Energy Security: Impacts of Lifting Alaskan North Slope Oil Exports Ban

Published by the Government Accountability Office on 1990-11-08.

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

    GA,           >              United States
                                 General Accounting    Office
                                 Washington,   D.C. 20648

                                 Resources, Community,             and
                                 Economic Development              Division


                                 November       8,199Q

                                   The Honorable Philip R. Sharp ”
                                   Chairman, Subcommittee on Energy and Power
                                I, Committee on Energy and Commerce
                              ” ‘. House pf Representatives
                                 Dear,Mr! Chairman:                           ”:”

                                 your letterof February 9,1989, asked us to assessthe impacts of
                                 removing export restrictions on Alaskan North Slope (ms) crude oil. In
                                 April of this year, while our study was ongoing, we testified with your                                               i
                                 concurrence on this subject before the Subcommittee on International
                                 Economic Policy and Trade, House Committee on Foreign Affairs.                                                        .I
/                                Although our analysis was completed prior to the August 1990 Iraqi
              0                  invasion ,of Kuwait, we have recognized the potential impact of the inva-                                           I
                                 sion in this report. As agreed with your office, our review focused-on                                            X ~
                                 likely,‘changes”m the Alaskan oil trade between now and 1996, both with                                               t

              2                  and without the ban, and how these changes will affect the nation’s       ’
                                 economy and energy security.         ’                                                                            #

    Results iii Brief          I If.the ban on exporting ,ANScrude oil remains in place, ANSproduction
                                 will, of course, continue to go to US. ports. However, because of
                      I          declining ANS production, shipments to eastern US. ports, i.e., those on
                                 the East Coast, the Caribbean, and the Gulf of Mexico, will probably
                          I      cease at, some time, in the next. several years. Producersof ANs crude
                                 prefer to sell theircrude to West Coast refiners, given the cost ‘of trans-
                                 porting it to-East Coast refiners.

                                 If the b& on exporting ANS crude oil i6 removed, some of it is likely to be
                                 exported to Pacific Rim countries. Since transportation costs to Pacific
                                 Rim ports are much less’than those to eastern U.S. ports, oil that is cur-
                                 rently transported to the eastern United. States is likely to be exported.
                                 In addition, some MS crude that would have gone to the U.S. West Coast
                                 may also be exported, since the cost of transporting oil to some Pacific
                                 Rim destinations is comparable to, if not lower than, the cost to U.S.
                                 West Coast ports. In this regard, the heavier weight of ANScrude is more
                                 likely to be attractive to,refiners in Pacific Rim countries than it is to
                                 US. West Coast refiners, who refine more of their oil into light products,
                                 such as gasoline.

                                 Page       1            GAO/RCJW91-21        Impacts   of Lifting   Alaskan   North   Slope   Oil Exports   Ban
                                           ::_,,                                                           ‘,,
                                               .          ‘I        ,,’

                        The probable economic effects of lifting the ban on ANScrude (as com-
                        pared with leaving it in place) will be to                                                              .

                    . increase the price of Ays crude’at’the wellhead-because of the .reduc-
                      tion intransportation’costs and the attractiveness of ANS crude to
                      Pacific Rim refiners;-a&,consequently,      the price that West Coast
                       refiners pay for crude oil;
                    l  promote economic efficiency by reducing transportation costs in the ANS
                       crude oil trade, increasing domestic oil production, allowing better use
                8      of refinery processing:,resources,and ensuring that ANSoil is allocated to
                       its ‘hi&h&i valued, &S;‘&d:        I’ :
                    . accelerate the declinein tanker demand and hurt the U.S; maritime’
                       industry because ANSexports are likely to be transported on foreign-flag
                      rather than U.S.-flag tankers. ”                                                                              /
                                                ‘il.  ),
           ,.           The energy supply disruption resulting fromIraq’s invasion of Ruwait
                        has focused attention on U.Si energy security and, in particular, our reli-
                        ance on imported oil. From an energy security standpoint,, the effect of
                        lifting.the ANSexport ban would probably be to increase total U.S. oil                                  ,
                        imports but possibly decrease net imports (total imports minus exports)
                        to the extent that refinery efficiency is improved and ANS oil production                               )
                        increases in response to higher prices. Finally, lifting the ban could also                             4
“./   .’                contribute to the integrated world market’s smooth and efficient
                        functioning.            : .’ ‘,    ,. ..,’

                         domestically produced crude oil transported through the Alaskan pipe
                         line may be exported from the United States.” The purpose of this ban
                         was to restrict~“the exportof goods where necessary to protect the
                         domestic economy from excessive drain of scarce materials and to
                         reduce the serious inflationary impact of foreign demand.” This provi-
                         sion of the law was part of the compromise that permitted the construc-
                         tion of the Trans-Alaska Pipeline. The act allows the ban to be lifted
                         only upon the President’s certification that the export of Alaskan oil is
                          in the national interest and meets’.several other specified conditions.


                         ‘The export of U.S. domestically produced crude oil is generally prohibited, including domktic crude
                         transported by pipeline over certain rights-of-way, petroleum produced from the Naval Petroleum
                         Reserve, and oil produced from the outer continental shelf.

                         Page 2                    GAO/RCED-91-21 Impacts of Lifting Alaskan North Slope Oil Exporta Ban            :
                                                                                                                                          ‘8 I

                                                    Fig’ure’ 1 illustrates the current pattern of               ANS ’
                                                                                                   crude oil shipments on
                                                     US. tar&e&with the b&i inplacei2 In 1989, these shipments totaled
Distrid&on                     ‘.                   ,about ‘lj700 tnotisand b&+&s ‘per day (MBD). About 1,300 MBD went to,
                        3*.          ‘,)    ,,, ‘1 :th& West Cb;ast;.about ,360 MBD to etitern US. ports via the Trans-
                         ‘7.                     / Panama Pipeline-the U.S. Gulf Cdast,~East Coast, and,Caribbean; and
                                                  : the remainder to ports in Alaska and :Hawaii.
Fi$ure 1: ‘Alaskan Oil’Exp&t~B&            in Place-1996                /                ‘.   .   _,

                                                  -?y                                                                                                       .


                                                                 \y         West Coast


                                    Hawaii I /

                                                                                                       qJ&J&                         U.S. Virgin Islands

         I   Refinery


                                                       20ur analysis focused on ANS crude shipped on U.S.-flag tankers only. SomeANS oil, however, is
                                                       shipped to the U.S. Virgin Islands on foreign-flag tankers and is an exception to the Merchant M&e
                                                       Act of 1920 (the Jones Act). In 1989,122 MBD, or about 6.6 percent of t&al ANS oil shipments, went
                                                       to the US. Virgin Islands on foreign-flag tankers.

                                                       Page 3                  GAO/RCED-91-21 Impacts of Lifting Alaskan North Slope oil Exports Ban.
                                    .         B-241225

                                            Since lQ87, the amount of ANS oil shipped to eastern ports has ;declined
                                     (I     as a result of decreasing ANS production and increasing:West Coast con-                                  )
                                            sumption. Because,transportation costs to eastern ports are consider-
                                            ably higher than those to the West Coast, Alaskan producers sell most of
                                          , their .oil to West Coast refiners.                     . .
                                             This trend is expected to continue, so that in the near future wscrude
                                             shipments, to eastern ports will cease.The exact timing of this develop-
                                             ment will depend,to, a large extent upon the rate~of’decline of Alaskan                                 ; ,
                                             production. Using the Energy Information Administration’s (EIA) base
                                           1 case assumption of Alaskan production, shipments to eastern ports
                                             could cease by.1992, even if West Coast demand for Alaskan production
                                             remains constant. Figure- 2 shows this distribution -pattern. --
sure 2: Alaskan Oil Export Ban in Place-1995

                          Prudhoe   Bay, North Slope

                                                Page     4    GAO/RCJXb-91-21   Impacts   of Lifting   Alaskan   North   Slope   Oil Exports   Ban
                                                                                                                                                                                                                       I   ’

Ban     d&Altikan
j)is~&c&w&n                                                     0
                                                                     Oil ,;_:,,,: ‘,
                                                                                          about 800 MBhWill       be exported to Pacific Rim countries. This will
                                                                                          occur, to ,a large extent,. becausesuch action would ,reduce transporta-
                    .                 ..‘,                                           1 :’ ~tioncosts~by!a considerablean%ount.~,Thisreduction in transportation
                                                                                          costs would increase the .amount MS producers receive for the oil.
                        7.                                            .            . ..’ ~Figure-3illustratesthe resultant .pattern of ..oil distribution, Concep
                                                                                          tually this figure illustrates”the~minimum
                                                                                                                              -      ., impact ,oflifting the;ban; .,                                                                 .
Finurc? !+ Alaskan
.   .=-.   -   -.       m -.-----..      Oil Exdart
                                         -..      ---P-.-----       Ban Lifted-Minimum
                                                                          --.--     -       ____     __.___   -__.   Ex~art
                                                                                                                     ---~--,-   Scenario
                                                                                                                                _--       .._.._



                                                                                  Prudhoe          Bay, ioflh        $lppe



                                               n Refinery

                                                                                                         In addition, some of the oil that is now shipped to the West Coast may
                                                                                                         also be exported, but opinions vary on how much. A possible maximum
                                                                                                         impact of lifting the ban might be one in which the only ANSoil that                                                                  ,

                                                                                                          Page 6                                   GAO/RCED-91.21   Impactsof Lifting   Alaskan North Slope Oil vrts           J3aq


                                         .,’   i  would conti&e..to )e shippe&io the West C&&t .i;r;ould@+qil,@Cd by ’
                                           ‘,.;;. ifitegmted oil compa&s, that js,..tho& that pioduce oil& Alaska and ./
                                                  transport it $0,their own rcef$qies on the Webt Coas!: $,, 198;F,J@esq ..-
                                                  companies usedaroun$.fi7!l qD qf AJ$3 crude. Figure 4 ilh.@rz$@,,&$.:’
                           .’                 , i tern of trade,ba$ed upon this asq+mption of exporting aboixt 1,000 MB$‘
                                 i       .‘: ,/ ,to Pacific Rim.c&ntriq              I)
             .T’                     !             ;   1i     I   ‘/       {,   ,.                 ,-       .i                        :
                                                                                                                           .;         ..,
Figure 4: Alaskan   Oil Export Ban Liftbd-Maxim&                           Export S+yiq                          :              i’.                    c
                                            ,.                                                                       ,.                                                                                      I

                                                                                            h’          Pipeline

                                                                                                                                                                                     U.S. tirgin Island

                                                              Page     6                  GAO/RCED-91-21                  Impacts         of Lifting       Alaskan   North   Slope   Oil Exporta       Ban       1


                                              .-Tobetter understand the’effect of lifting the Alaskan oil export ban, we
Res@@            EI.& z1                       requested that EIA carry out an analysis using ,its Transportation ‘and “’
Computer            MO&tiling          Of8 Refining of International Petroleum (%%Ip).computer.model. Thismodel~
Lift&$&     ‘the &la;&~~               oili   ‘simulates world petroleum activities, including crude oil production and
             6’                                transpoi-tation~ refineryoperations; .a;lidpetroleum producm d&iibu-
Export Ban 8’ : ‘, .:                          tion. .Model resultsindicate that’if the,ban had been lifted in i988, up to
                                          I.‘  1,600 MBD, or three-fourths,:of -Alaska’s crude oil production could have
                                               beenexported that yearand its price (refinery acquisition cost) on the                         *
                 ,.                  ,..       West Coast could haveincreased by over $2.00.per barrel.?.Cne reason
        ‘..                                    why   this-might have occurred is that Pacific Rim refiners may be willing.
            ,, ,, (I          .
                         ’ i.                  to  pay ‘more for’ Alaskan oil than West Coast refiners because it better
                                               suits their product demand~~andrefinery.) configurations;.
                                                      ‘_                   : L
             *1                                If, however, the ban ‘were lifted in l995, the model estimates that only
             ,I                 _ ..           400 MBri, out of an ANSproductionlevel of 1,300 MBD, would likely be
              ..                ‘: ‘_ :‘i      exported inthat (year and the price~would increase over the no-exports
                                               case by only about $0.20 per barrel. According to ELG,ANScrude exports                         ’
                                               would be cbnsiderably~smaller in 1996 than in 1988 for two main rea- ,
                                               sons. First, ANYproduction will have fallen from about 2,000 MBD to
            ,:                                 1,300’~~~: Second; production of other crudesthat are potential substi-                        *
                                               tutesfor ANSon the,West Coast is also expected to decline bysl996. The
                                               declirie~of production in relatively nearby countries and the relatively
             .*                                high transportation costsfromthe Middle East to the West Coast will
                                               make mscrude more attractive’to West Coast refiners in 1996 than in
                                                1988. There will not, however;,besuch a substantial increase in the
                                               price ofsms:crude from lifting the ban as there would have been in the
                                                1988 case.:Lower ANS’ production levels would have already brought the
                                               ANSprice closer to the upper bound of the world price even with the ban
                                               in place.                        _.

                                         While these model results are useful in analyzing the potential impact of
                                         lifting the ban, they are,only a guide for estimating actual changes in
                                         trade patterns that may take place; Appendix II contains a more
                                         detailed discussion of ‘the computer models we used and the limitations
                                         of the TRIPmodel. Appendix III contains tabulated results from the TRIP..                            I
                                         computer runs.

                                          3All dollar f&u-es are in constant 1988 dollars. /

                                          Page7                   GAOjTtCED-91-21     Impacts of Lifting Alaskan North Slope &l Exports Ban

                          Basically, lifting the ban would.have two general kinds of economic ’
Econotic Impacts of       effects. First, there,will be economic efficiency effects. In ‘awell func- j                                   *
Lifting the Alaskan .&l t*ioning oil‘market; ,economic efficiency means producing oil.so long as
ExpdqBan          .‘,  ,, ,incremental   benefits exceed incremental costs of production .andallo,-
                          eating oil to its .highest valued uses, in both national and international
                          ‘.       contexts. Second;,parties that. are involved in Alaskan oil trade’will be :
                   ..,’        ;   affected, creating both “winners’~ and “losers.”
        ‘.   I,.          ,I                     ,.
                                                :.,        . :
                                    .:,             -
Efficiency Increase’s              Lifting the ban may ,lead to potential economic efficiencies. Currently, a
                                   de.clining, but significant, amount of Alaskan oil is making its way to
                                   eastern U.S. ports. Lifting the ban would accelerate the disappearance of
                                   this trade because transportation costs would decrease significantly if
                                   the oil were bexported. Conseque,ntly, Alaskan producers would receive
                                   higher ,wellhead prices by,selling their oil to Pacific Rim countries.
                                   Tanker rates to Japan from Alaska in 1988, for example, were about
                                   $2.60 per barrel cheaper than to Houston. Avoiding the higher tanker
                                   rates would produce an eco,nomicefficiency gain.

                                   Another potential gain in economic efficiency could arise in the refining
                                   sector. Light crudes are more suited for the “light’! petroleum products,
                                   such as gasoline and diesel fuel, which are preferred on the West Coast.
             /                     Evidence suggests that U.S. ,West Coast refiners invested in additional
                                   “downstream”4 refining capacity to process medium-gravity Alaskan oil
                                   than that needed to process lighter crudes. This occurred because West
                                   Coast refiners were able to purchase Alaskan oil at a lower price than
                                   could refiners on the Gulf Coast.      *
                                   If the ban is lifted and ANS crude is sold on the world market, U.S. refin-
                                   eries are likely to pay more for, crude oil and might purchase lighter
                                   crudes instead of ANScrude. EIA% analysis suggests that this might allow
                                   refiners to free up,downstream processing capacity. If this occurs,
                                   refiners may be able to increase the volume of lighter petroleum prod-
                                   ucts they produce. This could produce a gain in economic efficiency by
                                   reducing refining costs.
                                   Lifting the ban could also promote economic efficiency by reducing the
                                   amount of heavy petroleum products, such as residual oil, produced by
                                   West Coast refiners. Residual oil can be used, among other things, to
                                   power ships and generate electricity. Because Alaskan oil is relatively

                                   4Tkwnstream” refining processei are those that occur after initial distillation in order to produce
                                   light products, such as gasoline,from heavy crude oil.

                                    Page 8                GAO/RCED-91-21 Impacts of Lifting Ala&n          North Slope Oil Rxports Ban       :
                                                                                                                                     ‘I I


              -1.                         heavy, refiners currently.~produce more residual oil than the U.S. West
                                          Coastrequires; This-supply of residual oil depresses its price and leads                                                 ”
                                          to more of its consumption than might otherwise occur; ~~~‘sanalysis
               ,_                         says that lifting the ban could reduce production of residual oil,if) *_.: ,. ;.
             ,,:;:, :,.‘,‘. .I
              ,.                          ‘refiners purchase lighter crudes that yield a smaller volume of residual
               ‘,                          fsel;.::,,.         ._,‘.    ’ !.       ,_
                      ‘.       ‘, :. .* ,‘I. $2 ; (. i ’ (’ _ ,, ,                 ) ,I
      \ ,’ .,   i ‘_,            :       -l?inally, a key aspect of economic efficiency deals with ensuring that
         :.                         , 1 products are allocated to their .highest valued uses. In this regard, both
                 ‘.                        the~United States, and its trading partners might be made better off by
                                           lifting the ban. Pacific Rim nations would have accessto Alaskan oil
                               ‘,          that has the potential of better fitting’their industrial needs,,and the
                                        5 West Coast woulclimport more light crudes, which better fit its needs
           :i                              for light end ‘products.
                                      :        .” ,, :        ,, “,              _
            I,  ,:-                                          .) -,
Potent&l Winners and                       Lifting the ‘ban could have important distributional effects-that is, it
                                           would produce winners and losers. Some oil producers in Alaska and
Losers ‘,                                   California would’particularly benefit if the ban is lifted. On the other                                               +
                                            handi lifting. the ban is likely to hurtboth independent oil refiners on the
                                            West Coast ,and the maritime industry. Effects on consumers are
                                            unclear.                                                                                                               d

Impact ori Producers                     A lifting of the ban may benefit some oil producers since the banhas
                      ,I                 ‘also affected wellhead prices for Alaskan and Californian oil and, as a
                                         result; Alaskan and%  Californian crude production. EIA madelhig results
                                         suggest that lifting t,he banmight ,increasewellhead prices for Alaskan
            /,                           oil by a$ much as $2.16 per barrel, depending on the amount exported.6                                                    ‘>-
                                         This may lead to some increases in production of both Alaskan and ‘Cali-
                                         fornian oil,. although the size of any increase is uncertain.

Impact on Refiners                       Independent California refiners are likely to be hurt if the .export ban is
                                         lifted because they will have to pay higher prices for Alaskanand Cali-
                                         fornian crudes. Unlike integrated producer-refiners, against whom they
                                         compete, the independent refiners will not benefit from.increases in                                                      .a
                                         wellhead prices. EU’S model analysissuggests that refineries can be
                                          expected to mitigate this loss’by purchasing lighter crudes; which are
                                          more ideally suited for producing gasoline. Lower costs of processing

                                          6EIA ran the model several times, constraining it to allow only certain levels of exports This gener-
                                          ated different prices for different export levels. The msximum price increase of $2,16 occurred when
                                          exporta were unconstraibed. See apps. II and III.

                                          Page    9              GAO/RCED-91-21     Impacts   of Lifting   Alaskan   North   Slope          Oil Exports   Be
                                     - these-light&nudes may, to some extent, help offset increases in the-’
                           !         : refiners’ crude.oil acquisition costs.
                                          ~                                /I
Impact on U.S. Shipping                 The U.S. maritime industry also stands to lose from,lifting the ban on
             -!     .,              _,,,ANScrude exports. As a result of the,Jones Act (the Merchant Marine
                                        Act ,of 1920), U.S.-flag tankers transp)ort virtually all ANs crude.6 If the
                                        ban ‘is lifted and some of this oil is exported, foreign-flag tankers,
                      ,.                because.their costs are lower,’ are likely to -transport that oil. This will ; ,
                                        accelerate the loss of U.S. ships,,which will be laid up, scrapped, or sold
                                      , anyway. as ANSproduction decreases.
                                                          ,’   :
                                       .Between 1989 and 1996, the&&&me Administration (M~RAD)estimates ’
                 .(                     that 32 ships will be lost because of declining ANS production, even if the
                                        ban stays in place., However, if the ban is lifted and exports begin in
                                         1991, the same losses occur as in the minimum exports case, but earlier.
                                        An additional seven ships are lost if there are maximum exports.
                                      The loss of these ships would also affect the national defense through
                                      reduced availability of U.S.-flag, “militarily useful” tankers; the federal
                                      budget through possible guaranteed loan defaults; and national unem-                                                      .
                                      ployment by threatening seafarers’ jobs. Appendix IV contains more
                                      details of our analysis of possible,.impacts on the US. m,aritime industry.

Impact on Consumers                   It is unclear to what.extent the refiners will be able to pass,,dong , :                                                      ~
                               ,’     increased crude oil costs to their customers in the form’of increased
                                      product prices. While.m’s modeling suggestsa possible substantial
                                      increase ,in the price of Alaskan cyude, it shows little change in con-
                                      sumer prices for gasoline on .the West Coast. We have identified at least
                                      two explanations. First, a switch by U.S. refineries to lighter crudes
                                      could mean more gasoline produced than under the ban. Second, the
                                      availability of imported gasoline may limit price increases for gasoline.

                                      The energy supply disruption resulting from Iraq’s August2, 1990,
Energy Security                       invasion of Kuwait has focused attention on U.S. energy security. The
                                      effects of this disruption show the potential economic implications of                                                    I
                                      relying on crude oil supplies from the Persian Gulf. For example,
                                      between August 1 and August 10,1990, average gasolineprices rose
                                      almost 18 cents per gallon, resulting in consumers’ paying about $63 mil-
                                      lion more per day for gasoline on August 10 than on July 31.

                                      %   1989,   U.S.-flag   tankei-s transpxted 93.4 percent of all ANS crude oil loadedat Valdez, Alaska.

                                      Page   10                   GAO/RcED-91-21       Impacts   of Lifting   Alaskan   North   Slope   Oil Exporta       Ran

                              Lifting the banon Alaskan crude exports would affect U.S. energy
                              security in three ways; First, it would increase total, or gross, U.S.
                              imports. Second, it would possibly lead to a decrease in net imports.
                              Finally; in an integrated world,oil,market, U.S. energy security depends
                            ’ in large p’art on this .market’s smooth and efficient functioning. Lifting
                              the ban could.contribute to this end.

                               Gross U.S. ‘imports. will increase because exports from Alaska will be                                I

                               replaced on the worldrmarket. It is difficult to tell with certainty where
                              theseimports will come from. However, on the basis of analysis pro-
                               vided, b$>ErA’S ,model, if the ban had.been lifted in 1988, most of the
                               increase would likely have originated from Latin America (particularly
                               Mexico and Ecuador), the Middle East (especially the United Arab Emir-
                               ates and Qatar), and Malaysia. If the. ban,were eliminated in 1996,
                               nearly all of the,increasewould likely come from the United Arab Emir-
                       i,      ates andiQatar. The shift ,in sources occurs because Middle East crudes
                               can more readily accommodate the 1995 increase in U.S. import demand                                  ’
                               because of declining production at alternate sources.
        1;I’                                 j, :                :. ‘,, :
                               U.S. net crude imports, that is, total imports less exports, will remain                              .
                               unchanged according ,to EI$ model.results. Net imports may decline,
                               however, if exports are not replaced on a barrel-for-barrel basis.                                        I
                               Imported crude might lead to improved refinery efficiency with the                                        ~
                            : result that refiners, particularlyXm.the West Coast, may be able to meet
                   .           the demand for light products with less crude. Furthermore, increased
                               U.S. crude production, arising from higher prices in Alaska and Cali-
                               fornia, might reduce the need for imports.
                              Worldwide oil market ‘efficiency also could improve to some extent if the
                              ban is lifted. This development,could contribute to U.S: energy security
                              in two ways. First, increased U.S. production would help diminish,, to at
                              least a small extent, world dependence on insecure oil supplies. Second,
                              greater security and diversification of supply would reduce the likeli-
                              hood that U.S. trading partners in the Pacific would bid up world oil
                              prices as sharply as they might otherwise in a disruption.

                              Lifting the ban on ANSoil exports could result in a substantial amount of
Observations                  Alaskan oil being transported to Pacific Rim countries. Such action
                              would probably end Alaskan oil shipments to the eastern United States.
                              However, these shipments will cease anyway over the next few years as
                              Alaskan production declines. Exports could also include oil that now
                              goes to the West Coast.

                              Page11           GAO/RCED-91-21   Impacts   of Lifting   Alaskan   North   Slope   Oil Exporta   Ban


           Lifting the ban would’probably leadto gains in economic efficiency and
           would benefit crude.oil producers in Alaska and California. However; it
,’       + ‘would also probably have:negativeTeffects on independent refiners on
           the, West Coast and the US: maritime industry, although much of the
           effect on the maritime industry will occur even if the ban remains in
            ,     because .of .declining ,Alaskan production.

           ‘morn an energy security perspective, lifting the ban would increase total
           U.S. oil imports but; possibly, decrease net imports as a result of
           increased oil.‘@oduction and improvements in refinery efficiency.
           Finally, lifting the .ban could ‘also contribute to the integrated world
           market’s smooth and efficient functioning.
           As specified in your request, we -focused our analysis on the energy and
           economic impacts of lifting.the ban on Alaskan crude oil exports. We
           restricted our review to’the period.ending in 1996.
                                 9’     :./    I..                I
           We interviewed government officials at the federal and state levels, as
           well as public; private,. andracademic authorities. Both the Department
         5 of Energy (DOE) and,MARADassisted in our data analysis. DOE utilized its
            TRIP computer model to develop possible changes in oil distribution pat-
            terns and their consequences.Model runs were made before the ,current
            Middle.East crisis. MARADassisted in analyzing maritime impacts. See
            appendix I for further details on ,.ourmethodology.
            We conducted our review in accordance with generally accepted govern-
            ment auditing standards. We discussed contents of this report with DOE                                         -
            and MARADstaff, who generally agreed with the facts as presented;,but
            as requested, we did not seek official,agency comments.
             Our review took place.between May 1989 and June 1990. Unless you
             publicly announce its contents.earlier, we plan no further distribution of
             this report until ? days after the date of this letter. At that time we will
             send copies to the Secretaries of Commerce, Energy, Transportation, and

            Page      12        GAO/RCED-91-21   Impacts   of Lifting   Alaskan   North   Slope   Oil Exports       Ban’       :

     \                     1,



               Defense; the Dii&ior, Office of Makgkment k&Budget;‘and ot’her
               interested parties. If you have any questions pr concerns, piease contact
               me at (202) 27&1441. Major contributorg to this l’eport are listed in
               appendix. V..


                                                                                                              (   .

               Victor S. Rezendes                .’
               Director, Energy Issues

               Page 13          GAO/RCED-91-21    Impacta of Lifting Ah&an   North Slope Oil Exporta ian

                                                                          :                                            1   >

&pen&XI               ”                                  ”       ‘,       ‘,.
Scope and             :             ,’                   ./., i
Methodology                    /
Appendix   II                                                                                                        18
Economic   Modeling       of   Overview of the
                               Eti Modeling    Computer Models’~~
                                             Procedure                                                               18
Lifting the ANS Oil            Limitations of TRIP,Modeling Effort ,.                                                20
Export Ban
Appendix III                                                                                                         23
Tabulated TRIP Model
Appendix IV                                                                                                           25
Impacts .of Lifting the        Background
                               Results of GAO’s Tanker Demand Analysis
ANS Export Ban, on             Related Maritime Issues                                                                28
the Maritime Industry.

Major C0ntributors        to
This. Report
Tables                         Table III. 1: West Coast Price-Quantity Relationships for                              23
                                   ANS Crude Oil in 1988 and 1996
                               Table 111.2:Changes in U.S. Crude Oil,Exports and                                      24
                                   Imports in 1988 and 1995
                               Table 111.3:West Coast Refinery Gate Product Prices in                                 24
                                   1988 and 1996
                               Table IV. l:.Export Ban in Place                                                       26
                               Table IV.2: Export Ban in Place, High Production                                       26
                               Table IV.3: Export Ban Lifted, Minimum Exports Scenario                                27

                               Page 14         GAO/RCJD91-21      Impacts of Lifting Alaskan North Slope Oil Exports Ban       1

                                                                                    I.                     ,.   I:,,                     “,,
                                                                                                                                  I_ ,         i

        :.                                Export Scqqrio
                  _.             :. ” Pi&q !I.& ,TUP I$d& Stscture      (.                            -                                                 19
    ,                                      I ,:  ”   .-
                                                     ,:         ,. ,.(’
                                                   _. ,’               ”

                       :.   ,,                                                                                                                                   i

                                                      I ~

                                      j    ,’

                                                            ;:                ‘.



                                    FANS   Al&kan North Slope
                                    BOE    Department of Energy
                                     EIA   Energy Information Administration
                                  : ‘GAO”  General Accounting Office
                                     MARAD Maritime Administration
                                     MBD   thousand barrels per day
                                     OTM   Oil Trade Model
                                     TRiP  Transportation and Refining of International Petroleum
                                               cotiptiter tiodel     I

                                    ‘Page 16                     GAO/RCED-91-21    Impacts   of Lifting   Alaskan      North   Slope      Oil Exports   E+
Appendix I
Scope.md Methodology                                          ,:’

                        We interviewed officials at the federal and state levels; as well as public,
                        private, and academic authorities. These included officials at the
                        Departments. of Commerce, Defense, and.Energy and.the Maritime
                        Administration @@AD);,state officials in Alaska and CaliforniB; repre-
                         sentatives of oil producing, trading, ,&id refining companies; maritime
                         and shipljing interests; economists; ‘and other academic authorities. We
                        ‘also,convened a panel of ‘federal and state officials and academics in
                  <”’   January 1990 to discuss issues.relating      to ANScrude oil exports.
                                                            ,‘I ,I‘,                                                ’ ,’
                        We examined numerousrel%i-ts,. studies, and other documents, including
                        the June’%&%“Report to Congress’on%skan Oil by a Department of
                        Commerce-chaired interagency task force and a previous GAO report,
                        Pros and Cons of Exporting Alaskan North Slope Oil (GAO/NSIAD&69,
                        Sept. 26,1983).

                        We requested that DOE'SEnergy Information Administration (EIA) use its
                        TRIPmodel to analyze how lifting the ban on ANS crude exports would
                        affect oil trade. TRIP simulates world trade in crude oil and petroleum
                        products and the refining of crude. EM analysts used TRIPto produce a
             I’         service report on the potential impacts of lifting the ban on petroleum
                        refining and trade using actual data from 1988 and projections for 1996.                      *
                        We hired a consultant, Dr. Hank Jenkins-Smith, with experience in using
                        computer models to analyze the oil export ban and also use’dthe latest
                        version of DOE'SOil Trade Model @TM) to check EIA’Swork. UTM is a
                        more aggregated and simplified version of TRIP,designed for use on a
                        personal computer. Appendix II contains additional details describing
                        the capabilities and functions of these DOEmodels. Appendix III contains
                        tabulated results from EIA TRIP computer runs.                                                 -

                        We do not view this modeling effort as a means of obtaining exact pre-
                        dictions,of ANS trade flows and prices. Subject to the inherent model lim-
                        itations, this. modeling suggests the direction and, to a lesser extent,
                        gross magnitudes of change.

                        MARADestimated future U.S.-flag tanker requirements, on the basis of
                        scenarios that we provided. With input from oil industry experts, mari-                        .
                        time industry representatives, and government officials, we postulated
                        four scenarios for future ANS oil distribution. All of the scenarios assume
                        that, as future Alaskan production levels decrease to levels forecast by
                        EIA, total shipments of ANYoil decreaseproportionately. We also
                        assumed that ANS oil shipments to the farthest, most expensive destina-
                        tions would be the first to decline. Shipment to. the East Coast, Gulf
                        Coast, and U.S. Caribbean ports, in that order, is more expensive than to

                        Page 16           GAO/RCED-91-21   Impacta of Lifting Alaskan North Slope Oil Exports Ban
 c                          West Coast ports: The least costly destinations would continue to receive
                            Aivs’oil at historical levels as long .asthere is a sufficient supply. Two of
r..:       ‘,               the scenarios thatwe developed ,assumethat the banon ANS crude
                            exports remainsin~place. One uses EIA:basecase estimates for Alaskan
,&’ .                       oil.production, and one uses EFA,high case estimates. The other two see-
 _,                         G&s assume that-the Alaskan oil,export ban is lifted and use EIA base
                            c&Ge~estimatesfor Alaskan oil-production. In one of these, we assumed
                            that theoiHhat would have :gonetoldistant ,U.S.ports is the only, oil that        : .
                            is exported. In the other, we assumed that these exports, plus all the oil
       .                    previously destined for West Coast ports and not controlled. by-mte-          A’J
                            grated ‘oil companies, @sent to foreign ports. On the.basu$of the ;, [
                            experts we consulted,‘we assumed that 670 MEiD would contmue to flow
                            to West Coast ports in this last case;:
                                                     ;    ;:%          ;-
                            We requested that -‘determine            the effects that each of these scena-
                          ’ .rios would. have on.future~US..-flag, tanker employment. MARADprovided
                            estimates-of tiinker requirements’needed to transport oil from Alaska to            ’
                             US. destinations and:of probable vessel,displacement. In developing
                .’           projected vessel displacements, MARADconsidered the age of the vessel; ’
                             carrying capacity relative to the requirements of the scenario, owner-             .
                             ship (either by an oil company or a tanker company), charter status
                             (long- or short&erm), ,andemployment history.
                             MARADalso developed estimates of operating costs of U.S.-flag tankers
                             and foreign2lag tankers on two international routes. We requested this
                     >,      in order to assess the capability of U.S.-flag tankers to operate in the
                             world oil trade incompetition with foreign-flag tankers. Appendix IV
                             provides further analysis of the,possible effects of lifting the ANScrude
                             oil export ban on the US. maritime industry as well as additional details
                             on national security and federal budgetary implications and possible
                             effects on seafarer employment.

                             Page 17          GAO/RCEIb91-21   hpacta   of Lifting Alaskan North Slope Oil Exports Ban   :
‘Appendix ,I1

Economie#MGdelingof Lifting, the.,&& Oil’ ,,, ,
Export Ban

                           EIA utiliied a large-scale linear programming model, the Transportation
                           and Refining of International Petroleum (TRIP) model, to simulate the
                           impacts of liftmg’the ~~$~oil~expoj ban. To help guide ELI’s modeling
                           effort and increase thelevel,,cf confidence in its results, we also per-
                           formed .our own simulations usinganother, smaller trade and refining
                           model, the Oil Trade Model, @TM) ,housed at DOE’S Office of Policy, Plan-
                           ning, ‘and AnalysisWe !took mto~account inherent model limitations
                           when using TRIPresults, in our analysis.
                              ,       ‘.   ,, .-
                ,, .
                           TRIP models world trade in crude oil, crude oil refining, and refined prod-
Overview of ,the :     ’   uctsfor a single period, usually a year. It divides the world into 33 geo-
Computer Models            graphical regions.’ Demand quantities for refined petroleum products
                           are treated as given data inputs, as are upper limits for the production                                        .
                           of all crude streams except Saudi Arabian Light. Saudi Arabia is consid-
                           ered the swing producer that can vary production at will to balance
                           supply and demand. Ten refined petroleum products and 49 crude oil
                           streams are represented; Crude suppliers are linked to refiners by
                           tanker routes or-pip,elines with fixed capacities and transportation
                           rates. Crude-refining ,activities are performed in submodels, repre-
                           senting aggregated refinery configurations and capacities in TRIP’S33
                           geographic regions. Refined petroleum products are then allocated to                                                I
                           meet demands in the regions.                                                                                    -. ,I
                                                                 :                                                                             (
                           The objective of the,model is to minimize the global cost of meeting
                           demands ,for petroleum products worldwide. The cost components                                                        1
                           include the cost of purchasing crudes and other materials2 and transpor-
                           tation and refining costs, which derive from fixed parameters supplied                                          p
                           as data inputs. As shown in figure 11.1,TRIPconsists of three main sec-
                           tors: crude production, or output; refining of crude into petroleumprod-                                              1
                           ucts; and the allocation of petroleum products for consumption. Marine
                           and pipeline transportation links connect these three sectors.

                           ‘Centrally planned economiesare depicted in the model with predetermined production and trade
                           20nly Saudi Arabian Light is explicitly priced in the data input. All other crudes are priced in relation
                           to Saudi Lit Gd reflect quality or location differences.

                           Page   18              GAO/WED-91-21       Impacts   of Lifting   Alaskan   North   Slope   Oil Exports   Ban         1
                                                                                   Economic Modeling    of Lifting the ANS oil                                 ‘8I
                                                                                   Export Ban

- TV---
           --- --
                   .._._  ___-            --_   Structure
                                                ----  ---.-

                        .o,...                        i

                       y.>,                x’             ,‘,I”’             ,‘,    .

    Data     -Cryde Outpq                                          .. Cr~dekhipment             ‘-    Refining 1..    ’                   Pioduct’ Shipment   Petroleum
    Inputs: ,Capacities                                               Capacities and.                 Configurations,                 ,   Capacities and      Product
                 .-*                                                                                                                                          Demands
                                                                      Gods                            Ckpacik, and                        casts
                         -4                                               ‘_.,                       ,,ciisfs   I                       .. ,:
                                                                       i .,       ‘.      r,’              ,‘,                        .: i.

                          0’                                                                                              I,     ‘8


          ,/!, 7       ‘2’.                                                        ‘$RIP ‘assumes that, over a llyear period, consumers, will not change their
                         :_,                                                       “planned” consumption tid crude@oduction capacities are not likely to                        ‘.
                                                                                   drastically change. On the basis of these assumptions, TRIP can deter-
                                                                                   mine the most economical pattern of allocating crude supplies to refin-
                                                                                   eries and allocating refined products to consumers for the:year for
                                                                                   which                                                                                        “’
                                    ‘..                                              ”     :the data have been provided. :.
                                                                                   OI’M and its data base are essentially.highly aggregated versions of their
                                                                                   TRIP counterparts, which makes. ,the’model small enough to run on a per-
                                                                                   sonal computer.3                                                                             *,

 Procedure                                                                         were available. We chose’ 199s because this is the last year withinthe
                                                                                   scope of this study and also be&use decreases in ANScrude oil produc-
                                                                                   tion between 1988 and 1996 are expected to lead to changes in shipping

                                                                                   3CnedifferencebetweenUI’MandTRIP,however,isthat the formerfeaturesnonlineardemand
                                                                                   curvesfor petroleumproducts,whileTRIPdemand
                                                                                                                             quantitiesfor petroleumproductsarepredeter-
                                                                                    4ANScrudeproductionisforecastto beabout1,300MBDin 1996,downfromabout2,000MBDin

                                                                                    Page 19               GAO/RCED-91-21
                                                                                                                       Impacts of Lifting Alaskan North Slope Oil Exports Ban
                             Economic Modeling of Lifting the ANS 011
                             Export Ban’

                              In both the 1988 and the 1995 simulations, EIA first ran the model to
                              establish the “business-as-usual” cases, with the ban in place. The 1988
                              “business-as-usual” ,run utilized actual data,lwhereas the’ 1996 run uti-
                              lized forecasts of crude production and of the consumption of petroleum
                              pi-oducts worldwide. Eq i’an the model againseveral times ‘for each of
                              the 2 years, easing the.export ban gradually in increments of 200 MBD up
                              to a’maximumYexport case. Thus, the model run with no exports was
                              followed by one with exports limited to a maximum of 200 MBD, then                                            ; ,
                              400 MBD, etc., until the maximum allowed export amount specified was
                              no longer binding. This maximum-export case is the same as one in
                             .which exports are totally unconstrained.: FXAadopted this approach
                              becausethe behavior of market participants in the industry is unlikely
                              to resemble the quick and complete adjustment to new conditions that
                              an unconstrained export case would suggest.

                             We usedXJIM for a number of 1988 simulations to study the hypothet-
                             ical effects of lifting the ban that year. Generally speaking, we found                                          .
                             CYlYMresults to be directionally consistent with those obtained from TRIP
                             and both generally consistent with the existing economic studies and the
                             observations of experts we consulted.

                             The TRIP model has various limitations, which stem from its inability to
Limitations of TRIP          model changing, dynamic relationships; its high level of aggregation; and
Modeling Effort              its potential for exaggerating the degree and speed of market

TRIP Is Limited in Its       TRIP is limited in its ability to analyze impacts over a period spanning
Ability to Analyze Impacts   several years because it is a static, single-period model.’ In modeling an
                             exhaustible resource, such as crude oil, changes occurring..between time
Over Several Years’          periods may be very important. Production in any one period depends
                             not only on price but also on the remaining stock of the resource. In our

                             6TRIP and Cl’M results for 1988 indicated that lifting the ban would result in the disappearanceof
                             ANS crude oil shipments to the US. Gulf Coast and other distant U.S. locations. TRIP, however,
                             allocated about 1,600 MBD of ANS crude exports to Japan and ‘other Asia,” while UI’M allocated                       I
                             about 960 MBD to Japan only. In both TRIP and CJTM,unconstrained ANS exports resulted ln a rise of
                             U.S. iVest Coast crude acquisition costs of about $2.00 per barrel. The direction of change
                                                                                                                      --~- in _-the price
                             of key petroleum products was generally the same, but the magnitude was greater in Ul’M. CJI’M
                             results, with respect to crude imports to replace ANS shipments to U.S. markets, did not seem as
                             plausible as TRIP results. In particular, Cl’M seemedto underplay the role of Middle East producers
                             as a source of replacement crudes, contrary to the expe&ations of observers we consulted. The differ-
                             ences between the two sets of results are probably due to the,greater degreeof detail in TRIP.

                             Page 20                 GAO/RCED-91-21 Impacts of Lifting Akwkan North Slope OR Exports Ban
                            Appendix II
                            EconomkModeling     otkifting   the ANS Oil                                     ‘II
                            Export Ban

                            modelmg.the ANScrude trade, however, production in any one year v&as
                            assumed independent of producti0.n in previous years.                                                   _

                            Inlmodeling’the ANScrude trade, EIA assumed that production in future                                       ~
                       :    years will be at the same levels, with or without the export ban,
                            although it was recognized that a likely rise in wellhead prices due to
                            exports might raise the level .of production over the next few yeari. If
                            ANScrude production is indeed sensitive to price changes in the range of
                            $1 to $2,1995 production may be higher than assumed in EIA’Smodeling
                            Similarly, the demands for petroleum products in TRIP are predeter-
                            mined as data inputs for each product by region, and they do not adjust
                            at all to changed market conditions.6 Crude supply quantities are also
                            somewhat rigidly specified in the model.

                            The model assumes that refinery capacities and configurations are static
                            over the period 1?88 to 1996 along with factors affecting transportation
                            rates between 1988 and 1995, except for the cost component attributed
                            to bunker fuel. ‘The cost of bunker fuel in the model is linked to the.price
                            of a standard or “marker” crude, which is a data input.

TRIP Is Limited by Its      Another important limitation of the model lies in its high level of aggre-
High Level of Aggregation   gation. One example is how TRIP models refinery configurations and
                            capacities. In reality, a refining region may have many individual refin-
                            eries with different sizes, degrees of sophistication (diversity of down-
                            stream units), and transportation accessto crude streams. Some.of these                                 ’
                            refineries may be owned by integrated oil companies, which produce
                            their own crude and operate their own tankers, while others are
                            independents. But TRIP aggregates all refineries in one region into one
                            refinery whose capacity in the various activities (e.g., crude and distilla-                        .
                            tion,’ thermal operations, hydrocracking, etc.) is the sum of the indi-
                            vidual refineries’ capacities for the same activities. The model does not,
                            therefore, capture the diversity in size and technical sophistication
                            among refineries, nor is it cognizant of possible ownership relationships                               -
                            that may influence the allocation of crude from producers to refiners.7

                            6Thislimitation,however,shouldnot b&considered a seriousonefor the purposeat hand.As men-
                            tionedabove,changes  in 1988pricesof refinedproductsdueto the lifting of the banareprobablytoo
                            smallto resultin anysignificantresponses onconsumption.
                                                                                  In 1996,pricechanges     areevensmaller.
                            ‘EIA dealtwith this limitationby testingthe sensitivityof modelresultsto progressive
                            the levelof exportspermitted.

                            Pageil                GAO/WED-91-21      Impacts of Lifting Alaskan North Slope Oil Exports Ban

                                   Appendix II
                                   Economic Modeling   of Liftin& the ANS oil
                                   Export Ban

TRIP May Exaggerate the’ .- TRIP ‘does.not explicitly~aticountfor..mstitut&onal factors that may limit
                            the extent and speed !with which petroleum markets adjust to changed
Degree and Speed of         conditions, For example,- . TRIP does not fully capture the influence of’
Adjustment      :           complex ownership relationships within the petroleum industry; nor
                                ‘4 ,does it explicitly model governmentregulations that may influence
                                    petroleum markets. EIA chose to run TRIP. scenarios with progressively
           J                        increasing uprjer limits on exports because the unconstrained export
                                    case might overstate the amount of ANS crude exports.                                                  ’ I


                                    Page 22               GAO/RCED-91-21        Impacts of Lifting   Alaskan North Slope Oil Exports Ban
Appendix III

Tabulated TRIP Model Re&ti ‘-:,.‘.                                                                                                                                      ‘,

                                                                                                                                                                            ’          .

                                                 The follov@ng tables containresul& from the TRIPmodel runs that are
               ‘;       .                        most relevant to’our’analysis. Note that the computer runs for 1988’had
              ?;                                 actual data as inputs and the 1995 runs relied on available estimates.
                            .                  , Also;-‘Riaximuin Allowable, +s Exports” refers to the constraint levels
                                                 EIA imposed during the computer runs. A maximum of 0 refers to the
                                                 export ban remaining in place, and,the “Unconstrained” case indicates
                            :                    that EtA’imposed no,limit on expo@ for that computer run.               8’
                                                                        :,   ..-                                                                                                               ’ 8
                                ‘..),                       1
                            ,                    Complete,resuits from the TRIP-model runs are contained in EIA Service
                                                 keport SR/EMEU/SO-3. Copies are available from the National Energy
                                                 Information Center, (202) 586-8800.
                                                                                                  ,,        I’

                                                                                                          . ..I _;
Table 111.1:West Coast Price-Quantity
Relationships for ANS Crude Oil in 1988         Maximum           ’                           ; , ’ ‘I: West Coast ANS
and 1995                                        allowable ANS                                       refinery acquisition                                RAC in&ease over
                                     ‘,         export levels, in           ANS shipments to             cost (RAC$ye;                                “NO Exports Case,”
                                                MBD                        West Coast, in MBD                                                                   per barrel                      .
                                                0                                               1,381                                      $13.49                                          *
                                                400                                             1,381                                      $13.74                               $0.25
                                                80&L                                            1;19j                                    ; $14.43                               $0.94
                                                Unconstrained                                     466                                      $15.65                               $2.16
                                                                                                                              .I YYO
                                                                                            1,255.                                        83n a~                                           a
                                            --400.                                            -855.                                       j21.17                                $0.19
                                                 Unconstrained              ”     :            849                                        $21.17 I’                             $0.19
                                        /       Nbtes: MBD=thousand barrels/day.          .‘T                       :
                                                All dollar figures are in 1988 dollars.
                                                aNo exports case.
                                                Source: EIA TRIP model solutions.

                                                Page   23                 GAO/WED-91-21         Impacts          of Lifting      Ala&m        North   slope   Oil Exports        Ban
                                                      Appendix III                                                                                                                 .       ,
                                                      TabulatedTRIPl@del&ulta                                .’                                                       ‘,;.

                                                           :                  ,..                                                                                            ’
                                                                                                                          .1           .
Table 111.2:Changes in U.S. ‘Crude Oil                         .’
Exports and Imports in. 1988 and 1995                                         ?.     .-    ‘,:‘, Maximumiallowable ANS crude oil exports                                                        .
                                    .’                              ,(.         ”
                                                                                ,’ i’ ’ , ‘:‘s.i a. il.988                       1995
                                                       ‘j .;                          409 : 800 .*e Unconstrained        400      Unconstrained
                                                       Destination/ source’ ..’       MBD’hlBD            I’.  case     MBD                case
                                                                     *               : ,*,
                                     .      .
                                                      tiepan.:,,.             “,       ;          '220 i '501                                        886      309                  315
                                     '. i                                                          ,'180‘. '.&)
                                                      'Other   Asia                                                                                  639       91                   91
                             ,'I                '_,    Total              ,                .,,”    490 800                                     1,525          400                  408
                                                                                                    ,,      I';      .,                         Imports
                                                       Ecli+or                               ~        :0.148                       "                 162          0                   0
                                                       Mexico                                       337 461                    '                     377 .        0                   0
           .,.                           .A.          .Tri,ni$ad                                     '0      38,.                                   ss            0                    0
                                                       Bolivia/Peru                                  63 111                                    , ,,149            0                   0
                 ,..                                   Kuwait                                           05    0 .t                                    0      (101)                 (101)       : _
       *                                        .
                                                       UAE/Qatar                                       '0    42                     ':,            502        515                   519
                                                       Iraq                                             0     0                                       0      (138)                 (143)
                              :. !                     Malaysia                                         0     0                                    238         99                    '98
                                                       All other .,                               ,o          0                                     29         23                    32         T
                                                       Total                                        400 800                        '.,.         1,525         388                  405
                                                       Notes: MBD=thousand bar&s/day.
                                                       Negative numbers (in parentheses) indioate a.ddclike in imports from these countries.
                                                       Source: EIA TRIP model solutions.

Table, 111.3:
           West Coast Refinery Gate
Product Prices in 1988 and 1995                                                                                                               Product price per barrel %
                                                       Maximum allowable ANS                                              ‘,..         ‘Premium                       87 octane
                                                       crude exports, in MBD ”                                                          gasoline                       unleaded
                                                       0                                                                                   ,$22.26                           $20.80
                                                       400                                                                                  $22.21                           $200.76
                                                       800                                                                                  $22.38                           $20.92
                                                       Unconstrained                                                                        $22.80                           $21.28

                                                       0                                                                                    $28.83                               $27.28
                                                       400                                                                                  $28.47                               $27.13
                                                       Unconstrained                                                                        $28.47           '_                  $27.13         '
                                                        Note: MBD=thousand          barrels/day.
                                                        Source: EIA TRIP model solutions.

                                                        Page24                      GAO/RCRD-91-21                 Impacts of Lifting Alaskan North Slope Oil Exports Ran
Appendix IV

lin~acti of Lifting the ANS Ekp&t Ban on the
Maritime Industry


                               .The,U.S.-flag tanker fleet relies heavily on the ANScrude oil trade.
              $;               M&D estimates show thatthe tanker fleet faces significant losses by
                                1995. These losses till occur with or wthout the ANS oil export ban in
             ,.:,              place because of declining Alaskan oil production. Lifting the ban, how-
             _‘.               ever, will accelerate tanker losses.                                      ‘.
                                  1                        1’     ,_
      /                                                      (.
                               Demand for U.S.-flag tankers rose With the need to transport ANScrude
Background                     to U.S. ports, but-recently, as ANS production has declined so has
                               demand for U.S.-flag tankers. porn, $88 to 1989, U.S.-flag tanker
                          .’   employment on ajl,,+~s routes has declined from the full-time equivalent
                               of approximately255 ships and 5.4 million deadweight tons to 39 tankers
                               totaling 4.6 million deadweight tons. Over these same years, ANScrude
              *,               oil production decreased from 1,974 MBD in 1988 to 1,832 MBD in 1989.
              :-!.                                               ” , _.
               *               NW crude oil production and the employment of U.S.-flag tankers are
               T.              linked because the Export-Administration ‘Act of 1979 effectively pro-
              “(’              hibits the export of ,ANS crude and the-Jones Act (the Merchant Marine
              -3%              Act of 1920) requires that U.S.-built, U.S.-flag vessels transport all car-
                               goes between U.S. ports.

                               Furthermore, the U.S. tanker fleet now relies heavily on the ANScrude
                               trade. U.S. tankers generally cannot compete on international routes
                               with foreign-flag tankers because.U.S.-flag tankers have higher associ-
                               ated costs and therefore higher rates.

                               Tables IV. 1-4 present MARAD’S‘estimates of future U.S.-flag tanker
Rehl its bf GAO’s              requirements based on four scenarios of ANSoil distribution that we sup-
Tank .er Demand                plied. Table IV.1 assumesthat the export ban stays in place and Alaskan
Analysis                       production declines according to EIA base case estimates. This decline
                               results in a loss of 7 tankers in 1990 ,and a loss of 25 more by 1995.

                               Page   26        GAO/RCED-sl%l   Impacts   of Liig   Alaskan   North   Slope   Oil Exports   Ban
                                          .Appendix N                                                                                            ,
                                           Jmpacta of Ufting the AT@ Export Ban onthe
                                           Maritime Industry

Table IV.l: Export Ban in Place              .--    ”
                                           %arly oil averagesin tv$D       :                                                            /
                                          ‘Scenario variables      -,-                                                1999       1990         1995
                                           Alaskan produdtion~e~tiriiatesa                                            2,010      1,960       1,290
                                        ’ Total ANS crude dil b&dings                                                 1,984      1,934       1,284
                                           WestCoast destinations                                                     1,258      1,258       1,005
                                           Total, US. destinatio&                                                     1,851      1,l301      1,131
                                           T&al exaorts ” j                                    :i                         0           0          .o

                                           US.-flaa tankers reatiired                                                    50          43     ‘_ ” 18

                                         ’ Sou&MARAD       estimates.
                                           *Alaskan produdtion estimates do not equal ANS loadings because the production estimates include
                                           notvANS drude produbed in-state and ANS oil that is consumed in-state and is not loaded onto tankers.
                                           bFigures include ANS crude shipped to Alaska and Hawaii.        .
                                             Table ‘IV.2 is based on EIA estimates that assume a higher world oil price
                                             in the future than in the base case’ and has a higher ANScrude oil pro-
                                             duction level and higher loading levels’in 1996 as compared with table
                                        ’ IV. 1. Table IV.2 reflects ‘a possible “best case” scenario for the maritime
                                             industry because the higher associated ,production levels would sustain
                                             higher levels of tanker employmentthan with the export ban in place
                                             and lower production level&This table shows a loss of 7 tankers in 1990
                                             and a net loss of 21 more by 1,996.
                                         ,;.,              ‘.j:,      j.
Table IV.2: Export Ban iv Place, High
Praduction Estimates                       Yearly’oilaveiages in’MBD      :             .
                                           Scenario variables                                                         1989       1990           1995
                                           Alaskan production estimatesa’                                          ._ 2,010      1,9@           1,530
                                           Total ANS crude oil loadinas                                               1,984      1,934          1,504
                                           West Coast destinations                                                    1,258      1,258          1,245
                                           Total, US. destinationsb                                                   1i851      1,801          1,371
                                           Total exoorts   ;                                                              0           0.             0
                                           U.S.-flagtankeis required                                                     50          43              22
                                           Source: MARAD estimates.
                                           aAlaskan production estimates do not equal ANS loadings because the production estimates include
                                           notvANS crude produced in-state and ANS oil that is consumed in-state and is not loaded onto tankers.
                                           bFigures include ANS crude shipped to Alaska and Hawaii.

                                           Table IV.3 assumesthat the export ban is lifted in 1990 and that ANS
                                           crude oil exports. occur in 1991. As in table IV.l, ANSproduction declines

                                           iEJA assumedworldcrudeoil pricesof $16.00in 1990and$20.60in 1996for its base case estimates
                                           and$18.00in 1990and$24.40in 1996for its highpriceestimates.
                                                                                                     Theseassumptions   weremade
                                           beforethe August1990Irwi invasionof Kuwait.

                                            Page 26                GAO/WED-91-21      Impacts of Lifting Alaskan North Slope Oil Exporta Ban
                                       Appendix Iv
                                       Impacts of Uftiug,the   AN@ Export Ban on’the
                                       Maritime Industry                 ‘,          ;,                                             ‘I t

              %                        is lifted, all the oil that would have’.gone‘via Panama to the eastern
              ‘)                       United States withthe banin;ijlace is .exported, instead in foreign-flag
             .,_,,- I
              ;                        tankers. The WestXoastcontinues to receive A.Ns.oilat the same level as
             :.i :                     in table IV.1. These assumptions result in tanker requirements in 1990
               .- ..
                                       and 1996 that are the same as those .in table IV. 1. This occurs becausein
     ..                                 I995 ANSproduction will have declined to the extent that no oil will be
                                       shipped to the e’&tern United:States,,either with or without the ban.
                                       However, tanker requirements during the period 1990-1996 would
                                       decline more siowly ,under’the scenario shown in table IV.1 than under
                   ,.!,                the table IV.3 scenario.         .:.:
               /     ,:                                    / ‘/ :,           ,_’
Table IV3 Export Ban Lifted, Minimum
Exports Scenario                       Yearly oil averages in MBD
                                       Scenario variables                                                           1959             199Q   1995
                                       Alaskan production estimatesa                                               2,010         1,960      1,290'
                                       Total Af$ crude oil lpadings                                                1,984         1,934      1,264
                                       West Coast destinations .,                                                  1,258          1.,258    1,005
                                       Total, U.S. destinatio,nsb,                                                 1,851          1,801     1.131
                                       Total exportsC                                                                    0      -’      0     ‘0
                                       U.S.-flag tankers requjred            ‘,                                        50             43         18

                                       ,Source: MARAD estimates.
                                       aAlaskan production estimates do not equal ANS loadings because the’ production estimates include
                                       non-ANS crude produced in-state and-ANS oil that ‘is consumed in-state and is not loaded,onto tankers.
                                       “Figures include ANS crude shipped to Alaska and Hawaii.
                                       CExljorts begin in 1991 at ,a level of 257 MBD and.de@ine to 0 by &2.
                                        Table IV.4 reflects the assumption that the export ban is lifted in 199d
                                        and that large amounts, 946 MBD, are exported beginning in 1991. We,
                                        assume that certain institutional factors, such as contractual agreements
                                        and producers supplying their own West Coast refineries, result in a
                                        minimum level,of 670 MBD continuing to flow to West Coast refineries. ”
                                        These assumptions result in a loss of 7 tankers in 1990 and 32,more by

                                        Page27                 GAO,‘RCED91-21.     Impacts of Lifting Ala&au    North Slops Oil Exports Bau
                                ..                                                                                                               I

                                               Appendix IV
                                               ImpactsofI.ifHngtheANSExportFhnonthe                     :
                                               Maritime Industry

.‘Table l\i;4: Export Ban Lifted, Maximum             ‘-        ’
 Exports Scenario              ,.                 Ybarly.oilaverages inMBD             2                        ,
                                                 “Scenario variables :.( !                                                 1989      1990 1995
                                                  Alabkan production’estimates?                I ,                         2,010     1,,960 1,290
                                       :    ’ Thtal ANS c!udesoilloadings ’                                                1,984     1,934 -1,264
                                                  West‘Coastdestinations L’                                                1,258     1,258   570
                                                  Total, U.S. destintitionsb                           :                   1,851     1,801   696
                                                  Total exljoit& :_i >            ,’ 4, .I.       :                            0         0   435 s ,’
                                                  US-flag.tankersjrequired’ ‘4:‘...:.,                  :                     50     ,43’      11
                                            .’ ’
                                                  Source:,Mr&$CI’e&&ies       ”         “.   .’
                                               aAlaqkan production estimatis dd not e&al ANS loadings becabse the prodtiction estimates include
                                               non-ANS crude produced in-state a?d AN8 oil that is consumed in-state and is not loaded onto tankers.
                                               bFi&res include ANS crude shiijped to Alaskaand HCaii..      ”       ‘,       :         j
                                               CExpqrts begin in 1991 at a level of 945 MB6
                                               ,Under all of the scenarios, the first tankers to feel the effects of                                   *
                                                decreased demand ‘will be,the tankers used on the most expensive
                                                routes: those transporting ANS crude to the U.S. East Coast, Gulf Coast;
                                                and Carribbean ports (in descending order from highest cost to lowest
                                                cost routes). This will occur because Alaskan producers effectively pay                                *
                                               .the price of transporting crude to their customers and when faced with
                                             ’ a diminishing supply of oil will cut off their most distant customers first.
                                                in order to rjay the smallest possible transportation costs.

                                               The loss of U.S.-flag tankers, whetherthe ban is lifted or not, will affect
Related Maritime                               the national defense, the federal budget, and seafarer employment.
                                                                   :.    .,
Effects on Thinker                             U.S.&flag tankers play an important role in U.S. defense plans, and the
Requirements for the                           loss of these’ships could also reduce the availability of U.S. tankers for
                                               national defense purposes. In.1988, the Commission on Merchant Marine
National Defense                               and Defense, a presidential commission made up of active and former
                                               government and industry officials, most recently defined the character-                                 I
                                               istics of a “militarily useful” tanker and the US. tanker requirements to
                                               support a global war. A “militarily useful” tanker is one of less than
                                                100,000 deadweight tons and is “coated,” i.e., capable of carrying mili-
                                               tary petroleum products. The 1988 requirements for coated tankers to
                                               support the’military and the economy to meet defense global war
                                               requirements was 9.9 million deadweight tons.

                                               Page28                  GAO/RCED-91-21     Impacts of Lifting Alaskan North Slope Oil Exporta Ran
                                  Appendix IV                                                                     .“,,            ..
                                  Impacts~of Lifiingthe   ANSExport   Banonthe            :,
                                  huhie     Industry                                                                                   .,


                .!’                                                                                                           ’

              _                   MARAD’sanalysis of the GAO.scenarios.shows that the demand for
              __                  tankers involved in transporting ANS crude oil from the eastern terminus                                          -
              x.,                 of the PanamaPipeline to U.S. East-Coast, Gulf Coast,and Caqibbean
                            ._.   portb is likely to .disappear; with or without the ban. The Commission
                                  expects :additional militarily useful tankers to be supplied by the Mili-                                     ”
                                  tary Sealift Command, the Ready Reserve Fleet, .theEffectlve -US;~Con-
                                  trolled Fleet, and those ships in the tanker fleet in addition to those !.+,
                                  employed in the transport of ANS crude to US. East Coast, Gulf Coast,
                                   and Carribbean ports.
                                                      .-ll .~     ._ *
                                                                .                                      ,/.~.
     BudgetaryF. Implications     As U.S.-flag tanker demand contmues’to decline, the federal government
                                  is exposed to possible loan defaults under the Title XI lo’an program2’Cf                                 .
                                  the 32 ships designated by MARADas likely to be lost under the minimum                                            :
                                  exports scenario,3 17 had outstanding loan balances totaling $493 mil-
              .                   lion at the end of 1989, but ,none of the additional ‘7that could be lost if
              i                   maximum exports occur had any outstanding balances. The possible
              \;.                 budgetary impacts from defaulted ioans wiil continue to decline over
                                  time as outstanding balances are reduced. For example, between June
                                   1988 and June 1989, the total Title XI outstanding balances declined
                                   from $962.7 million to $860.4 million.

     Maritime Employment           Declining tanker demand on domestic routes will have negative effects
                                   on seafarer employment. On the basis of export scenarios we supplied,
     Effects                       MARADestimates of ship losses would force the loss of from 797 to 961
                                   seafarer billets or employment losses of from 1,881 to 2,268 seafarers4
                                   The 797 billets represent’an estimate of the billets lost as a result of
                                   declining Alaskan production if the export ban remains in place or under
                                   the minimum exports scenario. The 961 billets represent the total loss if
                                   maximum exports occur.


                                    2LJnderTitle XI of the MerchantMarineAct of 1936,asamended in 1970,andits accompanying
                                    regulations,the MaritimeAdministrationisauthorizedto grantmortgageinsurance onshipsbuilt in
                                    3MARADexpectsthesesame32shipsto belostevenif the export banstaysin place.
                                    4MARADmultipliesthe numberof billetson a shipby 2.36in orderto estimatethe numberof sea-
                                    farersrequiredto fill abilleton a yearlybasis.

                                    Page 29                GAO/RCED-91-21   Impacts of: Lifting Alaskan North Slope Oil Exports .?an                    1
   Appendix V

   Major Contributors toTb                                @@it

                                Judy,A. En&and-Joseph, Associate Director
  Resource&                     Richard A. Hale,*Assistant Director.
  coI-qnIunity, and             tidrew J. Vogelsang, Evaluator-in4%arge ’
  EconQmic,.. ‘,:               Charles W. Bausell, Assistant Director for Econtimic
                              ’ Philip .G. Far&h, Staff Economist’

  Developmbnt Division,           ‘.
  Wa,&ingt~~nJ).C.-                                                  ” .:
                               Larry J. Calhoun, Regional Assiment Manager,
 San Francisco                 Frances H. Williams, Site Senior’
 Regional Office               Brad C. Dobbins, Staff Evaluator    .,       ,’                   ‘,

(308621)                      Page 30         GMVRCED-91-21   Impacts of Lifting Alaskan North slope oil Exports m