The Politics of Oil-Producer Cooperation

Citation preview

The Politics of

Oil-Producer Cooperation

The Political Economy of Global Interdependence

Thomas D. Willeft, Series Editor The Politics of Oil-Prodltcer Cooperation, Dag Harald Claes States, Banks, and Mnrkts: Mexico's Path to Fhancl'al Libermlimtion in Coitnyarativc F"erspecZive, Nancy Auehnilh The Political Economy of Eumpealz Monetary Utz$~atia~z, second edition, edited by Jelfrey A. Frieden and Garry Eicfiengreen

Peace Pmsperity, and Politics, edited b y Johrl Mueller Finnmial Mnrkt Reform i~ Ghirzn: Pmgress, Pmbkms, and Pmspetlts, edited b y Baizhu Chen, J. Kimball Dietrich, and Y i Feng l~zslifutions,Trarzsition Eco~zunzil;.s, and Ecorzollzic Ilewlopr~ent, b y "Em Yeager Gapitnl Cttsstrols iu Emsr;_piFzgEcmomitts, edited by Chsistine E? Ries m d Richard J. Sweeney Exlunge-Rn te Policies for Ernergiq Ma&t Econctmks, edited b y fiicharci J. Sweemy. Clas Wihlborg, and mamas D. Willett Political Capacity and Ecor?ombc Rehavhr, edited b y Marina Arbetman and Jacek Kugler Jlddging Econv~~ic Policy: Selecfed Writing of Gottfrid Haberler, edited b y Rcharcl J. Sweeney Edward Tower, m d m a m a s D, Willett Interest Grolrps and Monetary h~tegration:The Political Economy of Exchange Regirne Choice, Carsten Hefeker Crozuth, DCbt, and Puliiics: Economic Adjustment and the PoliticnZ Performance of Developing Collntries, b y Lewis W. Snider Establishing Mntzetar!j Stability in Emergiq Muvkef Ecorrumies, edited by Tholnas D. Willett, et al.

The PO Oil-Producer Cooperation Dag Harald Claes

/

' A Member of the Perseus Rook Croup

All rights reserved, Printed in the United Stater;of America, No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, incltiding photocopy, recording, or any infarmation storage and retrieval system, withnut permission in writkg from the publisher, Ccyyright Q 2001 by Westview P r ~ sA, Member of the Perseus Books Grotlp P~tbIishedin 2002 in tlw United Stater; of America by Wwtview Press, 5500 Central Avenue, Boulder, Colorado 80301-2877, and in tl-re United Kingdom by West-view Press, "L Hid's Copse Rc~ad,Cttrnnor HitX, Oxford OX2 9JJ Find us on the World Wide Web at wm.westviewpress.com Library of Ccmgress Cataloghg-in-Publication Data CIaes, Dag Harald The politics of oil-producer cooperationlkg HaraId CIaes. p. cm,----jn-repolitical ecclnomy of global interdependence) fr-rcludabibliograpl-ricalreferences and index, BBN 0-8133-6M3-X (#k.) I, Petrt~leurnindt~stryand tradelnternatic~nalcooperation-Pc~liticalaspects. I. Title, B, Series

The paper ttsed in this publication meets the requirements of the Amex-icanNational Stanf for Prir-rtedLibrary Materials 239-48-1984. dard far Pemanence c ~ Paper

Figure 2.2 T k Red Line, on page 55, is reprinted with the permission oi S i l n m & Muster from The Prize: The Epic Q~lesf fir Oil, Mo2zc.y 6.Power by D a ~ i eYergin. l Copyright O 1991,1992 by DmieX Yergk. Figure 2.15: E e ~ d sirz taxatio~z of yeholeurn prodzrcfs, on page 75, is of the PR & Infomat.iom Departxnent.of Ihe reprinted with the permissio~~ OIYC Secretariat from Shokri Ghanem: "The role and impact of 012~~'s market share" in OPEC Bulletin vol. XXIX, no G., 'June 1998, Figure 7.10: Cvzide oil prices, ncfzial alzd slkccessive Ivl fcr~zationalEnergy Workshop polls, cm page 27'4, is =printed with permission of iwr Press .from Tfie C e ~ i eout of the Bottle: Wodd Oil Since 1970 by Rlforris A. Addman. Copyright @ 1995 by the Massachusetts Enstitute of Technology

Contents List ofTnbles nad Figrrrres Acronyms AL.kr?uwledgrn~~zts

1.1 1.2 1 1.4 1.5 1

2

Market Context 2.1 22 2.3 2.4 25 2.6 2.7

3

a 3as hternational Politics, 2 Znternational Cooperation, 3 hternational Political Econamy, 6 The Multilevel Approach, 8 Relaticms to Economic n e o ~20 , A Nate on Sources, 27 Notes, 31

Market Strucbre, 36 Market Power, 44 The Era of the International Oil Companies, 51 'f'heOI~ECEra, 60 Structural Effects af the Price hcreases, 70 New Market-Power Relations, 77 Conclusion, 88 Notes, 91

f"olit;icalContext 3.1 3.2 3.3 3.4 3.5 3.6 3.7

Politiciaation, 96 a 3m d Scrcurity, 99 'f'he Irm-kaq War, 101 The Iraq-Kuwait War, 107 hternal nreats, 114 Politicai Cooperaticm Between the Gulf States, 121 Conclusion, 126 Notes, 127

xi XV

xvii

Con telzts

4.0 4.1 4.2 4.3 4.4 5

OPEC as m

Xnternatio~~al Organization fnstitutimal Approaches, 131 'I'he Organfzational Stntcturl, of OPEC, 142 hstitnxtio~~al Fm~ctionsof OPEC,144 Conclusion, 164 Notes, 166

Price and Produc.tion Policy of 5.1

OPEC Countries

'The Redundant Cartel: Policy; 1WI-1981,172 The Effective Cartel: a9ecQuota Policy, 1982-1985,182 The Defeded Cartel: OPEC Market-Share Policy, 198611998, 188 Notes, 198 OPEC Price

5.2

5.3

6

Country Case 1: Saudi Arabia-A 6.1 6.2 6.3 6.4 6.5 6.6 6 .

7

A Dynamic Approach to Hegeznony; 294 The Making of a Hegemon, 207 'The fncapable Hegemcm, 1973-1981,218 The Be~~evolent Hegemon: Swing Producer, 1982-11985,225 'The Coerilive Hegemon, 1986-1996,231 The Mixed Strategy; 1999-2080,233 The Relative Power of the Hegemon, 234 Notes, 237

OPEC:A. Successful Cartel? 7.1 7.2 7.3 7.4 7.5 7.6

8

Hegemcmic Power

Conditions for Cartels, 240 Cartel Behavior, 243 OPEC as a Price-Maker, 247 Quota Compliance, 253 The Profits of Cooperation, 269 A Cartel or a Success?, 273 Notes, 278

Extending the Cooperation: the Non-Op~cProducers

OPEC and

8.11 8.2 8.3

Producers c in General, 282 The N o n - o ~ ~ The OPEC-NOP~-C?I~EC Bargahing, 289 E ~ l a i n i n gNon-OPECCooperative Behavicrr, 292 Notes, 295

131

CLHZ tents

9

Country Case 12: Norway-From 9.1 9.2 9.3 9.4 9.5

10

ix

Free-Riding to Cooperation

Domestic Factors in International Relations, 299 'The Bqaining Process, 301 The Bargaining Game Between OPEC and Norway; 316 Explaining the NOrwegian Policy Chmge, 326 Conclusion, 346 Notes, 348

Oil-Producer Cooperation and the Shtdy of hternational Political Economy 40.1 10.2 18.3 10.4

Empirical Findings, 353 The Explanato~Mo&l Revisited, 361 The Value sl:Combinhg Political Science and Economics, 366 neoretical Implications, 371 Notes, 378

297

This page intentionally left blank

Tables and Figures

1.1 World proven reserves, production, growth, and decline in reserves, m d RIP ratios, 1970-1993 1,2 hvest~aentper additional daily barrt;l of capacity

Proved reserves at end of 1975,1985, and 1995 Refh~erycapacity and throughputs, 1995 The division of the Turkish Petroleum Company in 1928 Ownership shares in Middte East production distributed to compmies Producing countries"ependency on the majors, 1970 Taxonomy of tradhg I'rj,ncipal bmign downstream acquisit.ions ol petroleum-exporting countries, 198S19962 3.1 3.2 3.3

Saudi Arabian and Iraqi share of am@productio~~, July 1990 and 191-1994 Democracy hdexes for the ~ I ? E Ccountries, 197&1996 CLassiliicatinn of thmgimes of OPEC coumtries

4.1 4.2 4.3

Secretaries-genera1 of OPEC Key ministerial,committees of the OPEC conference Organizations and state power

5,1

Regional crude supply gains and lasses,

5.2

:Iraqi and Saudi Arabian production, July 1990, m d average, 1991-1994

6.1

Share oE total exports, selected countries, 1950-1965

7.1

Conditions affecting the incidence of castelization

Tables n ~ Xllusfmtimzs d

xii

Cartelization of intematio~~al oil orEe cartel strategies, 1973-1999 Cordation between price and production by time periods Cor~lationbeween price m d production by price movements OPEC quotas, l"382-1998 C o r ~ l a t i mbetween pmduction and quotas OPEC and Saudi Arabian production, quotas, m d overproduction, April 1982 to December 1995 OPEC members' averproductjon correlated with other mernbers' overproduction, March 1982-JuIy 19911 Relationship between ot;hersfoverproduction m d own overpmduction 8,l 8.2 828 9.1 9.2 9.3 9.4 9.5

Categories of non-OPEC produecrs Characteristics of selected n 0 n - m ~ producers, ~ 1985-1986 Selected non-~PEC producers"roduetion growth, from previous year Pressure exerted on. Norway by OPEC, 1%3-1987 Noway" accommodation of OI~EC, 198S1987 Nowegian conditicms for cooperation, 198S1987 Corretation between positions by major partiesin committee recommendations to pasliaxnentary propositions cm petroleum-~latedissues, 1970-3994 Cmmon positions between I:.,aborand the Cansen;ati:ve Party on petroleum-rcllated issues in parliwntary committees

1.1 Explanatmy model of the book 3.2 Model of the industrial organization of the hternational oil market

2.1 2.2 2.3 2.4

Market concentrations: 5, If), and 15 largest praduckg countries, 1965-1995 Market shares, 1965 Market shares, 1995 Trilateral oligopafy

Tables alzd Ilr'zrstrations

Conflictraal aspects of intergmup relations Cooperative aspects of intergroup re1atims The Red Line Enerm demand, 1950-1999 P r i m q energy consumption, 1950 and.1995, different fuels Selected OPEC cotmtries' production, September 1972 to April 197'4 Monthly oil prices, 1973-2000 Spot and official prices, July 197F1-July 1983 Oil consumption in barrels per thousand GDP, 1960-1993 (19W $) Oil. consumptian, 1965-1999 Trends in taxation of petroleum products Market shares, 19651999 OPEC member countries' estimated c ~ ~ r r e n t accomt balmce, 1W2-1998 Changes in control over the vertical prc,duc"cicmchain International oil companies' and producing countries%sharcl of product chain Categories of politicization kanian and Iraqi production, 1976-1983 Democracy index of ~ I ? E Cmembers by region, 1970-1993 Iranian, :lra@,Kuwaiti, and Saudi Arabian oil production, 1976-1995 Productim, quotas, and mmitoring Irmian, Iraqi, and Saudi Arabim monthly production, 1976-1983 OPEC production and spot prices, 1976-1998 b e r a g e production, Jmuary 1981-March 1982, and production over March 1982 quota 5aud.i kabian and other OPEC production, 1985-1986 Brelzt oil price, January 1999-June 2000 The hegemonic cost-minimization model OPEC and Saudi Arabia GDP, 196&1998 Saudi Arabian government budgets, s e k t e d years, 19"i"-1996 Saudi production and spot oil price, 1982-1985

xiv

6.5

Tables n ~ Xllusfmtimzs d

Share of OPEC production for Saudi Arabia, Iran, m d Iraq, 1905-1998 Kuwaiti, Nigerim, m d Libyan oil production, 1974-19175 IEA estimates of OPEC production, 1977 a d 1982, m d actual production OPEC production and qrrotas, April 1982-April 2000 N-person prisoner 'S dilemma OPEC comtries' overproduction and R/P ratio OPEC countrieskoverproduction m d share of total production N-person chickell OPEC revenue, price, and production, 1972-1995 W u e of OPEC membersf oil exports, 1960-1998 Crude oil prices, actual and successive International Energy Wrkstzop polls

8,1 8.2

8.3

8,4 9.1 9.2 9.3 9.4 9.5 9.6 9.7

Market share of am@,former Sobriet Union, and other non-OPEC,1988 and 1998 (70) Russim oil expwtwutside formr Soviet U~~ion, 1990-1997 us oil production, consumption, and.imparts, 1965-1999 Chinese oil prsductio~~ and c m s ~ ~ m p t i o ~ ~ , 198&1998 'The threat game-tree Actual, planned, and reduced. Norwegian oil production, 1986-1991 Norwegian oil production, 1971-1999 Bmnt Kfendoil price, 1976-1995 ($ m d NOK) The pet-rtlleum sector share of and total exports, 1972-1 999 Paid taxes from the petroleum sector, 1971-1999 Agreement in parliamentary committee remarks between Z,abor m d the Conservative Party regardhg petroleum issues and all issues, 1970-19919

Acronyms API

Aramco BNW BP CENT0 ETA

ENT FTC GATT

CCC 1EA

rcc XPC IPE

TPEX

mbd mrOc NOK NYMEX OECD OPEC X3DVSA

13EMEX

PLO

SoGal Statoil U AE

American Petroleum Institute Arabim h e r i c m Oil Company British National Oil Corporation British Petroleum Central Eeaty Organization Energy Tnformation Administration Ente Nazionale Jidrocarbusi Federal Trade Go General Agreement on Trade m d %riff Gulf Cooperallon Cuuncil International bert;y Agency hternational oil can~pmies Iraq Petroleum Company International Political Economy hternational Petroleum Exchange Million b a s ~ i per s day Nal;ional Irmim Oil Company Norwegian krone New York Mercantile Exchange Qrganization for Econornk Cooperation and Development Organization of the Petroleum Exporting Cvuntries Petritleos de Venezuela SA Petroleos Mexicmos Palestinian Liberation Orgmizatio~~ Standard Oil of California The Norwegim State Oil Company United Arab Emirates

This page intentionally left blank

Acknowledgments This project was initiated and completed while I was at the Fridtjal Nansen hstitute. Most of the work was done in the intermediate period while I was resemh kllow and assistant professor in the Deparlment of Political Science at the Universgy of Oslo. X thank both institutions for providhg necessary Funding For this project, Several people commented on different sectims of the manuscript. I thank you ail, not mentioned nor forgotten. Glenn Martin and Paula Reedy copyedited the manuscript and.improved it considerably In the pmcess from thesis to book I benefited from the stimulathg research environfnemt at Ihe ARENA program at the University of: Oslo, and from trying to tackle the penetrating comments of Professors Rams D. Willett a d Arthur Denzau. At Westview Press, David McBride and Christ h e Marra provided excellent editorial assistance. :None of the above-mentioned persons bear any respmsjbility for any misjudgmentwr errors that may be present in the final text. My wife, I:.,enelhad to put up with an uninsyired, inattentive, and tired huSband for several years, 1 can only hope that I will be able, to some extent, to make up for it. My daughter, Sofie, wap; bcrm in the middle of the work on the thesis. At that moment the importance of my doctoral thesis suf&red a drmatic relative deprivation, AXso at this very moment of writing I bear her calling, "'Pappa, I-"appa,'heminding me of the completeness of life, rather than the problems of conRjchg interests.

This page intentionally left blank

Introduction The Politics of Oil

VVhut debernzitzes the coopemCiue belruvior a m o q f-he oil-prodzrcit~gC I ) Z I ~ ~ Y ~ I ' S ? That is the questim this book. seeks to answer. The cooperation between the oil pmducers h s been the topic of many studies in the field oE energy economics. This study approaches this well-known topic from the theoretical field of pditical science, in particular from the subdiscipl.ine h s w n as htemational Political Economy (IPE). The study is based on the view that d e r s t a n d i n g the cooperation between, oil producers is a yuestjon of understanding interactions of politics m d wm0mics. n r e e analytical beliefs guide the skdy, First, the economic models of the oil market are not sufficie~~t in understanhg the cooperative behavior of the governments of oil-produckg states, ""Oil, unllke other commodities, has a uniwersality of sipificance MIhich in the real world places limits on the applicatian of economic ra.t.ionale to the evolution of thc industry" @&dell 1996:38). The economic models are valuable but need to be supplemented by moetels and approaches that art- capabte of integrating an understandi.ng ol thc pdicies of oil-producing states and the economic factors of the international oil. market. This study pmvides such m approach. Second, ou,r undcrstandjng of the oil producers' behavior, and fnr that matter m s t aspects of international politics, will increase if we are able to understand the dynamic relationship between factors at differmt analytical levels. To grasp how different factors interrelal.e, more complex models are r e q ~ i ~This d . study thus applies a multilevel approach in explaining the cooperative behavior of oil-producing states (Rosenau 1990). mird, &is is a study of' the policy of oil-producing states, not a testkg of their behavior against modeIs of optimal economic behavior, Nor is it a historical study =waling new empiriral facts. It is an exmple of what Bates et al, (1998) calls "analytic-al nanrative." This study is "motivated by a de-

Ivtfroducfion

2

sire to account for particular evcnts and outcomes" ((Bateset al. 1,998:3), while at the sarne time acknwledghg ""the benefits to be gained from the systematic use of theory'" (Bates et al. 1998). The reader wiU &us find that most chapters in this book are a combinat-ion of narrative and analysis.

1.1 Oil as International Politics Susan Strmge found energy to be a topic between the study of economics and the stlldy of political science, thus liemanding a political-economy approach: "what- is neccfed , . . is some analytical hamework for rclating the irrrpact of statesbctions on th markets for various sources of energy, with the impxt of these markets on the policies and actions, and imdeed the econornic developmetnt and nationai security of the states" (Strange P988:191). Economic models of the market behavior of oil-pducing states have been developed on the basis of the assunptions mderlying the behav ior of ecmomic entities (i.e., fims). The fact that the cooperatim between oit producers since 1971 has been a cooperation between states, not firmsl m k e s thcse models incapable ol fully explaining thcir b e h i o r . "Recognizing that the members of orEc (Organization of Petroleum. Exporting Countries) are national governments, not private businesses, conventiond econontic analysis of cartels m s t be Mended with potitical analysis" VilZett 1979a:52), The rationale of states is different from the rationale of firms. States have a more complw set of interests, and tend to place primacy on security interests. Fwthermore, states are more complex organizations that contain mort. potential for internal conflicts over aims and means. EinalTy, states have to pay attention to aspects at the societal level in a way that firms seldom have to do'-if for no other reason t:han that the leadership of the democratic states usually seeks reelection, and leaderships of nondemocratic states seek to avoid revolution. This makes it necessary to supple~xentthe economic understanding of the cooperation between the oil producers with an understanding of the behavior of oil-prtducing s t u t a &ly through a combination of political and economic analysis can one gain a better understmdirrg of the intrinsic world of international oil, as "the economics of oil influence 1986:360). This study seeks the politics of oil, and vice versa" s~tcha cornbination. The hternational oil market is, of course, not isolated horn the rest of the international economy. The changes in tbe oil market can be a result of not d y an individual actor's nnon-oil interests and behavior but also the actor's position in a more general intmational political structure, To capturc. this aspect of the intmationai oil market, Sirnon Bramleyr in his study of us oil policy, develops what he calls a conjunctural approacl-c,

combking realism with orthodox economics: "By . . . placing the developments of us oil policy in the context of its wider management of the world economy and natisn-s tate system, f the approach] demonstrates the c e ~ ~ t rrole a l which oil played in the organizdion of us hegemony" (Bromley 1991:242), His conchsion is that the explanation of oil-market development is connected with the overall development of the imtemationd p s t w a r ccononnic system: The ecmomic crisis can be more readily rmderstoc~das the results of general cmtradictions within the maturation of pc~sh-arcapitalism, evidenced also in an overall shift of US hegemony to an increasingly unilateral and predatory fcxrn; and secmd, the response of the United Stakes to these wider challenges played a significant role in the origins of the 0x1 price increases. QBromlq 1991:242-243)

A shilar propositia~~ is put forward by Karlsson (1093:12, my kmslatio~~): The fundamental . . . hypothesis in the study is that the cmtrcll and power in the internatic~naloil economy has been the key grrindbzalken in the vs-controlled world order, . . . When this key is disturbed so is the position of the hegemortie pclwer (the us), and its ability to cmtrsl the world order. The understanding of the international oil market is substantially limited if the question is not related to the control over the world order and the development and change clf the tzrrlrld economy.

As will become clear in the rest of this book I do not subscribe to this mderstmdinl;. E'irst, the role of the United States is weremphasized. 'The United States has been a policy-taker i,n Che internat-ional oil market since the beginning of the seventies. It has primarily struggled to compensate for its lack of influence in this market, using whatever p w w it found availabk: diplomatic skills, econon^ric=wards, and military force. 'The use of these instruments has been a sign of weaknesst not strength. Second, there are several political aspects of the international oil market worth strxdyjng that are not related to general int.ernat.ianal poljtics or supespawer rela6ons. Third., the interactinn betvveen the oil market and international politics can fruitfully be appma&ed through perspectives that focus on the pctlitical interests of markt actors like the oit-producing states. This is a basic idea in Chapter 3. 1.2 International Cooperation

This study is about the cooperation of states, more specifically about a certain group of states-the oil producers.2 Tnis cafls h some rclnarks

4

Ivtfroducfion

about what is meant by cooperatio~~. Axelrod m d Keohane (1986:226) argue that ""cooperation occurs when actors adjust their behavior to the actual or anticipated p~femncesof others.""Milner (1997:7) finds an agreement on this definition a "notable feature of the recent literature on hternational cooperation." It ffolows from the definition that actors" preferences are essential-both fie preferences of others and fie actorsf preference for adjusting its own policy to the others actors' preferences. This implies that the actors all assume that cooperation will be beneficial: ' / m a t counts as cooperation thus depends on the presmce of two elements: goal-directed behavior that seeks to m a t e mutual gai7.l~thSwgh policy ad~ustments"(MiZner 1.997:8). The definition dso implies that cooperation is impossible between mtors whose preferences are perktiy contradictious. In such cases of pure confiict, there is no room for adjustments, Nor is there m y need lfor adjustment of bebaviftr or cooperation if thr actors3references over outcomes are identicd, una&iguous, and percreived bp a11 actors. Most prominent are the cases where actors have both comrnon m d conRicting interests at the same time. In such cases cooperation mfght ~ a i i z oute comes that represent Pareto-hprovement, as opposed to the outcomes generated, by the actors%individual rationaX behavior. 'To arrive at the Pareto-optimal outcome quires that all actors eschew their dominant strategy. In addition, they must not g ~ e d i l yattempt to obtain their most prefrzrred outcome once they have settled at the unstaMe outcome they prefer to the stable equilibrium" (Stein 3990:32). Arifd tinderdal (1982175)points out that sellers h a market can ""smultmeously compete for market shares, face product prices as collective goods (externality), and stand a chance of increasing cost-efficiency by coordinating R&D or production or d i s t r i b ~ ~ factivities i o ~ ~ (synergy)." m e relationship between oil producers is a prominent exarnple of such a combination of common and contradicting interests. C)ne parti,cula,r issue in the theoretical debat.e in the s k d y of international mlatims is the role of international institutions in the promotion oi cooperation between states. Keohane (1989:3) argues that ""there is . . . ample evidence to conclude both that states have mutual interests and that institutimlalizatiop1is a variable rather than a constant in world poXitics. Given these cmditions, cclcrperaticm is possible but depends in part on iwitutional arrangement-S.A successful, theory of cooperatim must therefore take into account the effects of institutions," Even in cases where actors have identical prefe~nces,the need for some instiktions can emerge if, for instance, there are several eqvilifclrium points with equal payoff to the actors,Vhis creates a need for coordination, and s o w form of instifcutionaiization of the interxtion is likely.

There are several ways h which international hstitutians can increase cooperatim between states, First, instit-utionscm provide the actors with necessary infomtion about the other actcrrs"references. Second, institutio~~s can provide negotiation frameworks that reduce transaction costs and thus increase the effectiveness of inter-state transactions (Keohane 3989:113). Such frwwcrrks cm also heir, coordinate actors%expectations by establishhg conventio~~s* Third, institutio~~s can crihange the cost-benefit calculations of states, or the incompatibilitiesof their positions, by providing an arena for issue-lirrkages, mediating between states, and providintS;nst-rucmts for verifi.cation t-hat actors are abiding with agreements, or sanctions for those who are not. Fourth, institutions can ""frame"how akers view their collective options, as well as contribute to the creation of edlecthe identities. This list does not exhaust the possfile ways that institutions c m influence cooperation between states, but it highlights some of the aspects to be discussed in Chapter 4. Empirical skndies of cooperation might address differcat aspects of cooperation. Tl~roughoutthis study, four aspects will be prominent: First, the substa~tlvescope of the cooperation-that is, the parts of the producersf oif policies that are subject to cdective decisionmaking. In this study the mairr focus will, be the price and production policies of the ojl-producing cauntrics. Sccmd, the depth of cooperation. This pertains to the strength and autonomy of the collective hstikrtions compartld to the sovercigrlly of the individual state inthe fmmatim md,implementation of poliry This aspect can vary from loose targets for the pmducers' price and production policies to having a s u p r a n d i d hstit-ution deciding the price and production levels far a1 memher staks, n i d , the efecfz've~zessof the cooyeration, This aspect focuses cm the outcomc. oE the cooperation, and can be defh~edas the extent to which the oil producers reach their substantive objectives. In this study the effectiveness is m e a s u d against the nonrooperative outcme, that is, from a sitwtion of competitive behavior among the oil producers, If the producers3ncame rises above the assumed hcome in a competitive market situation, it becomes maningful to discuss the extent to which this increased hcome level is a result of cooperative behavior. Without such an hcreased income level, the question of effectjvcness becomes empirically less relevant. Fourth, the msfs connected. with the cooperation. 'This includes two types of costs: first, and mosf important, the ~f~portunity costs, which m the ""sacrifices" by the individual oil producer when bclnaving cooperatively compared with a situatim where it behawes competitivclly; and secmd, the transaction costs comected with conducting the cooperation, such as the costs of running the secretariat, holding mectilrgs, and so on, The beneficial effects will have to be evaluated against the existence of such costs.

Ivtfroducfion

1.3 International Political Economy

This is a study in the traetition called International Political Econc~my (IPE), as it focuses on the political aspect ol the rehtionship between states in an economic issue area, This is in line with the way Robert Gilpin (198;7:9)errtploys the concept of politicd economy simply to indicate a set of questions to be aamined by means of an eclectic mixture af analytic methods and theoretical perspectives. These questians are generated by the interaction of the state and the market as the ernbc>dirnent of politics and economics in the modern world. They ask how the state and its associated political processes affect the production and distribution of wealth and, in particular, hew political decisions and interests influence the location of economic actirvities and the distributian of the costs and benefits af these activities.

Although 1 d.o not subscribe to the notion of the markt-tt as the embodiment of economics but rather seek to identify the market actors in a mow cmcrete way, the key qztestion of -this study is, as for C;i,ipin, "how poltisal decisions and hterests influence economic activities." The developments in the oil market, especially 0r3~Cfs gradual contrd of it, created problms for the economic uadcrstanding of the market mechanisms; "there is currently no economic model that describes the prwess of price determination under OPEC" ('lburk 1977:322). 'With regard to theoretical issues, it remahs m open question how best to design a model of the behaviour of OITEC"(Gately 1984:1113),On the other hand, the theoretical problem Rad long been apparent, and internaily in the United States, o:ligopoly and effective cartel formation had been familiar phenomena: 'Teshaps the most remarkable fdurt. of modern value theory is its friability to explain the pricing, output and other rczlated decisions of the large, not quite monopoEstic f i r m wfiieh account for so large a proportion of our output" (Baumol1959:13)." The development of the oil market in the seventies led to disagreemetnt among economists how to hterpret the development. "Tfiert! are basically two explanations of what happened in 1973-7'4. The one most widely accepted by ecc,nomists is that OPEC effectiwety cartelized the world oil market, exploitjng its power to raise prices above competition levels by restricting producticm. . . . The other explanation argues that OPEC was largely irrelevant as an organjzation and that its melnhers acted competitively" "ately 1984:11Q1).5Even individual ecanamists' comprehension of OPEC as a market actor changed over a few years of market turmoil, Two quotations from works by Rcrbert Pindyck illustrate this point: "OPEC'sbehaviour is surprisingly predictable, since the

cartel is most likely to take only those actions that are in its best economic interest" ((Pil7dyck 1978~37);""economic rationalit y proba:bly applies even less to OPEC than to many other econornic agents" (Pindyck 1982:179). When economic models fail, economists resort to the explanation that pditical factors disrupted the "natural" market cltevelopment. 'fhis kind of disruption has occufred several times during the last thirty years, and made international oil a highfy fluctuating business, The major oil market ""shocks" of this period were (i) the 1973 Arab-Israeli W a ; (ii) the 1979 Iranian wolution; (iii) the 1980 outbreak of the Iran-kaq War; and (iv) the 1990-1993 Iraq-Kuwait War. During, or in the aftermath of, afl these political events, the price of oil changed dramaticalty, and the structure of trhe oil market was affected in a fashion ul-rexplainableby economic models of the oil market, The fact that economic rationality is unable to explain the actms' behaviclr leads to the questicm what kind of logic would actualy explain it. One possibility is that m r k e t behavior, unexplained by economic rationality, is in fact impossible to explain, and subsequently doomed to be left: to chance or idiosyncratic events. The tendency tt? highlight the so-called oil shocks pmmotes a picture of the oil market during the last thirty years as one left to chance. Such a view is just as wrong as the befief that economic models f d y explain the behavior of oil-producing states. What is called for is the development of models able to combine economic and political factors."he above-mentic,ned political events have to bo viewed together with the changes in the oi: market caused by eqraally dfamatic economic evelnts such as ( i )the Tehran-Tripoli agrclement of 1971; (ii)the establishment oi the OPEC quota system h 1982; (iii) OPEC'schanged market strategy of 1985; and (iv) Saudi. Arabia's stratea connected with the IraqKuwait lNar in 1990. The cooperation betwen oil producers is irrfluenced by tbr actors" interests and behavior outside the oif market. States ternd to have many interests outside a particular market; fjrms, few, One of the most promislent scholars of oil-market ecmomics admits that states have a complex set of interests, but disregards the possibiljty that non-oil i n t e ~ s t can s aMcct the state" oil-market behavior: Some models have go>vernments making oil production decisions for noneconomic reasons. This canfuses means with ends, and getting with spending. A state seeks first to survive; then, to cultivate its garden, ar spread the true faith, or bash its neighbors, or anything else. But whatever the r;biectives, the more wealth the better, A state that deliberately avoids wealth-maximizing is a special story, which had better be a gncd ane, (Aclelman 1993b:1;7)

Ivtfroducfion

8

Obviously Adelman's proposition "the more wealth the better" i s isorred, but for s t a t e s h d t h might be a way of achieving security and vice versa. It is an empirical question, when and how (Keohane 198.2:23). The statememt "the rnom wealth Ihc better" s e e m trivialfy true, but if the securi-ty of the state is threatened, no amomt of wealth can compensate for the possiblc cessaticm of its existence. In an earlier study X applied the concept of politicization to the retationshiy between OI~ECand r\iorway in the international oil market (Claes 3990). The actorshotiwes were expanded to encompass mcrtives unrelated to the oil m r k e t (read: pditical) m d likewise the actors' available means. h that study, politicizatim denoted the linkage of econmic matters tcr a political decisionmaking level, in line with the perspective outlined by Mirsch and Doyle (1977:11-13). I'oli.lriciaation was p e ~ e i v e dto occur when an actor h a given market allowed its political aims outside the market concerned to affect its economic objectives and behavior in the market; or whezz an actor in a given market employed political mems for purposes of achieving ecmomic obfectives in the market. This perspective will be fur&er developed in f3hapter 3. This study tries to cmbine ecmomic theories and theories of political sdence. The underlying assumptions regarding actorshatjves and interests in political science are different from those of ecmmics. Hveem f1996:37) has pointed out an ilnportanl aspect of this difference: ""economic theory tends to treat states and international organi.zations . . . exacrly as they treat indlriidds and firms. . . . Neoclassical economics does, in other words, not bother to analyze the actors' motives and interests."' An overriding aim of this book is to grasp the compIcxity of factors influencing states' motives, interests, and behavior. 'This calk for a multilevel approach, wl.rich means that theories at difi;?rentlevels of analysjs are included in order to explain the oil-producer cooperation. As painted out by Milner (1997:6): ""variations across countries, as well as over time and among issue areas within the s m e country suggest that- neither national nor the international level of analysis is suifjcient to explain the patterns of cooperation that we observe in world politics." h argument .for an approach that c d i n e s structurall m d behavinral avects is presented ;in the next section. 1.4 The Multilevel Approach

James Rosenau has argued im favor of more compla models in the study of internationd relathns. "Scientiiic convention stresses that theory should. be parsimonious, that the complexity inherent in the interaction of a large nurnber of variables prt-cludes meaningful theoretical insightsf' (Rosezzau 1990:23). X subscribe to this prhciple. However, parsimonious

models run into some difficulties: "Virtually by definition, parsimonious theories are compelled to i g o r e the mulSiple macro and micro levels at which the s o w e s of turbdence stir and gather momentum'~(Kosmau f 990). This book will fmsake passimony in order to ''achodedge multiple layers of causation" ((Rosenau 1990). This variety of theoretical appma""hewan lead to m incoherent approa,ch, Mokvcver, the aspects of oil-producer cooperation exmined here are not empirically arbitrary. As will become evident during the mpirical analysis in this study, the aspects discussed complement each other m d are bterrelated. mey are also theoretically cmpleme~~tary, as they highlight different aspects of what is called the situational andysis. ""Stuational analysis has as its prirnary objective the explanation of the actions underta,ken by political actors in given situations" (Farr 19KE49). There are two components of the situational analysis: the situation model and the principle of rationality. The principle of rationdity is discussed in section 1.4. The situatio~~al model contains the natural environment, the social environment, and the prdblem-situation inwhich the actor fjnds itseff (Farr 1987:50). "'The physical surroundings c m t a h aspects that are naturally given or given by prior human activities. . . . The social surroundings contain the relevant behavior of other actors, the social rela. . . The problem situation i s the ticms between actors and i~~stitutions. choice the actor faces" (Wovi,and Rasch f 996:74, my translation). A similar model of actors' behavior is given by Elster (1989c:13-14): A simple scheme for explajning an action is to see it as the end result of WC) successive filtering operations. We begin with a large set of all abstractly passible actions that an individual might undedake, The first filter is made up of all the physical, economic, legal and gsl)rcl.lolc>gicatconstraints that the individual faces. The actions consistent with these constraints form his opportunity set, The second filter is a mechanism that determines which action within the oppc>l"cunityset will actual1y be carried out. . . . In this perspective, actions are explained by vgorttxnities and desires-by what people can do and what they want to do.

The model presented by Elster is more general, than the one outlined by Farr and by Hovi and Rasch, as it does not specify the mechanism by which the =tor chooses m o n g diiferent. possible actions. -The situational model assumes that the actor makes a rational choice between possible actions. Some aspwts of the discussim in this book suggest that the oil producers in s m e instances have not adhercd to the principle of ra tiona2ity;T Based cm the sikational approach, the understandhg of structure in this book is not one of structure as a direct causal factor in itself, but as a

Ivt froducfion

20

consfrai~zton the behwim Clf the actors, individxlally and in common, and thus on the outcome of their cooperative efforts.8 These constrainjng aspects wilt be split in two parts: the stmctural tevel anci the institutional level. 'The structure is definitely something the actors rclgard as given. The institutims are more susceptlible to change due to changes in the behaviclr of the actors, However, in this book the mail1 focus will be the constrahhg e h e t s of the institwions, not haw or why the oil producers created tbe cooperative institutions in the first place (see Chapkr 4). The Sf ructzlral Level

EmpiricalZy the stmctura1 features of the international oil market tc:,be discussed in this study are factors such as the degree of horizontal and vertical integration, the demand elasticity, and the barriers to entry (see Chapter 2). Furthermore, as the cooperation among oil producen; is a matter of intergovernmenlal cooperation, the importance of the international political stmcturc, in particular for the security hterests of these states, is discussed in Cl~apter3. These structural aspects infiuence the uti:lity of the policy choices of all market actors, m d subsequently they limit the possibiliw of establishing successful oil-poducer cooperation. At any given point in time, when actors make decisions, they will tend to regard the structural aspects as given. This does not imply that the actors-ehavior does not influence the market structure. Several aspects of the actors" behavior discussed in this book can. be regarded as basic moves (Snyder 1"32:222). Basic moves are to be understood as an actor" bbehavior alterir\g the set of possjble outcomes of a negotiation, or, in this case, of the bargaining rdationship between the oil producert;. This points to the understmding in thc book of a dynamir relationship between structure m d behavior. A promi~~ent advocate-if not the prominent advocateof structural explanations is Ketnneth Waltz. He argues that ""systemic effects camat be reconstructed from the system" interactjng parts since the parts behave differently because they are parts of a system. The constraints and incentives of a system, its dynamics, change if its stru,dure changes or is transformed. To explain outcorns, we have to look at a system" ddyamics a d not just at the characteristics and the strategies of tbe unit$" (Waltz 1986:%2). However, structural explanations have little value UIIless it is shown how these factors have influenced an actor m d its intentions.9 Hoflis and Smith (1991:118) end their discussion of the strucbral approach like this: since the structure shows itself only in the behavic~raf the units, and since functional explanations must involve purposive behavior by the units, there

is no way of inferring that the units are mereXy dependent. . . . changes withh and between the units are the only plausible explanations of change in the system. . . . the case rests with Waltz's remark . . . that "the shaping and shc>vingof structures may be successt-ully resisted ."

As Hollis and Smith indicate, Waltz does not exclude the role of actorsf behavior and interaction in explaining the out.com. h fact, he seems to be in agreement with the mufilevel approach outlined here: ""Structures condition behavicrrs and outcomes, yet explanations of hehaviors and outcomes are indeterminate because both unit-level and structural causes are in play" ((Waltz 1986:343), FNc are probably at a point where d i f f e ~ n scholars t emphasize differmt aspects of a complex reality and thus reach diewent conclusions without- being in disagseement on the fundamental ontology Axelmd and Keohane (1986:253), for instance, regard the r d c of the units as more active, doing more than just resisting the "shapi,ng and shoving" of structustls: "We have also fomd that states are often dissatisfied with the structure of theis own environment. We have seen that gove ents have often tried to transfom the structures within. which they aperate," It makes a difference whether the actors are rnerely resisting the structure or tryh~gto transform it due to dissatisfaction with it.The latter hdicates more autonontous actors Chm the fortnec The statement by Waltz that ""states can more readily change their system than transform it" (198&342) fmplies that he believes that states can trmsform the system. Digeraces in emphasis do, however, have met-hodological irn,plications, as pojnted w t by George, FarIey, and DaIlin (1988:9): Of particufar interest here is the emphasis Axetroyd and Kec>hane[I9861 place on the need far moving beyond the structural form of the hypotheses to empirical analysis of the decisictn-making prcjcesxs of the actors and the strategic interaction beween them [they continue in a foc>tr?ro>te]. That is, not simply to make assumptions about what goes on in the Black Box of decision making and the Black Bclx clf strategic interaction but, insofar as passible, tcr study directly the prcycesses of decision making and interaction.

This is essential to the aim of this sbdy. Far too many studies of the international oil markt make crude t?ssumptions rclgardhg actors' interests, behavior, and interaction. The result is an overemphasis on structural aspects of the market.10 An approach that combines structural and behavioral asyects can yield new and better understanding of the int.ernati-l oil market, and such a thematically focused study can contribute to the understmdh~gof fnternational economirs and polirtics, of which the international oil market constitutes m interesthg and h p o r t m t aspect.

22

Ivt froducfion

Wlter Carls~zzscontributes to the age~zt-structuredebate by introduc-

ing time as a part oi the theoreticd model (Carlsn~s1992). This opens the posibbity to include both the impact of structure on agents and the impact of agents on structure, by modelirrg the relationship as a dynamic process. This pragmatic approach s e a s far better than trying to resolve ultirnatefy the causal relationship between actors and sh-ucture.The position taken in this study is that Che agent-structure debate is and will be unsolvable, It is triviaily true that ""bth agents and sociat stmcturef;interact reciprocally in determining the foreip policy behavior of sovereign states" (Carlszz~s1,992:250). This probatnly goes for all human decisions. We can discuss how and to what extent structures influence a particutar behavior and vice versa, but this is obwiously an empirial question. h Chapter 2 I try to expfain the passibility of oil-produccs cooperation on the basis of changes in the market smcture and the accompanying distribution of pawer among three groups of market actors: the producer countries, the consumes countkes, and the intematioslal oil cormpanies, First, the discussion concerns the extent to which the fundamental aspects of the market structure of the htemational oil market limit the oil producers' freedom of action in such a way that they influence the possible cooperaticm between them. Second, there is the question to what extent: the markt power of other actors, such as the cmsumhg corntries and the international connpanies, li~nitsthe success of cooperation between the oil pmducers. Cbanges in market structure and market-power distri:buti,onarc assumd to constrain the opportu~~ity set for possible successful oil-producer cooperation. Tbe argument is based m theories of industrial organfzation. In Figure 4.2 the model of industrial organization of the internationat oil market is intmdtlced. The figttre explic.ii-ly outlines the causal relationship between market stmcture, mariccrt bef-iavior, market power, and the so-called performance, vvhich in this hook is narmwed down to the possibility of producer cooperation. 'The elnpirjcal contents of these concepts are hrther dweloped in Chapter 2. It should be noted that the struchtral aspects of the oil market discussed in Chapter 2 form a basis for the discussion in all the following chapters, as the arguments in the other chapters are a l dependent on the structural chrmges identified in it. h Chapter 3, on the political. strudure, theoretical problems co~zneeted with the understanding of states as rational actors emerge. fn that chapter the actors are ~ g a r d e das having a complex set oE interests. It follows that the actors might have difficulties identifying thc optimal choice of action, or that their behavior in different situations might be contradictory. This leads us to the study of the behavior of states as compkx actors, which is a substantiai field in the Cheorctical Iiterature on forc.ig~n

policy, in particular following the seminal work of Graharn Allison (1971). In Chapter 3 this problem is confined to the possible contradiction between the welfare interests and the security interests of the oil-produck g states, and the implicatio~~ of this cmtradictio~~ for the coaperatio~~ betwecn such states, The orgmizational univ of the state is subseyumtl_y preservd in Cl~apter3, while the intert-stsare regarded as m m complex than in Chapter 2.

Tn Chapter 4 the discussion fcxruses on m intermediate level between the struckre, which the actors take as given, and thrir incJividua1 behavim or interaction. This level is occupied by institutions. The il-rstitutionsare created by the oil producers but still constrain the sam actors%behavior. "fnstitutions are the d e s of the game in a society or,more formaily, are the humanly devised cmstraints that shape hwman interaction. Xn cons+ yuencc they structure incentives in human exchange, whefher political, social or economic. Znstihationaf change shapes the way societies evolve through time and hence is the key tt? understmding historical &anger' (North 1990:3). hterstate cooperation oE certain degrees of time and space demands some forms of jnstitutimalizatictn,To some extent the institutions play the same role as the structure, k as much as institutions "consh-ain activity, and shape [actors" expectations" ((Kehane 1989:3). I-lowever, instifcutimsd e n are more easily influenced by the actors' behaviar. Institutions are created by the actors, and thus their autonomy has been a muck-debated issue in the study of internationd politics. h the case of the oil-producer coveration, the formal orgmization or9ecis the dorninant manifestation of this institutionalization.This organization has a low score on the scale of autonomy (see Chapter 9). The question is to wbat extent membership in the organization OPEC influences the memklers' behnvior, their objectives, and their collective identity Generally, institutions are regarded as at the same time constrainir-tg and c-mabljl~g actorsfbehavior. In Chapter 4, institu.tims art- prirnarib wen as constraining the actors' behaviar by their role in providing mformation, crcating decisionmaking rules, monitoring agreements, framfng decisions, and formimg co on identiities. This is to some extent contrary to the conve~ntionalview of ortic. The traditional approaeb to the study of OPEC has been th role of the orgmizatim in controlling the oil market and thus enabling the members to reap huge profits on their oil exports. 'These aspects are discussed in Chapter 2 as part of the relationship beheen the structure of the hternational ail market and the behavior of oil producers, oil cmsumers, and the jnternational oil companies. In the rest of the study the bcus is on Chc cooperation behoeen oil

Ivt froducfion

24

producers. Given this aim, the organization is directed ta the aim of achieving a collective good for the oil producers: a hifj;her oil price. Accorciingly, the role of the 0rg""ization is primarily to curb indivictual members' incentive to cheat or behave as free riders. The argument in Chapter 4 assumes that institutions rnight obtain sufficient autonomy to influence the actorsfbehavisr. Once m institution has been created, there with the dissol~~tion of the orgmizaare same additional costs co~~nected tion, regardless of its substantive value or importance. In the case of OPEC, this phenomenon is certainly present. Even in times when the members of the orgmization have disregarded the c o m m o ~policy ~ formulated during its deliberations, a dissolution of OITC would most certainiy have created a more severe and lasting market $isrupticm, a development most of the mentbers would regard as catastrophic. The organization has obtained a certain degme of autonomy and thus it has a certain indepencience from the will of the members. The kvel of State bzfea-ncti~)~~

h e to the above-mentioned hsufficieney of a purdy stmctural explanation of oil-producer cooperation, the attmtion is now turned to the interaction between the actors. ?"he importance of this level is argued by Art-hur A. Stein (1990:53): State behavicjr does not derive exclusively from structural factors like the distribution of power; neither can such behavior be explained solely by reference to domestic sectors and interests, Structure and sectors play a role in determining the constellation of actors' preferences, but structural and sectoral approaches are both incomplete and must be supplemented by an ernphasis on strategic interaction between states,

The meanjng of the word "cooperalion" indudcs an assumption ol such strategic hteraction between two or mre entities, As pojnted out above, fie barpining between the sii producers includes both cooperative and conflictllal ince~~tives.. Comparcd to the basic moves at the stmctural level, this level focuses m the negotiatory moves or communication moves (Snyder 1972:222). These arc. moves, among a set oE actors, within a stmctrarally constrained bargaining relationship. The market behaviolof the actors is influenced by changes at the structural level, but not completely determined by it. h Chpters 5 to K the actors' objectives are regarded as lirnited to their interests as oil pmducers-to maximize revenues from their oil production. :ill this chapter, cartel theory is combined with insight from the theory of collective action. 'The theory of coltective acticm, devdoped by

Mancur Ofson (1965), is used to discuss the issue of quota compliance, which is m e aspect of cartel theory.11 Chapters S to 8 are based on the secmd part of the situatimal analysis, the principle of rationatity, 736s concept needs some clarification. The principle inrplies &at an action is regarded as explahed when it can be shown that the action was adequate given the actor's pertleption of the situation (Hovi and Rasch 1"36:75). In itself this does not include assumptjons regardjng the actor" desires. However, Donald Davidson (19KO:Il) has argued that "at least in a vast number of typicd cases, some pro attitude must be assumed to be present if a statement. of an agent" rrelasons in acting is to be intelligible." This connects the action with the desires. Da\..idsonfsgeneral argument goes as follows: "Wt~enever someone does somethjng for a rcason . . . he can be characterized as (a) having some sort of pro attitude toward actions of a certain kind., and ( b ) be:tieving . . . that his action is of that kind" (Davidson 198O:34). These two aspects are n m e d the primry reason. Davidson clairns that "the primar)i reason for an action is its cause" (Davidson 1980:4).= Elster (1985:2-3) spetls out this point as fo1Iows: "we must =quire, first, that the reasons are reasons for the action; secondly, Chat the wasons do in fact cause t:he action for which they are masons; and thirdly, that the reasons cause the action 'in the right way' Implicit in these rttyuirements is &so a consistency requirement for the desires and beliefs themselves,"" This links the action with the reason, but the theory also tells us haw this reason emerges: ""An action, to be rational, must be the final result of three optimal decisions. First, it must be the best means of realizing a persods desire, given his beliefs. Next, these beliefs must themselva be optimal, given the evidence waitable to him. Finaily, the person must collect an optimal amount of evidencef"(Ester 1"389c:30). This outlines the understanding of the individual actor as a rational actor. The approach to the interactionat level is thus uneierstood as the meeting of two rational actors.13 The next step of this book is to ask whether we discover additional aspects of the d-producer cooperation if we modify- and supplement the principle of ratinnality It might be t-hat t-he principle of rationaljty is insufficient to explaixl the actors-behavior h particular instances, although it holds true that actors act rationatly most of the time. The behavior of an actor in a policy area can change as its interests bec m e relatively less important due to chmges of the opportuniy set in another poticy area. The world is complex, and it is impossible for the declisionmaker to have information about all issues in order to calculate the utility connected with each pdicy option, Rational choice theorists-assumptions regarciing the actor" knodedge seem too strong. Sirnon f1976:H) calls for loosetning the assumptions: "Rationality implies a com-

26

Ivt froducfion

plete, and unat.tahable, knowledge of the exact conscqztelrces of each choice, In actualiq, the human being never has more than a hagmentar). knowledge of the conditions surroundilzg his action.'" ?"his bounded rationality challa~gesthe p~sumptionof consistency: "Beliefs and desires can hardly be reasons for action unless they are consistent" ((Elster 19&5:4).Regarding the actor 'S desires, Elster (1985:C;)states that ""the consistency criteria for prekrences involve, minim&, tmnsifivity: if: X prrzfcr n to b and E) to c, I should prefer n to c. More complex consistency criteria are required when preferences are defined for options with a m m cornplex internal structure." h the study of interaction between actors, the rational choice is a framework for game theory. Afso in game-theoretical shtdies, simplifications are necessary: "a usual pmedure when applyjng gm-theoretical models is to perceive the players p~ferencesas exogenous. . . . This means for practical purposes that preferences arc assumed to remail1 unaffected by which mwes have been cmducted during the g m . For inslance, the prekrtnces are assumed to be unaffected if the actors h the game enter into an agreement" "ovi 2"32:16, my trasrslaticrn). As Hovi points out, the ahove assumption '"undoubtedly represents a ~ b s t a n t i a simpiification l in relalion to the real, world"" (Hovi, 1992:16), This does not imply any rejection of the principle of ratictnaliq. It does, however, call fr,r an inclusim of the pmcess of weighing the different il~terests.It also opens up the possi:bility that desires are more volati%eand. unpredictabl.e. In this study the parsimony of game theor). has been sacrificed in order tcr make a m m he-grained discussion possible and thus increase the validity of the study of adors%teractiolz.

The fourth level is the domestic aspects of the formation of states' policies in the imtemational oil market. 7he need for including even studies of individual actors is proposed by Mollis and Smith (l991:f 43). Hawing concluded that the structural approach is insufficient, they proceed by studying the units and their relations by discussing game theory. This still leaves at least two problems unsolved: One is the psycholagy of the individual human decision-makers and how it functions in srnaXl decision-making grc>ups.The other is the bureaucratic organization of the domestic process of making policy and translating it into decision and implementation. (l-tcrllis and Smith 1994:143)

The cooperation between oil producers is inflraenced by factors within the individual states. Several aspects of the domestic policy-making process of m n y oil,producers codd, be rclevant to this study It has been

necessary to limit this part of the book to two case studies: first, a case study of Saudl Arahja, focusing m the kingdom" r d e as a hegemonic power ammg the oil producers (see Chapter 6); and scrccmd, a case study of Norway' focusing on the consequmces of its increased market share for the country" ppoliey toward oil-producer cooperation (see Chapter 9). 'The assumption of a unitary actor, which has prevailed throughout the previous levels, is now abmdoned. The actors are ass~lrnednot only to have a complex set of interests, as is the case in Chapter 3, but also to be organizatitrlnally heterogeneous, modestly in Chapter 6, and more. fundamental1y in Chapter 9. Keohane and Nye (197E27)argue that "when there arc multiple issues on the agenda, many of which threaten the intert-sts of domestic groups but do not clearly threaten the nation as a whole, the problems of formulating a coherent and consistent foreign policy increase."' h Chapter 6, Gilpin" ((1981)theory of hegemonic stability is modified to explain better the fall of hegemony. In the discussion of Saudi Arabia's role in the oilproducer cooperation, the pdicy dilemma of the House of Saud becomes evident-the myal h m i b wishes to pmwieie external security through a military alliance with the United States, b ~runs ~ t the risk of provoking internal disruption by 'becomhg too Westernized. The ail, policy is balancing between the large exporter domfnating fie market and the large OPEC ally being exploited by the slnaller members. Tfne chapter does not include domestic factors as such, but shows the policy dilcnnma of an individual actor, and thus increases the complexity of the approach compared with that of the p ~ v i n u chapters. s The theory adapted in Chapter 9 is a combination of bargainixrg models of the relationship between ~ I ? E Cand Norwy, and models of the formation of public poficy in Norway. Tfne aim is to understand how the free-rider strateg~pursued by Norway in the seventies, became increasin& difficult to sustain as the Norwegian oil production and the share of the total market increased. Compared with Chapter 6, this chapter increases the complexity of the approach further by arguing that the organkational strucme of the individuai state also is complex, leadirtg not or11y to inconsistent interests but d s o to incor~sistencyinbehavior among different parts of the state. "Domestic politics and international relations are often somehow entanglttd, but our fieories have not yet sorkd out the puzzling tangle. It is fruitless to debate whether domestic plilics really determhe international relations, or the reverse, The answer to that question is clearly "both, sometimes""(t'utnam 1988:427). The model is hrther complicated because Norway was, and is, in a bargaining relationship with the OPEC m m b a s , and the fact that Norwegian domestic actors, like the executive and the legislature themscrlwes, are not unitary actors. 'The f0rmat.izc.d model of the two-levef game provided by Milner

28

Ivt froducfion

(1997) is thus used as a framework for the empirical discussion on the Noswegian policy toward OPEC, although it is not applied explicitly

Propositions and Model A multilevel approach such as tl-te ofie outlirred above unavoidably generates a complex set of explanatory propositions. 'This increases the possibility of incoherence and htemal contradictions in the approach chosen. This sectim will expticate the propositicms of the different chapters and estab)ish the relationship between them. It is argued that the multi-. level approach applied in this study is based on complernmtarity of the dif erent parts. The first propositian is that the sttccess of oil-producer cooperation has been constrained both by fundmental economic factors, such as demand elasticity a d lack of harriers to entry, a d by other actors' behavior in the international oil market, such as that of the oil-consuming countries and the international ail connpanies. The difference between t:he shortterm inelasticity and the long-term elasticity of crude oil demand inflicted the short-term beneficial oil-producer cooperation with cmsiderable costs in the long run. The strategy that was profitilble in the short term became costly in the long m. In ddition, other actors amplified this effect, making the oil producers' attempt to govern the international oil market even harder than the strucharal effects would have implied. These aspects of oil-poducer coopcrlration are discussed in Chapter 2. The second propositian discussed in this book is that the institutio~~alizatim of cooperation between oil producers has made the cooperation more successful @tan would otherwise have been the case. The establishment of OPEC and the subseque~~t worki.ng of this organizatim are assumed to have benefited the oil producers, Chapter 4 discusses how the members, through institutional mechanisms, have tried to overilome the hdividual mernbers%ce~ztivesto act as free riders. The third proposition is that the collective action of the oil producers has created an additional profit to the producers. The empiricai discussion in Chapter 5 first esta:bljshes the extent of collective xtion by the oil producers. Second, it focuses on the bargairzing process leading to such collective action. Zn Cl~apter7 the distribution of the costs and benefits resulting fmrn the cojlective stratee~yis also highlighted. Chapter 6 focwes on the role of Saudi Arabia, on the basis of the hypothesis that l a ~ proe ducers are more. likely to contribute to the provision of a cotlective good than are smaller actors. The fourth pmpositicm of this study is that the oil-related decisions made by oil-producing states artl influer~cedby interests outside the international oil masket. States have more on their minds than profit, n

FIGURE 1.1 Explanatory mc>delof the book

Chapter 3 the consequencedor oil-market behavior of states' security interests are examined, l'he question discussed is to what extent the security interests of the oil-producing states have bad a negative effect on their inccrntives to cooperate in tbe oil market, and their abililty to work jointly to achieve cooperative decisions and thus beneficial outcmes. The fifth propo"tion follows from the third. one, as Saudi Arabia, is identified as the actor havirng in certah periods carried a heavier burden than the rest of the members of OPEC. 'The proposition is Chat the OrEc cooperation has been dependent on Saudi Arabia" pedorming a role as a h e g e m i c power. This prcrposititm is discussed in Chapter Q. The sixth proposition suggests that the Zarger a country's share of the total market becomes, the less likely it is that this country will pursue a free-rider strategy The case of Noway itrustrates tbis poirnt and other bargaining aspects of the oil-producer cooperation. Since N o w a y has gone Irom being an FRsignificmt producer to becmjng the world's sec&-largest oil exporter, a case study of this country prwides a diachronic approach to the question of why an individual oil producer chooses to cooperate. These aspects arc discussed in Chapter 9, which aiso includes a cltiscussion of the proposition that changes in domestic .factors influence the willin,gness of a state's authorities to indulge in produces cooyeration. The relationship between the key variables included in this study can be illustrated in a model (see Figurc3 f .l).'The model suggests how the different aspects of the study are related. It does not imply that the study includes estimations of the direction and strength of causatity. ?"he model, thus has an itlustmtive function. As poinled out in section 1.2,the

Ivt froducfion

20

dependent variable in this book is the outcome of oil-producer coaperation, This outcome is determhed by the intentions and behavior of the oil producers. These, in turn, are determined by three independent variables: the market context; the politicat cant-ext; and the institutional, iunctions of the oil-producer organj.zation, OPEC, Besides hdicating the ~lationshipbetween the explanatory factors in the study; the model in Figure 1,1also indicates the dynamic approach of this book. Structure and behavior are intertwined, and the outcome of the oil-producer cooperation at time t, will influence the struckre at time t,. Et is left tc:,the discussion in the individual Chapters to suhstanliate the contents of the different explanatory factors. The aspects of the overraing research question and the model outlined in tbis section will be dealt with in the conclusion (Chapter 10), whae the fjndislgs of the dilferent Chapters are drawn together. The different chapters in this book alZ deal with the same empirical phezzomenon, the cooperation between oil producers durhg a thirty-year period, from. 1971 to 2UUO. Each chapter approaches different aspects of this phenomenon, from urtderlying structural ctmstraints, through the bargainkg between OPEC members, and to the strateu of an individml actor like Norway. The chapters are thus on different levels of andysis, and &:hey apply different theories. The folle,wing section phces this study in relation to exi,sting econolnic literatwe on oil-producer cooperation specifically and the international ojl market in general. 1.5 Relations to Economic Theories

In the study of the internationat cJil market, several economic modds have been applied. F~~ndamental assumptions, methadology and epistemology vary in the diiferent models. Robert Mabm (1992) has categorized the ecanomics of the oil market into four different lir~esof research: 1, 2. 3. 4.

those based on the understanding of ui.1as an ehaustible resource those focused on aspects related to industrial economics those making game-lheoreticld simulatio~z exercises those conducting econometr-ir testhg of oil-market price behavim

Although none of these lines of research providcs a satisfacbry answer about what determines the price of oil, Mahm obviously recognizes the merits of all of the four research streams. The present study relates to the economic lines clf research as folfokvs: Che fourth Line, dealing wjth econometric~and predictions, is of little relevance, as this study deals with past experiences and is aimed at etiplaining, not predicting. The third line, gante-theoretical aspects, forms the theoretical basis for Ihe diseussiczn i,n

Chapter 7 and parts of Chapter 9. The game theory presented in this study is, however, far from the one presenkd by M h m . The game theory in this study highlights some key aspects of the strategic interaction of specific actors, in particular s:iluations, in the history of oil-producer cooperation. The line of research mentioned by Mabm is a microecnncrmic wariant of game tl-tetrry that builds theoretical comprehensive models of behavior of all masket actors. In Chapter 2, concepts taken from the second line, industrial economics, will be applied. The theoretical foundation sf this line of research is outlhed in the next subsection. First it is necessary to present the arp;txment of those cvho rcfuk the first line of research, the one based on the understanding of oil as an exhaustible resource.

The first line of research mentioned above "has its true source in a seminal article published by Professor Hotelljng in 1931" ((Mabm 1992:2), Tn short, the f-loklling (1931) rule states that an exhaustible resource can be regarded as a fixed asset that the owner can &her extract or Leave in Ihe grolmd., The problem for the owner is to maxirnke the profit w e r the total extraction t h e . If oi is extracted immc.diately, the value oE it equals the interest rate (provided that the social discount rate equals the market interest rate), The only mason to leave the oil in the grwnd is a perception of future price increase. It follows that the price must rise at Lhe interest rate or the discount rate for the profiF of leaving the oil im the grolmd. to be equal to the profit of exh-actim.1" E?crtellWfs work has inspiwd a tremendous number of energy e c m mists. m e off il not the, most prominent energy economists, Morris A. Adelmn, has on several occasions a p e d against the applicability of the I-lotelling rule to the international oil market. In a m m recent cmtribution, he states that ""tie Hotelling Rule m d Hotelling bluation Principle are thoroughly discredited. A vdid theory was joined. to a w r o q premise, the fixed stock. It gave rc-rdts contrary to factf"(hdelman f993b:8). 'The argunncnt is partly based on an empirical observat-ion thatthe physical limit of oil is not a restriction on the consumption of oil. But it is also based on an understanding that the physical or geological amount: of oil is irrelevant to the w o r b g of the oil market cu,rrent:ly and lfor the foreseeable future. '*The total mheral h the earth is an irrelevant non-binding constraint. If expected finding-development costs exceed the expected net revenues, investment dries up, and the industry disappears, Whatever is left in the ground is unknown, probably unknowable, but s m l y unimportant: a geologia"tfact of nr,economfc interest'" (Adelman 1993a:220).

22

Ivt froduction

Growth, and Decline in TABLE 1.4 MrctrZd Proven Resemes, _Prc>ducti.ir)n, Resemes (all in 1Q x 9 barrels), and RIP Ratios (years), 19704993 Pmz?en Net Growth or Brcldz-lctimz Decline in Reserves R P Rla tio 17.4 45 18.3 22 33.2 2 9.3 -23 32.8 21.2 14 29.56 21.2 1I 27.9 22 28.4 20.2 21.9 -1 8 30.3 22.6 -7 27.2 22.9 2.2 26.0 23.7 -2 26.6 22.8 22 25.7 46 27.1 21.3 20.4 10 31.2 1 33.6 20.0 21.l 23 33.8 20.5 9 33.1 21.4 45 34.5 21.9 91 35.2 22.8 76 38.5 23.5 48 40.4 23.8 7 41'2 23.7 22 43.1 10 41.8 23.9 23.7 2 42.5

From a stra.ightfosward empi,rical perspetive, the overall s:iluation regardhg oil =serves has been calculakd by Peter W l l (1991), who argues, in accordance with Table 1.1, that the world is mming into oil, not out of oil. The basic assumption for the Hotetling arguxnent, the fixed stock, bas h fact been a substantidly increasing stock during the last thirty years, measured as proven rr.serves." The red world would be perverse according to thc Hotelling theorem, as it is a correlation betcveen incrttased production and increased reserves. Since 1945 the relationship has been as follows: The more oil that has been found, developed, and produced, the h;gher the estimation of rcjrnainng reserves has become. Geologicd knowledge and extraction technology have something to do with this development. W i t h u t going into detail on this

point, the casrclusio~~ is that the world has so far been moving away from the development prescri:bed by Hotelljng (1931). Geologically, there is a fixed amount of physical oil reserves in the world. 'The argument is that this fact is of no relevmce to the actors in the d market, In particular instances in the history of the oil industry, oil has been perceived as scme. This understandi~~g has not been based on any meas~~rememt of a substantial fal) in remaining oil reserves but rather on the experience of dramatic price increases, This fallacy is all too obvious: 'The price increase is understood as signaling a scartlity. 'The perception of scarcity leads to pmjedions ol continuous prices increases, which again are taken as evidence of scarcity. When the oil price increased ill 1973-3974 and 19763-1980, it was expected to continue to imcrease. As can easily be argued with the benefit of hjndsight, this was totally wrong. (This will be discussed in further detail in Chapter 2.) In this setting t.he HotelZillg theorem mi@t have some value, however, as it can explain the clecisionrnaking problem of an owner of an oil field or oil reserves, Tf the owner believes that prices will increase mre than the return on investments, he or she might tend to postpone extraction, "Nobody would sell his or her shares in Shell at E5 per share today if thcre was a belief that the price is going to rise to E6 in a month or so" "(Nabro 3992:3). This gives the Hoteiling theorem some c~dibility.However, the owners of the resources are often states, which tend to spend the money they earn. To forgo incom today often means hard political choices and bargaining with either other parts of a ruling elite or tho population in its capacity as voters*Thus the djfference betwee11 the logic of companies and the logic of states have implications for the relevance of the Hotef ling theorem. Adchan argues that the cost Of replacing the prodUced reser*s is a better measure of scarcity: ' m e early-warnhg signal of scarcity is a persistttnt rise in development costs and in the in-ground value of oil Rserves" "(All7nan 1993a:276). The trend shce 1945 is not clear-cut. The overall costs of developing oil have varid, m d produciRg countljm differ regarding the level of replacement costs and to what extent the Rplacemnt costs are rjsing or falling. A d e h a d s sample. of key omc countsies indicates that a large portion of the wltld1s oil reserves is bcated in countries with falling placement costs (see Tabk 3.2). It is also important to note that for a large part of the reserves being extracted today-as in most parts of the Mlddle East-the selling price is well above what is needed to cover costs and earn a fair profit. Thus, the incentives for cost-cut.ting in these areas are kw. tn other are=, such as the North Sea, where profit m a ~ i n have s come under pressure, substantial cast-cutting efforts have been successfully implemented. The conclu-

Ivt froducfion

24

Abu Dhabi 2,912 Algeria 4,570 Indonesia 5,908 Iran NA Iraq NA Kuwait 1,455 Libya 24,199 Nigeria 3,179 Qatar 1,331 Saudi 692 *All,dollar amounts in this bcrnk are in U.S. dollars. SOURCE: Adelman 1993b:6,

2,033 3,533 5,637 4,433 145 895

NB 2,6511 1,408 373

sion is thus that there are no signs today that overall replacerrrent costs are increasing. The other important questionable assumption in the Hotelling paradigm is the hear-cost assumption. Xt is understandable that F-XRtelling understood the wodd fn this way as he wrote his paper in the *irties, when most oil was extracted in low- and stable-cast areas with an abmdance of oil reserves. Toda~.,bowever, even in low-cost areas, fast increase in extraction is impossible or at least very costXy. Accordingly, no matter what the incentives for increased extraction m, Che physical cmditims and economics of increased extractjan prohibit such increases in depletion rates, at least fn the short run, Flarthermore, increasing prtlduction to a maximum dcpietjon rate takes time. Therefore tzre should exped to see production in the lowe~cl-rstareas grow faster than the high-cost. This is precisely what we saw before 1973. Then there was an abrupt turnaround. High-cost areas expanded drilling mightily while low-cost cut back, It was water Bowing uphill. The only theory tzrhich explains it is monopoly whether of one or a small grc~uptrying to act as one, to restrain output to maintain prices. (Adetman 1%E471

This powerfui. argument is important for thc focus of this study. In these terms it is tke degree of monopoly behavior by t h oil-producing corntries that is the focus of this sbdylh The conclusion that oil prices would continue to increase was also reached through anlalysis of the oil-producer cooperation, Dermot Gately (1995), with the benefit of the hindsight, cites Fesharaki (1981:3M): "'as to the extent of thc prjce increases [in the 198Cfsl,m e can only say with cer-

tahty that real prices will not be allowed to declke again." The following conclusion was reached by Oysteh Noreng (1978:301): A I c w level of oil imports into the OFCD (Organization for Economic Cooperation and Development) area will not silence the demands within ot3sc for price increases?as long as the oil price remains below the cost of alternative energy sources. . . . many c3rrFc countries with large populations and small oil reserves are likely to have trade balance deficits by the early 1980s. They will then have the choice of reducing their domestic programs of social and economic development, increase their oiX exports, or pressing for a hike in the oil price. The last option is the most acceptable to them because their ability to increase exports is limited and a loss of income is obviously undesirable.

Here it is the likely poticy of the oil producers that leads to the conclusion that prices will hcrease, This is in accordance with the mechmisms that, later in the [email protected], caused Adelman to reach the above-mentioned conclusit,n of '*waterflowing uphill.ffThese aspects of the oil-producer cooperation are luily develop4 in Chapter 5. T!ze Indzrsfrial Organizufion of the I~ztertzu tionul Oil Market

Having argued that the Hotelling tradition has limited relevance given the present state of the oil market, :l now turn to the second of Mabro's streams of research: industrial eco~~omics: The second stream cmsiuts of work in the mow tradikic~nalfield of industrial ecmomics. Its origins are much older [than those of the first stream]; and the most notable contributions preceded the oil shock of 1973. Recalf Paul Frankel (1946), Edith Penrose (1968), P-ielmut Frank (1966), Jack Hartshorn (1962) and Morris Adelman (1972), among others, who despite their fundamental differences in approaches and fundamental differences in views were all cmcerned with the ecmomic variables that matter (cr~sts;prices?behavior of firms; scale, vertical integration and industrial structure; transfer prices and tax optimization), and with the central issue of political economy, the re1a"tiomhipbetrvlreen government and industry. (Mabrt3 11i392:4)

Mabro finds that this tradition ""dindled to a trickle" "(1992:4) after 3973." However, Chapter 2 will discuss the stmch;lml conditions for oilproducer cooperatim front this perspective. But first s m e basic co~~ccpts in the Weratare of industrial organizations will have to be presented. Most textbooks on industrial organization start with a variant of a model describifig Che relationship betwee11 the basic charackristics of the

Ivt froducfion

26

hdustry, the market structure, the actors' behavior, md the ultimate outcome FR the forms of prices, patterns of profit, technological progress, and distribu.ticmai effects (Schert-r and Ross 1990:5; Gxer 1984:10; Shepherd 1,990). Except for Chapter 2, this study is about the behavior of a specific g r w p of market actors-the oil producers. En Chapter 2, the focus is on the context of the oif-pmducersf cooperation, 7'he context cmsists of two elelxents, which I have called market structure and market power. The market structurcl consists of a vertical and a horizontal structure. The market power is related to tbr distribution of ability to influence market development-s among three groups: the consuming countries, the international oil coqanitzs, and the producing countries. The p r o b l a then arises as to how basic characteristics, market stmcture, actors' behavior, and, for hstmce, price developments are related to each other. F i p r e 1.2 presents a standard tcrxtbook version of the industrial organization, including some key cmcepts used in this study* The factors included in Ihe nnarket structure box were tramionally considered exogenous given and not subject to possible influence by the market actors. Andysts soon discovered that this was not so. The "old" industrial econasnists hold on to the notion that the main flw in Figure 4.2 was going from structure through behavior to performance: The causation flows mainly downward,. . . At each point in time, the market" structure usually influences the behavior of the firms as they decide how strongly to compete or collude with each other. . . . Careful mainstream researchers have always recognized that cause and effect are mixed to some degree, but logic and business experience have strongly suggested that the causation usually flows mostly downward, (Shepherd 1930:7)

The ""new" hindustrial economists thus emphasize a number of serious deficiencies. In particular, while it may be possible to idcsntify a retatiomhip beween, for example, industry profitability and market concmtratim, there is no mason to believe that such a pwitive correlation tells us anything about causation. . . . M a t this points to is a diffieulv in the traditional approach in iidenti@ing which of the relwant economic phenomena are exctgenous and tzrtzich are endogenous, Developments in the new industrial economics suggest that most cof the factors that enter into market stmdure, conduct, and performance are endogenous, They are derived from the basic economic conditions that characterize the market under investigation and the strategic interactions of the players in those markets. . . . Firms do not merely react to given external conditiam, but try to make their economic environment as beneficial to t h e m e l v s as possible, taking into accc~untthat their competitcors will do likwiw. (Nc~rmanand La Manna 19921-2)

Demand condition: Supply conditions:

STRUCTURE Xne'iasticity CeagraphieaX concentration Market shares Entry barriers

BEHISVXQR

* Collusion with rivals * Strategies against rivals f i.e. the trilateraX ofigopoly)

PERFORMANCE

* Price patterns * Cost and profjt patterns FIGUM 4.2 Mc~delof the industrial arganiz;ation af the international oil market

This leads to the fundamentaI proposition of Chapter 2, namely that the aspects of the market stmcture cannot be separated from the aspects of the actors' behavior, The dynmic relationship created between structural change and the actors competing for market power is an important foundation for the rest of the book. A study of the success of the oil-pducer cooperation has to djscuss how this cooperation is affccted by and affiects the behavior of other actors and the market structure, Chapter 2 outtines the market structure, but the arguments in all the folloMiing chapters are based on the understanding of (i) a dynarnic relationship beheen the oil-pducer cooperation m d changes in the market stmcture, and (ii) the producers as a group in a strategic interaction with other market actors-the internationat oil companies and the oil-consuming countries, 1.6 A Note an Sources

As a comequence of the multilevel agprclach advocated in section 3.4, the empirical analyses hthe different parts of this study are partly based on secondary sources, ""Secondary research differs from p r i ~ u qveseatceh in that tbr colkction of the irrformation is not the respclnsiit7ility of the analyst. In secondary research, the analyst enters t-he picture after the data,

28

Ivt froducfion

colledion effort is over" "tewart and Kamins 1,993:3).There are obvious disadvantages comected with the use of secondar). sourccs: Data often are collected with a specific purpose in mind. . . . the data collected might be so extensive that the individual whose job it is to interpret the findings can pc~tentiallyarrive at many different, even cmflicting conclusions, all of which might be supported by sc:,me subset of the data. . . . the secondary data are aggregated in some form, and the unit of aggregation may be inappropriate for a particular purpose. . . . finally, secondary data are, by definition old data. (Stewart and Karnins 1993:6)

These aspects have to be given particular consideration when using secondary sources. S m e advantages of using secondary sources can also be noted: "it is less expensi:ve to use scleondary data than it is to conduct a p r i m r y research investigation" (Stewart and Kamins 1993:5). In this study the use of secondary sources has facilitated a more complex analysis than would have been possible if all aspects discussed had been based on primary research. This weighi~~g toward more variables and fewer prirnary sources has been a conscious choice. However, there ,?re several topics in this book that are based on primary research, meanhg that the data have been collected by me. In the following paragraphs the data used in the different chapters are outlined, and some methodological aspects discussed. First, however, some comments are necessary about the use of sourcebooks and cmpiled statistical material. This study covcrs a thirty-year time span,from 11972 to 2000. Several ecmonic factors are presented incharts. The data on prices and production are taken from reliable sources, widely used as references h the oil industry. Most price data in the hook arc based on the prjces reported in OPEC Nzllkiin. These monthly average price data are based m PlafffsOilgmm Price Report, wbich is regarded as reliable in reporting the spot crude oil prices. The monthly production data are taken from the publication Pefroleltim Economist, a well-bown source of information on the internationd oil market. Annuai production data have been taken from BP Silrtislical Review of Wodd Elzera* This publication is perhaps the mostcited reference for oil-market data, and is regaded as highly =liable. Production figures vary somewhat between different sources, but not in a way that shodd malte the di,scussion i,n this book uncertajn. In &&ion to these statistical reports, three sourcebooks have been used. E~zergyin fhe W r l d Ecrt~zolrzy* stufisfical rc.vit.ru qf trclzds in output, trade, and corzsu~rrplirltssince 1925 (Darmstadter et al..1,977.)is a compitation of stati.t;tical information on world energy until 1.970. mere is no analysis conducted in this s o m e . The same goes for Oil Ecolzomists' Naudbclok jJenkins 1986), which is a compilation of statistical informdion anci a dictionary of oil-

related terms. OPEC ancl the World E n r r r ~Nlurket-a c refer~~zce gzlide (Evans 1990) describes the events in the oil market and the decisions of ~ 7 1 and ~ ~ the ~ 3 members%ehavior in the market on a year-by-year basis fmm 1,966to 1990. The text prese~~ted in this book is mostly descriptive, providing "wide-ranging factual coverage . . . designed to pmvicrte a rounded historical perspectfvt.""(E\ians 1990:xxi). This publication is largely a textual prese~~tation, and the passibility of interpretations and errors is greater than in the above-mentioned statistical compilations. This has partly been compensated for by a substantial relimce on Oficial Rescllr.tf.ictt~s and Press Refmses-2 960-2 990 (OPEC 1,990). rhis compilation, published by The Secretariat of the Organization of the Petroleum Exporting Countries in Vienna, pr""ides the full texts of the decisions of OPEC, and c m in my opinion be regarded as just as goad as havhg the orighal texts. The data in Chapter 2, which form the basis for thc. inEerence of a causal relationship betkveen market structure, market power, and the possjbiliq of producer cooperation, are of two different kinds: statistlical data on the histrrrical developments in important economic variables cmstituting Che market structure, and a historical description of the relationship between the three groupwonstituting the trilateral oligopoly. The first kind of data is utilized though a descriptive economic met-had presenting t-he changes in the econmic variabks. C)n this basis, it is argued that the conditions fnr producer cooperation are better given a favorable development in such factors, The historicai narrative regarding the changes in market power is utilized in the same mamer, describing the changes in c m t d among the international companies, the consuming countries, and the producing countries, and suggesting that certaill cmste1Iations of pokver between thcse groups favor the establishment of cooperation beheen the oil producers. In Chapter 2, these two kinds of data and the connected reasoning are linked as the presentation is struct u ~ according d to the historical changes in these variables and the dependent o n e t h e oil-produccs cooperation, None of the data p ~ s e n t e d in this chapter are especially uncertain. 'I'hey are all very well estabtished m d largely noncontroversial in the literature on the oil market. Chapter 3 includes, in section 3.5, data from the Polity 111 project, which are used to describe the level of democracy among the OPEC countries. 'The validity and rcliabitity of these data arc discussed in Jaggers and Gurr (1995). The coding of the Democracy Indexes referred to in Table 3.2 obviously involties an element of subjectivity: "Giwen the paucity of dersnocracy as an analytical concept, the empirical varia,bility of current measures of regime type c m be attributed, at least in part, to dif-Eerencesamong rt?seartlhersUCfaggers anci Gurr 19%:469). Intcrsutnjective agreement seems a reasonable criterion for relying on the data, at

30

Ivt froducfion

least for the use made of them in this study: "The stronger the correiations between our measures and others3indicators of regime traits . . . the higher our confidence in the empirical valiclity of our i~~dicators and data"" gaggem m a n d Gwr 1995:473). The measttres applied are highly cop related with measures used by other scholars of democracy (faggers and Gurr 1995:475), The data in Chapter 4 are mostly p r h a r y data on the decisions made at the OPEC: conferences or by different ministerial, committees. f i s t of these arc. compied and published by the organizaticm itself (OPEC 1990). There is no reason to doubt the reliability of these data. However, they should. not be interpreted as telling the whole truth about the decisions. In every oqanization that i~~volves bargaining between states or other actors, there nnjght be hidden agt3nd.a~and dottble meanings wlated to both the bargaining pmess and the subseqrrent outcome. To some extent this has been taken illto account, as I have Fncluded both direct statements horn Che actors and the evaluat-ionof different events bp cometntators or h secondary literature. As in Chapter 2, the data used in the empirical discussion in Chapters 5 and 7 are a combhation of statistical data on ecmomic factors and historical narratives. Here the theme is the internal bargaining between oil producers, and not, as in Chapter 2, the market stmcturl, and the relationship between oil producers and olfier groups in the intemtimal oil market. The data on quotas are non-disputable, as they arc taken directly from OPEC resolutions and are reported verbatim in Table 7.6. The sipificance of the calculated OPEC revenue in Figure 7.8 is, however, more questionable, as no cost estinnates are included.. It is simply a multiplication of price and production cm a mclntJRiy basis, The =ason for this is that the intention regardbg this figure is only to give a getneral picture of the development, and that estimates of production costs are one of the most uncertain issues in the statistics of the international oil industry. For example, Norris A. Adel~nan,a prominent scholar, made estjmates on the basis of the follawing assumption: "It is assumed that drillhg costs are the same across all natims and are equivalent to the U.S. dritling costs since no country-specific drilling costs are available" "(Adelman and Shiahi 1989:3). Tl~eaccount of the development of the Saudi Arabian hegemony in Chapter 6 is based on secondaq literature, both on Saudi Arabin's role in the oil market and on the more general development of the state and the policy of the Saudi Arabian regime. Chapter 9 includes an account of the bargaining ~1at.ionship between orEc and. Norway in section 9.3, Here, newspapers are a prominent s o w e of information. In many cases these news reports refer only to statements made by the actors in the bargaining process, In such cases

the methodological uncertainty is related to the accllracy of these references. En cases where the journalists have irtterpreted aspects of the bargain% sihtaticm on the basis of interviews with actors inVoiVed in the process or their own judgmnts, the reliat?iliity is further kveake~zedby the journalists"possibk inaccuracy and misunderstandhg of avects of the situation, fn my use of such secondary sources, the primary aim has been to cite the facts drawn from primary sourcesf and as far as possiblt avoid the interpretation of the secondary author (see Moravcsik 1998:83). The insight into this bargaining process gained from mom official s o m e s is slim, as the offiejal statements m d e about: the bargahing with OrEc are all part oi the bargainixrg itself. The possibility of revea1in.g inhrmation to the oppment made most official statements vague and inane. In section 9.3 the weahesses of section 9.2 are somewhat countered as the ge12era1 aspects of She basgaining relalrimshiy bemeen OITC and Norway are discussed in terms of game theory. Both of these sections codd have been presented indjviduaily. My arguxnew is that the reliability and validity of the arguments in the chapter as a whole arc strengthened by discussthe s m e aspects both in a fine-grained empirical fashion and in a m m general, game-theoreticd way Section "34 deals with the domestic factors mflucncing the Norwegian policy toward oil-producer cooperation. Here, the data are of different k i ~ ~ dFirst, s . I use some economic data regarding the importance of the oil scl-ctor in the Norwegian economy. Second, data. on the political disagmments in the Norwegian parliament regading oil issws are presented. Third, some secondary sources and reported st..atememtsbp ministers are give~z.'These sources are =liable in the quoting of statements, The secondary sources range from graduate theses to renowned scholars on Norwegian oil policy.

Notes 2 . Environmental activists attacking oil installations and public criticism of relatic~nswith regimes abusing human rights are h.tro examples of situatiom whel-e oil companies are confronted by mcic?;tyin a manner similar to that characterizing the relations beh-een state and sc~ciety. 2, Even a further specification might be necessary. The study primarily deals with the net exporters of crude oil, in other words, those oil producers who operate on the international oil market. Much a"ttentic1n will be devoted to the members of the Organization of the Petrcdeurn Exporting Countries (OPEC), atthough Chapter 9 is a eaw study of a nan-OPEC producer, namely Ncjrway. 3. ""A point such that neither side can profitably move away from it is regarded as an equilibrium poi;r;rl and the strategies tzrhicl-r give it as eqzlilibn"utn strnfegiw" (Nicholson 1992:93). 4. Monopolizati had to a great extent been rendered a legal question-how legislation could prevent the formation of such groups. In the international oil

32

Ivt froducfion

market this legal aspect of the problem did not arise, What did emerge war; what we can call the idec~logicalattacks on OPEC for destroying the liberal world economic system of free trade. 5, This division of the ecmomics mc~delsis alsc~present in the oil-market madels in Crkmer and Zsfahani (1991) and in Mabro (1992). 6. The argument is not limited to the introduction of political factors. Approaches which combine the study of the oil market with factors other than politics can increase our understanding of the oit market. Bystein Nctreng (19527) has conducted a study of the relatianship between oil and religion. '7. Tn Chapter 4 the logic of appropriateness is presented as another way actors choose between different pc~ssibleactions, In Chapter 3 the complexity of states' interests modifies the actor's ability to make rational calculations of utitities connected with the different possible actic~ns,and thus the actor" rationality is weakened. Finally, in Chapter 9 the inciusian of the dmestic political structure in the analysis of a particular actor reveals a decisic3nmaking process sc~complex that the state cannot be regarded as a unitary actor, and thus a weaker principle of rationaliv is called for. 8. In this book the structure of the international oil market is applied simply, according to the meaning suggested by Edith Penrc3se f1987:8Q):""aramework within which activity is carried out and tzrt-rich is itself not easily changed" (i.e., factors the oiX prc>ducerstake as given). 9. Walter C a r l s n ~ s lecture, , September 43, 1991, Department of Political Science, University of Oslo. 10. 1am indebted to Bent SoEus Transy for pointing out this implication. 2 1. Adelman's argument, cited in section 1.2, that oil producers try to maximize wealth is the basis .Ear the discussitm in Chapter 5. The way the oif prtjducers achieve it is through colltzctive action in which the other producers influmce the individual actor" calculation of the rational choice. The rational choice thus beccjmes a mare complex calculation than in a situation where the actor makes its choice in isolation. The actor interacts strategically with its environment, and the environment thus cannot be treated parametrically in the rational decision process, 12. With this rmderstanding, Daviclson includes what Elster calls intentional models in the category of carnal models: ""fant to defend the . . . positian that rationalization is a species of causal explanation" "avidson 1980:3). This goes along the argument presented by Searle (1983:"f15): ""some philosophers have been so impressed by the peculiarities of human action that they have postulated a special kind of causation that goes with agents*Accr~rdingtcr them there are really two different kinds of causation, one for the agents and one for the rest of the rmiverse." "arle, however, argues that "since FoltesdaYs article [F~llesdal19"71j on the subject, it has been wictdy accepted that certain forms of causal statements are inten[t]ionalM"fibid.:11";7. Elster, however, argues that "it is clear that for practical purposes we may treat intentional and causal explanations as wholly distinct" "985:23). 13. John Harsanyi (1986:89) defines game theory as "the theory of rational behavior by two or more interacting rationat individuals, each of them determined to

maximize his OM-ninterests, whether selfish or unselfish, as specified by his c3wn utility function."" 14. This argument assumes that the income f r c oil ~ ~sales ~ is used in such a way as to generate social benefits equal to the general interest rate level, 15. Proven reserves are *'generally taken to be those quantities which geological and engineering information indicates with reasonable certainty can be recovered in the future from known resemoirs under existing economic and operating conditions'' (BP Anzoco Statistical Revieiczzu c$ World Energy 1999). 16. A mucbdcltbated theme in shod-term analysis of the supply side of the oil market is the study of grc3duction capacity and capacity utilization. This topic will not be a prc~minentpad of this study. Some readers might find this tack of emphasis on capaciv strange. Two arguments explain this: First, the public information on countries\ol-produdion capacity is inhel-ently uncertain*Sccjnd, the installed production capacity is important in understanding short-term constraints on producer behavior. In the longer run, as in this book, capacity for many producers becomes a question of investment capital for exploring and developing new prcjduction capacity Such dedsions are a matter of political considerations of the economic interests of the state. This is a topic studied in this book. But for these producers, capacity is not a constraining issue in the longer term. 17. This tvas partially due tc3 the impartance attributed to C ~ I ~ Eafter C this year, in particular the political aspects of the OPEC era, such as ideological attacks on CII'EC as a cartel and its role in the north+outh. divides and the Arab-Israeli conflict. "There is no point p ~ t e n d i n gin these circumstances that the study of ar3sc can be fully objective" (Mabro 1W2:5).

This page intentionally left blank

Market Context

The aim of this chapter is to discuss the structufal conditions for o i l - p ducer cooperation, The argument put forward. is that the success of the cooperation between the oil producen; has been constrained both by hciamentd econontic factors (such as changes in delnand and lack ol barriers to entry) and by the behavim of other actors (international oil companies and cmsuming countries). These two sets of factors (denoted as market structure and market power) interact during the time period in question, changing the conditions for the success of the oil-producer cooperatim. 'I'he market power of other actors and the strucbre of the hternational oil market are factors outside the direct cmtrol of the oilproducing countries"overnments, These governments accordinglp tend tcr rt.gard such factors as given when formkg their own oil policies. The oil producerskwn behavior, in turn, hfluences the other actars%behavior and the structure of the market, and thus creates changes in the conditions for the oil-producer cooperation itsetf at a later stage. As poinkd out by Carlsnxls (1992), the relationship betwee11 actors' beh,a,vior and structure should be regarded dynarnicalfy (see section 1.4). The studies of the international oil market during the last decades have to imchde a dynamic understanding of the relationship betwen structural 'factors and the behavior of market actors, This is the aim of sections 2.3 to 2.6, The concepts of ntarket structure and market power wiil be developed in sections 2.1 and 2.2, For the purpose of this studyf the most important aspects of the ma+ket structure are the demand elasticity and the nurrtber of oil producers, The m s t prominent aspect of the market power is the control over the oil resources, trading mechanisms, and the taxation strategies of consuming countries. I'he importance of these aspects is outlhed at the end of sections 2.1 and 2.2.

Market COBtext

2.1 Market. Structure

The aim of this section is to describe the structure of the international oil market. The concept of market structure generally refers to "certain stable attributes of the market that influence the firm" conduct in the marketplaceff(Caves 4980:64), or as "characteristics that are inherent to the product or relatively impervious to easy manipulation by policy" "reer 1984:9), Cohen and Cyert (1975:14) defhe the term "market structure" as "the number and size of buyers and sellers fn the market, the restrictions that may prevent firms and households from entering or feaving particular markets, and t:he availability of hfomation about potential buyers and sellers." Adelman (1972:78) defines the market structure as ""the number and size distrib~~tim of suppliers and the degree of vertical integration." "reer's definition is wider than the others. The fdlowing discussion will be in line with this wider cmcept, although most attention will be devoted to the aspects of market structure contained in the definition proposed by Cohen and Cyert. Changes in these aspects of the international oil market arc. suppoxd to constrain the market actors' behavior, but also to make some paths of action more feasible than others. h line with the aim of this study, it is the constrai~~ing and enabling of oil-producer cooperation that will be the focus. The most important aspect of the market structure is the concentration dimension, the extent of monopoly or competition. In this study the cmcept is emptoyed in a \zrider sense. 'The market structure consists of the institutional fra~neworkbuilt around a trmsxtion and is decisive for the negotiation (Williamson 197';1:239), The market stntcturt. has thus both a vertical and a horizontal eleme~~t.

Verfical Structure Since the beginning of the oil indush.y at the end of the nineteenth century, the vertical structure has been important for the concentration of market power. The Rackefeller empire was initially built on control over refining and transportation, not control over production, The market pc~werOf thC "Seven Siskrs'" after the % c o d world War was based on their control over several stages of the production chain, This section briefly describes the key characteristics of the vertiraX shucture of the international oil market. It forms a fomdatian for the discussion in the followhg sectior~sof this chapter, and in the rest of the shl,dy. The production cbaixl of oil can be divided, into five stages:

Markef Context

37

TABLE 2.3 Proved Resevves at End of 3975,1985, and 1995)(share of total) 1975 1985 1995 Ncjrth America 8.55 13.06 8.52 S, & C. America 3.88 4.93 7.76 Eurcjpe 4.29 4.00 1,74 12.06 8.60 5.61 (Former) USSR Middle East 55.25 56.14 64.85 Africa 9.76 8.00 7.19 Asia & Australia 6-22 5.26 4.34 Total World 100.00 400.00 1Q0.00 Of Which: QECD 12.68 17.03 10.22 OPEC 67.44 66.80 76.53 SCWRCE: BP S tat islz'cal Revkm of World Erzergy 1996.

I. 2, 3. 4.

exploration production transportation refining 5. distriibution

Successful exploration assumes actual presence of oil deposits. AIthough s m e advantage can be gained by technolsgy and know-how regarding the discovery and determination of oil fietdls, the location of lowcost oil deposits is the first determinant of the structure of the oil market. nble 2.1 shows the proved resenies at the end of 1975,1985, and 1995. Also at the production stage, the geographical location of the field is the main determinant of the pmduction costs. C)ffi;horedrilling hcreases costs cornpared with onshore activiq, and, for instance, Aiaskan or Siberian oil drillix~gtakes place in a far more difficult climate than drilling in the Saudi Arabian desert. Furthermore, there are different crude qualities, as well as differences h the location of deposits and their quantiv and quality, which witl cltetermine the costs cJf depletion of individual ail deposits, New drilfhg techiques have been develoiped to increase rate of depletion and lower the costs of production. Extraction rates have been improved by horizmtal lirilling and injection programs. In addition, the areas under consideration for exploration are constantly expanding due to technology enabling companies to prodwe oil in deeper waters and in harsher climatic conditions. The transportation ol oil can take place both beforra m d attes rcfinhg. Pipellne transportation often gives rise to what is known as a natural xno-

38

Market COBtext

TABLE 2.2 Refinery Capaciv and Throughputs, 2 995 Thomands of Barrels Per Day (th. bid) Capacity T!iro.szkgkputs Lltifizatl;on I%) United States 15,235 13,970 91.'E3 Canada 2,835 1,560 85-01 Mexico 1,520 1,315 86.52 S. & C]. 6,235 4,835 77.55 Europe 16,595 14,395 86.74 fc3rmer USSR 40,325 4,900 47.46 5,315 5,145 96.80 Middle East Africa 2,860 2,315 80.94 67.50 China 4,015 2,710 Japan 4,865 4,170 85.71 ahers 7,5125 7,600 99.67 'Tbtal Wilrtd 76,425 62,915 82.32 S~UI~CE:BP Statislicat Review qf 1.1Jorfd Energy 1996.

nopoly. However, most oil trmsport today is by ship, except for shortdistance transporta6on from producing wells to shipping terminals. At this stage of the production chain, the ecmomics of the oil industry are comected with Che shipping industry. Subsequently, the interests of the shipping hdustry and the pol-itical importance this industry plays in s o w countries might have an important influence on the oil industry. Tlze refil~eryindustry plays m important role in the oil-production chain, Che aspect has been the location of the refineries: ""Since the Sccond W r l d War there has been a steady trend towad locathg relfineries near consuming centers instead of near the sources of crudc-oil production, WI-rereas in 1939 some 70 percent oi refining capacity outside North America m d the Communist countries was near the oil fields, this had dropped to about half by 1951and to about 16 percelnt in 1965."" Another aspect of the refinery indust9 is the overall capaciw surplus during the last decades. ?Bbie 2.2 shows the distribution of mpacity and throughputs for some key areas. Since this stage of the producrion chain is possible to locate anywhere, contrary to the other stages, it is liable to the economic interests of states' national industrial policies. Subsequently argumellts regarding employment, national control over resources, and national economic and industrial development and growth becme important pditical facltors h the location and m n h g of the refhery industry. Unlike the other stages, the last me, distribution (which includes enduser sales), obviously has to take place whew the customer is. It is therefore? impossible for aclivit.ies at this stage of the production chain to es-

Markef Context

39

cape national jurisdiction and control. An international oil company can locate its activities at other stages more or less in a cwntry of its own choice, but this is not so with distributim and sales. Suhsequen* these aspects of the oil market are more influenced by particularities of national markets and local governmental regulations, For the purpose of this study, the different stages of the pm"dction chain rczprcsent different conditions for the relationsh* between Che actors in the oil market. The main focus of this study is on t:he crude oil market (i.e., the production stage). However, actors tend to use their power t?l. other stages to influence the development in the crude oil market. A p m i n e n t example is the way in which the producers attempt to increase consrtmpticm by rttducing the crude oil price, \Nhich is futile as long as the c m s ~ ~ m i ncountries g hcrease taxes on oil co~~sumption (see section 2.5).

As pointed out irr the introduction, the horizcmtal strucme has more direct innplica.t.ions for the oil-producer cooperation, as it reldes to the structure of the crude oil market as such. Prominent aspects of the structure of the cmde oil market are 1. market concentration 2. product differentiation 3. barriers to entry

In Figure 2.1 the market concentration is indicated by thr market share of the five, ten, and fiiteerz largest producer countries in the international oil market, Figure 2.1 indicates a modest change in market concentration. Although the five largest corntries have reduced their marktit share from 69 pescent in 1965 to 46 percent in 1995, the fifteen largest countsies have reduced their market share by a more modest 12 percmtage points, from 93 to 81, perce~~t. h his taxonomy of mnrketsl Wifiiam Shepherd uses the term "tight oligopojry" to describe a situation in which the leading four firms have 60 to 100 percent of the market (Shepherd 1990:lil). The other aspect of the market structure is the market share of individual producers, Figurc 2.2 lists the fifteen l a ~ e sproducing t countsies in 1965, while Figure 2.3 gives the same list for 1995. The followli,ng trends can be ext.racted from these rjgures. First, among the fijFteen collntries listed in 1995, only two are not on the 1965 list: the United Killgdom and Norway 'This indicates a stable pattern of market shnres. Scond, the differences between the top Hfteen countries have de-

FIGURE 2.3 Market shares, 1995 (%) Based m data in 81')Slatistiml Rcstiew of Wodd Ener~y,various issues.

SC)UNC'E:

Markef Context

43

creased. Third, disregarding the U~zitedStates and the former ussw, which have substantial domestic consumptlim, &e non-o13EccountsiesWxico, Norway, C h i ~ ~the a , United Kingdom, a d Canada-onstitute 21 percent of world production, compared wi..lh Canada, Mexico, and China" combined market share of 4.71 percent in 1965. It decreases C>I?EC'Sability to cmtrol the market when the main competition is concentrated in a few oher countries, ratha than having a large number of outsiders taking a substantial share of the market. Product differmtiation in the d e oil market is basically the differclnt oil qualities. These differences are express& h the prjce diffe~rztialsset by the extra costs of refning heavier crude oil. The margins can change sowwhat due to differences in demand for diffewnt qualities. The simple rule is that the m a h quality hdicator is the am (American. Petroleum Tnstitute) gravity-the higher the gravity, the higher the price. But the demand for differcmt yudities changes, seasonally m d over time. Also, different sectors of consumption can experience di.fft.rent demand paths, causixzg the composi.tion of demand for oil qualities to change, More important for the cooperative bargaining betwem the oil producers is that quality diffemtials can be an effective way to cover price djscounts. If a higher-priced oil quality is discounted, th price can in nomixlal terms be hif?;herthan prices of heavier oil types but still be valued as cheaper by the refinms. It thus imposes pressurcz on the price of thC other Oil qudities. Different oil regions have different compositions of oil quali.ties, making prices vary acmss ~gictns. There are few technical barriers to entry in the international oil market. There is, however, one fundamental barrier to entry in the location of resources. This was discussed in section 2.1. There are obviously different levels of techological skills m o n g oil producers. In particu,lar, when ojl is first discovemd in an area, the countries cmcerned usually lack the ability to develop the resources on their own. However, in the present situatian, with the international oil companies operating globally, the owner of the resource can buy the necessary expertise from them as long as it also provides sound investment concjition. As l o q as oil prices art. well jbovc development costs, other producers' behavior has little effect on development decisions of new potential producers, Given the existence of oil resources on their territory, states tend to encourage the de~relopmentof them, The concentration of oil production makes the crude market central to the understanding of the total oil industry "Supply is unlike demand. The morlnt of the oil demanded results from. independcnt: choices of millions of households m d firms. . . . a few governments make the most important supply etecisjons" "de1man 11993b:4). I-lowever, supply-sidecreat-ed price illcreases might triggt3r consttrner reactions, such as more

Market COBtext

44

efficient use of energy or substitution. As section 2.5 will show the demand side effects are important for understandhg the possii7lc success of oil-producer cooperation, Iblnrkf Slrucfzcre and Oil-Producer Cooperation

The market structure is important both for the market pctkver and for the possible success of oill-products cooperation as such. The vertical structure described in section 2.2 is an important featurt; for the changes in market power am011g the market actors (see sedio~l2.2). 1.1is also important for the development of new tradin.g mchasrjsms and thus the market power of the prcxrtucers (see scrction 2.6). The horizontal market stmcturt. influences the markt power m o n g the actors-for hstmce by inte11sify.Q competi"tivepressure on th producer or consumer side. The burizmtal stmcture, demand elasticity and the nun-tber of groducers also have a direct effect on the possibi%ityand success of oil-pmducer cooperation.. The demand elasticity is important for the success of oil-producer cooperation, as it &term&es the relationship between price and conswed volume. If demand is inelastic, small changes in production create large price changes, If dennand is m r e elastic, the price effect k m rclgulatim of production is smaller. Thus, in a situation of demand &elasticity oilproducer co~pem"ti.m more easilly creates large bellefits for the producers than a situation of elastic demand (ceteris paribusf. The recent history oi the international oil market sqgests that demand for crude oil is inelastic in the short: run and morc elastic in the longer run.. The reason for this is the lag between increased oil prices and consumersbadaptatim through cctnservatim a d substiktion. Tl~issuggests that oi producers cooperating by limiting output m able to reap substantial benelits in the short term, but the strategy can be less profitable in the longer term. The number of producers is obviously m hnportant aspect for the pot;sibility of successful cooperation between the producers" The larger the number of oil producers, the harder it becomes to establish a successful cooperation between them (Olson 1965:34-35). 'The number of producers has hcreased during the period in question, partly due to the price policy of OPEC, whjch has made new production areas profitable to develop. I'he established producers have not been at7lt-1to set up any harriers to such new ewriel;. This has m d e successfCul ojl-producer coopera.t.ionincreasing1y difficullt, 2.2 Market Power

As mentioned in the introduction, both fundamental economic aspects (as described h the previous section) and other actors' behavior co~~sti-

Markef Context

45

tute the e n v i r o ~ ~ msurrounding e~~t the oil-producer cooperation. 'This section deals with the other actors%ehavior. 'hpmposition is that oillproducer cooperation is constrained by other actors' behavior. Market p w e r is the basic concept developed to understand the inhence of other actors%ehavior un the oil producers9reedam of action in the international oil market. Market power is defined as ""the ability of a single, or group of buyer(s) or seller(s) to infIuence the price of the product or sevice in which it is trading. A perfectly competitke market in equiljbrium, ensures the complete absence of ma+ket power" "earc.e 3983:274). All the factors discussed in section 21 concer~ningthe market structure are important in d e t m i n i n g the market power of the sellers. The vertical strucme gives rise to actors' strategic positions in controlling other actors' outlet of oil or access to crude oil. m e market concentration is the basis for coherent action by a group of actors that leads to the realization of monopdy profit. Barriers to entry ensure this profit against outsiders. And, filzally, a high dernmd elasticity ~ I I S L E ~ that ~ S consumers will continue to buy the product Lalfhough the price is lncrcased due to the sellers" collecting a monopoly profit. It should be noted that this study concerns the m a r k t power of the sellers, The final consmers of oiX rider in the billions. Even the number of refineries a d oil companies around the world is very large. Thus, the evmts on the supply side of the market art. mare susceptible to individual or groups of actors' behavior than the demand side, where the large n u b e r of actors evens out differences in their behavior. This is not to say that the understanding of oil demand is less important, only that it is mare in line with traditional.ecmomic models of consumer behavior and less relevant to the core of the research question of this book. Given the assumption that the differe~~t groups of actors that obtah market power in the crude oil m r k e t will set the price to their own benefit, t'ne distribution of market pwe"mplies the distribution of the monopoly profit. 'The possible exjstence ol monopo:ly profit in t h industry iollows from the definition of market power, as the campetitfie market is defined as a market without any group of actors having market power. The )OSS of market power by a group of actors cottfd mean two different things: either that the market has become more competitive, or that another group of actors has fncreased its market power. Both aspects will be discussed in the empirical sections of this chapter. The market actors can hold tangi:bXe power resources like physical control over oil resources or production facilities; ownership of distribution channels or marketing facilities; and cmcessions to conduct business at the different stages of the production chain where such concessions are necessary. Actors might aiso hold htangible power r e w m e d i k e informatim about reserve deposits, production methods, or trading instruments. This emphasizes

Market COBtext

46

that market power is exercised in both the vertical and the horizmtal market stmcture. In the f01LoLVing subsectims, it is argued that an understanding of changes in the distribution of market power is the key to understandh g changes in the international oil market, This calls for some comments. Given the definiticm of market power, there is, in a sent;e, a fixed m o u n t of power at every stage of the vertical production chain. At each step the market actors are in a zero-sum. game over the ability to determine the price of the oil traded. It fntlows that the same must he true Em all the stages taken together. One actor's exercising of market power at one stage &us cmstrajns other actors' exercisjrng of their market p w e r at other stages of the production chain. At a given stage, market power is a relative concept. The higher the concentration is on the seller side, and the lower it is at the same time on the buyer side, the higher is the sellersbmket power, and vice versa. If there is low cmcentrati,on on both sides, market power is also low for all actors (see Pearce 1983:274). Wth a high concentration on both the seller and the buyer side, as for instance in a situation where. a mcrnop.oly is seIIing to a monopsony, the monopolist has the advantage if the buyer experiences an inelastic demand. These aspects will be discussed empjrically in sections 2.3 to 2.6. The Trilateral Oli.lrpoEy

The actors in the internat-ional oil market have been grouped as follows (Adelrnan 1977; Roncaglia 1985; Mikdashi 1986): I. oil-cmsumhg countries 2, irtternatimal private oil firms, upstream md. hwnstrearm 3. oil-producing countries, including state-run procluch7.gfirms Roncaglia (1985) describes the international oil market as a trilateral oZigc"p"ly made up of these three gmups of actors (Figurt-2.4). Inside the grottps, in each corner of this triangle, we find attenpb to coordinate behavior. The main mason for such cooperation is to strengthen the groups" position in refation to the other actors in the market. One =ason for establishing the cooperation betwem the oil companies was to ensure the strength of the jndividual company in relation to the authority of producing states.3 'The producers' main reason for establishing OPEC was to counter the market power of the major oil companies. The consunter countsies have established the International Energy Agency (IEA) to bdance the market p o w a of OI~EC.The tdaterai digopdy is ckaracterized by both inter- and intra-group cooperation and cmflict.

Markef Context

47

PRODUCERS

CONSUMEW

Figure 2.5 highrights the main conRictual aspects of the relations in the trilateral oligopoly. The ~iationshipbetween consumer and producer countries has a political-ideologicd dimension, as b y oil producers are Iformer colonies of the consumer countries. The main conflict beween producers and consumers has been connected with the supply disruptions followi,ng politicd confljcts in the Middle East and the ideological attacks on OPEC as a cartel: disrupting the international:free tradc liberalism a&ocated particularly by the URited States. Theoreticallyf Ihe companies' cooperation with the prodwers or the consumers would ~spectjvelyincrease or dccrease world oil prices. The companieAhave been attacked by the producers for siding with the cmsumers. As OPEC gained strengt.h.in the early seventies, the first goal was to reduce the companies3nfluence on the price policy. On the other hand, in the famous congressional hearing before the subcommittee on multinationaf corporations, the so-called Church contmission, in 1974-1975, representatives of the international companies met senators highly critical of the role of the cmpanirs, suggesting that they cooperated among themselves and with the producing com~triesta drive prices up and thus increase their own profit at the expense of the Amric m consumer. The cooperative aspects of the inter-group rcllali.ons highlighkd in Figure 2.6 also show how the relations have been perceked by the actors as beneficiaf. As will be further eteveloped in Cbapter 6 on Saudi Arabia, key producers for a long time have sought military support from Western consuming countries in exchange for a somewhat more consumerfriendily oil policy. I'he exchange between producing countries and the international companies is likewise a cornbindion of both conflictual and cooperative aspects. The consumer-company relationship is rather peripheral t~ the focus of this book, 'f'he ~lationshipwas more important in the fifties and sixties than it is today but this relatio~~ship also was char-

Market COBtext Polrtti-a1conflicts in {fie Middle East create supply disruptir>ns. The Opec ccarzd vialateh princip1es of economic likralism,

Compar-iics"cooperation with

\

FIGUM 2.5 ConEIietrral aspeds of intergroup relations

acterized by both cooperative and conflictual aspeds. The campmies to a certajn extent benefited from consumer countries" political backing in their operations in the pmducing countries, and the consurner countries could use the companies as a spearhead in establishing political ties wi& strategically important countries in the Middle East. This book is devoted to the study of the intra-group ~ l a t i o nbetween s members of one of the groups in Che trilateral oligopoly-the producers. Having this aim in mind, it seems fair to confine this empirical outline to the intra-group relations between the producers. In sertictns 2.3 to 2.6, intra-group aspects regarding consumers and companies will be discussed to the extent that they influence the cmditims for producer coopaation. Atso on this intra-group level, we find cooperative and cmRictual aspects..Three possible expl,anat.ions of cooperati,on, or lack of it, will be illustrated: l. the prcsence and role of a kgemon or leader 2, the constellation of hterests 3. the effect of exagenous hocks

The first aspect is based on the hegemonic stability thesis. FXegemony is loosely defined as a highly asymmetric distribution of power hfwor of a single actor"The hegemon is supposed to have hterests in line with the collective of actors, and the abi%ityto supply the "collective good." Ammg the producers, the role of Saudi Arabia in the recent history of the internat-ional oil market resembles that of such an actor. 'The hegemmic stability thesis also implies incmased codiet if it shodd happen that the hegemon is weakened. These aspects are discussed extensively in Chapter 6.

Markef Context

FIGURE 2.6 Cooperative aspects of intergroup rebtians

The second reason for the emergence of cooperative behavior is the fact that actors have c o m o n interests, in other words, the presance of an integrative potential. The question to be asked, is what the actors, hdividually, can gain from coopaation. The relationship between the pducers is close to a classic case of the prclblern of collective action (Malnes 1983).The point of departure for application of the ""Prisoner" DUilemma" is that all actors have an fnterclst in Limitation of prociuction in order to ensure a high oil price. At the same time, it is in the interests of all actors to sustain their own production volume, In such a case, all actors can gain by cooperating (ail reducing production), but individudy they can earn more by not cooperatkg while the others do. 'These aspeds are discussed extensively Fn Chapter 7. Understandin.g the oil price as a collective goad m o n g the oil pmducers implies a posibility of free riders. 'The large non-OPECexporters-MexicoI China, Norway; and Canada-but also some of the OPEC members acted as free riders during the ei&ties. These aspects of the relationship between the producers are discussed in Chapter 8, and Chapter 9 is a case study of Norway, emphasizing bargaining prOblcrns related to the Norwegian Free-rider strategy. The third aspect of the intra-group cooperation is ~ l a t e dto the existence of so-called external shocks-events outside the group that bring them closer together or pull them further apart, The formal pmducer-cooperative organization, OPEC, was fonned in 1960, but not as a resutt of my crisis. However, in the early eighties, OPEC made decisio~~s that actually restricted the individual member" sail policy by htroducing production quotas (see Chapter 5). This decision was a result of the effects of the previous crisis. OPEC had to compensate for the demand decrease m d the increase h non-orsc pmducers, whjch had emerged as a result of its own price i n c ~ a s e sconnected with fie lranian revolution and the Iran-lraq War (see Chapter 3).

Market COBtext

hlarket Poroer and OiI-Pmdzlcer Cooperafion I'he changes of power among the actors in the trilatr~rafotigopoly have implications for the possibility ol estabEshing a successt"ul cooperdion between the oil pmducers. In other words, the rehtimship between mernbers of one group is influenced by the relationship between the groups. This is the essence of the perspective develoiped in this section on market power. n r e e aspects will be given particular attention in the foflowing empiric.& sections. First, the co~~centratio~~ on the productio~~ side is much higher than the concentration on the consumption side, This gives the actors cmtmlljng the oil =sources a potential market power over other actors. The transfer of the control over these rcirsottrces from the international oil companies to the oil-producing states in the first half of the seventies fundmentally changed the basic conditions for any rczlationsh* between actors i n this market. Although this skndy is prharily concerned with the oil market after these changes, it is unavoidabk as a premise ior the dlscussims presented in most of the study m d in &is chapter in particular. Another aspect to be dealt with in this chapter is the role of different trading mechani.sms in the international oil market. The physical cargoes of oil can be traded in different ways, implying different levels of producer control over the price-setting*brig-tern Materai contracts with rigorous price clauses will bcnefit the selfer if prices fall, and.the buyer if prices inc:rease. A market characterized by m m transparent, market-determined contracts will not create the same asymmetry when prices change, as the actorshadaptation will be more incremental and cmtinuous. The changes in tracfing mechanisms during the eig:hties were in themselves not negative for the producers, but since they came at the same time as prices fell, they removed, some instmments with whirh the producers otherwise. c o d hawe sbwed down the price fatl. 'This slnows how market power can be exercised in the vertical structure, and has imptications for the effectkernss of oil-producer cooperation. A third aspect related tcr the market power among a gmup of actors is the taxation policy of consuming countries. X 1the producers' cartel policy fails and they attempt to increase the quantities of consumed oil by red oil price is passed ducing the price, it is necessary that the ~ d u c e crude on to the end-consumers. If instead the cons~~nning corntries ilncrease taxes, the cmd-consumer has no price incentive to increase cmsumption. This is exactly what has happened in the oil market during the last twel~tyyears- Thus, the oil-consuming corntries have increased their market p w e r through this mechanism. Havirrg outlined the cmcepts of market structure and market power, and idmtiiied the main groups of actors in the international oit market, it

Markef Context

51

is now time to turn to the essellce of this chapter-the dynamic hterrelationshiy between market stmcture, market powet m d oil-producer co~ aspects operation. 'The purpose of the following sections is tc:,s h how of the m r k e t structure created a situat-ion at the beginning of the sevellties that made it possible for OPEC to take control over the internatimlal oil market. 'The subsequent behawior of OPEC during the swenties for, more corrtlctly, until 1981)changed fu13dament.d structural aspects of the market. This once again changed the structuraX cmditions for the successful OPEC cooperation in the late eighties, making it harder and more costly for the producers to pursue a cooperative strategy m o n g themselves-a situation promptly taken advantage of by the other actors in the market. The stmcbral trmsformation of the oil market since the Second World War has hvolved a develoipment thmugh three stages: from an integrated market dominated by the large internatimlal oil companies, to a market domhated by the oil producers, and fhally to a m m complex market structure comprising only partially organized groups and division of markt-tt power among producers, consumers, and.the international companies. h sectio~~ 2.3 the empirical discussion is related to how aspects of the market stmcturc m d other actors' behavior, in particuhr that of the international oil companies prior to 1971, werc. important h forming the oil producers' key desire to gab1 control over the setting of the oil price. The hskumestt for achieving this was the cooperation between some owc mernbers in a joirrt strategy against the companies (see sectim 2.4). This cooperation changed the market structure and the power balance amollg the market actors (see section 2.5). The OPEC countries did, however, overexploit their position, leaving room for both the reof the compmies and an inercased position of the consuming tountrjes (see scction 2.6). 2.3 The Era of the International Oil Companies

The comercial production of oil started in the United States in 1859, in Titusville, E'ennsylvmia, Oil demand grew rapidly in the follovving w s twenty-five decades. At the turn of the century, world consumptio~~ tirnes the level oi that of 1870, Most of this expansion took place in the United States, where the market was soon dominated by the Rocke.fellerst Stixndard Oil, whose share of refinery capacity rose from 10 percent in 1870 to some 90 percent ten years Xater. The near-monopdy situation of Standard Oil in the us market was built on its position in tple vertical mrket structllre and &us the establishment ol a monopsony position toward producers. It is important to note that the role oi Standard C)il was "oased on a strong po"ition in the downstream segmmt, not on access and cmtrol over resources. For other cmpanies the co~~trol of re-

52

Market COBtext

serves became more important, as did the hancial strength to develop such reserves in undeveloped areas. Furthermore, the development of new transportation facilities was also an important means of competition for companies facing Standard Oil in t-he international oii market. At the begir.rnj.ng of the twentieth century the focus on the international market iz~tensifiedamow the companies, but: not m o n g states: The level of government involvement during this period was low due to the fact that oil was mainly seen as an unproblematic commodityt which strategic importance and exl.laust.ibilityhad not yet been placed on the agenda. . . . An exception to the low level of government involvement internationally during this period was the establishment of Angle>--Persian (now BP), where the British government entered as a major shareholder to secure the need for fuel oil for the British navy on the eve of the oncoming War, (Bergesen, Bj~rk,and Claes 1989:Il-42)

After the First WorId n"Br the irtwarb-lookhg attitude of the us government changed. The lack of access to foreign oil supplies was then regardcd a ""srious international problem facing thc Ulniled States" (Vergiin 1991:194). The same went for the French authorities (Yergin 1991:189). The area to look for oil was the Middle East. The British made Mesopotamia a British mandate under the L~agueof Nations. In comection with the San Remo ag~ement,an Anglo-French oil agreement was negotiated."'France wollld get 25 percent of the oil from Mesopotamia. . . . the vehjcle for oil development rcmained the Twkish Petroteu~n Company . . . and the French acquttred what had been the German share in it. . . . the French gave up tbeir territorial claim to Masd, Britain, for its part, made absolutely clear that any private company developing the Mesopotamian oil fierds woutd, very definitely be under its control" (Yergin 19%9:189-190). Access seemed closed to US inte~sts.However, with the breakdawn of the Ottoman Empire, the status of the Turkish I'etroleum Company concession was andear, and the companies started a long and bitter fight for influence in the formerly Turkish-domfnatclet area. The us governmnt responded by invoking "'the open door policy," which had three elemmts: 1. that the nationals of all nations be subject, in all mandated territc>ries,to equal treatment in law, 2, that no economic concessions in any mandated region be sc~ Large as tcr be exclusive, and 3. that no monclpolistic cmcessictn relating to any commodity be granted. The United States Government maintained that the war had been won by the Allied and Associated Powers fighting together, and that, cc>nsequently,any benefit, tzr hether in oil interests or ofhewise, should be available to the nationals of all Allied Pc>wers,

Markef Context

53

and should not be seized by those of any one particular power. Mcjreover, the United States asserted that the San Remv agreement discriminated against the rights of American nationals, that no rights in Iraq tzrere vested in the Turkish Petroleum Co., and that no valid cmcessions cuuXd come into existence thmugh the government of the people of the territory. The British point of view was that British nationals had ""acquired rights," that these rights must be respected, and that, although the United States had been an Allied Power, this fact gave its natianals no right to trespass upon acquired rights. The term "acquired rights" referred to the rights held by the Turkish Petroleum Co. and the rights prc>mised to that company lay the Ottoman Grand Vizier, as evidenced by his letter of J m e 28, 2924, to the British and German Ambassadors. (Federal E a d e Commission (~Tc)4952:51-52)

After yearlong negotiations, the United States, the United Kingdom, and France rcached a cornpromise in 3928. As shown in Bible 2.3, the American companies got a fourth of the Iraq Petroleum Company P C , fmmerly the Turkish. Petroleum Company) concession. The figtare also shows the substmthl UK and French governmatat stakes in the companies co~~stitnxthg the TurEsh Petroleum Company-a situation not likely to be the case for the us authorities. Furthermore, the companics and authorities also agreed to the soc d e d self-denying clause of 1914, stating that all parties should work jointly-and only jointly-in the region (Yergin 3993:204). The region included the Arabic peninsula (except Kwait), Iraq, and Twkey. This was the so-called Red Line Agreement (see Figure 2.7). In the meas inside the red line, the companies would pursue joint concessions. As soon as the US companies mentioned in Table 2.3 were inctdecf in the agreement, the open door policy was abandoned tand the door was shut to any new company, us-based or not, By 3928 more than 511 p e ~ e n of t oil productim outside the Mited States was controlled by Exxon, Shell, a t ~ dBritish Petroleum (RP). These companies met secrcltly and worked out a market-sharing deal, the socalled As-is-agreement. This was an agreement to keep the respective percentage market shares of sales jn various markets. Anolt-rer ifnportmt point in the agreement was the "Gulf plus pricing system," 'according to which crude was tc:,be priced as if pmduced in the Mexican Gulf regardless of actual origh. Cater the cmpanies also agreed to contrd prodtlction, Most of the other us companies joixled the agreement. The various agreements c0verc.d operations in all countries except the United States and the Soviet Union. By the end of the twenties, the companies had set up agreements governing their interrelations in the whole production chain.%e s c a ~ i t ythat had made the US companies so eager to get into the Middle East production area soon turned into a giant sur-

Market COBtext

54

TABLE 2.3 The Division of the Turkish Petroleum Cornuanv in 1928 Sarbsidia~ Ozunm Anglo-Persian Oil Co., in which the British D'Arcy Exploration Co. Ltd. government held 51% Royal Dutch/Shell (Royal Dutch: 60%; Shell: 4070)

Share 23.75

Anglo-Saxon ]Petroleum Co. Ltd.

23.75

Near Eastern Development Corp.

23.75

Compapie Francaise des PPtroles (CF13),in which the French govei-nment held 35% Standard Oil Co., New Jersey: 125%; Standard Oil Co. of New York: 25%; Gulf Oil Corp.: 36.66%; Atlantic Refining Co.: 16.66%; Pan American Petroleum and Transpc~rtCo. (subsidiary of Standard Oil of Indiana): 4G.%%

C. S. CulbenErian WUKE:

Participation and Investments Co.

5.00

Anderson 1981:I C).

plus as new discoveries were made both inside and outside the Red tine area. 'The downward pressure on prices increased. After the Second World War the international oil market was dominated by seven companies, populariy known as the ""Seven Sisters." These companies determined the order of the oil market. 'The consunnhg countries were dependent on the compmies~positimin the upstream sector. Tbis dependency coutd be exploited, and the consuming countriesf authorities have always had an alnbivalent atrt-ihde toward the integrated oil compal.lies,h The Seven Sisters accounted for virtually alf the oil produced outside the tlni-eed States and the Sr?condWorZd. They were integrated h the sellse that they controlled the entire production chain right from exploration to sale of the =fined products. As of 1953 these companieucontrolled 95.8 percent of the reserves, 90.2 percent of the production, 75.6 percent of the refi,ning capacity, and 712.3percent: of the product sales,' This created a stable structure as long as the oil market did not expmd: Only a few firms were capable of the risky search for oil in remute often harsh places. In each consuming country, refining and marketing was a small indust-q prc>tected by distance and gc>vernment,rna king entry diffi-

Markef Context

55

FIGURE 2.7 The Red tine Agreement, 1991:205.

liC3URCE: rergin

cult and unprofitable. Prc>ductionwas too risky without an assured outlet, known as "finding a hame for the crude." Refining was too risky without an assured supply crf crude. Hence in each country the few sellers tzrere emfronted by few buyers, and neither side wished to be at the mercy of the other. The obvious SO]utim was vertical integration. (Adelman 1995:44)

The Sisters also orgmized their operations in the Middle East through a consortium in which a13 the major corrrganies were engaged in at least two countries (see Table 2.4). :In this way, the Sisters stood stronger against possible regulation by the producing com~tries,as none of them were totally dependent on the will. of one g o v e r n & only. The Mid.dle East oil became increasingly ilnpcrrtant to the companies as it graciualty replaced oil from us domestic somes, hvlnicfi to a gxater extent was split

Market COBtext

56

TABLE 2.4 Ownership Shares in Middle East Production Distributed to Companies (%) Satrdi

Exxon T'xaco

lioCal Mobil Gulf B13 Shell CFP Iricon Grxlbenkian ~ U R C E :M N C Hearings 197'4,

5

up into several owner interests after American antimonopoly legislation had been enforced t?$ainst Standard Oil early in the twentiet.h century. Later, the antitrust laws mndered an extensive cooperation between the Sisters impossible in the Unfted States (Odefl4986:16). The experier~cesand military operations of the warring parties during the Second World. W r established the hamework for the poliiticd elements of the oil-market strucbre after the war. Havhg secured access to foreign petroleuln resotlrces as a vital part of their war s t r a t e the ~ ~ parties did nut see the peace settlement as a reason to dissolve their control over these ~sorarces.CX1 the c o n t r v , as consumption in industr)i and consumer markets increased, the companies had stlbstantial economic interest in maintaining control over the internationd oil markczt, Without the war the need for gwernmental involvement was perceived as less immediate. 'This, together with the overall establishment:of- a somwhat: liberal international trade regime, prompted the most important political t actors, the UK and US govemmmtti, to withdraw from d i ~ cinvolvement in the kternational oil market, leaving large room for the internatio~~al oil companies. In a period with an abundance of oil avaiiabk at low prices, the pclXiticat interference in the market disappeared during the fifties and. sixties. The next time pobtjcal actors directly interfe~di,n the international oil market, it was the oil producers who took advantage oi changes in the market strucbre that we= in their fwor. These chmges are the topic of the next section.. As Figure 2.8 shows, the general energy demand increased substantially after the Second World War. UI~til1973 there was an unhrlrken trel~din the demand increase. The price increases during the seventies

Markef Context

57

broke this trer~d(see section 2.4). =ring the eighties m d nheties the increase leveled off (see section 2.5). The demmd for different energy sources dso changed in the same period. 'The main feat.ure of this change was the reduction in the use of coal and the ixrcreased consumption of oil (see Figure 2.9). The demand for coal in the QECD area fncreased by same 25 percent from 3950 to 1987, while Ihe dcrnand for oil increased almost 490 percent in the s m e period. Natural gas and nuclear power ixrcrclased, but fornned a, minor past of energy liemand. As shown in Table 2.1, in 1995 two-thirds of the oil reserves were 360cated in the Middle East, The coal reserves were located in major consuming countries like the Unikd States, the United Kingdom, and Germnny. However, in the fast-expding transportation sector, coal was unusabIc, The changed fuel composiltion subsequently also meant that the energy trade increased. This gawe the integrated international oil companies a uniq~zeposition. Thanks to a cadination of rapidly increasing d consumption and low production costs in the Middle East, the international cmpanies =aped a considerable profit. This profit na.t.urally proved attractive to other private companies as well. as national oil-importing companies. There were two strategies open to these companies if they wished to gain entry into the market-tky could develop a new area for oil produclion outside the area controlled by the Sisters, or they could attempt to break the dominance of the "majors" in the Middle East. The French company CFP employed the former strdegy, opening production in Algeria in 1959, while the Italian national company ENI (Ente Nazionale Idmcarburj) chose to enter into a joint venturt~agreement with nlroc (the Natic,nal Irmian Oil Company) in 1957, providing ENXwith access to Irmian offshore territory, A number of similar agreements beheen exporting countries and smaller oit compmies followed (Schneider 3983:B-81). This created steadily increasing competition in the market, which in rum put pressure on prices. Among the main newcomers were cmpanies such as Getty Oil, Phillips, CFP, and Occidental. The fast played a cmcial role-in f 970 it becme the first to yield to Libya's p ~ s s u r eto raise prjces and increased tax rates to proctucer countries (Terzian 1985:120). This forced Shell" independent partners in the Oasis G o u p to follow suit, and only a month and a half later Shell gave in to I:.dbya's dmands. The developments ixr the international ail, market d u r M th sixties illustrate the general point that, as markets expand, cwpetitive p ~ s s u r t . increases. It is easier to be a monopolist in a m l l market than in a large one. The integrated structure created barriers to entry at the company level, but when the market grew, these barriers became impossible to maintain. Thus, the relations between one group of actors in the trilateral

FIGURE 2.9 Primary energy consumption, 1950 and 1995, different fuels (%) Darmstadter et al. 1071 :l3 and BP Slntislical Review cf Fnit~rfidE~zcrgy.

SOURCE:

Market COBtext

60

oligopoly, the hternational connpanies, changed the conditions for cooperation bcrtvveen the oil producers. The chmgcr infavor of OIJEC'Sinterests was enhanced by the continuous increase in oil demand dufing the sixties. In addition, there were no substitt~fesfor oil in the short term, and there had been no incentives to conserve energy, 2.4 The OPEC Era

The competition between the companies, between the majors and the newcomers, weake~~ed their common position. in the crude oil market. The lack of unity, combined with their jncreased dependency on Mid.dle East oil, reduced the companiesbarket power. The companies' position was to bc chnllenged by the producing colmtries. In this section, I will address the importance of this challenge for the oll market. The roIc oi pcriticai events, the institutional creation cJf OPEC, and the internal bargaining among the oil producers are discussed k~the following chapters.

Gaining M a r h t Poruer As described in Figure 2.8, the fiftjes was a time of rapid growth in demand. However, pmductim capacity grew even more rapidly As there was n o t h g holding this oil from the market, the official or "'posted" prices were undermined by competitive discounting, and the saying went: '"only fools and affiiiates pay pofied pricesf"tAdelman 1995:54). This cseated a dilemsna for Che companies' rdations with the host governments, as it was the posted price that was the basis for the producer countriesf taxes and royalties. Both in 1959 and 19611the international oil companies cut the posted prices. This a n g e ~ dthe oil-producing governments and triggered tlne establishment of OX~EC in September 1960. "'Afthough the cuts in posti~~gs (as finalized in n-tid-September 1960) were not rescinded, t-he formation of a united front by the main oil-exporting countries ensured that the majors made no future reductions in the unit =venue of MIddie Eastern host governments, regardfess of the continuing downward tend in market prices" "vans 1998:72). "'13 other words, the Middle Eastern governments werc being kept whole, while the companies absorbed ail the effects of the price cuts" (Uergin 1991:515f. 'I'he posted price &us became an artificial number ul-r~latedto market prices. As Addman (19YS:55) concludes: "Many writings on oii markets are spoiled by rdiance on meaningless posted prices." F-Xnkvever,on Ihe political scenc, the price cuts by the hternati.onal oil cmpanies were to trigger the formation of OPEC (the Organization of the Petroleum Exporting f3ountries).V~oliowingthe production cuts, diplomatic activities were htensified-in particular bp Juan Pablo Pkrez .At-

Markef Context

61

fonzo from Venezuela, and iZbdallah Tariki from Saucii Arabia, usually regarded as the foundjng fathers of o13Ec, Almg with Farman Farmanyan (Iran), 5. Nessim (United Arab Emirates), Ahmed El-Sayed Omar secret ge11(Kuwait), and Moharned Salman (Arab Leag~~e), they s i g ~ ~ ae d tlcmen's agreement known as the Maadi Pact in Cairo April 1959, The Maadi pact had limited ambitions: ""lssimply outlixled certain principles agreed to by the signatories and proposed a framework for periodic consultation"' (Terzian 1985:28),After a new round of price cuts by the companies in August 1960, the prtlducjllg countries took a further step. On 9ptembes 14, OPEC was formed in Baghdad. After the 1956 Suez crisis and the 1967' Arab-Israel War, North African oil exploration was intensified. The North African production held an important advantage over that in the Persjan Gulf since the oil did not have to be trmported througfn the conflict area arwnd the Gulf and the Suez Canal. As the Libyan cJil contained less sulfur than most Gulf oil qualitiesf it- was cheaper to refine and could therefm be priced higher than the heavier crudes of the Gulf region, '"ecause of these advantages, Libya received the highest per barrel payments of any Arab gave but most obsemers still considered Libyan oil undcspriced" (Schneider 1983:140),This, together with a new radical regime, meant that the road was open for a confl-mtation with the oil companies. On September 13, 1969, a coup df4tat took place in Libya; two weeks later, Muammar alQadhafi became president. Afaer a period of political consolidation, domestically and toward Libya's neighbors, the change in oil policy took form. Ch January 20,1970, the new oil mhister, Ezzedhe Mabrouk, told the oll companies operatirtg in Libya that the government wanted negotiations about a price rise as soon as possible. Algeria and Libya issued a joint statement demandi.ng m immediate increase in the price of oil, and Iraq proclaimed its support for the Libyan action (Terzian 1985:117). The Libyan authorities' strategy was to negotiate with fie companies individually, not as a bloc. By playing the independent Occide~~tal and the multinationals agahst each other, Libya, which had become a major crude exporter, managed to raise posted prices and the government take thexof. I:.,ibyawas in a somcwhat di,fkrcnt position than most: other WEC cou11tries. As Table 2.5 shows, Libya was less dependent on the majors, as the newcomers had almost 52 percent of the Libym oil production. After the I:,,ibyanaffair, lran and Vf4nezue:lajncrcased their share of profits and a "game of leapfrog began" "ergh, 1991:580),After some h t c m d di&rences, the companies united in a common front and sougl~tto negotiate wi& OPEC as a whole, When negotiatio~~s begm on January 19,1971, the CII'EC: representatives took the position that they would negotiate only for the Gulf corntries and not for the rest sf OI~EC.M ~ i l Yergin e (1991:582) emphasized the divisions among the ctr~ccountrjes, Terzian (1985:130)

Market Context

63

emphasized the divisions among the companies. Adelman (1995:85-87) emphasized the role of the US gove nt, particularly how rep~sentatives of the State Department advocated separatre negotiations between the companies and the oil producing cotxntries. Matever the cause, the conpanligs agreed to cmdwt two sets of negotiations, one with the Gulf exporters and one with the Mediterranean exporters. It &odd be noted that the market structure had not changed; there was no scarcity caused bp undcrlyhg changes in the relationship betn.een suXlply and demand: "From early 1971 to nearly the end of 1972, priccls incrrased despik continuing slrbstantial excess supply" (Addman 7,995:93). Un Febmary 14,1971, the so-cdled Tehran agreement beheen the internationd oil companies and the OPEC members exporting through tl-te C)n Aprjl 2, 1,971, a si.milar qreernent for the Persian Gulf was sig~~ed. OIYC members exporting through the Mediterranean was signed. The agreements covered tax and price inc:reases, inflation compensation, and a Eixing of such rates for future years. The eMwb of the agrcernmts were a 21 pescent price incrcase for Saudi Arabian crude (from $1.80 to $2.18), and an increase in rr-?venueof 38.9 percent. What m s mort? important, howevcr, was the fact that the producer cotlnt.rieshad nocv ga,i,ned firmer c m t d over the price-~etthg:~ Even as late as a few months ago, the very idea of the prc>dudngcountries of the Gulf achieving an across-the-board price increase . . . [as] agreed upon in Tehran a n the 14 February vvc>uldhave seemed almost inconceivable. . . . This victory-and t~ictoryit certainly was-was mainly due to two factors: firstly the tinprecedented degree of ~ i n i t yshown b y the QPEC member eauntries; and secmdly the great skill and nerve of the three-man ministerial committee which negotiated the deal on behalf of the Gulf states.!"

The distribution oE market power in the international oil market had changed: "U'njlateral tax increases were not new Before Tripoli. and Tehran, the OPEC nations had exerted defensive market power. Their excise taxes had put a floor of tax-plus-cost under the price. But raishg taxes in concert, to raise the worldwide price floor, was indeed ne'cvff (Adelman 1995:80). Taking the Libyan affair and the Tehran-Tripoli negotiaticms togefier, there was a combination of lack of unity among the companies and a new unity among the OPEC members. One could argue that both filctors were necessary, but neitl-ter of them alone was sufficient to explain the events of 1971. A unified ~ P E Cstrategy would not have succeeded unless the cmpanieshunity had begun to crack, as in the Libyan case. 'The lack of unity among the companies would have meant nothing unless the OPEC m e ~ ~ b ehad r s gaiined some abiliq to act in co~~cord.

64

Market COBtext

The relative importmce of these factors t e ~ ~ to d splace more weight on the lack of unit). among the companjes than on the ranit). among the oil producers- If is reasonable to argue &at with competition for production concessions among the companies, the producer countries would have been able to increase taxes Irtdividually, as, in Ifact, was the case in both tibya and Venezuela. This would not have happened through large negotiations and agreements like the Tr?hran-Tripoli agreernmts, b t ~the t result would most likely have been very similac The process started in 1969 and was driven primarily by individual countries. As Terzian (1985:112) notes, in oil circles, the seventies are h o w n as the "'decade of OPEC'soffensi~~e'. . . . tn fact, the offensive began in 1969, and almost wiChout or>ticrealiz-ing what was happening." The focus on the unity of the pmducer count-ries at later stages of the negotiat.ions has to be cowered by the pomibi.litp that the Tehran-Tripoli negotiations mi@t not have commenced, at- Lal had the companies played their cards better during the Lihyan affair. Relating to the concepts guiding this chapter, m r k e t structure and market power, the conclusion is that there was no structural change in the international oil market in the late 1960s ar early 19"i"s. It was a matter of a change iut the balance of power; or, even more accuratel~a change in actors%exercise of market power. The international companies would have been able to stop the producing countries from taking cmtml had they kept together. 'They had the cards but played them porly, a case of what David Baldurin has named "the paradox of unrealized pc)werf6(Baldwin 1979:463). With the international oil companies put on the sideline, the focus turned to what objectives the OPEC c~untries would pursue, given their a c k v e d market pwes.

October 17, the Organization aE the Arab Petroleum Exporting Cowtries (OAPEC) countries12 amom~cedtheir htention to reduce productio~~ by 5 percent per month until Israel retreated from the occupied territories and the rights of the Palestinians were restored (Blair 1976:ZM). Oil supplies to the Wded States and the Netherlands were to be stopped altoe t h e r due to their outspoken support of Israel, Saudi .Ara:hia%Sheik hmani later called the embargo a legitimate political action: "W watrcl~ed America and learned haw they use one's ecmomic power to meet political objectives, We studied this carefully" "obhson 1988:95). The embargo and t k production knitations, hwewer, removed very little oil .from the market, and they were sl.rort-iivcd @lair 1971i:266-268). Figure 2.10 shows the produrtjon of a selected nurnber of countries' prodraction from September 1973 to April 1974. As the figure shows, Saudi Arabia took a largo portion of the cutback. However, the Saudi, Arabia cutbacks

FIGURE 2.11 Msnlhfy oil prices in $/barrel, IW3-2000 r;cjuRcr: OPEC BuEIct h, various issues.

Markef Context

67

for December l973 were canceled, and on December 25, the Arab oil mktisters osdercd a 10 percent increase in production for January 197'4. By January, QI~ECproduction i n c ~ a s e dagain. No shortage of oil emccged, but expectations that the future?might lead to a supply shortage drove up prices: "Nobody knew bow long the cutback would last or how much worse it would get" "delmm 1995:110). Essjnger (1982a:873) finds the embargo "a sy;mbolic gesture of limited practical importance. . . . The true impact of the embargo was psychological. " The uncertainties about the future supply of oil did, hwever, create a highly favorable elnvimnrnent fnr price incseases. Ch C)ctober 6, the o13EC countries eJfected a first price hcrease witl-zout the involvement of the companies (Terzian 19@:170), and thus finally buried the Tehran and Tripoli agreements, which had been on their deathbed for several month. The enrbargo gave the OPEC countries an opportuniZfi to hcrezlse prices without suffering from rczduction in liemand. O13EC increased the government take from $1,77 to $3.05 in October 1973, frtrther to $7.00 horn January 1974, and to $20.66 by 1975 (see Figure 2.11). The concern of the OPEC nations war; to ratchet up the floor to prevent the down-wave from cancelling the up-wave and to make the price increase permanent. Since the C ~ P F Caction was anticipated, the price effect came before the cause. Tn December, Iran auctioned some crude oiX at more than $17 per barrel, after exaggerated repods of an imminent unprecedented jump in government take, . . . It became overwhelmingly clear in 4974 how little the market situation explained the price. (Adelman 1995:118)

By 1975 thr pmcess of changing the price stmcture was completed. The average governmetnt take had hcreased from $1.77 per barrel to Q0.66, and the posted price had increased. from $3.01 per barrel to $11.25 (Adelman 1995:120).Fur&em?ore, the trYhole cmcept of posted prices was substituted cvith t%le official sdling price. Was the price by 1975 raised too hi@ Such a question Obviously depen& on the basis b r waluation. The price of 1970 was '"above long-run competitive levels set by investmetnt requirements for continued expansion"' (Adelman 1995:133). In competitive market terms, the price increases from 1971 to 1975 seem unsustainable. fn 197'4, "spare capacity in OPECwas nearly 8 mbd" "oran 1981:254), so the price hcrease was hcmpatible with full: capacily utiiizatim, RTpi the second half of 1374, the world economy went into a severe =cession. Innation was mnning at double digits, thus quickly reducing &e red value of the price increases. There was also conaiderable cmfusivn about whether prices would conthue up or Ml back, In July 1974, Saulii Arabia omced m auc.t-ionof &out 1.5 rrrbd to be held in August. Such m auction would have provided a market-based

68

Market COBtext

signal of the futnxre price t r e ~ ~'The d . auction was posponed and subsequent@ cancelled, Pmrninent scholars differs on the reason fnr this, and &us on the role of Saudi Arabia. Morris Adelmm argues that Saudi Arabia cotxld have brought the price down single handedly, and that "the public record shows the Saudis leading the price rice. The auction was designed to raise them further. Wwn the hope disappeared, the auction was canceled" "(Adehzm 1995:127-128). meadore H. Morm daims that the Sat~di Arabians were ready to "accept whatever price the conditions of surplus would produc"If (Moran 1981:255). Furthermore, he refers to journals daimirzg that Ymmi, as early as in May 1974, proposed to cut the pasted price of Arabian tight to $9 per barel, md.hvited Iran to cooperate,lbnd that Ymani in September said that "OI~EC's posted crude prices [was] $2/bbl too high.""l"e Saudi Asabim pasitio~~ brought about swere internal tension (Moran 1981:257), :Measured against possible e f k t s on demmd, the price increases seemed not to be too high. The reductim in oil co~~sumption ~ Ithe I mid-1970s was due to economic recession m d slow p w t h (Adelmm 1995:446). In the lnnger run,effects becarme substantial. Wrlat, then, were the =actions amthe gmups of the trilateral oligopaly (see section 22), the consumhg countries and the companies? In the onsunning countries t h s e price increases were seen as a symptom of resource scarciv. It fitted wefl with a pubiication from the Club of Rome called Limits to Growk-Itt published in 1972. "Its arguments were a potent element in the fear and pessjrmism about impending shortages and res o m e constraints that became so pervasive in the 1 9 7 0 shaping ~~ policies and raponses of both oil-impctrting and oil-exporting colmtries'"Vergin 1991:569). Robert Pindyck (1978:3G) refers to a CIA report claiming that "a crisis is tikely to occw in the early 1980s as world energy demand exceeds supply, resulting in shortages of en(;?rmfrapidly rising prices, and economic contraction in all of the industrialized comtries, . . . This view has had an important role in forming the rationale for the Carter administrdiaxl's energy program." However, thcrfi?was no shortage; the price increase was a result of orBc exercising market power, not a lack of available resources. As Phdyck (1978:Sl) concludes: "The kind of worldwide enerm crisis of concern to the CIA and the Carter adminidration is unlikely to occur.'' Corrcerniw tbr international oil companies, their role in d i n g the international oil market was dimhished. h1 1972 the Iranim state-owned cmpany NICE took c m t d over Iranian oil production. Tn Saudi Arabia the oil production was cmtroIIed solely by the American ccmlpanies, organized in the consortium Ararnco (Arabian h e r i c a n CM Co.).li Saudi Arabia gradually took control over Aramco in the seventies, However, the cclmpany continued to be formlly directed fmm Detroit, a d it was not until 2988 that the company dected a Saudi Arabian chairmm.l".raq nationalized rrc (the Iraq Petroleum C1ompan)i) in 1972, and Kuwait was

Markef Context

69

assig~~ed 68 percent of the interests in KK (the Kuwait Oil Company) in 1974, Qatar and Abu Dhabi followed the Kuwaiti model. The international oil companies tbus had to look for oil elsewbrrc.. With the price increases, the change from posted price to offidal selfkg price, and the nationalization of production, the focus turned to the internal OPEC bargaining. ""Nolonger could OPEC gowe ents fix a price floor by fixing per barrel taxes and letting the companies coolpete freely above the fioor. They now had to fix prices in concert and to trust each other not to undermine tbose prices by trying to hard for additional sales. It was an endless exhausting struggle" "delmm 1995:143),

Terzian (1985:235) describes 197'5 as a turning point in the history of OI~EC-itmarked ""theend of the '"decade of the offensiveknd announced the bcghning of the organization's sul?s~quentdecline." At a summit meeting in Algiers, some OPEC members attempted to use the osgmization as a "qpcrhead of the Third World,'" something that wodd involve not or~lyeconomic but also political and ideological cooperation. These attempts &led, and OPEC remained m '"economic" oqganizatim, In the , concerning years that followed, disccnd in OPEC i w ~ a s e d particularly the questim of price increases. Saudi Arabia was the cent-rd oppnent of higher prices. This was explained as consideratiun for the industrialized countries' economies and OPEC'sdctpencjence m the growth of these. Other, more curious reasons we= dso gi\"c11. Rrzian (1985:244) cites a Dcr Spiegel interview in which Uarnanj. said that by limiting the price rise, Saudi Arabia was trykg to " p ~ v e n the t communists seizing power in France and Italy." In 1977, Iran joined Saudi Arabia in advocating a price freeze. Saudi Arabia and :Iran together accounted for a full 48 pexent of OPEC% total production. W e n both opposed price hcreases, there was little the other countsies could d.o.FXowever, sudden political events we= to play havoc with the oil market: The Shah of Iran was overthrom, and the fundamentalist Islamic rule under Ayatollah RuhoXlah momeixli commenced. Despite the fact that the other OPEC countries easily compensated for the disappearance of kanian oil, d m a n d increased as fie buyers wished to s e e m their access to crude oii in case of a likely future dcrnand surplus. As Figure 2.12 shows, a steep incwase in spot prices was followed by increased official prices. From December 1978 to October 1974 the spot priee raised from $13.8 per barrcll to $38.35. It was not until 1980 that the market begun to react to the countermeasures of the other OIZC countries and the s u ~ p l ysurplus was revealed. Rut here again, political eve~ntsunderxnined stahiliza.t.ic,n.011September 22, the Iran-lraq War broke out. However, this time there was a surplus in

Market COBtext

70

companies m d governments panicked m d bought oil to build up stocks at a very high price. We in OPEC took this false demand for real.. And. we made a mistake: we raised our prices hrther."'IT In March f 982, Che WEC members agreed to set their total production output at 18 mhd, and distf-ibuted,p d u c t i o n quotas among the mernbers. Saudi Arabia was not assigned a quota but was to act as a swing producer, and trhus vary production to balance the market at the $34/barrel price. Saudi Arabia performed the role of swing producer irt the folhtwing years by cutting back production, 'This codd not help the price from siiding, &eit slowly, as can be seen in Figure 2.11. The internal o13Ec bargaining aspects of this are discussed in detail irt Chapter 5. The importance in this chapter is how these actions by the amr memtters helped OPEC mainhin its control of the international oil market-a control the organization had gained nut by its own m d t s , but through favorable struchtral changes, and politicai events like revolu.tion and war. The Saudi Arlltoian role as swing producer allowed OPEC to maintain s m e controli of the price level, but from 198S1915-3:or)~c'sdominance and controt over pricr-setting crumbled (see Chapter 5). This was due to two impctrtmt developments. fn the first place, the growth i,n demand for oil came to a halt, due partly to higher prictrs, partly to a gmeral reduction in economic growth that occurred prior to OPEC'sprice increases, m d partly to nebv enerw sources such as natural gas and nuclear energy. Second, competition between oil producers irtcwased, in the same way as it had twenty years earlier between the companies. 'The new pmfitatriliq of oil production m d c fields in non-OPECcorntries profitable, and new producers took a larger share of the markfjt, Both factors came as a c m s e qumce of the market behavim of the 013~~ countries durhg the seventies. Their behavior during the pcrjod 7,971-1981 had lasting effects on the structure of the internaticmal oil market. Tbe iollowing sectim tries to capture h a t effects the price i n c ~ a s e scJf the scrventies had rcsgardint; both the underlying structural aspects (see section 2.5) and the behavior of the other market actors in,the tf-iiateraloligoy-oly the consuming corntries, and t-he international oil companies (see section 2.6). 2.5 Structural EffecZs of the Price Increases

As pointed out in secticm 2.3, energy consumption, and in particular oil consumption, rose sharply after the Second World War. Through the fifties the need for gasoline increased due to rapidly expanding private mcrtcrring and the growing transport sector in the industrialized cowtries. 'The economic growth in Che industridized countries has been an energy-dependent growth. Each FRcrease of one unit in Gmss Umestic

Markef Context

71

FIGURE 2.1 2, Spot and official prices, July 1978--July 1983 ($/barrel) OPEC BziEleti~z,various issues.

SOURCE:

the market, possibly as much as 2 million barrels per day (mbd). ban and Iraq took s o m 4 mbd frnm the mrket, while the remaining 2 rnbd were easily supplied by the other OPEC countries, which now had suficient spare productim capacity. Saudi k a b i a rclfused to cut its production until demand for orEc oil feU dramatically, which was partly heled by the oil companies' stock drawdown in the a u k of 1981. This caused Saudi Arabia to perfom a vc,lteface md, quite co~~trary to its earlier resistmce to production regujatiurr, the kingdom now actkely supported. it. Xn this way, Saudi Arabia, as the large" producer in Cr12EC, began to ~ g r t l a t epmduction to create market evilibriurn. At &is point, the OPEC countries had to prove their ability to function as a cartel and reduce production when demand decreased. Again, we can raise the w s t i o n about w h e ~ e the r price wap; raised too high. With the bemfit of hindsighl, the 1979-1980 case seems easier to conclude than the 197S1974 case In the words of Vamani himself, the price increases were a mistake: "We believe that we made a rrtistake in to the real sup197911980by raising our prices wi.lhout paying atte~~tion ply-demand situation. There was only a false demand, because the oil

74

Market COBtext

product (cnr) has required a certain n m b e r of units' increase in encrf~y consumption; this rath is the so-called e n q y intensity (see Figure 2.13). For ail, this ratio increased until 197'3?but has since then fallen, QECD economic growth is less oil-dependent:today than ever before. From f 979 to 1995 the ecmomies of the European Union (EU), Uf-Lited States, and Japm gww by %,M, and 55 percent, respectivety. The oil demand klI marginally in E14 and the United States, and grcw only marginally in Japan during the same period (see Figure 2.14). 'l'he explanations h r &is mduction in energy intensity are several. In the first place most industrialized countries today develop in the direction of a postindustrial society. The industrial sector" share of the cmS shrinks. This leads to a ~ d u c e need d fnr energy consumption in order to increase GDP. %cond, the price rises of the seventies made elncrgy conservation more profitable m d other energy sources more competitive. Since the oil-price falt in 1986 (see Figure 2.11), oil demand has hcreased only modestly; In fact, in 1993, total world oil demand was at the same level as in 1979, while the price was $17/barreI, in 1993 comparcd with $4C)/barrel in 1979. The price fall did not stimulate increased demand during the period f9K6-19"33,Subseqrxently, the oil producers have not been able to sustain their income level by compensating for the fall in price with increased outyut. Thjs lack of demand elasticity can, to a great extent, be asesibed to the consumirtg countf-ieshse of a rather effectbe tooX, nmely, the taxing of energy use. Fig- 2.15 illustrates how the pmduct price presented to EU and aEcn consmers has increased, while the crude price has decreasccf, The main explanation fur this is a substantial increase in the tax share of the oi product prices shown by tbr black area in the figure. When the reduced a u d e oil price is compensated for by increased taxes, the expected demand hcrcases due to the lowered price will not wcur, 'l'he consuming corantries' gove ents acbally have partial contml over a prominent kature of lfie present: market situation-the lack of dennand elasticity.

As outlined above, the ail market of the fifties and sixties was dominated by the Seven Sisters (see sedion 2,3). From 2971 until 1985, OPEC was the dominating actor in the oil market (see section 2.4). However, a nursrber of new oit pmducers emerged in the course of the seventies and eighties. As described in section 2.4, t h i s devdopmcnt was partly due to O~EC'S own pl-ice policy which created a considerable profit margin, and partly to technologicai advances that rendered feasible oil productjon in new areas. This underlnined OPEC'scontrol over the market development.

Markef Context

E&r@p@an Union $/~~KTC"/

l 20

$&arr#i

QECD

80

FIGUM 2.15 Trends in taxation of petmleurn products Ghanern (1998:5).

SOURCE:

Control of market devebgment thus required coordinated behavicrr by a l q e r nu~nberof actors than before. Siince the beginnil7g oE Ihe eighties, o r ~ has c repeatedly requested non-WECproducers' cooperation m production regdation in order tc:,strengthen the price. These aspects are discussed further inChaptw 8. The distribution of market shares among the producers is presented in Figure 2.16. The obvious change since 1973 is that while OPEC and nonomc producers shared the market equally them, $king the OPEC cowltries a substantial market control, OPEC h 25393 controlled only about 40 percmt of the market. Xn the period in between, its market p w e r was substantially lower; h the mid-eighties, the OPEC market share was approxbateb 30 percent. The trend is that the production of tbr traditional ~ W P E C producers, like that in Alaska and the North %a, has been sustained at a high level exceeding the oil industry" forecasts. "Mearly every non-OPECsupp:ly forecast tends to be pessimistic, with a near-term peak'" (Lynch 1995:l). In addition, new producing countries are presenting themselves in the international oil market. Several of these are corntries that were isolated from the world oil market due to the Cold War or, in some instmces, civil wars, The international political chmges created ecmornic opportu~~ities ior these countries and for the international coqaniczs willing to invest in these areas. Prominmt examples are h g o l a , Azerbaijan, Cambodia, Cuba (not open to all companies), Laos, Noznmbique, and Vieh7arn. The

Markef Context

77

important questim regardjng such new oil provinces is hukv to h a n c e investments in exploration and development. Here lies the role of the international oil companies, which for a long time have been so-called crude shorl, rneanifig they have refinhg capacity that- exceeds the crude oil they control. Their hancial positions, however, have been remarkably sound, especially those of Exxon and Royal &tch/Shell.lg The reaso1-r the interna.t.iona1companies sought oil in the North Sea and Alaska in the seventies was a need for alkrnathes to the Middle East, as the nationalizatim there made the companies' operation impossible. Alaska and the North Sea were high-cost alternatives to the Middle East. The production costs in the m w oil pmvinccs tie beheen the low-cost Middle East m d the high-cost areas such as Alaska and the :!or& %a. In addition, several of the new oil-producing countries wlcome private international oil companies and offer them better tax conditions than the established producer countries. ?"he newcomers among the produchg countries migM subsequently force the establ.ished producers to ease the cmpanies' conditions in traditional produring areas in order to facilitate new inveshnents by the &ernational oil companks.

SZrlictzlvnl Changes RedisfsibtttL.ivfurkt Power The leveling oft: of thc growth in oil consumption, togelher with the increase in non-oPEc production, totally changed the situation facing OPEC, The high prices achreved during the seventies ixnmediately came under pressure in the eighties. The favorable structural co~~ditions for oil-prodllcer cooperation at the beginning of the seventies had, turned into stmcturaliy unfavorable coditions by the begiming of the eig:htics. Chapters 5 to 8 discuss how OPEC met this chdlenge, and the eft-ects of this structural cbange m the cooperation between the oil producers. 'T"hese changes also affected the relationship betwen the groups of actors in the trilaterat oligopoty, the producers, consumers, and intemational companies. The reduced market power of the oil producers has improved Ehe position of the companies and the consuming countries. As poin.t.ed out in section 2.2' the rclductim of one group of actors' market power either makes the market m r e competitive or i~~crcases other actorsbarket power. 2,fi

Pifew Market-Power Relations

The factors discussed above are the so-called fundamel-rtals, factors that direct@cause changes in supply and demand. However, other important changes have taken place in the intmatimai oil market during the last decdes. 'This section focuses on changes in the vertical market stmcture

78

Market COBtext

(see section 2.1). 'The actors' behavior h the vertical structure is equally important for the changes of market power among the market actors,

New Tradirzg Mechanisms The old integrated oil-market strucbre atlowed multinational oil companies to balance world oil delnand and supply throt~gha vertically &egrated system, OI~EC" nationnalization of oil-industry assets Fntroduced new market mechanisms, primarily thmugh the system oE official selling prices and long-term contracts with a variety of oil companies, However, throughout the seventies and early eighties, the decliine in dtrmand and the appearance of new oil producers outside the cartel made it: very difficult for orEc to etnforce its offieid prices..As a cmseyuence, long-term contracts were gradually phased out. The conthuous mismatch betwen supply and demand created fertile ground for the developmnt. of a spot market (a market for single crude cargoes), cvhich is characterized by short-term contsacts, a high rate of turnover, and sensitivity to outside events. It is not unusual for a single cargo to be sold dozens of times belore it: fillally arrives at. the refinery. This m d e r s it difficult to establish to what extmt the transactions should be designated spot or contract sales, since a company may purchase oil from a producer according to a contract fomtula, only to scll s m e of this oil h the spot market afterward." The i n t e r n a t i d oil trade srabsequentiy became highly disintegrated. Under increasingiy competitiwe pressure, producers adopted prichg formulas linked to day-to-day fluct-Llations in spot prices.20 Under these market condjtims individual producers, both OPEC and nm-OI~EC, had no guarantee concerning the long-term loyalty of their customers- 'The daily market actors, the traders, obtained an increasingly important position in the market, From the time QI~EC'S oil irmdustry was naGonalized in the first half of the seventies until 1979, long-term co~~tracts were the domhmt form of crude sales. h 1979 the contract market constihted 85 percent of turnover. Sales on the spot m a r k t made up the rc-rmaining IS perc.ent (Estrada f 988:9). In the eighties, nonapEc productiol~incleascd, and new pmducers outside OITC l h k d their contract prices to the spot market price. The increased produc"cion, and the leveling off rtf demand, pu&hed the spot price downward (see section 2.5). This created a gap between thc spot price and the OPEC official price, a development that prompted m r e and more buyers tcr make their purchases on a spot basis (Mosswar-Kalzmmi f 986:5). This developme~~t also gave the spot-related contracts a contpetitive advantage. fn 1984,70 percent of traded cmde was sold either directly or indirectly through the spot market." l M l e the spot price in 1979 and 1980 had pressed the official price upward, it now contributed to the de-

Markef Context

79

TABLE 2.6 Taxonomy of Trading TVP~ Term Standard term EqttiZJr Marketrelated Spot Single syc~t

Daisy chain

Tertiary

Duratiorz

Price

Multiple cargoes

Mctnas or years

Govermentdetemined: Official Sales Price (QSP) Special arrmgernents Ad justable

Wet

Set by prevailing market conditions, below or above OS),

1f

It

11

Single cargo

11

Forwards

Multiple cargoes Multiple cargoes Single cargo

FuSures

Less reletrant

Netbacks

Derivatives Sways

13hysiml Conditiorz

blume

11

Years Months

One-time basis

Multiple trmsaetiions of same cargo Weeh or months 16

One-time basis

Less relevmt

11

PI

Discomted below OSX" ($3-2/barrel) Spot product pricerefated Set by expectations of future market crtnditic>ns It

Set to reduce price risks

" 1f

$1

"

$I

Paper

1f

cline of officid prices. The companies that signed long-term cllntrarts i,n the eady eigtnties also suffered., as they bad to pay the detemined price, while there was pknty of oil available in the spot market at substantiaify lawer prices. The spot market was soan supplemented by the ""tertiary market," so called by Mossavar-Rafnman.1(lf386:I),who defines it as a "hodgcpodge of sowwhat opaque transactions including barter, countertrade, processhg arrmgeme~~ts, cmde-crude or crude-product packaghg, and outright discounts" "(see Table 2.6). This occurred because the OPEC countries, due to the surplus on the market, did not wish to cwtribute Officially to a price collapse, but were at the same time eager to ellsure sales

80

Market COBtext

of their own crude. By the mid-eighties the tertiary market was being used by severd OPEC countries as a method of giving buyers concealed discounts it7 various ways. Mosswar-R m i (1986:12) calculated &is to constitut-cz s m e 5Cf to 55 percent of all orec oil sold, while the contracts covered some 30 to 35 percent, m d the traditional spot market only 10 to 35 percent (all 1985 figures). The actual price in this market is difficult to determhe, as all agreemel-rts between seller and buyer are veiled h secrecy. However, it can be assumed that the price has been sornewhert.between the spot price and the official price. This reflects the buyersf willingness to pay a little more b r ensuring deliverias rather than rely on spot sales alune, thoutgh not as m c h as the official price w d d indicate: With over half of all OPEC crude oiXs traded on a discounted basis, the pressure on the remaining crude oils grew, Even though the formal m3nc price structure tvas prevented from el-rllapsing to tertiary market Levels, through the end of 1985, official prices became increasingiy meaningless as the share of tertiary transactions continued to gain at the expense of term and spot sales. (Mcwsavar-Rahmani 19%:15)

This caused Saudi Arabia, which previously had sh-ongly opposed discount arrangements,22 to ilntroduce so-called nethack pricing.2WNback agremel-rtsare "The sale of crude oil with the buyer paying the producer a price which is &pendent on the proceeds obtahed from the sale of refilled productgf (Jenkins 3986:423). This method of price-fixing e n s w s the refineries a profit on. oil purchases, as the price is not decided before the refinery has sold the product and is fixed in such a wa)i that the c d e price equds the pmeiuct price mims refhing, trmspcrrt, insurancel loss, and hteresi: casts in processing of the oil (Mossavar-Rhmani 1986:19). The drawback of the netback pricixzg system was its cornpIexity and the transaction costs connected with negotiating individual agreements to match the circumstmces of different refiners (Roeber 1994:262). A simultaneous development diminished the need for netback agreements. "Through communication techology and the frequent publishing of oil-price data (e.g., in 121aft's Ililgmm News,and through Reuters), the spot price transparency increased. This caused the value of the crude oil in the product markets to be r m c t e d in fie spat price. The netback pricing was st~bsequentlyreplaced by market-related pricing. In the marketrelated. pricing system, the other features of the contract between the selter and buyer can vary, at the same time as the price is set according to the difterctnt spot crudes. 'The contract can be a long-term contract or single cargo deal, The important feature of spot-related, pricing is that the frequency of price changes is increased compared with the term market- Spot prices are changed almost constantly. The pr&li,em of a

Markef Context

81

mismatch between the contracted price and the market price is thus more or less resolved. During the latter half oE the eighties and into the nineties, one of the mast prominetnt features of oif trading has been the increased activity in the paper market or semi-paper market. The forward market is actually a market for spot transactions in MIhich oil is traded for delivzry at a future. time. 250 the farward market is a serni-paper market, as it is actual physical crude that is tracled. A more genuirte paper markt-tt is the so-called htures market.2"utures contracts are at the outset desiped for fh~mcial purposes. Thercl was no regdar futurcs trading inthe oil market until the seventies, as the major international oil companies bad no interest in such a lievelopmmt. The increase in futures trading in the early nfneties has been tremendous: "The WTI [Western Texas htermediate] crude cmtract was launched by NYMEX [New Vork Mercantile Exchange] in 1983 and is now [l9931 trading at a level eyuivalent to 100 times wrt prtlduction. . . . The Rrent contract on the I P E [ ~is traded at 40 times Brctnt pmduction during the first half of 1993" ((Roerber 19(33:48). The futures market prowides ""those in the physical market with the opportunity to offset the infnerent risks associakd with their business" (Banley 1989:27), The future market might also reduce price uncertaintJi and thus the possibility of arbitrage (Sykuta 1994:4). EledgiIlg is the pro"incmt hstrument for risk reductio~nh the futures market. h oil producer can buy himself insurmce against loss from a fall in the oil price by selling an amount of oil on the fixtures market, which can be rc.gad"d as a promise to szlpply oil on a future clate-When the price for physical oil falls, so cloes the futu.re price. By buykg back the option the producer gains a profit, which could counter his (potentid) tosses in the physical market. Sykut.a (1.994) gives an additiotnal rationale for the hturcs market by regarding it as an alternative form of cmtractilrg in the oil market: 'The futures ma+ket provides a nttxible form of contracting for the commcrdiv which can. be used in tandetn with, or as a s~~bstitute for, forward m d spot contracting to add flexibility to the firm" contracting and resource alfocatian decisions""(Sykuta 1994:s). This role of the futures market reduces buyers' costs of asszlrj,ng access to the commodit-y and likewise the producers' arress to market outIets.25 There are several paper barrels for each wet barrel, and with the price of even l o n g - t e r ~contracts ~ related to the price developments of Brctnt and wrr, the relatively few traders operatjrrg in these markets have a disproportionai influence on the day-to-day price developments. None of the OPEC countries, and only a few of the companies, are actually involved, in futures trading. Thus, they are d.epri:ved of infnrmation and possibilities to understmd the formation of prices. This, of course, makes it more complicated for omc and the oil companies to inflt~emcethe oil-

82

Market COBtext

price development. By the end of the eighties the short-term development of the oil price was left in the hands of the oil-market traders. Xf pesthe market, one simistic about the OI'EC countries-capacity to codd argue that the Iong-term oil price is nothing other than a scries of prices of short-term oil deals, wet- or paper-based. Xt is more fair to say, however, that the new trading mechanisms have introduced a new layer that the oil producers have to penctmte in order to gain contml over the oil price, Previously they could make secret deals with consume=, and discriminate a m g them. Now they have to p h y the game of convincing the short-term market dealers that they have reason to believe prices will g0 UP in the future, and they should therefore buy morc oil in order hedge agab~stthe price increase (Claes 1999a). However, this new oil-trading bushess has not made the relationship between oil producers and the companies unimportant altogether: In fact, the relationship between the actors mowing the physical oil also has undergone -mental changes durj,ng the last decades, changing the actors+position in the vertical stmcturc, This is the theme of the following subsections.

In ge~zeral,downstream investments have bee11 considered a natural m d important step in the industriali.zation and economic diversification of an oil-producing cowtry. For manyl downstream moves were ~ g a r d e das the way out of underdevelopnnent and dependence on Western influence: Domestic refineries have been regarded as an instrument to increase the value added of the oiX prcdduced. As a complement or as an alternative, some oil exporting countries have invested in refineries and distribution neworks in the consuming cormtries. The choice of downstream investments overseas is based not onXy on technicat and finandal considerations, but first and faremost on commercial strategy. Thus domestic refineries and international dotvnstream investments must be regarded as strategic choices both at a government and a company level. . . . The ability of an oil prc3ducing country to integrate downstream depends on its financial strength, managerial independence from pc~liticalauthorities, downstwam experience in domestic markets and qualified personnet with international experience. These factors seem to be more important for the downstream success than the size of the reserves, the uc)tumes of crude oil exported or the prices achieved. , CXaes 1989:27,29)26 (Bergesen, B j ~ r kand

Sifice the early eighties-den the market began to tighten-of the mgor oil-exporting countries have made importmt downstream invest-

Markef Context

83

rnents in Western Europe m d the United States through a ~rarietyof participation ag~ements.The purpose has been to g a b partial or total control ower companies that can refine and distribute part of their crude oil exports in the main cons~~lning countries"'This should h itself secure outlets for sales and thereby more stable revenues Ina volatile market, These exporters have also bought t d e r s , harhor and storage facilities, and petrocherni;cal plants incmsmming countries, In 1,987,orEc exporters accounted for 8.1 percent of the Western European market for oil products (Bergesen, Rjsrk, and Claes 4989:28). Among the largest downstream inbrestsrs are Kuwait, Venezuela, Saudi Arabia, Z,ibya, and Norway; Kuwait and Venezuela have refinhg capacity (domestic and ioreign) coverirrg 90 to 100 percent of tbeir production capacity; the carrespmding figure for Saudi Arabia is about 50 percent pinon 1991:264). Table 2.7 highfights s o m important foreip downstrcarn acqujsi."tons of oil-exporting countries durinl; the eighties. The pace of downstream htegration has slowed during the nineties ( H a m 1994:186). This might be explained by the weakened financial totaf external debt increased position of several oil producers. QI~EC's from $236.8 billion in 1986 to $343."3illion in 1992 (Gacheno~tr 1992:137717),813EC countries still would like to secure more outlets, but their financial troubles prohibit further investments in the refining and rnarkethg sectors.

?irming to the study of the cmpanies+ositim in the upstrem market, two f e a k r ~ are s particularly importmt: the ikancial sihtation of different companies, m d their crude position. The decline in refhirrg and distribution margins in the late seventies and early eighties led to a wave cJf closures, modtrmizations, and takwvers, especially h3 the Unikd Staks. During this period of gainful restructuring, hdependtmts and majms were competing for a litnited volume of known rr-?ser\iesthat more than tripled the median transaction value for reserves in-place h3 the Unikd States (Smi.L-het al, 1986).Much of this wave of reserve purchases and oll compan)i mesers and acquisitions was financed through debt. 'I'hus, it came as no surprise. that as soon as spot prices begm to show the first isrdicatio~~s of declke, companies with iinancial diffjculticstried to resell the s m e =serves bought some years before. f i m e m u s properties changed hmds ~peatedtyin a short period. In a sjmilar way, refiners who had upgraded their facilities at :high costs in the late seventies becarne caught in m economic sqtrreeze in the mid-eighties. Refinery upgrahs were based on the belief that the wide price spread between hcavy and li&t crudes w o d d conlinuc. It actudy

E-.

ac,

2

'.DC%

cam

E-.

ac,

Markef Context

85

narrnwed :mm $8.5/barrel in I982 to $wbarrel in 19%. Costly revamping to feed heavy cmde proved to be a bad investment in a business already beset by marginal economics. In several cases banks helped companies resist selling upstream and downstream properties by restnlcturing loans and advancing some loan fnrgiveness, but the price decline in 1986 forced many companfes to divest in order to fulfil1 debt obligations. This has reduced the nurnber of oil cmpanies and improved the position of those that were able to react quiclcly to changes in crude supply and product demand. U p s t ~ a m , there are similar problelns. Between 1%5 and 1987 the sevezn largest majors replaced only 40 percent of us reserves and.59 percent outside the United States through discoveries, extensions, and improved recovery. M e n revisiolns of oil reserves and purchases are taken into account, the m;Pjorsf 'Yepla~ement'~ was still 11 percent short compased with prodracticm. The majors-primarily hen, Shell, and n12-purchased mserves from slnaller companies that were either cutting back their oil activity or dropping out of the industry completdy, but the majors are still crude short, even if they are better off in this rczgarci than some of their smaller competitors. ALso, the late nineties meant rc?structurin&among the large hternational oil companies, In 1998 British Petroleum agreed to buy Amoco for $28 billion, and a $77'billion merger of Exxon and Mobil was mnou~nced,The companies seelned to be preparhg for a more competitive envirmmnt, whercr increased size is perccjved as necessary in order to take higher risks. 7'heoil price decline that started in 1997 and sent fie oil price down to $10 per barrel during 199963increased the pressure in the oil industry for cutting costs," h o t h e r problem facing the international oil cmpanies im the hte nineties was their l w rtth;lm on invested capital compared with that of other industries such as information techology: "oil shares have performed worse than any other group of industrial shares ower the past decaeie."2"us, regarding both fimncial and physical reserves, the nineties wsulted in a concentratioln among the intemational companies, putting the survivors in a stronger position in the international oil market than would oitherwise have been the case, This implies even stronger competition h the years to c m e , which gives reason to believe that the cmcentration of =sources within few oil companies will canthue. Far the OPEC countries the key problem is the ability; or rather lack of ability#to finance the necessary investments in their existing production facili-ties. Most OPEC countries are currently p'ducing close to capaciy (except Iraq, due to UN sandions). h y increase in production capacity will imply hvestmmts. As Figure 2.17 clearly shows, the financial reserves of the OPEC countries are no longer vvbat they we^ in the heyday of high oil prices. "Virtuatly all producing countries could use more li-

Markef Context

87

nancing, mare techology and more orgmization '"(Finon 1991:263). The role of the international companies in O I ~ Kproduction is on the increase. "Recently [by 19951 . . . mmy or2Eccorntries have started to revise their laws and regulations and become more open, to production-sharing agreements with for~ignfirms. Some Merrtbers, such as Algeria, knezuela, Iran and Iraq hawe now foliowed fndonesia, Nigeria, the UAE, Libya and Qatar, in opening up to f"c>rei&n equity sharing. Saudi k a b i a and Kuwait may foilow suit at a later stage" (Xsmail 1935:18),

This section has focused on the relationship between different market actors along the vertical djmension of the market st.ructure,The lactors discussed in the previous sections give rise to the idea of a ""New Deal" in the international oil market, whereby previous enemies cooperate and exchange control over different- market segments, creating a nekv, integrated, oil-market order. This new international petroleum order is based on a emergence of interest. 0x1the one hand, the large oil exporters search for secure outlets for lheir crude oil in order to protect trhmselves against future market volatility*Downstream integration-in one way or another-is an expression of risk-aversion on their part, which is understanctablt. given their experience in the oil market dLtring the eighties. This fits nicely into the predicaments of the oil industry at present. Companies with fbancial difficulties need new inwestments-for example, in their rdineries-while crude-short. companies are interested in arrmgements improving their access to oil reserves, In that way they will also hedge against future uncertainty. In this case the risk is the mirror i m q e of Che one the producers face. For the majors, such joint ventures provide protection against futurcl scarcity#which was the cmpanieshi#tmare of the seventies, 'This will open the way fctr the companfesf return to the updrean? masket, and subsequmtly part:ly =verse the structural change of the seventks, F i g m 2.18 illustrates this point, Regarding the future outlook for the structure of the international oil industry, Luciani (1995:49) expects producing cl-runtries to establish new rules under which foreign compmies will be allow-ed direct access to oil reserves under cmditions that will guarantee the govemxnrtnt%political control. I also expect the oil companies of the prc~ducingcountries tcr evolve gradually into integrated intei-nationaloil companies, with down~treamoperatic~nsincreasingly in line with their upstream assets, and resemes located abroad as well as in their country of origin. The process of vertical reintegration will not eliminate the spot and shod-term futures markets-.. . . As the process develops, the distinction be-

Market COBtext KF&qlE-RVFS PROIIVC'TION REFXUIbG TRANSPORT DISrR1BUrIC)?* I Y 5 t l s and 19htts

IY7t)r and IYSUs

t 9Yflc

FIGUM 2.18 Changes in control over the vertical productic~nchain

tween the various types of companies-national oil companies of the producing or impc~rtingcl-runtries international integrated nil el-rmpmies,small nonintegrated companies etc.--will fade out.

While the fifties and sixties were dominated entirc?lyby the major internati.onal oil companies, the upstrem section of the rnarket in the seventies was taken over by the main ojl-groduchg countries, These were first of all the QI~ECmernbers, but other producing countries also served a l q e share of €he upstream sector for national oil companies, as in Mexico (PEMEX), the United Kingdom (w), and.Norway (StatoiX), The mavl changes related to reserve control and producticm, The nccv structrxre that emrged in the nineties has given large exporters control over sone downstream outlets, and the mqor oif compa" nies have gained access and some control upstream (see Figure 2.39). Such a m a r k t structure wiil give both groups more long-term securjty. Commercial freedom to maneuver will be more limited, but it is unlikely that any ccmlpany would ccmlmit 1110 percent of its sales or purchases to long-term relationships,.Wthin a lmg-term security arrangement, a margin of Rexibifity will a h a y s be maintained, by continued trading on the spot and fubres markets.

The relationship between actors%behavior md the market structure is dynamic. What actors (producers or others) do at t, might change aspects of the market structure in a way that constrains the actors' behavior at t,. Furthermre?, the potential for successful producer cooperatio~z(i.e., a sit-

90

Market COBtext

uation where the net hcome of oil producers is raised due to collective action) is also constrained by the behavior of consumers and.companies. Finally prducers present fn the market at t, may behwe in a way that attracts new producers to the market at t,. All these aspects have been presented above as important in the development of the international oil market since the Secclnd WorZd War. 'l"llcr prirne focus of the mst of this book is Che period since 1971 and in particular the decade hln 1985 to 1995. This concluding section spells out some implications for the producer ccloperation, suggesting a l e s favorable stmch-lralenviro successfu1 producer cooperation h this period than inprevious decades. .As pointed out in section 1.5, both the theoretical contribution of I-lotelling a d the notion of s c m i v can be argued to be inadequate descriptions of structwal forces in the international oil nnarket. Consequently, these factors are of Lesser importance also for the oil-producer cooperatim. As will be shown later in this s t d y , m e countries have tried to use their large pmvm =serves as a bargaining card in internal orEc negotiations regarding production quotas. These attempts have usually been futile. However, the p d u c t i o n capacity is tc:,some extent based on the actors' reserves, and the production capacity is an important factor in the oil-producer bargaining, as it determines the possibiliw fnr an actor tc:,gain, or at least =duce losses, in a price war. This argument is obvious-only producers with spare capacity sland to g a h from a price war, as will be examixled morc deeply in Chapter 7. Also, the lwel of production is an important element in the discussion of colilective action in Chapter 7. Whether an actor increases its production by some percentage is assu~ledto have little effect on the politics oi that actcrr, but large and srnail oil producers arc. assumed to pwsue difkrent policies rcgardbg oil-producer cooperdion. Another important aspect influencing the oil-producer cooperation is downstream ifiegratirm. Oil pmducers with large d o m s t ~ a m assets are less affected by low crude oil prices. In, fact, to some extent the interests of these actors are sirnilar to those of consumer countries, since low input prices in refineries increase the profitabitity of these assets. The most prominent example is Kuwait, which during the nineties has increased its foreip d o w n s t ~ a m assets to a level equal to its income from cmde oil sales. 'Thfs enabtes the country to make mofiey in situations of both high and low oil prices. The changes in market power among the producers, consurners, and companievresented in sections 2.3 to 2.6 obviously constrain the patentiat for successfuX,producer cooperation..Before the newomen; broke the Seven SistersYominmce in the hternational oil market, there was little room fnr oil-producer cooperation. As will be untferlined in Chapter 3, the oil producem are not a pctlilicdy homogel~eousgroup. The only fac-

Markef Context

91

tors that hold these cowltries together are their common interests related to the oil market and, h fact, their opposing interests against the other group"; the internaticmal companies, and the consumer countries. The Middle Eastern countries needed the split betwcer.1 the companies (see section 2.4) in order to gain, the upper hand h the controX of the market. When this was achieved, the cJil-prcrducing countries experienced a decade of prosperity; from the beginning of the seventies to the early eighties. Then they had to face the rcvcnge of the markt forces m d the consumer countries as described in section 2.5. The demand slackened, and imrporting countries began taxing oil consumption, making the emduser insensitive to chang@sin crude oil prices. The price decrcase during the eighties has subsepentIy not increased demand, as consumers have been presented wiiSh sustaivled high product prices (see Figure 2.15). The aspects of rejntegration in the international oil market (presented in section 2.6) are a way for both crude-short c m p a " i e m d crude-lmg oilproduchg countries to be able to live wi& both high and low oil prices. For tbe oil-producer cooperation this also establishes an internal split between the oil producers with and withut downstrem assets. The djscussion ol t.ihese changes jn market pocver in this chapter has focused on hovv the ability of one of these groups to influence price-settingl and to capture the monopoly profit of ail production, has been determined bp structural changes and the behavior of the other gmups of actors. The empirical analysis has identified a dynamic relationship between the different groupmf actors and between actors3ehavior and structural effects. The rest of thjs book is dedicated to the study of the cooperathe behavior of one of these groups, the oil producers, Notes 1. The designation "the Seven Sisters" "-as first used by the Italian oil man Enrico Mattei, and was later used as the title RE Anthony Sampson" book about the seven largest oil companies (Sampsoyn 19?5:11). This group included Exxon, MObil, Standard Oil af California, Texaco, Gulf (all American), BP (British Petroleum; 51 percent of the shares were fc~rmerlyheld by the British go>vernment),Royal P>utrh/Shell(C;Oper-cent Dutch and 40 percent British), CFP (Compapie Franeaiw des Plittroles) is sometimes included in this group, despite representing a minimal share of world prc~duction(approximately 1.2 percent in 1950) (Schneicrter 1983:39). 2. W L. Newton's paper given at a seminar on the international petroleum industry at the School of Oriental and African Studies, University of London, 1957; cited in Penrc>se(1968:82, h. 2). 3, It should be noted that the explanations for the compmies" involvement and behavior in the Gulf region during the fiAies and sixties are mixed and not always strictly commercial.

92

Market COBtext

4. The San Rerno agreement was an agreement primarily between France and the United Kngdom on the mandates in the Middle East. France was awarded the mandate for Syria and Lebanon, and the United Kingdom received the mandate for Palestine and Mesopotamia (latel; Iraq), 5, Prices came u n d a pressure from discoveries in the United Shtes, in East Texas, and subsequently outside the international petmleum cartel" ccctntrr~l, 6- This was the case when US authorities dissolved the Standard Oil Cornpany in 1911, when the FTC investigated the cartel of international oil companies in 1952, and when a US S n a t e committee conducted hearings on international oil in 1974. 7. The figures refer to markets outside the United States and the Second World, and include CFP (Schnelder 1"383:40). 8. The members and organizational structure of OPEC are presented in Chapter 4. 9. Terzian (1985:143) refers to an anonymaus source in an American oil company" statement to Plnttts Oilgram Mews, February 3, 1971: ""they (the OPEC countries) could ask for $5-$6 and there is really nothing we could do about it." 10. Middle Enst Economic Surz~ey~ February 19, 1971:4, cited in Adelman f19C)Ei:E;Q),emphasis added by Adelman. The ministerial committee was made up of representatives famshid Arnuzegar (Iran), Sadoon Hammadi (Iraq), and Ahmed Zaki h m a n i (Saudi Arabia). By a resolution at the TwenQ-first OPEC Confewnce in December 1970, it was established to negotiate on behalf of Abu Bhabi, Iran, Iraq, Kuwait, Qatar, and Saudi Arabia. 14. The implicatic~nsof the embargo on the internal OPEC relations will be discussed in Chapter 5 . Here, the focus is on the market behavior of the Q13EC c o ~ ~ . tries collectively 12, OA13EC is the organization of Arabian petroleum exporting countries. It war; formed in 1968by Kuwait, Libya, and Saudi Arabia. Algeria, Bahrain, Egypt, Iraq, Qatar, Syria, and the United Arab Emirates have subsequently joined the organization (Jenkins 1986:429). IS. Middle Easf Ecurzumic Surztq May 34, 49574, cited in Moran 4981:254 note 33. 14, Oil and Gas Journal June 17,1974338, cited in Moran 1981:254 note 33. 15. The original partners in Aramo, establihed in 494& were Standard Oil of California (SoCal, subsequently Chevron) (30 percent), Texaco (30 percent), Exxon (30 percent), and Mtbil (10 percent). h February 1973, the Saudi Arabian government took over 25 percent of the company, and in 1974 increased its share to 60 percent (Scl~neider1983:40,393,40';7). 16, Fin~rzciailjlITimes, April 7, 1988. The company was not regarded as a Saudi Arabian state company, even though the American companies operated only on a contractor basis. The rc~leof Aramco in the Saudi Arabian oil industry has been subject to further changes due to the restructuring of the Saudi Arabian oil sector in the late eighties. 17. Midl-ne Enst Economic S u n r q July 18,1983, 18. Exxon's net income in 1995 was $6,4170 mi1tl"on;the corresponding figure f o x Royal UutchiShefl was $6,918 million (OPEC Annttnl Stntistiml Btallefin 1995). 19. Central in this marketing of oil is W N E X (New York Mercantile Exchange) and IPEX (International Petroleum Exchange). However; the market is in no way

Markef Context

93

bound to these bourses, as state-of-the-art- infc~rmationtechnolc>gymakes po~ssible the sale of these "paper barreIs""uik independent of geography 20, For a discussion of the role of new trading mechanisms, see Pcfuoleztr7t Infeltigetfce Weekly, April 22,1985; Fesharaki and Razavi 1986; and Mabrc~4987. 21. Plattls Oilgmm Nezus, August 29, 1985, article by Bijan Mossavar-Rahmani, Special Supplement, p. 1; Pefro.oleumIntelligence WeekIy, June 47, 1985, inteiview with Yamani; and Pet-mfetrmIrztelligelzce Weekly, April 25, 1985. 22. "1 think one of the very few cc~untriesthat is still selling on term contracts is Saudi Arabia." htnteiew with Sheik Yamani on June 7; reported June 17,1985, in 17efmleumInfelligence Weekly. 23. The concept had been used earlier by other prc~ducers(e.g., Nigeria), but not in the same strategic way that Saudi Arabia did in 1985. 24. ""1 iis in sc~meways helpful to see fc3rward markets as unripe futures markets. There is a logic in the evolutianary sequence that runs from spot to fomard, and an to futures" (Roeber 4993:45). 25. Sykuta Q1994:4T). Sykuta limits his argument to the purchasers' access to the commodity but as he claims elsewhere the argument "applies equally well to sellers of the commoditgr" "(ibid.:35, fn, 4). 26. It is also worth noting that in every case some type of reorganization of the national oil companies of the expc~rtingcormtries was necessary before it was po~ssibleto adopt a sizable internationalizatim prcjgram. Kuwait" "success" has largely been attributed tcr its organizakian, while the Saudi hesitatictn to invest downstream (in relation to the Texaco deal) indicates that the kingdom had to review its relationships with the Aramco parbers. In the case of Mexicr,, internal problems at Ipetr61eos Mexicanos (PEMEX) have made it very difficult h r the commercial managers to adopt an aggressive Ec>reign investment policy. 1n Venezuela the large investments in fareign refheries have created mounting political opposition to the Petr6lec1s de Venmuela SA (PDVSA) internationatizatjon program (Bergaen, B j ~ r kand Clacts 4989:29), 27. The Ecunomis;.t,No~vernber28,1998. 28. The Economist., December 5,1998,

This page intentionally left blank

Political Context

The ojt-producing corntries are, in addition to being actors in the internation& oil market, also part of the international system of sovereign states, Sovereign9 means there is no formal legitimde political ault.lority above the individual state, The jnternatimlal system of states has no central wlrrld government and is thus defined as anarchy*'This system is a self-help system hvhere the states arc indi,viduaily responsible for upholding their sovereignty, although they might pursue this fundamental goal in concert. By reproduca their sovereignty or surviving as independent political u ~ ~ i tthe s , states also reproduce the marchic structure (Buzan et al. 1993:132). While the market structure distrfiutes scarce resomes, the pcritical structurcz distributes power. 'The valiciity of an fntegrated notion of the wealth and power is not new to students of k t e n a tional relations, 1) Wealth is an absolutely essential means of power, whether far securiq or far aggression; 2) power is essential or valuable as a means to the acquisition ar retention of wealth; 3) wealth and power are each ultimate ends of national policy; 4) there is a tong-run harmony beween these ends, although in particular circumstances it may be necessary for a time to make economic sacrifices in the intel-ests af military security and therefore afsa of Ic~ng-rrm prc>sperity.(Viner 1948:10, cited in Kecjhane 1984:23)

Keohane's cornment on the fourth p ~ i in ~ this ~ t citation is that "in the short:run, tradeof; exist between the pursuit ol power and the pursuit of weal&" "meohane 1984:23). Paraphrasing the citation from Vher (1948), the question to be scrutinized in this chapter is to what cxtelzt security thleats cmstitule "'particulnr circumstnnces" that f o ~ ethe oit producers to make '"econonzic sacriftces i~7the intevclsts of military securify." h exmple could be if the v?;military's support of Saudi Arabia has made Saudi drabia reluctant to increase the oil price. Another case would be if the fran-

96

Political COBtext

ian antagonism agaiinst Western culture has made their oil industry less efficient than would. have been the case if they had been part of thle international oiI industry, The relationship between politics and ecmomics has been the subject of considerable dcbate among polit'tcal scientists and economists, Particular@ attempts to give one factor primacy before the other have created debate. "Unless definiitions of politics and e m o m i c s are arranged so that one category necessarjly includes all. fundamental phenomena, neither econclmic nor political determinism can explain events successfullyf" (Bergstelz et al. 1975:4). Keohane (1929:97-98) points out: " b e n peaple who accept close linkages between domestic and foreip policies frequmtly &W a sharp contrast between 'economics' and 'pofitics,' Tl~underous mock battles take place between analysts who believe that polti-. cal mothations are primary and those who contend that the 'real objectives' of policymakers are economfc." 'I'he interd ationship between politics m d economics is a ma.t.ter of "'hnw the hstihtions, fundamental assumptions and 'rules of the g a m h f political systems support or undermine differat patterns of allocation for econornic acthim as wet1 as ow the nature of economic activity affects the pdtical structure" weohme and Nye 1973:117). If politics and economics are totally inf;ertwined, the ~lationshipbetween them is impossible to detect or discuss. II order to understand the interaction between poiitjcs and economics, we need to separate the concepts analytically although we agree on their ontohgicai inseparability. Shce this study tries to explain the cooperation among oil producers in the fjelds of oil, we will limit this conceptual discussion to how political interests might influence the cJil-related behavior of the cJil-produchg states, A. co~zcrsptcapturhg this process is the term politicizatio~z.

Politicization is a concept commonly used., but which few have attempted to give a precise malytical cmtent. According to one defhGtion, politicizatian is to ""tmsform an. activity that seelns to have na political connotations, into one that is consciously bent towads political 4 s " (%ruton 1982). Here, politicization is attached tc:,the ob~ectiuesof the actoc Scondly, clne can examifie the action itself, or the type of means the actor emplnys. Tn this book, politicization will denote linkage of the economic matters to a political decisionmaking level (Hirsch a d Doyle 1977:11-13). Let us start out with the possibiliq that economic hterests are the single cause of market bchavior. Figure 3.1a illustrates this situation, Political factors can hfluence the actors market behavior in two ways. First, politicizatian. occurs when an ador in a given market allows

Political CC??~ text

97

his political aixns, ozrtside the market concerned, to aMect hir; behavior irz the market (Figure 3.lb). Secondly, politicization occurs when m actor h a given market employs political means for purposes of achieving objectives in the market (Figure 3.1c). In. the first case, pt.iticization can be an unconscious linkag as one" ppolitical intercsts predomjnate over one's ecmomic intert-sts and behavior in the market, fn the second case, politicization will have to be a conscious lkkage. Bergesetn defhnes pditicization as: '"simply the interference oi a g o v e m e n t in the international oil market'7Bergesen 198833). This also agrees with one of Wilsonfs f1982:1(3) delinitions. If we provide that the slate can ncver have any objectives other than political mes and all other market actors can never have politic& objectives, this will agree with the first defhition. If we provi,de Ihat the state has orlly political m a n s and no other actors have any political means, then this agrees with the second definition. However, it =ems mreasmable to argue that states by defir~itioncannot act based only on economic interests. It is also possi,bte that a state would like to abarzdon the politicization strategy. This means that a market behavior previously ixdluenced by political interests are set to be determined by o~nlyeconomic interests. This could be indicated by ietting the ecmonic intercsts deternine both the markt-tt behavior and the political interests (Figure 3.1d). Them might be an infinite nulnber of political factors behjnd the market behavior of oil producers, However, for mnst of the states in question, oil exporting is the dominant eccmamic sector. For some camtries the exporting of crude oil accou~ntsfor all of their foreign currency earniings, and the state income from taxing the oil production is the sole source of a t income. 117 these cases the ec~noMfcimportance of the petroar makes it an impctrtmt political issue, and onty other plllitical issues of m o s t importance will induce a change in what is assumed, by the state leadership, to be the optlmal petroleum policy. 'The survival of the state i s a prominent- issue of this kincf. For religious fmdmentalists, religious imperatives might also be regarded as more important than the ecmomic interests of the oil sector (see Noreng 1997). However, the security interest of the regime alld the cri.tizens is regarded as a, if not the, prominent politieal. [email protected] states, "No purpose of gave central than the protection of its citizens9physical securiw. Philosophicdy, m n y thinlcers have held that this, in fact, is the uftimate reason why hramankhd form governments" "make 1975:247). The claim in this chapter is that the development of the oil-producer cooperation is best u~nderstoodby studylng the relationship between actorskconomic and security i n t e ~ s t sand behavior. What is important is not that states have a hilJrarchy of interests, putting security at the top, but that the priority among hterests might vary over time and according

a) Unpoliticized market

c) Politicized market If

d) Depolit icized market

FIGURE 3.1 Categories of politicization

Political CC??~ text

99

to the actors' perceptio~~ of the situation surrom~dbgtheir decisions (see the situational analysis presented in section 1.4). h this chapter possible threats to the security of the regimes governing oil-pro"duci"g states are regarded as a motive that might change m kdividual country's oil policy The eznpirical discussion is organized almg the division between external security threats (sections 3.3 m d 3.4) and internal security- threats fsetion 3.5). Scction 3.3 discusses the consequences of the Iran-Iraq War for the oil pmd.ucers%behavior in the oil m r k e t and their oil-related cooperatim, and section 3.4 fotlows the same line of argummt in ~laticmto the XraqKuwait M r . Sctian 3.5 deals with the internal political stabaity of oil-producer regimes and the role of oil income in securing these ~ g i m e s while , scrction 3.6 highlights some potiticat cooperative aspects of the OPEC relations. First it is necessary to relate the concept of scl-czlrity to the characteristics of the oil-producing states. 3.2

Oil and Security

According to Buzan (l991:8), tbr concept of security has been dominated by the concept of power in the Itesature on these aspects of international relations: ""reduced to little mom than a synmym for power, security had little imdqeneient relevance in wider systemic terms." Ruzan v o t e s several delinitions that he finds point out imporlant aspects of natimal security, namely, "the centrality of values, the timixlg and intensity of threats a d the political nature oE security as an objectiwe of states" (Ruzan 1991:lW). Bt12;mdoes not seek an overall defhikion, but instead divides the security concept into five categories, related to the sectors affecting national security of which three are of particular importance to the discussion in this chapter:' Generally speaking, military security concerns the two-level interplay of the armed c3ffensive and defensive capabilities of states, and states"gercept.ic>ns of each ather" intentions. 130filicalsecurity concerns the organizatianal stability of states, ssytems af government and the ideologies that give them Zegitimacy. Economic security concerns access to the resources, finance and markets necessary to sustain acceptable Xevefs of welfare and state poweu: (Buzan 11i391:19) ?i, pursue

the fundamntal interest of suwival, states have to resort to power capabililies. I'hese caphilities can be physical (like oil =serves in the ground), or.t;anizationat, financiat, technological, or purely political (including military forces). Usually, nzilitav force, or coercive force, is regarded as an ultimate capability:

200

Political COBtext

Tf the securiv dilemma for all states was extremely acute, military force, supported by economic and other resources, would clearly be the dominant source of power, Survival is the primary goal of all state%and in the worst situations, fcxce is ultimately necessary to guarantee survival. Thus m%tary farce is always a central component of national power. (Meohane and Nye 49"i77:27)

As will b e c m evident in this and the next chapter, on the role of Saudi Arabia, states f a c a security thwats have a cclmplex menu of power capabsties; t h y need not resort to eoerche force immediately- In addition to such hndamental resources, states need the skill m d will to use their power resources effectively (Raldwin 1979:163). ?"he resources themsdves are ol no use unless the state is able and hvilling to use them.. For instance, Kuwait in 1990 had. no ability to stand up against the Iraqi invasion, A l h u g h K w a i t had been attacked before, it did little to buifd up military deterretnce against such attacks. h o t h e r problem is the context of power resources: ''the capabilities (or potential power) of an actor must be set in the conterct of a 'policy-contingeq-frameworkf specifying who is trWing (or might try) to get whom to do w:hatr' ((Baldkvln11979:l,tj4). Certain types of capalailities might not be effectively applicable in certain issue areas or & b a t h s . Baldwin argues that political power resources are relatively low in fungibility; for instance compared with money. Ofien it is impossible to distinguish betwem a country" security and economic interests regarding oil-market-rda.tcd behwior: ""Aside frorn our military &fen=, there is no prnjed of more central importmce to our national security and indeed. our independence as a sovereign naticzn [than energy saving, import independence, and ccmsun-ter cocrperatictn]" (Kisshger 2982b:xx). This staternetnt indicates that oil importing at reasonable prices is perceked as a question of national security, nut as a matter of economics h r . The discussion of separating the issues quickly becomes one of words, but this is nt:, less importanl, as Che use of words can i y l y changes in, actors3beavior: "By saying 'security' a state-representative mowes the particular case into a specific xea; claiming a special ri&t to use the means necessary to block this development" ( W w e r 1989~54, cited in Buzan 1991:li").Since the politicization of market hehawior seems obvious if the actor regards the oil market in the same way as Kjssinger, it might not:be wnrthwh,ie to distinguih the economic interests from the securiv i n t e ~ s t s.A . further restriction of relevant cases to the ones where actors give pricrrity to che security interests is thus necessary. The Mowing scl-ction will therefnrtl deal wjth cases where oil is related to potential or actual threats of intervention or subversion. In such cases it seems reasonable to assume that the security interest of the state takes priority over economic welfare interests..

Political CC??~ text

202

Most of the states hcluded in the discussion in this chapter are autocratic ~ g i m e sThe , state leaderships perceke threats both from internal oppositim and from other states. They themselves art? perceived by other political actors, domestic and foreign, as posing threats both toward other states and toward their own citizens.2 This makes the traditional external security concept tot>namow for comprehension of the securiy issue pertaining to the Middle East oil-producing states. In addition to demanding a mme comprehensive conception of security, this reality also loosens the assumptiw of states as unitary actors, which is a methodological basis for the most of the discussion in this book. In this chapter, the "black box" will be opened in discussing the role of oil in the state-society relations in selected oit-producin; countries (see particularly section 35). The transcending of the national-iaternationa1.level divide is even more important fur the next chapters, which include the case shtdies of Saudi Arabia and Norway (see Chapters h and 9). The followi.ng two sections will focus on the impications fos the oilproducer cooperation of two major military conflicts: the war between Iran and Eraq from 1989 to 19W (sect-ion 3.32, and then the war between Iraq and Kuwait frnm 1990 to 1991 (section 3.4). 3.3

The Iran-Iraq War

Iran and :Iraq have been involved in most military conflicts between OITC countries. This c m be shown with data from the data-set Mifitarized hterstate Disputes (htemational C-o~~sortium of Political Scie~~ce Research (ICI>SR data-file 9044). These data are based on coding pmcedurcls simjlar to those of the Corwlates of I/Var data, with two important exceptions. First, the criterion of number of deaths is zero, while for the war data-set the minhtlm number of deaths is set at 1,000, Second, the war data-set requires s w e military inkraction to have taken place, Mthile in the dispute data-set it is sufficient that military threats of some form have been uttered.. A total of forty-six disputes in this data-set involve two or more C>T9ECcountries, on opposite sides. With one excep~on-a border dispute between Saudi Arabi,a m d Qatm in 1992, k i l h g two border-gmrds-Iran or Iraq has been invoked in afl disputes between OPEC members since the Scond World War, E m and Iraq have never been on the s a w side as any other orEc member durjng the s m e period, except for the Kuwait-Iraq War of 1961 whert. Iran supported Kuwai.t. mese h & g s suggest that all militarized conflicts having effect m the C ~ I ~ Ecooperation C arcr related to co~zflictshvolving Iran or Iraq. Most prominent among these confl,ictsis the war between the two countries from 1%8 to 1988, The factors behind the Iran-Iraq War were many. Traditional conflicts between Sumite m d Shiite Muslims, historical cmSfici-sc o ~ ~ c e m hdisg

202

Political COBtext

puted land and river areas, m d expansionist objectives by former leaders had creakd clashes between the two countries for centuries. Thus, Iraq's declaration of war on :Iran m September 22,1980, was in line with a traditional conaid pattern in thit; region. Iraqi troops soon made substmtial advances into Iran, but then Iraq stopped and offered to negotiate. The Iranians rt?l"used, and at the same time gairted time to mobilize. At the end of 7,981, the lmqi advances c m e to a complete hatt. In the sprjng of 1982, the Imnian counterattacks had substantjal success, and the temptaticm to continue into Iraq became irresistible for Tehran. Although the Irmian forces gained some Iraqi territory; the war soon ran into a stalemate; neit-her of the parties was able to advance forcefully over a longer period. Iran ccmducted several offensives agaillst Iraq between 1984 and 7,987, but none of such a magrnitude to decisively etnd the war, h February 19M, as Iran started the first of these so-called final offensives, the war turned to the Persim Gulf waters, and the tanker war commenced. Iraq attacked shjps going into lranian ports, probably with the aim ol provoking m Iranian blockade of the Strait of Hsnxuz, which in turn would prompt the Weskrn countries to take mom sevext- action against Irm. Irm did not fall for the trap, but lawnckd attacks on tankers serving Kuwaiti and. Saudi Arabian ports in order to deter these countries from supporting Iraq. By 1985, the intensity and bmtality of the land war had increased again.'The use of ehernical weapolns contributed to a substantial increaser in civilian deaths; estimates range between 450,000 and 730,000 m the Irmian side, and 150,000 and 340,000 m the Iraqi side (Cordesman and Wagner 7,990:3).Adding the dilsect and indirect costs of the warfare, estimates have been put at $627 billion for Iran and $2561bjllion b r Imq.3 Most important for the oil market was the war in the Persian Gulf initiated in 1,984. Three hundred s h j p w e r e hit fmm 7,984 to 1988, Furthermore, Iraq tried to destroy Iranian oil-loadjng facilities, particularly on :Kharg Island. Although Iraq received laser-guided bombs and navigation equipment from France, the number m d small size of loadlng polnts at the Kharg hstallations made the bomblng ineffectfie, The Xrmim oil production varied during this period, but on average a substantial level was sustained. The Miest did not pay m c h attention to the conflict until 1984. Nat even the Iranian advances into Iraq caused much concern. One reasm was that until June 1982 there was virtually no activity at sea. The first sea was efforts were also characterized as small, sisrgle attacks and not part of a broader strategy of sea war, 'l'he oil proctuctim of the two befligercnt countries was already low, so this did not change the flow of oil horn the Persian Gulf. While Iran in 19% started using the ""hman wavef?actic, lraq escalated the war into the Guld hvaters. This obviously bmught the warfare

Political COBtext

203

closer to the other Gulf cou1ztries3erritories,and they were soon directly affected.. On May 14, two Kuwaiti tmkcrs were attacked, followed by an attack on a Saudi Arabian tanker on May 17, This development also brought the United States into the conflict, as Washingtan decided to supply Saudi Arabia with 400 Stinger antiaircraft missiles (Kechichian 1990:IOQ).Thus, Saudi Arabia was set to defend the Gulf waters. On June 5, 19M, a Royal Saudi Air Force jet shot down an Iranim F-4 Phantom i4ghter-bader. The kingdom tried not to let this incident bmadcn the war.W~oever,&er this point the other Gulf countries' ahifity to mediate or end the conflict by other diplomatic means was severely weakelzed (Kechichian 1990:101), The next major shift in the war directly affecting the other Gulf ccnantries was tbr change in the kanian sea war strategy. As Iraq intensified its attacks an marg Island and the Iranim shuttle traffic to the more distant islmds of Sirri m d Larak, the Iranians intensified their attacks on sea traffic to pro-Iraqi countries like Kuwait and Saudi Zrabia, Since Iran's air force was insitiqnificant,its light naval forces conducted hit-and-run attacks on tankers, a kind of naval guerrili7a warfare. 'f'hus, the ability t~ actuatly sink supertankers was lirmfted until Eran was supplied with heavier rockets, particularly by China. By 11986 the Tranjan attacks on ships in the Gulf increased, and the attacks became more severe. fn January 1987, K w a i t sought US naval protection of its tankers by suggesting Chat the ships be "re--flaggedf' to fly the "Stars and Stripes." Kuwait also leased three Soviet tankers, which the Soviet Union agreed to esccnt. By June, the first ships with US flags and protected by the us Navy entered the Gulf. 'This led to an increased hternationaIIzation of the c d i c t . Both superpowers were now directly invoked in the Guif War. In spring of 1987, :lranian forces attacked two Soviet tankers. fn May Iraq attack& the us frigate Sfarh; kiling thirty-seven mariners. This caused a political crisis in the United States, and Western public attention to the war increased dramatically, W-ren Iran laid out mines that reacl-red Kuwaiti parts, mhesweepers were sent to the area, and a joint operation with Saudi Arabia, using Saudi Arabim AWACS surveilZance planes, was established. On Jufy 211, 46387, the UN Security Council passed UN Resolution 548, calling for an immediate cease-fire. Saudi Arabia strengthened its position in the Gulf both as a result oi the US presence and t h r o q h its own rnilitary buildup. In particular, the Royal Saudi Air Force became the strangest air power in the region during this period,Waudi Arabia's increased self-confidence also m a n t that it became more willing to challenge Iran. Atter the eighth GCC (Gulf Cooperation Council) summit in 1987, the Saudi Arabian foreign minister, Prince Saudtal-Faysal, stated that "the aim of the ccc countries [was] to end the war and put an end to Iranian attacks against the GCC statesf' (Kechichian 11990:10&).While Kuwait and Saudi Arabia wanted ta take a

2 04

Political COBtext

tougher position against Iran, the United Arab E~airatesand a m a n sought continuous dialogue with Tehran (Kechichian 1990:106). Also, the tjmning down of 400 lranian pilgrims ir-t Mecca on July 31,1987, must be understoud as a result- of a more polent and confident Saudi Arabian regime. The war thus influenced the overall power relatims betwem key OPEC members. The next question is what effects the war had on the economic oil-related relations between the affected OPEC cotmtries* The price effect was short-term., and the lasting importance for the market was the disappearance of the two countries"roduction capacities. This made the hegernonic role of Saudi Arabia even more prominent, as these two countries, given their installed production capacity prior to the out'break of the Iranian revolution, were the possihle challengers to the Saudi Ara:bian posi.tion (see Chapter 6). Tbe Iranian oil production had been about 6 mbd in Septesnber 1978, while Zraq produced about 3.5 mbd in the first half of 1980. W-th the outbreak of the war, the Irmian production fell h m 1.3 mbd in August to 30,000 barrels in C)ctober. Iraqi production fell from 3.4 mbd. to 500,000 barrels over the same period. Figure 3.2 shows the devdopment of the two corantries' pwduction from I976 until 1989. The figure shows how Che Iranim production first dropped due to the Iranian revolution. It was restored at a lower level, but dropped again at the btl ing of the :Iran-lraq M r . The Iraqi ing of the lran-Iraq War" oil production atso clrnpped at the The problem to he raised in this section is whether the conRjct between the two corntries created prohlems for the economic cocrperaticm inside OPEC in the bargaiining process. Regarding the OPEC negotiatio~~s, a sttmmit of heads of stat scheduled for November 4 in Baghdad to celeersary of the organization in its founding city. brate the twentieth Plms were made for a renewal of the organization. 'The meethg never took place, However, on December 15,1980, the OPEC Conference met at Bali, with reprmentatives of two belligerent parties. 'I'he head of the Irmian delegation, CM Mnister favad Tondguyan, was in fact a war prisoncr captured by Iraqi troops at the beginning of the month. -Terzian (1985:28&287) describes how OPEC hmdled this first meeting after the war broke out: The Iranian delegation entered solemnly, bearing a pc~rtraitof their captured minister . . . installed the photo of Tondguyan on the chair reserved for the head of delegatbn and took their places behind it. Stxharto, the 13residentof Indonesia, arrived next and, after his inaugural speech, launched an appeal far a cessation of hostilities beh-een Iraq and Iran. . . . the Iraqi and Iranian delegatians were given the floor, They exchanged endless accusations, each cl-runtryblaming the other E c the ~ war. Finally, when the diatribes tzrere over, the ten other OPEC members adopted a resolution backing the call for peace

206

Political COBtext

issued by Indonesia. Once all these matters had been settled, the delegates moved on to the other items on the agenda, and it was business as usual.

The officid OPEC reaction was expressed in very diplomatjc terms: "The Conference endorsed the since= and honest appeal made by Hjs Excellence the President of the Republic of hdonesia in his inauguration speech to the two Member Cout~tries-Irm and Iraq-who are presently in dispute, to ytlickly seek the best possible solution to their c d i c t Leading to a peaceful settlement of their differencesf"(ar>~c1990:192). Due to the war, Iran became a pure price-taker, maximizing its oil exports. The success of this strategy is clearly illustrated in Figurc 3.2. Iranian oil production rose throughout 1982, from 1.1 rrtbd in March to 2.8 mbd in Dtzcernber. Ch April 10, 1982, Syria cut off the Iraqi pipeline to the Mediterranean, cutting 400,000 bd of Iraqi export capacity, Also illustrated in Figure 3.2, Iraqi production fell from 1.3 mbd in the first quarter to about 850,000 bd in the second half of 1982. The quotas allocated to the two countries in ApriX 1982 were 1.2 mbd each, At the Sixtyfifth O19EC Conferltnce in July 19132, Eran "demanded a quota of 2.5 m b/d and that its increase should be take11 from the Saudi. share'' "(Skeet 1988:188).The boldness of this demand was fueled by the success on the battlefield. fn 1982 Iran had stopped the Iraqi adt.ances and was ready for its cou~~terattaek, and only weeks after the meeting, Iran invaded Iraq. The resoktim of the Sixty-fifth Conference simply states that ""the Conference . . . decided to suspend the deliberations of the Conference until furt-hernaticef"or*~@ 1990:204).As Figure 3.2 shows, Iran did nat at all abide by its quota in the following months. Oxl the battlefield the Iranim offensive was soon stopped, and the war became a war of attrition. Saudi Arabia and the other GC@countries became more relaxed, no longer fearjng rapid Iranian expansion. They accodingly began to take a tougher stmd agai~~st Iran, a stand that extended to the Q13EC negotiations. A statement from the cc oil ministers' meeting in Salala on October 14, illustrates this: "The other producers should h o w that the cc Ministers expect them to shoulder their responsibilities and if they continue their misguided actions they will not be proteded by the rnetnber countries f r m the GCC from the consequences of these actions" ((Skeet 1988:189). h his openhg address at the followhg OPEC rneethg, the chairmm of the Nigerian delegation, Aahaji Yahaya Dikko, with a clear reference to the above statement, said that "thmats artl never a basis for ct,operationf" (Evans 1990:607). I;he Confertnce decided that the total quota b r 1983 should. be 18.5 mbd., but it was unable to agl-ee m the allocation of national quotas. 'I%rus,the 33.5 mbd ceilhg wap; not ccmvirrtchg to the market actors (see Chapter 7 br the crit.ecia for a successful cartel).

Political COBtext

207

The Sixty-sevcmth Co~zferellcewas held over t w h e days in Londnn, but only the last hours on the last day transformed it kzto a full.coslference. The Iranian yuota was lioubled, from 1.2 to 2.4 mbd. Iraq agreed to keep its 1.2 mbd quota ody on the condition that it should be revised upbvwd to parity with that of Iran as som as its production capacity allowed.. Saudi Arabia was not allocated a quota but was officially gken the role of "swing producer to supply the balmchg qumtities to meet market requirements" (OPEC 1990:208). Obviously, the following domswing of Saudi Arabi.a would have k m harcier had both Iran and. Iraq had their exporting capacity ildact. Given these facts, one shodd expect there to hiwe been a h.lrbuI n t situation when the war ended in 1988.However, by that time, the price had crasl~ed(in 1986), and the market to some extent had taken into accotmt the production capacity of the warring parties. As the war damage to oil iPlstalXations was reduced through the us presence inthe Gulf and the rnilitary buildup of the GCC countries, the production of Iran and Iraq picked up gradually, givbg room for a smoolher market adaptation. lie, conclude the discussion of the implications of the Iran-Iraq Miar on the OPEC cooperation, it is necessary to distinguish between the effect the wm had in changjng the power resomes of the belligerent parties and the effect on the OPEC negotiations as such. The OPEC negotiations were only modestty intluenced by the war. Aside from the incident involvifig the captured minister, and some exchange of accusations, the negotiations were business as usual. FXowever, the power basis for the diff-erent countriesfpositicms in the OPEC discussi~nswas dramatically changed by the war: Both Iran's s a n dIraq's oil production dropped significantly at the beginning of the war, from ;above 3 nrbd to under I. mbd each. Iran was able to =gain a prc,duc"cicml e d of about 2 mbd, while the Iraqi production remahed at about 1mbd until 1985, and then started to climb slowly (see Figure 3.2). This rttduced these two countries+positions in the OPEC bargaining compared with those of other countries like Saudi Arahia a d the other Gulf states. 3.4 The Iraq-Kuwait War

At the Arab Cooperation Council meeting in February 1990, Saddam Hussein told King Hussein of Jordan and President Muhark of Egypt to tell the (I;& states to forgive their loans to Iraq. Saddarn perceived the loans as payment for Iraq's defending the Gulf states agahst Iran, ' l e t the Gulf regimc.s knt,w that if they do not give this money to me, I will h o w how to get it," k said.6 At the Arab summit meeting in May 1990, Saddam argued, that the continued violation of orEc quotas by some Arab countries was eyuivalent to a declaration of war against Iraq. In a radio speech in July, his argument was even stro~zger:

208

Political COBtext

War is fought- with soldiers and much harm is done by explosions, killing, and coup attempts-but it is also done by economic means. Therefore, we would ask our brothers who d o not mean to wage war on Iraq: this is in fact a kind of war ag"rnst Iraq. . . . we have reached a point where we can n o longer withstand pressure,'

In mid-July the buildup of military forces and a diplomatic ofiensive began. OTI July 16, the Iraqi furcign minister, Tariq Aziz, delivered a mernormdum to the secretary-general of the Arab League, presenting the Iraqi indictment. h a n g otbcr accusatims, Kuwait and the U'xnikcd Arab Emirates were accused of having created a glut in the oil market, brhging prices down with devastating wonornic impact on the Middle East and Iraq. Furthermore, Kuwait was clahed to have robbed Iraq by setthg up oil instaktions in the southern section of the Iraqi Rurnaila oil field (Fwedmm and Karsh 199,7:4748). The followifig day Saddam itlcreased the tension by an address ta the nation. He accused Kuwait and the United. Arab Ernisates of conspiring with world imperialism and Zionism ta ""cut off the livelihood of the Arab nation," and threatened that "if words fail to afford us protectim, then we will hnve no choice but to resort to efEcctive action to put things right and ensure the restitution of our rights.'"Xwait ~ s p o n d e by d "rehting the Iraqi accusatims and expressing strong jndignation at Iraq's behavior" ( F r e e h a n and Karsh 1993:49).Wn July 31, representatives from Iraq and Kuwait met in Jeddjih for negotiations after Iraq had. plxed l(fO,,(>(lf) men along the Kuwaiti border. The talks etnded the next day. The parties d i s a g ~ e don the reawn lfor the bseakdom of the talks.1" On August 2, 1990, Iraq invaded and occupied Kwait. The Iraqi military operation was a sttccess: Kuwait was occupied in twelve hours." Contrar). to the Imn-haq Wr, the first phase of this c d i c t was iess characterized by changes m the battlefield and m m by those in the poli.tical and diplomatic arenas.Two developments made the invasion look less like a success for Saddam, The first was that the in.ternationa1 response, ing the Iraqi action, included most Arah countries. At the meeting of the Arab League on. August 18, Egypt proposed a resolution cmdemning tlne imasim, imposing economic sanctions m Iraq, and offering military support to Saudi Arabia. Twetve of the twenty delegates supported the resolution, m d only Liibya, the Palesthian Liberation Organizatjon (PLO), and Iraq voted against it. The other development was the US deployment of military foxes in Saudi Arabia. There had never been foreign forces in Saudi Arabia, and the thought- that the Cuslodjan of the Two Mosques s h u l d be defended by large us forccs seemed odd. I'he Unfted States, thc world's largest oil consumer, f& indirectly threatened by the prospect of Iraq havifig taken the Kuwaiti oil reserves, being

Political COBtext

209

able to put political or military pressure on Saudi Arabia. "President Bush became cmvlnced that the kingdom was vuherable and that the United States could not afford to see an Iraqi takeover'" (Freedman and Karsh f,993:86).When King Fahd al~reedto the establishment of us foxes on Saudi soil, the Operation B s e r t Shield started. Already on the night of August 2, the UN %curity Council had passed a manimous resolution (669) condernnhg the invasion. C h August 6, it decided to introduce economic smctians agajllst Iraq (Resolution 661)," and on h ~ s 18,t it sanctioned an embargo of lraq (Resduticm 67675).By mid%p&rnher, the us forces in Saudi Arabia had rf,?a&ed t,5C),Om.The Iraqis were distul;hed, and started building up forccs in Kuwait and southern t ""was becming impatimt Iraq. h early October it seemed P ~ s i d mBush and increasingly pessimistic With regard to the effectiveness of [the emOctober 30, a pronomic] sanctims'~(Frcedm.anand Karsh 1993:203).1" posal for the additionaf f0rc.e~needed to turn the operatim fnto an o f h sive orlc was presented to the president by Colin Powell, chairman of the Jcht Chiefs of Staff. "Bush reportedly concurred: "If that%wwh you need, we% do itm"Reedmm and Karsh 1993:208), On November 29, the UN Security Council passed a resolution sanctioning the use of afl necessary means unless fray did not withdraw horn Kuwait by Januarqi 15, 1991 (Resolution 678). Hectic fnternationai diplomatic activity folrowed until the Last mhutes before this deadlh~e.No solution was rea&ed, m d on Jmuary 17, the air strikes commenced. Operation Desert Storm had. begun, On Febmary 24, a ground attack was launched; it lated h r 100 hours, The ernbargo of lraq was set to continue unlil Xraq complied with all UN sanctions, inciudhg the one demanding the destruction of all. Trayi atomic, biological, and chemicai weapons. 7"he sanctions have caused substantial st~fferingm o n g Xraqi citizens, with m increase in diseases and child mortality. Although the embargo djd not include food and medicine, without oil expwts Iraq had no money to pay for such goods. In spring of 1996, negotiations commelrced in order to Let Iraq sdl some quantities of oil to pay for food and medicine to its suffering citizens. The UN Security Council boycott wsolution of lraq and the new regillle in Kuwait inAugust 1990 immedjately brought to a halt all oii flows from the two countries. The oil price skyrocketed from $18/barsel to almost $3C)/harrel fn a couple of weeks. Other pmducers, particularly Saudi Arabia, quickly filfed Che gap. "01119 August the Saudi Oil Miniskr, Hisham Nazer, annomred that his countsy muld. increaise proburtion by 2 mbd with or without tbe blesshg of QI~EC" (Freedman and Karsh 1993:182). There was no shortage of oil. Rfier a few weeks, the WEC production was back at almost its prewar level, Some additisnal barrels h-om n o n - m ~ c producers made the total market balance, atthough there were some elements of panic buying (see section 53).

210

Political COBtext

After conducting informal consultatio~~s among OPEC oil ministers in t'ienna from August 26 to 28,1990, and convening a meeting of the OPEC Mhisterial Monitoring Committee in Viema m Wednesday August 29, f 990, OPEC issued the followin&press release: After reviewing the trend of the oil market, and being of the opinion that it may wet! be prc>longed,thus increasing the uncertainty of oil supply in the fourth quarter of 19% and the first quarter of 2991, ot3sc,as a body which is fully aware of its mandate and respansibility to help ensure an adequate glc>balenergy supply; has independently decided to adopt the fc~llc>wini; position and interim course of action: 1Ci,clearly restate to the world that ar3nc stands for market stability and regular supply of oil to cmsumers, . . . That OPEC shall cc>nsequentlyincrease prc>duction,in accordance with need, in order to maintain the above-stated objectives. . . . Consumers, hoavveveu; need also to actively participate in the stabilization process, by executing the LEA Oil Sharing Agreement, which was specifically designed to meet eventualities such as the p ~ s e n situation, t and also by utilizing the present huge accumulation of stocks owned by the oil companies. . . . This is a temparary arrangement, applicable only until such time as the present crisis is deemed to be over, and this arrangemat shalt, not in any way compromise the provisions of the 1990 resoilutim, which is still valid. Once the present crisis is cmsidered to be mex; the Organization shall return to the July 1990 Resoiutim. Qo13nc195)Q:310-311)

Although OPEC wanted the TEAto initiate its sharing agreement, there was harmony between key OPEC mernbersyolitical viterests and.the possibility to act "responsibly"' by creating stability in the oil market. For Saulii Arabia and other producers with spare capacity, indivihal short-term, ecmonic inccntkes were satisfied through increased. productjon; for the countries with Less spare capacity, a price increase would be more beneficial. Iraq, Iran, and Libya accordingly advacated that OPEC should not increase production but l& the price increase. Tndonesia, Algeria, and Nigeria favored only moderate production increases. As the adopted resolution hardly put any restriction on the individual countries' increase in production, Saudi Arabia stood to gain the most from the organization's response (see section 6.5). Refore discussing the effects of the war on the oil-producer coopemtion, it is necessary to outline brie* the oil-related reasons behind the outbrt-ak and outcome of the Iraq-Kuvvait conflict. Two y?xestions arise rclgadi,ng the relationship between the oil il~terestsand the Gulf War. First, to what extent was the Kuwaiti uverpoduction (measured against its aflocated OPEC quota) an important reason for Saddam I-fussefn to invade Kuwait? *cond, to what extent was the fact that the mnexation of

Political CC??~ text

212

Kuwait gave Saddam control over about 20 percent of world oil reserves an important reason for the United States and Saudi Arabia to try to evict Saddam from Kuwait? As described above, Saddam, in his political attacks on the Gulf states during spring of 1990, fizcused nn both the r e p a y e a t of war loans from the Gulf states in connection with the Iran-kaq W a and the low oil price caused by the overproduction by particulady K w a i l and the United Arab Emirates. The Kuwaiti drilfjng activity in the Rumaila oil field, which crosses the border between Kuwait and Iraq?was also a cause for Iraqi concern. However, the question is not whether these factors were a part of the Iraqi argument in the buildup toward the invasion, but whether Iraq would have abstained from attackhg K w a i t had these factors not been present. In other words, we= these actors necessary conditions for the oufbreak of war?l"t is obviously problematic to distinguish the ecmomic motives from the political ones, since they in this instance, pull in the same djrectinn: Both lead to Ihe conclusion that Saddam stood to gaixl from invadjng Kuwait, The condusim offered by Freedmm and Karsh (4993:429) on this issue is i~~teresting to note: The mistake in the Arab world as well as in the West was not to recognize the desperation of Sadclam" situation. Such a recopiltic~nshould not have led to pressure on Kuwait and the United Arab Emirates to give generous1y to Tray, alf'hough that would probably have been the result. As the Kuwaitis recognized, if Saddam was rewarded this time he tzrould =on be back for more. They might have been mc~retactful, especially in the realm of honoring oil quotas or gratuitous snubs to Saddam's pride, but our j u d p e n t is that only a very large bribe tzrould have persuaded Saddam to pull back his troops once he had determined on his campaign of intimidation.

I'he previous support of Iraq fn the war against Iran, and the importance of Arab unity, szlggested that the other Arab countries would be rductant to interfere once the Xraqi invasinn was completed. Give Iraq" eexyansitrnist and imperial ambitirtns, the attack m Kuwait cannot be ascribed sdely to the lack of some billjons of dollars in loan remittance. Regarding Kuwait" and the United Arab Emirateskverproduction against t k i r OPEC quotas, one shouid note that Iraq itself produced substantially above its allocated quota, and in fact did not accept its quotas for most of the eighties (see section 7.4). This argument was part of the rlnetorieal game comccted with the invasion. The economic issues were used to 1egitirnize the invasion rather than the fmzdamemtal reason for it. Saddam. regarded. the international and regional political situation as one in which he c o d get may with thcl invasion of Kuwait, and he took advantage of this perceived opportunity. It turned out that either he xnis-

212

Political COBtext

read the situation or his opponents crihmged their mhds as they saw the implications of his policy This leads to the question about the mcrtives of the United States and the Gulf cottntries in opposing the invasion. Again, the problcnt of distinguishing different motives of individual actors appears. For the United States the normative arglament of punishhg the obvious breach of fundamental international norms of nonaggression, the political argument of securing the sovereipty of an important a f l y like Saudi Arabia, and the undesircrd situation ot: having Saddam in cmtml of one-fifth of world oil reserves all putled in the same disedion. 'The answer was to eject Saddarn horn Kuwait, For the other OPEC comtries, this was not so. The 0 1 2 1 2 ~advocates of prt-ssing the oil price higher got unexpected help from Saddarn, white the law-price advocates (the Gulf countries) focused on the negative effects of another period of oil-market turmoil. This division of the OIIEC: mc-)mbersfinterests was exposed. at the OPEC meeting in Vienna in late Aug~~si: 1990, where several non-Gulf cou~ztriesadvocated not increzlsing production in order to counteract the price rise after the invasion, For the Gulf countries, the security interests were of utmost importmce. Although it seemed unlikely that Saddarn would push farther into Saudi Arabia, gambling the nationshovereignty on the rationality of SadcIam seemed foolish. Fur&ermore, with an acceptance of the annexation of Kuwait, Saddarn w ~ t ~ 1bed in a position to increase the pditical and economic prGssure on the GuIf countries, even without military threats toward them. In the longer run this could have politicaw undermined the fragile regimes in the Culf countries and possibly pressed them to change t h i r oif policy according to the interests of Saddm. As srtggeskd in Chapter 6, the security fnterests of Saudi Arabia have in several instances been important in exflaining the kingdods oil policy. The Saudi kabian r e g h e found itself ixz a very tight position with Saddam's a m y on its n o h e m border, imposing thrtrats fn order to change Saudi Araf3ian oil policy This leads to the h a 1 issue in this sectim: Che implications of the Iraq-Kuwait War for the OPEC cooperation, After tbr war, Iraq was subject to an ernbargo; its oil productim was about 50U.UOo barrels a day until the end of 1996. Accordhgly, the pmduction-sharing between the other members was eased. In particular, Saudi Arabia's position had been strengthmed. Table 3.1 compares Saudi Arabids and Iraq's share of orEc production h Juiy 1990 with the pmduction from 1991to 1934 (see also Tabk 5.2 fnr the xlorninal figures). As Table 3.1 shows, the Saudi Arabim share rose from 22.45 percent to 34.22 percent of totat orEc output. 'The S a d i Arabian position inside OIYC became even stronger than before. The U N sanctions prohibiting Iraqi oil exports made the UN an "acting member oE OPEC," exercishg production limitation on behalf of Iraq, h o t h e r effect of the war was

Political Cmztext

213

TABLE 3.1 Saudi Arabian and Iraqi Share of OPEC Production, July 1990 and 1991-1 9% ("h) Jialu 1990 2 992-1 9911 Saudi Arabia 22.45 34.22 Iraq 13.51 1.C75 Total Share 35.96 35.87 SC)UIclCmlt"zimEcononiist; BP Statistical Review of WmId Energy.

that the Iraqi opposition internally in or)Fic was mndified. The earlier harsh criticism of tbr Gulf stateskil policy was no longer a viable strategy for Iraq in mec. Without any oil exports, its role in the organization was dubious, mreats from the Iraqi leadership to other countl-ies due to their oil policies would rapidly have been met by conde possibly smdions. At the end of section 3 3 there was a distinction made between the direct effect of the km-Iraq War on the OPEC negotiations and the effect on the basis for the beliigerent parties' position in OPEC. 'The result of the Irm-Iraq War was that the position oi the parties was weakened, but the ~ I ? E Cnegotiations were conducted in a business as usual fashion. After the Iraq-Kuwait War, the Kuwaiti oil-production capacity was som recovered, and.the countsy" position in OPEC has been strengthened, particularly because it was given carte Mmche to increase oil production above previous quota levels in order to acculnulate revezzues to recover from war damge. By June 1.992, Kuwaiti productim passed 1 mbd; in Srptember 1993, it leveled off above 2 mhd. When it comes to Iraq, its production reached 2 nnbd in April 1998, and 2.5 rnbd bp February 11,999. Accosding to the UN mandate, Iraq was in 1998 allowed to sell oil worth $5.25 billion. Having set the terms in $-value, the lower the price the more Iraqi oil would be allowed on the market. However, the lack of Iraqi production capacity did not bring the production up to the mandated l e d in $-value..As for fie Eraqi econorny in genclraf, the UN embargo following the war has clantaged the corntry's scverelyYIts position is thus weak and without any basis for economic or political infiuence h OPEC or in the Middle East region at large. At present it seems unlikely that the sitting regime in Baghdad will be ab1e to regaiin sufficient confidence among the other OI~ECproducers or the Arab states to play a major part in regional poii.tics. However, the poiitics of the MIddle East are unpredictable, so one cannot be certain on such issues. h o t h e r point is that if there should be a new regime in Iraq more hiendly toward the Gulf states, it would most likety be supported, financially and politricallyIboth by the c;rc countrim and by the United States,

Political CLHZ text

3.5 Internal Threats

In the previous sections the state was regardetet as a unitary actor, but with two prominent interests: security against external threats, and income f r m oil exports. This was a fruitful simplitfication, as the puvose of the prt-vious sections was to study the effect on fie oil cooperation of the military interaction between OPEC countries. In this s e d i o ~the ~ state is not viewed as a unitary actor, as the focus will be on the internal threats to the state. Such threats might often be as seric,us as externai threats. Ruzan (1991AlS123) distinguishes between politicd and societal threzlts, although he notes that they can be hard to disentangle. The political threats are "aimed at the organizational stability of the statef' (Buzan 1991:1fW), while the societal threats "amount to attacks on naticmal identity" @uzm 1991:122). Although both political and societal fireats can be external, the focus in this section will be on the htemal types of such threats. In, several oil-producing countries the political legitimacy of the regime is controversiat. The societies are polarized aiong religious, ethnic, or other dintetnsions, and the regimes are wart of their weak political foundation. The oil income in many cases plays an important role in reirtforcing fie pditical basis of Ehese ~ g i r n e s E .t is thus appropriate to ask to what extet~tthis inauences the oil.-rela.t.ed cooperation between such states.15 First, it is necessary to describe the relevant in.ternal political characteristics of the oif-pmducimg cowtries in vestion. Tc, descrihe l.he End of political regimes that haracterize the orEc countries, :l have utilized informatiun from. the PolitJJ IIE database on rt?girne type and poli.tical authority, and data from Freedom House on po:liti,cal, rights and civil liberties (Kwla,n 199fi).'"e Poity HI data,base includes indexes for calculating democracy-autocracy s c o r ~ sfor 161 states from 1946 to 1994.17 Democracy and autocracy art. measured on a 21-point scale from -c10 (dcmocracy)to -10 (autocracy). The global average of democracy measured in this way was -2.4 in 1977 and cliHlbed to -1-3 in 1994 (Jaggers and Gurr 19%:477), The Freedorn House data are from 19995 and 1996.1" Table 3.2 shows the average score on this index for twelve of the OPEC countries from 1970 "a 1994 (no data were available on Qatar). Of these twelve countries, only two had a positive delnocracy score h the period from 1970 to 199.11.There are substantiai regional differences between the OPEC mernbers. As Flgure 3.3 clearly shows, the demcrcratic OPEC countries are located in Asia m d South America, while the members located inAfrica m d the Middle East have on average very high negative scores. The global average on the Polity :llI index was -1-3in 1994; Africa in general was rated at 0, and the Middle East about -5 (Jaggers and Gurr

Political Cmztext

215

TABLE 3.2 Democracv Indexes for the OPEC Countries, 1970-1996 Cauut q PolikIII Freedo1rz House 1970-1 994 19% 1996 42 14 Algeria -7258 Ecuador 4.58 5 5 9 9 Gabon -8.59 Indonesia -7.00 13 13 Iran -7.08 13 13 Iraq -7.88 14 14 Kuwait -8.32 10 10 Libya -7.00 14 44 Nigeria -3.63 13 14 14 44 Saudi Arabia -10,00 Qatar NA 13 13 11 11 UAE -9.58 Venezuela 8.92 6 6 OPEC Average -5.26 11.46 11.38 WUIXCE: Polity 111 database; Kaplan 1996.

39%5:47). This shows that the level of autocracy is even higher among the OPEC members of these regio~~s than for the regions in ge~zeral. Several of the mnst important oil pmducers are monarcthical authoritarian states. Ruzan finds the political vulnerability of such regimes to be very high: In a brc~adseme . . . one might argue that the w h d e Zeitgeist. of the twentieth c e n t u ~has posed a poXitical threat to the legitimacy of monarchical rule. " f e shah of Iran was but the last in a long line of autocratic rulers to be swept away by mass-based political movements of various persuasions. Such events cause no puzzlement, although they may cause considerable surprise. The mystery i s how such anachrc>nisticforms of government manage to survive at all when the entire socio-pc~liticldenvironment of the times acts to corrode their legitimacy (Brxzan 1991:1 21)

The proposition that mmarchical ~ g i m e will s break d o m is supported by Riggs (1993:211;). fn his test of the fragility of mird World regimes, he finds that twelve of thirty-hnio monarchical regimes existhg in 1920 were still in power in 196,giving such regirnes a survival rate of 38 percent. Of the I11 constitutive regimes, 58, or 53 pexent, had survived until 1985." Among the twdve monarchical, regimes sumivjng as of 1985, six were oil-rich states (BArain, Brunei, Kuwait, Omm, Saudi Arabia, and the United Arab Emirates). Of Che monarchical regimes that lost power

Political COBtext

217

TABLE 3.3 Classificat.ic>nof the Regimes of OPEC Countries Regimes it2 Year of Xcgitn~?~ LBenr of h w ~ asr X~zdepen- Thnt Lost Regi~ze Monarchical

SingXe party

Kuwait Qatar Saudi Arabia UAE Algeria Iran

1961 1!97'1 1932

Iraq Libya Iran

1958 + 3 subsequent coups 1969 + 4 subsequent coup 1979 revolution

Ecuador Venezuela Nigeria

1966 -i- 5 subsequent coups

1971. 1965 l!97'9"

Presidentialist Parliamentary *Year of the Iranian revolution

during the same period, we find three oil-rich states: Libya, fray, and Iran.As T&le 3.3 suggests, several important oil-pmducimg countries art. monarchicd regines that have survived decades of instability among Third World ~ g i m ein s general. Oil plays an kportant role in sustainbg the regirnes oE these ccnantries. Autocratic reghes do not need to legitimize their decisions to the popuIatim in m y h d of election. Oil-rich autocratic regims do not need the support of the people in order to support tl-rernselves in economic terms. Rcnlicr states, those that own capital and derive all or most of their jncome from it, do not need the economic support of tbe population through taxation. In the case of oil-rich mcmarchies like Bahrain, Brunei, Kuwait, &an, Saudi Arabia, and the United Arab Emirates, the oil inc m e has fundamentaUy chianged the role of the state in the econorny*As was described in Chapter 2, the oil industries cJf these states were run by the international oil companies. When these oil industries were nationatized, the states took on a domhant mle in the national economies of the respective ofl-producing countries. T11e capital accumutation derived .from Ihe state-owned oil busincss by far exceeded the hconne fmm any other part of the economy. The state thus became a net provider of capital, and did not have to impose any form of taxation on the population. This points to a fundmental asped of state formation: the way rulers finmce the building of the state vlstitutions througfn taxation. The role of taxation in state formatilm was discussed by Schumpeter 0x8, 1954). Me s u m a r i z e s his description of the formation of Eriro-

218

Political COBtext

pean stdes by statjng that "without kandizl need the immediate cause lfor the creation of the modern state would have been absent, . . . Taxes not only helped to create the state. "They helped to form it. . . . Tax bill in hand, the state penetrated the private econamies and won increashg domi.nion over thenn"' (Schursrpetctr [l3181 1954:16-47). This created close litnks between the mmt,pdy of potver and the monopoly of taxation, but it also made the rulers depemdcnt on the citizelns in an economic sense. The citizen was regarded as an object of taxation: Atthough the precise politicat implicatic~nsof tax levying may vary according to the nature of the tax itself' in most cases the veration requires a brge degree of acceptance on the part of the population. . . . This establishes a link beween the ability to raise taxes and legitimacy, which is captured in the saying "no taxatian without representation.""(Luciani 1990:75)

In the ail-rich states this dependelncy does nat exist. 'The ruling elite derive their income dircctly by collecting rent from the oil activity on their soil, as described by Ji.tl Crystal: In the Gulf, the most important impact of oil and the rentier economy it produced was that it gave rulers direct access to external revenues, revenues generated outside the local economy Where once these revenues had to be squeezed from the population, through the merchants, who in turn exacted a political price, the rulers now received revenues independently, There are always moments when the state develops a high degree of autonomy from its social bases. . . . But oil-based states are unusuaf in that their higher degree of autonomy from other sclcial groupings is not the result of a momentary crisis, but part of a structurally. determined, ongoing grc3cess. This independence is almost uniquely peculiar to oil. Almost any other export-coffee, cottm-involves some accommodation between the rulers and the elite who control the workfs~rceand extract surplus revenues. Biii does not. The elites on whom the ruler Bqends are not local, but rather multinational oil companies. (Crystal lli39Q:G-7)

The political impficatims of the oil wealth fur the relationship between elite and society extend iurther, as oil "gave the regime =sources to develop new allies antong the national population though. distributhe pdicies" ((Crystal1"39:1(3).In several oil.-producing cwntries the state redistributes its oil income not to preserve justice or even out inequalities among societal groups, but to gain support and loyalty from ptentially challeng4ng groups in society (Noreng 1997). Maktabi (1993:170) argues that, "fidging from the ecmomic programs implemented. and the financial rewards distributed in post-war Kukvait, it seems that the rulership

Political COBtext

219

persisted buying the aegiance of tbc citizenry by distributi,ng econontic compensations to secure a legitimate basis for its rule." MMaktabi describes the way the al-Sabah regime pursued this strategy after the IraqKukvait War: "Immediately after the liberation, an amiri deeree announced that all personal bank loans were canceled.. Outstanding telcphcme, electricity m d water bills were ulled. . . . Afl employees in the public sector received back-dated salariesf"Maktabi l993:1'3"Q).Cash compensations of about $1,i"00 were given to all citizens that had end m d the war, and $34,500 to those f a d e s having sufifered losses during the war (Maktabi 1993:170)." Although it is reasmable to expect Ihe authorities to make some economic provisions for the population after a war, the extent of the aid in this case gives a clear impssion of a tme rentier state, reinforcing a "political cfier~telisticcontract" (Maktahi 1993:170)by economic mems.21 Using oil weaith to gain political support is a double-edged sword as not only the welfare but also the regime's political stability is dcpende~~t on the oil-derived incorne. It: may also stand in the way of the development of other instruments of h a n d k g the relationship between the elite and the society 'This is pointed out by al-Naqeeb (1990), referred to by Crystal (1994:269): "At first the state tried to depoliticize these [opposition] group&y buying them off. W e n these eft'crrts failed, ntlers increasingly resorted to terror, perhaps because owing to oil revenues, they %lad never before been forced to develop other, mom numced ways of dealing with opposition." Accordingly, the relationship betwen economic growth m d dernocratizatio~~ can very well be negative in oil-rich courztries, contrary to traditional assumptions (Rostow 1967; Hunthgton 1968). The tendency to use oil income to ga,in s ~ ~ p p of rrtm infiue~~ljal groups h the society opens the possibility for other groups to claim that the ~ g i m fosters e imequalities. Nort?ng (1993:7) points out that this creates a social-democraticfeature of the Islamic (undamentalist opposi.tion in Ihe Arab countries, since it is able to protest against the injustice of the rc-.gime%distribution of the oil wealth. According to Karawan (4%4:48), the Isiamic fundaxner1ta1ist.s "'have been successfui in gaining support a m g a yollnger generation of the urban poor, and among those segments of the middle class who were afflicted by the costs of econcrmic adjustment programs." M e n the oil income collapsed in 1986, it was a political as well as an economic challenge to the regimes in the Gulf countries. m e n they started to run budget def cits due to ~ d u c e dimcome, they soon reaiized that they bad to cut spending. This meant less money to distribute h order to gain political support for the regime. S~rensen(1996:377) q u e s that, incontrast to the European state-building process, the state formtion in the Thid World w s characterized by

220

Political COBtext

decolonization [giving] Ec>rmalsovereignty, or juriidicaf independence, t c ~ states that had very little in terms of substantial statczhood. The situatbn creates a peculiar security dilemma in the weak [Third World] states. . . . they are guaranteed juridical survival by the international sodety of states, no matter how deficient they are in substantial terms. . . . In earler days, state elite were compelled to create some substantial legitimacy of rule and thus gain support from their populatiuns to be able to weather perennial external threat, In the absence of substantial external threat, that is no longer necessary; state elites can go to any extreme in terms of violence against their peaple without paying the ultimate price: termination of the state.22

This means that p v e d s for strong internal democracy mwernents in the mird World are slim. Existence of political opposition. and regime change would not irnprove the legithacy of the regime. If one were to propose a generd conclusbn based on the empiricai studies referred above, it w~tUIdbe that there is no general trend. The m i d World states differ in terms of important factors influencing the regimes+otitical legitimacy. This is also tme of the states focused on in this book. 'They have one important aspect in common, their- oil revenue, but are highly different almg other variabfes. For instance, the tendency to buy political support is characteristic of weatthy authoritarian countries, such as the members of the c;cc.z"hmong the poorer count-ries, or societies w h a e wealth is mare equally djstributed, this strategy is not an option, This can have further impiicatims, as the fall in oil prices m y i n c ~ a s the e divergence between rich and poor oil prohcers. It m y also increase the potential threat of Islamic fundamentalists, as they can use the ecmomir situation to strengthen and broaden internal opposition. With the lack of nced b r both political support and hancial support, the oil-rentier state is self-supportive in times of high oil prices and thus in no need of the populatim for politicaf or economfc support. Furthermore, the population will to an increashg degree be dependent on the state, as most of the economic goods allocated are financed through the oil revenue generated by the state's position as o m e r of the oil resources. In economic terms the oil-rentier states are better off than autscracies in general, as they can compensate for lack of democratic legitimacy Zly selective allocation of economic welfare. When the oil revenue fell dramtically durjng the eigMies, this strateu was irnpossiblc to pursue, If the dassic hypothesis that discrepancy between the popuiatids expected need satisfaction anci actual need satisfaction creates potential for revolts, or at least gives ammunition to political opposition, one shodd. expect an increased instaility arnmg these regimes h tirnes of htw oil prices (Davies 3962). It further implies that autocratic oil-~ntier states shou.ld be more sta,t?lein times of high oil revenues, and less stable

Political CC??~ text

221

in times of low oil ~ v e l ~ u ethan s , arntocrades in general. 'This is because they are both economically and politically dependent on the oil revenue. The =duction in state oil revenues c~atec-ts pcri.tical &stability im addition to wonomic instability. Democratic oil producers w u l d not be politieaily jeopardized to the same degwe by lower oil income.24 It would lead too far astray from the topic of this book to go into any empiric& imvestigation of such hypotheses as those szlggested here. The findings do, however, have h p l i c a t h s for the argument in this study of the oil-producer cooperaticm, They point to the importance of bearing in mind that, for s o m of the world's most important oil producers, the oil income might be perceived as a matter of sel&p~servation against internal threats. T h i s does not, however, lead to any clear-cut argument regarding their cooperative behavior, As discussed in Chapter 7,actors fhding the oil income important might join collective efforts to incrclase or sustain prices, but might also compensate for falling prices with increased pmdwtim, the latter being a strategy that undcrmiaes the coiilective efforts of a producer cartel. What can be conchded is that the political importmce of oil income to the ~ g i m eof s these countries will tenli to make them place more importance on the oil industv and the international oil market &an they otherwise wnuld. They will thus follow whatever strategy they choose, cooperative or competitive, more eagerty and forcefully than regimes with other means to achicve poljtical legitimization. For them, politics and oil are even morc intertwined than for other types of ail producers, at; oil prices can be pexeived at; being a matter of regime exisknce. Taken together, the ext.ernal and internal threds discussed so far in this chapter paint a picture of rather conflictual ~lationshipsbetween Mid.dle East oil pmducers. The question arises as to what extcnt the oil prtlduction also has contributed to the emergellcc of cooperative pditical interests m o n g these comtries, Aspects dixussed in this section suggest a common interest among the rich against the poor, or ammg the Westernoriented sheikhs against the lslarnic ft~ndmentaists.'Thus, oil and politics go togethr h defining some common enemies and.hence a primary incentive for the estat3fishmmt of cooperative stmcturlts in the relations between some of the oil-producing Middle East countries. 'These political cooperatke aspects arc the topic of the next section. 3.6 Political Caoperation Between the Gulf States 11%February 1981, the Gulf monarchies-Saudi Arabia, Kuwait, Qatale, the United Arab Emirates, Bahrain, and Oman-formed the Gulf Cooperation Council (GCC), Both the external m d internal threats discussed above motiwated this cooperative effort. The establishmat of the GCC was triggered by the uprishg h Mecca in late 1 9 7 h n d the Iran-Iraq War. Later,

222

Political COBtext

the Iraqi attack on k w a i t identified yet another colnnon secttrity threat to the KC states. Let us start out with the externai thmats. Economically, the Iran-kaq War obviously was beneficial for all oil producers, as the oil price increased. Latel; the reduced production capacity of the belligerent states gave rcrom for increased production from other cowtries without decreasing the oil prjce. Politicaly, the other Culf countries found the war to be htheir interest. The new regime in Iran repmsented Islamic fundamentalism, which was perceived as a threat to the royal fanlily in Saudi Arabia and the a i r s in the other GuIf countries. The Culf countries provided Iraq with financial assistance (estimates vary between $25 and $65 billion) and sold so-called war-rczliftf oil on behalf of Iraq. The Gulf leaders perceived Iraq as the lesser of two evils, but wme "no less concerned over potential Iraqi hegemony in the Gulf region" (&chichim 1990:92). I-lowever, the Gulf countries were primarily concerned with preventhg a r spreading to their territories. In the first two years oZI the the ~ ~ fmm war, the GCC cowtries pursued a llieutral strategy in their official statements. Saudi Arabia a d Kuwait provided Iraq with fh~mcialsupport, while t-he United Arab Emirates tried to prcserve ""an open and friendly line of communication with Tehran" wechichian 1990:"3)).2" The neutral strategy was abandoned dufing 1982: "By November 1982.. . the council's attention was clearly focused on Iran's persistent intransigence in the war" wechichian 1990:97), Also, the rejection of mediation attempts by Iran led to the GCC countriesf official support of Iraq. 336s also meant that they kept a low p d i l e when it was revealed that Iraq used chexnical weapons (see section 3.3). At the b e g d i n g of trhe Irm-Iraq War, the KC counbies were militarily weak. This changed during the war, as the us nrtilgary guarmty to these countries was strengthened, and the KC countries' own maitaq power was enhanced. Zn addition, the KC as a security ownizatim had developed from notlrting into a militav aj)ia,nce with a c o m o n defense strategy weapons programs, and joint miljtar). exercises. ""For the first time in this century fnrces from all six states participated in coverative military activities aimed at defendkg t k i r territories. . . . Mescas the Gwtf War had vlitidly posed a threat to the ccc states, the end result was a strong=, more unified military structure" (Kechickian 1991):108). The external threat hposed by Xrm stretngthened the GCC'S kter~nalpolitical cohesian. The Iraqi attark on Kuwait in August 1990 demonstrated the habiliv of the smaller Gulf states to provide srtfficient military resistance against the larger colmt-ries in the region. The them-njne-year-oiid GCC coaperation bad. not changed this fact, The consequences of the Iraq-Kuwait W r were that the GCC cooperative efforts wert. pushed further: "'?"he second Guli crisis opetned a new era inthe GCC interstate relations. The small GCC

Political COBtext

223

countries, with the exception of Oman, became moro hclined to follow Saudi foreign policy. The ccc states have accepted the Saudi perspective on Gulf securityff(al-Alkkn 1994:89). mere are, however, political differences between the Guld comtries. Some of these, as long as they rclrnain unresolved, will represent a hindrance to the extension of cooperation between the Gulf states." To some commentators, the cooperation thus looks like one based on common fear rather than genuisle c o m m n interests: "The onIy bond holding the Sheikhs together was that their fears of Iran and :Iraq were w a t e r than their dislike of each other" "eikal1993:125). .Also, the perception of internal threats contributed to the formation of the GCC. 7'he members are all sheikhdoms, with a political front against countries With "revQjUtimalyl' leaderships, or socialist. or Masserite ideologies, The pan-Arabic idea was at this point dead: ""Divisions in the Arab world were notl-ting new, but signs of deeper frat;mentation emerged in 1980, when the shejkdoms formed wfiat mounted to a club for the rich, The Gulf Cooperation Council." ((Feikal1993:115).This interpretation implies that regionai cooperation stands in conflict with broader cooperation. 'This is not necessarily true, but in this case the ccc underlines deeper societal differences between the Arab countries, between the oil-rich, and mostly small, scarcely populated countries, and the poorer and heavily populated countries" The other aspect of the imtemal threats is the threat represented by Islamic fundamentalism, fostered by the regime in Tehran. The Islamic leaders do not need tanfts and troops to r e p ~ s e n at threat to the Gulf regimes, The possibility of internal fundamentatist groups c o n d u c t 4 sabotage or terrorist attacks, or challenging the secular ~ g i m e fn s other ways, is a more likely development. The bornt? explctsiorl in. Riyadh on November 13, 1995, which killed five h e r i c m citizens, mderlined the seriousness of these threats. Afthough this was the first terrorist action of this kind h the khgdom, it "suggested an ominous escalatio~~ of the conflict between the Saudi royal family and its Islamic opposition,"" The two external military threats perceived by the GCC countries-the threat of Iraq and the threat of Iran-are thus lhked to the two political and socictal th.reats: the threat of Arab radicalism, represented primarily by Nasser and to some extent Saddam Hussein, and the threat of Islamic fuxzdamentalis~~, kspired and supported by the r e g h e in Iran. Thus, the external and internal threats are to s o m extent linked." This pojnts to the complex interstate relatior-tsin the Middle East in gemral. According to Halliday (1.991:230),the Middle East: interstate campetition is more complex than anywhere else in the world because it involves not just a bipcrlar conflict . . . but- a set of intcerlc>ckingcon-

224

Political COBtext

fiicts-Arab-Israel, Iran-Iraq, Iraq-Syria, Iraq-Saudi Arabia, Saudi-Vemeni. . . . The history of formal and informal treaty organizations in the region is ane af tzreakness, incompleteness and failure: by the Arab League in the 1940s, by the Baghdad I""~c~-cENTc~ (Central Treaty Organization) in the 1950s and 1 9 6 0 ~ by~ the Twin 13ilfar%ayproach in the Gulf in the 1970s, by the Gulf Cooperation Council and the Arab Ccx~perationCouncil.

The development during the period discussed in this study is thus one of deterioratictn of the political coherence among the Arab oil producers. The oil hvealth has not lead to increased politic4 inter- or intrastate unity among this important subgrwp of oil producers, The question, then, remains: To what erttent do the political issues discussed above influence the oil policy of the Gulf st.ates? First, the securi-ty irtterests of the Gulf regixnes have been given more weight than the economic intermts. During intensified conflicts in the region, the economic expetnditure on arms, payment for foreign forces, and so on have been carried out regardless of economic costs, Oil has also been used to ensure alliances, such as when Saudi Arabia and Kuwait s d d oil, k m their neutral sector inorder to support the Iraqi war aga,inst Iran. The economic interests and the securjty interests have pulled. in the same direction. M e n this is the case, it is harder to establish the relative importance of the different .factors. Second, the warfare in the region bas destroyed production capacity in the beliigerent countries. Figure 3.4 compares c0mbint.d Irmim, Iraqi, and Ku'cvajti oil production with that of Saudi Arabia from 1978 to 1994The figure clearly illustrates how the warfare has reduced output from the affected countries. The Iranian production in spring of 1978 of a h o s t 6 mbd has never since been achieved, although the potential oil reserves are present. h spring of 1980, Iraq produced approxirnateIy 3.5 mbd; all though the production in the first half clf 1990 a p p c h e d this level, fie subsequent attaek cm Ku'cvajt.reduced Iraqi production to about. 0,s &d. The Kuwaiti loss of production has been more short-term. Cmtrary to what was predicted after the retreathg Eraqi forces had set a suhstantial part of the Kuwaiti oil fields on :ire, the comtry regained its produdion capacity during 1992, and in 1993was back to the prewar level, Third, tbr wars have made the internal OPEC negotiations over production quotas easier, as other producers have been unabte to prodtrce at- full capacity due to destruction of belligerent states"rod.uctim capaciv. C by FourG"~, the politiahantagmism cxated between the C ~ I ~ Emernbers the wartare seems to have had on:ly short-term effects orn the internal negotiations inside OPEC. Both Iran and Iraq, and later Iraq m d Kuwait, have accepted the presence of each other's delegations at the or9ecmeetings Apart from some incidents of pohtical confrontation, "business as

226

Political COBtext

usual" has been the norm at OPEC meetings conduded during military conflicts between members, In a h g e r - t e r n perspective the political cooperation between Arab countries that was prominat in the fifties and sixties, and particularly inspil-ed by Nasser, gave political support to the oil cooperation and the confrontation with the international oil compa"i"s W ~ e nthis potitical unity fell apart, so did the politicaf basis for the economic and oil-~lated cooperation,

I'he discussion in this &apt= has focused on the prominent role cJf security interests. This is in line with traditional understancfing in international relations of the importance of self-preservation in any state leadership. 'l'he analysis has been based both m trixditional aspects of security, namely external threats against the intrusion of forejgn military forces (sections 3.3 and.3.49, and on the possible thrcrats to state leadership from internal forces (section 3.5). The filzdings in this chapter s~~ggest that, in additio~zto the more direct efiects of political i n t e ~ s t son the productiosl capacity of belligerent oil producers, warfare has also changed their i n t e ~ s tin s participation in cooperathe eff0rt.s m o n g the producers. If the explanation of this is purety poIitical, it will have to be distinguisE7ed from the exgImation that wars have a tendency to cost mcmey, puttislg the country more in need of income, and thus making it less able or willing to reduce its production. The security argumnt implim that states support or counter producer cooperatimfor securitlq msons. It is highly problemalric to reach a defh~ite cmcltrsion about why a state leader did what he or she did in a particular instance. The discussion above does, bowever, suggest that at least during the Irm-Iraq War and the hay-Kuwait War, several Gulf countries allowed security interests to govern their oil-market behavior. Indications of this are the ~ c c ' attcmpt s to lower oil prices in order to hurt Iran, and its strategy to sell oif on behalf of Iraq in order to influence the course of the war, Also, the internal threats to the regimes and the role of oil in supporting the regimes politically arc fundarnentaf to understandjng the oifmarket behavior of the Gulf oil producers. 'I'he role of oil in these societies might also be an addi,tional obstacle to possible democratization. This aspect, however, points to questions outside the scope of this study (see Noreng 1997). Tlze arnthoritarian chrackrislics of the regimes of m n y of the orFc states might also discourage democratic oil producers from cooperation 'f'he general point to be made here is that coilective e c m m i c with 0r"~c.B interest does not make states disregard the political obstacles to coopera-

Political COBtext

227

tion. 'This represe~ztsa difference between cooperation among states and cooperation among finxs, Firmsf solely dedicated. to profit, can more easily join prcrfitahle cot>peratiwearrangements. States, or state leaderships, with politicat aims such as =election in mind, will forgo profitable economic relations that might have negative political consequences. n r e a t s to vuh~erableauthoritarian regimes can make them behave according to neither the logic of consequence nor the logic o.f appmpriateness (see discussion in, section 4.1). Their behavior beclomes less predictable anci more contradicto~compared with theoretical models of states%hekavior. Ecolzontic modeJs, designed for understanding cooperation between firms, and ixrstit.utional models, desiped for understanding behavior in the Light of norms, miss this point when they are applied to the economic coclperatio~zbetween such states. The answer is a closer, more detailed, more historical, and idiosyncratic perspective-in other words, case studies. This is the approach pursucd in Chaptus 6 and 9, The political conflicts between OPEC members discussed in this chapter were assumed to have had a negatiw effect on the srEc cooperation. An organization with members in direct cmflict is likely to suffer significant setbacks in its cooperative eMorts. Hocvever, this is not the conclusion regarding ~ P E C Some . aspects of the ideological and.political conflict between Iran anci Iraq seeped into the O12EC negotiations, but the gezzeral picture is that the oil-related cooperatio~zin axle@has been only marginally affected by the politicd conflicts between its members. The organization was to some extent immune to spillover from the ntilitary confiicts. Notes I. Buzan himself argues that the cmcept shcjrald be defined in view of empirical cases: "Attempts at precise definition are much more suitable directed towards empirical cases where the particular factors in play can be identified."" (Buzan 1991:20) 2. The '""o~hc~dctxkonception of security cannot comprehend either the threats or state structures or regimes that d o not emerge from Other states, or the threats that states and regimes can pose to their own citizens and societies" "rause 4996:320). 3. Karnran Mofid in Tke hzdt;rj,mdenC,July 20,1988, cited in Hiro 1989:251. 4. ""Sudi Arabia's brilliant chief of operations was shunted aside for allowing the intercept and kill of the Iranian F-4 (Cordesman and Wagner 1990:214,n. 7). 5. "Riyadl~upgraded its F-15 inspectors and in 1987 placed an order with Britain for 72 Tornado fighter-bombers, thus @ving it an almost insurmountable edge in the area. Further, the Saudis not c)nXy welcomed four US Air Force AWACS sentry aircraft in 1980 but purchased five of their own, deploying the first in 1986. The A W C S in tandem with the F-15s clearly gave Saudi Arabia,

228

Political COBtext

and presumably the GCC, a credible deterrent capabiliv." WfI(eichchxan 4990:107-1 08) 6. Tke Observel; October 21,1990, cited in Freedman and Karsh 1993:45. 7. Baghdad Radic), July 48,1990, cited in Freedman and Karsh 1993:46;. 8. Baghdad Radia, July 18,1990, cited in Freedman and Karsh 1993:48. 9. The Arab League tried to mediate and chose Hc~sniMubarak as mediator, He went to Ba$dad on July 24, Saddam Hussein assured ~Vubarakthat the military forces along the boder were meant for intimidation, but asked Mubarak not to let the Kuwaitis know: "&other t-tusni, d o not let the Kuwaitis rest easy before the meeting" Freedman and Karsh 1993:50). When Mubarak arrived in Kuw-ait, Iraq announced that the Kuwait issue had not been discussed in Baghdad, Mubarak war; annoyed and told the Kuwaitis that Saddam did not intend to invade. "When asked what Sadclam wanted the Egyptian leader replied: 'It =ems he needs some money"""ibid.), 10. The Iraqi version is that they asked for $10 billion in compensation for the Rumaila oil field, but the Kuwaitis humiliated them by offering only $9 billion. The Kuwaiti version is that Iraq was not cmtent with $10 billion, but demanded surrender of some disputed territories, as well as oil-pumping rights inside Kuwait (ibid.:59-60). 14. The attack involved about 140,000 men and 1,800 tanks, The 46,000-strong Kuwaiti army war; not fully mc~bjlizedin order not to provoke Iraq, and was subsequently not able to resist the Iraqi attack, The emir, however; managed to acape to Saudi Arabia with most of the royal family. The Kuwaiti air force had made same attacks an the advancing Iraqi divisions, and sc~mesporadic fighting occurred, particularly around the royal palace. 12. Yemen and Cuba abstained. 13. "Sanctions will take time to have their full intended effect. W shall continue to review all options wit11 CIUT allies, but let it be clear, we will not let this aggression stand. Iraq will not be permitted tcr annex Kuwait. And that" not a threat, it's sot a boast, it's just the way it's going to be" @faBC Radio 4,8ctc>ber9, 4990, reprinted in Freedman and Karsh 1993:204). 14, An even stmnger argument would be that these factors were sufficient to explain the outbreak of the war. Regarding the magnitude of factors influencing the conRictual relations between Middle East countries, it seems unlikely that such aspects should be the owerriding cause of this particular cmflict. 15, As will be emphasized below, one should be cautious not to generalize about at1 oiX producers"po1icies on the basis of studies of the politics of Arabian oil-producing countries. Important member countries are located outside the Middle East and have characteristlies different frorn those of the Middle East oil pmducers. 1C;, The data are registered as IGPSR 6695, produced by Keith Jaggers, Department of Political Scienccr?,University of Colorado, and Ted Robert Gurr, Center for International Development and ConRict Managemrtnl; University of Maryland. 17. The coding includes =ores on five .\rariabXes:Competitiveness of political participation, from competitive f 4-3) to suppressed 4-21, Regulation of political participation, from factional/restricted (-1) to restricted (-2). Competitiveness of executive recruitment, frorn election (2) to selection (-2). Openness of executive

Political COBtext

229

recruitment, from election (+l) to closed (-1 ). Constraints on chief executive, from executive parity or subordination (4-4)to rmlimited power on executive (-3). See Jaggers and Gurr 1995:3'72. 18, The data are derived by adding the scclres on the Political Rights and Civil Liberties indexes. Each scale goes from 2 to 7, and the combined scale thus goes from 2 to 14, with 44 indicating the least free regime and 2 the most free. Measuring difkrent years and combining difkrent sources are of no consequence to the argument in this section, because the characteristics of political regimes seldom change dramatically, and the arguments in the following subsections are based only on a general picture of the political regimes in the oil-prc3ducing states. 19, Among these there tzrere substantial differences, as single-party regimes had a survival rate of 80 percent, parliamentary regimes 69 percent, and presidential regimes O percent. 20. The negative economic consequences of the war were reduced by the fact that the Kuwaiti regime derives a large proportion of its income from investmenh abroad. As pointed out in section 2.6, Kuwait has been one of the most active OPEC countries in using its crude oil income to invest in the oil marketing and refining indwtry in consuming countries. 21. "Political clientelistic contract" i s a phrase borrowed from Maktabl 4993:170. 22. The cmcept of weak states is taken from Buzan (1991:96-10'7) and can be summarized as states with low sociopolitical cohesion. 23. The relationship between the ruling elite and lslamic or other kinds of ogpclsition. varies fram country to country; it is hard to generalize about the implieations of it for all Arab OPEC members' oil policies. Some more detailed cornments on this issue are pro>vided in Chapter 6, which is a case study of Saudi Arabian oil policy. 24. There are tor, few cases to make empirical investigations of this hypothesis pclssible. Also, there have been only a few years of low csil prices since the independence of the Gulf states, reducing the po~ssiblesignificance of a tirne-series study. 25. This strategy was made somewhat difficult by the fact that Iran occupied parts of the United Arab Emirates (UAE) islands Abu Musa, Greater Tunb, and Lesser Tmb. The island c o d i c t escalated in 2992 when the ""Xranian authorities on Abu Musa suddenly demanded that the civilians entering the island to Xive on the Arab side have Iranian visa stamps in their passports, . . . Afthough Iran backed down on the visa issue, the dispute continued in the years after 1992 through repeated exchanges of accusations between Tehran and Abu Dhabi. . . . The Iranian statements were often very strong; for example, after the December 1992 GCC summit meeting, Tehran declared that if the UAE tried to take the ~S~ islands, it would have to "cross a sea of blood' to do so'" (The New York E I I Z Becember 28,1992, cited in Kugh 1996:62), 26. "The border dispute that had the most damaging effect on the GCC was the Saudi-Qatari problem. . . . the crisis culminated [in Abu at-Khufsj in an armed clash in August 1992, in which two Qataris and one Saudi were kilifled'"(a1-Alkim 1994:81). 27. 'RwEcanomisf.,November 48,1995.

230

Political COBtext

28. FTc>wever,the outcome of the Iraq-Kuwait VVair reduced the Iraqi threat against the Gulf states. It seems improbable that Iraq will be able to represent a military threat to the Gulf states in the fc~reseealalefuture, 29. Arve j c ~ h x n the , director of Statoil in 4986, argued against Nctmegian coveration with OPEC, not for economic reasons, but because ol characteristics of the regimes in several OPEC countrim, Aficnpuste~z,February 6,1986.

OPEC as an

International Organization

While the p ~ v i o u schapters discussed the structural conditions for oifproducer cooperation, the ailn of this chapter is to examine to what extent institutional factors have contributed to Changing the bchavior of key OPEC members h a more cooperative direction that wodd. otherwise have been fie case (see sec-tion 1.4). The instit-utiond factors are applicable only to OPEC wrnbers, contrary to the structural aspects discussed in Chapter 2, which were assumed to influence oil producers fn genertral. 'f'hus, the question to be m s w e r d is: What aspets of orEc as an internat-ional or.t;anization make the mernbers behave m r e cooperatkely than they wwld in the absence of the institution?l h the empirical analysis in section 4.3, six aspects of OPEC as an international organization will be discussed. These aspects are based m a combination of different approaches to the study of fnstitutions. These approaches are dealt wi& in sectio~~ 4.1. Section 4.2 briefly describes the organizational characteristics of OPEC, 4.1 lnstitutianal Approaches

Iktiolzal C7huice and 111sts'tuiciolzd Cmzstraitzfs In, economics the debate between institutional and neoclassical economists is rooted in a fundament& mtological question of what actually motivates individual behavior, and in a methodological pest-ion of how to study the behavior of man.%e basic problem for the new hstitutionaiists is the assumptim made in neoctassical economics about rational behavior:

232

CXTC as an

X~ztemcktionaEOrganization

The objection to the economic man was, or is, not SO much his materiaijsrn or egoism, but the atrtnt of his knowledge and his virtually unboundd rationality. It iis not such an extreme oversimplification that people wn~zttcr maximize, or aim at maximizing, so~7teflzir-~g-(th~>ugh of course there are di fficulties about defining the maximand). What is extremely oversimplificatory is to assume that people generally lznue the knuwlcdge to maximize, (Hutchinso)n 1984 1991:3&--39)

John Shart Mill made strong assumptions regarding the impo&t""w of the concept of competition: "only through the principle of competition has politiral economy any pretensions to the characteristics of a sciencef' (Mill 1909:242, cited in Hutchinson E191341 1991139). Hutchinson's interpretation of Ml,l suggests that it was only in a competitive si,tuation that "the assumption about full ratimaliw and correct expectations [coutd] be assumed to hold" "ill 1909:242, cited in Hutchinson 119841 1991:39). Furthermore, the state of perfcct competition, shovlld it c o m into existence, would not function far long, because no actor wouXd have any incentive to invest: the incentive to invest depends in part on the knowledge of a lixnited campetitive supply from other firms, or the establishment of a belief that others do not possess the informat.ic>nregarding the opportunity that is available to the investor. (Hodgsrrm 1991:286)

In a state of perfect competition, such opportunities do not emerge. This is not to make the obvious point that perfect cornpetiticln does not arise, but, more irnpmtantlk; "that it: would not be viable if it d i g (Hodgson 1991:187). So, the limited rationality; taken together with "the impassibility of perfect competition," makes room for supplement to the orthodox nedassical economic theory3 March and Olsen argue that actors often act on the basis of normative behavioral prtlscriptions: In a lagic of appmpriateness . . . behavic~r(beliefs as well as adions) are intentional but not wilXfuX, They involve fulfilling the obXigations of a role in a situation, and so of trying to determine the imperatirves of holding a position. Action stems from a conception of necessity, rather than preference. (March and OIsen 1989:162)

One part of this institutional eccrnomics is the study cJf transaction costs. Coase (1960:15) describes trmsacltion costs as follows: In order to carry out a market transaction it is necessary to discc~verwho it is that one wishes to deaf with, to inform pclopte that one wishes to deal and

on what terms, to cmduct negotiations leading up to a bargain, to draw u p the contract, to unde&ake the inspection needed to make sure that the terms of the contract are being t~bseived,and sc~on.

Transaction costs thus consist of search and information costs, bargaining and kcisfon costs, anci p d c i n g and enforcement costs (Dahlman 1979:148). Although Dahlman finds this division umecessarily elaborate, as all these costs are caused. by the lack of in.formation, the division shows the need to idmtify transaction costs in atl parts of a b q a i n h g process-a point @My relevant to the empirical discrlssim in sectinn 4.3. The transaction costs emerge as a result of bounded rationality and opport-unism. Robert Keohane points to the reduction of transaceon costs as one prominent reasoll for Che estabhshment of i,nternational organizatiom incorporated into international economic regimes (Keohme 3984:89-92), ias pointed out by Caporascr (1992:6,111), there are several ways by which international organizations reduce transaction casts: They provide administrative help, an ongoing forum in which representatives of different states can meet, and a set of rules and procedures for dealing with problems. In additian, regimes provide valuable informational services that facilitate mutual contact. They collect information, standardize cmceptual categories . . . codify rules and. practices, and attempt to increase the transparent y of both cooperatbe and defectkg moves.

This points directly to the aim of the elnpirical discussion in section 4.3. By studying how OPEC has reduced transaction costs along the I.in.es pofnted out by Capwitso, it is possible to detect the effect of the ortymization on the cooperative bchavior of the member states, In the sociological institutional approach the mechanism by which the actors make their choices is different from the assumed rationality of the ratiol~al,or ecmornic, model: Institutional cmceptions of action . . . differ from rational models in a mare fundamental way . . . Within an institutional framework, "choicc;3," if it can be called that, is based more on logic of appropriateness than on the logic of emsequence that underlies cmeeptic~nsof rational action. Institutionalized rules, duties, rights, and roles define acts as appropriate (normal, natural, right, good) or inappropriate (uncharacteristic, unnatural, wrong, bad), (March and Olsen 1996:251-252)

Both the rational and the sociological institutionalism pay attention to the problem of incomplete information and the implications for ratimality h effort to cambhe the institutional perspectivevresented above

234

CXTC as an

X~ztemcktionaEOrganization

shot~1dthus develop this possible "common groundf%etween economics and sociology (Wlliamson 1988:161).

Bounded Xatioaalify: The Commoa Crorcnd The most persuasive criticism of the rational choice model is based on the information required for actors to choose rationally among possible choices of action. Lack of information crcates uncertainty regarding the pcssible consequences of d i f f e ~ nchoices. t Herhert Simon strongly Rnoulzces rational action h such cases: "Wherever such uncertahties are present, an enhanced opporhnnity is provided for unconscious, or only partly conscious, drives and wishes tc:,intluence deliiberation. . . . where. evidence is weak and conficting, a rationatity principle has little independent prcldictive p w e r " Firnon 1985:302)."This lack of information makes bargaining rc_.ll?tricmssubject to manipulation by opponents. Arild Underdal outlines the effects on the bargaining prclcess: In an essentially co-operative context, Party would probably try to reduce Opponent" uncertainty . . . by submitting more extensive or more precise information, or by trying to strengthen 9ponent" confidence in the idorrnation he has. Also in the context cjf largely distributive bargaining, a Party cmfident in its c>wnperception of Opponent" suncertainty would prc~babty take measures to reduce that uncertainty, but it is likely to do so by trying to exploit the situatim, by transEt3rming it into one of partly inaccurate infc~rmation, so as to enhance its c3wn bargaining position. Tn general, a belief that Opponent is, or can feasibly be made, uncertain =erns to be a necessary condition for purpasively making efforts to mislead. (IJnderrBal 1983:187)

What is importmt here is that, regardless whether an actor seeks to guide or misguide another actor, the use of language will be important. The lack of certainty or the weahess of the actor's percepti.on might make him or her receptive to rhetorical argments: ""persuasion rests upon the ability tc:,organize the expc~rienceof those who are to be perst~aded"(Paine 19631:lO). The uncertahty not only becomes a basis for satisfaction rather than maxhization (Sirnon 1W6) but also hcreases the actor's mcertainty wgarding the sirtcerity of other actorsktterances. The costs of verifying whct%teranother actor spea:ks the tnlth or not is fundamentally dift-erent from gaining information about aspects of the suhstantive wmld, and such verilication might be unobtainable. Furthermore, in some situation one cannot be satisfied with merely probable information about th sinceriv of others.5 If an actor wants s w e assurance regard@ the truth of anothrr actorfs sincerity trhis cvil have to be sought withotlt: rclkrence to t-hat actor. Look-

ing at the actor" reputation for truthfuhess, or asking third parties about the same actor" truthfuhess h other situations, might be a way of improving the informatim about the other actm's ssinerity. One prtlblem is the costs of pursuing I-his strategy; another problem is the possit?ility of creating endfess regress, as actor A might try to increase the possibili'cqi that actor B will believe actor A speaks the truth, a l t h o w actor A does not. This skategy b y actor .h cvill further increase actor R's costs of h d ing the truth, and so on. Tbe relationship between espionage and cornterespicmage and cmtra-counterespionage illustrates this point. The irnplicatim of incomplete information for economic behavior is that "all economic exchange cannot be orgmized by contracts and market" (Furuboh and Rchter 1991:4).6 At least, the contractors are unilble to foresee all possible future events that mi$ht: change the cmcfitions on which the contract was established. Bounded ratimaliQ would not have created problems hatl all the actors always been trustworthy- The incentive to cheat one another is &ways p~semt;actors are, "'in Willimson's phrase %elf-seekingwith guil.e\ , , and . . . it is . . . very costly to distintjlxish opportunistic from ncrnoppcrrtunistic acton;, cx anfe" "(~urubotn m d Richter 1991:5). The actors can gaiin advantages by ""selective or distorted information disclosure or . . . self-disbelieved promises regarding future conduct" wUiarnson 1975:26). Wifliamson assumes that self-enforcing commitments that would mnul the second type of opportunism camot be secured (Williamson 197526).The basic assumption underlying this apprcraclx is as follows: Imtitutiunal theories . . . assume that action is based on incomplete and possibly biased ircEomaSictn. They fcreus the behavioral and social bases af ircEormatiom. They emphasize the ways in which institutions influence the perception and comtruction of the reality wMhin which action takes place, Institutional history shapes the definition af alternatives. Institutional capabilities and structures affect the flow of infc~rmatiun,the kinds of =arch undertaken, and the intmpretation made of the results, (March and O l g n 1994:EI)

Athough section 4.3 provid,es an empirical analysis of"the effect of institutional ac;pects on OPEC members%behavior, let me briefly spell out the ~ l e v a n c eof the theoretical discussion above for this particutar case. hcomplete informtion has been a problem for the OPEC cooperation, both regarding the actual prices different countries cbarged. and.regarding the actual production volumes placed on the market, Mso, factors outside the organization-particularly future demand, the non-ortic producers" behavior, and the intentional oil companies' s t o c b g policy-cause mcertairtties regarding the optimal response (see Chapter 2). 'This has created pmblcms for the ortic countries' abiiity to fine-tm~etheir marlcet op-

236

CXTC as an

X~ztemcktionaEOrganization

erations in order to achieve the desired price. As pohted out above, incomplete information need not create problems for the effectiveness of the cooperaticm if the parties invcrked can trust each othrr and jointly adapt to the chmged market situation and the new infarmation. Then, none of the OI~ECmembers wodd take advantage of the situation and let another member bear a hrger burden. I-lowever, this has not always been the case among OPEC m e ~ ~ b e rAs s . described in Chapter 3, some members have actually been at war with each other (Iran-Iraq and IraqKwait). The political regimes in the different countries are very different, includir~gconservative Arabic monarchies, dernacratic countries, Islamir states, and communist states. The members are spread over three continents, and include some of the poorest and some of the richest countries in the world. Differences, more than similarities, characterize the members of OPEC, excerpt fnr one point: They are all substantial net expcnters of oil and are all to a substantial degree dependent on income from their oil exports. This, however, is not likely to make them Sully confident of each other % tmstwosthiness. The conclusion of this section is that an actor" behaviclr c m be influenced by utilizing weahesses in the information and uncertainty of the truth of the language used h the interaction. Given uncertain i n f m a tion, the actor%preferences might be influenced also by the way options are prese~ntedand perceived.

Tn the situational model outlined in section 1.4, institutions are a part oi the social environment defbing the opportmity set of the actors. 'fhe irn-

portance of inst..itutionsin defining Ihe opportunity set of actors' behavior is perfectIy consistent with the assumption of rational actors mlcing rational choices inside the opporh;lnity set. Mlhen the ins.titutionalframework around the actors' decisions changes, lhey gain new infarmatiorn and form new beliefs that change the calculation of costs and benefits and subsequently make rationai actors change their behavior. 'I'he question then becomes: M a t aspect of the actors' choices are influenced by the institutions? Et could be the payoff from different choices, the calculation of costs and benefits comected with the different choices, the utility comected with different strategies, the interests, or the identity of the actors, The theory of rational choice would regard. the actorsYnterests or desires as givm at the individual lwel-'"the unmoved mowers, reflecting Hume's dictum that 'reasnn is, and ought only to be the dave of the (aster 1984a:4). X n this chapter it is argued passions' [Hurne 1"7"39:41qU that instihtional factors change actorsf interests. The fact that the theory of rational choice regards the actors' interests as given crcates no cont-ra-

diction with such an argument. The theory of ration& choice does not include interest formation, and thus it is compatible with other theories explaining the lormation of ac"crsr inte~sts,as long as they are consistent. It is thus possible to corslbi~nethe rational and sociologica) perspectives. Starting with the situational model presented in Chapter I, the institutional perspective of March and C)lsen offers two supplements. First, it develops the institutional constraints included in the specification ol the sifzaafion,As pointed out by Hovi and.Rasch (1996:74), institutions arc part of the social environment included in the situation model. On thfs point there is no djffe~nce between the perspectives. The second point is the argument that actors make their choices not on the basis of ratimlal caieulations but on the basis of an abidance of n o m s created by institutions-that it;, the logic of appropriateness, 'This szlpplement establishes mother mechanism for the actorskchoices within the opportunity set, I'he actors do not seek the best action giLpen desires, but rather the right action according to the norms established. 011 this point there is a suhstantial difference between the perspectives. The next questicm is to what extent and how these f u n d w n t a i theories are applied to kterna.t.ional institutions.

Keohane d e h e s institutions as '"persistent and connected sets of rules (format and infomai) that prcrscribe behavioral roles, constrain activity, and shape expectations" "(1989:M). He thezn distinguishes between three Iforms of international hstitrxtions: l. formal il-ttergovernme~,ntal or cross-nationd nongovernmer.ntal organizations, which are "purposive entities . . . capab.le of nonitoring activity and rt;acthg to it, and are deiiberatrely set up and designed by states" 2, international regimes, which are "i~zstitutionswith expIicit rules agreed upon by go\remmentsff 3. conventions, which are ""informal institutions, with implicit rules and mderstandhgs"

In this chapter the focus will be m ctr~c as a formal intergovernmental organization, It is obvious that informal rdes coexist with or within formal organizations. Fomal organizations n-tigkt thus be based on existing cmvf.ntions or regimes. In, the case of omc, it can be argtred that the formal organization was established prior to the development of informal cooperative ties between the founding members (see section 42). In this chapter the informal institutional aspczcts will be regarded as underlying

238

CXTC as an

X~ztemcktionaEOrganization

aspects of orl-;c,while the formal organization is the dominant institutional feature of the relationship beheen the oil producers, U~lderlying illformal aspects will be included without the introduction cJf the concepts of regimes or cmvezztians. 'This makes the analytical distinction between the concepts of organization and institution less relevant to the discussion of the role cJf QI~FCin the oil-producer cooperation conducted in this chwtez Chapter 8 provides a more detailed examinati.onof the relationship bemeen oil producers inside and outside WEC, Another reason for downplaying the discussion of diffe~ntapproaches to explaining international institutjons is trhat the focus of this chapter will be the effects of hstitutions on states9ehavior: international institutions are important for states' actions in part because they affect the incentives facing states, even if those states9tinndamental intel-ests are defhed autonomously. International institutions make it possible for states to take acticyns that would otherwiw be inconceivable. . . . They also affect the costs associated with alternatives that might have existed independently Institutions may also affect the understandings that leaders of states have of the roles they should play and their assumptions about ojthers" mutivations and perceived self-interests. "fat is, international institutions have constitutive as well as regulative aspects: they help defhe how interests are defined and how actions are interpreted. (Keohane 1983:5-6)

The study of international orgmizationshas been characterized by two separate traditions, an empirical traditim studyjng formal hternational organizations, and a more theoretical one studying the institutioslalizcd cooperatiun of states. There seems to have been tittle contact between those studying the kterrratio~zalinstitutions under the regime perspectives and those conducting studies of formal international orgmizations. Probably the most influcmtial book in the latter tradition is Robert W Cox and Harold K, Jacobson's Ttrc A~~crtonzy of t'r~fluence,published in 1973. This book hardly mentions the then-emerging tradition established by Keohane and Nye (1972). In his tathook m international organization, Jacobsm devotes one paragraph to the perspective developed by Keohane and Nye Oacobson 1984:380). "RObest Keohane argued for 'questionhg traditional ccmceptione; of internatilrnal organizatims as highly institutionaXized enlitics with explicitly developed formal structures"' (Keohane 1975:361, cited in Rochester 19%:794f. Two years Iater, together with foseph Nye, he argued that ' k e need to think of intemationai orgarriadions less as institutions than as clusters of intergovcrnmntitl m d transgovernmental networks associated with the formal institutions" (Keohane and Nye 1972240). Stokke (1997) shows how the interdependence school, t0gelhc.r with the integration thcory, formed the basis for

the international regime literature but does not hclude the international organization literature as such a foundation. As pointed out above, aithough this study deds with a formal organiPFC-aspects of the relationhip between the oPec men?bers and the role of the organizatim in the international oil market are ascribed to mechanfsms related to fnternational regimes. This calls for a cornbination of the insights generated in the two traditions, but first s o m f d m e n t a l aspects of the two scboois wili be presented. A fundamental problem in explaining thr implications of a certain institutionalizati,on (formal or inlornnal) is l.he countesfaclud question what would the actors%ehavior have been without the present institutions77 Answering tbis question might be easier in this study than elsewhere, as the actors hvolved would have had to relate to one mo&er in the marketplace anyway. The competjtive market d g b t serve as the n o m for evaluating the impact of instihtinnali~ation~ On the other hand, this does not accom~tfor the possibility that, withouflhe present institutions, some other k h d of hstitutionalization would have occurred, implying yet mother pattern of behavim by the actors. This kind of secondorder counterfactuality leads beyond my present imaginative capahilities, m d the evaluation of the effects of irtstitutimalization will suhsequmtly be made incremmtaily over t h e inskad of by comparing the institutional situation with s o m other inagined situation. Turning to the relations between the organization t k o r p and the study of formal international organizations, :Ness and Brechin f1988:245) state: ""Thegap between the study of international organizatims and the sociology of organizations is deep and persistent. . . . each appears to run its own course, largety uninformed by the other.'" Intermtional organizatims are futl-fledged organizations similar to the national organizations of ministries, firms, or special-interest groups. Some aspects of bureaucratic: theory might even he more rdevant in international organizations rhan in national ones (Claes 1,994). Ness and Brechin see a similarity between the sociol.ogists9focus on organizational environment and thc. study cJf international regimes: ""Defining and ider~tifyingvariance in environments has provm djfficult and jllusive. There is . . . a strikingly parallel development in conceptions with the m m recent regi772e perspective in hternational relationsf"tNess and Brechin 1988:249). Describing different theomtical approaches and the relationship between them is of no value unless it improves the malysis conducted. Obviously, the choice between the perspectives outlined above will lead to different foci in the study of owc as an institution. The explicit aim of this chapter is to analyze how the institutionalization of the cooperation between the O ~ F Cmernbers has contributed to changhg the

240

CXTC as an

X~ztemcktionaEOrganization

individual members' behavior and steering it in a more cooperative direction. This analytical focus implies a certain mix of the djfferent theoretical insit;hts gained in the various approaches presented. Shdies of international regirnes generaily have been open to such a varicty of fheoretical. apprmches.

The classic, and now referred to as the consensus definition of international regimes, was put forth by Stephen Krasner in a spceial issue of the journal. OrfetmfiomlOrganization in 1982: international regimes are defined as principles, norms, rufes, and decisionmaking procedures around which actor expectations converge in a given issue-area, Principles are beliefs of fact, causation, rectitude. Norms are standards of behavisr defined in terms of rights and obligations. Rules are specific prescriptions or postscriptions far action. Becisian-making procedures are prevailing practices for making and implementing collective choice. (Krasner 3982:186)

Several authors have tried to modify and clarify this ctefinition, giving M e r e n t interpretations and meanings to the concept international regimes. I-lelen Milner argues that "even strong proponents of the [regime] concept a b i t the difficulty in defining it" (1993:493), and that the term "regime""has lost its earlier charm. Some scholars have thus retracted to the language of institutions, or institutionalism (Milm r 1993, Keohane 1989); others use concepts like "policy coordinationf"Haas 1992:371), or ""gvernance system" "oung 1994). But as FXclsencXever, Mayer, and Rittherger state: "the substantive questivns that define the regime-ady tical research agenda-whe ther couched in terms of 'regimes,' 'institutions' or otherwise-still count among the major foci of International Relations scholarship in both Europe and North America (Hasenclever et al. 1997:1). Relating the concept of regimes to the perspective developed in the previous sections seems to be a matter of labels. International regimes is a subgroup of htesnational institutions, of which international organizations is another more formalized and bureaucratic subclass. If one starts from the r@gimetradition, one would say that regimes can be accompanied by organizations; if one starts from thrr international organization tracfition, one would tend to say that organizations can include regi~aes, Such statements need, not have different empirical implications, and thus both concepts and different theo~ticaltraditions can contribute to the mderstanding of OPEC.

Hasenclever et al. (1997) divide the theories of international regimes into t h e e schools: power-based, interest-based, and howleae-based theories.

* Power-based themies of international regifnes can be described as 'realist theories of cooperation.' Not only collfiict but atso cooperatio~~ is explained by power m d the distribution of capabilities among states. * 'l'he interest-based theories are the mainstream of regime theories. They "emphaize the role of international regimes in helping states to realize c o m o n in.terests"' (Hasenclever et al. 1997:4).The focus is on situations where the constellation of actors' interests is so that they can only achieve beneficial outcomes through institutionalized cooperation, * The knowledge-based theories stress ideas and k n d e d g e as explanatory variables. The focus is partly on hocv "causal and normative beliefs form perceptims of international pro:blems and thus demand for ~ g i m e s "(Hasaclever et al. 1997:f 37). Almost any theory of international cooperation can be included in these three broad categories. 'lhe number of different themticai topics covered by Hasenclever et al. (1997) clearly indicates the lack of demarcations of what is to be counted as theories of hternational regimes, The book covers political market failure, situation and problem structures, institutimal bargaining, hegemony, djstributionai conflict, relative gains, ideas, arguments, m d social identities, If we relate the substantive defhd tion of rt;gimes as the norm-based convergence of expectations that leads to the production of explicit rules (Stokke 1,997:27), this wide scope of theories of international reghes obviollsly encompasses more than this. Sirnilarty, all theories explahing the existence, persistence, and effects of substantive regims cannot possibly be counted as theories ol international regimes, Obviously tkese are other explanations for the existence, persistence, and effects of international instih;ltions than what we reasmably would include in the category "theories of international regimeseff8 We thus need to limit the category "theories of irtternational regimes" to explanations that include a nornative core. The second aspect of the cmcept of rt?gimes that needs clarif-ication is thus its underlying normat-ive core. Levy b u n g , and Ziirn (1995:273) express this when they write: "Given the basic thmst of regime analysis as a tool for understanding international cooperation and the role of norms in the pursuit of cooperation there is a need to go beyond merely routinizd or patterned bekavior. The principal claim of regime anatysis is that states may generate in identifiable issue areas that affect their behavior m d foster institnxtio~~s

242

CXTC as an

X~ztemcktionaEOrganization

cooperation, even if short-term hterests would dictate deviation." E the observed cooperation is explajned by patterns of complementary interests and underlying distribution of p w e r , regimes have no effect and thus, in such cases, theories of internac-ianalregirnes does not contribute to the explanation of cooperation. The discussion hthis chapter will primarily be related to the effects of regimes. It then seems important to distjnguish hokv the rclgixnes ixnprove conditions for cooperation and the "degree to which a regime ameliorates the problem that prompted its creation," to use a phrase by Levy, Vrtmg, and ZGm (1995:293,)-In the case of oil pmducer cooperation, this would be the difference between bow the rules make oil producers cooperate and the extent to which this rute-based cooperation leads to a higher oii prjce than what w o d d otherwjse have been the case, In the empirical hvestigation in section 4.3, I will search for institutional effects changing behavior according to both the logic of rationality and the logic of appropriateness. All the factors discrassed can be included in the hroad definition of theories of international re-lgimes given by Hasenclever et al. (1997). However, the discussion will show that the norms of cooperation inherclnt in the concept of i,niernatio1zalregimes are not necessary to explain the institutional role of ~ I ~ EmC the cooperative behavior of oil-producing states, Before turning to these specific empirical analytical topics, the next section brieCly describes Che organizational structure of OPEC, 4.2 The OrganizaCionalt Structure of OPEC

I-lovi (1992:230) defh~esan intmaticmai oqanization as ""a brmal, permnnent: set of institutional structures, by whi,ch a group of sovcrcign states seeks to regulate or solve certain problems comected with their mutual relationsf"my translation). OPEC is an international organization according to this clefhition. A t ~ a t krmally y estabIishing OPEC as a permanent intergovernmental o r g d z a t i m was published in September 3960, and after ratification by the sipatory countries it was ~gisteredat the United Nations in November 1962. OPEC was ofjicially recognized as an international. organization by the UN Economic and Social Council in June 196, and subsequently became a regular participant in meethgs of various UN bodies, including thc UN Conference on Trade and Development, and in other hternational fomrns (Evms 1990;145)), The founding members are Iran, Iray, Kuwait, Saudi Arabia, and Venezuela,~Subsequentmelnbers have been Qatar (January 1961), Indonesia, Libya (fune 1962), Abu Dhabi (November 1.9672,10 Algeria (July 3969), Nigefiia (July 3970, Ecuador (associate member June 1973; full member November 1"33), and Gabon (associate member November

19773; full melnber June 1975). Ecuador abandoned its membership in 1992, m d Gabon abandoned its membership effectively from 1995. The organization had its headquarters in Geneva from 1961 to August 19G. Since thezz the headquarters have been located in Vienna. The principal aim of O13Ec is '"coordmation and unification of the petroleum policies of member countries and the determination of the best means for safeguarding their interests, indi\ridually and collectively" ~ interestingly enough, considera(Artide 2 of the Statutes, o 1 31990:32). tion of the interests of the other groupmf the trilateral oligopoly, djscussed in Chapter 2, is taken into the Statutes: "Due regard shall be given at all tinnes to the interests of the producing nations and to the necessity of securing a skady income to the prtducing countries; an efficient, economic and reg~~lar supply of petroleum to consuming nations; and a fair return on their capikal to those investilrg in t:he petroleum industryf"Artide 2 of the Statutes, OI'EC 1990:32). Howewer, individual cojluSim wjth these groups by member states is prohibited: "If, as a result of the application of any decision of the Orgmkation, sanctions are employed, directly or indirectv, by any interested company or companies against one or more? melnber countries, no other member s h d accept any offer of a benefjcial treatment, whether in the fom of an increase in oil exports or in an impmvement in prices, which may be made to it by such interested company or companies with the intention of discouraging the applkati.on of the decision of the Orgmizatian"' (Article 4 of the Statutes, OPEC 1990:32). The aiteria for wrnbership are defined according to the role of petroleum in the country in question: " h y other country with a substantial net export of crude petroleum, wbich has fundamentally sirniar i n t e ~ s t s to those of member countries, m y become a fuil mcmber of the Qrganizatiosl, if accepted by a majority of three-fourths of full memibers, including the concurrent vote of all famder members" "rticlo 7 of the Stabtes, OPEC 1998:33).11 OPEC has three organs: the Conference, the Board. of Governors, and the Srcretariat. 'The Conference is ~ Y E C ' Ssupreme authoriq It consistls of defegations representing the men7ber countries, each of whieh shodd be represented at all cmferences, Howeve&a quorum of three-quarters oi member countries is necessary to hold a conference. Each full member country has one vote. Ail decisions of the Conkrence, other than on procedural matters, require the unanixnous agreement of the full members ~ ~ (Article I1 of the Statutes, 0 1 ' 1990:34). The Conference meets twice a year. However, an extraordinary meeth g of the Conference may be convened at the request of a member country or by the secretary general, after consultatim with the president: and approval by a simple majority of the member colmtries. The

244

CXTC as an

X~ztemcktionaEOrganization

Conference elects a president at the beghning of its meeting*Tfne president holds office for the duration of the meeting of the Conference, and ~ t a i n the s title until the next meeting. As OPEC% supreme authority, the Confe~nceformulates the policy of the organization and decides on all substantial matters: "All matters that are not expressly assigned to other c~rgansof the Organization shall fall within the competence of the Conference" "rticle 16 of the Statutes, OPEC 1990:35). The Board of Governors is composed of governors nominated by the member countries and confirmed by the Conference. The Board of Governors directs the management of o p e c affairs and the imflernentation of the decisions of the Conference; in fact, it runs the organization between the conferences. The Board of Governors-ale leaves little authority s r autonomy to the Secretariat: 'The Secretariat shall carry out the executive iunctinns of the Organization in accordance with the provisions of this Statute under the direction of the Board of Governors, The Secretariat of the Organization shall consist of the Secretary Gelzeral, the Deputy Secretary General and. such staff as may be required. It shall function at the headquarters of the Orgmization" (kticles 25, 26, OPEC 1990:37). The persons who have served as secretaries-general of OPEC are listed in Table 4.1. in exterThe secretary-gemrd is to some erctent able to r e p ~ s mowc t nal relations: ""The Secretary General shall be the legally aut.lnorized representative of the Organization. The Secretary General. shall be the chief officr;r of the Scretariat, and in that capacity shall have the authoriw to direct the affairs of the Organization, in accordance wjth directions of the Board of Governors (Article 27 of the Statutes, OPEC 1990:37),12 In addition to the OPEC Confemnce (twice a year), some important ministerial committees have been established (see Table 4.2). These are constituted of members of the Conference, and =port directly to f d o w members of the Conference. The committees are important instruments in the lievebgrnent of OPEC~Spolicies.

There is little literature applyhg institutional theory to the study of OPEC, In fact, even authors discussing O t 3 ~ c % rrtlle in the wodd economy tend to dokmflay the autonomy of the organization in relatio~zto the interests of the member countries: Tt must be recognized that the character and direction of OPEC is determined not by what QPEC has done or intended to d o but failed to do, but by what each member government has done or proposed to do. . . . The history of OPEC is replete both with cases in which OPEC actions were inspired by cieci-

TABLE 41 Scretaries-Ceneriiii of OPEC Dr. Fuad Rouhani Iran AbQuZRahman Al- Bazzaz Iraq Kuwait Ashraf T. Lutfi Mc~hamadSaleh Joufidar Saudi Arabia Venezuela Francisco R. Parra Dr. EIrich Sanger Indonesia Ornar El Badri Libya Abu Dhabi Dr. Nadirrn Pachachi Dr. Abderrahman Khene Algeria Chief Meshach 0.Feyide Nigeria AZi Mc3hammed Jaidah Qatar Rene G. Brtiz Ecuador Dr. Marc S. Nan Nguema Gabcm Dr. Fadhit 5, At-Chalabi Iraq Indonesia Dr. Subroto Abdalla Satem El Badri Libya Dr. RiIwanu Lukrnan Nigeria WL'I~CE:Evans 1496:l50; OPEC Bulletiltz, January 1995,

Jan. 1%1-April15264 May 4964-April 1965 May 1965-Dec. 1966 Jan.-Dec. 1967 Jan.-Dec. 1968 Jan.-Dec. 1969 Jan.-Dec. 1970 Jan. 1971-Dec. 1972 Jan. 1973-Dec. 4974 Jan. 1975-Dec. 1976 Jan. 1977-Dec. 497% Jan. 1979-June 1981 July 1981-June 4983 July 1985June 1988 July 1988-June 1994 July-Dec. 1994 J a n u a 1995~

sions at~ltladytaken by individual members, as well as by instances in which individual members opted tcr disregard decisions made by C ~ P E C., . . The moment the perception exists that an individual country" interests and goals are not served by art QPEC decision . . . the interests of that country can be expected to supersede those of the organization" ceomman objectives. (Afnasrawi 1985:3)

In this chapter it will be argued that there are aspects to the history of that contradict this statement. 'The followhg sections provide an empirical account of the pl.esence of six instlitutional factors. The first aspect is the role of the organization in proVidillg an arena fos exchange of information. 'This aspect is grounded in the ratinnal, choice tradition of t ~ a t i n gstates5interests as given. The role of the organization is to influence the cost-benefit calculation of rational actors by being an arena where the members can exchange illformation and positions and search for new solutions. This aspect only suggests that, without a place where such exchange can take place, co0per;lticm becomes harder to establ.sh. In fact, wilhout any arena for communication, cooperation is impossible. The second avect is the role of the organization in prowidhg information &out the mernbers"ositions m d the oil market h general. This asOPEC

246

CXTC as an

X~ztemcktionaEOrganization

TABLE 4.2 Key Ministerial Committees of the OPEC Conference Ministerhl Cc)m?lzil-l.ec F o e Members Ministerial Committee on May 29178 Algeria, Iran, Iraq, Kuwait, Saudi long-term strategy Arabia, Venezuela. Ministerial Committee to Mar, 1982 To December 1984: Algeria, monltc3r the oil market Indonesia, UAE, Venezuela. From Jan. 4985: Algeria, Ecuador, Iran, Iraq, Libya, UAE. Ministerial Comrnittee on Oct. 1984 Initial members: Libya, Saudi price differentials Arabia, UAE. Additional members in December 4484: Algeria, Kuwait, Nigeria, Qatar. Ministerial Executive Council Dec. 1984 Indc~nesia,Nigeria, Saudi Arabia, UAE, Venezuela, Open to other on implementation of members. pricing and prc?duction agreement Ministerial Committee on June 198"7lgeria, Indonesia, Nigeria, Saudi price evolution Arabia, Venezuela. ("Committee of Five"") Ministerial Committee to Nigeria, Venezuela. June 198"i"ndc1nesia, motivate members to comply with the 3986 agreement on quotas (""Committee of Three") Nov. 1988 Algeria, Indonesia, Iran, Iraq, Ministerial Monitoring Committee Kuwait, Nigeria, Saudi Arabia, Venezuela. Open to all members and split into two subcommittees in July 1990. ?ir?t;~i:cr:: Evans 1%0:152-153.

pect also is grounded in the ratlional choice traclitim of t ~ a t i n g states' interests as given. The focus is m the r d e of the secretariat in prwiding informatio~~ about solutions and other actors' behavior. The importmce sl: information is connected with the consequences of uncertainty in the relationship between states: "if actors behave purpo&vely given the information available to them, perception-the information that actors possess about others-cm be a critical determinant of behavior" "teisr 3990:55). The mom informaticm the organizatim can proVide, the less ljkely are misperception and misunderstanding among the meJnber states. It is, however, an eznpirical question wfiether this leads to more or less cooperation. Information can rczveal conflictuai ir^lterests as we11 as cooperative ones.

The third aspect is the role of the orgmizatio~~ increatjng decisio ing rules that affect the subseqtlent outcome of the ox>~r: bargaining, Keohane argues the importance of rules as such: "short-rur.1self-interest is afkcted by constra.ints ixnposed on policy choices by agreed-upon rules; long-mn conceptions of self-intercst may be reshaped as a result, in part, of practices engaged in over a p"xtiod of time" (&ohme 19911:737). The decisionmaking rules are important as they imply that the dedsions, to some extent, are the result of the shared interests of member comtries, This is an essential k a m e of cooperation. The rules regutating the dmisionn?aLng in the organization xnight also he important for the form and substantial contents of the cooperative decisions reached by its members. The fourth aspect is the role of the orgmization ir-t monitoring agreements and verifying accusations rcgardhg situations where clluntries break agreements. The Fnrportance of m d t o r i n g and verificatiosl, which are also rooted in the economic instibtionalism, is based on the study of f o r d organizations and their rolc regarding state beha\lior (see Karns and Mingst 1990).Monitoring the actorsfb&avior presupposes some adrninistrative capacity. It is furthermore a necessary condition for the solidjty of the t?$reements: "institutionnl arcitngements affect . . . the jbility of governmentr; to monitor otherskompIia.nce and, to implement their w n comrrtitmclmts-hem their ahility to make credibte commitments in the first place" weohme 1989~2).The assumption is thus that the more verification there is, the more likely are states' conmihnents to cooperative agreements. The fjifthaspect is the role of the organization in framhg decisions and establishing shared perception among the ~ P E Cmembers. This point is stressed by Axelrod and Keohane (4986:X7): "The sipificance of perception, including belie& m d cog~nition,will come as no surprise to students of international politics. Yet it is wmth pointkg out once again that decision making in ambiguous settings is heavily innuenced by the way in which the actors think about their problem.'"is wodd mean that the way the bargaining problem is p ~ s e n t e dto the OI~ECmembers influences their cooperative behawior. Two assumptions will be discussed. First, if the bargaining problem is presented as an internal OPEC problem, the m e d e r s w o d d be more likely to c~mplyIcompared with a situntion where the bargaining probtem is p s e n t e d as an oil-producer problem in general, Secand, the more uniform the OPEC members' perceptions are, the more likely are their contributions to cooperative efforts. The h a l aspect is the role of the organizatim in establishing a collective identity m o n g OPEC metnbers. Alexander Wendt has claimed that international institutions also may perform a role in the formation of interests and collective identity amctng states (Wendt 1994). He seeks to study situatio~~s where statesf hterests are not given but are endogenous

248

CXTC as an

X~ztemcktionaEOrganization

to interaction: "through interaction, states might form collective identities and interests, redefining the terms of Olson" problem altogether" (Wendt 1994:3M). Wendt considers it an open empiricaf question when states' interests really are exogenous and can be adequately analyzed thruugh a rationalist approach and when this approach becomes hadequate due to a cltynmic relationship between statcssf interests and action m d the institutional flarmation of interests and identities.'"

A funbamental function of an intctmational organization is to provide a meeting place for the mernbers, a setting for members to exchange views and opi,nions on ksues rclated to the aims of the orgmization. Furthermore, arenas ""regulak the access of actors to problems and the access of problems to decision games. Moeover, they spec@ the official purpose as well as the rdes, locathn, and tinti,ng of the game" (Underdal 1994b:110--111). The arena is essential fur the exchange of opinions and information. In the case of OPEC, the meetings of the Conference are the most important arena .for exchange of views and opinions. The importance of WEC as an arena is thus linked to the frequency oi the meetings of rep~sentativesof the member countrks. 7'heConfermce convened 100 tjmes in thr! period from the first: meeting in September 1960 until June 1996. The oil ministers, which usually represent the m m ber countries, have met, on average, every fourth or fifth montb. The meetings usually last for two days, In addition are the meetings of the Board of Governors and those of the different ministerial committees listed in Table 4.2. This must be regarded as a fairly high frevency of meetings between representatives of the governments of lnetnber countries of an internaticmal organization, The participants c m thus be expected to be quite familiar with each other and the ~spectivecountrriesf positions, strategies, and so on. If nothing else, this howledge can be based on the behavior (statements, voting, etc.) of the same countries at the p ~ v i o u sConferclme, which cclnwened, at the most, six months earlier. In addition, a hjgh frequency of infasrnal contact is likely to folfow from the meetings of the Conference. The oil polijcies of individual countries are unlikely to change dramatically over the short period between the cderenees, unless dramatic- evmts take place. This is important in particular for the quota agreements (see Table 7.6). Agreements on individual quotas are made for an expticit future period, usually until the next Cont'erence. The possibility of- renegotialing one's quota reduces the long-term implications of these decisims for the countrieskil policies. This should make the countries more cooperative than they othercvise would be, as it is possible fos them to c o m back and argue fos a higher

quota ina few months' time. For Che negotiat-ims as such, this means that several aspects of the quota agreements do not have to be dealt with every time. In cases where previous quotas have not been rolled over into the next period (i.e., until the next Conk?rence), a marginal change for all members or individual changes for one or two mmbers have been the usual way of changing the level or distributim of quotas. With other aspects held cmstant, this has reduced transaction casts (see section 4.1). The existing quotas have been a startirtg point for the negotiations at the conferences, and thus several aspects of the b q a h i n g between the OPEC members are not discussed at every meeting. 'The mdtilateral approach also reduces transaction costs in the setting of quotas, Setting quotas bilaterally ammg the OPEC members would be an impossible v o j e c t . This leads to the aspect of permanency Robert Axelrod (1984:ZQ)has shown that if the actors are to meet an hfinite or u n h o w n number of tirtles, they will have the incentive to choose cooperatim.l"terated games provide "multiple opportunities for rctdiation" wnderdal 1994b:112). Should one actor give concessions at t, on the condition that others do the same, it could always change this strategy at t,. Every member knows that the quota, distribution could become a topic only six months iates, Toe t h e r with reliable information about individual membess"ductim, this provides crucial instruments for subsegumt sanctioning of quota vialation..These aspects are fully developed in Chapters 5 to 7.

FIovi (1992:23&238) distinguihes between three different forms of information: factual information about the characteristics of the member countries, information about their behavior, m d informatio~~ that elucidates causal connections, that is, the potential effects of the organization" decisions. The orEc Secretariat is an irnportmt provi,der of all these kinds of iniormation. The reseasch department of the Secretariat handles these functions, It has the role of pmviding the mernbers with informaticm about the internat-id oi: market. l-he reports have also offered informtion about the differcllnt member countries, particularly their oil-pnchg and production behavior. 7'he Secretariat pmvides the reprmentatives at the Conferences and in the differe~~t mhisterial cmmittees with informatio~~ about developments in the international oil market." Furthemore, as OPEC is one of the major actors influencing the international oil market, it has been necessary to study the likely effects ol the difierent strategy options available to its members. Information about merrrbers"ositions and behavior forms an important part of the mmitoring functions dealt with below

250

CXTC as an

X~ztemcktionaEOrganization

Although the Statutes provide for an independent role of the secretarygcrneral, and thereby the Secwtariat,'?its effectiveness has depended on the quality of its staff rather than any particular structure of the orgmization"" (Skeet 1988:237). In 1963, the first secretary-general, Dr. Fuad Rouhani, actually negotiated with the international ail, companies for increased royalties on behalf of the member countries. 'fhe negoliations, which wcre substantially delayed by Ihe companies' tactics, e ~ ~ d with ed a split orfymization,as some countrits signed a g ~ e m e n t wi.th s the companies and some did not. The issue was excluded from the Conference's agmda, and the roie of the secretary-general was subsepently reduced. Skeet (1988:237) concludes that the removal of Rouhani in 1963meant an "implicit admfssion . . . tbat OPEC and its secretary-generai would not act, as had been visualized at its creation, as an operating arm of its members." Some of the subsequent secretaries gmeral have been able to create sorne roorn to manewer through their capacities as politicians or oil expcrts, but any ixnportmt role as a mediator between members, or as a negotiator with other actors in the market or a spokesperson to the ixrternational comn-tunity, has either been handled by a key member of the Conference or bee13 clearly mmdated to the secretary general by the Conierence. The Secretariat has no independent role beyond providing ixrformation and carrying out pmcedural tasks in connection with the conferences, 'This lack of any permanmt, powrl"u1 autonomous body weakens the influence of the organization m the behavior of its mernbess,

Decisioa w~akingR LLles In all international organizations, &cisittnma.king rules are irnpmtmt: The decision rule is clearly an important determinant of the capacity to aggregate diverging preferences; other things being equal, aggregatim capacity reaches ib maximum in strictly hierarchical structures, and is at its minimum in systems requiring agreement (unanimity). . . . 13roceduraX arrangements may differ . . . with regard to differentiation into subpracesses (committe works vs. plenary sessitms), and the amount:of discretion vested in committee or conference chairs in, i~zterall'n, draftkg proposals (""negatiating texts"). (IJnderdal 4994b:Ill)

The decisionmakiq rule of o m c is described in Article 13. of the Statutes: "Each Full Member Country shall have one vote. AU. decisions of the Confe~stcle,other than proceduraf matters, shall requise the unanimous agreement of aXI Full Members" ((OPEC 1990:M). Thus, it follows from the vatation from Underdal above that the aggregation capacity is at its m i n i a m . According to the Statutes, the individual state has a veto.

TABLE 4.3 Organizations and Stafe Power 13lri:oerItzside f he Organh a t ion Pouter Outside the Orgaptizatiorz Large SmmIE Large indifferent reject. organizational decisions Small prefer organizational indifferent decisions

Obviously international orgmizations where the individual memkrs can veto decisions are less autonom~lt~s thm orgmizations where hdividual members face the possibiliw of losing a vote. However, the OPEC conferences have had a tendency to work twarci consensus anli thus avoid voting. As a consequence of this consensms strategy, deei.sions have been postponed from one meeting to the next, This was the case both with price-inc~asecltecisions during the seventies and with aspects of the quota system durilng the eighlies. Subcommittees have been used extensively to deal with specific issues in smaller and morc clandesthe settings. The veto right and the consensus ambition have, hwever, not aiways avoided havhg co~nfere~nces end with split decisio~ns,as happened at the important December 1976 meetkg in Uoha, described in section 5.1.. The formal v o t a power in OPEC is one country, one vote. It is not related to the market power of the m e ~ ~ b e rTsh.i s discrepancy between the distribution of power inside and outsi.de the organization may have implications for the membersf behavior. 'Xhe genertral proposition is that actors Witk a large or small amount of pakver both hside and outside the organization tend to be indifferent about whether or not issues are dealt with thrt,u& the decisionmaking procedures of the org""izatim, In this case, to not hclude issues in the OPEC deliberations mems lettjng them be decided through individual members" market policies. Camtries with little power outside the organhation (i.e., with little market pokver) will tend to hvor more pdicy devdopment through the bodies of orEc, wfiile countsies with greater market power will tend to keep issues away from the organization and possibly i p o r e decisions made by it. This is iliustrated in Table 4.3. In the case of the quota decisions, the members" ability to p d u c e in excess of the assiped quota represents c ~ n epossible power resome outside the organization. With a large spare production capacity; a country can undermine the collective decision by overproducing, Saudi Arabia has to some extent pursued such a strategy when dissatisfied with OPEC agreements (see section 5,l).The discrepancy between Saudi Arabia's im-

252

CXTC as an

X~ztemcktionaEOrganization

portance in the international market and its single vote h OPEC has to s o m extent been compensated for by the inclusion of Saudi Arabh in al.most all important ministerial committees (see Table 4.2). 7'heparticular rnle of SaudiArabia is fuily developed jn Chapter 6. Karns and Mingst (1986:467; 1990) have made an interesting cmtribution to the discussion of the relationship between international orgmizations and their member states"Their approach is to study the relationship between an intergave ental organization (ra?) and its members as a combination of the IGCI as a source of influence on the member countriesf foreign p o k y m d Che rc,o as an instruntcnt for the member countries" god achievement. When looking at OI~ECas an inshvment of states' goal achievement, it is fmportant to note that the decisions made in OPEC relate to prnblcms that the individual countries could in principle hnndle themselves, The interesting question is to what extent making decisions jointly hcreases the members;' net utility. This calculatim might naturally diMer from com-rtry to country and vary over time. Whm the room for coherent action in o13ec jncreased durirrg the sevmties as the producing countries nationatized, or gained control over their &l indrastries through other means, the role of Ehe organization b e m e obsolete-no coordination or institutional arrangement was necessary to serve the membersf interests (see section 2.4) Instead oE enforcing fie strength of orEc as a cartel, t-he price i n c ~ a s e ssom caused sevcre di,sagreernent inside WEC (see section 5.1). The seventies was a period when OPEC struggled with intern& differences rather t-hmgained cclntrol over the market and strengthened its position agaiinst the other actors, such as the international companies, the consuming countries, and producers outside the orgmization. The rule of unanimity gives the orgmiaation little autonomy in relation to the members, but it has also made the members a h at consensus. The differences ammg o12EC members have made s w e of the collective decisions v a g ~ ~and e , members have tended to disregard the decisions. So h, the analysis suggests that OPEC has been an arena for negotiations between sovereign states. The next three sections will m d i f y this picture to some extent"

An impmtmt elment in ensuring cooperation is the detection of defectors.16 An organization may play a mle in tbis by providing administrative staM and tecnnicat inslruments, or establishing im,partirstt inspection bodies. The reliability of the bformation gathered is m important aspect of the role of the oqanization in this ~ s p e c t'll-re . monitoring functions of the organization are meant to deal, with the possibility that rnernber

states behave opportunistically. As pointed out by Keohane, ""sme actors, may be dlshmest, and enter into agreements that they have no intention of fulfilling (1989:22f7). As pointed out above, Williamson f1975:25-26) does nut believe it is possible to establish self-enforcisrg cmmitments that would annul self-disbelieving promjses. Athough the ot extinguish opportunistic hehavior, it can irrcrease the probability that opportunjsrn wilt be detected. This should, ceteris paribus, make other actors more willing to enter into agreements with potential opportunists. OPEC took on a more substantial institutional role h the early eighties. This happened not through an activation of the established institutions or Statutes but through the creation of a new subcommittee, the MarlPcet Monitoring Committee, at the Sixty-third am@Conference in March 1982. This cmmitket established to moni.tor the production quotas ini.tiat& at the s m e time (see section 5.21, held its first meeting on April 23, f 982. At the Sixty-fotxrth OPEC Conference on May 20 and 21, the pmduction ceiling was maintahed, and was to be subject to continued surveil.lance by the monitoring commfttee. The Conference discussion was domc e its quota, an issue pursued by the hated by Ism's n o ~ ~ c m p l i m with monitorling committee after its metjng on July 7: Members of the committee expressed some concern about certain member countries' "over-production" mlative to their ceiling. . . . The 65th (extraordinary) meeting of the CIPECcmference opened in Vienna on July 9 to consider the monitorjng committee" recommendations. Differences over the economic issues involved, were, howweu; sharyened by political tensions, and on July 40 the meeting was suspended after reaching deadlock, . . . Iran claimed that there war; wide suppc~rtat the meeting for the principle of an increase in its quota to 3,000,000 bpd and that this represented a ""moral triumpfi" over Saudi Arabia. (Evans 4990:605406)

The Seventy-second OPEC Conference in December 1984 called for a system to pmvide full information m individual membersf production exparts and prices: Far this purpose, C ~ I ~ Ewill C create a Ministerial Executive Council. . . . The Ministerial Executive Council . . . is empowered to take any measures it deems necessary to fulfil1 its tasks. The Council will be assisted by one or mare reputable international auditing firms to provide a check on Member Countries"getroteurn sales, tanker nominations, shipments, prices, quantities, etc. The auditing firm will be empowered to send its representatives to Member Countries tcr check the bc~oks,invc~ices,or any other documents that are deemed necessar)r by the firm in fuFuXfilfment of its tasks. lEJikewisethe

254

CXTC as an

X~ztemcktionaEOrganization

Council may ct?oo)se any other means of check and controt, such as tanker tracking methods, to be unde&aken by consulting firms. The Council may also send its representatives to pay visits to the parts and loading terminals of Member Countries to gro>videcheck and controt, besides the auditing firm% srepresenl.al.ivr?s.rentve Member Countries undertake to make available to the auditors, their representatives and the representatives of the Council, all the required documents. They also undertake to send all the information on tanker nominations to the OPEC Secretariat. ( C ~ ~ B ~1990:226) ZC

This unanimous adoption. of this resolution was, in j~~risdictional terms, dramatic, 1Phe Ministerial Executive Council is empowered to apply any measures necessary, and the auditi~~g firms are also given wide authority regarding i n ~ e c t i o n isn the mentber countries' territories*However, the resolution did not spell out any sanctions to be implemented shvuld members break the agreed production v a t a s . 11%a ~eparatemeetiw cm January 27, the Ministerial Executive Council outlined a proposat for the monitoring scheme, and suggsted that a Dutch accomting firm, Klynveld Kraayenhof, should be hired to carry out the audits (Evans f9911:658459), The identities of quota violators werc not officiaiily published by the organization, but were raised during the internal bargahing at the cmfcsrences, as was reported for instance at the Seventy-second orEc Conference in Decernber 1984: "Scveral delegates expressed deep concern about the Organization's diminishing credibility in the markeylace, in which connection there was much blunt criticism of various menbers' mncornpiiance with Conference decisions" (Evans P990;645). I'rciblems emerged when this rewlutiOn was to be implemented. At the 9vent.y-fot~rthConference inAugust: 1985, the Ministerial Executive Cormeil infc3rmed the ccjnfewnce that five C ~ I ~ EmemC bers had declined to co-operate fully with the auditing procedures approved by the T 3 r d Conference meeting and made various recommendstions for improving the effectiveness of these praceduws. The five countries concerned (which were not publicly named) agreed to open their books to the auditors in the future and joined the other member countries in approving a strengthening of the Cc>uncillsmmotl-rring methods. The Conference did not, hc>wever;accept a prc>pc)salby some member countries that the Council should be empowered to recommend the imposition of sanctions on cauntries where ""malpractices"%ere revealed as a result of independent audits. (Evans 1990:665)

I'he monitoring efforts represented important imprtwements to the cohesion and effectiveness of the productio~lquotas, but the unwillinlgness to

impose threats of sanctions m d e the wanizatio11's efforts half-hcarted. The decision made in the second part of the Seventy-eighth Conference in August 1986, illustrates that the quota distribution and the individual members' aadfrrere~~ce to it were questions open for discussion. At this meeting, Iraq was rekased from quota restrictions, This was not only an ecmomic cmcession but also a political one (see secticm 3.6). 'l'he resolution also contajned a hrmulation of the status of the q.ilota decisian: "this temporary masure should not prejudice the discussion in the Conference concerning new national quotas b r ifs Members and ~ I ~ E C 'approS priate and rightful total production""^^^ 199(1:245). In, addition to the fact &at no sanctions were effected agaixlst quota-breakers, the temporality of the quota decisions was underlined. As the figures ol totd OPEC prod~rctimand quotas in Figwe 4.1 suggest, the production was reduced below the quota level during the monitorhg period. However, by comparing tbr fi8urc.s for Smdi Arabia and the other OPEC members it becomes clear that this reduction was entirely due to Saudi Ara:hia cutting its production. Rather than the effect of international instibtions monitoring state behavior, the role of Saudi Arabia shodd be emphasi.zed. Chapter h is ent.irel,y devoted to the hegemonic role of Saudi Arabia. Figure 4.7 also demonstrates how the change in the distribution of production quotas in RcTarclh 1983 was a transfer of quota .from Saudj Arabia to the rest Of the orEc mennbers. The jnternal bargahing over quotas is further discussed in Chapter 5.

The sihtational model presented in section 1.4 distinguished between aspects of the situation surrou~zdjngan actor m d the actor's actual behavior or choice. An importmt aspect between these two levels is the actor's perception cJf the environment surroundillt; his m her dctcisions. The organiza"cion can influence the actors' perceptions of the situation surrounding their decisions-in other words, the organization can frame member states2ecisions. Althoul;h this is not necessarily incompatible with the msumptions of the theory of rationd ehoice, it is fully in line with the sociological instihtional approach. As pointed out by March and Olsen (1994:2Ci3), ""institutions influence the pcrlrception . . . of the reality withh which action takes place.'" According to the theory of rational choice, the actors"preferences regarding differmt alternatives are not taken to be dependent cm the w v the alternatives are presented. This is the assumption of invariance. hportant studies in cognitive and social psychology have shown how the preferences of actors vary accorciing to the presentation of pos"ile choices (Tversky and Kahnernan 1986; Qx~attraneand Tversky 1988;

Tversky; Slavic, and Kahnernan 199Q).17Quattrone and Tversky (1988:735)conclude their article as follows: The descriptive failure of normative prindples, such as invariance and coherence, does not mean that people are unintelligent or irrational. The failure merely indicates that judgment and choice-like perception and memory-are prone to distod.ic>nand error. The significance of the results sterns from the observation that the errors are common and systematic, rather than idiosyncratic or random, hence they cannot be dismissed as noise.

Kemeth A r m (1983:29) =gads this as "the most d m i n g cri.ticism of risk-benefit anatysis.'Xlso, March and Olsen f 1989:154-155) argue that there are "numerous reasons for lamenting the tendency to treat preferences as given in theories of choice. Where prekrences arc arnbiguous and changhg endogenously, t~at-itlg them as consistent, clear, stable, and exogemous leads to predictive theories that are wrong and normat-ive theorits that arc misleading." Like uncertajn information, undecided pmferences will increase the actor's susceptibility to influence from olfner actors. This i s not contrary to the theory of rational choice, as this theory does not pretend to explain why the actor has the preferences he or she has. Richarcison fl990:5), however, regards this as a serious weakness of the rational choice theory: "the goals, p-eferences etc.-remain unexplained. The language of the expected utility offers a refomulatim of the explanandum not an explanation. . . . The thin theory of rationality, like the syllogism, is empty of content, it enters into every explanation, but in and of its& it can exgtain nothirzg.'TEfster f1983:15) &scribes the dilemma as follows: "to say that truth is necessary for rationd beliefs clearSy i s to require too much; to say that consistency is sufficient, to demand too little . . . for rational desires: the requirement of consisttmcy is too weak, that of ethical good~zesstoo strong*'f Mihm regarding framing of decisions, other actorskse of language becomes an important feature. It seems even more difficult to reveal frming than ies, How should Ihe actor become m a r e that his or her ekoice has been framed? It is possible to have the world presented to you in different ways, but what should bo the norm from which to decide which presentation is the real o~le?lB The skategy of the Market Monitoring Committee in 1983-1984 was a case of framing of decisions. As shown above, the Committee introduced the concwt of overproduction. The leader of the committee, DC Man& Said aI Otaiba, thus tried to establish the perception of the situation as a problem of distributim of production quotas rather than as a question of urhich relationship between price and total OPEC production

258

CXTC as an

X~ztemcktionaEOrganization

would present the best possible outcome for the OPEC ~ountriies.1-~ By changing the perception of the game structure into one of pure distribution, the i~~dividual memberskalculatisns were framed, as one member's overproduction wodd mean that another mmber woufd have to cut back. 'l"l~emonitoring committee met both in August a d September 1983, m d decreased its estimated demand for OPEC crude. Its recommendation was subsequently to maintain the 17.5 mbd production ceiling. In July 19133 the role of the MarkcJt Monitori~~g Committee was praised by the presidcw of the Seventieth orEc Conference: The mast effective instrument of success was the strict adherence to our agreement and the maintenance of cohesion of the Brganizat.ic>n.The work of keeping abreast of oil market developments and Member Countries+erformanee was facilitated by the Ministerial Monitoring Committee [Market Monitoring Cornmitt.ee]zo. . . . It has continued to maintain a cXose and vigib n t watch on market developments, advising Your ExceiXencies of the situaE Ctake the right decisions. tion at any given point in time and enabling C ~ I ~ to (CIPEC 2 990:225)

Focusing m overproduction led OPEC into a role as a rd~$j&alsupplier. The Market Monitoring Committee actually initiated the now-widespread way of reading the oil market: First one estimates the total, demand; then the nonar2ecsupply is taken as given, and accollnt is taken of likely stockhg or destockng; the rest of the demand is the so-called call on OPEC. The "'call on c313Ec" tthus neglects the possibility of d e m n d being fnfluenced by the price tevel, which in turn c d d be influenced by orec's production. It is also assmed that non-or~c producers prodwe at iull capacity regardless of price (i.e., that they arc genulne price-takers). This way to read the market is in line with the dominant firm model (Cr4mer and I s f h m i 1991:4245). With this perception of am@as a residual suppIier comes a reduced role of the mmopolist in the price-setting. Although OPEC does not become a price-taker in this model, the orgmization becomes a non-OPEC produ'ction-taker.What happened in the market was that the monopolist" ((ox.~c%s) market share dwreased as the frhge ( n ~ n - 1 ~increased ~~) its market share-all in line with the dominmt firm model. h theory, three strategies are open to the monopolist i,n such a situation, and o13~c did, to a variable degree, try to pursue all three:

I. cut prices h order to hcrease demand 2, put pressurt? on nan-opEc producers in order to include them in the prduction-limitation p m g m 3. cut one's own production

Since the first failed because of the increased taxation of oil consumption by the governmnts in oil-consuming countries (see Figure 2.15) and the second failed due to the lack of means to force the ~ - O I > E producC ers to comply (see Chapter B), OPEC came to rely on the third strateq. Xn the period. 1983--1985 the costs werc almost entirely borne by Saudi Arabia (see section 6.5). The role of the monitoring committee durhg 1982 was that it reduced the room for members to d e m n d a higher total ceiling, althougtn some members obvious:ly a$vocated this at tl-te conferences. Since m m weight was given to the total ceiling, the individual wrnbess' claims for higher quotas becme an entirely distributional baqaining process, in which one member" increase in prociucti~nwoutd have to be followed by another member's decreased production. Distributional bargains are for obvious reasons much tougher than bargains with integrative potential. p'duceion deciThe role of the co ittee fn actualty frming the a9r?c sions as a distributional issue contributed to the subsequent solution in March 1983, when Saudi Arabia reduced its ceiljng in exchange for an increase in the Iranian quota (see section 5.2). The committee strengthened the organization-thd is, the collective strength of the members-but it simultaneously b~creasedthe internal tension. W ~ e nthe quota system was established, the role of the Market Monitoring Cornsnittee was enlarged from providing information about not only the market but afso the individual members>yuta discifline. According to the head of the committee, urn Oil Minister Dr. Maneh Said a1 Otaiba, the compliance W= mmitored on the basis of a quarterly average (Evms 1990:633).Although the c m m i t k e improved. the information on memberskcompliance, it was unclear wfiat sanctims could be inflicted on non-abidhg members" This was clearly illustrated by the statement made by al Otaiba prior to the Seventieth OPEC Conference in July 1984: /'Dr. al Otaiba said at the end of the meeting [the committee meethg before?the ConPcrence] that it was necessary fox the Conference to 'give confidence to the oil market by askirzg the member countries to atnolish quota violations and direct and indirect price discounting"""Evans 1990:636, italics xnine). The Conference decided to send cnrsc delegations to visit member countries "with a view to hrther consolidating Member Countries' commitments to stabilize the market and defend the price structure" "(QPEC 1990:216). The perception of quota violations as overproduction and the market balance to create a residual named the ""cl1 on QI~EC" both frame the choices of the individual o r E c member. The notion of overproducticm makes it harder for FRdividual countries to engage in price wars agaillst nm-OT~EC suppliers, as a quota violation =presents an attempt to gain at the expcnse of fellow OPEC mexnbers. The call on o r F c covers the possible

2 60

CXTC as an

X~ztemcktionaEOrganization

effects on the non-orec production from a prjce fall facilitated by increased OPEC production, The non-0x2~~ producers are assurned to be price-takers, irnplying that their total production is also assumed to be insens-itive to price declines. This is in line with l.he institutional approach of March, and C)lsen (1389:45): "hgeneral, the predic.tims that can be made depenli on the fine &tail of the ways an instibtion organizes cmtact and experiences, . . . W h e rneanjng ~ develops witfiin such a system, there is a tedency to partition a population of indivjbuals into tjroupwr institutions that share interpretations and p~ferenceswithin groups but not acmss groupsMAIt:hou& all oil producers in this resped have similar interests in the oil market, the C313EC members tend to view the nrn-oI2Ecproducers as having different interests than theirs. G m m m perception and hterpretatiorn of the world are aspects of the instituticmalization of cooperation between states that cannot be understood through a purely rationalistic approach. The stressing of the difference between orEc and non-OrEc producers also has irnpctrtant role in the institutional explanations of the OPEC cooperation, as it contributed to the establishment of a coiifective identity among the OPEC mernbers,

t'stablklti~tgCollective Irie~ztify The aim of this section is to discuss the importmce of a cornmorn identity among the WEC members for the explanation of their cooperative behavior. Increased cooperative behavim m i e t c m e about as a ~ s dcJft several different processes of identity formation. Alexander Wendt (1994:3&&391)outlisres different causal mechanisms that ""promote collective state identities." The mechanisms operate on differetnt levels. The first level is the "kters~b~eclkf. s y s t e ~ ~structures ic [which] consist of the shared understandings, expectations, and social knowledge embedded in international institutions and threat complexes, in terms of which slates defhe (some of) their idetntities and interests." The second level is the sptemic proccsses, understood as "dpamics in the external contertC of state action," which lead to increased interdependence, which can imply both an ""increasein the ' d p m i c density' of interactions . . . land the emergence of a 'common Othe.cfThis reduces the ability to meet corporate needs unftaterally and increases the extent to which actors share a c m m n fate. As thc abi.lity to meet corporate needs unilateraib declines, so does the incenthe to hang onto the egoistic:identiticls that generate such policks, and as the degree of common fate increases, so does the incentive to idetntify with others" "e~ndt 1994:389). The third kvel is the strategic practice. Wendt develops two subcatetjories, the behawioral and the rhetorical. "'The behavioral suggests that rclpeated acts of cooperation will tend to have two effects on iderntities

and interests. First, the symbolic intemctionist concept of 'reflected appraisalsbuggests that actors form identities by learning, through interaction, to see fiemselves as others do. . . . Second . . . by engaging in coogerative behaviar, an actor will gradually change its own beliefs about who it is, helpirtg to internalize the new i d e n t q for itself, . . . ~ p e a t e dhteraction can transform an irrterdependence of outcomes into one of utility"7Wendt f,994:390). The rhetorical practice "may have eMects similar to those of behavioral practice but it does so through a different mode of communicatim, variousty enacted as consciousness risingf dialogue . . . discussion and persuasion . . . education, idco)ogical labor . . . political argument. . . symbolic action . . . and so on (Wendt 1994:391), Herc?, t w aspects will be discussed. First, an image of a c might increase the actors' perception of colnmon interests among themselves, increasing t h i r cooperative behavior, a mchimiszn Wendt calls the systemic proems level (49%:389). % c o d , an organization perc-ttived as a model far o&ers might itself g a l co~~fide~~ce, a mech,?nism W e ~ ~calls d t Che iirst type oi behavioral and rhetorical strategic practice (1994:390). Different aspects of cdlective identity an-tong states or societies have been important for the study of international relations for m n y decades, in particular in the field of integration tl-teory: "one of the crucial yue* tions in integration theory is the relation of a sense of community to behavioral interdependencer"Nye 1971:26). Jacob and Teune (19M:4-5) argue that "political integration generally implies a relationship of community . . . a feeling of identity and self-awareness." According to Ru,lS (1977:M) there is a relationship betwcen common interests m d common identity as a ""sense of common hterests . . . [that] in same cases . . . may express the ahility of the individuals or groups cmcerned to identify with each other to the extent of beating each other's interczsts as ends in themselves and not mercly as means to an md; that is to say, it may express a sense of c o m m vaIue~.~" The argument in this subsection is that internatio~~al, orgmizations can inhence sta.tes%ehavior in a m m cooperathe direction by contributing to the creation of collec-tive identities. Forming ide~ltitythrough a perception of a common enemy in this case means that the cooperative behavior of Q I ~ E Cmembers is d.eve1opehas a result of the members' perceiving some kind of threat from other actors (Wendt 1994:389). This threat nrtight change in time and space. Concluding his study '"Self and Other in International Relations," Neumann (19%:161i-167) argues that "the delineation of a self from an other is an actz'ue and ongoing part of identity formation. 'The creation of social boundaries is not a cmsequence of integration, but one of its necessary a pviitvi ingredients." This can be interpreted as meaning sornethimg like, "no self withwt an other." 'The resolutictns

2 62

CXTC as an

X~ztemcktionaEOrganization

adopted at the First OPEC Conference in Septe~sberI960 clearly indicate that the organization was established as an instrument h t:he ongoing fight with the internationai oil companies. Particularly, the price policy of the companies triggered strong reactions f r m the oil-producing countries. The first sentence of the first CXEC resolution reads as follows: "The members can no longer remain indifferent to the attitude heretofore adopted by the Oil Companies in effecting price modifications" ((OPEC 1990:1), The second resolution adopted established the "permanent eganization of the Petrdeum Expmting Countries for rclgular consultation m o n g its Members with a view to coordinating and unifying the policies of the Members." Later, at the Eighth Conference fn 1965, thfs was estabfished as the principd aim of the organization (OPEC 199Q:32).As the organization was established in opposition to the companies and with the aim to unify the policies of the member states, the us versus them xlationship was between the producing mostly private, international oil comtries inside OPEC and the foreig~~, companies. During the first half of the seventies most orEc members nationalized their oil prcrduction. This made the relationship to the international. companies of lesser im,portance, or, more correctly, made the international oil companies less im.portant for the develspments in the oil market. Consequently, ""the other," h v i n g been one of the main reasons for the establishment of the organization, was no longer an. important factor. Regardkg the relationship with other actors wanting to join the organization, the potential, new members were restricted by Article 7c of thc Statutes, which provides that ""any other corntry with a substmt-ial net export of crude petrcoleum which has fwdarnentaily sirnilar interests to those of Member Counlries, can become a F d M d e r of the W a n i z a ticm. . . . No country may be admitted to associate membership which does not hndamentally have iz~terestsm d aims similar to those of member countries" (WEC 1990:33). This is an example of what Cox and Jacobson (1.973) call a boundary-decision, creatixlg criteria for who can be incfrxded in t k organization and who cannot. The Statutes suggest that countries wi& a substmtial net export of crude petroleum have some objectives in common. They also set clear criteria mgarding what kind. of . statutes obV;oudy create a c m countries OPEC is set to ~ p m s e n tSuch man. ide17tity among the OPEC comtries. Creating cornmm identi.9 by being an example for others means that the members of ~ I ? E Cmight find. it harder to leave or obstruct the working of an organization that has becom an ideal for other Third World countsies seeking to better their position against the industrialized corntries through similar collective action.

When orEc succeeded in raising thc price of crude oil, it seemed the beginning of a broader change in the relationship between the raw materid producers in the south and the industriahed countries in thc. north. "After the events in the fall of 1977, ore@became a model, radicdly strengthening the belief in the effectiveness of producer-exporter cooperation as a mems of promoting the interest of the periphery countries"' (Hveem 197215). 'The orEc s u m i t of heads of state inAlgiers in March 1975 was the peak of the role of OPEC as "the shield of the mird World,"' as the banners in the city proclaimed (Terzian 3985:213). Algerian authorities submitted scverd '"&aft resolutions"' covering, mong other issues, an international cnnference on raw materials and development, a declaration dealing with the international monetary system, a kjnd of defense pact between orec members, and a pr4ect for an OrEc fund for development and international cooperation (Terzian 3985:213-214). The orEc special fund was set up in 1977, providing "interest-free long-term loans to developing countries" "(OPEC 1990:139). The support fsr the Tl~irdW r l d was stated on several occasions, as for instance in the eighteenth anniversary address of Secretary-General ATi M. Jaidah: Countriest being themselves developing countries, have always identified themseXves with the ei'Forts of the Third Wc3rtd to bring about a n w international economic order. In this spirit they have done their utmost to aid their brethren in the other developing countries . . . through steadily increasing financial contributions to alleviation of the latter grc>upts balance of paymats problems and to enable them to go ahead with development projects. (CII~EC1990:158)

CII~ECMember

The role as the shield or spearhead of the Third Miorld obvious built a on identity m o n g the OPEC members. Although the substantive eMort was moderate, and the role later petered out, it contrihvlCed to increashg the perception of commonality against the divergent interests emerging during the latter half of the seventies, particularly regarding the price issue (see section 5.1). The relationship between the menrbers of UI'EC is not like that of a community of states. The bmds between the OPEC mernbers are, in this respect, weak. mere is no ambition in the organization to try to develop the cooperation between the members into anything broader than the present speci.fir issue-oil cooperation. Hwever, as this section has shown, there are ele~sentsof colfective identity formation among the o13EC members. Aspects of the Statutes and the way o&er countries perceive fie OPEC members have contributed to this. Thus, to some extent,

2 64

CXTC as an

X~ztemcktionaEOrganization

the OPEC decisions take place "wi.Lhin a broader framework of rules, roles and identities" "arch and Olsen P994:25l).

The role of the Market Monitoring Committee illustrates a much debated aspect of internat-ional institutions-the extellt to Ml)nich such institutions have an autonomous role against the interests of the states constitutirng the organization. Rohert Keohane out1j,nes t w important aspects of the signif cance of institutions: "International institutions are important for states' actions in part because they aifcct t:he incentives facing states . . . [and1 they help define how intemsts are defined and how actions are interpretedff "eohane 1989:s-6). In 1982, the Market Moni-toring Committee helped define the problem facing the oil producers as one of intern4 OPEC etistribtrtictn of production cutbacks. This leads to a discussion of thc likely provision ol the collective good, cvhjch is the key topic in Chapter 7. The Market Monitorkg Committee actually reduced the size of the reIevmt group h m all oil producers to the members of mec. Considekg this from a purely economic or game-theoretical approach, there is no reason why reduced pmbction of an OPEC merrtber should have any other effect thm reduced production of an oil producer outside the organization, However, the committee created such a notion a m g the members of cnmc by focusing on their overproduction." This focused the bargaining problem on the internai distributio~~ of producrion cutback, At the same time, non-WECpmductim increased steadily leaving less room in the market for the B19EC membersf production. Other external factors also contributed, like the slackening ol demand and the increased taxation of imported oil by consuming countries (see section 2.5). Also, rather rigorous measrtrm were taken to persuade the members to comply with the a g ~ e d quota arrangements, but no explicit threats of sanctions were introduced. Individual members uttered their .intention to hcrease their w n production if others did not comply, but this was not the poficy of the organization as such. Thus, the cmclusion is that the implementation of the quota system h 1982 would not have had the effect it had without the insl;itul;ionalmechanjsm provided by the Market Monitori,ng Committee in est;xblishing the notion of the interests of OPEC producers as different from those of oil producers as such. The c m i t t e e put an institutional constraint on the behavior of the orEc mexnbers. 'The effect of the quota syste~non members actual production is further discussed in Chapter 7. Fur&ermore, the production-monitoring efforts made it harder for the individual rnelnbers ta get away wi& cheatkg umoticed. The decision

at the Seventy-fourth OPEC Conference hdicates that the cheating coulztries felt some embarrassment of being marked as non-compliers or not abiding with the monitoring proceclures. The combination of strong monitoril?g applied h accordance wi& a resolution of the Scvmty-third Conference and. the lack of willingness to sanction cheaters by the Seventy-fourth Conference c l e d y illustrates hoMi in this matter O19EC tread on the edge of what an international orgnnization can hope to achieve when confronted with the interests of sovereign states. Figure 1.1 also shows that the reduced werproduction in the m i t o r i n g perioci was atmost entirely due to Saudi Ar&ia reducing its production. Also the situation in 1983, when Saudi Arabia rttduced its quota while the others increased theirs, s h w s that the independent role of the quota system on production was r a h r limited. These topics are fwt:bcr discwscd in Chapters 5 to 8. The impclrtance of the institutional argument still hcrlds some validity, as the OPEC menbers have behwed differently from non-OPECproducers, large or small: Given that the so-called non-orsc prc>ducers,almost by definition, consistently maintain their output levels at full capacity, it obviously fcrifows that whatever spare capacity there is in the tzrorld supply system is located in the group of 12,ctrsc prcjducers, and that the vital job of regulating supply in order to defend p r i c e e i n other words keeping this spare capacity off the market-inevitably remains the pu~viewof C)E~EC.(Seymour 4996:3)

Even though WEC did not facilitate other oil pmducerskmtributim to the collective good, it did help to keep OPEC toget-her.A common e n r y increased internal coherence. Looking back at the introduction to this chapter md. the role or function of inf;emationd organizations, the empirical sections of this chapter clearly show how the international organization-OPEC-provides the members with injovrvrili.z'on regarding the market, dcfincs a common percepfiot-r of the membersf bargailling problem, helps to establish rules for production by settjng quotas, contribn.t.es to the enforcement of these mles by nzolzifori~~p the individ.ual mexnbershctual production, and has increased the c o h e ~ n c eof OPEC by enhancing the pe~eptionof a cornm m enemy in the non-OPEC producers, In 1982, the mmbers' percreption of the situation was partly formed, by the osganization; in 1984, a mmitclring effort was established and pursued by the organjzation; a d in 1986, the coherence of the members agahst the other oil producers was strengthened by the organization. On the basis of th discussion in this chapter, it is argued that, at these particular instances, the oil-producer cooperation w ~ t ~not l d have been the s m e without the institutional fac-

CXTC as an

2 66

X~ztemcktionaEOrganization

tors underlying the behavior of the orgmization, which partly cantradicted the members3interests. When OPEC was established in 1960, the idea of production-sharing agremmts was included in the Statutes, but was not i q l e n t e ~ ~ t eThis d . can be interprekd as indicating that the inslitutional basis for the implementation of this instrument was not present in 1960. When pmduction-sharing was ifiroduced in 1982, the commm institutional cxpesiance of the orEc members made them more prone to act in concert and implement both the quota system and, in 1984, the monitoring system in order to strengthen &c? organization at the expense of their individual sovereignty. The focus of this chapter has been on the expllcit effects oi institutimd factors on states' hehavior. Even when discussing the aspects of collective identity, the empirical argtxnnelrts were based on =solutions m d statements by reprcsmtatives of OPEC or OPEC mmber states. International organizations like CX~ECmight, hokvever, influence state behavior wifiout such explicit expressjon. 'The possihijity of institut.ianal factors influenth g state behavior through more implicit mechanisms than those discussed in this chapter is suggested by Rosenau and Czernpiel (4992:3): "To prcsune the presence of governance without government- is to conceive of functions that have to be perfomed in any viable human system irmspective of whether the system has evolved orgmizations and institutions expficitly charged Wifh perfomhg them," For the purpose of this study, this argument suggests that states" behavior rnight be influenced by institutimal factors trYitbout: any explicit action by the organization. Investigating such mechanisms would require other m d far more thorough methods than the ones applied in this study. 'The conclusions reached in this chapter cannot d e out that such implicit effects have infiuenced the coopesativc behnvinr of the member states of OPEC, in addition to the explicit ones discussed in this chapter. Fur*errnore, when it is argued in the next chapter that the cooperative, cartel-like behavior of the OPEC m e ~ ~ b eisr snot a result of institutio~~al hctors, it is with the reservation that implicit mechanisms of the kind mentioned here have not been active.

Notes 2 . This question is based on the assumption that by institutionalizing the cooperation in a fc~rmalorganization, the individual actors will tend to behave more cooperatively. Hovi (1992:229) points out that cmperation beween statczs might be gerfectty viabte without an organization, and the establishment and running of international organizations might be costly. The net gain from institutionalizing cooperation is thus an open empirical question.

2. It should be noted that the term "wonomic institutionatists" in the Eollc>wing pages refers to the "new" institutionalists, It is an important difference between the old institutionalisb as Thorstein Vebten, John Commons, and Wesley Mitchell and the new institutionalists discussed in this chapter. While the aspects of new institutionalism discussed here largely accept the primary role of the individual and thus ascribe to methodological individualism, the old institutionalists rejected this primacy of the individual: "Not only is the individual's conduct hedged about and directed by his habitual relations to his fellows in a group, but these relations, being of an institutional characteu; vary as the institutional scene varies" "eblen 1909: 245, cited in Fliodgson 19"3:209). In the discussion in this subsection such arguments tzrould place Veblen closer to the sctciological institutic>nalisrndevelctped below than to the economic institutionalism of this sectim, 3, 13hrasetaken from Hodgsrrm 2991:186. This approach suggests a fundamental difference (not only one of degree) between competitive markets and markets with some kind of imperfect competition, such as oiigopolies or monopolies, 4. "Rationality implies a complete, and unattainable, kncjwledge of the exact conxquences of each choice. In actuality the human being never has more than a fragmentary knowledge of the conditions surrounding his action, nor mc~rethan a slight insight into the regularities and laws that tzrould permit him to induce future consquences from a knc>wledgeof present circumstances" "imon 197&81). 5, Michael Dummett (19"i7J:298,cited in Davidson 19%:266) argues that "there is a general convention wher&y the utterance of a sentence, except in special contexts, is understood as being carried out with the intention af uttering a true sentence," Dmald Davidmn (19@:270), on the contrary, argues that "What is understood is that the speaker, if he has asserted something, has represented himself as believing it-as uttering a s e n t m e he believes true, then. But this is not a convention, it is merely part of the analysis of what assertion is. . . . there cannot be a cmventic~nalsign that shows that one is saying what one believe; for every liar would use it. . . . There is no convention of sincerity" According to Ramberg (1989:8), "Davidsc~n'skind af thecjry is one which tzrould enable us to specify the conditions under which a s e n t m e in a language is true, without telling us anything about when those cmditions prevail or how tc~determine tzrhether they do prevail." h line with Davidson's refusal of conventions regarding the truth of utterances, Rc>bert.Jervis (197"O:CiCi)claims that ""since signals derive their meanings fram tacit or explicit ragreemnt among actors, an actor can lie as easily as he can tell the truth. A signal used to cmvey an accurate message can also convey a misleading one, It is logically impc~ssibleto design a signaling system that does not have this attribute." Nothing in the language itseXf makes it possible to distinguish lies from truth. 6, As indicated above, the possibility of making more profit than your competitors is in a fmdamental way based on your knowing samething he or she does not know. " f w , assuming perfect information misses the whole point. '7. This questicyn is similar to the questions raised by the students of international regimes cancming regime-effectiveness (see b u n g 1982, 1994; Underdal 1992; Stokke 1997).

2 68

CXTC as an

X~ztemcktionaEOrganization

8. An example might illustrate this point. GATT (the Generat Agreement on Trade and Tariff) is often referred to as a prominent example of an international regime ( L e v Young ~ and ZGrn 1995:279). I guess not even the boldest advoycate of regime theory would include David Ricardo's theory of comparative advantage as a part of the theories of intesnatianal re@mes, although the mechanism specified by the thectr~i5in itself, both can make states change their behavior and start bilateral trade relations (hardly- a regime), but also make them change behavior and agree on common rules for free-trade (as in the case of GATT). 9, The founding members are ""those countries which were r e p ~ s e n t e dat the Ist Conference, held in Baghdad, and which signed the original agreement of the establishment of the Organization" (OPEC 4%0:33). 10. Tn 1974, Abu Dhabl" mmbership was transferred to the UAE, of which Abu Dl-rabi is a member together with Dubai, Ajman, Shaja, Umm al Qaiwain, Ras a1 Khaima, and Fujaira. 14. ""A net petroleum-exparting country which does not qualify for membership under [the above] paragraph . . . may nevertheless be admitted as an associate member by the Conference under such special conditions as may be prescribed by the Conference, if accepted by a majority of three-fourths, induding the concurrent vote of all founder members. No country may be admitted to associate membership that does not fundamentally have interests and aims similar to those of member countries" "rticle 7 of the Statutes, OPEC 1990:33). 12. As is normal for international gc>vei-nmenta1organizatic~ns,the bureaucracy shall not be rep~sentatirvesof their respective countries: "The staff of the Secretariat are international emplc~yeeswith an exclusively intei-natiortal character, In the performance of their duties, they shall neither seek nor accept instructions from any government or frc>many other authority outside the Organization. They shall refrain fram any action which might reflect on their position as international employyees and they shall undertake to carry out their duties with the sale object of bearing the interests of the Organhation in mind" (Article 32 of the Statutes, OPEC 1990:39). IS. Wendt defines self-interest and collectirie interest as effects of the extent to which social identities involve identification with the fate of the other (Wendt 4994:386). His positic~nis between the rationalistic and the holistic, as he regards the importance of collective and egoistic interests as an empirical question: "Thus, I am not suggesting that cotlective interests replace egoistic: ones as exogenously given constants in a rationalist model but, rather, that identities and interests be treated as dependent variables endogenously to interactic~n""(ibid.:387). 14. This implies that the strategy of noncooperation is not a dominant strategy if we view at1 sequences of the game as a whole. The structure is thus different in an iterated prisoner" dilemma than in a one-shot priscjner" dilemma, 25. This might seem an unnecessary task today, when daily market information is readily available through international energy infctrmation services around the world, It is important to note that this has not always been the case. Not until the mid-eighties was there transparency in key oil-refated economic factors such as prices, production, shipping' refining, and so on. Naturally, some aspects of this are today regarded as industrial secrets by prcyducers and other actors in the market.

16. This could be viewed as a part of the infc2rmation prc)blem, as it is a question of information about other actctrs~fitrtrecompliance with agreements. The fact that it is a question of$-llztl.e behavior makes monltsrjng, for the purposes of this chapter, a diffel-ent issue from the aspects of infc~rmaftiondiscussed in the subsection above on OPEC as an information agmcy. 17. Below, an illustration is presented of the respondentskcfi~ieebetween the following political programs, presented in two different, but substantially equal, ways:

Case 1 Unenrplaymefzf Program J 10% Prouam K 5%

bzfi~ t iouz f 2% 17%

Case 2 Emyloyrmnf Program J 90% 12rogramK 95%

I~$al-t'o~z 12% 17%

Tn case 1,36 percent chose program J and 64 percent prcjgram K. Tn case 2,54 percent chose program J and 46 percent program K (Quattrone and Tversky 1988:727), In case 1, the difference in unemployment is perceived as increasing costs, white in case 2, the difference in employment is pel-ceived as a reduction in gains. 18. For instance, it is impossible to decide which case is more true in note 17. 19. As will be explained in Chapter 7, this pric+production relationship was perceived differently among the OPEC countries because of fundarnentai differences in their prc>durtioncapacities, rexrves, and income needs. 120. As Figure 3.2 show-S, since the Ministerial Monitoring Committee was atablished in 4988, it is obvious that it was the Market Monitoring Committee that the president of the Conference had in mind. 21. "Members of the Committee expressed some cmcei-n about ce&ain member cauntries""overproduction' relatirve t'u their ceilings, it being made clear by the Venezuelan Minister of Energy and Mines that his cuuntr)~ (which had to turn away buyers in order to abide by its own production ceiling) would feel free to disregard that ceiling if such visEations cmtinued" "vans 1998:605).

This page intentionally left blank

Price and Production Policy of OPEC Countries

The previous thrce chaptctrs have studied constraints on the oil-related behavior of OPEC members. This chapter focuses on the behavior of states t%tennselvesfor, rnore precisely, on the interaction betweell oil producers. In line with the multilevel approach outlined in section 1.4, the fnllowing four chapters represent studies of the interaction of states. The follswhg discussion of the price and production policy of the o13EC members are organized into tZlree periods.1

I. The first period starts with the TekanTripoli agreements of I971 and lasts until the eve of the second oil-price shock around 1.981. In this period price-setting dominated OPEC policy and cooperation (see section 5.1). 2, The second period centcrs on the estnblishmmt of the quotas in 198243 and the foltwing swing-producer policy of Saudi Arabia untiS 1985. 'This period is of crucial importance for undcrstanding both the mistakes of the past and the future behavim of OPEC: (see section 5.2). 3. The third period is from 1986 to present. 'The key problem in this period is the nzaaagi~gof the quota system, caustng WEC countries to abaneion the ambition to set the oil price (see section 5.3). As mentioned, the discussion is based on cartel theory, which will be described and retated to the case of OPEC in Cl~apter7. Et should be emphasized that in this chapter only inkrnal aspects of the price dedsio~~s are relevant. The aspects of price increase that am connected with the relationship between the oil producers a d other market actors were discussed in Chpter 2. Thus, the first oil shmk is less rdetrant to the discus-

Price a ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

2 72

sion h this chapter. Aspeds relating to the relationship between oil producers inside m d outside of WEC are discussed in Chapter 8. 5.1 The Redundant Cartel: OPEC Price Policy, 1973-1981 T!ze Firsii Qil-Price Shock

The Tehran-Tripoli agreements of 1971 implied a 21 percent price increase for Saudi Arabian crude (from $1.80 to $2.18) and m increase in revenue of 38.9 percent. M a t was more importmt however, was that the producer countries gained controll over the price-setting (see section 2.4). In the early seventies there was a desircr among radicai Arab powers to use oil as a weapon against Israel, particularly through the United States%creased dependence on imported oil from WEC, This idea was not: a new one. As early as 1948, supplies of oil to consumer countric?~ were stopped in connectio~~ with the cortfljctbetween Israel m d the Arab countries. The s m e happened in 1956 and. 1967' (5ehneid.w 1983:21.2). Foremost m m g adtlocates of this policy were Iraq, Algeria, and Libya. The only country in opposition was m importmt on audi Arabia (see section 6.3). The decision to increase oil prices after the oil embargo of October 3973 was not an WEC dcdsio~~, one =ached by orEc mennhers indivjdt~ally (and to some extent by the OAPEC countries), or by some kind of tacit coopuation between OPEC members. The org""ization, OPEC, was not the key instrument for facilitating the price increase. It was probitbliy not even necessar)., as the countries most likely would have behaved in the same way without any institutional constraints on their behavior (see Chapter 4). Nor is it necessary to have an hstitutional arrangement like OI'EC to make oil producers increase prices as long as their individual production can be sustained at the same l e d with the new higher price.. "OPEC, therefore, found itself, laxgely by accident, with a price of $II instead of $5" "keet 1988:150), OPEC did make some reluctant members, particdarly Saudi Arabia, follow along, and restrained others, such as Iraq and Algeria, from increasi.ng prices even further.

Although oil consumption was rcgaded as inellastic, the market experienced a slowdtlwn in demand after the price increases of 1973-197J.. 'The o r E C tOuntrjes had to absorb it. all, and their production fell, by 3.5 rnbd horn 1974 to 1975. The industrialized economies experienced an economic recession creating high rates of k~flation.This reduced the value of the price increase, callsing some ~ E rneMbers C to call for further price

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

2 73

increases. Saudi Arabia, which had reluctantly agreed to the 1973 ernbargo, was strmgly opposed to further price increafies. ""Having devoted much of 1974 to the adjusment of the fiscal relationship between host eats and fort.ign concession . . . the OrEc corntries were f a c d in 1975 with. the need to adjust the official selling prices. . . . Against this background, attcntion focused on the relative market s h a ~osf t;he dif erent: mmbers of the Organization" (Evans 1990:466). In 1975, Saudi Arabia produced above 7 nzbd, and Xran produced 5.3 nnbd. meir combined p d u c t i o n constituted more than 40 percmt of total OPEC pduction. At an extmordjnary meeting of the orEc Conferelnce from Septfzmber 24 to 27,1975, Iran advocated a price increase of 15 to 20 percent, while Saudi Arabia argued for a price freeze. A 10 percent increase was agreed upon for the next twelve months. me13 the OPEC members met in May 1976, the orfyanizatim" Economic Commission had estimated that to compensate for inflation the price would have to be raised by another 20 percent. Eight countries favored a price increase: Iraq, Libya, Nigeria, ]:ran, Gabon, Qatar, Indonesia, and Ecuador. Three countries proposed a 10 percent increase: Mgeria, b a i t , and Vcmezuela. Only one cowtry opposed any price increase: Saudi Arabia. 'The decisim was postpolled until the meting in Doha in December 1976. At the Doha meeting Saudi Arabia proposed a S percent increase, supported by the Uilited Arab Emirates. The members did, not agsee, and the othcr ca~ntrkl;increased their oil price by 15 percent, while Saudi Arabia and the Unjted Arab Errzirates hcreased theirs by only 5 pexent. A two-tier price structure was thlrs established. After the meeting Saudi Arabian Oil Mirrister S:keik A h e d Zaki h a n i Ihreatened to increase production: '*We will m o v e the production ceiling of 8.5 million b a r ~ l as day. We will damp the market-it means the wholc strwctuce of price will collapse ali over the world" (cited in-Ferzian 1985:244). Saudi Arabian production did actually increase from 8.5 mbd in 197%to 9.2 mbd in 1977, In 1977 km, too, advocated a price k e z e . Whm these two count-ries, which together accounted for a full 48 percent of ~ I ~ E C 'total S production, oppowQP"ice fncreases, there was little the other cowtries could do. Hawever, unexpected events were to play havoc with the oil market.

The second oil-price revolution connected with OPEC ~ccurredin 1979 and 1980. Agaill, this had less to do with QI~ECthan with individual OPEC members and reaction among the other market actors (such as the international oil companies). Moreover, to understand the internal ortc prwess leading up to the quota decisim in ikfarch 1982, one needs to understand the way orEc handled the so-called second oil shock, that is, the

274

Price a ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

effects on the hiernational oil market of the Iranian revolution and the outlareak of the Iran-Xmy War (see section 3.3). In January 19% demonstrations begm in t.he holy city @m in :Iran. In the course of Octobet; 37,000 oil workers went: out on strike, reducing Iran" oil production dramatically, from some 6 mbd in August 1978 to appmximatefy 1250,000 bd in Februay 1979. Production rapidiy picked up to some 3 mbd, but this still meant a halvhg of Iran's pductiun. After the shah left Trm in January 1979, the new regime threw out all foreign oil companies and personnel, It had no in-house expertise to handle the loss of pressure that occz~rredin severat oil fields*The former production capaciv of well abwe 6 mbd was not possible to regain without new investments. Several OPEC corntries had at that time a deficit in their trade balance, and pressed for an incmase in OIJEC'S oofficial selling price. Saudi Arabia, not wanting such a development, increased production from 8.4 to 10.4 mbd, a figure that appaelncd or even exceeded sustainable prodtrction capacity. Moreover, most other OIJEC countries were already producing at almost full capacity I'hu?;, the other OPEC countries were unable to prevent a furCher price rise by increased prodLtction, even if fhey had wanted to do so. Mlith the disappearance of Iran"s production, stocks increased tremendously as consumers ensured supplies in case oE future shortage:" The rush to build inventories by oil el-rmpanies, reinforced by cmsumers, resulted in an additional three million barrels per day of "demand" &c)ve actual consumption. When added to the bvo million barrels per day of net lost supplies, the outcome was a total shortfall of f-ire million barrels per day, which was equivalent to about 10 percent of consumption. (Yergin 1991368'7)

This pushed the spot price sky-high. When the :Iranians went on strike on October 1,S,1,978,it took only five days tt?brkg the Iranian oil production to a complete standstill (Erzian 1985:257). However, the National Iranian (r>ilCompany (NW) managed to continue sdling oil off its stocks on the international market for some months. At the same time, Saudi Arabia m d some other OPEC countries incwased their saks of oil, f i l h g the gap left by the lranian shorthll (see Eiglare 5.1). nereforc., there was no actual pkysical shortage of crude oil during the rest of 197%.Mowever, the market actors perceived a h r t a g e as the price began to climb already in October 1978. In January 19;791, the Iranian oil sales came to an end. Total orEc production fell according)y to 28.4 mbd, 3.6 rnbd less than in September 1979, The spot market price increased rapidly, as Figure 2.12 shows. 'T'he OI'EC countries fncreased their official prices foltwing the increase in the spot price, alt.hough not to the sarnc level. The dis-

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

2%

crepancy betwee11 spot and offic-ial prices was to some extent fjlfed by different k h d s of commissions and ""black money" transfers to oil officials or authorities in the diferent produtling cowtries (Terzian 1985:26%265). As the events evolved in the first months of 1979, OPEC, on Febmary 21, 3979, summoned an extraorclinary meeting of the Confemnce cm March 26." week later, on February 28, the 9cretariat issued another statement: ""Conference decisions in setting crude oil prices do not prevent Wmber Countries from &kg an upward acljustment in light of their prevailing drcrurnst..ances. . . . Xn the present drcumstmces, the actio~zsof Member Countries in exercising their sovereign rjghts c m not be construed as pr4udickg the sotidarity and univ cJf OPEC" (WEC 3990:16,7). At the meetin,g on March 26, the rnarkcs crude price was raised from $13.335/baml to $14.546/harrel from April I. At this t h e the spot price was above $20/barrel, But the Conference confirmed the above statement regarding upward price adwst~aents:"Kesides this adust.xnent-[the changed marker crude], it is left for each Mcrnber Country to add to its price market premia which it deems justifiable in the light of its own circumstances" (OPEC 191)0:164). The Co~zferencealso made m appeal to the consming countries to restrain the international oil companjes: "the Conference calls upon all consuming countries to take such measures as to prevent oil companjes from charging them prices beyond the price decided upon by the OPEC Conference" ((OI'EC 1990:164). This shcrws that the organizaticm did not have any measures available to curb the upward pressart! on the oil price. While the WEC members were free to add to the agreed prices as they saw fit, the organizatim cailed upon the ccmsumers and companies to h d d back the price escalation. A w o r b g cartel has to be able to counter a price increasc with increased production, as well as to cut production when the price falls. The 3979 OPEC decision ~ f e r r e dto above shows the orgmizationfsinability to handle a si.tuatio11 where the oil price incrclased rapidy DiSfcrent authors eznphasize difierent factors explaining the severily of the s e c 4 oil crisis. Terzian (1985:263) suggests that statements Zly fanles khlesinger, the US Encsgy 9cretary, added to and su:bstantl.ally aggravated the crisis. Yergin (1993:68M87) emphasizes the stock-building as an important factor making the crisis more severe: "if companies thou@t that prices were goiing to go up, they bought more of today's cheaper barrels so that they would have to buy less of tomorrow" morore expensive oil. And that war; what happened, wit11 extraorciinary vengeance and .fury, inthe panic of 1979 and 1980.'' L w y (1982:IC)W)emphnsizes the lack of col-rerent political action on behalf of the importing countsies. Neither nationally nor internationdly, through the fnternational Energy Agency (IEA), did the c a n s ~ ~ m h countries' g governments try to restriet the price

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

2 77

increase or the panic-buying behaviar of the oil companies" "Mile it is,

of course, not certain whether any national or international arrangements, had they been made, would have succeeded, in actual fact mthing was even attempted to stem the tide oE upward-syiraling prices. It woufd, appear that most of the wounds of 1979 have been self-kfiicted" (Levy 1982:1000). Former Saudi Ar&im Oil Mnister Sheik Ahmed Zaki Yamani, in hindsight, views the events as follows: " ~ r o u g h a u t1979 continued high demand in the west and the peak prices being paid on the spot markets were creating chms. Oil cclmpanies were rushing in to buy. Rut this w a not ~ oil they needed. X t was for storage. X asked myself why and the only possibic answer was that they thougfiit there could be a totd cut-& in oil exports from :Iran. . . . So them was ail this panic buyingff(cited in Robinson 1988:225). He suggests th& the sdution wodd have been for the consuming countries"governments to forbid t:he oil companiefito trade in the spot market. An effective cartel organization would also have been able to prevent the crisis, as it could have prohjbited sales at prices bigher than what would clear tbe markt. As it turned out, only Saudi Arabia stood firm on official prices and traded mostly in the contract: market, but the orF;c Conference did not limit the upward price mwemmt, so the osgmizatiltn did not play any vital part in this ~ s p e c t . h May 1979, Iran was agah selljivlg oil on the international market (see Figure 5.1). The physical. hurtage had iasted only four months, OIEC met again in June 1979. Althou* Saudi Arabia had changed its poriition and was willing ta agree ta a price adjust~~ent ($18/barrel), the other cou~ztries wanted far more. The compromise was the following arrangement: Tn an endeavor to bring some stability to the market, the Conference decided on the fotlwing: 1. To adjust the marker crude price from the present level to ~ ~ $ per 1 8ban-el. 2. To allow Member Countries to add to the price of their crude a maximum market premium of vs$2 per barrel over and above their normal differential, if and when such a market premium was necessitated by market ccmbitions. 3. That the maximum prices that can be charged by Member Gctuntries shall not exceed US$ 23.50 per barrel, whether on account of quality and location advantage or market prernia. (CI~BEC IWO:465)

lan Skeet (1988:lh0) concludes that "in terms of price management these tcvo m e e t a s of OI~ECwere an adznissim that they neithrr had, nor wished to have, any control. Cur& bhnche was given, or taken, to obtain the highest possible price, The inclusion of a ceiling price in, the J m e communiquk was a weak effort to =strain the worst excesses of price greed, but it had no practical effect," 'The world obviuusiy look different

2 78

Price a ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

to different authors, as Terzian (1985:269) reads the same eveznts as follows: "This 'ceiling pricekas quite a development. . . . in this month oi June 1979, OI'EC was not only not shork~gup prices, it was actually hoZding them down by imposing a 'ceiling.'"' It is, hokvever, hard to HIld any evidence that the cei1in.g price had any effect whatsoever on OPEC members%ehavior. Skeet%hterprt-tation seems more adequate. h 9ptember 7,979, the private oil companies increased their stock, a step that led to a new upward mnvement of spot prices, This was iolcountries, Mthich i n c ~ a s e dtheir contractuaf prices by htwed by non-0121~~ adding substantia1surcharges to their selling prices (Terzian 153853271).In October 1979, the OITC countsies started breaking the ceiling of the June meeting, increasing their selling prices. On November 4, the US Embassy in Tehran was occupied, and a hostage-takhg crisis evolved, followed by a full us economic boycott oi Iran. On Paovember 20,100 Muslim fundamentalists occupied the M - H a r m al Sharif Mosque in Mecca. It took the Saudi governmernt two weeks to get them out. The stability of the Saudi regirne was caIled. into question, adding to the insta:billty of the oil-market situation. 'The orw meetit~gin Caracas in late December 1979 displayed a cmfliclt between Saudi Arabia and the United Arab Emirates on one side and Trim and the African producers on the other, with Venezuela and Kuwait trying to mediate (Terzian 1985:275).Saudi Arabia would not accept any official price higher than $26/barrel, while Iran would not accept any price lower than $32/barret. The meting did not resolve the dif-ferences, and the OPEC ministers left without any agreement. Yamani w m e d that Chc orEc production, at that time 3 mbd greater than before the 1raniar.l mvolution, was c ~ a t i n ga glut in the market. There was no increase in the underlying demand, so the increased production went into building of inverntories. When the oil compaslics felt rnore secure against future supply disruptions, these barrels would come onstream. M a n i prcrdicted a price f a : ''Iknew for sure that the oil held im stock would eventudy come back onto the market as a source of"supply"' (cited in Robinson 1988:216). Before leaving Caracas, Yamani dectared, ""There will be a surplus of oil on the marht: the pricemill fallr"(cted in Terzian 1985:275). h January 1980, Libya increased its oil price by $$/basreI. Smdi Arabia increitsed its oil price by $2/barrel to aliw its price with those af the other omc members. This f d e d , as the others immediately increased their prices by the same $2/barrel (Rrzian 1985:2?5). The same happened in May 19Kfl, bringing the official Saudi Arabiafi price up to $SO/barrel. In the second hdf of 1980, the increase in spot prices came to a halt. So did, the vlcrcases in the officid prices of the OPEC countries. The twentiministers met in Viema in the middle af Septembec With or.~c% eth anniversary coming up, it seemed a good ti,me to standardize prices.

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

2 79

Saudi Arabia once more increased its price by $?/barrel, and this time the others remavied unchanged. A s u m i t of heads of state was scheduled for November 4 in Raghdatl, with the pmspect of a renewal of the organization. It was never to take place, since, on September 22,1980, the Irm-lsaq W;ar broke out (see section 3.3). The outbreak of this war did not affect the oil market in the s m e mmner as did the revol~~tion in Iran. At the time the war broke out, there was a supply surplus in the market of about 2 mbd. The outbreak of war led to a =duction in both countries' production of some 4 n-tbd, leaving only 2 mbd of actual shortage. This was easily replaced by other OPEC and non-QITCcountries"roduction increases. The main eclonomic consequmce for the oif market was the more lasting disappeamw of prtlduction capacity of some 3 to 5 mbd (Bj~rk1988). The exceptionally high oil price accelerated the drop in consumption, This brought OPEC prduction down (t;igurc. 5.2). C)n December 15,1980, the orFc Conference met in Bali, with representatil~esof the two warring countries attending. A substantial surplus had emerged in the market, even with Iran and Iraq at war. At the same time, several OPEC countries used the political situation to increase their selling prices. This caused Saudi Arabia to use its idle production capacity as a threat, in order to achieve its long-sought goal of a unified price structure.

Saudi Arabia, at the time producing above 10 mbd, refused to lower its productim unless the other countries lowered their prices. Saudi Arabia was creating the surphs situation to force the other countries to b w c r their prices. This conflict cofiinued at OPEC'snext. meeting in Geneva FR May 1981. The press release issued &er the meeting read as follows: "The Conference . . . decided to mak~tair~ the deemed marker crude price at a ceiling of rrs$36/b with a maximun ortc price of us$ 4I/b until the end of the year, The majority of Member Countries decided to cut production by a minilnum of 10 percent, effective 1st June, 1~ ~ ~ " " ( R P1990:194). Ec Saudi Arhia refused, and Iran and Iraq were excluded, from the production cuts as they needed money fnr their warfare. Although Nigeria, Venezuela, Kuvvait, and @tar cut their production by mare than 10 percent, the market was still in a surplus situation. Saudi Arabia continued to produce about 10 mbd, making buyers sLvitch from the more expensive sellers to Saudi Arabia, which, as in past years, kept its prices weil below the others. T l e officiai selling price of Nigerian oil was at this time $40/basrel, and the Smdi Arabian price was $32/barrel. Subseyumtly, some of the other OPEC members experie~~ced an imposed cut in product.on, as there were no buyers for

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

282

their crude oil. Terzian (19635:288) puts these forced cuts in production at 30 percent for Algeria, $0 percent for Kuwait, and. 50 percent for Nigeria (all 1983 figures). In August it becme known that Nigeria was givillg the buyers a $3 discount, and thereby de facto reducing the selling price, The oil ministers were s u m m e d to a consultative meethg in Geneva, August 39 to 21, with the intention to convert the meeting into a full, c o n f e n c e if they reached an agreement, A unified price was the ubvious goal of the meeting. A compromise of $35/barrel was put on the table, and eleven corntries agreed. Iran refused and demanded $36/barrel, while Saudi Arabia. insisted on $34ibamel. The meeting ended. without any decision. However, the prohtems got worse. LZiith ~ d u c e ddemand for a9r?coil, and the indecisiveness of the orec meeting, the companies foreseeing lower prices started to =duce their stocks of oil. By the same logic when prices inmased, there was no rttason to keep large stocks in a s q l u s mnrket with falling prices- 'The destocking increased the d o m w a r d pressure on the OXTC official prices. By October even the high-price countries agreed that it was necessary to have a united QIZC. An extraordinary meeting of the Conference, held on. October 29, 1981, decided as hllows: "The Confemnce . . . recognizing the necessity to adopt a unjfied pricing system for OPEC crude . . . has resolved to set the official price of thc marker crude at US$ 34 per barrel" (OPEC1990:199). This decision was exacSly the same as the Saucti Arabian proposal, in Geneva. two months earlier, This m a r k d the end of this period. At last the Saudi &hian rjght for a unified OPEC price structure had succeeded, However, the reason for the other countriesf ahidance was not any kind of organizational discipline or the negotiating skills of the Smdis. It was the rnarlcet forces of ~ducreci demand for orBc oil, which left the countries with no other options. There was therefore no ""rle of OI'EC~"kin the process until 1981, in the sellse that the states9rice m d production behavim in the market was a result of hstitutional factors or collective action. The increase in the oil price in the first half of the sewenties was a result of individual countriesf utiiizing Che opportunities the market created (see Chapter 2). The market conditions were such that the actions performed by the members oi OPEC wodd most likely have been the same without the prtrsence of the organization. The period between 1978 and 1981, when the market changed, created conflict a m o n g the 0 1 ' members, ~ ~ and the coliective action irt 1981 c=ating a new unified price structure was also a result of the m r k e t forces. But by then there was a change in OPEC% role in the international oil market. WhiIe the market forces continued to press the price down, the

Price a ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

282

OPEC countries, havixlg experienced this second oil shack, had to cut back production to sustaisl prices, This activated the secmd. and third aspects of cartel themy outlined in section 7.2-the determfnation of a productim Level for the group and the docation of output among the members of OPEC,

5.2 The Effective Cartel: OPEC Quota Policy; 1982-1985 T!ze March 1982 Decision

The OI~ECdecisim of October 1981 was not sufficient to keep prices at $34/barrel. As can be seen from Figure 2.12, the spot price then was below the official price and thus pressed prices down, in the m e way as the spot market had increased prices during the frani.an revolution just three years earlier*hstead of surcharges, discounts became the order of the day At the Arab Energy Conference inDoXla, on Masch 6,1982, Yamani said, "Uestocking has ached unprecedented Levels: my experts estimate it at 4 million barrels a day. . . . We have to take m initiative. There can be na question of altering the price structure it took us so much trouble to set up again: on the other hand, r r y country is prepared to make an effort in terns of proda&ion'"(quoted inTerzian 1985:296). Two weeks later an extraordinary OPEC conference was heLd in Viema. The press rdease h m the Sixty-third or9r?cConfewnce read as follows: "After having examined the cursent market situation, it was decided by the Conference to reconfirm. the price of the Marker Crude, Arabian tight, 34 m", ex-Ras 'fmufa, at $34 per barrel . . . and to take the necessary measures for stabihing the market. For this purpose it was dccided that, as of April 1,1982, the total OITC production will have a ceiling of 18 d i m b/d. This ceiZing will he =viewed at the next Meeting of the Conkrence i,n thc light of market: dcvclopmcnts" ((or~c1990:2632). Thus' Ihe price target was kept intact while the production was cut. Saudi Arabia was not gken an exact quota, but was to act as a swing producer. Saudi Arabia also unilateratly cut: anot:her SC)(3,0t10 b/d shortly after the meeting "in order to emphasize that this cut-like all Saudi production becisitms-was ~ g a r d e das a sovereign act outsiete the formal scope of any OPEC agreement (Evans 1998:601). A1thougfn the international oil companies tried to put p ~ s s u r em the country they percri\/ed to be the weak link in the chain-Nigeria4-the unity of OPEC was more sdid than ever before. The trouble was that OrEc pesceived the probtems to be temporar)~Demand continued to be weak. I-lwing to live with temporary production restraints was tolerable, but havirtg to face such quotas for the foreseeable future and simdtaneousiy

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

283

havhg to cut prices was very difficztltt. As Fig~tre5 2 indicates, although o13EC reduced production substantially in the first half of the eightks, the price did not pick up. Ch the contrary it continued to slide. The quota sydem bad come to stay, and reduction of official priees was unavoidable. Several OPEC members had already lived with the situation of being unable to sd1 all the oil produced, and thus being forced to cut back on production.. From their point of view, it wodd be yrcferabk to let OPEC dwide this rather than leave it in the hands of the buyers. The stage was set for a rwiew of the intemaf C ~ I ~ Ebargaining C over production quotas. C)ne could expect the organization to have been instrumental in settjng up criteria fnr the allocation of production quotas. The fact is that the initial v o t a s of March 1982 were not: based on any criteria at all: When the prduction proFamme was adopted far the first t i m in 1982, it was not based on a careful study by CIPEC, in light of each Member Cc9mtry1s position in the world market and its oiX and development potential. Neither was the quota system based on certain objective criteria which could be accepted by all. Subsequent attempts tcr identify such criteria show that there are h a d t y any on which Member Countries uniformally [sic]agree. The system was first established as a result of negotiations made more or less on the spur of the moment and influenced, to a great extent, by Member C1ormti-y prc>ductionIevefs which prevail& at the time. (AI-Chalabi 1989:25-26)

Contrary to what Fadhil Al-ChalaE7i clairns, even tbr criterim of present production was only to some extent determining the first quota distribution. Figure 5.3 shows the previous production and the dationship between it and the initial quotas of March 1982. If there had been an eyual percentage cut the Zhe would be flat; if there had bee11 m u~~derstrndh~g that large producers should. get a disproportionately larger cut, the line would foll~wthe colu s in the figure. Na such correlation exists." At the morne~ntthe qzaotas wcre established, the allocation was disputed. Zrm did. not accept the 1.2 rnbd quota alocated at the March 1982 meeting. The country had the intention of increasing its oil production to 3 mbd "whatever the cost" "ranian Oil Minister Mohammed Gharazi, quoted in Terrzian 1985:301). The war with fray created a need for income, ~ ) ordered to s d i as and the Nation& Iranian Oil Campmy ( ~ 1 0 was much oil as possible, at m y price. "Oil was sold on the spot market at prices $4 or $5 below thtr official price for the eqttirralent Saudi crude" (Terzian 3985:31)2). Iray was hifidered from following the same poiicry, as its production capacity was reduced to about the quota level by Iranian attacks on shipments through the Persian Gulf and because Syria cut off the pipeline to tbe Mediterrmean. Iran was able to increase its production capacity and thus violated the qraota ag~ernent,

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

285

At the Sixty-fifth ai.-EcConference in 'July 1982, Iran ""deranded a quota of 2.5 m b/d and that its increase should be taken from the Saudi share" (Skeet 1988:ltiK). Iran did not abide by its quota in the followhg months. 'The slatentetnt issued by the cc countries after the oil ministers" meeting in Satala on October 14 illustrates this: "The producers should h o w that the KC Ministers expect them to shoulder their revonsibilities and if they continue their misguided actions they will not be protected by the member countries from the ccc from the conseqtlences of these acticms" (Skeet 1988:189). In his opening address at the following OPEC meeting, the chairman of the Nigerian delegat.im, Pilhnji Yahaya Dikko, with clear reference to the above statement, said that ""treats are never a basis for coopem"tmm(Evans 1990:607). The Conference decided that the total qztota for 7,983should be 18.5 mbd, but the Conference was unable to agree on the allocation of national quotas, Thus, the 18.5 mbd ceilirrg was not convinchg to the market actors. By the b e of the OPEC Confe~nceh flccember f 982, the m r k e t did not have suificient room for 18.5 mbd of o13Ec crude, Several otE.rer producing countries had increased their produdion fn the times of high prices, and there was nothing stopping these non-OPECcou~ztriesfrom producing at full capaciw. AIlhough the total production was set at 18.5 mbd, addkg the individual quota clailns at the December 1982 meeting would give a 6 g w of at?otll: 23-5 mbd. Iran elairncd a quota of 3.2 mbd and called. for a cut in Saudi Arabian production to below 5 mbd. Iran also argued that the following criteria sE-zoulLiguide the establishment of national production quotas: (i) historical ouirput levels, (ii) size of oil reserves, (iii) size of population, and (iv) extent of current economic and social needs (Evans 1990:61)7).Venezuela argued for a national quota of ""at least 2.1 mbd . . . Libya not less than 1.3" "erzian 1"35:310). Iraq argued for a quota increase e ~ atol whatever Iran ubtained (Evms 1990:607) OX. the historical level of 111 percent of prduction, which meant 2.3 mbd (Terzian f 985:330).%e tl'nited Asab Em.irates clcmanded a quota of 1.35 &d. The Decermber meetjng ended with a decision that "OI~EC production for the year 1988 should not exceed 18.5 million b/d. Howwer, an agremetnt of establishing natior-tal yuotas for the distribution of that total amount would mquire further consultations a m g the respective ents. . . . every effort should be made by each Member Country to preserve the price structure and to stabilize the market conditionsff (WEC 1990:206), In fact, there was no institutional p ~ s s u r e on the individual m e d e r countries to abide by the 18.5 mbd production ceiIing; still, the previously a p e d price level was sust..ained,Xn a declining market, this obviously could not hold for long, The price on the spot market fell, while at the same time OPEC members begm to relate contract cmde to the spot price (see section 2.6).

286

Price a ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

At a GCC meethg on January E, 1983, it was decided to hold a consultative meeting of the OPEC oil ministers in Geneva on January 23 and 2.1.. This meeting was a failure; not even a press release was issued. There was no Rasm for Yamani or the Saudi Arabians to bear the political costs of pressing for a price reduction, as they were convinced that the market, or, morc correctly, the non-oxJ~cproducers, would do it for them in the fnllowing months, if not weeks. Ch-t February 38, the &ritishNational Oil Corporation (BWK) r d c e d its price from $33.marrel to $31).5/barrel. Nigeria had to follow suit m d cut its price to $30; "they also made clear that they wollld match cent for cent any UK price cut below $30'" (Skeet f988:191). Yanani hvanted a $4 cut: h the marker crude prjce fmm $34 to $30. Nokt; with the Nigerian oil at $30, the marker would have to go behtw that level due to the different y?xatity cif the oil. "This was the situation when orr;c began a series of cnnsdtative meeti.ngs in Riyadh, Geneva, Paris, md, finally London, The London meeting lasted twelve days, with mly the last hours m March 34 transformed into an official Extraordhary Colnference, OPEC's67th" ((Skeet 195212:191). The Conference resolved: 1. To set t.hc official price of the Marker crude . . . at us $29 per barrel. 2. 7i> maintain the existing differentials among the various owc cnrdes at the same level as agreed upcm at the 63rd (Extraordinary) Meeting . . . with the te~aporaryexception that the differentials for the Nigerian crudes shuuld be us$l over the price of the Marker Crude. 3. To establish a ceiling for the total OPEC production of 17.5 mitlion barrels per day, within which individual Mernber Country quotas were allocated. This ceiling is to be observed as an werage for the remaking part of 1983. No qztota is allocated to the Kingdom of Saudi Arabia which will act as a swing producer to supply the balancir-zg qumtiticls to meet m a r k t wquirements. 4. The Member Countries shall avoid giving djscom-rts in any form whtsoever and refrah from dumphg petroleum products into the world cJit market at price which will jeclgarcjize the crude oil pricing strwtrrrc3. The Conference agseed that the recolnntelnded oil prices are floor prices and the national production quotas are ceiling figums, (OI~EC1990:2(18) As seen in Table 7.6, Iran obtained a doubling of its quota at the March 3983 meeting, while Kuwait, Libya, and Venezuela obtained suhstmtial increases. The March 1983 agxement was to last longer than the March

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

287

I982 clecision. As the ceiling was an average for the whole of 1983, it is worth noting that during the first half of 1983 the total OPEC production was wet1 below the 47.5 mbd ceiling, while the second hali pushed p m duction above this level, as Figure 5.2 indicates. Most amr: members cheated either on the price floor or on the production quota. This fed to new prtrssum on the agreed price level. The Saudi iarabian buildups of floating stocks espetial.ly created dowllward pressure on prices.7 The o13EC Market :Monitorirrg Committee tried to impose some discipline m the mernber countries, but was eslpecially attackd by the Saudi iarabians, who tolerated no hterfcrence in their oil poticy. .Although the conflict between Iran and Saudi Arabia surfaced at the OPEC rneetir~girr December 4983,Qhe quotas were. not changed until the auturnn of 1984. At an e x t r a o r h a y meeting from C)etober 29 to 31, the price floor of $29/barrel was defended, but the total production quota. was reduced by 1.5mbd from November 4 (OI~EC1990:ZO). Just days before?the meting, Norway had cut: its oil price (see sectinn 9.2). The focus was thus more on conflict between outside producers and OPEC than on internal O19EC disagreements (see Chapter 8).

Saudi Arabiu: Swinging Lon? OPEC'sOctober 19884 decision and Saudi Arabia" actions as a swing producer during 1984 and 19% created fairly sta:ble prices in this period (see Figure 5.2). However, Saudi Arabia had to cut back on its production to sustain prices. This policy was impossible to mairrtain indefinitely. The Saudi oil production fell simultaneously with the prices. The loss of market share also gave Saudi Arabia a more rnarginai mle fn world p~litjcs and in Middle Eastern &airs (see Chapter 6). 7;hc characteristic feature of this period was that OPEC was able to counter the downward pressart. on prices. There was liefinitely a ""mie of OI~EC'"in this pcrlriod, as the market out-come would most likely have been d i f f e ~ ncvithout: t the organization, However, randerlying the orgmization" sde in this period was the willinpess of Saudi Arabia to bear the costs of performing as a benevolent hegemon. It was Saudi Arabia that cut its own production to sustain pries. hoking only at the market behavior, it is thus fair to conclude that the "role of OI~EC'"was to a large extent the role of Saudi Arahia (see section 6.4). This would, hocvever, miss the indisect effcct of the organization, Without the institutional framework established withh the OPEC ortja"ization, it is less likely that Saudi Arabia would have followed the strategy it did. Without any quotas, the role as scving producer wndd have been meaningless, The fact that the strategy did not succeed for longer than a few years is due to a combination of stmctural factors (developed in C h p t e r 2) m d the orEc decisions. 'The price target of $32

288

Price a ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

flater, $29) was too high given the incsrrased (long-term) demand elasticity and the increased number of non-o12~c producers following a competitive strategy*We are back to the interaction between marltet structure m d market behavior e~nphasilzedin Chapter 2-It is briefly summarized by the following contesnporary comment: The story of the rise and fall of orsc is starkly illustrated in what has been happening to oil prices and to shiMing supply and demand in the world market. The falling oil cartel, like virtually every cartel before it, committed the blunder of setting its monopoly price too high. This simultaneously called for vast supplies from outside producers that the cartel could not contm1, and eventually induced consei-vatianby cmsumers. 9

h the summer of 1985 there were definite siws that the Saudi Arabians would no lmger accept production beyond the quotas by other OPEC countries, as tong as they themselves produced only half their quota." In October Saudi Arabia shifted policy and. increased, production by introducing the so-called netback pricing system (see section 2.6). 'This led to considerable over-production in the market m d heavy pressure on spot prices, a ciscumstance several countries tried to compensate for by increasing pmducticm, In this atmosphere OPEC gatherltd for its meeting in Geneva on 9 December, a meethg that was to change yet agaln the direction of developmnt in the oil market. 5.3 The Defected Cartel: OPEC Market-Share Palicy 19861998

Changes Mnrkef SlvategI 2 985 At the meeting in Geneva irr December 1985, OI~ECmsolved to change its market strategy from the dcfeflsc of a high oil price to the delense of the OPEC countrieshmket shares, ""Having cmsidered. t:he past and likely futurl, developments in the world oil market and the persistenty d e c k ing trend of orEc prodwtion, the CQII~CSC~C~ decided to secure m d defend for o12~c a fair share h the world oil market consistent with the necessary income for Member Comtriesf development.'"J The OPEC members drew just as much atte~~tion to the producers outside ~ P E Cas to those wit%iinthe organization who were fasing to keep to their quotas. 'Thus, strong complaints about nm-members, and threats of a price war against the~n,were expressed: "czrt;~still hr-lrbors Che hope that other producers will cooperate in trying to maintah prices by curbing their output. R u t irnplicit in the cornmuniqu4 issued yesterday is the threat of a price war if they do not"12 A confirnation of this intention OPEC

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

289

was given in a press release from Yamani some two months later: " M e n the OPEC member Conference, held in Geneva last December unanimously decided to secure and defend for tbr Organization a fair oil market share, specm,lativesales in the rnarkt began to push oil priees downward as it became clear to all that if non-OI~EC member states did not ~ v i s their e policies and give up some part of their market shares to O ~ ~ E C , there would be a production surplus which woufd lead to a fall in oil prices."lVhe o~anizalionnow had a twofold task - to ensure illternal discipline and to increase the creciibility a d the perception of the seriousmss of the threat posed by other oil pmducers.l" West Saudi Arabia first took on its role as a swing producer in 1982, the need for h t e d discipline eased. So did the importance of the m i toring committee (see section 4.3). However, as Smdi Arabia changed its market strateg in 1985, and the oil pl-ice fell in the spring of 1986, the need for hternal discipline and mmitming institutions returned. As Figurc 5.4 illustrates, the initial respctnse of the other orEc members w s not to reduce production as a result of the Saudi Arabian threat. On the contrary, the other OPEC countries hereased production during tbr spring of 11986 to compensate for the price fall. Nor did the non-OPECcountries change their policy as a result of the threat, As described in Chapter 2, the price fell dramatically, and the world oil market was in a situation of a price war between all oil pdu,ccrs. The threat had failed (see Chapter 8 and section 9.3). At the December 1986 meeting of tbr OPEC Conference, a new accord was agreed upon for the first half of 11987. Now thc price issue was rclsolved by the i n t r ~ d ~ coft i ~ ~ the WEC basket. As the basket included several different WEC oll yualitks, the ~ f e r e n c price e was &at of a composite of WEC oil, rather than the single rnarktr crude which, had been used as a rckrence inthe seventies-Furthermorer the price diffesentials were fixed., and a system for revision oi price policy and monitoring was estabtished. The reintroduction of Irag to the quota system was discussed, but lraq would not accept anything less than parity with the Iranim quota of 2.255 mbd, as of January 1987. The pevious Eraqi quota of October 1984 was 1.2 mbd. 'The deaeStock was resolved when Iran dsopped 2s demand &at Iraq be g i v a ~a binding quota, but Iran "made it clear that it reserved the right to reopen the issue if Iraq failed to respcllct a new 'ddt-med' quota'" (Evans 3991):7d41).

A b U . ~ d o ~ iQZIQ~US ~ t g D ~ i r i ~thex Iraq-Ktlwaif War In July 1998 the Iraqi leader Saddam Hussein accused Kuwait of "stealing" oil from Iraq and demnded repayment, declaring that "'Zrayis will never forget the maxi~nthat cutting necks is better than cutting thc. means of I,iving.""lVheIraqi .foreign minister wrote a letter to the secre-

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

291

TABLE 5.1 Regicjna'i Crude Supply Gains and Losses, August &-October 20, 1990

I ~ ~ t ptcr or US Europc japan Asia Kuwait/lraq (losses) - 1 , -1,020 -480 -425 Uthers (gains) -1-660 -1-830 +958 +622 Balance -510 -190 1-478 +l97 WCRCE: Z"ZetroleunzIn telligei-lce Weekly, October 22,1990.

Ofher

Total

-385 +350 -35

-3,480

+3,420 -60

tary-gmeral of the Arab League, stating that he regarded the stedhg as "a direct miitary attack on Iraq'band declaring the Iraqi "right to retum all the mmey that was stolen from uur oil fields."lh After the Eighty-sev&h OPEC Conference in July 1990, Iraq and K w a i t met for negotiations, which irnmecaiately broke down. A few days later, on August 2, Iraq invaded Kuwait m d subsequently declaed a "cmprchensive, eternat and inseparable mergerf3etween the two countries,'7The UN Scuriw Council declared a fult boycott of Iraq and the new regime in Kuwait. This immediately brought to a hak all oil Bows from the two countries. The oil price skyrocketed from $18/barrel to almost $3C)/barrel in a couple of days. C h August 29,1990, orEc issued a press rekasc stating that the organization would '"consequently increase production, in accordmce with need in order to maintain . . .market stataility and regular suppIy of oil to consumt.rs""(OPEC 1990:3111). This inplied an abandoning of the quota, system until further notice (see section 3.4, where the implications of the Gulf M7ar are discussed in greater detail). Other producers, particularly Saudi Arabia, quickly filled the gap. There was no shortage of oil. After a few weeks the WEC production was back at almost the prewar k v d . Some additional barrels from non-OPEC producers made the total market balmce. Chce agaiin there was some degree of panic buyhg. The Japanese oil traders increased their purchases beymd the prewar levelf creating a price rise as indicated in Table 5.1. The disappearance of oil from Kuwait and Iraq eased the pressure on the OIZC quota system, especially the Saudi restraints. The contijnous UN blockade of Iraq enabled the other OPEC countries tc-,continue a high pmduction poliry without dimpting the oil price. The loss of Xmqi production was almost entirely covered by Saudi Arabia, as Table 5.2 bows. Allhough the price Rurtuated over the period, with substantial weakenhg h both 1988 m d 1093, there is no obvious t r e ~ ~neither d, up'cvard nor downward, in the period from 1986 to 1936. When Saudi Arilbia, introduced its role as s w h g producer in 19M, the need to discipline the members through institutional mechanis~nrrswas reduced. The Iraq-

Price and Provodznetiolz Polr'c!! of OPEC Co&intries

292 TABLE 5,2 1991-1 994.

Iraqi and Saudi Arabian Production (mbd), July 1990, and Average,

Kuwait War and the subsequent UN embargo against Iraq bad the same effect. This gave the other OPEC producers r o m for increased production w i t h t r i s b g a fall in oil prices, as the Iraqi production-3.2 rnbd in July B90-was kept at almost zero.

MzclMZi~g'irhucrzlgh the 1990s At the meeting of the Ministerial Monitoring C a m i t t e e in February f 992, the quota allocations were reestablished, with K w a i t being allowed to produce as much as possible, Iraq was not allocated a quota but was fncluded at its currmt production level. In January 1993, new quotas were set, including one for Kuwajt (see Table 7.6). The UN sanctions agajnst Iraq continued to block ~ g u l a Iraqi r oil exports. The Scurity Council of the UN was in effect a part of the QI~ECproduction sharing efforts, and a very abiding one as . ~ wasl the UN curbed an ixrcrease in Iraqi exports that would otherwise have taken place. h the spring of 1996 the U N and Iraqi aufiorities worked out an oil-for-food deai. Iraq was to be alfokved to sell oil in order to pay for food and medicine under a UN-monitoreddistribution system. As tbe irnpkmentatim of the (fedseemed imminent, OPEC included Iraqi oil production of 1.2 mbd in the quota agrecmernt of June f 996 (see Table 7.6). This created an lncrcase in tthe total quota, as no other country cut back.lN Ch December 30, Iraq resumed export of oil under the oil-br-food agreement with the UN, Weather conditions, low levels of stock, m d the fact that traders had anticipated the return of Iraq for several months made the immediate price effects negligibk. The return of Iraq with a quota amund the Irmian quota of 3.6 rnbd would put the orEc quota agreements umder pressure, The reaction fmrn Saudi Arabia w d d be most important, as it was t-he country that repkced the Iraqi oil m t-he market (see Tablie 5.2). During the 1991)s OPEC also experience two cases of rnernbers leaving the organization. In November 1992, the OI~ECconference accepted the Ecuadorian suspension of membership effectively from January 1993.19 In July 1996, the orEc Conference accepted Gabon's ter.minat.ion of its

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

293

membership effectively from January 1995.2'"Amuzegar (1999:56) describes the Ecuadorian dilemma as follows: '"Facing an increashg need fnr fnmign exchange and having a small crude o u t p t Ecuador rczgularly exceeded its a r E c quotas, fclX behind in its metnbership fees and fhally left the organjzatim as of January 1993." What would be the reason for ally country to leave OIYEC? Given the prominent rnle of the production lirnitatiorls in the intemai OPEC bargains, it can be assumed that countries leaving the osmization believe they will be able to increase production further outside OPEC than fnside. Comparing the Ecmadnrim prodllction increase outsjde o r E c with the development of OPEC p r ~ d u c t i min the same period, such a strategy seems to have been une;uccesshl in that case. While the Emadorim production increased with 16.7 percent from 1992 to 1998, the OPEC com~triesh~creased their production with 17.9 percent in the same period., In the case of Gabon, which announced its intention to terminate its membership h 1994 a similar pattern can be observed. From 1994 to 1998, Gabon's soil production increased with 6 percent, while the O13Ec production in the same period has increased with 13 percent. 'These two countries were large ovcryroduccrs inside OPEC; measured in percentage of overprodzaction above allotted quotas, their small market shares made their lack of compliance a minor probtem. 'f'hese aspects will be discussed extensivq in aapLer 7%The bet that most: factors influencing the strategic xnarket policies of o i t p d u c i n g countries are the s m e for producers outside and inside of OPEC will be discussed in Chapters 8 and 9, A. more dramatic effect on the oil price followed the economic downturn in Asia in the autumn of 1997. The Mf in demand following this recession ilnplied a weaker, not stronger, market for OPEC crude. The sensiMe response would have been to cut production in order to balance the market at a lower level of demand. However, in November 19517, the OPEC countrie~increased the ~ o t a sThe . Brmt oil price fell from $19.88 per barrel in October 1997 to $13.11 per barrel in March 19998. M e n OPEC reduced the production ceiling again in March 1998, the new agreed production was not based on the quotas but on reduction from actual production the month before. Mabm (J998:10) argued that "or~c b d ] abandoned its production quota policy" "nce the price policy was abmdoned back in 1986, there was not much left of the market-goveming role of OPEC. Mabro (199613:lQ)thus concluded that: ""~PECtoday has neither a price nor a quota policy.ff t production reductim was iniaiated by In March 1998, an a g ~ e m e nat1 Mexico, Zienezuelr-t, and Saudi Arabj,a. It was set to reduce production with more than one million barrels per day The deal was immediately described as illsufficient to bring the market in balmce: ""Iholds out a reasonable promise of shrinking the global s ~ ~ r p l of u ssupply over de-

294

Price a ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

mand . . . what it probably won't do is pull glutted world oil market back into balmce overnight.""2 Iran, hdonesia, and Nigeria were not assumed to cut their production. Given its market share and pditicai importance, getting Iran onboard becane important. The Iranian hrc3ign minister visited Saudi Arabia and got s o m concessions. More innportant was that this was a signal of possible future oi-related cooperation between the two countries. In June 1998, prices dmpped again. This put pressure on the produccs cooperation, Now Saudi Arabia took the position than no new cuts should be initiated, llnless cmpliance with the earlier agreement was above 90 pescent," In October 1998, a meting of the so-called producerconsurner dialogue was held in Cape 'fbwn. Under the cover of this meeting the major oil exporters discussed further measures to strengtt-ten prices without any concrete measures being taken. The economic crisis in Asia, which began in the s m e r of 1997, had severc impact on the oil market. The Asian oil consun?ption constihkzs above 25 percent of total world demand. While the demand of North America and Europe has been fairly stal.lle dlariw the 1980s and 19905, the Asian demmd has increased by more thm 40 percent. ''h 7,998 . . . East Asian oil demand growth, which averaged 5.6% annually between 1990 and 1996, slowed sharplyf"(EEA 1999a). Instead of continued gsowth, the Asian demand fell marginally from 1997 to 7,998- The erconomic growth in Asia was seen as ""the rnotor for rising global demand fnr oil."'" OPEC,like most analystti, did not forcrscre the depth and persistence of the Asian economic crisis. However, the decision of the OPEC Conference in Novermber 1997 to increase quotas iced the pl-ice slide further. Wth the bemfit of hindsight, it is easy to see that the right answer in late 1997 wouM hnvc been to cut production, not increase it.

In March 1999, ~ I ~ Ea w C e d on substantial cuts in production." This agreement, a d the fact tbat it was negotiated before the meethg blew new life into the organization: "Wy exceedkg most assessntemts of the reduction required to balancer markets, OPEC has clearly shown its determination not to =peat last year" saies of cuts, which time again trailed the demand curveef"""e meeting was over in ten minutes. A. new negotiation strategy had been applied. Ministers from Algeria, Iran, Saudi Aracountry Mexico met two weeks before the bia, Venezuela, and non-OE~EC orEc conference and agrc3ed on a 2xnbd production cut for orEc and nono13EC countries. The OX~EC Conference was simply a confirmation of this agreement. Aka, by June, the implementation cJf the March 1999 agreement seemed better than ever before. "or~c'scrude oil production fig-

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

295

urcs for May have bsought morc good news for the prohcing countries, signaling a rise in the level of compliance with output cutbacks c o r n i t rnents by 10 OPEC meMber countries (excluding Iraq) to nearly 90% as compared with W% in April,"~"nother aspeet of the developments in 1999 was renewed talks about a firmer price target. In May,the Saudi Arabim Oil Mhister d i Nakni affirmed that Saudi Arabia was ready to defend a price target of $18 to $23per barrel.27 Later it was hdicated that this strategy was g a h b g support by several producers, both inside and outside ~ I > E C : A senior Gulf official who is familiar with Saudi oil policy has been quoted as affirming that members of CII'FC: together with some n o n - ~ exporters ~~~c are currently involved in consultations with a view to devising a strategy to stabilize crude oil prises in the range of $18-20 per barrel (for Brent) by intei-vening in the market whenever prices move outside the targeted band. . . . Saudi Arabia has taken a leading roXe in discussions with other optic and non-l=t131:cproducers on this strategy. . . . QPEC members plus non-L313ECMexico, Oman and Norway tzrere involved in talks to agree on a mechanism for market intervention.%

Both regarding pmduction and prices, the producer cooperatha seemed strengthened compared to the bleak prospects h3 1998. A long-time commentator on OPEC afhirs, Ian Seymour, lfound the conference to be more "cmdial than anythirmg experienced for some years pastf"and that '"ere is now a pre.tty strong conse~~sus . . . that defendk~gprices and revenues is more h p o r t m t . . . than boosting export volumes in an already sated mkt.'"ZgClr; in the words of the Algerian b e r g y Minister Uoucef Yousfi, acting presidel~tof am@: '"PEC is like a "hoel~ix risen from the ashe~."~M The renewed efforts in oil-producer cooperation, now including the nm-or9ec pmducers Mexico, Russia, and Nc.rrway, paid off. By the summer of 2000, the oil price reached $31)per barrel, It had gone from $10 to $30 in eighteen months (see Figure 5.5). With such a high price it beccrmes tempting fur the individual producer to sell some extra b a r ~ l beforc3 s the hsreased producrion from another producer weakens the market again, We are back at the classiral cOflf3~tIVeaction problem, which might be triggered once again among the oil producers. However, by mid-2000, many orEc m d e r s lacked spare capacity. Thus it seemed impossible to increase production in the short run, 'T'heonly countr?/ with srtbstantial spare capacity instded was Saudi Arabia. Thus the underlying strength, of the kingdom. increased during the price rise of 1999-2000. In March 21300, ~ 7 1 found ~ ~ ~ 1 the price to be too high to ensure a stable market developme~~t, m d thus reversed the March 1999 production cuts

Bn'ce land Pmduction Policy $OPEC Cc~zinfries

297

of 1.7 mbd.3 Iran disassociated itself from the a g r e e m e t r a r i l y because the leaderhip in Tehran disliked. the pressure exerted by the US ent on the OPEC countries. AS Figurt? 5.5 shows, the production increase had d y a temporary eMect on the oil price. 'Thus t?l. the planned but still extraosdinary Conference in June 2000, the quotas were raised fur&er by 708,000 barrels per day (see Table 7.69." This Conference also brought Iran formlly back into the quota system, Athough not explieitIy and fully published dficially, the two Conferences during the spring of 2 0 0 developed an agreement on a price band mechanism. The agreemetnt sets a target price of $25 per barrel and an obtigation to increase production by 500,000 b a r ~ l per s day if the price rises above $28 for twenty consecuf;ivetradhg days and likewise decrease pmducticm if the price falls below $22 per bard.33 t:n July 2000, the orEc S c ~ t a r i aissued t a statement in accordmce with such m agreement.34 This could be an hstrrtment fnr tighter market gwernance by OPEC, but by the t h e of this writbg ('July 2QQO),it is hard to tell if this is the case, F~~rthermore, before the Secretariat" statement was published, the dominant o13Ec mrrrber, Saudi Arabia, had unitaterally announced a 500,0(30 b a r d s per day h crease of its own. production (see section 6.6). From. OPEC's perspective, the wodd economic: outlook by mid-2000 seemed good, as the Asian economfes v v e ~ =covering, and the already u~lprccedentedsustainaetiity of the us gsowth seemed to go on. 'These trends might easily turn around and surprise market analysts, as they c does, howtend to do. 'Tlhe present agreement among the o r ~ members ever, seem less depende~lton increased demand than were previous agreements. The fact that several nm-or~c producers are ixrcluded in the negotiations also reduces the probability that non-OPEC producers should capture the increased demand. It could be that am@is finally coming around to tackle the long-term chaIhges, as outlined, by former Saudi Arabim oil minister &mad Zaki Yamani, Regarding the longer term, OPEC needs to devdop a strategy that keeps oil prices cmstartt in real terms at a level low enough to deter non-ojPEc producers from adding to their reserTes and keep worldwide oil demand grc>wing at a healthy rate, What this level shautd be, of course, is open to vigorous debate#but whatever it is it should ensure that the dernand for CIPECoil rises steadily to alasnin the short-term as requested by the organization. 15. Platf's Week,Juty 23. 16. Ibid. 2 7. X~zterrzcationaEHemid PiFzrne, August 2 0, 1990, 18. Gabon left the organization in 1996 (fctrmally, from January 4, 1995) and thus was no longer included in the quota distriibutian. 29. " m e Conference regretfully accepted the wish of Ecuador to suspend its full membership in the Organization. f-loweveu;recognizhg the current economic constraints facing that country, the Conference hopes that Ecuadctr will be able to overcome these difficulties and rejoin the Organization in the not tcm distant future." Ninety-second Meeting of the Conference press release no. 9/ 92-Vienna, Austria, Ncjvember 27,1992. 20. ""Te Conference has taken note of the notice submitted by Gabon on 20 December 1994, and cmsiders Cabon's Membership in the Organizatic~nterminated as of 1 January 2 995, in accordance with Article 8 of the Statute" Remlution No 100.350, July 7,1996. 21. ktrokcttm X~telli~ncr. Weekly* March 30, 19%. 22. Petroleum In fel'lipnce Weekly, August 3,1998. 23. Pefrot~umEwnumist April 1999:2, 24. Press release from the 207th OPEC Conference, Vienna, March 23,1999. 25. Pefrolrum I~afelligent-.e Weekly, March 29,199, 26. Middle East Economic Sunrq! June 2 4,1999, 27".Middle Eask Ecurzamic Szkrztq, May 17; 19%. 28. Middle Ensf Economic Sunrq! June 2 4,3999. 29. Middle East Ecalzurnic S~tlrvey,Marcl1 29,1999, 30. Middle East Ecorromic Sunrq! March 29,1999, 31. Press release from the Zmth OPEC Conference, Vienna, March 29,2080.

200

Price n ~ Prodznetl;olz d Polr'c!! of OPEC Cozrnt-ries

32.. Press release from the 130th (Extraordinary) OPEC Conference, Vienna, June 21,2000. 33. Middle Ensl Econolltic Sltrzley, June 26, 2000. Pefroleuvrt Inlelligelzee Weeklyf June 26,2000, interprets the deal as being more Ic~osely:"if prices gn outside the range for 20 consecutive days it will trigger a consultation of members to consider what action to take," 34. The statement has a more subtle formulation as it read: "production would be raised by 500,008 if the average price of the OPEC Reference Basket of crudes lez~elfora s~jec@cperiod of tinge" (my italics). It was furremained above a certai~~ thermore an early warning as it was stated that the cedain level and specific period of time tvas ""other things being equal . . . expected . . . [rot happen bef0l-e the end of the current month" (OPEC press release, July 7,2000). It did not. 35. Speech by Sheik Ahmed Zaki Yamani, at the Centre for Global Energy Shxdies and Oil and Gas Joamnl Joint Conference: Oil Price Challenges into the Next Century in Houston, Texas, September 9-40,1999.

Country Case I: Saudi ArabiaA Hegemonic Power

The aim of t b chapter is to answer Che folfo\vin.g pestion: I'o what extent has cooperation between oil producers been achieved by Saudi Arabia's performhg as a hegemonic p w e r in the international oit market? The hegemnic role of Saudi Ara:bia was one of the m s t prorninemt ieatures of the oil-producer cooperation, and thus the oll market, durtng the seventies and eighties. It is the clmventional trYisdom that Saudi Arabia holds the key to the success of OPEC and is, in fact, the most inRnemtia1 actor in the oil market. The argument presented in this chapter does not dispute this. However, some qualifications should be made. Two important factors are considered to constrain the Saudi Arabim role as a hegemonic power: changes in the market structure; and the Saudi Arabian rc-.gime%preoccupation with security, both internal and external. The underfyjng katures of these aspects were descrihcd in Chapters 2 and 3, and will, be taken as given h the discussion in this chapter, By focushg on one actor, it is possible to study h w oil interests and security interests interact in the f n r m t i m of a state3 policy. The empiricd discussion in this chapter tries to capture the dyrramk character of the Saudi Arabian hegemony. Accordingly, four liifferent phases will be outlined: 3, the period from 3973 to 1981, when Saudi hrabia sougbt to influelncc the oil policy of thc other orEc rnelnbers but was unable to do so (see section 6.3) 2, the pmiod from 3981 to 1985, when Saudi Arabia carried a disproportional share of the casts in order to provide the callective good-a higher oil price (see section 6.4) 3. the period from 1985 to the present, when Saudi Arahia first coerced the other orEc m e d e r s to chmge their market strategy in

202

Country Case X: Snzddi Arab&

1985-19636, then took advantage of the UN sanctions against Iraq to increase its own oil production (see section 6.5) 4. the situation in 1999-2000, when Saudi Arabia used both threats and promises in order to establish a new quota agreement (see section 6.6) The lundannental starting point is the thesis regarding the role of a hegemon in international cooperation: '"the presence of a single, strongly dominant actor in international politics lea& to collectively desirable o.ut-comesfor 1111 states in the international systern""(Snidal. 1985:579)." The hegemon is perceived to possess s m e structural p w e r resources, which are transform& "into bargair~ingleverqe cast in terms appropriate to the issues at stake in specific instrxnces of institutional bargaining" ( h u n g 1993:288),2 In applying this to the oil-producer relations, the point of departure is that all actors have an fnterest in a cotlective lirnitation of production in order to elnsure a lsigh oil price. At Che same time it is in the individual actors-interests to sustain their own production volume. tn such case, the actors can gain by cooperating (reducing production), but individxlab they can gain more by not cooperating bile the others do. This is the problem of collective action that will be discussed in Chapter 7. 'I'he argurnent pursued in this chapter is that the larger the dative size of the individual actor, the stmnger the incentive for this actor to contribute to the provision of the collective good. If one actor is so large that its consumption of the cdectke good is so profitable that it covers the costs of providing the collective good, this actor can be assumed to provide the collective good regardless of whether the other actors contribute or not. Olson states this point as follows: One point is imrnrtdiately evident. If there is some quantity of a collective good that can be obtained at a cost sufficiently low in rela"tic1n to its benefit that some one persc>nin the relevant group would gain from prcIviciing that good all by himself, then there is same presumption that the collective good will be prc~vided.(Olsc~n1965:22)

The starting point in the discussion of the role of a hegemon in estabfishing cooperation is thus: The larger the size barket sbare) of an actor, the stronger the tendency is to contribute to the provision of the colXective good (a higher oil price). It would, of course, be beneficial for the hrgemon to get other actors to cooperate, even though it would provide the collective goad without such cooperation. This, however, presupposes that the other actors have insufficient information. If they h o w tbat the hegemon will provide the good regardfess of their b&avi,or, nothkg would make them contribute-

Cc~zintryCase 1: Saudi Ambia

203

In such cases, the hegeznon would have to change the other actorsf costbenefit calculation by irnposixzg some kind of sanctions on nmcmtf-ibuting actors, It will become evident in the fobwing djscussion of Saudi Arabia thatthe hegemon has othr, noneconomic motivations for providing m economic collective good such as a high oil price- Chapter 3 revealed that t.he po:[itical m d secwity interests of Saudi Arabia caused it to follow a costly cooperative strategy toward the other OPEC mennbers after the Iranian rwolution and the outbreak of the Iran-Iraq Wr, This made fie country less sensitive to the ecoz~omiccosts of cooperation. It might even be argued that it was not until this strategy was perceived to have political costs, in addition to the economic ones, that the kingdom changed its oilh 1985. This pohts ta the importmt disthdian between market s t r a t e ~ a hegemon behaving benevolently and one behaving coercively (Snidal 1985:588): In the benevolent leadership model a greater absolute size of the largest actor means it has a greater intel-est in prc~vidingthe gocld. . . . In the ct~ercive leadership model, by cmtrast, it is relative size that is foremost. The key to centralized provision is the ability to force subordinate states to make contributions, and this ability rests primarily on the relative power of states. . . . The benevolent modet focuses primarily on interest, implying that capability follows, . . . the coercive model focuses on capability, implying that intel-est in providing the public gncd fc~llawsfrom the distributitic~nof capabilities. (Snidat 19&5:5813-589)

The two brms of hegemclnic p o w r to some extent presume different power resources. fn the case of a hegemonic oil producer aiming at sustaining a cartel or a looser producer cooperation, the benevolent skategy assumes the ability to cut back a substantial amount of total oil prtlduction in order to increase or sustah prices, 'The producer must therefore be an initially large producer. A coercive strategy assumes the ahiljt)r to increase production. fn the short-term this means available spare capacity; in the medium,-term, it means ability to incrclase production in existing fields from reserves readily developed. The coercive strategy also assumes an ability to survive the assumed low oil price likely to follow .from one's sown hcreased prodwtion. fn other \.vords,the production operating costs must be relatively l w cmpared with those of other producers. This distinction between benevolent and coercive hegemony is important for the foilowing empirical analysis, as it wilt be shocvn that Saudi Arabia is characterized by having a large market share, spare capacity, and low operating costs compared with other producers. The kingdnm has thus had the caphiljties to pursue both the benevolent and

204

Country Case X: Snzddi Arab&

the coercive strategy in different periods. The change of strategy over tirne calls for a dynamic approach to the Saudi Arabim hegemmy. 6.1 A Dynamic Approach t(7 Hegemany3

Robert Gilpfn f1983:30-14f has outlined five assumptions concerning states' behavior leadkg to what he calls international political chmge: 3.

An intematicmal system is stable if no state believes it pmfilable

to attempt to crihange the system. 2. .A state will attempt to chrrnge the htematioslal system if the expected benefits exceed the expected costs.. 3. A state will seek to change the international system until the marghal costs of furUler change are equal to, or greater than,the marginal benefits. 4. a c e an equilibrium between the costs and benefits of further change is reached, the tendency is for the economic costs of maintaining the status quo to rise faster than the ecmomic capacity to support the status quo. 5. If the disequilibrium in the international system is not resolved, then the system will be changed, a d a new equilibrium reflecting the redistribution of power will be established.

Gilpin argues that &re w i l l be a continuous struggle between "patentiat hegemonsf9fightingfor dominance. As soun as o m of them is in control, it will bring upon itself greater costs relative to the other actors. From the fourth assumption, it follows that the other actorsfposition will be stre~~gthened the longer the hegemon sits.. In, the end, the hegemon will fall, and, a new one will take its piace, Gilpin's line of argument seems rather precietermined, provoking a sear& for possibk altmative courses of action for the hegemon. This is because the hegemon faces mort. than one type of costs-it is costly to d e c h e as well as to go on as before. Gdpin" iourth asswpticm, stating that the costs cJf preserving the system will increase faster than the utility of it, is reasonable. More pdlematic is the hypothesis that, given this assurmption, the hegemon will have to give way to ( ~ n eof the other '"potential hegemons." The basis for the argwemt in this section is that changing the system, or letting other potential hegemms get acccss to the "tlnrone," will imply substantial casts for the hegemm, in addiZion tc:,the costs of preserving the sygem, These two cost functions are assumed to be collverse in the sense that if one increases, the other demases. Given that the hegemon wishes to miz~imizeits costs on both dimensions, the question will be how much change is needed to avoid, as much as possible of the costs oE

Cc~zintryCase 1: Saudi Ambia

205

preserving the sygem without carryhg more than the necessary costs of changing the system. Changing the system could k p l y that all regulation of actors3behavior is abmdoned, or that one or more other actors take the hegeman's place, so that the actors' behavior is still regujated, but by otlters than the original begemon. In ~ l a t i o nto the other potential hegemons, the existing hegemon may give concessions to possible rivals in order to conciliate their threats against the hegemn. This can be seen as a way in whiCh the h e g m m reduces some of the negative effflcts of prmervitlg the system by paying the costs of some chmges of it. But, there will always have to be m actor taking the hegernmic burdens if the collective goods are to be providcd and the system is to be preserved. Despite the hegemon's increasing costs of maintaining an htematiox~alsystel~,it will, be in the he gem ox^" own interest to carry some of these costs combined with some changes of the system. Tlx.~s,the hegemon's strategy options do not constitute a choice between carving and not carrying the costs; it is ralher a question of how to realize the collective goods with minhal costs, This makes the yue* tion of where and b w the hegemon will find it profitable to adjust an empirical questiox~,not a predetermined one as suggested by Gilph. 'This also implies that an essential resource for the hegemon in maintajning a hegemcmy is tbr ability to adjust. The argument above can be illustrded by a form1 model. In Figure 6.1, C(W) is the hegemods costs of changing the system; C(Z) is the hegemon" cccrsts of preserving the system. Change is measured from left to right, and the costs rtf chmging increase the morc one changesPreservation is measured from right to left, and the costs of pmservation increase the less one changes. En this case, the costs of preservation are set to be nominally @her than the costs of change; this, of course, is not a necessary condition, as the cost functions are assurned to be additisre. The third cost function, C ( Z )1; C(W), is the sum of the two nominal cost functions. The gexreral assunnption can thus be stated as folocvs: The hegemon will c h a q e the system until the sum of the costs of chantging and preserving the s y s t m are minimized. In the model, this is expressed in the point Xo*When X = X0 the hcgemon minimizes the c o d i n e d costs of preservation and change. This minimal cost point is found by fulfilling the follow conditions: (I)CP(W>+ Cr(Z)=z 0, and (2) C""() + C""() > 0,

The minimal-cost point is thus equal to the point where the marginal costs of preserving the system equals the marghal costs of changing the system. This argzlmnt su,pplemen.ts Gilpin's argzlmnt cited above

Country Case X: Snzddi Arab&

0

ange

FIGURE 6.1. The hegemonlc cost-minimiza tion model SC~URCE:CLaes 19%a:44.

(1981:lQ-11), as it considers the possibility that the hegemon does not seek the equilibrium point of marginal costs and benefits from further change, but the equilibrium point between the costs of preservation and the costs of change (see Gitpin" third assu~nptionabove). This might also be more Fnrportant than the ret1atimship between the ecmonic costs of maintaining the stabs qua and the capacity to do so (see Gilpin"~fourth assumption above). The cost curves in Figme 6.1 can incorporate the hegemnn"~capaciv to sustah the status quo, with the consequence that the slope of the curves will change over time. 'This puts pressure cm the flexibility of the hegemon to adapt to external changes of costs. Given such adaptability, it is the argument of this section that the hegemm can sustain its poriition, m power, even when costs and capacity change. The two c m e p t s introduced i,n t-he intmduction-benevolent and coercive hegemony-point to a chmge in the hegemon" sstrategy toward the other adorpi in the group; in this case, the oft producers. The formal model outlined in this section poislts to the costs inherent in the role as hegmon, and the cmseFences fclr the hegemnn"~position. Tn the following sections (6.2. to 6.5) the chmges over t h e of these aspects of Saudi Arabia's mle in the cooperation between the oil producers are discussed. The Saudi

Cc~zintryCase 1: Saudi Ambia

207

&hian change of strategy from a benevctle~~t one (see scc.lion 4.4) to a coercive one (see section 6.5) is an empirical case of a hegemon behaving in accordance with the model outlirred in this section. First, Saudi Arabia gained a hegemonic positim m o n g the oil producers. Then the country experienred unacceptabk costs from havhg this position, anct thus had to change the system. It did, hwever, counter this loss of hegemony by COexing other producers to cooperate in providing the collective good. Saudi Arabia tried to minimjze the costs of change and the costs of preservation at the same time. It sought to o b t a i ~the ~ Xvpoint in Figure 6.1. These aspects are malyzed empirically in sedions 6.4 m d 6.5. First, the historical background to Saudi Arabia" position is pmsented in section 6.2. In scliction 6.3, the Salldi Arabian pdicy from 1973 to 1981is discussed. This period is characterized by Saudi Ar&ia% seekhg a hegemo~~ic position but being iPlcapalrrle of doing so due to the market conditions and the poficies of other OI~ECmembers (see Chapters 2 and 5).

6.2 The Making of a Negemon This section briefly outlines some aspects sl"Saudi Arabim history that made the kingdom the most important oil producer in the world. The folhtwing modern Saudi saying caph;lres the essence of this development, viebved from the perspective of Smdi h a b i d s citizens: If you didn't became a Saudi in the days of King Abdulaziz 61952-19531, you will never be a Saudi. Tf ~ C I Udidn't become rich during the days of King Khalid 115375-19821, you will never be rich. If you didn? become poor during the days of King Fahd f4982- 1, you will never be paor. (cited in Wilsan and Graharn 1994:1'71)

The Fomiltz'on

Uf:the KinCqdomof

Snridi Arab-in

A common perspective m the relationship between Saudi Arabia alld Rritak h the interwar peri,od is one of imperialism. This is particularly so for political studles focusing on oil as a strategic commodit)r A recent example is Brodey (19911, where among others the foundation of Saudi Arabia ir-l f 927 is interpreted as follows: This erection af state organization% and thus the incipient extension of the nation-state system to the region, war; in part a particular strategy of territorial control deployed by the British to facilitate both internal pacificatiron by indigenous elite and administrative coherence for operation af oil capital. More generally, it reflected the changes attendant on the consoXidation of

208

Country Case X: Snzddi Arab&

formal cuXontalism under the New-Imperialism of the imperialist epoch, without taking a proper colonial farm. (Bromley 1992:108)

This argument is not valid regarding Saudi Arabia. 'The Kingdom of Saudi Arabia was proclaimed an independent state by a decree on September 23, 1932. Contrary to most other states in the Middle East, which were established by colonial powers, it was established by a local force that took control over the territory. However, the reiationship between the first h g , Abd al-Aziz, and Britaitl was paramoullt to the creaticm of the Kngdom: "Abdulaziz . . . h e w he needed British support to succeed in unitjng Ara:biam"ilson and Graharn 1994:89), Abd al-Aziz had obvious interests in establishing good relations first with Britah and later with the United States. To understmd this, a brief recapitulation of the ecmomic aspects of the founding of Saudi Arabia is necessary*From his first contacts with the British, Abd al-Aziz had emphasized his fir1ancia.l we&. From 1917 until 1924 he was subsidized with E5,000 per month, m d s o m ribs and ammunition ( b u n g 1983:4). The taxation of the pilgrimage to Mecca was the major source of imome. Until 1927, as world economy grew, so did the nunbes of pilgrirms; however, from 1927 until 1932, the number of pjXgrims droyped. from 132,000 to 29,500 (Slugiett and brolak-Slugiett 1982:46). The depression also hit Muslims, and the Saudi Arabian income from the pilgrimage fell dramatieally. The position of Abd aX-Aziz as ruler of both his traditional area, Najd, and the newly conquered Hijaz, also meant increased expenditures..The two areas were &&rent regarding socio-demographic and cultural traditions, Najd was the base area of Warnabism, while Hijaz, the area where Mecca is located, was a far m r e cclsmopolitan area. The Najd tribes also creat-ed the base for the t:khwm, the military force of Abd al-Aziz. His securlity interests thus implied, that he had to balance his modernization aims with the il^lterests of the mom pious and traditional Najd tribes..The Xkwan rrtnellion from 1927 to 1930 clearly illustrates the thin Ihe between modernization and traditionalism that Abd . oil was discovd-Aziz had to follow in the first years cJf his ~ i g nMrhen ered in Bahrain in June 1932, Abd al-Aziz grasped the possibibty to gain political ahantages. With the commercial exploitation of oil, '"he Saudi ruter was . . . able to purchase a r m a m a s and trmsport and to'buy' the loplty of the tribesmen. merclafter the ami,rts [prince's] potential to resist him was completely nullified." (Abir 1993:5). Sfince then, the religious leaders have had only a consultative role in Saudi Arabia, and their influence has been depelndent on the king's goodwill. hbir (1993:6) maintains that the Saudi royal family and the ulama have mutual interests in the preservatim of their historical alliance. It contributes to "the regime's legitimization, to stability and to national integration. the

Cc~zintryCase 1: Saudi Ambia TABLE 6.2 Share of 'Total Expods, Selected Countries ( X ) , 195@1965 Cauntrv 2 950 3 955 Z 960 Venezuela 30.1 26.3 123.6 Kuwait 6.5 43,1 13.3 Saudi Arabia 9.9 11.1 10.8 Neth. Antilles 45.0 93 5.7 United States 5.2 3.7 1.5 Total of the 5 66.7 639 54.4 WURCE: Darmstadter et al. 19"7:149.

209

3 965 18.0 44.2 11'2 3.9 0.7 45.0

other hand, it helps preserve the Wahhabi character of, and the role of the ulama in, the kingdom.""

After the Second World War the Saudi Arabian oil production grew rapidly. Chevron and Texaco, which held the concessions, agreed to include lVlobil and Exxm as additional shareholders in their daughter company conducting the oil pr~durtionin Saudi Arabia, hamco. Ry 1950, Saudi,Arabia was exporting approximately 10 percent of the total world oil exports. During the fifties the importance of other producers was reduced, as Table 6.4 indicates. 'l'he importance of the Middle East hcreased, and among the Middfe East producers the role of Saudi Arabia increased. h 1950 the enormous o l X production in the Middle East area attrarkd Western companies and governments. 'The attraction is clearv described in a State Department Poticy Paper of September 10, 7,950 (reyrinted in ivme 1974, part 7:123): This area [the Persian GuXf geosynclinej is a uniquely rich section of the earth's crust. Except for sorne submerged areas under the Persian Gulf, it is cl-rmpletely covered by cc>ncessionsbelonging to us and fc3reign ail companies. Proven oil resewes, apprc>xirnatefy40 billion barrels, equal those of all the rest of the world combined and are almost double US proven reserves. Only some 500 tzrells have been drilled in the Persian Gulf as against over one million wells in the us. us wells average 12 barrels per d a y Middle East wells 5000 barrels per day. It is further estimated that the 13ersianGulf area has an even greater propadion af the ""prc~bable'kr""pssible" oil reselves in the world, i.e., sorne 150 billion barrels."

The aim of the Saudi Arabian govenlmcmt in the early fifies was to increase the revenues from the Aramco operathns, The basis was the

220

Country Case X: Snzddi Arab&

model of Ve~~ezuela, which had achieved a 50 percent share of companiczshet income. Aramco sought to dcpress the companies3axable inentfs revenue never&eless rase "from around 28 come, but the gave cents a barrel in 1951 (while the posted price 'or Gulf exports was $2.75 per barrel) to around 80 cents per barrel in 1956 (posted price $1.93)"' (Evans 4991):342). From 4968 the strategy of the Saudi Arabian govemment changed to seeking state participation in the Aramco upstream operations, mreats of legislathe measures unliess the government was included in Armco led to the 1972 agreement in whick the state achieved a 25 percent ownership of rhe company"~upstream. producing assets. The gwernment put "rang prctssure on the Ararnco parhers, achiwing 100 percent particir)ation by late 1974, By then, Saudi Arahia's share of total world production was &out 14 percent. The resource base for Saudj Arabia" stmctural leadership or hegemony was thus in place by the early seventies. Saudi Arabia had by far the largest potential oil Rserves, its production costs WE among the lo\vest in the hvorld, its production capaci-ty was increasing rapidly, and the polit'tcal regime was dedicated to modemizatim-implyhg increased oil prodwtim. T'his was not sufficient for Saudi Arabia to determhe the future course of the world oil price. As described in section 5.1, the OI~ECbargairzing in the seventies was primarily centered m the price issue. Until the beginning of the seventies, actual prices for inctividwal cargoes of crude oil were a secret to those other than the buyer and seller of the single cargo. I'he price transparency was l ~ wand , there was room for price cfiscriminati.on. This price sc.crclcy m&it difficult for Saudi Arabia to use a low price as a way to limit exports Irom other buyers. The puhljcation of OPEC prices, the questiftn of the differentials, and the role of the marker crude all. were important factors that ellitbled Saudi Arabia to play its hegemanic role (see Chapter 5). In the eighties the question changed from price to production, as the transparency probtem emerged concemifig actual WEC member production and exports. As discussed in section 4.3, the verification of individual memberskexports becme a disputed topic. I'he main cooperathe instrument in the eigbties obviously was the quota arrangements, by \zrhich the indi.vidual member was allocated a certain production volume. Saudi Arabia" role as a swing g d u c e r made it possible for the country to balance the marZcet with some a priori howledge about the likely production of other OPEC members. T!ze State f2mf icipat ion Negu t iatimis

Saudi Arabia was one of the most impmtmt countries in the formation of an importmt role, both in the internal negotiaticms in orEc and in rclpresenting the wanizatio~lin negotiations with other aeOPEC. It has played

Cc~zintryCase 1: Saudi Ambia

22 1

tars h the international oil market. 'The reason for this carnot entirely be ascribed to the size of the country" oil reserves, production, or share oi OPEC pmductim, This section discusses a case where more intangible aspects of the Saudi Arabj,an hegemnny were prominent. The formation of Q I ~ E Cin 2960 is described in section 2.4. Terzian (1985:65) argues that the ""founding fatJ-rers of OIT~Care incontestabty the Venezuelm, Juan Pablo Alfonza m d the Saudi Abdallah Tarilkri." Tariki had become mjnister of petrolem and mixleral. resources as a result of the changes in the gove ent following actions by the so-called Liberal prirrces, who in June 1960, inspired by Arab nationalism, demanded cmstitutimal reforms. Being a dedicated nationalist; Tariki wanted mnre national cmtml over oil production, and OPEC was seen as an instmmcmt to achieve this in alllimce with other oil producers. In 1962 Tariki was replaced as Saudi oil minister by the more pragmatic Sheik Ahmed Zaki Yamani."~~the mid-sixties knezuela pressed for an experimmtal program for production planning in OPEC.Sarrdi Arabia made its support for the program conditional on an incmase in its own production but faited to gain acceptance for this demand, and withdrew its s ~ ~ p p ofor r t the program., 'Thus, OPEC'spolicy during the sixties did not hclude any joint actions on produclion or price policy (see Chapter 2). Through the Tehrm-Tripoli agreements of 1971, the a m c countries gained influence on price-sening at the expense of the international, oil s the two agreements made OPEC companies The price i n c ~ a s e following cancel further discussions &out the productim plmning pmgram ""unlil such tim.e as the Conference considers it necessary to counteract any element which might lead to i~~stability and deterioration in the oil market"" (OPEC 199Q:SS).In January 1972, negotiations begm between apEc producers in the Persian Gulf and.the comessionairrz companies on the issue of state participation. The leader of the OPEC delegation was Yamani. 111 February; Yamani was given a mandate to conthue the negotiations at the national level with representatives of Aramco. Aramco was totally opposed to the general principle of state participation but made an exclusive offer to Saudj Arabia that would have given the country a 50 percent stake in the development of unexploitcd proved oil fields wit.hin its concession area. '"This proposal was seen by Saudi Arabia as an attempt to undermine its commitment:to the Cuff states' cotlective negotiating p s i tion and was accordingly rejected" (Evans 1990:429). Yamani soubsequmtly requested that OPEC hotd an extraorclhq meeting of the Conkrence; &is request made k m c o accept the principle of 20 percent state participation. The Twenty-seventh C313EC Conference inMarch 1972 issued a statement threatening sanctions against any company that "may attempt to undermine the solidasity of the Organization by submitting to

222

Country Case X: Snzddi Arab&

the demand for participat-ion in s o m member countries and not in others"' (WEC 1990:98). Yammi was once again mandated to negotiate on behalf of the Gulf O19EC producers.6 On June 3, 3972, Iraq nationalized the Iraq Petrolcm Counpany and later that month the orEc Conference issued a statement in support of the Iraqi, government" action, This made I(amanifs threats t w d the companies more credible when he stated, "participation is our substitute .for nationalization'" mobinson 19813:67)b m a n i continued his negotiations with the senior oil companies for most oE the year. On October 25, tbe Gulf producers ( M u Dhabi, Kuwait, Qatar, and Saudi Arabia) and the cmcessionairc.colnpnnies approved the General Agreement on Participatim. The Thirtieth owc Conference in Oetcrher 1972 called the agreement a t-ufnimg point in oil h i s t q and a tribute to the role pfayed by Kkg hisal in "elimhating many nf the obstacles" "ring the negotjatims, By the end of 1972 the roles of Saudi Arabia, King Faisal, and m the operative stage Sheik Yamani, as leader of OPEC, were established. Yamani's role indicates his skills as an entrepreneurial leader ( h u n g 3991:293). Frrrthermortr, it illustrates bung's argument that the strength of a leader includes the ability to use both its structural position and its entrepreneurial skills (Vnung 1991:303). It also supports the important arglament put forward by, aanlong others, Baldwin (1979) that power resources in iz.lternaticmal politics have no value unless the actor also possesses the skill and will ta use them. The position acquimd by b m a n i and Saudi Arabia in the negotiations with a (for OPEC)common opponent like the international compmies was m importmt asset when the internal OPEC bargainixrg got tougher in the secmd half of the seventies, as it manifested Saudi Arabia" role as the leadk~gcountry in Or2Ec, Furthermore, to the outside world, OPEC became identified with Yamani, and his opinions had easier access to international media and attention than those of the other OI~ECmembers.

The ""Special Relafz'u~zship'~ with fhe U~zitedSlates

In May 1933, Standard Oil of California was granted a sixty-year concession for oil exploration and production heastern Saudi Arabia. The company was to lend the government f5C),(ZCr0,Mihich the gove not Obljged to repay in Full, while Stanctard Oil could deduct the loan from future gavemental revenues from possible oll production, As described abawe, this income was essential to Abd al-Azizfs fir~anciafposition at that time. Alt.hough thc pilgrimage picked up d u ~ i n gthe late thirties, the Second Mrorld War completely halted this saurce of income. The Srcond World War also constrained the oil production in Saudi Arabia. The British financial assistance thus constituted a larger part of the

Cc~zintryCase 1: Saudi Ambia

223

Saudis' income, and this gave the British a clear advmtage over us firms. On the political level the United States had not devefoped a policy toward the region. As late as 1944, Presidmt Frankiin D, RooseveLt had ruled against any aid to Saudj i2rabia. Mter considering a request b m the us companies operating in Saudi Arabia that the us g o v e m e n t give fh~mcialaid to Abd al-Aziz, Roosevelt instructed one of his aides in July f 941: "wifl you tefl the British . . . t hope they can take care of the King of Saudi Arabia, This is a little far afield for us" "ergisl 1991:394), The us policy changed during the war, partly due to the percepticln of an increased need .for imported 03, and partly due to the perceptjon of the lack of Britain" capacit). to secure Western geopolitical interests in the region. To balance the British inflwnce, the r:s companies operating in Saudi Arabia persuaded the us government in 1943 to extend "lendlease" 'aid to Saudi Arabia. This laid the foundation for the political ties between Saudi Arabia and the United States. Roosevelt and Abd al-Aziz met inearly 1945, and a us security guarantee to Saudi Arabia was established in 1947e7In October 19-50, President Harry S, Trurnm wrote to King Abd-al-Aziz: "Iwish to =new tc:,your majesty the assurances which have been made to you several times in the past, that the United States is interested in the preservation of the independence and territorial integrity of Saudi Arabia. No threat to your Kingdom could mcur which would not be a matter of immediate cmcern to the United States.'" "is relationship was subsequently developed both at the political level, through the military guarantee and weapons deliveries, and at the economic level, through the us campany Aramco. Contraq to the argument of B m l c y (1991:108) cited above, the preceding discussion shows that the Saudi ~ l a t i a n swith the Illlited Kingdom and the I,mited States were established on. the basis of mutual interests, not on the basis of exploitation in, an imperialistic fashion. Furthermore, I?ibd al-Aziz" pditical position fnside Saudi Arabia was strongly enhanced by the state revenues generated from the oil industry. The oil income not d y made the economic mdernization of Saudi Ara" bia possibte, it also gave the monarch political freedom of action, as he o ~ e r w i s would e have had to be more subordbate toward the piaus religious tribes. This dilemma b e h e e n traditionalism and modernism is "the most commm approach to the analysls of developments in Saudi Arahia" (Mihloclk 1982:75), and, as argued, the retationhip with the United. Kingdom and the United States fits right into it. In section 2.4, the relatimship between the international oil companies and the oil-produciz~gstates were discussed. The government of the United. States played no lesser rote in the events surrounding the n h r m and Tripoli agreemmts (rl' 1974 and the oil embargo of 1973. Based on the experience of Occidental being sqrxeei.,ed by Libya, the oil companies re-

224

Country Case X: Snzddi Arab&

alized they needed to negotiate jointly with the oil-produciz~gstates. Such openly coherent market behavior c d d violate the us antitrust legislation. h January 197ffbefore the TcJhran and Tripoli nego"ciationsbegan, the oil companies approached the us government m d got a so-called business review that, given the information available, the Justice Bepartment wodd not take crirninal action against the companies if they pursued a united negotiation strateu with the oil-produchg governments (Church 19% part 6:224). With this clearance the companies signed the Libyan Producers Agreement, a safety-net a p e m e n t that ensured other companies would p v i d e a cornpany with oil if this it was forced by a nt to reduce its production (Church 1974 part 6~224-230). its Twenq-first Conference in Decernber 197C),agrred cm a resol ~ ~ t i oadopthg n a 55 percent mhimum tax rate on oil cmpmies incme and a unifom generd increase h prices. The resolution demanded negotia~onswith the compmies w i & h thirty-me days. "I?herewas thus a time kame for when t-he negotia,t.ions ha,d to contcnencc. Cm January 16,1971, the companies was ready to negotiate on Lal issues, but only joinay, one team from the companies and one from C ~ I ~ E C'The . followirrg day Undersecretary oE State John N. lrwin met with the shah of Iran. Rased on this conversation, Irwin suggcrsted. that the State flepartment u q e the compa" nies ""to agree to negotiate in Tehran with the Gulf producerf" (Churclh 1974 part 5:167). The us gover ent had in effect hmged its position. The importance of this change in us gov olicy has been heavib debated ever since. Morris Adehan (15395. &at: "had &e us gavernment not desmyed the new-fomd sdidarity of the companies . . . Che Persian Gulf producers been Ifrustrated. . . . The kstorim c m helped the cartel in its hour of greatest only record that the US g need," A key actor in th 011s was the companies' adviser J o b J. be was not "too much impressed . . . by the attikde of ,"he also argues that tht. US policy "wasnft dwisive . . .I much by what the Gave ent's position was. I rather took for granted that Mr. MaclacArthur would side with the Shah" WfChurch 3974 part 5:26&26q. Krasner (1978:265) discusses how the us policy can be explaked. He partly attributes it rmce, but he also emphasizes the conflicting interests of the us gave t: The fact is that American policy did satisfy the r~bjediveof keeping the Shah and Faisal happy, even if it did not keep prices down. American policy-makers acted to preserve the stability of noneornmunist regimes even thaugh this strategy meant accepting higher price-nd opposing the preferences of the oil companies. It was onIy after 1973 that prices themselves became a matter of crlncern. (Krasner 19715:265)

Cc~zintryCase 1: Saudi Ambia

225

The political relationship between the us and Saudi Arabia had been growing since the Second World War. The us had grilnted Saudi Arabia substantial aid and loans and supplied mfiitav equipment since tbe Second world War (us Department of State 1993: 754-755). While the international. oil. companies naturally emphasized the economic relations, the ent had primarily focused on the pofi.ticai relationship, a d po:[itical in a meaning diMemnt frnm *-rilitaryxSali.ons*This is illust..rated by the following passage of a letter from the us h b a s s a d o r to Saudi Arabia to the State Department in 1959: Assistant Scretary Zmin, who a p p e a ~ d exceptionally well briefed, made as his central theme the paint that United States Government decisions to train the Saudi military forces and to grc3vide them with arms was basically a political one, Therefore, no matter what our military problems with the Saudis are they should be accepted up to the limit of our national interests. . . . He said one of the worries of Defence was that the Saudis might accuse us of failing to carry out our commitments. . . . I replied I thought Faisal was coming around more to understanding the value cjf the United States friendship. (vs Depadment of State 1993:;"$3)

h 1970, President &chard M. Nixon stated what became k n s m as the Nixon doctrine: Neither the deEense nor the development of other nations can be exclusively or primarily an American undertaking. The nations of each part of the world should assume the primary responsibility for their own well-being; and they themselves should determine the terms of that well-being, We shall be faithful to our treaty commitments, but we shall reduce our involvement and our presence in other nations' affairs. (Palmer 1992:87)

According to Pal~ner(1992:87), the Nixon doctrine had ""pofom~dimpact on American policy in the Middle East, and its consequences are still being felt in the 1990s.'" The strategy war; to have Iran, not the United States, replace the position of Britah in the region. By 1973, oil and politics were more mixed than ever before. At a meeting between Faisal and the Aramco parbers in ikfay 1973, Faisal, according to transcribed confidential noks, said the followi.ng: "Time is ruming out with respect to us interests in the Middle East, as well as Saudi position in the Arab world. Saudi Arabia is h the danger of being isolated m o n g its Arab friends because of the fail~~re of the us Govemmcnt- to give Saudi Arabia positive support . . . you will lose everything" ((Robinson 3988:89). To the companies, ' k v e r y ~ i ~ ~meant g ' " fie oil concessions;

22 G

Country Case X: Snzddi Arab&

to the US go~rernment,it memt an importmt political ally in the world's most tuibulent and important region, The 3973 ernbargo imposed a severe test on the relationship. The Saudi regime did not seek the embargo, but felt pressured by other Arab coulztries. In fact, the US government and the Saudi Arabjan leadership misread each other" situations fn connection with the Yom Kippur War and the fctllo\ving oil embargo: Washingtcm continued to view the Saudis as Arab moderators who could influence large sections of the Arab world thrc)ugh oil largess. By adopting such a stance, Washington misread Saudi (and Arab) public opinion, and more importantly, the ability of the al-Saud to disregard the strong currents flowing around them. The Saudis committed the same mistake vis-8-vis VVashingtcm. Must Amrisans supported Israel, and would not let it be Befeated or destroyed by its Arab enemies; this attitude, in turn' entered into American policy considerations. (Wilmn and Graham 1994:I W)

er 3974, the friendly rdationship with the mited States was =stored. 01 June 8, 11974, Saudi Arabim Deputy Printc Minister Fahd signed an agreement for extensive v.s.-Saudi military and economic cooperation. The agreement was described by Secretary of State Kksinger as "a milestone in u.s. relations with Saudi Arabia and Arab ccluntries in general,""t invtdved massive American assistance to the kingdom in planning and implementing its economic and military Bwelopment and in return called for Saudi cooperation in meeting the energy needs of the United States and its Western allies. (Gotub 1985:22)

The Carter administratictn's unforbnate hmdling of the Iranian crisis and the hostage affairs hveakemed its credibility in the eyes of the Saudi Arhian regime, which could vividlqi imagine itself in a position similar to that of the shah. The Reagan administration relied more one-sidedly on market forces, and was witi-ing to follow the market even in times of crisi.~:''In the event of an emergency preparedness plans caU fur relying primarily on m a r k t forces to atlocate energy supplies."'" The m i l i t v commit~aentto Saudi Arabia was, however, maintained, As discussed in Chapter 3, the United. States partly inrervened in the tanker was during the Iran-Iraq conflict by reflagging Kuwaiti and Saudi hraZnian oil tmkers and providing protection for Chese ships in the Gulf area. Also, Vice President George Bush embarked on a trip to Saudi Arabia in 1986 when oil prices m r r t e t e d . Being a Texan oil man, Bush savu the producer side of the coin, and did not emphasize the consumer interest in low oil prices:

Cc~zintryCase 1: Saudi Ambia

22 7

"on the eve of his trip Bush said that he would 'be selllng very hard' to persuade the Saudis 'of our own domestic interest and thus the interest of national securiq . . . I think it is essential that we talk about stabitiv and that we not just have a continued fsee fall lilse a pararhutkt jumping out wilhout a parachute.3ush was clearly saying that market forces had gme too farf"Yergin 1991:756).10 The incidctnt clearly iilustrates the complex domestic int.ere5t.s of the W e d States regarding oil, inclrading both consumer m d producer interests. In adcfition, oil supply is regarded as a security issue (Bull-&erg1987:34-56). The oil-price fall of 1986 made oiln toproducer interests more visible and Lhus a part ol the us f o ~ i g policy ward Saudi Arabia. The next major incidmt affecting the US-Saudi hrabian ~latictnship was the Iraqi attack on. Kuwait in August 1990. This was dealt with in section 3.4. The operation gave Saudi Arabia weapons deliverigs and military support on a tremendous scale during 1990-1991, but Opuaticrn Desert Stcrrrn also created some lasti,ng ncgative effects m the us-Saudi Arahim relationship. The war and the unsettled sit-uation in Iraq have meant a higher presence of US military personnel in the kingdom. 'This has provoked Islamic fundmentalists in both Iran and Saudi, ArabiaBorrtb amcks against us facilities and personnel in Saudi Arabia attest to the level of tension created. It worsens the diklnnla of the royal family, between the need for us support against external security threats and the possjbility that this support itsdf increases internal securi-ty threats. This s illustrates the probtem of the i n t e d sccuriv of autocratic ~ g i m e such as that in Saudi Arabia, as discussed in section 3.5, This section has pointed uut four cornerstones of the Saudi Arabjan hegemony*First, Saudi Arabia is the only state in the regicm that was established by itself- and not as a result of intperialistic expansi.m and subsequent decolonialitzatim This gives the kingdom a particular position. Srconli, the size of the Saudi Ar&ian oil reserves and production gives the kingdom power resources in the internatio~~al, oil market. mird, the negotiation skills of Yamani helped to position Saudi Arabja at the centcr stage of the political turmoil ccmnected with the oil market during the seventies. Fourth, the close and lasting relationship with the United States provides the kingdom with a miliLary guarantee that strengthens e external threats. the security of the Saudi hrabian ~ g i m against It is now time to turn to the dynamic chmges in the Saudi Arabian exercise of its hegemonic power in the oil-pmducer cooperation. Four phases will be discussed: first, the period from 1973 to 1981, when Saulii Arabia was un;tble to exercise hegemonic power amollg the oil produeers; second, the period from 1982 to 1 9 s ' when Saudi Arabia performed the mle of benevolent kgemcrn; third, the period from 1986 to 1996, when the Saudi k h a n hegentonic strategy can be perceived as coer-

228

Country Case X: Snzddi Arab&

cive; m d flnally the situation in 1999-2000 when Saudi Arabia cambhed the two hegemonic strategies. 6 3 The Incapable Hegeman, 1973-1981

During the Arab-Israeli war of 1967, several Arab countries advocated the use of oil-export restraints as a political weapon against Western conntries that s~~pported Israel. This strategy failed due to several factors: the m a h Israeli ally, the United States, imported only 5 percent of its tot& comumptim from the Arab producers; thr exports from non-Arab countries, like Iran and WnezueJa, increased as the Arilb pdu,cers cut back. Fur.thermore, the embargo was not a totd volunne restraint, m d the international companies were therefore able to swap shipments among differe~~t destinat-ions The main consquence of the embargo was a loss of revenue for the Arab producers, In the early seventies there was a desire m o n g radical Arab powers to once again use oil as a weapon against Israel, as the United States had experienced. an increased dependence on imported oil from OI~EC.Foremost among the aclvwates of this policy were Ira% Algeria, and Libya. 'The only cou~ztryin clear opposition to this strategy was Saudi Arabia. 'This opposition can be explained by several factors. Schneider (1983:213) mahtains that Saudi Arabia" security- interests wem particularly important. The Saudi authorities perceived two scparate threats to the safety of their state: (i) Israel's exjst% and potmtial expansion; and (ilArab ) radicaiism. n e s e wert. untferstood as being closely il^ltt.rtwined,as increased Israeli expansion kvolllid stremgthe~zArab radicalism, us support for Israel and its lack of support for conservative forces in the Arab world had, h the e y s of King Faisal, the same effect. Even though an Arab oil boycott could have solved this question, Saudi Arabia did not wish to confront the CJnited States, which they considered to be the only possible guarantor of Saudi Arabia%security (see section 6.2). ""As late as October 1972, King FaisaS said inan inkrviecv, 'It is useless to talk about the use of oil as an instrument of pressure against the u.5.-it is dangerous even to think of thatf.'"J m e n it turned out that the United States r9ected Saudi Arabia's wish for a more ""blmced policy" "obirrson 1988:73), and Egyptian President Anwar Sadat used pressure to Obtairr the support of Saudi Arabia in the ongoing codict with Israel,'VFaisd chmged his mind. On A u g ~ ~30, s t 1923,Faisizl declared: "We do not wish to place any restrictions on our oil exports to the United States, but America" complete support for Zionism against the Arabs makes it e x t ~ m e l ydifficult for us to continue to supply the us petroleum needs and to even maintain our

Cc~zintryCase 1: Saudi Ambia

229

friendship with the United States,"= The OAPEC meetbg of October 17, 1973, fnllwed a Saudi Arabian proposal, and decided to cut oil production by 5 percent and to cut it by a fur&er 5 percent each month until "Israeli forces have completely witf.ldrawn kom aXI Arab territories occupied. in June 1967 and the legitimate rights of the Palestinian peoyte are restort-d""(Evans 49510:440). "However, at a further meethg held on Decetnbrzr 24 and 25 the Conkrence modified its polieies after hearing a report by the Saudi Arabian and Al.gerian Oil Minjsters on their recent visits to various Western comtries, which had led &em to conclude that a more positive atltude should be adopted towards 'friendfy' cou~~tries" (Evms 1990:442). However, the o A r E c policy from October to December had dramatic effects on t-he oil price in the spot market. The or9ecmembers then, had the opportunity to raise the official selljng prices. At a meeting betvveen the Gulf oil ministers in Tehrm in Decennber 1973, Yamani strongly opp0""darge price increases, arguing that they might drive the hdustrialized cotmtrieskecmomiesinto recession. 011the other hand, the shak of Iran pressed Lfor maximum expXoitatim of the market situatim, and a price increase as large as possible. "Topreserve ~ I ? E Cunity, kmni agreed to raise the marker posting to W.651 for 34" crude f.o.b. Ras Tanura, However, he made it clear that this was not the preferred policy of Saudi Arabia, as later ccmfirmed by King Faisal. Having joined the embargo, Faisal gained prestige among other Arab countsies and rcllieved some of the internal p ~ s s u r eon the regime. However, the oil e m b q o did not chmge the us-friendly policy pursued by Faisal and Crocvn Prince F M . Already in June 1974, Fahd made an official visit to Washington, rt.sult.ing in a comprehensive agreement on economic, technical, and rnilitary cooperation (see sectim 6.2). ""Riyadh expressed its readiness to help maisrtain a regular supply of oil to the market and to cuib the rise in oil prices"' (Abir 1393:67), In JuXy 1974, Saudi Arabia anmunced an oil auction cJf 4.5 mbd, without any .minimum price, confident that- no one wodd pay over $t1.651/barrcl.14 As demand had slackened, an auction would most likely put the official OPEC prices under presswe. hfter intense pressm from other OPEC COWtries, the plan was eanceled. Mowver, che stage was set for tense orEc bargainjng over the price issue, which p ~ v a i l e dfor the rest of the seventies. In these negcrtialinns Saudi Arabia was to take a difierent position than most of the other wt;c m e ~ ~ b e rJn s . the foflowing years the most promjnent opponent of the Saudi Arabians was the shah of Iran, After the oil auctkn was called off, ""the Shah led the counterattack against Saudi Arabia, insisting that the cartel's principle be that producer revenue should not be allowed to decline (i.e., prices or taxes should rise if volumes droppecJ)'"(Moran 1981:2S5-256). This confrontation between Saudi Arabia and Iran was the beginning of a la~gthyeonfXict between

220

Country Case X: Snzddi Arab&

these two major OPEC producers on. the price issue. 'The Saudi Arabians were not able to dominate the shah" soil policy: ""kamani later insisted that if the United States wanted the kingdom to use its production capacity on behalf of moderation . . . the us government hvould have to play a more active role in rrtlieui~~g pressElre from Imn" (hiloran 19R1:Etj;italics mine). This was exactly what Ji y Carter did three years later, h 1977 fsee end of the fctflow-ing subsectiion).

When nationalization or dominant state participation was established in most C>I3ECcollntTit"~,t.he question of tax reference prices and their relationship to poded prices becarno irreievant. The new basis for price-setting was the OPEC official selling prices, of which tlne Saudi Arabian marker crude was to become the key instrument. According to Stevens (19122:217), the marker crude was a powerful instrunnent: "Before each o13Ec "price-settingbmeeting, Saudi Arabia has decided on the level at which this crude should be set. . . . Thus the Saudis enter the OPEC meetkg with the marker price fixed." As this section will show, this is a far too simplistic description. The other countries exercised considerable pressure on the Saudi Arabian control over prices by exp l o i k g the price differclmtials system, and by 11977Saudi Arabia actudy had to carry out a threat to h o d the mrket, a strategy that caused substantial damage tc:,the Saudi Arabian oil-pmduction facilitjes. The marker aude, the above-mentioned Saudi Arabian ljght 34" mr, was in 1973-1974 introduced as a standard cmde from which the other C>T9ECcrzldes were priced acceding to formulas based on the differences in gsavity, suffur content, fseig'nt advmtages, and other qvality factors. These factors constitute the differentials, which is the concept that defines the extent of heterogeneity of the d~ferentsellerskrude oil, Saudi Arabia took i m cmtrol over the setting of the rnasker crude, leaving less room. ior the other OPEC c~untriesin their pursuit of an aggressive price p o k y dufing the seventies. The differentials caused substantial problems for Saudi Arabia, as the other OPEC countries, h the weak market situation that merged. In 1974-1975, priced their oil just above the Saudi kabian marker crude. This had the effect that Saudi Arabia last market share to more ""valuablef\sil qualities. Durhg 1975 most ai-i;entionwas focused on the narrowing oi differentialsbetween Mediterranem and. Gulf crudes. W~en Eaisal died in 1975, he was =placed by Kalzled. The s t m g man in the ruling of Saudi Arabia was, however, the prime minister, Crown Prince Fahd. He restrained Vamani" freedom to maneuver and took complete control over the Saudi Arabian oil pdicy FahcS's policy was similar or even tougher than Yammifs.Dwrjng owc's Forty-fifth Confer-

Cc~zintryCase 1: Saudi Ambia

222

ence in 9ptember 1975, the Irmian delegation delnanded a 20 percelzt increase in oil prices. Falnd instructed h m i to preparc a communiytlk declaring that Saudi Arabia wodd disassociate itself from its OPEC partners should t h y go along with Chc Irmian demands (Terzian f985:238). The Saudi Arabian position within OI~EChad by 1975became excqtional in at least three respects: (i) the country bad huge reserves, making it nn actor wit.h a long-term pasjpective on the oil market; (ii)it had q a r e capacity, aUowing it to pose credible theats regardkg a price war, as it could cornpensate for l w e r prices with higher oravut; and (iii) it had a popula~onof kwer than 7 million, malting the internd economic situaion luxurious cornpaed with those of other, more populous CXEC countr-ies. C in Doha in December 1976, a diviAt the Forty-eighth C ~ I ~ EConference sion betweell the OPEC members was impossible to avoid. Prior to the meeting, the shah of Iran had announred he would accept nothing less than a 15 percent increase in prices, while Iraq proposed a 26 percent increase (Moran 1983,:258). When the Conference opwed, Vatmnni announced that "My government feels that prices should be frozen for a furfier six months" (Terzian 1985:2.1.l). The following d a y he proposed a 5 percent increase as a sign of great willingness to compromise. The price hawks were unwilling to meet the Saudi demand. h m m i even flew back to Riyadh to consult with Fahd, and rt;ttlmed with the firm position of a 5 percclrt increase, "take it or leave it." The break of OPEC unity was final; the pmss release from this Conference states: ""Eleven countries, withln the Cmference, decided to increase the price of $11.51 per barrel . . . to $1,2.70per barrcl as of January 1st f 977, and to $13.30 as of July l,st 1977. . . . Saudi Arabia and United Arab Emirates decided to raise their prices by five percent only""012~c 199C1:14). The two-tier price s p k m was a realjty. Yizrnani made the fdokving statement before Ieaving Doha: In the past they used to decide fc~rthe Saudi crude oil price and we acthey are deciding far their c>wncrudes and we cepted-.Now we refused. decide for our own crude. . . . We will remove the production ceiling of 8.5 million barrels a day. We will damp the market-it means the whcjle structure of prices will collapse all over the world. There will be no structure of prices if they d o not accept the price of the Saudi market crude, (Terzian 1985:244)

Both Iraq and Iran reacted strongly to the Saudi Arabian dedsion, The shah catled the planned auctim "an act of aggression'" against Iran (on French television hJanuary 1927, cited in Marm 1981:258). t inercase proThe Saudi strategy now was to implement the t h ~ a to duction and thereby bring the prices d a m . However, "when the Saudi governmelnt ordercd Aranco to start producing at full capacity in De-

Country Case X: Snzddi Arab&

222

c e ~ ~ b1976 e r 'it became evident' that the Company's maximuln capacity was only 9.3 million barrels a day, not the 11.8 million that it had boasted of. Fahd was n a t u f d y furious" (Terzian 1985:245-246). ?"he company was orde~cd. to do everything possihle to increase capacity, and ,?II technical limits were suspended. This made pipelines explode and caused damage to reservoirs as the subsoil press= dropped and some reservoirs experienced seepage of water. Before the next OPEC cor~ference,"the eleven" ' h a agreed to drop the S percent increase of July 1377, while Saucli Arabia and the United Arab Emirates announced a further 5 percent increase in their prices.. The OPEC unity was restored. The costs of restmillg the pipelines were estimated at $100 million; the damage to the ~ s e r v o i r was s harder to calculate. Saudi Arabia continued to produce above a policy that hcreased the glut on. the world oil market, reducing the room for further price increases, The Saudi Ar&im &reat to Rood the market was not credible, As the Saudi Arabian r e g h e was not aware of this, the events of 19777still prove that Saudi Arabia was willing, although not able, to use its resources to coert?e the other OPEC members into compliance with its price strategy. It was to take eight years until the coercive hegemorlic strategy was successfully implemented (see sectim 6.5). The four-year period f r m the abolition of the anti-us sanctions h July 19% until the outbreak of the Irmian ~ v d u t i o nin the awtunzn of 1978 nevectheiess reveals a fiigh-profile strategy by Saudi Arabia. The willrngness to take on the other OPEC members, and to throw OPEC into crisis to avoid price increases, shows that Yammi and the royal family WE willing to bear possifile pditical costs in the pursuit of a price policy that they perceived as being in the interest of Saudi riarabia. In ecmomk terms the strategy described above was a risk-aversion strategy; If demmd was slack, Saudi Arabia, keeping its prices below those of the other producers, would be the first to sell its oil. If etemand was tight, the kingdom would forsake a profit by maintahing lower prices. h November 1977, the shah of Iran visited President fimmy Carter in Washington. hfter this visit the shah changed his price policy and signaXcd that he wodd support a price freeze throughout 1978. Saudi Arabia and Iran combined produced 48 percent ol: OPEC output. Now that they both advocated a price f ~ e z ethe , other members had little influence. Elowever, d u ~ i n g1,9722other poliical evel~tswere to change the situation in the oil market compIetelJr.

"3w

Secbk~ifyThmis: The Iranian Xevoltltioa and the Iran-Iraq War, 7978-1981

Saudi Arabia and the other conservative Arah cowtries pmved unable to restrain the (jsowing lslamic Sundarnentatisnt. As pohted out h section

Cc~zintryCase 1: Saudi Ambia

223

5.1, the Iranian revolution memt a halving of Iran's production, m d seve r d OPEC countries pressed for a rise in o r " ~ c ' s official selljng prices. Saudi Arabia did not want a price inwase, anci countered the loss of Iranian production by increasing production b m 8.4 to 10.4 &d. The Saudi Arabian production thus approached the countrfs sustainable production capacity, 'fhe other OPEC countries were already producing at almost full capacity As the connpanies accmulated stocks to ensure suppties in case of future shortage, the spot price increased from. $12.8/har.re1 in September 1978 to $38.35/harrel by October 19721. The revolution h Irm, contrary to the 1973 crisis, involved no deliberate attempt by the actors to chmge the course of the oil market by political means, but it did affect their behavior. M a n i expressed it this way: "It had to influace your thinking*This was a rnajor politics) event. It caused oil consumers to ganir, but it did not change the pattern of cons u p l i o n . And that" what made the price chaos so dangerous'~Robinson 1988:214). The Iranian revolution also affected the internal politicaX clllmate in Saudi Arabia. As was pointed out in section 6.2, the leadership in Saulii Arabia has &ays had to balance tsad-itionatism and modemism, or in this particular respect, fundamenta.lism and secularism. "The Saudi mling ctass was seriously shaken by the collapse of the monarchy and the rise of fundamentalist regime in Iran. . . . a humbled Zaki Yamani accepted a substantial raising of oil prices advocated by ~ I ~ E Cf'hawkshnd 'S Saudi nationalists" "bir 1993:"i"). Yamani was pressured not only in the international arena but also on the national political scene, as 02-price hawks and.anti-Mstem interests strengthened their positions due to the Irmian revolution. The spread of Islamic fundame~~talism has always been seen as one of the most severe threats to the position of the Hcruse of Saud. The fundamentaiism was not only an Iranian phenomenon in the seventies: "'As the .frustration of the conservatives rapidly grew after the rnid 19TOs, the wave of 'neo-hndmentalisd began to spread from AI-Madina to the I m m MtIhammad ibn S a d Islamic University (Kyadh) and to the theo l o a facdty in Mecca. . . . Many of the 'alims (religious scholars) and the students . . . were openly critical of m y of th imovations and Wcstern influences introduced by the government" (Abir 1993:80). On October 19, 1929,400 to 500 fuxzdarnentalists occupied the Mecca mosque, seized the Ka'ba haram, and denounced the corrupt government of the Sauds, It took the regime two weeks to crush the rt;bellion. The rebels gained little sympathy among Ihe Saudi population, primarily because lhey had defiled the holiest shrhe of Islam, the Mecca mosque, Un November 28, an unrelated emption occurred among 5hiites in the Eastcrn Province, supported by Tehran's calls upon them to rise up agalnst their corrupt rulers

224

Country Case X: Snzddi Arab&

in Riyadh. In sum, these incidents co~~fimmed the rulling elite's perception of the Islamic fundamentalism as a major threat to their position. In the second half of 1980 the i n c ~ a s ein spot prices came to a halt, as did the increases in the officid prices of the WEC countries. The mhisters " anmet FR Vienna in the middle of September, With ~ I ~ E C twentieth niversary coming up, it seemed a good time to standardize prices. Saudi Arabia once more increased its price by Sibarrel, and this time the other members did not change their prices, A summit of heads of state was scheduied for November 4 in Baghdad, with the pmspect of a reneml for the organization. It was never to take place; on %ptember 22, 1980, the Iran-Iraq War broke out (see section 3.3).19 Saudi Arabia (and K w a i t ) supported Iraq financially and contributed to pipeline transmission of lraqi oil through its own territory to avoid the exposed Gulf area. Saudi Arabia was willing to forgo oil revenues to achieve a lower tension level in the war, or to limit Iran's military potential. Politically and economically, the period from 197%to 1981 was highly vdatile not only inthe oil market but also im the Middte East as such. The security hterests dominated the eco~~omic interests of several Gulf states in this period. In the case of Saudi Arabia, the developments had a profound impact on the regime and its conscicrusness regarding its own unstable po:titical foundation. As David G Q ~ states u ~ h fie introduction to his study, "Saudi decisionnnaklng has been in8uenced by a variety of political and economic forces. I will conclude that m o n g these forces, one kjnd has dominated-concerns about national and regime survival" (Golub 1985:l).The rclgime was thus pressured both by external forces, as the other militarily potent countries pressed for c h q e s in the Saudi Arabian oil poticy, and by internal .forces, as the possibiflty of political mrest incmased, inspired by the developments in Iran. "Security first'" by now meant oil appeasement with the other OPEC memhen;. At tbe OPEC Conference in Abu Dhabi in Dece~sber1978, the Saudi Arabian strategy was to create a consensus on the price issue, The interpretation of the Saudi Arahian policy diff;ersamong writus. Terzian (1985:2@-2fil) emphasizes the percehed imbalance between demand and supply, m d the limited production capacity of Saudl Arabia, m a h g the country mable to correct the market. Gatub (1985: 28-33) afgues that the diffe~nce between the Dokica meethg in 1976 (where Saudi Arabia hduced a split beheen the OI~ECcountries) and the Abu Dhahi meeting in Decersnber 1978 was politics. The perceived theat had three sources: '"Sadat's s i p ing oE the C m p David accords on Septelnber 17,19723; the incipient political. crisis in Iran; and the announcement of Syhan-Iraqi unj.Q plans on October 26 (1.9781'" (Golub 1985:29). Siding with Sadat and the United States against Iran, Iraq, and Syria became too much for the Saudi

Cc~zintryCase 1: Saudi Ambia

225

regime. "With the shock of the Shah's dqarture, Soviet encroachments in Soutfil Yemen, and uncertainty about us dependability Saudi leaders e to offend either the new ~ g i m in e Iran or the showed e x t ~ r n reluctance new1y p r e e ~ ~ h eIraq" n t " m m 1981:261). Thus, in 1978 Saudi Arabia took the first step toward. a new and mre conciliatory OPEC poticy 'I'his was seen as a way of reducing internal and externat political thrcats against the regime. As the price of oil skyrocketed during 1979 and 1980, the policy had few short-tctrm, costs. Allhough Golub (1985) shotvs how Saudi Arabia fueled the crisis, its traditional. allies and the cmsttlner countries never blamed Saudi Arabia for the second oil-price shock, Its Fnrage as moderator was intact, As this ~ e c t i mhas shown, the Saudi iarabian ~ g i m wanted e to influence OPEC price policy in a more moderate direction. Due to the large and increasing reserve base, and the special relationship with the United States, the lmg-term economic interests formed the Saudi Arabian preferellces regarding the price level. However, the Saudi Arabian regime did not possess the abilitJi to achieve this. The most kportant iactors behind this were the important pclsition of, and the oil policy pursued b y the Shah of Iran. The chiinge in the Iranian oil policy in 1977 was thus an important outside event, strengthening the Saudi Arabian influence in OPEC, However, it lasted only a year m d a half, until the Irmim revolution-an event that tri.ggered an increased emphasis on security issues in the Saudi Arabian rcgimrz, thus making it harder to pursue the ecmornic interests to a full extent. 6.4 The Benevolent Hegemon: Swing Producer' 1982-1985

In 1988, Saudi Arabia was set to cmtinue its OPEC-friendlypolicy initiated in 1978, but now the instrmmts had to change, and the conciliatory role was to become very costly for Saudi Ambia. After the political crisis was over, Saudi Arabia refused to lawer production unless the other countries lowercd their prices, and it created a srtrplus sit-uation to forcle them to do so, In Octtiber 1981, the Saulii Arabian fight for a u~~ified and nonirzcreasing OPEC price succeeded, as the o13EC members agreed on a n oificid price of the marker crude of $3*IZ/barre1 (see sec-tion5.1). However, the OPEC decisio~~ of October 1981 was not sufficient to stabilize the oil price. Tke spot price now was below the oificial price and thus pressing prices down, in the same way as the spot market had increased prices during l.he Iranian revolution just Chree years earli,er. Instead of surcharges, discounts became the order of the day. The contirruous high production h r n Saucli Arabia virtually flooded the market. fn Febmary 1982, the spot price fell dramat-icatly belokv the official prices. Fmm the

226

Country Case X: Snzddi Arab&

Saudi Arabian point of view, the unity that had taken so m c h effort to restczre in Geneva in October 1981 dready seemed threatened by competitive price reductions. At the Arab Ei~~ergy Cohrence in Doha in March 1982, OPEC htroduced the quota system (see section 52). Saudi Arabia was not given an exact quota but was to act as a swing producer, regulating production up and down in order to stabilize the oil price. Never belfom had Saudi been \.vi11bgto discuss prodllction levels with the otber OPEC h e r s . NOWYamani himself proposed a coordination of OPEC production. This volte-face by Saudi Arabia tlrrned OPEC h t o a cartel and made the OPEC cooperation more profou~ndthan ever before.The problem was that Yamani perceived the problems to be temporary Once the price was stahitized, the stock drawdown would halt and d m a n d wodd pick up. This did not happetn. In a &€lining market the connbination of m official price of $34/barrel m d a production level of 38.5 mbd did not hold for long. There was no instiktional pressure on the individual member country to adhere to the established 18.5 mbd ceiling. The price on the spot market fell, while at the same time OI~EC members begm to relate cmtract crude to the spot price. Umani stated: "It was a connp1et.e fait~xre,Quite honestly, I don't see a very bright future. Tn a few days we expect to see the price of Narth Sea oil come dawn by $2-3. And that will be the beginning of a chain reaction. . . . We have lost patie~ncewith the OPEC members that have chosen a short-term self intereat policy in preference to OI~EC" and their own long-term in.terestgr (statement cited in Robinson 1988:262). The Saudi Arabian behavior as swhg producer during 1984 and 1985 created fairly stable prices in this period. However, Saudi Arabia had to cut back on its production cmtinually to sustain prices. This policy was intpossibiie to nnaintain indefinitely. The Saudi oil revenue fell and decreased the Saudi Arabian cm,as illustrated in F i p r e 6.2, In 1982, Morris Adelmm cclnsidered the Saudi AraZnian chdlenge as follows: ""Nobody can predict how mu& the Saudis would be witling to reduce their output; nobody in OPEC wants to h d out, But if the current budget is around $88 billion, then there may be trtlutnte if real prices dectine below the current $32 and output drops below 7.5 mbd" "delman 1982:55). Four p a r s later the. price was below $30, and Saudi Arahim output was 1.5 mbd, and had been below 3 mbd. As Figure 6.2 shows, the increase in Saudi Arabian GDP peaked in 1980; for the other s x ~ countries c the GDI) continued to increase, aZthough at a slower pace. The Saudi Arabim GDP fell 53 percent from 1980 until it reached its lowest point h 1986, while the level of the other o13EC mmbers' cr,r was almost sustained in the same period, This was the effect of the swing-prducer role, as the lowered oil income during this period was a result of deliberately lowered produckn. 'The loss of

Cc~zintryCase 1: Saudi Ambia

227

market share also gabre Saudi Arabia a more marginal role hworld politics and in Middle Eastern affairs, and lower oil prices =ant less political power for the oil producers in general. This was expressed by Yamani on. Saudi television: In principle, w e must draw a line b e w e n ecmamics and pc~litics.In other words our political decisions should not affect:economic facts and laws. But crude oil is a poliecal power and no one can deny that Arab political power in 1973 was based an ail and that its inRuence reached its peak in the Western world in 1979 because of oil. At present we are suffering because of the weakness of Arab political power based on oil, These are elementary facts which are known even to the man in the ~ t r ~(Vergin t . 4991:1;"47)

The Saudi government budget was severely weakened by these events (see Figure 6.3). 111 1983and 1984 the budgeted deficits were in fact larger than the actual ones. However, in 1985 the Saudi authorities balanced the budget. 'This was a complet@Iyunrealistic maneuver, and :King Fahd actually rehsed to sanction any budget proposal for 3986. A similar nonbudget year ocsz~rredin 1991 during the Iraq-Kuwait W;;rr,.The actual figures for these years in Figure 6.3 are International Monetary Fund estimates, Two external factors worked against Saulii Arabia" atkmpt to balance the market througt-t its role as swing pdu,cer: the reduced oil demand and the increased n u b e r of producers (see section 2.5). The other OPEC members atso contributed to the w e a k % of the market, as they produced above thek quotas. The power resource necessary for a hegemon to pursue a benevolent strategy is the abffity to substantially reduce totaf output. As Figure 6.4 illustrates, Saudi Arabia cut production substantially from February I982 to September 1985. As the figure also shows, this did not prevent the oil price from continuing to fall throughout this period. In this respect the benevolent: strategy was a failure. It could be argued that the Saudi Arabian behavior prevented, an even steeper price fall. The events of the spring of I986 substantiate such a clairn (see section 6.5). On the other hand, the fact that priees were kept high i,n this pericrd was an importmt hctor behind the development of nm-QIZCproduction capacity in the North Sea and Ah&a (see sertictn 2.5). :lf OPEC had accepted a steeper price declhe in the early eightks, it could have prevented the intensified competition in the market and thus strengthened its control w e r the market. 5uch counterfactual speculation cannot:be empiricaliy substantiated. Furthermore, it is not likely that consideratio~~s regarding alternative market effects were the most important for the Saudi decision to abancion the benevolmt strategy. Figure 6.3 shows how the actual budget figures for the Saudi Arabim government changed during the first half of

FlGURE 6.3 Saudi Arabian gc3vernr SCIURCE: Ministry of Finance and Nat Saudi Arabian National Center for Fi

Cc~zintryCase 1: Saudi Ambia

231

the eighties. The go~rernmentwas actually operatJing at a deficit. Given the argument in section 3.5, this was, in addition to the economic cmsequmces fr,r the royal family, also a political risk. %-hekingdom's internal security rested on. the ability to use the oil revenue to gain political legitimacy The external thrt.at would also increase with lack of money as the importance of Saucii Arabia in the eyes of the Llnikd States wodd decrease if the county became an oil exporter on the level clf many others and was accordingly wilthout any particular power position in the oil market. Again, oil hall security implications for the rulers of Saudi Arabia. The MIillingness of their ally, the United States, to support the c o n try and its policy was perceived as dependent on their position in the oil market. In the eyes of the Saudi Arabian mgime, the benevolent strategy was a proven fail~~re. Oi;l and politics once again pohted in the s m e direction: to regain market shares. 6.5 The Coercive Hegemon, 1986-1896

In the s u m e r of 1985 Saudi k a b i a no tonger accepted production beyond quota by other OPEC countries as long as it produced only half its yuota.16 If we examirte the relation between quotas and production in the of 1985, we f h ~ dthat \Nhile Saudi Arabia in the third ~fuarterpmduced 1.85 mbd betow full quota, a number of other producers were exceeding their quotas at prices below official sales prices. txl October, Saudi Aritbia shifted policy and increased production. 'This led tc:,a cmsiderafile overproduction in the market and heavy pressuce on spot prices, a situation several countries tried to compensate for by irtcwasling production. At the OPEC Conference in Geneva in Decmber 1985, the member countries resolved to change t h i s market strategy h r n the d e iense of a high oil price to the defense of the o13Eccountriesharket shares (see section 5.3). As Figwe 6.2 hdicates, fmm SKli the GDP of other OPEC mennhers fell, and from 1987 to 1991 the Saudi Arabian GDE)irtcreased while the other OPEC membersf c,nr2on average flattened out.17 The economic interests of Saudi Arabia took precedence over the need for the sw-ing producer to ensure internal cohesion in OPEC, In some sense it was a depoliticization of the Saudi market b&avior (see section 3.1). &e possibte political =ason for cocrcing the other OPEC mernbers into quota djsciplinc was the need for a hegemonic power to determirte the conditions for the minor players"behavior, and not the other way around. "'!he hcgemm'?i costs from maiintaining the system increase faster than the hegemon's capacity to support it, The Saudi Arabian oil production i n c ~ a s e dfrom 2.6 mbd in 1985 to 5.2 mbd in 1996, This level was sustained until 1990. The spare capacity thus made it to some extelzt possible to counter the negative el"

232

Country Case X: Snzddi Arab&

kcts of the expected price fall following the decision to change market strategy Saudi Arabja did, however, posses further spare capacity. This was to be utilized as two other large OPEC procjucers suddenly disappeared from the market. Un August 2,1990, Iraq invaded Kuwait. The UN Security Council declared a full boycott of l r q and the new regime fn b a i t . This irnmediately brought to a halt all oil flows from the two countries. 'The oil price increased from $38/barml to a h o s t $3Q/barr.elina couple of days. Other producers, particularZy Saudi Arabia, quickly filled the gap. After a k w weeks the OPEC production was back at almost the prewar level. As descri:bed in section 3.4, King Fahd perceived a further attack on Saudi k & i a by Saddam Hussein as possible, particularly when large Iraqi fnrces in Kuwait were deployed close to the Kuwaiti-Saudi Arabian border. The Iraqi attack on Kuwait had demonstrated that the tacit ailiance between hay and the Gulf states dufing the Iran-Iraq War was bascd oa situational conditions that at least Saddam no longer regarded as important, The Saudi Arabjan hegemony in the region has never been certain, and both Iran and Iraq have continudy challenged the Saudi Arabian oil dictates inside and outside OPEC (see (3hapters 3 and S), The outbreak of the fraq-Kuwait War proved once again the Fnstability of interstate relations in the region. The security problem of Saudi Arabia again becme an q e n t - concern not only for the Saudi, regime but also lfor its ally, the United States (see section 6.2). On the oil scene the disappearance of Kuwaiti and lrayi production eased the pressure on the OPEC quota, system, especially the Saudi Arabian restraints. The continuous LJN blockade of Iraq enabled the other OPEC colantries to contirnue a high-producticm p o k y without: disruptirng the oil price. Although Saudi Arabia increased the oil r c v m e s through this increase in production, the increase was not at all comparable to what it would have been bad the Sautlis let the prices go and suwtairned the prewar production level. As in 1979, they were alone in having substantjal spare capacity However, security interests took precedence. The attack on Kuwit was perceived as a substantiat threat to the security of Saudi Arabia itself. The guarantor, the United States, was needed h m immediate military operation-Desert Shield. The Saudjs would have found it difAcult to secure this help it; they had sirnultanet.,usly squeezed the conswing countries economically by letting oil prices skyrocket. As in 1973 the political interests took precedence, but, contrary to the 1973 events, in favor of the United Stateshand the const~mers'interests. The prompt action of Saudi Arabja in filling the gap left by Kuwait and Iraq clearfy indicates that the c c r u n t ~had no intention of being the sucker in the market-share g a m .

Cc~zintryCase 1: Saudi Ambia 6.6

The Mixed Strategy, 1999-2000

With the =entry of Iraq in the oil market and the demand effects of the Asian crisis, the oil price c m e under pressure again in 1,998. Savdi Arabia did, not accept new production cuts unless the other memberskompliance with existing agreements was above 913 percent.l"~;airt, Saucii Arat3ia foflowed a coercive hegmonic strategy. However, in 1999, Saudi Arabia joined Venezuela and Mexico (a n m C>I?ECproduwr) in setting the stage for a ~ n e w e dproducer agreement (see section 5.3), In. this casc the hegenton used a new hstrurnent, forming a small, more effective, high-level group in order to govern the market. Not sufprisingly, the kingdom's political rivals in the Mid& East, Iran and Iraq,werc not part of the group (see section 6.7), Having designed the policy in advance, and having held, a, pre-Conference meeting among the largest producers, the deliberations at the One Hundred Seventh OPEC C-onfere~~ce in March 1,999 Ilasted less than ten minutes. 'The Saudi Arabian strategy on this instance was neither a threat of price war nor acceptance of carryiq the burden on behalf of the others. Rather, the Saudi Arabim Oil Mhistea; Ali al-Naimi, used a strategy of more or less secret negotiations with a few other producers. Annuuncing a fixed p r o p o ~ inadvance, h preempted potential opposition among other OPEC members. It became harder to argue against the proposed deal, since both Venezuela and Mexico were equal partners in the group. When prices went up from $10 to $30 per barrel (see swtion 5.31, Saudi Arabia strongly argued in favos of increased productjon. Rathcr high profile diplomacy by the us Energy Secretary Bill Richadson gave the Saudi Arabian and OPEC decision a politic& overtone. However, Saudi AraZnian long-term ecoaomic interests also suggested that an oil price of $30 per barrel was too high, as it could easily dampen demand increases and c e price make consumers substitute oil with other energy sources. S i ~ ~the did not come dokvn., fwtha production increases were agreed in June 2000, h July, Saudi Arabia announced an inknt-ion to increase production by another 500,000 barrels per day. This was a unilateral decision, dlhough it was said that it would be done in "consuftation with other producers.""Y The ittdependent strategy of Saudi Arabia was strongly opposed by other C313EC members, not least the president of OPEC and Venezuela's Minister of Energy and Mines, rZIj Rodriguez Araque. At the time of this writixlg (July2000), itis impossible to know if the proposed indepencient policy has been pursued further, The ewents of 1999-20UO do, however, it,lustrate how Saudi Arabia possessed the key ability of a hegemon as argued in the model in section 6.1, namely, the ability tc:,adjust its strategy and be flexible in its refations tc:,other pro-

234

Country Case X: Snzddi Arab&

ducers both inside and outside the OPEC frame. It thus constitutes a mix of the coercive and benevolent hegemonic power. W~ik the first half of tbr eighties saw Saudi Arabia swinging down, the first half of the ninetks was marked by Saudi Arabia's sMIing up, taking the mtirc room for production left by the UN boycott of Iraq. Obviously the swinging now was on Saudi Arabian terms. The hegemcm had regaiined stretngth, although the room to mmeuver was mare restricted due to factors discussed in section 2.5, in particular the taxation of oil by the consumer countries and the so-far unrestricted developmat of nonOPEC production capacity The period covered in this section shows bow Saudi Arabia tried. to regain its hegemony ammg the oil producers through a strong coercive strategy in the a u t u m of 1985. The benefits to the kingdom from this changed strategy were to a large extent enhmced by the Iraq-Kuwait War and the resufting UN smctions against Iraq. These factors gave room for increased Saudi Ar;Ebian oil reveznues: "Saddarn Htassein" tanks acc~mplished in twenty-four hours what Saudi oil policy had h i k d to achieve during the decade of the 1980s: they boosted prices mzd 5audi oil prtlduction" ((Wilson and Graham 1W:188)- Saudi Arabia took advantag of the situation, and, in line with the model in section 6.1, the fall of the hegemon was countered. FoiloMling this the Saudi Arabian position was strengthened as demmd increased m d made it possible to sustain a high Saudi Ara:hian oil production even as Iraq came back into the markt-tt, But aisc:,hecause the poii.tica1 relations toward Iran and Iraq fmpmved, as is the tlopic of the next section. 6.7 The Relative Power of the Hegemon Tn describjng a hegemon" posit.ion, it is kportant to specify the begeman% relative size, that is, the share of the market. However, this is not sufficient in understandi,ng the potential lhreats to trhe hegemon..e)ne also needs to compare the power resources of the hegemon, or leader, with the power =sources of other potential hegemonic states (see Nye 199(1:174). As outlined theoretically in section 6.1, it is the relathe power m a n g the large" states that dcfhes the dynamics of hegemonic cooperation. In the first (incapable)phase horn 1973 to 1981, the Saudi Arabian share of OPEC production was about the same as the franim and Iraqi production combined until Irmian oil pmduction dropped due to the revolution in 1978 (see F i g m 6.5). Althou& Saudi Arabia pmduced above 25 percent: of the total OPEC output, the relative differetnce betweeln Saudi Arabia and Iran was not particularly large, h t h i s respect, the position of Saudi Arabia was not so special. Frtrthermore, the Saudi Arabian attempt to domhnate OPEC was chaltelnged, inparticular by the shah of Iran.

Cc~zintryCase 1: Saudi Ambia

235

The benevolent phase commenced in 1981, when Saudi Arabia produced more than 40 percent of total OITC pduction. n e potential challengw" Iran and Iraq, were at war with each other, anci their oil prtlduction had f d e n from 25 percent- of OPEC p d u c t i o to ~ ~f O percent. Iran was no longer a potential hegemon, and posed no threat to the Saudi Arabim pcsition. As pointed out in section 5.2, Saudi Arabia viewed the price pressure of that time to be a short-term, prolslem. Thus, it m d e sense to take on a benevolent hegemmic role, Figure 6.5 shows the costs of this strategy, expressed as a fall in share of tot& OPEC output. The Saudi Arabian share fell from above 40 percent to just above 20 percent from 1981 to 1985, In accordance with Gilpin" theory presented in section 6.1, taking on the role as hegemon implied &tolerable costs. The coercive phase commenced in October 19635, when Saudi Arabia chmged its market skategy to regain its position (see section 5.2). This caused OPEC to foLlow suiZ in December 1485, and the oil price fell dramat-icalfy in the spring of 1986, As Figure 4.5 shokvs, Che Saudi share of o13Ec production did incrcase somewhat f r m 1985 to 1986, but was more or less stable until the Iraq-Kuwait War. NOW the reborn hegemon took advantage of the situation and hcreased produdion, covering the missing frayi and Kuwaiti output. Later on, the UN sanctions made it possible to sustain a high level of Sacrdi Arabian oil pmduction. At present, the Saudi Arabian policy is a con^rbjnation of the c o e ~ i v and e the benevolent strategies, When prices come under pressarc?, as they did h 1998, a coercive hegemon wit1 not =duce output, m will at least force others to cmtribute substantially to m k e room for the Iraqi oil, A benevdalt hegem m in this situation w d d cut back its own production, to stabitize the market. Saudi Arabia presently (mid-2000) does both. This chapter has shown hokv Saudj Arabia gahed its dominance hsi.de and outside OPEC, and the subsequent costs it jncurrcd from this role. The case of Saulii Arabia clearly iltustrates Gilpin" argurnent that "the tendency is for the economic costs of mai,ntaining the status quo to rise faster than the economic capacity to support the status yuo" (Gilpin 1981:11). However, the case of Saudi Arabia also substmtiates that the response to the rising net casts for the hegernonic power can be offensive, in the sense that the hegemon need not acquiesce and become a state in line with the others, but rather can find ways to regairt its position (see Figwe 6.1). In. the case of Sazldi Arabia, this happencd thmugh a change from a benevolent to a coercive strategy, and later by corrtblnjng the two strategies. In addition to oil-related factors, the politicat volatiiity of the internal and external affairs surrounding the Saudi Arabjan regime has made it mow difficult for the kingdom to utilize its market power. The separation of oil and politics has been the Smdi Arabian polity for most of the pe-

Cc~zintryCase 1: Saudi Ambia

237

riod. 'This was clearly argued by then-Deputy Oil Mb~isterSaud al-Faisal (now foreip minister) in 1972: "YOUalways hear that you can" sseparate oil from politics. I sirnply do not see Mthy not."'"" &e nothle exception was the 1,973ennlaargo. The problem is that the strategy of separating oil and politics is a viable strategy only as long as other actors accept it. T h i s was not the case with the United States hthe sixties, nor with some other OPEC members during the seventies, nor with the internal opposition during the eighties. Thus, the Saudi Arabian regime has continually had to consider the likely political consequences of its oil policy. Today, the consuming countries seem to accept. a fairly Kgb price in order to achieve stability in the market, Such a price might also c a h internal opposition in Saudi Arabia (see section 3.5). With a mom moderate Islamic rclgim in Tehran, and Baghdad no longer able to pose militmy threats against Saudi Arabia, the enviso ent is better than ever for depoliticizing the od-~o"J-cercooperatbn (see swtion 3.1). M a h g C->12ECa pure economic organization will rewirt. political will and ski21 from Che Saudi Arabim regime. On the other hand, the opportunity to do so might never be as good as it is as of this writing. New poiitical issues will most certahly influence the oil market again. As has been described in this chapter, the behavior of Saudi Arabia is based on the argument put forward in section 3.2. Without includhg the security considerations of the Saudi Asahian regime, one omits the most important motivation fur the ojl-related concessions given to other OPEC countries.zl The seeming w e h e s s of the hegemcmic power is due to an undesstmdable preoccupation wi& sec~~rity needs. Politics limit the economic options of the world's largest oil exporter. Notes 1.1 will use the term ""hegmony" ~ n o n y m o u s l ywith "leadership." Tn fact, hegemony comes from the Greek word for political leadership. 2. See also Underdalrs (1994a) argument that the hqemon exercises leadership both through rmilateral action and through coercive leadership. 3. This section is based on Claes 1986a. 4. %day the proven oil reserves of the Middle East are 660 billion barrels, compared with the above-mentioned 40 billion barrels in 3.9550. 5 , Sheik Ahrned Zaki Vamani was the Saudi Arabian minister of petroleum and mineral resources from 1962 until 1986. During this period he was the most visible Arab in the international, oil industry He was the most media-wise of the OPEC ministers, making his words more often cited and thus more often read than those of the other representatives of the organization. He also war; one of the most media-prominent Saudi Arabians, although not a member of the royal family. Disagreement with the royal family was the reason for his departure Ercxm office in 1986.

238

C o u try ~ Case l: Saudi Ambin

6. The companies now tried to make the best of the situation and demanded compensation for loss of future income, a demand that was flatly denied by Yamani (Evans 1990:430). 7. Newsweek, February 17,1992:22. 8. FRUS, 1950, vol. 5:1190-1191, cited in Yergin 1991:427428. 9. Summary of National Energy Policy Plan, US Department of Energy, Energy Insider, Washington, D.C., August 3,1981, p, 3, cited in Lieber 1992:157. 10. The statement of the vice president was in conflict with the official White House policy: "'Poor George,' was the way a senior White House official disparagingly talked about him, adding that Bush's position was not Administration Policy" (Yergin 1991:757). 11. Quoted from the Egyptian magazine al-Mzrsazmr in Middle East Economic Survey, and subsequently in Golub 1985:9. 12. On August 23, Sadat visited Faisal in Riyadh, and probably informed him of his plans for war with Israel (Schneider 1983:219; Robinson 1988:85-86). 13. Speech on American television, printed in Petroleum Intelligence Weekly, September 10,1973, and referred to in Schneider 1983:220. 14, There are different opinions about who initiated the auction plan. Moran (1981:255) claims that it was U5 Treasury Secretary Williarn Simon who "got the leadership of Saudi Arabia to announce publicly an auctionff; Robinson (1988:137) argues that it was Yarnanifsidea. 15. The outbreak of war between Iran and Iraq did not affect the oil market in the same manner as did the revolution in Iran, At the time the war broke out, there was a supply surplus in the market of about 2 mbd. The outbreak of war led to a reduction in both countries' praduction of some 4 mbd, leaving only 2 mbd of actual shortage. This was easily replaced by other OPEC and non-OPEC countries' production increases. 16. This is expressed particularly clearly in two interviews with Yarnani in the Jeddah newspaper Asharq a1 Azusnt, quoted in Middle East Economic Szirvq, June 3, 1985, and Petroleum lntellipce Weekly, June 17,1985. 17. The volatility in 1990-1991 was due to the dramatic consequences of the Iraq-Kuwait War for the CDP of these two countries. 18. Petroleum lntellipce Weekly, August 3,1998. 19. Statement by the Saudi Arabian Qil Minister Ali al-Naimi by the Saudi Press Agency on July 3,2000, reprinted in Middle East Economic Surl~ey,July 10,2000. 20. Petvolcttm Ivfelligence Weekly, November 20,1972, cited in Golub 1985:8. 21. Saudi Arabia's policy in the oil market is taken by more than one observer to be first and foremost decided by internal and external security interests (Golub 1985; Moran 1981; and Bjmrk 1988).

OPEC: A Successful Cartel?

The question guiding the discussim in this chapter is: ?i, what extent have the individual oil producers behaved cooperativeiy, and to what extent have the oil-producing countries increased their economic gains thruugh such cooperatim? It is the ailn of this chagter to explore the erctent that collective action has been a necessmy condition for the producers to excrcrise influence over the international oil market and thereby itncrease their economic gains, This raises the counterfacbal guesticm of what the oil producersf behavior, and thus the oil price, w u l d have been in che absence of cooperation. A cmpetitive international ail market could be taken as the noncooperative outcome. However, empirkally this is a second-best assumption. As discussed in Chapter 2, the international oil market has never been a competitive market, with the poss&Xe exception of sone months in 4986. Thert. is no solid basis for empirically testing l ~ much w the oil price has increased due to cooperation m o n g oil producers, Econontic models provide figures based on the assumptions entering the calculations. In reality we cannot know what the oil price would have k e n without oil-producer cooperation. In section 7.5, the djffercnce between the cooperative producers' price setting and income revenue is comparcd with assumed price and rt?venue in a competitive market, It is atso irnportant to investigate the differences between the OPEC members regarding their contribution to the collective good.1 The costs and benefits of coopwation can be unevenly distributed among the oil producers. 'Thus this &apter provides a partid explanation of the price and production poliq of the individual member countries discussed in Chapter 5. The discussim in this chapter will be organized according to some key aspects of cartel theory. The different: aspects of-cartel tkory correspond with the phases of the OPEC cooperation discussed in Chapter 5, which also highlights different aspects of the internal bargaining pmbIem facilng the orEc mernbers. The period horn, the estab)isfintent of OPEC in 1,960un-

OI~EC:A

240

Szr ccczssfir l Cartel?

TABLE 7.1 Outline of Conditions Affecting the Incidence of Cartelization, Graded on Basis of Feasibility and Necessity Fmsibility: Feasibility: Fensibilily: ExwEfcltzt Good Poor Illnr.ket Ccsndr'ti011or Need: unneeczssn~y Need: hclpf~rl Need: essential CFz~racterisfic Tacif COIJUS~OI~ Cartel Competition Number of firms Concentration ratio Type of product Rate of technological change Frequency (and size) of sales Opportunity for secret deals Rate of grow& ELasticity of demand Production costs across firms wrrricx.::Greer 19812:277,

Very few (2-5) Very high Standardized None

SveraX 42-25) High-medium Slightly different S1c>w-mc>derate

Many (30-t-) Low Differentiated Rapid

Frequent (small)

Often (medium)

lEJumpy(large)

None

Some

Great

Slow Medium Low (less than 0.5) Medium Identical Similar

Rapid High (2) Diverse

till971 is of little interest for the question discussed in this chapter, as the

orFc countries had little influence on the oil price durhg that period. (Someaspects of the OPEC countsieskooperation in this period. we= djscussed hsection 2.4.) 7.1 Conditions for Cartels A cartel can broadly be defined as " m explicit arrmgeznent among, or on behalf of, eatergrises in the same line of busisless that is designed to limit competition among them" "tacking and Watkins 1948:3, cited in Greer 19M:2h%).A more precise and specific oil-related definition is presented by Mead (1586:215): a ""professional definition of the cartel term-a single seller or a gmup of sellers operating in unison to reduce output below competiti,ve levels in order to obtain a price above competitive levels~" The factors presented in the model of industrial organizatim in section f ,5 give certah structural condjtions for Che feasibility and need for the establisheat of a cartel. Table 7.1 evaluates the market conditions affecting cartdizatim cJf a market. 'The hndammtal argument is that cartels are easiest to establih when they are least needed.

c~r~r:c: A SucmsJ111Cartel?

241

Some of the aspects ovlCIined in Table 7.1, (e.g., nu~nberof producers, frequency of sales, growth rate, and.elasticity of demand) are discussed in Chapta 2. To mderstand. the bcthavior of OPEC mernbers, one need?; to combhe the effects of the external conditions (Chapters 2 m d 3), the role of institutional arrangements (Chapter 4), and. the bargaining between members (Chapters 5 and 6). The concept of tacit collusion produces much the same outcome as cartelizatim or, in fact, m oligopolistic situation. The point is that with the market conditions spelled out in Table 7.1, there is no need for overt communication or agreelnient in order to produce high and stable prices. This can be done "solely through a rational calculation by each seller of what the consequences of his price decision would be, taking into account the probable or virtually certain reactions of his competitors" (Turner 19662, cited in Greer 1,9&:269), Some authors hold this to be the case for OPEC prior to 1982: Cartel theory has important irnplicatians for analyzing ot3sc,It m w t be applied with. care, however, for CWEC is nat a true cartel. Members ctlrre~ttly agree on a common price structure, but they have never reached any kind of formal agreement on sharing of production cutbacks, the hallmark of a fullfledged cartel. (Willett 1979b:584; italics mine)

Applying the insight gcrnerated from this general cartel theory to the oil industry suggest that tho stwcbral conditims for the cartelization have to various degrces been met in the oil market since optic was b r m d (see Table 7.2). The number of firms or, more correctly; the number of oil-reserve owners, increased durhg the period, and the conce~~triztion cmsequently decreased (see Chapter 2). The type of product did not change, although the new producers introduced new oil types or quaiities. 'This, however, does not r c p ~ s e n increased t variation in the oil &fferen.t.ia%s, but rather more oil qualities within the same scope or range of oil differentials. There were, of course, changes in the technotogical development in the cJil industry during this period. Che important consequence of this was the development of the Norwegian Continental Shelf as an oil pmvin.cc, which would have been impossible in the sixties. The speed of technological crhange has been relatkely slow, although it has becm somewhat h i g h in the nineties. Frequency of sale is an odd factor in the oil market, and in a sense not applicable. If one interprttts it more in line with the trading inslrumemts discussed in section 2.6, it illtastrates the change from long-term contracts to an increashg amount of spot, or spot-related, t was low fn the sales. It is suggested that the opporh;lnity for s e c ~deals sixties, greater inthe seventies, and low in the eighties. The reason for Ihe

242

OI~EC:A

TABLE 7.2 Cartelization of Internationa t Oil 1950s 2 970s Number of firms Few Several ("Sven Sisters'" (I3 OPEC members)

Szr ccczssfirl Cartel?

Concentratiron ratio Type of prc>duct

Vwy high Slighdy different

High-medium Slightly different

1980s nrld 2 990s Many (OPEC -t- incr. number of non.-QPEC producers) L&W Slightly different

Rate of technological change Frequency (and size) of sales Opportunity for secret deals Rate of grow& Elasticity of demand

Slowmoderate Lumpy Oarge)

Slowmoderate Often (medium)

Slowmoderate Frequent (small)

None

Some

None

Rapid Low (less than 0 5 ) Similar

Medium Medium

Slow High (2)

IncreasingXy diverse

Diveme

Prt>ductioncosts across firms

low oppmtunity in the sixties is quite different from the reason for the same result inthe eighties. h the sixties nnnst oil was traded by a handlul of vertically integrated international oil companies watching each other closely. The terms of trade were accordingly known to the insiders of these companies, m d the possibilities to go around this strudure were small (see section 2.3). With the OI~ECtakeover, the actual trading became more concealed, as it becan-te a relation betwren an oi-producing country and its buyers m d , to some extent, the internatio~~al companies. The relationship between the OI'EC producers was not at all as close as that between the companies. Thus, the room for secret deals increased. In the eighties, however, the masket situation in this regard banged again. The market became transparent due to a large nurnber of traders independent of producing countries or integrated companies. The international oil press also keeps a close watch on the clevelopme~ntsof new trading tactics by the producer countries. The possjbility today of closing secret deais and keeping them undiscowered by the market p s s or the traders is thus ratt.ler mall. The growth rate has slowed, as djseussed in section 2.5. The elasticity of demand was low in the short run, after 0 1 2 ~ ~ 'price s increases in the early seventies, but has been higher in the long run, manik?sted by the halt of the growth in consumption durjng the eigMies, The

c~r~r:c: A SucmsJ111Cartel?

243

~rariationof production costs has increased as the number of producers has increased, including producers in very different petmleum regions of the world. As Table 7,1 suggests, a cartel is easiest to establish when it is least needed. This fits perfectly with tbe WEC organi.zation, which in this perspective had little importance in the seventies, as the members were able to increasc prices and prodnctim without the help of the organization. FIowever, in the eighties this was no longer so, and.it becam necessary to strengthen the institutional armgements to achieve higher prices, or, more corwctly, avoid price decline. 'Thus, thc need of the producers for a cartel hcreases simultmeously, as it becomes more difficult to establish and maintain it, 7.2 Cartel Behavior

The structural conditims for the possibility and importance of cartels are, however, not enough to explain the actual existence and '"smccess" of cartelization, 'f'he success of a cartttf depends on four hctors ccmnected with the behavior of the cartel (Crkmer and fsfaihani 1991:38). A. cartel must

I. 2, 3. 4.

determine a price for the group as a whole; determine a production level for the group as a whole; allocato output among members; and detect and p u ~ ~ i cheaters. sh

The first two points have to do with thc.cartel's mlations to ""themarket,'hhiXe the last two points have to do with internal bargaining problems. 'f'he key problem for OPEC is rooted in an ambition to set both price and production levels at the same time. A cartel Ihat does not control all (or at least almost all) pmduction will work d y if it sets a price and.defends it by increasing productisn if the price is above the target and rttduchg production if the price is below the target. -This will, discmage other producers from entering the market, as the price wifI not go substantially above the target price, and it will secure inc:ome as production is reduced to keep the price at the target level.2 This illustrates the role of the cartel in the market, The internal allocation of production and the subsequent adherence to a set production level create addjtio~~al problems b r the cartel members. As poi~ltedout by Ml'itlctt (1979b:582), ""whil. all, of the oil exporting countries have a mutual economic interest: in restricting supply and keeping oil prices well above competitive levels, they havc substmial differences of inter-

244

OI~EC:A

Szr ccczssfirl Cartel?

est concerniing just how high prices should be and how great the supply restrictions of each individual producer should be." The followhg sections address the four factors one by one.

Determi~inga Pricefor the Group as a Whole The first problelx, the price, has probably been the most problelxatic issue in the OPEC bargaining. There arc several reasons for this. First, production levels did not genuinely support the official, or fixed, price until f 982. Second, the official omc price was discounted b y several individxlal producers when there was a surplus in the market, Third, the FRcreased nurnber of nm-OT~EC producers reduced QI~EC'Scontrol over the market priee. C)f: course, the aim for an indjvidual country is not the priee in itself, but the inmnne or revenue that a certain price level kplies, This is obviously depmdent not only on the price level but also on the individual country's production. However, the OPEC caoperatiol~described below shows that OITC d i s a g ~ e don the price issue long before the quota. debate started. Fur&ermore, &:he target price can be seen as a norm for evaluating the actual price in the market., which in itself can be seen as exogenously givcn, The possible production limits or expansion is thus an answer to discrepancies between the stated target price and the actual market price. 'The belnavior of the ( I P F ~comtries from 197%makes it a reasonable assurrrption that they pesccjved the price as a given in the first place, and that their subsequent production limitation was a response to the price decline.

Detrmrini~zga Pmducfiorz I,e?ielfor Z k Gmup US IE Whole The second probfem, the total quota level, is meant to be fixed according to the OT9EC production that wifl realize the price target, at t h e s when such a target has exjsted. C;iven an agreed understanding of the target price, the total possible OI~ECquota-or, in, oil jargon, the "cafl on o12~c"(see section 4.3) is a given figure, provided thercl is accurate infornation on demand and n o n - ~ production.Vhe ~~c problem arises when the members disagrcte on what this level is. Since the quotas were established in March 1982, the market conditions have confronted the 0 1 2 comtries ~ ~ with the choice either to cut prices to =gain their market shares or to cut prodrrctim to sustain the price level.When Saudi Arabia increased production in 1985-1986, one of the stated intentions was to make non-OI~EC productim qrofitabla and therclby stop it. This strategy worked well against the us producers, especially the so-called stripper-wells. Here, the investment costs are ftfw, and the variable costs of production are. relatively high cornpar& MIith the fixed costs. fn the North Sea, this relation-

c~r~r:c: A SucmsJ111Cartel?

245

ship is reversed. Here, the fixed costs, especially the kvestme~~t costs, are relatively high compared with the unit costs. Accordjngly, production continued even if the profit margin was low or wen nonexistent, as it was better for the i,nvestor to partially cover investmelrt costs than not at all, which would be the inlplication if production was stopped. Even fields not onstream, but almost develcllped, wodd be put in place; this is especially true of large fields, as was the case i,n the N m e g i a n sector of the North Sea. If this strategy by o13Ec had been sustahed for a longer period, new investments in the Narth Sea oil sectors would not have been made. The long development time of Norwegian fields suggests that prices would have had to be low for several years, It also fotlows that Norwegian cooperation with OPEC codd ccmsist only of increasing production at a lower pace, not cutthg back on the production level, since there was no economically defensible way of stopping already-pl production expansion. From the late 70s a m c let non-OPECproducers take their share of the market as competitive psice-takers, and its mmbers themselves filled the gap between the nm-OPECproduction and total demand, acting as swing producers. 'This, of course, gave OPEC little influence on. price, as it ~c was set by the relationship betvveen total dennand and the n o n - o ~ pro, OPEC members duction. As non-OPECproduction capacity i n c ~ a s dthe found themdves supplying less and less oil at lower and lower prices. The quota decision of 1.4182 was an attempt to change the price prohtem; the market-shrc-. decision of December 3985 was an attempt to handle the production problem.

From the outset, the quotas were disputed. Iran was particularly dissatisfied with its allocated quota. This has led to continuai bargaining inside OPEC over Ihe level of totd production reduction i,n the organization and the distribution of quotas. The quota bargainjng is special as the actors bargain at t w levels: iirst, ower what is to be the level of total quotas; and, second, over wt.tat is to be the quota for the imdhidual mernber. Then, t h r e is the prohlem of compliance, If one asslrrnes that the level of total quota^; is set to aehieve a certai~~ price in the market this level of total quotas is not negotiable, given the price target and assumed dmand. Whetfrter the agreed production level ~ a l i z e sthe price target is another matter. The merrtbers mi@t disagree about whether the assumptions are correct, and they rnight d i s a g ~ eon the price target. It follows that the quota-setting bargainjng is mostly distri:butive bargail7hg-allocating irrdividuai quotas within the limit of the total quotas, which, is set exogenously

246

OI~EC:A

Szr ccczssfirl Cartel?

While the criterion for setting Ihe total level of quotas is rather rigid, the criterion for individual allocation is virtually nonexistent. Obviously, histcrrical production level is important, but: production capacity (and spare?capacity) might prow to be the most intportant .factor.Also, income reytlirement, poverty, war costs, and reconstmctim programs have been directly or indirectly i~~troduced as argrtments for &creased quotas. 11% total, s m e forty different criteria have been introduced. Setting the quotas is an example of an iterated, game. As pointed out in sectioz~4.3, quotas are seldom set for more than one or two quarters. Thus, agreehg to one quota level at one m m e n t is not formally bindifig a at for more than some months. However, agrceixlg to a "low" ~ o t level one point in time might make it harder for the country to argue for higher quotas at a later point.

Detrcfiorz and Pur?lshmc)l~f of Che~ters Once the quotas are set, the game chang@sto one of collecthe action. If the total w t a level is correct@ set a d thus achi.ves the decided price level in the market the axle@countries find themselves in a traditional prisoner" dditernma situation. This could be analyzed on the basis of a high oil price being a clrliective good for the oil producers (Malnes 1983). Applied to the oi-producer relations, the point of departure?i,n the collective good is that all actors have an intercst in hihation of production to ensure a high oil price. At the same time, it is in the actorsf interests to sustah their own production volume. In such case, the actors c m gaiin by cooperatkg (reducing production), but individudly they can gain more by not cooperating while the others do. The oit price can thus be regarded as a collective goad. A. collective good is, in this sedion, understood in accordance with Olson's definition: "If person Xi in a group X, . . . Xi . . . XI>consumes it, it cannot feasibly be withheld from others in the grottp. In other hvords, thosc who do not purchase or pay lor any of the public or collective good cannot be exchded or kept from sharing the consumption of the goods, as they can where non-coil~tivegoods are concerned" "1sm 1965:314-15). Olson then distinguishes between exclusive and inclusive goods (01son 1965:38). Exclusive goods are goods of which one actor's consumption reduces other actors' consumption equally. Inclusive public goads are similar to pure public goads, as defined by Samekon: "each individcm?iumptionof such a good leads to no subtractim from any other ual"~ individual's consumption of that good, so Xll,I = X i + j simultaneously for each and every i individual and each collecthe consumption good" (Samuelson 1954:387). Obvbuslyf the price of oil can have a rivalrous effect on Chc producers, and thus fail to satisfy the defiIlitioln oE inclusive-

c~r~r:c: A SucmsJ111Cartel?

247

TABLE 7.3 OPEC Cartel Strategies, 19173-1 9953 Prz'ce-Seftip18 Dec. 75Feb. 79 Yes, fixed Feb. 79-*May 81 No May 83 -March 82 Yes, fixed Mar& 82-Dec. 85 Yes, fixed Dec. 85-Now 86 No Dec. %-Aug. 90 Target, weak form, not fixed Aug. W-Feb. 92 Target, weak farm, not fixed Feb. 92Target, weak form, not fixed

Prodl~cfz'onQuotas No No Nct Yes Yes Yes No, abandoned due to war Yes

ness. The oil price will consequently be regarded as m exclushe collective g w d for the cJi producers. The following four sections deal hvilh different aspects of the collective action by the OPEC mmbers, Both the price and production behavior oi OPEC members are relevant, but most emphasis will be put on the production behavior and its rektim to the qztota agreements. The final section discusses the OPEC: mmhers' economic gainS as such. 7.3

OPEC as a Price-Maker

If O13EC should behave in line with cartd theory, it should set a fixed price, based on. perceived demmd co~~ditions and the assumed production outside the cartel, and then increase production if the price rises above the agreed level, and decrease pmduction if the price falls bebw the same level. By varying production in this way, o r E c Should be able to maintah the agreed price (see above). This implies that the cartel's role in the market is based on two diment;ions: price-setting and production quotas," Table 7.3 shows how ~ P E Chas performed along these dimensions, The price-setting since 1986 has been a target price that indicates at *at OPEC should aim; it has not constitutd a binding or regulative figure in the OPEC deliberations, Mihether the price was set in a fixed form or the weaker, target form, one should expect a strong positive correlaticm between changes in price and changes h production. If orFc behaved like a cartel, it should have increased production when price ulcreased, and vice versa. Table 2.4 reports the cormlation between the individual member countries' production and the oil price from Jmuary 1973 to December 1995. The data are presented as time series, where the cases arc produrtjon and average price in individual months. 'The data have been divided into four shorter time periods: from Jmuary 1923 to December 1978; April 1982 to

c~r~r:c: A SucmsJ111Cartel?

249

December 1985; November 19886 to July 1998; and April 19992 to Dece~nber 1995. These four periods exclude the months durixlg, or in the atermath of, revolution and war, as well as very volatile periods, such as in 1986 (January 1979-March 1982; January lf3636actober1986; and August 1990-March 1992). The correlations for the period from January 1973 to Decen-tber 197% should be read with great care, as t-he price trend is t-he same throughotlC the perhd. The prjce dropped between only two of seventy-two months, Thus, the correlation figures for this period do not cover t-he cartel's ability tc:,reduce pmduclcion when t-he price falls. 'I'he last calznimn hTable 7,4 repo&ts the correlations for the whole period, includhg the m n t h s excluded in the other columns, when prices changed does not consfit-ute t-he sum of t-he other dramatically*Thus, this colu columns. As the figures in the last column s h w , o d y in the case of Saudi Arahia is there a small, positive, but insignificant correlation between production and price. AccordZng to the criteria for a successfd cartel, any price movement, positive or negative, large or small, should be accompanied by similar changes in production. As Table 7.4 shows, no such correlation can be the contrary, idenlified for any mentber for the period as a whole." for several members, there was a strong, negative, and significant carre;lation between price and production (Gabon, fran, Kuwait! Libya, Nigeria, Qata, and Vtnezuela). This suggests that the relationship between production and. price is the opposite of the cartcl-theory assramption. The ilncfividual producer tries to compensate for f a k g prices with increased production, and reduces production when prices increase above the target, This argument has been formdized in the economic model called the b a c k w a r - i n s u ~ p l y curve model (CrPmer and Isfahani 1991:45). The point is that- the producer would not try to maximize irtcome, but seek a more or less specified target for its revenue; thcr strategy can be called an iwome target strategy The mociel is based on the Hotellix~gtheorem (Hotelling 1931; see sectim 1.5). According to this theorem, the oil-exporting colxntq can choose between spending the oil Rvenues and investing them as a form of eyuity; the cltecision is one of weighing the uti:li,tp of current production against the utility of production at a later date. The producer may fjnd. its income so high that production is reduced as prices inwase. "T'his argument is, hwevw, fnfluenced by several other distorting lactars. First, the producer must anticipate that the oil price will increase more thm the existirrg interest rate. If not, it should instead produce the oil and save the income in other assets. Second, the producer must consider the gedogicat fact that it has a limited amount of oil reserves at its disposal as an actual cmstraint on its cltecision. :IE reserves increase more than produdion, the idea of the fixed stock becomes less obvious. Third, the producer must

250

OI~EC:A

Szr ccczssfir l Cartel?

TABLE Z.5 Correlation Between Price and Production by Price Movements Dccricnsz'tzg Price If2creasing Price Algeria Ecuador Gabon Indonesia Iran Iraq Kuwait Libya F.IIigeria Qatar S. Arabia UAE Venezuela OPEC r = currefation coefficient n =r number of c)bseivations (i.e., manths prcdudion) I? = two-tailed sipificance test ?~OUIduction Quota Over-productbn Over-pmduction (Yo)

L 7" 99 17,167 33 0.19

20,752 4 7,603 3,149 47.89

25,800 24,045 1,785 7.43

4,667 5,505 -838 -15.23

4,950 4,579 371 8.89

8,166 8,894 462 2.83

Snudi Arabia

Production Quo>ta Over-pmduction Ove~production(76)

OPEC fexcl. Snudi Arabia) Production 2 2,532 15,802 13,661 15023 Quo>ta Over-productiron 871 2,779 Over-prod uctiron (76) '7.4'7 21.34 WCI~CE:OPEC 1990; OPEC Buillefitl;i""efroleumEco~zomht,

17,634 16,011 1,622 10.13

tern h&is period was a failure. After the outbreak of the Iraq-Kwai't War, the quotas were abandoned until April 1992. 'The f03.tokvin.g phase, from Ap"il I992 until Uecember 1935, was characterized by the absence of one of tJle major OPEC p d u c e r s , namely, Iraq. The r ! boycott ~ left more room for the olher producers, and Chis in parficuhr was il:ld by SaudiArabia. As Tabte 7.8 shows, the total quotas in the first anlt second periods were al^ most the s m e , but substmtialiy increased in thc.third period. It is hard to determine to what extent- the OPEC colmt-ries' production was reduced due to the quotas. Several other reasons fur changes FR production are discussed in this book, From time to time large OPEC prtlducers have neglected their quotas, refused to accept alfocated quotas, or

260

OI~EC:A

Szr ccczssfirl Cartel?

simply announced another quota than what was agreed upon at OPEC meetixlgs. On the other hand, O I ~ E Ccountries"duction has been restrained by technical problems, internal political m d economical turmoil, and, not least, warfare among mernbers of the organizatim. The next subsections do, however, indicate a possible explanation for differences in quota compliance among OI'EC members.

The ExCelzf of Corrtpfinnn. Robert Axelrod (1984:lQ)shows that if the actors meet m hfhite or UIh o w n number of t h e s , they will have incentives to choose conditional cooperation because ""the shadow of the future"' influences the actors' strategic calcdations in the g a m being flayed. By mans of computersimulated tests, Axelrod shows that the strategy '*Tit for Tat" ((which invokes cooperalion in the first round, after which the action the opponent adopted in the prwious m m d is chosen) gives greater discomted utility than the skategy ""non-cooperation at all times," wwhicb is the strategy when conforming to the orlhodox prisoner's dilemma game. AnaXylTically, this is not very helpful because a t ~ yfeaible payoff vector can be supported as a Nash equilibrium outcome if there is a sufficient amount of discounting (Frieban 4971 and 19%:103). Furthemore, ''in ccases *ere subsets of the players find it collectively .~vorthbvhileto provide the pufilic good, there arises a quite different strategic problem, which results from some players having an incentive to ensure that the suhset which provides the public good does not hclude themselves" (Taylor 198282). This cdls for the inclusion of time as a factor in the study of t h W E C q u o t a - b a r game. fn the following discussion fie OI'EC quota cornpliance can thus be seen as n-person prisoner's diiemma it-rrakd m indefinite number of times. The itmation of games opens the possibility for both punishment and forgiveness, and for intcrnal bargains as to who should be allowed a free ride on the others" cooperative efforts. The strategic puzzle facixlg the individual oil producer is to what extent one's own abidance of y?xotas is likely to sustain or raise prices, and on the other hand to what extent one's okvn bscakhg of quotas is fikely to lead to a fall in prices. The answer to this is determined to a great extent by the but also by the value of the wide range of amount of overproductio~~, cmtextual factors discussed in CI:hapter 2.Mowever, it is also highly dependent on the internal strategic question: to what extent one"s own quota ahidance will make others cooperate, and on the other hand to what extent one's own breakhg of quota agreements will make others defect, This is the question to be discussed in this section. 7"he basic model cJf an n-person prisoner" dilernma is illustrated in Figwe 7.4. Seen from the perspective of player i the x-axis =presents the

clr3r:c:A SucmsJ111Cartel?

FIGURE 7.4 N-person prisoner's dilemma

number of other actors choosing to cooperate, while the y-axis measures utility.11 What happms when one moves from a two-person prisoner's dilemma to m n-person prisoner 'S diletnma is that the so-called k-point can be identified: '"here is some number, k, p a t e r than 1, such that if individuals numbering k m m m choose their unpreferrt-ld alternative and the rest do not, Chose who do arc better off than if they had all chosetn their preferred altmatives" (CSchelling 19713:218), The critical a m n t oi cooperation ( k ) illustrates the smalle?jtcoalition size that wilt profit by choosing cooperation rather than a situation where nobody cooperates. It is always better for the individual actor to choose to defect than to cooperate, no matter how many other actors have chasm to coapcrlrate. The value of such defection increases with the number of others that cooperate. This is the so-called free-rider hcentirie. From tbe quotas initiated in April 1982 until September 1999, the average ovesprodrtctjon was above 20 percent. Some mernbers obviausly have tricd to have a free rick on the otherskooperation. As pointed out by Taylor (l987:82), the aim is not to be part of the group that cooperates. The orEc members can be assumed to know that their &ationship will prevail in the foreseeable future,l%nd that the basis for the colective action, the quotas, can be changed in a few montfxsf time. 'The v o t a s are in principle open for renegotiafio~nat every a p E c meting. In, such iterated

262

OI~EC:A

Szr ccczssfirl Cartel?

games, as we have seen, the strategy of noncooperation may no longer be a dominant strategy. Can a situation where m members coopeate, while N-mdoes not, f o m an equilibriun-t in such a game? The answer is given by the folk theorem: Xt may for a s~~fficiently high discount factor. A "trigger strategy" can support such a pattern fnr a sufficiently high discount factor. In a triggm of this kind, each of the r f i players cooperate in the p=sent round as long as none of the rrt players defected in any previous mund.l" The kigher the discount factor-that is, tl-te m m value the pbyers attach to future gains-the hwcr the slrbset of cooperators (m) is needed in this equilibrium. For a given discaunt factor, cooperation can be achiwed by increasing the nulnber cJf players choosing cooperation, given others' cooperation. X 1 we assume that ten out of the Ibirteen OPEC members is sufficient, we will have 286 possible cooperative coalitions.l" I'he vestion of who s h o d constitute the cooperathg suhset remains open. Based on the discussion h this subsection, two hypotheses about which producers wodd be likely to cooperate can be put forth: the patient (with the highest discomt factor), and the largest (with the highest market share)(Olson 1965).1Wost of the discussion will be devoted to the latter, The Role of Patience

A patient producer is one with a low discount rate. Given the discussion above, the lower the discount rate, the moro kvil1ing the player should be to cooperate by cutting his own production in order to sustain or inncrcase prices. This vestion then arises: Which producer is likely to be patient? C)ne pssible criteria is the relationship b e m e n oil reserves and curretnt production.16 A pmducer with large reserves c o r n p a d to prctsent production knows that he will be in the market for a long time. n e total income from this producer's oil =serve will thus be stretched out into the iuture. The producer will have a longer time horizan than a producer with smdl resemes compared to current production. a short horizon on the extraction of one's oil reserves, the producer will put. 1iti.e value to the pl-ice of oil in the futurcl, since one will not be around as an oil producer at that time. Figure 7 5 shows the relationship between the reserveslproduction-ratios of the indkidual OPEC countries and their overproduction. The f i g w shows no clear pattern to the effect that higher R/P-ratios go together with lower overproductim. At least this way of opemtionalizing patience, or discount rate, does not provide an explanation as to which cowtries would be more prone to cooperate thm others. Let me therefore turn to the other factor: size.

OI~EC:A

Szrccczssfirl Cartel?

C)tson (1965) emphasizes the size of the group as one, if not the most, irnportant aspect determiaing the possibjlity of collective action: Even in the smallest groups, however, the collective good will not ardinarily be prcjvided on an optimal scale-.. . . This tendency tow-ards suboptirnality i s due to the fact that a collective good is, by definition, such that other individuals in the group cannclt be kept from cmsuming it once any individual in the group has provided it for himself. Since an individual member thus gets only part of the benefit of any expenditure he makes to obtain more of the collective good, he will discclntinue his purchase of the collective good before the optimal amount far the group as a whole has been provided-.Tn addition, the amounts of the collective gm8 that a member of the group receives free from other members will further reduce his incentive to provide more of that good at his c>wnexpense, Accordingly, the Iargcr the group, Clle an olltz'mal amozint ofa collective good. QOlson farther it will fa11 sfgort ofproz~idl'r?g 19(;5:34-35)

In this respect, OPEC is a small organization, with ten to thirteen members during the period studied in this book. Howeva, the O13EC members do not have an. equal share of the collective good, understood as a high oil price, as they differ in size; that is, their produced quantiy of oil differs. This difference fn size of the actors has implications for their wil,lingness to col~trihuteto the collective good. A starting point: in the discussion in Chapter 6, on the role of Saudi .Ara:hia, was Otsods argument (1965:22) that the larger the size of an actor (i.e., the market share), the more likely this actor is to contribute to the collective goad (Le., a higher oil price). This is so because the larger t:he actor is, the larger this actor's rdative hart3 is of the collectke good if this is provided. Thus, exacrly the s m e argument formulated above by Cllson also goes for the size of the indjvidud actor. A smalf group and a large actor give the same result: a larger shart? for this hdividual of ehe total collective good provided. Figure 7.6 shows the djstribution of the C313Ec members" market share and over)?roduction, Saudi Arabia, the largest proefucer, is the country least engaged in overgroduclion. In fact, Saudj Arabia produced less than its allocated quota during the period includcd in the fjgure. hang the other countrit.~,the figures artl to some extent in accordance with the collective action theory, as the smaller countries, those that most easily get away with cheating, arc less loyal to the agreement than the middle-range countries, albeit with s o w impcrrtmt exceptions that will be discussed bela\;v:

OI~EC:A

Szr ccczssfirl Cartel?

FIGUM 7.7 N-person chicken

As pointed out above, by choosing to cooperate, an. actor's utility increases with the nurnher of other actors choosing to cooperate. This implies that the actor's prcrferclnce for cooperation will increase with its =lathe share of the total good. For the oil prodwers, size in this respect is their production, and the relative size is their market share, or, for the OPEC members, their share of total OPEC output. If the actor% prcrduction is mall, the preference w i l l be never to cooperate (as in the prisoner" dilemma). At a certain size, on the other hand, the individual utitiv of realization crf the collectke good wilt increase, and the actor may regard cooperation as profitable if a nuntber of other actors cooperate. Thus, the larger actor does not have prisoner" dilemma preferences, but preferences represented in the "chicken game" (see Figure 7.7). hs pointed out: by Hovi (1992:1,01), if the large actor is sufficicntliy large, cooperatim might even be the dominant strategy; that is, the hegemon will cooperate even if nobody else does (see Chapter 6). The equilibrium point in such a game is that the hegemon provides the coilective good, white everybody else defects. In the chicken game it is beneficial for the actor to contribute only if fewer than k actors contribute (see Figure 73).

c~r~r:c: A SucmsJ111Cartel?

267

TABLE 7.9 OPEC Members' Overproduction Currelated with Other Members" O~rerprc>duction, March 1982-July 1990 Cnlegory Cuu~jlry Percent Share Currelnt io rz High Saudi Arabia 25.50 335 Middle Iran 42.44 s.387 9.6'7 .385 Iraq Venezuela 9-53 -,a15 UAE 7.58 .663 7.55 .435 Nigeria Indonesia 7.01 ,244 6.86 .701 Kuwait Libya 5.74 -.l10 Atgeria 4.12 .572 1.77 ,298 Low Qa tar Ecuador 1.34 .005 Gabon 0.92 .444 Notes: Production is measurd as a percatage share of total OPEC production.

The beta-coefficients referred to as "Correlation" hTable 7.9 are calculated as the werpmduction of the respective countries in ~ l a t i o nto the overproductim of all other OPEC members. .hhigh positive value is interpreted, as follows: The morc other mermbers overprodwe, the m r e this country owerproduces. This wlrutd give a crude indication cJf the countries-emde~~ciesto pursue a prisoner" ddemma strategy or a chickelz strategy A corntry foliowhg a prisoner" dilemma strategy will incrcase production no matter what others do; a country with chicken prt-ferences will decrease (increase) its own productio~~ as others increase (decrease) their production, As Tabk 7.9 indicates, only Iran and Libya, and to a minor extent, Venezuela, seem to have followed a chicken strategy. Table 7,10 c d i n e s the notim of the three phases of the quota cooperation htroduced in -Table 7.8 with the three categories of producer size introduced in Table 7.9. ?he tattle shows the relationship between others' and own overproduction among the three categories during the three phases, The coefficients in Table 7,10 clearly demonstrate the change in the Saudi Arabian strategy; from reducing production as others increase theirs to strungly increasing production as o t h r increase theirs. Note that the high-pmduc.tim category consists of only Saudi Arabia. The middleproduction category consists oE colmtries with different s c o ~ smaking , the average coefficient almost zero. Some of these countries are discussed indiwidudy below, as they rczpmsc-mt idi0s)incrasies. ?he low-production category performed as solid prisoner's dilclnrna actors in the two first

268

OI~EC:A

Szr ccczssfirl Cartel?

TABLE 7.3 O Refationship BeWeen Others' Overprc3duction and Own Qverprc~dudim

nnze firiods Apr. 82-Dec. 85 Nov. 86-July 90 Apr. %?--Dec.95

tow

2158 .3072 .Q527

Middle -.0565 .Q204 -,Q474

f-f@f;~ -.2686 .7185

,672

Al f -.l538 ,0867 -.0209

phases, whik the third phase needs further explanation, The number indicated migl~tsuggest that the small producers behaved m m cooperatively in the third pcriod. The explanation is Chat Ecuador left: omc effectively fPom January 1993, and Gabon left the organization effectively from Jmuary 3945. Accordingly, they are not included in the entire third period. Each left the organizatiol~stating they wodd pursue an independent oil-production policy, Their free-rider strategies have thus been even stronger than the others, without this being expressed in Table 7.10,

Apart from the strategic questions discussed, one can propose the hypothesis that the individual actor's quota compliaxlce is determined mort. by individual charactoristks than by considerations about the implications of other members%ehavior on one's own. hterests or of one's own behavior on the other members%intert.sts. Ti> some extent the factors proposed by member countries in the discussion about quota altocation can be used to explaifi why countries comply or not. Figure 7.6 above demonstrated the relationship between an individual member 'S production and quota compliance..The findings were mostfy in tine with the arg that larger producer are more prone to cooperate than small producers. The most im.portant exception is the United Arab Emirates. Here, there is an idiosyncratic explanation. The UAE consist of several kingdoms joincd together in a co~~federa.t.ion. The oil policy is not suhject to interference from the federal governmental level. There is therefore cotlective b q a i n i n g inside the UAE to allocate the total quota among the kingdoms. T'he major Ikingdom, Abu Dhabi, joi,ned OPEC in 1967. After the formation of the UAE in 1971, the Abu Dh;abi menrbership was transferred to the United Arab Emirates. At this time, the other kingdms were. rather s~nailoil pmducers, In the folfocving years their oil prodLtction increased. The other kingdms did not, however, feel the same obligations toward OPEC as Abu Dhabi. Bbu Uhahi is thus subject to presswe from the other h g d o m s to make room for their hcreased oil produdion under the UAE

c~r~r:c: A SucmsJ111Cartel?

269

quota, As a result, the Unites Arab Emirates face a collective actio~nproblem intcmdly before the country enters the o13Ec negotiations. This actually strenglthens the ~ l e v a n c of e the collective action problem, as the UAE might more accz~rately;in this perspectke, be hcluded 51Figure 7,6 as five small producers with a smaller share of production, but with the same percentage share of overprc,duc"cicm+ Another outlier is Iraq, The Iraqi overproduction is intilrrately connected with the war between Iran and Iraq, When the quotas were established in March 3982, Iran and lraq were allocated the same quota. A year later the Iranjan quota was douhled frnm 1.2 to 2.4 mbd. The mnin reason was that Iran, despite the war with fray, had been able to increase its production. This was not the case for Iraq, and the Iraqi quota was therefore not increased. However, the Iraqi government made it clear trhat it would accept only a quota that was on a par with that of Iran, regardless of what the actual Iraqi production was. Iraq was first able to increase oil output in 1985. Most of the Iraqi overproduction happened betcveen this increase in capacity and the omc decision to give Iraq parity with Iran from January 19139. Iraq chose not to comply with a quota the country had not accepted b m the outset. Two other countries can be mentioned here, namely, Kuwait and knezuela. As discussed in section 2.6, oil producers with large downstreafn assets stand to gahn evezz if crude oil prices are low, since their income is also derhed from refhnerics and product sales. Kuwait and Venezuela are tbe oil producers with the largest shares of downstrem activities in relation to their crude production. The Kuwaiti revenue is equaUy divided between crude sales and income from downstrcm activities. Gcmtrary to the other OPEC members, these producers can increase their d o w n s t ~ a mincol~ewhan crude oil prices faf, and thereby m k e up 'for the lost income .in the crude oil sector, substmtial vertical irrtegration, they also move their own oil downstrtlam and thus operate like multinatio~nalcompmies with bternal pricing. 'This price-sei-thg might exclude some aspects of costs in the different stages of the production chin, in order to make up for f l u c b ~ o n im s the cmde oil price. Thus, their cost-bmefit calcdatioxls are d i f f e ~ n from t those of t-he other producers, and they also have some heedom of action that the other producers lack, These idiosyncrasies suggest that other considerations might influace the countriesf hterests co~mectedwith cooperathg with other producers. 7.5 The Profits of Cooperation

This section will describe the development of revenues to the o13~c corntries durinl; the pericrd discussed in this book. How to attribute the profit to the OPEC countrieskooperative behavior is discussed h section 7 ~ 3 . 1 ~

270

OI~EC:A

Szr ccczssfirl Cartel?

Figure 7.8 indicates the movement of price and production m d the derived revenue. The vertical axis is the price, the horizontal is the owc production, while the iso-rwenue curves are drawn to in&ate dif erent =venue levels. Income increases diagonally .from the origh of coordinates. The line drawn inside the figure is a combination of o13~c production and price, and thus indicates the revenue level." m e picture prese~ntedin Figwrct 7 3 clearly shows how the rewnue Of the OrEc cou~ntries increased from 1973 to 1.979 (wifi the exception of 1975), without any reduction in the members+roduction. Oil demand was virtuarfy price-inelastic in the short-term (a few years). 'This had less to do with cartel behavior than with the increased oil dependency of the consuming countries, and a psychological notion of scarcity in the industrialked world. As described isl section 5.1, mast OPEC countries became exhilarated by their success and pushed for further price increases. Furthermore, as is shown in Figure 7.8, after the second d-price shock in 1979-1980, both OPEC production and price fell until 1985 (see section 5.2). In 1986, the price fell dramatically and o i ~ changed c its market strategy. Durir~gthe decacie from 1986 to 1996, the oit price fluctuated mademteb, but bllowed a downward tretnd. OPEC production increased steadily (sec section 5.3). This combhation left OI~ECrevenue fairly stable during this tm-year period, with an exception of 19911, when the oil price rose in the autulnn as a result of the Iraq-Kuwait War. Anuther indication of 013~~'s gairrs is the value of OPEC memberski1 exports. Figure 7.9 iliustrates hoMi this income has varied substantially, hcreashg until 19880 and then declking wit-;h1986 as a low point. The argument in this chaptcr and in Chapter 4, that the organization actuaIly begm working as a cartel in 1982 with the introduction of the quota system, means that the cartel has performed p o r l y in increasing or even in sustaining the revenue or value of exports for its members, Given an understartding of C)I'EC as a cartel with a competitive fringe (the noln-aPEC producers b&avi,ng as price-takers), the key variables determining the future oil price are the total demmd, the non-mac producerskapacity (as they are assumed to produce at h11 capacity at all times), t heffects of md, on the basis of these fadors, the ""calf on a ~ ~ @ . ' T ithe the franim revolution and the Iran-Iraq War on ail, prices, forecasts followed about the future development of the oil price that were to be proven totally wrong..A s was discussed in Chapter 2, the level of consumption had acted d y for a short period to the 1973-1974 price increase. In 1981, conswption was stitl perceiued to be price-inelastic. 'The non-OPECproducers were perceived to have limited reserves and thus as representing no long-term thmat to the position of ox~c, Increased demand and limited supply lead to assumptions of cmtinual sharp price hcreases (see Figure 7 .IQ).

c~r~r:c: A SucmsJ111Cartel?

273

After 1981, the tables turned, Although the forecasts stjll predicted higher prices in the longer tarn., the short-tarn. situatiun was different. The stock downeirakv and surplus production over a demand that was flattellkg out made the short-term outlook bleak for the OPEC com~tries. Dtlring the first half of the eighties, Saudi Arabia m d some of the other OPEC mernber~cut back their production and sustained prices well above the competitive level, This strategy, mmo in line with that of a cartel, was based m the perception that the weak prices were a short-tern prohiem, caused by the intemationai companies' reduction of stocks. %is strategy was understandable, as the message of the I981 forecasts by almost everybody in the industry was that a ~ chaving , increased the price of oil from $10 to $50 per barrel, could look forward to $70 per barrel in 1990 and $90 to $100 per barrel i,n the yeas 2CtOCI (all figures in 1990 $). There was no reason to believe that the fa11 in spot prices would continueit was just a question of riding out the storm (see Figure 7.10). Short-term production cutbxks would ease the pressttre on prices. However, as discussed in sectiosl2.5, oil demand did not pi& up, and non-sr~~c production increased further, beymd at1 forecasts. The price was adiditionally pressured, and so were the m e ~ ~ bcountries+evenues* er 7.6 A Cartel or a Success? "OIJEC is not the cartel that some of our friends outside like to portray us as; if we acted as a true cartel, prices woddnft be where they are right no\/v.!"Ig"Cartel is by far the most widelty used term to descrjbe OPEC. . . . However, there is general agreement that the textbook definition of a cartel does not apply to pre-1982 WEC, and specialists have spent much time identifying the i n t e n d features that make WEC successftzl despite this fact" (Cr4mer and Isfhani 1991:31).These quotes highlight the analytical problem of an evalrtatilie chapter like this one: m a t is the relationship between the cooperative or cartel-like behavior of o r E c and the oil price and thus OIW mmemberf"revenue? In the introduction to this cl-rapter two questions were. put forth cmceming the cxtent of OPEC countries' cooperative behavior and the possible gai.ns from such behavior. The previous section suggests that a nontrivial amount: of cooperative behavior has taken place. h m m precise measurement is unattainable because we can not h o w with certainty how these actors would. have behaved if they had not cooperated within the framework of OI~EC.It is e q u d y h d to deternine hcrw much of the revenues that can be attributed to cooperative behavior among the oil producers compared to other factors that increased the producers' profits, As pointed out i17 the introduction to this chapter, it is difficult to h o w what the oil revenue of the producers would have been had %herebeen no coop-

FIGURE 7.40 Crude oil prices, actual and successive International Energy Workshop polls SOURCE: Adelman 1995:18&.

c~r~r:c: A SucmsJ111Cartel?

275

eration between them. In the sprhg of 1986, whesz OPEC abandoned the strategy of defading the price level acrd began pursuing a strategy to protect its market share, the price fell to about $10/barrcl (see section 5.3). The sane levd was reahed during the price fall of 3998-1999 (SW scctim 5.3). Since the price only stayed at that fgvel for short periods of time, the actuai effect on investments and =placement of =serves was not empiricaily tested. A d h a n (1986) has estimated the competitive price of oilthat is, the price without producer cooperation in some fom-in 1986 to be $8 in the short run and $5 in the htnger run. He claims that these prices would mahtah investments hnew reserves m d capacity: At the high-cost end, it would take a price as Law as $4 to produce an immediate shutdc~wnof nearly half of capacity in the United States, and as 1c)w-as $2 tu d o the same in the North Sea. . . . At the low-cost end (assuming campetition and completely independent decisionmaking) a price of $5 would make it profitable b r the orst nations to expand their output to ower 60 million barrels daily. That price would be sustainable past 1995 at the least. (Adelman 1986:10)

On the other end, one can hardly claim that o13Ec has behaved as a Ifullfledged mmopolist, and thus gained the profit of a monogolist. 'The reason why this morropolist profit has not been reached by a cartel like OPEC is explained by Willett(3979a:55): As the members of the cartel raise prices towards the joint prc~fit-maxirnizatian level, the apected increase in profits, if the cartel held tugether, would have tcr be balanced against the increased risk that the cartel tzrouId split apart, reducing prc3fits for all the members. Because of this trade-off between maximum passible profits and the increased likelihood of a breakdown in the restrictive arrangements, the optimal price for a cartel would usually be Iess than the optimal full monopoly price.

tack of OPEC cohesion and miseacling of the "maximum posible profits'' fmccrd prices down. In the seventies they increased the price too much, and in the eighties they cut production too little, They were not at all alme in this misrcrading (see F i p r e 7,10). 'The coaclrnsion regarding the re1aionsSnjp between cooperation and =venue is thus a mixed one. The stated ambitions of C313EC have not been met, but the revenues obtaiz~edhave exceeded what would have been obtained in a coolpetitive h t e r n a t - i d oiX masket. The descrjptictn of the development of revenue and vahe of exports clearly shows that the OPEC members gaiz~edfrom tbe developments of the seventies. But this was not a result of:cooperative behavior: In the latter half of the seventies, djs-

271;

OI~EC:A

Szr ccczssfirl Cartel?

counting undermhed the of,llcial prices. hdividual OPEC me~nbercou~ztries secr~tlywdercut their own official price, This can hardly be regarded as disciphed cartel bcrhavior, but rather as the use of a competitive tool agakst other melnbers of the orgmiaation. The price increase from P978 to P982 was not caused by OPEC but was an effect of revolution and warfare (see Chapter 3). The picture becomes even more complicated when the argurnent is taken one step further: flow has this market contp.ol come about? Has it efforts, or has it been the rebeen provided by ddiberative c-crlrative sult of u~zcoordinatedbut similar and simultaneous behavior of individual rational actors based on favarable market conditions? As this study has shown, OPEC is not a single headed entity. "Different members of the cartel might prefer different prices because they differed either in theis degrees of risk aversion or their expecltatlions about the cohesiveness of the cartel" "illett 1979a:55). A solid test of the imgortmcc of cooperative behavhr is oniy possjbe when the indjvidual and the collective rationally diverge. This was not the case during the seventies. C)I-redoes not need to cooperate with others in order to raise prices when the produces at the same time can sustain the production level. This became apparent during the Iranian mvolution in 3978-1979 and with the o u t h a k of the Iran-Iraq War of 1980-1"31,, In these situations the restrictions agreed upon at the OPEC Conference werc abandoned in favor of a free-for-all price increase. The fact that the price codd not be raised witbout negative cmsequences for the market outlet for OPEC oil b e c m e evident in the beginning of rhe eighties, When the market-sharing mechanism-the yuotas-was introduced, the costs of cooperation followed suit as the free-rider incentive appeared. So, we are back at the argument in the introduction to this chapter: Cartels are l a s t effective when they are most needed. The internal cohesion is the key issue. U n d a competition, m a r k e t - s h a r i is automatic. Each operator produces all it can, up to the paint where the cost of additional output would exceed the price. But a cartet exists far only one reason: t o keep the price above the marginal cost. Thereby, the cartel shuts off the automatic market-sharing mechanism. A joint decision by the members must replace it. But marketsharing is a zero-sum game: any member's gain is another member's loss. (Adelman 1995:t;)

This study has, in line with Wilfett (1979a), elnphasized Ihe differences among the OPEC members, and in particular the special role of Saudl Arabia. The costs of cooperation were carried by Saudi Arabia from 1982 until 1,985. M e n Saudi Arabia fomd this intolerable, Che cooperation col-

c~r~r:c: A SucmsJ111Cartel?

277

lapsed. The situation in the first nine rnollths of 1986 is the closest the internationd oil market has ever been to a competitive market, OPEC" call. fnr help f r m the non-OPECpmducers illustrates how the cartef had lost its previous potential. Political circun?stances "helped" the cartel from August 1990-frst, as the Iraq-Kuwait W-ar stopped output f r m these countries and, later, as the U N sanctions prohibited Iraqi exports, leaving room for Saudi Arabia to sustah a hjgher level of production without causixzg a deterioration of the oil price. The connection between the market situation and the problem for the internal OPEC cooperation was the starting point of this chapter wf7en it was stated that cartels are easiest to establish when they are least needed. OPEC ckarXy illustrates this point, Durillg the seventies, OPEC a~ an organization played a minor role; the behavior of the OPEC members did, hawever, cause s~xbstmtialprice increases and, in fact, changed the structure of and mariccrt-power relations in the international oil market. During the first half of the eighties, OI~F-C, or, more accurately, Saudi Arabia, paid Che price for having overcxploited C changed (see section 2.5), the market, as demand and n m - ~ E supply with the effect of pressing prices down and leaving less room for OPEC production. Now, the wanizatianal facbrs m d e a difference, and the organization actually reduced the freedom of action of the members (see Chapter 4). Since 1482, OPEC has cut production significanttg and, in s o m periods, substant.iaily This, m o n g other factors, has kept the price above its competitiw level. As shown in section 7.5, this gave the OPEC countries increased revenue. The internai distrilslution of gains and costs of t-his cooperative policy has varied over time and a m n g members, Most hportant in the discussion in this chapter is the tcndency to cheat on agreed production levels, that is, the qrxota compliance. Apast: from the important exceptions melntioned in section 7.4, the o13Ec prducers fit the proposition that smaller countries tend to cheat m m than larger countries. 'This leads back to the role of Saudi Arabia discussed in Chapter 6. Saudi Arabia" role inside o13Ec has some similarities with the economic market model called ""dminant-firm.'" This model describes a market with one large producer and several small firms. The core assmgtion is that '"the dominant firm, sets the price and.allows the minor firms to sell alf they can at that price; the dorninmt firm sells the rest" "ohen and Cyert 1475:245). This makes the other producers b&ve competitively,and if the dominant fim's market share becams very small, the market "reverts to the coqetitive solution" "(ererner and Isfahani 1991:31). Saudi Arabja performed a role as dominmt firm, from 1982 to 1985*20This made the other producers insellsitive to the changes in d e m n d and non-orsc production (see section 2.5). They had experience with demand regulations. W e n they had experienced weak delnand in the mid--f 970s, they had adjusted prices (see

278

OI~EC:A

Szr ccczssfirl Cartel?

section 5.1)" Whe11 prices had to be defended with reduced productim in the early 1980s, Saudj Arabia took the burden. In a way the problem since then has been to make every O13EC member behawe as dominant firms-that is, to constantly consider the right output level in order to support agreed prices target, This price target should be a best estim.ate of what could be called "sustainahie mmopoly price.'" That is the highest price attaindle that does not stimulate alternative s~~pplies, co~~version into other m e s y forms, or reduce stable demmd tends, I f producers do this in cclncert, we are talking about a cartel, or "cartel-like behavior,"' nfter these four chapters on kternal OPEC affairs, the m i n ream11 for lack of success (i.e., prices below the sustajnahle monopoly price) is the individual members' compliance with price and production agreements. rZIready in the early 1980s the idea of compensating this lack of internal cohesion by increasing the scope of oil-producer cooperation emerged. I'he next two chapters adctress this issue..

Notes l . The phenomenon called coltlective p o d is discussed in sections 7.2 and 7.4. A high oil price is regarded as a collective good for the oil producers throughout this chapter. 2. This to some extent assumes stable demand level. The demand also will have to be somewhat inelastic. With an elastic demand, small price changes would necessitate large changes in a prc>ducticmlevel, which would be costly for the cartel members, 3. Non-OPEC producers here are assurn& to behave as price-takers maximizing profit witkhout considering that their production level could influence the world market price. 4. Prjssibly except for the period during and immediately after the Gulf VVar between Iraq and Kuwait in 1990-1 9541. 5. Tn addition there are two instrumental aspeds: allocation of output, and detection and punishment of cheaters (see section 7.2). 6. For some countries there are positive correlations during some of the individual peric~ds. '7. A s was discussed in further detail in Chapter 3, the Iranian prc>duction is susceptible to factors other than the oil policy and ecmchomic interests of the Iranian regime. Warfare and destruction of oil installatiom have had a detrimental impact on Iran's ability to produce oil. 8. ""Te "hawkish' Iraqis, who had demanded a 30 percent price increase in 1975, actually cut prices in 1976 to gain m a r k t share" "(AdeXman 19515:158). 9, ""There are two main aspects of crude o;il quality which inauence the price, the distillate content and the sulphur content . . . light crude oils do command I-tigher prices than the heavier crudeii and this may be expressed as a gravity differential" "Uenkr ns 198(;:335).

c~r~r:c: A SucmsJ111Cartel?

279

10.1 am indebted to teif Helland for his comments and suggestions, which strcmgiiy impmved the game theoretical parts of this sectic~n. 11 Letf(nl be the payoff for any player when he chooses cooperation (C) and n others choose C, and g(n) the payoff when he chooses to defect and E others choose C, Tl-ren an n-person prisoner's dilemma is defined by: Qi) gttd >f(nl far each value of n 2 0, and (ii)ffN-1)> g(0). (Taylor 19828384). 12, Unless, of course, a member decides to sustain the membership. 13. The discount factor has to satisfy the Ec)llow.ing condition i f defection should not pay o f f for a member l' in m: -.

Far =ample, if we assume that Defect yields 2 utilities more than Cooperatian, that isf gg,fni = n and f,lnl = E-2, and inxrt the number af OPEC members, we get the fallowing level of discount factor sustaining cooperation:

14. This number is given by the binomial coeffiFfident:

15. Since the OPEC members differ in size, defined by their market share, the actual number is less relevant than the combined market share of the cooperating members. 16. There are of course several other filctcjrs influencing a particular prt~ducer's patience, Some idiosyncratic factors are discussed below. 17. It should also be noted that I do not address the impc~rtantquestion af what the O13EC3members have done with the profit. This is a topic recently covered by others (Arnuzegar 1999 and KarlIW7). 18. Figure 7.8 includes no estimation of costs, Cost data are inherently uncertain or mostly unavailable. There is no reason to assume that inclusion of cost data would have dlangecl the overall picture presmted in the figure, although the shape of the line drawn inside the figure might have been tilted somewhat if costs had chnged over time. 19. OPEC's Secretary-General, Rilwanu Lukman, interviewed in Peltmleunrl Ecarzomisf.March 1999:8. 20. Willcstt (1979a:56) predicted this in 19'7% *%thereappears to be a reasonable expectation that aver the coming decade the Saudis, perhaps in collaboration with other members of the mganization of Arab Petroleum Expoding Countries (BAPEC) would have to accept: a disprcyportionate share of the grc>upfsrequired cul'back in pr~duction.~'

This page intentionally left blank

Extending the Cooperation: OPEC and the Non-OPECProducers

The previous four chaptcrs have dealt with the oil-producer cooperation C. It is time to turn the that has taken place inside the organization focus beyond the bordcrs of this organization. In 1,998, 58 percernt of world oil production took place outside OPEC. As described in section 2.5, OI~EClost c o n t d wer prices in the 1980s not only because of reduced oil demand, but afso due to increased oil supplirss from outside the organization, lost another 7-8 mbd of marketabte oiX in Eavor of these new- and numerous producers, who have benefited from the Organization" non-economic pricing system which enabled them to invest directly or through the oil companies in new oilfields. . . . OPEC helped make it easy for non-OPEC countries to invest profitably in oil. (AI-Chalabi 1989:42)

OPEC has

The loss of market share to noln-oPEc producers became a presskg prsblem for o13Ec in the beginning of the 1.980s. This chapter will descri:be s o w of the problems of e*nding the oil-producer caopcrlration beyond the organizationd bordms of ortic. .Although many structrural constraints and national interests are the same among producers inside and outside OPEC, the m-WEC producers as a group are even more heterogeneous than the OPEC members. fikcwjse, their policiss toward oil-producer cooperation vary wildly. A fuller description of a particular bargaining relationship betwen an OPEC and a non-or9~cproducer is prtrsented in section 9.2. Chapter 9 also provides a more thorough explanation ol a n o n - m ~ cproducer's (Norways) behavior. Chapter 9 can thus be seen as a case study of the more general issues discussed hboth this and the previous chapter.

Exfendirzg fhe Cc~opemf ion: C?PL"Cand NOB-OPEC

282

TABLE 8.1. Categories of Nctn-OPEC Prc>ducers Cntegtjy I CnEegoq 11 Prcjduction (mbd) Above 2 0.3 to 1

Share of non.-OPEC: Reserves Prcjduction Countries

Category 111

0.1 to 0.3

About 70% About 60%

Nearly 125% Above 20%

About 5% Above 5%

Canada, UK, Norway, US, Mexicc~,China, Russia

Brazil, Oman, Egypt, Argentina, Angola, Malaysia, India, Syria, Colombia, Australia, Kazakhstan, Yemen, Gabon, Ecuador

Congo, Denmark, Vietnam, Peru, Italy Azerbaijan, Brunei, filornania, Trinidad and Tobago, Turkmenistan, Carneroon

WL'ICCE: EIA (1999b).

8.1 The Non-OPECProducers in General

The non-OI~EC producers can be categorized according to their production. 'The U.S. Energy Information Administration (EM) (1999b) applies three categories that cover all non-negligible producers (see Table 8.1). In section 35 the pofitical heterogeneiy of the OI~ECcountries was underlined. A glance at Table 8.1 proves the heterogeneity of the non-OPEC countries along h o s t every indicator from political systent to economic growth, 'The previous chapters have discussed the problem of collective action among the now eleven ~ I ? E Cmembers, Extendir~gthe collective action to include all the thirty-two countries i,n Table 8.1 scelns impossible. The role of these countries in oil-producer cooperation has thus been more as bilateral relations with OI~ECor QI~EC-countries than as a collective entity However, there have been exceptions, and kformally some of these countries have bilateral. relations m m g themselves concerning oil issues. &ring the nfneties oil ministers from Iblexico and Norway bave had several bilateral nrreetbgs*It should also be noted that the non-OPEC group includes two countries that until 1992 (Ecuador)and 1995 (Gabon) were members of OPEC. In the rest of this subsection I will comment briefly on s o m characterjstics clf some of the non-oFEc producers in category E of Table 8.1. New oil reserves are conthuous:ly discowered within and outside of OPEC. However, OPEC still controls over se~rentypercent of the worlds

proven oil reserves. Concernhg produdion, the noll-opE@countries have dwreased their share from 66.1 percent in 1988 to 17.9 percent in 1998. I'hese figures conceal an almost halving of the market share of the former Soviet Union. As Figure 8.1 indicates, the other non-OPECcountries have sustajned their m a r k t sharc or, in fact, i n c ~ a s e dit by 1.5 percent, while the increased market share of C)I?EC has been at: the expense of the former Soviet Union. The palitiical and ecmomic changes in the Soviet Union sixrce 1985 arc of course beyonli the scope of this study. 'The fa11 in oil production was, however, accompanied by m even faster fall in oil consumption. During the 1 9 9 0 ~the ~ (now Russian) export outside the former Soviet Union countries has thus tumed amund, and increased steadily since 1992 (see Figwe 8.2). A paradox then emerged as Russia" rrole in the international market was strengthened at the same time that its production fell dramatically. Lack of inwestments during the decade also have made it unlikely that Russia wiLS be able to increase the oil production r a p a y . In Ihe longer term., Russia is among the most promising n o n - ~ ~areas ' ~ c for making larger new discoveries. "'The farmer Soviet republics may still offer the best possibilities of dewloping giant and even super-gimt fields olx tside the Middle East" "artshorn 1993:274)* Another large oil producer-in fact, the world's second largest: in f 998-the United States has experienced a tufbulence wfic.11 it comes to the relationship between own production and.consumption, Figure 8.3 shows the us conswption, production, and import ~ q u i r e dfrom 1965 to 1998. The us oil is thus not a part of the day-to-day international oil trade. Changes in us consumption or production, on the other hand, do have substantial impact on the world's overall sup*-and-demand balance. The development: of us oil consumption during the 19EUls and f 990s is also noteworthy. Despjte the economic growth and development of two decades, oil consumption as of 1998 was still below the peak of 1978. The import share has from time to time worried the us government. hcreasing imports in times of increasjng prices, as was the case of the early seventies, are worse than incrmsing imports in times of fillling prices, as was the case of the late nheties*111 November 1923, us President Rchard Nixon addressed the nation with the foliowing words: "Let us set as our national goal . . . that by the end of this decade we will have developed the potelztial to meet our own energy needs withou"f:ependbg on my ioreign energy sources. . . . VVe have an energy crisis, but the= is no crisis of the American spirit" (RulT-Berg 1987:3). tn 1998, when W cJit imports reached an unprrzcedented high of mare than 55 p e ~ e n t a' perception of an energy crisis in the us government was hard. to discover. The two largest rrcm-~ticproducers, b s s i a and the U ~ i t e dStates, are both large cmsumers;, The tlnited States is an increashgly laxge net im-

porter. Russia has cmtributed to the oil-producer cooperatio~~ by statkg its intention to cut production in 1986. In 1996, the then Soviet Union in.-creased its production to the QECD the following year (hl-Chalabi 1989:44). Mexico has been far mom willing and active in the dialogue with OPEC (see section 8.2). Having turned into a net fmporter of ail, a deep-drilling exploration program was law~ched,leading to Large new discoveries in 1972, and by 1374, Mexico was agaiR a net exportclr (Vergin 1991:666), thus coinciding with the price ir7crc.ase of tbe first oil price shock. Adelman (1995:15Q)argues that all the new Mexican fields of the early 1970s were prditable at pre-1974 prices. Unfortunately, the combination of discoveries and p'ice increases led Mexico to borrow a large amount of money abroad. Aclelman (2995:196) is rather harsh in his description of the Mexican oil hdustry and financial restraints: Tn Mexico, the national oil company PetroXeos Mexicanos (Fernex) was a gaudy exemplar of corruption and waste, The belief in ever-raising prices was a national disaster. With roughly 50 billion barrels in reserves and current price~"xme"ding$30 per barrel, it seemed safe to borrow $60 billion abroad. The resuit was a national financial crisis and masshe disinvestment in ail productic~n.. . . Moreover, many of the tzrelfrj were unecmamic, drilled to provide local jobs and contracts. . . . The policy war; sli>wlyreversed, especially after 1988, but revival and cleaning up the debris of long mismanagement is painful and slow.

Although the Mexican oil reserves have deched during the last decade, the production i n c ~ a s e dfrom 2.9 mbd in 1988 to 3.5 in 1998. nllt;hough the production Level of: Canada and the United Kingdom brings them into the first category in Table 8.1, their net export makes them less significant in the market of internationally traded oil. "T'hus they wjll not be commented on fuztrtha here, China is, in this regard, i,n the s m e situation as the U ~ ~ i t eSt.ates d as a major producer, but also a net imporkr. The Chinese case is exactly opposite the cltevelopmmt in Russia. China has become a net importer due to a s t r o ~ ~ hcrease g h oil cmsumptiun during the last decade (see Figure 8.4). Taking both market share and net export into consideration, this sketchy description of the largest non-OPECprodtlcers leaves us with three important substantial net-exporters: Russia, Mexico, and Norway Together with some of the millor non-OI"ECproducers, these three countries have been the target for QPEC'S Svivitatians to cooperatio~~. This will be klnsther discussed in the next section.

8.2 The a ~ ~ c - N o n - oBargaining ~~r:

In 1965, OPEC had a market share of about 45 percent. In 1998, the share was 42 perce~zt.In between these almost identical figures, OPEC'smarket share was as high as 53 percent (1973) and as low as 29 percent (1985). With a market share of 53 percent, produwrs outside the organizaticm were haray a concern for OPEC at att. With a market share of 29 percent, the non-c?13scproducers werc almost the only concern. official prices came under pressure in 1981 (see section 5.21, W ~ e O19EC n the non-orec producers caused probl.erns for OPEC w k n they led in open price reductions in, the years 1982 to 1984. In March 1982, the British oil compay RNOC reduced its reference price with $4 to $31, $5.5 b e l w sinnilar Nigerian crude quality Norway and the Soviet Union, followed the British move, .Although the price was subsequently increased, the same procedure of reduction happmed again in February 1983. In Octtr ber 1984, Norway took the lead by abandoning the reference price and introducQ market-mlated prices. "Spokesmen for Statoil said, however, that tl-te company%intention was not to lead the market but rather to formalize an existing situation which had been disguised by the use of 'unreal" oificial pnces7CErsans 1991:639; see section 9.3). The Nor& Sea producersbmoves were particularly h a d to swatlow fnr the weak link in the orEc chah-Nigeria. .A poor nfrican country with high dependency on oil income-and. a similar crude qualiv as the Nosth Sea p d u c e r s Nigeria had a h a d time placing its c d e cJiX on the market, and had to shndocv the price redwtions of the North Sea prodwers. Thus' the market-driven price behavior of the non-oX'Ec producers had immediate repercussions into the coherence axnong the O13EC producers. As discussed in section 6.4, Saudi Arabia carried the bwrden clraring these years by cutting its production in an attempt to balance the market. When this was impossible to achieve, Saudi Arabia changed its strategy to &fend its market share. So did OPEC in Decentber 1985. Then the bargaining g a m t w a r d the non-OPECproducers also changed character. Now an actual price war commenced In a press conference held after the Uecember 19885 am@Conference, the Saudi Arabim oil minister, Ahmad Zaki producers ~~~ when answerYamani, made a clear warning to the n o n - c ? ~ ing a p s t i o n about the likeiihood of a price war: "It all depends on what the non-ctrec producers will dn."?The Norwegian response is discussed in further detail in section 3.2; here, :l shall discuss the general picture regarding the nonar.ec Esponse to the situatim that emerged in December 1985% Table 8.2 lists the most irngortant non-OPECproducers, their production and net exports, and their signaling of a coopem"tive or noncooperative attitude toward the OPEC invitations to cooperate in 1,985. If Norway

Exfendirzg fhe Cc~opemf ion: C?PL"Cand NOB-OPEC

290

TABLE 8.2 Characteristics of Selected Non-OPEC Producers, 1985-1986 United Kingdom Argentina Canada Trinidad Nctwa y USSR Mexico

2,655 490 4,810 175 815 2 2,035 2,910 890 2,505 500 445

1,025 nc 90 nc 320 ne 175 nc 615 nelc 3,675 C 1,6170 C Egypt 475 C China 695 C Oman 500 c Malaysia 2.50 C Total 9,490 Note: Prc;lducticdn is in th, b/d in 1985, and attitude is denoted by c = cooperative and /or ne = noncoc~perative wrrricx.::Production and Consumption figures taken fmm BP Statistical Revl'ei-uof World Er~etgy. 4,630 400 1,490 NA 2.00 8,3140 1,240 445 1,810 NA 195

is fncluded in the noncooperative group, OPEC ~ c e i v e dcooperative sign& :mm exporters with productio1n m d net exports of 7.2 &d, with the noncooperative group constituting 2.2 mbd. This suggests that when the nm-OT~EC producers were confronted with the collective action problem through the orEc offernsive in spring 1986, several of them did consider cooperation rather than contjnuation of a he-rider strategy, Among the OPEC countries, the larger producers tended to cooperate more than the minor members. This to some externt confirms Olson's general hypothesis that the larger an actor is, the more likely it is that it will cooperate (see section 7.4). This trend is not found among the nana1zc producers. In 1%8, among both smaller and larger non-OPECproducers, there were countsies of strategy inctjned to cooperate and corntries that resisted the invitation from OPEC. The categarizatians h Tables 8.2 a d 8.3 are based on politically stated intentions regardilng cooperation with OPEC.The next question is: How did the non-WECproducers actuaXly behave? proTable 8.3 cclmpares the cooperative and noncooperative ~o~-OI'EC ducers' production grokvth rate from 1985 until 1987. Interestingly, as Ihe figure shows, the growth rate of the cooperative group incmased horn 0.13 percent in 19M-W85 to 2.10 percent in 1985-1986. The growth rate of the noncooperathe group 'ell fsom 2.67 to 0-84 percent in the same period. Since Norway subsequently made some concessions to OI~EC(see Chapkr g), the table also includes the noncooperathe group, excludfng Norway. The effect of this is that the noncooperativists actuallly reduced

TABLE 8.3 Stectect Non-OPEC Producers' Productian Growth, from 13revious Year (%) 2 984-1985 2 985-1986 1986-1987 Cooperative g o u p 0.13 2.10 2.72 Nc>ncoc>perative group including N o w a y 2.67 0.84 2.50 Noncooperative group excluding N o m a y 1.88 -0.78 0.00 Cooperative group: USSR, Mexico, E g y t , China, Oman, and Malaysia Noncooperative group: United Kingdom, Argmtina, Canada, and Trinidad wrrricx.::Based cm data fmrn BP Statistiml Review of World E~lergy,

their production in 1986 compared with 1985, Norway, aft- slating its cooperative strategy in 1986, increased its production, and the other noncooperativists kept a flat pmduction profile from 1986 to 1987, This i1,lustra.t.e~ the problem of a discrepancy betwee11 words and deeds in ixrtemational cooperation as such. Althuugh it is a matter of correlation and not causality it is a fact that those nona1zc producers that sipaled their cooperative attitude behaved noncooperatively;, and those that signaled a noncooperative attitude actually contributed to WEC% price-sta" bilizhg efforts. The most formally hstitutionalized attempt of colilective non-OPECaction was the response to WEC'S call for cooperation from seven n m - o r ~ c countries in 1988. The group of seven included: h g o l a , China, Colnmbia,, Egypt, Malaysia, Mexico, and Oman. At a meeting in London on March 9, six of the seven countsies (Caliomkaia opted out) offered to cut their export by 5 percent, given a proportional response from OPEC (Evans 199(3:721).This wodd have meant some 183,OO barrels per day for the seven n m - o r ~ ccomtries, and same 700,000 barrels per day for OPEC. The proposal led to a joint meeting in April 1988 oE reprc.smtati\les of six OPEC c~tlntries,the group of sevm, and an ohserver from Norway. o13EC turned this down, hut hdicated its willingness to match the nonOPEC proposal barrel for barrel-that is, OPEC, too, would make a reduction of about 200,00(1 barrets per dayWeverthei;ess, the meeting was a move in the direction of a broader producer cooperation in the market." The noMPEC producers continued meeting alone and with OrEc until the outbreak of the Gulf: War in August f 991). Then the prohlern was how to increase production fast enough (see section 3.4). As described in section 5.3, afer the GuIf War, OPEC did not apply a particdafly coherent market strategy Thus, the orgmization did not invite the non-opEc producers to cooperate further. Wth tray back in the market (see section 3A), and reduced demand due to economic recession in Asia (see section 5-31, the pressme on prices increased. orEc had to revitalize its producticm

Exfendirzg fhe Cc~opemf ion: C?PL"Cand NOB-OPEC

292

policy. 'This also brought to the forefront the issue of cooperation with producers outside the organization, h 15498, the ox~~c-non-m~c relations were reactkated. OPEC appmwhed several non-OPEC producers in order to stretngthem the coclperative efforts. This time, Mexico took the lead on the n o n - ~ Eside, C initiating diplomatic activit.ies toward other n m - o r ~ c producers ~~ producers like Norway. I'hus, in March 1999, fnur n o n - 0 ~ (Mexico, N'orwy, Rwsia, and Oman) were exflicitly included in the oIJEC scherne of pmductim cuts.Wexico was particularly instrumental by joining Saudi Arabia and Venezuela in prcrparing the plan for production cuts before the March 1999 OPEC Conference. Subsequently;the division of producers into srEc and non-or)~cproducers have less importance than before. The bargainhg, negotiatim, and policy preparation now takes place across the orgmizational boarder, This implies a higher efiectiveness in infiuencixzg the narke et. It might, however, threaten the importance and role of OPEC as an organization. In an interview with Petroleum Ecol-zc,misf,OPEC'ssecretary-general rcjects that the interaction with outside producers had adermined OPEC: On the contrary; Venezuela, Saudi Arabia and Mexico got together on an informal basis to discuss our common problem beween o ~ ~ r ;and C non-OPEC, After they reached some understanding it was up to C ~ P F Cto implement the decisions and they came to us. . . . Consultations are going on aXt the time, on a formal and informal basis, . . . Mistakenly people have talked abaut the fact that a new organizatim is going to emerge from this, . . . The impartant thing is that we are now getting our friends outside OPEC to cooperate by restraining themselves . . . some of them are cutting. . . . fNf3 hope they cantinue to coc~perate. . . the mare we coc~peratein normal times, the better we will be able to cooperate when times are hard.5

8.3

Explaining Non-OPEC Cooperative Behavior

It is of corarse impossible to go into a etiscussion of the motives and dispositims for cooperation or lack o.E such for all or even only the key nonorEc producers. An entire chapter is devoted to one non-BI'ECproducer-Norway (see Chaptcr 9). Here, I can only sketch some general pro7lc.m of cooperatiozn between an organisation like OPEC and individual countf-ieslike the non