oversight

Alaskan North Slope Oil: Limited Effects of Lifting Export Ban on Oil and Shipping Industries and Consumers

Published by the Government Accountability Office on 1999-07-01.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                  United States General Accounting Office

GAO               Report to Congressional Committees




July 1999
                  ALASKAN NORTH
                  SLOPE OIL

                  Limited Effects of
                  Lifting Export Ban on
                  Oil and Shipping
                  Industries and
                  Consumers




GAO/RCED-99-191
United States General Accounting Office                                             Resources, Community, and
Washington, D.C. 20548                                                          Economic Development Division



           B-282694                                                                                         Letter

           July 1, 1999

           Congressional Committees

           This report responds to the mandate in Public Law 104-58, title II, that GAO review Alaska and
           California energy production and the effects of lifting the ban on exporting Alaskan North Slope oil.
           As agreed with your offices, this report addresses the effects of lifting the export ban on (1) Alaskan
           North Slope and California crude oil prices and production and (2) refiners, consumers, and the oil
           shipping industry (including the tanker fleet, the tanker building industry, and the tanker repair
           industry) on the U.S. West Coast. To put the effects of lifting the ban in context, the report covers
           changes in Alaska and California production during the past decade (1989 through 1998). This report
           also discusses export-related environmental issues related to lifting the export ban.

           We are sending copies of this report to the Honorable William M. Daley, Secretary of Commerce; the
           Honorable Bill Richardson, Secretary of Energy; the Honorable Bruce Babbitt, Secretary of Interior;
           and the Honorable Rodney E. Slater, Secretary of Transportation. Copies will also be made available
           to others upon request.

           If you have any questions about this report, please contact me at (202) 512-3841. Major contributors
           to this report are listed in appendix III.




           Susan D. Kladiva
           Associate Director, Energy,
            Resources, and Science Issues




                      Leter
United States General Accounting Office
Washington, D.C. 20548



           B-282694                                                                     Letter

           List of Committees

           The Honorable Frank H. Murkowski
           Chairman
           The Honorable Jeff Bingaman
           Ranking Minority Member
           Committee on Energy and Natural Resources
           United States Senate

           The Honorable Don Young
           Chairman
           The Honorable George Miller
           Ranking Minority Member
           Committee on Resources
           House of Representatives

           The Honorable Tom Bliley
           Chairman
           The Honorable John D. Dingell
           Ranking Minority Member
           Committee on Commerce
           House of Representatives




                      Leter         Page 2             GAO/RCED-99-191 Alaskan North Slope Oil
Executive Summary



Purpose              Over 12 billion barrels of crude oil have been produced on the Alaskan
                     North Slope since oil was discovered there in 1968. Initially, U.S. tankers
                     transported Alaskan North Slope oil to California and other U.S. refineries,
                     partly because the Congress banned exporting such oil to foreign
                     countries. The ban, intended in part to reduce U.S. dependence on foreign
                     oil, was controversial from the beginning. Advocates of lifting the ban
                     argued that selling the oil in the world market would increase the demand
                     for it. Increased demand, in turn, was expected to increase the price and
                     production of Alaska and California crude oil, thereby increasing the states’
                     revenues. Opponents argued that increased crude oil prices resulting from
                     lifting the ban would reduce some refiners’ profit margins and force some
                     to become dependent on Alaskan North Slope oil because they would have
                     no practical access to cheaper foreign oil. Opponents also argued that
                     lifting the ban would take business from the U.S. shipping industry because
                     operators of oil tankers would use low-cost foreign tankers and crews to
                     export Alaskan North Slope oil and have tankers repaired in low-cost
                     foreign shipyards.

                     Legislation enacted in 1995 allowed Alaskan North Slope oil to be exported
                     (P.L. 104-58, title II). That legislation also required GAO to review Alaska
                     and California energy production and the effects of lifting the export ban.
                     As agreed with the Senate Committee on Energy and Natural Resources
                     and the House Committees on Resources and on Commerce, this report
                     addresses the effects of lifting the export ban on (1) Alaskan North Slope
                     and California crude oil prices and production and (2) refiners, consumers,
                     and the Alaskan North Slope oil-shipping industry (including the tanker
                     fleet, the tanker building industry, and the tanker repair industry) on the
                     U.S. West Coast. For the purpose of this report, the U.S. West Coast
                     includes Alaska, California, Hawaii, Oregon, and Washington State. To put
                     the effects of lifting the ban in context, this report discusses changes in
                     Alaska and California production during the past decade (1989 through
                     1998). This report also discusses export-related environmental issues
                     resulting from lifting the ban (see app. I).



Background           In 1968, billions of barrels of crude oil were discovered in Prudhoe Bay on
                     the Alaskan North Slope, significantly affecting U.S. oil and oil-shipping
                     industries. Alaska became a major U.S. oil-producing state. U.S. West
                     Coast refiners retooled to efficiently process Alaskan North Slope oil,
                     which accounted for about 43 percent of all crude oil that was refined on
                     the West Coast in 1998. New tankers were also built to transport the oil



             Leter   Page 3                                  GAO/RCED-99-191 Alaskan North Slope Oil
        Executive Summary




        from Valdez, Alaska, on Prince William Sound, to West Coast and other
        refineries. The first commercial tanker carrying Alaskan North Slope oil
        left Valdez, Alaska, for the West Coast on August 1, 1977. The first
        commercial tanker exporting such oil left Valdez for Asia on May 31, 1996,
        approximately 6 months after the legislation lifting the ban was enacted.1
        (See fig. 1.)


        Figure 1: Locations of Alaska Oil Fields and Tanker Routes From Valdez, Alaska, to
        Refineries That Received Alaskan North Slope Oil, 1977-98

                                                      Arctic Ocean

                                                             Prudhoe Bay
                                                     North
                                                     Slope

                                                Brooks
                                               Mountains
                                                                                              A r c ti c C ir c l e



                                                                              Trans-Alaska Pipeline




                                              Cook Inlet
                                                                             Valdez
                                                             Ni
                                                                  k i s ki
                                                                                U.S
                                                                                   .
                                                                                  Ea
                                                                                      st a
                                                                                        nd
                                                                                             Gu




                                                ia
                                                                                              lf C




                                              As                                                                                  U.
                                                                                                  oa




                                                                                                                                     S.
                                                                                                     st




                                                                                                                                        W
                                                                                        U.




                                                       wa
                                                       ii




                                                                                                             an
                                                                                                        s




                                                                                                                                         es
                                                                                           S.




                                                                                                                  d
                                                     Ha                                                               Pu                          tC
                                                                                                                                                        oa
                                                                                              V




                                                                                                                             er                              st
                                                                                              ir g




                                                                                                                                  to
                                                                                                  in




                                                                                                       sl                              Ri
                                                                                                            an                              co
                                                                                                     I




                                                                                                                 ds                              v ia
                                                                                                                      v ia                              Pa
                                                                                                                                                           na     ma
                                                                                                                             Ca
                                                                                                                                   pe
                                                                                                                                        H or
                                                                                                                                                  n


        Source: Alyeska Pipeline Service Company, BP-Amoco, Alaska Department of Natural Resources,
        and Energy Security: Impacts of Lifting Alaskan North Slope Oil Exports Ban (GAO/RCED-91-21,
        Nov. 8, 1990).



        1
         The Nov. 28, 1995, legislation authorized the export of Alaskan North Slope oil unless the President
        found, within 5 months of the date of enactment, that exporting the oil was not in the national interest.




Leter   Page 4                                                      GAO/RCED-99-191 Alaskan North Slope Oil
                   Executive Summary




                   GAO conducted statistical and economic analyses of crude oil price and
                   production data, reviewed related studies, and obtained the views of
                   federal, state, oil industry, shipping industry, and other officials to
                   determine the effects of lifting the export ban. To determine the effects of
                   lifting the ban on oil prices, GAO developed a time-series model. Because
                   oil prices are influenced by many factors in addition to removing the ban,
                   GAO controlled for these other factors by modeling the difference between
                   the prices of West Coast oils and the prices of similar oils in other markets.
                   Furthermore, where applicable, established economic concepts and
                   theories were applied to predict the likely effects on Alaskan North Slope
                   and California crude oil production in the future. When important price,
                   production, refining, and shipping data were unavailable because they were
                   proprietary, GAO attempted, to the extent possible, to obtain such
                   information from alternative sources. However, because of proprietary
                   data limitations, GAO was unable to determine the full effects of lifting the
                   export ban on cost increases for refiners using Alaskan North Slope or
                   comparable California oil or on the U.S. West Coast market in general.



Results in Brief   Lifting the export ban raised the relative prices of Alaskan North Slope and
                   comparable California oils between $.98 and $1.30 higher per barrel than
                   they would have been had the ban not been lifted.2 To date, these price
                   increases have not had an observable effect on Alaskan North Slope and
                   California crude oil production. Nevertheless, future oil production should
                   be higher than it would have been because higher crude oil prices have
                   given producers an incentive to produce more oil. According to
                   projections by the Alaska Department of Revenue and to oil industry
                   officials, new oil fields developed in Alaska since the ban was lifted are
                   expected to increase Alaskan North Slope oil production by an average of
                   115,000 barrels per day for the next two decades. However, it was not
                   possible for GAO to separate the effects of lifting the ban on expected
                   production from the effects of broader oil market changes occurring at the
                   same time. For example, relatively high world oil prices in 1996 and 1997
                   encouraged oil producers to expand exploration and development
                   activities, while low prices in 1998 caused producers to close wells and
                   reduce development activities. Moreover, this expected production
                   increase will not reverse the decade-long decline of Alaska and California



                   2
                    The price of Alaskan North Slope and comparable California oils rose in comparison to selected,
                   widely traded world oils commonly used as benchmarks for comparing oils and setting prices.




                   Page 5                                               GAO/RCED-99-191 Alaskan North Slope Oil
                             Executive Summary




                             oil production, which is expected to continue as aging oil fields become
                             depleted.

                             Lifting the export ban increased some refiners’ costs but had limited effects
                             on consumers and the oil-shipping industry on the West Coast. While
                             higher prices for Alaskan North Slope and comparable California oil
                             increased the costs of some individual refiners using that oil, it was not
                             possible to determine the extent of cost increases for those refiners or the
                             West Coast market in general. Despite higher crude oil prices for some
                             refiners, no observed increases occurred in the prices of three important
                             petroleum products used by consumers on the West Coast--gasoline, diesel,
                             and jet fuel. Lifting the ban has also had a minimal effect to date on most
                             oil tanker operators that transport Alaskan North Slope oil, the U.S.
                             shipbuilding industry, and the West Coast ship repair industry. However,
                             shipbuilding and ship repair industry officials on the West Coast are
                             concerned that Alaskan North Slope oil tanker business may shift in the
                             future to low-cost foreign shipyards.



Principal Findings

Lifting the Export Ban       Lifting the ban caused the relative prices of Alaskan North Slope and
Increased Oil Prices and     California oils with comparable characteristics to be between $.98 and
                             $1.30 higher per barrel than they would have been had the ban not been
Should Increase Future Oil   removed, according to GAO’s analyses. 3 In addition, lifting the ban led to
Production                   exports to Asia, which allowed oil companies to reduce their shipping costs
                             for the oil that was exported because Asian ports are closer than the ports
                             for the U.S. Gulf Coast and U.S. Virgin Islands, where some Alaskan North
                             Slope oil was shipped before the ban was lifted. However, lifting the ban
                             has not led to a large volume of exports--only about 5 percent (60,000
                             barrels per day) of all Alaskan North Slope production has been exported
                             to foreign countries since the ban was lifted. Furthermore, oil production
                             in Alaska and California has had no observable increase to date as a result
                             of lifting the export ban.



                             3
                              In conducting these analyses, GAO selected three world oils that are commonly used as benchmarks
                             for comparing oil prices or with characteristics (weight and sulfur content) comparable to Alaskan
                             North Slope oil. The lighter the weight and lower the sulfur content, the higher the quality of crude oil
                             because it costs less to refine this oil.




                             Page 6                                                 GAO/RCED-99-191 Alaskan North Slope Oil
                                 Executive Summary




                                 GAO believes future production should increase because the ban was
                                 lifted, although not enough to reverse the decade-long decline in oil
                                 production as aging oil fields become depleted. Higher market prices and
                                 lower shipping costs have given oil producers more incentive to develop
                                 new oil fields. Industry and government officials told us that the
                                 development of new Alaskan North Slope oil fields increased during the
                                 period after the export ban was removed. These new fields are expected to
                                 average about 115,000 barrels per day between 1999 and 2020. Some oil
                                 officials attributed part of this increase to the effects of lifting the ban,
                                 while others said that these effects could not be separated from broader
                                 market conditions. These officials cited high world oil prices in 1996 and
                                 1997 as one of the factors that encouraged them to open new fields in
                                 Alaska. Conversely, oil officials said that low oil prices in 1998 caused
                                 them to close California wells to avoid costly maintenance and to modify
                                 their plans for the future development of the Alaskan North Slope. GAO
                                 could not separate the effects of lifting the ban on expected production
                                 from these broader market changes. Nonetheless, while production is
                                 expected to increase, the increase will not reverse the overall long-term
                                 decline in Alaska and California oil production as aging oil fields become
                                 depleted. Production in both states decreased almost every year from 1989
                                 through 1998 and is expected to fall further in the future.


The Effects of Lifting the Oil   Lifting the export ban generally had limited effects on refiners, consumers,
Export Ban on Refiners,          and the shipping industry on the West Coast. Higher market prices for
                                 Alaskan North Slope and comparable California oils translate directly into
Consumers, and the               higher costs for refiners buying these oils. However, not all refiners were
Shipping Industry on the         affected equally, as illustrated in the following hypothetical cases. For
West Coast Have Been             refiners that used large volumes of Alaskan North Slope and comparable
Generally Limited                California oils, costs would have risen when the prices of these oils rose. If
                                 the refiners bought only these oils at the market price, costs would have
                                 risen by exactly the amount the price increased as a result of lifting the
                                 ban--about $.98 to $1.30 per barrel. For refiners that did not use significant
                                 quantities of these oils, cost would have been less affected by price
                                 increases. Finally, for refiners that used mostly oil that came from their
                                 own companies’ wells, information was not available to determine how
                                 price increases affected these companies’ costs. Because data on all
                                 refiners’ crude oil purchases and internal transactions are proprietary, GAO




                                 Page 7                                   GAO/RCED-99-191 Alaskan North Slope Oil
Executive Summary




could not determine the increase in refiners’ costs that was due to higher
Alaskan North Slope and California oil prices.

Despite higher crude oil costs for some refiners, no observed increases
occurred in West Coast consumer prices as a result of lifting the export
ban. GAO analyzed three important petroleum products used by
consumers, which accounted for about 80 percent of the products
produced by West Coast refiners, and found no significant increases in
prices. According to GAO’s statistical and economic analyses, the prices of
gasoline, diesel, and jet fuel on the West Coast did not significantly change
as a result of lifting the export ban. Moreover, consumer groups and
industry experts GAO contacted were unaware of any adverse effects on
consumers from lifting the ban.

To date, lifting the export ban has had limited effects on most oil tanker
operators that transport Alaskan North Slope oil, the shipbuilding industry,
and the ship repair industry. The effect on tanker operators has been
limited because most Alaskan North Slope oil--about 95 percent--has
continued to be shipped to the U.S. West Coast. Officials of charter
shipping companies carrying the exported oil said that lifting the ban had
benefited their business by slightly increasing the demand for tankers. In
1996 and 1997, according to GAO’s analysis, exports increased the demand
for U.S. tankers by one or two and created an estimated 58 to 115 new U.S.
crew jobs on tankers used to transport Alaskan North Slope oil. This was
because U.S.-registered tankers with U.S. crews used to export Alaskan
North Slope oil to Asia replaced foreign-registered tankers with foreign
crews carrying such oil to the U.S. Virgin Islands.4 These new jobs partially
offset overall job losses in the fleet resulting from declines in Alaskan
North Slope production during the past decade.

The U.S. shipbuilding and ship repair industries also have experienced few
effects. According to oil industry officials, foreign-built tankers have not
been used to export Alaskan North Slope oil, and U.S. shipbuilders have
not lost orders for new tankers to foreign shipyards. Although U.S.
shipbuilders expected at least 10 new tanker orders in the 1990s, only 3


4
  The 1995 export legislation mandated that U.S.-documented (including U.S.-registered and -crewed)
and U.S.-owned but not necessarily U.S.-built tankers be used to export Alaskan North Slope oil.
Foreign-built tankers with foreign crews are permitted to carry such oil to the U.S. Virgin Islands under
an exception in the Jones Act, which, along with several related trade laws, requires that any vessel
transporting cargo between U.S. ports must be U.S.-built, U.S.-flagged (registered), U.S.-owned, and
U.S.–crewed.




Page 8                                                 GAO/RCED-99-191 Alaskan North Slope Oil
                  Executive Summary




                  have materialized to date. Thus, half of the Alaskan North Slope oil tanker
                  fleet consists of older, single-hulled tankers built in the 1970s or before. 5
                  Furthermore, while exporting crude oil has given tanker fleet operators an
                  added incentive to repair Alaskan North Slope oil tankers in low-cost Asian
                  shipyards during export trips, there has not been a trend toward more
                  foreign repairs since exports began, according to U.S. Customs’ repair data.
                  However, officials in the U.S. shipbuilding and West Coast ship repair
                  industries said that they are concerned that business may shift in the future
                  to foreign shipyards. For example, West Coast tanker repair industry
                  officials told us that a trend toward more foreign repairs could be
                  beginning. They cited as an example a U.S. charter shipping company that
                  used a U.S. tanker to carry an export shipment of Alaskan North Slope oil
                  to Korea and then had the tanker repaired in a Korean shipyard at a cost
                  well below estimated prices for repair in a U.S. West Coast shipyard.



Recommendations   This report makes no recommendations.




Agency Comments   GAO provided a draft of this report to the Department of Energy, including
                  its Energy Information Administration and Office of Policy, for review and
                  comment. GAO discussed the report with Energy Information
                  Administration officials, including the Director, Petroleum Division, and
                  Office of Policy staff. While the Department did not take a position on the
                  findings presented in the report, it provided clarifying comments that GAO
                  incorporated, where appropriate.




                  5
                   The Oil Pollution Act of 1990 requires that all single-hulled tankers be phased out of operation by 2015,
                  depending on the tanker age, and that all new tankers built be double-hulled to reduce the effects of oil
                  spills in the event of an accident.




                  Page 9                                                 GAO/RCED-99-191 Alaskan North Slope Oil
Contents



Executive Summary                                                                                   3


Chapter 1                Alaskan North Slope Oil Discovery Was Largest in U.S. History             13
                         Alaskan North Slope Oil Discovery Changed the West Coast Oil
Introduction               Industry                                                                14
                         Original Ban Was Debated and Ultimately Removed                           17
                         Objectives, Scope, and Methodology                                        19


Chapter 2                Prices of Some West Coast Oil Rose as a Result of Lifting the Ban         22
                         Shipping Costs Are Lower for Exported Oil                                 23
Lifting the Export Ban   Improved Economic Conditions Have Had No Observable Effect on
Increased Oil Prices       Oil Production to Date                                                  25
                         Future Oil Production Should Be Higher as a Result of Lifting the Ban     25
and Should Increase      Despite Increases in Future Oil Production, the Long-Term Production
Future Oil Production      Decline Will Continue                                                   27
                         State Oil Revenues Were Affected                                          28


Chapter 3                Some Refiners’ Crude Oil Acquisition Costs Rose, but the Extent Is
                           Uncertain                                                               30
Effects of Lifting Oil   Consumers Were Not Significantly Affected by Lifting the Export Ban       31
Export Ban on            Exports Have Had a Limited Effect on Alaskan North Slope Oil
                           Shipping                                                                31
Refiners, Consumers,
and Oil Shipping
Industry on the West
Coast Have Generally
Been Limited
Appendixes               Appendix I: Lifting the Oil Export Ban Has Had Little Impact on the
                           Alaskan Environment                                               41
                         Appendix II: Methodology for Estimating the Effect of Lifting
                           the Alaskan Oil Export Ban On Crude and Petroleum Product Prices 43
                         Appendix III: GAO Contacts and Staff Acknowledgments                56


Tables                   Table 2.1: Cost to Ship Alaskan North Slope Oil, by Destination, 1996   24
                         Table 3.1: Destinations of Alaskan North Slope Oil Tankers and
                           Volumes Carried, 1994-98                                              33
                         Table II.1: Results of Price Analysis for Alaskan North Slope (ANS) Oil 47



                         Page 10                               GAO/RCED-99-191 Alaskan North Slope Oil
          Contents




          Table II.2:   Results of Price Analysis for Line 63 (L63) Oil                 47
          Table II.3:   Results of Price Analysis for Kern River (KERN) Oil             48
          Table II.4:   Results of Price Analysis for THUMS Oil                         49
          Table II.5:   Analysis of Jet Fuel (JET) Prices                               50
          Table II.6:   Analysis of Diesel Prices                                       50
          Table II.7:   Analysis of Gasoline (GAS) Prices                               51


Figures   Figure 1: Locations of Alaska Oil Fields and Tanker Routes From
            Valdez, Alaska, to Refineries That Received Alaskan North Slope
            Oil, 1977-98                                                                 4
          Figure 1.1: Map of Showing Locations of Alaska Oil Fields                     12
          Figure 1.2: Percent of Oil Removed From Alaskan North Slope and
            Percent of Production Rights in Prudhoe Bay Oil Field, by Major Oil
            Company, in 1998                                                            14
          Figure 1.3: Shipping Routes for Alaskan North Slope Oil
            Tankers, 1989-98                                                            16
          Figure 2.1: New Fields Will Add to Daily Alaskan North Slope Oil
            Production                                                                  26
          Figure 2.2: Annual Alaska and California Crude Oil Production,
            1989-98                                                                     28
          Figure 3.1: Number of Alaskan North Slope Tankers Scheduled to
            Be Phased Out, 1999-2015                                                    36
          Figure 3.2: Number of Overseas Repairs of Alaskan North Slope
            Tankers, 1989-98                                                            39
          Figure II.1: Proportion of Alaskan North Slope Oil Sold in the West
            Coast Market                                                                44
          Figure II.2: Alaskan North Slope Oil Destinations Other Than the
            West Coast                                                                  45
          Figure II.3: ANS Wellhead Price Has Risen Compared With WTI                   52




          Abbreviations

          ARCO             Atlantic Richfield Company
          BP-Amoco         British Petroleum-Amoco



          Page 11                                   GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 1

Introduction                                                                                                    Chap1
                                                                                                                    ter




               In January 1968, oil was discovered in Prudhoe Bay on Alaska’s North
               Slope—an 88,000 square-mile frozen landmass extending from the foothills
               of the Brooks Mountain Range to the Arctic Ocean, as shown in figure 1.1.
               The Prudhoe Bay area, located about 250 miles north of the Arctic Circle
               and about 1,200 miles south of the North Pole, had no local road system
               and was inaccessible by tanker most of the year because extremely cold
               temperatures freeze the nearby Arctic Ocean. Consequently, oil companies
               began planning the construction of the Trans-Alaska Pipeline System—an
               800-mile pipeline to transport oil from the frozen Alaskan North Slope to
               Valdez, on Alaska’s Prince William Sound, for shipment to distant
               refineries. The Congress approved pipeline construction in November
               1973, and construction was completed in July 1977. The first commercial
               tanker carrying Alaskan North Slope oil from Valdez left for the U.S. West
               Coast on August 1, 1977.


               Figure 1.1: Map of Showing Locations of Alaska Oil Fields

                                                   Arctic Ocean

                                                               Prudhoe Bay
                                              National        North       Arctic
                                              Petroleum       Slope       National
                                              Reserve
                                                                          Wildlife
                                                                          Refuge
                                                  Brooks
                                                 Mountains
                                                                                        A r c ti c C ir c l e




                                                                            Trans-Alaska Pipeline




                                                 Cook Inlet
                                                                           Valdez


                                                                         Prince William Sound




               Source: Alyeska Pipeline Service Company, BP-Amoco, and Alaska Department of Natural
               Resources.




               Page 12                                                GAO/RCED-99-191 Alaskan North Slope Oil
                          Chapter 1
                          Introduction




Alaskan North Slope       Alaska contains huge quantities of crude oil. The Prudhoe Bay discovery
                          was the largest in North America. Oil companies estimate that the state
Oil Discovery Was         had at least 41 billion barrels of oil in place at the time of the North Slope
Largest in U.S. History   discovery. According to Alaska Department of Natural Resources data,
                          updated May 1998, an estimated 19.5 billion barrels were extractable using
                          today’s technology and under prevailing economic conditions (commonly
                          referred to as proven reserves).

                          Of the 19.5 billion barrels of proven reserves, about 13.8 billion barrels have
                          already been produced by 22 fields. Thirteen Alaskan North Slope fields
                          that contained an estimated 18.2 billion barrels of proven reserves have
                          produced about 12.5 billion barrels. Prudhoe Bay, the oldest and largest
                          field on the Alaskan North Slope, accounted for about 73 percent of those
                          reserves and about 80 percent of total production. The remaining proven
                          reserves are contained in nine Cook Inlet fields that have already produced
                          about 1.2 billion barrels. Since 1978, the first full year of Alaskan North
                          Slope oil production after the completion of the Trans-Alaska Pipeline,
                          Alaska has accounted for between 14 and 25 percent of U.S. crude oil
                          production and has ranked among the largest U.S. crude oil-producing
                          states every year.

                          The Alaska Department of Natural Resources’ estimates, however, did not
                          include all Alaska oil. The estimates excluded Alaskan North Slope oil
                          fields in various stages of development that had not produced measurable
                          quantities of oil by 1998. They also excluded the Alaska National
                          Petroleum Reserve, the Arctic National Wildlife Refuge, and undeveloped
                          Outer Continental Shelf areas. Oil analysts believe these areas contain
                          billions of barrels of proven reserves.1

                          British Petroleum-Amoco Corporation (BP-Amoco), Atlantic Richfield
                          Company (ARCO), and Exxon have controlling interests in most Alaskan
                          North Slope oil production. As shown in figure 1.2, in 1998 these three
                          companies owned production rights for over 90 percent of the Prudhoe Bay
                          field and accounted for over 90 percent of all the oil removed from the
                          Alaskan North Slope.2 Fourteen other companies also had production


                          1
                            Alaska also has trillions of cubic feet of natural gas that Alaskan North Slope oil producers would like
                          to commercialize. The oil industry is exploring ways to convert the natural gas to liquid to be
                          transported off the North Slope, possibly through traditional pipelines such as the Trans-Alaska
                          Pipeline.
                          2
                              BP and Amoco merged in 1998. In April 1999, BP-Amoco and ARCO confirmed plans to merge.




                          Page 13                                                GAO/RCED-99-191 Alaskan North Slope Oil
                        Chapter 1
                        Introduction




                        interests in the Alaskan North Slope in 1998, including companies owned
                        by native Alaskan groups.



                        Figure 1.2: Percent of Oil Removed From Alaskan North Slope and Percent of
                        Production Rights in Prudhoe Bay Oil Field, by Major Oil Company, in 1998
                        120% Percentage


                        100%



                         80%



                         60%



                         40%



                         20%



                          0%

                                  Oil removed      Prudhoe Bay



                                           Other

                                           Exxon

                                           ARCO

                                           BP-Amoco

                        Source: Alaska Department of Natural Resources and Energy Security: Impacts of Lifting Alaskan
                        North Slope Oil Exports Ban (GAO/RCED-91-21, Nov. 8, 1990)




Alaskan North Slope     The addition of Alaskan North Slope oil production to the oil produced in
                        California and other West Coast states meant that, for the first time,
Oil Discovery Changed   production on the U.S. West Coast was greater than West Coast refiners’
                        demand for crude oil.3 Consequently, oil producers in Alaska looked to
the West Coast Oil      other markets. Figure 1.3 shows the historical shipping routes for Alaskan
Industry                North Slope oil and the location of potential refining markets. This figure
                        illustrates the principal difference between these potential markets—
                        namely, the distance between these markets and the Port of Valdez.




                        Page 14                                            GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 1
Introduction




Generally, shorter shipping distances translate into lower transportation
costs and higher profits for oil producers, although other factors, such as
tanker size, also affect costs.4 The West Coast is the closest domestic
market for Alaskan North Slope oil, and Asia is closer than most other U.S.
markets, such as the U.S. Gulf Coast and U.S. Virgin Islands. However, the
Congress had banned the export of Alaskan North Slope oil. Therefore,
Alaskan North Slope oil producers took oil not sold on the West Coast to
more distant domestic markets.




3
 For the purpose of this report, the West Coast includes Alaska, California, Hawaii, Oregon, and
Washington State.
4
 Because crew size and operating costs other than fuel are basically constant regardless of tanker size,
the per-barrel cost to transport oil can be less for larger tankers than for smaller tankers.




Page 15                                               GAO/RCED-99-191 Alaskan North Slope Oil
                                          Chapter 1
                                          Introduction




Figure 1.3: Shipping Routes for Alaskan North Slope Oil Tankers, 1989-98




                                                         North Slope

                                                              Prudhoe Bay



                                      Nikiski      Trans-Alaska Pipeline

                                                Valdez




                 Asia
                                                           West Coast                                                                              East Coast




                                                                Pipeline


                        Hawaii

                                                                                                                        Gulf
                                                                                                                        Coast
                                                                                                                                                                        Puerto
                                                                                                                                                                          Rico



                                                                                                                                                            U.S. Virgin Islands


                                                                                                                                                        Trans-Panama Pipeline




                                                                       U.
                                                                            S.
                                                                                 V ir
                                                                                        gin
                                                                                              Isla
                                                                                                     nd
                                                                                                          sv
                                                                                                               ia C
                                                                                                                      ape
                                                                                                                            H or
                                                                                                                                   n
    Refinery


                                          Source: Energy Security: Impacts of Lifting Alaskan North Slope Oil Exports Ban (GAO/RCED-91-21,
                                          Nov. 8, 1990).




                                          Page 16                                                                                      GAO/RCED-99-191 Alaskan North Slope Oil
                            Chapter 1
                            Introduction




West Coast Refineries       The proximity to Valdez, along with the ban on exports, made the West
Retooled to Process         Coast the preferred destination for the sellers of Alaskan North Slope oil.
                            Because this oil’s characteristics (weight and sulfur content) differed from
Alaskan North Slope Oil     those of foreign oil, refiners had to invest in additional refining equipment
                            to handle the Alaskan North Slope oil. After West Coast refiners retooled
                            to efficiently process that oil, Alaskan North Slope oil took the place of
                            much of the foreign oil that West Coast refiners had imported. In 1998,
                            Alaskan North Slope oil constituted about 43 percent of all crude oil refined
                            on the West Coast.


New Tankers Were Built to   The discovery of oil on the Alaskan North Slope, along with the export ban,
Transport Alaskan North     also had an effect on the U.S. oil-shipping industry. U.S. shipyards built
                            over 50 tankers in the 1970s and 1980s to carry crude oil from Valdez to
Slope Oil                   distant refineries. Until the Congress lifted the ban on exporting Alaskan
                            North Slope oil, tankers transported the Alaskan North Slope oil to U.S.
                            ports. As a result, the tankers were required to comply with the Jones Act.
                            The Jones Act, along with several related trade laws, require that any vessel
                            transporting cargo between U.S. ports must be U.S.-built, U.S.-flagged
                            (registered), U.S.-owned, and U.S.-crewed. Under an exception in the
                            Jones Act, foreign-built tankers were allowed to transport oil from Valdez
                            to the U.S. Virgin Islands.



Original Ban Was            The Congress banned exporting Alaskan North Slope oil when it authorized
                            the construction of the Trans-Alaska Pipeline in 1973. The legislation,
Debated and                 which was enacted in the midst of the Arab oil embargo, amended the
Ultimately Removed          Mineral Leasing Act of 1920 and restricted the export of U.S. oil transported
                            over a federal right-of-way. Exports were allowed only if the President
                            found that they would not diminish the quantity or quality of oil available to
                            the United States and were in the national interest. The Energy Policy and
                            Conservation Act of 1975, the Export Administration Act of 1979, and
                            various other laws provided additional restrictions on Alaskan North Slope
                            oil exports. These restrictions were intended, in part, to reduce U.S.
                            dependency on foreign oil, ensure that Alaskan North Slope oil would be
                            used to benefit U.S. citizens, and protect the U.S. economy from a drain of
                            scarce resources.

                            The export ban was controversial from its beginning, and the pros and cons
                            of lifting it were debated in congressional hearings and in other discussions
                            for years. In addition, several studies addressed the likely effects of lifting



                            Page 17                                  GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 1
Introduction




the ban. At issue was who would benefit and who would not benefit from
lifting the ban.

Advocates of lifting the export ban argued that it created a surplus of
Alaskan North Slope oil on the West Coast, in turn depressing price and
production and limiting state governments’ revenues.5 For example, the
Department of Energy concluded in 1994 that lifting the ban on exporting
Alaskan North Slope oil would (1) increase the price of the oil by
expanding its markets, (2) increase Alaska and California revenues through
increased royalties and taxes, and (3) generate new economic activity and
employment in Alaska and California.6 Moreover, these benefits were
expected to accrue without an increase in gasoline prices.

Opponents argued that lifting the ban would have adverse consequences.
For example, in a 1995 report prepared for the Coalition to Keep Alaska Oil,
consultants agreed with the Department of Energy’s 1994 conclusions that
the price and production of Alaskan North Slope oil would increase. But
they also predicted that oil companies’ export-related revenue and
production gains would be small and of short duration because the West
Coast would become dependent on foreign imports.7 The consultants also
predicted that refiners that only refine crude oil and do not produce oil
(commonly referred to as independent refiners) would become dependent
on Alaskan North Slope oil because they would have no practical access to
cheaper foreign oil and their profit margins would decrease. Furthermore,
the report stated, consumers’ prices would increase because crude oil
prices would be higher. Finally, allowing companies to export oil on
foreign-built tankers instead of more costly U.S.-built tankers was expected
to hurt the U.S. shipping industry.

In 1990, we reported that lifting the ban would likely increase the price of
Alaskan North Slope oil. 8 We reported that some oil would likely be
exported to Asia instead of being shipped to the U.S. East and Gulf Coasts,


5
 States receive severance tax for oil extracted from the ground for consumption in other states, royalty
revenue based on the value of oil, income tax, and property tax.
6
    Exporting Alaskan North Slope Crude Oil: Benefits and Costs, Department of Energy (June 1994).
7
 National Consequences of Exporting Alaska North Slope Oil, Wilson Gillette & Co., Petroleum
Economics and Logistics Consultants (May 1995), prepared for the Coalition to Keep Alaska Oil—a
Washington, D.C.-based group that opposed exporting Alaska oil.
8
Alaskan Crude Oil Exports (GAO/T-RCED-90-59, Apr. 5, 1990) and Energy Security: Impacts of Lifting
Alaskan North Slope Oil Exports Ban (GAO/RCED-91-21, Nov. 8, 1990).




Page 18                                                GAO/RCED-99-191 Alaskan North Slope Oil
                         Chapter 1
                         Introduction




                         the U.S. Virgin Islands, Puerto Rico, and possibly some U.S. West Coast
                         ports because transportation costs to Asia were lower. We also reported
                         that lifting the ban would promote economic efficiency by increasing
                         domestic oil production and allowing better use of refinery resources.
                         Finally, we stated that lifting the ban would increase the decline in demand
                         for U.S. tankers because Alaskan North Slope oil would be exported on
                         foreign-flagged instead of U.S.-flagged tankers.

                         In 1995, the Congress lifted the ban on exporting Alaskan North Slope oil
                         (P.L. 104-58, title II). The 1995 act eliminated the export restrictions in the
                         Mineral Leasing Act of 1920 and various other statutes and regulations.
                         The act also requires that oil tankers transporting Alaskan North Slope oil
                         to foreign destinations be U.S. documented (including U.S.-registered and
                         U.S.-crewed) but not necessarily U.S.-built. According to the conference
                         report accompanying the 1995 legislation, the purpose of lifting the export
                         ban was to allow Alaskan North Slope crude oil to compete with other
                         crude oil in the world market under normal market conditions. The first
                         commercial tanker exporting Alaskan North Slope oil left Valdez for Asia
                         on May 31, 1996, about 6 months after the ban was lifted.9



Objectives, Scope, and   The 1995 law required us to review Alaska and California energy
                         production and the effects of lifting the ban on independent oil refiners,
Methodology              consumers, and shipbuilding and ship repair yards on the West Coast and
                         Hawaii. As agreed with the Senate Committee on Energy and Natural
                         Resources and the House Committees on Resources and on Commerce,
                         this report responds to that mandate and addresses the effects of lifting the
                         export ban on (1) Alaskan North Slope and California crude oil prices and
                         production and (2) refiners, consumers, and the oil-shipping industry
                         (including the tanker fleet, the tanker building industry, and the tanker
                         repair industry) on the U.S. West Coast. To put the effects of lifting the ban
                         in context, this report discusses changes in Alaska and California
                         production during the past decade (1989 through 1998). This report also
                         discusses export-related environmental issues resulting from lifting the ban
                         (see app. I).

                         To assess the effect of lifting the export ban on Alaskan North Slope and
                         California crude oil prices and production, we collected and analyzed


                         9
                          The Nov. 28, 1995, legislation authorized the export of Alaskan North Slope oil unless the President
                         found, within 5 months of the date of enactment, that exporting the oil was not in the national interest.




                         Page 19                                                GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 1
Introduction




crude oil price and production data from the Department of Energy, the
Alaska Departments of Natural Resources and of Revenue, the California
Departments of Conservation and of Revenue, selected oil producers and
refiners, the Alyeska Pipeline Service Company—the organization that
operates the Trans-Alaska Pipeline System—and Platts Oil Prices Data
Base as reported by Standard & Poor’s DRI. We also reviewed previous
GAO reports, studies, and other available literature. In addition, we
interviewed federal, state, and oil industry officials to obtain their views on
the effects of lifting the ban. Furthermore, we conducted statistical
analyses using oil-price data before and after the ban was lifted to
determine how lifting the export ban had affected the prices of Alaskan
North Slope and California oil. A complete discussion of our statistical and
economic analyses for determining the effects of lifting the export ban on
Alaskan North Slope and California crude oil prices is in appendix II.

To assess the effects of lifting the export ban on refiners, consumers, and
the oil-shipping industry on the West Coast, we interviewed West Coast
crude oil-refining officials, consumer groups, and oil-shipping industry
officials to obtain their views on the effects of lifting the ban. We also
conducted statistical analyses of the effects of lifting the export ban on the
prices of key petroleum products used by West Coast consumers. These
analyses were similar to those used to determine the effects of lifting the
ban on oil prices. Furthermore, to review the effects of oil exports on the
U.S. oil-shipping industry, we talked to Alaskan North Slope oil industry
officials, tanker fleet operators, shipbuilding and ship repair industry
officials, maritime union representatives, state environmental groups, and
state and federal officials. We contacted federal agencies, including the
U.S. Maritime Administration and U.S. Coast Guard within the Department
of Transportation and the U.S. Customs Service. We also interviewed state
officials in Alaska, California, Oregon, and Washington State, and industry
officials in these states (including officials with oil companies that refine
oil in Hawaii) and in Washington, D.C. From these officials, we obtained
and analyzed selected data and records to understand trends in the Alaskan
North Slope shipping, shipbuilding, and ship repair industries and to
identify the impact of oil exports on these industries.

In addition, where applicable, we applied established economic concepts
and theories to predict the likely effects on Alaskan North Slope and
California crude oil production in the future. When important price,
production, refining, and shipping data were unavailable because they were
proprietary, we attempted, to the extent possible, to obtain such
information from alternative sources. However, because of proprietary



Page 20                                  GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 1
Introduction




data limitations, we were unable to determine the full effects of lifting the
export ban on cost increases for refiners using Alaskan North Slope or
comparable California oil or on the U.S. West Coast market in general.

We provided a draft of this report to the Department of Energy, including its
Energy Information Administration and Office of Policy, for review and
comment. We discussed the report with Energy Information
Administration officials, including the Director, Petroleum Division, and
Office of Policy staff. While the Department did not take a position on the
findings presented in the report, it provided clarifying comments that we
incorporated, where appropriate.

We conducted our work from July 1998 through June 1999 in accordance
with generally accepted government auditing standards.




Page 21                                  GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 2

Lifting the Export Ban Increased Oil Prices
and Should Increase Future Oil Production                                                                                     Chap2
                                                                                                                                  ter




                        Lifting the export ban raised the prices of Alaskan North Slope and some
                        California oils between $.98 and $1.30 higher per barrel than they would
                        have been had the ban not been lifted. To date, these price increases have
                        not had an observable effect on Alaskan North Slope and California crude
                        oil production. Nevertheless, future oil production should be higher than it
                        would have been had the ban not been lifted because higher crude oil
                        prices have given producers an incentive to produce more oil. According
                        to oil industry officials, new oil fields developed in Alaska since the ban
                        was lifted are expected to increase Alaskan North Slope oil production by
                        an average of 115,000 barrels per day for the next two decades. However,
                        we could not separate the effects of lifting the ban on expected production
                        from the effects of broader oil market changes occurring at the same time.
                        For example, relatively high world oil prices in 1996 and 1997 encouraged
                        oil producers to expand exploration and development, while low prices in
                        1998 caused producers to close wells and reduce development. Moreover,
                        this expected production increase will not reverse the decade-long decline
                        of Alaska and California oil production, which is expected to continue as
                        aging oil fields become depleted.



Prices of Some West     While world oil prices have been volatile since the export ban was lifted,
                        the price of Alaskan North Slope and some California oil sold in the West
Coast Oil Rose as a     Coast market is higher than it would have been had the export ban not
Result of Lifting the   been removed. Allowing exports to Asia meant increased demand for
                        Alaskan North Slope oil and higher prices. To determine the effect of lifting
Ban                     the ban on oil prices, we developed a time-series model. Because oil prices
                        are influenced by many factors other than removing the ban, we had to
                        control for these other factors. We did this by modeling the differences
                        between the prices of West Coast oils and the prices of similar oils in other
                        markets. Our analysis indicates that the market price of Alaskan North
                        Slope oil rose compared with the prices of three oils--Brent Blend, Nigerian
                        Forcados, and West Texas Intermediate.1 The price increase for Alaskan
                        North Slope oil relative to these three oils ranged from $0.98 to $1.30 per
                        barrel.

                        The effect of lifting the ban on California oil prices depends on the type of
                        oil examined. Light-weight oil with a low sulfur content is higher quality


                        1
                         Brent Blend and West Texas Intermediate oils are widely traded oils and are commonly used as
                        benchmarks for the purpose of comparing other oils and for setting prices. Nigerian Forcados oil is
                        similar in characteristics to Alaskan North Slope oil.




                        Page 22                                               GAO/RCED-99-191 Alaskan North Slope Oil
                         Chapter 2
                         Lifting the Export Ban Increased Oil Prices
                         and Should Increase Future Oil Production




                         and more valuable than heavy oil with high sulfur content because
                         high-quality oil costs less to refine into gasoline and other light petroleum
                         products. Alaskan North Slope oil is lighter weight and has a lower sulfur
                         content than most California oils. Our analysis indicates that the price of
                         “Line 63” oil in California, which is similar in quality to Alaskan North Slope
                         oil, rose by $1.28 per barrel compared with the price of West Texas
                         Intermediate oil as a result of lifting the ban. However, the effect of lifting
                         the ban on the prices of two other Californian oils we examined (Kern
                         River and THUMS) was insignificant. These two oils are heavy in contrast
                         with Alaskan North Slope and Line 63 oil, which may explain why their
                         prices did not respond to the removal of the export ban in the same way.
                         Appendix II explains the methodology we used to estimate these price
                         increases as well as the economics explanation for why oil prices were
                         expected to increase when the ban was lifted.



Shipping Costs Are       Lifting the export ban also resulted in lower shipping costs for oil exported
                         to Asia. For example, total transportation cost in 1996 for oil sold in Asia
Lower for Exported Oil   was about $4.51 less per barrel than for oil sold on the U.S. Gulf Coast.
                         Overall, shipping costs fell by at least $15 million in 1996, $28 million in
                         1997, and $22 million in 1998 from what they would have been had oil not
                         sold in the West Coast market continued to go to other domestic markets.
                         Like higher oil prices, lower shipping costs improve oil companies’
                         incentives to produce more oil.

                         Table 2.1 shows the differences in length of tanker voyages, pipeline tariffs,
                         and total transportation costs per barrel for oil shipped from Valdez,
                         Alaska, to Asia and the U.S. Gulf Coast, the U.S. Virgin Islands, and the
                         Mid-Continent in 1996.2 As the table shows, an average tanker trip to Asia
                         took 30 days, while the average trip to the Gulf Coast took 41 days. In the
                         case of oil sold in the Gulf Coast and the Mid-Continent, shippers paid
                         pipeline tariffs in addition to tanker costs.3 The additional pipeline tariff
                         was approximately $0.82 per barrel for Gulf Coast shipments and $2.17 per
                         barrel for Mid-Continent shipments. U.S. Virgin Islands shipments went by


                         2
                           For the purposes of this report, the Mid-Continent includes Indiana, Illinois, Iowa, Kansas, Kentucky,
                         Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, Oklahoma, South Dakota, Tennessee,
                         and Wisconsin.
                         3
                          Oil destined for the U.S. Gulf Coast was typically taken by tanker from Valdez, Alaska, to the Pacific
                         side of Panama and then transported by pipeline to the Atlantic side of Panama, where it was reloaded
                         onto tankers for the trip to its final destination, such as Louisiana. Shipments to the Mid-Continent
                         went from Valdez to Los Angeles and from there by pipeline to their final destinations.




                         Page 23                                                GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 2
Lifting the Export Ban Increased Oil Prices
and Should Increase Future Oil Production




tanker from Valdez around Cape Horn. This route took an average of 84
days, or about twice as long as the next shorter route. However, the
shipping costs to the U.S. Virgin Islands were slightly lower than for the
much shorter journey to Asia because the oil companies used larger foreign
tankers with foreign crews to transport the oil to the U.S. Virgin Islands. 4
Foreign tankers are much less costly to build, and operating costs for
foreign-crewed vessels are lower than for U.S.-crewed vessels. Although
the 1995 law does not prohibit exports on foreign-built tankers, all
shipments of Alaskan North Slope oil other than to the U.S. Virgin Islands
have gone on U.S.-crewed tankers.5 Table 2.1 also shows the average costs
for West Coast shipments in 1996. As the table shows, the West Coast is the
lowest cost destination for Alaskan North Slope oil.



Table 2.1: Cost to Ship Alaskan North Slope Oil, by Destination, 1996

                                                         Pipeline tariff per
Destination                           Tanker days                    barrel             Cost per barrel
Asia                                              30                    $0.00                     $2.64
Gulf Coast                                        41                    $0.82                     $7.15
Virgin Islands                                    84                    $0.00                     $2.35
Mid-Continent                                     16                    $2.17                     $3.80
West Coast                                     5-18a                    $0.00                     $1.63
Note. The cost figures are from 1996, the last year of sales to the Gulf Coast and the only year in
which sales to Asia coincided with sales to the other destinations shown.
a
 Shipments to Nikiski, Alaska, and Hawaii took 5 and 18 days, respectively. Shipments to all other
West Coast destinations took between 6 and 17 days.
Sources: BP-Amoco, U.S. Maritime Administration, and Alaska Department of Natural Resources.




4
    To date, oil companies have used U.S.-built and U.S.-crewed tankers to export oil to Asia.
5
 One exporter told us that while shipping costs to the U.S. Virgin Islands were lower than to Asia, sales
to Asia are still more profitable than sales to the U.S. Virgin Islands because the price paid by the
Islands refiner was below the price paid in Asia. This exporter told us that its revenue minus shipping
costs is between $1 and $2 per barrel higher for Asian sales than for sales to the U.S. Virgin Islands.




Page 24                                                  GAO/RCED-99-191 Alaskan North Slope Oil
                        Chapter 2
                        Lifting the Export Ban Increased Oil Prices
                        and Should Increase Future Oil Production




Improved Economic       Higher market prices for Alaskan North Slope oil and lower shipping costs
                        for exported oil have given oil producers incentive to produce more crude
Conditions Have Had     oil. To date, however, this incentive has not had an observable effect on
No Observable Effect    Alaskan North Slope or California crude oil production. Oil industry
                        officials told us that any effects on production would not occur
on Oil Production to    immediately. There is a lag between the time producers begin to receive
Date                    higher prices for Alaskan North Slope oil and the time it takes for
                        additional development activities to produce more oil.



Future Oil Production   Oil companies began developing several new fields after the export ban
                        was lifted, and production from these fields is projected to add significantly
Should Be Higher as a   to future Alaskan North Slope production. Figure 2.1 shows the expected
Result of Lifting the   impact—starting in 1999—of the development of new fields since the
                        export ban was lifted on production levels. The bottom line in the figure
Ban                     shows the current projected production of fields that existed prior to the
                        lifting of the export ban. The top line shows the current projected
                        production of all fields--including those that were developed and those for
                        which development has been planned and approved–-since the export ban
                        was lifted. The additional projected production between 1999 and 2020
                        from these new fields is about 115,000 barrels per day, on average. Some
                        oil industry officials told us that some of these new developments were in




                        Page 25                                       GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 2
Lifting the Export Ban Increased Oil Prices
and Should Increase Future Oil Production




response to the removal of the export ban, while others said it was difficult
to point to one factor to explain the change.



Figure 2.1: New Fields Will Add to Daily Alaskan North Slope Oil Production

1400 Thousands of barrels per day



1200



1000



 800



 600



 400



 200



   0



       1998     2000    2002     2004       2006   2008     2010   2012   2014   2016   2018   2020
        Years

                All Fields: 1998 Forecast
                1995 Fields: 1998 Forecast

Source: Alaskan Department of Revenue, Revenue Sources Book: Forecast and Historical Data, Fall
1995 and 1998.


We found no evidence of a similar increase in oil production in California.
Overall oil production in California has continued to decline in the years
since the ban was lifted, and we did not observe an expansion of
development activity. While an increase in the market price of some
California oils would be expected to lead to increased levels of production,
none of the oil producers contacted said they had increased their
production as a result of lifting the ban.

We could not separate the effects of lifting the export ban on expected
production increases from the effects of broader oil market changes



Page 26                                                   GAO/RCED-99-191 Alaskan North Slope Oil
                         Chapter 2
                         Lifting the Export Ban Increased Oil Prices
                         and Should Increase Future Oil Production




                         occurring at the same time. Among the other factors positively affecting
                         production decisions were generally high oil prices in 1996 and 1997 and
                         improvements in oil exploration and recovery technology. Higher oil prices
                         encourage greater investment in production and exploration. Average
                         market prices for Alaskan North Slope oil in 1996 and 1997 were $17.74 and
                         $20.90 per barrel, respectively, compared with $15.86 in 1998. Similarly,
                         improved production and exploration technology has lowered production
                         costs, providing greater incentive to produce more oil. More recently, low
                         oil prices in 1998 caused California oil producers to close some oil wells to
                         avoid maintenance costs. The low prices also caused Alaska oil producers
                         to delay planned investments and development. Oil company officials,
                         government analysts, and industry experts told us that separating the
                         effects of lifting the export ban from such other factors is difficult if not
                         impossible.



Despite Increases in     The expected increase in Alaskan North Slope oil production from lifting
                         the ban will not reverse the long-term decline in oil production in Alaska
Future Oil Production,   and California as aging oil fields in these states become depleted. As
the Long-Term            shown in figure 2.2, crude oil production in both Alaska and California
                         decreased almost every year from 1989 through 1998. During that period,
Production Decline       Alaska production decreased by about 35 percent, or about 696,000 barrels
Will Continue            per day, primarily because increased production in new, relatively small oil
                         fields did not offset decreased production in large aging fields. New fields
                         and fields that had been closed but were reopened during that period added
                         about 236,000 barrels per day in 1998, which was less than the production
                         decrease in the Prudhoe Bay field, the oldest and largest oil field on the
                         Alaskan North Slope. By 1998, the Prudhoe Bay field was about 74-percent
                         depleted, and production was about half the 1989 level—about 713,000
                         barrels per day versus about 1.43 million barrels per day. California
                         production also decreased by about 9 percent during that period, or about
                         94,000 barrels per day, because production in new fields did not offset




                         Page 27                                       GAO/RCED-99-191 Alaskan North Slope Oil
                     Chapter 2
                     Lifting the Export Ban Increased Oil Prices
                     and Should Increase Future Oil Production




                     decreased production from aging fields. Low oil prices in 1998 also
                     discouraged California production.



                     Figure 2.2: Annual Alaska and California Crude Oil Production, 1989-98
                     2.5   Millions of barrels per day




                     2.0




                     1.5




                     1.0



                     0.5




                       0



                           1989      1990      1991      1992   1993   1994   1995    1996     1997    1998
                           Years
                                      Alaska

                                      California


                     Source: Alaska Department of Revenue and California Department of Conservation.




State Oil Revenues   Alaska revenue rose because of the higher market prices and lower
                     shipping costs that resulted from lifting the export ban. Alaska’s petroleum
Were Affected        revenue comes from severance taxes, royalties, corporate income tax,
                     property tax, and petroleum rent and lease bonuses. Royalty, severance
                     tax, and income tax revenue are based on the value of oil after excluding
                     pipeline tariff and transportation costs. In April 1998, the Alaska
                     Department of Revenue estimated that the annual increase in revenue
                     resulting from higher West Coast market prices for Alaskan North Slope oil
                     was $40 million. The officials also estimated that the annual increase in




                     Page 28                                            GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 2
Lifting the Export Ban Increased Oil Prices
and Should Increase Future Oil Production




revenue from lower shipping costs to Asia was $10 million. These effects
were the direct result of lifting the export ban.

California revenue comes from a share of federal royalties, income taxes,
and property taxes. California officials told us that they receive relatively
little revenue from these sources. Consequently, there was no significant
change in revenue as a result of lifting the export ban.




Page 29                                       GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 3

Effects of Lifting Oil Export Ban on Refiners,
Consumers, and Oil Shipping Industry on the
West Coast Have Generally Been Limited                                                                    Chap3
                                                                                                              ter




                          Lifting the oil export ban has had limited effects on refiners, consumers,
                          and the oil-shipping industry—including Alaskan North Slope fleet
                          operators, shipbuilders, and tanker repair yards. Higher market prices for
                          Alaskan North Slope and some California oil increased some refiners’ costs
                          but had no or an unclear effect on other refiners’ costs. Despite higher
                          crude oil costs for some refiners, West Coast consumers appear to have
                          been unaffected by lifting the ban because the prices of important
                          petroleum products they use have not increased. There have also been
                          minimal effects on the shipping industry to date, although shipbuilding and
                          repair industry officials are concerned that business may shift in the future
                          to low-cost foreign shipyards.



Some Refiners’ Crude      While higher prices for Alaskan North Slope and comparable California oil
                          increased the costs of some individual refiners, we could not determine the
Oil Acquisition Costs     extent of the cost increase for these refiners or for the West Coast market
Rose, but the Extent Is   in general. Proprietary data needed to make the determination were not
                          available. The impact of rising costs on refiners depends on their ability to
Uncertain                 pass these costs on to consumers by raising the prices of the petroleum
                          products they sell.

                          Higher market prices for Alaskan North Slope and comparable California
                          oil translate directly into higher costs for refiners buying this oil on the
                          market. However, not all refiners are affected equally. We looked at three
                          hypothetical cases. First, a refiner buying large volumes of Alaskan North
                          Slope and comparable California oil would experience cost increases when
                          the prices of such oil rise. In the case in which a refiner buys nothing but
                          this oil and always at the market price, costs would rise by exactly the
                          amount the price increased as a result of lifting the ban—about $.98 to
                          $1.30 per barrel on the basis of our analysis. Second, the costs for a refiner
                          buying little or no Alaskan North Slope or comparable California oil would
                          be largely unaffected by increases in the market prices of this oil. Finally,
                          for some refiners that refine mostly oil that comes from their own
                          companies’ wells, the effect of the increase in the market price of the oil
                          they produce and refine is unclear because their oil is not sold in the
                          market.

                          Data on refiners’ crude oil purchases and the prices paid are unavailable
                          because they are proprietary. Therefore, we could not determine the
                          increase in refiners’ costs because of higher Alaskan North Slope and
                          California oil prices that resulted from lifting the ban. Some refiners we
                          contacted said they pay higher prices for this oil, some said they were



                          Page 30                                  GAO/RCED-99-191 Alaskan North Slope Oil
                         Chapter 3
                         Effects of Lifting Oil Export Ban on Refiners,
                         Consumers, and Oil Shipping Industry on the
                         West Coast Have Generally Been Limited




                         unaffected, and others said it was analytically impossible to determine the
                         effect. However, none of the refiners shared specific cost data with us.

                         The extent to which refiners can pass higher costs on to consumers
                         determines how their profits are affected by increased crude oil prices.
                         The ability of West Coast refiners to pass rising crude oil costs on to
                         consumers may be constrained by competitive oil market conditions. All
                         refiners were not affected equally by increasing oil costs. Therefore, those
                         refiners whose costs increased the most may not be able to increase their
                         product prices to fully recoup the costs without losing sales to those
                         refiners whose costs did not rise by as much. Increases in crude oil costs
                         not passed on to consumers in the form of higher prices will reduce profit
                         margins for refiners. West Coast refiners we contacted did not reveal the
                         extent to which they passed on increased acquisition costs for crude oil to
                         consumers.



Consumers Were Not       We analyzed the differences between the prices of West Coast petroleum
                         products and the prices of the same products in other U.S. markets. Our
Significantly Affected   analysis indicates no significant changes in the prices of regular unleaded
by Lifting the Export    gasoline, diesel, and jet fuel as a result of lifting the export ban. In 1998,
                         these three products accounted for more than 80 percent of the total output
Ban                      of West Coast refineries, as well as the bulk of consumers’ expenditures on
                         petroleum products. These products were chosen because they are good
                         indicators of any potential change.



Exports Have Had a       Lifting the oil export ban has had a limited effect on the Alaskan North
                         Slope oil tanker fleet, the U.S. shipbuilding industry, and the West Coast
Limited Effect on        tanker repair industry. Overall, most tankers carrying Alaskan North Slope
Alaskan North Slope      oil continue to take the oil to the U.S. West Coast, and the demand for U.S.
                         tankers to transport Alaskan North Slope oil has continued to decline,
Oil Shipping             although exports have slightly offset the decline. Foreign-built tankers
                         have not been used to export Alaskan North Slope oil, and U.S.
                         shipbuilders have not lost orders for new tankers to foreign shipyards.
                         Furthermore, there has not been a trend toward more foreign repairs of
                         Alaskan North Slope tankers since exports began. Nevertheless, U.S.




                         Page 31                                          GAO/RCED-99-191 Alaskan North Slope Oil
                              Chapter 3
                              Effects of Lifting Oil Export Ban on Refiners,
                              Consumers, and Oil Shipping Industry on the
                              West Coast Have Generally Been Limited




                              shipbuilding and West Coast repair yard officials are concerned that they
                              may lose future business to foreign shipyards in part because of oil exports.


Effects on Tankers            Lifting the oil export ban has not greatly altered the number and routes of
Transporting Alaskan North    tankers used to transport Alaskan North Slope oil to date. While the 1995
                              law that lifted the ban does not require companies to use U.S.-built tankers
Slope Oil Have Been Limited   for export shipments, the fleet serving the Alaskan North Slope remains
                              basically domestic, both in vessel registration and shipment destinations.
                              Moreover, this fleet is almost entirely owned by, or under long-term-charter
                              to, the major Alaskan North Slope oil producers. The number of tankers
                              used to transport Alaskan North Slope oil from Valdez has been decreasing
                              steadily in the 1990s, as a result of the downward trend in Alaska oil
                              production. In 1998, the Valdez fleet had 30 tankers, compared with over 50
                              in 1990.

                              Lifting the ban has not significantly altered Alaskan North Slope shipping
                              operations. Most of the oil produced continues to be shipped to West Coast
                              refineries. A small percentage—about 5 percent—of the oil has been
                              exported since the export ban was lifted. The major oil producers in
                              Alaska ship most of their oil to West Coast states, particularly Washington
                              and California—to refineries around Puget Sound, San Francisco, and Los
                              Angeles. In 1998, the average volume shipped to West Coast refineries was
                              a little over one million barrels per day, carried by 30 tankers in 465
                              shipments. In comparison, only one major producer—BP-Amoco—has
                              been a significant exporter. Since exports began in May 1996, it has
                              exported an average of about 60,000 barrels per day. For example, in 1998,
                              five different tankers chartered to BP-Amoco took a total of 20 shipments
                              to Korea, China, Japan, and Taiwan. An Exxon tanker also took one
                              shipment to Japan in 1997 and one in 1998. Recent trends in major
                              destinations and volumes shipped are shown in table 3.1.




                                                                                                            (continued)



                              Page 32                                          GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 3
Effects of Lifting Oil Export Ban on Refiners,
Consumers, and Oil Shipping Industry on the
West Coast Have Generally Been Limited




Table 3.1: Destinations of Alaskan North Slope Oil Tankers and Volumes Carried,
1994-98

Thousands of barrels per day
Destination                                   1994         1995          1996         1997          1998
California/Washington                        1,279        1,229         1,205         1,113        1,026
Hawaii                                           53           49            50           39               43
Alaska                                          28            32            34           36               26
U.S. Gulf Coast via Panama                      75            62            4a             0               0
                       b                                                                   a
U.S. Virgin Islands                             95            88            48            5                0
Asia                                              0             0          36a           68               53
Total                                        1,530         1,460        1,377         1,261       1,149c
a
    On a full-year basis; for the partial year during which oil was shipped, per-day volume was higher.
b
    All shipments were made on foreign-flagged tankers.
c
    Does not add due to rounding.
Source: U.S. Maritime Administration.


As shown in table 3.1, the volume of oil shipped to Washington/California
and Hawaii has decreased gradually in recent years, while the volume
shipped to Alaska increased from 1994 through 1997, then decreased in
1998. At the same time, the volume shipped to the U.S. Gulf Coast via
Panama and to the U.S. Virgin Islands around Cape Horn fell to zero after
the export ban was lifted. According to federal maritime and industry
officials, both the U.S. Gulf Coast and U.S. Virgin Islands destinations were
declining even without the influence of exports because, compared with
U.S. West Coast destinations, they involve high shipping costs, especially
the shipments to the U.S. Gulf Coast. Some officials said that export
shipments in effect replaced the trade with the U.S. Virgin Islands and
accelerated its end.

Exports have affected some tanker operators more than others. Officials
of ARCO and Exxon, which have subsidiaries that own and operate tankers
in the Alaskan North Slope trade, said that because they have made few, if
any, export shipments, lifting the export ban has had little or no effect on
their Alaskan North Slope tanker fleets. However, officials of BP-Amoco
(which is not a U.S.-owned corporation and therefore is not permitted to
own tankers engaged in the U.S. domestic trade) said that exports to Asia
allow the company to lower its transportation costs and thus provide an
important new market. In addition, officials of the charter shipping
companies that carried exports for BP-Amoco said that the export
legislation benefited their business. These officials said that exports have



Page 33                                                  GAO/RCED-99-191 Alaskan North Slope Oil
                               Chapter 3
                               Effects of Lifting Oil Export Ban on Refiners,
                               Consumers, and Oil Shipping Industry on the
                               West Coast Have Generally Been Limited




                               slightly increased the demand for U.S. tankers to carry Alaskan North Slope
                               oil. According to officials of two companies, because of exports, a few of
                               their tankers that might otherwise have been unused were active in the
                               Alaskan North Slope fleet. Our analysis confirmed that while overall fleet
                               size continues to decrease, exports may have slightly increased the demand
                               for U.S. tankers in the Alaskan North Slope trade in 1996 and 1997.

                               Exports have led to the disappearance of foreign-registered tankers from
                               the Alaskan North Slope fleet and may therefore have caused an increase in
                               jobs for U.S.-tanker crews. Foreign tankers with foreign crews carried
                               Alaskan North Slope oil from Valdez to the U.S. Virgin Islands under a
                               long-standing exception in the Jones Act. As shown in table 3.1, before the
                               ban was lifted, oil was shipped from Valdez around Cape Horn to refineries
                               in the U.S. Virgin Islands. Several foreign-registered, foreign-crewed
                               tankers made these trips. According to our analysis, lifting the ban caused
                               these foreign tankers and crews to be replaced by U.S.-crewed tankers
                               going to Asia. Tankers carrying Alaskan North Slope oil from Valdez to
                               Asia to date have been U.S.-documented (including U.S.-registered and
                               U.S.-crewed) and U.S.-owned, as required by the 1995 legislation that lifted
                               the export ban. As a result of this change in destinations, the equivalent of
                               one or two additional U.S. tankers were used to carry Alaskan North Slope
                               oil in 1996 and 1997, creating an estimated 58 to 115 U.S. tanker crew jobs.1
                               These jobs partially offset the overall decrease in U.S. tanker crew jobs in
                               the Alaskan North Slope trade during the past decade caused by declining
                               crude oil production and fleet size.


Effects on U.S. Shipbuilding   To date, lifting the oil export ban has also had a limited effect on the U.S.
Have Been Limited              shipbuilding industry. Demand for new tankers for the Alaskan North
                               Slope trade—either U.S. or foreign-built—appears to be minimal at present
                               and driven primarily by factors other than exports. Since the export ban
                               was lifted, Alaskan North Slope tanker operators have had the option of
                               exporting oil in foreign-built tankers, but to date they have not done so.
                               Likewise, U.S. shipyards have not lost orders for new Alaskan North Slope
                               export tankers to foreign shipyards. Although several U.S. shipyards are
                               equipped to build Alaskan North Slope tankers, no U.S. shipyard has
                               delivered one since 1987. According to industry officials, U.S. shipbuilders
                               have been at a price disadvantage in the world commercial shipbuilding


                               1
                                Based on U.S. Maritime Administration estimates of 25 billets per tanker and 2.3 crew members per
                               billet.




                               Page 34                                             GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 3
Effects of Lifting Oil Export Ban on Refiners,
Consumers, and Oil Shipping Industry on the
West Coast Have Generally Been Limited




market because of, among other reasons, higher costs and less-modern
production methods.

U.S. shipbuilders and other industry officials expected 10 or more new
orders in the 1990s for tankers to serve the Alaskan North Slope. These
expectations resulted in part from the enactment of the Oil Pollution Act of
1990, in response to the Exxon Valdez accident.2 The act mandated, among
other things, the phaseout of single-hulled tankers and the transition to
double-hulled tankers by 2015, in order to reduce the effects of oil spills in
the event of accidents. However, only three orders have materialized so
far. All three orders were from ARCO for tankers to be built by Avondale,
Inc., of Louisiana, and to be delivered between 2000 and 2002. Additionally,
a proposed order from BP-Amoco for three tankers to be built by the
National Steel and Shipbuilding Company, of San Diego, was deferred
indefinitely in October 1998. According to industry officials, factors in the
lack of orders to date include falling oil prices in 1998 and their effect on
Alaskan North Slope planning and development, as well as the price of new
tankers—in some cases up to three times as much in U.S. shipyards
compared with overseas yards.

Despite the lack of tanker demand to date, there could be some demand for
new Alaskan North Slope tankers in the next decade, according to
shipbuilding and oil company officials. As shown in figure 3.1, under Oil
Pollution Act of 1990 requirements, 26 Alaskan North Slope tankers are due
to be phased out of the fleet by 2015.




2
On Mar. 24, 1989, the Exxon Valdez ran aground on a reef, spilling about 11 million gallons of Alaskan
North Slope crude oil into Prince William Sound.




Page 35                                              GAO/RCED-99-191 Alaskan North Slope Oil
                                               Chapter 3
                                               Effects of Lifting Oil Export Ban on Refiners,
                                               Consumers, and Oil Shipping Industry on the
                                               West Coast Have Generally Been Limited




Figure 3.1: Number of Alaskan North Slope Tankers Scheduled to Be Phased Out, 1999-2015
4.0   Number of tankers




3.0




2.0




1.0




0.0


        1999    2000      2001   2002   2003   2004   2005    2006     2007    2008    2009     2010   2011   2012   2013   2014   2015
        Years


                                               Source: U.S. Maritime Administration.


                                               As shown in figure 3.1, 19 tankers serving the Alaskan North Slope are to be
                                               phased out by the end of 2006. Some of these tankers, but not all, would
                                               need to be replaced, assuming that Alaskan North Slope production
                                               continues to decline. Oil companies would have replacement alternatives
                                               to new U.S.-built tankers, including (1) extending the life of existing
                                               tankers by converting the hulls and (2) using existing or new foreign-built
                                               tankers for exports. Oil company officials told us that their needs for
                                               future U.S. tankers will depend on various oil industry and market factors.
                                               Although introducing foreign-built tankers into the Alaskan North Slope



                                               Page 36                                            GAO/RCED-99-191 Alaskan North Slope Oil
                           Chapter 3
                           Effects of Lifting Oil Export Ban on Refiners,
                           Consumers, and Oil Shipping Industry on the
                           West Coast Have Generally Been Limited




                           trade to carry exports is an option, oil company officials told us they have
                           no plans to do so in the foreseeable future.

                           Nevertheless, officials in the U.S. shipbuilding industry said they are
                           concerned about losing future Alaskan North Slope tanker orders to
                           overseas shipyards, in part because of exports. They contend that the
                           export option gives oil companies an added incentive to further postpone
                           orders for new U.S.-built tankers. According to these shipbuilding officials,
                           foreign-built tankers to export Alaskan North Slope oil are a possibility
                           within a few years, if not immediately. If so, jobs in U.S. shipyards could be
                           affected. According to company officials, each tanker order postponed or
                           lost to a foreign competitor costs about 1,000 U.S. shipyard jobs for the 18
                           months it takes to construct a tanker.

                           In addition, postponed tanker orders contribute to the aging of the Alaskan
                           North Slope fleet, with a potential impact on fleet safety. Because no new
                           tankers have entered the fleet since 1987, half of the fleet consists of
                           single-hulled tankers built in the 1970s or before. Even though the oldest
                           tankers have been phased out of service, the phaseout has been so gradual
                           that, on average, the remaining fleet has gotten older. The average age of
                           the fleet has increased since the Oil Pollution Act of 1990 was passed—
                           from about 16 years old in 1990 to 21 years old in 1998. 3


Effects on West Coast      The ability to export Alaskan North Slope oil has given tanker operators an
Tanker Repair Yards Have   added incentive to repair tankers overseas rather than on the West Coast
                           because they can reduce costs by combining oil shipments to Asia with less
Been Limited               expensive Asian repairs. However, since the export ban was lifted, there
                           has not been a trend toward more overseas repairs. Tankers serving the
                           Alaskan North Slope undergo major, scheduled “drydock” repairs about
                           twice every 5 years at a cost of $1 million to over $10 million each. A
                           drydock repair can take a tanker out of service for several weeks. Exact
                           information on the number of Alaskan North Slope tanker repairs for
                           recent years was unavailable. However, according to data supplied by
                           industry officials, and on the basis of recent fleet size, we estimate that




                           3
                             The 1990 fleet included four tankers rebuilt in 1983 or earlier and the 1998 fleet included three such
                           tankers.




                           Page 37                                                GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 3
Effects of Lifting Oil Export Ban on Refiners,
Consumers, and Oil Shipping Industry on the
West Coast Have Generally Been Limited




about 10 to 15 such repairs have occurred annually for tankers serving the
Alaskan North Slope in recent years.

On average, repairs have been decreasing in the 1990s at a rate that is
commensurate with the decline in Alaskan North Slope production and
fleet size. Three West Coast repair yards, in California, Oregon, and
Washington State, compete with several Asian yards for the Alaskan North
Slope tanker repair business. These West Coast yards are situated near
Alaskan North Slope shipping lanes and destinations. However, according
to industry officials, the U.S. repair yards are at a competitive disadvantage
because Asian yards may charge less than half of what a U.S. yard would
charge for a comparable tanker repair.

Combining an oil shipment to Asia with a less expensive Asian repair
allows tanker operators to avoid the extra cost of going without oil cargo to
Asia for a repair. Overseas repairs of U.S. ships are subject to U.S. Customs
duties of 50 percent of certain repair costs levied on the vessel operator.
According to U.S. Customs and shipping industry data, overseas repairs of
Alaskan North Slope tankers have not increased significantly since the ban
was lifted, as shown in figure 3.2.




Page 38                                          GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 3
Effects of Lifting Oil Export Ban on Refiners,
Consumers, and Oil Shipping Industry on the
West Coast Have Generally Been Limited




Figure 3.2: Number of Overseas Repairs of Alaskan North Slope Tankers, 1989-98
8   Number of repairs



7



6



5



4



3



2



1



0


    1989    1990        1991   1992   1993   1994    1995   1996   1997   1998
    Years


Source: U.S. Customs Service.


As shown in figure 3.2, overseas repairs of Alaskan North Slope tankers
have averaged between three and four a year. No significant trend toward
more overseas repairs has developed since exports began. Of the nine total
overseas repairs since 1996, seven involved the tankers of one oil company
that has historically repaired its tankers overseas and has not been an
exporter of Alaskan North Slope oil.

Officials of the West Coast tanker repair industry said that their recent
experience raised concerns that a trend toward more foreign repairs of
Alaskan North Slope tankers could be beginning to develop, with exports
as a contributing factor. They cited two foreign repairs of Alaskan North
Slope tankers in Asia in 1998. In one of these cases, a tanker that
transported crude oil to Korea underwent a scheduled drydock repair in a
Korean shipyard before returning to the United States. According to West
Coast repair industry officials, this case illustrates how exports may be



Page 39                                             GAO/RCED-99-191 Alaskan North Slope Oil
Chapter 3
Effects of Lifting Oil Export Ban on Refiners,
Consumers, and Oil Shipping Industry on the
West Coast Have Generally Been Limited




starting to harm the West Coast ship repair industry. In the other case, the
tanker went without cargo to Singapore for a scheduled drydock repair.
According to operators involved in the two cases, a major factor in having
repairs done overseas was the significantly lower cost in Asian repair yards
compared with U.S. West Coast yards, even when U.S. Customs duties are
added and even without carrying cargo, as in the latter case. According to
West Coast repair industry officials, the two lost repairs represented
several million dollars in business and potential lack of employment for
over 500 workers a day for each repair.




Page 40                                          GAO/RCED-99-191 Alaskan North Slope Oil
Appendix I

Lifting the Oil Export Ban Has Had Little
Impact on the Alaskan Environment                                                                            ApIenxdi




               According to our discussions with Alaskan North Slope oil industry
               officials, state of Alaska officials, and environmental groups, lifting the
               export ban has had little effect on the Alaskan environment to date. Their
               opinions are based in part on the fact that export shipments from 1996 to
               1998 were more limited than projected by some analysts. Export
               shipments have not resulted in the use of larger tankers or new, potentially
               more hazardous shipping routes to Asia and have not significantly
               increased the risks of invading nonindigenous marine species, such as
               plankton and other organisms, into Alaskan waters.

               Prior to lifting the oil export ban, according to some projections, exports of
               200,000 barrels of oil per day or more were to be carried in Alaskan North
               Slope tankers that were of larger than average size—vessels with a capacity
               of 200,000 deadweight tons or more.1 In addition, some environmentalists
               foresaw the use of new, potentially more hazardous shipping routes to Asia
               along the Aleutian Islands coast instead of in deep water. However, this
               projected new tanker traffic has not materialized. There have been a
               relatively small number of export shipments to date—a total of 21 in 1998,
               for example--and existing tankers have been used. In addition, export
               tankers have been avoiding near-Aleutian routes in favor of already existing
               routes for which contingency planning and spill response teams are already
               in place. Furthermore, officials for the oil industry, Aleyska Pipeline
               Service Company (the organization that operates the Trans-Alaska Pipeline
               System), and U.S. Coast Guard said that they have not changed their
               operations as a result of the ban’s being lifted. Rather, any changes made
               resulted from the Exxon Valdez accident in 1989. The existing routes
               involve escorted transits out of Prince William Sound into deeper ocean,
               where tankers then head toward the U.S. mainland, Hawaii, or Asia.

               Nonindigenous marine species may come to Alaska in a tanker’s ballast
               water, which is routinely exchanged (taken on board and/or discharged
               from the vessel) either in port or under way, in order to maintain vessel
               stability. Potentially invasive species include plankton, crustaceans, and
               other organisms that could disturb the local marine ecology. However, the
               export-related invasion of Asian marine species does not appear to be a
               significant problem to date. Alaska officials and two representatives of an
               environmental group told us that this is an issue in Alaska, but not
               necessarily because of Alaskan North Slope oil exports. According to oil
               company officials, the operators of export tankers have added a second


               1
                   Deadweight tons are a measure of a ship’s cargo capacity.




               Page 41                                                 GAO/RCED-99-191 Alaskan North Slope Oil
Appendix I
Lifting the Oil Export Ban Has Had Little
Impact on the Alaskan Environment




mid-ocean ballast water exchange while under way toward Valdez in
addition to the one exchange required by federal law—the National
Invasive Species Act of 1996. They said this step helps to avoid the
invasion of nonindigenous species. In addition, according to an interim
report commissioned by the Regional Citizens’ Advisory Council of Prince
William Sound, species in ballast water from Asia may not as readily occur
or survive in Alaskan waters as species from Washington and California
ports.2 A final report on this matter is due to be issued later in 1999.




2
 Biological Invasion of Cold-Water Coastal Ecosystems: Ballast Mediated Introduction in Port
Valdez/Prince William Sound, Alaska—1998 Progress Report, Dec. 3, 1998, for Regional Citizens’
Advisory Council of Prince William Sound.




Page 42                                              GAO/RCED-99-191 Alaskan North Slope Oil
Appendix II

Methodology for Estimating the Effect of
Lifting the Alaskan Oil Export Ban on Crude
and Petroleum Product Prices                                                                                                   ApIpx
                                                                                                                                   Ien
                                                                                                                                     di




                        This appendix provides details on the statistical analysis we used to explain
                        and estimate how lifting the ban on Alaskan oil exports has affected crude
                        oil and petroleum product prices on the U.S. West Coast. We conducted
                        statistical analyses using data on oil prices before and after the ban was
                        lifted to determine whether the effects on price predicted by previous
                        studies were borne out. We used a similar analysis to determine what
                        effect the ban’s lifting had on the prices charged for gasoline, diesel, and jet
                        fuel on the West Coast.

                        In summary, our statistical analysis indicates that lifting the export ban led
                        to relative prices for Alaskan crude oil and for comparable California crude
                        oil that were higher than they would have been had the ban remained in
                        force. However, lifting the ban had no statistically significant effect on the
                        prices of gasoline, diesel, or jet fuel. These results are consistent with
                        predictions of earlier studies. In addition, we found that sales of Alaskan
                        crude oil destined for the U.S. Virgin Islands and the U.S. Gulf Coast ended
                        and sales to the Mid-Continent fell abruptly once exports became a feasible
                        alternative. Relatedly, we found that the proportion of Alaskan oil sold on
                        the West Coast has risen over time as Alaskan and California production
                        have fallen.



Prior Studies Found     Several analyses by the Department of Energy, GAO, and the private sector
                        found that West Coast prices of Alaskan oil were lower as a result of
That the Ban Resulted   banning exports of this oil.1 These studies concluded that this ban resulted
in Lower Oil Prices     in an abundance of crude oil relative to refining demand on the West Coast,
                        leading to lower crude oil prices there. The studies contend that lifting the
                        export ban would result in Alaskan oil being sold in Asian markets, causing
                        its price to rise on the West Coast. Because California oil is also refined on
                        the West Coast, higher Alaskan prices would tend to increase refiners’
                        demand for California oil, pushing up its price as well. The effects of lifting
                        the ban on oil production, refining costs, and consumer prices were also
                        predicted in these studies. For example, higher oil prices were expected to
                        encourage more oil production while also leading to higher costs for
                        refiners buying that oil and possibly higher prices for consumer petroleum
                        products.



                        1
                         Exporting Alaskan North Slope Crude Oil: Benefits and Costs, U.S. Department of Energy (June 1994),
                        Alaskan Crude Oil Exports (GAO/T-RCED-90-59, Apr. 5, 1990) , and Samuel A. Van Vactor, Time to End
                        the Alaskan Oil Export Ban, Policy Analysis, May 18, 1995, No. 227.




                        Page 43                                             GAO/RCED-99-191 Alaskan North Slope Oil
                              Appendix II
                              Methodology for Estimating the Effect of
                              Lifting the Alaskan Oil Export Ban on Crude
                              and Petroleum Product Prices




Alaskan Oil Is No Longer      Sales of Alaskan oil to the U.S. Gulf Coast ended about 5 months before the
Sold in the U.S. Gulf Coast   first exports to Asia, and sales to the U.S. Virgin Islands ended about 8
                              months later. Sales to the Mid-Continent states dropped off abruptly when
and U. S. Virgin Islands      the ban was lifted. Before the ban was lifted, total sales to these markets
                              were falling as Alaskan oil production fell from its peak in 1988. Whether
                              the lifting of the ban, rather than falling production, caused the cessation of
                              sales to the Gulf Coast and Virgin Islands is unclear. However the abrupt
                              drop in sales to the Gulf Coast, Virgin Islands, and Mid-Continent states
                              coincided with the ban’s removal.

                              Figures II.1 and II.2 illustrate the observations listed above. As production
                              has fallen, the proportion sold to West Coast refiners has risen, and sales to
                              alternative markets have declined. Finally, since the ban was lifted, some
                              sales to these other markets have been replaced by Asian sales.



                              Figure II.1: Proportion of Alaskan North Slope Oil Sold in the West Coast Market
                              120% Proportion of oil



                              100%



                               80%



                               60%



                               40%



                               20%



                                0%



                                        1994           1995        1996            1997            1998
                                         Years


                                                   Alaskan North Slope oil sold in other markets


                                                   Alaskan North Slope oil sold in West Coast

                              Source: U.S. Maritime Administration Vessel Loading and Destination reports.




                              Page 44                                                  GAO/RCED-99-191 Alaskan North Slope Oil
                         Appendix II
                         Methodology for Estimating the Effect of
                         Lifting the Alaskan Oil Export Ban on Crude
                         and Petroleum Product Prices




                         Figure II.2: Alaskan North Slope Oil Destinations Other Than the West Coast

                         30   Millions of barrels of Alaskan North Slope oil



                         25




                         20




                         15




                         10




                          5




                          0
                                1996                       1997                   1998
                                Years

                                         Asia

                                          Mid-Continent

                                         Virgin Islands


                         Note: The 1996 figures for Asia are for May 31 through December 31, corresponding to the period in
                         1996 during which oil was exported to Asia.
                         Source: U.S. Maritime Administration Vessel Loading and Destination reports.




Statistical Analysis     As predicted by the studies described above and as we found in our
                         analyses of the prices of oil before and after the ban was lifted, the price of
Indicates That Lifting   Alaskan North Slope oil has risen relative to other oil as a result of lifting
the Ban Resulted in      the export ban. We also found that the price of a comparable blend of
                         California oil--Line 63--rose as a result of lifting the ban.
Higher Prices for
Alaskan and              To determine the effect of lifting the ban on oil prices, we developed a
Comparable California    time-series model. Because oil prices are influenced by many factors other
                         than removing the ban, we had to control for these other factors. We did
Oil                      this by modeling the differentials between the prices of West Coast oil and
                         the prices of similar oil in other markets. Modeling the price differentials



                         Page 45                                               GAO/RCED-99-191 Alaskan North Slope Oil
Appendix II
Methodology for Estimating the Effect of
Lifting the Alaskan Oil Export Ban on Crude
and Petroleum Product Prices




between two crude oils is a way to control for all the factors that affect the
two oils similarly—such as changes in global supply and demand—while
capturing changes in the local markets of the two oils that affect only one
of the oils. To control for local factors affecting the prices of individual
comparison oils, we examined three price differentials for Alaskan North
Slope (ANS) oil and three for Line 63 (L63). The comparison oils, making
up the other part of the differentials are Brent Blend (BB), Nigerian
Forcados (FOR), and West Texas Intermediate (WTI). For each of the six
differentials, the price differential is defined as the price of the comparison
oil minus the price of the West Coast oil—be it Alaskan North Slope or Line
63 oils. We used daily spot price data as reported in Platts Oil Prices Data
Base. The data series we used run from January 8, 1992 ,through December
4, 1998, and are in nominal dollar terms.2

To compare the price differential before and after the export ban was lifted,
we included in the regression a dummy variable to indicate when the ban
was removed. Diagnostic tests indicated that the price differential series
was auto-regressive of order one. Therefore, we included a lagged price
differential as a regressor. Diagnostic tests also revealed that the residuals
were auto-regressive conditionally heteroskedastic. To correct for this, we
included a GARCH(1,1) component to the regression.3 The estimated form
of the equation of the price differential is
DIFFt = α + βDIFFt−1 + γEXPORTt + ε t ,

where DIFF is the difference between the prices of the comparison and
West Coast oils; EXPORT is a dummy variable to indicate the removal of
the ban; α, β, and γ are parameters to be estimated; and ε is a random error
term.4 The subscript t denotes time. The parameter γ does not measure the
marginal effect of lifting the ban on the price differential because of the
auto-regressive property of the model.5 The results of maximum likelihood
estimation are shown in tables II.1 and II.2.



2
 We used daily spot price data rather than constructing weekly or monthly averages of this data
because we wanted to examine prices that reflect actual transactions. It is possible that using weekly
or monthly averages would change the results of the statistical analysis but we did not explore this
approach.
3
 The details of the diagnostic tests and the choice of the final form of the regression model are
discussed in detail below.
4
 The dummy variable takes on the value of unity for dates on or after May 28, 1996, and is otherwise
equal to zero.




Page 46                                                GAO/RCED-99-191 Alaskan North Slope Oil
Appendix II
Methodology for Estimating the Effect of
Lifting the Alaskan Oil Export Ban on Crude
and Petroleum Product Prices




Table II.1: Results of Price Analysis for Alaskan North Slope (ANS) Oil

Coefficients                                    WTI-ANS                  BB-ANS                FOR-ANS
CONSTANT                                            0.037*                  0.019*                  0.022*
                                                   (0.003)                 (0.005)                 (0.006)
DIFFt-1                                             0.986*                  0.976*                  0.982*
                                                   (0.002)                 (0.003)                 (0.003)
EXPORT                                             -0.018*                 -0.024*                  -0.02*
                                                   (0.004)                 (0.007)                 (0.008)
 Γ
-Γ                                                   1.295                   0.979                     1.102
Adj. R-Squared                                       0.976                   0.946                     0.952
Note: Standard errors are in parentheses. An asterisk denotes significance at the 5-percent level.


The long-term increase in the relative price of Alaskan North Slope oil
relative to the three comparison oils ranges from $0.98 to $1.30 as shown in
the -Γ cells. All coefficient estimates are significant at the 5-percent level.



Table II.2: Results of Price Analysis for Line 63 (L63) Oil

Coefficients                                     WTI-L63                  BB-L63                FOR-L63
CONSTANT                                            0.051*                  0.017*                  0.025*
                                                   (0.007)                 (0.008)                 (0.009)
DIFFt-1                                             0.986*                  0.987*                  0.986*
                                                   (0.003)                 (0.003)                 (0.003)
EXPORT                                            -0.018*                   -0.005                     -0.012
                                                  (0.006)                  (0.009)                      (0.01)
 Γ
-Γ                                                  1.282                         a                         a
Adj. R-Squared                                      0.975                   0.956                      0.959
a
    Not applicable. The export coefficient was not significant, so we did not calculate the Γ value.
Note: Standard errors are in parentheses. An asterisk denotes significance at the 5-percent level.




5
 The auto-regressive structure of the model requires that the long-run impact of lifting the ban be
calculated as follows:

            1
Γ =γ
           1− β
where Γ is the long-term effect of lifting the ban on the differential, and γ and β are as defined above.
The negative of Γ can be interpreted as the increase in the price of the West Coast oil relative to the
comparison oil. These values are listed in tables II.1 and II.2 for the cases in which the EXPORT
variable was statistically significant.




Page 47                                                  GAO/RCED-99-191 Alaskan North Slope Oil
                             Appendix II
                             Methodology for Estimating the Effect of
                             Lifting the Alaskan Oil Export Ban on Crude
                             and Petroleum Product Prices




                             The price of Line 63 oil rises significantly compared with West Texas
                             Intermediate but does not rise significantly compared with either Brent
                             Blend or Nigerian Forcados. In the latter two cases, the direction of change
                             is the same and the order of magnitude of γ is similar to the significant case.
                             However, because the coefficients are not statistically significant, we did
                             not calculate the long-term effect.



No Effect Found on the       We conducted a similar analysis using the prices of two heavy California
Prices of California Heavy   oils—Kern River and THUMS—and found no significant changes when the
                             export ban was lifted. We used daily spot price data as reported in Platts
Crude Oils                   Oil Prices Data Base. The data series we used run from January 8, 1992,
                             through December 4, 1998, and are in nominal dollar terms. We compared
                             the prices of these two oils to the prices of Duri, Indonesia (DU); Shengli,
                             China (SH); and West Texas Intermediate oils. West Texas Intermediate
                             was chosen to provide some consistency between these and the previous
                             regression results.6 The other two oils were chosen because, like Kern
                             River and THUMS, they are heavy oils and therefore their prices should be
                             expected to be affected similarly by global market conditions. The
                             regression results follow.

                             Kern River oil shows no significant change in price compared to any of the
                             comparison oils, as shown in table II.3.



                             Table II.3: Results of Price Analysis for Kern River (KERN) Oil

                             Coefficients                                  WTI-KERN                  SH-KERN                DU-KERN
                             CONSTANT                                            0.186*                  0.048*                 0.0574*
                                                                                (0.041)                 (0.013)                  (0.014)
                             DIFFt-1                                             0.969*                  0.984*                  0.982*
                                                                                (0.006)                 (0.004)                 (0.004)
                             EXPORT                                              -0.013                   0.002                  -0.001
                                                                                (0.019)                 (0.012)                 (0.013)
                              Γ
                             -Γ                                                        a                       a                       a
                             Adj. R-Squared                                       0.917                   0.958                     0.952
                             a
                                 Not applicable. The export coefficient was not significant, so we did not calculate the Γ value.


                             6
                              In the case of the WTI-Kern and WTI-THUMS differential models, we also ran regressions that
                             included a world “light-heavy” differential (specifically, Brent Blend and Duri, Indonesia) to control for
                             structural changes in the light-heavy differential in the mid-1990s. Adding this term in the regression did
                             not change the significance of the EXPORT coefficient, so we do not report these results.




                Leetr        Page 48                                                   GAO/RCED-99-191 Alaskan North Slope Oil
                             Appendix II
                             Methodology for Estimating the Effect of
                             Lifting the Alaskan Oil Export Ban on Crude
                             and Petroleum Product Prices




                             Note: Standard errors are in parentheses. An asterisk denotes significance at the 5-percent level.


                             As with the Kern River results, there was no significant change in the price
                             of THUMS relative to any comparison oil, as shown in table II.4.



                             Table II.4: Results of Price Analysis for THUMS Oil

                             Coefficients                                WTI-THUMS                SH-THUMS                DU-THUMS
                             CONSTANT                                            0.144*                 0.027*                      0.031*
                                                                                (0.033)                  (0.01)                    (0.011)
                             DIFFt-1                                              0.97*                 0.986*                      0.986*
                                                                                (0.006)                (0.004)                     (0.004)
                             EXPORT                                              -0.016                 -0.005                      -0.009
                                                                                (0.018)                (0.011)                     (0.012)
                              Γ
                             -Γ                                                        a                      a                         a
                             Adj. R-Squared                                       0.913                  0.956                       0.95
                             a
                                 Not applicable. The export coefficient was not significant so we did not calculate the Γ value.
                             Note: Standard errors are in parentheses. An asterisk denotes significance at the 5-percent level.




Statistical Analysis         Our analysis of prices of three key consumer products—gasoline, diesel
Indicates No Effect on the   fuel, and jet fuel—indicated that lifting the ban had no statistically
                             significant effect on them. Because higher crude oil prices translate
Prices of Petroleum          directly into higher refiner costs for refiners, it would not be unusual to
Products                     find that some of the increase in costs was passed on to consumers in the
                             form of higher prices for petroleum products. However, refiners whose
                             costs rose as a result of removing the ban may not have been able to pass
                             these costs on to consumers because of competition from imported final
                             products or from other West Coast refiners whose costs did not rise.

                             To determine the effect on petroleum prices of removing the ban, we
                             conducted an analysis of petroleum prices that was similar to the crude oil
                             price analysis described above. We used daily spot price data as reported in
                             Platts Oil Prices Data Base. The data series we used run from January 8,
                             1992, through December 4, 1998, and are in nominal dollar terms. We
                             compared West Coast prices of regular unleaded gas, diesel fuel, and jet
                             fuel with prices of these same products in other markets. The time series
                             of the petroleum product price differentials and the crude oil price
                             differentials behaved similarly, making the same model structure
                             appropriate. Our analysis revealed no statistically significant change in
                             West Coast petroleum product prices that was not explained by similar



                             Page 49                                                   GAO/RCED-99-191 Alaskan North Slope Oil
Appendix II
Methodology for Estimating the Effect of
Lifting the Alaskan Oil Export Ban on Crude
and Petroleum Product Prices




changes in the prices of these products in other markets. Specifically, we
did not find any change in the prices at the time the ban was lifted,
indicated by the absence of a statistically significant estimated β coefficient
for the export dummy variable. The results of our regressions are listed in
tables II.5 through II.7.



Table II.5: Analysis of Jet Fuel (JET) Prices

Coefficients                                                 LA-CHIJET                        SF-CHIJET
CONSTANT                                                            0.622*                          0.581*
                                                                   (0.079)                         (0.081)
DIFFt-1                                                             0.965*                          0.966*
                                                                   (0.005)                         (0.005)
EXPORT                                                              -0.027                           -0.02
                                                                   (0.043)                         (0.045)
 Γ
-Γ                                                                        a                               a
Adj. R-Squared                                                      0.922                              0.921
a
    Not applicable. The export coefficient was not significant, so we did not calculate the Γ value.
Note: Standard errors are in parentheses. An asterisk denotes significance at the 5-percent level.


Lifting the ban had no significant effect on the prices of jet fuel in Los
Angeles (LA) or San Francisco (SF) when compared with the price of jet
fuel in Chicago (CHI).



Table II.6: Analysis of Diesel Prices

Coefficients                                                                          SF-GULFDIESEL
CONSTANT                                                                                            0.417*
                                                                                                   (0.083)
DIFFt-1                                                                                             0.972*
                                                                                                   (0.005)
EXPORT                                                                                               0.028
                                                                                                   (0.051)
 Γ
-Γ                                                                                                        a
Adj. R-Squared                                                                                         0.941
a
    Not applicable. The export coefficient was not significant, so we did not calculate the Γ value.
Note: Standard errors are in parentheses. An asterisk denotes significance at the 5-percent level.


Diesel fuel prices in San Francisco were also unaffected by lifting the
export ban when compared with diesel prices in the Gulf Coast (GULF).



Page 50                                                  GAO/RCED-99-191 Alaskan North Slope Oil
                           Appendix II
                           Methodology for Estimating the Effect of
                           Lifting the Alaskan Oil Export Ban on Crude
                           and Petroleum Product Prices




                           Table II.7: Analysis of Gasoline (GAS) Prices

                           Coefficients                                              LA-GULFGAS                      SF-GULFGAS
                           CONSTANT                                                            0.236*                          0.101*
                                                                                              (0.029)                         (0.031)
                           DIFFt-1                                                             0.961*                          0.969*
                                                                                              (0.004)                         (0.005)
                           EXPORT                                                              -0.011                             0.044
                                                                                              (0.054)                             (0.06)
                            Γ
                           -Γ                                                                        a                                a
                           Adj. R-Squared                                                      0.934                               0.94
                           a
                               Not applicable. The export coefficient was not significant, so we did not calculate the Γ value.
                           Note: Standard errors are in parentheses. An asterisk denotes significance at the 5 percent level.


                           Again, there was no significant change in the West Coast prices of regular
                           unleaded gasoline as a result of lifting the export ban.



Further Evidence of an     Higher market prices for Alaskan North Slope oil and lower shipping costs
Increase in the Price of   mean higher revenues for producers of this oil. The impact of lifting the
                           ban on Alaskan North Slope producers’ revenues net of transportation
Alaskan North Slope Oil    costs can be seen in figure II.3. This figure shows the prices received at the
                           “wellhead” for two oils—West Texas Intermediate and Alaskan North
                           Slope—as well as the difference between their respective prices. These
                           so-called wellhead prices reflect the revenue the producer receives net of
                           transportation and shipping costs. Rising market prices for Alaskan North




                           Page 51                                                  GAO/RCED-99-191 Alaskan North Slope Oil
                             Appendix II
                             Methodology for Estimating the Effect of
                             Lifting the Alaskan Oil Export Ban on Crude
                             and Petroleum Product Prices




                             Slope oil and falling shipping costs both contribute to the shrinking of the
                             gap between the two oils’ wellhead prices.



                             Figure II.3: ANS Wellhead Price Has Risen Compared With WTI
                             25      Dollars per barrel




                             20




                             15



                             10




                                 5




                                 0


                                       1994                   1995             1996                  1997                  1998
                                       Years

                                                   WTI wellhead price

                                                   ANS wellhead price

                                                   WTI-ANS differential

                             Source: GAO’s analyses of Energy Information Administration data.




Tests for Stationarity and   To ensure that we estimated the correct model of the price differentials, we
Diagnostics: Development     checked for stationarity of the price differential time series and did some
                             standard diagnostic testing. These tests helped us to develop the final form
of the Correct Statistical   of the model and give us confidence that the results of the estimations are
Model                        not spurious. A detailed description of the development of the model
                             follows.7



                             7
                              The process described for choosing the correct statistical model was followed for each oil price
                             differential and for each petroleum price differential.




                             Page 52                                               GAO/RCED-99-191 Alaskan North Slope Oil
Appendix II
Methodology for Estimating the Effect of
Lifting the Alaskan Oil Export Ban on Crude
and Petroleum Product Prices




We chose a period of study of January 8, 1992, through December 4, 1998,
and used daily spot prices as reported by Platts Oil Prices Data Base. This
period was chosen in order to encompass the removal of the export ban
with sufficient time on either side of this event. An arithmetic mean of the
low and high spot prices was used for the analysis. All tests for stationarity
and diagnostics were performed on data prior to May 28, 1996.

The simplest model of the price differential is a standard ordinary least
squares (OLS) regression of the price differential on a constant term with a
dummy variable to pick up the effect of lifting the export ban. Test
statistics derived from an OLS regression of a time series will not be
reliable if the time series is not stationary.8 Therefore, the first step is to
test for stationarity of the series of price differentials.

To test for stationarity, we used the Augmented Dickey-Fuller (ADF) test.
First, we estimate the following equation using OLS:
                                                      ρ
                        ∆PDt = α + φPDt−1 + ∑ β t − j ∆PDt− j + ε t
                                                      j =1

where PD is the difference between the prices of the comparison oil and
Alaskan North Slope oil; ∆ is the difference operator; α, φ, and β are
parameters to be estimated; and ε is a random error term. The subscript t
denotes time. The number of lags of PD (denoted by ρ) is chosen by
starting with a large number of lags and sequentially dropping the
statistically insignificant lags from the highest lag down. Lags greater than
ρ have a statistically insignificant effect on the regressand.

A test for nonstationarity amounts to a test of the null hypothesis that φ is
equal to zero. If φ is equal to zero, the time series PDt has a unit root and
will behave analogously to a random walk, which is nonstationary.
Alternatively, if φ is negative and statistically significant, then the time
series (expressed as deviations from the mean) will converge to zero in
response to shocks. Using the period prior to the lifting of the export ban,
the ADF test rejects the null hypothesis of nonstationarity at the 5-percent




8
 A time series is stationary if its mean, variance, and autocovariances are independent of time and
nonstationary otherwise. If the price differential we model is stationary, then we would expect to see
that the mean of the differential is unchanging over time except when it changes in response to an
event, such as the removal of the export ban.




Page 53                                               GAO/RCED-99-191 Alaskan North Slope Oil
Appendix II
Methodology for Estimating the Effect of
Lifting the Alaskan Oil Export Ban on Crude
and Petroleum Product Prices




level.9 We also performed the Phillips-Perron test and were able to reject
nonstationarity at the 5-percent level.

Results of these tests indicate that standard regression techniques are
appropriate for modeling the price differentials. Next, we performed some
diagnostic tests to determine whether lagged values of the differential
affect current values. More specifically, we checked the order of the model
by examining the correlogram and partial correlogram. The correlogram
showed a steady decline in the size of the coefficients after the first lag of
the dependent variable, suggesting an auto-regressive order 1 (AR(1))
process. The partial correlogram revealed a strong partial correlation
between PDt and PDt-1 but very small partial correlations for PDt and PDt-k
for k>1. This further indicated an AR(1) process. Next, we estimate the
AR(1) model using OLS:

PDt = β k PDt− k + ε t

We performed some diagnostic tests of the residuals of the regression.
Specifically, we tested for autoregressive conditional heteroskedasticity
(ARCH) using a Lagrange Multiplier test. ARCH residuals have the
characteristic that high values of the estimated residuals bunch together
temporally. ARCH residuals are quite common in time series analysis of
economic variables. The Lagrange Multiplier test allows us to reject at the
1-percent level the null hypothesis that the residuals are not ARCH.

We re-estimate the model using the maximum likelihood method and
including a GARCH(1,1) component to correct for the ARCH residuals.10
This is the most standard version of this type of model. The correlogram
and partial correlogram of the residuals of the new specification reveal no
further ARCH terms.11

Finally, we estimate the model over the entire time period, including the
dummy variable for exports. The model was estimated using the entire
sample and adding an indicator variable for dates after the export ban was


9
  We used the period before the ban was lifted to test for stationarity because if we had included the
lifting of the ban and the period following that event, then the differential might change as a result of
lifting the ban. This, in turn, could cause us to incorrectly conclude that the time series is
nonstationary.
10
  The GARCH(1,1) estimation includes an auto-regressive term and a moving-average term to correct
for the existence of ARCH residuals. The price differential and variance equations are estimated
simultaneously.




Page 54                                                 GAO/RCED-99-191 Alaskan North Slope Oil
Appendix II
Methodology for Estimating the Effect of
Lifting the Alaskan Oil Export Ban on Crude
and Petroleum Product Prices




lifted. (We used May 28, 1996.) The equation is estimated using maximum
likelihood methods and including a GARCH(1,1) component to correct for
ARCH residuals. The final price differential model is

PDt = α + βPDt−1 + γδ t + ε t .




11
  To further ensure that the price differential model did not have a unit root or was otherwise
misspecified, we performed a test for model specification developed by Davidson, Godfrey, and
MacKinnon (1985). This is equivalent to the Plosser-Schwert-White (1982) differencing test. The test
statistics for the Davidson, Godfrey, and MacKinnon test do not allow us to reject the null hypothesis
that the differenced and undifferenced models result in identical parameter estimates. This is strong
evidence that our original model is properly specified.




Page 55                                               GAO/RCED-99-191 Alaskan North Slope Oil
Appendix III

GAO Contacts and Staff Acknowledgments                                                              AIpIenxdi




GAO Contacts           Daniel Haas (202) 512-9828




Acknowledgments        In addition to the above, Charles Bausell, Dave Brack, Rodney Conti, Byron
                       S. Galloway, Sterling Leibenguth, and Frank Rusco made key contributions
                       to this report.




(141229)       Leetr   Page 56                                GAO/RCED-99-191 Alaskan North Slope Oil
Ordering Information

The first copy of each GAO report and testimony is free.
Additional copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order made
out to the Superintendent of Documents, when necessary, VISA and
MasterCard credit cards are accepted, also.

Orders for 100 or more copies to be mailed to a single address are
discounted 25 percent.

Orders by mail:

U.S. General Accounting Office
P.O. Box 37050
Washington, DC 20013

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting Office
Washington, DC

Orders may also be placed by calling (202) 512-6000
or by using fax number (202) 512-6061, or TDD (202) 512-2537.

Each day, GAO issues a list of newly available reports and
testimony. To receive facsimile copies of the daily list or any list
from the past 30 days, please call (202) 512-6000 using a touchtone
phone. A recorded menu will provide information on how to obtain
these lists.

For information on how to access GAO reports on the INTERNET,
send an e-mail message with “info” in the body to:

info@www.gao.gov

or visit GAO’s World Wide Web Home Page at:

http://www.gao.gov
United States                       Bulk Rate
General Accounting Office      Postage & Fees Paid
Washington, D.C. 20548-0001           GAO
                                 Permit No. GI00
Official Business
Penalty for Private Use $300

Address Correction Requested
PAGE 59   GAO/XXXX-98-??? NAME OF DOCUMENT
Contents




Table 2.1: Cost to Ship Alaskan North Slope Oil, by Destination, 1996   24
Table 3.1: Destinations of Alaskan North Slope Oil Tankers and Volumes
  Carried, 1994-98                                                      32
Table II.1: Results of Price Analysis for Alaskan North Slope (ANS) Oil 47
Table II.2: Results of Price Analysis for Line 63 (L63) Oil             47
Table II.3: Results of Price Analysis for Kern River (KERN) Oil         48
Table II.4: Results of Price Analysis for THUMS Oil                     49
Table II.5: Analysis of Jet Fuel (JET) Prices                           50
Table II.6: Analysis of Diesel Prices                                   50
Table II.7: Analysis of Gasoline (GAS) Prices                           51




Page 60                                                     GAO/XXXX ?????
Contents




Page 61    GAO/XXXX ?????
Contents




Figure 1: Locations of Alaska Oil Fields and Tanker Routes From Valdez,
  Alaska, to Refineries That Received Alaskan North Slope Oil, 1977-98 4
Figure 1.1: Locations of Alaska Oil Fields                                12
Figure 1.2: Percent of Oil Removed From Alaskan North Slope and Percent
  of Production Rights in Prudhoe Bay Oil Field, by Oil Major Company, in
  1998                                                                    14
Figure 1.3: Shipping Routes for Alaskan North Slope Oil Tankers, 1989-98
  16
Figure 2.1: New Fields Will Add to Daily Alaskan North Slope Oil Production
  26
Figure 2.2: Annual Alaska and California Crude Oil Production, 1989-98 28
Figure 3.1: Number of Alaskan North Slope Tankers Scheduled to Be
  Phased Out, 1999-2015                                                   36
Figure 3.2: Number of Overseas Repairs of Alaskan North Slope Tankers,
  1989-98                                                                 39
Figure II.1: Proportion of Alaskan North Slope Oil Sold in the West Coast
  Market                                                                  44
Figure II.2: Alaskan North Slope Oil Destinations Other Than the West Coast
  45
Figure II.3: ANS Wellhead Price Has Risen Compared With WTI               52




Page 62                                                       GAO/XXXX ?????