Energy Policy: Ranking Options to Improve the Readiness of and Expand the
Strategic Petroleum Reserve (Letter Report, 08/18/94, GAO/RCED-94-259).

This report discusses several near- and long-term options for improving
the readiness of and expanding the Strategic Petroleum Reserve.
Resolving problems affecting the Reserve's readiness, such as the
buildup of geothermal heat and gas in stored crude oil, and replacing
equipment at the end of its design life would overcome the significantly
degraded ability of the Reserve to respond to oil disruptions.  As a
result, GAO ranked these options as high priorities. Continuing to fill
the reserve to its current capacity of 750 million barrels or expanding
the reserve to hold one billion barrels would both cost much more.
Considering the limited potential benefits and the higher costs of
filling or expanding the Reserve, GAO gave this option a relatively low
priority.  The benefits of increasing the Reserve's daily drawdown
capability are less clear; however, implementing this option would
increase the nation's ability to respond more flexibly to oil
disruptions and would likely entail more moderate costs.  GAO ranked
this option as a medium priority.

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

 REPORTNUM:  RCED-94-259
     TITLE:  Energy Policy: Ranking Options to Improve the Readiness of 
             and Expand the Strategic Petroleum Reserve
      DATE:  08/18/94
   SUBJECT:  Petroleum storage
             Fuel supplies
             Emergency preparedness
             Oil resources
             Cost effectiveness analysis
             Crude oil
             Petroleum industry
             Petroleum refining facilities
             Energy consumption
             Fuel conservation
IDENTIFIER:  Strategic Petroleum Reserve
             Gulf Coast (LA)
             Louisiana
             Texas
             Bryan Mound (TX)
             West Hackberry (LA)
             DOE Life Extension Program
             SPR
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


Cover
================================================================ COVER


Report to the Chairman, Subcommittee on Energy and Power, Committee
on Energy and Commerce, House of Representatives

August 1994

ENERGY POLICY - RANKING OPTIONS TO
IMPROVE THE READINESS OF AND
EXPAND THE STRATEGIC PETROLEUM
RESERVE

GAO/RCED-94-259

Energy Policy--SPR Readiness and Expansion Options


Abbreviations
=============================================================== ABBREV

  DOE - Department of Energy
  EIA - Energy Information Administration
  GAO - General Accounting Office
  GNP - gross national product
  SPR - Strategic Petroleum Reserve

Letter
=============================================================== LETTER


B-257314

August 18, 1994

The Honorable Philip R.  Sharp
Chairman, Subcommittee on Energy
 and Power
Committee on Energy and Commerce
House of Representatives

Dear Mr.  Chairman: 

On May 25, 1994, we testified before the Subcommittee on the results
of the work we performed at your request on several near- and
long-term options for improving the readiness of and expanding the
Strategic Petroleum Reserve (SPR).\1 Specifically, you asked our
position on the relative priority that should be given to the options
of (1) eliminating problems related to the buildup of the heat and
gas content of some crude oil in the SPR; (2) replacing existing
facilities and systems to extend the useful life of the reserve; (3)
filling the SPR to the current 750-million-barrel capacity or
expanding and filling the SPR to 1 billion barrels; and (4)
increasing the daily drawdown rate from 4.5 million to 6 million
barrels for the SPR's current size, as well as for a 1-billion-barrel
reserve. 

In conducting our analyses, we used a Department of Energy (DOE)
model that is designed to examine the costs and benefits associated
with a variety of size and drawdown issues.  The benefits come
largely from replacing with SPR oil imported oil whose supply has
been disrupted, thereby dampening oil price increases and their
resulting impact on the nation's economy.  The model allows us to
make different assumptions about the probability and length of
disruptions, oil prices and the quantities of oil available in the
marketplace, market price elasticities, the impact of oil price
increases on the gross national product, discount rates, and other
parameters.  As agreed with your office, this report presents the
results of our analyses concerning the relative priority that should
be given to four near- and long-term options presented above for
improving the readiness and expansion of the SPR, as well as a
detailed description of the model and the related analyses of its
methodology and sensitivity that we performed.  (For more details
about the model, our assumptions, and the results of our analyses,
see app.  I.)


--------------------
\1 Energy Policy and Conservation Act Reauthorization
(GAO/T-RCED-94-214, May 25, 1994). 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

We found that resolving problems that affect the SPR's readiness,
such as the buildup of geothermal heat and gas in stored crude oil,
and replacing equipment that is at the end of its design life would
correct or prevent the significantly degraded ability of the SPR to
respond to oil disruptions.  Because of this, we ranked these options
as high priorities.  Continuing to fill the reserve to its current
capacity of 750 million barrels or expanding the reserve to hold 1
billion barrels would both entail much higher costs.  Considering the
limited potential benefits and the higher costs of filling or
expanding the SPR, we gave this option a relatively low priority. 
The benefits of increasing the SPR's daily drawdown capability are
less clear; however, implementing this option would increase the
nation's ability to respond more flexibly to oil disruptions and
would likely entail more moderate costs.  We ranked this option as a
medium priority. 


   BACKGROUND
------------------------------------------------------------ Letter :2

The SPR, which is authorized by the Energy Policy and Conservation
Act to store up to 1 billion barrels of crude oil for use during a
disruption in the oil supply, provides insurance against oil market
shocks and their potentially significant effects on the economy.  In
October 1991, DOE, which manages the SPR, completed the development
of facilities for storing 750 million barrels of crude oil in
underground Gulf Coast salt domes in Louisiana and Texas and in a
marine terminal on the Mississippi River at St.  James, Louisiana. 
DOE is working toward achieving a maximum design drawdown rate of 4.5
million barrels a day.  Currently, SPR facilities are designed to
draw down crude oil at a rate of about 4.3 million barrels a day. 
Since fiscal year 1976, the Congress has appropriated about $21
billion (or about $33 billion when adjusted for inflation to 1994
dollars) for SPR programs and activities.  As a result, almost 600
million barrels of crude oil has been stored. 

After developing a storage capacity of 750 million barrels, DOE
shifted its attention to improving the readiness of the reserve for a
drawdown of the existing inventory.  In 1992, the agency established
as top priorities ensuring this readiness and extending the useful
life of the SPR's present systems beyond the end of this decade.  DOE
has also conducted studies, analyses, and public hearings to produce
the reports mandated under the Energy Policy and Conservation Act
Amendments of 1990 (P.L.  101-383) to pick sites and complete an SPR
Plan Amendment to expand the reserve to 1 billion barrels. 

The SPR's Program Office in Washington, D.C., is responsible for
managing the overall program and planning activities to achieve the
program's goals and objectives.  The Assistant Secretary for Fossil
Energy has overall programmatic responsibility for achieving these
goals and objectives.  The Project Management Office, located in New
Orleans, Louisiana, carries out day-to-day project activities,
including the management and operation of five underground storage
sites and the one marine terminal. 


   CORRECTING PROBLEMS OF HEAT AND
   GAS BUILDUP SHOULD BE ASSIGNED
   HIGH PRIORITY
------------------------------------------------------------ Letter :3

In the SPR, both the amount of oil available for drawdown and the
total daily drawdown rates have been lessened because the temperature
of the stored crude oil has been elevated by geothermal heating, and
gas, primarily methane, from the surrounding salt formations has
mixed with some of the oil.  Both the elevated temperatures and the
high gas content diminish DOE's ability to draw down the oil because
they have raised the vapor pressure of the oil above safety and air
pollution limits.  The vapor pressure must be reduced to ensure that
crude oil from the SPR can be delivered at the proper specifications
for commercial transportation and refining. 

As early as 1984, DOE had indications that crude oil stored in the
SPR had elevated temperatures and excessive gas content.  However,
the results of early tests to determine the scope and impact of the
problems were inconclusive.  After becoming convinced in 1993 that
the problems were significant and widespread, DOE established a Vapor
Pressure Task Force--consisting of SPR personnel, various
contractors, and representatives of DOE's national laboratories--to
define the full extent of the problems and develop corrective
actions.  While tests continue, DOE now estimates that about 400
million barrels, or two-thirds of the oil, is affected.  DOE further
estimates that by blending affected and unaffected oil and observing
certain operating restrictions, it could draw about 520 million
barrels out of the reserve without further corrective actions, but
only at a maximum daily rate of about 2 million barrels per day, or
less than one-half the maximum design rate for the reserve.  Without
blending affected and unaffected oil and observing the operating
restrictions, only about 800,000 barrels could be drawn down per day,
according to DOE's estimates. 

DOE plans to permanently install heat exchangers to dissipate the
excess heat as the reserve is drawn down.  DOE intends to complete
this work by April 1995 at a cost of about $19 million.  To reduce
the gas content of the oil to acceptable levels, DOE will bring about
144 million barrels to the surface, degas it, and then return and
blend it with other oil in the caverns.  DOE plans to complete this
work by November 1997 at a cost of about $45 million. 

Using a DOE model, we estimated that the benefits of reducing the
heat and gas content could be substantial.  The net present value of
the benefits ranges from about $2.9 billion to $16.7 billion,
depending on the extent to which the reserve's drawdown capability is
restricted and which disruption scenario is evaluated.  The benefits
are substantial because the heat and gas significantly impair the
SPR's drawdown capability and therefore limit the SPR's ability to
dampen oil price increases and their economic impact.  Because of the
large potential benefits and relatively low total cost of this
action--about $64 million--we would assign it a high priority. 


   REPLACING AGING FACILITIES
   COULD AVERT FUTURE LOSS IN THE
   SPR'S CAPABILITY
------------------------------------------------------------ Letter :4

The SPR's drawdown rate is also threatened by major problems in the
mechanical, civil, and electrical systems--problems that DOE believes
will become progressively worse.  The United States has spent over
$3.9 billion (or about $6 billion when adjusted for inflation) for
SPR storage sites and related distribution systems.  These systems
were installed in the late 1970s and early 1980s, with a designed
life span of 20 years.  DOE plans to replace and upgrade the drawdown
and distribution systems through its Life Extension Program and at
the same time simplify and standardize equipment to reduce future
maintenance costs.  These changes will also enhance DOE's ability to
test equipment under maximum usage rates. 

Using DOE's evaluation of the SPR system's availability without a
Life Extension Program, we estimated that the daily drawdown rate
could drop to about 3 million barrels per day, or about 67 percent of
the system's design capability, within 10 years.  The most severe
drop would occur at the SPR's two largest sites, the Bryan Mound,
Texas, and West Hackberry, Louisiana, sites, where over 420 million
barrels of SPR oil are stored.  DOE estimates that the Life Extension
Program will take 7 years to complete and cost about $375 million (or
about $315 million in present value terms).  The program will extend
the useful life of the reserve to the year 2025. 

Our analysis shows that if the drawdown rate is lessened to about 3
million barrels per day, the net benefits of carrying out the Life
Extension Program could be as high as $1.6 billion across the various
disruption scenarios under which the reserve would be needed.  These
results do not include the additional maintenance costs of keeping
the present SPR facilities and systems operational if the life
extension projects are not done.  Given the extent to which the SPR's
capability could be degraded without the Life Extension Program and
the potential net benefits of preserving the large investment in
equipment and capability to date, we would also assign this activity
a high priority. 


   FILLING THE SPR TO ITS CURRENT
   CAPACITY OR EXPANDING THE
   CAPACITY WOULD BE COSTLY
------------------------------------------------------------ Letter :5

Because of budget constraints, DOE is not requesting any funds in
fiscal year 1995 to continue to fill the reserve from its current
level of about 600 million barrels to its capacity of 750 million
barrels.  DOE estimates that filling the reserve to capacity would
cost a total of about $4.2 billion (or $3.7 billion in present value
terms).  Also, DOE has informed the Congress that because of extreme
demands on the federal budget, the administration does not foresee
that the reserve can be expanded to the authorized 1 billion barrels
within a meaningful planning horizon.\2 DOE estimates that expanding
and filling the reserve to 1 billion barrels would cost between $10
billion and $11 billion (or between $6.4 billion and $7.1 billion in
present value terms), depending on the price of oil and the sites
selected for the expansion.  Any expansion of the reserve would also
have to address the potential for heat and gas to build up in the
expanded portion of the reserve. 

Our analysis shows that filling the reserve to its 750-million-barrel
capacity could produce net benefits of as much as $3.3 billion, which
occurs under the severe disruption scenario, in which large,
longer-lasting disruptions are more likely.  However, under a milder
disruption scenario, in which disruptions are shorter, the costs
would exceed the benefits by as much as $1.3 billion.  Because the
costs of expanding and filling the reserve to 1 billion barrels would
be very high, our analysis shows that the costs would exceed the
expected benefits under all but the most severe disruption scenario. 
Such costs would exceed the benefits by as much as $2.8 billion. 
Given the high costs of purchasing additional oil and expanding the
reserve's capacity and the relatively few disruption scenarios under
which a larger reserve would produce benefits, we would assign these
options a relatively low priority. 

Some SPR staff believe that cost savings might be possible if the
size of the SPR were officially capped at its current size of about
600 million barrels.  This cap could enable DOE to consolidate and
take out of operation unneeded sites and limit life extension
projects to only those sites remaining.  SPR staff are examining this
issue but have not yet reached any conclusions. 


--------------------
\2 DOE has conducted studies, analyses, and public hearings to select
expansion sites.  However, the agency has forwarded to the Congress a
bill to amend the Energy Policy and Conservation Act, as amended,
that will require continued planning to expand the SPR to 1 billion
barrels only when a pattern of appropriations develops that would
provide for filling the existing facilities within 5 years. 


   INCREASING THE DAILY DRAWDOWN
   CAPABILITY COULD INCREASE
   FLEXIBILITY
------------------------------------------------------------ Letter :6

Increasing the daily drawdown capability, either while increasing the
total amount of oil in the reserve or separately, would increase the
nation's ability to replace disrupted crude oil supplies and dampen
oil price increases and their impact on the economy.  DOE has
estimated the cost of increasing the maximum daily drawdown
capability from the existing 4.5 million barrels per day to 6 million
barrels per day as the reserve is expanded to 1 billion barrels. 
These estimates range from $2 million to $196 million, depending on
which sites DOE selects to expand.  DOE has not estimated the cost of
increasing the daily drawdown capability for the existing reserve
without an expansion.  SPR program officials we spoke with said,
however, that such capability could be added, and, if required, could
logically be included in the Life Extension Program.  Any increase in
drawdown capability also assumes that the problems of heat and gas
buildup are resolved. 

Our analysis shows that increasing the SPR's drawdown capability to 6
million barrels per day, for the current or an expanded reserve,
produces net benefits under most scenarios only for the expanded
reserve.  However, we cannot fully evaluate the potential advantages
of increasing the drawdown capability because of constraints in the
model that limit the amount of SPR oil that can be released at any
given time.  For example, the model will not readily allow the user
to increase, or "surge," the initial drawdown and then lower it if a
disruption seems likely to last longer than originally anticipated. 
Such a surge capability could help to meet DOE's stated intention of
quickly injecting large amounts of oil into disrupted markets to
dampen oil price increases.  The model does show that if disruptions
are relatively short, increasing the daily drawdown capability
provides more net benefits than increasing the current size of the
reserve.  Also, increasing the daily drawdown capability would likely
cost less than purchasing large quantities of additional oil or
expanding the reserve, and it would give DOE more flexibility to
respond to a wider range of disruption scenarios.  Consequently, we
would assign this option as applied to the reserve's current size a
medium priority. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :7

We discussed the facts of this report with the Associate Deputy
Assistant Secretary Strategic Petroleum Reserves and his staff and
with officials of the Oak Ridge National Laboratory who developed
DOE's DIS-RISK model.  They agreed with the facts presented and
suggested minor changes that were incorporated where appropriate. 
However, as requested, we did not obtain written agency comments on a
draft of this report. 


---------------------------------------------------------- Letter :7.1

In conducting our work, we held extensive interviews at DOE
headquarters and the SPR project office and reviewed numerous related
studies and evaluations.  In the project office, we reviewed DOE's
Vapor Pressure Task Force and its contractors' records in order to
analyze short-term and long-term corrective actions, as well as
estimates of the cost, time, and impact of DOE's proposed actions to
correct the SPR's problems of heat and gas buildup.  For the life
extension and expansion programs, we reviewed cost projections
prepared by contractors for the work DOE is proposing to undertake. 
As noted earlier, appendix I describes the DOE model we used in our
analyses to determine the relative priority of the various options
for improving the readiness of and expanding the SPR.  We performed
our work between May 1993 and May 1994 in accordance with generally
accepted government auditing standards. 

As arranged with your office, unless you publicly announce its
contents earlier, we plan no further distribution of this report
until 7 days after the date of this letter.  At that time, we will
send copies to the Secretary of Energy.  We will also make copies
available to others upon request. 

Please call me at (202) 512-3841 if you or your staff have any
questions.  Major contributors to this report are listed in appendix
II. 

Sincerely yours,

Victor S.  Rezendes
Director, Energy and
 Science Issues


DESCRIPTION OF DOE'S MODEL AND OUR
EVALUATION OF SPR READINESS AND
EXPANSION ISSUES
=========================================================== Appendix I

This appendix discusses the results of risk analyses that we
performed, using a model developed and provided to us by the
Department of Energy (DOE) to evaluate the Strategic Petroleum
Reserve's (SPR) readiness and expansion issues.  The first section
provides an overview of DOE's "DIS-RISK" model.  The second section
contains a detailed technical examination of the model's structure,
tracing the model's logic from the generation of a random oil supply
disruption to the calculation of SPR costs and benefits, given the
simplifying assumptions underlying the model.  The third section
contains our comments on some of the underlying assumptions made in
the model.  The fourth section briefly describes the SPR readiness
and expansion issues we examined.  In the fifth section, we describe
five scenarios--collections of assumptions about the nature of oil
supply disruptions and oil market responses--that we used to examine
each of these SPR issues.  The sixth section contains estimates of
the net benefits under each scenario for each readiness or expansion
issue.  In the last section, we compare the benefits and costs of
enhancing the SPR's drawdown capabilities with those of filling the
SPR. 


   OVERVIEW OF THE DIS-RISK MODEL
--------------------------------------------------------- Appendix I:1

As the basis for examining SPR issues, we used a model developed for
this purpose by DOE.  DOE provided us with a model known as DIS-RISK,
a risk-analysis version of the DIS-SPR model first developed during
an interagency study evaluating SPR expansion issues.\1 DIS-RISK is
designed to evaluate the incremental benefits of changes to the SPR
in an uncertain environment.  Given that the future course of oil
market disruptions is uncertain, this modeling approach is useful
because it generates a range of possible outcomes that can be
expected under various scenarios.  The model incorporates important
simplifying assumptions about the oil market's and policymakers'
responses, and alternative assumptions can be chosen and the results
can be examined to determine their sensitivity. 

Many uncertainties surround SPR issues, and they must be addressed in
evaluating the benefits of expanding the SPR's size and capabilities. 
These uncertainties include whether the oil supply will be disrupted,
when and how severely it will be disrupted, and how long it will be
disrupted.  Further uncertainty surrounds the oil market's response
to a supply disruption--how high oil prices will rise, for example,
and how quickly they will return to more normal levels.  In addition,
the effects of higher-than-anticipated oil prices on the economy as a
whole are uncertain, as are the government's policies for drawing
down the SPR and for adding oil to the reserve to meet specified
targets. 

The DIS-RISK model addressed these uncertainties explicitly by
requiring assumptions about the following: 

  the nature of supply disruptions, characterized by (1) the length
     of disruptions in calendar quarters, assumed to be known in
     advance, and (2) the underlying probability distribution
     describing the likelihood that a disruption of a given size will
     occur;

  the condition of the oil market in the future, including (1) oil
     prices and quantities, (2) measures of the price elasticity of
     demand, (3) measures of alternative oil supplies, including
     available foreign stocks, and (4) the rate at which oil prices
     will return to expected levels after a shock;

  the impact of an unanticipated oil price increase on the gross
     national product (GNP), expressed in elasticity form as the
     percentage decrease in GNP resulting from a 1-percent increase
     in the price of oil; and

  the policies governing the use of the SPR. 

DIS-RISK compares one potential SPR configuration, specified in terms
of target size, drawdown and distribution capability, and fill rates,
with an alternative SPR configuration.  For instance, one SPR
configuration may be based on the SPR's current size of approximately
600 million barrels, and the alternative may represent an expanded
SPR of 750 million barrels.  Both SPR configurations are subjected to
the same set of oil market disruptions, as well as the same set of
assumptions about the condition of the oil market, the effects of
oil-prices on the economy, and SPR drawdown rules. 

For each year through year 2020--the end of the analysis period--the
model generates a gross oil market disruption as a random value
derived from the specified probability distribution.  In conjunction
with the assumptions about SPR policies reacting to oil disruptions,
oil market conditions and responses, and economic impacts, it is
possible to define SPR costs and benefits for both of the SPR
configurations.  In most years, large disruptions do not occur, but
the government, as owner of the reserve, bears the costs of adding
oil to the SPR (at least if the SPR is less than its target size). 
In years in which large disruptions do occur and the SPR is drawn
down, the government will receive revenues from selling SPR oil.  If
the SPR oil cannot fully offset the disruption, however, disruption
costs will be incurred. 

The model calculates these disruption costs as the sum of lost GNP
costs, incremental oil import costs, and deadweight losses of oil
consumers.\2 By calculating the costs of the two SPR configurations,
as well as comparing their responses to the same set of disruptions,
the DIS-RISK model estimates the net benefits of increases in the
SPR's size or capability.  In most years, a larger or an enhanced SPR
will provide little or no incremental benefits because the SPR will
not be drawn down, even though the expansion or enhancement was
costly in terms of additional facilities, oil acquisition, or both. 
However, in some years, perhaps a very small number of years, a
larger or enhanced SPR will better mitigate a supply disruption than
a smaller SPR.  The benefits of a larger SPR can be expressed in
terms of smaller GNP disruption costs, incremental oil import costs,
and deadweight losses.  The net benefits of SPR expansion are the
present value of these benefits, less the present value of
incremental oil acquisition costs, capital costs, and operating
costs.  More specifically, net benefits are defined as the average of
the net benefit values obtained for each of the large
number--1,000--of iterations we performed for each readiness or
expansion issue under each scenario. 


--------------------
\1 Participants in the interagency study, which was chaired by DOE,
included the Departments of State, the Treasury, Defense, the
Interior, and Commerce; the Office of Management and Budget; the
Central Intelligence Agency; the National Security Council; the
Economic Policy Council; the Council of Economic Advisers; the
Federal Emergency Management Agency; and the Energy Information
Administration.  The study, the 1990 Interagency SPR Size Study,
examined the costs and benefits of expanding the SPR. 

\2 Deadweight losses are those losses in welfare by oil consumers
that are not offset by gains to oil producers.  The deadweight loss
and incremental oil import cost categories reflect disruption costs
in the oil market.  GNP losses reflect the frictional and cyclical
costs incurred as other sectors of the economy adjust to an oil price
shock. 


   STRUCTURE OF THE DIS-RISK MODEL
--------------------------------------------------------- Appendix I:2

One way to highlight the structure of DIS-RISK is to trace the
model's linkages in the event of an oil supply disruption.  In
general, these linkages reflect the basic set of assumptions derived
from the 1990 Interagency SPR Size Study.\3 We take this underlying
structure as given.  These core assumptions include, for example, the
uncertainty characterizing supply disruptions, the basic structure of
the oil market, the rule defining SPR usage, and the definition of
benefit categories.  Within this framework, however, the DIS-RISK
model permits great flexibility in terms of specific
parameterization.  For instance, disruptions can be characterized by
a variety of underlying statistical distributions. 

Each iteration of the model produces for each year through 2020 a
gross disruption, expressed in terms of barrels per day. 
Specifically, each iteration produces for each year a value from a
specified Weibull distribution,\4 although other distributions could
be used.  This random value represents a percentage of the volume of
the world's oil market in that year, translated into barrels per day. 
The model limits the size of the gross disruption in two ways. 
First, the gross disruption cannot exceed 50 percent of the world's
oil market, and, second, the gross disruption cannot exceed 20.25
million barrels per day.  Disruptions are assumed to be of known
duration and are expressed in terms of quarters, so a given
disruption that lasts for 2 quarters is more severe than a disruption
of equal magnitude that lasts for 1 quarter. 

In the DIS-RISK model, attempts are made to offset the gross
disruption.  Major sources of offsets include fuel switching
capability, slack capacity, and foreign and private oil stockpiles. 
Reference paths specify amounts for these offsets for each year
through 2020.  If the disruption is larger than the available
offsets, then the SPR is drawn upon. 

A key assumption of the DIS-RISK model is that the SPR will be drawn
down in an attempt to offset fully any remaining oil shortfall,
subject to two important constraints.\5 One constraint is the
exhaustion rate, defined as the amount of oil in the SPR in a given
year divided by the number of days in the disruption.  Because
disruptions are assumed to be of known duration, any given random
disruption represents a known amount of oil.  If this amount of oil
exceeds the amount of oil stored in the SPR, then the SPR cannot
fully offset the disruption even if the reserve is emptied.  Thus,
the exhaustion rate can be viewed as a size-based limit on the SPR's
ability to mitigate disruptions--the SPR cannot release more oil than
is stored, and the exhaustion rate is simply the daily drawdown
amount that would, by the end of the disruption, drain the SPR.  The
second constraint is the distribution and drawdown constraint.  The
amount of oil released to the market is further limited by the SPR's
drawdown capabilities--the SPR cannot release more oil than can be
accommodated by its system of pipelines, pumps, and terminals. 

If a net disruption does not exceed the exhaustion and drawdown
limits, the SPR is drawn down by the amount of the disruption.  An
oil price increase is thus averted because the SPR oil is substituted
for the oil whose supply is disrupted.  The government, as owner of
the reserve, receives payment for the oil it sells from the SPR. 
However, if a net disruption exceeds either the exhaustion or the
drawdown limits, the SPR does not fully offset the disruption even if
it is drawn down as much as possible.  In this event, the price of
oil increases to induce decreases in the quantity of oil demanded by
oil users. 

The estimated costs of disruptions depend on various conditions in
the oil market, as well as on relationships between the oil market
and the production of goods and services more generally.  One key
factor is, of course, the magnitude of an oil price increase, which
depends on the size of the net oil shortfall and the price elasticity
of the demand for oil.\6 Another important issue is the extent to
which a given increase in the price of oil negatively affects the
production of goods and services.  Another key factor is how quickly
oil prices return to normal--that is, to the reference oil path.\7

The precise amount by which oil prices rise depends on oil market
conditions, specifically on oil market demand elasticities.  The
DIS-RISK model estimates the world price elasticity of demand, which
in turn depends on an estimate of the U.S.  price elasticity and the
U.S.  share of world oil demand.  A more inelastic oil demand means
that any given supply disruption will lead to a larger increase in
the price of oil.\8

One major category of SPR benefits is measured in terms of losses in
the GNP that are avoided.  To the extent that SPR oil is substituted
for oil whose supply has been disrupted, the economic costs of
reallocating resources among sectors of the economy are avoided.  In
the 1990 Interagency SPR Size Study, GNP losses resulting from an
unanticipated oil price increase are represented by a simple
elasticity relationship between the oil price increase and the GNP
losses.  This relationship simplifies the effects of many linkages
from the oil sector to other parts of the economy, including the
potential reactions of the monetary authorities.  In the DIS-RISK
model, GNP losses are incurred only for the duration of the
disruption. 

Two other categories of SPR benefits also involve disruption costs
avoided.  These are incremental oil import costs and deadweight
losses of oil consumers.  During a drawdown of the SPR, oil released
from the reserve is substituted for imported oil.  However, if the
disruption is large enough to cause an increase in oil prices, a
higher price is paid for the oil that is still imported.  Likewise, a
higher price is required to induce consumers of oil to reduce the
quantity of oil they use.  Some expenditures for oil are transferred
to oil producers and are already accounted for as a disruption cost
to the extent that these expenditures flow to foreign suppliers of
oil.  However, the price increase also creates a deadweight loss. 
These disruption costs can linger beyond the duration of the
disruption because these costs are relevant as long as oil prices
exceed the anticipated oil price--that is, the reference price
assumed to hold in the absence of a disruption.  Hence, assumptions
made about the rate at which oil prices return to the reference price
path after a disruption affect the calculation of disruption costs
avoided. 

Table I.1 presents values for some of the key parameters developed
during the 1990 Interagency SPR Size Study. 



                          Table I.1
           
            Values for Key Parameters Developed in
             the 1990 Interagency SPR Size Study

Parameter        Low case       Mid case       High case
---------------  -------------  -------------  -------------
Discount rate    8 percent      10 percent     12 percent

Offsets\a        2.9 million    6.2 million    7.5 million
                 barrels per    barrels per    barrels per
                 day            day            day

Disruption       32.5 percent   32.5 percent   32.5 percent
probabilities    probability    probability    probability
                 of disruption  of disruption  of disruption
                 of greater     of greater     of greater
                 than 1         than 1         than 1
                 percent of     percent of     percent of
                 world market   world market   world market
                 and 0.52       and 1.0        and 1.44
                 percent        percent        percent
                 probability    probability    probability
                 of disruption  of disruption  of disruption
                 of greater     of greater     of greater
                 than 15        than 15        than 15
                 percent of     percent of     percent of
                 world market   world market   world market

Maximum          \b                            \b
disruption

Percent of                      50 percent
market                          20.25 million
Barrels per day

Oil price        \b             38.2 percent   \b
adjustment (per                 of difference
quarter after                   between
disruption)                     actual price
                                and reference
                                price

Demand           -0.10          -0.14          -0.17
elasticity\a

GNP elasticity   -.020          -.025          -.040
------------------------------------------------------------
\a Average of values over the years 1994 through 2020. 

\b Low-case and high-case parameter values are the same as in the mid
case. 


--------------------
\3 A complete description of the DIS-RISK model and the model's
relationship to the 1990 Interagency SPR Size Study is provided in
Paul N.  Leiby and Donald W.  Jones, "DIS-RISK Model for SPR
Analysis, Model Documentation and Benchmarking Results," Oak Ridge
National Laboratory (Dec.  1993). 

\4 The Weibull distribution is a continuous, two-parameter
distribution that is often used, for example, to study the
reliability of systems. 

\5 Actually, the model requires a fixed response rule, such as
"attempt to offset fully."

\6 The size of the shortfall depends in turn on the severity of the
disruption, the condition of the supply side of the oil market in
terms of available offsets, and the size and drawdown capability of
the SPR. 

\7 Reference path refers to the oil price and quantity values that
are specified for each year through 2020. 

\8 The price elasticity of demand is assumed to change over time. 
Throughout the period of the analysis, elasticity values are assumed
to fall in the inelastic range of the demand schedule; less inelastic
values are assumed to obtain later in the period being analyzed. 


   UNDERLYING ASSUMPTIONS MADE IN
   THE DIS-RISK MODEL
--------------------------------------------------------- Appendix I:3

The DIS-RISK model calculates economic benefits associated with the
SPR, focusing on the fundamental uncertainties surrounding
disruptions in oil supply.  Many of the assumptions made in this
model are designed to simplify the complicated linkages among sectors
of the economy and to permit analytical tractability.  Additionally,
the assumptions are not intended to describe fully the complicated
policy choices about the rates for filling and drawing down the SPR. 
In this section, we discuss three key aspects of the DIS-RISK model's
structure:  the role of price signals, the government's SPR policy,
and the model's treatment of drawdown issues. 

The model assumes that the SPR is drawn down whenever a gross
disruption cannot be offset by other sources of energy, including
foreign stocks and fuel switching.  Furthermore, if the SPR
successfully offsets the disruption, even if it is drained of oil in
the most extreme case, the price of oil will not increase.  These are
important assumptions.  For one thing, oil price increases would
likely be necessary to signal fuel switching and to bring slack
production on line.  Additionally, it is quite unlikely that oil
prices would not increase, at least temporarily, during a disruption
that was large enough to engage the SPR. 

SPR policy assumptions in the DIS-RISK model include the automatic
drawdown rule, as well as fill and refill rates.  Drawdown that is
automatic does not describe the actual conditions under which the SPR
could be drawn down.  The Energy Policy and Conservation Act provides
that SPR drawdown may not occur unless the President determines such
action is necessary due to a severe energy supply disruption. 
Furthermore, recent fill rates have been far smaller than those
assumed in the 1990 Interagency SPR Size Study.  Fill rates are
particularly important in examining SPR expansion issues. 

Other important assumptions come into play in examining drawdown
issues.  The smallest unit of time that is relevant is 1 quarter (3
months), so that any variation in economic or policy variables that
would occur within a 3-month period during a real disruption is not
reflected in the model.  As mentioned earlier, the drawdown
constraint may not effectively limit the SPR's performance during a
disruption.  In particular, if disruptions are assumed to be longer
or SPR's size is assumed to be smaller, the exhaustion rate rather
than the drawdown rate constrains the SPR's performance.  Within the
constraints of the DIS-RISK model, it may not be possible to examine,
for example, the effects of surge drawdown that would be possible if
the SPR's drawdown capability were expanded. 


   SPR READINESS AND EXPANSION
   ISSUES THAT WE EXAMINED
--------------------------------------------------------- Appendix I:4

We examined four issues relevant to investments in the SPR. 


      ELEVATED TEMPERATURES AND
      GAS CONTENT
------------------------------------------------------- Appendix I:4.1

  First, we examined the effect of elevated temperatures and gas
     content on the availability for use of some crude oil stored in
     the SPR.  We obtained information from DOE about the
     implications for the SPR's size and drawdown capabilities if
     these impairments are not addressed.  DOE estimates that, by
     blending the affected and unaffected oil and observing certain
     operating restrictions, it could draw about 520 million barrels
     out of the reserve without further corrective actions.  However,
     it could draw out only about 2 million barrels per day.  DOE
     further estimates that if it did not blend the affected and
     unaffected oil and observe the operating restrictions, it would
     have all of the oil available, but it could draw down only about
     800,000 barrels per day.  For the purposes of estimating net
     benefits, we defined the impairment in two ways on the basis of
     these alternative descriptions. 


      LIFE EXTENSION
------------------------------------------------------- Appendix I:4.2

  Second, we obtained information from DOE about the size and
     drawdown implications of DOE's proposed Life Extension Program. 
     According to DOE, without the proposed life extension projects,
     the drawdown and distribution capabilities at some of the SPR
     sites will be impaired and some sites will fall well below 70
     percent of their design capabilities. 


      SPR SIZE ISSUES:  FILL TO
      CURRENT 750-MILLION-BARREL
      CAPACITY AND EXPAND TO 1
      BILLION BARRELS
------------------------------------------------------- Appendix I:4.3

  Third, we examined the issue of filling the SPR to a target size of
     750 million barrels and then expanding the SPR's capacity from
     750 million barrels to 1 billion barrels.  Since DOE has already
     developed the capacity to store 750 million barrels of oil, the
     remaining costs of filling to this capacity would largely be for
     additional purchases of oil.  To expand the reserve to 1 billion
     barrels, additional sites and caverns would have to be developed
     and additional oil would have to be purchased. 


      DRAWDOWN ENHANCEMENTS
------------------------------------------------------- Appendix I:4.4

  Fourth, we examined two aspects related to increasing the SPR's
     drawdown capability.  We examined increasing drawdown and
     distribution from the current 4.5 million barrels per day to 6
     million barrels per day, assuming, first, that the SPR would
     remain at its current size and, second, that it would be
     expanded to a target size of 1 billion barrels. 


   SPECIFIC SCENARIOS USED IN OUR
   ANALYSES
--------------------------------------------------------- Appendix I:5

To provide a range of estimates of SPR net benefits and show how
investments are sensitive to the major uncertainties, we defined five
scenarios, or collections of assumptions about oil supply disruptions
and oil market conditions, to use in examining each of the four SPR
expansion and drawdown issues.  Key aspects of these scenarios are
presented in table I.2. 

We characterize the five scenarios as (1) the base case, (2) severe
disruptions, (3) mild disruptions, (4) responsive oil markets, and
(5) low oil prices.  In general, we relied heavily on the assumptions
made in the 1990 Interagency SPR Size Study and maintained in the
DIS-RISK model.  Importantly, we relied on the interagency study's
assumptions about disruption probabilities and oil market conditions. 
We used 1993 DOE forecasts of future oil prices and quantities that
are incorporated in the DIS-RISK model. 

Our base case includes assumptions similar to those used in the
interagency study with one notable exception.  We use a 4-percent
real discount rate rather than a 10-percent real rate selected by
DOE.  The 4-percent real discount rate better reflects the
government's current borrowing costs over the period of the analysis. 
We believe, in general, that this represents the appropriate discount
rate to use in analyzing government investments.  The 4-percent
discount rate, which we use in all of our scenarios, places a higher
present value on net benefits received in the future than a
10-percent discount rate would do. 

The severe disruption scenario differs from the base case in its
assumptions about the probability and duration of disruptions.  In
particular, disruptions are assumed to last for 9 months and there is
a greater chance that a very large disruption will happen.  In other
respects, this case is similar to the base case. 

The mild disruption scenario also differs from the base case only in
its assumptions about the probability and duration of disruptions. 
In this scenario, disruptions are assumed to last for 3 months.  The
severe and mild disruption scenarios demonstrate that beliefs about
the likelihood and severity of disruptions are important in examining
SPR issues. 

The fourth scenario--responsive oil markets--incorporates a different
set of assumptions about how oil markets respond to a disruption. 
This scenario makes the same assumptions about the probability and
duration of disruptions as the base case, but it posits more elastic
demand relationships, quicker oil price adjustments, and smaller
impacts on GNP from an oil price shock.  This scenario examines the
possible effects on the role of the SPR if innovations in the oil
market, such as the development of futures markets, tend to reduce
the impacts of oil shocks. 

Finally, actual world oil prices have been lower than those
forecasted by DOE and used in the base case scenario.  The low oil
price scenario, also developed by DOE, demonstrates the importance of
low oil prices on SPR investments, particularly on oil acquisition
costs. 



                          Table I.2
           
                      Five GAO Scenarios


                                                      Low
                                            Responsi  oil
Parameter           Base    Severe  Mild    ve        price
------------------  ------  ------  ------  --------  ------
Disruption length   2       3       1       2         2
(in quarters)

Disruption          DOE     DOE     DOE     DOE mid   DOE
probability         mid     high    low     case      mid
                    case    case    case              case

Oil price           38.2    38.2    38.2    66.7      38.2
adjustment          percen  percen  percen  percent   percen
                    t       t       t                 t

GNP elasticity      -.025   -.025   -.025   -.015     -.025

Offsets             DOE     DOE     DOE     DOE       DOE
                    mid     mid     mid     high      mid
                    case    case    case    case      case

Elasticity          DOE     DOE     DOE     DOE       DOE
                    mid     mid     mid     high      mid
                    case    case    case    case      case

Oil prices          EIA\a   EIA     EIA     EIA mid   EIA
                    mid     mid     mid     case      low
                    case    case    case              case

Real discount rate  4       4       4       4         4
                    percen  percen  percen  percent   percen
                    t       t       t                 t
------------------------------------------------------------
\a Energy Information Administration. 


   BENEFITS OF NEAR- AND LONG-TERM
   SPR READINESS AND EXPANSION
   ISSUES
--------------------------------------------------------- Appendix I:6

For some readiness and expansion issues, our estimates of net
benefits are not particularly sensitive to the choice of scenario,
while for other issues, our estimates of net benefits may be positive
under only one scenario.  In general, the benefits of filling and
expanding are larger if disruptions are assumed to be long-lasting,
as in the severe disruption scenario, while the costs exceed the
benefits if disruptions are assumed to be short, as in the mild
disruption scenario.  Conversely, the benefits of drawdown
enhancements are more likely to be larger if disruptions are assumed
to be shorter, as in the mild scenario.  Across the expansion and
readiness issues, the responsive scenario generally produces lower
net benefits than the base case because the responsive scenario
assumes that oil markets are better able to adjust to disruptions and
that the economic effects of a price shock are smaller.  Conversely,
the low oil price scenario generally produces higher net benefits
than the base case because the low oil price scenario assumes that
the costs of filling (and refilling) the SPR are lower. 


      ELEVATED TEMPERATURE AND GAS
      CONTENT
------------------------------------------------------- Appendix I:6.1

The benefits of reducing the SPR's elevated temperature and gas
problems are substantial.  Our results, summarized in table I.3,
suggest that, under all scenarios examined, the present value of net
benefits exceeds $2.9 billion, and in some scenarios incorporating
the more extreme definition of impairment, this value exceeds $15
billion.\9



                          Table I.3
           
              Benefits of Reducing Heat and Gas
                 Problems With SPR Crude Oil

                    (Dollars in billions)


Scenario                        Case 1\b            Case 2\c
--------------------  ------------------  ------------------
Base case                           $6.4               $14.7
Severe                               2.9                15.6
Mild                                 4.6                 7.5
Responsive                           3.0                 5.9
Low oil price                        6.9                16.7
------------------------------------------------------------
\a Analysis evaluates net benefits of increasing actual SPR
capabilities from impaired state, as defined, to design capabilities. 

\b Impaired SPR defined as 520 million barrels of oil available and
drawdown rate of about 2 million barrels per day. 

\c Impaired SPR defined as all stockpiled oil available and drawdown
rate of 800,000 barrels per day. 


--------------------
\9 Dollar figures are expressed in constant 1994 dollars. 


      LIFE EXTENSION
------------------------------------------------------- Appendix I:6.2

The benefits derived from SPR life extension are less clear, that is,
whether the benefits exceed the costs depend on the scenario.  DOE
officials characterize the Life Extension Program as correcting
impairments in the SPR's drawdown capabilities that are likely to
become more pronounced over time without the program.  The design
drawdown capability of 4.5 million barrels per day is the sum of the
design drawdown capabilities at the five SPR sites.  Given
assessments by DOE of the extent of the impairment in drawdown
capability likely to be experienced at each site at different times
in the future, we developed an impaired drawdown capability
schedule.\10 DOE estimated that the actual capability at three of the
five sites would be less than 70 percent of the design capability. 
On the basis of further discussions with SPR staff, we interpreted
this open-ended assessment to mean an overall reduction of 50 percent
of the design capability at these sites, which resulted in a
systemwide daily drawdown capability of approximately 3 million
barrels. 

The net benefit estimates presented in table I.4 reflect Life
Extension Program costs of about $300 million.  Any additional
operating costs that would be incurred in the absence of the Life
Extension Program have not been quantified.  Modest net benefits are
estimated under four of the five scenarios.  Only under the severe
disruption scenario, in which disruptions are assumed to last for 3
quarters, do the costs exceed the benefits.  This result is due in
part to the way in which the DIS-RISK model constrains the release of
oil from the SPR in the event of a disruption.  For a 3-quarter
disruption, the exhaustion rate constraint rather than the drawdown
constraint is binding, and no measurable incremental benefits accrue
from being able to draw down 4.5 million barrels per day rather than
approximately 3 million barrels per day. 



                          Table I.4
           
           Benefits of Replacing SPR Facilities and
           Systems Through a Life Extension Program

                    (Dollars in billions)

Scenario                                      Net benefits\a
------------------------------  ----------------------------
Base case                                               $0.9
Severe                                                  -0.3
Mild                                                     1.6
Responsive                                               0.2
Low oil price                                            1.0
------------------------------------------------------------
\a Net benefits assume that the present value of the Life Extension
Program's costs is about $300 million.  The benefits are derived by
improving daily drawdown capabilities from 2.9 million barrels (after
10 years) to the design capability (4.5 million barrels).  The SPR is
assumed to have about 600 million barrels of crude oil in storage. 


--------------------
\10 We did not attempt to assess the improvement in daily drawdown
that could be obtained if oil from sites whose drawdown capability
was relatively impaired were reallocated to sites whose actual
capability was closer to their design capability.  While costs would
be associated with transferring oil between sites, such transfers
could mitigate the drawdown impairment. 


      FILL AND EXPANSION ISSUES
------------------------------------------------------- Appendix I:6.3

The net benefits of filling the SPR from its current size to 750
million barrels are positive under some scenarios, including the base
case, but not under others.  Net benefits are largest under the
severe disruption scenario, in which disruptions are more likely and,
importantly for SPR size issues, longer.  Under the responsive oil
markets scenario, net benefits are negative.  The net benefits of
expanding the SPR from 750 million barrels to 1 billion barrels are
generally negative.  Unlike filling to 750 million barrels, expanding
beyond this size requires additional capital costs to prepare
additional cavern capacity.  These additional costs could be expected
to range from $300 million to over $1.5 billion, depending on
site-specific characteristics.  The net benefit figures presented in
table I.5 assume that the capital costs are $1 billion. 



                          Table I.5
           
              Benefits of Filling the SPR to its
              Current Capacity or Expanding and
                 Filling Additional Capacity

                    (Dollars in billions)


                             Fill to 750         Expand to 1
                         million barrels   billion barrels\a
--------------------  ------------------  ------------------
Base case                           $1.5               $-1.5
Severe                               3.3                 1.7
Mild                                -1.3                -2.8
Responsive                          -0.2                -2.4
Low oil price                        2.3                -0.4
------------------------------------------------------------
\a Assumes that the present value of the capital costs of expanding
the SPR is $1 billion. 


      DRAWDOWN ENHANCEMENTS
------------------------------------------------------- Appendix I:6.4

The costs of increasing drawdown capabilities are modest compared
with the capital and oil acquisition costs of significant SPR
expansions.  Increasing the distribution and drawdown capability
would allow more SPR oil to be extracted in a given period and would
afford operational flexibility by permitting a quicker initial
response.  However, in the DIS-RISK model, the potential advantages
of increasing drawdown capabilities may not be fully realized because
the exhaustion rate constraint rather than the drawdown constraint
may limit the amount of SPR oil that can be released.  If the amount
of oil assumed to be stored in the SPR is large and if disruptions
are assumed to be short, increases in the SPR's drawdown capability
are more likely to result in net benefits, as shown in table I.6. 



                          Table I.6
           
             Benefits of Enhancing Daily Drawdown
              Rate to 6 Million Barrels per Day

                    (Dollars in billions)


                            600-million-    1-billion-barrel
                              barrel SPR                 SPR
--------------------  ------------------  ------------------
Base case                          $-0.1                $1.3
Severe                              -0.1                -0.1
Mild                                 0.9                 1.0
Responsive                          -0.1                 0.5
Low oil price                       -0.1                 1.5
------------------------------------------------------------
\a Net benefits assume that the present value of the incremental
costs of enhancing the daily drawdown rate is $100 million. 


   COMPARING THE BENEFITS AND
   COSTS OF DRAWDOWN AND FILL
   OPTIONS
--------------------------------------------------------- Appendix I:7

Table I.7 compares the benefits and costs of filling the SPR with the
benefits and costs of adding drawdown enhancements.  Clearly,
assumptions about the duration of any disruptions are important in
choosing among these alternatives.  For this analysis, we changed
slightly our scenario definitions to emphasize the importance of a
disruption's duration.  We did this by defining the length of the
disruption in all five scenarios first as 1 quarter and then as 2
quarters.\11 All other assumptions about oil prices, oil markets'
responsiveness, and disruption probabilities are the same as in our
previous analyses.  In general, if disruptions are short, drawdown
enhancements are valuable, and if disruptions are longer, increased
oil stocks are more valuable. 



                          Table I.7
           
            Comparing Benefits of Enhancing Daily
                Drawdown With Filling the SPR

                    (Dollars in billions)



                         1-         2-         1-         2-
                    quarter    quarter    quarter    quarter
                  disruptio  disruptio  disruptio  disruptio
                          n          n          n          n
----------------  ---------  ---------  ---------  ---------
Base case              $2.1      $-0.1      $-1.2       $1.5
Severe                  3.0       -0.1       -1.1        2.7
Mild                    1.0       -0.1       -1.3        0.2
Responsive              0.6       -0.1       -1.3       -0.2
Low oil price           2.5       -0.1       -1.0        2.3
------------------------------------------------------------
\a Net benefits assume that the present value of the incremental
costs of enhancing the daily drawdown rate is $100 million. 


--------------------
\11 In our other analyses, disruption lengths were assumed to be 2
quarters in the base, responsive markets, and low oil scenarios; 3
quarters in the severe scenario; and 1 quarter in the mild scenario. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II

RESOURCES, COMMUNITY, AND ECONOMIC
DEVELOPMENT DIVISION, WASHINGTON,
D.C. 

Jim Wells, Associate Director, Energy and Science Issues
Gregg A.  Fisher, Assistant Director
Mehrzad Nadji, Assistant Director
Joseph A.  Maranto, Evaluator-in-Charge
Stephen Brown, Senior Economist

RELATED GAO PRODUCTS

Energy Security and Policy:  Analysis of the Pricing of Crude Oil and
Petroleum Products (GAO/RCED-93-17, Mar.  19, 1993). 

Oil Reserve:  Some Concerns Remain About SPR Drawdown and
Distribution (GAO/RCED-91-16, Nov.  28, 1990). 

Oil Reserve:  DOE's Management of the Strategic Petroleum Reserve
(GAO/RCED-87-171BR, July 17, 1987). 

The Strategic Petroleum Reserve:  An Overview of Its Development and
Use in the Event of an Oil Supply Disruption (GAO/RCED-85-134, Sept. 
30, 1985). 

More Assurance is Needed That Strategic Petroleum Reserve Oil Can be
Withdrawn as Designed (GAO/RCED-85-104, Sept.  27, 1985). 

Evaluation of the Department of Energy's Plan to Sell Oil From the
Strategic Petroleum Reserve (GAO/RCED-85-80, June 5, 1985). 

Analysis of Oil Withdrawal and Distribution Tests for the Strategic
Petroleum Reserve (GAO/RCED-85-115, May 8, 1985). 
