case studies relative to the budget scoring rules,
reviewed relevant Congressional Budget Office (CBO) scoring reports,
and met with agency, CBO, and OMB officials and staff. All of the
partnership case studies we reviewed were executed before OMB's 2003
changes to its instructions on the budgetary treatment of lease-
purchases and leases of capital assets. According to OMB staff, some
of these partnerships may have been scored differently under the
revised instructions.
To analyze ESPC costs, we reviewed the final delivery orders of each of
our six ESPC case studies.[Footnote 71] Cash flow schedules associated
with these delivery orders specified the case studies' expected savings
and costs, such as principal payments, interest payments, measurement
and verification (M&V) fees, operations and maintenance (O&M) fees, and
energy service company (ESCO) mark-ups. Using these documents, we
identified costs, on a present value (PV) basis (using A-94 guidance),
that agencies would not necessarily have incurred had they financed the
asset acquisition through timely, full, and up-front appropriations
instead of ESPCs. Although we had planned to perform a similar cost
analysis of our partnership case studies, we were unable to do so
because comparable data were not available.
To allow us to describe how ESPCs and partnerships are implemented and
monitored, we met with OMB, GSA, VA, Department of Defense (DOD), and
FEMP officials and staff. We also spoke with representatives of certain
ESCOs to help us understand how agencies negotiate and monitor ESPC
contracts. Finally, we reviewed agencies' documentation of how ESPC
baselines were estimated and how actual savings would be determined.
Written comments from DOD, DOE, GSA, and VA are included and addressed
in appendixes IV through VII. OMB provided oral comments. We have
incorporated changes as a result of these comments throughout, as
appropriate.
Our work was done in accordance with generally accepted government
auditing standards, from September 2003 through November 2004, in
Washington, D.C., Atlanta, Ga., Oak Ridge, Tenn., and Port Hueneme,
Calif.
[End of section]
Appendix II: ESPC Case Studies:
The Energy Policy Act of 1992 and Executive Order 13123 require federal
agencies to reduce their consumption of energy in federal buildings.
The act set a goal for the agencies of lowering their consumption per
gross square foot by fiscal year 2000 to a level 20 percent below
fiscal year 1985 baseline consumption levels.[Footnote 72] Executive
Order 13123 requires a 30 percent reduction from 1985 levels by the
year 2005 and a 35 percent reduction by 2010. ESPCs allow federal
agencies to acquire energy conservation measures (ECM) to meet these
goals and implement energy-efficiency projects without having to
request the full amount of appropriations from the federal
budget.[Footnote 73] Under an ESPC, the ESCOs assume much of the up-
front capital costs associated with the improvements. The government
then uses a portion of annual energy-related cost savings attributable
to the improvements to repay the ESCO for its investment over time,
which may be as long as 25 years. This means that, although the
government's energy use may drop immediately, its expenses are
generally not significantly reduced until after the ESPC is paid off
(see fig. 8). The ESCOs guarantee the performance of the equipment,
within certain parameters, for the term of the ESPC. Agencies
frequently acquire multiple or "bundled" ECMs through ESPCs so that
ECMs yielding more dollar savings can subsidize those yielding less
savings.
Figure 8: ESPCs Reallocate the Federal Government's Payments for Energy
and Energy-Related Operations & Maintenance Expenses (E+O&M):
[See PDF for image]
Note: The proportion of E+O&M cost savings depicted may be more or less
depending on the ECMs installed and the terms of the contract.
[End of figure]
Within DOE, the FEMP provides assistance to agencies seeking project
financing through a number of methods, such as ESPCs. According to FEMP
guidance, energy-related savings result from reduced energy use,
improved patterns of energy use, avoided renovation, and reduced
operations, maintenance, and repair costs. Thus, if agencies can avoid
scheduled renovations or maintenance of older equipment by implementing
ESPCs, they may use those avoided costs to "buy-down" the
ESPCs.[Footnote 74]
To streamline the procurement process, agencies have awarded multiple-
award, indefinite-delivery-indefinite quantity (IDIQ) contracts to a
number of ESCOs in different regions of the country. With these
multiple-award contracts in place, federal agencies can place and
implement delivery orders[Footnote 75] against the contracts in a
fraction of the time it takes to develop a stand-alone ESPC because the
competitive selection process has already been completed and key terms
of the contract, such as maximum markup ceilings,[Footnote 76] have
already been negotiated. FEMP's IDIQ contract, known as a super ESPC,
is used by many agencies. Figure 9 shows the distribution of agencies
using FEMP's super ESPC from fiscal years 1998 through 2003. The bar on
the right of the figure shows in more detail the 25 awards at other
agencies using FEMP's super ESPCs.
Figure 9: Number of FEMP Super ESPC Awards by Agency, Fiscal Years
1998-2003:
[See PDF for image]
[End of figure]
GSA generally writes delivery orders against FEMP's super ESPC, while
the Navy writes delivery orders against a variety of IDIQ contracts,
referred to as contracting vehicles. Figure 10 shows the distribution
of Navy ESPC awards by contract vehicle from fiscal years 1998 through
2003.
Figure 10: Number of Navy ESPC Awards by Contract Vehicle, Fiscal Years
1998-2003:
[See PDF for image]
[End of figure]
FEMP has issued guidelines and offered training and other support to
help agencies use its Super ESPC. A typical suite of FEMP services,
costing an estimated $30,000, includes, but is not limited to, a review
of the ESCO submittals and advice on:
* detailed energy surveys (DES)[Footnote 77] and related energy
baseline data;
* appropriateness of (M&V)[Footnote 78] plan for proposed ECMs;
* technical and economic feasibility of proposed ECMs;
* pricing and financing of ECMs and post-installation services
submitted in price schedules;
* issues to address during agency/ESCO negotiations;
* commissioning and postinstallation M&V reports and advice on project
acceptance; and:
* annual M&V reports to verify annual energy savings or issues to
resolve before resuming payments.
GSA sometimes uses FEMP's services. Because the Navy has a centralized
technical and contracting team that is familiar with ESPCs, it uses its
own staff rather than FEMP's contracting officers and facilitators to
support projects. However, the Navy has used some FEMP's services,
such as reviews of initial project proposals, free of charge, and has
purchased other FEMP support in special circumstances.
The process for selecting and implementing ESPCs varies among agencies.
GSA and the Navy generally delegate the decision of whether to finance
ECMs through full, up-front appropriations or ESPCs to regional
coordinators and installations' commanding officers, respectively.
Once the decision to use an ESPC has been made, GSA and the Navy build
support and consensus for the project inside the agency. FEMP guidance
also suggests agencies meet informally with prequalified ESCOs before
selecting an ESCO for the contract. The Navy may invite vendors to
participate in oral presentations covering a range of topics, including
their qualifications and past performance with ECMs as well as their
technical approach for projects. Based on these presentations, the
local facility makes its selection.[Footnote 79]
After the contractor has been selected, the project team schedules an
initial meeting, referred to as a "kick-off" meeting, to discuss, among
other things, the scope of an initial energy survey, payback terms and
restrictions, O&M requirements, M&V approaches, and site-specific
information. The contractor then performs a preliminary site survey,
based on a building walk-through and spot metering as well as an
analysis of various data, such as utility rate structure and energy
consumption statistics.[Footnote 80] Upon completion of the energy
audit, the contractor submits an initial proposal that includes a
summary of ECMs investigated and cost and savings estimates. Based on a
review of this initial proposal, the agency decides whether or not to
proceed with the ESPC. If it decides to go forward, the agency
transmits a letter confirming its intention to award the delivery order
to the ESCO (the Notice of Intent to Award) and issues a delivery order
request for proposal. It is only after the agency issues the Notice of
Intent to Award that the ESCO's expenses may be recoverable from the
government.
The contractor then performs a DES and submits a report that is the
basis for the project's contractually guaranteed savings, M&V, and O&M.
The DES is the ESCO's comprehensive audit of facilities and energy
systems at the project site. The DES augments, refines, and updates the
preliminary site survey data and provides the information needed to
update the feasibility analyses of the various ECMs under consideration
for the project. The agency's project team reviews the proposal and
submits its comments to the ESCO. Based upon these comments, the ESCO
develops a final energy project proposal. For projects with a
cancellation ceiling in excess of $10 million, the agency must notify
Congress of the project no later than 30 days before the task order
award.[Footnote 81] After the agency approves the project and
negotiates the price or determines the price to be fair and reasonable,
the project is awarded.
During the project execution phase, the contractor completes the
project design and then installs the ECMs. Title of the equipment and
systems built or installed under the delivery order is transferred from
the ESCO to the agency at the time of delivery order award,
installation, or contract closeout.[Footnote 82] ESPCs do not involve
change orders[Footnote 83] since contractors guarantee certain levels
of performance and are obligated to make changes necessary to achieve
those levels at their own expense. However, the agency and the
contractor may modify the delivery order to, for example, authorize the
installation of additional ECMs. Once the project has been successfully
completed and accepted by the government, payments begin. Contractors
are required to guarantee equipment performance; therefore, some level
of M&V[Footnote 84] is required to ensure guaranteed performance is
realized. M&V is performed by the ESCO that installs the equipment. The
level of M&V is negotiated between the government and the contractor
and must be specified in the signed delivery order. Agency officials
said that the type of M&V employed depends on the interaction of
various factors, including climate, the people using the facilities,
the facilities' mission, and the operation of the equipment. The
contractor may also perform the O&M for the ESPC-installed equipment.
Facilities with in-house expertise may take on these responsibilities
themselves. As part of the DES, the ESCO and agency negotiate a
responsibility matrix specifying the O&M duties to be performed by each
party. FEMP guidance calls for negotiating a responsibility matrix
across a comprehensive set of issues, O&M duties being only one.
During the term of the contract, an ECM's energy cost savings are used
to pay the ESCO. If annual savings resulting from the ESPC exceed
annual contractor payments, agencies other than the DOD and GSA may
retain 50 percent of this excess; the other 50 percent must be returned
to Treasury.[Footnote 85] GSA may deposit all of the excess savings in
the Federal Buildings Fund. As of fiscal year 2004, the Navy may retain
100 percent of its excess funds, a change from the previous requirement
to return one third of the excess savings to Treasury while retaining
the other two thirds.[Footnote 86] The Navy is required to use one half
of the retained savings for energy reduction projects or water
conservation activities and the other half for Navy welfare, morale,
and recreation projects, among other things. An official responsible
for the Navy's Shore Energy program said that, prior to 2004, the Navy
had not returned funds to Treasury since, to their knowledge, there
were no cases in which actual ESPC savings exceeded those guaranteed by
the contractor. However, according to other Navy officials, the Navy's
Comptroller has not issued guidance on the return of savings to
Treasury. Because the Navy does not maintain a central system to track
savings, it would be difficult to determine whether the actual savings
generated by an ESPC have ever exceeded guaranteed savings. In other
words, the Navy does not know if it should have paid some portion of
its energy savings to Treasury.
Implementing M&V strategies is required for ESPCs to verify the
achievement of guaranteed energy cost savings each year. According to
FEMP, M&V involves three major steps: baseline definition,
postinstallation verification, and regular-interval verification.
Annual M&V only needs to show that the overall savings guarantee has
been met, not determine actual savings for each ECM.
M&V methodologies are grouped into four categories and may vary
depending on the ECM installed. When choosing among M&V methodologies,
agencies must balance the accuracy of their energy savings estimates
with the costs of verifying those estimates. For example, where the
performance of the installed equipment is relatively certain, as is the
case for lighting retrofits, it may not be cost effective to measure
actual energy use throughout the term of the contract. In this case,
postinstallation and baseline energy use is estimated using engineering
calculations or system models. As long as the potential to perform is
verified, the savings are as originally claimed and do not vary over
the contract term. Alternatively, for projects with large elements of
uncertainty, such as chillers and chiller plants, contractors might
continually measure the energy use of equipment throughout the
contract. Continuous monitoring may greatly reduce uncertainty that
savings are actually being achieved, but will also cost more than less
rigorous methods of M&V.
M&V strategies allocate risk between the ESCO and the agency in
advance. Both ESCOs and agencies are reluctant to assume responsibility
for factors they cannot control. For example, the ESCO generally does
not assume responsibility for risk related to operational factors, such
as weather, how many hours the equipment is used, and maintenance
practices. Alternatively, the agency typically does not assume the
responsibility for risk associated with equipment performance since the
ESCO selects, designs, and installs the equipment.
If the actual annual savings are less than the annual guaranteed
savings amount, the ESCO must correct or resolve the situation or
negotiate a change in the contract. For two ESPCs, an annual M&V report
identified performance problems with installed ECMs. In these cases,
the ESCOs resolved the performance problems by either replacing or
installing new equipment. Even without the ESPC's guarantee, the
manufacturer's warranty would have indemnified the government in at
least one of these cases. However, according to Navy officials, without
the M&V process of ESPC, the equipment deficiencies might have gone
unnoticed during the equipment warranty period.
The rest of this appendix contains summaries of our six case studies
from GSA and the Navy. Following is a list of these ESPCs.
Navy:
* Navy Region Southwest, California:
* Patuxent River Naval Station, Maryland:
* Naval Submarine Base Bangor, Washington:
GSA:
* Gulfport Federal Courthouse, Mississippi:
* North Carolina bundled sites:
* Atlanta bundled sites, Georgia:
For each case study, we describe the contract terms and status, the
ECMs acquired, ancillary benefits according to the contractor, and any
implementation/monitoring issues identified in M&V reports. We also
include a table showing our cost analysis of the ESPC.
Navy Region Southwest, California:
The primary contractor for the ESPC for the Navy Region Southwest,
California, was NORESCO, ERI Services Division. The delivery order for
the contract was awarded on September 26, 2001, and specified that
title to all equipment installed by the contractor would be transferred
to the government upon project acceptance. As of December 18, 2003, all
ECMs had been physically installed and accepted by the government and
were operating and yielding savings. NORESCO and the Navy share
responsibility for the proper O&M of ECMs installed under this delivery
order.
According to the final delivery order, the Navy Region Southwest ESPC
consists of five ECMs: [Footnote 87] microturbine, heat recovery, and
variable frequency drives; an irrigation systems upgrade; a compressed
air systems upgrade; heating, ventilating and air conditioning systems
improvements; and a solar photovoltaic system.
In addition to reductions in energy use and operating costs, NORESCO
claimed the ESPC provided the following ancillary benefits:
* A world-class solar photovoltaic system that was, at one time, the
largest photovoltaic installation in the United States and the largest
covered parking solar photovoltaic system in the world. This covered
parking also allows sailors to leave their cars in a protected
environment while at sea (see fig. 11).
* A demonstration project for the Navy to determine the environmental
benefits of microturbine technology.
Figure 11: Covered Parking Photovoltaic System at Navy Region
Southwest, California:
[See PDF for image]
[End of figure]
The term of the contract is 10 years, with an interest rate of 9.32
percent. As shown in table 3, the PV cost of the ECMs financed through
an ESPC is approximately $14.7 million, approximately $1 million more
than the estimated cost of the ECMs financed through timely, full, and
up-front appropriations. The PV guaranteed cost savings specified in
the delivery order was about $6.8 million less than the PV of the
ESPC's total contract cycle costs, including O&M payments. However,
according to Navy officials, over the life of the equipment, the
projected cost savings will exceed the projected costs. Navy Region
Southwest used the $6.9 million in special energy project funds it
received from the Office of the Secretary of Defense (OSD) and the Navy
to buy down the project's principal balance upon acceptance. The OSD
funds were appropriated to help offset the cost of energy projects in
California, in an effort to address the energy supply shortages in the
state.
The ECMs under this delivery order were installed and accepted over a
number of months, with final acceptance of all ECMs in December 2003.
An initial M&V report issued in March 2004 provided a baseline to
ensure that all installed ECMs were performing as guaranteed. The
initial verification process did not reveal any major maintenance or
operational issues that would negatively affect performance. Verified
savings through the end of the first year amounted to roughly $1.4
million, which exceeded the savings guaranteed in the delivery
order.[Footnote 88]
Table 3: Cost Analysis of Navy Region Southwest ESPC:
Dollars in millions, present value.
Installation cost;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $13.66;
Cost of ECMs financed through ESPCs[B]: $11.92[C].
Interest payments @ 9.32%;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $2.54.
M&V payments;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $0.23.
Total;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $13.66;
Cost of ECMs financed through ESPCs[B]: $14.69;
Difference: $1.03.
Source: GAO analysis of ESPC delivery order files.
[A] This column represents the installation and construction price for
the ECMs. It does not include O&M expenses, since these costs typically
are appropriated annually. The price could be viewed as a proxy for the
amount Congress would have had to appropriate had the ECMs been
financed through timely, full, and up-front appropriations rather than
an ESPC. However, because it is difficult to predict what the true cost
of the asset would have been had it been financed differently, this is
not a precise measure.
[B] This column represents the installation and construction price for
the ECMs under an ESPC. In addition, it includes the interest and M&V
costs that must be paid under an ESPC. It does not include O&M
expenses, since these costs typically are appropriated annually.
[C] The present value of the installation cost is lower for an ESPC
than an ECM funded through full up-front appropriations because the
ESPC payments are spread over time, thus resulting in a lower present
value. These lower installation costs are more than offset by the
higher interest payments incurred by the government under the ESPC.
[End of table]
Patuxent River Naval Air Station, Maryland:
The prime contractor for the ESPC for Patuxent River Naval Air Station
was Energy Assets; the prime subcontractor was Co Energy Group. The
delivery order for the contract was awarded on September 28, 2000, and
specified that title to all equipment installed by the contractor would
be transferred to the government upon project acceptance. As of April
10, 2002, all ECMs had been physically installed, were operating, and
were yielding energy savings.
The Patuxent River ESPC consists of three ECMs: ground source heat pump
(GSHP) installation at nine buildings; process cooling water system
modification at one building; and lighting efficiency improvements at
seven buildings.
In addition to energy savings and capital improvements, Energy Assets
claimed the ESPC provided ancillary benefits:
* There was enhanced personnel safety and landscape aesthetics in the
station's Logistic Industrial Complex. Prior to the ESPC, steam
distribution piping leaks created a safety hazard and were an eyesore.
* There was environmental compliance for one building at the station.
Prior to the ESPC, a cooling water system at the building was
configured to discharge chlorinated water into the Chesapeake Bay.
The term of the contract is 20 years, with an interest rate of 9
percent. As shown in table 4, the PV cost of the ECMs financed through
an ESPC is approximately $5.8 million, approximately $1.4 more than the
estimated cost of the ECMs financed through timely, full, and up-front
appropriations. The PV guaranteed cost savings specified in the
delivery order was about $1.9 million less than the PV of the ESPC's
total contract cycle costs, including O&M payments. However, according
to Navy officials, over the life of the equipment, the projected cost
savings will exceed the projected costs. To reduce financing costs, the
Navy made a $2.3 million down payment on the project after awarding the
delivery order. According to Navy officials this down payment was
equivalent to the sum of one-time avoided costs resulting from the
project, such as avoiding environment-related upgrades to existing
cooling water systems.
According to the postimplementation and first annual monitoring and
verification reports issued by the contractor in August of 2002 and
December of 2003, respectively, all ECMs were operational and
performing as expected. Although energy savings goals were met in the
first year, one ECM did not perform according to the performance
requirements of the contract. During the summer of 2002, it was
reported that a process cooling water system installed under the ESPC
did not meet the demands of Navy laboratory test equipment. The Navy
and the contractor agreed that the resolution of the problem was the
responsibility of the contractor and that the contractor would not, at
any time, bill the government for costs incurred to resolve the
problem.
Table 4: Cost Analysis of Patuxent River Naval Air Station ESPC:
Dollars in millions, present value.
Installation cost;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $4.33;
Cost of ECMs financed through ESPCs[B]: $3.11[C].
Interest payments @ 9.00%;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $2.44.
M&V payments;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $0.22.
Total;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $4.33;
Cost of ECMs financed through ESPCs[B]: $5.77.
Difference: $1.44.
Source: GAO analysis of ESPC delivery order files.
[A] This column represents the installation and construction price for
the ECMs. It does not include O&M expenses, since these costs typically
are appropriated annually. The price could be viewed as a proxy for the
amount Congress would have had to appropriate had the ECMs been
financed through timely, full, and up-front appropriations rather than
an ESPC. However, because it is difficult to predict what the true cost
of the asset would have been had it been financed differently, this is
not a precise measure.
[B] This column represents the installation and construction price for
the ECMs under an ESPC. In addition, it includes the interest and M&V
costs that must be paid under an ESPC. It does not include O&M
expenses, since these costs typically are appropriated annually.
[C] The present value of the installation cost is lower for an ESPC
than an ECM funded through full up-front appropriations because the
ESPC payments are spread over time, thus resulting in a lower present
value. These lower installation costs are more than offset by the
higher interest payments incurred by the government under the ESPC.
[End of table]
Naval Submarine Base, Bangor, Washington:
The prime contractor for the ESPC at the Bangor Submarine Base in
Washington was Johnson Controls. The delivery order for the contract
was issued on September 27, 2001. As of March 1, 2003, all ECMs had
been installed, were operating, and were yielding savings. Upon
government acceptance of the completed project, title to the equipment
installed under the ESPC was transferred to the government.
Six ECMs were installed at the Bangor Submarine Base, including chiller
plant modifications, air handling unit modifications, chilled water
supply and pumping modifications, and lighting modifications for
various buildings.
The term of the contract is 9 years, with an interest rate of 7.44
percent. As shown in table 5, the PV cost of the ECMs financed through
an ESPC is approximately $5.34 million, approximately $1 million more
than the estimated cost of the ECMs financed through timely, full, and
up-front appropriations. The PV guaranteed cost savings specified in
the delivery order was about $1.3 million less than the PV of the
ESPC's total contract cycle costs, including O&M payments. However,
according to a Navy official, over the life of the equipment, the
projected cost savings will exceed the projected costs. The Navy was
authorized by the Office of the Secretary of Defense (OSD) to use a
fiscal year 2001 supplemental appropriation for the western power grid
crisis to make an up-front payment of roughly $1 million to reduce
ESPC-related financed costs. The government also made a roughly
$214,000 up-front payment using savings generated during the
construction period.
According to the postimplementation report issued by the ESCO, the
projected cost savings for the first year of the project were roughly
$752,000, exceeding the ESCO's savings guarantee of about
$634,000.[Footnote 89] These projected cost savings include about
$734,000 from reduced energy consumption and about $17,000 from avoided
O&M costs.
Table 5: Cost Analysis of Naval Submarine Base, Bangor ESPC:
Dollars in millions, present value.
Installation cost;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $4.33;
Cost of ECMs financed through ESPCs[B]: $3.53[C].
Interest payments @ 7.44%;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $1.40.
M&V payments;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $0.41.
Total;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $4.33;
Cost of ECMs financed through ESPCs[B]: $5.34;
Difference: $1.01.
Source: GAO analysis of ESPC delivery order files.
[A] This column represents the installation and construction price for
the ECMs. It does not include O&M expenses, since these costs typically
are appropriated annually. The price could be viewed as a proxy for the
amount Congress would have had to appropriate had the ECMs been
financed through timely, full, and up-front appropriations rather than
an ESPC. However, because it is difficult to predict what the true cost
of the asset would have been had it been financed differently, this is
not a precise measure.
[B] This column represents the installation and construction price for
the ECMs under an ESPC. In addition, it includes the interest and M&V
costs that must be paid under an ESPC. It does not include O&M
expenses, since these costs typically are appropriated annually.
[C] The present value of the installation cost is lower for an ESPC
than an ECM funded through full up-front appropriations because the
ESPC payments are spread over time, thus resulting in a lower present
value. These lower installation costs are more than offset by the
higher interest payments incurred by the government under the ESPC.
[End of table]
GSA Gulfport Federal Courthouse, Mississippi:
The prime contractor for the ESPC at GSA's Federal Courthouse in
Gulfport, Mississippi, was Sempra Energy Services. Unlike our other
ESPC case studies, the Gulfport ESPC did not retrofit existing systems
but installed ECMs in new construction. GSA's Office of General Counsel
determined that using ESPCs to finance ECMs was appropriate for the
costs of improvements over the "baseline" design (e.g., the difference
in cost between a standard chiller and a highly efficient chiller)
rather than the entire cost of the improved system. The delivery order
for the contract was awarded on September 28, 2001. As of September 19,
2003, all ECMs had been physically installed and were operating and had
the potential to deliver the guaranteed annual savings as reflected in
the ESPC delivery order.
According to the final delivery order, the ESPC for the Gulfport
Federal Courthouse consists of 14 ECMs, including variable frequency
drives for chilled water pumps and hot water pumps, lighting controls,
energy efficient chillers, and occupancy controlled ventilation.
The term of the contract is 17 years, with an interest rate of 8.4
percent. As shown in table 6, the PV cost of the ECMs financed through
an ESPC is approximately $2.5 million, approximately $0.9 million more
than the estimated cost of the ECMs financed through timely, full, and
up-front appropriations. The PV guaranteed cost savings specified in
the delivery order was about $90,000 less than the PV of the ESPC's
total contract cycle costs, including O&M payments. However, according
to GSA officials, over the life of the equipment, the projected cost
savings will exceed the projected costs. The government also made an
$88,000 up-front payment to cover certain O&M expenses.
Since GSA installed ECMs in new construction, there was no historical
baseline to which the performance of the Gulfport Federal Courthouse
equipment could be compared. Thus, GSA hired a consulting firm to model
a conceptual building and, from that model, determined baseline energy
consumption for the new building with less efficient equipment. The
first annual M&V report for the project was issued in December 2004,
after we had completed our analysis.
Table 6: Cost Analysis of Gulfport Federal Courthouse ESPC:
Dollars in millions, present value.
Installation cost;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $1.60;
Cost of ECMs financed through ESPCs[B]: $0.64[C].
Interest payments @ 8.4%;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $1.78.
M&V payments;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $0.08.
Total; Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $1.60;
Cost of ECMs financed through ESPCs[B]: $2.50;
Difference: $0.90.
Source: GAO analysis of ESPC delivery order files.
[A] This column represents the installation and construction price for
the ECMs. It does not include O&M expenses, since these costs typically
are appropriated annually. The price could be viewed as a proxy for the
amount Congress would have had to appropriate had the ECMs been
financed through timely, full, and up-front appropriations rather than
an ESPC. However, because it is difficult to predict what the true cost
of the asset would have been had it been financed differently, this is
not a precise measure.
[B] This column represents the installation and construction price for
the ECMs under an ESPC. In addition, it includes the interest and M&V
costs that must be paid under an ESPC. It does not include O&M
expenses, since these costs typically are appropriated annually.
[C] The present value of the installation cost is lower for an ESPC
than an ECM funded through full up-front appropriations because the
ESPC payments are spread over time, thus resulting in a lower present
value. These lower installation costs are more than offset by the
higher interest payments incurred by the government under the ESPC.
[End of table]
GSA North Carolina Bundled Sites:
GSA awarded the ESPC for multiple GSA-owned buildings in North
Carolina, the Greensboro IRS Building, the Greensboro Federal
Courthouse, the Raleigh Federal Building and Courthouse, the Winston-
Salem Federal Building and Courthouse, and the Wilmington Federal
Building to DukeSolutions (now AmerescoSolutions). The delivery order
for the contract was awarded on September 27, 2000. As of November
2001, GSA found the work performed under the ESPC to be sufficiently
complete.
According to the September 20, 2000, contract, the ESPC involved
multifaceted ECMs for multiple federal buildings within North Carolina.
The 10 ECMs include lighting retrofits, the installation of energy
management systems, replacement of motors for mechanical equipment and
air flow fans, and replacement of chillers.
The term of the contract is 19 years, with an interest rate of 8.59
percent. As shown in table 7, the PV cost of the ECMs financed through
an ESPC is approximately $1.93 million, approximately $0.54 million
more than the estimated cost of the ECMs financed through timely, full,
and up-front appropriations. The PV guaranteed cost savings specified
in the delivery order was about $1.1 million less than the PV of the
ESPC's total contract cycle costs, including O&M payments. However,
according to GSA officials, over the life of the equipment, the
projected cost savings will exceed the projected costs. As part of the
$3.1 million contract cycle costs, GSA made a $1.2 million up-front
payment. These funds had already been appropriated for energy
efficiency improvements to GSA buildings involved in the project.
However, the improvements became unnecessary after GSA accepted the
ESPC.
According to the M&V report issued in April 2003, all ECMs were
operating and would continue to operate as intended.
Table 7: Cost Analysis of North Carolina Bundled Sites' ESPC:
Dollars in millions, present value.
Installation cost;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $1.39;
Cost of ECMs financed through ESPCs[B]: $0.57[C].
Interest payments @ 8.59%;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $1.25.
M&V payments;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $0.11.
Total; Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $1.39;
Cost of ECMs financed through ESPCs[B]: $1.93;
Difference: $0.54.
Source: GAO analysis of ESPC delivery order files.
[A] This column represents the installation and construction price for
the ECMs. It does not include O&M expenses, since these costs typically
are appropriated annually. The price could be viewed as a proxy for the
amount Congress would have had to appropriate had the ECMs been
financed through timely, full, and up-front appropriations rather than
an ESPC. However, because it is difficult to predict what the true cost
of the asset would have been had it been financed differently, this is
not a precise measure.
[B] This column represents the installation and construction price for
the ECMs under an ESPC. In addition, it includes the interest and M&V
costs that must be paid under an ESPC. It does not include O&M
expenses, since these costs typically are appropriated annually.
[C] The present value of the installation cost is lower for an ESPC
than an ECM funded through full up-front appropriations because the
ESPC payments are spread over time, thus resulting in a lower present
value. These lower installation costs are more than offset by the
higher interest payments incurred by the government under the ESPC.
[End of table]
GSA Atlanta Bundled Sites, Georgia:
GSA awarded an ESPC for multiple sites in Atlanta, Georgia, including
the Richard B. Russell, Peachtree Summit, and Court of Appeals
buildings, to NORESCO (formerly ERI Services, Inc.). The delivery order
for the GSA Atlanta "bundled sites" was awarded on September 30, 1999,
and as of May 31, 2000, all ECMs had been installed, were operating,
and were yielding energy savings. Upon acceptance of the project, GSA
took title to all equipment, and NORESCO will be responsible for
maintenance and repair services for all ECMs.
The Atlanta bundled sites ESPC consists of five ECMs: energy efficient
lighting upgrades, variable frequency drives, two chiller plant
upgrades, and outside air reduction. The term of the contract is 20
years,with an interest rate of 8.50 percent. As shown in table 8, the
PV cost of the ECMs financed through an ESPC is approximately $7.8
million, approximately $1.6 million more than the estimated cost of the
ECMs financed through timely, full, and up-front appropriations. The PV
guaranteed cost savings specified in the delivery order was about
$500,000 less than the PV of the ESPC's total contract cycle costs,
including O&M payments. However, according to GSA officials, over the
life of the equipment, the projected cost savings will exceed the
projected costs. A one-time energy-related operations and maintenance
payment in the amount of $900,000 was paid upon completion and
acceptance of all ECMs. The $900,000 represented funds that had already
been appropriated to replace a chiller for one of the buildings
involved in the project.
NORESCO submitted annual verification reports for the first 3 years of
the project. Based on postinstallation and annual M&V activities,
project savings were verified and exceeded the guaranteed actual
savings by approximately 5 percent in years 1 and 2, and 1.6 percent
for year 3 of the contract.
Three of the ECMs installed under the ESPC experienced problems at one
point during the first 3 years of the contract. According to the M&V
reports, the contractor worked with GSA Atlanta to resolve two of these
problems. However, absent the ESPC's guarantee, it may be that the
manufacturer's warranty would have covered the problems. The contractor
determined that the third problem was part of a larger issue and,
consequently, was outside of the scope of the ESPC. According to a GSA
official, GSA does not always receive the savings guaranteed by the
ESCO. When actual savings fall below the guaranteed level, the ESCOs
might claim that they are not at fault and, consequently, are not
financially responsible.
Table 8: Cost Analysis of Atlanta Bundled Sites' ESPC:
Dollars in millions, present value.
Installation cost;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $6.15;
Cost of ECMs financed through ESPCs[B]: $2.46[C].
Interest payments @ 8.5%;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $5.20.
M&V payments;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: N/A;
Cost of ECMs financed through ESPCs[B]: $0.12.
Total;
Cost of ECMs financed through timely, full, and up-front
appropriations[A]: $6.15;
Cost of ECMs financed through ESPCs[B]: $7.78;
Difference: $1.63.
Source: GAO analysis of ESPC delivery order files.
[A] This column represents the installation and construction price for
the ECMs. It does not include O&M expenses, since these costs typically
are appropriated annually. The price could be viewed as a proxy for the
amount Congress would have had to appropriate had the ECMs been
financed through timely, full, and up-front appropriations rather than
an ESPC. However, because it is difficult to predict what the true cost
of the asset would have been had it been financed differently, this is
not a precise measure.
[B] This column represents the installation and construction price for
the ECMs under an ESPC. In addition, it includes the interest and M&V
costs that must be paid under an ESPC. It does not include O&M
expenses, since these costs typically are appropriated annually.
[C] The present value of the installation cost is lower for an ESPC
than an ECM funded through full up-front appropriations because the
ESPC payments are spread over time, thus resulting in a lower present
value. These lower installation costs are more than offset by the
higher interest payments incurred by the government under the ESPC.
[End of table]
[End of section]
Appendix III: Public/Private Partnership Case Studies:
Unlike ESPCs, which are fairly uniform in their structure, the term
partnership can be used to describe many different types of
arrangements since partnerships may take a variety of forms. For the
purposes of this report, these arrangements typically involve a
government agency contracting with a third party to renovate,
construct, operate, maintain, or manage a facility or system, in part
or in whole, which provides a public service.[Footnote 90] Under these
arrangements the agency may or may not retain ownership of the public
facility or system, but the private party generally invests its own
capital to design and develop the properties.
Congress has already enacted legislation that provides specific
agencies[Footnote 91] with statutory authority to enter into
partnerships. Four of the five partnerships we reviewed were done under
a specific law[Footnote 92] that enabled VA to enter into a type of
partnership known as enhanced use (EU) leases. An EU lease is an asset
management tool used by VA that includes a variety of different leasing
arrangements (i.e., lease/develop/operate, build/develop/operate). EU
leases enable VA to outlease VA-controlled property to the private
sector or other public entities to be improved for either VA's use or
non-VA uses. In return, VA receives fair consideration (monetary or in-
kind) that enhances its mission or programs.[Footnote 93] Agencies
without specific partnership authority, such as DOE, have used other
authorities as the basis for partnerships.[Footnote 94]
Potential benefits of partnerships include:
* attainment of efficient and repaired federal space,
* reduction of costs incurred from using functionally inefficient
buildings,
* development of underutilized federal real property,
* in-kind benefits, and:
* access to private sector expertise.
Critics of partnerships caution, however, that these ventures are not
the least expensive means of meeting capital needs, although in the
short term they may appear to be. For example, OMB staff have indicated
that where there is a long-term need for property by the federal
government, it is doubtful that a partnership would be a more
economical means of financing than directly appropriating funds for
renovation.[Footnote 95]
Partnerships must conform to budget scorekeeping rules and OMB
instructions published in OMB Circular A-11. According to A-11,
partnerships should not be used solely or primarily as a vehicle for
obtaining private financing for federal construction or renovation
projects. Ultimately, partnerships need to be evaluated on a case-by-
case basis to determine the best economic value for the government.
In some instances, case study agencies used partnerships to acquire
capital assets without having to obtain congressional appropriations
for the full costs up front. For example, one agency used existing
authority specifically to work around OMB budget scorekeeping
instructions, allowing the agency to obligate annual lease payments
rather than the full cost of the project it would have otherwise needed
to obligate up front.
This appendix contains summaries of our five case studies from two
agencies: DOE and VA.
DOE:
Oak Ridge National Laboratory, Tennessee:
VA:
Atlanta Regional Office Collocation, Georgia:
Medical Campus at Mountain Home, Tennessee:
Vancouver Single Room Occupancy, Washington:
North Chicago Energy Center, Illinois:
For each case study, we describe the background and structure of the
partnership, its risks, costs and benefits, and budget scoring issues.
We were unable to compare the costs agencies would have incurred had
they financed the assets through timely, full, and up-front
appropriations instead of partnerships. Although the government incurs
higher interest costs compared to up-front financing, we were unable to
evaluate claims that other factors, such as lower labor costs and fewer
bureaucratic requirements available to private partners, may have
reduced costs. Thus, we were unable to judge whether partnerships could
be less expensive overall.
DOE Oak Ridge National Laboratory, Tennessee:
DOE used existing law to structure a partnership that enabled it to
obtain the use of facilities for up to 25 years without recording large
up-front obligations and outlays. DOE's contractor, UT-Battelle, LLC,
obtained about $70 million in private financing for new office and
research facilities at Oak Ridge National Laboratory (ORNL). ORNL is
DOE's largest science and energy laboratory. ORNL was established in
1943 as part of the Manhattan Project to pioneer a method for producing
and separating plutonium. Today, ORNL leads the development of new
energy sources, technologies, and materials and the advancement of
knowledge in the biological, chemical, computational, engineering,
environmental, physical, and social sciences. Since April 2000 ORNL has
been managed and operated by a private, limited liability partnership
between the University of Tennessee and Battelle Memorial Institute,
UT-Battelle, LLC. As ORNL's management and operations (M&O) contractor,
UT-Battelle, LLC's primary client is DOE. The M&O contract with DOE is
a 5-year performance-based contract with an option for DOE to renew for
an additional 5-year term.
Shortly upon winning the M&O contract, UT-Battelle, LLC, submitted a
Strategic Facilities Revitalization Plan for construction of a total of
11 new facilities and renovation of existing facilities at ORNL during
the first 5-year phase of the program. Given the magnitude of needed
facilities improvements and the historical funding levels, this plan
proposed a partnership to secure funding for new construction and
renovation of existing space through a combination of federal, state,
and private funds--about $225 million, $26 million, and $70 million,
respectively.
One key component of this proposal was the transfer of land ownership
from DOE to a special-purpose entity to allow for construction and
lease of buildings by the private sector. Section 161(g) of the Atomic
Energy Act[Footnote 96] permitted DOE to transfer at no cost, via
quitclaim[Footnote 97] deed, 6.6 acres of land at ORNL to a special-
purpose entity, UT-Battelle Development Corporation (UTBDC), for the
construction of three buildings.[Footnote 98] UTBDC, a 501(3)(c)
nonprofit corporation, was created for the sole purpose of implementing
the privately financed elements of the revitalization project, and the
selection of UTBDC was not based on a competitive process. However,
upon receiving the land, UTBDC issued a 25-year ground lease for the
6.6-acre site to Keenan Development Associates, of Tennessee, LLC
(Keenan), a private developer competitively selected to build the new
facilities. As described below, UTBDC provided design specifications
and construction oversight and functioned essentially as a pass-through
entity for the land, financing, and leasing of the three buildings
reviewed in this case study.
Once private financing (bonds) was obtained and construction of the
three buildings was completed, Keenan implemented three prenegotiated
facility leases for these buildings with UTBDC for a term of 25
years.[Footnote 99] UTBDC, in turn, implemented subleases of the three
facilities to UT-Battelle, LLC, for DOE's ultimate use, with a lease-
term up to 25 years.[Footnote 100] Accordingly, DOE reimburses UT-
Battelle, LLC, for the sublease payments, which flow back to Keenan to
pay off the outstanding bonds. Figure 12 depicts the full partnership
arrangement used to revitalize ORNL, including the financing just
described.
Figure 12: Partnerships and Financing of ORNL's Revitalization:
[See PDF for image]
[A] The 2-acre transfer was for two of the four state-financed
facilities. A subsequent land transfer will be necessary for the
remaining two facilities.
[End of figure]
Risk:
Ultimately, the principal and interest on the bonds are covered by
DOE's sublease payments (if the subleases run for their full 25-year
term), as specifically recognized by Standard and Poor's (S&P) A+
rating of the bonds. However, DOE officials told us they neither
reviewed the private bond-offering memorandum nor asked to see it.
According to DOE's Counsel, a DOE review of the bond-offering
memorandum would not have been consistent with the fact that this was a
private transaction between Keenan and private investors. In addition,
UTBDC's Counsel said he was careful to make clear that the sublease
contained a 1-year termination clause to ensure DOE's involvement was
not misrepresented; instead UTBDC and its backers bear the risk
associated with the bond repayments. Furthermore, UTBDC officials told
us that in excess of $1 million of UTBDC private funds were expended in
support of the construction effort. Although UT-Battelle officials were
unwilling to provide us a copy of the bond-offering
memorandum,[Footnote 101] S&P's A+ bond rating report states that its
rating was based, in part, on a pledge of DOE rent payments, since DOE
was unlikely to vacate the facilities.
Costs and Benefits:
DOE officials could not provide us with any documentation showing that
the agency had performed an independent cost-benefit or business case
analysis[Footnote 102] of the private financing arrangement. The Chief
Financial Officer (CFO) of DOE's Oak Ridge Operations office asserted
that private financing and construction would be less expensive than
appropriations because the accelerated completion time[Footnote 103]
would enable them to more quickly vacate dilapidated buildings that
were expensive to operate and maintain. According to DOE officials, in
addition to dollar costs, the obsolete buildings were affecting the
recruitment and retention of top-quality scientists needed to further
DOE's mission. Although no appraisal was made of the land prior to
turning it over via quitclaim deed, which was given to UTBDC free of
charge, several DOE officials said they believed the land was without
value.[Footnote 104] Moreover, although unable to provide supporting
documentation, several officials said they believed that DOE's then CFO
would have looked into the costs and benefits. According to UT-
Battelle, LLC's Deputy Director for Operations, the former CFO received
summary analyses addressing the cost-benefit study performed by UT-
Battelle, LLC. The Deputy Director further explained that the former
CFO visited ORNL for a full day briefing and walk-around to review the
proposed project. The summary information provided by UT-Battelle, LLC,
to the former DOE CFO was, in our opinion, not sufficient for a
detailed, business case analysis. The type of underlying data needed to
perform such an analysis had been prepared by UT-Battelle, LLC, and was
readily supplied to us upon request. UT-Battelle, LLC, officials said
that DOE also would have been provided this data had it been requested.
According to UT-Battelle, LLC's estimates prior to the start of the
project, the privately-contracted construction that was used would have
cost about $45 million compared to about $101 million had DOE
contracted directly for the construction itself.[Footnote 105]
The value of the bonds issued by Keenan to construct the three
buildings totaled about $70 million. However, the actual cost to
construct, according to UTBDC's analysis, was about $54 million. Our
analysis of DOE's subleases shows that the 25-year PV that DOE will pay
to lease the three privately financed buildings will total about $96
million.
Budget Scoring:
According to DOE and UT-Battelle officials, a key concern of the
financing arrangement was ensuring that it would score as an operating
lease[Footnote 106] and thus only require the annual sublease payments
(plus cancellation costs) to be obligated and shown in the
budget.[Footnote 107] Had it not met the operating lease criteria, DOE
would have had to obligate in the first year of the sublease sufficient
budget authority to cover the PV of the government's sublease payments
over the full 25-year lease term. Given this concern, the terms of the
partnership were carefully constructed to ensure it would be scored as
an operating lease. For example, by giving the land to UTBDC, DOE
ensured that the project would not be located on government property.
Also, by establishing short-term subleases, UT-Battelle, LLC, ensured
the lease term did not exceed 75 percent of the estimated economic life
of the asset. Had the buildings been constructed on government land or
the lease term exceeded 75 percent of the economic life of the asset,
the arrangement might have been treated as a capital lease and DOE
would have had to obligate the full costs of the project up front.
VA Atlanta Regional Office Collocation, Georgia:
In 1998, VA used its EU lease authority to leverage more than $32
million in private financing for the collocation of the Atlanta VA
Regional Office (VARO) on department-owned property adjoining the VA
Medical Center in Dekalb County, Georgia. The collocation provides both
benefits and medical services on a single campus resulting in increased
convenience to veterans receiving services from VARO and the VA medical
center. Previously, the Atlanta VARO had been located in a GSA-
controlled building in midtown Atlanta. With its lease scheduled to
expire, VA considered moving the VARO into a new GSA-controlled
building in downtown Atlanta, known as the Atlanta Federal Center
(AFC). However several factors prompted VA to research other
alternatives. The move to the AFC would have (1) tripled VARO's rent to
GSA; (2) separated VARO's offices over more space than required; and
(3) according to the Georgia VA Commissioner, provided inadequate
parking access for disabled veterans.
VA ultimately collocated the VARO onto property adjoining the VA
Medical Center in Dekalb County by entering into a 35-year partnership
with the Dekalb County Development Authority (the Authority).[Footnote
108] According to the EU lease, the VARO project would increase
employment and expand economic development in Dekalb County. Under the
partnership, VA outleased six acres of VA-owned property to the
Authority for a 35-year period. In exchange for the lease, the
Authority agreed to finance, develop, own, operate, and maintain a
furnished and equipped office building and parking garage. To finance
the project, the Authority issued revenue bonds in excess of $32
million. VA then leased back the office space needed from the
Authority's developer through 2-year operating leases, which
automatically renew for up to nine consecutive terms unless VA takes
positive action to terminate the automatic renewal clause.[Footnote
109] At each renewal of the lease, VA maintains the right to reduce the
amount of office space it occupies if its requirements change.
Risk:
According to the EU lease, the construction of the VARO building and
parking was a private undertaking of the Authority and not an
undertaking of VA. Additionally, the Authority bears the risk and
responsibility to operate, maintain, repair, and replace assets in the
event of a causality loss, and holds the title to the office building
and parking garage as long as the revenue bonds are outstanding. At any
time during the ground lease, VA has the right to acquire the
improvements from the Authority for a purchase price equal to the sum
necessary to make the payment of the bonds. Upon expiration or
termination of the EU lease, title to all improvements on the land
automatically transfers to VA.
To mitigate the risks to the Authority if VA does not renew the lease
or reduces the amount of space it occupies in the VARO building, VA
deposited $1.8 million into a "renovation reserve fund" when the lease
was executed. The Authority may draw from this fund to renovate or
reconfigure rental space for new tenants should VA vacate some part of
the VARO building during the term of the bonds. VA officials said that
by agreeing to a reserve fund, VA's rent would be reduced because bonds
were sold to investors at a lower interest rate, thus reducing the
Authority's debt service and, in turn, VA's rental payments. VA also
mitigated the Authority's risk by agreeing not to replace the VARO
building with another regional administration or headquarters building
in Georgia using its EU lease authority during the term of the bonds.
According to VA, the Authority passed on additional risk mitigation
savings to the department by obtaining bond insurance that guaranteed
timely payment of principal and interest to bondholders.
Costs and Benefits:
Thirty-five-year PV life-cycle costs for the project were estimated to
total about $43 million, while the Authority bond issue totaled about
$33 million. Although VA compared the cost of the collocation to the
costs of moving into the AFC,[Footnote 110] the department decided not
to compare the costs of construction via EU lease to full, up-front
financing. According to VA officials (1) VA did not believe it would
receive up-front appropriations for VARO construction and (2) internal
VA protocol did not require economic comparisons between all possible
acquisition alternatives.
Budget Scoring:
Because the VARO collocation project was scored as an operating lease,
VA's 2-year lease payments to the Authority are the only aspects of the
arrangement reflected in the budget. VA officials said this scoring
treatment is appropriate since VA may decide to vacate part or all of
the building before the 35-year ground lease expires. The officials
explained that VA may need less office space when electronic filing
systems eventually replace existing paper-based systems. Also they
noted that a short-term lease provides flexibility in the event that
VA's field structure changes over time. However, it is not clear that
VA's need for the space is necessarily short-term. For example, prior
to the collocation of the Atlanta regional office with the medical
center, the regional office had occupied offices in the Atlanta area
for over 25 years. Scoring the EU lease as an operating lease assumes
that its need is short-term, even though VA has no current plans to
vacate the space.
VA Medical Campus at Mountain Home, Tennessee:
In 1998, VA entered into an EU lease with its local affiliate, James H.
Quillen College of Medicine of East Tennessee State University (ETSU),
and the State of Tennessee.[Footnote 111] Under the EU lease, VA
transferred to ETSU the long-term maintenance and development
responsibilities of 31 acres of land, including nine buildings, on VA's
medical campus property in Mountain Home, Tennessee, for a term of 35-
years. After the expiration of the lease, VA may transfer the fee
simple title to the property to the State of Tennessee; however, VA
made no guarantees to any such transfer. The EU lease superseded
existing leases of buildings and land where ETSU occupied the
facilities while VA was responsible for the maintenance and capital
improvements, without reimbursement from ETSU.
Risk:
Under the EU lease, the federal government retains a fee simple
ownership interest in the property. However, during the term of the
lease, VA is not responsible for damages to the property or for
injuries to persons on the property, except as provided for by
applicable law. In the event that any part of the property is damaged
or destroyed, other than as a result of VA's negligence, the state is
obligated to repair, restore, or rebuild the property.
Costs and Benefits:
VA projected that the initiative would result in a cost avoidance of
approximately $34.6 million (PV) in capital management costs over the
lease term. Also, through the term of the lease, an additional $6.3
million (PV) of "in kind consideration" medical services and possible
groundskeeping services would be provided to VA by ETSU staff and
residents each year. According to VA's capital asset management study,
these benefits were equivalent to the $40.9 million value of the
capital assets to be leased to ETSU.
According to VA officials, ETSU assumed responsibility for the
maintenance of the buildings so that it could make capital improvements
that VA was unwilling to undertake, such as the renovation of labs and
heating, ventilation, and air conditioning systems. VA's business case
states that VA would also benefit from the improved medical school
facilities since the improvements would increase funding of research,
including equipment, supplies, and technicians for VA physicians
working at the medical school, where such resources are not provided by
grants.
Budget Scoring:
Because the arrangement did not involve cash transfers, the EU lease is
not reflected in the budget.
VA Vancouver Single Room Occupancy, Washington:
In 1998, VA outleased about 1.4 acres of vacant, undeveloped
land[Footnote 112] adjacent to the Vancouver Division of the Portland
VA Medical Center. The 35-year, no-cost outlease was awarded to the
City of Vancouver's Housing Authority (Housing Authority).[Footnote
113] The Housing Authority subsequently financed, designed, and built a
126-bed single room occupancy (SRO) structure on the property in order
to provide transitional and permanent housing for single homeless
individuals of southwest Washington. The Housing Authority agreed to
give veterans referred by the Portland VA Medical Center priority
placement for at least 50 percent of the occupancy of the SRO property.
Risk:
The EU lease required the Housing Authority to bear all costs and
responsibility for developing and constructing the SRO. In addition,
the lease made the Housing Authority (1) responsible for all repair and
maintenance costs associated with the SRO and (2) subject to the risk
of loss or damage occurring on the property.
Unless VA decides to dispose of the property, the Housing Authority
must surrender the SRO and other improvements on the property to VA
upon termination or expiration of the lease. The Authority is
responsible for the development, construction, repair, and maintenance
of the SRO. Although in our opinion VA bears minimal risk, the
construction on the property represents an opportunity cost to VA.
Costs and Benefits:
According to a VA official, the property would likely have remained
unused without the EU lease arrangement. Although priority placement
enables VA to reduce costs by expediting the release of patients from
its medical center,[Footnote 114] VA stated it had not previously
considered using the land until the Housing Authority approached VA
with the SRO idea.
Budget Scoring:
Because the transaction did not involve cash consideration, it was not
reflected in the budget.
VA North Chicago Energy Center, Illinois:
In 2002, VA used its EU lease authority to initiate the development of
a cogeneration facility that could provide chilled water, electricity,
and steam to its 190-acre campus in North Chicago, Illinois. Prior to
this EU lease, VA purchased electricity from the local utility and
steam from an adjacent Navy facility. According to a study, VA
determined that VA's energy costs in North Chicago were 60 percent
higher than average and VA determined that the rates charged by the
Navy for steam were above market rates.[Footnote 115]
According to VA's analysis, VA determined that using an EU lease for
the development, construction, and O&M from a third party of an energy
center on the campus would be the most efficient and cost-effective way
to meet the energy requirements of the North Chicago VA Medical Center,
compared to federally contracted construction or purchasing energy
(steam and electric) from local sources. In 2002 VA signed a 35-year EU
lease with Cole Taylor Bank as trustee of the North Chicago Energy
Trust (Trust) to lease approximately 1 acre of land appraised at
$110,000. In return the Trust hired Energy Systems Group as the
developer to (1) develop, design, equip, construct, operate, and
maintain the Energy Center; (2) engage in sales of energy services to
third parties; (3) provide energy-related services including operating
and maintaining the VA Medical Center's systems for chilled water,
electricity, steam, and its respective distribution systems; and (4)
undertake energy savings initiatives at the VA Medical Center and other
VA facilities in the area. The Trust's trustee borrowed on behalf of
the Trust through the Illinois Development Finance Authority
(IDFA)[Footnote 116] about $37 million in bonds, secured by the
leasehold interest and its improvements. VA is the sole beneficiary of
the Trust.
Risk:
The energy service agreement between VA and the Trust sets the
standards and terms for VA to purchase steam and electricity generated
by the energy center. VA may terminate the agreement under default
provisions if the developer cannot complete the development or perform
and supply energy as specified in the agreement. According to a VA
official, VA then would in sequence (1) request that the Trust select
another firm to run the center or (2) take over operation of the
center. The bonds were issued by IDFA and the Owner Trust is
responsible for repayment of the bonds in the event of a failure to
perform or any other breach of the agreement. The EU lease holds the
developer responsible for any loss, cost, or liability of an
environmental nature that arises out of the developer's acts or
omissions in conjunction with the property. The Owner Trust is
responsible in the event there is a loss of assets, such as through
fire. If VA vacates the campus as part of its mission, VA has no
responsibility or liability for any future payment of the bonds. In
accordance with its legislative authority, VA may elect to transfer its
interest in the land that the energy center was built on to the Owner
Trust, so that the Trust may continue operations or pay off the bonds.
According to VA officials, they initiated the Trust arrangement for
this EU lease to protect the government in the event of bankruptcy or
foreclosure of a developer/operator and if the developer did not or
could not complete the project. By using a trust, the bond revenues
belong to the Trust and are paid to the developer.[Footnote 117] The
Trust maintains title to improvements during the lease term, afterwards
title transfers to VA at the end of the lease term.[Footnote 118]
Costs and Benefits:
VA contributed no funding for the development of the energy
center.[Footnote 119] According to VA officials, constructing and
running energy centers is not within VA's mission and it would not have
put forward a request to do so. VA estimated that a new energy center
would save VA about $12.7 million (net present value) over 10 years
compared to its current costs for electricity and steam. The new energy
center system runs parallel to the local utility, which gives VA a
backup source of electricity. At one point before VA entered into the
EU lease, it was paying the Navy over $3 million for steam and the
local utility over $1 million for electricity. VA accounted for 14
percent of the steam generated by the Navy facility and, according to
VA's business plan, VA could end its steam purchases without creating a
substantial loss to the Navy. The Navy would be able to reduce its
steam pressure and in turn increase its efficiency. Navy officials
confirmed that while the Navy had lost revenue, VA steam consumption
was not a significant portion of its business.
Budget Scoring:
The utility costs for the VA complex are reflected in the budget on an
annual basis but no other scoring for this project is reflected in the
budget. After this EU lease was signed, OMB has stated that under new
scoring instructions the costs associated with trusts should be scored
up-front since (1) VA maintains control of the assets acquired through
the trust and (2) VA bears the risk for these assets.
[End of section]
Appendix IV: Comments from the Department of Defense:
OFFICE OF THE UNDER SECRETARY OF DEFENSE:
ACQUISITION TECHNOLOGY AND LOGISTICS:
3000 DEFENSE PENTAGON:
WASHINGTON, DC 20301-3000:
Ms. Susan Irving:
Director, Federal Budget Analysis:
Government Accountability Office:
Washington, D.C. 20548:
OCT 25 2004:
This is the Department of Defense (DoD) response to the GAO Draft
Report, 'CAPITAL FINANCING: Partnerships and Energy Savings Performance
Contracts Raise Budgeting and Monitoring Concerns,' (GAO Code 450262),
dated 24 September 2004. Detailed comments on the report are enclosed.
The Department is concerned that the draft report reflects an
incomplete analysis and an incorrect understanding of the Energy
Savings Performance Contract (ESPC) program.
I urge your strongest consideration to revisit this report and
comprehensively address the issues we have raised prior to issuing this
report in final form. My staff is prepared and willing to assist your
team to more accurately assess the ESPC program. We would also
appreciate an additional opportunity for review prior to any final
publication.
Signed by:
Philip W. Grone:
Principal Assistant Deputy Under Secretary of Defense:
(Installations and Environment):
Enclosure.
As stated:
GAO DRAFT REPORT - DATED SEPTEMBER 24, 2004 GAO CODE 450262/GAO-05-55:
"CAPITAL FINANCING: Partnerships and Energy Savings Performance
Contracts Raise Budgeting and Monitoring Concerns"
DEPARTMENT OF DEFENSE COMMENTS TO THE RECOMMENDATIONS:
Recommendation 1. The GAO recommends that the Director of OMB work with
scorekeepers to develop a scorekeeping rule that would ensure that
funds are obligated to reflect the full commitment of the government,
considering the substance of all underlying agreements, when third
party financing is employed. (p. 38/GAO Draft Report):
DoD Response: We partially concur. We would concur in full if the
recommendation was modified to also properly consider guaranteed
savings. We believe that OMB and CBO should review their current
scorekeeping procedures and develop an approach suitable to a program
such as ESPC such that the contractually guaranteed savings are also
equitably considered and given due credit. Of significance is that no
increase in appropriations are required for ESPC contracts.
Recommendation 2. The GAO recommends that the Secretary of Defense
perform business case analyses and ensure that the full range of
funding alternatives, including the technical feasibility of useful
segments, are analyzed when making capital financing decisions. (p. 38/
GAO Draft Report):
DoD Response: We do not concur. A business case analysis suggested by
the report would only translate to an increased administrative cost to
the Department in the absence of a viable option to directly finance
energy conservation projects. To make this recommendation applicable
and useful to the decision maker, GAO should add language encouraging
the Congressional appropriation and authorization committees to support
the President's budget, line item "Energy Conservation Investment
Program" in the Military Construction (MILCON) bill. Using direct
appropriations is not a realistic, available option without imposing an
impact to some other aspect of the budget. The current recommendation
essentially promotes eliminating a program that has accounted for well
over 50% of our energy reduction against the Congressionally-mandated
goals without providing a viable, alternative approach for Congress, or
the Department to pursue. Also of note is that prior to any ESPC
contract award, a complete life cycle cost analysis is conducted,
including all cost benefit considerations, to ensure that each contract
is financially favorable and meets the short and long term needs of the
Government.
GAO DRAFT REPORT - DATED SEPTEMBER 24, 2004 GAO CODE 450262/GAO-05-55:
"CAPITAL FINANCING: Partnerships and Energy Savings Performance
Contracts Raise Budgeting and Monitoring Concerns"
DEPARTMENT OF DEFENSE TECHNICAL COMMENTS TO THE REPORT:
Comment: On the cover sheet in the second paragraph under the heading,
"What GAO Found", it is mentioned that "there is insufficient data to
measure" opportunity costs.
DoD Response: We strongly non-concur. We believe the analysis conducted
in this draft report falls well short of properly responding to the
issues of concern provided by the Federal energy management community.
The GAO audit team could conduct a few economic assessments that would
translate the projected annual savings of a given energy conservation
project into future year dollars which would represent the "value"
associated with lost opportunity costs for an otherwise viable,
desirable project delayed for a given period of time. We also question
the statistical relevance of the one project that indicated a cost
increase of 56 percent as a result of using alternate financing. If
that project was an exception and had unique circumstances, it
certainly does not present a fair representation of the true increased
cost of alternate financing, which we would argue is much closer to the
lower end of the range provided. A more probable range of cost
increase, given the contracts analyzed, is 25 to 35%, assuming direct
funding is available in a timely manner. It also appears that "risk"
was not duly considered in the cost analysis. If an energy conservation
project was funded with direct appropriations, the performance risk
falls entirely on the government. While the contractor or supplier will
guarantee the proper functioning of their product, they do not
guarantee its performance with respect to energy efficiency.
Conversely, the risk of performance in an alternatively financed
arrangement such as an ESPC resides with the contractor and requires
guaranteed energy savings. The cost associated with transference of
risk should be considered and accounted for.
Comment: On page 16, the report concludes that "ESPC commitments are
not fully recognized up-front in the budget."
DoD Response: We do not concur. The report does not properly articulate
that the "commitments" are paid for from the existing, current
appropriation levels, which are recognized up-front in the budget. The
report misleads the reader to believe that an ESPC contract has
committed Congress to funding above and beyond what is budgeted for
each year in the President's budget and provided in the subsequent
Congressional appropriation, which is not accurate. It should be
highlighted that without these ESPC contracts, the opportunity to
reduce future levels of required commitments via installation utility
bills is lost. While we agree some form of scoring, that perhaps only
considers cancellation or termination fees, is appropriate, the current
strategy for scoring is inadequate to properly account for the savings
that are generated and ultimately fund the financial obligations of
concern.
Comment: On page 20, it is asserted that, "acquiring capital through
ESPCs is more expensive than acquiring the same capital through full,
up-front appropriations in our case studies."
DoD Response: We strongly non-concur. This report falls short in its
due diligence to inform Congress of the "added cost" through the lost
opportunity for energy reductions to actually lower its future funding
levels. The final report and recommendations are a reflection of an
incomplete analysis that gives cursory mention to facts that might
support the opposite conclusion, that ESPCs are a better value to the
government. Particularly, most every "pro" while mentioned, was
provided the same common responses, either "there is insufficient data
to measure this effect" or "we did not analyze these claims."
Government officials provided valid suggestions that the audit team
consider and incorporate in their analysis including the cumulative
effect of aggregating energy conservation measures, the lost
opportunity costs of delays while sitting on an "unfunded" list for
scarce, highly competitive direct appropriations, the minimized
administration and overhead associated with contracting and
administration, decreased equipment maintenance costs, and the
decreased risk of performance to the government. Additionally, and
perhaps not considered at all by this report, ESPC contracts provide a
venue for the private sector to recommend technical solutions that the
government would not otherwise be aware of and in a position to pursue
through direct appropriations. The ESPC contract vehicle provides an
incentive for the private sector to research, identify, propose and
implement the most suitable technology, in the form of an ECM, and
guarantee that the technology's performance will achieve results in the
form of energy savings, all while minimizing the administration and
overhead (increased costs) that the government would otherwise have to
retain. Last, without ESPC authority for the entire past year, there
are numerous examples of "unfunded" or "stalled" energy conservation
projects that could be considered in a calculation of the "lost
opportunity costs" when "awaiting direct appropriations." All of these
points received cursory consideration at best, yet could have a
significant impact on the conclusions and findings. We are concerned
that this report does not present a fair and unbiased assessment of the
ESPC program.
Comment: On page 21, it is stated that, "M&V resulted in higher
sustained savings but is an expense that would not be incurred if the
ECMs were acquired through full, up-front appropriations."
DoD Response: We do not concur. Measurement and Verification has a
direct relationship with energy and cost savings and risk of
performance. The application of direct appropriations would not result
in the cost of M&V. However, in that case, the anticipated energy and
cost savings are not guaranteed, are not measured and verified, and are
less likely to result to the same magnitude, changing the payback
portfolio and extending it over a longer period, in some cases
potentially beyond the effective life of the equipment. The other
significant aspect of M&V is that it removes risk of performance from
the government and places the burden on the contractor to properly
estimate and maintain performance over the life of the contract. While
it may be an added direct cost, it is easily paid for through the
increased guaranteed savings. The report's statement that it is an
"additional cost" is misleading since it, like the entire investment,
is paid for from savings. This point, if included in the analysis,
could verify or discount mathematically the added value of M&V to ECM
performance and payback periods.
Comment: On page 24, it is stated that, "The performance of ECMs
installed through the use of ESPCs are guaranteed to reduce energy use
during the term of the contract so that payments to the contractor can
be made from the savings from lower utility bills. ESPCs contain
assumptions for such things as hours of operation and ECM efficiency
which, taken together, determine estimated savings. However, if the
assumptions are incorrect and estimated savings are not achieved, the
agency is still required by contract to pay the ESCO the agreed-upon
savings specified in the ESPC."
DoD Response: We strongly non-concur. The report has drawn an incorrect
conclusion. If there is a reason to believe that non-performance is
attributable to the government's actions, contractually the ESCO is
bound to present and defend any evidence prior to any payments being
authorized.
Comment: On page 31, there is speculation that, "a business case
analysis might have demonstrated that sufficient funds were available
to purchase the ECMs in smaller, useable segments, when technically
feasible."
DoD Response: We strongly non-concur. The GAO report should go beyond
speculation. There is no evidence that the analysis attempted to
quantify the increased administrative cost to the agency of conducting
a business case analysis for every energy conservation project they may
contemplate, nor does it demonstrate the added value a business case
analysis might provide to a decision maker confronted with limited
direct appropriations at their disposal to meet mandated energy
reduction goals.
Comment: On page 32, it is stated that buy-downs, "imply opportunities
exist to acquire ECMs in smaller, useful segments, when technically
feasible."
DoD Response: We strongly non-concur. This is an inappropriate and
incorrect conclusion stemming from an incomplete analysis. First, buy-
downs are typically only an extremely small percentage of the overall
contract and certainly not a reflection that the ECMs could be
otherwise financed in this manner. Also, the opportunities for buy-
downs are not predictable. They typically result from the unique
circumstances such as those described in this report. Most often, buy-
downs only occur when previously programmed projects can be cancelled
due to the timing of the ESPC and their funding then redirected. It
would appear that these same cases where buy-downs occurred could have
been analyzed by the GAO team from the aspect of projects that had been
deferred and had been "awaiting funding" for several years thus
presenting and quantifying the "lost opportunity" costs that this same
report concluded could not be analyzed for lack of evidence.
Comment: Page 37 states that, "in a federal setting, even the
appearance of a problem such as a conflict of interest is of concern
because it can erode the public's confidence in the government and
ultimately degrade an agency's ability to carry out its mission."
DoD Response: While we agree in concept with the statement, we do not
understand its relevance in this report. There is no evidence presented
in the analysis that suggests that a conflict of interest has existed
or currently exists. ESPC contracts are awarded and administered by
warranted contracting officers with the appropriate responsibility
entrusted to them. If the GAO team has recommendations for improving
the ESPC contractual relationship, we would welcome that input.
Otherwise, this conclusion is unfounded and irrelevant to the report.
The following are GAO's comments on the Department of Defense's letter
dated October 25, 2004.
GAO's Comments:
1. We do not agree that our recommendation needs to be modified to
further consider savings. We compared acquisition costs for a given set
of ECMs. Therefore the savings should not vary. Regardless of how they
are financed, the same given set of ECMs acquired for the same energy
reduction projects should yield the same energy savings.
2. Only by doing a business case analysis can the government ensure
that it selects the best alternative and that taxpayers' interests are
protected. Life-cycle cost analysis is only one part of a business case
analysis, which includes economic and financial analyses such as cost-
benefit and comparative alternative analyses. We recognize that using
full, up-front appropriations to fund ECMs would likely affect some
other aspect of the budget. It is not the intent of this report to
discourage or to eliminate energy conservation efforts or partnerships
with the private sector. However, recognizing the full commitment up-
front in the budget enhances transparency and enables decision makers
to make appropriate resource allocation choices among competing demands
that all have their full costs recorded in the budget. One of the
primary purposes of budgeting is to make resource allocation decisions
among competing claims that all have their full costs recorded in the
budget.
3. Measuring opportunity costs requires estimates of how long the delay
in obtaining ECMs would be and the cost of that delay. To do that, GAO
would have to make an assumption about when obtaining an ECM would be
of sufficiently high priority in comparison to the other programs for
which an agency would request full funding. Our report does not address
agencies' resource allocation decisions, but points out that the
decision to acquire ECMs through ESPCs is more expensive than through
timely, full, and up-front appropriations--a point with which agency
officials agreed. We believe that such an analysis is more
appropriately conducted by agencies as part of the business case
analysis of alternatives.
4. Given our case study approach, our report does not imply a
statistical relevance to any of our case study findings. We selected
case studies based on their cost and data availability. The case study
result that DOD questions is a delivery order awarded by GSA. Because
this ESPC involved new construction, GSA noted that the result of this
case would be unique but interesting; GSA agreed with our selection of
cases. We note DOD's comment that these ESPCs usually increase the
government's costs by 25 to 35 percent. While this is less than the 56
percent in the GSA case study, it is not an insignificant cost
differential.
5. Our report acknowledges that M&V of savings acts as a type of
insurance and that M&V strategies allocate risk between the agency and
the ESCO. M&V is an explicit cost in ESPCs. However, agencies could
choose to purchase M&V for ECMs financed through full, up-front
appropriations if, after conducting a business case analysis, they
believed it was in the best interest of the government. With respect to
the savings guarantee, as discussed on p. 29, ESPCs contain assumptions
that determine estimated savings. If the assumptions are not correct
and savings are not achieved, the agency is still required to pay the
ESCO the agreed-upon savings specified in the contract. Finally, we
clarified that the M&V comparison is to estimated savings.
6. We agree that appropriations are recognized in the budget. However,
ESPC commitments are not in fact fully recognized up front in the
budget. OMB has scored the acquisition costs of assets acquired through
ESPCs annually, over time, even though ESPCs represent long-term
commitments of the government. For example, agencies generally retain
control of the assets acquired for the entire life of the asset. Also,
agencies' termination liabilities for ESPCs typically correspond to the
outstanding principal balances due to the ESCOs.
7. We did not analyze the validity of DOD's claims because DOD could
provide no data supporting its claims. For example, the Navy does not
maintain a central system to track savings. Therefore, it could not
provide data on the amount of cost savings attributable to the use of
ESPCs. DOD also did not provide sensitivity analysis reflecting the
opportunity costs of waiting for appropriations. We did review the one
Oak Ridge National Laboratory cost study recommended to us, Evaluation
of Federal Energy Savings Performance Contracting--Methodology For
Comparing Processes and Costs of ESPC and Appropriations-Funded Energy
Projects (March 2003). In addition to our own review of the study, we
interviewed the authors of the study and talked with agency officials
about the study's methodology. Based on our analyses we found two major
flaws with the study: (1) we agreed with the study authors that the
sample size was too small and was not applicable to the entire federal
sector and (2) the study compares the costs and savings across various
types of ESPCs installed in several different federal facilities,
making it difficult to compare energy savings because the savings would
depend upon too many unpredictable factors. Also, as discussed on pages
29 and 30, we did discuss with GSA and Navy officials their historical
funding experiences. We note that DOD itself (see comment 4) says that
ESPCs increase the cost to the government by 25 to 35 percent compared
to timely, full, and up-front funding.
8. Our analysis recognizes the value of the technical expertise
provided by ESCOs by assuming that detailed energy surveys would be
needed and purchased under either funding scenario. We include the cost
of this type of service in our proxy for the amount Congress would have
to appropriate had ECMs been financed through timely, full, and up-
front appropriations. Further, our report notes that agencies' heavy
reliance on the ESCOs to recommend, install, and perform M&V to verify
results on their recommended ECMs creates potential conflicts of
interest that require active participation and scrutiny by agencies.
9. See comment 3.
10. See comment 5.
11. This information was taken directly from a FEMP document that FEMP
provided in one of its training courses offered to agencies (Super ESPC
Agency Project Binder, July 2004). On page 6 of the chapter entitled,
"Introduction to M&V for DOE Super ESPC Projects" it says, "In the
event that the stipulated values overstate the savings or reductions in
use decrease the savings, the agency must still pay the ESCO for the
agreed-upon savings."
12. In our opinion, because large buy-downs indicate the availability
of funds in the first year of the contract, they imply there may have
been opportunities to purchase ECMs in smaller, useable segments, when
technically feasible. However, because DOD prepared no business case
analysis to determine the viability of this alternative, it cannot be
known whether this would have been cost effective or not. Business case
analyses are well accepted as a leading practice among public and
private entities. OMB requires all executive branch agencies to prepare
business case analyses for major investments as part of their budget
submissions to OMB.
13. For our case studies, buy-downs did not always represent an
"extremely small percentage of the overall contract." As discussed on
page 39 of the report, three of the six case studies we reviewed
obligated and paid a significant portion of the total cost of the ECMs
in the first year of the contract. These three case studies used one-
time savings to pay down about 7 percent, 38 percent, and 39 percent of
contract cycle costs. We believe this shows that opportunities exist to
acquire ECMs in smaller, useful segments when technically feasible. The
other three case studies are not used to support our conclusion because
the up-front payments on these delivery orders stemmed from federal
funding unexpectedly made available to mitigate energy shortages in
California during fiscal year 2000 or because the up-front payment was
minimal. We are not asserting that these opportunities exist in every
case but we remain of the view that they should be explored as part of
a business case analysis.
14. As stated on pages 35 and 36, we found that GSA and the Navy took
an active role in negotiating case study ESPCs to protect the
government's interest. However, the potential for problems has been
demonstrated through numerous Army Audit Agency reports issued over the
last several years on ESPCs awarded by the Army. These reports stated
that energy savings baselines established by the ESCOs were faulty,
resulting in overpayments to the ESCO. Accordingly, we believe our
conclusion is both warranted and relevant.
[End of section]
Appendix V: Comments from the Department of Energy:
The Under Secretary of Energy:
Washington, DC 20585:
October 25, 2004:
Susan J. Irving:
Director, Federal Budget Analysis:
Strategic Issues:
Government Accountability Office:
Washington, DC 20548:
Dear Ms. Irving:
Enclosed are the Department of Energy's comments on the Draft Report
entitled "Capital Financing: Partnerships and Energy Savings
Performance Contracts Raise Budgeting and Monitoring Concerns" (GAO-05-
55). The Department appreciates the Government Accountability Office's
review of the matters addressed in the Draft Report.
Please contact Dreda Perry of my staff regarding any questions you may
have. Ms. Perry can be reached at 202-586-0561 or
dreda.perry@ee.doe.gov.
Sincerely,
Signed by:
David K. Gannan:
Acting Under Secretary for Energy, Science and Environment:
Enclosure:
DOE Comments On Draft GAO Report "Capital Financing: Partnerships and
Energy Savings Performance Contracts Raise Budgeting and Monitoring
Concerns" (Sep 2004):
The Department of Energy (DOE) is grateful to the Government
Accountability Office (GAO) for reviewing certain capital financing
issues. We appreciate this opportunity to provide comments on the draft
report entitled "Capital Financing: Partnerships and Energy Savings
Performance Contracts Raise Budgeting and Monitoring Concerns" (Draft
Report).
The draft report begins by invoking the 1967 Commission on Budget
Concepts in faulting DOE's arrangements under which it has acquired
(under specific legislative authority) building energy conservation
improvements and contract rights to general office space in private
construction at Oak Ridge, Tennessee. The draft report seems to posit
that these arrangements contravene the 1967 Commission's injunction
that "borderline ... transactions" be included in the President's
budget, and that a separate capital budget approach to budgeting be
rejected. The "borderline" transactions about which the Commission was
actually speaking were the operations of the social security,
unemployment, highway, medicare, and civil service retirement trust
funds, and their "exclusion ... from the present [1967] administrative
budget[.]" 1967 Comm'n Report at 25 & 26. And as to the Commission's
rejection of separate capital budgets, what it faulted was "exclud[ing]
outlays for capital goods from the total of budget expenditures," 1967
Comm'n Report at 33, rather than the amount of expenditures (such as
those for operating lease legal obligations) actually included in the
budget. Nothing in the 1967 Commission's report addressed operating
leases such as those at Oak Ridge, and even as to true capital
expenditures indicated that such "expenditures" be displayed by an
accrual method under which "goods and services acquired under contract,
as in construction ... result in reporting expenditures ... as the work
is actually performed to Government specifications." 1967 Comm'n Report
at 39. This element of the 1967 Commission report said nothing at all
about recording obligations, let along calculating them by reference to
speculation about the future when no legal obligations have been
assumed by an agency.
To briefly summarize our comments, we agree with one of the two
recommendations in the Draft Report, and we strongly disagree with the
other. DOE agrees with the recommendation that "agencies should perform
business case analyses and ensure that the full range of funding
alternatives, including the technical feasibility of useful segments,
are analyzed when making capital financing decisions." However, DOE
strongly disagrees with the recommendation that "the Director of OMB
work with scorekeepers to develop a scorekeeping rule that would ensure
that funds are obligated to reflect the full commitment of the
government, considering the substance of all underlying agreements,
when third party financing is employed."
Fundamentally, the Draft Report does not suggest any serious defects
with the current scorekeeping rules, which seem to us to be a
thoughtful, carefully developed, well understood, and fairly easily
administered set of factors well designed to determine what the "full
commitment of the government, considering the substance of ...
underlying agreements" actually is, and to score transactions
accordingly. Likewise, the Draft Report does not suggest any specific
factors that should be added or substituted. Rather, to the extent that
it proposes anything, it seems to be suggesting a "totality of the
circumstances" approach that would lead to subjective judgments about
"the substance of all underlying agreements" divorced from a list of
factors that can be understood and applied to different transactions.
Additionally, in the case of Energy Savings Performance Contracts
(ESPCs), such a scorekeeping and budgeting rule likely would discourage
or prevent agencies from entering into ESPCs, a result that is contrary
to statutory authority, good facility management, energy efficiency,
and the recently-expressed will of Congress. In the case of public/
private alternative financing arrangements, such a scoring rule would
not lead to more accurate or more transparent scoring or accounting,
and would discourage or prevent agencies from using their statutory
authorities to accommodate a myriad of different transactions involving
capital assets in a way best suited to advancing the federal
government's needs and interests in particular situations.
The Draft Report intermingles two completely separate and distinct
alternative financing methods - ESPCs, and the separate mechanism of
entering into public/private arrangements that relate to capital
assets. We believe that while both ESPCs and public-private
arrangements use private sector funding, their inherent differences
merit different analyses. To avoid confusion and misapplication of
principles, these different mechanisms should be analyzed separately.
ESPCs essentially involve improvements to existing federal assets and
are structured to capture energy savings generated through the
implementation of energy conservation measures by private sector
companies. These projects are relatively small, large in number, exist
throughout the federal system, and are specifically authorized under a
government-wide, statutory scheme. In fact, only in the last few days,
both houses of Congress approved an extension of the authority for
government agencies to enter into ESPCs (see footnote 20). Public-
private arrangements, on the other hand, are generally larger
transactions designed to accomplish particular mission needs. These
arrangements are complex, few in number, location-dependent, unique in
transactional structure, and require reliance on specific agency
legislative authorities. These projects require review by the agencies'
executive management, and a determination of best economic value (and
course of action) made consistent with the principles of sound
financial and budgetary decision making.
Therefore, our comments on the Draft Report will be bifurcated into
these two separate topics. A section on public/private arrangements
will provide general comments on three major subcomponents --business
case analysis, scorekeeping requirements, and legal interpretation of
transactions. The ESPC section will address the Draft Report in light
of the legislative history of that program. Specific comments on Draft
Report language will be set out in an Appendix.
At the outset, however, we recognize that both of these mechanisms deal
with the problem of how agencies manage capital improvements and
infrastructure development in an era of limited resources, limited
access to appropriations, and growing mission demands. In that context,
GAO has detailed the deteriorating condition of the federal real
property portfolio in recent reports concluding that the "federal
portfolio is in an alarming state of deterioration." [NOTE 1] Not
surprisingly, GAO designated federal real property as a high-risk area
requiring executive attention to resolve, noting that the government's
current real property practices "...have multibillion-dollar cost
implications and can seriously jeopardize mission accomplishment." DOE
agrees with these previous GAO's assessments.
Comments Related to Public/Private Arrangements Involving Capital
Assets:
1. Agencies Should Perform Business Case Analyses and Ensure That the
Full Range of Funding Alternatives Are Analyzed When Entering Into
Public/Private Arrangements Involving Capital Assets:
The Office of Management and Budget (OMB) has provided the agencies
with a comprehensive framework for conducting the business case
analysis. This framework is based on two major OMB Circulars:
1. OMB Circular A-11, Preparation, Submission and Execution of the
Budget; and:
2. OMB Circular A-94, Guidelines and Discount Rates for Benefit-Cost
Analysis of Federal Programs.
Circular A-11 provides the agencies executive-level guidance on the
planning, budgeting, and acquisition of capital assets, and provides
the budgetary scoring rules to be applied to the alternatives in the
range. In addition, GAO has issued its Executive Guide: Leading
Practices in Capital Decision-Making, [NOTE 2] which supplements
Circular A-11's Capital Programming Guide. [NOTE 3] Together, these
documents form the capital asset planning and budgetary foundation for
the federal government.
Circular A-94 provides agencies the requirements for conducting a
financial analysis of the alternatives in the range. The financial
analyses employed by OMB in Circular A-94 are, with minor exceptions,
identical to those used in the private sector. Using A-94, agencies
calculate the present value life-cycle cost of the alternatives in the
range, and, when possible, other values such as net present values,
internal rate of return, accounting payback, etc. DOE notes that OMB A-
94 discourages the use of operating leases to satisfy long-term needs
as operating leases are generally more costly than acquisition.
The business case for any capital investment decision should consider
the full range of alternatives. At a minimum, this range should include
appropriated funding, private-sector solutions, and continuing the
status quo.
DOE agrees with GAO that the agencies should perform a business case
analysis of the full range of alternatives, and that this analysis
should be used by Executive Management to reach a reasoned business
decision. A complete business case analysis considers many factors,
including the life-cycle cost of the alternatives, the commitment of
the federal government, the federal government's needs and any
uncertainty about future needs, risk borne by the parties, and the
likelihood that any particular alternative can be achieved. Costs,
while an important part of the analysis, need to be weighed against
these other factors by Executive Management.
DOE agrees that upfront appropriated funding can result in the lowest
life-cycle cost to the taxpayer, but recognizes that in many instances
third party financing can provide a lowest life cycle cost for a given
asset. Nevertheless, for a variety of financial and operational
reasons, federal agencies, just like entities in the private sector, do
not always want to or need to "acquire" the capital assets that they
may need to use for a period of time. For example, there may be many
circumstances where the government needs the use of a particular asset
for a period of time, but does not anticipate a long-term need for it
or cannot ascertain with any degree of certainty whether such a need
will continue to exist. In those circumstances, it might be possible in
retrospect to say that up-front appropriated funding would have been
the least expensive option, but that may not have been clear at the
time that advance decisions needed to be made. In short, a reasoned
business decision must include all factors, including the feasibility
of any particular alternative, to determine the best economic value for
an agency when it determines its capital asset needs and budgets.
In its Executive Guide, GAO provides guidance to the agencies for
capital investment planning. GAO states:
From a federal agency's point of view, however, full funding (for
capital assets) can be problematic, especially under periods in which
budget caps constrain spending. [NOTE 4]
The federal government's financial outlook has not improved since GAO
made this observation. In fact, in numerous reports published since the
mid-1990s, [NOTE 5] GAO continually questions the feasibility of
obtaining appropriations to acquire or service mission critical assets
due to fiscal constraints. Current budget projections show that the era
of federal budget constraints will continue for years.
DOE questions why the Draft Report does not include the cost of
maintaining the status quo in the range of alternatives in the business
case analysis for Executive Management consideration. While the cost of
obtaining space through alternative financing may be marginally higher
than obtaining space through appropriated funds, the cost of
maintaining the status quo is, in many cases, substantially higher.
Executive Management must consider these costs, weighing them against
the feasibility of obtaining appropriated funds, the risk to the
government, and other factors. The Draft Report's exclusion of the
status-quo alternative ignores this critical element for Executive
Management consideration. Exclusion of this alternative incorrectly
limits the choices faced by federal agencies, and understates the range
of true costs to the taxpayer. [NOTE 6]
DOE's Office of Management, Budget, and Evaluation/Chief Financial
Officer and Office of General Counsel are currently working on a draft
policy statement that will require a business case analysis be prepared
for public/private alternative financing arrangements, which would
include a cost-comparison analysis based on OMB A-94. This new DOE
policy [NOTE 7] will, when finalized, put DOE in compliance with GAO's
recommendation.
2. The Draft Report's Recommendation to Score Operating Leases as
Capital Leases is Inconsistent with Financial Accounting Standards and
OMB's Guidelines:
The Draft Report recommends that OMB "work with scorekeepers to develop
a scorekeeping rule for the acquisition of capital assets to ensure
that the budget reflects the full commitment of the government,
considering the substance of all underlying agreements, when third
party financing is employed"(Draft Report at 38) [emphasis added]. This
recommendation apparently seeks to have budget scorekeepers look beyond
current factors set out in OMB Circular A-11 to judge the true
"substance" of a transaction in order to capture and require full
upfront budget scoring for transactions that now qualify as operating
leases and do not require scoring under OMB Circular A-11. GAO
apparently believes that implementation of this recommendation would
improve budgetary transparency. [NOTE 8]
This recommendation to score a transaction according to the
"substance", as determined by criteria nowhere identified in the Draft
Report rather than using the well-established and fairly easily applied
factors set out in OMB Circular A-11, is inconsistent with both
private-sector and public-sector financial accounting standards. [NOTE
9] From a budgetary viewpoint, OMB A-11 is concerned with the
government's legal obligations and the actual allocation of risk in
transactions, and appropriately requires the full amount of the
government's legal obligations to be scored up front. Notably, the OMB
A-11 standard for determining whether a lease is a capital or operating
lease is more stringent than that used in the private sector. The Draft
Report acknowledged that the operating lease resulting from the ORNL
third-party financing arrangement met the stringent budgetary standard
established by OMB in A-11, and was correctly scored.
Moreover, the Draft Report's assumption that the ORNL transaction is
"really" the acquisition of a capital asset demonstrates the lack of
standards and the potential considerations that would be ignored in the
approach GAO proposes. In fact, the ORNL transaction allocated risk and
costs differently from the way they would have been allocated had it
been a true acquisition because it was NOT a true acquisition. Rather,
it reflected DOE's decision NOT to acquire a new building at the time
it entered into the transaction, in significant measure because DOE did
not want to make the commitment of resources to the acquisition of a
new building at the time of the transaction because there were genuine
uncertainties about the forecast need for the use of these facilities.
It did, however, have sufficient interest in a new building to be
prepared to make its property available for construction of a building
that it could use, and to enter into a year-by-year lease for the
building, but subject to its right to terminate the lease should funds
prove scarce and other mission needs take priority. DOE's rights and
obligations reflected that it was making a more limited commitment that
fell well short of an acquisition, and it would not have made sense to
score the transaction as if it had been an acquisition.
If adopted, this recommendation would raise the specter of obstacles to
the use of third-party financing arrangements that could readily be
interposed on a subjective basis, thereby in practice making these
arrangements very difficult to enter into. That would be unfortunate.
These arrangements provide the government flexibility to address
certain types of challenges associated with the management of its real
property in a timely manner while shifting capital and transactional
risk to the private sector. Moreover, implementation of the
recommendation would mean that budget scoring would impose obligations
on agencies and their Executive Management that exceed the actual legal
obligations they are incurring through their transactions, thus
incorrectly over-stating the obligations that an agency actually has
incurred. By severing budget scoring from the allocation of actual
legal and economic risk, this GAO recommendation would degrade budget
transparency.
Moreover, the Draft Report's recommendation also is inconsistent with
GAO's desire to ensure budgetary transparency while providing greater
manager flexibility. [NOTE 10] Because it does not contain any specific
standards, it is hard to see how the Recommendation will promote
transparency. And it is clear that it will hinder management
flexibility. In the Executive Guide: Leading Practices in Capital
Decision-Making, GAO encourages agencies to pursue innovative
approaches that both balance the need for budgetary control and provide
managerial flexibility when funding capital projects. In the past, GAO
has recognized the need to bring private-sector solutions to the
government's real property management challenges, and GAO has
identified a number of innovative approaches for providing the agencies
the managerial flexibility necessary to use these approaches. DOE
questions the Draft Report's apparent retreat on this issue.
DOE agrees with GAO that Executive Management must consider the full
range of alternatives in a business case analysis as dictated by OMB A-
94. Accordingly, the present value life-cycle costs of all the
alternatives, including appropriations, operating leases, maintaining
the status quo, and so on, must be evaluated equivalently. Executive
Management's decision, after weighing all factors including the
feasibility of the alternatives, may result in a determination that the
lowest cost alternative does not deliver the best value.
3. The Draft Report Incorrectly Attempts to "Looks Beyond" the Oak
Ridge Transaction Documents to Make a Subjective and Non-Legal Judgment
That the Transaction Should be Categorized and Scored As a Capital
Lease Under OMB Circular A-11.
The Draft Report characterizes the arrangement at Oak Ridge National
Laboratory (ORNL) as a transaction in which the legal instruments
created operating leases, but underlying factors led GAO to believe
that the commitment was really a long term obligation of the federal
government and that the scorekeeping rules should be changed to score
such operating leases as capital leases. In essence, GAO is
recommending that OMB "go beyond" the legal instruments (and the actual
legal and business obligations) of the third party financing
arrangements and go beyond Circular A-11 factors which substantiate
scoring as an operating lease, to find the existence of a capital
lease. GAO suggests "going beyond the strict terms of a proposed
transaction and scoring based on the substance of the deal," (Draft
Report at 36), without specifying what factors or weighting might be
appropriate to use in determining that "substance." [NOTE 11]
The Draft Report and its conclusions reflect a fundamental
misunderstanding of the ORNL transaction and the operating lease that
was the subject of the OMB analysis. The report ignores both the legal
structure and the underlying economic substance of the operating lease
at ORNL, and substantially departs from well-accepted methods of budget
scoring and of private sector accounting. The Draft Report and its
recommendations to OMB represent a misplaced shift in emphasis in
determining budgetary treatment of alternatively financed transactions
from the risk-based analysis of OMB Circular A-11, that focuses on
assessing the government's actual obligations or commitments and the
actual allocation of risk between the public and private sector, to a
conclusory and speculative emphasis on whether the transactions involve
long-term governmental needs (Draft Report at 2).
In fact, the ORNL transaction was deliberately structured so as not to
commit the government to purchasing a capital asset, in substantial
part, at least, for business reasons: DOE was not sure, in light of
evolving mission needs, whether it actually wanted to buy three new
buildings at the time it authorized the transaction. Accordingly, its
business needs were better met by renting the buildings. Therefore, the
transaction is set up so that it is the private sector developer and
its backers - not DOE - that bear the financial risk that DOE will not
use the buildings long-term by making DOE's only obligation a sublease
terminable at DOE's option with one-year's notice. DOE and its
Management and Operating (M&O) contractor have the specific right for
any reason whatsoever to terminate use of the new facilities with only
one year's notice. Accordingly, DOE (and its contractor) are only
committed for a term of one year. Therefore, DOE's risk is limited to
one year's rent, related expenses, and the value of 6.6 acres of land.
The public offering materials fully disclosed the one-year termination
provision and the full one-point increase in the interest rate on the
public offering after the events of September 11, 2001, reflected the
market's understanding of the transaction. Further reflecting the
actual allocation of risk is the fact that there were reserve funds
established by the private-sector Bond Trustee to cover any shortfall
in rent beyond the one-year period in the event of termination. [NOTE
12] Although the Draft Report cites a series of factors that make it
unlikely that DOE will exercise the option to terminate the lease, the
Draft Report's analysis depends on a series of possibilities that may
or may not come to pass (e.g. "ORNL's Project Manager told us that,
even if ORNL's mission was downsized, it was unlikely that DOE would
terminate any of the leases of the three new, state-of-the art
buildings to reoccupy the now empty, dilapidated buildings," Draft
Report at 17), but the risk of which DOE deliberately chose not to
bear. In that connection, it would be well not to forget that DOE
terminated the Super Collider and the Clinch River Breeder Reactor
projects even after a very substantial direct investment had been made
by the government for which there was little commercial market.
In short, it is a legal absolute that under the subleases, with one
year's notice, DOE is completely free of any further obligation to use
or pay for the new facilities, and that the termination provision
served a legitimate business purpose of DOE'S. Only by ignoring this
key provision, as well as the legitimate business purpose for this
provision, can GAO assert that there is something wrong with the way
the transaction was scored, even while admitting that it was properly
scored under current scoring rules. Neither GAO's recommendations, nor
budget scorekeeping rules, should be based on non-legal speculation
about what an agency may or may not do in the future, or the exercise
of subjective judgment "based on the substance of the deal" that
ignores the federal government's clear legal rights, obligations and
actual risks arising from the transaction at issue.
Comments on the Draft Report's Analysis of ESPC's,
The Draft Report's Conclusions Concerning the Budgetary Treatment of
"Long Term Obligations" Under ESPCs Does Not Reflect the Congressional
Intent For Which the ESPC Legislation Was Enacted and Continued.
The GAO premise in the Draft Report is to analyze ESPCs from a
traditional government contracts perspective as an acquisition of an
asset, and that the agency somehow needs to score as budget authority
the whole multi-year term obligation under the contract. This view is
misplaced at best, and at worst would result in nullifying the
Congressional purposes that the law was enacted to accomplish. The ESPC
mechanism was intended by Congress and the Executive Branch to be the
primary tool for the federal government to reduce energy consumption at
federal facilities and meet mandated energy savings goals. [NOTE 13] We
oppose the use of administrative budget scoring processes to negate the
Congressional objective of promoting energy conservation through the
use of ESPCs. A summary of the legislative history of this mechanism
would be helpful in explaining the errors in the Draft Report's
analysis.
First authorized in 1985, [NOTE 14] the unique ESPC long-term,
contracting mechanism was instituted by Congress to reduce energy
consumption at federal facilities. Under this mechanism Congress
authorized agencies to enter into up to 25-year arrangements with
private sector energy service companies in which the private sector
company would both install energy-efficient equipment and provide
energy-management services at federal sites for the purpose of
implementing energy conservation measures (ECM) at those sites.
Congress recognized the unique nature of these arrangements by
exempting these contracts from the Anti-Deficiency Act and allowing
these private-sector energy service companies to install their own
energy-efficient measures into federal facilities "at no-cost to the
federal customer" under which the private company "risks its own
capital in return for a share of value of the energy savings resulting
from the improvements" without the need to obligate total contract
costs up-front. [NOTE 15] In that context, it is a misnomer for GAO to
assert that ESPCs are being used by agencies for the purpose of
acquiring an asset.
To encourage greater use of this mechanism, Congress, in 1992, [NOTE
16] provided expanded authority for energy savings projects. In
recognizing the unique budgetary treatment of ESPCs, [NOTE 17] the
House and Senate conferees on the Energy Policy Act of 1992 stated:
Energy savings performance contracts are a mechanism through which
private sector funds can finance Federal energy efficiency
improvements. The Conferees recognize that these contracts differ
significantly from traditional Federal procurement contracts. Under
these contracts, the contractor is expected to bear the risk of
performance, make significant capital investment, guarantee
significant energy savings to the government agency, and from these
savings the agency, in effect, makes payment to the contractor. [NOTE
18]:
Congress specifically intended that these capital improvements were not
to be funded up-front by the federal government; rather the risks of
financing and performance were to be borne by the private sector. The
federal obligation would only attach to energy savings actually
incurred on an annual basis by the contractor's "guaranteed" energy
savings. The agency only obligates its annual appropriation (otherwise
provided for utility acquisition or building maintenance) when it makes
payment to the contractor to the extent that savings actually arise
from the contractor's performance.
Moreover, Congress, in reauthorizing ESPC authority through FY 2006 (in
the Ronald W. Reagan National Defense Authorization Act for FY 2005),
[NOTE 19] was concerned that this contracting mechanism was being
underutilized by the federal government. Therefore, Congress broadened
the allowable scope of these contracts and imposed a requirement on DOE
to implement administrative changes to "increase [ESPC] program
flexibility and effectiveness." [NOTE 20] This Congressional concern
runs counter to the Draft Report's recommendation that this method of
contracting should be constrained (if not eliminated entirely) by
requiring full, up-front, budget recording of all potential obligations
under the multi-year term of the contract. This would, in effect,
cripple the ability of the federal government to make needed capital
improvements to meet the energy savings goals imposed by Congress, and
completely misses the fact that the financial risks of this type of
contracting are borne by the private sector and not the federal
government.
In addition, the Draft Report analysis concludes that obtaining up-
front appropriations for these improvements would be less costly, and
therefore preferable, to ESPCs. This conclusion does not reflect the
realities of the agency budget and Congressional appropriation
processes, especially when the appropriation process is compared to the
timing needs of the improvements and to the extent and scope of
hundreds of energy conservation projects throughout the federal
infrastructure. Furthermore, it seems to mistakenly assume that in each
case in which an ESPC agreement and resulting improvements exist, the
relevant agency would seek the necessary appropriation, obtain it from
Congress, and implement the energy conservation measure. Again, this
assumption has no basis in reality. Waiting first to see if
appropriations can be obtained for each project before proceeding under
the ESPC process will cause serious, costly, and irreparable harm to
federal energy and infrastructure goals and would not save money or
energy. This would be an impractical and counterintuitive approach, and
contrary to GAO's own findings that federal infrastructure
deterioration is fast becoming a high-risk area in crisis.[NOTE 21] It
is virtually certain that this approach would result in fewer energy
efficiency improvements, fewer energy conservation measures, and
foregone energy savings. The ESPC authority is available as a creative,
practical, and timely method to assist the government in stemming this
deterioration and promoting energy conservation.
APPENDIX to DOE Comments on the Draft GAO Report, "Capital Financing:
Partnerships and Energy Savings Performance Contracts Raise Budgeting
and Monitoring Concerns"
DOE Comments on Specific Draft Report Language:
Discussion of the Oak Ridge Transaction:
1. The UT-Battelle modernization program implemented through funding by
the State of Tennessee and by alternative financing did not constitute
an acquisition by the federal government of capital assets. DOE has not
acquired any of the assets that were the subject of the Oak Ridge
transaction. At the end of 25 years, depending upon the intervening
circumstances, it may acquire them. The risk for the venture, other
than the cost of the land, is borne entirely by the bond investors and
private parties.
2. The UT-Battelle alternative financing approach did not result in a
partnership with DOE. The UT-Battelle project does not meet the
definition of a "partnership" as defined in the draft report or in the
August 2003 GAO report cited in the draft. As stated on page 6, last
paragraph, partnerships generally entail a government agency "engaging
a third party to, among other things, renovate, construct, operate, or
maintain a public facility." DOE did not engage UT-Battelle to
construct the three buildings, and those buildings are not "public
facilities." However, the draft report repeatedly cites the existence
of such a partnership as related to the UT-Battelle project, most
noticeably with the first statement on page 70. DOE's role was no more
than transferring a 6.6 acre parcel of land. UT-Battelle requested the
land transfer from DOE. Subsequently, DOE authorized UT-Battelle to
enter into a real estate lease for space, a routine act that is
performed on multiple occasions by UT-Battelle in the past four years.
As shown by the August 2000 letter from Dr. Madia to Mr. Malosh, and as
approved by DOE in March 2001, DOE was not the initiator of this
project. A partnership as defined in the August 2003 GAO report may
well exist for the other projects reviewed in the report, but does not
exist for the UT-Battelle project.
3. The UT-Battelle subleases do not violate the 75% rule. The three
facilities constructed and subleased to UT-Battelle are subject to a
ten year sublease with three five year options. Assuming that all of
the options are exercised, the sublease will end after 25 years. The
buildings have an estimated economic useful life of no less than 39
years, and as evidenced by the existing facilities at ORNL, will likely
be used for more than 50 years. Thus, the 25 year lease with options
could not exceed 75% of the expected useful life of the facilities.
GAO's prediction that "[i]t seems unlikely that the agencies will
vacate or abandon these assets before the end of their economically
useful lives, thus exceeding the 75 percent criteria for an operating
lease" invents an entirely new theory of the 75% rule, under which
compliance would be measured not by legal rights and obligations under
the lease as actually entered into, but by someone's subjective
prediction about what one of the parties is likely to do. This would
turn the 75% rule on its head.
4. The UT-Battelle transaction did not "establish [a] long term
commitment[] of the government," contrary to the assertion on p. 36
that all the transactions studied in the Draft Report established such
commitments.
5. Purpose of Transaction. On p. 34 the Draft Report asserts that the
ORNL transaction was undertaken "to obtain project financing." That
misunderstands the transaction. The transaction was undertaken
fundamentally because UT-Battelle suggested it as a means by which DOE
could obtain something of value to it - use of a new building on
property convenient to it that it might eventually decide to buy -
without having to commit to buying the building at the time of the
transaction.
6. Reference to DOE legal opinion. The last sentence on page 34
references a memorandum issued by DOE's general counsel. The DOE legal
opinion that was issued by the Assistant General Counsel for General
Law dealt with whether the lease was an operating lease. The purpose of
the opinion was not to address the issue for which the Draft Report
cites it.
7. The draft report suggests a conflict of interest and potential for
fraud or wrongdoing in the UT-Battelle project. On pages 7 and 30, the
draft report suggests a conflict of interest existed between UT-
Battelle and UT-Battelle Development Corporation (UTBDC), and the
dealing of the two entities with DOE. Obviously, there are overlapping
interests between UT-Battelle and UTBDC. Therefore efforts were made to
reassure DOE that neither UT-Battelle, its two owners (the University
of Tennessee and Battelle Memorial Institute), nor UTBDC would gain any
financial advantage from this arrangement. The DOE Certified Realty
Officer approved the lease as being consistent with market value for
comparable facilities in the Oak Ridge/Knoxville area. UTBDC and DOE
agreed that UTBDC would simply pass on to UT-Battelle the lease costs
imposed on UTBDC through the Facility Leases from Keenan Developers of
Tennessee. UTBDC incurred costs exceeding two million dollars for the
project for which it has not been reimbursed by DOE. UT-Battelle is a
non-profit LLC, and UTBDC is a tax-exempt entity under the rules of the
Internal Revenue Code. Both undergo annual audits by an independent
auditing firm.
Specific Comments on Draft Report's Discussions of ESPCs:
1. Throughout the report, GAO notes that there is insufficient data to
measure the effect of delayed appropriations on foregone energy and
maintenance savings. See, e.g., Draft Report at 25-26.
Response: GAO has drawn its conclusions and recommendations based on
examining 6 ESPC projects, zero direct-funded projects, and an analysis
based on the assumption that operating projects can be achieved either
way in the same amount of time. The ORNL study examined 71 ESPC
projects finding, on average, 28 months to an operating project; and 12
direct-funded projects finding, on average, 63 months to an operating
project. Throughout the Draft Report, GAO has based its recommendations
and conclusions on inadequate analysis and insufficient data.
Additional information on delays between initial requests and eventual
receipt of direct funding for energy efficiency projects is needed to
support a realistic assumption about how long agencies wait for energy
efficiency project direct funding. There is no evidence that GAO
attempted to gather data from OMB and agencies on such things as 1) the
amount of direct funding requested by facilities organizations within
agencies for energy efficiency projects annually in recent years
compared to such requests that made it into the overall agency
requests; 2) requests for the Office of Energy Efficiency and Renewable
Energy (EE) project direct funding in overall agency requests that
cleared OMB and were included in agency requests to Congress; and 3)
requests to Congress for EE project direct funding compared to
appropriations actually received.
2. Page 6, second sentence: "Also, for our ESPC case studies, the
government likely incurred additional costs for the M&V."
Response: Measurement and verification ("M&V") does result in
additional costs. M&V is a best practice that should be performed at a
practical level regardless of how a project is funded. Thus, to be
balanced, the lack of M&V to protect the government should be cited as
a weakness in appropriated projects as often as the cost of M&V in
ESPCs is projected as a detriment.
3. Page 2, 1st paragraph, last sentence. "Agencies receive the same
program benefits regardless of the financing approach used, assuming
they purchase the same capital equipment." Page 12, second paragraph,
first sentence: "Critics of ESPCs, however, point out that direct
purchase of more efficient energy systems would allow all future
savings to accrue to the government, rather than paying out a
percentage of the savings to private contractors."
Response: Implicit in GAO's analysis is the assumption that buying an
energy conservation measure and placing it into service, whether
through direct funding or ESPC, results in the same benefits over
service life. This assumption is wrong. Without M&V, problems go
undetected, equipment is not maintained as well and savings decay.
Costs of M&V should not be counted against an ESPC, unless the same
costs also are counted against direct-funded projects.
4. Page 11, second paragraph, last sentence: "Consequently, no increase
in appropriations is required."
Response: GAO is correct; with ESPCs no increase in appropriations is
required. When we consider budget outlays for both capital and energy
and related operating expenses, ESPCs are revenue neutral.
5. GAO notes that `for our six ESPC case studies, the government's
costs of acquiring assets increased 8 to 56 percent by using ESPCs
rather than full upfront appropriations."
Response: This assertion appears to be based on incomplete data. We
would like to work with GAO to review its analytical approach to ensure
its adequacy, which is unclear at this time.
DOE Comments GAO Draft Report, CAPITAL FINANCING: Partnerships and
Energy Savings Performance Contracts Raise Budgeting and Monitoring
Concerns (GAO-05-55) 11/23/2004 Page 14:
In the meantime we have used our analysis to compare ESPCs to direct
funding for the six projects. The "best case" for direct funding was
defined as follows: operating projects are achieved as quickly as with
ESPCs, they are financed by the Treasury, and they bear no survey/study
or M&V cost adders to project cost. Rather than charge M&V costs to
ESPCs, yet claim direct-funded projects with no M&V achieve the same
savings benefits over the long-term, as GAO did, our analysis drops M&V
costs from ESPCs as well. For consistency and transparency, our
spreadsheet was used to calculate the present value of costs for the
ESPC and direct funding best cases and the maximum possible % increase
in PV costs attributable to ESPC was calculated relative to the direct
funding "best case." The following table provides a summary comparison
of GAO's analysis and our analysis.
Agency Site: Navy Region Southwest;
GAO's Estimate Of % Increase in PV Cost Of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 8%;
Maximum Possible % Increase in PV Cost of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 12%.
Agency Site: Patuxent River NAS;
GAO's Estimate Of % Increase in PV Cost Of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 33%;
Maximum Possible % Increase in PV Cost of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 13%.
Agency Site: Naval Submarine Base Bangor;
GAO's Estimate Of % Increase in PV Cost Of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 23%;
Maximum Possible % Increase in PV Cost of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 4%.
Agency Site: GSA Gulfport, MS;
GAO's Estimate Of % Increase in PV Cost Of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 56%;
Maximum Possible % Increase in PV Cost of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 26%.
Agency Site: GSA North Carolina Sites;
GAO's Estimate Of % Increase in PV Cost Of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 39%;
Maximum Possible % Increase in PV Cost of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 12%.
Agency Site: GSA Atlanta, GA Sites;
GAO's Estimate Of % Increase in PV Cost Of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 27%;
Maximum Possible % Increase in PV Cost of ECMs Financed Through ESPCs
Over Cost If Direct Funded: 16%.
[End of table]
GAO's estimates of percent increases in energy conservation measure
costs with ESPCs exceed the maximum possible percent increases
calculated based on "best case" assumptions for direct funding. With
real world assumptions for direct funding (projects bear realistic
adders for survey/study costs, delays, etc.) ECM costs may or may not
be greater when ESPCs are used.
6. Page 16. Section heading: "ESPC Commitments Are Not Fully Recognized
Up-Front in the Budget".
Response: ESPC commitments are already recognized in the budget because
a facility's operating expenses are recognized in the budget.
Therefore, the budget already recognizes ESPC commitments because they
are satisfied from savings to operating expense budgets. ESPCs also
have guarantees and contractual recourse --should savings fall short,
the government can demand a cure and withhold payments in the meantime.
Recognizing ESPC commitments again up-front would be double counting.
7. Page 22. "As shown in Figure 6, for all six ESPC case studies,
contract cycle energy cost savings specified by the contractor did not
fully cover total contract cycle costs because agencies made upfront
payments ".
Response: The avoided costs are true savings and should be shown.
8. Page 24, second sentence "ESPCs contain assumptions for such things
as hours of operation and ECM efficiency which, taken together,
determine estimated savings. However, if the assumptions are incorrect
and estimated savings are not achieved, the agency is still required by
contract to pay the ESCO the agreed-upon savings specified in the
ESPC."
Response: While it is true that contracted ECM cost savings may differ
from actual ECM cost savings in a given year (because, for example, the
weather was not typical or energy rates were not as forecasted), the
fact remains that the government has remedies against the relevant
contractor (the ESCO) if the verified contracted cost savings do not
match or exceed the guaranteed cost savings each year. When appropriate
assumptions and choices are made, annual contracted and actual savings
will be reasonably similar, and over the contract term contracted and
actual savings tend to converge. The alternative to using simplifying
assumptions for the purpose of calculating savings - having the ESCO
take the risk that factors such as the weather, future energy rates,
and the government's own operating hours and non-project-related loads
will affect savings - would be a poor and expensive choice for the
government. Direct funded projects are the ones where the government
pays no matter what (upon acceptance), and owns future problems.
9. Page 26, first sentence: "Since the agencies did not request
additional appropriations or adjust their plans to accommodate needed
capital investments, it cannot be known whether agencies were correct
in assuming that timely appropriations would not be available".
Response: Agencies make rational decisions based on their past
experience. GAO fails to acknowledge the past experience of these
agencies, which provides the basis for estimating average wait-times
for appropriations, and average costs incurred (in the form of surveys
and studies) for requesting appropriated funds. For its 2003 study,
ORNL examined 12 direct-funded projects finding, on average, 63 months
to an operating project. For its 1993 investigative report on the In-
House Energy Management Program (DOE/IG-0317), the DOE Inspector
General examined 93 direct-funded projects finding, on average, 73
months to an operating project. About all that can be said based on
available information is that ESPCs can achieve operating projects in
28 months on average, and while individual agency experience may vary,
direct funding generally takes considerably longer, if it happens at
all.
10. Page 29, last sentence, "Employing best practices in using ESPCs
also may provide opportunities to better ensure the government receives
the best value for its investment."
Response: DOE agrees that employing best practices and using available
government-side expertise is important with ESPCs and with the
preceding sentences that describe the importance of DOE FEMP/agency in-
house technical assistance on projects.
11. Page 32, section title "Large Buydowns of Principal Raise Questions
About the Need for ESPC Financing."
Response: Agencies partnered with contractors and developed ESPCs,
which take on average 15 months from kickoff to award and another 13
months to become operating projects accepted by the government. During
the 15-month window, agencies make the best value choice and minimize
financing costs by applying available funds as a pre-performance period
payment (P4) to reduce financing costs over term. Borrowing from GAO's
own case studies, the Office of the Secretary of Defense used a
supplemental appropriation in 2001 to help offset the cost of energy
projects on the Western power grid, in an effort to address the energy
supply shortages in the West. So Navy Region Southwest and Naval
Submarine Base Bangor WA applied P4 payments from these funds of $6.9M
and $1M respectively on their ESPCs to reduce financing costs. If these
sites had done nothing while waiting for appropriations, the project
development cycle would have started rather than ended in 2001 and
nothing would have been placed in the ground fast enough to address the
temporary energy supply shortages.
12. GAO's ESPC case study analysis in Appendix 1I assumes it costs
agencies nothing to request and receive appropriations for energy
efficiency projects.
Response: Agencies must spend funds on surveys and studies to define
projects in order to request direct funding for them. Sometimes surveys
and studies fail to find cost-effective projects to recommend. When
projects are found, they are often not funded in the first year,
updating requests year-to-year involves additional cost, and some
recommended projects are never funded. For its 2003 study, ORNL
examined 12 direct-funded projects finding, on average, that survey/
study costs equaled 25% of project costs. For its 1993 investigative
report on the In-House Energy Management Program (DOE/IG-0317), the DOE
Inspector General examined $121.7 million worth of direct-funded
projects over 7 years finding, on average, that survey/study costs
equaled 14% of project costs. GAO's recommendation of elaborate
business case analyses would further increase these costs. Yet in the
Appendix II case studies, GAO assumes agencies incur zero costs to
secure direct funding.
13. Appendix 11, page 44 - appropriations cannot achieve operating
projects on the same schedule as ESPC and energy and maintenance cost
savings are foregone in the meantime.
Response: GAO's assumptions should include lost savings from delays. It
is inconsistent to charge M&V costs to ESPCs, yet claim direct-funded
projects with no M&V achieve the same savings benefits over the long-
term. The avoided costs, which are the source of what GAO refers to as
buy-downs, are true savings that have been left out of the analysis.
ESPCs are revenue neutral, as the GAO report correctly notes on page
11. Why do the case studies not show this?
NOTES:
[1] GAO attributes the condition of the portfolio to a number of
factors, including: scoring rules that force recognition of the full
costs of commitments up-front; the inability of agencies to obtain
appropriated funding to maintain and/or acquire mission critical assets
due to fiscal constraints; and the lack of reliable and useful data for
strategic decision making. See generally, GAO-02-46T, PUBLIC-PRIVATE
PARTNERSHIPS - Factors to Consider When Deliberating Governmental Use
as a Real Property Management Tool (October 2001); GAO-02-622T, FEDERAL
REAL PROPERTY-Views on Real Property Reform Issues (April 2002); GAO-
03-609T, GENERAL SERVICES ADMINISTRATION - Factors Affecting the
Construction and Operating Costs of Federal Buildings (April 2003);
GAO-03-1011, BUDGET ISSUES -Alternative Approaches to Financing Federal
Capital; and GAO-04-119T, FEDERAL REAL PROPERTY, Actions Needed to
Address Long-Standing and Complex Problems.
[2] GAO EXECUTIVE GUIDE - Leading Practices in Capital Decision-Making,
GAO/AIMD-99-32, December 1998, Page 49:
[3] OMB Circular A-11, Part 7, Planning, Budgeting, Acquisition, and
Management of Capital Assets, Supplement to Part 7-Capital Programming
Guide. GAO's Accounting and Information Management Division provided
agencies with its EXECUTIVE GUIDE - Leading Practices in Capital
Decision-Making, GAO/AIMD-99-32 (December 1998), as a supplement to
OMB's Capital Programming Guide.
[4] Page 49, GAO EXECUTIVE GUIDE - Leading Practices in Capital
Decision-Making, GAO/AIMD-99-32 (December 1998):
[5] See footnote 1:
[6] Moreover, in GAO's view, alternative financing arrangements are of
concern because higher interest rates and other factors may increase
the cost of third-party financing compared to full, up-front budget
authority. However, in the draft Report, the GAO repeatedly concedes
that it has not fully evaluated the various potential cost savings that
may be associated with the use of the alternative financing vehicles
profiled in the draft report (see, e.g., pp. 4, 6, 24-25).
[7] This policy will not apply to ESPC's which are subject to
requirements set out in 10 C.F.R. Part 436.30.
[8] GAO'S ORIGINAL CONCERN WITH OPERATING LEASES WAS THAT AGENCIES WERE
EFFECTIVELY INCENTIVIZED TO CHOOSE OPERerating leases in-lieu-of
ownership since operating leases were less costly in any given year, as
stated in GAO -04-119T, FEDERAL REAL PROPERTY: Actions Needed to Long-
Standing and Complex Problems (October 2003). The operating lease that
formed the basis for GAO's concern was a standard operating lease
without the potential for ownership. GAO has recognized that the
current and future fiscal environment will be constrained, and
appropriations for general purpose real property will be severely
limited.
[9] OMB has adopted the private sector's financial accounting standard
for determining the status of a lease, either capital or operating. The
private-sector standard requires that each of four conditions be met;
failure of any one of these conditions requires that a lease be
considered a capital lease. OMB has added two additional conditions to
the private sector standard, ((i) The asset is a general purpose asset
rather than being for a special government and is not built to the
unique specification of the government as lessee, (ii) there is a
private sector market for the asset), resulting in a government
standard that is more stringent than that used in the private sector.
The private sector (and OMB) standard is based on the legal obligations
of the parties. In the private sector this accounting standard protects
the shareholder by ensuring disclosure of the full extent of the
company's legal obligation. OMB's enhanced standard serves the same
purpose for the taxpayers. As an additional test, OMB requires the
agencies to assess the degree of risk assumed by the parties to bolster
its scoring decision.
[10] GAO /AIMD-99-32, Executive Guide: Leading Practices in Capital
Decision-Making (December 1998):
[11] It is not beyond the realm of possibility that there could be
additional factors that may bear on this question beyond those
identified in Circular A-11, although DOE believes the Circular's
current factors are appropriate and work well. The Draft Report,
however, makes no attempt to delineate what such underlying factors
might be, or how they should be viewed for scorekeeping purposes. DOE
believes it is important that any list of factors be few in number, be
objective, and be easily understood and administered, as is the case
now. In any event, DOE believes the reality of the ORNL transaction was
well captured by the current list of factors, and that it is in fact
properly understood as an operating lease.
[12] The Draft Report cites Standard and Poor's A+ rating of the ORNL
bond issuance, pointing to "a strong lease revenue stream from DOE, for
a period of up to 25 years," that "would be pledged as security for the
payment of the bonds." Draft Report at 17 (emphasis added). In fact,
DOE did not pledge this stream and was not a party to the issuance of
the bonds. Rather, UT-Battelle, UTBDC, and Keenan Developers executed
an assignment of rents they might receive to the Bond Trustee, but
those rents remain subject to the one-year termination clause, to the
extent they have as their source potential payments from DOE. The Draft
Report also fails to note the participation of independent, outside
directors on the Board of the Development Corporation and in the chart
on page 31, appears to incorrectly suggest that the Development
Corporation supplied financing for this project directly to DOE.
[13] Executive Order 13123 (June 8, 1998) "Federal Efficient Energy
Management". The federal government is mandated to achieve energy
savings performance requirements for federal buildings. 42 U.S.C. §
8253. Even higher energy conservation goals for federal buildings were
imposed under Executive Order 13123.
[14] Pub. L. No. 99-272 (1985), codified at 42 U.S.C. § 8287.
[15] H.Rep. No. 99-453, 99TH Cong., 15th Sess. at 441-443 (1985) (Comm.
On Conference). The Conferees stated that this mechanism would:
"[A]llow the Federal government to enter into long term contracts, not
to exceed twenty-five years, without the necessity of obligating the
total cost of the contract, including termination cost at the time of
contract award." Id. (emphasis added)
[16] Section 155 of Pub. L. No 102-486 (1992) (Energy Policy Act of
1992) codified at 42 U.S.C. § 8287.
[17] The 1985 Act referred to the contracts as "Shared Energy Savings
Contracts." The 1992 amendments changed the name of these contracts to
"Energy Savings Performance Contracts."
[18] H.Rep. No 102-1018, 385 (1992) (conference committee), reprinted
in 1992 U.S.C.C.A.N. 2476:
[19] Pub. L. No______, [not yet signed by President]
[20] Section 1090(f) of Pub. L. No___, [not yet signed by President]
"Not later than 180 days after the date of the enactment of this Act,
the Secretary of Energy shall complete a review of the Energy Savings
Performance Contract program to identify statutory, regulatory, and
administrative obstacles that prevent Federal agencies from fully
utilizing the program. In addition, this review shall identify all
areas for increasing program flexibility and effectiveness, including
audit and measurement verification requirements, accounting for energy
use in determining savings, contracting requirements, including the
identification of additional qualified contractors, and energy
efficiency services covered. The Secretary shall report these findings
to Congress and shall implement identified administrative and
regulatory changes to increase program flexibility and effectiveness to
the extent that such changes are consistent with statutory authority."
[21] GAO-04-119T (Oct. 1, 2003) "Federal Real Property - Actions Needed
to Address Long-standing and Complex Problems"
The following are GAO's comments on the Department of Energy's letter
dated October 25, 2004.
GAO's Comments:
1. Our report does not state that the 1967 Commission on Budget
Concepts explicitly addressed operating leases such as those at Oak
Ridge. As stated, the budgetary principle advanced by the 1967
Commission on Budget Concepts is that the federal budget should be as
comprehensive as possible; that is, all activities of the federal
government should be shown within a unified budget. GAO has long
supported an inclusive budget that discloses up-front the full
commitments of the government.
2. Ensuring that the full commitment of the government is recognized in
the budget will provide greater transparency for effective
congressional and public oversight. Moreover, it ensures that decision
makers have the information needed to make the trade-offs inherent in
allocating resources among competing demands. Further, this report
recognizes the difficulty in ensuring the validity of agencies' long-
term plans, but scoring that is based on the substance of a transaction
could result in a better reflection of the government's full
commitment.
3. It is not the intent of this report to discourage or to eliminate
energy conservation efforts or partnerships with the private sector.
Given recent congressional action to extend ESPC authority through
fiscal year 2006, we have revised our draft to recommend that OMB
require, and suggest that Congress consider requiring agencies that use
ESPCs to present to Congress an analysis comparing total contract cycle
costs of ESPCs entered into during the fiscal year with estimated up-
front funding costs for the same ECMs. However, recognizing the full
commitment up-front in the budget enhances transparency and enables
decision makers to make appropriate resource allocation choices among
competing demands that all have their full costs recorded in the
budget.
4. In our August 2003 report (Budget Issues: Alternative Approaches to
Finance Federal Capital, GAO-03-1011) we identified 10 capital
financing approaches that have been used by federal agencies to finance
capital. Subsequently, as requested by the Senate Chairman, Committee
on the Budget, we analyzed in greater detail two examples of these
alternative approaches: public/private partnerships and Energy Savings
Performance Contracts. Although this report includes our findings both
on ESPCs and partnerships, our analysis of these two financing
mechanisms was prepared separately and considered the unique
circumstances of each case study.
5. Financing asset acquisition through the United States Treasury
always has a lower interest cost than third-party financing. We looked
only at acquisitions because we recognize that if the need for an asset
is short-term, the government would not need to acquire it. Also, see
comment 2.
6. In a constrained budget environment, agencies need to prioritize
their projects needing resources and request funds for those of the
highest priority. The excerpt DOE cites from our 1998 Executive Guide
recognizes--as we do on page 2 of this report--that from an agency's
point of view the ability to record acquisition costs of a capital
asset over the life of that asset can be very attractive. However, from
the point of view of the government as a whole, these provisions may
increase costs. It is the Congress' role to allocate resources across
agencies. The same paragraph cited by DOE further states, "some
strategies currently exist at the federal level that allow agencies a
certain amount of flexibility in funding capital projects without a
loss of fiscal control. These strategies include budgeting for stand-
alone stages…."
7. Our report does not suggest excluding the status quo from
consideration when agencies evaluate the full range of alternatives in
business case analyses. Our focus is on acquisition costs and whether
or not alternative financing arrangements increase or decrease the
total cost of capital acquisition. We do not question agencies'
decisions to acquire assets and assume that the same assets would be
acquired regardless of how they are financed.
8. DOE and VA officials have stated that lower labor costs and fewer
bureaucratic requirements could make partnership financing overall less
expensive than financing through full, up-front appropriations. Despite
this assertion officials were not able to provide documentation to
support these claims. DOE contractors did provide a cost-benefit
analyses for the financing arrangement at ORNL; however, it was
inconclusive because the analyses compared private financing versus
receiving federal appropriations over a 10-year period and did not
compare receiving full, up-front appropriations.
9. We commend DOE on drafting a policy that will require a business
case analysis for public/private partnerships.
10. As we have testified and reported in the past,[Footnote 120] we
believe the budget should reflect the full commitment of the
government, considering the substance of all underlying agreements,
when third-party financing is employed. According to OMB staff, some of
the partnerships we reviewed may have been scored differently under the
revised A-11 guidelines. However, even given the 2003 revisions, we
believe the scorekeeping rules should continue to be refined to ensure
that the full commitment of the government is considered in the budget.
In its comments on our draft report, OMB agreed in concept with this
recommendation and stated that reflecting the full commitment of the
government has always been its goal.
In our conclusions, we discuss agencies' long-term capital plans as
indicative of their long-term capital requirements and as useful in
determining the substance of underlying agreements to obtain capital.
We recommend that the scorekeepers develop rules that would include
consideration of these plans. We recognize that ensuring the validity
of such plans may pose implementation challenges such as the need to
validate agencies' long-term capital requirements.
11. The report has been changed to indicate the partnership was scored
according to OMB's interpretation.
12. We recognize that DOE did not purchase the three privately financed
buildings at ORNL. However, the buildings were clearly constructed for
DOE's benefit and we do not believe it likely DOE would abandon these
state-of-the art buildings to reoccupy the currently dilapidated
buildings.
13. See comments 3 and 10. We disagree with DOE about whether only the
legal commitment should be reflected in the budget or whether the
underlying substance of the deal should be reflected.
14. As discussed throughout our report, we believe up-front recognition
of the full cost enhances budget transparency. Further, we believe the
specific standards to be incorporated in any change in scorekeeping
guidelines would most appropriately be established by the scorekeepers.
Consideration of the substance of all underlying agreements should be
part of any specific standards, as we recommend.
15. GAO suggests that agencies' long-term capital plans be considered
in determining the substance of the underlying agreement.
16. We fully understand that the ORNL transactions were deliberately
structured to be considered operating leases. As is clear from DOE's
earlier comments, we simply disagree with DOE about whether only the
legal commitment should be reflected in the budget or the underlying
substance of the deal. During our review DOE and UT-Battelle, LLC,
officials reviewed and agreed with our description of the ORNL
transaction included as figure 12, "Partnerships and Financing of
ORNL's Revitalization."
17. Standard and Poor's A+ bond rating analysis explicitly discussed "a
strong lease revenue stream" from DOE for a period up to 25 years and
that the trustee would have a valid security interest in the rent
stream." Therefore, the private sector clearly viewed this as a long-
term commitment by DOE.
18. See comment 16. Also, neither DOE nor its contractors provided GAO
with documentation to support its assertion that reserve funds were
established by the private sector Bond Trustee to cover any shortfall
in rent beyond the 1-year period in the event of termination.
Additionally, Standard and Poor's bond rating did not state reserve
funds were established to cover any shortfall in rent beyond the 1-year
period in the event of termination. Rather, it cites a "debt service
reserve fund equal to one month's base rent."
19. See comments 14 and 15.
20. Our report did not seek to analyze ESPCs from a traditional
government contracts perspective. Our report does analyze ESPCs and
partnerships from a budget scorekeeping perspective. Also, see comment
3.
21. Whether or not Standard and Poor's was correct in citing that DOE
had pledged a lease revenue stream as security for the payment of the
bonds, such a statement could have affected bond investors' decisions
to purchase. Furthermore, clearly the private sector considered the
substance of the underlying agreements in making this assessment. We
have clarified in figure 7 that the Development Corporation arranged
for the private financing which, as shown in figure 12, was secured by
Keenan Development Associates of TN, LLC, through Banc of America.
22. All of our six ESPC case studies paid more to obtain energy
conservation measures through ESPCs than they would have paid through
full, up-front appropriations. Given FEMP's assertions that 18 federal
agencies and departments have implemented ESPC projects worth $1.7
billion, we believe that "no cost to federal customers" is a misnomer.
Furthermore, agencies do acquire assets through ESPCs. As stated in
FEMP guidance, ownership of the asset usually transfers from the ESCO
to the agency when the equipment has been installed and accepted and
after initial confirmation of the guaranteed savings. Nonetheless,
given recent congressional action to extend ESPC authority through
fiscal year 2006, we have revised our draft such that ESPCs have been
deleted from the first recommendation. Instead, we have suggested that
Congress should consider requiring agencies that use ESPCs to present
to Congress an analysis comparing total contract cycle costs of ESPCs
entered into during the fiscal year with estimated up-front funding
costs for the same ECMs.
23. Given the mandates to reduce energy consumption, we do not believe
the ability of the government to invest in needed capital improvements
would be crippled by the requirement to recognize costs up front in the
budget. Congress can decide what constitutes priority claims on
resources. Also see comment 3 and 22.
24. We acknowledge that waiting for funds to be appropriated may result
in opportunity costs. However, DOE did not provide sufficient
documentation to support its assertion that waiting for appropriations
before proceeding with the ESPC process will cause serious, costly, and
irreparable harm to federal energy and infrastructure goals. As stated
in our report, we recommend to the heads of case study agencies,
including DOE, that business case analyses be performed and that the
full range of funding alternatives be analyzed. Such an analysis could
include the effect of not obtaining timely appropriations.
25. Our report does not state that DOE acquired any of the assets that
were the subject of the Oak Ridge transaction, nor did our analyses
look at UT-Battelle's modernization program implemented through the
State of Tennessee. Rather our report states that DOE used existing law
to structure a partnership that enabled it to obtain the long-term use
of facilities that was arranged through private financing.
26. Our report takes a broad definition of partnerships. Also, we
specifically identified the transactions at Oak Ridge National
Laboratory as an example of a public/private partnership in our report,
Budget Issues: Alternative Approaches to Finance Federal Capital, GAO-
03-1011 (Washington, D.C.: Aug. 21, 2003), pages 5, 48, and 52-53.
27. We clarified our report to remove any implication that the 75
percent criteria for operating leases was violated.
28. We disagree. Our draft report states the financing approaches used
in many of the case studies were structured to include features that do
not require up-front budget recognition even though they established
long-term commitments of the government. As stated in our report, UTBDC
implemented subleases of three facilities to UT-Battelle, LLC, for
DOE's ultimate use, each with a lease term of up to 25 years.
29. According to a UT-Battelle, LLC, official, UTBDC was created for
the purpose of securing private financing. Thus, our report does assert
that the ORNL transaction was undertaken to obtain private financing.
We do not see this as inconsistent with DOE's comment that the
"transaction was undertaken…as a means by which DOE could obtain
something of value to it--the use of a new building."
30. References to DOE legal opinions in our draft are based on
documentation given to us by the DOE Chief Counsel in Oak Ridge based
on the relevance to the Oak Ridge National Laboratory public/private
partnership. All statements in our report are within the context of how
the citations were written. We clarified the author of the legal
opinion and the subject of the memo.
31. Our report does not say that a conflict of interest and potential
for fraud or wrongdoing existed in the Oak Ridge National Laboratory
partnership. Rather, our report states that partnerships require
monitoring because of the complicated relationships involved.
32. Our report states we used a case study approach and notes that this
does not allow us to generalize our findings across the government. To
analyze ESPC costs, we reviewed the delivery orders, given to us by GSA
and the Navy, for each of our six ESPC case studies. In the course of
our audit work we reviewed the ORNL study (Oak Ridge National
Laboratory, Evaluation of Federal Energy Savings Performance
Contracting-Methodology For Comparing Processes and Costs of ESPC and
Appropriations-Funded Energy Projects, March 2003), interviewed the
authors of the study, and talked with agency officials about the
study's methodology. Based on our analyses we found two major flaws in
the study: (1) as agreed with the study authors the sample size was too
small and was not applicable to the entire federal sector and (2) the
study compares the costs and savings across various types of ESPCs
installed in several different federal facilities, making it difficult
to compare energy savings because the savings would depend upon too
many unpredictable factors. Also, as discussed on page 30, we did
discuss with GSA and Navy officials their historical funding
experiences.
33. It is the executive branch's long-standing position that the levels
of internal executive branch funding requests are predecisional
deliberative documents and therefore unavailable to us.
34. We acknowledge in our report that officials we spoke with said they
believed M&V results in higher sustained savings. In this report, we
take no position on whether M&V should be purchased, but agency
officials said that measurement and verification is an expense that
would not be incurred if the energy conservation measures were acquired
through full, up-front appropriations. Additionally, representatives
from energy service companies said that their private sector clients do
not always purchase M&V and verification, and if they do, it is for a
shorter period than contracts secured by the government. The lack of
M&V being purchased in the private sector suggests it is worth
exploring whether the amount of M&V purchased is a necessary expense
for the government to incur whether the project is directly funded or
obtained with an ESPC.
35. We include M&V in ESPCs costs because they are a required component
of ESPCs to demonstrate that annual savings generated by ECMs meet or
exceed contract payments. However, we exclude M&V from our proxy
estimate of the cost if the ECMs were acquired through timely, full,
and up-front appropriations because agency officials told us that M&V
is an expense that would not be incurred if the ECMs were acquired
through timely, full, and up-front appropriations. See also comment 34.
36. While it may be that ESPCs do not result in an increase in
agencies' utility bills, the ECMs acquired by all of our six ESPC case
studies were more expensive than if ECMs had been acquired through
timely, full, up-front appropriations. Thus, we do not consider ESPCs
to be budget neutral over the long-term.
37. Our analyses were based on data obtained from final awarded
delivery orders from case study contract files given to us by the Navy
and GSA and was reviewed as part of technical comments on the
appendixes by both the Navy and GSA.
38. It is unclear to us how DOE derived the "maximum possible percent
increase (PV) financed through ESPCs over cost if direct funded." Using
the data contained in the case studies' delivery orders, we attempted
to replicate DOE's "best case assumptions" by subtracting M&V costs out
of the comparison. We found that M&V represented a relatively small
percentage of the ESPC contract cycle costs and thus did not
significantly affect the increase in the costs attributed to ESPC
financing versus timely, full, and up-front appropriations. For
example, in the case of the federal courthouse in Gulfport,
Mississippi, removing M&V costs from the comparison caused the
percentage difference to decline from 56 percent to 51 percent, not to
26 percent as DOE suggests. The 51 percent difference reflects interest
costs that GSA must pay to the ESCO over the course of the contract.
39. Funds for ESPCs are obligated on an annual basis; therefore, the
budget does not recognize the government's full long-term commitment up
front, when the decisions are made.
40. Since we used the delivery order to derive our proxy estimate, any
avoided costs that were counted as a saving in that delivery were also
counted as a saving in our proxy estimate. Our analysis assumes
agencies acquire the same assets and avoid the same costs regardless of
funding approach. Also, the savings from the avoided costs were used to
make up-front payments in the contracts.
41. We do not contest agencies' use of simplified assumptions in M&V
strategies. Rather, our concern is focused on the statement contained
in FEMP's July 2004 Super ESPC Agency Project Binder. On page 6 of the
chapter entitled, "Introduction to M&V for Super ESPC Projects" it
says, "In the event that the stipulated values overstate the savings or
reductions in use decrease the savings, the agency must still pay the
ESCO for the agreed-upon savings." It does not discuss other remedies
against the ESCO nor does DOE's comment elaborate on what these
remedies might include.
42. Our report acknowledges agencies' past experiences on page 33. Also
see comment 32.
43. Large buy-downs indicate the availability of funds in the first
year of the contract and so imply opportunities exist to acquire ECMs
in smaller, useful segments, when technically feasible, with full, up-
front appropriations instead of through ESPCs. Navy and GSA officials
indicated to us that they typically did not consider financing ECMs
through useful segments before deciding to use ESPCs. Moreover, we did
not include the up-front payments made by Navy Region Southwest or
Naval Submarine Base Bangor in this statement because these payments
were made from an unexpected federal appropriation, which will not
likely occur again.
44. As stated in our draft report agencies did not request full, up-
front appropriations for the case studies reviewed in our report. Thus,
it cannot be known how much might have been needed to develop surveys
and studies to define projects in order to request direct
appropriations. Business case analyses are well accepted as a leading
practice among public and private entities. OMB requires all executive
branch agencies to prepare business case analyses for major investments
as part of their budget submissions to OMB. Also see comment 32.
45. See comments 24, 34, and 35.
[End of section]
Appendix VI: Comments from the General Services Administration:
GSA:
GSA Administrator:
November 5, 2004:
The Honorable David M. Walker:
Comptroller General of the United States:
Government Accountability Office:
Washington, DC 20548:
Dear Mr. Walker:
The General Services Administration (GSA) appreciates this opportunity
to submit agency comments on the Government Accountability Office (GAO)
"Draft Report to the Chairman, Committee on the Budget, U.S. Senate,
Partnerships and Energy Savings Performance Contracts Raise Budgeting
and Monitoring Concerns," GAO-05-55 (Draft Report).
Energy Savings Performance Contracts (ESPC's) are an indispensable way
for GSA to meet the aggressive energy goals set forth in Public Law
102-486 and Executive Order 13123, "Greening the Government through
Efficient Energy Management."
Since GSA does not normally engage in any other type of public-private
partnership, we will comment only with respect to ESPC's. We will
address GAO's recommendation that Federal agencies undertake business
case analyses before making capital financing decisions, and then we
will address two of GAO's findings.
1. Business Case Analyses:
GAO recommends in its draft report that "the Secretaries of Energy, VA,
the Navy and the GSA Administrator perform business case analyses and
ensure that the full range of funding alternatives, including the
technical feasibility of useful segments, are analyzed when making
capital financing decisions."
GSA's policy is to perform the business case analysis required by the
Office of Management and Budget (OMB). OMB's July 25, 1998, memorandum
"Federal Use of Energy Savings Performance Contracting" directs that
factors such as whether the agency has a long-term need for the
building; best value has been realized through competitive contracting
procedures; and the ESPC agreement guarantees that the minimum savings
to be generated by the improvements will cover the full cost of the
Federal investment should be considered when deciding whether to use an
ESPC. GSA has promulgated supplemental internal procedures to be
followed when entering into an ESPC in its July 2, 1999, memorandum
"Alternative Financing of Energy Savings Projects." These procedures
require the performance of a life cycle cost analysis as part of the
ESPC evaluation process.
GSA does not, however, always consider the full range of funding
alternatives for energy conservation measures concurrently. GSA seeks
appropriated funding for energy conservation projects before ESPC
financing is considered. If an energy conservation measure is part of a
larger whole-building modernization, it is considered for line-item
prospectus-level repair and alterations funding. Funds are allocated to
line-item prospectus-level projects by the National Office of Real
Property Asset Management based on the business merits of the entire
project, not just the energy conservation components of the project.
Alternatively, an energy conservation measure may be submitted for
funding through GSA's line-item energy appropriation. Projects funded
in this manner are selected by the Energy Center in the National Office
Expert Services Division based on criteria established by the Energy
Policy Act of 1992, including simple payback, savings to investment
ratio, and life-cycle cost. [NOTE 1] Energy conservation projects that
do not receive appropriations either as part of a larger line-item
prospectus level project or from GSA's line-item energy appropriation
are generally considered for financing through an ESPC.
II. Additional Comments:
GSA has the following two additional comments on the draft report.
A. GAO suggests that opportunities may exist to acquire energy
conservation measures in smaller segments using direct appropriations
rather than third party financing. Often, energy conservation measures
cannot be undertaken as stand-alone projects because they are
synergistic. For instance, chiller capacity and the "heat load"
generated by lighting and other equipment are interdependent. We will,
nevertheless, revise our energy conservation project evaluation process
to include consideration of whether components of a larger project
would reduce energy consumption even if the other components were never
completed.
B. GAO states on page 22 of its report that "contract cycle energy
costs savings specified by the contractor did not fully cover total
contract cycle costs because agencies made up-front payments." We would
like to clarify that, aside from ancillary up-front costs, such as
asbestos abatement, that must be incurred in order to carry out the
project at all, contractors for all of GSA's ESPC projects guarantee
that project savings, including additional operations and maintenance
costs, repair costs, and parts replacement costs that GSA would have
incurred had the old equipment remained in place, will meet or exceed
project costs during the contract term. GSA enforces these guarantees.
Thank you for the opportunity to comment on the draft report. Should
you have any questions, please contact me. Staff inquiries may be
directed to Mr. Mark Ewing, Office of Applied Science, Expert Services
Division, Public Buildings Service, at (202) 708-9296.
Sincerely,
Signed by:
Stephen A. Perry,
Administrator:
cc: Susan J. Irving:
Director, Federal Budget Analysis
Strategic Issues:
NOTE:
[1] Recently, line-item energy appropriations have been limited: as GAO
notes in its draft report, GSA's budget authority for energy efficiency
projects declined from $20 million in fiscal year 1999 to $4.2 million
in fiscal year 2004, and it received no funds in fiscal years 2002 and
2003.
The following are GAO's comments on the General Services
Administration's letter dated November 5, 2004.
GAO's Comments:
1. While we recognize GSA's current procedures to perform life-cycle
cost analyses as part of its ESPC evaluation process, life-cycle
costing is only one aspect of a business case analysis. Both OMB's
guidance and our Executive Guide to Leading Practices in Capital
Decision-Making stress the importance of alternatives analysis as
another component of building a business case. Such an analysis would
consider the full range of funding alternatives. GSA does not analyze
the full range of funding alternatives and therefore has an incomplete
business case analysis.
2. We asked GSA staff in Atlanta whether GSA had requested
appropriations for any of the case study ESPC projects we reviewed and
were told that the field office had not submitted a request to
headquarters because, given the $6 billion backlog of repairs and
alterations needed, the field office considered it unlikely that such
funding would be approved. At headquarters, GSA budget officials told
us that they do not specifically request funds up-front for ECMs
because they are financed over time through ESPCs.
3. We recognize that it is not always possible to undertake energy
conservation measures as stand-alone projects. Accordingly, our
recommendation asks that the technical feasibility of useful segments
be considered when making capital financing decisions. We commend GSA's
decision to revise its energy conservation project evaluation process
to include consideration of useful segments.
4. We agree that ESPC delivery orders are written to specify that
project savings meet or exceed financed costs. However, the ancillary
up-front costs also are specifically included in the contract and thus
are a part of total contract cycle costs.
[End of section]
Appendix VII: Comments from Veterans Affairs:
THE SECRETARY OF VETERANS AFFAIRS:
WASHINGTON:
October 25, 2004:
Susan J. Irving:
Director:
Strategic Issues:
U. S. Government Accountability Office:
441 G Street, NW:
Washington, DC 20548:
Dear Ms. Irving:
The Department of Veterans Affairs (VA) has reviewed your draft report
CAPITAL FINANCING: Partnerships and Energy Savings Performance
Contracts Raise Budgeting and Monitoring Concerns, (GAO-05-55) and has
several comments regarding the conclusions and recommendations. VA
disagrees with the report's conclusions and recommendation to the
Office of Management and Budget (OMB). Additional analysis should be
undertaken to fully appreciate the contribution these asset management
programs make to VA's mission and to the delivery of services to our
Nation's veterans.
Implementation of the report's scoring recommendation to OMB would
limit, discourage, and possibly eliminate the enhanced-used (EU) lease
and energy savings performance programs. The demonstrated benefits of
these programs and resulting services for veterans would be lost. As
you know, VA uses EUs to transform underutilized or unused assets into
services for our Nation's veterans, such as transitional housing,
assisted living centers, and co-location of benefits offices with
medical centers.
VA's use of state-of-the-art alternative financing programs and
structures minimizes the impact on the Department's budget while
achieving the infrastructure and programs that enhance VA's mission and
maximize health care and benefits to veterans. Again, implementing the
recommendation regarding additional scoring threatens to eliminate this
enhancement to the veterans programs.
The enclosure provides additional comments to the draft report and
otherwise addresses GAO's conclusions and recommendations. VA
appreciates the opportunity to comment on your draft report.
Sincerely yours,
Signed by:
Anthony J. Principi:
Enclosure:
The Department of Veterans Affairs (VA) Comments on the Government
Accountability Office's (GAO) Draft Report: CAPITAL FINANCING:
Partnerships and Energy Savings Performance Contracts Raise Budgeting
and Monitoring Concerns (GAO-05-55):
GAO recommends that the Director of OMB work with the scorekeepers to
develop a scorekeeping rule for the acquisition of capital assets to
ensure that the budget reflects the full commitment of the government,
considering the substance of all underlying agreements, when third
party financing is employed.
Comments - OMB has scoring rules in place regarding Energy Savings
Performance Contracts (ESPC) and partnerships. These scoring rules were
recently updated to capture the different types of alternative
financing being used across the federal sector. All significant
projects (those with a total net present value over $7M) are reviewed
by OMB for compliance with this scorekeeping. The recent changes
related to public/private partnerships are set forth to protect the
government's interests, thus offsetting the need for full scoring of a
project before its benefits and purpose become useful. Changing the
scoring rules again would have a negative impact on all agencies
seeking to use alternative financing mechanisms. Additionally, the
private sector could view more changes to the scoring rules as
increased risk. The end result could be discouraging private sector
financing by creating more uncertainty in the private capital markets.
In effect, the recommended changes would make alternative financing to
the federal government unavailable.
GAO recommends the Secretaries of Energy, VA, and the Navy and the GSA
Administrator perform business case analyses and ensure that the full
range of funding alternatives, including the technical feasibility of
useful segments, are analyzed when making capital financing decisions.
Concur - VA has in place a process to evaluate thoroughly business case
and alternatives analyses for above-threshold capital investments. The
Department's Capital Investment Program incorporates these, as well as
risk, cost effectiveness, and earned value analyses and provides a
useful tool to gauge a project's validity. The Office of Asset
Enterprise Management is the lead office for both cost and technical
implications pertaining to the Capital Investment Program for both ESPC
and EU partnerships.
Additional Comments:
GAO's analysis was confined to the government's cost and was not a
cost-benefit analysis.
Comment - GAO states that it did not use a cost-benefit analysis
approach when considering its recommendation that scoring of all ESPCs
and partnerships should be performed for all projects. It is unclear to
VA how GAO can make such a recommendation without factual basis (i.e.,
a lifecycle cost analysis) to substantiate the report's findings,
conclusions, and recommendations. Without a lifecycle cost analysis,
there is no real assessment of the overall true economic value and
long-term savings associated with these projects, whether financed
through alternative means or funded through appropriations. A lifecycle
cost analysis would show that a long-term alternative financing
scenario has a lesser cost impact to the budget than a single, 1-year
scoring of a project. Significant savings in the operation of these
alternative financing projects would be realized and included in the
overall lifecycle cost analysis. With notification to Congress, as
described in the response below, this ensures that budgetary decisions
are made in concert with the Office of Management and Budget (OMB) to
achieve VA's objectives and mission.
ESPC commitments are not fully recognized up-front in the budget.
Comment - VA disagrees with this statement. Throughout the draft
report, GAO makes the assertion that information regarding total
costing of projects for ESPCs and partnerships was invisible in the
budget process. This finding/conclusion is erroneous. The case studies
that GAO cited as part of its research on the enhanced-use leasing
process suggest that this assertion is erroneous. Inherent in VA's
enhanced-use leasing process were two notifications to congressional
oversight committees for all projects. In these notifications (the
first being a 90-day waiting period, and the second being a 30-day
period), there is a full disclosure of the projects being considered
and the budgetary impacts for both the short and long term. Congress
has the opportunity at those two instances to inform VA that these
projects should not be pursued through enhanced-use leases, or should
be inserted into the Department's budget and receive appropriated funds
for accomplishment. Likewise, before a contract is finalized for ESPCs
that total over $10M, Congress is again notified for a period of 30
calendar days of the budgetary impact of the project for the short and
long term. Again, Congress has the opportunity to inform VA to include
those projects in its annual budget. Also, starting with the FY 2004 VA
budget submission, all enhanced-use leasing projects are identified by
line item. In fact, VA had previously submitted several projects
through the budget process, requesting appropriated funds, but the
requests were disapproved --with the direction to pursue enhanced-use
leasing. There has been absolutely no hiding of any budgetary issues
from Congress. In addition, both VA's appropriation and authorization
committees are fully aware of VA's enhanced-use lease program and its
projects. Congress has continued to encourage VA to increase our use of
enhanced-use leasing. In addition, VA reports these activities on its
financial statements and is upgrading this reporting for FY 2004 to
increase the budget transparency of these activities.
Partnerships indicate a long-term commitment by the government.
Comment - VA disagrees with this statement. GAO asserts that regardless
of how partnerships were structured, they had features that indicate a
long-term commitment by the government. This statement inaccurately
describes the financial and management structure used for VA's
enhanced-use lease projects. The transactions are structured relative
to a performance-based contract. If the prescribed performance is not
achieved, VA has the right to default the lease without future payment.
Furthermore, if VA should decide to cease operations, the payments are
not required. VA is committing itself to 2-year agreements for the use
of space or the purchase of energy products, but VA is not committing
or implying a long-term commitment to those purchases. This approach
allows VA the flexibility to adjust its infrastructure to meet changing
workload requirements. The approach also provides VA with significantly
more flexibility to react to market changes than traditional
government-funded direct appropriated projects.
The following are GAO's comments on the Veterans Affair's letter dated
October 25, 2004.
GAO's Comments:
1. Given recent congressional action to extend ESPC authority through
fiscal year 2006, we have revised our draft to recommend that OMB
require, and suggest that Congress should consider requiring agencies
that use ESPCs to present to Congress an analysis comparing total
contract cycle costs of ESPCs entered into during the fiscal year with
estimated up-front funding costs for the same ECMs.
We have better emphasized and clarified in the report that OMB updated
and revised its instructions in 2003 to address lease-backs from
public/private partnerships. According to OMB staff, some of the
partnerships we reviewed may have been scored differently under the
revised guidelines. However, we still believe the scorekeeping rules
should continue to be refined to ensure that the full commitment of the
government is considered in the budget.
2. We have added language to clarify further that our report looked at
the government's cost of acquiring assets. Evaluating the benefits of
the assets was not one of our objectives. We assume that the same
assets would be acquired regardless of how they are financed and thus
they would have identical benefits and operating costs. Given our
objectives, focusing our analysis on the government's cost of financing
the assets' acquisition was the appropriate approach for this report.
Recognizing costs up front does not prohibit discussion of future
benefits when requesting appropriations.
3. The statement that ESPC commitments are not fully recognized up-
front in the budget does not refer to VA's enhanced use leases.
Further, although congressional notification is an important and
valuable process, it does not constitute recognition in the budget.
4. Our report does not state that VA or other agencies included in our
review have deliberately hidden budgetary information from Congress.
Nor do we dispute that Congress has continued to encourage VA's use of
enhanced-use leasing. Clearly, enhanced-use leases were explicitly
authorized by law. Rather, when these leases are structured such that
developed property is leased-back in short-term increments, OMB's
interpretation of the budget scoring rules permitted only the short-
term costs associated with these assets to be scored in the budget.
Finally, with respect to prior requests for appropriated funds, VA
officials explained to us that while requests had been submitted for
regional offices, requests had not been submitted for the Atlanta
regional office or other case studies in our review.
5. The lease-back agreements we reviewed had features that indicated a
long-term commitment by the government. They were structured such that
the government's legal commitment was confined to short-term periods.
Accordingly, OMB's interpretation of the budget scorekeeping rules
required that only these short-term costs to be recognized up-front in
the budget. However, recording only the 2-year legal commitment
understates the likely longer term costs of the government. For
example, prior to the enhanced-use lease used to develop a collocated
regional office in Atlanta, the regional office had occupied offices in
that area for over 25 years. While it is certainly possible that VA may
choose to discontinue operations in the Atlanta area, in our opinion
reflecting zero cost beyond the 2-year legal commitment overstates the
chances of this occurring.
[End of section]
Appendix VIII: GAO Contact and Staff Acknowledgments:
GAO Contact:
Christine Bonham, (202) 512-9576:
Acknowledgments:
In addition to the contact person named above, Carol Henn, Maria
Edelstein, Sandra Beattie, Dewi Djunaidy, Scott Farrow, Hannah Laufe,
David Nicholson, and Adam Shapiro also made significant contributions
to this report.
(450262):
FOOTNOTES
[1] GAO, Budget Issues: Budgeting for Federal Capital, GAO/AIMD-97-5
(Washington, D.C.: Nov. 12, 1996); Accrual Budgeting: Experiences of
Other Nations and Implications for the United States, GAO/AIMD-00-57
(Washington, D.C.: Feb. 18, 2000); and Fiscal Exposures: Improving the
Budgetary Focus on Long-Term Costs and Uncertainties, GAO-03-213
(Washington, D.C.: Jan. 24, 2003).
[2] For purposes of this report, capital assets exclude investments in
high technology assets, such as information technology, and assets
owned by state and local governments, such as highways.
[3] In this report, alternative financing mechanisms refer to ways of
financing capital assets other than through full, up-front
appropriations. For more information on this, see GAO, Budget Issues:
Alternative Approaches to Finance Federal Capital, GAO-03-1011
(Washington, D.C.: Aug. 21, 2003).
[4] The federal budget uses budget authority, obligations, and outlays
to measure costs.
[5] In cases that involve only noncash transactions, an agency may
never incur a monetary cost that is recognized in the federal
government's cash-and obligations-based budget.
[6] Budget authority is the authority provided by law to incur
financial obligations that will result in outlays.
[7] GAO, High-Risk Series: Federal Real Property, GAO-03-122
(Washington, D.C.: Jan. 2003).
[8] Budget scorekeeping rules or guidelines are used by the House and
Senate Budget Committees, the Congressional Budget Office (CBO), and
the Office of Management and Budget (OMB) (the scorekeepers) to measure
compliance with the Congressional Budget Act of 1974, as amended, and
the Balanced Budget and Emergency Deficit Control Act of 1985, as
amended. The purpose of the guidelines is to ensure that the
scorekeepers measure the effects of legislation on the deficit
consistent with established scorekeeping conventions and with the
specific requirements of those acts. These rules are reviewed annually
by the scorekeepers and revised as necessary to adhere to the purpose.
They cannot be changed unless all of the scorekeepers agree. In
addition, OMB publishes instructions on the budgetary treatment of
lease purchases and leases of capital assets.
[9] See figure 1 on page 12.
[10] GAO, Public Buildings: Budget Scorekeeping Prompts Difficult
Decisions, GAO/T-AIMD-GGD-94-43 (Washington, D.C.: Oct. 28, 1993) and
Federal Aircraft: Inaccurate Cost Data and Weaknesses in Fleet
Management Planning Hamper Cost Effective Operations, GAO-04-645
(Washington, D.C.: June 18, 2004).
[11] Another option would be for agencies to establish capital
acquisition funds to pursue ownership where it is advantageous, from an
economic perspective. We discussed this option in 2000. See GAO,
Federal Aircraft: Inaccurate Cost Data and Weaknesses in Fleet
Management Planning Hamper Cost Effective Operations, GAO-04-645
(Washington, D.C.: June 18, 2004). See also GAO, Accrual Budgeting:
Experiences of Other Nations and Implications for the United States,
GAO/AIMD-00-57 (Washington, D.C.: Feb. 18, 2000).
[12] GAO, Budget Issues: Alternative Approaches to Finance Capital,
GAO-03-1011 (Washington, D.C.: Aug. 21, 2003).
[13] An ESPC is a contracting method that allows a contractor to incur
the up-front costs of implementing energy savings measures, such as
lighting retrofits and ventilation systems at federal facilities, and
for the government to repay these costs over time through related
energy savings (42 U.S.C. § 8287). To streamline the procurement
process, the U.S. Department of Energy's Federal Energy Management
Program (FEMP) awarded indefinite-delivery, indefinite-quantity (IDIQ)
contracts-Super ESPCs-to a number of energy service companies (ESCO).
With these umbrella contracts in place, federal agencies can place and
implement delivery orders against the contracts in a fraction of the
time it takes to develop a stand-alone ESPC. See appendix II for
examples.
[14] Partnerships tap the capital and expertise of the private sector
to improve or redevelop federal real property assets. Partnerships are
sometimes used when excess capacity exists within an asset and existing
government facilities do not adequately satisfy the current or
potential future needs. Ideally, the partnerships are designed such
that each participant makes complementary contributions that offer
benefits to all parties. In some instances, Congress has enacted
legislation specifically authorizing partnerships (e.g., The Public
Buildings Cooperative Use Act of 1976, as amended. 40 U.S.C. § 3306).
In other circumstances, an agency may rely on its existing authorities
to enter into a partnership (e.g., DOE has its own authority to
transfer land. 42 U.S.C. § 2201(g)). See appendix III for examples.
[15] See GAO, Budget Issues: Alternative Approaches to Finance Federal
Capital, GAO-03-1011 (Washington, D.C.: Aug. 21, 2003).
[16] Since the budget measures only cashflows, the benefits with which
these costs are compared, based on policy makers' judgment, must be
presented in materials that are supplementary to the budget, in a cost-
benefit analysis. Such an analysis compares the costs and benefits of
investments, programs, or policy actions in order to determine which
alternative(s) maximize net benefits. Cost-benefit analysis attempts to
consider the net present value of costs and benefits, regardless of
whether they are reflected in market transactions.
[17] 42 U.S.C. § 8287.
[18] VA used its enhanced-use lease authority, 38 U.S.C. §§ 8161 - 8169
and DOE used its authority under the Atomic Energy Act to transfer
land, 42 U.S.C. § 2201(g).
[19] Implementing M&V strategies is required in federal ESPCs. Since
energy savings are guaranteed, the legislation requires the contractor
to verify the achievement of energy cost savings each year.
[20] GAO, Executive Guide: Leading Practices in Capital Decision-
Making, GAO/AIMD-99-32 (Washington, D.C.: Dec. 1998).
[21] In this report, contract cycle costs refer to the total costs the
government is committed to paying over the life of the contract.
[22] OMB Circular A-11 defines a useful segment as an economically and
programmatically separate component of a capital investment that
provides a measurable performance outcome for which benefits exceed the
costs, even if no further funding is appropriated.
[23] Obligations are binding agreements that result in outlays
(payments), immediately or in the future. Budgetary resources must be
available before obligations can be incurred legally.
[24] A capital lease is any lease other than a lease-purchase that does
not meet the criteria of an operating lease. Lease-purchase means a
type of lease in which ownership of the asset is transferred to the
government at or shortly after the end of the lease term. Such a lease
may or may not contain a bargain-price purchase option.
[25] See GAO, High-Risk Series: Federal Real Property, GAO-03-122
(Washington, D.C.: Jan. 2003).
[26] See appendix II for a more detailed description of ESPCs.
[27] We plan on issuing a report in 2005 to the House Committee on
Government Reform that will provide governmentwide information on the
goals, results, and issues surrounding the use of ESPCs.
[28] The Omnibus Reconciliation Act of 1985, Pub. L. No. 99-272,
amended the National Energy Conservation Policy Act, Pub. L. No. 95-
619, to authorize federal agencies to enter into ESPCs. The Energy
Policy Act of 1992, Pub. L. No. 102-486, further amended the ESPC
authority. Agencies' authority to enter into ESPCs was renewed on
October 28, 2004 in Pub. L. No. 108-375 and is scheduled to expire
October 1, 2006.
[29] 42 U.S.C. § 8253(a)(1).
[30] The legislation authorizing agencies to enter into Energy Savings
Performance Contracts authorizes multiyear contracts for up to 25 years
provided funds are available to pay for the first year of the contract.
The legislation provides that "A Federal agency may enter into a
multiyear contract . . . for a period not to exceed 25 years without
funding of cancellation charges before cancellation, if . . . funds are
available and adequate for payment of the cost of such contract for the
fiscal year . . . ." (42 U.S.C. § (a)(2)(D)(ii)). If a contract
includes a cancellation clause in excess of $10,000,000, the agency
must provide written notification to the Congress (42 U.S.C. §
(a)(2)(D)(iii)). The legislation also stipulates that an ESPC "may be
paid only from funds appropriated or otherwise made available to the
agency for fiscal year 1986 or any fiscal year thereafter for the
payment of energy expenses (and related operation and maintenance
expenses)" (42 U.S.C. § 8287a). Congress makes budget authority
available on a fiscal year basis for energy expenses from which an
agency pays the ESPC contractor based on the estimated savings to the
government. The ESPC legislation thus permits an agency to spread the
costs of the contracts over a number of years.
[31] DOD and GSA may retain and use 100 percent of all savings without
further appropriation (see 10 U.S.C. § 2685(b) and 40 U.S.C. § 592(f)).
[32] 42 U.S.C. § 8256(c)(5)(A).
[33] The relevant bills scored by CBO include H.R. 1346, 108TH Cong.
(2003); H.R. 1837, 108TH Cong. (2003); H.R. 6, 108TH Cong. (2003); H.R.
1644, 108TH Cong. (2003); S. 14, 108TH Cong. (2003), S. 2400, 108TH
Cong. (2004), and H.R. 4200, 108TH Cong. (2004), all of which would
extend agencies' authority to enter into ESPCs.
[34] See appendix III for a more detailed description of partnerships.
[35] 20 U.S.C. § 107.
[36] 40 U.S.C. § 3306.
[37] 16 U.S.C. § 470a.
[38] GAO, Executive Guide: Leading Practices in Capital Decision-
Making, GAO/AIMD-99-32 (Washington D.C.: Dec. 1, 1998).
[39] All of the partnerships case studies we reviewed were executed
before OMB's 2003 changes to its instructions on the budgetary
treatment of lease-purchases and leases of capital assets. According
to OMB staff, some of these partnerships may have been scored
differently under the revised instructions.
[40] Jacob J. Lew, Acting Director, Office of Management and Budget,
Memorandum for the Heads of Executive Departments and Establishments on
Federal Use of Energy Savings Performance Contracts, (Washington, D.C.:
July 25, 1998).
[41] See appendix III for a full discussion of this case study.
[42] A legal instrument used to release one party's right, title, or
interest to another without providing a guarantee or warranty of title.
[43] The quitclaim deed ensured the buildings would not be constructed
on government land. According to budget scoring rules, if the project
was constructed or located on government land, it will be presumed to
be a special purpose asset of the government, and thus potentially a
capital lease. However, according to OMB officials, this transfer of
land may not have been necessary, since construction on government land
is just one of several factors OMB considers when determining if an
arrangement should be scored as a capital lease or an operating lease.
[44] Budget scoring rules state that, for operating leases that include
a cancellation clause, agencies need only obligate an amount sufficient
to cover the first year's lease payments, plus cancellation costs.
Alternatively, when agencies enter into a capital lease contract or
lease-purchase, budget authority is scored in the year in which the
authority is first made available in the amount of the net present
value of the government's total estimated legal obligations over the
life of the contract.
[45] DeKalb County is located in the state of Georgia and includes a
portion of the city of Atlanta.
[46] See appendix III for a full discussion of this case study.
[47] VA officials informed us that they have changed this practice such
that future leases will require VA to take positive action to renew
rather than terminate.
[48] Termination costs are represented through a "renovation reserve
fund." To mitigate the risks to the Authority if VA reduces the amount
of space it occupies in the VARO building, VA deposited $1.8 million
into the fund upon the date of full execution of the lease. The
Authority may draw from this fund to renovate or reconfigure rental
space for new tenants should VA vacate some part of the VARO during the
term of the ground lease. VA officials said that the fund effectively
reduced VA's rent since it reduced the Authority's risk and, thus, the
amount the Authority had to borrow.
[49] See appendix III for a full discussion of this case study.
[50] See appendixes II and III for a more detailed description of ESPCs
and partnerships.
[51] Because it is difficult to predict what the true cost of the asset
would have been had it been financed differently, this is not a precise
measure.
[52] This comparison of costs represents a budget analysis, not a
broader cost-benefit analysis. The analysis takes into account the
costs incurred by the federal government, rather than total social
costs and net benefits. It also assumes that an agency could acquire
the same ECMs for the same price, regardless of how its acquisition is
financed.
[53] For the ESPC case study at Gulfport, where up-front payments
represented 2 percent of the total contract cycle costs, ECMs were
installed in a newly constructed building. Thus, savings from avoided
repair and renewal, which could be used to buy down principal, did not
exist. Also, because it was new construction, GSA and the ESCO
calculated both the baseline performance and expected savings amounts
based on a model.
[54] See figure 12 on page 81.
[55] According to a DOE official, part of the timing difference is due
to the appropriations cycle--agency requests are submitted about 2
years before appropriations are received. One official of DOE's M&O
contractor added that some of the difference is due to construction
time. He said that construction overseen by private developers using a
commercial model is faster than construction overseen by DOE because
DOE is bound by certain inefficient labor agreements and processes that
do not apply to the private sector. Both DOE and contractor officials
agreed that faster construction enables DOE to more quickly vacate
obsolete and dilapidated buildings, which are expensive to maintain.
[56] UT-Battelle, LLC prepared an analysis of these costs, however DOE
did not review this analysis. The analysis did not include an analysis
of timely, up-front appropriations as an alternative.
[57] According to GSA and Navy officials, even if they acquired ECMs
through timely, full, and up-front appropriations, they might contract
with ESCOs to obtain technical expertise on the ECMs to be installed.
However, with an ESPC, ESCOs not only provide expertise, they also
measure and verify whether guaranteed savings are met.
[58] The process for implementing an ESPC is described in greater
detail in appendix II.
[59] A typical suite of FEMP services costs agencies about $30,000.
[60] See U.S. Army Audit Agency, Energy Savings Performance Contracts:
U.S. Army Joint Readiness Training Center and Fort Polk, AA 01-471
(Alexandria, VA: Sept. 24, 2001); Energy Savings Performance Contracts:
XVIII Airborne Corps and Fort Bragg, AA-02-098 (Alexandria, VA: Dec.
14, 2001); Energy Savings Performance Contracts: U.S. Army Military
District of Washington, A-2002-0288-IMO (Alexandria, VA: July 24,
2002); and Lessons Learned from an Energy Savings Performance Contract:
U.S. Army Garrison, Alaska, A-2004-0068-FFP (Alexandria, VA: Dec. 10,
2003).
[61] A representative from one of these ESCOs said that his company
preferred that agency officials accompany them during the M&V process
to validate savings.
[62] We anticipate we will issue a report in 2005 that will further
explore this issue.
[63] UTBDC, a special purpose, nonprofit entity, was established for
the sole purpose of securing private financing of three general-use
buildings on the ORNL reservation. The transfer of land was an integral
component of the private financing plan. This arrangement is described
in detail in appendix III.
[64] See figure 12 in appendix III for an illustration of the financing
stream.
[65] GAO, Executive Guide: Leading Practices in Capital Decision-
Making, GAO/AIMD-99-32 (Washington, D.C.: Dec. 1, 1998).
[66] Two other ESPC case studies also paid a significant portion of the
total contract cycle costs in year 1. These payments stemmed from
federal funding unexpectedly made available to mitigate energy
shortages in California in fiscal year 2000.
[67] OMB Circular A-11 defines a useful segment as an economically and
programmatically separate component of a capital investment that
provides a measurable performance outcome for which the benefits exceed
the costs, even if no further funding is appropriated.
[68] Two of the five partnerships we reviewed were not done to obtain
project financing. Rather, they were made in response to an opportunity
to achieve other benefits. See summaries of VA's partnership
arrangements at Mt. Home, Tennessee, and Vancouver, Washington, in
appendix III for details.
[69] Our August 2003 report identified alternative financing approaches
based on prior GAO reports and more current research. See GAO, Budget
Issues: Alternative Approaches to Finance Federal Capital, GAO-03-1011
(Washington, D.C.: Aug. 21, 2003). While that work was not intended to
result in a comprehensive list of all capital financing approaches, we
believe we identified the major approaches used.
[70] An ESPC is a contracting method that allows a contractor to incur
the cost of implementing energy saving measures at federal facilities
with the agency repaying the contractor over time using the resulting
savings in utility costs. To streamline the procurement process, FEMP
awarded indefinite-delivery, indefinite-quantity (IDIQ) contracts--
Super ESPCs--to a number of energy service companies (ESCO). With these
umbrella contracts in place, federal agencies can place and implement
delivery orders against the contracts in a fraction of the time it
takes to develop a stand-alone ESPC. Many delivery orders have been
written against ESPCs established by other agencies, such as the Army
Corps of Engineers.
[71] We did not include modifications added to ESPCs after the delivery
orders were signed because we were most interested in decisions made at
the point the government's commitment was established.
[72] 42 U.S.C. § 8253(a)(1).
[73] 42 U.S.C. § 8287 (a)(2)(D)(ii).
[74] Although FEMP's guidance does not include the term "buy-down," it
is a term used within the industry and agencies. In this report, buy-
downs include prepayments of principal, typically resulting from
avoided renovation or maintenance of older equipment.
[75] Delivery order is a term used by FEMP in its ESPC IDIQ contracts.
It is used for agencies ordering ESPC services under FEMP IDIQ
contracts and is interchangeable with the term task order, used for
agencies ordering services under DOD IDIQ contracts. The Federal
Acquisition Regulation (FAR) defines delivery order as "an order for
supplies placed against an established contract or with Government
sources." Originally, FEMP contracting officers considered this to be
the best definition for an ESPC, so it was adopted as the term used in
the IDIQ.
[76] ESPC markup ceilings were established on a competitive basis by
FEMP. Markups average about 29 percent and include overhead, benefits,
sales, legal expenses, and profit.
[77] The detailed energy survey is the ESCO's comprehensive audit of
facilities and energy systems at the project site. It augments,
refines, and updates the preliminary site survey data and provides the
information needed to update the feasibility analyses of the various
ECMs under consideration for the project.
[78] Measurement and verification is the process by which ESCOs
determine that equipment is performing as guaranteed.
[79] The Defense Federal Acquisition Regulation requires that the
contracting officer provide all preapproved contractors a fair
opportunity to compete for the contract. See 48 CFR § 216.505-70.
[80] Energy consumption statistics are measured in British thermal
units (BTU).
[81] 42 U.S.C. § 8287 (a)(2)(D)(iii). Prior to the enactment of the
Energy Act of 2000 (Pub. L. No. 106-469), the cancellation ceiling
threshold was $750,000.
[82] According to FEMP officials, most agencies transfer title at the
acceptance of the installation and the postinstallation M&V report,
after confirmation of the guaranteed savings.
[83] Change orders involve alterations to the design of an individual
ECM.
[84] FEMP has issued M&V guidance to agencies.
[85] 42 U.S.C. § 8256(c)(5)(A).
[86] 40 U.S.C. § 592(f) and 10 U.S.C. § 2865(b) authorize GSA and the
Navy, respectively, to retain excess savings.
[87] After the delivery order was signed in 2001, a number of
modifications were made to the contract that expanded the total scope
of work. For example, a sixth ECM was added in October 2002 to install
a Compressed Air System Upgrade at the Naval Station in San Diego.
[88] These savings were realized in fiscal year 2004. As of fiscal year
2004, the Navy was not required to return excess savings to Treasury.
[89] These excess savings were realized in fiscal year 2004.
Accordingly, the Navy may retain these "excess" savings.
[90] GAO, Public-Private Partnerships: Key Elements of Federal Building
and Facility Partnerships, GAO/T-GGD-99-81 (Washington, D.C.: Apr. 29,
1999).
[91] This additional management tool has been authorized for VA, DOD,
and the National Aeronautics and Space Administration.
[92] 38 U.S.C. § 8161-8169.
[93] See figure 3 on page 18 for a list of the basic elements of an EU
lease.
[94] See page 18 for a more detailed discussion about other
authorities.
[95] GAO, Public-Private Partnerships: Factors to Consider When
Deliberating Governmental Use as a Real Property Management Tool, GAO-
02-46T (Washington, D.C.: Oct. 1, 2001).
[96] Section 161(g) of the Atomic Energy Act, codified at 42 U.S.C.
2201(g), authorizes the Secretary of Energy to "acquire, purchase,
lease, and hold real and personal property…and to sell, lease, grant,
and dispose of such real and personal property. . . ."
[97] A legal instrument used to release one party's right, title, or
interest to another without providing a guarantee or warranty of title.
[98] The three privately constructed buildings are the Computational
Sciences Building, the Research Office Building, and the Engineering
Technology Facility.
[99] The facility leases provide that UTBDC may sublease any part of
its premises, and it may assign its leases of the facilities to "an
entity other than DOE or its designee." Therefore, if DOE chooses to
terminate any of its subleases, UTBDC may sublease the property to
another organization and still ensure bond payments are covered.
[100] The subleases from UTBDC have an initial term of 10 years
followed by three 5-year renewals, for a total of 25 years. Pursuant to
the quitclaim deed, DOE reserves the right to repurchase any part of
the land conveyed for a nominal consideration, provided that no
subleases have been terminated during the 25 years.
[101] UTBDC officials stated that the Battelle Memorial Institute has
spent millions developing this financing structure as well as the
language of the bond offering. The officials stated that this
information is proprietary and provides them with a business advantage
over others competing for DOE revitalization projects at other
campuses.
[102] A business case analysis is a tool for planning and decision
making that projects the financial implications and other
organizational consequences of a proposed action. The overriding
purpose of a business case analysis is to make transparent to decision
makers all the objectives to be met by a facilities investment, the
underlying assumptions, and the attendant costs and potential
consequences of alternative actions. The overriding purpose of these
analyses is to allow decision makers to (1) see and understand all the
objectives to be met by a facilities investment and the potential
consequences of facilities investment decisions and (2) make informed
choices about owning, leasing, reinvesting in, or constructing
facilities.
[103] It is periodically argued that privately contracted construction
can be completed faster than federally contracted construction. While
we could not find any formal studies of this, there is some evidence to
support this theory. For example, although the three buildings
constructed privately were conceptualized at the same time as DOE's
highest priority construction on the ORNL reservation, the privately
constructed buildings were completed and occupied in the summer of
2003, while the federally contracted building is not scheduled for
completion until the summer of 2005. In addition, UT-Battelle, LLC,
officials provided summary data on the construction costs per square
foot for the privately constructed versus estimated government
construction, which showed that private construction costs were roughly
30 percent less expensive for comparable space.
[104] In 2002, the city of Oak Ridge appraised the transferred land for
$79,400.
[105] UTBDC assumed that full appropriations for DOE-contracted
construction would be made available over a 10-year period.
[106] See figure 1 on page 12 for the definition of an operating lease.
[107] On December 26, 2000, DOE/Oak Ridge Operations' Chief Counsel
opined that OMB Circular A-11 did not require coverage of leases
entered into by DOE contractors. However, it had been DOE's policy to
apply Circular A-11 to such leases.
[108] The Authority was created by the Georgia General Assembly for the
purpose of promoting trade, commerce, industry, and employment
opportunities for the public good and to promote the general welfare of
the state.
[109] VA officials informed us that they have changed this practice so
that future leases will require VA to take positive action to renew
rather than terminate.
[110] Thirty-five-year PV lease payments for less space in the AFC
would have totaled roughly $105 million, without the added benefit of
parking, furnishings, and equipment.
[111] VA manages the largest medical education and health professions
training program in the United States. VA facilities are affiliated
with 107 medical schools, 55 dental schools, and more than 1,200 other
schools across the country.
[112] In January 1998, appraisers estimated the value of the land at
$350,000.
[113] The Vancouver Housing Authority is a public municipal corporation
that derives its authority from Washington State Law RCW 35.82. It is
governed by a six-member Board of Commissioners appointed to staggered
5-year terms, with the exception of the Resident Commissioner who is
appointed to a 2-year term. All are appointed by the Mayor of
Vancouver, Washington, and abide by state laws governing conflicts of
interest, open public meetings, and rules of conduct for public
officials.
[114] It is generally accepted that about one-third of homeless people
are veterans.
[115] According to VA officials, VA determined there was no commercial
market supply for steam in North Chicago. VA officials also stated they
determined VA was paying above-market rates for steam by surveying the
marketplace in downtown Chicago.
[116] The Illinois Development Finance Authority issued the bonds with
the North Chicago Energy Trust as the borrower through its trustee.
[117] If the developer needed to be replaced, funds would still be
available in the trust to hire a new developer and complete the
project. This did not become an issue and the project was completed and
operational in 2003.
[118] A provision allows VA to obtain the title earlier by paying the
balance of the secured indebtedness, all reimbursement obligations, and
interest and redemption premium amounts on the bonds.
[119] The only VA investment in the EU lease is the outlease of real
property to the Owner Trust valued at $110,000.
[120] GAO, High-Risk Series: Federal Real Property, GAO-03-122
(Washington, D.C.: Jan. 2003); Public Buildings: Budget Scorekeeping
Prompts Difficult Decisions, GAO/T-AIMD-GGD-94-43 (Washington, D.C.:
Oct. 28, 1993); and Budget Issues: Budget Scorekeeping for Acquisition
of Federal Buildings, GAO/T-AIMD-94-189 (Washington, D.C.: Sept. 20,
1994).
GAO's Mission:
The Government Accountability Office, the 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.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through the Internet. GAO's Web site ( www.gao.gov ) contains
abstracts and full-text files of current reports and testimony and an
expanding archive of older products. The Web site features a search
engine to help you locate documents using key words and phrases. You
can print these documents in their entirety, including charts and other
graphics.
Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as "Today's Reports," on its
Web site daily. The list contains links to the full-text document
files. To have GAO e-mail this list to you every afternoon, go to
www.gao.gov and select "Subscribe to e-mail alerts" under the "Order
GAO Products" heading.
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, Waste, and Abuse in Federal Programs:
Contact:
Web site: www.gao.gov/fraudnet/fraudnet.htm
E-mail: fraudnet@gao.gov
Automated answering system: (800) 424-5454 or (202) 512-7470:
Public Affairs:
Jeff Nelligan, managing director,
NelliganJ@gao.gov
(202) 512-4800
U.S. Government Accountability Office,
441 G Street NW, Room 7149
Washington, D.C. 20548: