UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Arbitration in WTO disputes : the forgotten alternative Jacyk, David William 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-0444.pdf [ 8.4MB ]
Metadata
JSON: 831-1.0077735.json
JSON-LD: 831-1.0077735-ld.json
RDF/XML (Pretty): 831-1.0077735-rdf.xml
RDF/JSON: 831-1.0077735-rdf.json
Turtle: 831-1.0077735-turtle.txt
N-Triples: 831-1.0077735-rdf-ntriples.txt
Original Record: 831-1.0077735-source.json
Full Text
831-1.0077735-fulltext.txt
Citation
831-1.0077735.ris

Full Text

A R B I T R A T I O N I N W T O D I S P U T E S : T H E F O R G O T T E N A L T E R N A T I V E by D A V I D W I L L I A M J A C Y K L L . B . , The University of Manitoba, 1993 B . A . , The University of Manitoba, 2003 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R OF L A W S in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Law) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A August 2007 © David Wi l l i am Jacyk 2007 A B S T R A C T The creation of a binding adjudication system under the Dispute Settlement Understanding ("DSU") is one of the major successes of the W T O . However, while the Dispute Settlement Body ("DSB") has experienced a high level of compliance with its rulings, there have been enough failures to raise concerns about compliance with W T O rulings. This in turn endangers the long term viability and legitimacy of the W T O as a decision-making body. This thesis explores the possibility o f more effective integration of arbitration as a means of dealing with a small number of problematic cases where compliance with a ruling is doubtful. It considers arbitration as an alternative to what has effectively become an institutionalized litigation system involving panels and the Appellate Body, and as an adjunct to the diplomatic resolution of disputes, particularly for policy driven cases where compliance with W T O rulings is more doubtful. While proposals for the use of arbitration made during the Uruguay Round of negotiations leading to the creation o f the W T O have been realized in the provisions o f the D S U , arbitration has never been effectively tested as a true alternative. Further, arbitration as an alternative to the litigation system has been almost entirely ignored in the context of the current debate over reform of the W T O dispute settlement system. After over a decade of W T O decision making, it is now an opportune point to consider meaningful institutional reform that more fully incorporates arbitration as an alternative form of dispute settlement at the W T O in politically difficult cases, and that builds on the existing but underused arbitration provision in Article 25 of the D S U . This thesis challenges the predominant bias towards the litigation system involving panels and the Appellate Body as a one-size-fits-all solution. It explores the potential role of arbitration, in the context of compliance theories, a historical review of the negotiations during the Uruguay Round, and an analysis of the shortcomings of the current D S U that contribute to the problems of non-compliance. i i TABLE OF CONTENTS A B S T R A C T i i T A B L E OF C O N T E N T S i i i A C K N O W L E D G E M E N T S v D E D I C A T I O N v i C H A P T E R I: THESIS O U T L I N E , O B J E C T I V E S A N D C O N C E P T S 1 1.1 Introduction 1 1.2 Methodology & Objectives 3 1.3 Reconsidering the Legalism of the D S U 6 1.4 Working Concepts 9 1.4.1 Compliance and Implementation of Policy Change 9 1.4.2 Distinguishing Arbitration From Judicial Settlement in the W T O 10 1.4.3 Features of Arbitration Generally and in the W T O Context 12 1.5 Finality of Decisions and Party Control as Distinguishing Features 15 1.6 Chapter Description 19 C H A P T E R II: C O M P L I A N C E T H E O R I E S : P O W E R , R U L E S A N D R E P U T A T I O N ..22 2.1 Introduction 22 2.2 International Relations Theory and the Elements of Power and Rules 24 2.2.1 Traditional Realism & Jackson's Power vs. Rules Paradigm 25 2.2.2 Keohane's Institutionalist Regime Theory 29 2.3 International Law Theory 33 2.3.1 Societal Theories of Franck and the Chayes 33 2.3.2 Koh ' s Transnational Legal Process 46 2.3.3 Contextualism - Jackson's Power vs. Rules Paradigm Revisited 49 2.4 Conclusion 54 C H A P T E R III: T H E D E V E L O P M E N T O F T H E D S B I N T H E U R U G U A Y R O U N D . . 5 7 3.1 Introduction 57 3.2 History of G A T T Prior to the Uruguay Round 58 3.3 The Beginnings of the Uruguay Round 66 3.4 The Initial Proposals and the Arbitration Alternative 68 3.5 The Impact of U . S . Unilateralism on Uruguay Round Negotiations 77 3.6 Proposals on Arbitration, Adoption and Appellate Review 80 3.7 Arbitration Proposals Stall 83 3.8 Arbitration Under the D S U 89 3.9 Lessons from The History of G A T T and the Uruguay Round 90 3.10 Conclusion 94 C H A P T E R IV: J U D I C I A L S E T T L E M E N T A N D A R B I T R A T I O N I N T H E C U R R E N T D S U 96 4.1 Introduction 96 4.2 The Judicial Settlement System 97 4.3 The Objectives of the D S U 100 4.4 The Limited Effectiveness of Remedies Under the D S U 105 4.5 Non-Compliance and Shortcomings of the Judicial Settlement System I l l 4.6 A n Over Enthusiasm for the Judicial Settlement System 113 4.7 Specific Arbitration Provisions in the D S U 116 4.7.1 Timeframe Arbitration 117 4.7.2 Concessions Arbitration 125 4.7.3 Issue Arbitration 130 4.8 Conclusion 135 C H A P T E R V : R E F O R M 139 5.1 Introduction 139 5.2 Surveying the Debate on D S U Reform 141 5.2.1 Fixation on Compensation and Remedies 142 5.2.2 Procedural Legalism 144 5.2.3 Direction of Sutherland Report 146 5.2.4 Recommendations for Reform of Consultations 149 5.3 The Potential Util i ty of Arbitration - A Framework for the Future 150 5.4 Integration of Arbitration Through Dispute Diversion 154 5.5 A Politicization of the D S U and Return To G A T T ? 156 5.6 The Feasibility of Diversion - De-Politicizing Judicial Settlement 158 5.7 Policy-Charged Cases in the Judicial Settlement System 161 5.8 Identifying the Risk Cases 164 5.9 Further Considerations and Refinements 168 5.10 Thesis Conclusion and Further Research ..171 B I B L I O G R A P H Y 176 iv A C K N O W L E D G E M E N T S I would like to take this opportunity to acknowledge those persons who have provided their assistance, enabling me to complete this thesis and the L L . M . program at the University of British Columbia. I would like to thank Dr. Janis Sarra for her guidance and patience as my thesis supervisor. I would also like to thank Professor Ljiljana Biukovic as my second reader, who provided her time and insight in her review of the draft, and who oversaw the research relating to Chapter 3 . I thank Valerie Hughes, of the Department of Finance, Canada, and Professor Michael Hahn, University of Waikato for their general comments and encouragement in pursuing the topic of this thesis. Finally, I would like to thank Professor Wesley Pue and Joanne Chung, Graduate Secretary for their patience and tolerance in assisting me in completing the L L . M . program on a part-time basis. v D E D I C A T I O N This thesis is dedicated solely to my lovely wife Nicola. Completion of this work would not have been possible without your continuous encouragement, patience and support. C H A P T E R I - T H E S I S O U T L I N E , O B J E C T I V E S A N D C O N C E P T S 1.1 I n t r o d u c t i o n The international trade system has been regulated for decades by two distinct forms of process - the legalistic and the political or diplomatic. The trade system established by the General Agreement on Trade and Tarriffs in 1947 ("the G A T T system"), which predated the World Trade Organization ("WTO") , was often criticized as being too political, and in its later years, largely ineffective for dispute resolution. One o f the touted accomplishments of the W T O was the creation of a binding adjudication system under the Dispute Settlement Understanding ( "DSU") 1 resulting in the legalization of the dispute settlement mechanism. This legalized system has been used frequently and with great success. B y the end of 2006, over 260 disputes had been subject to the W T O dispute settlement process, resulting in 108 circulated panel decisions 4 and 67 circulated Appellate Body decisions. 5 The Dispute Settlement Body ("DSB") has experienced a significant rate of success with respect to compliance with its rulings - experts generally suggest that 80 per cent of cases 1 The Understanding On Rules And Procedures Governing The Settlement Of Disputes, Annex 2 to Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations and Marrakech Agreement Establishing the World Trade Organization, April 15, 1994. Also known as the Dispute Settlement Understanding ("the DSU"). 2 Richard H. Steinberg, "Judicial Lawmaking at the WTO: Discursive, Constitutional and Political Constraints" (2004) 98 Am. J. Int'l L. 247 at 247. 3 Kara Leitner & Simon Lester, "WTO Dispute Settlement 1995-2006—A Statistical Analysis" (2007) 10 J. Int'l Econ. L. 165 at 172. 4 Ibid, at 172. 5 Ibid, at 173. 1 are implemented within a reasonable period of time 6, though it has been suggested that it could be as high as 90 per cent. The relatively high rate of compliance with D S B decisions is high water mark for the international trading system, and for international law generally. Although this new system has resulted in the automatic right to adjudication of disputes and the virtually automatic adoption of rulings by the D S B , it has still not resulted in automatic compliance in every case.9 While it is generally recognized that the compliance rate with D S B rulings has been high, there has been a steady problem with compliance in cases involving politically sensitive matters.1 0 There have been enough failures with respect to implementation of rulings, 1 1 particularly in a handful of high profile disputes, to warrant some concern about long term compliance as expressed in W T O rulings. In short, there is room for improvement, and there is good reason to consider it now. Because the D S B has been relatively effective to date, there have been increasingly high expectations of its capacity to resolve all trade disputes through the issuance of binding decisions. The balance between the legalism of the current system, and the flexibility and consensus building of the diplomatic system continues to be a source of discussion. 6 William J. Davey, "The WTO Dispute Settlement System: The First Ten Years" (2005) 8 J. Int'l Econ. L.17, at 47 [Davey, "First Ten Years"]; Debra P. Steger, "WTO Dispute Settlement: Systemic Issues" in John M . Curtis & Dan Curiak, eds., Trade Policy Research 2005 (Ottawa: Minister of Public Works and Government Services, 2006), 51 at 67 [Steger, "Systemic Issues"]. 7 Bruce Wilson, "Compliance by W T O Members with Adverse W T O Dispute Settlement Rulings: The Record To Date" (2007) 10 J. Int'l Econ. L. 397. 8 Eric A. Posner & John C. Yoo, "Judicial Independence in International Tribunals" (2005) 93:1 Cal. L. Rev. 1 9 Brendan P. McGivern, "Seeking Compliance with W T O Rulings: Theory, Practice and Alternatives" (2002):36 Int'l Law. 141 at 157. 10 Steger, "Systemic Issues" supra note 6 at 67. 11 William J. Davey, "Looking Forwards" (2006) J. Int'l Econ. L. 17 at 11 [Davey, "Looking Forwards"]. 2 Proposals range between two extremes. There are those that would see a virtual return to the blocking system of rulings in the G A T T system, while others seek the further legalization of the W T O system by changing voting rules to facilitate more legislation, 12 * though it is unlikely that either is politically feasible. Overall, the current discourse over reforms has focused disproportionately on the legalistic means rather than the political means for resolving trade disputes. The tendency has been towards an overly legalistic view of the role of the W T O - a form of juridification focusing too much on the function of litigation within the W T O . This thesis explores the possibility of more effective integration of arbitration as a means of improving implementation in the "handful of major, politically sensitive cases that test the limits of the system". 1 4 It considers arbitration as an alternative within the current dispute settlement system for specific cases, and as a possible middle ground between the extremes of a power based system and rules based system. 1.2 M e t h o d o l o g y & O b j e c t i v e s Ultimately this thesis works towards two broad conclusions. First, it suggests that through the development of a bias for the judicial resolution of disputes, the W T O system has, at least in a practical sense, inappropriately excluded arbitration as a truly alternative form of dispute resolution. Second, it proposes that arbitration may be a more appropriate form 12 Steinberg, supra note 2 at 249. 13 Joseph H.H. Weiler, "The Rule of Lawyers and the Ethos of Diplomats" (2001) 35 J. World Trade 191, at 198-200. 14 Steger, "Systemic Issues", supra note 6 at 67. 3 of dispute resolution in politically difficult cases where implementation of policy change is more doubtful. This thesis uses several contexts to consider arbitration as an alternative mechanism for dispute settlement within the W T O . It considers the major compliance theories, to determine i f the current bias towards judicial settlement over arbitration is justified from a compliance perspective. Next it reviews the historical development of the G A T T system and the creation of the W T O , focusing on the negotiations during the Uruguay Round, in order to determine i f there are any historical reasons for the limited integration of arbitration as an alternative means of dispute settlement. It then considers the dispute system within the current D S U for two reasons: first, to analyze the shortcomings of the legalized system of dispute settlement directed at remedies; and second, to analyze the limited use of arbitration provisions within the D S U to demonstrate the potential of arbitration as a true alternative. Last, it considers the current discourse of reform of the dispute settlement system focusing on improving remedies and procedure, over creating more space for negotiations in difficult cases. I demonstrate that arbitration has been neglected as an alternative dispute settlement mechanism, not because of any particular shortcoming, but rather by virtue of inertia of legalism that began during the Uruguay Round. This thesis ultimately proposes that the arbitration provided for in Article 25 of the D S U , but to date used only once, is uniquely suited to constitute a middle ground between the diplomatic and legalistic means of dispute resolution. It ultimately proposes a vetting 4 system within the dispute resolution system that would force disputes down one of two separate tracks: one for "ordinary" trade disputes, and another for politically sensitive cases where implementation of change in accordance with underlying obligations is unlikely to result. It thus proposes forcing a small number of politically difficult disputes down an Article 25 arbitration track as an alternative to the predominant system of litigation in the W T O , described in this thesis as the "judicial settlement system". 1 5 The purpose of replacing the current system with Article 25 arbitration in such cases is to obtain an objective ruling that would inform the discourse around treaty obligations and push the disputing parties towards negotiation. This is distinct from the judicial system, where the objective seems more and more, to establish a legally impenetrable declaration of obligations that is intended to be enforced to its letter. I suggest that this is an inappropriate approach for the small amount of politically difficult cases where compliance with such a ruling is unlikely to occur. This thesis therefore challenges the predominant bias towards the system of judicial settlement involving panels and the Appellate Body as a one-size-fits-all form of dispute resolution in all cases. M y ultimate objective is to encourage the member states to reconsider the concept of arbitration as part of the current discourse over reform of the D S U . 15 A term used by the G A T T Secretariat during the Uruguay Round to distinguish arbitration from the process of resolution though court like litigation. See G A T T , Negotiating Group on Dispute Settlement, Concepts, Forms and Effects of Arbitration, Note by the Secretariat, G A T T Doc. No. MTN.GNG/NG13/W/20, (22 February, 1988) at 3. 5 1.3 R e c o n s i d e r i n g the L e g a l i s m o f the D S U While proposals for the use of arbitration made during the Uruguay Round have been realized in the provisions of the D S U , arbitration has never been effectively tested as a true alternative within the W T O . Many proposals have been made over the best way to reform dispute settlement at the W T O in the last few years. Negotiations for improvements were originally scheduled to be finished by 2003, but have yet to be completed. 1 6 In June 2003, then W T O Director-General Supachai Panitchpakdi attempted to address systemic reform by establishing a Consultative Board to identify the institutional challenges of the W T O and to make recommendations for its improvement. 17 The Consultative Board released the "Sutherland Report" in January of 2005, and amongst other topics, sets out proposed reforms for the dispute settlement system. While it has been almost two years since its release, no outcomes for dispute settlement have materialized. Negotiations have focused on reforms that would further legalize the 18 judicial settlement system - third party rights, remand authority of the Appellate Body, 16 A full review of the dispute settlement rules and procedures was to be completed within four years of the creation of the WTO. The process commenced in 1998, and a reform package was to be approved at the Seattle Ministerial Conference in 1999, before it collapsed. See The Sutherland Report, infra note 17 at 51. At the Doha Ministerial Conference in 2001, the Ministers provided a new mandate in the Doha Ministerial Declaration, setting a deadline of May 2003 for the completion of negotiations. In that year the Chairman of the negotiating committee presented two separate drafts, but consensus could not be reached on either. W T O , Dispute Settlement Body Special Session, Minutes Of Meeting (held on 22 February 2006), WTO Doc. TN/DS/M/30, para.4. The deadline for completion was extended to May 2004 and again to August 24, 2004 without result or further deadline. See "Hong Kong Ministerial: Briefing Notes", online: W T O <http://www.wto.org/english/thewto e/minist e/min05 e/brief e/brieflO_e.htm> (last accessed 15 August, 2007). 17 Peter Sutherland at al, "The Future of the WTO: Addressing Institutional Challenges in the New Millennium", Report by the Consultative Board to the Former Director-General Supachai Panitchpakdi (2004), online: W T O <http://www.wto.org/english/thewto e/lOanniv e/future wto e.htm> ["The "Sutherland Report"]. The Consultative Board was chaired by Peter Sutherland, a former G A T T director general who was materially involved in the creation of the WTO. 18 Davey, "Looking Forwards", supra note 11 at 19. 6 sequencing of procedural steps and time periods for dealing with implementation of rulings and more regulation of panels. 1 9 A l l of these suggested reforms would lead the W T O down a path of even more formal legalism. Arbitration has been almost entirely ignored in the context of the current debate over reform of the W T O dispute settlement system. The proposals in the Sutherland Report illustrate a bias towards a legalistic form of dispute resolution. The direction of reform illustrates a belief that the G A T T system was an inept tool that has only been improved over time, without considering any of its merit. This approach to reform assumes that progress in the area of dispute settlement in international trade is linear. B y this logic, because the system has been improved by the legalism of the current system, further improvement can be achieved by increasing rules and adopting more legalistic measures. However, as one scholar has suggested, the evolution of dispute settlement may be more like a pendulum, swinging between the extremes of a power oriented system and a rule oriented system. This thesis contests the direction of reform taken in the Sutherland Report, with the notion that a form of arbitration may represent a midpoint of sorts between the two extremes of legalism, and the political side of the W T O , and that would be more suitable for the most politically difficult cases. 19 "Hong Kong W T O Ministerial 2005: Briefing Notes", online: W T O <http://www.wto.org/english/thewto e/minist e/min05 e/brief e/brieflO e.htm> (last accessed 15 August, 2007). From a North American perspective, the recent settlement of the softwood lumber dispute between Canada and U . S . 2 0 has illustrated the need to reconsider the role of the dispute settlement system. There are many who might consider the outcome an example of failure - focusing on the refusal of the only economic hegemon to comply with the rulings of the W T O , and the use of economic power to override international obligations established under the rules based system. Conversely, it might be considered a monumental success - a situation where the D S B fulfilled its role in bringing to an end a difficult and policy charged dispute that had raged for over two decades, and aiding the disputing parties in overcoming the internal domestic politics within their own respective boundaries. This example forces certain questions. H o w should failure and success be defined? Is compliance with a formal ruling the ultimate objective? If not, then what is the significance of a D S B ruling, and what should the expectations of compliance be? Should those expectations be the same irrespective of the nature of the dispute and the parties involved? These issues logically require a consideration of the best mechanism for resolving individual'disputes. 20 Canada was largely successful in a series of W T O and N A F T A decisions in their ongoing dispute with the U.S. over softwood lumber. Despite Canada's success in establishing that the U.S. were not justified in applying anti-dumping and countervailing duties on Canadian imports, the U.S. position appeared to be intransigent on the idea of returning all of the duties, resulting in ongoing references to the WTO. Finally, Canada and the U.S. entered into the Softwood Lumber Agreement between the Government of Canada and the Government of the United States of America, executed 12 September 2006 and amended 12 October 2006, under which Canada agreed to forego the return of $1 billion of the $5 billion of U.S. imposed duties. The softwood lumber treaty was quickly implemented into Canadian domestic law as the Softwood Lumber Products Export Charge Act, 2006, receiving Royal Assent on December 14, 2006. Full descriptions of the W T O disputes WT/DS236, WT/DS247, WT/DS257, WT/DS264, WT/DS277 and WT/DS311 can be found online: W T O < http://www.wto.org/english/tratop e/dispue/dispu subjects index e,htm#lumber>. 8 1.4 Working Concepts 1.4.1 Compliance and Implementation of Policy Change Before embarking on those specific tasks, it is first important to establish a working concept of compliance and implementation of policy change. In essence, this thesis follows the lead of Robert Hudec, who viewed the dispute settlement system in a political 21 way, as a mechanism to lead governments to concessions in wake o f rulings. A s such, one can draw a distinction between perfect compliance with a D S B ruling and compliance with the primary obligation of a member state under the W T O agreements. In the case of the latter, one may have compliance under the surface brought on by the 22 dispute settlement system. The compliance theories of international law and the objectives of the D S U support such an approach. In the most difficult, politically sensitive cases, where non-compliance is likely, a flexible approach to the nature of D S B rulings could be crucial for the long term survival of the W T O . A settlement negotiated by the disputing parties, that has been informed or guided by an arbitration decision, even i f not in conformity to the letter of that ruling, is superior to a legal ruling that has withstood appeal, but that is never 21 Robert E. Hudec, Enforcing International Trade Law: The Evolution of the Modern GATT Legal System. (Salem, New Hampshire: Butterworth Legal Publishers, 1993) at 360 [Hudec, "Enforcing International Trade Law "]; Marc L . Busch & Eric Reinhardt, "Testing International Trade Law: Empirical Studies of G A T T / W T O Dispute Settlement" (2004), in Daniel M Kennedy and James D Southwick, (eds.), The Political Economy of International Trade Law: Essays in Honour of Robert Hudec (New York and Cambridge: Cambridge University Press, 2002) 457 at 458 [Busch & Reinhardt, "Testing International Trade Law"]. 22 Carlos Vazquez & John H. Jackson, "Some Reflections on Compliance with W T O Dispute Settlement Decisions" (2002) 33 Law & Pol'y Int'l Bus. 555 at 566. 9 implemented. However, it is also important that any form of reform does not dilute the absolute binding effect of D S B rulings in most cases, so as not to undo the relative success of the dispute settlement. 1.4.2 Distinguishing Arbitration from Judicial Settlement in the W T O H o w does one distinguish the main dispute mechanism under the D S U from arbitration? This is an obvious starting point before trying to compare the mainstream dispute settlement system to "arbitration" under the D S U . Some observers might consider that the main dispute mechanism under the D S U is itself a form of arbitration. 2 4 The assumption likely emanates from the fact that panels under the G A T T system were more 25 similar to a form of arbitration. Furthermore, the panel system is now formally overseen by the D S B , a political body which is comprised of representatives of all o f the members of the W T O . The decisions of panels appointed to hear disputes result only in recommendations that require adoption by the D S B . In practice however, both panels and the Appellate Body are the decision-makers in the mainstream dispute resolution process and are "judicial tribunals in the international law sense". 2 6 23 Steger, "Systemic Issues", supra note 6 at 67-72; Davey, "First Ten Years", supra note 6. 24 Dora Marta Gruner, "Accounting for the Public Interest in International Arbitration: The Need for Procedural and Structural Reform", Col. Journal of Transnational Law 41 (2002-03) at 92. 25 Posner & Yoo, supra note 8 at 44. 26 Joost Pauwelyn, Conflict Of Norms In Public International Law: How WTO Law Relates To Other Rules Of International Law (Cambridge: Cambridge University Press, 2003) at 442 [Pauwelyn, "Conflict Of Norms"]; Weiler, supra note 13 at 201-202. 10 The main dispute settlement system has therefore been compared to a system of litigation 27 by those who refer to the "court like aspects of the dispute settlement in the W T O " . Comparatively, arbitration in the context of international disputes has been described as "an alternative to the dispute settlement by diplomatic means or by recourse to international courts." 2 8 The D S U itself distinguishes the mainstream litigation process from the secondary processes of arbitration by including processes under Articles 21.3, 29 22.6 and 25 referred to as "arbitration". This thesis therefore departs from the premise that the main form of dispute resolution under the D S U is a form of judicial process or court litigation, 3 0 or "judicial settlement" that is distinguishable from arbitration. I use the term "judicial settlement" as a means of describing the legalized dispute settlement system in the W T O , as it was described by the G A T T Secretariat in a report prepared during the Uruguay Round negotiations. 3 1 In that report, the G A T T Secretariat distinguished "arbitration" from "judicial settlement", the latter being a process that 32 involved the reference of a dispute to an international standing tribunal. 27 Amelia Porges, "Settling W T O Disputes: What do Litigation Models Tell Us?" (2003-2004) 19:1 Ohio St. J. Disp. Resol.141 at 141. 28 Ernst Ulrich Petersmann & Meinard Hilf, "Strengthening the G A T T Dispute Settlement System: On the Use of Arbitration in G A T T " , in Ernst Ulrich Petersmann & Meinard Hilf, eds., The New GATT Round of Multilateral Trade Negotiations (Deventer & Boston: Kluwer Law and Taxation Publishers, 1988) 323 at 337. 29 There are other forms of Arbitrations provided for under other W T O agreements such as the General Agreement on Trade in Services to resolve the amount of compensation if a state wishes to withdraw from commitment entirely, and under the Agreement on Subsidies and Countervailing Measures regarding prohibited subsidies, in Article 4.7. See David Palmeter and Petros C. Mavroidis. Dispute Settlement in the World Trade Organization: Law, Practice and Procedure. (Oxford ; New York: Cambridge University Press, 2004) at 208-209. 30 Weiler, supra note 3 at 203. See also, Posner & Yoo, supra note 8 at 10 & 45, referring to the more "court-like" dispute settlement mechanism of the WTO, and refer to it as one of several "global" courts. 31 G A T T , Negotiating Group on Dispute Settlement, Concepts, Forms and Effects of Arbitration, Note by the Secretariat, G A T T Doc. No. MTN.GNG/NG13/W/20, (22 February, 1988). 32 Ibid, at 3. 11 1.4.3 Features of Arbitration Generally and in the W T O Context In the world of private law, commercial arbitration is a form of adjudication, for which authority is derived from the agreement of the disputing parties. 3 3 It has been succinctly described as an alternative form of dispute settlement which is "based on the consent of the parties and not the coercive force of the formal adjudication system". 3 4 Arbitration in both domestic and international arenas has evolved from a level of dissatisfaction from 35 the dominant legal system. In essence, arbitration in any context must keep a reasonable distance from the litigation process in order to retain its value as a desirable alternative. 3 6 The reasons for, and the advantages of, arbitration can vary depending on the parties in dispute, the system(s) within which the parties operate and the nature o f the dispute. Traditionally, private law arbitration provides parties with an informal forum facilitating speedy resolution, protection of the confidentiality of information within the proceedings, and greater opportunity for effective enforcement of international awards in foreign 37 • courts. A l l o f these factors provide rationale for selecting arbitration in a private commercial setting, particularly in complex cases involving international trade issues. In order for arbitration within the W T O to remain a suitable option for dispute settlement, it must offer some form of significant advantage over the main adjudication option 33 Gus Van Harten, "Judicial Supervision of N A F T A Chapter 11 Arbitration: Public or Private Law?" (2005) 21 Arb. Int'l 493 at 497; Alan Redfern & Martin Hunter, Law and Practice of International Commercial Arbitration (London: Sweet & Maxwell, 1999), at 7. 34 Catherine A. Rogers, "Emerging Dilemmas In International Economic Arbitration: The Vocation Of The International Arbitrator" (2005) 20 Am. U. Int'l L . Rev. 986 at 1017. 35 Francisco Orrego Vicuna, "Arbitration in a New International Alternative Dispute Resolution System" (2002) 57:2 Disp. Resol. J. 64. at 64-65. 36 Ibid, at 64. 37 Redfern & Hunter, supra note 33 at 8-11, 15, 26-30. 12 available to disputants. However, the advantages of arbitration in the W T O wi l l necessarily be different than the advantages of private law arbitration, given the different contexts. The dispute settlement mechanisms adopted by the W T O in the D S U are distinct from international arbitration involving private interests, as they deal with international dispute resolution between states only . 3 8 A n analysis of the advantages of arbitration in the W T O context must therefore look more specifically at the features of state to state arbitration. According to Meinard H i l f and Ernst Ulr ich Petersmann, former secretary for the G A T T Secretariat and legal advisor to the W T O , international arbitration between states is distinguished from the modes of dispute resolution including judicial settlement, or what they describe as "judicial resolution", by the following features:3 9 the parties determine the arbitrator's jurisdiction; the dispute is resolved by a binding decision of the arbitral tribunal, and not a negotiated settlement; the arbitral body is constituted to hear the particular case only; the decision maker is chosen by or on behalf of the parties; the parties have control over the procedure to be followed in the arbitration; the arbitration award is final, and not subject to appeal. 38 Frank W. Swacker, Kenneth R. Redden & Larry B. Wenger, "WTO & ADR" (2000) 55:3 Disp. Resol. J. 36. 39 Petersmann & Hilf, supra note 28 at 337. 13 It is only the last four features on which this thesis w i l l focus, in distinguishing between arbitration and judicial settlement in the W T O . The first feature is not applicable as there is little room for party control over the choice of law under the provisions of the D S U . In all cases, arbitrators are required to apply the principles of the underlying agreements. Article 3.2 of the D S U clarifies that rulings of the D S B cannot enhance or diminish the rights under the agreement, and is applicable to arbitration. 4 0 Indeed, the failure to apply W T O law would defeat the purpose of a multilateral rules based trading system. A s such, the arbitrators have no jurisdiction outside of the substantive law found within the scheme of W T O agreements and obligations. 4 1 With respect to the second feature - the resolution by a binding decision of the arbitral tribunal - there is no distinction between judicial settlement and arbitration. However, one premise of this thesis is that the issuance of an arbitral ruling ought to be directed at fostering ongoing negotiation towards resolution. Indeed, I suggest that the possibility of negotiations around a single arbitrator ruling, rather than a decision which has passed through a formal legal appeal system, is more suitable for politically difficult cases. A n arbitration ruling provides an analysis of obligations that can inform further discussions, but does not provide the parties with a continuing appeal process designed to obtain a "legally correct" interpretation, which can eviscerate any room for further negotiations or practical resolution. In other words, the more legalized the procedure, the less there is any reason or political ability for a successful party to make any form of concessions towards a resolution that w i l l be will ingly implemented by the losing party. 40 Boisson de Chazournes, Laurence. "L'Arbitrage A L ' O M C " (2003) Revue de l'arbirtage, 949 at 956. 41 Ibid, at 955-6. 14 The third, fourth and fifth features listed by H i l f and Petersmann are interrelated. Two of them, party choice over the decision-maker and control over the procedure of the arbitration are forms of party control. The third feature - the overall objective of resolving the particular case only - is also a function of the party's ability to choose the decision-maker. The decision-maker is chosen in the interest of the parties to resolve the dispute, and not necessarily to create or develop principles of general application. These three features can therefore be conflated into one feature - party control over procedure and the decision-maker. In examining the question o f the potential advantages o f arbitration, I therefore consider two main criteria of differentiation in the W T O dispute settlement context: first, the principle of finality, or expressed differently, the limitation of appeal avenues; second, the element o f party control over process and the decision-maker. These are arguably the two basic principles underlying arbitration. 4 2 This thesis ultimately suggests that although arbitration has not been effectively utilized in the current D S U , arbitration may nevertheless be more suitable for certain types o f disputes within the W T O because of these two features distinguishing it from judicial settlement. 1.5 The Finality of Decisions and Party Control as Distinguishing Features The first point o f distinction is the limitation of rights of appeal or judicial review in arbitration proceedings. The creation o f the Appellate Body and legal review of panel decision is the most definitive move towards U .S . legalism. 4 3 In the case of private law, 42 George Zekos, "An Examination of G A T T / W T O Arbitration Procedures." (1999) 54:4 Disp. Resol. J. 72 [Zekos, "An Examination of GATT/WTO"] . 43 Michael K. Young, "Dispute Resolution in the Uruguay Round: Lawyers Triumph Over Diplomats" (1995)29 Int'l. Law 401 at 403. 15 it has been suggested that the liberation of arbitration involves limiting the judicial scrutiny of awards. 4 4 For the purpose of this thesis, the arbitration proposed would not include a right to an appeal of the arbitrator's decision, and the decision of the arbitrator would be final, as it currently is under the D S U . 4 5 There are two main disadvantages of the right of appeal in cases involving politically charged issues. First, they do not encourage settlements, and indeed may serve to entrench the position of the parties 4 6 and foster inflexibility within domestic political factions. Some proposals for reform of W T O dispute settlement in fact seek to reduce access to the Appellate B o d y . 4 7 Second, the added stages of appeal and various aspects of compliance review also compounds the problem of an extended delay before the provision of a definitive legal statement of the rights and obligations of the disputing 48 parties, and may serve only to prolong the dispute. Conversely, eliminating the right of appeal in politically difficult cases would put the parties back on a negotiating track, with the benefit o f an objective interpretation around which further discussions can develop. The thesis thus draws a distinction between a ruling resulting from the judicial settlement 44 Thomas E. Carbonneau, "Diversity Or Cacophony?: New Sources Of Norms In International Law Symposium: Article: Arbitral Law-Making" (2004) 25 Mich. J. Int'l L. 1206. 45 Article 25(4) of the DSU. 46 Porges, supra note 27 at 182 - where it is noted that the chances of any settlement during Appellate Body proceedings are unlikely. See also Joost Pauwelyn, "The Limits of Litigation: 'Americanization' and Negotiation in the Settlement of W T O Disputes" (2003-2004) 19:1 Ohio St. J. Disp. Resol. 121 at 124-128. [Pauwelyn, "Americanization"]. Pauwelyn discusses the general effects of legalization "spurred on" by the creation of the Appellate Body. 47 See e.g. Claude Barfield, Free Trade, Sovereignty, Democracy: The Future of the World Trade Organization. (Washington D.C.: American Enterprise Institute, 2001). 48 Busch & Reinhardt, "Testing International Trade Law", supra note 21 at 478; Yuka Fukunaga, "Securing Compliance Through The W T O Dispute Settlement System: Implementation of DSB Recommendations" (2006) 9 J. Int'l Econ. L. 383 at 385. See also Porges, supra note 27 at 145, who recognizes the practice of states to delay to protect obviously illegal measures by insisting on "the full extent of time-consuming procedural niceties". 16 system intended to be followed to its letter, and an arbitration ruling which, while still final, is intended to force the parties into a negotiated resolution, but without the same systemic delay of, or strain on, the judicial settlement system. In the case of the arbitration track, compliance, or implementation, is more in keeping with Robert Hudec's outlook on W T O dispute settlement. Given that a legal appeal was introduced into the W T O dispute settlement system as a means of ensuring the legitimacy of rulings, removal of the appeal logically would require an adjustment to preserve the legitimacy o f decisions. This would be balanced by the parties' ability to choose procedure and the decision-maker within the arbitration. When compared to judicial settlement, arbitration generally limits the involvement of the third party decision-maker, as the particular panel is established only to settle the one dispute, and w i l l follow procedures agreed to by the parties who retain control over procedure, and in some instances other than the W T O , the substantive law to be applied. 4 9 This presents a significant advantage in politically difficult cases, as it allows parties to select parties who may be more familiar with political situations within the disputing states. Based on the component of choice and consent, the current panel system is much closer to a court than arbitration. The D S U provide the disputants with only limited involvement in the choice of a panel 5 0 and none over the Appellate Body 51 52 members hearing an appeal from a panel decision, which has occurred in most cases. 49 Posner & Yoo, supra note 8 at 9. 50 Articles 8(5) (6) & (7) of the DSU. 51 Other than the original appointment of individuals to the permanent Appellate Body, which requires unanimity amongst the member states. 52 Over 70 per cent of panel decisions have been appealed. See Leitner & Lester, supra note 3 at 173. 17 One potential disadvantage of the element of party control over selection of an arbitrator is that arbitration may not have the same effect in developing broadly applicable principles, and thus may not promote predictability to the same degree as the judicial settlement system. Ultimately, I suggest that this does not present a problem in the context of dispute settlement in the W T O , and might be seen as an advantage of arbitration. First, since the arbitration track contemplated in this thesis would be used for only a limited amount of disputes, it would not disrupt the creation of principles even though it may complicate the consistent application of those principles. There would continue to be a body of developed principles, and a precedent for reference to those principles, to the extent it is appropriate for the resolution of the individual dispute. Second, as w i l l be discussed in Chapter 4 , the overriding objective of the D S U is not the creation and development of legal principles but rather the resolution of specific disputes. A s w i l l be seen later in that chapter, W T O arbitrations have already diverged from developed principles in the context of arbitrations in the W T O , in order to resolve the issue between the parties. 18 1.6 Chapter Description In reviewing the possible utility of arbitration in the D S U , each chapter that follows considers a different aspect. Chapter 2 reviews the main theoretical perspectives on compliance in international law. A review of these theories provides only an entry point, but a necessary building block, for a discussion of reform of the W T O dispute settlement. Ultimately, this thesis suggests that within these paradigms, the system of arbitration is more appropriate in politically difficult cases than the judicial settlement system now in place within the D S B . Admittedly this is the most that can be accomplished here. Compliance theories of course are unlikely to provide specific proof for establishing when a particular form of dispute resolution would prove to be superior to another. However, it would be pointless to propose the function of one form of dispute settlement over another without addressing how compliance arises in international law. More importantly, this exercise discounts the possibility that arbitration as a system has been disregarded on any theoretical perspective of compliance. I w i l l leave it for the following chapters to elaborate on how arbitration has been virtually excluded as an alternative dispute settlement mechanism. Chapter 3 w i l l track the history of dispute resolution through the transformation from the G A T T system to the W T O . It focuses largely on the negotiations during the Uruguay Round. A review of the development of the G A T T system and the evolution towards the W T O demonstrates that the negotiating parties saw a potential for the use of arbitration as an alternative to the legalized panel system that had developed within G A T T system. 19 However, the discussion of this option ended somewhat prematurely in the Uruguay Round, when the focus of the negotiating parties turned to the need to create a more structured system of judicial settlement with a full appeal system. Chapter 4 wi l l examine the forms of arbitration formalized by in the D S U as part of the creation of the W T O under the Marrakech Agreement Establishing the World Trade 53 Organization. It considers the present place of arbitration within the D S U , against that of the judicial settlement system, and is divided into two parts. The first part considers the current judicial settlement system, its objectives, and the limitation of its legal remedies. The second part of Chapter 4 then looks at the three provisions in the D S U that refer to "arbitration" - Articles 21.3, 22.6 and 25 of the D S U . The first two arise only after a dispute has been adjudicated, and where the remaining issue is compliance with a panel or Appellate Body ruling. In this sense, arbitrations under Articles 22.6 and 21.3 operate only as adjuncts to the judicial settlement system and are not in that sense pure forms of arbitration. However, they do at least provide an increased form of party control over procedure and the decision-maker. Further, they set a precedent for the use of arbitration to resolve political matters that are immune from appeal. Lastly, Chapter 4considers arbitration under Article 25, which does create a true alternative to the judicial settlement process, but one that is currently underused. It is this process that the W T O can further integrate to exploit the benefits of arbitration within the D S B . 53 The "WTO Agreement" is in fact a package comprised of a web of 30 interconnected agreements. The main underlying agreements setting out the substantive obligations of the member states are the General Agreement on Trade and Tariff 1994, (the "GATT"), the General Agreement on Trade in Services ("GATS") and the Agreement on Trade-Related Intellectual Property Rights ("TRIPS"), all of which provide for the primary obligations of the member states in respect of the overall duty of non-discrimination in trade. Disputes arising under these governing agreements are generally regulated by the DSU. 20 Finally, Chapter 5 considers the current discourse for reform, including the predominantly legalistic reforms proposed in the Sutherland Report. It considers the solutions for encouraging negotiated settlement, which is in fact the original intention of the G A T T system, and the primary objective of the D S U . 5 4 It then proposes the possible uses of arbitration despite the current institutional bias towards judicial settlement within the W T O . Here, I hope to illustrate the unexplored potential of arbitration within the D S B and thus to demonstrate why the alternative of arbitration should not be excluded from the current discourse over the reform of the W T O dispute settlement framework. While maintaining many aspects of the D S U to build on the successful record, Chapter 5 proposes the creation of a two tiered system for dispute resolution that involves a vetting process of disputes at the consultations stage by the Secretariat, based on the political nature of the dispute. It is suggested that an experiment with such concepts may well promote a realistic balance between member state compliance with their international obligations, and the necessity of maneuvering through domestic pressures. The next set of reforms of the dispute settlement system may well determine the long term viability of the W T O and the preservation of the rules based trade system. I conclude that the possibility of utility of arbitration should not be excluded from this important discourse. 54 "The Sutherland Report", supra note 17 at 50 (para. 223). 21 C H A P T E R I I : C O M P L I A N C E T H E O R I E S : P O W E R , R U L E S A N D R E P U T A T I O N 2.1 I n t r o d u c t i o n Given that compliance is the goal of any discussion over reforming the D S B , the question of the role of arbitration in the W T O cannot be explored without fully canvassing the underlying theoretical perspectives of compliance in international law. This chapter therefore proceeds to consider the relevant theoretical perspectives of compliance, particularly addressing the elements of power, rules, institutions, reputation and the ultimate objective of compliance. Theories of compliance in international law usually consider two binary elements: the role of power and the modifying effect of legal rules and institutions. This chapter considers a wide sampling of the theories of compliance, as each theory tends to offer a useful perspective on the relevant concepts or factors of compliance within international law. Before considering international law compliance theories in detail, I w i l l briefly address two theories of international relations: traditional realism and institutionalist regime theory. Notwithstanding a traditional rift between the two factions, 5 5 they are useful in informing the discourse over the use and role of rules in the D S B , and in laying a contextual foundation for the more legal based compliance theories. Indeed, these 55 Robert O. Keohane, Power and Governance in a Partially Globalized World (London & New York: Routledge, 2002) at 7 [Keohane, "Power and Governance"]. Keohane suggests at the extent of the animosity between the two by warning that academic disagreement should never deteriorate to "conflicting schools of warring scholars", which could hinder progress of understanding. Keohane seems almost apologetic about this development - suggesting that there is much that can be drawn from both realism and institutionalism. 22 theories of international relations reinforce the continuing significance of power in what has undoubtedly become a rules oriented system of litigation. A consideration of all the theories helps to provide numerous explanations for non-compliance that inform the discussion of reform of the D S U for the purpose of enhancing compliance. Such a review enables a better conceptualization of the different reasons why a state may not comply with a ruling interpreting its obligations: 1) It does not need to (realist perspective) or it is not in its best interests to comply (institutionalist regime theory); 2) The process of determining the obligation was illegitimate or unfair (Franck). Here, as has been observed by others, constituents are more likely to accept an adverse political decision i f it is made through legitimate political institutions, although the procedures for resolution must be seen as fair. 5 6 3) It is incapable of complying (Chayes); 4 ) It is politically difficult to comply because the government of the day is unable to manage factions of domestic pressures (Koh and van den Broek). Conversely a binding ruling can sometimes give the government the ability to 57 make a credible claim that its hands are tied, making concessions more 58 politically palatable; 56 Young, supra note 43 at 408. 57 Busch & Reinhardt, "Testing International Trade Law", supra note 21 at 474. 58 Marc L. Busch & Eric Reinhardt, "Bargaining in the Shadow of the Law: Early Settlement in G A T T / W T O Disputes" (2000-1) 24 Fordham Int'l L. J. 158 at 165 [Busch & Reinhardt, "Bargaining in the Shadow"]. 23 A review of these theories also helps to establish some basic premises for this thesis. First, while power can of course be moderated to a large degree by a rules based system, it w i l l continue to have a role in the resolution of international trade disputes irrespective of the existence of a legalized system. A s such, not every international dispute can be resolved through a legal mechanism. A n effective system must recognize this reality in order to be effective. Second, reputation necessarily remains a major coercive force towards compliance. It is therefore important to find the right balance between power and rules, where the element of reputation as a pressure towards compliance is engaged. Last, the current discourse of reform directed mostly towards further legalism is not justified by the prevalent theories of compliance in international law. Compliance theories do not provide a convincing basis for excluding the use of arbitration within the D S B system, but conversely, provide reasons to further integrate it for appropriate cases. 2.2 I n t e r n a t i o n a l R e l a t i o n s T h e o r y a n d the E l e m e n t s o f P o w e r a n d R u l e s The two perspectives of traditional realism and institutionalist regime theory provide a logical entry point. While somewhat one-dimensional, traditional realism's emphasis on power reminds us that the political element pervades every aspect of the trade law system, including compliance with W T O rulings. Furthermore, critique of realist theory is indeed the starting point of certain compliance theories of international law. On the other hand, institutionalist regime theory emphasizes the role of reputation as a force towards compliance, while focusing on the role of institution in regulating principles, norms, rules and decision-making procedures. In considering reform of the dispute 2 4 settlement system, it is important to recognize to what extent power continues to affect compliance, and at the same time, to recognize the importance of reputation as a coercive force that can overcome power imbalances. One cannot assume that more legalistic rules will translate to better compliance in each and very case. Disputes involving policy issues which go beyond mere commercial or trade concerns are more likely to result in implementation problems.59 In these cases in particular, it is important not to view litigation as the only course for resolution. 2.2.1 Traditional Realism & Jackson's Power vs. Rules Paradigm Traditional realism is a good starting point for this analysis, as it represents "a first cut" at understanding international politics,6 0 and is the predominant paradigm within international relations.61 Traditional realism is based on several longstanding principles 6 2 It focuses on the sovereignty of the state as the only relevant player. States make decisions solely on the basis of self interest. Cooperation with other states is explained by the idea that cooperation is often in the interest of the dominant states. States seek increases in power and wealth as a zero sum gain - any increase in wealth or power results in an equivalent decrease in the wealth or power of others. States thus seek to increase relative gains. 59 McGivern, supra note 9 at 156. 60 Robert O. Keohane, "Power and Governance", supra note 55 at 6. 61 Abram & Antonia Chayes, The New Sovereignty: Compliance with International Regulatory Agreements (Cambridge: Harvard University Press, 1995) at 9-10. 62 See generally Hans Morgenthau, Politics Among Nations: The Struggle for Power and Peace (New York: A A Knopf, 1948), and subsequent editions. 25 Realism is a power based theory that suggests that more powerful states are able to impose their will, and will do so out of self interest. It is thus for the most powerful state(s), or hegemonic entity(ies), to maintain the international order. The hegemon will do so only when its own self interest dictates the maintenance of that system. Ultimately, some realists question the existence of any system of "law" where compliance is entirely dependent on the interests of the powerful states, and not supported by enforcement mechanisms. Applying this theory rigidly, the WTO judicial system and underlying agreements do not constitute a legal system. This is an extreme view that would likely not find much support today from either international relations experts or international lawyers. The significance of the realist power based theory is illustrated in the writing of John Jackson on the "crumbling institutions of the liberal trade system".63 In this seminal work, Jackson identifies the two main guiding forces of the international system: power orientation and rule orientation. The role of rules in an international system is to overcome the influence of power, a primary purpose of any legal system.64 It is this basic paradigm that necessarily underlies any discussion considering how to best enhance compliance in international law. 63 John H. Jackson, "The Crumbling Institutions of the Liberal Trade System", (1978) 12 J. World Trade L. 93 at 103 [Jackson, "The Crumbling Institutions"]. 64 Vazquez & Jackson, supra note 22 at 567. 26 It is possible to extract three underlying tenets from Jackson's dichotomous model. First, power orientation and rule orientation are mutually exclusive. Second, a power oriented system of international order is maintained by the self interest of the powerful states. Jackson contrasts this with a rule oriented system in which the institutional regime eliminates the influence of power, and thus allows disputes to be resolved by the norms which are reflected in the creation of an objective rule structure. Finally, Jackson suggests that the movement from a power oriented system to a rule oriented system in international law has occurred by gradual evolution. 6 6 A system based on rule orientation however eliminates the influence of power through an objective rule structure. Jackson's power orientation extreme falls more in line with the realist perspective. However, treating rule orientation and power orientation to be mutually exclusive is at best an oversimplification. It has been suggested that Jackson's model has been distorted by the realist perspective, by its injection of a very narrow definition of power. 6 7 In fact, 68 Jackson did not give a clear definition of power. This realist revision results in a binary approach with relatively simplistic terms: the international system is either power oriented and thus decisions are determined by structural power, or it is rule oriented, where power has no influence on the final outcome of a dispute. 6 9 The obvious difficulty with applying Jackson's analysis this way is that it is incapable of explaining any 65 Matthew S. Dunne, "Redefining Power Orientation: A Reassessment of Jackson's Paradigm in Light of Asymmetries of Power, Negotiation, and Compliance in the GATT/WTO Dispute Settlement System" (2002) 34 Law & Pol'y Int'l Bus. 277 at 283. 66 Ibid, at 283. 67 Ibid, at 280. 68 Ibid, at 286. 69 Jackson, "The Crumbling Institutions", supra note 63 at 103. Jackson himself recognizes the possibility that many would consider the legal system in terms of an all or nothing approach, where rules are either useless, or alternatively, bind the hands of nations in all cases. 27 anomalies in the overly simplistic assumption that more power equals victory. If power could be viewed in static and absolute terms only, then one could predict the outcome of any trade dispute by merely determining which party held the balance of power. With the development of the GATT system and eventually the WTO, the prospect of doing so has become increasingly problematic. As one example, both the U.S. and the European Union (the "E.U.") have stronger records of compliance in relation to smaller nations than they do with each other.70 Therefore, while traditional realism theory may to some degree explain the disintegration of the GATT system in the 1970's (explored more in the next chapter), it is inadequate in explaining compliance in the context of the WTO. It is doubtful that even the most evolved rule based trading system, or any form of international law for that matter, would satisfy a pure realist of its status as a true legal system. Thus even the current dispute settlement, in all of its legalistic glory of binding rulings, legal appeals and retaliation measures, would still have a fundamental defect in the eyes of the realist - the inability of any member states to truly enforce the binding rulings of the DSB against the successful members state. Traditional realism would therefore not classify it as a legal system. The development of the highly successful dispute settlement regime of the WTO and its good 71 record of resolution and compliance thus belies the power theories of compliance. 70 Naboth van den Broek, "Power Paradoxes in Enforcement and Implementation of World Trade Organization Dispute Settlement Reports: Interdisciplinary Approaches and New Proposals" (2003) 37 J. World Trade 127 at 148. 71 Ibid, at 142-148; Dunne, supra note 65 at 289. 28 Nevertheless most theories of compliance in international law would, to some degree, acknowledge some of the tenets of realism, particularly the significance of power within any system of international law. Indeed, the fact that the implementation of rulings continues to be a major focus of the DSB and the ongoing discussion of reform, testifies to the practicality of traditional realist theory. The push towards further legalization within the DSB might well be characterized as an ongoing and continuous backlash against traditional realism - an aversion against any function or influence for economic power within the dispute settlement. The realist perspective thus still has some relevance, given that the W T O dispute settlement system continues to have an element of power 72 orientation where disputes are resolved by negotiations and consent. It is therefore more useful to consider the presence of power in the context of Jackson's power oriented vs. rule oriented paradigm as explained by Matthew Dunne, which is explored in further detail later in this chapter. 2.2.2 Keohane's Institutionalist Regime Theory The elements of rules and power are more carefully balanced within the institutionalist 73 regime theory, of which Robert Keohane is one of the authoritative voices. As such it provides a logical intersection between traditional realism and the theories of compliance in international law, in its focus on institutions which create, maintain and enforce principles, norms, rules and decision making procedures. It is a rationalist theory that acknowledges the world view of realism by assuming that the international system is 72 Meinhard Hilf, "Power, Rules and Principles - Which Orientation for W T O / G A T T Law?" (2001) 4 J. Int'l Econ. L. I l l at 115. 73 Harold H. Koh, "Why do Nations Obey International Law" (1997) 106 Yale L . J. 2599 at 2625. 29 generally anarchic, and that the State seeks to further it own interests. States, guided by rationality, nevertheless pursue regimes of cooperation, as it is in their best interests to 75 seek absolute gains rather than relative gains. International regimes are manifestations of the necessary international cooperation for this common objective of economic prosperity and development. As such, this fundamental theory addresses the function of rules, and provides some insight, from an international relations perspective, into the function of institutions such as the WTO in enforcing those rules. As institutionalist regime theory is based primarily on a critique of hegemonic stability theory, it has been criticized as minimizing the impact of power within international regimes, seemingly denying what to realists, is an irrefutable notion. Keohane takes issue with this "silliest" of criticisms, suggesting that power still pays a role in his theory.76 While institutionalist regime theory rejects the theories of hegemonic stability of the realist camp, it acknowledges that the hegemon may have a role in setting up a regime, 77 but eventually loses its influence. Indeed, institutionalist regime theory ultimately denies the notion of international lawyers that law can affect outcomes irrespective of politics.78 The regime theorist suggests however that is difficult for states to obtain their objectives through the mere exercise of power, because of the increasing interdependence of states. As the international community has now reached a high level of 74 van Den Broek, supra note 70 at 132. 75 Robert Keohane, After Hegemony: Cooperation and Discord in the World Political Economy. (Princeton, New Jersey: Princeton University Press, 1984) [Keohane, "After Hegemony"]. 76 Keohane, "Power and Governance", supra note 55 at 6. 77 Keohane, "After Hegemony", supra note 75. 78 Keohane, "Power and Governance", supra note 55 at 29-30. 3 0 interdependence, international regimes reduce the transaction costs of creating, regulating and enforcing rules. 7 9 Institutionalist regime theory offers a more nuanced outlook when compared to realism, and is therefore more useful as a critique of the level of legalism that now pervades the W T O system. It also offers some explanation for compliance under the W T O regime by establishing the function of rules in overcoming power differentials, and in defining the role of the institution in resolving that tension. A natural question for the theory is what promotes compliance with rules where self interest may dictate non-compliance? Here regime theory is helpful in introducing the element of reputation as a coercive force towards compliance. The maintenance of reputation is of significant importance, as others w i l l be reluctant to cooperate with states whose actions challenge the integrity of 80 underlying norms and rules. Reputation can thus impact the development of the institutions that help to set and enforce norms and monitor their implementation, and, which reduce "information asymmetries". 8 1 International institutions provide the other actors with the forum through which they can establish the parameters of the underlying principles, and monitor the behaviour of other states.82 In considering dispute settlement in the W T O , there are a few other lessons to be learned from institutionalist regime theory. Regime theory would suggest that the creation of the W T O itself represents a consensual negotiated regime that has resulted from the exercise 79 Ibid, at 4 & 30. 80 van den Broek, supra note 70 at 135. 81 Men Honghua, "Critiques of the Theories of International Regimes: The Viewpoint of Main Western Schools of Thought", at 5 online <http://www.irchina.org/en/pdf/mhhl.pdf.>. 82 Ibid. 31 of self interest by its individual members. Further, Keohane's view on the significance of the hegemon in creating the system is equally insightful. A s wi l l be seen in the next chapter, the domestic interests of the U .S . eventually changed the direction of negotiations over dispute settlement in a dramatic fashion. However, while the creation of the judicial settlement system that now includes a full right of legal appeal to a standing appellate body was a staple of the negotiation in the creation of the W T O , the influence of the U.S . within a newly stabilized, rules based system has declined. 8 4 This leaves room to now fully revisit the value of arbitration as an alternative to judicial settlement for certain disputes, an option whose potential has never been explored fully. The theory also provides some explanation for the occurrences of the failure of retaliation as an effective mechanism for inducing compliance with W T O ruling. Although the rule orientation of the D S B has resulted in forms of retaliation approved by the D S B , the underlying trade interdependence between disputing states has rendered the imposition of the full brunt of the sanctions impossible. 8 5 In this way, retaliation as a remedy is very limited in terms of improving the overall regime. Institutionalist regime theory is increasingly valuable in informing the international law theories of compliance. While regime theorists do not typically refer to law in explaining the international system in rationalistic terms, others have noted that the classical 83 Gail E . Evans, Lawmaking Under the Trade Constitution: A Study in Legislating by the World Trade Organization (The Hague; Boston: Kluwer Law International, 2001) at 55; Robert Keohane, "After Hegemony, supra note 75. 84 van den Broek, supra note 70 at 131. 85 van den Broek, supra note 70 at 147, referring to the dispute between the U.S. and the E.U. in relation to the "Foreign Sales Corporations" dispute. See infra note 335. 32 definition of regimes as "principles, norms, rules and decision making procedures" is what international law is all about. 8 6 It has been argued that the work of regime theorists such as Keohane has created the "theoretical space for international law within 87 international relations theory". 2.3 International Law Theory 2.3.1 Societal Theories of Franck and the Chaves A s mentioned earlier, it is widely accepted that the D S B has been largely successful in resolving a majority of disputes resulting in full or partial implementation. The societal theories of Thomas Frank and Abram and Antonia Chayes provide some explanation of the relative success of the D S B by emphasizing the importance of process. They also assist in identifying weaknesses and shortfalls in the current system. The ability to reconcile international law theory to institutionalist regime theory is most obvious in Franck's work. The main tenet of Franck's theory is that the presence of a sophisticated system of rules, in and of itself, creates a "compliance pul l" on parties belonging to that system. 8 8 Members develop a sense of obligation from the presence of a fair process and coherent 86 Chayes & Chayes, supra note 61 at 1 (footnote 3). 87 Koh, supra note 73 at 2625. 88 Thomas Franck, The Power of Legitimacy Amongst Nations (Oxford: Oxford University Press, 1990), at 16 [Franck, "Power of Legitimacy"]. 33 sense of principles. This line of thought rejects the notion that compliance is dependent solely on power. Indeed such a theory has proven to be inadequate in the context of the W T O , given that both the U .S . and the E . U . have stronger records of compliance in relation to smaller nations than they do with each other. 9 0 The sophisticated rules based system under the W T O has therefore, as Franck might argue, exerted a compliance pull on participating states by fostering a belief that the institution has been created and operated based on generally accepted rules. 9 1 This in turn has granted the D S B a level of legitimacy. The key elements that pervade Franck's work are the concepts of legitimacy, fairness and communitarian peer pressure. 9 2 Legitimacy has its own influence in instilling obedience, and is defined by Franck as "a property of a rule or rulemaking institution which itself exerts a pull towards compliance on those addressed normatively" because of a belief that the rule or institution has been created and "operates in accordance with generally accepted principles of right process." 9 3 His general answer to the question of why nations obey the rules is that states perceive that the rules and institutions have a high degree of legitimacy, where the institution has been created, and operated in accordance with principles to which parties have consented. 9 4 The compliance pull varies widely and thus legitimacy itself has a varying degree. 9 5 The higher the perceived level of legitimacy, the 89 Americo Beviglia Zampetti, "Democratic Legitimacy in the World Trade Organization: The Justice Dimension" (2003) 37 J. World Trade 105 at 107. 90 van den Broek, supra note 70 at 148. 91 Franck, "Power of Legitimacy", supra note 88 at 19. 92 Ibid, at 40. 93 Ibid, at 24. 94 Ibid. 95 Ibid, at 26. 34 greater the compliance pull. In order to ensure the level of legitimacy, one must consider not only the policy results or "outputs", but also the "inputs" of due process and fairness. 9 6 Franck offers four factors that determine the varying level of legitimacy: determinacy; 97 symbolic validation; conceptual coherence; and adherence. Determinacy is the level of clarity with which rules are expressed and understood, and ultimately the level of transparency.9 8 It includes the malleability of express terms in treaties, and the extent to which they can be manipulated. Where the context can be determined with relative ease and certainty, 9 9 there w i l l be higher degree of legitimacy. Clarity therefore refers not only to the inherent simplicity of principles, but also the ease with which the precise meaning can be established through legitimate process, either by third party adjudication, or by the parties themselves. The second factor of legitimacy is symbolic validation through rituals and processes that serve to express the accepted methods of enforcing the authenticity of the ru le , 1 0 0 thus legitimizing the system itself. 1 0 1 Franck's third factor of legitimacy is conceptual coherence between the rules and the underlying principles to the decision-making process. Here Frank clarifies that coherence is not synonymous with consistency. Coherence can actually endure 96 Debra P. Steger, "The Struggle for Legitimacy in the WTO", Trade Policy Research 2003 (Ottawa: Minister of Public Works and Government Services, 2003) at 2 [ Steger, "Struggle for Legitimacy"]. 97 Franck, "Power of Legitimacy", supra note 88 at 40. 98 Ibid, at 50-90. 99 Ibid, at 64. 100 Ibid, at 91-134. 101 Ibid, at 93-4. \02 Ibid, at 135-182. 35 inconsistency of treatment, as long as the distinctions which are made are justifiable on a 103 principled basis. Coherence thus relates to the connection between the rules and the rationale or principle underlying those rules. The fourth factor is adherence - the nexus between the primary rule of obligation and the secondary rule of process. 1 0 4 Here, Franck draws on the conclusion of H . L . A . Hart that in form, the international system is primitive in that it lacks "secondary rules". 1 0 5 Hart suggests that a system of law requires a hierarchy of rules. The primary rules are the primary obligations that can often be determined from the text of treaties itself. In the case of the W T O , this would be principles such as the overriding rule of non-discrimination. Secondary rules are those that define the process for developing or changing the primary rules. For the W T O , this would include the procedural rules in the D S U . The secondary rules thus create the manner by which rules are formalized through a valid exercise of power, and provide the facility within which those primary rules are legitimized, such as a court. In Hart's hierarchy, there is an overall recognition that rules are binding. However, because the international system did not always have secondary rules, international rules were merely respected out of a sense of obligation, but were not obeyed. 1 0 6 Franck agrees that the compliance pull of rules w i l l be much more forceful when the primary rules are supported with secondary rules, though this does necessarily mandate the 103 Ibid. 104 Franck, "Power of Legitimacy", supra note 88 at 183-194. 105 See H.L .A. Hart, The Concept of Law (Oxford: Claredon Press, 1961). 106 Ibid; see also Koh, supra note 73 at 2616, describing the "theoretical critique of the obligatory force of international law". 36 overdevelopment of rules. Franck recognized that the international system had changed since Hart's writings, suggesting that there were stronger obligations which arose by virtue of states' memberships within international institutions. 1 0 7 In looking at the four factors, it is difficult to detect where judicial settlement has a 108 significant advantage over arbitration that is now institutionalized under the D S U , assuming that the latter provides the parties with a greater deal of control over the decision maker and the process. The element of "clarity" reflects the ease with which primary rules can be interpreted. The necessity of an appeal within the judicial settlement process would itself illustrate a problem with interpretation. Furthermore, while an appeal process may be justified on the basis of promoting consistency, it is possible that such consistency does not reflect the true meaning of the obligation, or worse, that it merely promotes what is broadly perceived as a biased interpretation. The third and fourth factors of coherence and adherence both deal with the relationship between the principles and the process, important for both arbitration and judicial settlement. The aspect of symbolic validation is likely much more consistent with the creation of a regular court rather than a system of arbitration. However a fully matured W T O may invoke this symbolic validation in a general sense, while arbitration under the auspices of that institution may provide only slightly less value as a symbol than does the judicial settlement system. 107 Franck, "Power of Legitimacy", supra note 88 at 185-186. 108 This process of the institutionalization of arbitration by virtue of Article 25 of the DSU is more fully discussed in Chapter 3 below. 37 In his later work, Franck refers more to fairness than legitimacy. He suggests that nations have little incentive to obey rules that are not perceived to be fair, and that the level of fairness is determined by the process of discourse. A s such, the system as a whole is more so a process than a system of rules. The obedience to rules arises from a cost benefit analysis, but the ultimate compliance pull emanates from communitarian peer pressure. 1 1 0 Overall Franck still views legitimacy as the link to compliance, a significant point of contention with Keohane's institutionalist regime theory. 1 1 1 The work of Thomas Franck is therefore useful in explaining the relative success of the D S B . The sophisticated rules based system under the W T O has, as Franck might argue, exerted a compliance pull on participating states by fostering a belief that the institution has been created and operated based on generally accepted rules. The success of the D S B can of course be partially attributed to the creation of primary rules during the Uruguay Round - the W T O treaties, in which the founding members participated. The well developed legal framework of the W T O has resolved a majority of disputes resulting in full or partial implementation, largely due to the development of secondary rules. This in turn has granted the D S B a level o f legitimacy. The existence of a set of primary rules, and a sophisticated process for regulation would exist whether or not a particular dispute is resolved by adjudication through a standing judicial body, or an arbitrator who applies primary rules through secondary rules chosen by the disputing parties, as secondary rules need not be sedentary. However, secondary rules ensure the process is fair and thus 109 Thomas Franck, Fairness & International Law and Institutions (Oxford: Oxford University Press, 1995) [Franck, "Fairness & International Law "]. 110 Koh, supra note 73 at 2642. 111 Robert Keohane, "Power and Governance", supra note 55 at 121. 112 Franck, "Fairness & International Law", supra note 109 at 19. 38 legitimate, and are designed to foster the consensus building process within the W T O . Arbitration that is already incorporated into the D S U simply provides an alternative form of secondary rules. It is not evident that the use of a developed arbitration process for specific disputes that allows individual disputants more control over process or secondary rules, would be any less beneficial from the judicial settlement system in this respect. Some might argue that outside parties may have less confidence in such a system, though it is the disputant parties that either w i l l or w i l l not comply with a specific rulings or award in a specific case. It is in fact arguable that the opposite is true - that the disputing parties would be more confident in the legitimacy or fairness of the secondary rules that they helped design in respect of the specific dispute at issue. This may be even more true for many of the developing and least-developed nations that have joined the W T O after the Uruguay Round, or who may have felt pressured into more rigid form of "one size fits a l l " dispute settlement during the Uruguay Round by the threat of U .S . unilateralism. A s discussed in the next chapter, the actual participation of other states in the creation of the W T O dispute resolution system was marginalized by this factor. B y itself, the compliance pull theory fails to explain all incidents of non-compliance in cases of judicial settlement. One example is the dispute between Canada and Brazi l over 113 the regional aircraft industry. Canada has generally demonstrated a high regard for the principles of the W T O . Its leaders have referred to the importance of the binding nature 113 See e.g. Canada - Export Credits and Loan Guarantees for Regional Aircraft - Recourse to Arbitration by Brazil under Article 22.6 of the DSU, WT/DS222/ARB, 17 February, 2003 online: W T O <http://www.wto.org/english/tratop e/dispu e/cases_e/ds222 e.htm >. 39 of such decisions of international bodies in the context of the U . S . refusal to adhere to decisions relating to N A F T A complaints relating to softwood lumber products. 1 1 4 Canada has been one of the primary users of the D S B , yet has failed to implement the D S B ruling in favour of Brazi l . If Canada perception's of the W T O is one of a high level of legitimacy, why has it failed to fully implement its decision in even a single case? Franck's theory is limited in this respect, unless fairness or legitimacy can be seen as varying from case to case - where the same set of rules and process are seen as less fair or legitimate depending on the context, or perhaps importance, of the case at issue. Franck's work is often logically compared to that of Abram and Antonia Chayes. A common feature between the two theorists is that they provide convincing explanations of the important role of fairness in explaining compliance in international l aw . 1 1 5 The Chayes propose a cooperative mode of compliance similar to that of Franck, but emphasize the role of discourse in their "managerial model". Their theory begins with a critique of real ism. 1 1 6 L ike Franck, the Chayes challenge the realist tradition, contending that it is through treaty regimes that states comply, and not through the threat of any form of sanction. 114 Peter O'Neill & Stewart Bell, The Vancouver Sun 1 October, 2005. In response to the U.S. reaction after a NAFTA panel decision favouring Canada in the longstanding softwood lumber dispute, Paul Martin, then Prime Minister of Canada was quoted as saying: "Free trade depends on a dispute settlement procedure that is respected by all parties in letter and in spirit, and unfortunately the reaction to ("of) the U.S. in the face of the latest NAFTA panel decision on softwood lumber mocks that principle, and in so doing it sends the wrong message to the world." 115 Marco Bronckers & Naboth Van den Broek. "Financial Compensation in the WTO: Improving the Remedies of WTO Dispute Settlement" (2005) 8 J. Int'l Econ. L. 101 at 111. 116 Chayes & Chayes, supra note 61 at 9-10. 40 According to the Chayes, the natural propensity to comply with international obligations is founded on three main factors. First, they argue that it is more efficient for states driven 117 • by bureaucratic decision making machines to comply. The investment of time and resources into the initial development of the obligation is substantial. This is surely no truer than in the case of the obligations under the W T O underlying agreements, which were the subject of complex negotiations for over seven years during the Uruguay Round. The Chayes reason that it is too costly to revisit such decisions by choosing not to follow the course of conduct to which a state has committed. Secondly, the Chayes point out that the decision to incur the obligation is subjected to a wide and complex process at both the 118 domestic and the international levels. The rigorous processes provide some assurance that the end result is representative of the interests of the state as a whole, as the policy has been exposed to high level of scrutiny. Third, the Chayes point to the development of "binding" principles of international law such as pacta sunt servanda,119 the obligation of states to follow the clear wording of treaty obligations in good faith. Here the Chayes cross paths with the regime theorists. Regime theory would also focus on the cooperative efforts that occur within the web of norms and rules, although it would deny the Chayes' 120 contention that they would often centre around the treaty itself. The Chayes' critique of realism next engages the question of why nations do not always comply. While they acknowledge the most obvious examples of breaches of international law as cases of bare defiance, they consider these to be the exception rather Ul Ibid, at A. 118 Ibid, at 4-7. \\9Ibid. at 8-9. \20 Ibid, at 1. 41 than the rule. In explaining the majority of cases of non-compliance, they identify three main factors: first, determinacy and ambiguity; second, the limitation of capacity of state; and last, the variable time lag that naturally arises after commitments are formalized. 1 2 2 The first factor is similar to Franck's concept of determinacy, as the Chayes identify the fine balance that must be struck within the specific wording of treaty obligations. While ambiguity in treaty wording may arise from the inability to foresee all possible conflicts, it may also result from the deliberate design of two parties who are unable to resolve a 123 difference. In the case of the latter, ambiguous language is used to create the obligation, and then it is left for the parties to propose their respective interpretations, when and i f the occasion arises. A s such, the Chayes' theory can be used to question the pursuit of an objective, legally correct interpretation of obligations in the wording of the W T O underlying agreements. This point is significant in considering a second factor for non-compliance: the time lag between a state's commitment and its ultimate compliance. 1 2 4 The Chayes acknowledge that transitional periods are required for many treaties, particularly ones involving matters such as human rights, where the level of adaptation may be great for certain 125 states. They emphasize however that initiating the process though an express 121 Ibid, at 10. 122/to', at 10-17. 123 Ibid, at 10-11. 124 Ibid, at 14-15. 125 Ibid, at 16-17. 42 commitment should be viewed as a significant step in and of itself, and ought not be 126 dismissed as "aspirational" only. The third factor is the limitation of capacity. Here the Chayes's focus on a state's economic inability to meet an obligation. The inability to comply may also arise from the limited resources of developing countries, or the inability of government to fully control 127 the conduct of individuals who are the actual targets of certain treaties. Naboth van den Broek expands on this factor, suggesting that in policy driven cases, there can be too 128 much pressure from domestic interests to permit the state to comply with the ruling. While the Chayes might agree with Franck that compliance is not elicited through the threat of sanction, they emphasize the "iterative process of discourse" rather than Franck's touchstones of legitimacy or fairness, as that main thrust for inducing compliance. A s in the case of institutionalist regime theorists, reputation plays a prominent role as a pressure towards conformity to rules. However where the institutionalist regime theorist views the regime as encompassing a set of prohibitive norms, the Chayes see the institution as constituting the forum for management of i • 130 discourse. The Chayes advocate for softer forms of law designed to engage reputation. Thus, for example, transparency is necessary, as it facilitates a system of monitoring behavior and 126 Ibid, at 17. Ml Ibid, at 13-14. 128 van den Broek, supra note 70 at 149. 129 Koh, supra note 73 at 2601. 130 Chayes & Chayes, supra note 61 at 228; Koh, supra note 73 at 2638. 43 reporting on specific instances of non-compliance. The institution's role is to ensure transparency and to provide a form of dispute settlement that can generate an authoritative interpretation of obligations. In this regard, the Chayes are unimpressed by the international lawyer's "obsession" with binding adjudication, and are more 131 favourably disposed to compulsory conciliation. This softer form of dispute settlement fosters the iterative process of discourse, making space for the underused methods of 132 persuasion, or "jawboning". This iterative process of discourse takes place on several levels: amongst the parties, within the organization and before the publ ic . 1 3 3 The Chayes' managerial approach, like Franck's compliance pull , thus focuses on process. Using the concepts of H L A Hart, the Chayes' and Franck focus more on secondary obligations rather than primary obligations. The role of the institution is to manage the process of discourse and provide the forum for a discussion which converges around treaty norms. 1 3 4 This process fosters an environment where any deviation from norms require explanation and justification. 1 3 5 This process of discourse and justification itself reinforces those norms, and ultimately engenders a bias towards compliance. Non-compliance is explained not as a failure of the system, but as attributable to a state's lack 136 of capacity to implement decisions considered to be binding. 131 Ibid, at 24-25. Ml Ibid, at 25. 133 Ibid, at 25. 134 Ibid, at 340. 135 Ibid, at 118-123. 136 van den Broek, supra note 70 at 14, 151-152. 44 The managerial approach can thus be used to challenge the notion that non-compliance with the letter of a D S B ruling should be considered a failure. Rather, the ruling should foster further discourse around the ruling that leads to a negotiated result, whether or not the ultimate resolution conforms to the letter of the ruling. The managerial model of the Chayes, like the contextualist approach of Matthew Dunne discussed later, is thus helpful in drawing the parameters of an appropriate role for the D S B . A s for the explanation for compliance, there seems to be little difference as between a judicial settlement system and an institutional arbitration system. The underlying principles do not necessarily prefer a decision with a process before a fully institutionalized standing body with legal appeal avenues over a single decision of an arbitrator. The theory of the Chayes emphasizes the process as a means to an end, not an end itself. The key element is the establishment of a process around which treaty norms are discussed, thus facilitating the iterative discourse. The treaty combined with the third party's interpretation of the obligation under the treaty forms the basis around which negotiations can converge. The purpose is thus not to produce an elusive, objective interpretation that may not exist, but rather to provide guidance as to the merits of each party's position on the interpretation of the obligations. A system of judicial settlement may provide this forum in a formal sense, but institutional arbitration w i l l facilitate the discourse around treaty norms between the parties as well . One main distinction with the judicial settlement system is that in an arbitration system, the discourse may occur in fewer stages when there is no appeal right. This does not 45 necessarily detract from the quality of that discourse, or the extent to which deviation from that ruling requires explanation and modification. Conversely, ongoing appeal rights tend to overemphasize the "correctness" of the ruling, and thus leave less room for any further negotiation. 2.3.2 Koh ' s Transnational Legal Process Harold K o h praises what he terms the societal theories of Franck and the Chayes, but finds that they fail to address an important aspect: the manner in which global norms 137 become internalized as domestic law. According to K o h , the important linkages with domestic systems and the process of norm internalization forms part of a "transnational 138 legal process". He observes that while both Franck and the Chayes correctly conclude that compliance is based partially on the perception of legitimacy and fairness or on reputation, that there is another compelling motivation: the internalization of norms into 139 domestic systems. This process of internalization is initiated by an actor who provokes an interaction, ultimately forcing the issue toward an authoritative interpretation that generates a legal ru le . 1 4 0 The actor who has initiated the process seeks to have the opposing party internalize the determinative interpretation. Domestic institutions establish proceedings 137 Koh, supra note 73 at 2646. \3SIbid. at 2646. 139 Ibid, at 2646-9. UOIbid. at 2649. 46 which then induce and maintain regular conformity. The repetition of this process is important in broadcasting and reinforcing the norm. The involvement of additional actors such as international governmental organizations (IGOs) like the W T O enhances this process, as the IGO provides the forum for norm enunciation and elaboration. 1 4 2 This transnational process ultimately enmeshes the norm into the domestic legal system. The norm first becomes widespread and is embedded in the social fabric. This leads to a formal policy that embeds it politically. Finally it is embedded legally by way of formal incarnation into domestic l a w , 1 4 3 and thus becomes enforceable. Koh ' s theory is insightful in addressing the role of domestic systems in relation to compliance and enforcement. Naboth van den Broek builds on this in his combined theory of compliance, which considers the process of norm internalization. In so doing, van den Broek relies on empirical data from the first ten years of decisions from the W T O . Indeed, one of his explanations for non-compliance is the conflicting norms of international and domestic arena and the slow rate of internalization of the "higher" norms of international l a w . 1 4 4 The transnational legal process provides a framework for understanding how the dynamics of domestic politics fits within the issue of compliance with W T O rulings, emphasizing the limitations of a self-standing international law system. Without the internalization process described by K o h , the ability to enforce a ruling is limited. Yet the 141 Ibid, at 2654. Wllbid. at 2650. 143 Ibid, at 2651. 144 van den Broek, supra note 70 at 148-151. 47 internalization process he describes is an inherently political process that is affected by pressures from domestic constituents. This is often a major hurdle to implementing change in politically difficult cases. A ruling by the D S B in effect creates external pressures on the state government, 1 4 5 which can be used as tool for justifying negotiation on an issue that is resisted by the domestic constituents. Indeed, while D S B rulings are said to be binding, the W T O still lacks an autonomous enforcement power. A s such, state governments can only credibly claim to its constituency that it is bound by the ruling when there is a lack of information about the consequences of non-compliance. 1 4 6 This is an extremely tenuous basis on which to manage domestic pressures, and it is suggested, unrealistic on a long term basis in democracies with effective opposition parties. In this respect, an arbitration ruling that is rendered by an arbitrator mutually agreed upon by the parties, as distinct from a ruling from a permanent judicial body, should be equally useful in manoeuvring through the domestic politics to justify a negotiated resolution. The enforceability of rulings from each are equally impaired. The result from either is the same - an objective assessment of primary obligations has clarified the breach. Within the domestic sphere of that state, a change in policy can be justified on the basis that to do otherwise would be to breach an international obligation under a treaty, and jeopardize its reputation and the benefits that state enjoys under that treaty, unless an alternative resolution can be negotiated. Both a judicial settlement system and an arbitration system can lead to this result. 145 McGivern, supra note 9 at 153-4. 146 Bush and Reinhardt, "Testing International Trade Law", supra note 21 at 474. 48 2.3.3 Contextualism - Jackson's Power vs. Rules Paradigm Revisited A s can be seen by the previous discussion, the compliance theories consistently look at the binary elements of power and rules, generally include a critique of the realist's obsession with power. The work of legal scholar Matthew Dunne takes the fundamentals of Jackson's paradigm of power orientation vs. rule orientation, and revisits its concepts as applied by the realist perspective, while incorporating many concepts from the compliance theories of international law. Dunne proposes an alternative perspective to the realist view that power orientation and rule orientation are mutually exclusive categories. He advocates for "contextualism" and "flexible rule orientation", concepts which produce a more nuanced view of the nature of institutions and states. 1 4 7 This synthesis of power orientation with the rule structure is reminiscent of institutionalist regime theory, as Dunne suggests that rules do not constrain the exercise of power, but 148 rather serve to legitimate some behaviour while illegitimating other behaviours. In the end, the more powerful nations are still able to resist the application of norms that run contrary to its own interests, but pressures by other mechanisms induce compliance in the majority of cases. B y this theory, compliance is not enforced exclusively by hard law measures such as sanctions, but rather more effectively by a mixture of hard law, and forms of soft law such as monitoring of conduct. Maintaining transparency is important as it exposes the 147 Dunne, supra note 65 at 310-332 & 336-340. US Ibid, at 329. 49 non-compliance of any particular state to a W T O ruling. The utility of soft law is thus similar to its place in the managerial model proposed by Chayes. More importantly, Dunne's contextualism refines the notion of power within the W T O , suggesting that outcomes of disputes are determined by a number of factors: the nature of the issue; the behaviour of the parties in question; and the context within which the dispute arises and is contested. 1 4 9 While it is likely that the state with more power wi l l be successful in a given dispute, power cannot be defined in static terms. To Dunne, power must be considered in the context of the particular issue in question, not overall structural power. Dunne uses the example of the level of commitment as an independent variable of power. 1 5 0 While larger countries may have less time to devote to one of many concerns or interests, a smaller, less powerful country may be much more devoted to a favourable outcome on an issue that is its primary concern. 1 5 1 This serves to refine the definition of power within the W T O system. It is neither an absolute concept, nor one that is based only on economic power in the classical sense. Power can, as international lawyers would hope, be derived from the rules based system. In the case of W T O , this explains how developing countries can indeed exercise a form of power. Secondly, Dunne proposes that the behaviour of the parties within the dispute, which 152 cannot always be predicted, is itself a factor on the outcome. Here, economic power is 149 Ibid, at 317-321. 150 Ibid, at 317-319. 151 Ibid. Here Dunne, uses the example of Ecuador in the Bananas dispute with the E . C . . Ecuador would have been much more concerned about the banana industry, a staple of its economy, while the E.C. would have undoubtedly had numerous other trade matters of much higher concern. \52 Ibid, at 317-321. 50 subservient to the elements of reputation and perceived reality. While a more powerful country may be able to draw on its broader economic influence, a smaller party may be able to engage tactics such as public conformation to "shame" the more powerful , 153 party. Dunne considers the context of the dispute as the third factor affecting the outcome of disputes. 1 5 4 The available rules, such as the provisions authorizing a party to exercise retaliation, combined with the surrounding circumstances, may give the smaller, weaker country a form of leverage. While negotiations on a particular trade dispute may be bilateral, it is resolved in a multilateral context. Again, using the E.C. Bananas III dispute 1 5 5 as an example, Dunne points out that Ecuador, a developing state, was able to obtain the D S B ' s authorization for a form of "cross retaliation". This allowed Ecuador to impose sanctions against the European Community (the " E . C " ) in the area of intellectual property, a pressure point for the E . C . 1 5 6 Ecuador's action ultimately contributed to the settlement of the case. The concept of power has also been used to include other contextual components: the ability to influence the decision-making of other states through "communicative measures"; the ability to influence the formation of internalization consensus and the creation of an international law; and a state's ability to • 157 insulate itself from the influence of other states in determining its own policies. 153 Ibid, at 318-319. 154 lb id. at 320-321. 155 See infra note 410. 156 Dunne, supra note 65 at 320-321. 157 van den Broek, supra note 70 at 134. 51 Dunne's concept of contextualism thus suggests that outcomes in trade disputes w i l l always be uncertain and unpredictable to a degree, because the behaviour of the parties and the reaction of the international community can never be predicted with total 158 accuracy. He rejects the simplistic equation of "economic power equals victory". Dunne further suggests that the shift toward rule orientation is neither absolute, nor evolutionary. It is more akin to a pendulum, which swings from one end to the next, while never reaching either extreme. 1 5 9 One might therefore conclude that where the system has swung too far in favour of rule orientation, there w i l l eventually be a backlash towards the influence of power. The flexible rule orientation model, while recognizing the use of sanctions as incentives for compliance, also considers the tactics and tools of parties within international institutions. Dunne's approach therefore intersects with the interests of the societal theorists: Franck's focus on issues of legitimacy, and the role of the institutions in the managerial approach of the Chayes. The international institution becomes the instrument for channeling political pressures, engaging the element identified by Thomas Franck, Harold K o h and Robert Hudec as communitarian pressure. 1 6 0 Jackson himself recognizes that rules that are negotiated may often have little meaning, but at a minimum, create a system in which disputes can be resolved by a combination of rule application and 158 Dunne, supra note 65 at 332-335. 159 Ibid, at 332-335. 160 Koh, supra note 73 at 2642; Robert Hudec, "Broadening the Scope of Remedies Under the WTO" in Freidl Weiss, ed., Improving W T O Dispute Settlement Procedures: Issues and Lessons from the Practice of Other International Courts & Tribunals (London: Cameron May Ltd., 2000) 369 at 400 [Hudec, "Broadening the Scope"]. 52 n e g o t i a t i o n . 1 6 1 The negotiations take place w i t h i n the context o f rules, and b y reference to 162 the l i k e l y interpretation o f the obl igations o f the transgressor. T h e rule orientation contemplated by Jackson, as revised b y D u n n e , also provides a basis to observe h o w the j u d i c i a l settlement system has been used excessively . D u n n e ' s analogy o f the p e n d u l u m provides a counterpoint to the assumption that evo lut ion o f dispute settlement i n G A T T and the W T O is l inear. U s i n g the p e n d u l u m analogy, perhaps it has s w u n g too far, or has at least reached its l i m i t , before reaching the extreme o f total l ega l i sm. The trend towards over- legal ism through extensive rules and procedures fosters a false sense o f confidence and expectation i n a one-size-f its-al l f o r m o f dispute resolut ion. D u n n e ' s adaptation o f Jackson 's paradigm is an important u n d e r l y i n g f ramework f r o m w h i c h to consider and conceptual ize effective re form. W h i l e D u n n e ' s refinement o f Jackson 's paradigm w o u l d not support reversion to a p o l i t i c a l or p o w e r based system, it helps conceptual ize re form as the search for the ideal balance between rule orientation and power orientation. T h i s balance can be best achieved i n an alternative system o f dispute settlement for the smaller number o f disputes for w h i c h the legal ist ic means w i l l not result i n implementat ion o f p o l i c y change. Disputes i n v o l v i n g p o l i c y i m p l i c a t i o n s w h i c h go b e y o n d the issue o f trade are less l i k e l y to result i n implementat ion based o n the appl icat ion o f rules o n l y . 1 6 3 161 Jackson, "The Crumbling Institutions", supra note 63 at 103. 162 Ibid, at 99. 163 McGivern, supra note 9 at 156. 53 2.4 Conclusion Records of compliance within G A T T and W T O cannot be explained exclusively by the power based theories of realism, economic rationalism, or compliance pull of international l a w . 1 6 4 A s such it is important to canvass a spectrum of theories. The common element of these theories is an emphasis on process as a means of facilitating a discourse amongst states, as opposed to a system for obtaining an objectively correct interpretation of the primary obligations. In all of these theories, the role of the institution is not to provide top down objective interpretation of obligations, but rather to provide a forum and process for resolution. The rules provide guidance, which promote a legitimacy and sense of fairness thus promoting compliance. Secondly, they collectively emphasize the importance of reputation in the process, an element which does not necessitate a fully legalized system of judicial settlement for every case, irrespective of its nature. A n arbitration ruling that provides an objective interpretation of obligations equally requires the unsuccessful party to contend with damage to reputation that necessarily flows from non-compliance. The ultimate role of the D S B is to create the framework and a process of secondary rules that focuses discourse around the interpretation of the primary rules - the obligations of the parties. While a judicial settlement system is useful for this process, it has not worked in every case. A n arbitration system that injects more party autonomy over the decision-making and the retention of the process might be one way of addressing cases where non-164 van den Broek, supra note 70 at 141-2. 54 compliance is more likely due to a variety of factors - economic feasibility, overhaul of fundamental values, or management of domestic factions. The importance of reputation is a recurring theme in this thesis, as it contends that any structure of the current D S B must leave room for, and engage, the element of reputation into the dispute settlement process. The utility of arbitration is in creating an alternative rule framework, as contemplated by compliance theorists such as Thomas Franck and Abram and Antonia Chayes. In difficult cases, arbitration could be just as effective as the system of judicial settlement in providing a ruling against which an offending state must measure its own behaviour. It has been argued that the goal of high level panels of the judicial settlement system is to obtain an objective analysis of high profile dispute to unlock a political stalemate. 1 6 5 If that is indeed the main purpose, it would seem that an arbitration ruling, without the necessity of a permanent judicial institution with full legal appeals, could well fulfill that objective, and in some cases, more efficiently. While this may not necessarily result in a negotiated resolution in all difficult cases, it would at least not compound the damage on the judicial settlement system where states fail to comply after an extended appeal process. The very presence of the arbitration ruling would thus set the stage for further discourse as to how the losing party wi l l respond, while at the same time, engaging scrutiny and communitarian pressure of the W T O member states. While arbitration under the D S U takes place outside the formal judicial settlement process, and is not subject to appellate 165 Joost Pauwelyn, "The Sutherland Report - A Missed Opportunity For Genuine Debate On Trade, Globalization And Reforming The WTO" (2005) 8 J. Int'l Econ. L. 329 [Pauwelyn, "A Missed Opportunity"]. 55 review, the parties must notify the D S B of the outcome, and any D S B member may raise any point relating thereto. 1 6 6 There is thus an institutional process of discourse, taking place in the context of the D S B , which scrutinizes the conduct of the disputing parties and reinforces the norms of the W T O , and creates a greater force towards compliance. The elimination of an appeal in a few politically difficult cases would reinforce the notion of the negotiated resolution. This is explored in more detail in Chapter 5, after the review of arbitration proposals in the Uruguay Round in Chapter 3, and the discussion of the current judicial settlement system and arbitration provisions in the D S U in Chapter 4. 166 Article 25.3 of the DSU. 56 CHAPTER III: THE DEVELOPMENT OF THE DSB IN THE URUGUAY ROUND 3.1 Introduction The purpose of this chapter is to consider the arbitration alternative in the context of the historical development of the judicial settlement system within the G A T T system and the W T O . In order to consider the potential and utility of arbitration as an alternative, this chapter necessarily looks at the development of the panel and appeal system, which is now the primary means of dispute resolution firmly entrenched in the W T O . In tracking the development of arbitration in the context of the historical development of G A T T and W T O dispute settlement, I make the following points: 1) Arbitration in the context of the G A T T system and the W T O has been conceived as an alternative to the system of judicial settlement now in place. 2) Throughout the existence of the G A T T system and the negotiations in the Uruguay Round, arbitration as an alternative to judicial settlement has received widespread recognition. A s such, there has been the political w i l l to consider arbitration in the context of the G A T T system and the W T O dispute settlement process. 3) The circumstances surrounding the development of the D S U were such that the concept of arbitration, while seriously considered as an alternative to the dominant mechanism of judicial settlement, was not fully developed in the Uruguay Round. 57 In making these points, I demonstrate that the process of arbitration has a place in the W T O and indeed untapped potential, while having historical support as an alternative to the predominant judicial settlement model. Further, this historical review would suggest that the development of the dispute settlement system, in terms of use and records of compliance, has been pendulum-like and not linear. The history of dispute settlement in G A T T has established that there have been periods of high compliance, following the strengthening of the rule system, followed by an eventual trailing off of utility, thus necessitating new reform. This pendulum-like nature of development necessitates a consideration of the direction of reform, and challenges the assumption that a more legalistic approach of the judicial system is what is needed for effective reform of the dispute settlement system. In so arguing, I suggest in later chapters that arbitration may well be a suitable model to strike a middle ground between a rules based orientation and a power based orientation, in particular cases. 3.2 History of GATT Prior to the Uruguay Round The creation of the current dispute settlement system was a long evolutionary process beginning in 1947 with the creation of the G A T T system as a by product of the failure of the International Trade Organization (the "ITO"). The ITO was conceived in the Bretton Woods Conference of 1947, and was meant to fall within the framework of the United Nations. The Conference sought to address the world wide problem of economic nationalism leading up to World War II, and ultimately led to the creation of the 58 International Monetary Fund and the World Bank. The Havana Charter proposing the creation of the ITO sought to monitor and regulate economic activity, including the problem of escalating tariffs that had reached unprecedented levels in the 1920's and 167 1930's. , D / It was a well developed and comprehensive document that included explicit obligations based on the principles of non-discrimination. The Havana Charter was not only a trade agreement, but included areas such as competition and labour. The Havana Charter demonstrated a widespread desire to incorporate several forms of dispute settlement, including both judicial settlement and arbitration. Chapter 8 of the Charter contained the provisions for adjudication and arbitration of trade disputes. Article 93 contained a specific provision that the executive could refer disputes to arbitration "upon such terms as may be agreed between the Executive Board and the Members concerned". Article 94 replicated this wording for a determination of the presence of any nullification or impairment of benefits. In addition, Article 96 provided a form of judicial settlement, by creating a reference for an advisory opinion to the International Court of Justice. Dispute settlement under the Havana Charter was thus conceived as a multi-tiered system, with the benefits of two alternative forms of dispute settlement processes involving decisions from third parties, though without clear guidelines as to how each would apply in any particular case. The difficulty of integrating these two forms of dispute settlement was therefore not resolved by the text. 167 Par Hallstrom, The GATT Panels and The Formation of International Trade Law. (Stockholm: Jurisfdrlaget, 1994) at 22. 59 The Havana Charter was ultimately rejected by the U .S . Congress over concerns that the 168 other members did not share the U.S . vision of free markets. Consequently, the Havana Charter was never ratified, and the ITO was relegated to "still-born" status. 1 6 9 While the negotiating parties failed to create an institutional structure for a trade agreement, the parties did manage to salvage the remnants of the trade agreement itself. However, in order to circumvent the scrutiny of the U.S . Congress, the agreement was limited to trade i 170 issues only. On October 30, 1947, the 23 Contracting Parties, as they were referred to in the treaty, 171 signed the General Agreement on Trade and Tariffs ("the G A T T " ) and the Protocol of Provisional Application putting the G A T T into force on a provisional basis - where the 172 legal basis was accepted without actual ratification. While it remained a provisional agreement with a tentative legal basis, it nevertheless established a reduction of tariffs 173 and fixed binding tariff rates. The dispute settlement provisions were contained in Articles X X I I and X X I I I of the G A T T , which set out only a vague framework. Article X X I I required the Contracting Parties to give "sympathetic consideration" to the matters affecting the operation of the 168 Ibid, at 23. 169 G A T T , Negotiating Group on Dispute Settlement, Concepts, Forms and Effects of Arbitration, Note by the Secretariat, G A T T Doc. No. MTN.GNG/NG13/W/20, (22 February 1988). 170 Hallstrom, supra note 167 at 25. 171 U N Doc E/PC/T/214 October 13, 1947, entered into force Jan.l 1948, 61 Stat. A5, 55 U.N.T.S. 187. 172 Hallstrom, supra note 167 at 25-6. 173 The obligations and customary practices were later more formally set out in the G A T T , Understanding Regarding Notification Consultation, Dispute Settlement and Surveillance, 3 December 1979, G A T T Doc. No. L/4907 (the "Tokyo Understanding"), signed by 99 Contracting Parties at the conclusion of the Tokyo Round of G A T T negotiations in 1979. 60 G A T T , and provided a vague process for consultations, thus reinforcing the political element of the organization. Article X X I I I provided the process for legal complaints and represented the only teeth to the dispute settlement system. However, it constituted a basis for the eventual development of the panel system of the G A T T system (hereinafter referred to simply as " G A T T " ) and, indirectly, the more formalized judicial settlement system of the W T O . Article X X I I I allowed the Contracting Parties to rule on governmental action alleged to have violated obligations under the G A T T , to make recommendations, and to authorize an aggrieved state to withdraw concessions. For the first few years, disputes were addressed through the formation of "working parties", which were to investigate complaints, guide disputants towards solutions and make recommendations for resolutions. 1 7 4 Overall, the two dispute settlement articles provided very little in terms of a procedural framework, and have been best described as "a diplomat's concept of legal order". Interestingly, while Article X X I I I tracked much of the language of the Article 93 o f the Havana Charter, it omitted any reference to an arbitration alternative. This has led to the conclusion that it may have been deliberately excluded. 1 7 6 If true, this would only reinforce the significance of the specific discussion of an arbitration alternative in the Uruguay Round and the eventual inclusion in the W T O dispute settlement process. 174 Young, supra note 43 at 393. See also Porges, supra note 27 at 148-149. 175 Hudec, "Enforcing International Trade Law ", supra note 21 at 7. 176 Valerie Hughes, "Arbitration Within the W T O " in F. Ortino & E. -U. Petersmann, eds., The WTO Dispute Settlement System 1995-2003 (The Hague: Kluwer Law International, 2004) 75 at 76. Notably, while a report prepared by the G A T T Secretariat during the Uruguay Round notes the exclusion of arbitration from the text, it does not suggest that the exclusion was deliberate, and goes on to refer to the fact that many G A T T parties had already included ad hoc arbitration in bilateral agreements. Supra note 15. The fact that some G A T T members felt that arbitration was always available notwithstanding its exclusion from the text may suggest that the drafting parties did not feel it was necessary to include any explicit reference to mutually agreed arbitration. 61 Despite the uncertain and fragile basis for the legal order in Article X X I I I , the panel system was used frequently and with great success in the first 12 years. Even with the absence of details in Article X X I I I , the panel system of adjudication had developed as a 178 practice by 1955. Article X X I I I provided no specific rules entitling a Contracting Party to a panel, nor did it provide any guidance as to the adoption of panel decisions by the Contracting Parties. B y practice, panel decisions would only be adopted by consensus. This left room for any Contracting Party, particularly the losing party in the dispute, to block the adoption of the panel report. Nevertheless, in the period between 1948 and 1959, approximately 40 complaints were made within the dispute settlement provisions, and approximately 80 per cent of the complaints were successfully resolved. 1 7 9 A l l of this was achieved despite the fact that no complaint was ever referred to the International Court of Justice for an opinion. The first 12 years thus laid the foundation for G A T T as an independent quasi-institution based primarily on a political or diplomatic framework. It maintained its own self-standing dispute settlement apparatus within a structure that was based primarily on customary practice. From these early successes resulting in the rudimentary rule formation in G A T T , the development of dispute settlement was very much akin to a pendulum swinging back and 177 Robert E. Hudec, Adjudication of International Trade Disputes. (Trade Policy Research Centre, London: Ditchling Press Limited 1978) at 8 [Hudec, "Adjudication of International Trade Disputes"]. 178 The first dispute referred to a Panel of experts was a dispute between Norway and Germany in 1952, however it became a practice by 1955. See Terence P. Stewart, (ed.) The GATT Uruguay Round: A Negotiating History (1986-1992) (Volume II: A Commentary) (Deventer & Boston: Kluwer Law and Taxation Publishers, 1993) at 2676; Hudec, "Adjudication of International Trade Disputes", supra note 177 at 8; Jackson, John H. The World Trading System: Law and Policy of International Economic Relation.(Cambridge, Massachusetts & London, England: The MIT Press, 1989) at 95 [Jackson, "The World Trading System "]. 179 Hudec, Adjudication of International Trade Disputes", supra note 177 at 8-9. 62 forth. Indeed, the pendulum swung back in the next fifteen years, which proved to be an 180 unproductive period of decline for G A T T . Far fewer complaints were filed, most of 181 which were initiated by the U .S . Only ten were filed in the period between 1960 and 182 1969. The infrequency of recourse to the panel system has been attributed to the loss 183 of the confidence in the system. Indeed, by the 1970's, it was clear that there was changing landscape for several reasons. The breakdown in the consensus was tempered by a change in the overall balance of economic power. The development of the European Economic Community, (the " E E C " ) and the emergence of Japan created two new 184 economic powers as a counterbalance to U . S . hegemonic power. A period of protectionism set in, characterized by the development of voting blocs within the expanding membership of G A T T , and the proliferation of new tariffs and non-tariff barriers. 1 8 5 Tensions mounted, and reference to the dispute system was perceived as hostile. The disputes were often lengthened considerably by disputes over procedure. The need for changes set the agenda for the Tokyo Round of negotiations commencing in 1973. The objectives of the Tokyo Round were to develop clearer rules on the disputes, and marked the first time that dispute settlement was the subject of formal negotiations. 1 8 6 In essence, the Understanding Regarding Notification Consultation, Dispute Settlement and Surveillance finalized by the Contracting Parties in 1979 (the 180 Porges, supra note 27 at 149. Porges refers to a 17 year period of decline. 181 Hudec, "Adjudication of International Trade Disputes", supra note 177 at 13. 182 Stewart, supra note 178 at 2679. 183 Hudec, "Adjudication of International Trade Disputes", supra note 177 at 13; Stewart, supra note 178 at 2680. 184 Hudec, "Adjudication of International Trade Disputes", supra note 177 at 22; van den Broek, supra note 70 at 131. 185 HallstrOm, supra note 167 at 37; Stewart, supra note 178 at 2680. 186 Porges, supra note 27 at 150. 63 "Tokyo Understanding") also codified the customary practices relating to the panel process that had developed over the previous 25 years. A t this point the resolution of the individual dispute was the primary objective over the development of enduring principles. This bias towards the resolution of individual disputes can be seen in the dispute settlement system. Separate dispute settlement processes were developed within nine different areas, or "codes", creating a patchwork of processes for different areas of trade. The Tokyo Understanding also emphasized the conciliatory role of G A T T as well as the resort to arbitration and to the panel system. It addressed the use of non tariff barriers, the improvement of the panel process and the process of surveillance of the implementation of panel reports. It clarified that recommendations were to be implemented, preferably through the withdrawal of measures; then, i f not forthcoming, through the provision of compensation, and as a last resort, through retaliation. While the legal status of the Tokyo Understanding was uncertain, the crystallization of the framework now formalized in a written text seemed to increase the confidence of the Contracting Parties in the panel system. Recourse to panels increased after the Tokyo 187 Understanding. In 1982, the Contracting Parties issued a Ministerial Declaration stating that while there was no need for major changes from the Tokyo Understanding, improvements were needed in the areas of consultations, the establishment of panels, the • • 188 procedural framework of the panel process and the enforcement of panel decisions. The declaration also affirmed the role of the G A T T Secretariat in assisting panels with 187 Though the frequency may have equally represented the increased tension, the fact that parties increasingly sought fit to address that tension by referring complaints to the panel system surely suggests that the process presented a viable option. 188 G A T T , Ministerial Declaration, Contracting Parties 38,h Session, G A T T Doc. No. L/5424, (29 November 1982) at 6-8. 64 189 the "legal, historical and procedural matters dealt with". In the eyes of some observers, G A T T ' s transformation to legalism can be traced back to this period of time - this change in direction laid the foundation for the eventual establishment of the legalistic dispute settlement system within the institutional framework of the W T O . 1 9 0 The transformation of G A T T in the mid-eighties is undeniable. Between 1985 and 1989, 69 complaints were filed. Once again, approximately 80 per cent of the cases were successfully resolved. 1 9 1 However, G A T T ' s improvement was not limited to the quantity of complaints made or resolved - the improvement transcended to the quality of the decision-making. Robert Hudec has noted that in this period, G A T T produced higher 192 quality decision makers, and dealt with much more difficult or sensitive issues. The success of G A T T in this period of time can likely be attributed to the increased involvement of the G A T T Secretariat contemplated in the 1982 Ministerial Declaration. The involvement of officers of G A T T ' s Secretariat as independent legal researchers and advisors led to a more legalistic approach to panel decision-making, resulting in a more 193 cohesive body of law and analysis of the obligations. G A T T had therefore shifted from a diplomatic institution to a rule oriented, legalistic system where procedures were more mibid. at 7. 190 Robert E . Hudec, "The New W T O Dispute Settlement Procedure: An Overview of the First Three Years", (1999) 8 Minn. J. Global Trade 1 at 4 [Hudec, "The New W T O Dispute Settlement Procedure"]; Jackson, "The World Trading System ", supra note 178 at 98, who refers to the 1982 ministerial meeting as resulting in considerable stress in the direction of "rule integrity". 191 Hudec, "The New W T O Dispute Settlement Procedure", supra note 190 at 8-9. 192 Hudec, "EnforcingInternational Trade Law", supra note 21 at 199 & 206-208. 193 Hallstrom, supra note 167 at 19-20. 65 clearly defined, and where panels sought to interpret the obligations under the G A T T with more precision and consistency. However, another noticeable development during this period was the practical limitations of G A T T . While the legal framework had become more cohesive, the political commitment of the Contracting Parties in the face of the consequences of this renewed legalism was more doubtful. 1 9 4 In the period of 1985 and 1986, five of six panel reports were blocked from adoption. 1 9 5 The disputes typically involved the U .S . and the E . C , and covered more sensitive issues such as agricultural products. Retaliation proved to be very limited as a coercive force towards compliance. B y 1988, there had been only one case where the Contracting Parties used Article X X I I I to authorize the suspension of concessions as a result of a failure to implement panel recommendations, and the authorization was never acted upon. 1 9 6 3.3 The Beginnings of the Uruguay Round A t the time of the commencement of the Uruguay Round, G A T T still had a fairly good 197 compliance record. Nevertheless, with the growing sensitivity of disputes and the 194 Hudec, "EnforcingInternational Trade Law", supra note 21 at 201-203. 195 ibid. 196 Jackson, "The World Trading System ", supra note 178 at 96; Hudec, "New W T O Dispute Settlement Procedure", supra note 190 at 5. The Contracting Parties approved the suspension of concessions by Netherlands against the U.S. in its dispute over the barrier on importation of grain, however Netherlands never implemented any suspension. See also John Croome, Reshaping the World Trading System: A History of the Uruguay Round. (The Hague: Kluwer Law International, 1999) at 279. 197 By 1988, 233 complaints had been made, and 73 reports completed, most of which were adopted and implemented by the losing parties. See Jackson, "The World Trading System ", supra note 178 at 98-99. 66 increasing problem of non-compliance with panel recommendations, the Contracting Parties agreed to include dispute settlement in the new round of negotiations. The Uruguay Round was launched in Punta del Este in September of 1986. The Ministerial Declaration from Punta del Este included both general and specific objectives - to "improve and strengthen the rules and procedures of the dispute settlement process", and to develop arrangements for the "monitoring of the procedures that would facilitate 198 compliance with adopted recommendations". The Uruguay Round was an extremely ambitious undertaking. Not only did it involve newer obligations and commitments in contentious areas such as agriculture, it looked to expand into entirely different sectors such as services and intellectual property. Negotiations were broken down into fifteen separate negotiating groups, including the Negotiating Group on Dispute Settlement. Despite the breadth of areas of trade that were covered, the negotiations were conducted on a "single undertaking" basis - members were required to consent on all agreements simultaneously. 1 9 9 The negotiations were further complicated by the growth in membership. B y the beginning of the Uruguay Round, 86 states were members of G A T T , with membership steadily increasing throughout the round. 2 0 0 Given the unprecedented magnitude of these negotiations, it is 198 G A T T , Ministerial Declaration on the Uruguay Round, September 20, 1986, 33 GATT/BISD 19 (1987). 199 Peter Sutherland, "Concluding the Uruguay Round - Creating the New Architecture of Trade for the Global Economy" (2000-1) 24 Fordham Int'l L.J. 20. 200 Ibid, at 20. Membership started at 86 at beginning of the round, increased to 100 by the end of 1990, and then to 125 by the end of the Uruguay Round. Many of the states who joined during the Uruguay Round did so in anticipation of the creation of a new institutional body for which they could be founding members. 67 not surprising that the Uruguay Round eventually spanned eight years, despite the fact that the initial deadline for completing negotiations was the end of 1990. The Negotiating Group on Dispute Settlement (the "Negotiating Group") focused on the birth defects of the original G A T T dispute system: the need for stricter deadlines for the completion of the panel process; the potential of an appeal process for review panel decision; the need for a new system for the adoption to address the problem of blocking; the need for measures requiring implementation of reports after adoption and surveillance; and the need to clarify the roles of conciliation, arbitration and retaliation. The Negotiating Group also sought to create a unified system of dispute settlement for all areas of dispute, to reverse the fragmentation of the dispute settlement resulting from the Tokyo Round. In addition, several Contracting Parties made a push to bring these new legal rules within an institutional setting, a concept that was still met by much resistance from the U .S . The Round did ultimately achieve the important symbolic effect of clarifying, once and for all , the legal status of the formal decisions resulting from the panel process, but the road to that point was a long and rocky one. 3.4 The Initial Proposals and the Arbitration Alternative The tension between the political or diplomatic characteristics of G A T T and the legal aspects were apparent from the outset of the Uruguay Round. The negotiations relating to 201 John H. Jackson, The Uruguay Round, World Trading Organization and Problem of Regulating International Economic Behaviour. (Carleton University, Ottawa: Graphic Services, 1995) at 9 [Jackson, "The Uruguay Round"] 202 Ibid, at 9. 68 the dispute settlement mechanism revealed divergent perspectives in the philosophical approaches amongst the U.S. , Canada, the E . C . and Japan, the so called "quad members". While the U.S . and Canada viewed the objective of the dispute system as establishing the correct and objective legal interpretation of treaty obligations, the E . C . and Japan saw the system geared towards settlement, confidential negotiations, and obtaining solutions to 203 the specific disputes in question, irrespective of the specific legality of the obligations. While these divergent views of dispute settlement have often been associated with the political and the legal dichotomy, it is also a distinction between judicial settlement and arbitration. A s suggested earlier, the primary goal of arbitration is the settlement of the dispute as between the affected parties, and less so the creation of any form of legal precedent. Early in the Round, it became clear that the concerns of the U .S . would drive the direction of the discussions on the dispute settlement system. In mid-1987, the U .S . submitted a proposal for the improvement of the dispute settlement system. 2 0 4 The proposal included practical changes, such as tighter timeframes for resolution of disputes. Most importantly, it addressed the major problem of report-blocking, proposing that panel reports become binding automatically, and subject to retaliation measures i f the offending measures were not withdrawn and compensation not provided. The U .S . acknowledged there could be the problem of legally flawed reports, and suggested that reports be made binding unless there was a consensus against adoption. The U .S . alternatively proposed that adoption of the report exclude the parties involved in the 203 Mora, infra note 247, at 128-136; Young, supra note 43 at 389-390; Croome, supra note 196 at 125. 204 Improved Dispute Settlement: Elements for Consideration. Discussion Paper Prepared by United States Delegation, G A T T Doc. No. MTN.GNG/NG13/W/6 (25 June, 1987). 69 dispute process. The proposal concluded by suggesting that the dispute settlement system could only function successfully i f the Contracting Parties "come to view dispute settlement not as a contest of wills but as an essential element in the management of the 205 world trading systems". The possibility of arbitration as an alternative to a system of judicial settlement was presented early on during the Uruguay Round. The U . S . proposal suggested that the Contracting Parties should consider a system for binding arbitration as an alternative means of resolution that would co-exist with the panel system. The U .S . noted that arbitration was a "widespread and common form of dispute settlement in international trade" and could be used in lieu of the normal panel process in certain classes of disputes, 206 for simple issues that were taking "too long and becoming too political". The arbitration system proposed would exclude any approval process, but would require the consent of both parties to the dispute. A failure to implement recommendations would 207 automatically give the aggrieved party a right to compensation or retaliation. The proposal thus conceived of a two-tiered system providing more than one form of dispute settlement for different types of cases. A s wi l l be seen later, arbitration was conceived within subsequent proposals as a true alternative to other possible means of dispute settlement. 205 Ibid, at 6. 206 Ibid, at 2. 207 G A T T , Negotiating Group on Dispute Settlement, Summary and Comparative Analysis of Proposals for Negotiations, Note by the Secretariat, Revision G A T T Doc. No. MTN.GNG/NG13/W/14/Rev.l, (26 February 1988), at 30, para. 81. But see Hughes, supra note 176 at 77 where she describes the U.S. proposal as intending that a party would not be "compelled" to implement, leaving compensation or retaliation as options. 70 Other members however did not embrace the U.S preference for legalistic means of dispute settlement. In September of 1987, the E . C . submitted its own proposal, asserting its underlying philosophy that the primary goal of dispute resolution should be a negotiated settlement. The proposal further asserted that the legal aspects ought not 208 • become the "key element". While the proposal made many suggestions for improvement, it confirmed a fundamental disagreement on the principle of consensus and the adoption of panel reports. On the issue of panel reports, the E . C . recommended maintaining the consensus principle for both the adoption of the panel report, suggesting that the quality of panel reports should simply be improved. 2 0 9 The implementation of recommendations would be subject to time deadlines, after which the successful party would be entitled to compensation or retaliation. The E . C . proposal also grappled with the idea of arbitration having utility for specific disputes. The E . C . observed that a mandatory arbitration process did not require the approval of G A T T Council (the "Council"), and had always been available to the Contracting Parties. 2 1 0 Nevertheless, the E . C . favoured the institutionalization of the arbitration process. 2 1 1 The proposal noted however that it would be difficult to define the categories of disputes where arbitration should be prescribed in place of the panel process, but suggested that it should be limited to factual issues only and not to situations 212 involving questions for interpretation that could a constitute a legal precedent. Once 208 Communication From The EEC, G A T T Doc. No. MTN.GNG/NG13/W/12 (24 September 1987) at 2. 209 Ibid, at 4. 210 Ibid, at 3. 2\\ Ibid. at3. 212 lb id. at 3. 71 again, the issue of how to categorize disputes as more suitable for arbitration as opposed to the judicial settlement system was identified, though this did not lead to much discussion. Such a discussion may well have been useful, as both U.S and E . C . proposals suggested that arbitration would be useful in only limited situations, where the issues were relatively easy to settle or factual. A t this point, arbitration began to be considered as a possible solution for the shortcomings of the panel system. The exchange of proposals by the U . S . and the E . C demonstrated that the issues of adoption of panel reports and arbitration were linked. The U.S . position seemed to assume that any panel system might retain some inefficiency and thus strongly advocated for a more efficient alternative system of binding arbitration. The link between these issues was reinforced by a Swiss proposal made on September 18, 1987 that suggested an arbitration process in the event that the Council failed to adopt a panel report. 2 1 4 215 A letter proposal by Japan incorporating suggestions on arbitration demonstrated that it shared the recognition of the other quad members that an alternative process for binding dispute settlement for specific types of disputes could be useful, despite the difficulty of identifying the types of disputes that would be suitable for such a regime. The proposals 213 Hughes, supra 176 at 78. 214 Communication From Switzerland, G A T T Doc. No. MTN.GNG/NG13/W/8 (18 September, 1987) at 3. 215 Communication From Japan, G A T T Doc. No. MTN.GNG/NG13/W/21 (1 March, 1988) at 7-9. In March of 1988, Japan included the possibility of arbitration as an alternative in its supplementary proposal, conditional on the neutrality of the arbitral body and the consistency and transparency of the process. Notably, a third condition was that any decision must be discrete to the specific dispute, and binding only on the parties to this dispute, so that it would not establish any principle of general application or interpretation of G A T T . 72 of other Contracting Parties appeared to accept this, but raised a variety of concerns. It was suggested that the Council should necessarily play a role in defining the scope of the arbitration and approving the resolution, in order to ensure coherence with G A T T 216 217 principles, and to protect third party interests that might otherwise be affected. Other proposals suggested that the Council have the ability to monitor and reject the arbitration 218 based on the grounds defined by the parties. The next stage of discussions revealed two things - first a commitment to exploring arbitration within an institutional context, and second, an uncertainty as to how arbitration was distinguishable from the panel system, and thus about its place within G A T T . In fact the members of the Negotiating Group themselves appear to have diverging perspectives on the basic concept of arbitration within the G A T T , as there was almost no point of reference. Despite the E . C . contention that arbitration had always been available under G A T T , it had been rarely used. In 1963, the Netherlands and the U .S . agreed to refer a dispute to arbitration. The terms of reference required the tribunal to determine the value of certain U .S . exports of poultry to Germany. A s a precondition to the arbitration, both parties agreed to be bound by the decision, and the award was indeed implemented. 2 1 9 This was a rare example of the use of arbitration in a G A T T dispute. 2 2 0 216 See Communication From The Nordic Countries, G A T T Doc. No. M T N . G N G / N G 13/W/l0 (18 September, 1987) at 2. The communication was submitted by Norway on behalf of the Nordic Countries. 217 Communication From Canada, G A T T Doc. No. M T N . G N G / N G 13/W/l 3 (24 September, 1987) at 10; Communication From Korea, G A T T Doc. No. MTN.GNG/NG13/W/l9 (20 November, 1987) at 3. 218 Communication From Switzerland, G A T T Doc. No. M T N . G N G / N G 13/W/8 (18 September, 1987) at 3-4; Communication From Canada, supra note 217, at 10. 219 The terms of reference of the "Chicken Wars" arbitration are described by the Secretariat in its report G A T T , Negotiating Group on Dispute Settlement, Concepts, Forms and Effects of Arbitration, Note by the Secretariat, G A T T Doc. No. M T N . G N G / N G 13/W/20, (22 February, 1988) at 9. See also Mora, infra note 247 at 140. 73 Despite all o f the uncertainty surrounding the issue, the Negotiating Group was clearly intrigued about the arbitration alternative. In early 1988, the Negotiating Group requested the Secretariat to prepare a background paper on the concepts, forms and effect of 221 arbitration. The paper attempted a few tasks: to clarify the different concepts, forms and features of arbitration; to outline issues that should be addressed in the context of the 222 negotiations; and to summarize the proposals to date on the use of arbitration. The report started by reinforcing the two main branches of G A T T , the legal means of 223 dispute settlement and the diplomatic means, the latter characterized by the flexibility of procedures, the control over disputes and avoidance of binding decisions. While the draft identified arbitration as a form of legal means, it distinguished it from judicial settlement by referring to the ability to choose the decision-maker, to define the scope of the dispute and the jurisdiction of the tribunal, and to choose the applicable rules and procedures. 2 2 4 Arbitration thus shared some of the flexibility and benefits of the diplomatic means of dispute settlement. The report of the Secretariat did not entirely accept that arbitration was indeed properly 225 available under the previous G A T T regime. In particular, the report suggested that the 220 However, a few Contracting Parties had agreed to refer their G A T T disputes to arbitration pursuant to clauses under their bilateral agreements. See Hallstrom, supra note 167 at 15. 221 G A T T , Concepts, Forms and Effects of Arbitration, Note by the Secretariat, G A T T Doc. No. M T N . G N G / N G 13/W/20, (22 February, 1988). 222 Ibid, at 1. 223 Ibid, at 3. 224 Ibid, at 3. 225 Ibid. See paragraph 21 of the Report where the Secretariat writes "The need for additional G A T T rules on arbitration depends on the prevailing views on the availability of arbitration under the current G A T T rules." 74 Negotiating Group consider whether or not an arbitral award binding on the parties would require the approval of the Council . The Secretariat also set out two questions that went to the heart of the future of arbitration in the dispute settlement process of G A T T and any superseding institution: first, it asked whether or not the Contracting Parties should recognize a right to request an ad hoc panel with an arbitration mandate; second, whether or not there was a need for additional G A T T rules recognizing the availability and 226 promoting the use of arbitration amongst the Contracting Parties. A s can be seen later, given the inclusion of certain provisions in the D S U , it would appear that the answer of the Negotiating Group to both questions was "yes". While the utility of arbitration as an alternative gained momentum, the shape and scope of the arbitration option remained unclear. Nevertheless, the option for arbitration found its place in the Negotiating Group's draft of December 9, 1988 completed shortly after 227 • the mid-term review meeting in Montreal in December of 1988. The three articles on arbitration in the draft demonstrated a broadly conceived notion of the utility of arbitration within the G A T T . The first article confirmed the utility of "expeditious arbitration" as an alternative means 228 of dispute settlement for issues that were "clearly defined". The second article left the specifics of arbitration largely up to the disputing parties, establishing that resort to arbitration would be based only on the mutual agreement of the parties, and not the 226 Ibid, at 10. 227 The draft was put on hold until the meeting in April 1989, although the text in the arbitration provisions that resulted from the Ministerial meeting was the same. G A T T , Trade Negotiations Committee, Meeting at Ministerial Level, G A T T Doc. No. MTN.TNC/7(MIN), (9 December, 1988). 228 Ibid, at 28, para. E l . 75 approval of the Council . The second article also required notification of the arbitration to all Contracting Parties, thus providing any third party rights with a chance to seek participation rights. Most importantly, the third article required that the parties agree to abide by the arbitration award. The draft made no mention of many other aspects that had been previously discussed: an arbitration clause; a classification of disputes eligible for arbitration, the application of G A T T principles or law, or the Council 's involvement in the process. The broad wording of the draft appeared to allow for the determination of any issue, including an issue of law or interpretation of the provisions of G A T T . This represented a much broader concept of the utility of arbitration than that envisioned by the initial proposals of the U .S . and the E . C . The draft was comparatively more problematic with respect to the judicial settlement process. The draft still did not confirm a right to establish a panel, and merely left the 229 decision to the Council upon a party's request for a panel. Another controversial feature of the draft was the preservation of the practice of adoption of panel reports by a consensus decision that included the losing party. The only proposed solution for the problem of report blocking was found in a portion of the text that stated that delays in the 230 dispute settlement process were to "be avoided". 229 Ibid, at 32. 230 Ibid, at 32, para. G.3; Croome, supra note 196, at 129. 76 3.5 The Impact of U.S. Unilateralism on Uruguay Round Negotiations While the draft presented at Montreal left much unanswered in respect of the panel system, the next stage of negotiations became much more focused on judicial settlement, to the detriment of the development of the arbitration alternative. The absence of an agreement on the consensus issue was a focal point at the mid-term review meeting in Montreal. The draft reflected the fact that the majority of members sought to retain the consensus requirement, including the ability to block adoption of reports and retaliation measures. The U.S disaffection with those proposals was a factor in propelling the development of its infamous domestic 301 laws in the period leading up to the mid-term review in Montreal. These laws originated with the U.S. Trade Act of 1974,231 which was designed to provide some form of diplomatic protection to citizens who were unprotected under international 232 * law from the effect of unfair or distortive trade practices. The Act gave a right of redress to U . S . citizens, as any complaints would require an investigation. If the complaint was shown to be valid, the legislation required government action, including the possibility of retaliation measures against the offending state. The 301 laws permitted complaints to be referred to G A T T i f appropriate, but also allowed the executive branch to act unilaterally i f necessary. The 301 legislation therefore purported to permit the U .S . to impose a system of unilateral reprisal in response to international trade law violations. 231 19U.S.C. § 2 4 1 1 (1990), as amended. 232 Hudec, "Enforcing International Trade Law, supra note 21. 77 The U .S . justified the 301 law by noting that the G A T T process was slow and ineffective. 2 3 3 Prior to the Montreal mid-term, the U.S . enacted the Omnibus Trade and Competition Act of 1988, which was designed to enhance the 301 laws. The amendment delegated the power to enforce the 301 laws from the executive down to the U.S . trade representative, • * • 235 removing the executive's discretion and making the process mandatory. The new "Super 301" law was also more proactive. It required the government to conduct a survey of restrictions on a world-wide scale to identify trade distorting practices and create a list of the worst trade offenders for investigation. It required the government to implement 236 unilateral action against those offenders so identified. The amendment also added a "Special 301" law, which extended the responsibilities of the U . S . government into the 237 area of intellectual property. The nature of the U.S . legislative changes became a true challenge to the direction of the negotiations on dispute settlement. The 301 laws were particularly antagonistic to the principles of G A T T , as the legislation adopted its own definition of unreasonable trade 233 Hudec, "Enforcing International Trade Law, supra note 21 at 193. 234 19 U.S.C. § 2901 (2000). 235 Jean Heilman Grier, "Section 301 of the 1974 Trade Act", online: United States of America Department of Commerce <http://www.osec.doc.gov/ogc/occic/301 .html >: Stewart, supra note 178 at 2760. 236 Jean Heilman Grier, "Section 301 of the 1974 Trade Act", online: United States of America Department of Commerce <http://www.osec.doc.gov/ogc/occic/301.html >; Stewart, supra note 178 at 2760. 237 Jean Heilman Grier, "Section 301 of the 1974 Trade Act", online: United States of America Department of Commerce <http://www.osec.doc.gov/ogc/occic/301 .html >; Stewart, supra note 178 at 2761. 78 practices, and was not bound to the concepts in international law. Further, the legislation made retaliatory responses mandatory in certain cases, and indeed resulted in 239 retaliatory action by the U . S . in several instances. While the 301 laws unquestionably represented a form of unilateralism and undermined the principles of G A T T , the U.S justified the measures on the basis that the practice of blocking G A T T reports was itself a form of unilateralism. 2 4 0 The U . S . suggested that it would not be required to resort to its 301 laws as long as the G A T T dealt more effectively with disputes. 2 4 1 Its status as the world hegemonic power can therefore be seen as a major influence on the shaping of the D S B . With the new 301 amendments aggressively challenging a rules-based multilateral trading system, the U .S . actions became a focus for discussion at the Montreal meeting. A t a G A T T Council meeting held a few months in advance of the Montreal meeting, the E . C . expressed its "grave concern" with the laws, while several other members including Japan expressed their own dissatisfaction. 2 4 2 A t this point, the danger of U .S . unilateralism was taken much more seriously by the G A T T Contracting Parties, 2 4 3 and the prohibition of unilateral action became a priority for the negotiations. 2 4 4 The E . C . in particular viewed the ongoing prospect of unilateral action as a threat, and was concerned 238 Hudec, "Enforcing International Trade Law, supra note 21. 239 Jackson, "The World Trading System", supra note 178 at 106. 240 Croome, supra note 196 at 228. 241 Croome, supra note 196 at 225; Stewart, supra note 178 at 2762. 242 G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting (held on 22 September, 1988), G A T T Doc. No. C/M/224, (17 October, 1988) at 28-34. See also Croome, supra note 196 at 225. 243 Hudec, "EnforcingInternational Trade.Law", supra note 21 at 234. 244 Croome, supra note 196 at 225. 79 that such unilateral action could risk the disintegration of the G A T T system. While the U.S . maintained its position on the 301 laws, it was also resistant to the idea of the creation of a formal institution to oversee the rules-based trading system. 3.6 Proposals on Arbitration, Adoption and Appellate Review The next stage of negotiations demonstrated the linkage between the Negotiating Group's consideration of arbitration, and the introduction of the concept of an appeal mechanism. The option for arbitration and the treatment of the adoption process were replicated in the negotiations party's Improvements Decision of Apr i l 12, 1989 ("Improvements of 1989"). 2 4 6 This was the first attempt at regulating the resort to binding arbitration within G A T T . 2 4 7 It has been suggested that the purpose was to include a form of resolution that was more adjudicative than what already existed, to resolve clear-cut, non-political issues. 2 4 8 Indeed, this theory may find some support in the fact that it was the U.S . proposal of 1987 that raised the idea of arbitration early in the Uruguay Round. However, as w i l l be seen later, arbitration was eventually included in the final act despite the development of a much stronger adjudication system in the D S U . Despite the discussion of the use of arbitration for only particular types of disputes, the form of arbitration introduced on a trial basis, pending the completion of the Round, was 245 Porges, supra note 27 at 153-154; Hudec, "EnforcingInternational Trade Law", supra note 21 at 228. 246 G A T T , Negotiating Group on Dispute Settlement, Improvements to the GATT Dispute Settlement Rules and Procedures, (Decision of 12 April 1989), G A T T Doc. No. L/6489. Known as the "Improvements of 1989". 247 Miquel Montana I. Mora, " G A T T with Teeth: Law Wins over Politics in the Resolution of International Trade Disputes" (1993-1994) 31 Colum. J. of Transnat'l L. 103 at 139. 248 Ibid, at 139. Mora also suggests that provisions for arbitration were specifically included despite its previous availability under the G A T T scheme, in order to encourage its use. 80 relatively open-ended. The Improvements of 1989 reproduced the three paragraphs on arbitration in the draft approved in Montreal: the first paragraph set out arbitration as an alternative process for any issues that were "clearly defined"; the second confirmed that arbitration,, and the procedure to be followed, were subject to the mutual agreement of the disputing parties; and the third provided that the disputing parties alone would agree to be bound by the award, and that third parties were permitted to participate, but only with the consent of the parties. The notion of broad based use of arbitration was apparent in the text. The use of arbitration was not restricted to factual disputes, but could indeed be used for the interpretation of legal principles. The broad conception of arbitration was not only apparent from the text, but also from the context of negotiations leading up to the final version. The language adopted during the Ministerial meeting in Montreal, and formalized in the Improvements o f 1989, was in fact very similar to the wording and conceptual underpinnings of a proposal submitted by Mexico in June of 1988, and revised in October 198 8. 2 4 9 The one key difference was that the Mexican proposal contained a fourth paragraph that provided " I f arbitration works 250 and proves useful in practice, its use could subsequently be extended". These words were obviously not intended for inclusion in a formal treaty, and indeed, they were not incorporated into the Improvements of 1989. Nevertheless, the fact that the text drew heavily from the Mexican proposal suggested that the Contracting Parties had indeed contemplated the possibility of expanding the use of arbitration, i f it proved successful. 249 Communication From Mexico, G A T T Doc. No. MTN.GNG/NG13/W/26 (23 June, 1988), revised in Communication From Mexico, G A T T Doc. No. MTN.GNG/NG13/W/26/Rev.l (10 October, 1988). 250 ibid. 81 The progress with respect to the judicial settlement system was comparatively slower up to Apr i l of 1989. While the Improvements of 1989 specifically maintained the practice of adopting panel reports by consensus, they also, for the first time, ensured the creation 252 of panel as a matter of right by eliminating the right to veto, a monumental change. Nevertheless, the draft lacked two key elements that would become priorities for negotiation: a change in the consensus process for adoption that preoccupied the U.S . ; and recognition of an appellate review process. In addition, the Improvements of 1989 did not address implementation of reports in great detail. It provided neither a specific timeframe within which a losing party would be required to comply with a panel recommendation, nor any provision on retaliation. Despite the fact that the Improvements of 1989 did not include an appellate procedure, the draft undoubtedly had a major effect on the ultimate creation of a judicial settlement system. The research of March Busch and Eric Reinhardt has found that the U.S relied more heavily on G A T T / W T O dispute resolution as opposed to unilateralism after 1989. 2 5 3 Further, the formalization of the arbitration alternative also correlated with the creation of a right to a panel, which disputing parties were undoubtedly more wil l ing to utilize, given the development of that system within G A T T . While the arbitration alternative still required consent of the parties like the previous panel system, the panel system had opened up with the Improvements of 1989. It has been suggested that the inclusion of arbitration took a step towards support for a formal system of adjudication 251 Ibid, at 4, para. G l . 252 Posner & Yoo, supra note 8 at 45. 253 Busch & Reinhardt, "Testing International Trade Law", supra note 21 at 475. 82 for dispute resolution. The arbitration provisions in the text may have indeed acted as a catalyst for the development of a more judicial system. Ironically, i f this was the effect, it was at the expense of the development of arbitration as a truly independent form of dispute resolution. 3.7 Arbitration Proposals Stall The next stage of negotiations demonstrated a shift in focus by the Negotiating Group from arbitration to a legalized process of judicial settlement that included an appeal process. After this point, the proposals developing the concept of arbitration stalled completely. After the Improvements of 1989, the only significant development in the arbitration proposal during the spring and summer of 1989 was a Swiss proposal that sought to strengthen the arbitration provisions. The objective of the proposal was to ensure that arbitration procedures became "a harmonious part o f G A T T ' s multilateral 255 framework without calling into question the principle that a final award is binding". The Swiss proposal alerted the Negotiating Group that the Improvements of 1989 only had the barebones of an arbitration system, and that the system needed to be "moulded to 256 the multilateral content of the G A T T dispute settlement". Indeed, the content of the draft itself supported the Swiss perspective - it was a proposal that had not matured through the negotiation process of the Uruguay Round. 254 Young, supra note 43 at 398. 255 Communication From Switzerland, G A T T Doc. No. M T N . G N G / N G 13/W/33 (19 July, 1989). 256 G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting (held on 28 September, 1989): G A T T Doc. No. MTN.GNG/NG13/16 (13 November, 1989) at 7. 83 The reaction of the Negotiating Group to the Swiss proposal confirmed that arbitration was no longer a focus of negotiations, given the other major issues such as panel procedure, adoption and appellate review. Certain delegates questioned the need to elaborate on the process of arbitration, while others expressed the concern that the proposal may "inappropriately encourage resort to arbitration at the expense of the more 257 traditional forms of dispute settlement". Over the next several meetings held in the period between September 1989 and September 1990, little else was said about arbitration as an independent form of dispute settlement. A t the February 1990 meeting, one delegate advocated a wait and see approach, suggesting that the use of arbitration 258 * • should be considered by its degree of effectiveness. Meanwhile, the arbitration 259 alternative was put to little use. The attention shift of the negotiating parties was apparent in the meeting in A p r i l of 1990. At this meeting, both the U .S . and the E . C . presented their respective proposals for change. Each addressed a form of appellate review, and the use of compensation and retaliation. 2 6 0 The content of the E . C . proposal highlighted the major tension in the negotiations - the commitment against unilateral measures. In its proposal, the E . C . indicated that an "unequivocal and irreversible" commitment against the use of unilateral 257 ibid. 258 G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting (held on 7 February, 1990): Note by the Secretariat, G A T T Doc. No. M T N . G N G / N G 13/18 (21 March, 1990) at 3. There were also further discussions of conflicts that would affect the binding nature of arbitration awards, see Stewart, supra note 178 at 2773. 259 Hughes, supra note 176 at 79. 260 Statement by the Spokesman of the European Community at the Meeting on 5-6 April, G A T T Doc. No. M T N . G N G / N G 13/W/39 (5 April, 1990); Communication From the U.S., G A T T Doc. No. M T N . G N G / N G 13/W/40 (6 April, 1990). 84 measures was a key precondition to strengthening the dispute settlement, an obvious reference to the U . S . position. The E .C. ' s insistence on a commitment against unilateral 262 retaliation outside of G A T T was supported by many of the delegates. It was thus clear that the focus was now on the strengthening of the judicial settlement system in exchange for a commitment from the U.S . not to resort to its 301 laws. A s one delegate suggested, it was also apparent that the creation of an appellate review producing binding decisions could affect the Contracting Parties view's on the utility of an alternative arbitration 263 system. It was equally apparent that between Apr i l 1989 and September of 1990, the arbitration option was simply not a major source of discussion, despite the fact that it had not been particularly well defined in the Improvements of 1989. Rather, the focus of the group became the strengthening of this system through a new adoption process and the creation of an appellate review system, in order to appease the concerns of the U .S . In September 1990, the Negotiating Group instructed the Secretariat to summarize the discussions regarding improvements in a memorandum, in preparation for the Ministerial meeting in Brussels in December of 1990. 2 6 4 After making informal revisions to the draft, the Chairman of the Negotiating Group would submit the "Chairman's Text on Dispute 261 Statement by the Spokesman of the European Community at the Meeting on 5-6 April, supra note 260 at 1. 262 G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting (held on 5 April, 1990): Note by the Secretariat, G A T T Doc. No. M T N . G N G / N G 13/19 (28 May, 1990) at 3. 263 Ibid, at 5. 264 G A T T , Negotiating Group on Dispute Settlement, Draft Text on Dispute Settlement, G A T T Doc. No. M T N . G N G / N G 13/W/45 (21 September, 1989). 85 Settlement" (the "Chairman's Text") to Arthur Dunkel, Chairman of the Trade 266 Negotiating Committee and Director General of G A T T . If approved by the Trade Negotiating Committee, the Chairman's Text was to be incorporated into an omnibus draft covering all o f the round's negotiations, to be presented at the Brussels meeting. Although the Secretariat memorandum contained several suggestions for the improvement on the use of arbitration, most sought only to clarify the G A T T Council 's role in the process. However, there is some indication of a desire to maintain some distance between the arbitration alternative and the judicial settlement system. One recommendation was that the arbitration body was to follow standard panel procedure 267 unless it otherwise decided. This suggestion was not incorporated into the Chairman's Text, which preserved a completely open-ended procedure for arbitration. The only guideposts on procedure for arbitration remained whatever was mutually agreed upon by the disputing parties. A t the same time, the notion of the integration of arbitration into an institutional context took on a newer direction. The Secretariat memorandum also referred to two separate and specific forms of arbitration as a last resort for resolving disputes relating to the implementation of rulings. First, it provided for arbitration regarding the "reasonable period of time" within which the losing party was required to implement the 265 G A T T , Chairman's Text on Dispute Settlement, G A T T Doc. No. CGT/807-14 NG13 (19 October, 1989) ("Chairman's Text"). 266 G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting for 24 September to 11 October 1989, Note by the Secretariat, G A T T Doc. No. M T N . G N G / N G 13/23, (24 October, 1989). The covering letter to the Chairman of the Negotiating Group clarified that the document was a Chairman's text because it did not necessarily represent the consensus of the Negotiating Group. 267 Supra, note 264 at 9. 86 recommendations or ruling. Second, in cases where retaliation was proposed in response to non-compliance, it set out an arbitration process to determine i f the level of the proposed retaliation was excessive. 2 6 8 The Secretariat also summarized the proposal to add the concepts of compensation and retaliation as measures available to encourage prompt implementation. These proposals were eventually incorporated into the Chairman's Text, though as it w i l l be discussed in the next chapter, these forms do not represent a form of alternative within the D S U , and it is perhaps a misnomer to refer to them as arbitration at all . Otherwise the main arbitration provisions in the Chairman's Text remained mostly unchanged from the Apr i l 1989 draft. The Chairman's Text followed the wording from the Negotiating Group draft and added a requirement for arbitration awards to be "notified to the Council where any Contracting Party may raise any point relating thereto". 2 6 9 In considering the importance of reputation as a coercive force towards compliance, this was an important development for the process of arbitration. The requirement would mean that this dispute settlement mechanism would engage the element of reputation as much as any process of judicial settlement. The Chairman's Text also added wording to clarify that implementation and surveillance provisions for panel and appellate review reports would apply mutatis mutandis to arbitration awards. 268 These were in fact proposals made by Canada 3 months earlier. After the Montreal Meeting, the Negotiating Group on Dispute Settlement met ten times after the April 1989 draft, up to October 1990. Though a similar proposal was made by Mexico in the same timeframe, it did not recommend arbitration in these situations. Canada's suggestions for arbitration with respect to the reasonable timeframe and the level of retaliation were eventually incorporated into the Secretariat's memorandum, and then into the draft presented in Brussels. See Communication From Canada, G A T T Doc. No. M T N . G N G / N G 13/W/41 (28 June, 1990) at 6 & 7. 269 Chairman's Text, supra note 265 at 16, para. E.3. 87 The amended text of this draft marked a turning point in the Uruguay Round negotiations, where efforts were redirected towards the judicial settlement system. While the arbitration provisions stayed virtually the same, the Secretariat's memorandum contained two monumental changes to the panel system from the previous year: first, the creation of formal appellate review process; and second, four separate options for the adoption of panel and appellate review decisions, only one of which maintained the traditional consensus approach. On the issue of adoption, the Chairman's Text specifically deleted the portion of the Improvements of 1989 that provided for the maintenance of the consensus rule for adoption. This language was surrounded by square brackets, indicating that it had not been finally resolved. The new provisional language thus proposed that panel reports: shall be adopted at a council meeting unless one of the parties formally notifies the Council of its decision to appeal or the Council decides [otherwise] [by 270 consensus] [not to adopt the report]. (Bold added) The provision for the adoption of appeal decisions was worded similarly: An appellate report shall be adopted at a council meeting and unconditionally accepted by the parties to the dispute unless the Council decides [otherwise] [by consensus] [not to adopt the appellate report].271 (Bold added) 270 Ibid, at 9. 271 Ibid, at 11. 88 While the draft language on adoption and automaticity had been "square-bracketed", the disposal of the wording maintaining the consensus rule coupled with the updated wording clearly favoured a reverse consensus rule. This sudden reversal on the issue of automaticity did not represent the political will of the majority, but was rather the product 272 of the pressure by the U.S. and its 301 laws. The Chairman's Text was essentially approved in the same form at the Trade Negotiating Committee meeting of November 26, 1990. It was then incorporated into the four 273 hundred page omnibus Draft Final Act presented for consideration at the Ministerial meeting in Brussels. Although, the Draft Final Act was never approved at the Brussels meeting as a result of a stalemate regarding the agricultural issues,274 the dispute settlement provisions were reproduced in another draft in December of 1991. The Understanding on Rules and Procedure Governing the Settlement of Disputes that 275 formed part of the package signed in Marrakech in April of 1994, essentially replicated 276 the text that had been agreed to in December of 1991 with only slight modifications. 3.8 Arbitration Under the DSU The Uruguay Round had thus succeeded in creating a new dispute settlement process. Despite the presence of the judicial settlement system as the primary method of dispute 272 Hudec, "New W T O Dispute Settlement Procedure", supra note 190 at 14. 273 Draft Final Act Embodying The Results of the Uruguay Round of Multilateral Trade Negotiations, G A T T Doc. No. MTN.TNC/W/35/Rev.l , (3 December, 1990). 274 Stewart, supra note 178 at 2786. 275 Annex 2 to Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations and Marrakech Agreement Establishing the World Trade Organization. 276 Young, supra note 43 at 399. 89 settlement, three separate procedures that are referred to as "arbitration" survived into the 277 D S U . The two that were introduced after the Improvements of 1989 however are really only adjuncts to the panel process regarding implementation issues. First, Article 21.3 of the D S U provides for arbitration to determine the "reasonable" period of time to comply with a panel or Appellate ruling, failing other methods. Second, Article 22.6 of the D S U provides for an arbitration regarding the level of concessions that can be suspended by an aggrieved state should the offending state fail to comply or provide compensation. However, the last example, Article 25 of the D S U , provides a much broader form of arbitration. These provisions w i l l be discussed in more detail in the next chapter. 3.9 Lessons From the History of GATT and the Uruguay Round There are several points that can be drawn from the historical review of the discourse regarding the concept of arbitration within the G A T T and W T O systems. First, there has long been a widely shared belief in the utility of arbitration within the multilateral trade system. The option of arbitration in Article 93 of the original Havana Charter eventually found a new form in the D S U . A s has been seen, it is arguable that the ability to refer to arbitration within the G A T T system has always existed. While Article 25 was specifically added to the D S U , this may have been rendered necessary by the provisions precluding 278 use of dispute settlement processes outside of the D S U . 277 There are still other forms of arbitrations provided for under other W T O agreements. Supra note 29. 278 Young, supra note 43 at 401. 90 Second, arbitration has mostly been conceived as an alternative to, and not as a complete replacement for, judicial settlement within GATT. Invariably, the proposals made within the Negotiating Group during the Uruguay Round contemplated a two tiered system where arbitration could be resorted to as an alternative form of dispute settlement. A variety of members have expounded on the virtue of arbitration as an alternative form of dispute settlement. It has been described as: "an effective means of resolving disputes" 279 and a "useful tool in trade policy"; as an "instrument properly adjusted to G A T T working parties could - in clearly defined cases - be available to parties in a dispute"; 2 8 0 and as a system that "might facilitate resolution of certain disputes basically of a factual . , , 281 nature . One member of the Negotiating Group seemed to have captured the idea that the dispute settlement system should not seek to impose a "one-size-fits-all" approach to disputes: Experience shows that disputes brought before the G A T T are multifarious and often comprise trade problems that have not been fully addressed in the past for which no precedents exist. The dispute settlement system in G A T T should therefore be designed so as to respond adequately to the different nature of dispute cases. This suggests that the parties to a dispute should have the choice between a number of alternative and/or complementary techniques and 282 mechanisms. Third, there is still room for, and at least a historical recognition of, the utility of arbitration within the current system. If it has any future in the WTO, it is of course 279 Communication From Canada, supra note 217 at 10. 280 Communication From The Nordic Countries, supra note 216 at 2. 281 Communication From Korea, supra note 217 at 3. 282 Communication From The Nordic Countries, supra note 216 at 1. 91 important that there is a political w i l l to retain arbitration as an alternative form of dispute resolution. Although the Uruguay Round resulted in the creation o f an effective judicial settlement system that includes a form of legalized appellate review, stronger implementation measures and a reverse consensus rule, the arbitration alternative conceived in the Improvements of 1989 nevertheless survived into the text of the D S U . This strongly suggests that the member states still saw the utility in having arbitration available as an alternative in certain cases. While there were discussions about limiting arbitration to specific types of disputes, the final text of Article 25 is not so restrictive. Indeed, the arbitration contemplated within the D S U has maintained much flexibility. Fourth, the dynamics of the Uruguay Round made it difficult for the Contracting Parties to focus much attention on developing the proposals for arbitration made after December of 1988. B y that time, the energies of the parties were focused on creating a new and effective binding judicial settlement system in order to appease the U .S . and to avert the potential disintegration of the G A T T system. There was a new urgency to address issues such as appellate review and the modification of the consensus rule. It is apparent from the evolution of the negotiations that while arbitration was generally recognized as a useful and desirable alternative, it was never fully developed or explored. B y 1989, while the Swiss delegation recognized that the arbitration system could be much more developed, there seemed to be no desire to entertain further discussions in light of these other more pressing issues. The discussions for developing that system were effectively ended fairly early into the Uruguay Round. A s such, the rather vague text that became Article 25 remained largely unchanged from the Improvements of 1989. 92 Fifth, it is an appropriate time to consider an amendment to enhance the use of arbitration as an alternative. Again it is worth considering the institutionalist regime theory of Robert Keohane as it relates to the role of the hegemon in creating the system. While the U . S , undoubtedly had a major influence in diverting negotiations on dispute settlement from arbitration to a fully legalized system of judicial settlement with a standing appellate body, its position of power and influence is necessarily reduced by the authority of the D S B as an institution. This rather predictable result may well explain the U.S. 's reluctance to agree to a formal institution during the Uruguay Round. N o w that the judicial settlement system has some traction and a reasonably good record of compliance, it is unlikely that new proposals for improvement would jeopardize the entire multilateral trading system. In building in a requirement of a review of the D S U every four years, 2 8 3 there was in fact a collective recognition of the limits of litigation to resolve disputes. Given the fact that experience has shed light on some of the weaknesses of the judicial settlement system, the prospect of an alternative mechanism may become more attractive to all members states. Further to this last point, while arbitration was clearly treated as a viable alternative during the Uruguay Round, the Contracting Parties at the time did not have benefit o f measuring the success of the judicial settlement system that they were creating. They could only predict where it might be successful, and it where it may fail. Indeed, there have been some failures. There has now been over twelve years of decisions, and a solid history that would assist in determining what types of amendments would strengthen the 283 Porges, supra note 27 at 153-4. 93 dispute settlement system. The history of the development of G A T T also demonstrates a pendulum-like evolution - a period of stability and effectiveness (1948-1960), followed by a period of stagnation (1960-early 1970s') followed by reform (Tokyo Round agreement by 1979) and new found stability and effectiveness (1980's), followed by stagnation (late 1980's and the problems with report blocking) and again reform (the creation of the WTO and the DSU in 1994). The effectiveness of the institutions naturally shift reliance away from a power based system, until they prove to be less effective. As Dunne might suggest, these shifts resemble swings of the pendulum, where the G A T T has vacillated between the extremes of power orientation and rule orientation. 3.10 Conclusion The history of negotiations during the Uruguay Round reinforces the concept of a system with a third option for alternative dispute settlement, other than consultations and judicial settlement. Arbitration in the context of WTO dispute settlement has been conceived as different from the judicial system now in place. The proposals made in the Uruguay Round indicate that arbitration had received widespread recognition as an available alternative, and that alternative even materialized as Article 25 of the DSU, although as discussed in the next chapter, it has never been truly tested. However, because of the political dynamics of the Uruguay Round and the fear of U.S. unilateralism, the arbitration alternative became a neglected part of the multifaceted and politically charged negotiations. Nevertheless, the goal of the negotiators of the Uruguay Round was not to 94 create expansive judicial lawmaking, a fact reflected in the objectives of the D S U considered in the following chapter. For these reasons, it is appropriate to now consider the possibility of this alternative in the present context. To do so, it is necessary to next analyze the current structure of the D S U , the successes and shortcomings of a D S B that is oriented almost exclusively towards the judicial settlement system, and both the current and potential role of arbitration within the D S B . 284 Steinberg, supra note 2 at 250. 95 CHAPTER IV: JUDICIAL SETTLEMENT AND ARBITRATION IN THE CURRENT DSU 4.1 Introduction The D S U may have struck a reasonably good balance between rules and power resulting in the success of the W T O over the first twelve years. This seems to have led to an assumption that because the W T O is more effective than G A T T , that more legalistic procedures wi l l continue to improve compliance rates. This chapter w i l l consider that bias, in the context of the D S U , which now includes three forms of arbitration. This chapter is split into two parts. The first part considers the present judicial settlement and enforcement system under the D S U , looking at the current procedural and remedial provisions against the objectives of the D S U . It w i l l consider the limitations of seeking a binding decision that is to be enforced as the ultimate objective of the judicial settlement system, and the limitation of remedies as the ultimate enforcement tool within this system. The second part of the chapter analyzes the use of three provisions in the D S B that refer to "arbitration": Articles 21.3, 22.6 and 25 of the D S U . I w i l l identify the useful developments in these three forms of arbitration that illustrate the potential benefit o f a more general use of arbitration as a means to resolve difficult political disputes. A s w i l l be seen, Article 22.6 and 21.3 arbitrations arise only after a dispute has been adjudicated, and where the remaining matter is compliance, and are barely distinct from the process of judicial settlement. However, while they do not create a true alternative to judicial settlement, they do demonstrate two interesting points. First, a process that at least 96 resembles an arbitration process, and for which there is no appeal, can be used to resolve political issues. A s such arbitration within the D S U has not been restricted to factual disputes only. Second, these weaker forms of arbitration may in some ways demonstrate the potential benefit of Article 25 arbitration upon the application of compliance theories. 4.2 The Judicial Settlement System In formalizing the panel system, the member parties have in effect created a court with 285 compulsory jurisdiction. Overall, the main goal of the W T O underlying agreements is to ensure access to markets. Where a party alleges a breach of the obligations under one of these agreements, that party may initiate a complaint with a request for consultations under Article 4.3 of the D S U . Should the dispute not resolve through these preliminary steps, the member can request the establishment of a dispute panel. B y litigation 286 standards, the panel dispute settlement process is swift. The D S U contemplates that all disputes w i l l be resolved within 15 months after the filing of the complaint. The litigation can still involve complicating features, including the filing of expert reports and third party intervention by way of amicus curiae briefs. The panel ultimately hears evidence and argument to determine i f there are any breaches of the principles of the underlying agreements. Where it is determined that a responding state has failed to meet an obligation in one of the underlying agreements, the panel then makes a general recommendation that the offending party bring its law or measures into 285 Posner & Yoo, supra note 8 at 45-46. 286 Zekos, "An Examination of GATT/WTO", supra note 42 at 73. 97 conformity with the specific provisions that have been breached. The reports of the panel 287 must be adopted by the D S B within 60 days of circulation and routinely are, due to the reverse consensus rule. Decisions can be appealed to the Appellate Body on grounds of 288 error in law. Appellate panels are typically composed of three members from the 289 permanent standing body who are selected randomly. While decisions are not technically binding in other cases, previous decisions are frequently cited in support of the existence of legal principles, thus creating a form of defacto system of stare , . . 290 decisis. The disputants within the judicial system usually have little choice over procedure and the decision-maker. Article 8 of the D S U provides a top down system for the panel selection. Article 8(4) provides for the establishment of an indicative list that is maintained by the Secretariat for panel selection. While the panel system is referred to as an ad hoc system, there is only minimal party input into panel selection. Article 8(6) sets out that the Secretariat proposes nominations, and provides that the parties shall not oppose that selection absent "compelling" reasons. The ultimate decision is not that of the parties, as Article 8(7) provides that absent an agreement within 20 days, the panel shall be determined by the Director General. It is thus the Secretariat who, as under the 287 Article 16(4) of the DSU. 288 Article 17(6) of the DSU. 289 Article 17(1), (2) & (3) of the DSU; Posner & Yoo, supra note 8 at 45. 290 Hilf, supra note 72 at 116 & 129. While decisions of the Appellate Body do not formally have precedential effect, as Hilf says they tend to "create expectations", and indeed the system has evolved to a system based on principles developed in previous decisions. Reports of the Appellate Body thus tend to "verify" the principles; Pauwelyn, "Americanization", supra note 46 at 125. 98 previous G A T T system, continues to have defacto control over the selection of the , 291 panel. More importantly, the appeal process to the Appellate Body largely mitigates any choice over the decision-maker that does exist at the panel level. B y virtue of Articles 17.6 & 17.13, the Appellate Body can effectively overturn panel decisions on issues of law. Further, many of the proposals regarding the reform of the dispute settlement process advocate for a fully institutionalized panel system, which would reduce even further a state's ability to participate in the choice of the decision-maker. 2 9 2 Conversely, Article 25 provides for a form of arbitration of any matter that is "clearly defined by the parties", on the basis of mutual agreement, thus permitting an element of party control. With respect to control over procedure in panel proceedings, Article 12 provides that panels shall follow the working procedures established by Appendix 3 of the agreement unless the panels decide otherwise, after consulting with the disputing parties. While Article 12(2) provides that procedures should be sufficiently flexible, this does not enshrine the ability of the disputing parties to choose. 291 Weiler, supra note 13 at 202. 292 The debate is fully expounded upon in a series of commentary in the Journal of International Economic Law. See William J. Davey, "The Case for a W T O Permanent Panel Body" (2003) 6 J. Int'l Econ. L. 177. Thomas Cottier, "The W T O Permanent Panel Body: A Bridge too Far?", 6 J. Int'l Econ. L . 187. Andrew W. Shoyer, "Panel Selection in W T O Dispute Settlement Proceedings", (2003) 6 J. Int'l Econ. L . 203 ; Jacques H. J. Bourgeois, "Comment on a W T O Permanent Panel Body" (2003). 6 J. Int'l Econ. L . 211 Michael Cartland, "Comment on a WTO Permanent Panel Body" (2003) 6 J. Int'l Econ. L 214; Seung Wha Chang, "Comment on a W T O Permanent Panel Body" (2003) 6 J. Int'l Econ. L. 219; Luzius Wasescha, "Comment on a W T O Permanent Panel Body" (2003)6 J. Int'l Econ. L. 224; Julio Lacarte, "Comment on a W T O Permanent Panel Body" (2003) 6 J. Int'l Econ. L. 227. Frieder Roessler, "Comment on a W T O Permanent Panel Body: The Cobra Effects of the W T O Panel Selection Procedures" (2003) 6 J. Int'l Econ. L. 230. 99 With respect to the issue of compliance with D S B rulings, the D S U demonstrates a true mixture of the legal and political aspects of the W T O . While Article 19 of the D S U permits the panel or Appellate Body to make recommendations as to how the states could bring the measure within conformity with their obligations, in practice this has rarely 293 • • been done. The manner by which States comply with a ruling, and the decision whether the required fix is administrative, regulatory or legislative, is effectively left to the offending State to determine after the adoption of the panel or Appellate Body report. In practice, specific recommendations are rarely made and panels have usually rejected requests to make such recommendations. 2 9 4 There is therefore an obvious sensitivity to the prerogative of the state to determine how it w i l l address the failure to meet its obligations under the W T O agreements - a preference in the current system for less forceful means for encouraging compliance. Nevertheless, full implementation is the 295 preferred method under s. 22.1 of the D S U . 4.3 The Objectives of the DSU While there has been a preponderance of writing regarding reforms of the D S B , most observers agree that the level of compliance engendered by the D S B has marked an unqualified success. The rates of dispute resolution and compliance with decisions of the D S B have been h igh , 2 9 6 and mark a significant improvement over the previous G A T T 293 See Porges, supra note 27 at 146. 294 Fukunaga, supra note 48 at 399. 295 Vazquez & Jackson, supra note 22 at 558. 296 Wilson, supra note 7; Donald McRae, "What is the Future of WTO Dispute Settlement?" (2004) 7 J. Int'l Econ. L . 3 at 5; Davey, "The First Ten Years", supra note 6; McGivern, supra note 9 at 156. 100 system/*' This has been a remarkable achievement in international law, which has generally struggled with compliance with the decisions of international tribunals. 2 9 8 Are the current processes and remedies under the D S U consistent with the objectives of the system? In order to answer that, one must consider the objectives explicitly set out in the D S U . Article 3.2 confirms the principles of security and predictability in the multilateral trade system, while clarifying that the D S B cannot add to or diminish rights in the underlying agreements. This appears to promote a certain efficacy in the dispute settlement system, 2 9 9 through the development of rules for other disputes. However Articles 3.3 and 3.4 promote the prompt settlement of disputes, and establish that the recommendations made by the D S B are aimed at achieving a satisfactory settlement, in individual disputes. 3 0 0 This seems particularly clear from Article 3.4 of the D S U , which provides that: Recommendations and rulings made by the D S B shall be aimed at achieving a satisfactory settlement of the matter in accordance with the rights and obligations under this Understanding and under the covered agreements. 297 Marc L. Busch & Eric Reinhardt, "The Evolution of G A T T / W T O Dispute Settlement" Trade Policy Research 2003 (Ottawa: Minister of Public Works and Government Services, 2003) 143 at 157. [Busch & Reinhardt, "The Evolution of GATT/WTO"] . For the period between 1980 and 2000, compliance levels for developed states increased from 40% under G A T T to 74% under the W T O DSB regime. Though the results for developing countries are far less impressive, they also improved substantially, increasing from 36 % to 50%. Contra see Posner & Yoo, supra note 8. 298 Contra see Posner & Yoo supra note 8. 299 Vazquez & Jackson supra note 22 at 563. 300 Ibid. 101 The U.S . itself has recently confirmed that the settlement of disputes is primary over the creation of legal principles, despite its role in establishing the Appellate Body and the legalistic direction of the D S B : ..the purpose of the dispute settlement system is not to produce reports or to "make law," but rather to help Members resolve trade disputes among them. W T O adjudicative bodies should avoid making findings that are not aimed at resolving the dispute. It is useful to bear in mind that such bodies are not permitted to render authoritative interpretations of the covered agreements. 3 0 1 The objective of positive resolution of dispute is further emphasized by Article 3.10, which provides that members engage in dispute resolution procedure in good faith, in an effort to resolve the dispute. Therefore, as much as Article 3.2 promotes predictability through a developing body of law and decisions, the comments of the U . S . would seem to detract from any notion that the D S B ' s role is the creation of a body of authoritative law. There is therefore good reason to conclude that the overall primary purpose of the dispute settlement system is to settle the specific dispute and remedy any injury, and not simply to secure compliance irrespective of the injury. 3 0 2 The D S U attempts to capitalize on the element of reputation, as is the role of an institution according to compliance theorists such as the Chayes. The D S U contributes to the creation o f norms against which the conduct o f disputing parting can be measured, by providing additional pressure for the parties to resolve issues through negotiations 301 Communication From The United States, WTO Doc. No. TN/DS/W/82/Add.2 (17 March, 2006). 302 Fukunaga, supra note 48 at 386; Marc L. Busch & Eric Reinhardt. "Fixing What 'Ain't Broke' ?: Third Party Rights, Consultations and the DSU" online: Georgetown University <http://www9.georgetown.edu/facultv/mlb66/DSU.pdf> [Busch & Reinhardt, "Fixing What Ain't Broke"]; Contra Hilf, supra note 72 at 116. 102 towards the settlement of the dispute. This pressure takes several forms. Article 19 of the D S U provides that the panel w i l l ultimately make "recommendations", even though they are still considered to be legally binding. 3 0 3 Article 21.6 o f the D S U provides a further political tool of pressure, as it allows any party to raise the issue of compliance at any time. This permits any member to address the matter within the D S B , putting public light and attention on the delinquent state's inaction. The notion of political resolution therefore permeates the D S U and is a dominant theme of the rules themselves, and to some extent, the practices of the D S B . The D S B ' s reluctance to make specific recommendations for an offending party to bring their measures within compliance, leaves the parties to attempt to agree on an appropriate resolution. Enforcement mechanisms are still largely dependent on the involvement of the parties themselves, as demonstrated by softer mechanisms such as monitoring, reporting and ministerial meetings, which seek to encourage compliance, 3 0 4 even where retaliation is approved. However, while the D S U remedies are somewhat softer, they do not contemplate that the member state has a choice to provide compensation or to endure retaliation in the place o f elimination of the offending measure, contrary to the belief o f some. 3 0 5 The nature of compensation is meant to be temporary, and cannot be used to replace compliance. Indeed, it is telling that the Sutherland Report itself has sought fit to 303 Hudec, "Broadening the Scope", supra note 160 at 372-373. 304 Christopher Arup, "The State of Play of Dispute Settlement 'Law'at the World Trade Organization" (2003) 37 J. World Trade 897 at 902. 305 The Sutherland Report, supra note 17 at 54. 103 clarify this point. The express reference to such a trite point is a good indicator that the perceptions of many have gone far adrift of the original objectives of the D S U . There is thus some ambivalence surrounding the nature of the D S B and its system of 307 remedies, as there seems to be constant tension between the creation of legal principles and the resolution of the specific dispute. One observer has noted that although the practical availability of remedies has marked an improvement over G A T T , these 308 improvements are in fact less than what one expects from an "effective" legal system. It has even been suggested that while the W T O system appears to be a stronger legal system, the responsibility for breaching an obligation is actually more limited than under general international l a w . 3 0 9 Good examples of this growing problem are decisions involving the U.S and the E . C . Both the decision respecting the European ban on beef hormones in European Communities - Measures Concerning Meat and Meat Products ("Beef Hormones") 3 1 0 and the ruling in U.S. Tax Treatment for Foreign Sales Corporations ("Foreign Sales Corporations") 3 1 1 failed to result in compliance with adopted rulings, despite the approval of retaliation measures. They are examples of disputes that encompass much more than ordinary commercial concerns. The fact that 306 ibid. 307 McRae, supra note 296 at 7. 308 Hudec, "Broadening the Scope", supra note 160 at 399. 309 Vazquez & Jackson, supra note 22 at 560. 310 European Communities - Measures Concerning Meat and Meat Products (Hormones, Report of the Appellate Body, WT/DS26/AB/R, WT/DS48/AB/R, 16 January 1998; and see European Communities -Measures Concerning Meat and Meat Products (Hormones)- Original Complaint by the United States -Recourse to Arbitration by the European Communities under Article 22.6(c) of the DSU, Decision By the Arbitrators, WT/DS26/ARB, 12 July 1999. 311 United States Tax Treatment for Foreign Sales Corporations, Report of the Appellate Body, WT/DS108/ASB/RW, 24 February 2000; and see United States - Tax Treatment for 'Foreign Sales Corporations' - Recourse to Arbitration by the United States under Article 22.6 of the DSU and Article 4.11 of the SCM Agreement, WT/DS 108/ARB, 30 August 2002. 104 such disputes go into areas such as health concerns or control over the tax base, put them 312 into the category of disputes that are "deep-rooted in political complexities" where non-compliance is more likely. Given the objectives of the D S U , it is more appropriate to consider the political and legal aspects of dispute settlement as two extremes of a continuum, and not as a dichotomy where the system is either one or the other. The system is geared towards the resolution of specific disputes, though should, to the extent possible, develop legal principles to assist in avoiding disputes on the same issues. 4.4 The Limited Effectiveness of Remedies Under the DSU There is an inherent contradiction in the legalized judicial settlement system that is based on remedies such as retaliation, as economic power continues to act as an overriding factor. Ultimately, the enforcement of an award under the D S U depends on the economic strengths of the aggrieved state and not any applicable rules or principles of international commercial arbitration. 3 1 3 The weaknesses in the system of remedies suggest that the coercive force of reputation might better address the implementation of changes in politically difficult cases. The remedies provided for under Article 3.7 o f the D S U include compensation, and the suspension of concessions or obligations to the losing party, a remnant of the former 312 Sungoon Cho, "The Nature of Remedies in International Trade Law" (2004) 65 U. Pitt. L. Rev. 763 at 784. 313 Zekos, "An Examination of G A T T / W T O " , supra note 42 at 74-5. 105 Article X X I I L 2 of the G A T T . Although the W T O prefers the term "counter-measures",314 retaliation is an apt term. While retaliation already existed as a remedy in international law, the formal inclusion of retaliation in the W T O agreement fulfilled the purpose of 315 "taming retaliation", by controlling its use by requiring parties to obtain authorization first, thus proscribing unilateral action. The D S U contemplates countermeasures equivalent to the level o f nullification or impairment of benefits of the state under the underlying agreements. The inclusion of retaliation and other specific mechanisms has been touted as an innovation that recognizes the problem of compliance, and which at 316 least attempts to address those problems within the treaty itself. In many ways, the remedies of compensation and retaliation are "deeply flawed" and "dysfunctional". The pecking order of preference in Article 3.7 confirms that these two remedies are included only to provide incentives for the state to withdraw the 318 offending measure and comply with obligations. Article 3.7 provides first for a mutually acceptable solution in accordance with the agreements, and the withdrawal of the impugned measures. Entitlement to compensation arises only i f immediate withdrawal is impracticable, and then only as a temporary measure. However it requires the mutual agreement of the disputing parties, and thus has been criticized as solution in theory on ly . 3 1 9 One argument for the preference for compensation is that this does not 314 Fukunaga, supra note 48 at 418. 315 McGivern, supra note 9 at 143. 316 Joost Pauwelyn, "Enforcement and Countermeasures in the WTO: Rules are Rules - Towards a More Collective Approach" (2000) 94 A.J.I.L. 335 at 339. [Pauwelyn, "Countermeasures in the WTO"]. 317 John H. Jackson, "International Law Status of W T O Dispute Settlement Reports: Obligation to Comply Or Option To Buy Out" (2004) 98 Am. J. Int'l L. 109 at 123. 318 See also McGivern, supra note 9 at 144. 319 Bronckers & van den Broek, supra note 115 at 103. 106 c o m p o u n d the p r o b l e m o f trade distort ion l i k e retaliation. O f course it is equal ly arguable that compensat ion facilitates trade distort ion, as compensat ion is t y p i c a l l y • 321 understood as trade concessions and not actual monetary compensat ion. Further, even f inancia l compensat ion c a n be trade distortive, at least indirect ly , i f v i e w e d as a means o f 322 b u y i n g out responsibi l i ty to c o m p l y w i t h obl igat ions. C o m p e n s a t i o n has i n any event p r o v e n to have l i m i t e d pract ical effect i n the W T O context. S i m i l a r l y , retal iat ion is permitted o n l y as a last resort, where compensat ion is unsuccessful . T h e scheme is therefore not intended to be remedia l , but is rather directed at i n d u c i n g c o m p l i a n c e . 3 2 4 W h i l e retal iation is directed towards this purpose, it is nevertheless antithetical to the pr inciples u n d e r l y i n g the W T O agreements, as it results i n an increase, rather than a reduct ion, i n trade barriers. It u l t imate ly benefits groups i n 325 favour o f protect ionism and does not enhance the l ibera l i zat ion o f trade. Indeed it has 326 been suggested that the o v e r a l l objective o f retal iat ion i n the W T O is unclear. 320 Fukunaga, supra note 48 at 412. 321 ibid. 322 The Sutherland Report, supra note 17 at 54. 323 Fukunaga, supra note 48 at 412, where it is noted that up to February 2006, compensation had been agreed to in only three cases - in two cases it took the form of tariff reduction, and in one case an actual monetary amount of compensation agreed upon. 324 van den Broek, supra note 70 at 139. 325 McGivern, supra note 9 at 152. 326 Holger Spamann, "The Myth of "Rebalancing" Retaliation in W T O Dispute Settlement Practice" (2006) 9 J. Int'l Econ. L . 31 at 60. 107 Furthermore, retaliation as a sanction has been criticized as the weakest part of the D S U 327 and as bad policy, as it requires member states to effectively punish its own citizens. A decision to retaliate w i l l necessarily have a negative effect on the consumer citizens of the complaining state, as trade barriers w i l l tend to create higher prices in that particular sector. Conversely, retaliation may not even provide relief to the industry in the 328 complaining states. The complaining state is put it in a position to choose its own poison - allow the discriminatory practice to continue, thus adversely impacting the exporters, or implement countermeasures on other imported products and hurt its own consumers. It can only hope that the proper choice wi l l enhance the prospect of inducing compliance, or the development of a politically acceptable solution. Developing states have a more fundamental problem, as the effect on domestic factions 330 tends to affect smaller economies even more and the benefits of retaliation can be 331 outweighed by the cost of litigation alone. Attempts at retaliation may have even more 332 devastating consequences, such as the reduction of development aid. Overall, for either developed states or developing states, the difficulty is thus not establishing the legal 333 authority for retaliation, but rather the political w i l l to implement it. 327 Mitsuo Matsushita, Thomas J. Schoenbaum, and Petros C. Mavroidis. The World Trade Organization: Law, Practice, and Policy. (Oxford ; New York: Oxford University Press, 2003) at 92-93; McGivern, supra note 9 at 153. 328 McGivern, supra note 9 at 152; Bronckers & van den Broek, supra note 115 at 103. 329 McGivern, supra note 9 at 153. 330 Steger, "Systemic Issues", supra note 6 at 69. 331 Bronckers & van den Broek, supra note 115 at 106. 332 Joost Pauwelyn, "Enforcement and Countermeasures", supra note 316 at 338. 333 Busch & Reinhardt, "The Evolution of G A T T / W T O " , supra note 297 at 159-160. 108 The inclusion of retaliation has been less successful than anticipated by the drafters of the D S U 3 3 4 and has proven to have limited impact. In the Foreign Sales Corporations dispute, the E . C . was provided with the authority to impose over $4 bil l ion in retaliation in 335 response to the sustained non-compliance of the U.S . . This was the largest retaliation 336 * request in international law, and thus represented a milestone in the use of retaliation as a measure to induce compliance. However, the underlying trade interdependence with the U . S . and a resistance from the European industry made it impossible for the E . C . to 337 impose the full extent of the sanctions approved by the D S B . Some have wondered what such outcomes have achieved other than simply awarding huge retaliation 338 authorization for its own sake. Similarly in the case of the E.C.'s ban on imports of beef raised with growth-promoting hormones in the Beef Hormones dispute, both Canada and the U .S . obtained D S B authorization to retaliate in 1999. 3 3 9 Both parties implemented retaliation measures of 100 per cent duty on particular agricultural products. 3 4 0 Despite having applied a "fairly 334 McGivern, supra note 9 at 157. 335 United States -Tax Treatment for 'Foreign Sales Corporations' - Recourse to Arbitration by the United States under Article 22.6 of the DSU and Article 4.11 of the SCM Agreement, WT/DS108/ARB, 30 August 2002. 336 Spamann, supra note 326 at 74; Steinberg, supra note 2 at 272. 337 van den Broek, supra note 70 at 147. See also Bronkers and van den Broek, supra note 115 at 104. 338 Spamann, supra note 326 at 74. It should be noted however that U.S. Congress has made two sets of legislative changes in response. See Wilson supra note 7 at 430; online: W T O <http://www.wto.org/english/tratop e/dispu e/cases e/dsl08 e,htm> (last accessed 25 August, 2007). 339 European Communities - Measures Concerning Meat and Meat Products (Hormones)- Original Complaint by the United States - Recourse to Arbitration by the European Communities under Article 22.6(c) of the DSU, Decision By the Arbitrators, WT/DS26/ARB, 12 July 1999; European Communities -Measures Concerning Meat and Meat Products (Hormones) - Original Complaint by Canada - Recourse to Arbitration by the European Communities under Article 22.6(c) of the DSU, Decision By the Arbitrators, WT/DS48/ARB, 12 July 1999. 340 McGivern, supra note 9 at 145. 109 calibrated retaliation strategy", this failed to result in proven compliance by the E . C . and the dispute has carried on for several years. 3 4 2 While the D S B certainly marks an improvement over the dispute resolution under the G A T T system, initial results under the W T O as compared to the G A T T system have suggested it is no better at inducing compliance for adverse decisions in cases involving the E . C . and the U . S . 3 4 3 Further, one observer has postulated that the arbitrariness of authorizations for suspension of concessions such as this one have in fact merely represented a diplomatic solution rather than a reasoned calculation based on the D S U measure of equivalence. 3 4 4 It has been suggested than there is in fact no evidence that retaliation actually works . 3 4 5 For these type or reasons, it has been suggested that direct retaliation may be far less important in enforcing international law than the effect of reputation. 3 4 6 It ought not be surprising that the D S U had not included stronger remedies. A s some have observed, the limitation of remedies was effectively a tradeoff for establishing the institutional context of the W T O . 3 4 7 There has now been a movement toward a preference 341 Jidi Nzelibe, "Credibility Imperative - Political Dynamics of Retaliation" (2005) 6:1 Theor. Inq. L. 215 at 227. 342 In late 2005 a hearing of the E.U.'s request to review the retaliation measures of both the U.S. and Canada commenced. The E .U. has claimed that it has now brought its measures within compliance, and have sought an end to the authorization of continued retaliation measures. The Panel report was expected in June of 2007. See online: WTO < http://www.wto.org/english/tratop_e/dispu_e/cases_e/ds321_e.htm> (last accessed 15 August 2007). See also Wilson, supra note 7, at 402. 343 Busch & Reinhardt, "Bargaining in the Shadow", supra note 58 at 166 & 171. Contra see Wilson, supra note 7 at 397 who suggests the E.C and the U.S. have "generally succeeded in bringing themselves in compliance...except for some residual compliance difficulties in a small number of cases." 344 Spamann, supra note 326 at 58, although Spamann suggests that it is difficult to conclude that the E.C. Hormones award "was as wrong as other Article 22.6 decisions were". 345 Steger," Systemic Issues", supra note 6 at 71, where she suggests that even economic modeling has proven to be far too speculative in establishing the effectiveness of retaliation. See also McGivern, supra note 9 at 152-153. 346 Spamann, supra note 326 at 78. 347 Vazquez & Jackson, supra note 22 at 562. 110 for forms of compensation over retaliation as a means of enforcing. Given the recognized shortcomings of compensation and retaliation as a means of inducing compliance with trade obligations, there are gaps in the current D S U as to the means of enforcing rulings, particularly in politically difficult cases. Perhaps a form of compensation is suitable for cases where developing countries are the complainants. However, most of the difficult political cases that have resulted in retaliation awards are cases where developed countries have failed to comply with D S B rulings over complaints brought by other developed countries. 3 4 9 4.5 Non-Compliance and Shortcomings of the Judicial Settlement System The tightening of the legal procedures and even the inclusion of enforcement mechanisms have undoubtedly contributed something to the success of the D S B . It is therefore easy to understand why discussions regarding reform focus on developing other remedies and increasing the authority of the panels or the Appellate Body. However, this dialogue ought not drift too far from the context of the actual objectives of the D S U . Compliance based on coercion is neither a realistic objective, nor one which is contemplated by the D S U . Nor is the provision of compensation for transgressions an overriding goal. Had it been so, the D S U would have likely included more severe penalties and punitive measures for past breaches. 3 5 0 348 Bronckers & van den Broek, supra note 115; Steger, "Systemic Issues", supra note 6; McGivern, supra note 9 at 157. 349 See generally Wilson, supra note 7. 350 Vazquez & Jackson, supra note 22 at 565. I l l Given this retreat to legalism, it is increasingly important to critically assess precisely why countries have failed to comply with specific rulings, even i f this has occurred only in a minority of cases. It has been argued while parties use the processes available to them, jthey make pragmatic decisions that are guided as much by political and economic 351 decisions as legal conclusions. If this is so, then one has to consider the assumption that better compliance wi l l result from better litigation system with more rules and remedies, and increased Appellate Body authority. In reviewing the recent decisions o f the W T O , Naboth van den Broek has described 352 several possibilities for non-compliance. One of the factors he identifies is the political inability of a member state to comply. In policy driven cases, there can be too much pressure from domestic interests to permit the state to comply with the ru l ing . 3 5 3 A second but related reason is the conflicting norms of international and domestic arena and the slow rate of internalization of the higher norms of international law. He also notes that the role of non-state actors creates additional political pressures, noting that the E . U . and the U .S . are the members who are most highly influenced by non-governmental organizations ( " N G O s " ) . 3 5 4 Given these political pressures, it is therefore important to recognize that not all disputes can be resolved in the context of the legalistic mechanisms of W T O dispute resolution. Despite the current prevalent ambition to have disputes resolved by binding and 351 Arup, supra note 304 at 919. 352 van den Broek, supra note 70 at 148-151. 353 Ibid, at 149. 354 Ibid, at 150. 112 enforceable decisions, the "engine" of dispute settlement has always been the early settlement of cases. 3 5 5 The research of Marc Busch and Eric Reinhardt suggests that the advent of the D S U under the W T O has not led to a significant increase in early 356 settlements. This research furthermore underlines why negotiations can be so important. The research has shown that cases involving disputes between the E . C . / E . U . and the U.S . , the likelihood of concessions is reduced dramatically where the parties fail to settle before a 35T panel ruling, regardless of the ultimate decision by the D S B . The Beef Hormones and Foreign Sales Corporations disputes can be distinguished from other cases by virtue of the non-commercial interests at stake: the health risks surrounding the use o f hormones in U . S . cattle; and domestic income tax policy, an integral aspect of economic sovereignty. The fact that such disputes involve non-trade concerns makes non-compliance more 358 l i k e l y , J J O and more predictable. 4.6 An Over Enthusiasm for the Judicial Settlement System While the D S U has incorporated forms of arbitration in the final text, it has undoubtedly established a predominant framework for the resolution of disputes that resembles a sophisticated judicial system. Despite the D S B ' s less than stellar record of resolving disputes between the U .S . and the E . U . , the overall high level of compliance under the 355 Busch & Reinhardt, "Evolution of G A T T / W T O " , supra note 297 at 176. 356 Ibid, at 165. 357 Ibid, at 160-1 & 164. 358 McGivern, supra note 9 at 156. 113 D S U regime has made attractive the notion of increasing the power of panels and the Appellate Body. The logic is simple - increasing the power of the decision-maker w i l l engender even better compliance with it decisions. Using a similar logic, such reform 359 focuses on "improving" the remedies for breaches, rather than focusing on encouraging settlement. This line of thinking ultimately hinges on the effective use of retaliation or compensation as the main tool of the W T O against non-compliance, over the element of reputation or the "ultimate remedy" of community pressure. 3 6 0 The review of compliance records of member states however has suggested that compliance is more likely in cases 361 resolved by a negotiated settlement rather than a decision of the D S B . The limitations of the system addressed above represent the reason why the W T O should look more toward a means of dispute settlement that enhances negotiations for the resolution of politically difficult cases. Although compliance theories might support maintaining a detailed framework for dispute resolution to create the compliance pull , the legalistic model has instilled a single minded approach to adjudication of disputes. The D S B has been resorted to much more frequently than was ever anticipated. 3 6 2 Indeed, the growth can also be attributed to the 363 • * • ability of states to litigate cases involving political sensitivities. But the ability to bring such cases does not mean that the system w i l l adequately resolve such cases. I agree with 359 See e.g. The Sutherland Report, supra note 17. 360 Hudec, "Broadening the Scope", supra note 160 at 399-400. 361 Busch & Reinhardt, "Evolution of G A T T / W T O " , supra note 297 at 145. 362 McRae supra note 296 at 4; The Sutherland Report, supra note 17 at para. 22. See also Leitner & Lester, supra note 3 at 166, who report that as of January 1, 2007, 356 complaints have been made, though there has been a noticeable decline in the last 3 years. 363 Pauwelyn, "Enforcement and Countermeasures", supra note 316 at 338. 114 the proposition that the frequency of use of the judicial settlement system is not appropriate for the level of political aspects and diplomacy that should be maintained. 3 6 4 Robert Hudec has noted that although the practical availability of remedies has marked an improvement over G A T T , these improvements are in fact less than what one expects 365 • • from an "effective" legal system. He concludes that it is best to leave room to maneuver for politically acceptable solutions. This is important for at least two reasons. First, the legal system cannot resolve also disputes. Second, caution should be exercised to ensure that expectations of the current dispute settlement do not exceed what can be reasonably delivered. The legitimacy of the W T O is somewhat fragile, 3 6 6 and failing to meet expectations w i l l serve only to further detract from the body's moral authority. While arbitration might be considered a legalistic method of dispute settlement, it is nevertheless a softer form of third party resolution that is more amenable to ongoing negotiations. Unfortunately, as can be seen in the second part of this chapter, it has not yet been integrated or utilized to its full potential. 364 McRae, supra note 296 at 5. 365 Hudec, "Broadening the Scope", supra note 160 at 399. 366 See e.g. Robert Howse & Kalypso Nicolaidas. "Enhancing W T O Legitimacy - Constistutionalization or Subsidiarity" in Marco Verweij & Tim Josling (eds.), Deliberately Democratizing Multilateral Organization, special issue of Governance (2003) online: University of Michigan <http://facultv.law.umich.edu/rhowse/>. 115 4.7 Specific Arbitration Provisions in the DSU The three separate forms of arbitration fall into two categories. Article 21.3 of the D S U , provides for arbitration for the timeframe for compliance, and Article 22.6 of the D S U provides arbitration over the level of concessions that can be suspended by an aggrieved state should the offending state fail to comply or provide. These were introduced in the 367 Secretariat draft following a Canadian proposal, and have been aptly characterized as 368 "sui generis''' forms of arbitration, as they are unique adaptations of the concept of arbitration. They are effectively mandatory forms of dispute settlement that merely pick up where the panel dispute settlement system ends, and do not provide an alternative process. In all likelihood, they might have been included as part of the panel system, but for the fact that the shape of the panel system was still uncertain at the time that these arbitration provisions were included in the Chairman's Text. However they do provide two key distinctions from the process for panel decisions: first, there is no appeal from determinations of the arbitrator; and second, arbitration gives the disputing parties better control over procedure and the arbitrator, although in the case of latter, not as much as appears on its face. Conversely, Article 25 maintains the broad language included in the Improvements of 1989 draft, with only minor adjustments and permits arbitration for any clearly defined issue. Unlike the "sui generis" forms of arbitration, it does not require a previous decision from a panel or the Appellate Body. Article 25 arbitration provides a purer form of party 367 Supra note 268. 368 Boisson de Chazournes, supra note 40 at 953. 116 control, though the parties are still bound by the substantive law of the underlying W T O agreements. The arbitrator arguably exercises relative independence, as the award does not require any adoption or approval by the D S B . While this is the only form of arbitration that exists as an alternative to the judicial settlement process before panels and 369 the Appellate Body, it has so far only been used in one case. 4.7.1 Timeframe Arbitration Though not an alternative form of dispute settlement within the D S U , Article 21.3 provides an interesting study in the use of arbitration in the D S U . While it is referred to as "arbitration", as it was when it was introduced into the Negotiating Group draft in September 1990, it functions as an adjunct to the judicial system, as it operates only after a D S B ruling has been adopted. Yet it could not have been clear precisely how it would fit in with the system of judicial settlement in 1990, as the Uruguay Round negotiations had still not developed many of the staples of the dispute settlement system including the automatic adoption of D S B rulings, and the appellate system. Today, the utility of article 21.3 begins precisely where that of the panel and Appellate Body has ended. 370 The D S U requires prompt compliance with the D S B recommendations and rulings. Once a panel or Appellate Body decision is adopted by the D S B , the offending state is required to report to the D S B at a meeting within 30 days after the adoption of the report 369 United States - Section 110(5) of the U.S. Copyright Act - Recourse to the Arbitration under Article 25 o / ^ e D S V , WT/DS 160/ARB25/1, 9 November 2001 ["U.S. Copyright']. 370 Article 21.1 of the DSU. 117 as to its intentions to comply. The offending state w i l l often claim that immediate compliance is not possible and wi l l ask for a "reasonable" period of to implement the necessary change, as provided for in Article 21.1. The different possible changes are often categorized into three categories: administrative changes, regulatory changes and 372 legislative changes. The reasonable period of time can be determined in three ways: first, the offending state can propose a period with the approval of the D S B ; second, the offending state can make an agreement with the aggrieved state within 45 days; or, third, i f these first two methods are unsuccessful, through binding arbitration to take place within 90 days of the adoption of the report. 3 7 3 Timeframe arbitration is therefore a course of last resort - a default provision where all else fa i l s . 3 7 4 Given that it is initiated by only one party as a default measure, a referral to arbitration does not require the consent of both parties. Article 21.3 does not provide for any specific working procedures and unlike other arbitrations, there is usually no organizational meeting and the arbitrator draws up a working schedule. 3 7 5 While the disputing parties are permitted to agree on an arbitrator, in the absence of an agreement with 10 days, an arbitrator is appointed by the W T O 371 Article 21.3 of the DSU. 372 Hughes, supra note 176 at 84. 373 Article 21.3 of the DSU. 374 Hughes, supra note 176 at 82. 375 Pierre Monnier, "Working Procedures Before Panels, the Appellate Body and Other Adjudicating Bodies of the W T O " (2002) 1 Law & Prac. Int'l Courts & Trib. 481 at 517 [Monnier, "Working Procedures"]. 118 Director General. In most cases, a single arbitrator is appointed, usually one from the 377 permanent Appellate Body. This practice is not required by the rules, and has likely developed not only for the sake of efficiency of appointment, but also due to member preference for Appellate Body experience and the maintenance of consistency and 378 coherence in awards. These latter interests are simply not as important with respect to the use of arbitration that I propose. I w i l l later suggest that the practice of reliance on Appellate Body experience out of a sense of familiarity, or to promote consistency, ought not be continued in the context of Article 25 arbitration, in order to maintain distance with the mainstream judicial settlement process. The extent of the jurisdiction of the arbitrator is brought into question by a peculiar reference to the appropriate timeframe in Article 21.3, which provides: .. .a guideline for the arbitrator should be that the reasonable period of time to implement a panel or Appellate Body recommendation should not exceed 15 months from the date of the adoption... However that time may be shorter or longer depending upon the particular circumstances. There has been much debate in the awards as to whether the 15 months constitutes a default period o f time, where a party would have to provide a reason to deviate from that 379 380 norm either way, or an outer limit. However, it has generally been accepted that the 376 Ibid. See also Hughes, supra note 176 who points out that of the first 15 cases, 10 resulted in agreement by the parties. 377 Palmeter & Mavroidis, supra note 29 at 272. See also Hughes, supra note 176 at 83, where she suggests it is difficult to say why parties have preferred Appellate Body members. 378 Hughes, supra note 176 at 83. 379 Japan — Taxes on Alcoholic Beverages, Award of the Arbitrator under Article 21.3(c) of the DSU, WT/DS8/15, WT/DS10/15, WTVDS11/13, 14 February 1997 at para. 27; 119 arbitrator has relative discretion to determine the period based on a consideration of the impact of the "particular circumstances". This is always governed by the overriding principles of reasonableness and prompt compliance, but nevertheless gives the arbitrator 381 a fair amount of discretion to consider the individual circumstances. The arbitrator is ultimately guided by the principle that the reasonable period of time should be the 382 shortest period possible in considering the offending state's legal system. A trend toward progressively shorter periods in the awards demonstrates that this principle has gradually crystallized. Effectively, while 15 months had been treated as the standard until mid 1998, by 2000, the average reasonable period was eight months for a case requiring administrative change and twelve months for cases requiring legislative changes. 3 8 3 The arbitrator has no jurisdiction to make recommendations as to the means by which the offending state should implement the suspension, or whether or not a particular countermeasure actually conforms to the principles of the agreement or the 384 recommendations and rulings of the panel decision. The choice of method of implementation is that of the offending state, subject to review by the normal dispute European Communities - Regime for the Importation, Sale and Distribution of Bananas Award of the Arbitrator under Article 21.3(c) of the DSU, WT/DS27/15, 7 January 1998 at para. 19. 380 EC Measures Concerning Meat and Meat Products (Hormones) Award of the Arbitrator under Article 21.3(c) of the DSU, WT/DS26/15, WT/DS48/13, 29 May 1998 at para. 48. 3 81 United States - Measures Affecting The Cross-Border Supply Of Gambling and Betting Services Award of the Arbitrator Under Article 21.3(C) of DSU, WT/DS285/13, 19 August 2005 Arb-2005-2/19, ("U.S. Gambling") at para. 44. 382 Canada - Term of Patent Protection, Award of the Arbitrator under Article 21.3(c) of the DSU, WT/DS170/10 28 February, 2001 at para. 39 ["Canada - Term of Patent Protection"]. Also U.S. Gambling, supra note 381 at para. 33. See also Article 3.3 & Article 21.1 of the DSU, which set out the principles of the necessity of prompt settlement of disputes & prompt compliance with decisions. 383 Pierre Monnier, "Time to Comply with W T O Rulings" (2001) 35 J. World Trade 825 at 831 [Monnier, "Time to Comply"]. 384 U.S. Gambling, supra note 381 at para. 33. 120 settlement process. This gives rise to an unusual situation - it is possible that the measure proposed by the offending state to implement is totally inadequate, or is unresponsive to the recommendations made by the panel. A s such an arbitrator could be called upon to determine the period for which the state must comply, even i f the proposed measure does not conform to the general principles of the D S U that ultimately governs the arbitration. Article 26.1 allows the arbitrator, upon the request of one of the parties, to determine the amount of benefits nullified or impaired by the offending measure, and to suggest ways that the matter can otherwise be resolved. However, these suggestions are 386 not binding, and arbitrators are in any event reluctant to do so. A s of January 2007, there have been 21 arbitration awards circulated pursuant to Article 21 .3 . 3 8 7 While I would suggest that this form of arbitration lacks a true element of party control, one remarkable aspect of this procedure is the ability of an arbitrator to issue a ruling that considers the legislative function of the member state. Timeframe arbitration often centers on what "particular circumstances" would justify extending the period for compliance. This is where the scope of the arbitrator's role is most obvious and broad. Indeed, the arbitrator arguably exercises a quasi-supervisory role over a state's legislative or regulatory process. This is illustrated by the case of the U.S . complaint against Canada 388 in Canada - Term of Patent Protection. 385 Article 21.5 of the DSU; Palmeter & Mavroidis, supra note 29 at 248; Fukunaga, supra note 48 at 401. 386 Hughes, supra note 176 at 83. 387 Leitner and Lester, supra note 3 at 174. 388 Canada - Term of Patent Protection, supra note 382. 121 In the Canada - Term of Patent Protection arbitration, the arbitrator was to determine the reasonable period of time for Canada to comply with a panel ruling requiring Canada to bring its patent legislation into conformity with its obligations under the Agreement on 389 Trade-Related Aspects of Intellectual Property Rights ("TRIPS"). The panel had found that Canada's compulsory licensing scheme allowing generic pharmaceutical producers to manufacture patented drugs did not provide the minimal 20 year period of patent protection for the original producer as required under TRIPS. Canada argued that the reasonable period of time to comply was almost 15 months because of the "contentiousness" of the measure. The arbitrator considered that Canada's proposed timeline for amending the legislation appeared to depend upon the priority given by the Canadian government. 3 9 0 The arbitrator further determined that the contentiousness of the measure for reform and the likelihood of strong opposition was irrelevant. 3 9 1 The timeframe arbitration review of the parliamentary process delves into the area of state sovereignty. Though the arbitrator in Canada - Term of Patent Protection was 392 • reluctant to accept all of the legislative restraints as factors, an arbitrator w i l l generally, on a case by case basis, consider the nature of the changes and the legislative process to some degree. While the contentiousness of legislation is not a relevant factor, the 393 complexity of legislation can be a factor. Whether or not the Parliamentary schedule of 389 Annex 1C to Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations and Marrakech Agreement Establishing the World Trade Organization, April 15, 1994. 390 Canada - Term of Patent Protection supra note 382 at para. 64. 391 Ibid, at para. 60; See also Hughes, supra note 176 at 84. 392 Canada - Term of Patent Protection, supra note 382 at paras. 52- 60. 393 U.S. Gambling, supra note 381 at para. 48; Canada - Term of Patent Protection supra note 382 at para. 58. 122 a particular state w i l l be relevant w i l l depend on the circumstances. Further the arbitrator also must take into account the special circumstances of developing 395 countries. A developing country found to have breached its obligations may be given 396 extra time to comply when its needs justify that extra consideration. In most cases, the result has been either implementation or an agreement between the parties as to the ultimate resolution. 3 9 7 It is telling that during the Uruguay Round, the Negotiating Group on Dispute Settlement had saw fit to engage a process referred to as an arbitration in order to deal with what is in essence, an issue of enforcement. Further, while timeframe arbitration addresses an issue that is very narrow, it is hardly the purely factual types of issues contemplated by the proposals in the Uruguay Round. On the contrary, it engages fairly lofty responsibilities for an arbitrator, despite reluctance to make recommendations on the method of implementation. Despite the apparent exclusion of "political factors" as a 398 consideration, timeframe arbitration indeed involves political considerations relating to the feasibility of policy changes with the offending party's domestic system. A s a further example of this, the D S B recently agreed to an amendment to Article 21.3(c) obliging the arbitrator determining the reasonable prior of time to "take into account the particular problems and interests of developing-country Members in interpreting the 394 U.S. Gambling, supra note 381, para. 52 and Canada Patent Term, supra note 382 at paras. 65-66. 395 Article 21.2 of the DSU. 396 However, it is still unclear whether this principle could affect the determination of a reasonable period of time when the aggrieved state is a developing country. See U.S. Gambling, supra note 381 at para. 63. 397 Fukunaga, supra note 48 at footnote 77. 398 Canada - Term of Patent Protection, supra note 382 at para. 60. 123 particular circumstances" that determine the reasonable period of time pursuant to the principle of special and differential treatment. 3 9 9 A s previously set out, not all cases of implementation require legislative changes, but possibly only administrative or regulatory changes. However, while arbitrators have avoided making recommendations on the means of implementation, 4 0 0 they must necessarily assume a method of implementation. Further, it has been noted that in the result, arbitrators have provided a period of time without any explanation for the period u 401 chosen. In short, arbitration under Article 21.3 establishes a precedent for a form of political decision-making through arbitration, even though the decision is not subject to appeal. Given that such a process has been adopted with little controversy, 4 0 2 it is therefore possible to envision a system in which politically difficult cases that are unlikely to be resolved by a binding ruling by a panel and by the Appellate Body, might be systematically diverted out of the litigation track, to an arbitration track. Arbitration would be directed at bolstering the efforts for prompt implementation and therefore ought not be rigid, but should merely facilitate interactions between parties. 4 0 3 The D S B acknowledges that the timeframe order by the arbitrator can be modified upon the 399 WTO, Minutes Of Meeting Special Session of the Dispute Settlement Body (held on 13 July 2006), W T O Doc. TN/DS/M/34, para. 10. 400 Hughes, supra note 176 at 83. 401 Ibid, at 84. 402 Ibid, at 86. 403 See Fukunaga, supra note 48 at 403. 124 agreement of the disputing parties. The main purpose of arbitration would therefore be to facilitate the iterative discourse towards resolution. The same principle could apply with respect to Article 25 arbitrations that involve the main dispute itself. 4.7.2 Concessions Arbitration It is equally telling that during the Uruguay Round, the Negotiating Group on Dispute Settlement agreed upon a second form of "sui generis" arbitration that arises after the D S B ' s adoption of the report of a panel or Appellate Body, and that is directed at the enforcement of the D S U ru l ing . 4 0 5 Once the reasonable period for implementation has been established, the offending state now has a timeframe for making the necessary administrative, regulatory or legislative changes in order to bring its offending measures within conformity to its obligations under the W T O agreements. Where that has not happened within the timeframe prescribed by the Article 21.3 award, the aggrieved state is entitled to mutually acceptable compensation. If the parties do not agree to a form of compensation within 20 days of the expiry of the reasonable period, the aggrieved state can seek D S B approval for a suspension of concessions 4 0 6 The concessions subject to suspension are those benefits that the offending state otherwise enjoys pursuant to the underlying agreements, generally a form of guaranteed and equal access to markets. 404 ibid. 405 It should however be recalled that at the time that both of these provisions were proposed by Canada in June of 1990, the Negotiating Group on Dispute Settlement had not settled on the current panel/Appellate Body litigation system. Nevertheless, it is telling that for enforcement, the Negotiating Group gravitated towards arbitration, and those provisions remained even after negotiations evolved to include automatic adoption of panel decisions, and the creation of appeals to the Appellate Body. 406 Article 22.2 of the DSU. 125 Article 22.3 of the D S U sets out the principles for priorizing the different forms of retaliation. Article 22.4 of the D S U establishes that the proper measure of retaliation is that of equivalence to the level of nullification or impairment, a concept that is different than proportionality. 4 0 7 A s such the measure for retaliation is not purely quantitative, 4 0 8 but to some degree, qualitative. The level of retaliation must be measured against a notional effect on the damaged market. The arbitration under s. 22.6 is thus more complicated and delicate than a mere number crunching exercise. However, it has been suggested that this form of arbitration has utterly failed at determining a reasonable measure of equivalence, or striking any form of rebalance after the trade distorting effects of the breach, due to political considerations. 4 0 9 In determining what concessions wi l l be suspended, the state must comply with the principles in Article 22.3. First, it must consider a suspension of concessions within the same sector. For example, i f the offending measure affected services under the General Agreement on Trade in Services ( "GATS") , the concessions suspended would be within the same division of services, such as financial services within G A T S . Second, where this is not practicable, the aggrieved state must consider a suspension in another sector within the same agreement. Where the first two options are impractical, the aggrieved state can consider suspension in respect of concessions in an entirely different agreement from that under which the initial breach arose (in this example, TRIPS) , or "cross retaliation". 4 1 0 407 Fukunaga, supra note 48 at 419. Am Ibid, at 423. 409 Spamann, supra note 326. 410 For an example where cross retaliation was approved, see European Communities Regime for the Importation, Sale and Distribution of Bananas. Recourse to the Arbitrator by the European Communities under Article 22.6 of the DSU, WT/DS27/ARB/ECU, 24 March 2000. ("£C Bananas ///"). In that case, 126 The arbitration under Article 22.6 arises where the offending state disagrees with the proposed level of retaliation. If the offending party objects to the level o f suspension proposed (as distinct from the form of suspension), it can then seek to have the matter referred to arbitration. The arbitration under Article 22.6 determines only whether or not the proposed suspension goes beyond the level of nullification and impairment of benefits resulting from the original offending measure. The arbitrator can therefore only determine i f the level of suspension is appropriate, and not the "nature" of concessions or obligations to be suspended. 4 1 1 However, where the arbitrator determines that the selection procedure and principles in Article 22.3 have not been followed, the arbitrator is required to give an estimate as to the level of suspension considered to be equivalent to nullification or impairment 4 1 2 In one arbitration decision, the arbitrators went so far as to find that they could make a determination referred to in Article 23.2(a) as to the consistency with the W T O agreements of new measures implemented by the offending states to comply with the panel or Appellate Body report, though that was in exceptional circumstances. 4 1 3 while the violation by the E.C related to goods under G A T T and services under G A T S , Ecuador proposed a suspension of concessions under TRIPS, an entirely different agreement. The arbitration accepted the proposal, while expressing some concerns in respect of Ecuador's choice of TRIPS sectors within which suspensions would apply. 411 Article 22.7 of the DSU. 412 Palmeter & Mavroidis, supra note 29 at 270; "EC Bananas HF, supra note 410. 413 European Communities Regime for the Importation, Sale and Distribution of Bananas. Recourse to the Arbitrator by the European Communities under Article 22.6 of the DSU, WT/DS27/ARB, 6 April 1999. See however Freidl Weiss, ed. Improving WTO Dispute Settlement Procedures - Issues and Lessons from the Practice of Other International Courts & Tribunals (London: Cameron May Ltd., 2000) at 86, where it is suggested that the decision of the arbitrator was influenced by the fact that one of the parties had refused to appear before the panel, and that the finding was ultimately, ultra viries. Thus the position that the arbitrator had the jurisdiction to determine the W T O consistency of the new measures appears to be in some doubt. 127 The offending party's request for arbitration under s. 22.6 would preempt the aggrieved party's request for approval of the D S B under Article 22 .2 . 4 1 4 The arbitrator therefore proceeds without a decision from the D S B with respect to the nature of the proposed suspension. Article 22.6 does not set out any specific procedures for arbitration, other than requiring that the arbitrators shall consist of the original panel members hearing the substantive dispute, 4 1 5 thus removing any party control over the decision-maker. The concessions arbitration must be done within 60 days after the expiry of the reasonable period. The fact that the D S U dictates that the arbitrators shall be the same as those hearing the substantive dispute demonstrates how this form o f arbitration is currently an extension o f the judicial settlement track. Since the arbitrators are generally members of the original panel, the procedures are typically similar to those of the panel process. 4 1 6 For example, the burden of proof for a concessions arbitration is the same that exists in any other D S B proceeding. This burden follows the assumption that parties act in conformity with their obligations and it is therefore for the alleging party to prove a breach. 4 1 7 In concessions arbitration, it is the offending state that must therefore prove that the proposed retaliation is disproportionate. The adoption of this principle of onus proves awkward, as it means that the offending state must prove that the suspension of concessions proposed by the aggrieved state is disproportionate to the nullification or impairment, without information as to the effect on the aggrieved state. Arbitrators have resolved this issue by requiring 414 Palmeter & Mavroidis, supra note 29 at 271. 415 Under Article 22.6 of the DSU, the Director-General would designate replacements for any of the original panel members that are unavailable. 416 Monnier, "Working Procedures", supra note 375 at 514-515. 417 Palmeter & Mavroidis, supra note 29 at 272. 128 parties to submit a preliminary argument outlining the methodology for the aggrieved 4 1 8 state's calculation of the concessions. A s of March 2007, there have been 9 awards circulated in respect of Article 22.6 arbitrations, 4 1 9 eight of which resulted in approval of retaliation 4 2 0 In one respect, the concessions arbitration sets an example for one feature that might distinguish arbitration from litigation in a proposed two-tiered dispute settlement system. Like the process under Article 21.3, there is no provision for any appeal from this decision, and the parties are required to accept the decision as final, and cannot seek a second arbitration 4 2 1 The D S B must be promptly informed of the results of the arbitration, but while there is no adoption process for the D S B , the aggrieved party must still seek the D S B approval of suspension in the amount approved by the arbitrator. Article 22.6 thus sets another precedent for a system where a right to appeal does not attach to every W T O dispute, even where the arbitration may not have all the advantages attendant on arbitrations in the traditional sense, such as choice o f the arbitrator. A review of the use of concessions arbitration however reveals a second important point. Arbitrators hearing Article 22.6 arbitrations may not have applied the wholly legal analysis suggested by the formal principles that have purportedly developed. It has been suggested that the decisions under this form of arbitration have typically sought a middle 418 See e.g. European Communities Regime for the Importation, Sale and Distribution of Bananas. Recourse to the Arbitrator by the European Communities under Article 22.6 of the DSU, supra note 413. 419 Leitner and Lester, supra note 3 at 174-175. 420 Wilson, supra note 7 at 399. 421 Article 22.7 of the DSU. Some have suggested consideration of making awards subject to review by the Appellate Body. See e.g. Steger, "Systemic Issues" supra note 6 at 73. 129 ground between the parties' positions, and thus consistently present diplomatic solutions. 4 2 2 I do not view this as a negative aspect of this form of arbitration. Robert Hudec has suggested that where an award has the effect o f providing a reasonably objective ruling, it may well "persuade the relevant audiences in both countries that a neutral tribunal had made an objective judgment of equivalence". 4 2 3 While speaking particularly of the potential political benefits of the use of arbitrary measures in Article 22.6 arbitration, Hudec's observation speaks volumes about the potential use of arbitration for establishing a single unappealable ruling, within the W T O context generally. Therefore, while concessions arbitration has been criticized in cases of non-compliance as merely prolonging the dispute, 4 2 4 i f it has been used in this manner, it may well set an example for the use of arbitration of the underlying dispute in the more politically difficult cases. 4.7.3 Issue Arbitration The third form of arbitration in the D S U is found under Article 25, which provides for arbitration for any issue that is clearly defined by both parties. A s discussed in the previous chapter, some G A T T members believed that it was always open to G A T T parties to agree to a form of arbitration for the settlement of particular disputes, given the absence of any unifying and mandatory dispute settlement system. 4 2 5 B y this perspective, the option presented in Article 25 only became necessary by the inclusion of Article 23 of 422 Spamann, supra note 326 at 75-76. 423 Hudec, "Broadening the Scope", supra note 160 at 391. 424 Fukunaga, supra note 48 at 385. 425 Young, supra note 43 at 401. 130 the D S U , requiring the parties to submit any allegation of a violation giving rise to a dispute in accordance with the D S U procedures. 4 2 6 Nevertheless the formal recognition of a procedure that acts as an alternative to the mainstream judicial system is a significant step, irrespective of the manner of its introduction into the D S U . In effect, arbitration as a true alternative has been integrated into the institutional context. While it is a separate procedure, it proceeds within the general auspices of the D S B and has a degree of institutional support. This is unlike the previous G A T T system, where there was not even a consensus amongst members as to the availability of arbitration. Issue arbitration under Article 25 more closely resembles arbitration in the classical sense. It can be used to resolve any issue, and does not require a previous decision from a panel or the Appellate Body like arbitration under Articles 21.3 or 22.6. The arbitrator arguably exercises relative independence, as the award does not require any formal adoption or approval by the D S B . 4 2 7 Indeed its flexibility has left it open to criticism from legalists who have wondered what contribution arbitration can make to the overall 428 development of a consistent and cogent body of law. While this is the only form of arbitration that exists as a true alternative to the litigation processes before panels and the Appellate Body, it has so far only been used in one case. The case of United States-Section 110(5) of the U.S. Copyright Act {"U.S. Copyright")429 involved an E . C . complaint that the United States Copyright Act failed to protect the 426 ibid. 427 Article 25(3) of DSU only provides that awards must be notified to the DSB. 428 Young, supra note 43 at footnote 57. 429 U.S. Copyright, supra note 369. 131 exclusive copyrights of the E . C right holders of music, thus causing a loss of royalties. Ultimately, the Appellate Body found that section 110(5) of the United States Copyright Act breached Article 13 of TRIPS, and therefore recommended that the D S B request the U.S . to bring its law into conformity with its obligations under TRIPS. Ironically, the case used the Article 25 procedure for resolving the level of nullification and impairment, and thus did not stray far from the parameters of subject matter of concessions arbitration. The arbitrator was asked to calculate the level of E . C . benefits that had been nullified or impaired as of the date of the arbitration, and thus involved a very factual issue, though not completely divorced from the application of legal principles. The particular issue was whether or not it was reasonable for the E . C . to calculate its losses for all potentially realizable income. The U.S . argued that it would be impossible to implement and enforce a licensing system for collecting royalties for every conceivable rights user (i.e. use of copyrighted music in restaurants and bars) and that any calculations had to recognize the practicalities and costs for collecting license fees in respect of any potential use. 4 3 0 The arbitrator ultimately determined that the level of impairment was the amount that could have reasonably been collected through licensed users only, thus rejecting the E . C . methodology of calculating the potentially realizable income without regard to the practicalities of licensing and collection. 4 3 1 The ultimate calculation based on lost profits marked a lack of consistency with decisions in s. 22.6 arbitration, which have typically used the measure of trade effects, or lost trade. 4 3 2 430 U.S. Copyright, supra note 369 at paras. 3.6-3.35. 431 Ibid, at para. 3.33. 432 Spamann, supra note 326 at 40-41. 132 Whereas Article 25 seems to be broad enough to cover almost any issue, this was not the most inventive use of issue arbitration. The issue defined by the parties could have been determined by an arbitrator without a request under Article 25, under Article 22.6. Article 26.1(c) would have also permitted the timeframe arbitration to address such a matter, upon the request of one of the parties. 4 3 3 There are no specific procedures set out in Article 25. Article 25.2 explicitly provides that the parties must agree to the procedures and to the arbitrator. Here again, the parties involved in the U.S. Copyright case did not demonstrate a great deal of creativity and elected to have the original panel that decided the substantive dispute. A s such, it did not demonstrate the full potential or breadth o f choice over the decision-maker. When it was determined that two members of the original panel were unavailable, the Director General appointed replacements, in accordance with a set of specific working procedures circulated ahead of time 4 3 4 The arbitrator in U.S. Copyright determined that it could consider the question of its own jurisdiction 4 3 5 In finding that it had the jurisdiction to resolve the dispute, the arbitrator made particular note of the fact that the dispute did not affect the rights of any other members. 4 3 6 433 Monnier, "Time to Comply", supra note 383 at 842. 434 Hughes, supra note 176 at 81. 435 U.S. Copyright, supra note 369 at paras. 2.1 - 2.7. 436 Ibid, at paras. 2.6 - 2.7. 133 The parties requested the use of principles adopted in arbitration for Article 22.6, which was not surprising given that the substantive issue resembled one for concessions 438 arbitration. The specific working procedures adopted in the arbitration largely followed those used under previous concessions arbitration, 4 3 9 and required the U .S . to submit a preliminary paper outlining the methodology used to calculate the benefit and impairment to the E . C . The parties then submitted concurrent written submissions and concurrent rebuttals. 4 4 0 The procedures ultimately adopted bore a substantial resemblance to those used within panel proceedings, with some modification. 4 4 1 The arbitration in this case demonstrates the potential of the arbitration process, as it had the effect of providing guidance to the parties for an agreement as to compensation. 4 4 2 While the award may have not provided the precise objective solution, the parties eventually reached an agreement on the appropriate amount of compensation based on the award. 4 4 3 Indeed it has been noted that the manner in which the arbitration facilitated negotiation on the issue was "an interesting and constructive precedent, likely to be followed in the future." 4 4 4 While I share this enthusiasm for the prospect opened up by this experiment, unfortunately it has not been followed as a precedent. 437 Ibid, at para. 4.4; See also Spamann, supra note 326 at 71. 438 Hughes, supra note 176 at 81. 439 Pierre Monnier, "Working Procedures", supra note 375 at 512-3. 440 U.S. Copyright, supra note 369 at para. 1.7. 441 Pierre Monnier, "Working Procedures", supra note 375 at 513. 442 Fukunaga, supra note 48 at 412. 443 Ibid, at 414; Communication from the U.S. WTVDS160/23,26 June 2003; Spamann, supra note 326 at 79. 444 McGivern, supra note 9 at 157; see generally Bronckers & van den Broek, supra note 115. 134 4.8 Conclusion There are a number of aspects of the current use of arbitration under the D S U that may inform the discussion of the further integration of arbitration. First, there is still an incoherency in the integration of arbitration as an alternative with the D S U . Indeed, two of three forms of arbitration in the D S U are merely extensions of the judicial settlement system and hardly arbitration at all . On the other hand, arbitration under Article 25 is more of a self-contained alternative dispute mechanism, even though it has been used only once. This incoherency reinforces the notion that the shaping of the arbitration alternative was not fully developed in the Uruguay Round. This is not a striking revelation. Cleary the creation of the W T O and an effective dispute settlement system involving over 100 member states is itself remarkable. It was surely a first cut at the creation of a system that was designed to be refined over time and after experience. This of course does not mean that arbitration cannot be integrated more effectively into the D S U or that it cannot have some utility. It ought not be forgotten as an alternative to the more legalistic system of judicial settlement. However, this incoherence may have obscured the true benefits of arbitration as a possible alternative for specific disputes. It is difficult to view it as a true alternative when it is not carefully defined conceptually, or i f it is not clearly distinguished from the judicial settlement system of dispute, at least in practice. 135 Second, the current structure of the D S U suggests that there is nevertheless a belief that arbitration, in some form, has a role to play, perhaps an important role, particularly in respect of the ultimate resolution of the dispute and implementation of some form of change in the event of a violation of W T O principles. The two forms of mandatory arbitration at least add an extra and distinct layer to the judicial settlement system. The arbitration provisions may well be ideal as a useful tool within the D S U as an example of a modified rule system designed to promote the discourse necessary to facilitate a resolution. The strength of the arbitration provisions is not to create a formal system o f enforcement that mirrors the private commercial arbitration system but rather in fostering further iterative discourse that converges around treaty principles, as contemplated by the Chayes managerial model. They can do so by fostering a discourse on how and when the offending party w i l l alter its conduct to conform to its obligations, and by providing some objective measures for the value or cost of non-compliance. These forms of arbitration do not overreach by providing an appeal right on legal issues. Furthermore, where compliance does not follow a D S B ruling, the arbitration process provides another process to determine the empirical value to non-compliance. A s such, the secondary rules creating the arbitration process provide the framework for enhancing the legitimacy of the D S B rulings in and of itself, and for reinforcing the idea of objective standards, thus engaging the element of reputation as a pressure towards compliance with rulings. Section 21.3 and 22.6 can be seen as forms of expression that 136 advise the parties clearly that they must ultimately comply with the finding of the D S B . 4 4 5 This provides an example of the potential benefits of using Article 25 arbitration to resolve more politically difficult cases. Third, while some of the discussions in the Uruguay Round suggested that arbitration ought to be limited to only factual disputes, the use of processes under Articles 21.3 and 22.6 demonstrates that arbitration can, and has been used as a means of resolving political issues, without causing great concerns. Even though these sui generis forms of arbitration may not be true forms of arbitration, they have set a precedent for member states to accept a system of dispute resolution that does not provide for a right of appeal in every instance. This last point leads to the contention in the next chapter - that arbitration might be seen as a mechanism that falls somewhere between the extremes of diplomatic means and legalistic means of dispute settlement, and thus worthy of consideration as part of the solution of addressing the most politically difficult disputes. The creation of a distinct arbitration system that is independent of the judicial system, and that gives the parties some measure of control over the procedure may well provide the requisite legitimacy, while promoting negotiated resolution. From the perspective of the offending party, it w i l l have been provided with an opportunity to fully argue the merits of its case before an independent third party, thus promoting Franck's key elements of legitimacy and 445 Young, supra note 43 at 405. 137 fairness. That process would also allow an arbitrator, chosen by the parties, to consider the reasons required for a certain period to implement a ruling. The idea of using arbitration as a middle ground between the diplomatic and legalistic means of resolving disputes is premised on the simple notion that judicial settlement w i l l be incapable of inducing compliance in every case. While arbitration could clearly not lay such a claim, it may be a more suitable form of dispute settlement over the judicial settlement system in difficult cases, as explored in the next chapter. 138 CHAPTER V: REFORM 5.1 Introduction While the commencement of the Uruguay Round in 1986 may have marked an appropriate time to strengthen the dispute settlement system, the pendulum-like movement of G A T T between rules orientation and power orientation may suggest that this is now a time to consider "weakening" the system in some small measure, in order to strike a better balance for politically difficult cases. The question of reform must therefore be considered in light of the delicate balance that the D S U attempts to strike between the legalization of the system with its emphasis on "enforceable" remedies, and the purpose of engaging political pressures. Compliance theories suggest that the utility of objective decisions is primarily as a force toward normative condemnation and mutual resolution, and not to obtain enforceable judgments. This is a major difference between compliance in international law and compliance in domestic law. A s the court without a bailiff, the D S B is unable to truly oversee the actual enforcement of monetary judgments or retaliation. 4 4 6 Arbitration, as distinct from judicial settlement may be uniquely positioned to achieve the right blend of these elements in a few, select cases. In this chapter, I first consider the current direction of reform, and then propose a vetting system that seeks to identify the 446 It has been said that W T O rulings are not binding in the "traditional" sense since there are no traditional enforcement powers such as a police force or injunctive relief. See Judith Hippler Bello, "The W T O Dispute Settlement Understanding: Less is More" (1996) 90 Am. J. Int'l L. 416, though further explained in (2001) 95 Am. J. Int'l L. 984 at 986-987. See also John H. Jackson, "International Law Status of W T O Dispute Settlement Reports: Obligation to Comply or Option to Buy Out" (2004) 98 Am. J. Int'l L. 109 at 123. As Jackson concedes, the remedies of compensation and suspension are in some ways "deeply flawed" and "dysfunctional". 139 politically difficult cases, and that diverts them outside the judicial settlement system and directs them down an arbitration track. Given the relatively good record of dispute resolution, it is anticipated that this would be a minority of cases, which could generally involve disputes between developed states, particularly the superpowers. One might naturally ask how an arbitration ruling would induce better compliance than a ruling in the judicial settlement system? After all , the judicial system was created around an appeal system. From the U .S . perspective, the appeal system was intended to foster compliance with the new obligations negotiated in the Uruguay Round, whereas from the E . C . perspective, it was hoped that it would curb U .S . unilateralism. 4 4 7 Ultimately, I do not suggest that an arbitration award wi l l have any additional coercive force than a D S B ruling. However, I would suggest that for the purposes of politically charged cases, it would not have any less. Second, I do not suggest that that such a system would result in better compliance with the strict letter of the award or ruling. However, it would have the advantage of diverting a dispute where compliance with a ruling is unlikely in any event, putting it more quickly on a negotiation track, after the benefit of an objective third ruling around which the iterative discourse can take place. Rather than just a binding decision, the judicial settlement produces a decision which is a purportedly objective statement of obligations, and legally unimpeachable, leaving little room for any further bargaining or discussion. This can be counterproductive in a dispute that is politically charged. Conversely, there is 447 Steinberg, supra note 2 at 250. 140 more negotiation room after a single arbitration ruling, as opposed to a decision within the institutionalized panel system with legal appeals that further entrenches the negotiating position of the disputing parties. The proposal is thus intended to prescribe a more appropriate form of dispute settlement based on the nature of the dispute, considering the theories of compliance and the objectives of the D S U . 5.2 Surveying the Debate on DSU Reform The current debate regarding reform of W T O dispute settlement illustrates three important points. First, the discourse over reform demonstrates a bias towards increasing the legalism of the system, despite a recognition of the limitations of that system. Secondly, after 12 years since the creation of this formal legal system of dispute settlement, there continues to be some ambivalence as to the ultimate objective of the dispute settlement system. Reform proposals tend to be directed at either creating more rules for governing litigation and remedies, or to negotiations. None seem to attempt to blend these two objectives. The use of the arbitration may be the logical intersection. Third, a discussion on the potential for arbitration has been virtually absent from this debate. This is despite the fact that arbitration has been historically identified as an alternative means of resolving disputes, and is more suitable for certain types of disputes. 141 5.2.1 Fixation on Compensation and Remedies Almost two years after the passing of the original deadline for negotiations on dispute settlement reform in May of 2003, the Consultative Board appointed by the W T O Director-General Supachai Panitchpakdi released the Sutherland Report . 4 4 8 The issue of effective reform of the D S U has continued to be a source of debate. This discourse has focused mostly on procedural "improvements" and broader enforcement mechanisms, usually centering on compensation to the complaining party. 4 4 9 A s a whole, the reforms that are generally proposed are directed at making the dispute settlement system more judic ia l - l ike . 4 5 0 The reforms proposed by the Sutherland Report included a recommendation of experimentation with damage awards where the offending state had failed to comply with a D S B ru l ing . 4 5 1 The primary argument for the use of damage awards is that it is a more accessible form of remedy for developing countries that have less economic ability to exercise any rights of retaliation. 4 5 2 For this reason, some have proposed that developing countries should be able to choose monetary compensation. 4 5 3 It has been suggested that 448 The Sutherland Report, supra note 17. 449 See e.g. Fukunaga, supra note 48 at 413-414. 450 Davey, "Looking Forwards", supra note 11 at 19. 451 The Sutherland Report, supra note 17 at 50 (para. 223). 452 Steger, "Systemic Issues", supra note 6 at 69. 453 William J. Davey, "The Sutherland Report on Dispute Settlement - A Comment" (2005) 8 J. Int'l Econ. L. 321 at 321-3 [Davey, "The Sutherland Report"]. 142 specific compensation for the affected sector of the complaining state is in any event more consistent with the goals of the W T O . 4 5 4 The current dialogue of reform has also touched on the more specific concept of retroactive damages. This remedy is supported on the basis that it would discourage the "hit and run" form of discrimination - designed to affect the exporting country for a short period of time in a peak season. Conceptually the tactic protects the domestic market during a high season. B y the time the complaint proceeds, it has become a moot issue. While retroactive damages might provide some disincentive for the "hit and run" tactic, it can also provide more incentive for a complaining party to pursue the matter to a final decision, rather than considering negotiation. 4 5 5 Another recommendation that has been proposed is a form of fine that would escalate over the passage of time. The escalating fine could be in accord with the size of the economy of the particular party. 4 5 6 Another suggestion has been that of collective and punitive retaliation - retaliation in which all of the members would participate. However this form of retaliation would only compound the commercial impact on a broader group of innocent bystanders. 4 5 7 However these solutions suffer from the same limitations as proposals for compensation. Enforcement measures are unlikely to improve a situation 454 Nzelibe, supra note 341 at 242. 455 See Porges, supra note 27 at 178. 456 Matsushita, supra note 327 at 94. See also Davey, "Looking Forwards", supra note 11 at 22-23. 457 McGivern, supra note 9 at 156. 143 where they have no effect on politicians, or where responding party has failed to implement a recommendation for political reasons. 4 5 9 While recognizing the emerging problem of compliance, the Sutherland Report itself focused on the alternatives of monetary compensation and the potential of cost awards. In doing so, it acknowledged the obvious danger that the use of monetary compensation would allow developed countries to simply "buy out" of their obligations, contrary to the principles found in Article 3.7 of the D S U . Nevertheless, it concludes with a recommendation for "experimentation" by allowing the complaining party to substitute monetary compensation with compensatory market access, while proposing that the D S B should exercise great care to ensure it is only used as a temporary fallback. 4 6 0 5.2.2 Procedural Legalism The Sutherland Report's predisposition to legalism in the D S B goes beyond the discussion of compensation and other monetary measures. It is also apparent in the procedural recommendations, such as a permanent roster of panel members with expertise on procedural and evidentiary issues 4 6 1 and increasing the authority of the 458 Nzelibe, supra note 144 at 229. 459 Cho, supra note 312 at 787-788; see also Benjamin L. Brimeyer, "Bananas, Beef and Compliance in the World Trade Organization: The Inability of the W T O Dispute Settlement Process to Achieve Compliance from the Superpower Nations", (2001) 10 Minn. J. Global Trade 133 at 162 for discussion on the role of politics in W T O compliance. 460 The Sutherland Report, supra note 17 at 54. 461 Davey, "Looking Forwards", supra note 11 at 21. 144 Appellate B o d y . 4 0 2 In the most difficult cases these reforms could simply increase procedural wranglings, compound the possibility of endless appeals, and prioritize the quest for the objectively correct decision, on the assumption that this w i l l induce compliance. While a decision that is "objectively" and legally correct may be very useful in most cases, it has limited benefit where political considerations would mean that the unsuccessful party fails to implement D S B recommendations. While such a decision may well create a precedent for the interpretation of the specific W T O obligation for other situations in future cases, a party's failure to implement the ruling would likewise create a precedent and justification for the non-observance of obligations in other cases. A s Joost Pauwelyn points out, the commentary in the Sutherland Report, also reveals a partiality towards judicial activism over consensus bui ld ing. 4 6 3 This qualified endorsement o f judicial activism appears to confirm the institutional inertia towards legalism and a preference for a de facto system of stare decisis. The Report itself demonstrates an interesting ambivalence with respect to the issue of the precedential value of the decisions of the Appellate Body. While acknowledging that the principle of stare decisis does not apply, the Report trumpets the virtues of jurisprudence of the Appellate Body as being "extraordinarily rich and detailed for a body in existence only ten years". 4 6 4 It thus provides a qualified approval of the creation of a body of law to provide authoritative interpretations that w i l l resolve or obviate disputes between parties in future matters. 462 See e.g. Steger, "Systemic Issues", supra note 6 at 68-69; Davey, "Looking Forwards", supra note 11 at 23, for discussions regarding the power to remand dispute back to the Panel system and thus elongate the litigation even more. 463 Pauwelyn, "A Missed Opportunity" supra note 165 at 344-5. 464 The Sutherland Report, supra note 17 at 51. 145 The Report also refers to a "gap-filling" responsibility with respect to the law. While the Report notes that it might not be an appropriate role for the D S B , it alludes to the lack of successful negotiation at the W T O , and to the hope that the Doha Round wi l l "correct the imbalance between law-making and any tendency toward creative law enforcement through the dispute settlement system." 4 6 6 This suggests that in the absence of sufficient negotiated change within the diplomatic realm, that the D S B might continue the practice of gap-filling. The Report ironically suggests that such improvements may provide a disincentive to the over use of the D S B . 4 6 7 This approach, taken to certain lengths, would violate the principle under Article 3.2 of the D S U that D S B rulings cannot add to or detract from the rights or obligations provided in the underlying W T O agreements. 5.2.3 Direction of Sutherland Report The proposals in the Sutherland Report and the surrounding debate on reform fail to adequately address the political reasons for non-compliance. It also forgets or ignores that weaker remedies were used as a precondition for a stronger dispute resolution system, 4 6 8 and thus departs from the delicate balance struck between the political and legal aspects of the W T O 4 6 9 The limitation of remedies may be the price of the legalized institution, as 465 See e.g. Steinberg, supra note 2 at 251-252, for discussion of the practice of gap-filling. 466 The Sutherland Report, supra note 17 at 55. 467 ibid. 468 Pauwelyn, "A Missed Opportunity", supra note 165 at 344-345. 469 Ibid, at 338. 146 it was contemplated that more onerous remedies were not necessary because of high level of procedural development in the D S B . 4 7 0 The Sutherland Report does however concede that compliance may ultimately depend on the attitudes of members and not improved remedies 4 7 1 While this may be considered by some to be an unsatisfactory commentary, it is perhaps one o f its more realistic observations, as the question of reform and the use of intrusive remedies has become as much a political question as a legal one. 4 7 2 It acknowledges that further legalism and "improved" remedies may not be the answers. 4 7 3 Remarkably, the Sutherland Report does not go any further in analyzing the use of the negotiation provisions already in the D S U in order to enhance the framework within which attitude change may be realized. This is all the more ironic in light of the explicit content of Article 4.1 that affirms the members resolve to improve the effectiveness of the consultations procedures. The utility of arbitration under Article 25 is not addressed in the Sutherland Report, and has been all but ignored within the negotiations over reform within the Dispute Settlement Negotiation Committee of the W T O . A proposal of Australia from 2003 contemplated the use of a mandatory form of expedited arbitration under Article 25 upon the request of a third party that would "serve to determine the right of a third party to negotiate compensation, as well as the level o f that compensation." 4 7 4 Others have suggested the use of Article 25 for the purpose of determining compensation, following 470 Vazquez & Jackson, supra note 22 at 562. 471 The Sutherland Report, supra note 17 at 54-55. 472 Vazquez & Jackson, supra note 22 at 565. 473 Davey, "The Sutherland Report", supra note 453 at 324. 474 Communication From Australia, WTO Doc. No. TN/DS/W/34 (15 January 2003). 147 the precedent of the U.S. Copyright case . 4 0 While these of course are worth considering, there does not seem to be any consensus on the appropriateness of compensation. While text enhancing compensation was included in the Chairman's draft of 16 M a y 2003, it does not appear to be a priority in current negotiations. Establishing a level o f compensation as a form of remedy for developing countries in politically difficult cases may be a useful suggestion. However, as mentioned from the outset of this chapter, the vetting system with the arbitration track would be directed at the politically difficult cases, most of which have involved developed countries. This overall approach of the Sutherland Report fails to recognize that while improvements in procedural rules can indeed develop the general framework that w i l l assist in bringing issues to a head and contribute to the resolution of disputes, rules cannot cover everything, particularly in international law. A t best, it can be hoped that they wi l l induce the preferred behaviour. 4 7 6 None of the reforms discussed directly address the problem of compliance in the context of the changing political pressures in the international arena. Where there are no direct political implications, reforms cannot lead to any significant changes to a level o f compliance. 4 7 7 The failure to account for political sensitivities w i l l inevitably lead to more decisions where compliance is simply not forthcoming. The absence of any recognition in the Sutherland Report of this problem is perilous, as it ignores the inherent fragility of the D S B . 475 See e.g. Hughes, supra note 176. 476 Zampetti, supra note 89 at 123. 477 van den Broek, supra note 70 at 155-6. 148 5.2.4 Recommendations for Reform of Consultations Others have suggested that formal remedies are only one aspect of the legal enforcement of W T O decisions. In the face of the growing legalism of the W T O system, some have emphasized that consultations and negotiations are still the "bedrock" of the system despite remaining relatively unchanged since the transformation from the G A T T system 478 to the W T O . B y this perspective, effective reform measures must serve to induce the participation o f both disputants in the actual resolution. The backlash against unilaterally legalistic reform emphasizes the incentives or pressures for early settlement. This is a positive development i f one accepts the research that indicate that early settlement results in the better policy outcomes overa l l . 4 7 9 A negotiated settlement before hearing can be more beneficial than a decision with perfect compliance with a legal ruling, as it provides each of the parties more control over the specific outcome. 4 8 0 One suggestion for improving negotiations is to allow any state to unilaterally engage mediation facilities referred to under Article 5 of the D S U as a matter 481 of right. Another suggestion that has been made is to improve participation of the parties at the consultations stage. While the consultations stage is a mandatory process for initiating any dispute resolution within the W T O , it has largely become a pro forma 478 Busch & Reinhardt, "Fixing What Ain't Broke", supra note 302 at 9. 479 Though this is not necessarily so for disputes involving developing countries. See Busch & Reinhardt, "Evolution of G A T T / W T O " , supra note 297 at 144. 480 Cf. Porges, supra note 27 at 174. 481 See Pauwelyn, "Americanization" supra note 46 at 135, where Pauwelyn suggests that Article 24.2 of the DSU that provides the least developed states with a right to unilaterally engage the mediation and good offices facilities ought to be extended to all developing states. Pauwelyn does not mention developed states, but there appears to be little reason why a developed state should not be able to engage the process unilaterally against another developed state. 149 exercise, in some cases, lasting no more than an h o u r . w Yet this initial process can be a very useful tool. It provides an opportunity for parties to reconsider initial positions, before too much time and political effort is invested in a litigation process, thereby embedding their positions. 5.3 The Potential Utility of Arbitration - A Framework for the Future While the recommendations and efforts for bolstering the consultations process is a positive development, it nevertheless represents a current thinking to consider reform along two poles - diplomatic resolution on one side and the legal on the other. Indeed it is difficult to attempt a blend of these aspects. On the one hand, the judicial system has worked relatively well in most cases. Injecting the legal system with political considerations could well undermine the legitimacy of the entire system. On the other hand, as emphasized throughout this thesis, obtaining an objectively "correct" interpretation wi l l not assist where political considerations make such a resolution unrealistic. More importantly, effective reform must find way of maintaining the ultimate 483 tool for enforcement of international rulings - community pressure. I propose that arbitration is a form of dispute settlement that may fall somewhere between the extremes of Jackson's rule orientation and legal orientation, thus representing a third option that deserves to be explored. 482 McRae, supra note 296 at 9; Porges supra note 27 at 160. 483 Hudec, "Broadening the Scope", supra note 160 at 400. 150 Given the current bias towards the judicial settlement system and the underuse of Article 25 arbitration, the integration of arbitration would require a vetting process to identify the policy charged cases where compliance problems are more likely to arise. Once identified, these difficult cases could be automatically diverted to the arbitration track. The arbitration track would exploit the main advantages of arbitration within the W T O -the lack of a legal appeal, and the control over procedure and decision-maker. Although lacking any appeal, arbitration nevertheless creates a process for obtaining an objective ruling with legitimacy, but which also gets the disputing parties more quickly back on track for negotiations in the shadow of concerns over reputation. A s discussed in Chapter Two, this institutional process of discourse need not revolve around a process that seeks incessantly to produce legally impenetrable declarations of legal obligations. The presence of a ruling itself can set the stage for further discourse as to how the losing party w i l l respond, while at the same time, engaging the scrutiny and communitarian pressure of the W T O member states. While arbitration takes place outside the formal litigation process, and is not subject to appellate review, the parties must notify the D S B of the outcome, and any D S B member may raise any point relating thereto. 4 8 4 One limitation with the judicial settlement system is that it encourages parties to think solely in terms of obtaining an enforceable ruling. If a panel's decision is perceived as "wrong", it encourages further appeal for a more favourable judgment to secure or avoid a remedy of compensation or retaliation, as the case may be. Conversely, an arbitration, 484 Article 25.3 of the DSU. 151 while purporting to give an objective answer, gives limited protection against a decision that is "wrong". The pressures to comply with the exact letter of the decision are lessened somewhat where there is no appellate review, and where the correctness of the decision might still be the source of some discussion after the award. Even a decision that does not withstand the analytical scrutiny of an appeal gives the losing state a justification to its domestic constituencies for considering difficult policy change. Ultimately an arbitration in specific cases would simply seek to "broaden the space for political debate" 485 within the dispute settlement system in these difficult cases. Decisions through arbitration would still provide the benefit o f a third party interpretation, a further basis for iterative discourse, and the engagement of domestic political factions. Diverting select cases from the judicial settlement system would have the further benefit o f directing that discourse towards negotiated resolution much more quickly, precluding parties in such politically charged cases from using the legal system 486 to delay the inevitable discussion around treaty norms. It would thus reduce the protracted legal maneuvering and potentially endless litigation, which has its role in cases where compliance is a possible or likely outcome, can be counterproductive where implementation is unlikely for political reasons. A t that point, the offending state may 487 have entrenched its position of non-compliance with its domestic audience by fighting each fight after its original non-compliance, investing both time and expense, and making 485 Pauwelyn, "A Missed Opportunity", supra note 165 at 340. 486 See supra note 48. 487 See Pauwelyn, "Americanization", supra note 46 at 125 & 127; See also Cho, supra note 312 at 787-788. 152 policy reversal less likely.'* 6 8 This is of course the danger of a litigation track with full appeal rights, when there are political pressures internal to the state party that mitigate against compliance. Further in many cases, the impact of the decision itself may impel a government to appeal. 4 8 9 Rather, it would be important to provide the parties with only limited access to third party decision-making, and proceed more quickly to negotiations that reflect on the objective ruling of an arbitrator. Some might question the legitimacy of a decision of arbitrators that does not withstand analytical scrutiny, 4 9 0 particularly that of an appeal process. However, assuming Franck's legitimacy of the dispute resolution system is a key influence on the compliance pull , preserving a right o f appeal is not the only way of ensuring legitimacy or procedural fairness. The control over procedure provides a form of legitimacy to compensate for the absence of any appeal. Article 25 provides the parties with an extra element of consent over the decision-maker and the procedures to be followed. A s a general principle, the prospect of compliance may be greater where the parties have greater control over the process. 4 9 1 In these politically difficult cases, the scrutiny of an appeal process to ensure the correctness of the decision can be unhelpful. The dimension of control over the decision-maker and the arbitral procedure provides states with more reason to both accept and defend the legitimacy of the process, even where it more dramatically affects internal policy in sensitive areas. The parties' ability to 488 Porges, supra note 27 at 168. 489 Ibid, at 169. 490 McGivern, supra note 9 at 151. 491 Cf. Porges, supra note 27 at 174-5, in relation to the prospect of compliance after settlement. 153 select the arbitrator would have some advantages over the current panel system. First, the DSB could adopt a form of arbitration where each party is entitled to select one of the board members, who would then be require to appoint a third party by agreement within a short designated period of time. This system could allow the parties to select an arbitrator from its own jurisdiction, who may have a deeper understanding of the political challenges within that particular legal system. This is an advantage that cannot be met within the panel system, as it does not permit the appointment of individuals from either of the disputing states. The type of experience that is useful in the most politically difficult cases may not be extensive knowledge of treaty obligations, D S U procedures or the application of principles developed from previous decision. Rather, experience in the political system of the disputing parties may be the most important form of experience. 5.4 Integration of Arbitration Through Dispute Diversion In order to improve compliance for politically sensitive cases, it is important to provide different procedures to identify and deal with those cases differently. However, even i f one accepts that arbitration can be the middle ground between political negotiations and judicial settlement, there is another important question: why would diversion to Article 25 arbitration be accepted by the W T O members, given that its infrequent use suggests that they do want to use it? The logical explanations for the infrequency of use do not suggest at any insurmountable hurdle to the further integration of Article 25 arbitration in politically difficult cases. 154 Valerie Hughes, a former Director of the Appellate Body Secretariat has described the member state's failure to take advantage of the flexibility of Article 25 arbitration as "curious". 4 9 2 In offering possible explanations, 4 9 3 she first suggests that matters that are distinct or narrow bilateral issues, "for which arbitration is best suited", are few and far between. However, the notion that arbitration may be suitable for other types of disputes may not have been ever fully considered by the member parties o f G A T T or the W T O . Secondly, she refers to the fact that awards may not be grounded in legal principles. Last, she refers to the lack of appeal process. 4 9 4 There may have been little incentive for responding parties to agree to arbitration for these reasons, and thus no prospect for mutual agreement to arbitration. These explanations for the member states' lack of motivation to utilize Article 25 arbitration undoubtedly ring true, likely arising from the momentum of bias towards the judicial settlement system. However, i f there is indeed any widespread vision of the utility of arbitration restricting it to simple "factual" disputes, this limited vision developed at a time before the experiment with a fully developed, institutionalized system for judicial settlement proved to be inadequate for certain disputes. There is no reason to restrict arbitration to purely factual disputes as contemplated by some states in the Uruguay Round. A s a recent review o f international arbitrations has noted, state to state arbitrations have addressed "a wide range of disputes, from controversies over border and damage to property during wars to collision between ships at sea". 4 9 5 492 Hughes, supra note 176 at 85. 493 ibid. 494 ibid; see also Pauwelyn, "Americanization", supra note 46 at 138. 495 Posner & Yoo, supra note 8 at 9. 155 Further, the widespread acceptance of the timeframe arbitration process under Article 21.3 suggests that member states recognize the legitimacy of the arbitrator's role, despite the inevitable intrusion into political sovereignty. A s has been seen by the use of arbitration processes under Articles 21.3 and 22.6, member states have accepted these proceedings despite the absence of an appeal process. It may now be recognized that in politically difficult cases, a formal legal framework with a protracted legal appeal process but no formal coercive force is unlikely to result in a resolution that w i l l be will ingly implemented by a losing party. 5.5 A Politicization of the DSU and Return To GATT? This proposal leads to an obvious question - how would the arbitration track be any different than G A T T , and thus affect a re-politicization of the D S U ? While this is natural question, it should first be recognized that G A T T was fairly successful in terms of international tribunals, and indeed, according to some research, just as successful as the W T O . 4 9 6 As suggested in Chapter 3, the G A T T system was able to deal with more and more politically sensitive cases, and had some success. 4 9 7 O f course, the number and diversity of interests of member states has changed dramatically since its period of success. More importantly, amongst its deficiencies, was a problem that it shared with the current system - it represented a one-size fits-all form of dispute settlement. 496 ibid. 497 Ibid, at 48. In their review of the differences in compliance rates between G A T T on the one hand, and the W T O between 1995 & 2000 on the other, Posner and Yoo conclude that the differences are not statistically significant. 156 While the arbitration track directed at a negotiated resolution proposed would in some ways resemble the old G A T T panel system, I do not propose a return to the G A T T system for any class of dispute. First, irrespective of the decision of the Secretariat as to the appropriate track for dispute, parties could not block the establishment of a panel or arbitrator, and thus avoid the issuance of an objective ruling by a third party, as it could before 1989. Where a settlement is not reached in the consultations stage, the parties would have, as a matter of right, access to a proceeding for obtaining a ruling of a third party, either through judicial settlement or through a form of arbitration. Second, while the nature of an arbitration ruling would be more directed at generating a negotiated settlement, it would nevertheless be issued in the context of a fully institutionalized setting. It would thus be distinct from the ad hoc form of arbitration which was arguably available in the previous G A T T system. The institutionalization of arbitration as a method of dispute settlement was a component of both of the initial proposals from the E . C and the U.S . in the Uruguay R o u n d 4 9 8 and indeed was realized by the Improvements of 1989, even though arbitration was rarely used afterwards. The formal institutionalization of the process is itself, a component for increasing the compliance p u l l . 4 9 9 498 Supra notes 204 & 208. See also Hughes, supra note 176 at 77, where it is noted that in their 1987 proposals, while the E.C. used the phrase "institutionalized", the U.S. refers to it as a "formally available technique". 499 See generally Franck, "Power of Legitimacy", supra note 185-186. I believe that this is an appropriate and reasonable application of Franck's general notion of compliance pull. I note that in a critique of Franck's theory, Posner & Yoo {supra note 8 at 72) note that Franck theorizes that "adjudication by authentic international courts contributes to the legitimacy of international law" (Franck, "Legitimacy", supra note at 88 at 29-33). While this may be true, I suggest that the compliance pull described in Franck's theory would apply as well to a formalized system of arbitration within the institutionalized setting of the W T O , given Franck's observations suggesting that in the past, stronger obligations arose as a function of membership in international institutions. See supra note 88. 157 Thus the pressures towards a resolution with all disputing parties w i l l be higher, and the ability to justify any required changes to the domestic factions would be notionally stronger. While disputes would proceed down one of two tracks, they would nevertheless be connected within the same agreement, and serve to maintain the same principles. Indeed they may in some ways serve to maintain one another. For example, the connection between the two tracks could prove useful in considering other forms of sanctions, such as a form of suspension of rights of access to the judicial settlement track in the event of any unresolved cases in the arbitration track. Finally, this would not represent a retreat to a power oriented system to the disadvantage of developing countries. While there have been some criticisms of developing countries concerning the influence of power in arbitration, based on the history of the W T O , most of the politically difficult cases are disputes involving developed countries, and in particular, transatlantic disputes between the super powers. 5 0 0 5.6 The Feasibility of Diversion - De-Politicizing Judicial Settlement The D S B ought not encourage the use of judicial settlement where there is little prospect of implementation of that ruling. A system of institutional diversion of politically difficult cases is both logical and feasible. O f course, widespread recognition of the utility of arbitration amongst member states would be an important first step in establishing the legitimacy of the process. However it is clear that influential member 500 Brimeyer, supra note 459 at 167: Steinberg, supra note 2 at 267 & 275; see also Wilson, supra note 7. 158 states have identified the process of arbitration as a distinctly beneficial system. A s discussed, several proposals in the Uruguay Round included arbitration as an alternative, and even suggested its use in a specific class of disputes. Such a proposal may presently garner widespread consideration for a few reasons. First, the system is no longer under a realistic threat of U .S . withdrawal from the system, 5 0 1 as it was when the Appellate Body was created during the Uruguay Round. Although the U.S . today is likely still the only single nation superpower, the emergence of other potential economic superpowers such as India and China would likely mean that 502 a rules based system w i l l continue to be in the best interests of the U . S . Indeed, given the recent U . S . admonition against any tendency of the Appellate Body towards law-making, 5 0 3 it is far from clear that the U.S . would necessarily oppose a proposal to end appeals in particular cases. Second, Article 21.3 and 22.6 establish two forms of arbitration that are not consensual, but are rather mandatory, upon the request of one party. In effect, this illustrates how cases where compliance becomes an issue are already being shifted to arbitration when parties disagree as to the reasonable period for compliance, or when the successful party is forced to consider suspension of concessions. Concessions arbitration, despite acting as an adjunct to the judicial settlement process, has arguably been guided more by diplomatic considerations than any rationale legal calculation of the appropriate measure 501 Steinberg, supra note 2 at 267. 502 Davey, "Looking Forwards", supra note 11 at 17. 503 Communication From The United States, WTO Doc. No. TN/DS/W/82/Add.2 (17 March, 2006). 159 for countermeasures. This demonstrates that the reality of the limitation of the legal process is not lost on W T O decision makers. To the extent that such considerations remain in subterfuge, the dispute settlement system cannot meet its full potential. Worse, any surreptitious recognition of such political considerations w i l l ultimately undermine the overall integrity of the entire dispute settlement process. In light of this, weakening the independence of the decision-maker in the politically difficult cases may well better enhance the chance of the long-term survival of the overall system of dispute settlement. 5 0 5 A s some observers have already noted, there is a similar selection process that is already exercised by the Appellate Body. When presented with political issues, the Appellate Body has sometimes taken on the issue, while other times it refuses to address it squarely, 5 0 6 a practice known as "issue avoidance". 5 0 7 Furthermore, the Appellate Body may have already engaged in a form of conciliatory behaviour, to minimize the risks of 508 * institutional damage arising from non-compliance. These observations emphasize two points. First, even advocates of the legal system have recognized its limitations, and the danger of trying to resolve political cases within that system, at the risk of adversely affecting its overall legitimacy. Indeed, the Appellate Body is likely not suited for 504 See generally Spamann, supra note 326. 505 Posner & Yoo, supra note 8 at 74. 506 Steger, "Systemic Issues", supra note 6 at 65-6. 507 See e.g. William J. Davey, "Has the W T O Dispute Settlement System Exceeded its Jurisdiction." (2001)4 J. Int'l Econ. L. 79. 508 Geoffrey Garrett & James McCall Smith, The Politics of Dispute Settlement. Online: Yale University <http://www.vale.edu/leitner/pdf/1999-05.pdf>. 160 making "difficult political calculations". 3 u y Second, there is already an unspoken vetting system taking place within the legalized system. In order for the Appellate Body to maintain support from its members, particularly the more powerful ones, it must necessarily operate within its "political constraints". 5 1 0 The legal system should be as immunized as possible from political influence lest its credibility as a legally objective decision-maker be seriously undermined. 5 1 1 A two tiered system involving arbitration o f politically difficult cases would assist in maintaining some transparent separation of the purely legal mechanisms of the judicial settlement system, and the arbitration track proceedings that apply legal mechanisms to political considerations, where practically required. 5.7 Policy-Charged Cases in the Judicial Settlement System Ultimately the judicial settlement system has not coped well with politicized disputes, as demonstrated by the high profile cases between the superpowers that remain unresolved. It might be suggested that the softwood lumber agreement recently struck by the U.S . and 512 * ' Canada might in fact prove the effectiveness of judicial settlement system, based on the model I have proposed. In other words, the series of W T O rulings while not implemented to the letter, set the parameters for a negotiated settlement to a longstanding 509 Steinberg, supra note 2 at 274. 510 Steinberg, supra note 2 at 268-270. 511 Weiler, supra note 13 at 195. Weiler refers to the danger of persistence of diplomatic practices within the legal context that may undermine the rule of law and the benefits arising from the judicial settlement system. 512 Softwood Lumber Agreement between the Government of Canada and the Government of the United States of America, executed 12 September 2006 and amended 12 October 2006. 161 dispute between the two parties. It may be that the extensive litigation has allowed both countries to implement a palatable settlement, and the U.S . to manage its influential domestic pressure groups. However, one must wonder - what is the attendant cost to the integrity of the W T O of the series of unimplemented decisions? There can be no question that other members have a general systemic interest in such outcomes, a reflection of the inherent reciprocal nature of rights and obligations. 5 1 4 Viewed this way, the judicial settlement system has failed, as the outcome has failed to accord with the objective legal obligations as pronounced though the full rigour of the legalized appeal system. While the ultimate result of a negotiated settlement in a longstanding and complex trade dispute is a favourable outcome, one cannot underestimate the toll taken on the judicial settlement system. Further, it is clear that the dispute settlement has not always worked this way in light of other disputes, such as the regional aircraft disputes between Brazi l and Canada. 5 1 5 A t different points, both states had demonstrated that the government programs of the other constituted breaches of W T O obligations, each obtaining D S B authorization for the suspension of concessions against the other. After a panel ruling confirming Canada's breach, the Canadian Trade Minister stated that Brazi l needed a victory before they could 513 Online: Foreign Affairs and International Trade, Canada, <http://w01.international.gc.ca/minpub/Publication.aspx?isRedirect=True&publication id=384690&langua ge=E&docnumber=l 57> (last accessed 25 August, 2007). There were other panel rulings under the provisions of the North American Free Trade Agreement. 514 Fukunaga, supra note 48at 388&391. 515 Online: W T O <http://www.wto.org/english/tratop e/dispu e/cases_e/ds222_e.htm >. <http://www.wto.org/english/tratop e/dispu e/cases_e/ds70 e.htm> & <http://www.wto.org/english/tratop_e/dispu e/cases e/ds46 e.htm>. See generally Helen Sullivan, "Regional Jet Trade Wars: Politics and Compliance in W T O Dispute Resolution", (2003) 12 Minn. J. Global Trade 71 for full description of the disputes. 162 sit down and negotiate a settlement, an interesting juxtaposition with a later statement of Canada's Prime Minister regarding the U.S. failure to implement WTO and NAFTA 517 rulings regarding the softwood lumber dispute. A press release posted on the website of the Department of Foreign Affairs and International Trade after the arbitration award approving Brazil's suspension of concessions, reinforced this notion: The award marks an end to all current proceedings at the WTO surrounding the aircraft financing dispute between Canada and Brazil, enabling the two countries to concentrate on negotiating an end to this dispute. ... This award will not affect the financing of the Air Wisconsin, Air Nostrum and Comair purchases. The award could allow Brazil to raise tariffs on Canadian exports to Brazil but it is not expected that Brazil will impose the countermeasures, given the reciprocal 518 damage this could cause to bilateral trade. Brazil's success has not proven beneficial to any negotiations, as suggested by the Minister's statement. Despite the many decisions of panels and the Appellate Body, and Canada's claim in December 2002 that negotiations with Brazil had progressed,519 no agreement has resulted.520 It appears that the WTO decisions have had the opposite effect, creating a perception that each party was successful, and thus entrenching domestic resistance towards making any concessions. If anything, the various tides and 516 Steven Chase, "Pettigrew Sees Silver Lining in WTO Loss" Globe and Mail (28 January 2002), B l . See also Sullivan, supra note 515 at 102. 517 See supra note at 114. 518 Online: Foreign Affairs and International Trade, Canada, <http://w01.international.gc.ca/minpub/Publication.aspx?isRedirect=True&FileSpec=/Min Pub Docs/1058 Q8.htm&Language=E> (last accessed 25 August, 2007). (Posted on December 23, 2002). 519 Ibid. 520 As of March 1, 2007, see Wilson, supra note 7 at 403. There is no publicly available information of any agreement reported to the WTO as of 20 August, 2007. See online: WTO <http://www.wto.org/english/tratop e/dispu_e/cases e/ds222 e.htm >. <http://www.wto.org/english/tratop e/dispu e/cases e/ds70 e.htm> & <http://www.wto.org/english/tratop e/dispu e/cases e/ds46_e.htm>. 163 turns of an acrimonious dispute that was protracted by the judicial settlement^' may well have realized a recognized danger of the judicial settlement system - increasing domestic pressures against any negotiated resolution, thus locking-in the adversarial positions of 522 the disputing parties. 5.8 Identifying the Risk Cases If the W T O was to use a process of vetting disputes, it requires a way for identifying policy charged, problematic or high risk cases. It has been argued that trade decisions 523 themselves are inherently political, thus raising a key question: how can one determine i f a case is politically charged or difficult? The history of the W T O dispute settlement has demonstrated that certain disputes can be distinguished over others as "politically charged". 5 2 4 Amel ia Porges, a former legal officer of the G A T T Secretariat, has identified policy cases as ones that are brought for reasons of principle, with " low stakeholder involvement or with overwhelmingly strong governmental direction, where the stakes are 525 sometimes symbolic". I propose that these cases should be identified by a vetting process engaged at the consultations stage. There are a few reasons for this. First, one of the practical purposes 521 See generally Sullivan, supra note 515. 522 See Pauwelyn, "Americanization", supra note 46 at 126-7. 523 Young, supra note 43 at 408. 524 Busch & Reinhardt, "Fixing What Ain't Broke" supra note 302 at 13; se also McGivern supra note 9 at 141, who refers to the "highly politicized" disputes where there is non-compliance. 525 Porges, supra note 27 at 155. 164 of the consultations stage is to identify the impugned measure and the issues. A recent proposal has suggested the expansion of the consultations process for better information 527 exchange, while others have encouraged a more active role for the Secretariat in the consultations process. A n increased involvement of the Secretariat in the consultations process would create an opportunity to more carefully assess the parties' positions. This would allow the Secretariat to gain full knowledge of the respective positions of the disputants, and to identify and promote common ground. The Secretariat would exercise the authority to refuse judicial settlement, placing the political dispute on the arbitration track. The exercise of such a power would involve a tacit admission that the D S B is unable to effectively resolve cases with all of the entanglements of multilateral public interests of modern disputes. While somewhat heretical in this golden age of legalism at the W T O , this is not an overreach in the context of the current system. During the consultations process, the Secretariat would require briefs describing the main issue, the complaining party's proposal for possible resolutions, as well as the defending state's recitation of any political challenges to comply. Introduction at this stage has the advantage of identifying both the breach and the potential solutions as administrative, legislative or regulatory, as they are categorized at the compliance review stage. This process would require the responding party to make a written proposal as to how it w i l l change the impugned policies should it be unsuccessful after a hearing. In this context, the information gathering at the consultations stage would require the responding party to, amongst other things, answer certain questions such as: 526 Ibid, at 157. 527 Porges supra note 27 at 147, 171-174, 180-181. 165 - If the complaint is successful, how would it affect your economic and political system? - What steps could be taken to implement that decision? Once again this approach is not without some precedent in the W T O . Article 21.3 requires losing parties to inform the D S B of its intentions for implementation within 30 days after the adoption of the ruling. This is obviously a time period within which a state in unlikely to be able to navigate the potential of legislative changes. If a member state is expected to do so after a decision, it is equally plausible that it can put forward some form of plan at the consultations stage. This step would assist in assessing the political aspects of the dispute and permit a meaningful assessment of the risk of non-compliance. A requirement to provide extensive positions in writing would produce a frank discussion around such issues, expose the risk of the dispute, and facilitate a reasoned decision as to the risk of non-compliance. The system would likely require the level of confidentiality 528 already suggested by others, in order to ensure a candid discussion, and avoid the creation of any appearance of admissions. This is not difficult, as Article 4.6 of the D S U already provides that consultations are to be confidential and are conducted on a without prejudice basis. Indeed there is support for the notion that confidentiality still has a place 529 within the consultations process. 528 See e.g. Busch & Reinhardt, "Fixing What Ain't Broke" supra note 302 at 16. 529 Weiler, supra note 13 at 202; Bush & Reinhardt, "Fixing What Ain't Broke" supra note 302 at 16. 166 T h e process o f ident i fy ing h i g h r isk disputes m a y not be as di f f icul t as it w o u l d first appear. T h e observations o f H e l e n S u l l i v a n o n the case o f the C a n a d a - B r a z i l dispute over regional aircraft are ins ight ful : T o gain insight into the importance o f the aerospace industry to B r a z i l , it is desirable to understand Sao Jose dos C a m p o s , and an area Northeast o f S a n P a u l o . . . . . . T o B r a z i l i a n s , E m b r a e r is more that a corporat ion, it is a s y m b o l o f B r a z i l i a n 530 success i n a First W o r l d Industry, thus representing their hope for the future." In descr ib ing the dispute, S u l l i v a n aptly identifies the p o l i t i c a l baggage and indicators o f p o l i t i c a l issues w i t h i n the dispute. T o B r a z i l , the regional aircraft industry represented the hope o f d e v e l o p i n g its e c o n o m y to include trade i n technolog ica l ly sophisticated 531 goods. D u r i n g the bitter dispute, C a n a d a was perceived as a developed country w h o s i m p l y sought to prevent a deve loping country f r o m engaging i n trade b e y o n d basic sectors. C a n a d a became an enemy i n the eye o f B r a z i l i a n s , as its pos i t ion incited 533 widespread p u b l i c anger that went beyond the pol i t i c ians . T h i s type o f domestic p o l i t i c a l d y n a m i c w o u l d have had a major effect o n B r a z i l ' s ab i l i ty to implement the adverse decis ions against it. G a i n i n g a ful ler understanding b y e l i c i t i n g that i n f o r m a t i o n dur ing the consultations w o u l d i n a l l l i k e l i h o o d uncover the p o l i t i c a l l y charged tensions u n d e r l y i n g the dispute. T h i s process m a y w e l l have exposed the di f f icult ies w i t h any r u l i n g that was adverse to B r a z i l , the potential o f increased a c r i m o n y dur ing the dispute, 530 Sullivan, supra note 515 at 75. 531 Ibid, at 75, see discussion at footnote 37. 532 Ibid, at 94. 533 Ibid, at 97. 167 and the corresponding effect this could have on the compliance of either party with the ultimate ruling. 5.9 Further Considerations and Refinements A s mentioned, an important aspect of the arbitration alternative is ensuring that the procedure is perceived as legitimate, a key to the ultimate acceptance of any ru l ing . 5 3 4 In order to ensure that the arbitration process is viewed as legitimate by domestic political constituencies, the actual decision stage of the process would need to be transparent, in contrast to the exchange at the consultations stage. This is a relatively minor obstacle. In a recent proposal to the Dispute Settlement Negotiating Committee, the U .S . has already suggested increasing the transparency of arbitration, proposing amongst other things, that Article 18 of the D S U be amended to include a reference to arbitrations. 5 3 5 Undoubtedly, a diversion to arbitration would also raise a concern about the protection of the interests of other states. Arbitration does not necessarily exclude third party participation, and more specific rules could be developed to provide third parties with a form of access. In fact, there are already certain protections, given that parties to an arbitration are required to notify the D S B of the arbitration award, and members are 534 Young, supra note 43 at 408-9. 535 Communication From The United States, WTO Doc. No. TN/DS/W/86 (20 April, 2006). Indeed, one could argue that the fact that the proposal specifies inclusion of Article 25 arbitrations necessarily recognizes the continued utility of the proceeding. 168 permitted to raise any concerns before the Council or Committee of the relevant . 536 agreement. Another question is whether or not the compliance process under Article 21.3 and 22.6 would apply to a ruling by an arbitrator in the main dispute. Article 25.4 currently provides that Articles 21 and 22 are equally applicable to arbitration awards, meaning that these arbitration proceedings are currently available after a ruling under Article 25. Should these processes continue to be available in cases diverted to the arbitration track? A n Article 21.3 arbitration would continue to have some use, as the original arbitrator 537 could be used to resolve the issue of the reasonable implementation period. The investigation by the same arbitrator into the political system of the losing party indeed fits logically with the review of the political aspects of the original dispute. Further, arbitration under Article 21.3 has the advantage of extending negotiations around the original ruling, so that negotiations continue around when and how, and not if, the offending state w i l l implement an outcome acceptable to the complainant. Conversely, given the limitations of retaliation and the difficulties surrounding the calculation and implementation of countermeasures, it is questionable whether or not a s.22.6 arbitration would have any purpose under the arbitration track proceedings. However the arbitration track could be used as a means of establishing compensation, as it could provide another step in encouraging the offending state in coming to a negotiated settlement to bring its measures in line with its W T O obligations. 536 Article 25.3 of the DSU. 537 Hughes, supra note 176 at 80. 169 Ultimately, I have made a broad proposal for change, and there is much potential for refinement and discussion over the precise ways of structuring such a system. This might include the possibility of a non binding opinion on the merits of the dispute rather than a binding arbitration ruling. It may also include the possibility of implementing financial compensation in politically difficult cases, as a mandatory form of remedy as has been 538 suggested by others. This option could be particularly useful in the rare occasion where a developing state is involved in case diverted to the arbitration track. In such a case, where the complainant is a developing state, the possibility of obtaining a tangible compensation award where implementing retaliation would be impossible might act as a counterbalance to any perception or complaint about the influence of power in the arbitration proceedings. 538 See e.g. Bronckers & van den Broek, supra note 115. 170 5.10 Thesis Conclusion and Further Research While the dispute settlement system under the W T O and the G A T T has been relatively successful in the field of international adjudication, history has proven that neither the political solutions of the previous G A T T system nor the legalistic mechanisms of the W T O can resolve the myriad of issues that could be potentially addressed within the W T O . Arbitration may be the best suited instrument to supplement the D S U and enhance compliance with W T O rulings in a few politically difficult cases. A s a result of the multifaceted nature of the Uruguay Round negotiations, while arbitration was seen as a useful alternative to the litigation system, the concept did not fully develop and has since fallen off the radar screen. However, the time is ripe to truly explore the arbitration alternative as one possibility of dealing with the most difficult disputes. The creation of the W T O represents a shift from the traditional international dispute 539 resolution model of alternative dispute resolution toward a system of binding decisions. This process has seen the legalistic effectively replace the diplomatic. 5 4 0 While the high level of compliance with W T O rulings is a remarkable achievement, it is neither reasonable nor prudent to seek perfect compliance, an artificial goal that was never intended. 5 4 1 It is important to maintain realistic expectations of the D S B and to resist the urge to allow inertia push the W T O system more towards overly legalistic mechanisms. It is simply not reasonable to expect that as state to state trade disputes become more 539 McRae, supra note 296 at 9. 540 Steger, "The Struggle for Legitimacy", supra note 96 at 121. 541 Vazquez & Jackson, supra note 22 at 565. 171 multifaceted and policy-charged, that they can still all be resolved by way of a one size-fits-all system of binding decision-making by a supranational court. 5 4 2 Based on the history of negotiations leading to the creation of the D S B and the stated objectives in the D S U , the dispute settlement system was not designed for that purpose. Further, the quest for perfect compliance may not promote the long term stability of the W T O . 5 4 3 The growing collective compulsion to expand the range of available remedies and increase the power of the D S B to address compliance issues, rather than promoting negotiations that are informed by objective rulings can be counterproductive. In considering reform to the D S B , one cannot assume that stricter rules for improved remedies w i l l result in greater compliance. Proposal for measures that are directed at enhancing a system of compensation and retaliation seem to be disconnected from the objectives of the D S U . The current discourse of D S B reform too often incorporates these approaches, while seemingly overlooking that in international law, normative condemnation represents a significant incentive for compliance. 5 4 4 Some have proposed that judicial minimization giving more deference to the position of member states would allow the D S B to better account for the political sensitivities that drive the more policy oriented disputes. 5 4 5 A system vetting the most politicized disputes to a gentler and more consensual form of decision-making would have the same effect. That is not to recommend a return to the previous G A T T system. However it means a 542 Pauwelyn, "Americanization", supra note 46 at 140. 543 Fukunaga, supra note 48 at 426. 544 Bush & Reinhardt, "Evolution of G A T T / W T O " , supra note 297 at 147. 545 Pauwelyn, "A Missed Opportunity", supra note 165 at 346. 172 candid recognition by international lawyers of the limitations of compensation and retaliation, and the importance of normative condemnation 5 4 6 as the primary pressure towards compliance in the international trading system. It is therefore important to understand the limitations of adopting measures from domestic systems that are based on different objectives and on a different basis for enforcement. Arbitration may offer an effective alternative rule framework for specific cases, one that lies somewhere between the fragile diplomatic basis on which G A T T was originally based, and the unbridled legalism following the creation of the W T O . In this middle ground, the component of the reputation of member states is still engaged, but with a lesser risk of damaging the reputation of the judicial settlement system achieved through the relative successes of the last twelve years. Every instance of non-compliance from W T O rulings undermines the legitimacy of the adjudication system as whole , 5 4 7 particularly when that ruling purports to provide an objectively correct interpretation of W T O obligations through a legal appeal system. The history of G A T T has demonstrated that a crisis of confidence in the dispute settlement process can risk the disintegration of the entire system, or at least, render it o f limited use, as was the case for G A T T during distinct periods. On the other hand, it may be that certain policy driven disputes cannot reasonably be resolved through the simple application of legal principles. A s such, a vetting system that redirects the politically difficult cases away from the judicial 548 settlement system may serve to "protect the judicial integrity" of the W T O dispute 546 Busch & Reinhardt, "Evolution of G A T T / W T O " , supra note 297 at 147. 547 Jackson, "The World Trading System", supra note 178 at 85. 548 Cho, supra note 312 at 784. 173 system. When confronted with policy driven disputes that may be difficult to resolve, the preservation of the integrity of the adjudication system alone is a worthy objective. The proposal of this thesis may well appear to be heretical to those proponents of the legalized dispute settlement system of the W T O , or those who have witnessed first hand the successes of that system. Further, there are many challenges for the vetting system proposed, not the least of which is the danger that such a proposal goes much further than mere "fine-tuning". A n y undertaking to make improvements to the current system should be taken with caution. 5 4 9 Indeed, it is important to ensure that any reform does not overreach and jeopardize the legitimacy of the system as a whole. In particular, the vetting system would require some protections to minimize any bias, or perception of bias, towards developed states over developing countries states, or to states demonstrating a weaker respect for panel or Appellate Body rulings over those who are more compliant. There is much research that would be required in this and other respects before moving this proposal beyond its current fledgling state. Should the W T O adapt a body of institutional arbitration rules, and i f so which ones, and which elements? Should arbitration be used to determine a form of compensation in politically difficult cases involving complaints from developing states? Could the arbitration system be used to generate non-binding advisory opinions that would be effective in engaging normative condemnation in these cases? I suspect that to the extent that the proposal in this thesis is 549 Jaqueline Krikorian, "Plane, Trains and Automobiles: The Impact of the W T O "Court" on Canada in the First Ten Years" (2005) 8 J. Int'l Econ. L. 921 at 967; Busch & Reinhardt, "Fixing what Ain't Broke". supra note 302. 174 considered by experts in this area, that it would generate much criticism, debate and hopefully, further research for this and many other reasons. I would welcome this attention, and would suggest that the current discourse for reforming and improving the dispute settlement system would only benefit from any consideration or discussion of this option. 175 BIBLIOGRAPHY Monographs Barfield, Claude. Free Trade, Sovereignty, Democracy: The Future of the World Trade Organization. (Washington D . C . : American Enterprise Institute, 2001). Chayes, Abram & Antonia Handler Chayes. The New Sovereignty: Compliance with International Regulatory Agreements (Cambridge: Harvard University Press, 1995). Cottier, Thomas, Petros C. Mavroidis & Patrick Blatter, eds. The Role of the Judge in International Trade Regulation (Ann Arbor: University of Michigan Press, 2003). Croome, John. Reshaping the World Trading System: A History of the Uruguay Round (The Hague: Kluwer Law International, 1999). Evans, Gai l E . . Lawmaking Under the Trade Constitution: A Study in Legislating by the World Trade Organization (The Hague; Boston: Kluwer Law International, 2001). Franck, Thomas. The Power of Legitimacy Amongst Nations (Oxford: Claredon Press, 1990). Franck, Thomas. Fairness & International Law and Institutions (Oxford: Oxford University Press, 1995). Garcia, Frank. Trade Inequality, and Justice: Towards a Liberal Theory of Just Trade (Ardsley, N . Y . : Transnational Publishers, 2003). Hart, H . L . A . The Concept of Law (Oxford: Claredon Press, 1961). Hallstrom, Par. The GATT Panels and the Formation of International Trade Law (Stockholm: Jurisforlaget, 1994). Hudec, Robert E . Adjudication of International Trade Disputes (Trade Policy Research Centre, London: Ditchling Press Limited 1978). Hudec, Robert E . The GATT Legal System and World Trade Diplomacy, 2 n d ed., Salem, (New Hampshire: Butterworth Legal Publishers, 1990). Hudec, Robert E . Enforcing International Trade Law. The Evolution of the Modern GATT Legal System (Salem, N e w Hampshire: Butterworth Legal Publishers, 1993). Jackson, John H . The World Trading System: Law and Policy of International Economic Relations (Cambridge, Massachusetts & London, England: The M I T Press, 1989). 176 Jackson, John H . The Uruguay Round, World Trading Organization and Problem of Regulating International Economic Behaviour (Carleton University, Ottawa: Graphic Services, 1995). Keohane, Robert O. After Hegemony: Cooperation and Discord in the World Political Economy (Princeton, N e w Jersey: Princeton University Press, 1984). Keohane, Robert O. Power and Governance in a Partially Globalized World (London & New York: Routledge, 2002). Matsushita, Mitsuo, Thomas J. Schoenbaum & Petros C. Mavroidis. The World Trade Organization: Law, Practice, and Policy (Oxford; New York: Oxford University Press, 2003). Morgenthau, Hans. Politics Among Nations: The Struggle for Power and Peace (New York: A A Knopf, 1948). Palmeter, David & Petros C. Mavroidis. Dispute Settlement in the World Trade Organization: Law, Practice and Procedure (Oxford; N e w York: Cambridge University Press, 2004). Pauwelyn, Joost. Conflict Of Norms In Public International Law: How WTO Law Relates To Other Rules Of International Law (Cambridge: Cambridge University Press, 2003). Redfern, A lan & Martin Hunter, eds. Law and Practice of International Commercial Arbitration, (London: Sweet & Maxwel l , 2003). Stewart Terence P., ed. The GATT Uruguay Round: A Negotiating History (1986-1992) (Volume II: A Commentary) (Deventer & Boston: Kluwer Law and Taxation Publishers, 1993). Stewart, Terence P. & A m y S. Dwyer. Handbook on WTO Trade Remedy Disputes: The First Six Years, 1995-2000 (Ardsley, N . Y . : Transnational Publishers, 2001). van Houtte, Hans. The Law Of International Trade (London: Sweet & Maxwel l , 1995). Weiss, Freidl, ed. Improving WTO Dispute Settlement Procedures - Issues and Lessons from the Practice of Other International Courts & Tribunals (London: Cameron M a y Ltd. , 2000). Wilkinson, Rorden. Multilateralism and The World Trade Organisation: The Architecture And Extension Of International Trade Regulation. (London; N e w York: Routledge, 2003.) 177 Journal/Newspaper/Online Authored Articles Alvarez, Henri. "Arbitration under the North American Free Trade Agreement Chapter 11" (2000) 16:4 Arb. Int'l 393. Arup, Christopher. "The State of Play of Dispute Settlement " L a w " at the World Trade Organization" (2003) 37 J. World Trade 897. Bello, Judith Hippler. "The W T O Dispute Settlement Understanding: Less is More" (1996) 90 A m . J. Int'l L . 416. Bel lo, Judith Hippler. "Book Review"(2001) 95 A m . J . Int'l L . 984. Blessing, Marc. "State Arbitrations: Predictably Unpredictable Solutions?" (2005) 22 J. Int'l Arb . 435. Boisson de Chazournes, Laurence. "L'Arbitrage A L ' O M C " (2003) Revue de l'arbirtage, 949. Brimeyer, Benjamin L . "Bananas, Beef and Compliance in the World Trade Organization: The Inability of the W T O Dispute Settlement Process to Achieve Compliance from the Superpower Nations", (2001) 10 M i n n . J. Global Trade 133. Bronckers Marco & Naboth Van den Broek. "Financial Compensation in the W T O : Improving the Remedies of W T O Dispute Settlement" (2005) 8 J. Int'l Econ. L . 101. Busch, Marc L . & Eric Reinhardt. "The Evolution of G A T T / W T O Dispute Settlement" in John M . Curtis & Dan Curiak, eds., Trade Policy Research 2003 (Ottawa: Minister of Public Works and Government Services, 2003) 143. Busch, Marc L . & Eric Reinhardt. "Bargaining in the Shadow of the Law: Early Settlement in G A T T / W T O Dispute"(2000-1) 24 Fordham Int'l L . J. 158. Busch, Marc L . & Eric Reinhardt. "Fixing What ' A i n ' t Broke'?: Third Party Rights, Consultations, and the D S U " , online: Georgetown University <http://www9.georgetown.edu/faculty/mlb66/DSU.pdf>. Busch, Marc L . and Eric Reinhardt. "Testing International Trade Law: Empirical Studies of G A T T / W T O Dispute Settlement" in Daniel M Kennedy and James D Southwick, (eds.), The Political Economy of International Trade Law: Essays in Honour of Robert Hudec (New York and Cambridge: Cambridge University Press, 2002) 457. Byers, Michael . "Custom, Power, and the Power of the Rules - Customary International Law from an Interdisciplinary Perspective"( 1995-6) 17 M i c h . J. Int'l L . 109. 178 Carbonneau, Thomas E . "Diversity Or Cacophony?: N e w Sources O f Norms In International Law Symposium: Article: Arbitral Law-Making" (2004) 25 M i c h . J. Int'l L . 1183. Chase, Steven. "Pettigrew Sees Silver Lining in W T O Loss" Globe and Mail (28 January 2002), B l . Cho, Sungoon. "The Nature of Remedies in International Trade L a w " (2004) 65 U . Pitt. L . Rev. 763. Davey, Wi l l i am J. "Has the W T O Dispute Settlement System Exceeded its Jurisdiction." (2001) 4 J. Int'l Econ. L . 79. Davey, Wi l l i am J. "Implementation in W T O Dispute Settlement System: A n Introduction to the Problem and Possible Solutions" (2005) 8 J. Int'l Econ. L . 1. Davey, Wi l l iam J. "The W T O Dispute Settlement System: The First Ten Years" (2005) 8 J. Int'l Econ. L . 17. Davey, Wi l l i am J. "The Sutherland Report on Dispute Settlement - A Comment" (2005) 8 J. Int'l Econ. L . 321. Davey, Wi l l i am J. "The W T O : Looking Forwards" (2006) 9 J. Int'l Econ. L . 3. Dunne, Matthew S. "Redefining Power Orientation: A Reassessment O f Jackson's Paradigm In Light O f Asymmetries O f Power, Negotiation, A n d Compliance In The G A T T / W T O Dispute Settlement System" 34 L a w & Pol'y Int'l Bus. 277. Franck. Susan D . "The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions" (2005): 73 Fordham L . Rev. 1521. Fukunaga, Yuka . "Securing Compliance Through the W T O Dispute Settlement System: Implementation of D S B Recommendations" (2006) 9 J. Int'l Econ. L . 383. Garrett, Geoffrey & James M c C a l l Smith, "The Politics of Dispute Settlement", online: Yale University <http://www.yale.edu/leitner/pdf/1999-05.pdf>. Giardina, Andres and Americo Beviglia Zampetti. "Settling Competition Related Disputes - The Arbitration Alternative in the W T O Framework" (1997) 31:6 J. World Trade 27. Gruner, Dora Marta. "Accounting for the Public Interest in International Arbitration: The Need for Procedural and Structural Reform" (2002-03) 41 Colum. J. Transnat'l L . 923. 179 Hilf , Meinhard, "Power, Rules and Principles - Which Orientation for W T O / G A T T Law?" (2001) 4 J. Int'l Econ. L . 111. Honghua, Men. "Critiques of the Theory of International Regimes: The Viewpoints of M a i n Western Schools of Thought", online: IR China Online <http://www.irchina.org/en/pdf/mhhl.pdf>. Howse, Robert and Kalypso Nicolaidas. "Enhancing W T O Legitimacy -Contistutionalization or Subsidiarity" in Marco Verweij & T im Josling, eds., Deliberately Democratizing Multilateral Organization, special issue of Governance (2003), online: University of Michigan <http://faculty.law.umich.edu/rhowse/ >. Howse, Robert and Kalypso Nicolaidas. "Legitimacy & Global Governance - Why Constitutionalization the W T O is Taking A Step Too Far", online: University of Michigan <http://faculty.law.umich.edu/rhowse/>. Howse, Robert. "How to Begin to Think About the 'Democratic Deficit ' at the W T O . " , online: University of Michigan <http://faculty.law.umich.edu/rhowse/>. Hudec, Robert. "Broadening the Scope of Remedies Under the W T O " in Freidl Weiss, ed., Improving WTO Dispute Settlement Procedures: Issues and Lessons from the Practice of Other International Courts & Tribunals (London: Cameron M a y Ltd. , 2000) 382. Also found online <http://www.worldtradelaw.net/articles/hudecremedies.pdf>. Hudec, Robert E . "The N e w W T O Dispute Settlement Procedure: A n Overview of the First Three Years", (1999) 8:1 Minn . J. Global Trade 1. Hughes, Valerie. "Arbitration Within the W T O " in F.Ortino & E . - U . Petersmann, eds., The WTO Dispute Settlement System 1995-2003 (The Hague: Kluwer L a w International, 2004)75. Jackson, John H . "The Crumbling Institutions of the Liberal Trade System" (1978) 12:93 J. World Trade 98. Jackson, John H . "International Law Status of W T O Dispute Settlement Reports: Obligation to Comply Or Option To Buy Out" (2004) 98 A m . J. Int'l L . 109. Kearns, Jason E . and Steve Charnovitz. "Adjudicating Compliance in the W T O : A Review of D S U Article 21.5" (2002) 5 J. Int'l Econ. L . 331. K o h , Harold H . "Why do Nations Obey International L a w " (1997) 106 Yale L . J. 2599. Krajewski, Markus. "Democratic Legitimacy & Constitutional Perspectives of W T O L a w " 35:1 (2001) J. World Trade 167. Krikorian, Jaqueline. "Plane, Trains and Automobiles - The Impact of the W T O "Court" on Canada in the First Ten Years" (2005) 8 J. Int'l Econ. L . 921. 180 Legum, Barton. "Emerging Fora For International Litigation (Part 1): The Innovation O f Investor-State Arbitration Under N A F T A " (2002) 43 Harvard Int'l L . J . 531. Leitner, Kara & Simon Lester. " W T O Dispute Settlement 1995-2006—A Statistical Analysis" (2007) 10 J. Intl Econ. L . 165. McGivern , Brendan P. "Seeking Compliance with W T O Rulings: Theory, Practice and Alternatives" (2002) 36 Int'l Law. 141. McRae, Donald. "What is the Future of W T O Dispute Settlement?" (2004) 7 J. Int'l Econ. L . 3. Montana I. Mora, Miquel . " G A T T with Teeth: Law Wins over Politics in the Resolution of International Trade Disputes" (1993-1994) 31 Colum. J. of Transnat'l L . 103. Monnier, Pierre. "Working Procedures Before Panels, the Appellate Body and Other Adjudicating Bodies of the W T O " (2002) 1 L a w & Prac. Int'l Courts & Trib. 481. Monnier, Pierre. "Time to Comply with W T O Rulings" (October 2001) 35 J. World Trade 825. Nzelibe, Jidi . "Credibility Imperative - Political Dynamics of Retaliation" (2005) 6:1 Theor. Inq. L . 216. Pauwelyn, Joost. "The Sutherland Report - A Missed Opportunity For Genuine Debate On Trade, Globalization A n d Reforming The W T O " (2005) 8 J. Int'l Econ. L . 329. Pauwelyn, Joost. "The Limits of Litigation: 'Americanization' and Negotiation in the Settlement of W T O Disputes."(2003-2004) 19:1 Ohio St. J. Disp. Resol. 121. Pauwelyn, Joost. "Enforcement and Countermeasures in the W T O : Rules are Rules -Towards a More Collective Approach" (2000) 94 A.J . I .L . 335. Pauwelyn, Joost. "Adding Sweeteners to Softwood Lumber: the W T O - N A F T A 'Spaghetti B o w l ' is Cooking ' " (2006) 9 J. Int'l Econ. L . 197. Petersmann, Ernst-Ulrich. " W T O Negotiators Meet Academics: The Negotiations on Improvements of the W T O Dispute Settlement System" (2003) 6 J. Int'l Econ. L . 237. Petersmann, Ernst Ulr ich & Meinard Hilf . "Strengthening the G A T T Dispute Settlement System: On the Use of Arbitration in G A T T " , in Ernst Ulr ich Petersmann & Meinard Hilf , eds., The New GATT Round of Multilateral Trade Negotiations (Deventer & Boston: Kluwer L a w and Taxation Publishers, 1988) 323. 181 Porges, Amelia . "Settling W T O Disputes: What do Litigation Models Tel l Us?" (2003-2004) 19:1 Ohio St. J. Disp. Resol. 141. Posner, Eric A . & John C. Yoo . "Judicial Independence in International Tribunals" (2005) 93:1 Cal . L . Rev. 1. Rogers, Catherine A . "Emerging Dilemmas In International Economic Arbitration: The Vocation O f The International Arbitrator" (2005) 20 A m . U . Int'l L . Rev. 957. Spamann, Holger. "The Myth of "Rebalancing" Retaliation in W T O Dispute Settlement Practice" (2006) 9 J. Int'l Econ. L . 31. Sullivan, Helen. "Regional Jet Trade Wars: Politics and Compliance in W T O Dispute Resolution", (2003) 12 Minn . J. Global Trade 71. Steger, Debra P. "Appellate Body Jurisprudence Relating to Trade Remedies" 35 (2001) J. World Trade 799. Steger, Debra P. "The Struggle for Legitimacy in the W T O . " Trade Policy Research 2003 (Ottawa: Minister of Public Works and Government Services, 2003), 111-141. Steger, Debra P. " W T O Dispute Settlement: Systemic Issues." in John M . Curtis & Dan Curiak, eds., Trade Policy Research 2005 (Ottawa: Minister of Public Works and Government Services, 2006), 51-76. Online: <http://www.dfait-maeci.gc.ca/eet/research/TPR-2005/TPR-2005 Chapter 03 -_Steger.pdf>. Steinberg, Richard H . "Judicial Lawmaking at the W T O : Discursive, Constitutional and Political Constraints" (2004) 98 A m . J. Int'l L . 247. Sutherland, Peter. "Concluding the Uruguay Round - Creating the N e w Architecture of Trade for the Global Economy" (2000-1) 24 Fordham Int'l L . J . 15. Swacker, Frank W. , Kenneth R. Redden & Larry B . Wenger, " W T O & A D R " (2000) 55:3 Disp. Resol. J. 35. van den Broek, Naboth. "Power Paradoxes in Enforcement and Implementation of World Trade Organization Dispute Settlement Reports: Interdisciplinary Approaches and New Proposals"(2003) 37 J. World Trade 127. Van Harten, Gus. "Judicial Supervision of N A F T A Chapter 11 Arbitration: Public or Private Law?" (2005): 21 Arb. Int'l 493. Vazquez, Carlos & John H . Jackson. "Some Reflections on Compliance with W T O Dispute Settlement Decisions" (2002) 33 Law & Pol 'y Int'l Bus. 555. 182 Vicuna, Francisco Orrego. "Arbitration in a New International Alternative Dispute Resolution System" (2002) 57:2 Disp. Resol. J. 64. Weiler, Joseph H . H . "The Rule of Lawyers and the Ethos of Diplomats" (2001) 35 J. World Trade 191. Werner, Jacques. "The Trade Explosion and some Like ly Effects on International Arbitration." (1997) 14 J Int'l Arb. 5. Wilson, Bruce. "Compliance by W T O Members With Adverse W T O Dispute Settlement Rulings: The Record to Date" (2007) 10 J. Int'l Econ. L . 397. Young, Michael K . "Dispute Resolution in the Uruguay Round: Lawyers Triumph Over Diplomats" (1995) 29 Int'l . Law. 401. Zampetti, Americo Beviglia. "Democratic Legitimacy in the World Trade Organization: The Justice Dimension" (2003) 37 J. World Trade 105. Zekos, George I. "Arbitration as Dispute Settlement Mechanism under U N C L O S , The Hamburg Rules and W T O " (2002) 19 J Int'l Arb. 497. Zekos, George I. " A n Examination of G A T T / W T O Arbitration Procedures." (1999) 54:4 Disp. Resol. J. 72. 183 WTO Decisions Canada - Export Credits and Loan Guarantees for Regional Aircraft - Recourse to Arbitration by Brazil under Article 22.6 of the DSU, WT/DS222 /ARB,17 February, 2003. Canada - Term of Patent Protection Award of the Arbitrator under Article 21.3(c) of the DSU, W T / D S 170/10, 28 February, 2001. European Communities - Regime for the Importation, Sale and Distribution of Bananas Award of the Arbitrator under Article 21.3(c) of the DSU, WT/DS27/15, 7 January 1998. European Communities - Regime for the Importation, Sale and Distribution of Bananas. Recourse to the Arbitrator by the European Communities under Article 22.6 of the AS£/ ,WT/DS27/ARB, 6 Apr i l 1999. European Communities - Regime for the Importation, Sale and Distribution of Bananas. Recourse to the Arbitrator by the European Communities under Article 22.6 of the DSU, W T / D S 2 7 / A R B / E C U , 24 March 2000. European Communities - Measures Concerning Meat and Meat Products (Hormones, Report of the Appellate Body, W T / D S 2 6 / A B / R , W T / D S 4 8 / A B / R , 16 January 1998. European Communities - Measures Concerning Meat and Meat Products (Hormones)-Original Complaint by the United States - Recourse to Arbitration by the European Communities under Article 22.6(c) of the DSU, Decision B y the Arbitrators, W T / D S 2 6 / A R B , 12 July 1999 European Communities - Measures Concerning Meat and Meat Products (Hormones) -Original Complaint by Canada - Recourse to Arbitration by the European Communities under Article 22.6(c) of the DSU, Decision B y the Arbitrators, W T / D S 4 8 / A R B , 12 July 1999. E.C. Measures Concerning Meat and Meat Products (Hormones) - Award of the Arbitrator under Article 21.3(c) of the DSU, WT/DS26/15, WT/DS48/13, 29 M a y 1998. Japan - Taxes on Alcoholic Beverages, Award of the Arbitrator under Article 21.3(c) of the DSU, WT/DS8/15, WT/DS10/15, WT/DS11/13, 14 February 1997. United States - Measures Affecting The Cross-Border Supply Of Gambling and Betting Services Award of the Arbitrator Under Article 21.3(C) of DSU, WT/DS285/13, 19 August 2005 Arb-2005-2/19. United States Tax Treatment for Foreign Sales Corporations, Report of the Appellate Body, W T / D S 108 /ASB/RW, 24 February 2000. 184 United States - Tax Treatment for 'Foreign Sales Corporations' - Recourse to Arbitration by the United States under Article 22.6 of the DSU and Article 4.11 of the SCM Agreement, W T / D S 108/ARB, 30 August 2002. United States - Section 110(5) of the U.S. Copyright Act - Recourse to the Arbitration under Article 25 of the DSU, W T / D S 160/ARB25/1, 9 November 2001. 185 GATT/WTO Documents Chairman's Text on Dispute Settlement G A T T Doc. No . CGT/807-14 NG13 (19 October, 1989) . Communication From Australia, W T O Doc. No . TN/DS/W/34 (15 January 2003). Communication From Canada, G A T T Doc. No. M T N . G N G / N G 1 3 / W / 1 3 (24 September, 1987). Communication From Canada, G A T T Doc. N o . M T N . G N G / N G 1 3 / W / 4 1 (28 June, 1990) . Communication From The EEC, G A T T Doc. No . M T N . G N G / N G 1 3 / W / 1 2 (24 September, 1987). Communication From The EEC, G A T T Doc. N o . M T N . G N G / N G 13/W/22 (2 March, 1988) . Communication From Japan, G A T T Doc. N o . M T N . G N G / N G 13/W/21 (1 March, 1988). Communication From Korea, G A T T Doc. No . M T N . G N G / N G 1 3 / W / 1 9 (20 November, 1987). Communication From Mexico, G A T T Doc. No . M T N . G N G / N G 13/W/26 (23 June, 1988). Communication From Mexico, G A T T Doc. No . M T N . G N G / N G 1 3 / W / 2 6 / R e v . l (10 October, 1988). Communication From The Nordic Countries, G A T T Doc. N o . M T N . G N G / N G 1 3 / W / l 0 (18 September, 1987). Communication From Switzerland, G A T T Doc. No . M T N . G N G / N G 13/W/8 (18 September, 1987). Communication From Switzerland, G A T T Doc. N o . M T N . G N G / N G 1 3 / W / 3 3 (19 July, 1989) . Communication From the United States, G A T T Doc. No . M T N . G N G / N G 1 3 / W / 4 0 (6 A p r i l , 1990). Communication From the United States, W T O Doc. No . W T / D S 160/23 (6 June, 2003). Communication From The United States, W T O Doc. No . TN/DS/W/82/Add.2 (17 March, 2006). 186 Communication From The United States, W T O Doc. No . TN/DS/W/86 (20 A p r i l , 2006). Draft Text on Dispute Settlement, G A T T Doc. No . M T N . G N G / N G 13/W/45 (21 September, 1989). G A T T , Draft Text on Dispute Settlement, G A T T Doc. No . M T N . G N G / N G 13/W/45 (21 September, 1990). G A T T , Draft Final Act Embodying The Results of the Uruguay Round of Multilateral Trade Negotiations G A T T Doc. No . M T N . T N C / W / 3 5 / R e v . l (3 December, 1990). G A T T , Group "Framework", Note by the Acting Chairman of Group, 12 December 1978, G A T T Doc. No . MTN/FR7W/20 . G A T T , Ministerial Declaration, 38 t h Session, Adopted on 29 November 1982, G A T T Doc. N o . L/5424. G A T T , Ministerial Declaration on the Uruguay Round, September 20, 1986, 33 G A T T / B I S D 19 (1987). G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting (held on 22 September, 1988), G A T T Doc. N o . C/M/224 (17 October, 1988). G A T T , Negotiating Group on Dispute Settlement, Meeting at Ministerial Level, G A T T Doc. No . M T N . T N C / 7 ( M I N ) , (9 December, 1988). G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting for 24 September to 11 October 1989, Note by the Secretariat, G A T T Doc. No . M T N . G N G / N G 13/23 (24 October, 1989). G A T T , Negotiating Group on Dispute Settlement, Minutes of Meeting (held on 28 September, 1989): Note by the Secretariat, G A T T Doc. No. M T N . G N G / N G 1 3 / 1 6 (13 November, 1989) G A T T , Negotiating Group on Dispute Settlement, Note by the Secretariat, G A T T Doc. No . M T N . G N G / N G 1 3 / W / 4 / R e v . l (10 November, 1989). G A T T , Negotiating Group on Dispute Settlement, Meeting of 7 February, 1990, Note by the Secretariat, G A T T Doc. No . M T N . G N G / N G 1 3 / 1 8 (21 March, 1990). G A T T , Negotiating Group on Dispute Settlement, Meeting of 5 April, 1990: Note by the Secretariat, G A T T Doc. No . M T N . G N G / N G 1 3 / 1 9 (28 May , 1990). G A T T , Report of the Consultative Group Of Eighteenth to the Council of Representatives, 31 October, 1977, G A T T Doc. N o . L4585. 187 G A T T , Secretariat, Concepts, Forms and Effects of Arbitration, Note by the Secretariat, G A T T Doc. No . M T N . G N G / N G 1 3 / W / 2 0 , (22 February, 1988). G A T T , Secretariat, Summary and Comparative Analysis of Proposals for Negotiations, Note by the Secretariat, G A T T Doc. No . M T N . G N G / N G 1 3 / W / 1 4 / R e v . l (26 February, 1988). G A T T , Trade Negotiations Committee, Improvements to the GATT Dispute Settlement Rules and Procedures, Decision of 12 Apr i l 1989, G A T T Doc. No . L/6489. G A T T , Understanding Regarding Notification Consultation, Dispute Settlement and Surveillance, 3 December 1979, G A T T Doc. N o . L/4907. Improved Dispute Settlement: Elements for Consideration. Discussion Paper Prepared by United States Delegation, G A T T Doc. No . M T N . G N G / N G 13/W/6 (25 June, 1987). Statement by the Spokesman of the European Community at the Meeting on 5-6 April, G A T T Doc. N o . M T N . G N G / N G 1 3 / W / 3 9 (5 A p r i l , 1990). Sutherland, Peter et al. "The Future of the W T O : Addressing Institutional Challenges in the New Mil lennium", Report by the Consultative Board to the Former Director-General Supachai Panitchpakdi (2004), online: W T O <http://www.wto.orR/english/thewto e/lOanniv e/future wto e.htm>. W T O , Dispute Settlement Body, Minutes Of Meeting (held on 22 February, 2006), W T O Doc. T N / D S / M / 3 0 . W T O , Minutes Of Meeting Special Session of the Dispute Settlement Body (held on 13 July 2006), W T O Doc. T N / D S / M / 3 4 . 188 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0077735/manifest

Comment

Related Items