The reasons requirement in international investment arbitration: critical case studies

The Reasons Requirement in International Investment Arbitration Critical Case Studies Guillermo Aguilar Alvarez W. Mich...

Author: Guillermo Aguilar | William Michael Reisman

118 downloads 1130 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

The Reasons Requirement in International Investment Arbitration Critical Case Studies

Guillermo Aguilar Alvarez W. Michael Reisman EDITORS

Library of Congress Cataloging-in-Publication Data The reasons requirement in international investment arbitration : critical case studies / editors Guillermo Aguilar Alvarez, W. Michael Reisman. p. cm. Includes index. ISBN 978-90-04-16632-5 1. Investments, Foreign (International law)—Cases. 2. Arbitration and award, International—Cases. 3. Investments, Foreign—Law and legislation—Cases. I. Aguilar, Guillermo. II. Reisman, W. Michael (William Michael), 1939– K3830.R43 2008 346.07—dc22 2008015759

Copyright © 2008 Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishers, IDC Publishers, Martinus Nijhoff Publishers, and VSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Brill provided that the appropriate fees are paid directly to the Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. Manufactured in the United States of America

CONTENTS Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Chapter 1: How Well Are Investment Awards Reasoned?. . . . . . . . . 1 Guillermo Aguilar Alvarez and W. Michael Reisman Chapter 2:

The Final Award in Mondev International v. United States of America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Julien Cantegreil

Chapter 3: The Final Award in Feldman v. Mexico . . . . . . . . . . . . . . 63 Dennis Clare Chapter 4: The Decision on Liability in LG&E v. Argentina . . . . . 117 Rocío Digón Chapter 5 The Award in Saluka Investments v. Czech Republic. . . . 149 George Stephanov Georgiev Chapter 6:

The Decision on Jurisdiction in Tokios Tokeles v. Ukraine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Nartnirun Junngam

Chapter 7: The Final Award in Occidental v. Ecuador. . . . . . . . . . . 211 Aloysius P. Llamzon Chapter 8:

The Awards in Wena Hotels Limited v. Arab Republic of Egypt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Julie Maupin

Chapter 9: The Award in Thunderbird v. Mexico. . . . . . . . . . . . . . . 261 Santiago Montt Chapter 10: The Final Award in Loewen v. United States . . . . . . . . . 291 Dirk Pulkowski Chapter 11: The Award in Petrobart Limited v. Kyrgyz Republic . . . . 323 Galina Zukova Table of Legal Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349 Table of Cases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361 iii

ACKNOWLEDGMENTS This book could not have been completed had it not been for the extraordinary efforts of Rocío Digón and George S. Georgiev. They reviewed and edited each of the case studies, in addition to contributing their own. In the final stages, Rocío Digón played a particularly important role. Nor could this book have been completed without the patience and skill of Cina Teixeira Santos, who managed the production of the manuscript with her customary efficiency, grace, and patience. The support of Dean Harold Koh and the Yale Law School was indispensable. On behalf of all the contributors to this volume, we acknowledge our gratitude. G.A.A. & W.M.R.

v

CHAPTER 1

HOW WELL ARE INVESTMENT AWARDS REASONED? Guillermo Aguilar Alvarez and W. Michael Reisman

This collection of papers emerged from a seminar on international investment law that we taught jointly at the Yale Law School in 2005–2006. The students brought a rich experience and, as important for a subject like this, a rich national diversity. A considerable part of the seminar involved close reading of recent international investment arbitral awards. These decisions have emerged, for the moment, as the most important engines of legal development in this field. Most of the major policies animating international investment law are now quite well known, so it is not particularly difficult to appraise the substantive outcomes of particular awards by reference to those standards. But our students were repeatedly struck by the number of cases in which the reasons purportedly supporting normative démarches or innovative formulations were either scanty or simply did not parse. In a few cases, the most careful reading of the award failed to reveal key factual findings, major or minor syllogistic premises, or normative judgments that were necessary to reach a conclusion. In one case, critical factors for determining damages—the bottom line—were not identified or discussed. Interestingly, in almost all (but not all) instances, the students felt that the right decision had probably been reached. But without the building blocks that reasons reflect, one could not reconstruct or “reverse engineer” the reasoning of the tribunal. From this experience, we and our students concluded that it would be a useful exercise to examine the adequacy of reasons in some of the most important recent international investment law awards in order to see if there were significant trends with policy implications. The studies in this collection represent the best of the seminar.

1

2 • The Reasons Requirement in International Investment Arbitration

I.

THE REASON FOR THE REASONS

Reasons seem to be particularly important in the area of international investment law. Unlike international commercial arbitration, which is conducted entirely by and for professionals and whose awards are only rarely published, international investment awards may have a major political impact on an entire country. In some cases, the economic consequences of investment awards may have long-term implications for the economic vitality of the state concerned. Citizens, as “stakeholders” who may have to bear the consequences of an award, are entitled to know why a decision was made. Similarly, the political opposition in democratic states will wish to understand why a decision was made. The quality of internal political debate about investment policy, which is surely one of the most vital issues for developing countries, can hardly proceed if the reasons for decisions that are being taken are not clear. Thus, the reasons requirement, as a control mechanism in other sectors of arbitration, acquires a greater importance in international investment arbitration. Indeed, these considerations should lead to the drafting of awards that are accessible to reasonably intelligent people and are not so recondite that none but a small group of specialists can comprehend them. Investment arbitrations are frequently conducted under the auspices of the International Center for the Settlement of Investment Disputes (ICSID). This was particularly important for the students’ inquiries because ICSID arbitration was designed exclusively for international investment disputes. Moreover, Article 48(3) of the ICSID Convention1 and Article 47(1)(i) of the ICSID Arbitration Rules,2 like the default rule in general arbitral law (e.g., Article 52(2)(i) of the ICSID Additional Facility Rules3 and Article 32(3) of the U.N. Commission on International Trade Law (UNCITRAL) Arbitration Rules),4 requires that a 1 Convention on the Settlement of Investment Disputes between States and Nationals of Other States, art. 48(3), ICSID (World Bank), opened for signature Mar. 18, 1965, 575 U.N.T.S. 159. 2 International Centre for Settlement of Investment Disputes Convention, Regulations and Rules art. 47(i), (Oct. 14, 1966). 3 International Centre for Settlement of Investment Disputes, ICSID Additional Facility Rules, art. 52(2)(i). 4 UNCITRAL Arbitration Rules of 1976, adopted by the U.N. General Assembly on Dec. 15, 1976 art. 32(3), 15 I.L.M. 701 (1976).

How Well Are Investment Awards Reasoned? • 3

tribunal provide reasons for its opinions. The sanction under Article 52(1)(e) of the Convention for failing to comply is severe. Either party may request annulment of the award by an application in writing addressed to the Secretary-General on one or more of the following grounds: . . . that the award has failed to state the reasons on which it is based. The progress of the drafters’ thinking on this issue, which is tracked in the legislative history of the Convention, indicates that requiring reasons was important for the drafters from the very genesis of the Convention. The preliminary draft of Article 52(1) (Document 24) simply reproduced Georges Scelle’s formula in the International Law Commission’s (ILC’s) Draft Code on International Commercial Arbitration, which did not make express the need for reasons but that, as the ILC’s commentary made clear, included the requirement of a reasoned award among minimum procedural standards. The first ICSID draft (Document 43) disaggregated the minimum procedure requirement by turning a “failure to state the reasons for the award, unless the parties have agreed that reasons need not be stated” into an express ground for annulment. In other words, at that stage, the drafters were sufficiently concerned about the need for reasoned awards to require them explicitly, but they treated it as a dispositive requirement: reasons would be required in ICSID arbitrations unless the parties decided to dispense with them, as they were entitled to do. The Revised Draft (Document 123), which became Article 52(1) of the Convention, suppressed the dispositive option and transformed the need for reasons into an explicit requirement out of which the parties could not opt by private agreement. Hence, whatever general international law may hold on the matter of party-optional regimes with respect to reasoned awards, the text of the ICSID Convention makes the requirement of a reasoned judgment a peremptory lex specialis. It is unclear from its drafting history how much detail in reasoning the Convention requires. In particular, it is not certain whether the requirements of Article 52(1)(e) incorporate Article 42(1) and Article 48(3) and thus require the expression of reasons or motivation, by explicit reference to the applicable law, of every question. Article 42(1) provides: The Tribunal shall decide a dispute in accordance with such rules of law as may be agreed by the parties. In the absence of


such agreement, the Tribunal shall apply the law of the Contracting State party to the dispute (including its rules on the conflict of laws) and such rules of international law as may be applicable. Article 48(3) provides: The award shall deal with every question submitted to the Tribunal, and shall state the reasons upon which it is based. Some possible support for a more expansive scope for Article 52(1)(e) may be gained from the comment by the representative of the Philippines at the Convention.5 Then again, the authoritative “Users’ Guide” to treaties, the Vienna Convention on the Law of Treaties, admonishes us to read Article 52 on its own terms and not to seek to attach one of the varied views expressed at conferences as if it were the authoritative reading. But that is not how all the ad hoc committees called upon to apply that provision viewed the matter. II.

ICSID PRACTICE

The review system in ICSID grappled with this question in the most detailed way in the first series of annulment petitions under Article 52. In 1971, the West German multinational company, Klöckner, and the government of Cameroon contracted in a “Basic Agreement” with an ICSID Arbitration clause, to create a fertilizer company called SOCAME. Klöckner owned 51 percent and was responsible for the technical and commercial management of SOCAME, and the government of Cameroon owned the remaining 49 percent and would guarantee payment for the company. In the following years, Klöckner and Cameroon signed a “Supply Contract” in 1971 and a “Convention” in 1973, defining additional rights and obligations and again including an ICSID arbitration clause. Klöckner’s management responsibilities were previously left vague, but were later specified in a management agreement in which there was an International Chamber of Commerce (ICC) arbitration clause. Once the factories began production, SOCAME’s output was continually below original projections. The government of Cameroon was

5 Documents Concerning the Origin and Formulation of the Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (History, Vol. 2, Part 2, at 856).


advised by its subsequent consultant to redesign the factory to improve the processes. The government of Cameroon then asked Klöckner to participate in the redesign and to make additional capital contributions, but instead Klöckner yielded its majority shareholding. Even after the redesign, the factory proved unworkable and SOCAME, now with the government of Cameroon as a majority shareholder, refused to pay Klöckner for the factory. Klöckner initiated an ICSID arbitration against the government of Cameroon for payments guaranteed under the “supply contract” that stated Klöckner would provide just a factory, which it had done, but which did not specifically require the factory’s workability. Cameroon counterclaimed a violation of obligations and sought to bring up management issues. Klöckner responded that management issues should be brought to the ICC, and not to ICSID, as provided by the arbitration clause in the management contract. The ICSID tribunal6 was not convinced by Klöckner’s arguments. In effect the award was a victory for Cameroon. Klöckner’s party-appointed arbitrator in his dissenting opinion considered the award as null due to “important mistakes, the numerous contradictions and failures to state the grounds and the misrepresentation of contractual clauses.” Klöckner immediately applied to the Secretary-General of ICSID for nullification of the award under Article 52 of the ICSID Convention. This was the first application for annulment ever lodged against an ICSID award, and the ad hoc committee undertook a searching and detailed examination of every aspect of the ICSID review process. Our interest is only with respect to reasons. In deciding on a tribunal’s obligation to state reasons, the Klöckner committee held that there should be a formal rather than a substantive test of reasons. In the view of the committee, as long as the award’s “answers seem tenable and not arbitrary, they do not constitute a manifest excess of power,” and in case of doubt, analysis would be in favor of the validity of the award. But that presumption could prove quite fragile. It is not for the Committee to imagine what might or should have been the arbitrators’ reasons, any more than it should substitute “correct” reasons for possibly “incorrect” reasons, or deal “ex post facto” with questions submitted to the Tribunal which the 6 Klöckner Industrie-Anlagen GmbH and Others v. Republic of Cameroon, 2 ICSID REP. 9 (1983).


Award left unanswered. The only role of the Committee here is to state where there is one of the grounds for annulment set out in Article 52 of the Convention, and to draw the consequences under the same Article. In this sense, the Committee defends the Convention’s legal purity.7 With a theory of “legal purity,” the Klöckner committee established a very high standard for reasons. Consistent with such a theory, the committee’s methodology did not include an assessment of the gravity of the defect in the reasons. Nullification of awards seemed to be on a hair trigger. The critical paragraph 58 of its decision provides: In the opinion of the ad hoc Committee, the provisions of Article 42 could not be interpreted as stating simple advice or recommendations to the arbitrators or an obligation without sanction. Obviously, and in accordance with principles of interpretation that are recognized generally—for example, by Article 31 of the Vienna Convention on the Law of Treaties—Article 52 on the annulment of awards must be interpreted in the context of the Convention and in particular of Articles 42 and 48, and vice versa. It is furthermore impossible to imagine that when they drafted Article 52, the Convention’s authors would have forgotten the existence of Articles 42 or 48(3), just as it is impossible to assume that the authors of provisions like Articles 42(1) or 48(3) would have neglected to consider the sanction for noncompliance. In paragraph 68, the committee elaborated its own reasoning: First, it should be asked whether the arbitrator’s duty under Article 42(1) requiring the arbitrator to apply “the law of the Contracting State” is or can be fulfilled by reference to one “basic principle,” and what is more, without making any more precise reference. This may be doubted if one considers the difference between “rule” and “principle” (and in particular “basic principle”) and the classic definition of law in the objective sense as a body of rules. It will also be noted in this context that Article 42(1) itself distinguishes between the concepts of “rules of law” and “principles of law.” 7 Klöckner Industrie-Anlagen GmbH and Others v. Republic of Cameroon, May 3, 1985, English translations of French original in 1 ICSID REV.—FILJ 89, at ¶151 (1986).


In paragraph 69, it proceeded to explain: Furthermore, the reference to “other national codes which we know of,” to the “particularly appropriate” character of the rule “in more complex international ventures, such as the present one” (p. 105) and to the particular importance that “universal requirements of frankness and loyalty . . . be applied in cases such as this one” seem to indicate that the Tribunal may have wanted to base, or thought it was basing, its decision on the general principles of law recognized by civilized nations, as that term is used in Article 38(3) of the Statute of the International Court of Justice. It is not impossible that the Tribunal was prompted to do so by the reference in Article 42(1) in fine to the “principles of international law as may be applicable” although these are not to be confused with “general principles.” Such an interpretation is conjectural and cannot be accepted. Article 42 of the Washington Convention certainly provides that “in the absence of agreement between the parties, the Tribunal shall apply the law of the Contracting State party to the dispute . . . and such principles of international law as may be applicable.” This gives these principles (perhaps omitting cases in which it should be ascertained whether the domestic law conforms to international law) a dual role, that is, complementary (in the case of a “lacuna” in the law of the State), or corrective, should the State’s law not conform on all points to the principles of international law. In both cases, the arbitrators may have recourse to the “principles of international law” only after having inquired into and established the content of the law of the State party to the dispute (which cannot be reduced to one principle, even a basic one) and after having applied the relevant rules of the State’s law. Article 42(1) therefore clearly does not allow the arbitrator to base his decision solely on the “rules” or “principles of international law.” According to the Klöckner decision, even a reference to a “principe de base” would not meet the standards of Article 52 unless it was accompanied by some adequate documentation. Thus in paragraph 71, the committee asked itself: Does the “basic principle” referred to by the Award (p. 105) as one of “French civil law” come from positive law, i.e., from the


law’s body of rules? It is impossible to answer this question by reading the Award, which contains no reference whatsoever to legislative texts, to judgments, or to scholarly opinions. In this respect the contrast is striking between Section 2 (on the “duty of full disclosure”) and Section 3 (on the exceptio non adimpleti contractus, pp. 109–114 and pp. 118, 124, 126, etc). Section 3 contains a great number of references to scholarly opinion (doctrine) as well as, directly or indirectly, to case law (jurisprudence). One could therefore assume that in the case of Section 2, regarding the duty of frankness, the arbitrators either began a similar search for authorities but found it unproductive or, more likely, thought that a search for positive law was unnecessary. A contradiction of reasons is, according to paragraph 116 of the Klöckner decision, another ground for nullity: As for “contradiction of reasons,” it is in principle appropriate to bring this notion under the category “failure to state reasons” for the very simple reason that two genuinely contradictory reasons cancel each other out. Hence the failure to state reasons. The arbitrator’s obligation to state reasons which are not contradictory must therefore be accepted. Establishing the existence of such a contradiction may certainly give rise to difficulties, for example if one of the reasons involves a principal claim, while the other involves a counterclaim. This, however, cannot in itself warrant passing over the question of contradiction, at least in terms of admissibility. It should also be noted that, in the event that contradictory reasons lead to the conclusion that there was a failure to state reasons, it may be asked whether this failure causes any harm to the party seeking annulment (cf. the principle “no annulment without grievance”) and whether the award is not sufficiently well founded by other reasons stated in the award. Similarly, in the view of the Klöckner committee, “hypothetical” reasons are invalid: It was incumbent upon the Applicant for Annulment to show that the contested Award is based, on one point or another, on a


simple hypothesis, instead of on facts or definite legal arguments8 (emphasis in original). As to the sufficiency of the reasons’ content, the Klöckner ad hoc decision set a teleological rather than quantitative test. It is worth noting that the “reasons” referred to in Article 52(1)(e) are, as indicated more clearly in the English and Spanish texts [of the Washington Convention], “reasons on which it is based ” or “los motivos en que se funde.” The reasons should therefore be the basis of the Tribunal’s decision, and in this sense “sufficient.” The latter notion should obviously be approached with special caution if the application for annulment under Article 52 is not to serve as an appeal in disguise. One illustration of this danger is found in the Application for Annulment where, criticizing the “inadequacy of reasons because of misconstruction” (dénaturation) (pages 17, 24) (a concept known in French law but absent from the Convention’s text), the application puts forward a variety of considerations, some of which belong to an appeal proceeding and are consequently inadmissible. Interpretation of the concept of “failure to state reasons” is therefore decisive. It is especially delicate because of the absence of any previous interpretation of the Washington Convention and the lack of sufficiently clear or consistent indications from prior international practice. The text of this article requires a statement of reasons on which the award is based. This does not mean just any reasons, purely formal or apparent, but rather reasons having some substance, allowing the reader to follow the arbitral tribunal’s reasoning, on facts and on law. A middle and reasonable path is to be satisfied with reasons that are “sufficiently relevant,” that is, reasonably capable of justifying the result reached by the Tribunal. In other words, there would be a “failure to state reasons” in the absence of a statement of reasons that are “sufficiently relevant,” that is, reasonably sustainable and capable of providing a basis for the decision.9 (emphasis in original) 8

Id. at ¶ 126.

9

Id. at ¶¶ 118–120.


Hence, the Klöckner ad hoc committee concluded: There would be a “failure to state reasons” if no reasoning or explanation whatsoever, or no “sufficiently relevant” or “reasonably acceptable” reasoning could be found for some conclusion or decision in the Award.10 The Klöckner committee decision constitutes the most thorough discussion of this particular ground of nullity in the jurisprudence of ad hoc committees. It sets an extremely high, indeed unprecedented standard. Not surprisingly, the next two ad hoc committees had to address the issue of the reasons requirement and they introduced significant variations on the Klöckner theory. In AMCO v. Indonesia, AMCO, a U.S. company, accepted a guarantee with the government of Indonesia to invest a specific amount of capital in a hotel in Jakarta. In time, Indonesia concluded that AMCO had not invested the capital that was promised and so, on the night of March 30, 1980, a force affiliated with the government of Indonesia forcefully removed AMCO’s hotel management and seized the hotel. Later on, the government revoked AMCO’s license citing a breach of contract. Following these events and unsuccessful efforts to settle the dispute in Indonesian courts, AMCO exercised its right to ICSID arbitration. The tribunal’s award was decided in AMCO’s favor, but it was based on reasoning that was laconic in some parts and opaque and puzzling in others. Notably, the award did not include a determination as to the question under which legal system the issue of materiality was being tested. Indonesia, relying on Klöckner v. Cameroon, applied for annulment under Article 52 and was successful with respect to parts of the award. The Amco committee, unlike the Klöckner committee, flatly rejected the notion of automatic nullification as a consequence of any mechanical or technical discrepancy. In its view, an ad hoc committee should refuse to nullify if the defect in reasoning is de minimis or an obiter dictum. Rather, an ad hoc committee should concentrate on the question of whether there were any material violations in the award. In response to the question of annulment for lack of reasons in the award, the Amco v. Indonesia committee departed from Klöckner:

10

Id. at ¶ 130.


If it be true that a full control and review of the reasoning followed by an ICSID tribunal would transform an annulment proceeding into an ordinary appeal, it is also true that supporting reasons must be more than a matter of nomenclature and must constitute an appropriate foundation for the conclusions reached through such reasons. Stated a little differently, there must be a reasonable connection between the bases invoked by a tribunal and the conclusions reached by it. The phrase “sufficiently pertinent reasons” appears to this ad hoc Committee to be a simple and useful clarification of the term “reasons” used in the Convention.11 This involved a drastic reduction of the reasons requirement as it had been formulated by the ad hoc committee in Klöckner. In 1971, Guinea and the Maritime International Nominees Establishment (MINE) created a joint venture with a Guinea-appointed president and a MINE-appointed vice president. After considerable disagreement between Guinea and MINE, Guinea contracted with an outside company. MINE thereupon initiated an ICSID arbitration citing a breach of contract. The tribunal awarded in MINE’s favor and, following the lead of Klöckner and AMCO, Guinea applied for a review, citing a failure, inter alia, to state reasons and a failure to address certain allegedly pivotal questions. The MINE ad hoc committee produced its own formulation of the reasons requirement: The Committee is of the opinion that the requirement that an award has to be motivated implies that it must enable the reader to follow the reasoning of the Tribunal on points of fact and law. It implies that, and only that. The adequacy of the reasoning is not an appropriate standard of review under Paragraph (1)(e), [of Article 52] because it almost inevitably draws an ad hoc Committee into an examination of the substance of the tribunal’s decision, in disregard of the exclusion of the remedy of appeal by Article 53 of the Convention. A Committee might be tempted to annul an award because that examination disclosed

11 Amco Asia Corp. and Others v. The Republic of Indonesia, Decision on Annulment (May 16, 1986), 1 ICSID REP. 509, 520–21.


a manifestly incorrect application of the law, which, however, is not a ground for annulment. In the Committee’s view, the requirement to state reasons is satisfied as long as the award enables one to follow how the tribunal proceeded from Point A. to Point B. and eventually to its conclusion, even if it made an error of fact or of law. This minimum requirement is in particular not satisfied by either contradictory or frivolous reasons.12 Challenges to investment awards on the ground of failure to comply with the reasons requirement have also been lodged before national courts when the ICSID Convention was not the jurisdictional vehicle. Thus, in the Metalclad v. Mexico North American Free Trade Agreement (NAFTA) arbitration, Mexico sought to have the award set aside in the Supreme Court of British Columbia.13 With respect to the reasons requirement, Mexico alleged: (a) the failure of the Tribunal to deal with numerous questions related to the finding by the Tribunal of a breach of Article 1105; (b) the failure of the Tribunal to address the question of whether Metalclad acted in bad faith in bringing the claim; (c) the failure of the Tribunal to deal with deceptions in Metalclad’s memorial when it asserted that it did not become aware of the need for a municipal construction permit until December 1995 and when it asserted that it had expended $20.5 million in connection with the landfill facility at the Site; and (d) the failure of the Tribunal to set out the precise manner in which it calculated the damages to which Metalclad was entitled, including the failure to address the evidence that the value of COTERIN’s assets was declared by Metalclad in 1996 for tax purposes to be only $136,339.14 Justice Tysoe reviewed the Klöckner, Amco, and MINE decisions, which we considered earlier, for illumination of the ICSID Additional Facility 12

MINE v. Guinea, Decision on Annulment (Dec. 22, 1989), 4 ICSID REP. 79, 88.

13

United Mexican States v. Metalclad, 2001 B.C.S.C. 664, Per Justice Tysoe.

14

Id. at ¶ 119.


Article 53(1).15 In his opinion, which has been noted for its remarkably zealous review of an international arbitral award, Justice Tysoe introduced a very modest standard with respect to the reasons requirement: The tribunal must answer the questions that have been submitted to it and give its reasons for its answers. In other words, the tribunal must deal fully with the dispute between the parties and give reasons for its decision. It is not reasonable to require the tribunal to answer each and every argument which is made in connection with the questions which the tribunal must decide.16 In the more recent decision of Mitchell v. Democratic Republic of Congo,17 the ad hoc committee set aside the tribunal’s award based on a failure to state reasons and a manifest excess of power in the decision to accept jurisdiction. The case involved the alleged expropriation of a legal consulting practice which Patrick Mitchell, a U.S. national, had set up in the Democratic Republic of Congo (DRC). In March 1999, a domestic Military Court ordered an “intervention” on the premises of the firm, which involved putting the office under seal, seizing documents belonging to the firm, and incarcerating two of the firm’s lawyers.18 The tribunal’s award, issued in February 2004, found that Mitchell had been a victim of expropriation by the DRC in violation of the U.S.-DRC bilateral investment treaty (BIT), and it ordered the DRC to pay Mitchell compensation and interest, and assume responsibility for his legal fees. The DRC applied for an annulment of the award claiming that the tribunal had manifestly exceeded its powers in its decision on jurisdiction and with regard to the law applicable to the dispute, and that the tribunal had failed to state reasons with regard to (1) accepting jurisdiction ratione materiae, (2) finding that the measures under dispute constituted expropriation, and (3) the amount of the compensable damages.19

15 “The award shall be made in writing, shall deal with every question submitted to the Tribunal and shall state the reasons upon which it is based.” ICSID Additional Facility Rules, supra note 3, at art. 53(1). 16

United Mexican States v. Metalclad, supra note 13, at ¶ 122.

Mr. Patrick Mitchell v. the Democratic Republic of Congo, Decision on Annulment, Nov. 1, 2006, available at http://ita.law.uvic.ca/documents/mitchellannulment.pdf. 17

18

Id. at ¶ 1.

19

Id. at ¶ 17.


The ad hoc committee considered the relevant legal standard and found that: an annulment proceeding is different from an appeal procedure and . . . does not entail the carrying out of a substantive review of an award. [T]he grounds invoked by the Respondent in respect of the Award are sometimes based on arguments which touch upon the merits of the case. The ad hoc Committee is fully aware of the risk to which the nature of this case exposes it. Nor is there any doubt that the grounds for annulment set out in Article 52 must be examined in a neutral and reasonable manner, that is, neither narrowly nor extensively. An ad hoc Committee should not decide to annul an award unless it is convinced that there has been a substantial violation of a rule protected by Article 52.20 The ad hoc committee noted its view that “a failure to state reasons exists whenever reasons are purely and simply not given, or are so inadequate that the coherence of the reasoning is seriously affected.” It adopted and quoted the MINE standard cited above and further stated that “a seriously contradictory reasoning would be equivalent to a failure to state reasons under Article 52(1)(e), provided that the contradiction is at the same time apparent, to a point such that the ad hoc Committee cannot be reproached for engaging in an analysis of the merits.”21 Specifically, the committee found that the award was “incomplete and obscure” with regard to its finding that the investor’s interest was in fact an investment within the meaning of the ICSID Convention and the BIT:22 The Award . . . refers to various fragments of the operation, without finally indicating the reasons why it regards it overall as an investment . . . Such an inadequacy of reasons is deemed to be particularly grave, as it seriously affects the coherence of the reasoning and, moreover, as it opens the door to a risk of genuine abuses, to the extent that it boils down to granting the qualification as investor to any legal counseling firm or law firm estab20

Id. at ¶ 19.

21

Id. at ¶ 21.

22

Id. at ¶ 40.

23

Id.


lished in a foreign country, thereby enabling it to take advantage of the special arbitration system of ICSID.23 As the committee was unable to decipher the tribunal’s reasoning in concluding that “Mitchell & Associates” constituted an investment, it annulled the award, in part, for a failure to state reasons. The ad hoc committee’s finding of a lack of adequate reasoning with regard to the existence of investment had decisive implications for the DRC’s second claim of “manifest excess of power.” The committee admitted that, without the finding of insufficient reasoning, this claim would not have succeeded24 but ultimately found that “the combination of flaws in the Award is such that an excess of power on the part of the Arbitral Tribunal must be acknowledged.”25 At paragraph 48, the committee concluded that it is “obligated to annul the Award of February 9, 2004 due to the failure to state reasons and manifest excess of powers committed by the Arbitral Tribunal in its decision to declare itself competent.”26 Notwithstanding, the committee decided to continue its review of the tribunal’s decision: given the importance of certain arguments debated between the parties in respect of these criticisms, the ad hoc Committee deems it useful to discuss the other grounds briefly. Needless to say, the discussion which follows is written as if the ‘Mitchell & Associates’ firm did constitute a genuine investment and the Arbitral Tribunal properly considered that it had jurisdiction.27 After making these findings and annulling the award, the tribunal considered the remainder of the DRC’s arguments. Operating under the assumption that Mitchell & Associates was an investment, the committee evaluated the claim that the valuation of compensation was not supported by adequate reasons and asserted that Article 52(1)(e) applied to findings on the merits: Article 52(1)(e) may be applicable, even in respect of questions on the merits, when the reasoning of the Arbitral Tribunal is 24

Id., ¶ 45.

25

Id., ¶ 46.

26

Id., ¶ 48.

27

Id.


inadequate or contradictory. It must nevertheless be observed, first, that the reasoning of the Arbitral Tribunal as to the valuation of the compensation is lengthy and detailed, covering six pages of the Award, and it appears from its very wording that the parties had ample opportunity to express themselves on all of the issues relating to that valuation. The Committee will examine the various points raised by the DRC in a summary manner, within the limits of a control of the coherence of the reasoning, and without expressing itself on any possible poor decision which it is not called upon to correct in any event.28 In the above paragraph, the committee attempted to distinguish between an appellate review of the merits—which would allow for the correction of a “poor decision”—and an Article 52 review of the merits— which would stay “within the limits of a control of the coherence of the reasoning.” The committee found that the following are outside the limits of control: revision of incorrect findings, introduction of new arguments, and consideration of fullness of reasoning. None of the respondent’s claims regarding the tribunal’s failure to state reasons with respect to the valuation of compensation were accepted; the committee was sufficiently satisfied that the reasoning requirement was met within the confines it previously set forth. Thus, two standards were proposed: a high standard, enunciated by the ad hoc committee in Klöckner and a lower standard enunciated by the Amco and MINE committees, and now adopted by the Mitchell committee. The general principle enunciated in the more modest formulations in Amco, MINE, and Mitchell—that the reasons for an award should, at the very least, permit an understanding of the reasoning of the tribunal— seems correct to us, as both a textual and a teleological interpretation of the ICSID reasons regime, for it serves as a mode of control on the arbitral tribunal’s discretion and a confirmation that the tribunal respected other explicit requirements in Article 52. If the reasons do not provide enough of an insight into how a tribunal decided, there is no way for a reviewer to know whether the tribunal did what it was required to do under its rules and the applicable law. In short, in the absence of reasons, there can be no reasoned or principled control. As this book goes to publication, the ad hoc committees in the Lucchetti v. Peru and the CMS v. Republic of Argentina cases have issued decisions on 28

Id. at ¶ 65.

29

Industria Nacional de Alimentos, S.A. and Indalsa Perú, S.A. v. The Republic of


annulment.29 The investor in the Lucchetti case alleged, inter alia, that “the Tribunal compounded its other errors by presenting genuinely contradictory reasons for its ruling, and by failing to deal with critical questions raised by Lucchetti, both of which constitute a “failure to state reasons” within the scope of Article 52(1)(e) of the ICSID Convention.”30 More specifically, The Claimants allege that the Tribunal presented contradictory reasons for its Award by basing itself on different and inconsistent standards such as the “same subject matter” and the “same origin or source” of the disputes, whether certain elements were “central to the dispute,” when the dispute “crystallised,” and whether, “as a practical matter,” it was the same dispute. These standards were contradictory and unclear, and they made the Award contradictory and unclear. The tribunal also failed to deal with Lucchetti’s arguments: (a) that the preamble to Decree 259 was merely a pretext and that the real reasons for the Decree were political, (b) that the 1998 judgments were substantially correct, and (c) that there was a distinction between the dispute with the municipal authorities and the dispute with the Republic of Peru.31 As for Peru, it contended that “the reasons in the Tribunal’s award are clear and coherent,”32 and continued: Even if the Tribunal had failed to address the three questions raised by Lucchetti, the Tribunal would not have committed an annullable offence. But in any case, the Tribunal dealt with the questions, the two first questions in determining whether the dispute was a new dispute and the last question—although it was

Peru, ICSID Case No. ARB/03/4, Annulment Proceeding (Sept. 5, 2007) [hereinafter Lucchetti v. Peru]; CMS Gas Transmission Co. v. Argentine Republic, ICSID Case No. ARB/01/8, Annulment Proceeding (Sept. 25, 2007). 30

Id., Lucchetti v. Peru at ¶ 33.

31

Id. at ¶¶ 51 and 52.

32

Id. at ¶ 62.


not argued by Lucchetti—by the general finding that it did not have jurisdiction over Decrees 258 and 259.33 The ad hoc committee relied on MINE as its principal authority34 but also derived guidance from the annulment decision in Vivendi, which it read as holding that annulment under Article 52(1)(e) should only occur in clear cases and that a failure to state reasons, in order to lead to annulment, must not only lack in any expressed rationale, but the relevant point must also be necessary to the tribunal’s decision.35 A majority of the Committee concluded that The Ad hoc Committee has found above that the Award does not give a full picture of the various elements which should be taken into account for treaty interpretation under the Vienna Convention. However, in order to establish whether it was the dispute from 1997–1998 that continued after the BIT had entered into force, the Tribunal did refer to various standards adopted in international case-law and doctrine and set out the elements which the Tribunal found conclusive. The Committee cannot find in the Tribunal’s reasoning any contradiction or lack of precision such as to leave a doubt about the legal or factual elements upon which the Tribunal based its conclusion. Moreover, the Committee is satisfied that the Tribunal examined all Lucchetti’s arguments and finds that it dealt with them in the Award to such an extent and in such a manner as could reasonably be required.36 Sir Franklin Berman dissented on this point. He distinguished between the reasons requirement at the jurisdictional phase of a case and, by implication, the reasons requirement at the merits phase: when a Tribunal proposes to non-suit a Claimant at the initial stage, i.e. so as to preclude any airing of the claims on their merits (or demerits), the grounds for doing so must be clear and 33

Id. at ¶ 63.

34

MINE, supra note 12, at ¶¶ 5.08, 5.09; Lucchetti v. Peru, supra note 29, at ¶ 127.

35

Lucchetti v. Peru, supra note 29, at ¶ 128.

36

Id. at ¶ 129.


strong, and in particular that they must be clearly explained and justified, so as to enable the Claimant (not to mention other consumers of the ICSID system) to understand what the Tribunal has done and why. Where, on the other hand, the case is not sufficiently clear as to enable the issue to be convincingly determined in limine, the proper course is plainly that provided for in Article 41(2) of the Convention and in the Arbitration Rules, namely to join the preliminary objection to the merits, and determine it then on the basis of full and complete argument. The converse of this proposition is of course that, if a Tribunal chooses to decline jurisdiction at the preliminary stage without adequately explaining the reasons why, then one is at once within the area of annullable error—if not on the basis of an excess of powers, then at least on the basis of a failure to give reasons (though, as the ad hoc Committee correctly observes at paragraph 72 of the Decision, it is quite possible to conceive of circumstances in which two grounds for annulment should not be thought of as operating in isolation, but instead as reinforcing one another).37 Sir Franklin was not, however, proposing a rigid or extreme (or, one might say, a trivial) standard. For example, he agreed with the majority that the failure to describe “what rules of treaty interpretation it was applying” would not be a ground for annulment if there was no indication that, in fact, the tribunal had disregarded any significant element of the well-known rules of treaty interpretation.38 But, for Sir Franklin, “the real question . . . is whether [the members of the tribunal whose award was under review] adequately explained what they were doing in the interpretative process, and did so specifically with a very particular care needed from a tribunal proposing, on the basis of the interpretative outcome, to decline jurisdiction altogether.”39 On this point, Sir Franklin found the award “so defective that I would be prepared to annul it.”40 In CMS Gas Transmission Company v. Argentine Republic, the ad hoc committee annulled, in part, the tribunal’s award of May 12, 2005. That award had found that Argentina had breached the fair and equitable 37

Id. at ¶ 4.

38

Id. at ¶ 5.

39

Id. at ¶ 6.

40

Id. at ¶ 6.


treatment and umbrella clauses of the Argentina-US BIT, for which it had awarded U.S. $133.2 million to CMS. The committee annulled the tribunal’s finding on the umbrella clause for a failure to state reasons under Article 52(1)(e) of the ICSID Convention, but rejected all of the remaining applications for annulment. In some ways, CMS, though citing to a number of prior ad hoc committee decisions, represents a departure with respect to the reasons requirement of Article 52 and, more broadly, from the more modest role for ad hoc committees that had succeeded the rather extensive conception elaborated by the Klöckner committee. Before entering into the substance of the claims, the CMS committee associated itself with the “limited function” of ICSID annulment proceedings and set forth the standard of review it would apply to determine whether an award was based upon a manifest excess of powers or failed to state reasons. The committee explicitly associated itself with the annulment decisions in MTD v. Chile, MINE, Vivendi, and Wena Hotels. With respect to a manifest excess of powers, the committee distinguished between manifest excess of powers and manifest error of law, as established in MINE: “Disregard of the applicable rules of law must be distinguished from erroneous application of those rules which, even if manifestly unwarranted, furnished no ground for annulment.”41 Concerning the failure to state reasons, the committee adopted the Wena committee’s standard of explaining errors in a tribunal’s reasoning that do not amount to a failure to state reasons: It is in the nature of this ground of annulment that in case the award suffers from a lack of reasons which can be challenged within the meaning and scope of Article 52(1)(e), the remedy needs to be the annulment of the award. The purpose of this particular ground for annulment is not to have the award reversed on its merits. It is to allow the parties to understand the Tribunal’s decision. If the award does not meet the minimal requirement as to the reasons given by the Tribunal, it does not necessarily need to be resubmitted to a new Tribunal. If the ad hoc committee so concludes, on the basis of the knowledge it has received upon the dispute, the reasons supporting the Tribunal’s conclusions can be explained by the ad hoc Committee itself.42

41

MINE, supra note 12, at ¶¶ 5.03–5.04.

Wena Hotels Ltd. v. Arab Republic of Egypt, Decision on Application for Annulment, ¶ 83 (Feb. 5, 2002), 41 I.L.M. 933 (2002). 42


The Wena standard is not without ambiguity. It could mean that an ad hoc committee will not annul when the minimum reasons requirement has not been met as long as the committee can infer the reasons underlying the tribunal’s decision. But it could also mean that the committee will not annul if it concludes, from its own analysis, that the correct conclusion was reached for the wrong reasons. As we will see, the CMS committee apparently understood the dictum in Wena in this second sense. The committee found that the tribunal failed to state its reasons in finding that the Argentine Republic had violated the umbrella clause of Article II(2)(c) of the BIT. The CMS tribunal’s discussion on the umbrella clause had been brief. In its award on jurisdiction,43 the tribunal had noted the claimant’s position to the effect that both the BIT and the whole process of investing in TGN was related to the privatization of the gas industry in Argentina, a process which was the subject of specific guarantees and commitments by the Republic of Argentina. These guarantees included measures of legal stability and economic mechanisms aimed at ensuring the financial feasibility and the success of the investment, not just the question of the quality of the shares.44 The tribunal seems to have read this assertion as relating to “whether this dispute arises directly from the investment.”45 Finding that minority shareholders could bring an action under the BIT,46 the tribunal concluded Because, as noted above, the rights of the Claimant can be asserted independently from the rights of TGN and those relating to the License, and because the Claimant has a separate cause of action under the Treaty in connection with the protected investment, the Tribunal concludes that the present dispute arises directly from the investment made and that therefore there is no bar to the exercise of jurisdiction on this count.47

43 CMS Decision on Objections to Jurisdiction (July 17, 2003), 42 I.L.M. 788 (2003). 44

Id. at ¶ 39 (emphasis added).

45

Id. at ¶ 41.

46

Id. at ¶48.

47

Id. at ¶68.


In its award on the merits,48 the tribunal seems to have viewed the above passages as resolving the jurisdictional issue of the umbrella clause.49 As to the merits, the tribunal held that not all contract breaches result in breaches of the Treaty. The standard of protection of the treaty will be engaged only when there is a specific breach of treaty rights and obligations or a violation of contract rights protected under the treaty. Purely commercial aspects of a contract might not be protected by the treaty in some situations, but the protection is likely to be available when there is significant interference by governments or public agencies with the rights of the investor.50 Citing but not discussing seven recent decisions,51 the tribunal applied the distinction to the dispute and concluded that “None of the measures complained of in this case can be described as a commercial question as they are all related to government decisions that have resulted in the interferences and breaches noted.”52 Two measures that uniquely violated the umbrella clause were those that violated, first, “the obligation not to freeze the tariff regime or subject it to price controls” and, second, “the obligation not to alter the basic rules governing the License with TGN’s written consent.53 The committee could not follow the tribunal’s reasoning in concluding that Article II(2)(c) applied to CMS. The claimant had submitted to the ad hoc committee what the committee described as “a literal interpretation” of Article II(2)(c), to wit that “Argentina entered into legal obligations under the License, which were obligations ‘with regard to investments’ under that Article.” After enumerating possible interpretations of the Article II(2)(c) provisions in paragraph 95, the committee concluded that

48

CMS Award of May 12, 2005, 44 I.L.M. 1205 (2005).

Id. at ¶ 299. We say “seems to” because the merits award does not indicate to which part of its jurisdictional award it is referring. 49

50

Id. at ¶ 299.

51

Id. at ¶ 300.

52

Id. at ¶ 301.

53

Id. at ¶ 302.


it is quite unclear how the Tribunal arrived at its conclusion that CMS could enforce the obligations of Argentina to TGN. It could have done so by the above interpretation of Article II(2)(c), but in that case one would have expected a discussion of the issue of interpretation referred to above . . . In these circumstances, there is a significant lacuna in the Award, which makes it impossible for the reader to follow the reasoning on this point. It is not the case that answers to the question raised ‘can be reasonably inferred from the terms used in the decision’; they cannot. Accordingly, the Tribunal’s finding on Article II(2)(c) must be annulled for failure to state reasons.54 This would seem to go beyond the Wena standard, which the CMS committee had explicitly adopted, for, as the committee itself acknowledged, a literal interpretation was a plausible explanation of what the tribunal had done; moreover, literal interpretations, which are the default interpretive mode in international law, are ipse dixit reasons for which one would not expect more than the iteration of the text. But as one studies the committee’s decision, it becomes clear that the committee’s problem was not the absence of reasons so much as the committee’s conviction that even the plausible literal reasoning was incorrect! The umbrella clause had stated that “Each Party shall observe any obligation it may have entered into with regard to investments.” It will be noted that the word “obligation” in the umbrella clause is unqualified. But the committee chose to read the word “obligation” in the umbrella clause as if it were preceded by the word “legal” as if it meant only “legal obligations” while excluding from its ambit “legitimate expectations.”55 Curiously, this holding, which is outcome-determinative, is unsupported by any reasons. On this basis, the committee proceeded to reason: (a) In speaking of “any obligations it may have entered into with regard to investments,” it seems clear that Article II(2)(c) is concerned with consensual obligations arising independently of the BIT itself (i.e. under the law of the host State or possibly under international law). Further they must be specific obligations concerning the investment. They do not cover general requirements imposed by the law of the host State. 54

CMS Annulment, supra note 29, at ¶¶ 96, 97.

55

Id. at ¶ 89.


(b) Consensual obligations are not entered into erga omnes but with regard to particular persons. Similarly the performance of such obligations or requirements occurs with regard to, and as between, obligor and obligee. (c) The effect of the umbrella clause is not to transform the obligation which is relied on into something else; the content of the obligation is unaffected, as is its proper law. If this is so, it would appear that the parties to the obligation (i.e., the persons bound by it and entitled to rely on it) are likewise not changed by reason of the umbrella clause. (d) The obligation of the State covered by Article II(2)(c) will often be a bilateral obligation, or will be intrinsically linked to obligations of the investment company. Yet a shareholder, though apparently entitled to enforce the company’s rights in its own interest, will not be bound by the company’s obligations, e.g. as to dispute settlement. (e) If the Tribunal’s implicit interpretation is right, then the mechanism in Article 25(2)(b) of the ICSID Convention is unnecessary wherever there is an umbrella clause. (f) There is no discussion in the award of the travaux of the BIT on this point, or of the prior understandings of the proponents of the umbrella clause as to its function.56 This then was the basis for the annulment for lack of reasons: 96. In the end it is quite unclear how the Tribunal arrived at its conclusion that CMS could enforce the obligations of Argentina to TGN. It could have done so by the above interpretation of Article II(2)(c), but in that case one would have expected a discussion of the issues of interpretation referred to above. Or it could have decided that CMS had an Argentine law right to compliance with the obligations, yet CMS claims no such right; and Argentine law appears not to recognize it. 97. In these circumstances there is a significant lacuna in the Award, which makes it impossible for the reader to follow the 56



reasoning on this point. It is not the case that answers to the question raised “can be reasonably inferred from the terms used in the decision.” Thus, the CMS committee relied on “right reasons” to dismiss a plausible construction of the tribunal’s reasons for its decision on the effect of the umbrella clause.57 This reliance on “right reasons” may signal a subtle constitutive change in the function of the ad hoc committee procedure under ICSID Article 52 from review to quasi-appeal, which will have major significance for the meaning of “reasons” in future cases. The penchant of the CMS committee to go beyond the functions of Article 52 is manifest in its review of the tribunal’s state of necessity assessment. Here, the committee did not find a failure to state reasons or a manifest excess of powers; however, it did find instances where it sought to “clarify certain points of substance on which, in its view, the Tribunal made manifest errors of law.”58 In these instances, the committee offered right reasons in addition to evaluating the adequacy of the tribunal’s reasons. According to the committee, the tribunal’s first manifest error of reasoning—one that did not amount to a failure to state reasons—concerned the tribunal’s conclusion that the conditions of necessity under Article XI of the BIT were not met59—because the conditions had not been satisfied under Article 25 of the ILC’s Articles on State Responsibility (ILC Article 25): The problem is, however, that the Tribunal stopped there and did not provide any further reasoning at all in respect of its decision under Article XI. The motivation of the Award on this point is inadequate. The Tribunal should certainly have been more 57 The CMS committee’s treatment of umbrella clauses is itself scantily reasoned; it is more declaratory than reasoned and manifests minimal consideration of the range of arbitral views that have been expressed on the thorny problem of interpretation of umbrella clauses. 58

Id. at ¶ 45.

We would like to note that neither the tribunal nor the ad hoc committee addresses the possibility that Article XI of the BIT is a provision intended to apply to matters of national security while ILC Article 25 is intended to apply to state of necessity. This distinction between national security and state of necessity, which has been drawn in international law, is not raised in the CMS awards. Because BIT Article XI makes no express reference to a state of necessity, the tribunal reads Article XI alongside Article IV(3) in order to establish a link between national security and “state of national emergency,” and, thus, state of necessity as defined in ILC Article 25. 59


explicit in specifying, for instance, that the very same reasons which disqualified Argentina from relying on the general law of necessity meant that the measures it took could not be considered ‘necessary’ for the purposes of Article XI either . . . [A]lthough the motivation of the Award could certainly have been clearer, a careful reader can follow the implicit reasoning of the Tribunal as indicated in paragraph 124 above.60 The committee found a similar lack of clarity in its analysis of whether the tribunal manifestly exceeded its powers in its assessment of state of necessity. According to the committee, the tribunal made manifest errors of law in its failure to analyze the individual requirements of BIT Article XI and ILC Article 25 and the relationship between the two articles. After identifying these two errors, the committee devoted two paragraphs to the analysis the tribunal could have undertaken and concluded: These two errors made by the Tribunal could have had a decisive impact on the operative part of the Award. As admitted by CMS, the Tribunal gave an erroneous interpretation to Article XI. In fact, it did not examine whether the conditions laid down by Article XI were fulfilled and whether, as a consequence, the measures taken by Argentina were capable of constituting, even prima facie, a breach of the BIT. If the Committee was acting as a court of appeal, it would have to reconsider the Award on this ground. The Committee recalls, once more, that it has only a limited jurisdiction under Article 52 of the ICSID Convention. In the circumstances, the Committee cannot simply substitute its own view of the law and its own appreciation of the facts for those of the Tribunal. Notwithstanding the identified errors and lacunas in the Award, it is the case in the end that the Tribunal applied Article XI of the Treaty. Although applying it cryptically and defectively, it applied it. There is accordingly no manifest excess of powers.61 One wonders whether such comments, which exceed the ambit of ICSID Article 52, will erode the authority of an award that has, ironically, been confirmed and must be implemented by the respondent state.

60

Id. at ¶¶ 125, 127.

61

Id. at ¶¶ 135–136.


Whatever may be one’s view, the issue of the reasons requirement continues to be a lively one in the jurisprudence of international investment law. III. THE APPROPRIATE STANDARD If commentators agree that reasons are necessary, they do not agree on what makes reasons sufficient or adequate. One group of scholars, which includes Carlston, Fiore, Balasko, and Reisman, has set a relatively low threshold to test the sufficiency of reasons. Carlston states: arbitration is a judicial process. The requirement that a decision be reached by a formally stated process of reasoning would, therefore, seem to be essential. It need not be in meticulous detail; a statement indicating in a general way the legal reasons upon which the award is based will be valid and binding. The circumstance, however, that upon certain aspects of the opinion reasons were lacking cannot reasonably be considered to result in the nullity of the entire decision.62 Reisman has written: A detailed justification of every premise, every step in a process of inference, and every subsidiary conclusion, is unfeasible. Excluding per curiam decisions, which are a rare international phenomenon, few, if any, international judgments and awards have been “fully reasoned.”63 Prior to the Klöckner decision, some case law appears to have implemented this view. In the Case Concerning the Arbitral Award Made by the King of Spain on 23 December 1906 (Honduras v. Nicaragua), which was decided by the International Court of Justice in 1960, Nicaragua sought to impugn an award rendered by the King of Spain on a number of grounds, including inadequacy of reasons. The King’s award was seven and one-half pages long including some five pages of reasoning.64 The Court dealt summarily with the claim based on allegedly insufficient reasons:

62

K. CARLSTON, THE PROCESS OF INTERNATIONAL ARBITRATION at 53 (1946).

63

W. MICHAEL REISMAN, NULLITY AND REVISION at 618 (1971).

I.C.J. Pleadings, Case Concerning the Arbitral Award made by the King of Spain on 23 December 1906 (Honduras v. Nicaragua), Vol. II, at 18–26. 64


an examination of the Award shows that it deals in logical order and in some detail with all relevant considerations and that it contains ample reasoning and explanations in support of the conclusions arrived at by the arbitrator. In the opinion of the Court, this ground is without foundation.65 Another group of scholars, most of whom completed their studies before the ratification or drafting of the ICSID Convention, appears to demand that reasons be much more specific, though none set the bar as high as the Klöckner ad hoc committee. The grounds for reasoned decisions in circumstances in which an appeal may be required are obvious: a key part of the appeal is about the cogency of the reasoning of the first instance of decision. Without reasons, the appeal would be perforce a de novo trial. In Carlsen and Others v. Council of the European Union, the Court said: it is settled case-law that the statement of reasons required by Article 190 of the Treaty must be appropriate to the nature of the measure in question. It must show clearly and unequivocally the reasoning of the institution which enacted the measure so as to inform the persons concerned of the justification for the measure adopted and to enable the Court to exercise its powers of review.66 Without reasons, the Court could hardly exercise its powers of review. But the applicability of this requirement in international commercial arbitration is uncertain. In the arbitration model of the New York Convention,67 appeal has been eschewed. Review in all secondary jurisdictions is limited to the grounds set out in Article V. That provision does not include a reasons requirement and it is difficult to find one by implication. In a particular instance, Article V may, to be sure, acquire a reasons requirement if (1) the parties elect arbitral rules that require a reasoned award or (2) the law of the country in which or under the law of which the award was made requires reasons. Not all of the extant rule 65

1960 I.C.J. 216.

Carlsen and Others v. Council of the European Union, Case T-610/97 R Court of European Communities, at ¶ 38 (1997). 66

67 Convention on the Recognition and Enforcement of Foreign Arbitral Awards, June 10, 1958, 21 U.S.T. 2517, 330 U.N.T.S. 38.


systems in international commercial arbitration require reasons. ICC and ICSID Rules do require reasons, but American Arbitration Association (AAA) Rules for International Arbitration, London Court of International Arbitration (LCIA) Rules, and UNCITRAL Rules permit the parties to waive what would otherwise be the requirement of a reasoned award. And not every national system requires reasons or prohibits parties from dispensing with a reasons requirement. International investment arbitration, as we said, may present special circumstances with respect to a reasons requirement. Unlike international commercial arbitrations, which occur between two private parties both of whom have elected arbitration because they wish confidentiality and for both of whom the issue is essentially commercial, international investment arbitration involves a much wider cast of players. In addition to the government of the host state, which is the invariable respondent, and its loyal opposition, international investment arbitrations involve local industries, non-governmental groups, which may be quite active within the state, and, in a significant number of cases, broad strata of the population, which may be profoundly affected by the outcome of the award. Precisely because awards in international investment law cut so deeply into the national political economy and have such profound and potentially continuing effects on national political processes, their acceptability depends on their persuasiveness, that is, the extent to which they manifest a good faith application of the law. In this respect, a higher threshold for the reasons requirement than might obtain for international commercial arbitration, rather than a lower or more modest one, would seem to commend itself. IV. THE FUTURE OF REASONS Wholly apart from the ICSID requirement, international investment law is evolving, and it is clear that investment tribunals have become major agencies of its evolution. Broad terms such as “the international minimum standard” or its functional equivalent in bilateral investment treaties, “fair and equitable treatment,” appear, at the present time, to be more and more the fulcrum of decision. Like virtually all students of international investment law, we take it for granted that tribunals charged with decision must assign a contemporary meaning to these terms rather than apply an anachronism from the past. But if tribunals do evolve the term, the multiple constituencies of international investment law are surely entitled to an explanation of why a particular tribunal reached the decision that it reached. Insofar as a particular


decision involves a derivation from some older principle, which was already a beneficiary of a general legal consensus, an identification of the major principle and a pellucid logical exercise would appear to be a minimum requirement. In those circumstances in which the problem presented to the tribunal requires an adaptation or even an innovation in policy—and we accept that such situations may sometimes fall to investment tribunals—a purported exercise in logical derivation, far from explaining what is being done, can only conceal what is being done. In these latter circumstances, one might do worse than consider the wise advice of Professor Myres S. McDougal: The essence of a reasoned decision by the authority of the secular values of a public order of human dignity is a disciplined appraisal of alternative choices of immediate consequences in terms of preferred long-term effects, and not in either the timid foreswearing of concern for immediate consequences or in the quixotic search for criteria of decision that transcend the world of men and values in metaphysical fantasy. The reference of legal principles must be either to their internal—logical—arrangement or to the external consequences of their application. It remains mysterious what criteria for decision a “neutral” system could offer.68 This approach seems to recommend itself for its candor and might be preferred over the usual resort to so-called “teleological interpretation,” for it should be apparent to every student of the law that the objects and purposes of any instrument or, indeed, any area of law can be cast so broadly as to justify any of a wide range of interpretive conclusions. Students of arbitration recognize that there may be circumstances in which there will be powerful pressures on the members of the tribunal to be obscure rather than explicitly rational, in Professor McDougal’s sense. There may be circumstances in which an award is drafted so as to be minimally offensive to a respondent government, and there may be an effort to practice judicial parsimony and reach conclusions based on firm principles of law, obviating elaborate reasoning for other claims. And, of course, one cannot exclude the accommodations that must sometimes be reached among the members of three-person tribunals that can lead to distortion in reasoning. 68 Myres McDougal, Perspectives for an International Law of Human Dignity, 1959 AM. SOC’Y INT’L L. PROCS. 107, 121.


These exceptional circumstances notwithstanding, the operational presumption of the collective exercise conducted by our students was that parties to international investment disputes are entitled to fully reasoned awards as are their various internal constituents. Nor do we exclude the quality control that is inherent in an effective reasons requirement. Even without a formal instance of review, a requirement of reasoned decision may act as a potential form of control. Indeed, in many contexts, the demand for reasons is part of an explicit control mechanism. For example, Part III of the Trade-Related Aspects of International Property Rights (TRIPs) Agreement requires state parties to provide “expeditious remedies for infringement.”69 The remedies are to be “fair and equitable”70 and the rulings about TRIPs infringements must be in writing and “reasoned.”71 Similarly, the North American Free Trade Agreement (NAFTA) requires state-to-state arbitration decisions to include findings of fact and a determination as to the measure’s consistency with the treaty.72 In a section entitled “The Future of Reasons,” it is appropriate to acknowledge that our students are a key part of that future. As teachers, we respect our students’ views and did not try to impose our own. Our appraisals of their analyses of awards were not based on agreeing with a student’s conclusions but on the quality of the analyses. It should be no surprise that people from different legal and economic systems look at international investment law quite differently.

69 Agreement on Trade-Related Aspects of Intellectual Property Rights, Marrakesh Agreement Establishing the World Trade Organization, Annex 1C, Legal Instruments —Results of the Uruguay Round, Apr. 15, 1994, 1869 U.N.T.S. 299 art. 41.1 (1994). 70

Id. at art. 41.2.

71

Id. at art. 41.3.

72

Art. 2016(2).

CHAPTER 2

THE FINAL AWARD IN MONDEV INTERNATIONAL V. UNITED STATES OF AMERICA Julien Cantegreil*

I.

INTRODUCTION: A MISSED OPPORTUNITY?

The reasoning provided by the tribunal in the Mondev v. United States1 case has certainly helped structure the chain of North American Free Trade Agreement (NAFTA)2 investment law awards. Inter alia, it has comforted the understanding of the minimum standard of treatment as a matter of evolutionary customary international law. It has settled that the judgment of what is “fair and equitable” cannot be made in the abstract but “must depend on the facts of the particular case.” With respect to causation, it has followed the path taken by the Pope & Talbot 3 and the S.D. Myers4 awards to relax the procedural requirements for bringing a * Ecole normale supérieure and Sorbonne (Paris). Former Member of the Cabinet of the Ministre de l’Economie, des Finances et de l’Industrie (France). Ecole normale supérieure (economics), Agrégation (philosophy), Sorbonne and Yale Law School (law and LL.M.). Substantial improvements are due to Professors W. Michael Reisman and Guillermo Aguilar Alvarez’s comments. Many thanks to Santiago Montt for friendship and thoughtful discussions. Voor Meta van Adelberg, haar impliciete redeneringen. 1 Mondev Int’l Ltd. v. United States of America, ICSID Case No. ARB (AF)/99/2, Final Award (Oct. 11, 2002) [hereinafter Mondev]. See also Mondev Notice of Arbitration (Sept. 1, 1999) [hereinafter Notice of Arbitration]; U.S. Counter Memorial on Jurisdiction (June 1, 2001) [hereinafter U.S. C-Mem] and U.S. Supplementary Memorial on Jurisdiction (Oct. 1, 2001) [hereinafter U.S. Rej.]. These documents are available at http://www.naftaclaims.com and http://www.state.gov/s/l/c3758.htm. Mondev Memorial and Mondev Counter Memorial are not available for analysis. The transcripts of the hearings on the case are also available at http://www.naftaclaims.com [hereinafter Hearings]. 2

U.S.-Can.-Mex. (Dec. 17, 1992), 32 I.L.M. 605 (1993) [hereinafter NAFTA].

Pope & Talbot, Inc. v. Canada, Interim Award on Damages, ¶ 74 et seq. (May 31, 2002), 41 I.L.M. 1347 (2002). 3

4

S.D. Myers, Inc. v. Government of Canada, Second Partial Award at ¶¶ 140–160

33


claim under NAFTA Articles 1116 and 1117. Substantial parts of the tribunal’s reasoning were explicitly endorsed, as early as 2003, by other tribunals.5 On a more critical note, although many concerns were expressed about whether the U.S. position as defendants in the case influenced the outcome, no in-depth analysis of the tribunal’s reasoning on this entire case has been provided. The case was difficult, as exemplified by the fact that the claimant reshaped his claim at each step of the arbitration and by the contradictory concluding note of the arbitrators. Although they dismissed the entire claim, they expressed “some sympathy for Mondev’s situation.” This sympathy was more merit-focused than in the S.D. Myers majority and the Pope & Talbot awards, as the tribunal added that “in the end, the [respondents] succeeded, but only on rather technical grounds” and it “exerciz[ed its] discretion” in making no order for costs or expenses,6 considering that this may take Mondev’s situation “fairly [. . .] into account.”7 Yet the complexity of the case and what has been termed, in a broader context, as “powerful pressures on the members of the tribunal to be obscure rather than explicitly rational”8 could not justify what the following pages present as inadequate reasoning of the critical factual findings and major premises in the award. After presenting the factual background and the judicial proceedings of the case in Section II, the adequacy of the award will be discussed, showing, in Section III, that the tribunal has not well reasoned its Article 1105 claim, and arguing, in Section IV, that had it done so, it would have (Oct. 21, 2002), available at http://www.naftaclaims.com/Disputes/Canada/SDMyers/ SDMyersAwardDamages.pdf. 5 Beginning with ADF Group Inc. v. United States, ICSID Case No. ARB(AF)/ 00/1, Final Award (Jan. 9, 2003). 6 To put this in perspective, see Professor Bryan Schwartz’s Separate Opinion on Costs in the S.D. Myers case (supra note 4), available at http://www.naftaclaims.com. 7 Mondev, supra note 1, at ¶ 159. Similarly, the Loewen tribunal concludes that there had been a miscarriage of justice by the U.S. courts vis-à-vis the Canadian investor, but decided nevertheless not to “use the [NAFTA] weapons at hand to put it right.” It does so even after posing the rhetorical question: “What clearer case than the present could there be for the ideals of NAFTA to be given some teeth?” See Loewen Group, Inc. & Raymond L. Loewen v. United States, ICSID Case No. ARB(AF)/98/3, 42 I.L.M. 811, 850 (2003). 8 W.M. Reisman & G. Aguilar Alvarez, How Well Are Investment Awards Reasoned?, supra Chapter 1 at 30.

The Final Award in Mondev International v. United States of America • 35

decided the expropriation claim in a substantially different way. The key question relates less to the quality of the reviewing court than to Mondev’s entitlement to compensation in the event that the reviewing court deprived Mondev of profits realized upon its legitimate expectations based on NAFTA. The sufficient treatment of the Article 1102 claim is discussed in Section V. Since NAFTA awards are among the real engines of development in foreign investment law and since “it is not unreasonable to assume that the legal issues in [this] context may, for the time being, be dominated by the definition of expropriation,”9 there is no doubt that only a plain and explicit reasoning on all Mondev issues could have given real breadth to this case and, partant, to the field of international investment law. II.

FACTUAL BACKGROUND

In December 1978, in order to rehabilitate one of its dilapidated areas, the city of Boston (City) and the Boston Redevelopment Authority (BRA), which has planning regulatory responsibilities for the urban renewal project, entered into a commercial real estate development contract with Lafayette Place Associates (LPA). LPA was formed by a company incorporated under the laws of Canada, Mondev International, Ltd. (Mondev) and its joint-venture partner, Sefrius Corporation, to develop, build, own, and manage suitable projects. The 1978 Tripartite Agreement between the parties (Agreement) was governed by the laws of the Commonwealth of Massachusetts. While granting LPA exclusive development rights, the Agreement provides for the development of the area in two successive phases and grants LPA a conditional option to purchase a Phase II property upon notice of the City’s intention to demolish a parking structure.10 LPA completed the first phase by 1985 and exercised its option on time on July 2, 1986, but without success—allegedly because of bad faith on the part of the local authorities.11 Although the parties amended the Agreement to extend the closing deadline,12 various delays and difficulties 9 R. Dolzer, Indirect Expropriations: New Developments?, 11 N.Y.U. ENVTL. L.J., at 66 (2002). 10 The claimants asserted that in early 1979 it was understood by both parties that the contingencies would be eliminated, but the respondent rejected this assertion. An appraisal mechanism was to provide the information necessary to calculate the purchase price. An appraisal mechanism was allegedly allowed to fill in the details of the purchase and sale contract as to the contours of the parcel. Compare the opinions of both parties, Hearings, supra note 1, at 42 and 476, Days 1 and 3, respectively. 11

Notice of Arbitration, supra note 1, at ¶ 45.

12

Mondev would have already been coerced at this stage. See Hearings, supra note


in completing Phase II, which are at the heart of Mondev’s expropriation claim, prevented them from finalizing the plans for this second phase within the given time frame.13 Further, on January 1, 1989, LPA negotiated a sale and a subsequent lease of the deal14 to the Canadian corporation Campeau (Campeau). Campeau began to negotiate with the BRA and the City to pursue its own development plan but filed soon after for bankruptcy. The interest and rights under the Agreement then reverted to LPA, whose investment went sour. Following domestic law, LPA filed a suit against the City and the BRA (in March 1992), notably for breach by the City and BRA of their contractual obligations under the Agreement and for the tortious interference of BRA with LPA’s contractual relations with Campeau.15 In 1995, a trial judge affirmed a jury’s verdict for breach of the Tripartite Agreement against the City but set aside the jury’s verdict against BRA on the ground that, although BRA’s conduct constituted a tortious interference,16 pursuant to the Massachusetts Tort Claims Act (MTCA) a public employer is immune from lawsuits that arise from an intentional tort claim, including interference with contractual relations.17 Both the City and LPA next sought direct review by the Massachusetts Supreme Judicial Court (SJC) which granted the request and delivered a long and wellarticulated decision at a time when NAFTA had already entered into force. Two of the SJC holdings are crucial to the NAFTA proceedings: 1, Day 1, at 865–66. Compare with the coercion referred to in CME Czech Republic B.V. v. Czech Republic, UNCITRAL ad hoc case, Final Award, ¶ 114 (Mar. 14, 2003). 13 On the alleged bad faith by the City and the BRA, see Hearings, supra note 1, Day 1, at 39–40, 58 et seq., 388 et seq. Compare with respondent’s view. Id. at 472 et seq. 14 It is unclear whether the drop end results from a coercion (Mondev in Hearings, supra note 1, Day 1, at 20) or not (Mondev’s complaint). On this “strange lease” (Judge Stephen dixit), see, for example, Hearings, id., Day 3, at 483–84, and compare claimant’s and respondent’s arguments, in Hearings, id., at Day 1, 87 and Day 3, 875 et seq. 15 On this domestic litigation, compare both positions of Mondev (Notice of Arbitration, supra note 1, at ¶ 144, and Hearings, supra note 1, Day 1, at 115 et seq.) and the United States, Hearings, Day 3, at 486 et seq. 16 Compare Mondev’s view with the U.S. focus on the 56-day delay when the LPA needed the BRA’s approval for the sale to Campeau. See Hearings, id., at 20 and 957. 17 See Tort Claims Act (Massachusetts) Pub. L. No. 258, at § 10(c). The BRA has not raised its immunity defense until after the case has essentially been tried, but the judge has decided that it is still available under Massachusetts law.


First, the SJC held that although a binding contract was formed between the two parties,18 LPA had not “manifest[ed] the willingness and ability to perform”19 and therefore not fulfilled its contractual obligation. The SJC absolved BRA of all liability on this single ground, whose rationale Mondev would later contest. Second, the SJC exclusively considered whether BRA met the definition of a “public employer” within the MTCA. Its affirmative answer made this court uphold the trial court’s rulings that BRA should benefit from immunity.20 As a follow up, even Mondev acknowledged that no domestic recourse was available to LPA even remotely likely to alter the results of the SJC decision.21 Both its petition for rehearing before the SJC on both claims and its writ of certiorari to the U.S. Supreme Court in respect of its contract claim were denied.22 Thus, in 1999, the City and the BRA were both relieved of all liability and LPA had exhausted all domestic remedies. In 1999, Mondev filed a claim under NAFTA Chapter 11.23 Focusing on the decision of the SJC,24 Mondev framed its claim as a U.S. violation 18 Lafayette Place Assocs. v. Boston Redev. Auth. & City of Boston, 694 N.E.2d 820, 825–831 (Mass. 1998) [hereinafter Lafayette Place Assocs.]. 19 Note that this reversed the jury’s finding of fact (Hearings, supra note 1, Day 5, at 923 et seq.) on the non-application of the appellate review’s standards by the SJC and the “square corners rule” (pronouncing that private parties need to be assiduous in dealing with government entities). See Hearings, id., Day 5, at 933. 20 The court additionally holds that the trial judge properly applied the MTCA’s provision and that the cause of action for unfair or deceptive trade practices does not apply to BRA since this later is motivated by legislative mandate. Lafayette Place Assocs., supra note 18, at 835–36. On the SJC decision, compare Notice of Arbitration, supra note 1, ¶ 97 et seq. and U.S. C-Mem., supra note 1, at ¶ 17. 21

See Notice of Arbitration, supra note 1, ¶ 114 et seq.

In order to make a case for the U.S. Supreme Court to grant the certiorari petition, LPA had styled the SJC decision as a sharp alteration of settled common law regarding the due process and taking clauses of the U.S. Constitution. The City of Boston argues to the contrary that LPA has never raised any questions of federal law before the SJC until it asserts a due process clause claim in its petition for rehearing and further that the SJC’s decision has not constituted a sharp change in the common law of Massachusetts. See U.S. C-Mem., supra note 1, at ¶ 18. 22

23 On the choice between an Article 1116 and an Article 1117 claim, see Professor Crawford’s mild lecture in Hearings, supra note 1, Day 3, at 608–609. 24 Although Mondev’s memorials are not accessible, the Hearings (id., Day 3, at 548) confirm the United States (U.S. C-Mem., supra note 1, ¶ 1) when they argue that Mondev “took a different track [. . .] in its Notice of Arbitration, focus[ing] the bulk of its Memorial not on what happened in court but on what took place in Boston Conference rooms and public offices in the late 1970s and 1980s.”


of Articles 1102 (National Treatment), 1105 (Minimum Standard of Treatment), and 1110 (Expropriation and Compensation).25 The NAFTA tribunal found that it does not have jurisdiction over the claims on national treatment and expropriation, and devoted the bulk of its opinion to the remaining claim for denial of justice. It held that the City and the BRA cannot be liable for conduct that occurred before January 1, 1994, and that only the dismissal of LPA’s claims by the JSC (May 20, 1998) and the denial of certiorari by the U.S. Supreme Court (March 1, 1999) would eventually constitute a breach of NAFTA. Considering that it is not itself a domestic appellate body and that the SJC merely applied the law and did not commit any “arbitrary or discriminatory or unjust” act, the tribunal denied Mondev’s denial of justice claim with no further inquiry on any sort of treaty-based legitimate expectations of Mondev. In other words, it follows the United States when they asserted that the central claim of the case is “whether the decision of the highest court of Massachusetts violated the minimum requirement of customary international law and whether that decision constituted a denial of justice, as that term is understood in international law.”26 By focusing on the single issue of the quality of the reviewing court, the tribunal entirely missed the issue of compensation and expropriation. The two most unreasoned parts of the award (Articles 1105 and 1110) will be dealt with in Sections III and IV, below. The chapter concludes with a short examination of the tribunal’s well-reasoned dismissal of the national treatment claim in Section V. III. THE UNDERREASONED DENIAL OF JUSTICE CLAIM The tribunal approached the merits of the Article 1105(1) claim on the ground that the alleged violations occurred after NAFTA entered into force. When Mondev alleged that the acts by the City and the BRA, through their treatment of LPA in the 1980s, violates Article 1105(1) requirements,27 the question is not whether the City’s and BRA’s treat25 The respondent objected to the competence of the arbitral tribunal on the grounds that Mondev did not own the rights at issue, that most of its claims were time barred, and that it lacked standing under Article 1116. A last objection, concerning a lack of a final judicial act, was later withdrawn by the respondent. 26

Hearings, supra note 1, Day 3, at 447.

See Notice of Arbitration, supra note 1, ¶¶ 141–143 and 165–168. See also S. Schwebel, On Whether the Breach by a State of a Contract with an Alien Is a Breach of International Law, in INTERNATIONAL LAW AT THE TOME OF ITS CODIFICATION: ESSAYS IN HONOR OF ROBERTO AGO at 410 (1987) and U.S. C-Mem., supra note 1, at ¶ 35. See, as well, the convincing U.S. arguments on Mondev’s faulty understanding of Article 27


ment of Mondev are wrongful in domestic law—in general mere breaches of contract are not per se violations of international law—but whether this act is internationally wrongful. Only the absence of redress in domestic law, which happened between 1994 and 1999,28 constitutes non-compliance with the requirements of international law.29 The tribunal ultimately dismissed each of the grounds supporting this Article 1105(1) claim, finding that the SJC decision is not so arbitrary as to rise to the level of denial of justice under customary international law. This arbitration, the second claim brought against the U.S. judicial system under Article 1105(1) for denial of justice,30 was the first to base part of its claim upon the actions of a supreme court. Although the tribunal unsurprisingly claims not to be a court of appeal,31 it conducted a very extensive review of the lower court’s adjudication. Although Mondev

1105(1)’s guarantee of “full protection and security.” Id. at ¶ 37. Note that Mondev made in its Notice of Arbitration the startling point that the SJC itself committed an expropriation by reversing the lower court judgment in LPA’s favor (see Notice of Arbitration, supra note 1, at ¶ 74), but Mondev appears to have abandoned this argument later. 28 A serious question (which will not be reviewed here) relates to how NAFTA is concerned, since it gives the investor the choice of an international remedy straightaway without exhausting local remedies.

See ALWYN FREEMAN, INTERNATIONAL RESPONSIBILITY OF STATES FOR DENIAL OF JUSat 539–570 (1938); LOUIS SOHN & RICHARD BAXTER, HARVARD DRAFT CONVENTION ON THE INTERNATIONAL RESPONSIBILITY OF STATES FOR INJURY TO ALIENS, in particular, arts. 6–8 (1961). According to Chattin v. United Mexican States (4 R.I.A.A., at 282 (Mex.U.S. Gen. Cl. Comm’n 1927) [hereinafter Chattin]), in order to establish that the decision was manifestly unjust, Mondev would have had to show that the conduct of the Massachusetts courts amounted “to an outrage, to bad faith, to willful neglect of duty, or to an insufficiency of governmental action recognizable by every unbiased man.” 29

TICE,

30 NAFTA is by no means the first international agreement to provide a forum for arbitral decisions on denial of justice. One mixed claims commission in the later part of the 19th century addressed claims arising from allegedly deficient judicial processes. See MOORE, HISTORY AND DIGEST OF THE INTERNATIONAL ARBITRATION TO WHICH THE UNITED STATES HAS BEEN A PARTY, at 3073–3234 (1898), quoted in A. Bjorklund, Reconciling State Sovereignty and Investor Protection in Denial of Justice Claims, 45 VA. J. INT’L L. 809, 888 (2005). On the distinction between acts affecting the administration of justice and other wrongful acts implicating the responsibility of a state in international law, see, e.g., United States (L.F. Neer) v. United Mexican States, 4 R.I.A.A 60, 61–62 (Mex-U.S. Gen. Cl. Comm’n 1926) and Chattin, supra note 29, at 286–88, this latter limiting the standard to acts of the judiciary. 31 On the nature of international review of national courts, see R. Ahdieh, Between Dialogue and Decree: International Review of National Courts, 79 N.Y.U. L. REV. 2029 (2004).


failed to establish a violation of Article 1105(1),32 the tribunal’s reasoning proves insufficient for several reasons.33 Prior to appraising the quality of the reviewing court with respect to the applicable rules of international law, the tribunal first had to explain its own conception of the rule and subsequently address the following issues. 1. The tribunal was certainly explicit enough about the restriction of the scope of applicable law, which had a tremendously adverse effect on Mondev’s case. It concluded with no surprise that the Free Trade Commission’s (FTC’s) July 31, 2001, statement,34 which changed the minimum standard of treatment in Article 1105 from that of “international law” to “customary international law,”35 is not an amendment but an interpretation—therefore binding.36 The tribunal had then to define the scope 32 See Hearings, supra note 1, Day 3, at 683. Mondev does not dispute that denial of justice rules are relevant standards under customary international law minimum standard of treatment but, as emphasized by the U.S. defense, Mondev attempted to establish that Article 1105 incorporates a subjective standard under which arbitrators could hold sovereign states to have violated international legal obligations (the so called “smell test”), but clearly failed to demonstrate that such a subjective test is part of the customary international law standard of treatment set forth in Article 1105. Mondev quoted Chattin (supra note 29), which sets a very high threshold under customary international law for a denial of justice, and Amco-Asia Corp. et al. v. Republic of Indonesia (ICSID Case No. ARB/81/8, 24 I.L.M., at 365 (1985)), which applies, inter alia, the stringent Chattin test, and Elettronica Sicula S.p.A. (United States v. Italy), 1989 I.C.J. 15, 28 I.L.M. 1109 (1989) [hereinafter ELSI], to support a procedural denial of justice claim. While these cases do not support Mondev’s subjective test, most of the other cases quoted by Mondev do not even address customary international law standards at all: ELSI interprets the term “arbitrary” in the treaty at issue and does not set up a customary law obligation of non-arbitrariness. The Maffezzini tribunal (Emilio Agustin Maffezini v. Kingdom of Spain, ICSID Case No. ARB/97/7, Decision on Objections to Jurisdiction (Jan. 25, 2000), 16(1) ICSID REV.—FOREIGN INV. L.J.) (in the matter of compliance with rather unusual provisions of a Spain-Argentine BIT) does not address customary international law standards. The S.D. Myers (supra note 4) and Metalclad (ICSID Case No. ARB(AF)/97/1 (Aug. 30, 2000), 40 I.L.M. 36 (2001)) tribunals do not cite any customary international law authority in holding that violations of other NAFTA provisions amounted to a violation of Article 1105(1); thus neither case was decided on the basis of a violation of a customary international law rule. 33 See Azinian v. United Mexican States, 39 I.L.M. 537 (1999). Notwithstanding the lack of an overt claim of denial of justice, the Azinian court considered, at some length, the historical and modern standards of denial of justice and the appropriate application of those standards under Chapter 11. 34

Notice of Arbitration, supra note 1, at ¶ 31.

35

Id. at ¶¶ 101–125.

36 See NAFTA art. 1131 and Mondev, supra note 1, ¶¶ 105, 119–120. See also Pope & Talbot Inc. v. Canada, supra note 3, at ¶ 66.


of “customary international law” under Article 1105(1) and this FTC interpretation. It cannot be criticized for considering that the relevant standard is customary international law “as it has evolved” since this understanding endorses the view of the United States37 and does not contradict Mondev’s position to allow the tribunal to decide as it sees fit without calling for any particular course of action.38 However, the tribunal reached its conclusion on the customary nature of the Article 1105(1) requirement in a rather obscure way: while it may be difficult to define opinio juris in 1994,39 it could not conceive a definition of the content of international law without referring to opinio juris and solely to elements repeated in time40 without falling into a circularity problem.41 Further, when the tribunal had to define the applicable standard for denial of justice under NAFTA,42 it did not make explicit its understanding of the content of the international law minimum standard of treatment.43 The minimum standard of treatment is an umbrella concept incorporating sets of rules crystallized into customary international law in very specific contexts. As Professor Brownlie puts it, under the international minimum standard of treatment, “there is no single standard, but different standards relating to different situations.”44 The tribunal only provided that the standard is not “frozen in amber” but is an evolutionary standard.45 Considering the obvious evolution of these standards during the last 15 years (for instance the notion of transparency between the Mondev, Thunderbird, Techmed, and Maffezzini cases), the tribunal

37

Confirmed in Hearings, supra note 1, Day 3, at 660 et seq.

According to Mondev, “the standard adopted in article 1105 was that as it existed in 1994, (. . .) as it had developed to that time”; “We also agreed, like all customary international law, that the international law minimum standard has evolved and can evolve.” See Mondev, supra note 1, at ¶ 111. 38

39

Notice of Arbitration, supra note 1, at ¶¶ 124–125.

40

Id. at ¶ 113.

41

On this complex issue, see the demonstration provided id. at ¶¶ 100–125.

42

Id. at ¶ 94.

43


44

IAN BROWNLIE, PRINCIPLES OF PUBLIC INTERNATIONAL LAW at 531 (6th ed. 2003).

Notice of Arbitration, supra note 1 at ¶¶ 123–125. Canada in an amicus curiae submission has suggested that the tribunal use the meaning of those provisions as established by a case before the Mexican Claims Commissions in 1926 (Mondev, supra note 1, at ¶ 114) and the Neer case (supra note 32, cited in Mondev, supra note 1, at ¶ 556). 45


should have explicitly indicated whether it was applying the minimum standard of treatment requirement as of 1994,46 1998, 1999, or 2002. 2. Without explaining its own conception of the rule, it remains of secondary importance for the tribunal to address other questions such as the eventual margin of appreciation in the application of the Article 1105(1) standard. The first flaw in the reasoning led to further inadequacies when the tribunal took a deeper look into the SJC’s handling of the appeal of Mondev’s civil suit against the City and BRA. On the surface, the tribunal’s reasoning is sufficient. Mondev claimed that the tribunal departed significantly from its established jurisprudence by dismissing LPA’s contract claim against the City after explicitly finding that the parties had entered into a binding contract. In sharp contrast, the respondent argued that the SJC decision was an “extensively reasoned”47 decision, written with “an abundance of thoroughness.”48 Such a discrepancy in the claims led to two different questions, which the tribunal thoroughly addressed. The first question arose from the dismissal of the contractual claim against the City. To begin with, the SJC essentially reasoned that the law of Massachusetts requires that the non-performing party is put in breach,49 that is, that any buyer must demonstrate that he is ready, able, and willing to perform by setting a time and place for passing papers or making some other concrete offer of performance.50 According to the Mondev tribunal, this judgment involved a “significant and serious departure” from case law, which was exacerbated by its failure to consider whether this allegedly new approach should apply retroactively.51 There 46


47

See Hearings, id., Day 3, at 449.

48

See Hearings, id., Day 3, at 524.

On the specific question whether LPA can maintain a claim against the City for breach of the bilateral contract for the purchase and sale of the parcel, see the SJC’s decision, at 427 Mass. 509, Part 2B (1998). 49

50

See Hearings, supra note 1, Day 3, at 457.

51 On this claim, see U.S. C-Mem, supra note 1, at 46. On the recognition by international law that states have the authority to prescribe laws in different ways, including by the common law method of adjudication, whereby, in uncodified areas of the law, the applicable rules evolve with and as a result of decisions in successive cases, see The Sunday Time case, Eur. Ct. H.R. 6538/74, at 245, ¶ 47 (1979).


is no reason to criticize the arbitrators’ reasoning that it is doubtful that the Massachusetts court applied the law in a novel fashion52 and its conclusion that, even if it had done so, it was only a judicial error that would not constitute so grave an error as to render the decision manifestly unjust—“its decision would have fallen within the limits of common law adjudication.”53 Then, the tribunal was undoubtedly right in asserting that the SJC correctly decided that the LPA was not excused from its obligation to put the city in default. The SJC decided that there had been no “outright refusal” by the City to perform and that the evidence did not reveal a notice sufficiently specific to satisfy the law as established in a precedent from 1954.54 Because this case was “a complex and heavily regulated transaction [. . .], where public entities and public and elected officials with changing policies and constituencies are involved and the transaction spans many years,” the SJC explicitly referred to Justice Holmes’s dictum according to which “men must turn square corners when they deal with the Government.”55 Since neither the LPA nor Campeau had turned such corners, the SJC decided that the LPA was not excused from its obligation to put the city in default. Although the arbitrators did not mask that the Holmesian pronouncement “stands in some tension” with the proposition that “governments are subject to the same rules of contractual liability as are private parties,” they nonetheless concluded that the SJC’s determination of “whether an agreement in principle to transfer real property is binding and whether all the conditions for the performance of such an agreement have been met” is nothing more than common law adjudication.56

52 The Martini (Italy v. Venezuela (May 3, 1930)), Rihani (U.S.-Mex., Decision 27C (1948)), and Jalapa (Decision 13-E (1948)) cases quoted by Mondev in its Reply at ¶¶ 104–106 (according to Rejoinder on Competence and Liability of Respondent United States of America (Oct. 1, 2001)), only reflect procedural or substantive denials of justice that may be found only in extreme circumstances such as inexcusable delay or bad faith. See Judge Tanaka dicta in the ICJ Barcelona Traction (1970 I.C.J. 3)case: “it is an extremely serious matter to make a charge of denial of justice vis-à-vis a state. It involves [. . .] a moral condemnation of [its] judiciary.” 53

Mondev, supra note 1 at ¶ 133.

See Leigh v. Rule, 331 Mass. 664, 668 (1964). See Hearings, supra note 1, Day 3, at 519 et seq. and 527 et seq. on the failure to perform its obligation/repudiation. 54

55

Mondev, supra note 1, at ¶ 130.

See, in particular, LeBlanc v. Molloy, 335 Mass. 636, 638 (1957) and Mayer v. Boston Metropolitan Airport, 355 Mass. 344, 244 N.E.2d 568 (Mass. 1969). 56


The second question—whether the SJC should have remanded the case to a jury to decide whether the LPA was ready and willing to perform its contractual obligations57—brings an additional element of complexity in making the tribunal address the question of whether an insufficient review by a domestic court may constitute a denial of justice. Obvious reasons explain why this tribunal, like others, proved ultimately reticent to find liability for the decisions of national courts.58 While not persuaded that the SJC’s conduct was improper, it reasoned that even in that case the application of local procedural rules such as questions of appellate fact-finding could constitute a denial of justice only in the most extreme circumstances:59 a particular result may be surprising, but the shock or surprise occasioned to an impartial tribunal should not lead to justified concerns as to the judicial property of the outcome.60 Although the Mondev tribunal articulates a stringent test for denial of justice, this test is not impossible to meet.61 If the tribunal dismissed the investor’s claim in its entirety, it certainly provided some very useful dicta about the minimum standard provision which had vexed the Pope proceedings. Yet the historic dividing line between what constitutes a mere error and what constitutes a manifest injustice often remains unclear. Taking this into 57 Notice of Arbitration, supra note 1, at ¶ 141. Mondev offers yet no support to establish that customary international law confers on it a right to a jury trial to determine a question of civil damages. According to Freeman, “States are permitted to order their judicial systems as they choose: in fulfilling [the requirement to provide an adequate judicial protection for the rights of aliens] each States enjoys a plenary margin of liberty. The organization of its courts, the procedure to be followed, the kinds of remedies instituted, the laws themselves, are left to the State’s own discretion,” FREEMAN, note 29 supra, at 78–79. 58 As noted by the U.S.-Mexico Claims Commission in the 1868 Garrison case, “it is a matter of the greatest political and international delicacy for one country to disacknowledge the judicial decision of a court of another country, which nevertheless the law of nations universally allows in extreme cases,” 3 MOORE’S INTERNATIONAL ARBITRATION 3129 (1995). 59

See Hearings, supra note 1, Day 3, at 719 and U.S. C-Mem, id., at 56–58.

See the ICJ reasoning in the ELSI case, supra note 32 (“a wilful disregard of due process of law . . . which shocks, or at least surprises, a sense of judicial propriety”). Note that there is indeed no customary international law establishing that a jury must make any determinations. Such civil jury trials are not the norm in many jurisdictions. In TP and KM v. the United Kingdom (Grand Chamber, No. 28945/95, ECHR 2001–V (2001)), the European Court of Human Rights gives further support for the conclusion that determinations such as that made by the SJC in LPA’s case cannot give rise to an international claim 60

61

Compare with Loewen, supra note 7.


account, “denial of justice” remains a flexible concept. The “shock the conscience” or “manifest injustice” standard is often unsatisfactory for the same reason that review based on equitable rather than legal principles has been viewed with suspicion. Because the number of investor-state cases involving denial of justice claims is likely to increase, the formulation of a coherent concept is essential to and more far-reaching than the Mondev case. 3. Beyond these two first questions, more or less well reasoned by the tribunal, the reasoning should be criticized for its appraisal of a third question related to immunity. The tribunal decided that although statutory immunity may not shield a state party from suit for a substantive breach of NAFTA, the mere fact that a state party extended statutory immunity to a regulatory authority does not constitute a breach of Article 1105(1). This conclusion is not supported by reasons. Under this rationale the United States may perfectly avoid international responsibility by local enactments purporting to confer immunity on governmental bodies. The SJC did not even attempt to de-emphasize the direct effect of the immunity, noting for instance that “it is perfectly possible for a governmental entity to engage in dishonest or unscrupulous behaviour as it pursues its legislatively mandated end.”62 The tribunal concedes that the SJC has not only correctly resolved the issue of Massachusetts law raised by these two claims but also that both the trials court’s and the SJC’s conclusion that the BRA is a public entity immune from liability for intentional tort under the MTCA is correct. Yet, Mondev challenged not only the SJC’s rationale but the law itself in asserting that such immunity effectively deprives the LPA of access to the courts and thereby denies justice. The Article 1105 claim is related to the fact that the MTCA violates customary international law. The tribunal carefully reviewed this issue63 in looking at analogous cases to determine the effect of BRA’s immunity and also in examining 62 Lafayette Place Assocs., supra note 18, at 836 (1998) (quoted in Notice of Arbitration, supra note 1, at ¶ 113. Eventually, the proscription of “unfair or deceptive practices in the conduct of any trade or commerce” was established by a Massachusetts statute, and prior case law under that statute held that it did not apply in terms to parties motivated by legislative “mandate, not business or personal reasons.” Poznik v. Massachusetts Med. Prof’l Ins. Assoc., 417 Mass. 48, 52, 629 N.E.2d 1 (1994), which led to the provocative statement above. 63

Public authorities engaged in commerce being in general not entitled to statu-


the rationale behind exempting public authority from liability for intentional torts.64 It grounded its minimal understanding of the issue by stating that “circumstances can be envisaged where a conferral of a general immunity from suit for conduct of a public authority affecting a NAFTA investment could amount to a breach of Article 1105(1) of NAFTA.” While stating that statutory immunity from tort action could also violate Article 1105(1), the tribunal suggests that there may be sound reasons why a regulatory authority should be protected against claims brought by disgruntled private parties with respect to the specific tort of interference with contractual relations. Stating that having to defend such actions might be deemed unacceptably distractive, the tribunal concludes that although statutory immunity may not shield a state party from suit for a substantive breach of NAFTA, the mere fact that the United States has extended statutory immunity to one of its regulatory authorities does not constitute a breach of Article 1105(1). It concludes that this “is a matter for the competent organs of the States to decide.” The limited immunity provided by Massachusetts law to BRA is therefore not a denial of justice.65 But in order to conclude that customary international law does not require statutory authority to be generally liable for their torts and that, therefore, the immunity of the BRA does not infringe Article 1105(1), the tribunal had to describe the applicable customary international law.66 The tribunal does not delve into this question. Although the hearing transcripts leave the impression that the previous Mondev cases poorly tory immunity for intentional torts, Mondev claims that the extension of immunity by the municipal government in the limited circumstances of the present case would violate the customary international law minimum standard of treatment of aliens. The U.S. counterargues that the international authorities cited by Mondev provides no support for its claim; that state practice on domestic sovereign immunity shows no international consensus on when a state must subject itself to suit in its court; and ultimately that any limited immunity granted to by the United States to BRA do not contravene Article 1105(1). (Mondev, supra note 1, at ¶ 140). See Hearings, Day 3, at 460. 64

Notice of Arbitration, supra note 1, at ¶ 145–150.

65

Id. at ¶ 154.

66 See id., at ¶¶ 127 and 142, claiming, in short, that both the trial court and the SJC used the MTCA and Massachusetts General Laws chapter 93A to shield BRA with sovereign immunity that they ruled could not be waived, thereby categorically excluding the application of due process in this or any other similar case and allowing the City, acting through BRA, to expropriate anyone’s investment in this and future case without the expropriated person having any judicial recourse even for intentional misconduct. See Hearings, Day 1, at 141–52.


addressed this issue as well,67 it remained a duty for the tribunal to state whether current practice would support Mondev’s contention that municipal sovereign immunity is internationally wrongful68 or whether customary international law leaves the state subject to a claim for denial of justice at an international level.69 Although it was certainly risky for Mondev to refer to emerging international norms governing the jurisdiction of courts over foreign states to support its view that municipal sovereign immunity in tort is internationally illegal,70 the question had to be dealt with as such by the tribunal. At least two ways out were possible. First, the tribunal could have addressed the fact that while customary international law requires states to provide certain protections to aliens, no international authority requires states to provide a cause of action for the tort of intentional interference with contractual relations. Had it been proven that customary international law requires redress for such a wrong, it certainly leaves the form of the remedies to the state. Following this first option, the tribunal could have improved its reasoning while also dismissing Mondev’s claim.71 It was enough for the tribunal to rule on the fact, noted by the United States in its Counter Memorial, that U.S. law had not left the LPA without remedies: although immune from suit in intentional tort under Massachusetts laws, the LPA was not immune from suit under the U.S. Federal Civil Rights Act. Had it so desired, the LPA could have asserted a claim against the BRA under that

67 From an unpublished decision by an English court of first instance (Matthews v. Ministry of Defense, [2003] 1 A.C. 1163) to a misreading of the 1929 Harvard Research Project’s Draft Convention, which on the mentioned point related to access to justice to resolve private wrongs and not issues of government liability. 68 On the necessity for Mondev, not the United States, to bear the burden to establish that state practice support the proposition that municipal sovereign immunity is internationally wrongful, see U.S. Rej., supra note 1, at 16–17. 69 For a “rough barometer,” see Ashingdane v. United Kingdom (7 Eur. H.R. Rep. 528 (1985)), even if this case relates to an obligation pursuant to an international convention and not to customary international law, and even if the European Court of Human Rights has developed constructs in interpreting its Article 6(1) provision that are not necessarily based on customary international law. See infra note 75. 70

Hearings, Day 4, at 774 et seq.

71 Under general international law, before pursuing an international claim, the claimant is bound to exhaust any local remedy that is adequate and effective so long as the remedy is not obviously futile, that is, so long as it is not clear in advance that the municipal courts of the state concerned will not provide redress for the injured investor. See the Interhandel case (Switzerland v. U.S.), I.C.J., 27 I.L.R. 475 (Mar. 21, 1959).


act;72 the Federal Civil Rights Act provides relief for harm caused by extreme conduct by government officials and those acting in concert with them.73 Second, in the event that no remedy was available, the tribunal had no other option than to address in depth the question of whether immunity could be granted.74 Lauterpacht, for instance, merely acknowledges that the doctrine of absolute sovereign immunity has already been jettisoned by the majority of states, but so has the Massachusetts legislature jettisoned absolute state immunity in Massachusetts courts.75 Although the MTCA supports this proposition that the doctrine of full immunity has been abandoned, the tribunal has failed to address clearly its conception on relative immunity. As noted by Paulsson, the tribunal should have indicated whether a line of demarcation exists between cases of “acceptable” immunity and other immunity.76 International legal devel72 See U.S. C-Mem, supra note 1, at 53–54 and nn.71–72. See, e.g., Raso v. lago, 958 F. Supp. 686 (D. Mass. 1987), aff’d, 135 F.3d 11 (1st Cir. 1998) (adjudicating claim against BRA under federal civil rights act for, among other things, depriving plaintiffs, without due process, of an alleged property right to an absolute housing preference); Lewis v. Boston Redev. Auth., 1996 WL 208473 (D. Mass. 1996); Hurney v. Boston Redev. Auth., 1996 WL 658 (D. Mass. 1978) (recognizing that local governmental entities are subject to suit under federal civil rights act). 73 Pursuant to Moran v. Bench, 353 F. 2d 193, 194 (1st Cir. 1965), “[I]f defendants employed their official powers for the purpose of injuring the plaintiff, rather than to serve the proper ends of their governmental duties, plaintiff might well have a claim under the civil rights statutes.” 74 The practice of sovereignty in tort law has a longstanding tradition. While it has been relaxed to some extent in most of the world’s jurisdiction, the waivers are limited and subject to various exceptions, a domain that might have been necessary for the tribunal to reexamine. See G. Eörsi, Private and Governmental Liability for the Torts of Employees and Organs, in 11 INTERNATIONAL ENCYCLOPAEDIA OF COMPARATIVE LAW, at 83–85, 120–31 (1976). 75 E. Lauterpacht, The Problem of Jurisdiction Immunities of Foreign States, 98 BRIT. Y.B. INT’L L. 220 (1951). 76 See JAN PAULSSON, DENIAL OF JUSTICE IN INTERNATIONAL LAW, at 142 (2005). In this respect, the respondent’s reference, in its rejoinder to the decision of the European Court of Human Rights and to the case law under Article 6(1) of the European Convention on Human Rights is misplaced (cf. supra note 69). These cases illustrate no more than the manner in which the balance might be struck between laws limiting access to courts and the citizen’s right to have a fair hearing within the meaning of Article 6(1) of the Convention. It leaves open, of course, the question of whether a national’s right to a fair hearing within the meaning of Article 6(1) is the same substantively as a foreign investor’s right to treatment that is fair and equitable


opments have certainly transformed the ability of states to circumscribe legal recourse by “general legislation,” and Paulsson is correct to state that a legislative attempt to avoid legal responsibility by creating immunity for public bodies will more likely be characterized as a denial of justice if the legislation is special rather than general, if it emerges after a conflict has arisen and seems calculated to have an effect on it, or if it is discriminatory in its effect. According to the United States, the contemporary practice for government liability,77 far from supporting Mondev’s assertion on a customary international law bar to sovereign immunity against the conduct at issue, governs the liability of a state in its municipal courts by different rules than those applied to private persons78 and expresses no general consensus as to what acts or activities may or may not be immune in municipal courts. When explaining the degree of deference towards different national solutions with respect to the liability of public bodies, the tribunal may have attempted to address de Visscher’s analysis in Le Déni de justice en droit international (1935) taking current practice into consideration.79 Although it is difficult to square the issue in Mondev with the conclusion of de Visscher, the tribunal should have attempted this reconciliation. According to de Visscher, since international law requires at least one court to examine the merits of the claim, the denial of justice would only qualify the adjudications declining jurisdiction when no other avenue for complaint is available.80 When immuand in accordance with full protection and security as set forth in NAFTA Article 1105. See McElhinner v. Ireland, 34 Eur. H.R. Rep. 322 (2001). 77 On the frequency of this sort of exception for the tort of intentional interference with contractual relations in municipal regimes, see U.S. C-Mem., supra note 1, at 53, n.70. In short, in common law countries, the tendency has been to base government liability on the law governing the private wrongs with a number of exceptions. In France, government liability is regarded as a matter of public law and is entrusted only to the administrative courts, which demonstrates that the fact that Massachusetts law provides different rules of liability for the BRA and for private persons is in no way dissident with state practice today. Moreover, as stated by the U.S. Rej., supra note 1, at n.30, the law of many legal systems throughout the world would (like the MTCA) restrict a state’s liability under the precise circumstances presented here. 78 In the Hearings, supra note 1, Day 4, at 744 et seq., the United States shows that although the past half century has seen a relaxation in the scope of municipal sovereign immunity, the prevailing state practice today continues to recognize sovereign immunity in tort for selected tortious acts and activities. 79 See Ch. de Visscher, Le Déni de justice en droit international, 54 R.C.A.D.I. 370, at 395–96 (1935). 80

On the “fork-in-the-road problem,” see P. Bekker, The Use of Non-Domestic Courts


nity is granted under local law such as in the Mondev claim, the tribunal may have followed de Visscher who considered that the absence of any jurisdiction competent to hear a complaint for damages caused by prejudicial administrative acts is internationally wrongful if the absence is due to “a measure of discrimination against [a] foreigner.”81 This is particularly important when the state has entered into a contractual relationship with an alien, in which case it must “place at his disposal internal legal means for ensuring its fulfilment.”82 IV. THE MISSED CREEPING EXPROPRIATION CLAIM The previous developments on immunity lead to the main failure in the Mondev tribunal’s reasoning, which consists of understanding the expropriation claim much too narrowly. Mondev should indeed be entitled to compensation regardless of the quality of the reviewing court if the SJC’s decision, by granting immunity, deprived the claimant of profits realized upon its legitimate expectations based on the NAFTA. As suggested in dicta in the Azinian v. Mexico case, the judicial decision would otherwise constitute an expropriation or a measure tantamount to an expropriation, as Mondev, unlike the investors in Azinian,83 had explicitly claimed.84 Instead of addressing this issue, the tribunal recommitted for Obtaining Domestic Relief: Jurisdictional Conflicts Between NAFTA Tribunals and U.S. Courts, 11 ILSA J. INT’L & COMP. L. 331 (2005) (“As long as NAFTA tribunals continue to adhere to a strict interpretation of the local remedies/judicial finality and diversity of nationality requirements along the lines of international law, there should be no jurisdictional conflicts. Investors will need to file a petition for certiorari in the U.S. Supreme Court lest they find themselves struck out under NAFTA Chapter 11 based on the non-fulfillment of the rule of judicial finality. Most importantly, however, if the domestic courts (arguably) got it wrong under international law and if the rule of judicial finality is satisfied, NAFTA remedies should be available and such suits will not be frivolous or abusive.”). 81

Id. at 395.

82

Id., at 396, n.1.

83 In Azinian (supra note 33), American shareholders of a Mexican corporation claimed that a Mexican city expropriated their investment by wrongfully terminating a waste management contract. The Mexican Federal Circuit Court upheld the city’s decision to cancel the contract, finding that the contract was invalid under Mexican law. Although the investors did not challenge the Mexican judicial decision as expropriatory, the tribunal noted that a government authority “cannot be faulted for acting in a manner validated by its courts unless the courts themselves are disavowed at the international level,” for instance, if there was evidence that the Mexican judicial decisions were arbitrary or malicious. 84

See Notice of Arbitration, supra note 1, at ¶ 124 (“it is also a settled principle of


itself at least procedurally to respecting the already pressed boundaries of the international takings cases.85 It considered only an expropriation of Mondev’s option in the Hayward Parcel and held that should it have occurred, it did so on the date the option lapsed (January 1, 1989)—and since Mondev did raise only a breach of contract claim and not any expropriation claims in the Massachusetts courts, the actions of the courts could not have advanced the expropriation. It determined further upon a similar rationale that any expropriation of the Hayward project as a whole was complete before 1994 as well (by the date of foreclosure in 1991), and consequently dismissed any expropriation claim as time barred. But the issue of “what resembled an expropriation”86 deserved that “a higher threshold for the reasons requirement [. . .], rather than a lower, or more modest one, would seem to command itself.”87 Had the tribunal ruled on the merits of this claim, its reasoning should have thoroughly discussed the following elements. Four factual precisions from the hearings are critical to properly address the indirect expropriation claim. First, there can be no doubt that the Agreement specified the terms of the option, including a detailed formula for determining the purchase price if the LPA exercised the option and a binding arbitration procedure to resolve potential dis-

international law that a State is responsible for the acts of its judiciary”). See IAN BROWNLIE, SYSTEM OF THE LAW OF NATIONS: STATE RESPONSIBILITY, Part I, at 144 (1983) (“the judiciary and the courts are organs of the state and they generate responsibility in the same way as other categories of officials”); FREEMAN, supra note 29, at 28 (“the fundamental principles which underlie [sic] international responsibility apply with equal force to the responsibility of the State for acts and omissions on the part of its judicial organs”); M. WHITEMAN, DAMAGES IN INTERNATIONAL LAW, at 7 (1937) (“the wrongful acts of the respondent state, which may be in the nature of positive acts or consist in the failure of the state to prevent the injury . . . may be committed through . . . the judicial authorities”). See also Fields of Texas, Inc. v. Islamic Republic of Iran, Award No. 258-43-1 (Oct. 8, 1986), 12 IRAN-U.S. CL. TRIB. REP. 308, 318 (1988) (a state may be held responsible for a court judgment or decision that works to wrongfully deprive an alien of its property rights); and Decision No. 196, France-Italian Conciliation Commission (Dec. 7, 1955), 13 R.I.A.A. 422, 438 (1964) (concerning the interpretation of Article 79, Section 6, lit.C, of the Treaty of Peace (France v. Italy)). 85 See bibliography, in Ch. Hansen, Mondev International Ltd v. United States: A Case Study of the Potential Risks of NAFTA’s Everexpanding Arbitration Provisions, N.C.J. INT’L L. & COM. REG. 351, 380–81 & n.109. 86 See T. Weiler, Current Legal Trends in the Americas: NAFTA Chapter 11 Jurisprudence Coming Along Nicely, 9 SW. J.L. & TRADE AM. 245, 247 (2002). 87

Reisman & Aguilar Alvarez, supra note 8, at 29.


putes between the parties regarding the sale of the property.88 Second, Mondev undertook his high risk project in 1978 only because the second phase was to proceed on very favorable terms.89 Third, concerning the various delays and difficulties in realizing the second phase,90 the City balked when executing the transfer because it was undoubtedly “engaged in a series of delaying and obstructing maneuvers”91 that were an attempt to avoid the now unfavorable contractual terms.92 In particular, although the LPA added a so called “drop dead” date93 and considered alternatives in order to protect the second phase of the project to save something of value,94 it had to lease its rights in the project to Campeau when faced with the impending expiration of the option. The hearings make the 88 On the three-person appraisal board, the SJC concluded (Lafayette Place Assoc., supra note 18, at 518–19) that “to borrow Justice Holmes’s metaphor, the machinery was built and had merely to be set in motion.” Accordingly, it concluded the Agreement did create an enforceable contract with respect to the Hayward Parcel. 89


90 See Hearings, id., Day 1, at 39–40, on the City and BRA’s bad faith. See Hearings, id., Day 2, at 388–393, and U.S. C-Mem., supra note 1, at 39–42 (describing that the City acted appropriately in fulfilling their mandate of promoting economic development on terms compatible with sometimes competing interests, such as reducing environmental impacts, ensuring the availability of affordable housing, minimizing negative effects on traffic circulation and preserving the city’s character). 91


92 At the time the City and LPA entered into the Agreement, the contractual terms seemed to benefit each party—the City would finally realize its goal of urban renewal of the unseemly Combat Zone, and LPA could potentially exercise its contingent option to purchase the Phase II property at a pre-determined rate if Phase I proved successful in enhancing the value of the area. Neither party, however, foresaw the Boston real estate boom of the 1980s and the dramatic increase in property values that ensued. There can be no doubt that the value of the Hayward Parcel increased as a result of Phase I and the affiliated City improvement to the area, and that LPA would be able to purchase the Hayward Parcel rights at a figure well below current market value when it closed on the transaction (“as of Jan. 1, 1989, the value of the Hayward Parcel was 19.1 million. [. . .] Under the formula set out under the Tripartite Agreement, LPA was entitled to purchase that parcel for 2.68 million, which was Mondev’s incentive in going forward with this procedure” Hearings, id., Day 1, at 48.). See, e.g., Hearings, id., Day 5, at 923–24. 93 It ruled that “the Developer shall lose its rights hereunder to proceed with an acquisition if a closing has not occurred by January 1, 1989, unless the City and/or the Authority shall fail to work in good faith with the Developer through the design review process to conclude a closing.” 94 See Hearings, id., Day 2, at 394, and Day 1, at 98 et seq. on the reason for the Agreement.


case that this procedure itself (a lease) was chosen to avoid any approval by the BRA.95 By the use of its governmental authority, the BRA deprived the LPA of its right to sell its interest in the project to Campeau. Fourth, Campeau also encountered objective resistance from City authorities. Although the mayor, Mr. Flynn, inherited the now unfavorable terms of the Agreement from his predecessor, the City was nonetheless contractually bound by those terms.96 The city undoubtedly knew that if the dropdead date expired, the appraisal price provision would no longer be binding, and the arrangement would change to its own advantage.97 For example, it is only after the January 1, 1989, deadline had expired and Campeau had consequently agreed to pay the current market price that the City approved Campeau’s plans. As previously noted, the SJC observed that “it is perfectly possible for a governmental entity to engage in dishonest or unscrupulous behaviour as it pursues its legislative authority mandated ends.”98 Consequently, Mondev’s investment in the LPA had been destroyed by mid-1991. The City’s and BRA’s conduct has resulted in, to adopt the language of the tribunal in the CME v. Czech Republic award, “the evisceration of the arrangements in reliance upon which the foreign investor was induced to invest.”99 This deprivation does not differ from a formal expropriation: Mondev was deprived of the economic benefit that it reasonably expected to enjoy under its contract because of the direct, seemingly foreseeable, and intended result of the course of conduct on which the City and the BRA had embarked. This deprivation of its investment occurred step by step—in what Mondev’s lawyer has coined as a “death by thousands cuts.”100 The failure to compensate made this deprivation a continuing act until March 1999, when all hope of compensation had gone, apart from those present in the NAFTA proceedings. The idea of past conduct giving rise to present liability or to the notion of a contin95

Id., Day 1, at 98.

In January 1984, just after the City notified LPA of its intent to discontinue use of the garage, Mr. Flynn replaced Mr. White as Mayor and served in that capacity until July 1992. Seen through the fresh eyes of a new administration, the Agreement seemed to heavily favor LPA. 96

97

Hearings, supra note 1, Day 3, at 518 et seq. and at 527 on the price paid.

98

Id., Day 2, at 395.

Id., Day 3, at 611. See also CME Czech Republic B.V. v. Czech Republic, supra note 12. 99

100



uing wrongful act is nothing unusual or novel, as claimed by the U.S. Rejoinder. Both are indeed recognized in international law. As stated by Dolzer in reference to the Tippetts, Biloune, and Metalclad cases, the dismissal of this case as time-barred101 contrasts sharply with the fact that “several cases decided by arbitral tribunals in the past two decades had explicitly focused on the effect on the owner as the dominant or exclusive criterion that delineates the line between a taking and a regulation.”102 This trend had been exemplified with the explicit and unequivocal pronouncement in favor of the “sole effect doctrine” in the Metalclad award of 2000.103 In order to reason this NAFTA Article 1110 claim on the merits, the tribunal should have carefully reasoned whether the four criteria of “expropriation” under NAFTA were satisfied. This analysis would have certainly been difficult, particularly with respect to the three remaining requirements of an expropriation claim. First, concerning the investment, for the purposes of NAFTA Chapter 11, “investment” is defined in exceedingly broad terms in Article 1139. It covers almost every type of financial interest, direct or indirect, except certain claims to money.104 There is no doubt that Mondev’s investment comes within the scope of Article 1110: it is made of its wholly owned local partnership, LPA, in the Lafayette Place project or, if not LPA, of the bundle of contract rights held by LPA (i.e., the right in the physical property that constitutes the first phase; the contract right to the option 101 See the Feldman award on jurisdictional issues (Preliminary Award on Jurisdiction, ¶ 39 et seq., Dec. 6, 2000, available at http://naftaclaims.com/Disputes/ Mexico/Feldman/FeldmanPrelimAwardJurisdiction.pdf), which held that, when explaining the meaning of Article 1117(1) (identical to the meaning of Article 1116 in this respect), “NAFTA itself did not purport to have any retrospective effect. Accordingly, the Tribunal may not deal with acts or omissions that occurred before January 1, 1994.” See, however, Professor Crawford’s welcome clarification, in the Hearings on Day 3, at 540. 102

Dolzer, supra note 9, at 86.

Note that “in the absence of special agreements, current international law still recognizes that investment agreements between a host state and a foreign corporation are, in principle, founded in the national law of the host state rather than in an undefined international legal order.” Dolzer, id., at 67. W.M. Reisman, The Regime for Lacunae in the ICSID Choice of Law Provision and the Question of Its Threshold, 15 ICSID REV.—FOREIGN INV. L.J. 362 (2000). 103

104 See Feldman v. Mexico, ICSID Case No. ARB(AF)/99/1, 42 I.L.M. 625, 644 (2003). On the diversity requirement, see Mondev, supra note 1, at ¶ 100.


and other contract rights that are basically the option right to purchase, to develop the second phase; and thereby, thirdly, to complete the whole project, which has an extra value rather than the value of its composing part).105 Second, concerning the expropriation, NAFTA’s broad reference to a “measure tantamount to . . . expropriation” (Article 1110(1)) is ambiguous because the treaty does not provide any definition of the concept of “expropriation.” Since the definition is ambiguous,106 the NAFTA tribunals have thus far approached the NAFTA expropriation regime cautiously107—only one claimant has ever succeeded in recovering compensation for an alleged expropriation. The absence of a definition of creeping expropriation under NAFTA should have led the Mondev tribunal to provide its own definition through dicta. Several options were available to the tribunal to frame this Article 1110 claim: a creeping expropriation, as the various cases cited suggest; neutralization of the benefits of the property; interference in the use of the property; or the enjoyment of its reasonably-to-be-expected benefits. By addressing this issue, the tribunal would have incidentally redefined the canon as to what can constitute a regulatory taking and would have most likely extended that concept to judicial decisions.108 To this end, from the Iran-U.S. 105 See Hearings, supra note 1, Day 1, at 189; Day 2, at 376 (U.S. position as summed up by Mondev); and Day 3, at 619 (on Mondev’s ownership of the rights with respect to the Hayward Parcel, which forms the basis of its claim before this tribunal). See also id., Day 3, at 636 (on the development by LPA, but if there is here a tension from within the claimant argument, it is not clear that there is inconsistency). 106 See the extremely broad standard of regulatory expropriations for which compensation is required in Ethyl Corp. v. Canada, NAFTA, UNCITRAL (June 24, 1998), 38 I.L.M. 708, 726 (1998) (“something other than a ‘law,’ even something in the nature of a ‘practice,’ which may not even amount to a legal stricture, may qualify”). In Metalclad, supra note 32, 40 I.L.M. 54, the seminal regulatory expropriation decision under NAFTA and the only one in which a compensable expropriation was found, the tribunal similarly resorted to a broad definition of “expropriation” (“coverts incidental interference with the use of property which has the effect of depriving the owner, in whole or in significant part, or the use or reasonably-to-beexpected economic benefit of property even if not necessarily to the obvious benefit of the host state”). 107 See, e.g., the dismissal of Methanex claims in their entirety (Methanex Corp. v. United States, Final Award (Aug. 9, 2005), available at http://www.state.gov/documents/organization/51052.pdf). 108 See V. Been & J. Beauvais, The Global Fifth Amendment? NAFTA’s Investment Protections and the Misguided Quest for an International “Regulatory Takings” Doctrine, 78 N.Y.U. L. REV. 30 (2003).


Claims Tribunal in Starret Housing v. Iran109 to the Metalclad v. Mexico NAFTA case,110 several authorities would have supported the proposition that de facto or “indirect expropriation” (as Mondev put it) is subject to expropriation claims, that is, that taking of property may be direct or indirect or that it may take place outright, in stages, or through successive acts or omissions. Third, concerning compensation, both the taking and deprivation of Mondev’s investment must have occurred and the wrongfulness of the pre-treaty deprivation must have continued into the period when NAFTA was in force.111 Only the fact that Mondev’s deprivation of property has remained uncompensated could create a breach of the international obligations upon the state.112 The tribunal’s reasoning should have left no room for ambiguity on the fact that the wrongful act was not dead or 109 The tribunal in Starret famously ruled that “it is recognized in international law that measures taken by a state can interfere with property rights to such an extent that these rights must be deemed to have been expropriated, even though the state does not purport to have expropriated them, and the legal title to the property formally remains with the owner.” Starret Housing v. Iran, 4 IRAN-U.S. CL. TRIB. REP. 122 (1983). 110 See supra note 105. See also S.D. Myers v. Canada, supra note 4 (“the drafters of the NAFTA intended the word ‘tantamount’ to embrace the concept of so-called creeping expropriation, rather than to expand the internationally accepted scope of the term expropriation. . . . The term ‘expropriation’ in article 1110 must be interpreted in the light of the whole body of state practice, treatises and judicial interpretations of that term in international law cases”). Finally, see CME v. The Czech Republic, supra note 12 (“De facto expropriation or indirect expropriation, i.e., measures that do not involve an overtaking, but that effectively neutralize the benefit of the property of the foreign owner are subject to expropriation claims. This is undisputed under international law. Furthermore, it makes no difference whether the deprivation was caused by action or by inaction.”). Professor Y. Dinstein used the term “deprivation,” in LIBER AMICORUM OF JUDGE ODA, at 855 (2002). 111 It is not undisputed (as claimed by the United States in U.S. Rej. and their Hearings, supra note 1, Day 1, at 259) that the “tribunal is competent to hear only claims for alleged breaches of Chapter Eleven based on acts or omissions of the United States that occurred after NAFTA’s entry into force.” In Mondev’s view, while a breach of NAFTA can only occur once NAFTA is in force, such a post-NAFTA breach does not have to be based only on acts or omissions which also post-date NAFTA. 112 In Foremost Tehran, Inc. v. Islamic Republic of Iran (10 IRAN-U.S. CL. TRIB. REP. 228 (1985)) for instance, the Iran tribunal held that it was not an expropriation up to the date of the cut-off of its jurisdiction. The American court subsequently held that subsequent events completed the expropriation. In other words, a state can be worse off when it creepingly expropriates as compared to when it overly expropriates. Yet, there is no intertemporal issue in the Foremost case as in the Mondev case, since the United States dealt with the Foremost case under a rule relevant at the time.


stale but continuing after January 1994 because the alleged misconduct should be conceived as being coupled with the failure to offer redress.113 The International Law Commission, in its explanation of what was to become Article 28 of the Vienna Convention on the Law of Treaties (VCLT), clarified that the treaty shall not apply to acts or facts that are completed, or to situations that ceased to exist before the treaty entered into force.114 In the event the wrong was the failure to pay compensation, the tribunal should have thus made explicit whether the possibility to obtain compensation through the normal applicable legal procedures ceased to exist on May 20, 1998 (SJC’s decision on the substance of the case) or on March 1, 1999 (Supreme Court’s date of decision).115 The tribunal should have faced the following issue: whereas the formal expropriation—where the state formally and by legislation nationalizes or expropriates a whole category of property—is accompanied by legislative provisions laying down procedures for compensation, this would not happen in the present case. The tribunal should have clearly reasoned what constitutes an appropriate compensation in the present case. NAFTA seems to adopt the Hull formula, albeit with different language, thus subjecting the three contracting parties to a higher standard of compensation for expropriation116 than the “appropriate compensation” standard on customary international law, under which the state decides for itself under its own laws and circumstances, how much compensation is reasonable for a given expropriation.117 IV. THE WELL-REASONED NATIONAL TREATMENT CLAIM According to Mondev, BRA’s refusal to fulfill its contractual obligations under the Agreement results from an anti-Canadian “naked animosity,” which would have turned into treating the Canadian company less favorably than its similarly situated American counterpart. The 113

See Hearings, supra note 1, Day 1, at 270–71.

114

See Professor Crawford’s useful distinctions, id. at 273–74.

See Hearings, id., Day 1, at 31, and Professor Crawford’s hypothetical, Hearings, id., at 410. 115

116 See J. Fowles, Swords Into Plowshares: Softening the Edge of NAFTA’s Chapter XI Regulatory Expropriation Provision, 36 CUMB. L. REV. 83, 89 et seq. (2005). 117 This standard has been adopted by the U.N. General Assembly Resolution on Permanent Sovereignty Over Natural Resources, G.A. Res. 1803, U.N. GAOR, 17th Sess., Supp. No. 9, U.N. Doc. A/1903 (Dec. 14, 1962), reprinted in 2 I.L.M. 223, 225 (1963). On this difference, see O. Schachter, Compensation for Expropriation,” 78 AM. J. INT’L L. 121 (1984).


tribunal well reasoned its dismissal of the Article 1102 claim on jurisdictional grounds arguing that the alleged discriminatory statements were made before NAFTA went into effect. Another reason for which there is no need to linger on the reasoning of this dismissal is that the tribunal took care, and did it perfectly well, to clarify in dicta that, if decided on the merits, the Mondev allegations would fail.118 The hearings leave no doubt that this alleged antiCanadian animus and the following violation of Article 1102(2) are of limited scope: they arose from a single statement allegedly made by a former director of the BRA in 1987 and from three statements of the City’s lawyers, some at the time of the court proceedings and others after them.119 The tribunal suggested that although it certainly does not condone the City’s conduct, a state’s refusal to abide by contractual obligations combined with occasional disparaging remarks is insufficient to establish a successful claim of discriminatory treatment. This reasoning follows in the footsteps of the usual interpretation of Article 1102 requirements by other NAFTA tribunals requiring either a showing of discriminatory intent or, at a minimum, discriminatory effect.120 The Mondev award does not contain any novel interpretations of Article 1102. On the contrary, its careful reasoning on both the jurisdiction and the merits of the case usefully reinforces the requirement of a high threshold of impropriety before international liability may attach to state conduct.

118

Notice of Arbitration, supra note 1, at ¶¶ 64–65

The record does not support Mondev’s claim that the Massachusetts court was biased. See U.S. C-Mem, supra note 1, at ¶ 62–67. Mondev itself makes it very clear that none of the four statements relates to the treatment accorded to LPA by the Massachusetts courts. Startlingly enough, neither in the notice of arbitration nor during the hearings, did Mondev seriously attempt to prove what was largely an issue of fact: for instance, while the records include evidence of the opposite when the treatment given to Campeau is taken into account, Mondev does not even attempt to demonstrate that an investment in like circumstances of a U.S. investor would have been treated differently than an investment of an Canadian investor, which is a necessary condition to claim that LPA had received less-favorable treatment. See U.S. CMem, id. at ¶¶ 63 and 67 and U.S. Rej., id. at ¶¶ 53–54. 119

120 It is then impossible to compare Mondev with the S.D. Myers case, where the finding of an Article 1102 violation is based on extensive evidence of blatant discrimination against an American investor in order to guarantee the preeminence of the Canadian industry.


V.

CONCLUSION: TIME REMEMBERED

Beyond the Mondev case itself, the international investment law community would have benefited from such an eminent tribunal addressing the question of expropriation in an extensively reasoned way.121 As noted by R. Ahdieh, like the Final Award in Loewen, the Mondev tribunal’s decision is suggestive of a “dialectical pattern of engagement.”122 Even if there had been no appeal, the Mondev tribunal looked to domestic legal norms (including the exceptions to the Federal Tort Claims Act’s abrogation of immunity) for guidance on the questions of interstitial lawmaking and sovereign immunity that were presented or a suggestion of some deference to the assessment of the jury, given its direct access to the evidence. Finally it rooted its adjudication on norms emerging from the pre-NAFTA bilateral investment treaties. Moreover, the tribunal had to explore in depth a number of issues not directly presented but potentially relevant to future domestic cases. 121 In this respect, it would have been interesting to reference Teodoro García and M.A. Garcia (United Mexican States) v. United States of America, 4 R.I.A.A. 119 (1926)) a case in which an American soldier shot at a Mexican family crossing the Rio Grande in a raft, thinking that the family was involved in smuggling. One bullet killed a child. The U.S. military court decided in favor of the family, convicting the soldier. But the President of the United States, using his presidential powers, reversed the decision, deciding in favor of the soldier. The Commission decided that there was no denial of justice. “In order to assume such a denial there should be convincing evidence that, put to test of international standards, the disapproval of the sentence of the court-martial by the President acting in his judicial capacity amounted to an outrage, to bad faith, to willful neglect of duty, or to insufficiency of governmental action so far short of international standards that every reasonable and impartial man would readily recognize its insufficiency. None of these deficiencies appears from the record.” Notwithstanding the rejection of the denial of justice claim, the Commission awarded damages. It held:

The only problem before this Commission is whether, under international law, the American officer was entitled to shoot in the direction of the raft in the way he did. The Commission makes its conception of international law in this respect dependent upon the answer to the question, whether there exists among civilized nations any international standards concerning the taking of human life. The Commission not only holds that there exists one, but also that it is necessary to state and acknowledge its existence because of the fact that there are parts of the world and specific circumstances in which human practice apparently is inclined to fall below this standard. 122 See Ahdieh, supra note 31, who refers his own perspective on non-conventional forms of judicial communication to the works of A. Bickel, R. Cover, A. Aleinikoff, and G. Calabresi.


Most significantly, this included a lengthy discussion of the permissible scope of a judicial definition of sovereign immunity, or, symmetrically, the notion of a creeping expropriation under NAFTA. Professor Dolzer wrote that “in the 1970s and 1980s, the battles before the courts and in academic writing were fought primarily in order to determine the standard of compensation and the measuring of expropriated value.”123 Today, the same author notes that: a widespread assumption in both the business and legal communities is that the international takings doctrine is in disarray, the jurisprudence is inconsistent, and the results are often unpredictable. To some extent, this belief can be founded on the review of the relevant international tribunal and judicial decisions124 [. . .] But in the domestic orders, the legal solution is also considered to be far from clear, and legal doctrine does not allow hard and fast rules and predictions.125 Were the arbitrators to have better reasoned the Article 1105 claim and decided to rule on the merits of the Article 1110 claim, they would have had a great occasion to clarify the rules on this paradigm case of an indirect or creeping expropriation or deprivation by state organs of a protected investor’s investment. This alternative was complex and discussed at length by the two parties during the hearings—and even at the very end, the arbitrators still remained unclear about the real nature of Mondev’s claim.126 It had to be dealt with in the award itself. ICJ President Roslyn Higgins once said that “time plays an important role in the international legal system.”127 Because time was crucial as well in Mondev, the inability of the tribunal to address the issue of the “continuing” nature of the wrongful acts makes the entire award flawed. By

123 R. Dolzer, New Foundations of the Laws of Expropriation of Alien Property, 75 AM. J. INT’L LAW 553 (1981). See also R. Higgins, The Taking of Property by the State: Recent Developments in International Law, 176 R.D.C.A.I. 259 (1982). 124 Professor Dolzer refers, for comparison, to CME v. Czech Rep., supra note 12, and Lauder v. Czech Rep., Final Award (Sept. 3, 2001). 125

R. Dolzer, supra note 9, at 66.

Mondev Memorial and Mondev Countermemorial are unfortunately not available for analysis. See U.S. Rej., supra note 1, at 2, and Professor Crawford’s ultimate demand for clarification in the Hearings, id., at 950. 126

127 R. Higgins, Time and the Law: International Perspective on an Old Problem, 46 INT’L COMP. L.Q. 501–20 (1997).


refusing to address Mondev’s takings claim as it should have done, the tribunal sidestepped what could have resulted in a further relaxation of the NAFTA expropriation standards. In any case, the tribunal would have set a useful precedent on related questions such as creeping expropriation, compensation and immunity in the NAFTA context.

CHAPTER 3

THE FINAL AWARD IN FELDMAN V. MEXICO Dennis Clare*

I.

INTRODUCTION

Marvin Feldman, a U.S. citizen, was the sole owner of Corporación de Exportaciones Mexicanas, S.A. de C.V. (CEMSA), a Mexican company that purchased cigarettes to export from Mexico. CEMSA’s export of cigarettes was economically viable because as an exporter it could obtain rebates of the large excise tax it paid when it purchased the cigarettes. Because of these deep rebates, CEMSA could profit from selling cigarettes abroad at prices lower than it had originally paid for them in Mexico. Thus, when Mexico declared CEMSA ineligible to obtain the rebates, Feldman claimed several North American Free Trade Agreement (NAFTA) violations, including expropriation and discrimination. Mexico claimed that it was merely enforcing laws with which CEMSA had not been properly complying, but Feldman argued that the manner in which Mexico ultimately put him out of business was not just tax code enforcement, but creeping expropriation. Feldman accused Mexico of numerous questionable patterns of behavior, including arbitrariness, discrimination, unreasonable treatment, inconsistency, and non-transparency, all of which the arbitral tribunal verified, but these behaviors did not in its view amount to an expropriation of CEMSA. The Feldman tribunal did find that Mexico discriminated against CEMSA in violation of NAFTA Article 1102 on National Treatment. However, this discrimination did not rise to the level that would constitute an expropriation under NAFTA, even when combined with other inappropriate actions on the part of Mexico.1

* J.D., June 2006, Yale Law School. 1 Marvin Feldman v. Mexico, ICSID Case No. ARB(AF)/99/1, Final Award (Dec. 16, 2002) [hereinafter Award].

63


This essay examines the tribunal award in the case of Feldman v. Mexico. It aims less to discuss the accuracy of the tribunal’s decision as it does to assess the arguments the tribunal puts forth for its decisions. Thus, this essay analyzes the tribunal’s reasoning in reaching many of its conclusions, including its main decisions on expropriation and discrimination. It points to numerous flaws in this reasoning, from arguments that are not sound because of false premises, such as failure to cite sources for key principles of international law, to arguments that are not valid because of poor inferences, caused by failure to adhere to requirements of the burden of proof, and by inattention to the parties’ specific claims. In addition, this essay shows that the tribunal award in Feldman suffers from an overall tendency to reach conclusions based upon numerous considerations weighed as a totality rather than based upon specific, dispositive points of argument. While the elements of an argument may not be compelling on their own, the tribunal maintains, when considered on the whole, the factors it cites serve as justifications for its decisions. In examining these elements, however, this essay suggests that this form of aggregate argumentation is problematic. Not only are the various parts of these arguments often interrelated, such that if one part fails, so too do the others, but also there is no reason to conclude that an aggregate mass of less than persuasive arguments takes on the effect of a controlling or compelling argument when considered as a whole. Poor arguments remain poor arguments even when assembled together. II.

FACTS OF THE CASE

The Feldman tribunal emphasizes in its award that for various reasons many of the facts of the case remain unclear. In some cases, Mexican ministry records have been destroyed as a matter of policy. In others, the tribunal suggests Mexico was less than forthcoming in providing relevant information. In addition, Feldman alleges that certain key agreements were made orally. Therefore, there is little documentary evidence of such agreements. Moreover, where correspondence does exist between Feldman and ministry officials, it is often in the form of responses to letters that are now missing, thus providing little insight into what had been previously discussed. What is clear is that the production and sale of cigarettes in Mexico have long been subject to provisions of the nation’s Impuesto Especial Sobre Produccíon y Servicios (IEPS) law, an excise tax that has ranged from 139.3 percent to 85 percent over the years. The IEPS tax rate on exports of cigarettes, however, was 0 percent during all times CEMSA

The Final Award in Marvin Feldman v. Mexico • 65

exported cigarettes. Furthermore, according to Article 4 of the IEPS law, exporters who had paid the sales tax when purchasing cigarettes in Mexico were entitled to rebates of the tax amount if the cigarettes were later exported, as long as they submitted receipts with their rebate claims specifically indicating the amount of IEPS tax paid in the sales price. Feldman’s company, CEMSA, began exporting cigarettes from Mexico in 1990. The record indicates that CEMSA received IEPS rebates for exports that year. Although Feldman claims that these included rebates for exports of cigarettes, Mexico insists that CEMSA only received rebates for exports of alcoholic beverages that year. Amendments to the IEPS in 1991 then made it impossible for CEMSA to obtain the rebates and, hence, to export cigarettes. Under the new regulations, only producers of cigarettes and their buyers/distributors were permitted to claim rebates for the IEPS tax. Companies like CEMSA, which exported cigarettes purchased from big box retailers such as Walmart and Sam’s Club, were now excluded. Feldman challenged these 1991 changes in an Amparo action in a Mexican court, which ruled, inter alia, that the Mexican Ministry of Finance and Public Credit (Secretaría de Hacienda y Crédito Público (SHCP)) did not have the authority to issue these changes. Maintaining that the changes were made at least in part because Mexican business mogul and major Mexican cigarette producer, Carlos Slim, pressured SHCP to drive CEMSA out of the cigarette export business, Feldman also filed a criminal complaint against certain SHCP officials for abuse of authority and conspiracy. While the original court ruling in favor of CEMSA was still on appeal, the Mexican congress amended the IEPS law, effective January 1, 1992, to once again permit the IEPS rebates for reseller exporters. Thus, CEMSA again began to export cigarettes, and although the relevant records have been destroyed, Feldman claims rebates were provided to CEMSA for its cigarette exports throughout 1992. In January 1993, CEMSA was once again forced to stop exporting cigarettes, this time because it could not comply with IEPS Article 4, requiring the taxes paid in retail purchases to be separately stated on receipts submitted for export rebates. Because CEMSA purchased cigarettes from big box retailers instead of buying them directly from the cigarette producers, the IEPS tax those retailers paid in their original purchase of the cigarettes from the producers was included in CEMSA’s purchase price but was not separately noted on the receipts CEMSA received. Therefore,


CEMSA could not provide SHCP with receipts specifying the IEPS amount it paid in its cigarette purchases. With no ability to get the rebates, CEMSA could no longer profit from cigarette exports. In August 1993, the Mexican Supreme Court of Justice ruled unanimously in favor of CEMSA on the pending Amparo appeal, declaring that “measures allowing IEPS rebates only to producers and their distributors violated constitutional principles of tax equity and non-discrimination.”2 In the tribunal’s view, “[t]he court did not discuss or rule explicitly on any other relevant issues, such as whether the Claimant was entitled to rebates notwithstanding the Claimant’s inability to produce invoices stating the tax amounts separately.”3 From 1993 to mid-1996, CEMSA could not export cigarettes because SHCP insisted that it fulfill the receipts requirement that it could not fulfill. Then, in June 1996, CEMSA began exporting cigarettes yet again. According to Feldman, a series of negotiations with Mexican officials in 1995 and 1996 led to an oral agreement whereby CEMSA could obtain rebates for cigarette exports. Although Mexico strongly denies that such an agreement was made, CEMSA did receive rebates for the cigarettes it exported from June 1996 until approximately December 1997. On or about December 1, 1997, rebates to CEMSA were terminated, and CEMSA did not receive the rebates expected for exports during October and November 1997.4 Amendments to the IEPS law in December 1997 then restricted rebates for exports to the first buyer of cigarettes in Mexico, once again denying resellers such as CEMSA a right to IEPS rebates on exported goods. The amendments also required exporters of specific goods, including cigarettes, to register in the Sectorial Exporters Registry in order to be entitled to the 0 percent tax rate on exports. In 1998, CEMSA’s submission for export registration was denied. Mexico claims CEMSA’s registration was denied because it was the subject of an ongoing audit, which investigated CEMSA’s export rebates from June 1996–September 1997. The audit led SHCP to demand that 2

Award, supra note 1, at ¶ 16.

3

Id.

4

Feldman challenged this denial of rebates in a Mexican court and ultimately

lost.


CEMSA repay the approximately U.S. $25 million in rebates it received over this period, plus interest and penalties. CEMSA challenged this tax assessment in an action for declaratory judgment in a Mexican court, which remained pending at the time the tribunal award was issued. As a U.S. owner of a Mexican enterprise, Marvin Feldman filed for arbitration against Mexico in accordance with NAFTA Article 1117 on April 30, 1999. Feldman claimed violations by Mexico of NAFTA Articles 1110 (Expropriation) and 1105 (Minimum Standard of Treatment) and later added a claim under Article 1102 (National Treatment). In accordance with NAFTA Article 1120, the case was heard by an arbitral panel in accordance with the International Center for the Settlement of Investment Disputes (ICSID) Additional Facility Rules. The ICSID arbitral tribunal gave no deference to how these issues— including issues of Mexican law as to whether Feldman had a legal right to the IEPS tax rebates—were being resolved in ongoing proceedings in Mexican courts. The tribunal rejected the expropriation claim on various grounds, including that Feldman never had a legal right to the rebates.5 However, on a comparison of only two companies in like circumstances, the tribunal did find that Mexico had violated its national treatment obligations by discriminating against CEMSA relative to domestic companies. In a dissenting opinion, the minority tribunal member concurred with the award on the expropriation claim but dissented with regard to the discrimination finding. The dissent claimed, inter alia, that if CEMSA could not claim expropriation because it had no right to the rebates, it should also not prevail on the discrimination claim for the same reason.6 III. JURISDICTIONAL ISSUES The tribunal made several initial decisions on jurisdiction in a preliminary award, but attached other jurisdictional issues to the award on the merits. These issues are worth examining because the tribunal’s analysis here informs its rationale for its decisions on expropriation and 5 The tribunal’s rejection of the expropriation claim also made Feldman’s NAFTA Article 1105 claim moot. Because Article 1105 is not applicable in tax cases, a claim under this article could only have arisen under Article 1110(1)(c), if an expropriation had been found. 6 Marvin Feldman v. Mexico, ICSID Case No. ARB(AF)/99/1, Dissenting Opinion (Dec. 16, 2002) [hereinafter Dissent].


discrimination in subsequent sections of the award. Thus, in what follows, this essay will examine the tribunal’s reasoning in its three jurisdictional decisions on estoppel, in its decision on exhaustion of local remedies, and in its decisions on other jurisdictional matters. In this examination, special attention will be paid to the tribunal’s disposition toward Mexican courts in deciding matters of Mexican law. A.

The Tribunal’s Decisions on Estoppel

1.

Overview of Feldman’s Three Estoppel Claims

As noted above, Feldman claims to have reached an agreement with Mexican tax officials in negotiations over the course of 1995–1996, which culminated with “assurances” that CEMSA would receive IEPS rebates for exports even though it could not provide receipts separately indicating the specific amount of IEPS taxes it had paid. Feldman claims he was told to calculate those figures himself based on current IEPS rates.7 He was then to submit the calculations to tax officials to receive CEMSA’s rebates. Feldman not only claims Mexican officials gave CEMSA specific authorization to export and receive tax rebates without the receipts required by Article 4 of the IEPS law, he also submits that his alleged agreement with tax officials should restrict Mexico’s defenses to his NAFTA claims in three ways. First, Feldman argues the agreement should serve to toll the three-year time limitation set in NAFTA Article 1117(2) for bringing damage claims for as long as the agreement was in place. Second, because Mexico gave him assurances that he would receive rebates for exports, Mexico should be estopped from relying on a time limitation defense altogether. According to Feldman, a finding of equitable estoppel or “tolling” is appropriate in a case such as this one where a lawsuit was discouraged by the actions of a defendant. Although the clearest example is where a defendant has expressly agreed not to raise a defense based upon a statute of limitations, other representations, promises or actions will suffice to estop a party from invoking a statute of limitations.8 7 Marvin Feldman v. Mexico, Claimant’s Memorial, at ¶ 184 (Mar. 30, 2001) [hereinafter Memorial]. 8

Id. at ¶ 187.


Thus, Feldman insists that two things should happen: (1) Article 1117’s three-year limit on bringing NAFTA claims should extend backward for approximately 32.5 months to account for the period from which the agreement was allegedly made in June 1995 until February 1998, when he received what he understood as formal notice that the agreement was terminated;9 and (2) Mexico should be prevented from invoking a time limitation at all. Third, Feldman argues that this agreement estops Mexico from asserting that CEMSA was not entitled to rebates for the period 1996–1997. Thus, according to Feldman, Mexico is also estopped from denying the very basis of his claim. The same principle of estoppel, which calls for a tolling of, and/or a bar from invoking, the time limitation, Feldman submits, bars Mexican officials from denying now what they had confirmed to him earlier, namely, CEMSA’s entitlement to IEPS rebates without the receipts. Feldman claims he relied upon such assurances by Mexico to his detriment when CEMSA bought and exported cigarettes in expectation of receiving those rebates. Thus, Mexico should be barred both from reclaiming the rebates it paid to CEMSA in 1996–1997, and from denying rebates to CEMSA for exports in October and November 1997. Mexico strongly denies any such agreement with Feldman was made and notes a lack of hard evidence to support his allegation. It claims any assurances it gave Feldman regarding CEMSA’s entitlement to rebates concerned only its intention to comply with the Supreme Court’s Amparo ruling, which it insists guaranteed resellers the same tax rate as producers but did not reach the issue of exporters’ entitlement to rebates without submitting the required receipts. Hence, on Mexico’s interpretation of the Amparo decision, resellers were entitled to the tax rate and rebates, but they still had to comply with the other requirements of the IEPS law, such as providing the proper receipts. Moreover, Mexico argues that any agreement that called for it to disregard provisions of its tax code, such as the receipts requirement for export rebates, would not be binding as a matter of Mexican law. Furthermore, Mexico also rejects Feldman’s appeal to estoppel derived from international law since international law 9 As the tribunal points out, extending the time limitation backwards an extra 32.5 months from the time Feldman filed his notice of arbitration on April 30, 1999, would make the relevant date not April 30, 1996, but some time in mid-August 1993, before NAFTA even came into force. However, because the tribunal rejected Feldman’s estoppel claim, this was not a significant issue.


ultimately derives from state practice, and under the law of all three states signatory to NAFTA, estoppel cannot be invoked to prevent enforcement of the tax code. 2.

Analysis of the First Estoppel Decision

The tribunal rejects each of Feldman’s three claims. First, it claims that NAFTA does not contemplate any tolling of its time limitations. 10 Nowhere in the treaty, the tribunal notes, are any conditions for such tolling stipulated. In addition, the international law conditions that might call for such tolling are not present. The reasoning the tribunal employs to make this argument, however, is somewhat questionable. For example, the tribunal asserts the following: Even under general principles of law to be applied by international tribunals, it should be noted that in several national legal systems such suspension is provided only in the final part of the limitation period (e.g. in the last six months) and only in either cases of act of God or if the debtor maliciously prevented the right holder from instituting a suit (see e.g. German Civil Code para. 203; Greek Civil Code Article 255). In this case no such unavoidable events have been pleaded.11 The wording of the first sentence is confusing on its face. Strictly read, it states only that under general principles of international law, the practices of various states should be noted. However, this sentence functions in the tribunal’s award to assert something much more concrete. It means that: 1. 2. 3. 4.

General principles of international law apply here; Those principles direct us to examine the laws of various states; A look at the laws of two states, Germany and Greece, gives us the dispositive rule, applicable in this case; Suspension of time limitation periods can only take place within, and towards the end of, the limited period, and may only be based upon acts of God or malicious acts by a wouldbe defendant towards a would-be plaintiff.

10


11

Id.


Thus, the tribunal essentially rules upon Feldman’s claim not only without reference to a clear rule of international law, but also without sufficient justification for the standard it invokes. This is not to say, of course, that the tribunal may not be correct. It is simply to highlight the cursory treatment the tribunal gives to the task of stating the sources for its principle of decision making. The tribunal then adds to its explanation in the same paragraph, stating the following: Basically, the Claimant maintains that a lawsuit was “discouraged” by Respondent’s actions . . . However, “discouraging” a lawsuit does not amount to preventing it. The decision whether, and when, to bring a lawsuit lies with the prospective plaintiff, who also bears the respective benefits and risks. Among the various factors to be taken into consideration is the running of the period of limitation and its interruption as well. Nothing in the file shows that the Claimant . . . was prevented from taking into consideration all relevant factors.12 This statement is problematic for two reasons. First, it waffles as to the criteria it employs. It begins by suggesting that estoppel might apply when a lawsuit is “prevented” but not when one is just “discouraged.” It then suggests that the criterion might not be whether the lawsuit itself is prevented, but whether a would-be claimant was prevented from considering the time limit when deciding whether to file suit. Second, and more important, the tribunal once again fails to cite a source for its criterion that prevention of a lawsuit, not just discouragement of a suit, is necessary for an assertion of estoppel. Thus, although the tribunal flatly contradicts Feldman’s claim that “a finding of equitable estoppel or ‘tolling’ is appropriate in a case such as this one where a lawsuit was discouraged by the actions of a defendant,” it does not explain why this is the case or cite a source of law that supports its understanding. Such a citation is crucial here because of Feldman’s specific claim: “Although the clearest example is where a defendant has expressly agreed not to raise a defense based upon a statute of limitations, other representations, promises or actions will suffice to estop a party from invoking a statute of limitations.”13 The tribunal has in essence deter12

Id.

13

Memorial, supra note 7, at ¶ 187.


mined that “other representations, promises or actions” will not suffice, but it has not stated why it understands the actual prevention of a lawsuit as the only grounds for estopping application of time limitation. The tribunal has therefore rejected Feldman’s claim without actually addressing it. 3.

Analysis of the Second Estoppel Decision

On the issue of estopping Mexico from using any time limitation at all, the tribunal takes a similar approach to the one it took on a tolling of the time limitation. First, it claims once again that “NAFTA Articles 1117(2) and 1116(2) introduce a clear and rigid limitation defense which, as such, is not subject to any suspension, prolongation or other qualification.”14 The tribunal then immediately changes course, stating that Of course, an acknowledgement of the claim under dispute by the [state] organ competent to that effect and in the form prescribed by law would probably interrupt the running of the period of limitation. But any other state behavior short of such formal and authorized recognition would only under exceptional circumstances be able to either bring about interruption of the running of limitation or estop the Respondent State from presenting a regular limitation defense. Such exceptional circumstances would include a long, uniform, consistent and effective behavior of the competent State organs which would recognize the existence, and possibly also the amount, of the claim. No such circumstances were presented to the Tribunal in this case.15 This reasoning is also thoroughly problematic for five reasons. First, if it acknowledges the existence of exceptions to the NAFTA time limitation, the tribunal should not insist that there are no such exceptions just prior to describing the exceptions. Second, the tribunal offers no reason why a “formal and authorized recognition” of a claim by a state party could possibly estop that party from using a time limitation defense. Again, this is not to say that this is not true, but rather to say that if it is true, the tribunal should cite the international law authority from which it derives this understanding. 14


15

Id.


Third, the tribunal offers no suggestion as to what would constitute a “formal and authorized recognition” by a state party or how such recognition would be identified from an international law perspective. Fourth, even if one had the means to identify a “formal and authorized recognition” by a state party “organ competent” to do so, it does not follow that the formality of the recognition therefore creates an exception to NAFTA Article 1117(2). That the tribunal mentions the state party “organ competent” to recognize the claim suggests that perhaps a state’s laws will indicate the criterion for what is “formal and authorized” recognition for that state, but this does not mean that international law, upon which NAFTA claims are decided, thereby instructs adjudicators to look at a state’s law to determine if an unspoken NAFTA exception kicks in. Fifth, if there are additional exceptions to NAFTA Article 1117(2) beyond a competent state organ’s formal and authorized recognition of a claim, where do they come from? The tribunal states clearly that “exceptional circumstances include a long, uniform, consistent and effective behavior of the competent State organs which would recognize the existence, and possibly also the amount, of the claim,” but does not indicate where it found, or from what authority it derived, this standard. The tribunal states that no exceptional circumstances were presented to it in this case. But this statement dismisses Feldman’s factual assertions without saying why they fail to qualify as the type of “exceptional circumstances” the tribunal would require. Feldman’s argument regarding his alleged agreement with Mexico can easily be understood as precisely the claim that SHCP’s behavior in rebating CEMSA’s IEPS taxes from June 1996–September 1997 was “a long, uniform, consistent and effective behavior of the competent State organs.”16 Indeed, it would be hard to argue that SHCP’s accurately repaying CEMSA’s rebates for 16 straight months did not entail the exact factors the tribunal cites in its criterion above, that is, the competent government authority consistently and uniformly recognizing both the existence and the amount of CEMSA’s claims. Perhaps sensing this, the tribunal offers no argument. Instead, it acknowledges that “some assurances on CEMSA’s entitlement to IEPS tax rebates were given to Claimant and CEMSA at various times by various middle- and high-ranking SHCP officials, and with varying content” before simply asserting without any further support that SHCP’s assurances to Feldman “never amounted to either an authorized and formal 16

Id. at ¶ 64.


acknowledgement of the claim by the Respondent or to a uniform, consistent and effective behavior of Respondent. Therefore the Tribunal does not deem that the Respondent is estopped from invoking the threeyear limitation period under NAFTA Article 1117(2).”17 One is thus left to wonder which component of the tribunal’s apocryphal criterion Feldman’s claim fails to satisfy. 4.

Analysis of the Third Estoppel Decision

The tribunal goes on to apply this same criterion to reject, for the same reasons, Feldman’s third and final jurisdictional claim—that his agreement with SHCP estops Mexico from denying the very basis of his claim, CEMSA’s legal right to rebates. “Here again,” the tribunal declares, “the criterion is a long, uniform, consistent and effective behavior of the competent State organs.”18 It then repeats, in part verbatim, the same reasoning from the previous decision on estopping the time limitation: The tribunal recognizes again that some assurances on CEMSA’s entitlement to IEPS tax rebates were given to Claimant and CEMSA at various times, probably over a longer period, by various middle- and high-ranking SHCP officials, and with varying content. However, the Tribunal misses the uniform, consistent and effective character of such behavior as well as it connection with the competent State organs at all times.19 More interesting, though, is what is missing—a justification from the tribunal as to how it reached this conclusion, regardless of its correctness. Parsing this language, it seems that in making the last two decisions, the tribunal conflates its own two exceptions to NAFTA Article 1117(2). It may be the case that Feldman fails to meet the first exception, for other than what Feldman claims was an oral agreement, there is no explicit “formal and authorized recognition” by SHCP that CEMSA was entitled to export rebates without submitting the required receipts. But it is not clear why, in terms of the second exception, when looking for uniform, consistent, and effective behavior by Mexico, the tribunal examines only those same assurances alleged to constitute the agreement between Feldman and SHCP (i.e., the evidence in support of the first 17

Id. at ¶ 63.

18

Id. at ¶ 64.

19

Id.


exception). Why does the tribunal look only at these assurances and totally disregard SHCP’s uniform, consistent, and effective paying of CEMSA’s rebates between June 1996 and September 1997 as the kind of behavior that could satisfy the second exception? For, as noted above, a good argument can be made that SHCP’s practice of providing rebates to CEMSA during that period was precisely such behavior, that is, the competent state organ recognizing both the existence and amount of claims by uniformly, consistently, and effectively paying them to CEMSA. Again, the tribunal offers no analysis here. Still, hints as to how the tribunal decided this matter may be gleaned from the two sentences it tags on to the end of the paragraph rendering this third estoppel decision. Having stated that Mexico’s behavior does not meet the criterion for the second exception, the tribunal adds, as if it somehow follows or even obliquely relates: In this respect, the Tribunal also takes into consideration that in any state governed by the rule of law there is no way to impose, to reduce, to claim, to recuperate, or to transfer any tax burdens by agreements with some tax officials not provided by the law. Such agreements would necessarily have a quasi private and could neither bind the State nor be enforced against it.20 By beginning the first sentence above with “In this respect” the tribunal is attempting to link its foregoing considerations to the principle it now introduces. That language is misleading. The principle the tribunal introduces is entirely distinct from its previous considerations. In fact, to the extent that this principle is true and applicable here, it renders those considerations moot. For if no agreement can alter the clear prescriptions of the law, the tribunal has no reason to examine possible exceptions to the law based upon such alleged agreements.21

20

Id.

This, however, raises the question of how a tribunal would address a situation in which there was evidence that representatives of a state enticed an investor to make an investment by alleging certain laws presently on the books would not be applied. If the Feldman tribunal is correct that no such agreement could alter the law, then the tribunal would seem to be endorsing a policy of “caveat investor,” suggesting that such claims by state representatives would not affect the outcome of an international arbitration. This would be problematic because states would seem to have license to make various promises to entice investment which they then would not be bound to uphold. 21


In addition, the legal significance of noting that a tribunal takes a certain principle “into consideration” is unclear. Either the principle was one of the grounds on which the decision was made, in which case the tribunal should simply say so and explain how, or it was inapplicable to the decision, in which case the tribunal’s citing it only muddies its reasoning. The tribunal seems to try to have it both ways, suggesting that this principle did not determine the decision but somehow informed it in a significant way. Although determining what role this principle plays or does not play in the tribunal’s third estoppel decision may not be possible, the tribunal’s citing it does reveal one key aspect of the tribunal’s understanding of the case—that the agreement Feldman alleges took place would require parties to enter into “agreements with some tax officials not provided by the law.” In other words, the tribunal sees the agreement Feldman alleges as an agreement to disregard the valid provision of the tax law requiring submission of invoices with IEPS tax amounts separately stated in order to obtain rebates. Quite importantly, however, this is not what Feldman alleges. Feldman does not claim that he reached an agreement with Mexican tax officials to ignore the law. On the contrary, he claims his agreement with SHCP officials was in regard to how the law would be applied subsequent to his successful Amparo action. As noted, Feldman had challenged provisions of the 1991 IEPS law denying rebate opportunities to resellers. The court ruled that “measures allowing IEPS rebates only to producers and their distributors violated constitutional principles of tax equity and non-discrimination.”22 Feldman interpreted this ruling to reach the receipts requirement too, rendering the requirement invalid due to the inability of resellers to obtain the receipts. Thus, on Feldman’s view, the Amparo decision invalidated all measures that discriminate between resellers and producers/distributors with regard to obtaining export rebates. Both Mexico and the tribunal interpret the Amparo decision differently. Mexico claims, and the tribunal agrees, that the Amparo decision simply did not reach the receipts issue. According to the tribunal, the court only addresses the specific provisions of the 1991 IEPS law being challenged: “The court did not discuss or rule explicitly on any other relevant issues, such as whether the Claimant was entitled to rebates

22

Id. at ¶ 16.


notwithstanding the Claimant’s inability to produce invoices stating the tax amounts separately.”23 This enables us to reevaluate the tribunal’s consideration of Feldman’s alleged agreement with SHCP officials. On the tribunal’s interpretation of the Amparo decision, the agreement that Feldman alleges allowed him to obtain export rebates would have required tax officials to not enforce the receipts provision of the IEPS law. This explains the tribunal’s mention of the principle that agreements to take actions not provided for by the law cannot be binding on or enforced against states. What it does not explain, and indeed, a large question that it raises is, why, if the tribunal understood the alleged agreement this way, it entertained Feldman’s estoppel arguments at all. Why did the tribunal not reject Feldman’s estoppel claims from the start on the grounds that they would require the tribunal to treat as binding alleged agreements that cannot be binding because they are contrary to law? In other words, we are left again to wonder why the tribunal merely affixed this principle to the end of its estoppel considerations, saying it took this principle “into consideration” rather than using it as the grounds for a decision to reject Feldman’s “agreement”-based estoppel arguments altogether. Unfortunately, the tribunal’s final comments on the third estoppel decision only complicate its reasoning. The tribunal notes that although Feldman’s alleged agreement with SHCP does not merit estopping Mexico’s defenses to his claims, certain actions of Mexican officials may be brought to bear in the tribunal decisions on other sections of the award: Notwithstanding this [estoppel] finding, the Tribunal will consider such behavior of several SHCP officials while examining the bases of “creeping” or otherwise relevant form of expropriation, or effective denial of national treatment, under NAFTA Articles 1110 and 1102. Indeed, it is possible that behavior of some State organs such as the ones under consideration here may have led the claimant to initiate, or to expand, his investment and, thus, may have contributed to the occurrence or amount of his damage, if any.24

23

Id.

24

Id. at ¶ 65.


This statement seems to contradict the principle just cited above. It is difficult to understand—especially on the tribunal’s rather Archimedean view of what is and is not lawful—how actions could have no binding effect in one context because they are extra-legal, but could nonetheless have important legal impact in another context. B.

The Tribunal’s Decision on Exhaustion of Local Remedies

The tribunal’s decision on exhaustion of local remedies sheds light on the critical issue of the applicability of Mexican law and the tribunal’s role as interpreter of that law. Indeed, this essay suggests that the decision on exhaustion of local remedies carries within it a decision on CEMSA’s right to receive IEPS rebates under Mexican law and, thus, a decision on whether Feldman had an investment in accordance with NAFTA. This issue is insufficiently examined in both the pleadings of the parties and in the analysis of the tribunal. As the tribunal explains, rather than claiming Feldman did not have an investment in accordance with NAFTA, Mexico “basically denies that the Claimant has any right to receive IEPS rebates as a matter of Mexican law.”25 This is critical to the overall case because as Mexico argues, there “is no international legal right to IEPS rebates; it is purely a question of domestic law.”26 Hence, claims Mexico, “if CEMSA did not have the ‘right’ at domestic law that it claims was expropriated, there can be no expropriation at international law.”27 Importantly, whether CEMSA had a legal right to IEPS rebates and whether such rebates required compliance with the receipts requirement were both matters pending in separate proceedings in Mexican courts at the time the tribunal issued its award. Therefore, Mexico claims that the tribunal, in accordance with NAFTA Article 1131, should have applied the exhaustion of local remedies rule and refrained from substituting its own interpretation of Mexican law for that of Mexican courts on issues presently pending in those courts.

25 Id. at ¶ 69. This, in turn, would have meant that Feldman did not have an investment in accordance with NAFTA, but to its detriment Mexico did not make this explicit in the pleadings, nor did the tribunal did not recognize this lack of investment as a jurisdictional matter. 26

Marvin Feldman v. Mexico, Respondent’s Rejoinder at ¶ 30 (June 25, 2001).

27

Id. at ¶ 31.


The tribunal responds first by arguing that NAFTA Article 1121(2)(b) contemplates a general derogation from the customary international law rule of exhaustion of local remedies. The tribunal asserts: Article 1121(2)(b) requires, for recourse to arbitration to be open, that the disputing investor waive his right to initiate or continue the other domestic proceedings. Therefore, in contrast to the local remedies rule Article 1121(2)(b) gives preference to international arbitration rather than domestic judicial proceedings, provided that a waiver with regard to the latter is declared by the disputing investor. This preference refers, however, to a claim for damages only, explicitly leaving available to a claimant “proceedings for injunctive, declaratory or other extraordinary relief” before the national courts. Thus, Article 1121(2)(b) . . . substitutes itself as a qualified and special rule on the relationship between domestic and international judicial proceedings, and a departure from the general rule of customary international law on the exhaustion of local remedies.28 It is not clear, however, that NAFTA’s requirement that parties waive domestic damages in some types of proceedings in order to arbitrate “gives preference to international arbitration rather than domestic judicial proceedings.” Rather, it seems to give no preference to one or the other, leaving either option open, and merely insisting that a party not attempt to get two bites at the apple by simultaneously trying claims in both fora. The tribunal simply goes too far in suggesting that Article 1121(2)(b) is meant as a general departure for NAFTA from the local remedies rule. Because Article 1121(2)(b) does not address all remedies, nor even all judicial remedies, it should not be interpreted as a general substitute to the local remedies rule. Rather, it should only be a substitute for the local remedies rule with regard to the claims it requires to be waived. Legal actions Article 1121(2)(b) does not apply to should still be presumed to be subject to the local remedies rule. Here, Feldman was seeking a declaratory judgment to invalidate the tax assessment calling for CEMSA to repay rebates, a remedy to which Article 1121(2)(b) explicitly states it does not apply. Therefore, it would be incorrect to assume Article 1121(2)(b) overrides customary international law rule with respect to this remedy. Instead, we should presume the local remedies rule does still 28



apply and that the tribunal should allow Feldman to conclude his pursuit of a declaratory judgment. We should also presume that it would behoove an arbitral tribunal purporting to determine whether an investment was expropriated to refer to the decision of the sole forum capable of determining whether there was such an investment in the first place—the Mexican courts. Indeed, it would be awkward if a tribunal interpreted a state’s domestic law in such a way that led it to rule that a claimant’s right was expropriated before a domestic court then resolved the same issue by ruling that as a matter of national law, the claimant never even possessed the right that the tribunal already ruled had been expropriated. It would be even more questionable if a claimant had to waive a right to proceed in a domestic case that would decide such an issue in order to participate in the NAFTA arbitration that would then decide the issue instead, with or without deference to the most recent decision of the domestic courts on the issue. The tribunal does not seem to be troubled by these awkwardly contradictory possibilities. Instead, it states confidently that questions as to whether Mexican law as determined by administrative authorities or Mexican courts is consistent with the requirements of NAFTA and international law are to be determined in this arbitral proceeding, and we are not barred from making that determination by the fact that not all of the issues have yet been resolved by Mexican courts.29 This statement does not resolve the kinds of contradictions contemplated above. Indeed, it merely offers a contradiction of its own, suggesting that a tribunal can somehow interpret whether domestic law “as determined by . . . Mexican courts” is consistent with NAFTA before that law has been determined by those courts. This is precisely the type of problem Mexico sought to avoid by emphasizing that the “juridical fact necessary to establish the existence of something that could be expropriated at international law, viz., a right established at domestic law (assuming it is an investment protected by Article 1139), is undecided in present Mexican jurisprudence.”30 29


Marvin Feldman v. Mexico, Respondent’s Countermemorial at ¶ 363 (undated) [hereinafter Countermemorial]. 30


To further justify its decision, the tribunal also cites the truism that an action determined to be legal under Mexican law by Mexican courts is not necessarily legal under NAFTA or international law. But this is misleading because what is at stake here is purely an issue of domestic law. As Mexico argues, “where the elements of the international wrong require the denial of a recognized right in domestic law, the domestic court’s resolution as to whether the alleged right exists is authoritative.”31 In other words, even if the tribunal insisted on deciding this issue solely on NAFTA and/or international law grounds, it would discover that those bodies of law direct it to decide the issue by interpreting Mexican law, which, as the tribunal itself acknowledges in the quotation above, demands examining the role Mexican courts have played in determining that law. The Feldman tribunal disregards this crucial consideration, and thus not only employs very poor reasoning on its exhaustion of local remedies decision, but also does not take into account that what is at stake here—namely, whether CEMSA had a right to IEPS rebates, and, in turn, whether Feldman even had an investment in accordance with NAFTA Article 1139—may be dispositive of the entire case.32 C.

The Tribunal’s Decisions on Other Jurisdictional Constraints

The issue of the tribunal’s deference toward Mexican court decisions on matters of Mexican law is raised again in the section of the award called “Other Jurisdictional Constraints.” Here the inquiry is whether the tribunal has the authority “to grant declaratory relief with respect to the validity or legality of the 1998 audit and the corresponding tax assessment by SHCP vis-à-vis CEMSA.”33 Because Feldman had challenged these actions in Mexican courts, the tribunal examines what deference it should give the Mexican court decisions on these issues, which include whether CEMSA had the right to receive rebates for its cigarette exports notwithstanding its inability to submit receipts indicating the amount of IEPS tax paid. Although the tribunal ultimately rules that such declaratory relief is within its authority, it once again decides not to give deference to the Mexican courts on issues of Mexican law, and once again its justifications for doing so are insufficient. 31

Id. at ¶ 364.

As I note in my analysis of the expropriation decision, the tribunal interpreted Mexican law as holding that CEMSA did not in fact have a right to the rebates. Thus, the tribunal could have decided that it had no jurisdiction because Feldman had no investment. 32

33



The tribunal offers several reasons why it is “not inclined to give [the Mexican court decisions] significant weight.”34 First, it states that neither party has suggested these decisions are controlling. This is problematic for two reasons: first, because the tribunal is obliged to make proper decisions on matters of law and jurisdiction regardless of whether the parties have argued they are controlling; and second, because Mexico’s argument demonstrates that the issues at stake in whether receipts are required for rebates may indeed be controlling. As the discussion of local remedies above indicates, what is at stake in the tribunal’s reading of Mexican court decisions may be controlling because it may determine whether Feldman has an investment under NAFTA. If CEMSA never had a right to export cigarettes, then Feldman may have never had an investment in accordance with Article 1139 and thus may have no standing to bring a NAFTA claim.35 Mexico’s claim that to disregard its courts’ decisions would be to “usurp the jurisdiction of the Mexican courts” is precisely the suggestion those courts’ decisions are controlling on this issue.36 Indeed, to usurp, by definition, is to improperly take control; hence, the basis of the claim that Mexican jurisdiction would be usurped is the idea that Mexican courts’ decisions on the matter are controlling. For the tribunal to say that neither of the parties suggested Mexican courts’ decisions are controlling is either to misunderstand Mexico’s claim or to disregard it. Second, the tribunal notes that these proceedings are not final. This, however, confuses the issue. The question of whether to defer to Mexican court decisions on matters of Mexican law is not one of whether to apply decisions from proceedings that are not final. Rather, it is a question of whether to wait until those decisions are final before deferring to them as a means of assessing and applying the law. Therefore, it is ironic that the tribunal offers this explanation because in the previous section of the award the tribunal itself made the decision that it did not have to wait for local actions to be completed before ruling on questions of Mexican law. The tribunal’s third rationale, that deference is unwarranted because the tribunal cannot resolve apparent conflicts in the various rulings of 34

Id. at ¶ 84.

On this view Mexico is also correct that to the extent that Mexican courts’ decisions determine the law as to whether CEMSA has a right to rebates, Feldman’s claims based upon that law, which is still being determined in cases sub judice, are not yet ripe. 35

36

Countermemorial, supra note 30, at ¶ 575.


Mexican courts in separate proceedings, suffers from the same confusion.37 It is not a reason to disregard Mexican court decisions altogether, but rather a reason to wait until the courts resolve these conflicts by applying the law they are still in the process of determining. The tribunal’s fourth justification, which it claims is “probably the most important,” is actually the most misleading.38 As in its decision on local remedies, the tribunal again states that “Mexican courts are applying Mexican law while this Tribunal must apply the provisions of NAFTA and international law, which do not necessarily provide the same results as under Mexican law.”39 While in general it is true that international law and Mexican law do not provide the same answers to legal questions, this is false when, as is the case here, international law stipulates that Mexican law is controlling. The tribunal’s final comment on the issue is also unpersuasive. The tribunal asserts that “the Claimant has not challenged any of the Mexican court decisions, even those unfavorable to the Claimant, as breaching the international law standard for denial of justice, and it is premature to consider any question of possible non-compliance of a Mexican court decision by the Respondent, since the issue of compliance has not yet arisen.”40 The reader presumes this statement is in reference to Feldman’s claim that the denial of rebates to CEMSA violated the 1993 Amparo decision, because in Feldman’s view the decision’s language on the equality of taxpayers meant that exporters such as CEMSA had a right to rebates even though they might be unable to provide IEPS receipts. This statement seems to indicate that because it is bound to apply international law, the tribunal cannot examine Mexico’s alleged non-compliance with court decisions unless Feldman challenges those actions as a denial of justice. While it is true that Feldman might challenge alleged non-compliance by Mexico on denial of justice grounds, it does not follow that this is the only situation in which the tribunal should examine Mexican compliance with its courts’ decisions. Here yet again the tribunal seems to disregard the courts’ role as adjudicators of what the law is in favor of their role as adjudicators of how the law should be applied to various sets of facts. But courts play both roles. Thus, if it is the 37


38

Id.

39

Id.

40

Id.


case that the law as determined by the courts calls for rebates to be issued notwithstanding the non-submission of receipts, the tribunal can examine whether the denial of those rebates due to non-submission of receipts amounted to an expropriation as an improper denial of a valid claim to monies. The result of these considerations is the tribunal’s decision that because “the validity or recovery of these tax rebates functionally have an impact on, and belong to, the Tribunal’s evaluation whether a ‘creeping’ or any other relevant . . . form of expropriation has taken place,” it does have the authority to issue declaratory relief relating to the tax assessment.41 In addition, the tribunal declares that “any decision by this Arbitral Tribunal [regarding CEMSA’s entitlement to tax rebates] is bound to have, under the terms of NAFTA Article 1136(1), a direct bearing upon any domestic litigation (pending or final) on the entitlement to tax rebates.”42 Thus, on the tribunal’s view, we are left not only with the bizarre possibility that the tribunal will reach a different conclusion than Mexican courts on an issue of what rights parties have under Mexican law, but also with the rather perverse proposition that the tribunal’s decision would trump any contradictory determination by Mexican courts of what Mexican law is. D.

Summary of Jurisdiction Considerations

The issue of whether Marvin Feldman had an investment in accordance with NAFTA was not explicitly raised in either the pleadings or the award sections on jurisdiction. As became clear in the discussion of the tribunal’s decision on local remedies, this is at least in part because the tribunal failed to defer to Mexican courts as to whether Feldman had an investment in accordance with Mexican law. Thus, not only has the tribunal at times failed to provide competent reasons for its decisions, but in so failing it has neglected to recognize what is at stake in CEMSA’s right to receive rebates under Mexican law, that is, the very existence of an investment by Feldman that qualifies as an investment under NAFTA Article 1139.

41

Id. at ¶ 88.

42

Id.


IV. THE TRIBUNAL’S DECISIONS ON EXPROPRIATION A.

Views of the Parties and Applicable Law

As the tribunal notes, Feldman’s “key contention is that the various actions of Mexican authorities, particularly SHCP, in denying the IEPS rebates on cigarette exports to CEMSA, resulted in an indirect or ‘creeping’ expropriation of the Claimant’s investment and were tantamount to expropriation under Article 1110.”43 Feldman makes several related allegations, including that Mexico’s actions were arbitrary, confiscatory, and discriminatory, and that Mexico’s non-compliance with the 1993 Amparo decision was a denial of justice. As already discussed in the section on estoppel, Feldman interprets the 1993 Amparo decision to require rebates for exports even if exporters cannot provide receipts separately proving the IEPS taxes paid. Mexico denies this, asserting that the decision guarantees resellers a 0 percent tax rate but does not reach the issue of whether rebates must be provided despite the non-submission of receipts. Mexico therefore argues that the receipts requirement of the IEPS law remains in effect despite the Amparo decision and adamantly denies ever reaching an agreement with Feldman to the contrary. Furthermore, Mexico argues that Feldman cannot even demonstrate ownership of an investment that would give him standing to bring an action under Chapter 11.44 The tribunal considers this latter claim by Mexico in its section on applicable law. Unfortunately, the tribunal gives the issue rather short shrift, stating only the following: A threshold question is whether there is an ‘investment’ covered by NAFTA. The term investment is defined in Article 1139, in exceedingly broad terms. It covers almost every type of financial interest, direct or indirect, except certain claims to money. The first listed item under ‘investment’ is an ‘enterprise.’ There is no disagreement among the parties that Corporación de Exportaciones Mexicanas, S.A. (CEMSA) is a corporate entity organized under 43

Id. at ¶ 89.

44 Although this essay has argued that the question of whether Feldman had an investment in accordance with NAFTA underlies many of the tribunal’s jurisdictional decisions, it is only in the tribunal’s discussion of Feldman’s expropriation claim that this issue is specifically considered. This is perhaps due to the fact that only here does Mexico explicitly challenge Feldman’s claim to have an investment.


the laws of Mexico, essentially wholly owned by the American investor Marvin Roy Feldman Karpa. Among the dictionary definitions of ‘enterprise’ are ‘a unit of economic organization or activity; esp. a business organization’ (Webster’s New Collegiate Dictionary, 1977 ed.). As such, the Tribunal determines that CEMSA comes within the term ‘enterprise’ and thus is an investment under NAFTA.45 This highly formalistic analysis understands the “threshold” in a way that almost any purported investor would be able to pass. Indeed, it requires little more than filing paperwork with whatever government agency oversees the formation of legal entities in a host state. Nonetheless, even under the tribunal’s liberal understanding of what constitutes an investment, CEMSA still may not qualify. For example, even if CEMSA is a registered legal entity in Mexico, it does not follow that this company is an “economic organization” consistent with the stated definition of “enterprise” or that it is engaged in economic (or any) activity at all. Many companies are created as so-called “shelf companies” and have no existence other than on paper. It is hard to imagine that such companies could be considered investments under a treaty meant to encourage trade and investment on the grounds that such activity can be beneficial for member states and their citizens. Moreover, it is unclear that such a company would be an investment if it turned out not to be permitted to conduct the kind of business or economic activity it was created to conduct. Thus, the tribunal’s reasons for declaring CEMSA an investment are largely unsatisfying. B.

The Tribunal’s Four Arguments Regarding Feldman’s Expropriation Claims

The tribunal is correct to point out that expropriation under NAFTA includes both direct and indirect expropriation, the latter including both creeping expropriations and measures tantamount to expropriation. It is also correct when it states that it is difficult to draw the line between valid governmental regulatory activity and indirect expropriations. Because CEMSA was not actually taken over by the Mexican government, Feldman’s claims fall under the category of indirect expropriation or measures “tantamount to expropriation.” As the tribunal states, “[b]y their very nature, tax measures, even if they are designed to have the effect of expropriation, will be indirect, with an effect that may be tantamount to expropriation.”46 45


46

Id. at ¶ 101.


The tribunal’s savvy discussion of the various types of expropriation, however, only heightens one’s disappointment in its analysis of whether an expropriation has taken place. For this analysis seems to neglect this very framework the tribunal has limned. Instead, the tribunal renders its decision that no expropriation has occurred by offering several inapposite comments: This Tribunal’s rationale for declining to find a violation of Article 1110 can be summarized as follows: (1) As Azinian suggests, not every business problem experienced by a foreign investor is an expropriation under Article 1110; (2) NAFTA and principles of customary international law do not require a state to permit “gray market” exports of cigarettes; (3) at no relevant time has the IEPS law, as written, afforded Mexican cigarette resellers such as CEMSA a ‘right’ to export cigarettes . . . ; and (4) the Claimant’s investment, the exporting business known as CEMSA . . . remains under the complete control of the Claimant. . . . While none of these factors alone is necessarily conclusive, in the Tribunal’s view taken together they tip the expropriation/regulation balance away from a finding of expropriation.47 Although they may be intended to respond to certain arguments made by the parties, three of these four points do almost no explanatory work at all. After examining how these points are developed in more detail by the tribunal, it will become evident that the first two points leave the question of indirect expropriation unanswered, and the fourth point addresses a question that is irrelevant to this case. 1.

The Tribunal’s First Expropriation Argument

The tribunal expands upon the first point in a section entitled “Many Business Problems are Not Expropriations.” Here, it twice repeats the truism cited above: 1.

2.

“[T]he Tribunal is aware that not every business problem experienced by a foreign investor is an indirect or creeping expropriation under Article 1110.”48 “To paraphrase Azinian, not all governmental activity that makes it difficult of impossible for an investor to carry out

47

Id. at ¶ 111.

48

Id. at ¶ 112.


a particular business, change in the law or change in the application of existing laws that makes it uneconomical to continue a particular business, is an expropriation under Article 1110.”49 Repeating this point, however, does not answer the question the tribunal is charged with answering—whether in this instance the business problem is in fact an expropriation. It is merely to state the context in which the question is being asked and to describe why the question can be difficult to answer. Thus, it does not advance the inquiry. Rather, it essentially repeats a premise, one that the tribunal itself has already called attention to in its previous discussion of applicable law by stating the following: Recognizing direct expropriation is relatively easy; governmental authorities take over a mine or a factory, depriving an investor of all meaningful benefits of ownership and control. However, it is much less clear when governmental action that interferes with broadly-defined property rights—an investment under NAFTA, Article 1139—crosses the line from valid regulation to a compensable taking, and it is fair to say that no one has come up with a fully satisfactory means of drawing this line.50 It is indeed difficult to draw the line, and perhaps no one has put forth a convincing method of doing so that can be employed in all cases of this sort. However, in each case in which an indirect expropriation is alleged, drawing this line with respect to the facts of that case is nonetheless a requirement, and stating the difficulty of fulfilling this requirement is not an excuse for avoiding it. Yet this is what the tribunal does, skipping from this description of the background of its task to its conclusion that no expropriation has taken place without indication of how it moved from the former to the latter. The only arguably explanatory comment in this section is the tribunal’s statement that “[t]he IEPS law on its face (although not necessarily as applied) is undeniably a measure of general taxation of the kind envisaged by Restatement [of the Law Third, the Foreign Relations of the United States] Comment g.”51 The tribunal here refers to the Restatement’s 49

Id.

50

Id. at ¶ 100.

51

Id. at ¶ 113.


effort to distinguish between indirect expropriation and valid governmental regulation. The tribunal had previously cited this comment in its discussion of applicable laws, adding its own emphasis as follows: A state is responsible as for an expropriation of property under Subsection (1) when it subjects alien property to taxation, regulation, or other action that is confiscatory, or that prevents, unreasonably interferes with, or unduly delays, effective enjoyment of an alien’s property or its removal from the state’s territory. . . . A state is not responsible for loss of property or for other economic disadvantage resulting from bona fide general taxation, regulation, forfeiture for crime, or other action of the kind that is commonly accepted as within the police power of states, if it is not discriminatory.52 The tribunal’s claim that “the IEPS is a measure of general taxation of the kind envisaged by Restatement Comment g” might somewhat illuminate the grounds for its decision that there was no expropriation, except for the fact that the manner in which this claim is phrased is self-undermining. For when the tribunal adds its parenthetical comment, “(although not necessarily as applied),”53 it raises precisely the question of whether the manner in which Mexico applied this measure is in fact acceptable in accordance with Comment g. Having just acknowledged in the same paragraph that Feldman “has been treated in a less than reasonable manner,”54 the tribunal’s parenthetical note suggests that what might otherwise have been the enforcement of a “bona fide general taxation” was perhaps in this instance a means of discrimination, thus derogating from the non-discriminatory condition noted in the Restatement, and in turn, raising again the question of whether the taxation measure was in fact bona fide regulation or an expropriation. The tribunal rests on the fact that the measure is a means of taxation, but eludes analysis of the remaining, difficult question—whether the manner in which that taxation was implemented made it expropriatory. The tribunal’s final comments in this section are un-illuminating as well. In fact, they primarily serve to distract the reader from what should be the tribunal’s central concern, explaining its decision on expropriation. The tribunal’s assertion is that Feldman 52

Id. at ¶ 105.

53

Id. at ¶ 113.

54

Id.


could have availed himself early on of the procedures available under Mexican law to obtain a formal, binding ruling on the invoice issue from SHCP, but apparently chose not to do so. Despite the legal uncertainties of the issues upon which the success of the business depended, the Claimant asked for clarification of the legal issues under Article 4 of the IEPS law only when effectively forced to do so, in April 1998 after SHCP denied the Claimant’s request for tax rebates for the October 1997–January 1998 exports, and in March 1999 when as a result of a tax audit SHCP demanded return of rebates, plus interest, inflation adjustment and penalties, for rebates received in 1996 and 1997.55 These remarks about what could have happened offer the tribunal no guidance in its task of assessing what did happen nor do they justify that assessment in any way. To lament that Feldman did not take an action that might have obviated the tribunal’s need to undertake the task of explanation that it here fails to carry out is neither to explain nor support the tribunal’s conclusion. In other words, that Feldman did not make it easier for the tribunal by asking Mexican authorities to address this issue does not mean that the tribunal is not now required to do so. This claim is also heavily ironic due to the fact that the tribunal has twice rejected giving deference to decisions in contemporaneous actions in Mexican courts that were in the very process of ruling on the matter. Indeed, these proceedings, which the tribunal refused to consider (as already discussed in the jurisdiction section), are authoritative over the formal rulings of administrators which the tribunal suggests Feldman could have sought to obtain in order to decide the issue. 2.

The Tribunal’s Second Expropriation Argument

As an explanation for its decision on expropriation, the tribunal’s observation that neither NAFTA nor international law requires a state to permit “grey market” exports is as much of a non-starter as repeating the truism that not all business problems are expropriations. Again, the fact that the law does not require certain types of exports to be permitted suggests nothing about whether those exports are in fact permitted nor about whether Feldman’s company is or is not an authorized exporter. Readers of the tribunal’s award are left with two unsupported suggestions: (1) that Feldman was not in fact a permitted exporter; and (2) that since a state can limit who exports from it, regulation of these exporters 55

Id. at ¶ 114, citations omitted.


can take place in any manner the state deems appropriate. What is missing from this line of reasoning, of course, is a perspicuous explanation of why CEMSA necessarily operates in the “grey market,” and is thus not a permitted exporter, and an analysis of why the restrictions imposed on CEMSA (including the manner in which they were imposed) were not violations of NAFTA. Instead, the tribunal cites a letter by the U.S. Competent Authority that, if anything, calls into question, rather than justifies, the conclusion the tribunal reaches here. In accordance with NAFTA Article 2103(6), the U.S. official had approved some of Feldman’s expropriation claims for arbitration and rejected another. The official declared that while Mexico’s 1998 law limiting rebate availability to the “first sale” in Mexico was not an expropriation, Feldman’s claims about Mexico’s previous actions, its denial of rebates, and alleged non-compliance with the 1993 Amparo decision could proceed to arbitration to determine if they amounted to an expropriation. Because the tribunal’s reasoning with regard to this letter and its implications is so flawed, it is worth citing paragraph 116 of the award in its entirety: The conclusion that neither NAFTA nor rules of customary international law require a state to permit gray market exports is to some extent reinforced by the determination of the U.S. Competent authority that Mexico’s action in enacting legislation effective January 1, 1998, which restricted the availability of rebates of excise taxes to those who purchase cigarettes in the “first sale” within Mexico (i.e., the sale from the producer to the producer’s customer, but not any subsequent resales) was not an expropriation under Article 1110 of NAFTA. The effect of this 1998 IEPS amendment had exactly the same objective as the 1991 IEPS amendment that denied resellers the availability of the zero percent tax rate for their exports. (This was the 1991 IEPS amendment that was held unconstitutional in the Amparo decision by the Mexican Supreme Court in 1993.) The U.S. Competent Authority letter attempts to de-link the 1998 measure to the earlier measures by stating that “No inference should be drawn concerning my views of the views of the United States government regarding whether the first two measures described above [the alleged refusal of Mexico to implement the Amparo decision and its refusal to provide IEPS rebates] is an expropriation under Article 1110 of the NAFTA,” but the comparison is inescapable. At a minimum, it suggests that tax law and policy


changes are intended to be given relatively broad leeway under NAFTA, even if their effect is to make it impractical for certain business activities to continue.56 Here, to support its claim, the tribunal cites a letter that explicitly rejects the kind of inference the tribunal makes based upon it. The tribunal claims that this inference is “inescapable,” but this is so only to the extent that it does recognize why the inference should not be made. Indeed, the tribunal suggests that because the 1991 and 1998 IEPS laws had essentially the same objective, if the second law does not amount to an expropriation, then neither should the first. But here again, the tribunal makes the same mistake it made in earlier sections of the award—conflating the permissibility of what a regulation attempts to do with the permissibility of the manner in which the regulation does it. Moreover, even if the letter from the U.S. Competent Authority could be read to stand for the principle that “tax law and policy changes are intended to be given relatively broad leeway under NAFTA,” the tribunal would still have to offer reasons for why Mexico’s regulations of CEMSA are within that leeway. It does not do this. Instead, it once again merely cites a broad principle, presuming that the facts of this case can be seen to fall within that principle without any specific argument to that effect. 3.

The Tribunal’s Third Expropriation Argument

The closest the tribunal comes to offering a sound argument for its decision that no expropriation took place is in the discussion of the third factor in support of that decision. In short, the claim is that because CEMSA never had a “right” under Mexican law to receive rebates for exported cigarettes, the denial of rebates cannot constitute an expropriation. As noted, in order to obtain rebates of the IEPS tax paid in the purchase price of goods, Article 4 of the IEPS law required exporters to submit receipts separately indicating the IEPS tax paid to SHCP. Thus, to actually benefit from the 0 percent tax rate for exports, exporters who purchased the goods they exported had to comply with Article 4. However, because it purchased cigarettes from big box retailers in Mexico, the IEPS tax those retailers paid in their original purchase of the cigarettes from producers was not included on CEMSA’s receipts. Nor was CEMSA able to convince the producers to provide it with this information. According to Feldman, this is because the producers wanted to 56

Id. at ¶ 116.


maintain a monopoly on exports of cigarettes from Mexico. In any event, CEMSA was never able to obtain the proper receipts and was thus unable to comply with Article 4 of the IEPS law. Therefore, according to the tribunal, the claimant never really possessed a “right” to obtain tax rebates upon exportation of cigarettes but only a right to the 0 percent tax rate. This is important, because as far as the tribunal can determine, “the only significant asset of the investment, the enterprise known as CEMSA, is its alleged right to receive IEPS tax rebates upon exportation of cigarettes, and to profit from that business.”57 Feldman argues that a right to the 0 percent tax rate is meaningless if CEMSA is unable to obtain the rebates. Thus, he interprets the 1993 Amparo decision as meaning that submission of receipts is not required. According to Feldman, SHCP’s denial of rebates to CEMSA based on Article 4 unlawfully limits the scope of the Amparo decision and is therefore both an expropriation of his investment and a denial of justice for failure to implement a court ruling. The tribunal considers Feldman’s interpretation of this decision but ultimately is unconvinced by it: There is language in the opinion that condemns discrimination between producers and other sellers generally, which is not limited to the 0% tax rate. Also, there is some inherent logic behind the Claimant’s position; if the Claimant were correct, this would be a strong argument for finding a creeping expropriation or denial of justice. If the Amparo resolves only the 0% tax rate, but the Claimant cannot satisfy the other requirements of the IEPS law, including Article 4 regarding invoices, there is no possibility of CEMSA’s benefiting from that decision with regard to cigarette exports, as the company is still prevented from carrying on its cigarette export business. 121. The problem for the Claimant is that a careful reading of the Amparo Supreme Court decision reveals no mention of Article 4; the discussion is confined solely to the availability of the 0% tax rate under Article 2 of the IEPS law to resellers as well as producers, and to a general assessment of the unconstitutionality of discrimination . . . There is no indication in the opinion that the Supreme Court intended to abrogate or modify this critical provision of the IEPS law, since it apparently did

57

Id. at ¶ 118.


not even consider the issue, and the Tribunal has no way of guessing what the result would have been had the Article 4 issue been squarely presented to the Supreme Court.58 Given these considerations, the tribunal goes on to conclude that “the court decision did not resolve the claimant’s problems with obtaining tax rebates on cigarette exports because the claimant failed to challenge Article 4 of the IEPS law.”59 While it is true that the tribunal would have no way of guessing what the Mexico Supreme Court would have said had the Article 4 issue explicitly been raised, it is also true that this very Article 4 issue was then currently being examined by Mexican courts. Thus, as noted above, the tribunal could have waited to defer to the Mexican courts’ rulings on the matter. Instead it chose not to wait and rendered its own interpretation of the Amparo opinion. On this interpretation, compliance with Article 4 was still required in order to obtain rebates, and thus, according to the tribunal, CEMSA never had an actual right to the rebates. Therefore, says the tribunal, SHCP’s denial of rebates to CEMSA was not an expropriation. This brings to light a key observation of this essay: If the reason Feldman’s investment was not expropriated was because he never had a right to the rebates he was denied, it would have made more sense for the tribunal to have declared as a matter of jurisdiction that Feldman did not have an investment in accordance with NAFTA Article 1139. As has been argued above, the tribunal’s ruling that he did have an investment is problematic, for it is based merely on the fact that he incorporated a company called CEMSA. The tribunal decided this company was an “enterprise” in accordance with Article 1139, even though it also acknowledges that “the only significant asset of the investment, the enterprise known as CEMSA, is its alleged right to receive IEPS tax rebates upon exportation of cigarettes, and to profit from that business.”60 Thus, the tribunal would have it that a company with no assets and no right to conduct the only business it was intended to conduct nonetheless qualifies as an investment under NAFTA.

58

Id. at ¶¶ 120–121.

59

Id. at ¶ 122.

60

Id. at ¶ 118.


Feldman’s claim of expropriation, however, is not only dependent on the tribunal’s interpretation of the Amparo decision. In addition to his claim on the basis of his interpretation of that decision, Feldman also claims that he had reached a specific agreement with SHCP officials regarding this decision and that they had given him assurances that CEMSA’s request for rebates would be honored despite its inability to provide the invoices. This was discussed above, as was the fact that Mexico adamantly denies such an agreement was made. The tribunal acknowledges that if such an agreement had been made, Mexico’s failure to comply with it “could be evidence of a denial of due process or fair and equitable treatment and support a conclusion that the IEPS law was intentionally being administered in a manner designed to destroy CEMSA’s export operations.”61 The tribunal also acknowledges that SHCP offered some assurance to Feldman regarding CEMSA’s right to rebates after the Amparo decision. However, according to the tribunal, it is unclear whether these were assurances that CEMSA could receive rebates without submitting the receipts. Although there is some evidence that Feldman reached some sort of agreement with SHCP, the tribunal asserts, this evidence is minimal and offers no indication of the scope or contents of that purported agreement. The tribunal goes on to declare that it “does not consider the invoicing requirements to be a mere formality or patently unreasonable, to be waived by officials based on their discretion.”62 Taken together, these comments indicate that the tribunal misinterprets Feldman’s claim. He does not claim that his agreement was contrary to the law or required waiving certain provisions of the law on the basis of someone’s discretion. His claim is that he and SHCP agreed that the Supreme Court’s Amparo decision rendered a certain provision of the law, Article 4 of IEPS, invalid and unenforceable. Thus, it seems clear that the tribunal’s own interpretation of the Amparo decision as not reaching the invoices issue not only leads it to reject Feldman’s expropriation claim, but also colors its interpretation of the claim itself. This misinterpretation lessens the burden of the tribunal to such an extent that one wonders if perhaps the tribunal is dissembling somewhat. For on this view, the tribunal is able to avoid a crucial and difficult ques61

Id. at ¶ 125.

62

Id. at ¶ 129.


tion, namely, whether SHCP did concur with Feldman’s interpretation of the Amparo decision and assured him that CEMSA would receive rebates regardless of whether it provided receipts. This, in turn, raises the question of whether the IEPS law, as the tribunal phrased it, “was intentionally being administered in a manner designed to destroy CEMSA’s export operations.”63 Despite a clear pattern of SHCP practice consistent with Feldman’s claims and at least some evidence of an agreement, the tribunal’s chosen standard of evidence enables it to rather easily resolve this central issue in the case. The tribunal also gives little consideration to whether Mexico’s waffling enforcement of the Article 4 receipts requirement created a legitimate expectation on the part of CEMSA that Article 4 was no longer to be enforced. If so, such sudden enforcement of this IEPS provision after 16 months (June 1996–September 1997) of non-enforcement might be, in the tribunal’s words, “so arbitrary as to constitute expropriatory action.”64 The tribunal notes that it “has some sympathy with the Claimant’s position here,” but goes on to assert that a reasonable person, given the complex and exacting nature of tax laws and regulations, and the ambiguity of statements by and correspondence with SHCP officials, should have sought expert tax counsel if it was not already available to him. Had this occurred, the Tribunal doubts than [sic] any competent tax attorney would have confirmed the Claimant’s right to rebates in the absence of proper invoices showing the tax amounts separately, given the text of Article 4 of the IEPS law and the lack of apparent legal authority on part [sic] of SHCP officials to waive this requirement.65 Here again the tribunal represents Feldman’s claim about an agreement with SHCP as one that required tax officials to waive provisions of the law, and here again, the tribunal makes the bizarre argument that part of what justifies its finding of no expropriation was Feldman’s failure to take actions that might have obviated its need to make such a finding at all. In short, the tribunal misrepresents the claims it is assessing and then does a poor job of justifying its assessment; or rather, it presents an irrel63

Id. at ¶ 125.

64

Id. at ¶ 132.

65

Id.


evant set of facts about what might have happened instead of offering a justification for its determination of what did not happen, that is, an expropriation. The tribunal continues by stating that “[w]hile the transparency in some of the actions of SHCP may be questioned, it is doubtful that lack of transparency alone rises to the level of violation of NAFTA and international law.”66 This may be true, but it is also true that lack of transparency alone is not the allegation. Indeed, the tribunal itself acknowledges that at various times SHCP’s treatment of CEMSA was also “arbitrary,” “less than reasonable,” “inconsistent,” and discriminatory.67 Thus, the issue is not whether lack of transparency alone is a violation but whether these patterns of behavior as considered in the aggregate constitute a violation. Indeed, the Restatement—which the tribunal itself cites in its section on the applicable law of expropriation—notes that the presence of these can tip the balance from valid regulation to actions tantamount to expropriation.68 Instead of making this assessment, the tribunal once again asserts that the Claimant would have been wise to seek a formal administrative ruling on the applicability of Article 4 of the IEPS, and court review if the ruling were adverse . . . Regardless of the results of the ruling process the Claimant would have been better off. If he had received a favorable ruling on Article 4, it would have been much easier for him to defend his rights under Mexican law and before this Tribunal. If he had lost, he could have at least avoided the uncertainties of his alleged right to rebates during much of the 1992–1997 period, and could have brought a NAFTA claim under Chapter 11 much earlier.69 From Feldman’s perspective, however, reaching an agreement with SHCP as to what the Amparo decision meant was precisely such an effort to avoid uncertainties. Because the tribunal does not assess Feldman’s agreement claim for what it was alleged to be, that is, an agreement as to 66

Id. at ¶ 133.

67

Id. at ¶¶ 143, 113, 109 and 188, respectively.

68

See supra note 45.

69



the interpretation of the Amparo decision and not an agreement to waive still valid laws, it does not acknowledge this. Instead, by shifting the focus to what Feldman might have done, the tribunal explains less about what Feldman could have avoided than it reveals about what the tribunal is presently avoiding, a well-reasoned explanation of its denial of Feldman’s claim. 4.

The Tribunal’s Fourth Expropriation Argument

The tribunal’s fourth argument regarding Feldman’s expropriation claim is that no expropriation took place because “the regulatory action has not deprived the Claimant of control of his company, CEMSA, interfered directly in the internal operations of CEMSA or displaced the Claimant as the controlling shareholder.”70 These facts may be true. However, they are not relevant to Feldman’s claim. Whether the ownership or leadership of a company has been usurped would be a fitting inquiry for examination of claims of direct expropriation, but as the Feldman tribunal itself indicates, “There is in this case no allegation of a direct expropriation or taking under Article 1110.”71 The purpose of allowing for claims of indirect expropriation is to acknowledge that an investment may be improperly interfered with by means other than those traditionally associated with direct expropriation. The tribunal also makes this point in its discussion of indirect expropriation, stating that “the ways in which governmental authorities may force a company out of business, or significantly reduce the economic benefits of its business, are many.”72 These means do not necessarily involve taking physical or operational control of a company. As the tribunal notes, “In the past, confiscatory taxation, denial of access to infrastructure or necessary raw materials, imposition of unreasonable regulatory regimes, among others, have been considered to be expropriatory actions.”73 The expropriation claim very clearly centers on tax measures, and the tribunal notes that “[b]y their very nature tax measures, even if they are designed to and have the effect of expropriation, will be indirect, with an effect that may be tantamount to expropriation.”74 Thus, the tri70

Id. at ¶ 142.

71

Id. at ¶ 108.

72

Id. at ¶ 100.

73

Id.

74

Id. at ¶ 101.


bunal seems to acknowledge that inquiring whether the claimant is still in control of his company may offer no insight for claims of indirect expropriation. So why does the tribunal, which clearly seems to appreciate what is at stake in claims of indirect expropriation, offer this comment as a final thought on its expropriation analysis? A possible answer is found in the section that follows: the tribunal seems eager for its analysis to correspond to the frameworks of analysis employed in previous NAFTA arbitrations involving expropriation claims. Indeed, the tribunal adds a separate section at the end of its expropriation analysis to discuss the consistency among its decision and several previous NAFTA cases. The tribunal first cites Metalclad,75 primarily as a case in which government assurances to an investor were better documented than those assurances Feldman alleges receiving from SHCP. However, the tribunal also notes that part of what distinguishes Metalclad from this case is that “Metalclad was deprived of all beneficial use of its ‘property’ despite the fact that in this case there has been no such deprivation alleged.”76 The tribunal continues this line of reasoning in its references to other cases. It asserts that in S.D. Myers,77 “the [Meyers] tribunal noted that expropriation normally constitutes a taking of property, with a view towards transfer of ownership, a situation that did not occur in this case or that one.”78 The tribunal then quotes from the Pope and Talbot decision, which stated that in determining “whether a particular interference with business activities amounts to an expropriation, the test is whether that interference is sufficiently restrictive to support a conclusion that the property has been taken from its owner.”79 In each of these previous NAFTA cases, the tribunals’ standard seems to contemplate a physical taking of tangible property or a forcible transfer of the ownership or control of a company. However, since no such actions were even alleged in Feldman, the tribunal’s noting the absence of such actions does not provide reasons for its decision on those allega75 Metalclad Corp. v. United Mexican States, ICSID Case No. ARB(AF)/97/1 (2000). 76


77

S.D. Myers, Inc. v. Canada, 40 I.L.M. 1408 (2001).

78


Id. at ¶ 151 (quoting Pope & Talbot Inc. v. Government of Canada, Interim Award (June 26, 2000), 40 I.L.M. 258, at ¶ 81). 79


tions. If the tribunal had intended to apply a criterion for expropriation that required the physical taking of tangible property, it could have made its decision at a much earlier stage, given that it made a point of stating that “the only significant asset of the investment, the enterprise known as CEMSA, is its alleged right to receive IEPS tax rebates upon exportation of cigarettes, and to profit from that business.”80 As the tribunal notes, CEMSA hardly had anything that could have been taken. Thus, unless the case should have been dismissed at the jurisdictional stage for Feldman’s failure to state a claim a valid claim, or a claim upon which relief could be granted, reference to this taking-of-property criterion used in previous NAFTA arbitrations is an unhelpful distraction. V.

THE TRIBUNAL’S ANALYSIS OF FELDMAN’S NATIONAL TREATMENT CLAIM

A.

Introduction NAFTA Article 1102(2) on national treatment reads as follows: Each Party shall accord to investments of investors of another Party treatment no less favorable than it accords, in like circumstances, to investments of its own investors with respect to the establishment, acquisition, expansion, management, conduct, operation, and sale or other disposition of investments.81

Feldman claims that Mexico discriminated against CEMSA by affording it less favorable treatment than it afforded domestic companies in like circumstances. However, as already noted, examinations of alleged violations of NAFTA with respect to Feldman’s investment should first consider whether Feldman even had an investment. As discussed in the sections on jurisdiction and expropriation, if Feldman never had a legal right to the rebates, then the best approach for the tribunal would have been to reject Feldman’s claims on the ground that he never had an investment.82 Here, however, the tribunal, does not reexamine whether 80

Id. at ¶ 118.

As the tribunal notes, “Article 1102(1) is the same except that it refers to ‘investors’ rather than to ‘investments of investors;’ under Article 1102(3), the obligation applies to state/provincial governments as well, but this is not relevant here” (Award, supra note 1, at ¶ 165). 81

82 Although neither explicitly claims that Feldman did not have an investment, both Mexico and the dissenting tribunal member make similar arguments. Mexico argues that its law disallows SHCP officials to provide rebates to exporters without the proper receipts. Therefore, “it would be highly inappropriate for the Tribunal to find


Feldman had an investment in accordance with NAFTA. Instead, it proceeds to examine Feldman’s discrimination claim on the presumption that CEMSA does qualify as an investment. Feldman’s discrimination claim is that Mexico violated NAFTA Article 1102 by doing three things: (1) allowing domestic companies to receive rebates for exports without receipts in the 1998–2000 period, a period after CEMSA’s rebates had been denied and its cigarette exporting operations shut down; (2) neither denying rebates to, nor attempting to recoup rebates from, domestic companies for the periods of 1996–2000; and (3) denying CEMSA registration as an export trading company when it had approved domestic companies. Although Mexico argues that a discrimination claimant must show that the discrimination alleged took place because of the claimant’s status as a foreign national, the tribunal rejects this assertion. Because this is a case of de facto discrimination, the issue is not whether discrimination took place because of Feldman’s nationality but “whether rebates have in fact been provided for domestically owned cigarette exporters while denied to a foreign re-seller, CEMSA.”83 Thus, as a matter of law, claimants have only to demonstrate the existence of, not the reasons for, discrimination perpetrated against them. Despite certain insufficiencies of evidence and reasoning that will be described, the tribunal nonetheless concluded that Mexico violated its national treatment obligations under NAFTA Article 1102 by discriminating against CEMSA. According to the tribunal, rebates were paid to domestic companies in periods when CEMSA was denied rebates. In addition, the tribunal determined that Mexico made efforts to recoup rebates given to CEMSA when it did not make the same efforts with regard to the domestic companies. Finally, the tribunal asserts that no domestic reseller/exporter was denied export registration in the manner CEMSA was. Thus, the tribunal basically accepted each of Feldman’s discrimination claims. But as this essay argues, neither the evidence that a violation of national treatment based on the failure of SHCP to provide a benefit which they had no authority under Mexican law to provide” (Award, supra note 1, at ¶ 160). The dissenting tribunal member, in turn, insists that Feldman’s claims of discrimination should be analyzed in light of the fact that Feldman’s business was “based on premises that clearly violate Mexican laws: to obtain rebates from the Government without being entitled to them” (Dissent, supra note 6, at 2). 83



Feldman presents to advance his claims nor the reasoning the tribunal uses to justify its decisions are sufficient to support these conclusions. B.

Domestic Companies in “Like Circumstances” as CEMSA

The problems pertaining to evidence arise from the very start. As noted, finding a violation of NAFTA’s Article 1102 on national treatment requires demonstrating that a foreign investor was improperly discriminated against relative to domestic investors in “like circumstances.” In this case, however, the tribunal notes that “only a limited amount of relevant factual information has been presented” in regard to other companies that might have been in like circumstances with CEMSA.84 According to the tribunal, “neither party suggests that there are any other foreign owned reseller/exporters other than the Claimant.”85 The record indicates that there may have been five to ten other companies registered to export cigarettes from Mexico, but it is unclear who owned these companies and whether they were able to submit receipts specifying payments of IEPS taxes. The tribunal cites a Mexican official, Enrique Díaz Guzman, as confirming “that at least three trading companies (i.e., not producers) received IEPS rebates for cigarette exports at various times between September 1996 and May 2000.”86 However, there is little indication of the activities of these companies and no indication of whether they were foreign or domestically owned. Feldman’s discrimination claim focuses on CEMSA’s treatment as compared to the so-called Poblano Group companies, Mercados I and Mercados II, owned by Caesar Poblano. Feldman’s claim is that like CEMSA, the two Poblano Group companies were unable to purchase cigarettes from domestic producers and had to buy them from volume retailers such as Walmart and Sam’s Club. This means that like CEMSA these exporters were also unable to submit to SHCP the receipts required by Article 4 of the IEPS law to obtain rebates. Thus, Feldman’s discrimination allegation rests on his claims that CEMSA was treated differently than the Poblano Group companies with regard to payments of rebates, efforts to recoup rebates paid, and permission to export. However, it is important to note that Feldman presented no evidence as to the business activities of the Poblano Group companies—neither evidence of cigarette 84

Id. at ¶ 167.

85

Id.

86

Id.


exports, rebate payments, nor audits. He did, however, present evidence that Mexico was conducting a personal audit of Cesar Poblano, which he submits as evidence that Mexico began an audit of Mr. Poblano’s business activities only long after it targeted his own. Mexico responds by making various related claims. First, it argues that because Poblano’s owner lent money to Feldman for CEMSA, CEMSA and the Poblano Group are essentially the same investment and therefore cannot be compared for the purposes of assessing Feldman’s Article 1102 allegation. Second, Mexico claims that the Poblano Group is in fact the subject of an ongoing audit with respect to tax issues and may as a result be asked to return rebates it was given as well. Hence, it would be inappropriate to find a violation of Article 1102 simply because CEMSA happened to be audited first. Mexico notes that its tax system is set up to address all improprieties within a period of five years and must be given some latitude from case to case. Third, according to Mexico, CEMSA was denied export registration because it was being audited at the time it tried to register. Because the Poblano Group was not being audited when it registered as an exporter, Mexico argues it was not in like circumstances with CEMSA in regard to that issue. The tribunal makes clear that “the ‘universe’ of firms in like circumstances [with CEMSA] are those foreign-owned and domestic-owned firms that are in the business of reselling/exporting cigarettes.”87 Based on this principle, it decides that those firms “for the purposes of this case are CEMSA and the corporate members of the Poblano Group.”88 Thus, despite allusions in Mr. Guzman’s testimony to other companies that may have exported cigarettes, received rebates, or been audited, the tribunal indicates that for this case only comparisons of CEMSA to the Poblano Group companies will be valid. As we shall see, however, the tribunal waffles from this principle, thereby greatly undermining its basis of decision making. C.

The Substance of the Discrimination Claim

As mentioned above, the tribunal finds discrimination on the grounds that CEMSA was denied rebates when the Poblano Group was not and was ordered to repay rebates when the Poblano Group was not. However, the tribunal arrives at this conclusion only after making a 87

Id. at ¶ 171.

88

Id. at ¶ 172.


number of questionable moves. In the tribunal’s discussion of audits, we find several of them. Unfortunately, these questionable moves later inform the tribunal’s further analysis of discrimination. 1.

Discrimination Based on Failure to (Timely) Audit

The tribunal begins its analysis of Feldman’s discrimination claims by stating that “even if Mexico is auditing Mr. Poblano, this process was begun long after the audit of CEMSA,” and “there is no documentation that the audit continued after approximately March 2000, or that it even involved IEPS rebates.”89 At that point, the tribunal asserts, CEMSA’s rebates had already been denied, and despite Mexico’s claims to the contrary, the tribunal concludes that no such action has been taken with regard to the Poblano Group. The tribunal therefore finds that CEMSA was audited well before any other domestic reseller/exporters and suggests that this alone is prima facie evidence of discrimination.90 Deciphering the tribunal’s argument here is difficult, but it basically runs as follows: CEMSA has been audited. Because Mexico has only claimed that audits of companies in like circumstances with CEMSA are ongoing, but has not actually presented documentary evidence of such audits, we can presume that no such audits have taken place or did not take place until recently. Because no such audits have taken place, or at least have not taken place until recently, the early audit of CEMSA is “itself evidence of discrimination.”91 This argument is problematic for three reasons: First, there is no reason to assume the audit of CEMSA was improper or illegitimate and thus that auditing CEMSA was necessarily discriminatory if the Poblano companies were not audited. Second, the tribunal did not adequately address Feldman’s evidence, nor did it properly consider Mexico’s nonsubmission of documents as an Article 2105 issue. Third, because the tribunal mishandled the evidentiary issues, it also misapplied the burden of proof and unfairly concluded that, at a minimum, the Poblano Group companies were not audited in a reasonable time frame and that perhaps they were not audited at all. These points should be examined individually.

89

Id. at ¶ 174.

90

Id., citations omitted.

91

Id.


If there were evidence that CEMSA was specifically targeted as the subject of an unwarranted audit, perhaps this would be evidence of discrimination. However, there was no particular reason for the tribunal to conclude that the audit of CEMSA was illegitimate. Indeed, the tribunal itself emphasizes that Feldman’s rebate calculations had been grossly erroneous at times. Analyzing Feldman’s description of how he calculated the rebates he claimed for CEMSA and his assertion that this method was approved by an SHCP, the tribunal states that it “does not find the Claimant’s testimony on this issue to be credible.”92 The tribunal then adds: “It is inconceivable to the Tribunal that . . . [SCHP] would have given the Claimant or any other taxpayer carte blanche to over-estimate the amount of the rebates, in flagrant violation of the IEPS law.”93 Therefore, that CEMSA was audited for a period in which Feldman claimed and received rebates overstated by as much as 93 percent cannot be considered necessarily discriminatory or even unreasonable. On the contrary, if Mexico had not approved Feldman’s calculations as he alleges, an audit of his company for overstated rebate claims would have had to have been considered likely or even expected. Second, it is questionable for the tribunal to conclude, based merely upon a lack of documentation, that Mexico had not carried out audits of the Poblano Group. After all, Mexico insisted that all instances of improper rebate claims were in fact being investigated but that its privacy law prevented officials from revealing details of specific audits. NAFTA Article 2105 seems to support Mexico’s position here. Article 2105 clearly grants state parties leave from providing the kinds of information the tribunal laments Mexico has not provided, that is, “information the disclosure of which would impede law enforcement or would be contrary to the Party’s law protecting personal privacy or the financial affairs and accounts of individual customers of financial institutions.” This suggests that Mexico was not required to disclose which companies or persons it is auditing nor the issues or amounts in controversy in an audit. That Mexico did not explicitly cite this article as the reason for the limited extent of the information it provided should not matter. It remains the responsibility of the tribunal to conduct the arbitration in accordance with NAFTA provisions and to call for submissions from

92

Id. at ¶ 131.

93

Id.


the parties with regard to the scope of articles such as Article 2105 if necessary.94 Third, a few comments about the burden of proof in such cases will help illuminate the flaws in the tribunal’s position. The generally accepted standard is that if a claimant presents a prima facie case for his claim, the burden of proof is then shifted to the respondent. But here it is unclear how, if at all, Feldman, made a prima facie case for discrimination. Instead, he has lamented his company’s being audited and merely suggested, without presenting any concrete evidence, that the Poblano Group companies have not been similarly audited.95 In fact, Feldman did not even present evidence that the Poblano Group companies have sought or received rebates for cigarette exports. The tribunal nonetheless totally neglects Feldman’s lack of evidence and carries on as if he has somehow made his case by simply making his accusation: “If Mexican authorities are auditing or intend to audit other taxpayers who are in like circumstances with CEMSA, the Government of Mexico, as the only party with access to such information, has not been particularly forthcoming in presenting the necessary evidence.”96 However, it is critical to understand that just because one party may be the only one with access to certain information does not mean that that party therefore has the burden of presenting it. In the comment above, the tribunal is equivocating with respect to two different rubrics for what is “necessary evidence.” First, there is the evidence that it is nec94 Mexico did make reference to this article in its request for interpretation of the award and in its effort to have the tribunal’s decision annulled in Canadian courts. The tribunal refused to offer an interpretation, claiming that reference to Article 2105 meant that Mexico “effectively is seeking a new decision.” All three tribunal members agreed on this point. Two levels of Canadian courts also rejected Mexico’s effort to have the decision annulled on the grounds that the tribunal’s inferences were in violation of Article 2105. In this author’s view, these courts were wrong to defer to the tribunal’s decision on interpretation of the award, as they did not fully consider that Mexico’s claim was that the tribunal had not followed NAFTA provisions, a claim the tribunal would obviously be likely to deny. Thus, the tribunal was wrong both not to apply Article 2105 and to reject Mexico’s request for interpretation; and the Canadian courts failed to understand what was at stake in Mexico’s claim and why deference to the tribunal (on whether the tribunal itself fulfilled its duty to employ Article 2105) was inappropriate. 95 As noted above, however, even if there were clear proof that CEMSA had been audited and the Poblano Group companies had not, this would still not necessarily mean that CEMSA had been discriminated against. 96



essary for Feldman to present in order to make a successful prima facie claim. Second, there is the evidence Mexico must necessarily provide in its defense if the prima facie case has been made. These are two different categories that should not be conflated. Thus, the tribunal’s suggestion that it is Mexico’s responsibility to provide information about which companies it is auditing in order for the tribunal to assess Feldman’s bald discrimination allegation improperly reverses the burden of proof.97 A burden of proof means precisely that it is not simply the party with the best information that must present it; on the contrary, a burden of proof means that certain procedural steps must be completed before a party with information can be compelled to provide it. Here, the tribunal has erred by requiring little, if any, evidence from Feldman before shifting the burden to Mexico. Having shifted the burden to Mexico, the tribunal then errs twice more: first, by not recognizing Mexico’s refusal to submit records as an Article 2105 issue; and second, because of this, failing to rule properly on what evidence NAFTA requires Mexico to produce. As a final note, even if one assumed the tribunal’s bizarre inferential reasoning is correct, that is, that any audit of the Poblano Group by Mexico necessarily began much later than its audit of CEMSA, the tribunal’s assessment that such timing is itself discriminatory leaves one to wonder what discretion a host state retains in deciding whether and when to audit taxpayers. The tribunal offers no standard for appraising whether the timing of an audit is discriminatory but simply states that in this case it seems to have been so, thereby rejecting Mexico’s own suggestion that a five-year period is reasonable for tax authorities to investigate recent claims. 2.

Discrimination Based on Denial of Export Registration

Next, the tribunal goes on to suggest that CEMSA was discriminated against when it was denied export registration as well. According to the tribunal, an unsigned memorandum which reasonably could have been generated only in SHCP indicates that registration was being denied on the basis of the audit of the Claimant’s rebate pay97 As will be evident below, the tribunal’s conflation of these two distinct requirements leads it to bungle the burden of proof on other claims as well.


ments. There is no evidence that any domestic reseller/exporter has been denied export privileges in this manner.98 In Mexico’s view, the reason no exporter/reseller was denied registration in “this manner” is simply because no exporter/reseller was being audited at the time it tried to register. By claiming to find discrimination, the tribunal is thus rejecting Mexico’s argument that the companies were not in like circumstances with respect to being qualified for export registry. Clearly, the tribunal does not view CEMSA being audited as a legitimate reason for denying its registration as an exporter. However, this denial of CEMSA’s export registration cannot be an independent ground for finding discrimination because the tribunal has already stated that the audit is itself evidence of discrimination. Thus, the argument with regard to export registration turns in large part upon the tribunal’s flawed views about the audit. This only highlights that fact that, once again, the tribunal seems to pass judgment on what discretion tax officials retain to make these kinds of administrative rulings. And once again, the tribunal indicates that officials have exercised that discretion discriminatorily without offering any standard for what is discriminatory. 3.

Discrimination Based on Denial/Recouping of Rebates

Having discussed the timing of the alleged audits and the denial of CEMSA’s export registration, the tribunal then shifts its attention to the core points of the dispute regarding discrimination. These are whether the Poblano Group companies were given tax rebates in periods in which CEMSA was denied rebates and whether, if they were, SHCP attempted to recoup these improper rebates by an audit or other means, as it did with CEMSA. Hence, the tribunal’s view of the audit of CEMSA as itself discriminatory largely informs its considerations here. The tribunal acknowledges that “the extent of the evidence of discrimination on the record is admittedly limited.”99 In fact, the tribunal admits that “there are only a few documents in the record bearing directly on the existence of differing treatment, particularly the statement of Mr. Diaz Guzman, the ‘mystery’ memorandum from SHCP’s files, and the tax registration statement for Mercados Regionales, owned by the Poblano Group.”100 Hence, reaching a finding of discrimination 98


99

Id. at ¶ 176.

100

Id.


on this evidence may seem unlikely. The tribunal even states that “[o]ne member of this Tribunal believes that this evidence on the record is insufficient to prove discrimination.”101 But the tribunal is undeterred. By waffling from its decision on companies in like circumstances with CEMSA, by continuing to adhere to its mistaken view of the burden of proof, and by subsequently inferring a number of facts based upon that misapplied burden, the tribunal claims to justify its finding that discrimination against CEMSA took place: The majority’s view is based first on the conclusion that the burden of proof was shifted from the claimant to the respondent, with the respondent then failing to meet its new burden, and on assessment of the record as a whole. But it is also based on a very simple two-pronged conclusion, as neither point was ever effectively challenged by the respondent: a. No cigarette reseller-exporter (the Claimant, Poblano Group member or otherwise) could legally have qualified for the IEPS rebates, since none under the facts established in this case would have been able to obtain the necessary invoices stating the tax amounts separately. b. The Claimant was denied the rebates at a time when at least three other companies in like circumstances, i.e. resellers and exporters apparently including at least two members of the Poblano Group, were granted them.102 Here the tribunal is quite clearly equivocating. The majority view is not, as it suggests, based on two separate grounds—its conclusion that the burden of proof was shifted and “also” its “very simple, two-pronged conclusion.” On the contrary, the tribunal’s acceptance of the two-pronged conclusion is merely its justification for shifting the burden of proof in the first place. Thus, the two suggested grounds are in fact separate parts of the same single view—if one accepts that companies in like circumstances with CEMSA received rebates when CEMSA did not, then Mexico 101

Id.

Id., citations omitted. It is unclear why, in articulating the burden of proof for this investor-state investment dispute, the tribunal cites a standard from the World Trade Organization (WTO), a state-to-state trade settlement body. However, the basic point about shifting the burden of proof once a prima facie case has been made is well taken. Whether the prima facie case has in fact been made here, however, this author disputes. 102


inherits the burden to show there was no discriminatory treatment. But has Feldman shown that other companies in like circumstances received rebated when CEMSA did not? This question is what needs to be analyzed in detail. Unfortunately, the tribunal almost entirely neglects this issue of what Feldman has proved. Instead, it merely asserts that “the Claimant in our view has established a presumption and a prima facie case that the Claimant has been treated in a different and less favorable manner than several Mexican owned cigarette resellers.”103 Thus, with little explanation the tribunal has moved rather quickly from viewing the evidence of discrimination as “admittedly limited” to understanding it as “sufficient to raise a presumption that what is claimed is true.” In both the two-pronged conclusion and the cursory remarks above it, however, the tribunal reveals a major error in its reasoning. In both instances the tribunal derogates from its earlier decision that for the purposes of this case, only CEMSA and the Poblano Group companies are to be considered in like circumstances. In the two-pronged conclusion, the tribunal mentions that CEMSA was denied rebates “at a time when at least three other companies in like circumstances . . . were granted them.” In the statement above this, the tribunal also refers to CEMSA’s being treated less favorably than “several Mexican owned cigarette resellers.” Therefore, in both cases, the tribunal is clearly considering evidence that pertains to companies other than just CEMSA and the Poblano Group. If it were clear that companies other than the Poblano Group were in like circumstances with CEMSA, these other companies should have been considered as such all along. However, despite their being alluded to in the statement of Enrique Díaz Guzman, the similarity of circumstances of these other companies to CEMSA was not at all clear. Thus, why Mr. Guzman’s statement should be included as one of the three documents mentioned in support of a discrimination finding is also unclear. Moreover, if the tribunal was prepared to accept at face value the comments of Mr. Guzman in support of Feldman’s allegations, it would likely also have to accept Mr. Guzman’s claim that these other companies were also being audited. Of course, this would impugn the discrimination argument altogether.

103

Id.


The tribunal’s waffling with respect to which companies were in like circumstances with CEMSA illuminates its lack of rigor in assessing what should count as evidence of discrimination. If the tribunal was not even confident that certain companies were in like circumstances with CEMSA, it should not have considered testimony about how those companies were treated as evidence in support of Feldman’s prima facie case. The tribunal is trying to have it both ways. As it stands, however, vague testimony about companies that may or may not have been in like circumstances with CEMSA cannot alone support Feldman’s prima facie case for discrimination. Instead, the tribunal’s merely undermines the rationale for its decision by bringing in evidence that should have been excluded as not clearly pertaining to the universe of companies in like circumstances. Thus, once again, as in its discussion of expropriation, the tribunal’s effort to consider the “record as a whole” is essentially a “fudge factor” that serves to obfuscate, not clarify, its reasoning. The tribunal then elaborates on the issue of the prima facie case and the burden of proof by stating that how Mexico responded to Feldman’s allegations indicates that the allegations must have been correct: In weighing the evidence, including the record of the five day hearing, the majority is also affected by the Respondent’s approach to the issue of discrimination.104 If the Respondent had had available to it evidence showing that the Poblano Group companies had not been treated in a more favorable fashion than CEMSA with regard to receiving IEPS rebates, it has never been explained why it was not introduced. Instead, the Respondent spent a substantial amount of its time during the hearing and in its memorials seeking (unsuccessfully in the Tribunal’s view) to demonstrate that CEMSA and the Poblano Group were related companies (as there could be no discrimination, presumably within a single company group). Yet, if the Poblano Group firms had not received the rebates, that evidence of relationship would have been totally irrelevant. Why would any rational party have taken this approach at the hearing and in the briefs if it had information in its possession that would have shown that the Mexican owned cigarette exporters were being treated in the same manner as the Claimant, that is, denied IEPS rebates for cigarette 104 Here the tribunal says “also,” as if this explanation were in addition to the explanation just offered, rather than merely another poorly reasoned justification in support of the first explanation.


exports where proper invoices were not available? Thus, it is entirely reasonable for the majority of this Tribunal to make an inference based on the Respondent’s failure to present evidence on the discrimination issue.105 This reasoning is atrocious. Again, the whole point of a burden of proof is that a party that does not have the burden does not have to introduce certain evidence or explain why it has not introduced it until after the burden is shifted, and its actions until such time cannot be interpreted as a reason to shift the burden. Therefore, the tribunal’s asking why Mexico did not introduce certain evidence is just not appropriate. Even if NAFTA Article 2105 is totally disregarded, it would be enough to presume that Mexico wanted Feldman to bear the burden of proving his case. Second, the fact that a party that does not bear the burden of proof has introduced certain evidence does not on that ground make it “entirely reasonable” for the tribunal to speculate upon the reasons why that evidence was introduced. Nor does it allow the tribunal to speculate about why other evidence was not introduced. The fact that Mexico claims the Poblano Group and CEMSA should not be treated as separate companies for purposes of Article 1102 analysis in no way gives the tribunal reason to infer that the two companies must have been treated differently. It would be inappropriate on any logical or legal analysis to claim that a party that has a burden of proof has made a prima facie case for shifting that burden, not because of the evidence that the burdened party has adduced as required, but because of facts allegedly inferable from the limited evidence the other party introduced. To do this is not only to misapply the burden of proof, but to do so ironically, by shifting even the burden of making the prima facie case to the party that would inherit the burden of proof if that prima facie case had been made. Here, the tribunal is simply making an end-run around what is lawfully required. It knows Mexico has better information than Feldman on certain issues such as taxpayers and audits, so it dissembles in an effort to find a way to place the burden upon Mexico to provide that information. However, this is incorrect. The law requires Feldman to make his case before the burden can be shifted, regardless of what information Mexico has that might make Feldman’s prima facie case easier for him to prove.

105



If the party with the best evidence were always compelled to provide it, there would never be any need for a burden of proof. In addition, because enforcement of any law, including tax laws, can be difficult for all states, it would seem rather formalistic for a tribunal to claim violation of national treatment under NAFTA due to isolated instances of underenforcement that states later made efforts to rectify. For example, it may be the case that in some months CEMSA obtained exports rebates without receipts, and Poblano Group members did not. In other months, the reverse situation might have applied. Such discrepancies might go occasionally unnoticed by SHCP or may be addressed in audits not yet carried out. Therefore, to make an accurate assessment of whether a certain company has been discriminated against in violation of NAFTA, one might expect a tribunal to examine broader patterns of treatment than the Feldman tribunal indicates are necessary. However, the discussion above suggests that the evidentiary record is not sufficient to examine these broader patterns of enforcement.106 Although the tribunal claims that Mexico “never effectively challenged” the claim that Poblano Group members received rebates when CEMSA did not, the tribunal’s reasoning by inference seems to demonstrate that, on the contrary, such a claim was never effectively made in the first place.107 The tribunal thus avoids confronting the problem of insufficient information to properly assess discrimination claims under NAFTA. For example, the tribunal would need specific information covering extended periods of time regarding whether Mexico audited other companies or attempted to recoup rebates paid to them. Instead of acknowledging that such a case cannot be made here, the tribunal simply lowers the bar for making a discrimination claim. Just as it lowered the burden of what Feldman had to show in order to allow the case to go forward, 106 According to the dissent, the claimant is not only unable to prove different treatment over significant periods, but is unable to demonstrate different treatment at any time. On the dissent’s view, “the Claimant cannot even prove one single element showing the treatment given by the Mexican Government to the Poblano Group with regard to its acceptance or denial of tax rebates.” The tribunal’s decision is therefore based on its chain of inferences and not on the documentary record from which “it cannot be established what kind of treatment has been accorded to the Poblano Group, not even if the Poblano Group has obtained any tax rebate at all” (Dissent, supra note 6, at 6). 107



the tribunal now lessens the burdens of what it has to consider in order to find discrimination—any, even an isolated, instance of unequal treatment by a state of domestic and foreign companies in like circumstances. Although the tribunal does not succeed at conclusively demonstrating even an isolated instance of unequal treatment, its claim to have done so sets the bar quite low for prospective claimants. Almost anything short of perfect enforcement of a state’s law might be found to be a discriminatory violation of NAFTA. This examination of the discrimination claims in Feldman demonstrates that the appropriate burden of proof can be difficult for claimants to fulfill. This is especially true in tax cases where claimants might not have access to how other taxpayers were treated, even though the host state likely does have those records. However, the very notion of a burden or proof is premised on the likelihood of an asymmetry of information between parties to a dispute. Thus, burdens of proof are not meant to be disregarded or taken lightly simply on the grounds that such an asymmetry is found to exist. Rather, they are meant to prevent bodies like the Feldman tribunal from using strings of inferences to move them from claims like this—“The extent of evidence of discrimination on the record is admittedly limited”—to claims like this—“the differing treatment of CEMSA and the Poblano Group is obvious.”108 The burden of proof means precisely that it is for the claimant to present evidence and suggest the adequate logical links. Here, the claimant simply did not do so. VI. CONCLUSION This essay draws attention to numerous flaws in the reasoning of the Feldman award, from the use of equivocal language and apocryphal principles to inapposite arguments and fallacious logic. In the section on estoppel (Section III.A) it questions where the tribunal derived the principles of international law it appeals to. It also questions the tribunal’s efforts to scrutinize the parties’ claims and its efforts to show how the standards used fit the facts of the case. In the section on exhaustion of local remedies (Section III.B), it criticizes the tribunal’s disposition toward state court decisions, suggesting that state courts should have sole authority over determining rights provided by state law. Moreover, this essay claims that NAFTA does not generally deviate from the exhaustion of local remedies rule of international law. Instead, 108

Id. at ¶¶ 176, 179, respectively.


it suggests that the rule can be waived only for claims under Article 1121(2)(b). Therefore, claims that need not be waived, such as Feldman’s action for declaratory judgment in a Mexican court, should still be exhausted before an international tribunal takes up the issue. In the discussion of expropriation (Section IV), this essay analyzes the four “factors” the tribunal claims justify its finding when “taken together” and finds that three of these are in fact non-starters, while only one factor offers guidance for how the expropriation decision should be made. In its discussion of the tribunal’s discrimination finding (Section V), this essay criticizes the tribunal’s handling of the burden of proof and shows how other considerations the tribunal presents as separate arguments were actually dependent upon this treatment of the burden of proof. Finally, this essay suggests at various points (Sections III.D., IV.A, and V.A.) that the tribunal fails to recognize a crucial dispositive element of this case: CEMSA does not qualify as an investment in accordance with NAFTA. Thus, the tribunal neglected the simplest means of resolving this dispute, stating that Feldman does not have standing to bring a claim under NAFTA.

CHAPTER 4

THE DECISION ON LIABILITY IN LG&E V. ARGENTINA Rocío Digón* I.

INTRODUCTION

This paper examines the adequacy of reasoning in the decision on liability in LG&E Energy Corp., LG&E Capital Corp., and LG&E International, Inc. v. Argentine Republic,1 one of a series of ICSID claims that has arisen out of Argentina’s economic collapse in 2001. I will first demonstrate how the tribunal’s reasoning on the fair and equitable treatment claim is well-reasoned under Article 48(3) of the ICSID Convention2 and will subsequently argue that inadequate reasons are provided with respect to three of the remaining four claims: arbitrary and discriminatory treatment, indirect expropriation, and the state of necessity. The inadequacies of reasoning in this award are not caused by an absence of reasons, as in several of the other awards reviewed in this volume, but instead, by a lack of internal coherence. In these instances, the reasons threshold of the MINE ad hoc committee—that the award allow the reader to “follow how the tribunal proceeded from Point A to Point B and eventually to its conclusion”3—is not met. Where reasons are offered, there is often a discrepancy between the sources the tribunal is * J.D. 2008 (expected), Yale Law School; LL.M., public international law, Leiden University, 2005. This paper was written as part of the Reasons Project, directed by Professors Michael Reisman and Guillermo Aguilar Alvarez, Yale Law School. For helpful comments, I thank W. Michael Reisman, Guillermo Aguilar Alvarez, and George Stephanov Georgiev. 1 LG&E Energy Corp., LG&E Capital Corp. and LG&E Int’l Inc. v. Argentine Republic, ICSID Case No. ARB/02/1, Decision on Liability (Oct. 3, 2006), 46 I.L.M. 40 (2007) [hereinafter LG&E—Liability]. 2 “The Award . . . shall contain . . . the decision of the Tribunal on every question submitted to it, together with the reasons upon which the decision is based.” Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (ICSID) (W. Bank), opened for signature Mar. 18, 1965, 575 U.N.T.S. 159, art. 48 [hereinafter ICSID Convention]. 3 MINE v. Guinea, Decision on Annulment (Dec. 22, 1989), 4 ICSID REP. 79, 88 (1997).

117


purportedly relying on and the actual reasons. Simply stated, there is a troubling gap between the tribunal’s analysis and the tribunal’s conclusions in three of the four claims I will evaluate. This review of an award’s internal coherence does not reach (nor does it aspire to) a level of appellate review; there is an assessment of the consistency within each individual section and the overall award and not a determination of whether the tribunal’s findings were correct or incorrect.4 I am not suggesting that there are no coherent reasons that could possibly support the tribunal’s decisions—which may have been the right ones—but only that the tribunal failed to provide these reasons on several occasions. This paper attempts to point out the critical gaps in the tribunal’s reasoning and outline the dangerous consequences of incoherent reasoning, including its potential to be seen as a failure to comply with the ICSID Convention. II.

THE PROCEDURAL AND FACTUAL BACKGROUND OF THE DISPUTE

A.

Procedural Background

Claimants5 LG&E filed their request for arbitration against the respondent, the Argentine Republic, on December 28, 2001. Claimants supplemented their request on January 24, 2002, to account for further violations to the U.S.-Argentine Bilateral Investment Treaty (BIT) resulting from Argentina’s enactment of the Emergency Law. This ICSID tribunal issued a Decision of the Arbitral Tribunal on Objections to Jurisdiction on April 30, 2004, concluding that ICSID had jurisdiction

4 This is consistent with the AMCO v. Indonesia committee’s interpretation of the reasons requirement, as quoted in the introduction to this volume: “If it be true that a full control and review of the reasoning followed by an ICSID tribunal would transform an annulment proceeding into an ordinary appeal, it is also true that supporting reasons must be more than a matter of nomenclature and must constitute an appropriate foundation for the conclusions reached through such reasons. Stated a little differently, there must be a reasonable connection between the bases invoked by a tribunal and the conclusions reached by it. The phrase ‘sufficiently pertinent reasons’ appears to this ad hoc Committee to be a simple and useful clarification of the term ‘reasons’ used in the Convention.” Amco Asia Corporation and Others v. The Republic of Indonesia, Decision on Annulment (May 16, 1986), 1 ICSID REP. 509, 520–21 (1993). 5 “Claimants” includes three corporations: LG&E Energy Corp., LG&E Capital Corp., and LG&E International Inc. [hereinafter LG&E].

LG&E v. Argentina • 119

over the dispute because the claimant had jus standi as a foreign investor, the issue before the tribunal was a legal dispute, both parties consented to ICSID jurisdiction, and all of the other jurisdictional requirements pursuant to the ICSID Convention and the BIT had been fulfilled. As this paper will focus on the tribunal’s subsequent decision on liability, I will not go into further detail on the jurisdictional award. The tribunal addressed five claims in the liability phase of the proceedings: fair and equitable treatment, discriminatory and arbitrary treatment, the umbrella clause, considerations on indirect expropriation, and the state of necessity. The decision on liability, rendered on October 3, 2006, partially granted LG&E’s claims; the tribunal concluded that Argentina had breached the fair and equitable treatment standard, adopted discriminatory measures, and was liable under the umbrella clause, yet dismissed the expropriation claim and found no violation of the arbitrary treatment standard. Additionally, the tribunal concluded that Argentina was in a state of necessity from December 1, 2001, through August 26, 2003; as such, the respondent was exempt from the payment of any damages incurred during that period. B.

Factual Background

The factual background of this dispute is very similar to the other existing and pending ICSID claims against Argentina.6 LG&E is an American corporation with a shareholding interest in three gas distribution companies in Argentina operating under license after the privatization of Gas del Estado, the national natural-gas transport and distribution services monopoly: Distribuidora de Gas del Centro (Centro), Distribuidora de Gas Cuyana S.A. (Cuyana), and Gas Natural BAN S.A. (GasBan). Specifically, “[i]n February 1997, LG&E purchased a 45.9% interest in Centro and a 14.4% interest in Cuyana,” followed by a “19.6% interest in GasBan”7 in March 1999. The dispute between LG&E and Argentina arose from this shareholding interest.

6 See, e.g., CMS Gas Transmission Co. v. Argentine Republic, ICSID Case No. ARB/01/8, Award (May 12, 2005), 44 I.L.M. 1205 (2005); Sempra Energy Int’l v. Argentine Republic, ICSID Case No. ARB/02/16, Award (Sept. 28, 2007), available at http://www.investmentclaims.com/decisions/Sempra_Energy-Award.pdf; Houston Indus. Energy, Inc. and Others v. Argentine Republic, ICSID Case. No. ARB/98/1, Award (Aug. 24, 2001) (unpublished). 7

LG&E—Liability, supra note 1, at ¶ 52.


In the early 1990s, Argentina enacted three laws and one decree of particular relevance to this dispute: the State Reform Law, the Convertibility Law, the Gas Law, and Decree 1189/92. The State Reform Law called for the privatization or concession of large government-owned businesses, the Convertibility Law pegged the austral (and later the Argentine peso) to the U.S. dollar and banned price or valuation indexation, and the Gas Law created ENARGAS to oversee the natural-gas transport and distribution services industry, whose responsibilities included setting and adjusting maximum tariffs. Decree 1189/92 called for the privatization of Gas del Estado. The privatization plan sought to attract foreign investors and offered incentives such as the ability to purchase shares with guarantees, the power to grant licenses instead of concessions, and enhanced legal protections. LG&E entered into transfer agreements with the respondent based on these guarantees, which the Argentine government could ultimately not uphold. This failure to realize the guarantees formed the basis for this dispute. An economic crisis developed in Argentina during the late 1990s that prevented the government from implementing the terms of the various privatization laws. Furthermore, the government enacted the Emergency Law which converted peso valuation back to a floating exchange rate, abolished all clauses calling for tariff adjustments in U.S. dollars or other foreign currencies, and called for the renegotiation of private and public agreements to adapt them to the new exchange system. Decree 293/02 required the mandatory renegotiation of transport and distribution of natural gas licenses: “Centro, Cuyana, and GasBan have been involved in the renegotiation process under threat of rescission of contract.” As the conditions existing prior to the Emergency Law were not restored and no compensation for incurred losses was received, the claimants brought this ICSID claim. III. ASSESSMENT OF THE INDIVIDUAL CLAIMS IN THE DECISION ON LIABILITY A.

Fair and Equitable Treatment Standard

The tribunal’s reasoning in assessing whether Argentina implemented measures that violated its obligation of fair and equitable treatment to LG&E’s investment allows the reader to “follow how the Tribunal proceeded from point A to point B.”8 The internal logic of this section is evident in the tribunal’s presentation of the factual findings, clear ana8

MINE v. Guinea, supra note 3, at 88.


lytical framework, and application of the framework to the facts of the dispute. Notwithstanding, minor weaknesses in the tribunal’s analysis that do not undermine the adequacy of its overall reasoning will be highlighted in this section. First, the tribunal established four factual findings regarding the guarantees to investors set forth in the Gas Law and its implementing regulations that had been violated: (1) the calculation of tariffs in U.S. dollars before conversion into pesos; (2) the semi-annual adjustment of the tariffs according to the Producer Price Index (PPI); (3) a sufficient income to cover all costs and a reasonable rate of return provided by the tariffs; and (4) compensation for any freezing or price controls under the tariff system.9 These four violations of the Gas Law and its implementing regulations, set forth in “The Tribunal’s Conclusion Concerning Argentina’s Guarantees to Investors” ultimately provided the factual grounding for the tribunal’s conclusion that Argentina acted in an unfair and inequitable manner towards the claimants. After stating the factual underpinnings of the present dispute, the tribunal cogently presented the analytical framework and standard for interpretation required of fair and equitable treatment pursuant to the BIT. As the BIT does not expressly define fair and equitable treatment, the tribunal first examined the BIT’s preamble and then the decisions of prior ICSID tribunals, in accordance with Article 31 of the Vienna Convention on the Law of Treaties.10 The tribunal found that the parties 9


Vienna Convention on the Law of Treaties, 1155 U.N.T.S. 331 (1969). Article 31 states: 10

1. A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose. 2. The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes. . . . It is also interesting to note that preambular language has taken on an important role in the interpretation of bilateral and multilateral treaties. [A] teleological approach to interpreting bilateral or multilateral treaties should be based on the assumption that these treaties have been negotiated to facilitate and promote foreign investment, which is often reflected in the wording of the preambles. Thus, it might be concluded that, when in doubt, these treaties should be interpreted in favorem investor, stressing and expanding his rights so as to promote the flow of foreign investment.


entered into the BIT in order to accomplish three objectives relevant to fair and equitable treatment: (1) “maintain a stable framework for investment and maximum effective use of economic resources;” (2) “promote greater economic cooperation;” (3) stimulate the flow of private capital and the economic development of the parties.”11 From this BIT language, the tribunal derived the following broad standard of fair and equitable treatment: “the stability of the legal and business framework is an essential element of fair and equitable treatment in this case, provided that they do not pose any danger for the existence of the host State itself.”12 Thus far, the reader’s expectations of reasoning have been satisfied: the tribunal has established a standard from the language of the BIT. Next, the tribunal reviewed prior ICSID and North American Free Trade Agreement (NAFTA) tribunal decisions in order to present two more specific components of fair and equitable treatment that have evolved as internationally recognized standards: investor expectations and transparency. Both of these components were first mentioned in Tecmed,13 and have since “been adopted by a succession of tribunals,”14 including CMS, Occidental, and Waste Management, Inc.15 The tribunal in this dispute adopted nearly identical language to that of the Tecmed tribunal when it found that: “the fair and equitable standard consists of the host State’s consistent and transparent behavior, free of ambiguity that involves the obligation to grant and maintain a stable and predictable legal framework necessary to fulfill the justified expectations of the foreign investor.”16 According to the tribunal, this interpretation is supported by the BIT and in the decisions of the aforementioned tribunals. The tribunal established the specific components of the broad Rudolf Dolzer, Indirect Expropriations: New Developments?, 11 N.Y.U. ENVTL. L.J. 64, 73 (2002). 11


12

Id. at ¶ 124.

Técnicas Medioambientales Tecmed, S.A. v. United Mexican States, ICSID Case No. ARB(AF)/00/2, 43 I.L.M. 133 (2004). 13

14


See CMS, supra note 6; Occidental Expl. & Prod. Co. v. The Republic of Ecuador, LCIA Case No. UN3467; Waste Management, Inc. v. United Mexican States, ICSID Case No. ARB(AF)/98/2, 40 I.L.M. 56 (2001). 15

16



framework of fair and equitable treatment by first considering the BIT provisions and subsequently adding more specific meaning through the evolution of jurisprudence constante. At this point, the tribunal could have strengthened its reasoning by explicitly indicating that the standard of fair and equitable treatment should be defined according to conventional treaty law, as set forth in the BIT. When the BIT language is broad, the arbitral tribunal has the discretion—and, in fact, the interpretive power—to elaborate the meaning of a standard. Finally, in order to satisfy the reader’s expectations of reasoning, the tribunal must apply the framework of fair and equitable treatment it has established to the particular facts of the dispute. These expectations are successfully met in the tribunal’s conclusion that Argentina violated the BIT provisions of fair and equitable treatment: Having created specific expectations among investors, Argentina was bound by its obligations concerning the investment guarantees vis-à-vis public utility licensees, and in particular, the gas-distribution licensees. The abrogation of these specific guarantees violates the stability and predictability underlying the standard of fair and equitable treatment . . . Specifically, it was unfair and inequitable to pass a law discarding the guarantee in Decree 1738/92 that the tariffs would be calculated in U.S. dollars and then converted into pesos . . . Argentina went too far by completely dismantling the very legal framework constructed to attract investors.17 As demonstrated in the quotation, the tribunal clearly indicated both the standard it was applying and the specific facts of LG&E’s claim to find that Argentina acted unfairly and inequitably. B.

Discriminatory and Arbitrary Treatment: Discrimination Against the Investor or Against an Entire Sector?

Had the tribunal faithfully applied its own reasoning, the tribunal should not have concluded that LG&E suffered from arbitrary and discriminatory treatment. But this was not the outcome of the tribunal’s analysis; it held that Argentina did treat LG&E in a discriminatory fashion. The tribunal provided both reasons and reasoning but did not adequately base its conclusions on this reasoning. Article II(2)(b) of the BIT 17

Id. at ¶¶ 134, 139.


provides that “[n]either Party shall in any way impair by arbitrary or discriminatory measures the management, operation, maintenance, use, enjoyment, acquisition, expansion, or disposal of investments.” In order to define “discriminatory treatment,” the tribunal relied on prior arbitral jurisprudence, which has developed beyond the non-discrimination principle in customary international law.18 The tribunal initially defined a discriminatory measure as one whose “intent is to discriminate . . . or if the measure has a discriminatory effect”19 and adopts the ELSI standard, which it summarizes as: “[I]n order to establish when a measure is discriminatory, there must be (i) an intentional treatment (ii) in favor of a national (iii) against a foreign investor, and (iv) that is not taken under similar circumstances against another national.”20 Under the ELSI 21 standard, a discriminatory effect is not necessary to find a violation of the non-discrimination principle; intent to discriminate based on nationality grounds is sufficient. The ELSI decision held: 18 In customary international law, non-discrimination is composed of two elements: “first, the measures directed against a particular party must be for reasons unrelated to the substance of the matter . . . second, discrimination entails like persons being treated in an inequivalent manner.” In other words, an action is discriminatory when: (1) there is no correlation between the measures taken against a party and the actual matter, and (2) the party receives treatment different from other like persons. The definition of “like persons” has not yet been codified in customary international law and requires a comparative analysis: “[discriminatory treatment] calls for more sophisticated treatment in order to identify unreasonable (or material) discrimination as distinct from the different treatment of non-comparable situations.”

When a state violates this principle of non-discrimination, responsibility is incurred. According to Brownlie, “where a state acts within what is prima facie a right, power or privilege, but there is evidence that the precise occasion or mode of exercise of the right . . . was based upon a selection contrary to the principle of non-discrimination, responsibility will arise on the ground of unlawful discrimination.” See A.F.M. Maniruzzaman, Expropriation of Alien Property and the Principle of NonDiscrimination in the International Law of Foreign Investment: An Overview, 8 J. TRANSNAT’L L. & POL’Y 57, 59 (1998); IAN BROWNLIE, PRINCIPLES OF PUBLIC INTERNATIONAL LAW 546–47 (6th ed. 2003); Borzu Sabahi. National Treatment—Is Discriminatory Intent Relevant?, in T. WEILER ED., INVESTMENT TREATY ARBITRATION: A DEBATE AND DISCUSSION at 1(2008). 19 LG&E—Liability, supra note 1, at ¶ 146, citing KENNETH J. VANDEVELDE, UNITED STATES INVESTMENT TREATIES: POLICY AND PRACTICE at 77 (1992). 20


Case Concerning Elettronica Sicula S.p.A (ELSI) (United States v. Italy), 1989 I.C.J. REP. 15. 21


The effect of the word particularly, introducing the clauses “(a)” and “(b),” suggests that the prohibition of arbitrary (and discriminatory) acts is not confined to those resulting situations described in “(a)” and “(b),” but is in effect a prohibition of such acts whether or not they produce such results. It is necessary, therefore, to examine whether the requisition was, or was not, an arbitrary or discriminatory act of itself. . . . It is common ground that the requisition order was not made because of the nationality of the shareholders. . . . There is no sufficient evidence before the Chamber to support the suggestion that there was a plan to favour IRI at the expense of ELSI, and the claim of “discriminatory measures” in the sense of Article I of the Supplementary Agreement must therefore be rejected.22 Thus, under this standard, a measure is discriminatory if the state intended to discriminate, even where no discriminatory effect is produced. At this point in the LG&E award, the reader would expect the tribunal to indicate whether the respondent’s actions against LG&E satisfy the conditions it has set forth for a finding of discriminatory treatment. Unfortunately, the tribunal did not coherently undertake this review. There was no discussion of evidence of “intentional treatment in favor of a national against a foreign investor.”23 The tribunal instead seemed to revise the phrase to “intentional treatment in favor of one industry sector against another industry sector.” The only evidence of discrimination offered distinguished between sectors, “Even though it was not proved that these measures had been adopted with the purpose of causing Claimants’ foreign investments damage, discrimination against gas distribution companies vis-à-vis other companies, such as water supply and electricity companies, is evident.”24 Why did the tribunal expressly contradict the ELSI standard in its conclusion without explanation? The discussion of variances across sectors, without discussion of nationality, is not germane under the ELSI standard and the tribunal made no effort to justify its decision for introducing this reason. At the very least, the tribunal had three possible options to render the reasoning in this section coherent: a responsibility to demonstrate 22

Id. at 72.

23


24

Id.


(1) that national companies were privileged over foreign investors such as LG&E regardless of industry sector, (2) that there was discrimination within the gas distribution sector, or (3) to make a clear statement, that the ELSI standard has to be read to include this broad application due to an evolving arbitral jurisprudence. Although none of these options is set forth in the award, any of these alternatives would have allowed for a more reasoned decision on this point. I will discuss options (2) and (3) in greater detail. If the tribunal were to assess discrimination in the gas distribution sector, it could have first set forth the composition of the gas distribution sector and subsequently signaled evidence of discrimination. The factual section of the award indicated that: The blocks of shares of Centro, GasBan, and Cuyana that were subject to privatization were awarded to private investors: 90% of the shares of Centro were awarded to a consortium formed by Società Italiana Per Il Gas S.p.A. (“Italgas”), an Italian company, and Sideco Americana S.A., an Argentine Company (“Sideco”); 70% of the shares of GasBan were awarded to a consortium formed by Gas Natural SdG S.A., a Spanish company, Compañía General de Combustibles S.A. and Manra S.A., both Argentine companies; and 60% of the shares of Cuyana were awarded to a consortium formed by Italgas and Sideco.25 Unfortunately, neither the Factual Background nor the Tribunal’s Analysis on Discriminatory and Arbitrary Treatment offered evidence of such discrimination in the gas distribution sector. It is important to note that no other tribunal—including the various NAFTA awards that have considerably developed this standard—has applied the LG&E standard of discriminatory treatment across sectors without consideration of whether the companies were national or foreign. Even the Occidental tribunal, which argued against a narrow interpretation of the “claim to no less favorable treatment,” based its finding of a violation of discriminatory treatment on the effect (and not the intent) of the respondent’s actions on foreign versus national companies: “in like situations” cannot be interpreted in the narrow sense advanced by Ecuador as the purpose of national treatment is to 25

Id. at ¶ 47.


protect investors as compared to local producers, and this cannot be done by addressing exclusively the sector in which that particular activity is undertaken. . . . In the present dispute the fact is that OEPC has received treatment less favorable than that accorded to national companies . . . [T]he result of the policy enacted and the interpretation followed by the SRI in fact has been a less favorable treatment of OEPC.26 The Occidental tribunal expanded the scope of “no less favorable treatment” to include cross-sector analysis in evaluating discrimination and did not undertake an analysis of Ecuador’s intent. NAFTA jurisprudence likewise accepts this expanded cross-sector analysis while maintaining the nationality requirement; again, this jurisprudence does not support the LG&E tribunal’s analysis and conclusion.27 The LG&E tribunal should 26

Occidental, supra note 15, at ¶¶ 173, 177.

The Feldman v. Mexico award best summarizes the various components of the non-discrimination standard, which have been clarified and developed individually in other awards. The Feldman tribunal provided: 27

the hurdles for 1102 are several. They include (a) which domestic investors, if any, are in “like circumstances” with the foreign investor; (b) whether there has been discrimination against foreign investors, either de jure or de facto; (c) the extent to which differential treatment must be demonstrated to be a result of the foreign investor’s nationality; and (d) whether a foreign investor must receive the most favorable treatment given to any domestic investor or just to some of them. Marvin Roy Feldman Karpa v. United Mexican States, ICSID Case No. ARB(AF)/99/1, ¶ 166 (Dec. 16, 2002), 42 I.L.M. 625 (2003). The Pope & Talbot award set forth a clear definition of “in like circumstances,” which emphasized the importance of a contextual analysis. The tribunal wrote, a formulation focusing on the like circumstances question, on the other hand, will require addressing any difference in treatment, demanding that it be justified by showing that it bears a reasonable relationship to rational policies not motivated by preference of domestic over foreignowned investments. That is, once a difference in treatment between domestic and a foreign-owned investment is discerned, the question becomes, are they in like circumstances? It is in answering that question that the issue of discrimination may arise. Pope & Talbot Inc. v. Canada, UNCITRAL (NAFTA), Award on the Merits of Phase 2, ¶ 78 (April 10, 2001). In S.D. Myers, the tribunal found that differences in treatment will presumptively violate Article 1102(2), unless they have a reasonable nexus to rational government policies that (1) do


have explained its reasons for further expanding the discriminatory treatment standard. Regrettably, this tribunal did not pursue any of these alternatives, leaving the reader dissatisfied and perplexed by the discussion of the discriminatory treatment standard. In paragraph 147, the tribunal indicated, While the Tribunal concludes that based on the evidence presented, Respondent treated the gas-distribution companies in a discriminatory manner, imposing stricter measures on the gas-distribution companies than other public utilities sectors, Claimants have however not proven that these measures targeted Claimants’ investments specifically as foreign investments.28 According to this quotation, the effect of the respondent’s measures was discriminatory. But were the tribunal to have applied the ELSI standard of intent to discriminate it purportedly adopted, there should have been no finding of discriminatory treatment. But the tribunal concluded otherwise. In paragraph 148, the tribunal stated “Even though it was not proved that these measures had been adopted with the purpose of causing Claimants’ foreign investments damage, discrimination against gas distribution companies vis-à-vis other companies, such as water supply and electricity companies, is evident.”29 As Sabahi aptly notes, On the one hand, it stated that in order to establish whether a measure is discriminatory, one has to give equal weight to both intent and effect; it also seems to indicate that proof of either of them would be sufficient for establishing that a measure has been discriminatory. . . . This swinging between intent and effect is especially evident. . . . The result, however, as stated, is a finding of discrimination solely based on the effect of the measure.30

not distinguish, on their face or de facto, between foreign-owned and domestic companies, and (2) do not otherwise unduly determine the investment liberalizing objectives of NAFTA. S.D. Myers Inc. v. Canada, UNCITRAL (NAFTA), First Partial Award, ¶ 161 (Nov. 13, 2000), 40 I.L.M. 1408. 28


29

Id. at ¶ 148.

30

Sabahi, supra note 18, at 22.


Moreover, in the final section of the award, Decisions of the Tribunal on Liability, the tribunal provided that “Argentina breached the standard of fair and equitable treatment, no less favorable treatment than that to be accorded under the international law, and adopted discriminatory measures, causing damage to LG&E.”31 From this language, it appears that the tribunal found the respondent liable for a violation of discriminatory treatment, without any requirement of intent. As this is the tribunal’s final statement on liability for discriminatory treatment, without reference to the standard applied, it must be concluded that the respondent discriminated against LG&E. The confusion stemming from the discriminatory treatment finding is evident in scholarly updates and investment law updates summarizing the tribunal’s conclusions in LG&E. There is disagreement between commentators as to whether the tribunal actually found a violation. The most troubling are those sources that have found that the tribunal did not find a violation of discriminatory treatment, The Investment Treaty Update of the Watson, Farley & Williams arbitration group states that “[T]he tribunal found that Argentina had violated the fair and equitable treatment provision and the umbrella clause, but not the clause prohibiting unlawful expropriation or arbitrary/discriminatory treatment.”32 Likewise, two different articles published in Transnational Dispute Management analyzing LG&E’s state of necessity argument provide that “Although other utilities sectors were not treated similarly, all gas-distribution companies were subject to the same unfavorable tariff regimes and the Claimants had not shown discrimination against their investments specifically. Therefore the claim of discriminatory treatment was rejected”33 and “The CMS and LG&E tribunal also concurred in their assessment that the Argentine measures, though unfair and inequitable, did not amount to arbitrary or discriminatory measures.”34 This confusion among commentators is troubling and serves as addi-

31


Watson, Farley & Williams, Investment Treaty Update at 2 (Winter 2006), available at http://www.wfw.com/pubs/WFW%20Investment%20Treaty%20Update%20Winter %202006.pdf. 32

33 Greg Falkof, “State of Necessity” Defence Accepted in LG&E v. Argentina ICSID Tribunal, 3(5) TRANSNAT’L DISPUTE MGMT. 7 (Dec. 2006). 34 August Reinisch, Necessity in International Investment Arbitration—An Unnecessary Split of Opinion in Recent ICSID Cases? Comments on CMS v. Argentina and LG&E v. Argentina, 3(5) TRANSNAT’L DISPUTE MGMT. 5 (Dec. 2006) (internal citations omitted).


tional evidence of the tribunal’s inadequate reasoning in the discriminatory and arbitrary treatment section of the award. In contrast, as the tribunal’s subsequent finding that Argentina did not adopt arbitrary measures appears to be satisfactory in terms of reasons, I will not address it in further detail. C.

Considerations on Indirect Expropriation

The reasoning supporting the tribunal’s conclusion that LG&E’s investment in the Argentine gas-distribution sector was not indirectly expropriated leaves the reader unable to proceed from Point A to Point B. This section’s inadequate reasoning is particularly egregious due to the absence of a “mechanical formula to determine when measures attributable to the host state breach the dividing line between legitimate regulation and compensable indirect expropriation”35 in international investment law. Instead, tribunals have often relied on prior awards for guidance36 and undertaken detailed factual and critical analyses in order to determine whether or not there has been an indirect expropriation. Paulsson describes a finding of indirect expropriation as a two-phase process: The first stage of the analysis should focus on the nature or magnitude of the interference to the investor’s property interests in its investment caused by measures attributable to the Host State to determine whether those acts amount to a taking. The second stage should determine whether this taking or interference rises to the level of an expropriation by reference to the relevant treaty standard. If this second stage results in a finding of expropriation, it remains for consideration whether it is nonetheless 35 Jan Paulsson & Zachary Douglas, Indirect Expropriation in Investment Treaty Arbitrations, in ARBITRATING FOREIGN INVESTMENT DISPUTES: PROCEDURAL AND SUBSTANTIVE LEGAL ASPECTS at 145 (NORBERT HORN ED., 2004). 36 “Just as the quest for the alchemic formula no longer attracts research funding, the search for the chimerical perfect rule on indirect expropriation need not preoccupy counsel and arbitrators. The more arduous but realistic approach suggested by Professor Christie is the way forward.” Id. at 147. Professor Christie stated that “It is evident that the question of what kind of interference short of outright expropriation constitutes a ‘taking’ under international law presents a situation where the common law method of case by case development is pre-eminently the best method, in fact probably the only method, of legal development.’” G.C. Christie, What Constitutes a Taking of Property under International Law?, 38 BRIT. Y.B. INT’L L. 307, 338 (1962) (quoted by Paulsson, id. at 146).


lawful because it is for a public purpose, on a non-discriminatory basis, in accordance with due process of law and on payment of fair compensation.37 The reader’s initial expectation that the LG&E tribunal would effectively address these two phases of analysis is unrealized as the tribunal ultimately engaged in a superficial analysis that further confounds the existing interpretations of indirect expropriation. Because there is no definition of indirect expropriation in the BIT, the tribunal accurately identifies its task as one of looking “to international law in determining the relevant criteria for evaluating this claim.”38 The subsequent discussion of these “relevant criteria” is confusing and reveals a litany of reasons that are largely irrelevant to a cogent determination of indirect expropriation. The tribunal’s first attempt at defining indirect expropriation occurred at paragraph 188: Their [creeping and de facto expropriation] common point rests in the fact that the host State’s actions or conduct do not involve “overt taking” but the takings occurs when governmental measures have “effectively neutralize[d] the benefit of property of the foreign owner.” Ownership or enjoyment can be said to be “neutralized” where a party no longer is in control of the investment, or where it cannot direct the day-to-day operations of the investment.39 This definition combines language from prior ICSID tribunal awards in CME and Pope & Talbot respectively.40 But a close reading of the combined language reveals a contradiction. In the first sentence, indirect expropriation is defined as the effective neutralization of an investor’s property due to a governmental measure. In the second sentence, neutralization is defined as loss of ownership or enjoyment over property caused by a loss of control in the investment or a loss of control over the day-to-day operations of the investment. This definition inevitably raises 37

Id. at 148.

38


39

Id. at ¶ 188.

40 CME Czech Republic B.V. (The Netherlands) vs. The Czech Republic, UNCITRAL, Final Award (Mar. 14, 2003), available at http://ita.law.uvic.ca/documents/ CME-2003-Final_001.pdf; Pope & Talbot, supra note 27.


the questions: Can there be a situation in which an investor still directs the day-to-day operations of its investment, even after a government has effectively neutralized its benefits? If so, does this constitute an indirect expropriation? Are these examples of neutralization the exclusive instances in which it can occur or are they the only ones relevant to the present case? Herein lies the first instance of incoherent reasoning in the tribunal’s definition of indirect expropriation. The tribunal set forth what it believed to constitute a taking but did not apply it to the facts of the LG&E dispute. Instead, the tribunal continued in its attempt to define the criteria necessary to establish indirect expropriation. This time, the tribunal established an effects test that balanced “the degree of the measure’s interference with the right of ownership and the power of the State to adopt its policies.”41 Specifically, the tribunal indicated that this “degree of interference” is determined through an evaluation of the measure’s economic impact against a reasonable investor’s expectations and the duration of the measure42 and that the severity of the economic impact determines whether there is a “need for compensation due to expropriation . . . The impact must be substantial in order that compensation may be claimed for the expropriation.”43 This effects test is problematic in several ways: (1) it combines the first and second stage of Paulsson’s indirect expropriation analysis without completing a thorough investigation of either stage independently, and (2) it suggests that there are instances of expropriation in which compensation is not required. With respect to the first issue, the tribunal “appear[ed] to confuse two distinct analytical steps for a finding of expropriation by conflating the questions as to whether there has been a taking attributable to the Host State and whether the Host State is under an obligation to compensate that taking.”44 With respect to the second issue, the tribunal should have provided reasons for expropriation without compensation and explicitly referred to the distinction between regulatory and expropriatory measures. The tribunal ignored the consequences of these conclusions and incorrectly referred to the Tecmed decision for support that only certain thresholds of expropriation should be compensated. According to this 41


42

Id. at ¶ 190.

43

Id. at ¶ 191.

44

Paulsson & Douglas, supra note 35, at 148.


tribunal, Tecmed set a high threshold for compensation from expropriation: “To establish whether the Resolution is a measure equivalent to an expropriation . . . it must be first determined if the Claimant, due to the Resolution, was radically deprived of the economical use and enjoyment of its investments, as if the rights related thereto—such as the income or benefits related to the [investment] had ceased to exist.”45 But if paragraph 115 of Tecmed is read in its entirety, it becomes clear that this threshold is a guide to distinguish between an expropriation and a regulation, and not between a compensable expropriation and a non-compensable expropriation, as the LG&E tribunal implied. The Tecmed tribunal stated, This determination is important because it is one of the main elements to distinguish, from the point of view of an international tribunal, between a regulatory measure, which is an ordinary expression of the exercise of the state’s police power that entails a decrease in assets or rights, and a de facto expropriation that deprives those assets and rights of any real substance. Upon determining the degree to which the investor is deprived of its goods or rights, whether such deprivation should be compensated and whether it amounts or not to a de facto expropriation is also determined.46 According to Tecmed, a determination of the degree to which an investor is deprived of his rights leads to a determination of whether or not there was an expropriation and whether there should be compensation. It follows that if the degree of deprivation leads to a finding of expropriation, the investor should receive compensation. As previously indicated, the LG&E tribunal did not arrive at this conclusion and did not explicitly consider questions of regulation versus expropriation; this is a clear example of inadequate reasoning. The tribunal provided one final attempt to establish a definition of indirect expropriation and apply it to the facts of this dispute. Unfortunately, this attempt also fell somewhat short. First, the tribunal blurred the distinction between indirect expropriation and expropriation: Similarly, one must consider the duration of the measure as it relates to the degree of interference with the investor’s rights. Generally, the expropriation must be permanent, that is to say, 45

LG&E—Liability, supra note 1, at ¶ 192, citing Tecmed, supra note 13, at ¶ 115.

46

Tecmed, supra note 13, at ¶ 115.


it cannot have a temporary nature, unless the investment’s successful development depends on the realization of certain activities at specified moments that may not endure variations.47 This statement suggests that no expropriation may be indirect, as it does not allow for expropriation to occur as a “process”—which comprises the definition of indirect expropriation. An indirect expropriation is “a process which, notwithstanding that it may be aimed at other legitimate regulatory objectives and does not involve a single instance of an outright taking, nonetheless has the effect, often degree-by-degree, of depriving an owner of fundamental rights of property,”48 according to Fortier and Drymer. With this conflation of terms, the tribunal rendered three terms virtually inoperative: expropriation, indirect expropriation, and regulation. If the tribunal had seriously considered the interplay between these terms, it would have carefully set forth the vast and unsettled landscape that seeks to distinguish between regulation and expropriation. Fortier and Drymer review conventions such as the Harvard Draft Convention on the International Responsibility of States for Injuries to Aliens and the Restatement (Third) of Foreign Relations Law of the United States, and ICSID, Iran-United States Claims Tribunal, and NAFTA decisions such as Pope & Talbot, Tecmed, and Starrett Housing,49 and indicate that there are at least nine different descriptions of the required level of interference for a regulation to be considered an expropriation.50 These nine descriptions include the “sole effects” and “purpose” doctrines, and the Tecmed tribunal’s “calibration of the ‘purpose’ doctrine . . . as the proportionality

47


L. Yves Fortier & Stephen L. Drymer, Indirect Expropriation in the Law of International Investment: I Know It When I See It or Caveat Investor, 19 ICSID REV.—FOREIGN INV. L.J. 293, 294 (2004). 48

49 Louis B. Sohn & Richard Baxter, Draft Convention on the International Responsibility of States for Injuries to Aliens (Draft No. 12 with Explanatory Notes, Apr. 15, 1961) (“Harvard Draft Convention”), 23 AM. J. INT’L L. SPEC. SUPP. 131 (1929); RESTATEMENT (THIRD) OF FOREIGN RELATIONS LAW OF THE UNITED STATES (1987); Pope & Talbot, supra note 27; Starrett Housing Corp. v. The Government of the Islamic Republic of Iran, Interlocutory Award No. ITL 32-24-1 (Dec. 19, 1983), 4 IRAN-U.S. CL. TRIB. REP. 122 (1983); Starrett Housing Corp. v. The Government of the Islamic Republic of Iran, Award No. 314-24-1 (Aug. 14, 1987), 16 IRAN-U.S. CL. TRIB. REP. 112 (1987). 50

Fortier & Drymer, supra note 48.


of the measure.”51 The conclusion of this exhaustive review is clear: a finding of regulation does not require compensation, while a finding of expropriation does. The test for determining whether a state’s measure is a regulation or expropriation is still “in flux”: In most cases, the determination of when State conduct crosses the line between non-compensable regulation and compensable indirect expropriation tends to involve a balancing of several considerations. That many of these considerations are particular to a given case may be frustrating to those who seek not only clarity but definitiveness in international legal norms. Yet it is perhaps the only reasonable means of approaching an issue that touches both public and private interests and that can arise in such innumerable circumstances.52 The tribunal attempted to define indirect expropriation without ever directly referencing regulation, and in so doing, offered confusing reasoning and missed an opportunity to contribute to the evolution of the indirect expropriation standard. Second, the tribunal created an analytical overlap between fair and equitable treatment and indirect expropriation, which has become a recent trend in international investment law.53 In paragraph 194, the tribunal provided, “It is this Tribunal’s opinion that there must be a balance in the analysis both of the causes and the effects of a measure in order that one may qualify a measure as being of an expropriatory nature. It is important not to confound the State’s right to adopt policies with its power to take an expropriatory measure.”54 This balancing test of the causes and effects of a measure for a finding of expropriation echoes elements of the tribunal’s test for fair and equitable treatment, in which the tribunal stated “the stability of the legal and business framework is an essential element of fair and equitable treatment in this case, provided that they do not pose any danger for the existence of the host State 51

Id. at 326.

52

Id.

See, e.g., Alex Genin, Eastern Credit Ltd., Inc. and A. S. Baltoil v. Republic of Estonia, ICSID Case No. ARB/99/2, Award (June 25, 2001); Waste Mgmt., Inc. v. The United Mexican States, ICSID Case No. ARB(AF)/00/3, Award (Apr. 30, 2004); Noble Ventures, Inc. v. Romania, ICSID Case No. ARB/01/11, Final Award (Oct. 12, 2005). 53

54



itself.”55 In assessing fair and equitable, the tribunal must weigh maintaining the stability of the legal and business framework against the existence of the state; in assessing indirect expropriation, the tribunal must weigh the state’s rights to adopt policies against the expropriatory effects of its measures. Several scholars have noted the increasingly broadened scope of the fair and equitable treatment standard, which seems to include notions of indirect expropriation and “eclipse all other duties towards investors.”56 Dolzer likewise remarks upon the broadened scope of fair and equitable treatment in BITs: [I]t will be noticed that [BITs] typically include general clauses such as “fair and equitable treatment,” “full and constant protection” and “national treatment.” Not surprisingly, the precise substance and reach of such broad rules in the specific context of the protection of foreign investment has not been clarified and will evolve in a casuistic manner. Thus, it is also far from clear whether and to what extent such clauses may be interpreted to overlap with the takings issue and whether they may be considered to provide protection to the foreign owner under certain circumstances.57 As Dolzer intimates, such clauses have, in fact, overlapped with the takings issue. As a corollary to this overlap, it is interesting to note that several tribunals—including this tribunal—have found violations of fair and equitable treatment and no violation of indirect expropriation.58 This leads to the question: in the absence of a clear means of distinguishing between a non-compensable regulation and an indirect expropriation, are tribunals instead relying on a finding of unfair and inequitable treatment in order to decide in favor of claimants? The tribunal’s final error occurred when it did not apply the series of balancing tests it established to the specific facts of the LG&E dispute. 55

Id. at ¶ 124.

Vaughan Lowe, Regulation or Expropriation?, 55 CURRENT LEGAL PROBLEMS 447, 453–54 (2002). 56

57

Dolzer, supra note 10, at 67.

See, e.g., Saluka Investments BV v. The Czech Republic, UNCITRAL, Partial Award (Mar. 17, 2006), available at www.pca-cpa.org/ENGLISH/RPC/#Saluka. 58


The tribunal presented an almost perfunctory dismissal of the indirect expropriation claim: In the circumstances of this case, although the State adopted severe measures that had a certain impact on Claimants’ investment, such measures did not deprive the investors of the right to enjoy their investment. . . . [T]he true interests at stake here are the investment’s asset base, the value of which has rebounded since the economic crisis of December 2001 and 2002. Further, it cannot be said that Claimants lost control over their shares in the licensees, even though the value of the shares may have fluctuated. . . . Thus, the effect of the Argentine State’s actions has not been permanent on the value of the Claimants’ shares, and Claimants’ investment has not ceased to exist. Without a permanent, severe deprivation of LG&E’s rights with regard to its investment, the Tribunal concludes that these circumstances do not constitute expropriation.59 According to this conclusion, LG&E’s interests were not expropriated because (1) the state’s measures did not deprive the investors of the right to enjoy their investment, (2) the value of the asset base has rebounded since the economic crisis, (3) the claimants maintained control over their shares in the licensees despite fluctuating share prices and (4) Argentina’s actions did not permanently affect claimants’ investment and the claimants’ investment has not ceased to exist. The brevity of this conclusion ignores the individual factual claims of the parties (including the claimants’ assertion that the respondent’s abrogation of the guarantees in the tariff system led to a more than 90 percent reduction in LG&E’s holdings60). The conclusion also ignores the two definitions of indirect expropriation in the “Parties’ Positions,” and the factual specificity of prior tribunals such as Metalclad 61 59

LG&E—Liability, supra note 1, at ¶¶ 198–200.

60

Id. at ¶ 177.

61 The Metalclad tribunal was significantly more specific in articulating the facts that led to its conclusion on indirect expropriation:

As determined earlier (see above, para. 92), the Municipality denied the local construction permit in part because of the Municipality’s perception of the adverse environmental effects of the hazardous waste landfill and the geological unsuitability of the landfill site. These measures, taken together with the representations of the Mexican federal government, on which Metalclad relied, and the absence of a timely, orderly or substan-


and Feldman62 in their analyses of expropriation. In short, the tribunal should have been more precise in applying the balancing test of the “degree of the measure’s interference with the right of ownership and the power of the State to adopt its policies” to this particular dispute. The tribunal missed an opportunity to fill in a “‘large and sketchy’ area of ‘large lacunae’ in the law” of which “governmental measures . . . will almost always give rise to a finding of indirect expropriation, and hence to compensation”63 through its analysis of indirect expropriation as it applies to LG&E. As noted throughout this discussion on indirect expropriation, the tribunal could have addressed the shortcomings in reasoning in the following ways: 1.

2.

identified clearly the broad definition of indirect expropriation in customary international law and the awards of other tribunals; acknowledged the existing norm that if there is an expropriation, a claimant must be compensated;

tive basis for the denial by the Municipality of the local construction permit, amount to an indirect expropriation. Although not strictly necessary for its conclusion, the Tribunal also identifies as a further ground for a finding of expropriation the Ecological Decree issued by the Governor of SLP on September 20, 1997. This Decree covers an area of 188,758 hectares within the “Real de Guadalcazar” that includes the landfill site, and created therein an ecological preserve. This Decree had the effect of barring forever the operation of the landfill. Metalclad Corp. v. United Mexican States, ICSID Case No. ARB(AF)/97/1, Award at ¶¶ 102, 107, 109 (Aug. 30, 2000), 5 ICSID REP. 200 (2002). The Metalclad tribunal then distinguished these facts from those of prior decisions. 62

In Feldman, the tribunal concluded that: In the past, confiscatory taxation, denial of access to infrastructure or necessary raw materials, imposition of unreasonable regulatory regimes, among others, have been considered to be expropriatory actions. At the same time, governments must be free to act in the broader public interest through protection of the environment, new or modified tax regimes, the granting or withdrawal of government subsidies. . . . Reasonable governmental regulation of this type cannot be achieved if any business that is adversely affected may seek compensation, and it is safe to say that customary international law recognizes this. . . .

Feldman, supra note 27, at ¶ 103. 63

Fortier & Drymer, supra note 48, at 327.


3.

4.

established a means of analysis for indirect expropriation that confronted the existing interpretive ambiguities (instead of ignoring them); and focused specifically on the particular facts of this dispute in order to reach its conclusions.

D.

State of Necessity

1.

The BIT Provisions: Articles XI and IV(3)

The tribunal’s analysis of whether a state of necessity existed in Argentina pursuant to Article XI of the BIT demonstrates two flaws: incoherent reasoning and circular reasoning. Ultimately, these flaws prevent the reader from adequately understanding how the tribunal arrived at its conclusion. Article XI of the BIT provides: This Treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.64 A plain meaning interpretation of this article simply indicates that the treaty does not prevent parties from taking “necessary measures” to (1) maintain public order, (2) maintain or restore international peace or security, or (3) protect its own essential security interests. Thus, a finding that a party breached its Article XI violations requires a finding of a “necessary measure” in one of these three categories. It would logically follow that the tribunal would first provide a factual account of the sociopolitical environment in Argentina at the time of the alleged breach and then indicate whether these facts constituted any of the three conditions set forth in the treaty. But instead of undertaking this analysis, the tribunal began this section with an assertion of its conclusion: “Argentina was in a period of crisis [from December 1, 2001, until April 26, 2003] during which it was necessary to enact measures to maintain public order and protect its essential security interests.”65 The reader now expects the tribunal to explain how Argentina was in a period of crisis and how the measures 64


65

Id. at ¶ 225.


enacted by the state maintained public order and protected its essential security interests. These expectations are met but only fleetingly. The entire section on state of necessity is a series of starts and stops—several well-reasoned paragraphs immediately followed by unexpected unreasoned paragraphs. The first example of these abrupt starts and stops occurred in the opening paragraphs of the tribunal’s analysis. Paragraphs 227 and 228 established the start and end dates of Argentina’s “factual emergency.” One would expect paragraph 229 to pinpoint the specific facts in Argentina that led to this emergency. Instead, paragraph 229 notes, “Thus, Argentina is excused under Article XI from liability for any breaches of the Treaty between 1 December 2001 and 26 April 2003.”66 How did the tribunal reason from the existence of a state of necessity, which it had not yet factually demonstrated at that point in the analysis, to exemption from liability (and thus compensation) under Article XI of the BIT? At this point, the tribunal had two options to remedy this error of reasoning: discussing liability and compensation under Article XI after fully developing the factual circumstances of the state of necessity, or pausing to explain how Article XI exempted a party from liability and compensation. In the absence of explicit treaty language on exemption from liability and compensation, the tribunal should have turned to international law. The tribunal appears to have selected the first option, as it proceeded with its factual analysis and deferred its interpretation of Article XI. Next, the tribunal carefully detailed the “severe crises in the economic, political and social sectors”67 that contributed to the state of necessity. The clarity and depth of this factual exposition is commendable. In paragraph 237, the tribunal stated: “All of these devastating conditions—economic, political, social—in the aggregate triggered the protections afforded under Article XI of the Treaty to maintain order and control the civil unrest.” It subsequently refuted the claimants’ assertions that economic crisis is not an essential interest of the state and that the economic package was not the only means necessary to respond to the crisis. There are two minor stops in the reasoning, as identified by Reinisch. First, the tribunal could have more explicitly indicated how the economic crisis in Argentina met the high threshold set forth in the Russian 66

Id. at ¶ 229.

67

Id. at ¶ 231.


Indemnity Case, which “is usually considered to reflect the idea that an extreme economic emergency may give rise to a state of necessity.”68 Second, the tribunal shifted the level of analysis by merely stating that some economic response was the “only way” of coping with the crisis. This solution, while practically keeping the necessity defence available in situations of economic necessity, is also unsatisfying because it would allow the invocation of necessity even if the actual emergency measures adopted were wholly inadequate to respond to the crisis.69 This raises the issue of the burden of review on the tribunal with respect to emergency measures. Finally, the tribunal, without explanation in either the treaty or international law, places the burden of proof on the claimant: “Similarly, the Government deemed that freezing gas tariffs altogether during the crisis period was necessary, and Claimants have not provided any reason as to why such measure would not provide immediate relief from the crisis.”70 This unsupported presumption that the burden of proof lays with the claimant arises at least one more time in this section. The tribunal’s discussion of the applicable treaty provisions concludes that Argentina did not violate Article IV(3). This article provides that Nationals or companies of either Party whose investments suffer losses in the territory of the other Party owing to war or other armed conflict, revolution, state of national emergency, insurrection, civil disturbance or other similar events shall be accorded treatment by such other Party no less favorable than that accorded to its own nationals or companies or to nationals or companies of any third country, whichever is the more favorable treatment, as regards any measures it adopts in relation to such losses.71 68 Reinisch, supra note 34, at 9. The language of the Russian Indemnity case is the following: “l’obligation pour un Etat d’executer les traités peut fléchir ‘si l’existence même de l’Etat vient à être en danger, si l’observation du devoir international est . . . self-destructive.’” Affaire de l’Indemnité Russe (Russian Indemnity Case), 11 R.I.A.A. 431, at 443 (1912) (quoted in Reinisch, id. at n.40). 69

Id. at 12.

70


71

Id. at ¶ 243.


According to this article, national and foreign companies should receive equal treatment on the territory of a party in the event of a state of emergency. The tribunal did not find a violation of Article IV(3) because “during that period, the measures taken by Argentina were ‘across the board.’”72 The tribunal’s internal logic appears to be adequate; however, this conclusion contradicts the tribunal’s earlier finding that the claimant was subject to discriminatory treatment. (See Section IV, below.) 2.

International Law and the Draft Articles on State Responsibility

In Paragraph 245, the tribunal shifted its analysis from the specific provisions of the BIT to international law. This shift becomes problematic at the point when a tribunal applies international law in lieu of, or as a means of deflecting, the particular provisions of a BIT. It is important to note that the tribunal had not yet provided any reasoning for why Article XI of the BIT precludes liability and compensation in a state of necessity. Instead, the tribunal merely continued reiterating this statement as fact: While the Tribunal considers that the protections afforded by Article XI have been triggered in this case, and are sufficient to excuse Argentina’s liability, the Tribunal recognizes that satisfaction of the state of necessity standard as it exists in international law (reflected in Article 25 of the ILC’s Draft Articles on State Responsibility) supports the Tribunal’s conclusion.73 After the tribunal concluded that Article 25 of the Draft Articles on State Responsibility further supported the tribunal’s conclusion that a state of necessity existed under Article XI of the BIT, the tribunal purportedly considered the long-deferred issue of whether a finding of a state of necessity exempts a party from liability and compensation under international law. The tribunal disposed of this issue in two short paragraphs, standing in stark contrast to the 13 paragraphs devoted to the state of necessity in international law. The tribunal stated: With regard to Article 27 of the United Nations’ Draft Articles alleged by Claimants, the Tribunal opines that the article at issue does not specifically refer to the compensation for one or all of

72

Id. at ¶ 244.

73

Id. at ¶ 245.


the losses incurred by an investor as a result of the measures adopted by a State during a state of necessity. The commentary introduced by the Special Rapporteur establishes that Article 27 “does not attempt to specify if compensation is payable.” The rule does not specify if compensation is payable during the state of necessity or whether the State should reassume its obligations. In this case, this Tribunal’s interpretation of Article XI of the Treaty provided the answer. Following this interpretation the Tribunal considers that Article XI establishes the state of necessity as a ground for exclusion from wrongfulness of an act of the State, and therefore, the state is exempted from liability.74 These two paragraphs contain several examples of inadequate reasoning. First of all, they provide a clear demonstration of circular reasoning. The tribunal began with the false premise that Article XI of the BIT exempts a party from liability, indicated that international law does not specify whether this exemption is supported, and ended with a resort to its original premise that Article XI exempts liability. Second, the tribunal was too quick to conclude that international law does not provide guidance on the issue of liability under the state of necessity. Article 27 of the Draft Articles on State Responsibility provides that “The invocation of a circumstance precluding wrongfulness in accordance with this chapter is without prejudice to: (a) Compliance with the obligation in question, if and to the extent that the circumstance precluding wrongfulness no longer exists; (b) The question of compensation for any material loss caused by the act in question.”75 A literal reading leads to the conclusion that there are circumstances precluding wrongfulness that may be compensated. As Article 27 does not expressly indicate which circumstances are compensable, the tribunal was correct to review the Commentary on the Draft Articles. However, this review was unsatisfactory. The tribunal’s statement that Article 27 “does not attempt to specify if compensation is payable,” does not relieve the tribunal of its responsibility to evaluate when, in fact, compensation is payable.

74

Id. at ¶¶ 260–261.

75 International Law Commission Articles on State Responsibility, Official Records of the General Assembly, Fifty-sixth Sess., Supp. No. 10, A/56/10, chap. V, art. 27 (2001).


A further review of the ILC Commentary on Article 27 reveals that Article 27(b) was framed as a without prejudice clause because it was not possible to specify in general terms when compensation was payable. The ILC recognized, however, that without the possibility of such compensation “the State whose conduct would otherwise be unlawful might seek to shift the burden of the defense of its own interests or concerns on to an innocent third State.” It also noted the ICJ’s interpretation of this principle in the Gabcíkovo-Nagymaros Case, where the Court held that “Hungary expressly acknowledges that, in any event, such a state of necessity would not exempt it from its duty to compensate its partner.”76 This quotation demonstrates that the tribunal should have conducted a more in-depth analysis of whether a state of necessity precludes compensation, as the question is unsettled in international law. One possibility would have been for the tribunal to distinguish the present dispute from Gabcíkovo-Nagymaros (a dispute in which the ICJ held that Hungary’s abandonment of the Gabcíkovo-Nagymaros Project was not an essential state interest under the state of necessity) by indicating that the Court merely affirmed Hungary’s position on the duty to compensate under a state of necessity without further consideration. Unfortunately, the tribunal avoided direct engagement of this question, relying instead on its tenuous conclusion that “Article XI of the Treaty provides the answer.”77 The tribunal’s final reference to Article XI and the Draft Articles on State Responsibility in the section, “Consequences of the State of Necessity” is puzzling, as the tribunal appears to admit—for the first time—that Article XI does not provide grounds for exemption from liability: As established in the Tribunal’s Analysis, Article 27 of the ILC’s Draft Articles, as well as Article XI of the Treaty, does not specify if any compensation is payable to the party affected by losses dur76 Reinisch, supra note 34, at 19, citing Commentaries to the Draft Articles on Responsibility of States for Internationally Wrongful Acts, adopted by the International Law Commission at its Fifty-third session (2001), Official Records of the General Assembly, Fifty-Sixth session, Supplement No. 10 (A/56/10), 211; and GabcíkovoNagymaros Project (Hungary/Slovakia), 1997 I.C.J. REP. 7 (internal citations omitted). 77



ing the state of necessity. Nevertheless, and in accordance with that expressed under paragraphs 260 and 261 supra, this Tribunal has decided that the damages suffered during the state of necessity should be borne by the investor.78 As demonstrated above, paragraphs 260 and 261 do not adequately explain the tribunal’s assumption that a state of necessity precludes compensation. 3.

CMS Gas Transmission Company v. the Argentine Republic

Fifteen months before this tribunal rendered its decision on liability in LG&E, another ICSID tribunal handed down an award in a dispute between CMS and the Argentine Republic.79 Despite an almost identical set of facts and reliance on comparable, if not the same, analytical sources, the CMS tribunal came to the exact opposite conclusion from this tribunal on the state of necessity. The tribunal in LG&E had no obligation to discuss the findings in CMS; however, there are compelling reasons for such a discussion to have occurred. Most importantly, Argentina occupies a unique position in the ICSID docket: as of this publication, there are numerous existing claims against the state whose published final awards have been challenged in annulment proceedings such as: Aconquija, Azurix, and CMS.80 In light of all of these existing claims, it would have been instructive for the LG&E tribunal to distinguish its findings from those of the CMS tribunal. Even though there is no stare decisis in international arbitration, it is common practice for arbitral awards under investment treaties to compare and distinguish themselves from previously issued awards. In addition, the LG&E tribunal’s consideration of CMS would likely have strengthened the adequacy of reasoning provided in the state of necessity section and would have removed the burden of untangling the contradictions between these two awards on future arbitrators. Finally, the coexistence of these two con-

78

Id. at ¶ 264.

It should be noted that the CMS tribunal’s finding on the umbrella clause was annulled for a failure to state reasons under Article 52(1)(e) of the ICSID Convention. CMS Gas Transmission Co. v. Argentine Republic, ICSID Case No. ARB/01/8 Annulment Proceeding (Sept. 25, 2007). 79

80 Compañía de Aguas del Aconquija, S.A. & Compagnie Générale des Eaux (Vivendi) v. Argentina, ICSID Case No. ARB/97/3, Award (Nov. 21, 2000); Azurix Corp. v. Argentine Republic, ICSID Case No. ARB/01/12, Award (July 14, 2006).


flicting awards invites larger questions of whether the ICSID system requires modification. Possible solutions include appellate review and the consolidation of proceedings.81 IV. OVERALL INCONSISTENCIES GENERATED FROM THE INDIVIDUAL CLAIMS: THE TRIBUNAL’S CONCLUSIONS ON UNFAIR AND INEQUITABLE TREATMENT ARE INCOMPATIBLE WITH THE STATE OF NECESSITY FINDING Thus far, incoherent reasoning in the individual sections of the award have been examined. But there are also instances in which coherent reasoning in separate sections of the award become incoherent when read as a unified whole. In these instances, the conclusions are contradictory, and it is difficult, if not impossible, to reconciled them. The first two errors of consistency occur between the conclusions on fair and equitable treatment and the conclusions on the state of necessity claim. The first error occurs with respect to the finding that Resolution 38/02 was unfair and inequitable. The tribunal asserted that Argentina also has acted unfairly and inequitably in forcing the licensees to renegotiate public service contracts, and waive the right to pursue claims against the Government, or risk rescission of the contracts . . . Likewise, the Government’s Resolution No. 38/02 issued on 9 March 2002, which ordered ENARGAS to discontinue all tariff reviews and to refrain from adjusting tariffs or prices in any way, also breaches the fair and equitable treatment standard.82 The conclusion that Resolution 38/02 breached the fair and equitable treatment standard requires the question: How can Resolution 38/02 breach the standard if it was enacted during Argentina’s state of necessity on March 9, 2002? In a footnote only two paragraphs before the mention of Resolution 38/02, the tribunal indicated “As described more fully below, Argentina is excused from liability for the measures taken during the extreme circumstances of December 2001 until April 2003 in order to maintain public order and protect its essential interests. It was fair that during that period of time, Argentina suspended the guarantees of the Gas Law and postponed the PPI tariff adjustment until such time as the Government could manage to resume its obligations.” How then, can 81

See, e.g., Reinisch, supra note 34, at 26.

82

LG&E—Liability, supra note 1, at ¶¶ 137, 138.


Resolution 38/02 be construed as unfair? Following the tribunal’s reasoning as reflected in the footnote, the tribunal should have concluded that Resolution 38/02 was fair from the date of enactment in view of necessity—March 9, 2002—until the end of the state of necessity in 2003, at which point any subsequent implementation of Resolution 38/02 would have been unfair and inequitable. The tribunal’s second error concerns compensation for unfair and inequitable treatment. In the section on fair and equitable treatment, the tribunal implicitly suggested that the claimant should have been compensated for the respondent’s breaches: “Argentina took these steps without compensating Claimants and forcing Claimants to renegotiate (a process in which an Argentine official recommended that investors waive their claims against the Government relating to the licenses) or face rescission.”83 Again, the same questions apply as in the previous paragraph. How can compensation be an issue if these steps were taken during the state of necessity, which eliminates the respondent’s liability and need to compensate? It will be interesting to note how—or if—the tribunal resolves these inconsistencies when it determines damages in the next phase of its proceedings. This incompatibility between the tribunal’s conclusions on fair and equitable treatment and the state of necessity is indicative of inconsistencies in the award’s overall clarity. V.

CONCLUSION

As demonstrated in the tribunal’s award in LG&E, inadequate reasoning can take on numerous forms. These forms range from a flat out failure to provide reasons, to a lack of internal coherence within individual sections, to an error of consistent reasoning between sections. In each instance, the reader is prevented from “proceeding from Point A to Point B”84 and the reasoning requirement of the ICSID convention is not satisfied. Yet the tribunal is uniquely positioned to resolve some of these inconsistencies because the award on damages is still outstanding. I am not calling for the tribunal to conduct a self-review or revision of the individual sections in the decision on liability, but instead to address and clearly explain what I have identified as apparent contradictions among the sections on fair and equitable treatment, discriminatory treatment, 83

Id. at ¶ 120.

84

MINE v. Guinea, supra note 3.


and state of necessity. This explanation in the next phase of the proceedings is crucial, as LG&E is one of a small number of concluded ICSID awards on liability against the Argentine Republic. The consequences of inadequate reasoning resonate beyond this particular award and the pending claims against Argentina, and onto international investment law more generally. Inadequate reasoning threatens the expediency and equity of arbitration, the most egregious inadequacies calling into question its very legitimacy. Hopefully, the ad hoc committee’s decision to annul the award in Mitchell v. Democratic Republic of the Congo 85 for a failure to state reasons will serve as an important reminder that a conscientious commitment to fully comply with the reasons requirement of Article 48(3) of the ICSID Convention should and will be taken seriously.

85 Mr. Patrick Mitchell v. the Democratic Republic of Congo, Decision on Annulment (Nov. 1, 2006), available at http://ita.law.uvic.ca/documents/mitchellannulment.pdf.

CHAPTER 5

THE AWARD IN SALUKA INVESTMENTS V. CZECH REPUBLIC George Stephanov Georgiev*

I.

INTRODUCTION

This chapter analyzes the award issued over the claim that the Czech Republic had violated its bilateral investment treaty with the Netherlands through actions following the failed privatization of one of the biggest Czech banks. The chapter focuses on the award’s reasoning and finds that, overall, the tribunal provided sufficient information about the grounds for its decision. The treatment of the jurisdiction claim, the fair and equitable treatment claim, the non-impairment claim, and the full security and protection claim is sophisticated, cogent, and thorough. In these areas, the tribunal completely fulfilled its obligation to state reasons under Article 32(2) of the Rules of the U.N. Commission on International Trade Law (UNCITRAL). Some other aspects of the tribunal’s analysis, including that of the unlawful expropriation claim, exhibit flawed or attenuated reasoning, but do not threaten the overall coherence of the award. The discussion here also demonstrates the benefits of well-reasoned arbitral decisions by noting the award’s contribution to the negotiated settlement of this and other related disputes. The Saluka arbitration arose out of the failed privatization of the Czech Republic’s third largest bank, IPB (Investicní a Postovní banka a.s.). In 1998, the government sold a controlling block of IPB shares to Nomura Europe, which in turn transferred them to Saluka Investments BV (Saluka), a Nomura special-purpose company incorporated in the Netherlands. While the bank’s performance improved considerably under its new owner and management, the IPB still lacked sufficient operating capital and was burdened by a large amount of non-perform* J.D., Yale Law School, June 2007. For helpful comments, I thank Guillermo Aguilar Alvarez, Rocío Digón, Santiago Montt, and W. Michael Reisman. Any errors or omissions are my own. The paper discusses the status of the dispute as of September 1, 2007. Contact information: [email protected].

149


ing loans. In 1999 and 2000, the Czech government continued to privatize state-owned banks and began to supervise the sector more closely. After repeatedly failing to comply with new and more stringent regulatory requirements, in June 2000 the IPB was put under forced administration by the Czech National Bank and then sold to another banking group. Nomura lost managerial control over the bank, and the IPB shares it held through Saluka were rendered worthless. In July 2001, Saluka commenced arbitration proceedings under the bilateral investment treaty (BIT) between the Netherlands and the Czech Republic.1 Saluka alleged that the Czech Republic had deprived Saluka of its investment unlawfully and without just compensation, and that the Czech Republic had failed to accord Saluka’s investment fair and equitable treatment. After several delays caused by procedural maneuvers, actions in Czech courts, administrative investigations, and parallel international arbitration proceedings, the tribunal issued a detailed partial award on March 17, 2006.2 The tribunal accepted that it had jurisdiction to hear the dispute and found the Czech Republic liable under the fair and equitable treatment and the non-impairment clauses of the BIT, but not liable under the expropriation clause or the full security and protection clause. This chapter examines the adequacy of the tribunal’s reasoning in the partial award, as measured by the award’s internal logic, coherence, and consistency. The correctness of the tribunal’s arguments and the overall merits of the decision shall not be considered. Because this project posits that the merits can be properly evaluated only if a tribunal provides adequate reasons,3 the analysis here would nonetheless be the necessary first step if we were to determine whether Saluka was adjudicated correctly.

1 Agreement on Encouragement and Reciprocal Protection of Investments Between the Kingdom of the Netherlands and the Czech and Slovak Federal Republic, signed on Apr. 29, 1991, available at http://www.unctad.org/sections/dite/iia/ docs/bits/czech_netherlands.pdf [hereinafter Czech-Dutch BIT, or BIT]. 2 Saluka Investments BV (The Netherlands) v. The Czech Republic, UNCITRAL, Partial Award (Mar. 17, 2006), available at http://www.pca-cpa.org/ENGLISH/RPC/ #Saluka [hereinafter Saluka Award or Award]. 3 See Guillermo Aguilar Alvarez & W. Michael Reisman, How Well Are Investment Awards Reasoned?, supra Chapter 1.

The Award in Saluka Investments v. Czech Republic • 151

The chapter proceeds as follows: Section II considers the content of the requirement to state reasons under the UNCITRAL Arbitration Rules; Section III provides further factual and procedural information on the dispute between the Czech Republic and Saluka; Section IV delineates certain desirable characteristics of a well-reasoned award and finds that the tribunal was successful in fulfilling the reasons requirement in the sections on jurisdiction, fair and equitable treatment, non-impairment, and full security and protection; Section V argues that the tribunal fell short of providing adequate reasons in the discussion of expropriation and of the so-called “put option.” Section VI highlights the benefits of a well-reasoned arbitral decision by presenting changes in the parties’ negotiating positions after the release of the award; and Section VII concludes. II.

THE CONTENT OF THE UNCITRAL REASONS REQUIREMENT

The Saluka arbitration was conducted under the UNCITRAL Arbitration Rules of 1976, which contain a default provision requiring that the “arbitral tribunal shall state the reasons upon which the award is based, unless the parties have agreed that no reasons are to be given.”4 Because most of the contributions in this volume discuss cases from the International Center for the Settlement of Investment Disputes (ICSID), it is necessary to note that the UNCITRAL reasons requirement is similar but not identical to its ICSID counterpart. The ICSID Convention requires that “[t]he award shall deal with every question submitted to the Tribunal, and shall state the reasons upon which it is based.”5 The ICSID Convention does not allow for a waiver of the reasons requirement (i.e., it is a mandatory, rather than a default provision). Furthermore, Article 52(1)(e) of the ICSID Convention explicitly lists a tribunal’s failure to state reasons as grounds for annulment of the award. By contrast, the UNCITRAL Rules allow for a waiver of the reasons requirement and do not prescribe a formal control mechanism (in the form of an ad hoc committee or otherwise) that could set aside an award based on a failure to state reasons.

4 UNCITRAL Arbitration Rules of 1976, adopted by the U.N. General Assembly on Dec. 15, 1976, 15 I.L.M. 701 (1976), art. 32(3) [hereinafter UNCITRAL Rules]. 5 Convention on the Settlement of Investment Disputes Between States and Nationals of Other States, ICSID (World Bank), opened for signature Mar. 18, 1965, 575 U.N.T.S. 159, art. 48(3) [hereinafter ICSID Convention] (emphasis added).


Despite these procedural differences, the actual content of the UNCITRAL and ICSID reasons requirements does not differ greatly in practice: it is highly unusual for parties to an UNCITRAL investor-state arbitration to waive the reasons requirement, so UNCITRAL awards, like ICSID awards, ought to contain reasons. Furthermore, in most cases the losing UNCITRAL party could challenge an award under lex arbitri, arguing that the failure to state reasons conflicts with public policy. One source of domestic law regarding international arbitration proceedings is the 1985 UNCITRAL Model Law on International Commercial Arbitration, which has been adopted (with or without modification) in 48 countries and six U.S. states.6 The Model Law includes “public policy” as grounds for setting aside an award. Moreover, the Explanatory Note to the Model Law lists “serious departures from fundamental notions of procedural justice” as against public policy. Thus, if the arbitration is in a country that has adopted the Model Law, and if a court of that country finds that a failure to state reasons is a “serious departure from fundamental notions of procedural justice,” the court could set aside a poorly reasoned award, much like an ICSID ad hoc committee would. These provisions in the UNCITRAL Model Law function as an indirect control mechanism that mirrors, albeit imperfectly, the more specific set-aside provisions of Article 52(1)(e) of the ICSID Convention. The proceedings in Saluka illustrate the interplay between domestic laws governing international arbitration and the UNCITRAL Rules. The Saluka arbitration used Switzerland as its seat and was therefore governed by certain provisions of the Swiss Federal Code of Private International Law (CPIL), which applies to arbitrations seated in Switzerland.7 Article 180(2)(e) of the Code allows a domestic court to set aside an arbitral award, if it is not compatible with Swiss public policy (ordre public). As already noted, it is far from inconceivable that a failure to state reasons would be against domestic public policy. Thus, if the Saluka award did 6 U.N. Commission on International Trade Law, Model Law on International Commercial Arbitration, U.N. Doc. A/40/17, Annex I, available at http://www.uncitral.org/pdf/english/texts/arbitration/ml-arb/MLARB-english_revised%2006.pdf. For a list of countries and territories that have adopted legislation based on the UNCITRAL Model Law, see http://www.uncitral.org/uncitral/en/uncitral_texts/arbitration/1985Model_arbitration_status.html. 7 Bundesgesetz über das Internationale Privatrecht vom 18. Dezember 1987 [Federal Code of Private International Law], art. 176(1), available at http://www. admin.ch/ch/d/sr/2/291.de.pdf.


not adequately state reasons, there was a legal mechanism that could have set it aside, despite the absence of a set-aside provision in the UNCITRAL Rules. The Swiss CPIL itself contains a default reasons requirement: a tribunal should “set forth the reasons on which [the award] is based.”8 This requirement does not apply in cases where the parties have agreed upon a different “procedure” or “form” through which the award should be rendered.9 This provision was not waived in the Saluka case and, by virtue of being seated in Switzerland, the tribunal was bound by two separate requirements to state reasons, one under Article 32(2) of the UNCITRAL Rules, and one under Article 189(2) of Switzerland’s CPIL. It is not unusual for tribunals to be subject to a second reasons requirement under lex arbitri, since Article 31(2) of the UNCITRAL Model Law contains a default provision concerning reasons that mirrors Article 32(2) of the UNCITRAL Arbitration Rules. III. FACTUAL AND PROCEDURAL BACKGROUND OF THE DISPUTE A.

The Failed Privatization of IPB

Many of the underlying causes of the Saluka dispute can be traced to the reorganization and privatization of the Czech banking system in the early 1990s. After the end of communist rule in 1989 and the dissolution of Czechoslovakia at the end of 1992, the Czech government undertook gradual reforms in the state-owned banking sector. As of 1994, there were four large commercial banks which included the IPB. The banks were owned and controlled either directly by the Czech state or by the National Property Fund (NPF), a state agency, while the Czech National Bank (CNB) served as the banking regulator.10 The banks’ balance sheets were burdened by large amounts of non-performing loans, some granted to insolvent state enterprises during the communist era, and some granted thereafter to buttress important newly privatized firms as a matter of public policy. By the end of 1999, the stock of non-performing loans held by banks amounted to one third of total loans, or 26 percent of the Czech Republic’s gross domestic product (GDP).11

8 Id., art. 189(2) (“Der Entscheid ist schriftlich abzufassen, zu begründen, zu datieren und zu unterzeichnen.”) 9

Id., art. 189(1).

10

Award, supra note 2, at ¶¶ 33–35.

11

Id., ¶¶ 36–37.


Nomura’s acquisition of a controlling block of IPB shares constituted the first big bank privatization by the Czech government and thus the first case in which the bad debt problem had to be worked out by a private investor. The bank was attractive to Nomura not only because of its core business, but also because of its exclusive license to provide banking services at thousands of branches of the Czech Postal Offices as well as its valuable industrial holdings portfolio, which included a substantial amount of shares in the Pilsner Urquell beer brewery.12 Nomura was somewhat of an insider since it had been active in the Czech Republic since 1990 and had advised the government and the banks (including IPB) on a number of transactions.13 It first expressed an interest in acquiring the state’s IPB shares in mid-1996, but the Share Purchase Agreement (SPA) was not accepted by the government until July 23, 1997, and the actual shares were not transferred until March 8, 1998. After the transfer, Nomura held 46 percent of the IPB’s shares, which gave Nomura effective (albeit minority) control over the bank.14 During the lengthy share purchase negotiations, Nomura was assured by the then-Minister of Finance that he intended to privatize the other big banks under conditions no more favorable than those offered to Nomura and, specifically, that future bank sales would not be preceded by state aid intended to solve the banks’ bad-debt problems. However, he also indicated that he could not bind future governments and hence these assurances were not included in the SPA.15 Also during the negotiations, Nomura emphasized to the NPF that it was not entering into IPB as a strategic partner, but only as a limited recourse equity investor, that is, a portfolio investor.16 Consistent with this, Nomura set up Saluka, a special purpose entity incorporated under the laws of the Netherlands which was to hold Nomura’s IPB shares.17 After acquiring the shares, Nomura transferred them to Saluka in two tranches, one in October

12

Id., ¶ 34.

13

Id., ¶ 43.

14

Id., ¶¶ 54, 62, 65.

15

Id., ¶ 56.

16

Id., ¶ 58.

17 It appears that the Netherlands was chosen not because of its BIT with the Czech Republic, but because of favorable laws governing special-purpose vehicles which can incorporate as tax-advantaged Stichtings, or “charitable trusts.”


1998 and one in February 2000.18 These Saluka-held IPB shares were at the center of the arbitration.19 Before acquiring the IBP shares, Nomura initiated a notable and highly profitable series of transactions with IBP, which involved the “put option” discussed in Section V.B. In February 1998, Nomura concluded a Cooperation Agreement with IPB that aimed to combine IPB’s Pilsner Urquell shares with Nomura’s Radegast Brewery shares and thus effect a merger between the two breweries that had been previously rejected by the Czech competition authority. Through a complex chain of cashless stock-for-stock transactions, the IPB’s Pilsner Urquell shares were sold to an IPB-controlled special purpose entity, which then sold them to a Nomura-controlled special purpose entity, which in turn sold them to Nomura. Nomura exercised the put option it had negotiated, and paid for the valuable Pilsner Urquell shares with IPB shares.20 Nomura then combined its Radegast shares with the Pilsner Urquell shares and sold them to a Dutch company it owned, which in turn was sold to South African Breweries at a huge profit.21 These transactions ran parallel to the events described below. Nowhere in its bank acquisition filings with the Czech authorities did Nomura disclose its plans for the Pilsner Urquell and Radegast breweries.22 Nomura’s acquisition of the IPB shares in March 1998 coincided with a tightening of bank supervision and the imposition of strict capital adequacy requirements by the CNB. Added to the bad debt problems and the lack of effective legal mechanisms for debt recovery, this put the banking sector in a very precarious position. The government intervened 18 Award, supra note 2, at ¶ 71. The transfer was structured as a complex sale that is irrelevant for present purposes. The SPA between Nomura and the Czech Republic explicitly allowed for a transfer of the shares to a special purpose entity. 19 This chapter—like the parties’ submissions and the Award—will at times speak of Nomura’s control over the IPB and of Nomura’s “ownership” of IPB shares, even though the shares were nominally held by Saluka. 20 This was structured as a deferred payment and by the time payment was made, the IPB shares were already worthless. See Section V.B. 21 Award, supra note 2, at ¶¶ 59, 68–69. This schematic description of the transaction structure is a summary from the already abridged description provided by the tribunal. The sales surrounding the put option were the subject of a separate arbitration, Torkmain Investments Ltd et al. v. Pembridge Investments BV et al. That award has not been made public. 22

Id., ¶ 68.


by purchasing non-performing loans from the state banks in order to accelerate their privatization. Aid to the three big state banks amounted to 19 percent of the Czech Republic’s GDP in 1999, and it led to the successful sale of each of them to a large international banking group.23 Even though the IPB faced the same problems as the other big banks, it received no state assistance after Nomura became the controlling shareholder.24 By mid-1999, the IPB was experiencing serious difficulties; inspections by the CNB also revealed major financial deficiencies and irregularities. The bank tried to remedy the problem by lobbying for state aid, while Nomura sought a strategic partner, but both efforts proved unsuccessful. The CNB and the Ministry of Finance repeatedly asked Nomura to provide the IPB with additional capital, but Nomura refused, claiming that it was not interested in acting as a strategic investor. The situation continued to deteriorate and after the IPB suffered a brief bank run in February-March 2000, the Finance Minister “lost trust” in Nomura and was no longer willing to meet with its representatives. Henceforth, all meetings between the government and Nomura were unofficial, conducted by deputies under a “soft mandate,” and off the premises of the Finance Ministry or CNB.25 None of the government’s various proposals were acceptable to Nomura and vice versa. As the situation deteriorated further, in May 2000 the CNB identified forced administration, or nationalization of the bank, or the revocation of its license as more likely outcomes than stabilizing the bank with the help of a private investor and state aid.26 About the same time, the government began talks with CSOB, another big Czech bank controlled by the Belgian KBC. At a meeting in Paris, CSOB, the Minister of Finance, and the Governor of the CNB discussed a proposal (the “Paris Plan”) that envisioned the forced administration of 23

Id., ¶¶ 76–83.

Even though not discussed by the tribunal, the refusal of aid to IPB after 1998 and the government’s tougher stance towards Nomura could have been politically motivated. A new Social Democrat government came to power in 1998, and the IPB management was believed to be closely linked to the outgoing Civic Democrats. For more on the political aspects of the IPB privatization, see Vladimír Balas, Saluka Investments B.V. (The Netherlands) v. The Czech Republic: Comments on the Partial Award of 17 March 2006, 7(3) J. WORLD INV. & TRADE 371 (June 2006). 24

25

Award, supra note 2, at ¶¶ 85, 90–91, 100–101.

26

Id., ¶ 113.


IPB, a narrowly prescribed mandate for a Forced Administrator,27 and a quick transfer of IPB’s assets and day-to-day business to CSOB, accompanied by substantial state guarantees. A June 1, 2000, audit found that the IPB was far short of complying with capital adequacy requirements. Media attention and statements by CNB and Ministry of Finance officials on June 8–9, 2000, triggered another, much more sizeable bank run. Nomura presented a new proposal for the IPB and, after talks with the government, was under the belief that a solution would be reached shortly. However, on the fourth day of the bank run, June 15, 2000, the Czech Cabinet met and decided to put the IPB under forced administration. Simultaneously, the Czech Securities Commission imposed an immediate suspension of trading in IPB shares.28 On June 16, the Forced Administrator assumed the powers of the Board of Directors, and armed police entered the IPB’s headquarters to remove all bank managers. Three days later, the IPB was transferred to CSOB at the price of one Czech crown. The sale was accompanied by substantial state aid largely in line with the Paris Plan. This transformed CSOB into the largest bank in Central Europe.29 On June 30, 2000, Saluka attempted to transfer its 46 percent of IPB shares to Nomura, but the government did not approve the transaction.30 Saluka repeatedly and unsuccessfully attempted to challenge the suspension of trading in IPB shares until it lost standing to do so under a new amendment to the Czech Securities Act in January 2001.31 Around the same time, the Czech police carried out a search of Nomura’s Prague office and seized a number of documents.32 27 The role of the Forced Administrator is akin to that of a trustee in a bankruptcy (id., ¶ 133). 28

Id., ¶¶ 109–111, 118–135.

29 See Matthew Karnitschnig, Meltdown in Prague, BUSINESS WEEK, Aug. 7, 2000, available at http://www.businessweek.com/2000/00_32/b3693153.htm. 30 When this award was released in March 2006, Saluka still held the IPB shares, even though as of February 2004, their registered owner has been CSOB (Award, supra note 2, at ¶¶ 72, 152, 163). 31

Id., ¶¶ 136–156.

The Czech Constitutional Court subsequently held that the search had violated the company’s fundamental rights and ordered the return of the documents (id., ¶ 157). 32


B.

Arbitration Proceedings

Saluka (“claimant”) initiated an arbitration against the Czech Republic (“respondent”) under the Czech-Dutch BIT on July 18, 2001. The Czech government filed a counterclaim, which asserted that Nomura had breached its various contractual obligations with regard to IPB and was therefore liable for compensation. After a hearing, the tribunal issued a well-reasoned decision holding that it was without jurisdiction to decide on the respondent’s counterclaim.33 The partial award discussed here was issued about two years thereafter and largely favors Nomura. Because the place of arbitration was Geneva, the Czech Republic filed a challenge in a Swiss court under lex arbitri, claiming the tribunal had failed to distinguish properly between Saluka and Nomura. This challenge was rejected in September 2006.34 On November 30, 2006, Nomura and the Czech Republic signed an agreement announcing their decision to settle the dispute.35 The partial award cannot be analyzed adequately without an overview of the numerous parallel administrative, domestic, and international proceedings related to the IPB privatization. The Czech Parliament’s investigation into the transfer of IPB to CSOB and the approval of state aid found a number of violations of Czech law.36 Various decisions of Czech courts granted relief against administrative measures taken by the government, but such relief was usually only temporary.37 Nomura also called on the European Commission to review the legality of the state aid granted to CSOB and, after this request was denied, appealed to the European Court of First Instance (CFI).38 33 Saluka Investments BV (The Netherlands) v. The Czech Republic, UNCITRAL, Decision on Jurisdiction over the Czech Republic’s Counterclaim (May 7, 2004) available at http://www.pca-cpa.org/ENGLISH/RPC/#Saluka [hereinafter Decision on Jurisdiction]. 34 Czech Republic v. Saluka Investments BV, Swiss Federal Supreme Court, Urteil über die Staatrechtliche Beschwerde gegen den Teilspruch des Schiedgerichts Genf (Sept. 7, 2006); available at http://ita.law.uvic.ca/documents/Saluka-SwissChallenge. pdf.

Damon Vis-Dunbar, Czech Republic and Nomura Securities Reach Settlement, INVESTTREATY NEWS, Dec. 1, 2006, available at http://www.iisd.org/pdf/2006/itn_dec1_ 2006.pdf. 35

MENT

36 Award, supra note 2, at ¶¶ 144–149. The Parliamentary Commission Report did not have the effect of delaying or reversing the transaction. 37

See, e.g., Award, supra note 2, at ¶¶ 156–157.

38 Id., ¶ 161. Case T-233/05, Nomura Principal Investment Plc and Nomura International Plc v. Commission was filed at the CFI in 2005.


The Czech Republic initiated a separate UNCITRAL arbitration against Nomura in Zurich, seeking recovery of the state aid it gave to the IPB after the forced administration ended.39 No final award has been rendered, but interim decisions were seen to favor the Czech Republic. Finally, CSOB-controlled funds filed an arbitration against Nomura-controlled funds, disputing Nomura’s right to pay for its acquisition of the Pilsner Urquell brewery with worthless IPB shares. That tribunal decided in favor of Nomura and refused to invalidate the put option agreement.40 IV. THE TRIBUNAL’S OVERALL SUCCESS IN MEETING THE REASONS REQUIREMENT Neither the UNCITRAL Arbitration Rules, nor the ICSID Convention gives any specific guidance on what constitutes adequate reasoning in an arbitral award. Nevertheless, if we accept that the raison d’être of the reasons requirement is the ability to reverse-engineer a decision and check whether it was correct, the following, non-exhaustive list of desirable characteristics can be drawn: internal logic, coherence, consistency, clear discussion of the applicable laws, sufficient background information and detail, and unbroken logical links between the presentation of facts, the arguments accepted by the tribunal, and the decision reached. It should also be remembered that new awards do not enter a vacuum but instead populate the evolving field of international investment law. Therefore, when an award deviates from a well-settled interpretation, it should take extra care to elaborate the legal theory and the specific facts which justify the deviation. The only conceivable “benefit” of an unreasoned award is that it is difficult to appeal and reverse, since the losing party cannot latch onto any specific legal, logical, or factual errors made by the tribunal. The drafters of the ICSID Convention and—to some extent—of the UNCITRAL Rules and Model Law, have wisely foreclosed this possibility by making failure to state reasons grounds for annulment of the award. 39 The National Property Fund of the Czech Republic and The Czech Republic v. Nomura Principal Investment plc (Japan), UNCITRAL, ad hoc, Zurich (limited public information available.) The notice of arbitration was filed in December 2002, the tribunal accepted jurisdiction, and then issued an interim award in September 2004. The Swiss Federal Supreme Court declined to set aside the award. Oral hearings on the merits concluded in September 2006. See Czech News Agency, Chronology of IPB’s Forced Administration, Related Disputes, Nov. 28, 2006; available through Lexis. 40 Torkmain Investments Ltd et al. v. Pembridge Investments BV et al. (the tribunal’s decision has not been made public). For limited case details, see Czech News Agency, Nomura Wins Arbitration with CSOB Funds over Ceske Pivo Payment, Apr. 26, 2002; available through Lexis.


A.

Craftsmanship of the Award

Before delving into the tribunal’s discussion of the specific claims raised under the Czech-Dutch BIT, it is useful to consider the overall craftsmanship of the award. First, the tribunal did a superb job of setting out the relevant facts of the case in a comprehensive manner, not only in the introductory sections, but also throughout its discussion of the case. At critical points in the rather lengthy award, the reader is either reminded of important events relating to the IPB privatization, or directed to the sections that discuss them. To this end, the award contains 19 internal cross-references. Second, with the notable exception of the expropriation section discussed in Section V.A, the award considers the text of the Czech-Dutch BIT in great detail and frequently backs its interpretation by reference to well-selected international investment law cases and public international law cases. In doing so, it conforms to the widely accepted hierarchy of authorities in international law contained in Article 38(1) of the Statute of the International Court of Justice (ICJ).41 The tribunal paid very close attention to all aspects of the parties’ pleadings: it either referenced or discussed some 24 cases that were briefed, including the majority of recent ICSID, North American Free Trade Agreement (NAFTA), and UNCITRAL investment law cases, and it also considered various secondary sources on international investment law, including law journal articles and treatises. Third, again with the notable exception of the expropriation section, the award is very careful to decide contested issues based on the applicable law, as selected by the contracting parties. At each stage of the analysis, the tribunal considered what body of law would be applicable, justified its choice, and then clearly presented the legal standard that would be used. This was particularly important in the case of Saluka, since Article 8(6) of the Czech-Dutch BIT provides the following broad and non-exhaustive list of sources of law available to an arbitral tribunal: “the provisions of [the treaty], and other relevant Agreements between the Contracting Parties; the provisions of special agreements relating to the investment; the general principles of international law.”42 Despite the 41 Statute of the International Court of Justice art. 38(1) June 26, 1945, 59 Stat. 1055, 1060, available at http://www.icj-cig.org/icjwww/basicdocuments/ibasictext/ibasicstatute.htm. 42

Art. 8(6), Czech-Dutch BIT (bullet points replaced with semi colons). This pro-


treaty’s open-endedness with regard to the applicable law, the tribunal was—for the most part—methodological in its derivation and discussion of the various legal standards relevant to the case. Last but not least, most of the award is written in a lucid manner and is organized very logically, with legal and factual issues clearly separated in different sub-sections. Even though easier to do, the tribunal did not address the parties’ arguments sequentially because such an approach would have resulted in redundancy and could have obscured the award’s reasoning. Instead, the tribunal synthesized the key issues and organized the award around them, for the most part making special effort to address all arguments presented by the parties. This assessment of the award’s craftsmanship is shared by at least one longtime observer of the legal disputes surrounding the failed IPB privatization, who wrote that “the Tribunal’s description is really very well done and allows those that have been familiar with the case from the very beginning to see also that the Tribunal[‘s] members know much more than they point at. Such a well-elaborated award also allows readers to read in between the lines.”43 B.

Decision on Jurisdiction

The detailed discussion of the respondent’s objections on jurisdiction is one of the strongest aspects of the award. Since jurisdiction is a threshold inquiry, this part also helps buttress many of the subsequent sections of the award. The tribunal began by summarizing the various jurisdictional arguments raised by the parties in each of their submissions. It distinguished between the respondent’s notice to dismiss, which alleged that the tribunal lacked jurisdiction to hear Saluka’s claims, and the respondent’s procedurally unorthodox countermemorial, which raised counterclaims against Saluka and Nomura. The tribunal then explained its decision to join the notice to dismiss with the present award on the merits, and to

vision of the treaty can be contrasted with NAFTA’s Article 1131(1), which contains a much more limited choice of applicable laws. A NAFTA tribunal must “decide the issues in dispute in accordance with [the NAFTA] Agreement and applicable rules of international law.” North American Free Trade Agreement, Dec. 17, 1992, 32 I.L.M. 605 (1993). 43 Balas, supra note 24, at 372.


consider the countermemorial in the separate decision on jurisdiction, noted in Section III.B.44 The tribunal set out the Czech Republic’s eight key jurisdictional objections in paragraph 199 of the award and then grouped them under two rubrics—arguments that Nomura/Saluka had not made a qualifying “investment” under Article 1(a) of the BIT, and arguments that Saluka does not fit the definition of “investor” under Article 8(1) of the BIT. While the order of the claims was changed to avoid redundancy, the tribunal addressed all of the arguments raised by the parties, even though they themselves had not always been fully consistent in the submissions and in the hearings. Under the rubric of “investment,” the Czech Republic had argued that the purchase of IPB shares was not an investment, since Nomura/ Saluka had invested nothing in the IPB, i.e., that a portfolio investment can become a “real” investment only if it actively contributes capital or otherwise “invests” in the enterprise after acquiring its shares. The tribunal rejected this argument by referring to the BIT, which expressly lists as investment “shares, bonds and other kinds of interests in companies and joint ventures, as well as rights derived therefrom.”45 Furthermore, the tribunal noted that nothing in the BIT makes investor motivation part of the inquiry into whether an “investment” had been made. The tribunal could have easily stopped there, but it also addressed with sophistication the Czech Republic’s complex inter-textual argument, which looked to the syntax of the chapeau of Article 2 of the BIT to argue that active “investing”—for example, making a substantial contribution to the local economy or the well-being of the company—is required in order for an “investment” to arise.46 The tribunal considered the respondent’s second challenge under the rubric of investment in the same methodological manner. The Czech Republic had argued that Nomura’s purchase of IPB shares was unlawful, because it constituted a “dishonest scheme to secure enormous profits” through the Pilsner Urquell transactions.47 Nomura had allegedly violated a CNB regulation by failing to mention its intentions with regard 44


45

Art. 1(a)(ii), Czech-Dutch BIT, reproduced at ¶ 198 of the award.

46


47

Id., ¶ 212.


to the brewery in the business plan it submitted when acquiring the controlling block in IPB. The tribunal dismissed this claim by examining the CNB regulation and noting that it could not be read to raise a legal obligation for the investor to disclose its long-term plans beyond the immediate transaction. As further reasons to deny this claim, the tribunal pointed out that the Czech Republic had consistently treated Nomura as the rightful owner of the shares, and that even if the purchase had been somehow tainted by unlawfulness under Czech law, the shares were now held by Saluka and not Nomura.48 In a second set of claims, the Czech Republic disputed the tribunal’s jurisdiction under the BIT by arguing that Saluka did not qualify as an “investor.”49 Pointing to the closeness of Saluka’s relationship with Nomura, the Czech Republic argued that Saluka was fully owned and controlled by Nomura Europe and that it was the same as Nomura Europe. The latter, being an U.K. entity, could not have recourse to the Czech-Dutch BIT. The Czech Republic also alleged that it was Nomura that made the investment, while Saluka was merely a conduit or surrogate, and as such it could not be a bona fide investor under the BIT. The tribunal dismissed these arguments by referring to the language of the treaty and noting that had the parties envisioned such an interpretation, they could have easily adopted a definition of investor not covering wholly owned subsidiaries. Moreover, the tribunal recognized that even though Nomura and Saluka were closely related, such corporate arrangements were “commonplace in the world of commerce.”50 At the same time, in another marker of the high quality of reasoning in the award, the tribunal did not reject unequivocally the Czech Republic’s corporate veil-piercing argument. Instead, it explained that veil-piercing could be appropriate as an equitable remedy in certain cases where the corporate structure had been used to perpetuate fraud or other malfeasance, but that this standard had not been met in the instant case.51

48

Id., ¶¶ 213–218.

The Czech-Dutch BIT defines “investors” as “legal persons constituted under the laws of [the Netherlands]” (art. 1(b)(ii)), and grants the tribunal jurisdiction over “[a]ll disputes between one Contracting Party and an investor of the other Contracting Party concerning an investment of the latter” (art. 8.1) (reproduced at ¶ 194 and ¶ 197 of the award). 49

50


51

Id., ¶ 230.


The Czech Republic also alleged that Saluka was not an investor under the BIT because it did not act in good faith and because it made its IPB investment through “abuse of rights.” While these arguments mirrored in part claims already raised by the Czech Republic, the tribunal discussed them in sufficient detail and found that Nomura’s failure to disclose its long-term plans did not constitute a lack of good faith and that execution of the put option did not violate international bonos mores. Even if this were the case, however, the culpable party would be Nomura and not Saluka—the sole claimant in the arbitration.52 The Czech Republic’s final attack on Saluka’s status as an investor under the BIT was predicated on the entity’s lack of factual, social, or economic links to the Netherlands. Such an entity, it argued, cannot qualify as a Dutch investor under the BIT and should not be able to take advantage of the treaty. The tribunal noted that it had “some sympathy” for this claim, which—if true—allows for “abuses of the arbitral procedure, and to practices of ‘treaty shopping’ which share many of the disadvantages of the widely criticized practice of ‘forum shopping.’”53 In view of the clear provisions of the BIT, however, the tribunal refused to “in effect impose upon the parties a definition of ‘investor’ other than that which they themselves agreed” when they had complete “freedom of choice in this matter.”54 It therefore held that Saluka qualified as a Dutch investor. The tribunal’s reasoning on the question of jurisdiction was exemplary because it gave thorough consideration to each of the respondent’s arguments. The terms of the BIT are not ambiguous and another tribunal could have quickly (and rightly) concluded that the “investor” and “investment” requirements were satisfied based on the plain meaning of the text. The Saluka tribunal, however, did not short circuit the process and provided strong analytical support for its decision to accept jurisdiction over the dispute. It also displayed the necessary discipline and consistency in doing so. It found that Saluka qualified as an investor under the BIT by treating Saluka and Nomura as separate entities; correspondingly, it did not extend its jurisdiction over Saluka to then cover Nomura, despite the closeness of the corporate relationship. Instead, the tribunal emphasized that it refused jurisdiction “in respect of any claims of 52

Id., ¶¶ 231–238.

53

Id., ¶ 240.

54

Id., ¶ 241.


Nomura, or any claims in respect of damage suffered by Nomura and not by Saluka, or any claims in respect of damage suffered in respect of the IPB shares before October 1998 when the bulk of those shares became vested in the Claimant.”55 C.

Decision on Fair and Equitable Treatment

The discussion of Saluka’s claims under fair and equitable treatment (F&ET) is perhaps the finest example of clear and manifest reasoning in the award. Before reaching a decision on this question, the tribunal was very careful to locate the appropriate F&ET standard under the CzechDutch BIT, to consider whether the customary international law standard modifies the treaty standard in any way, and then to apply the correct standard to the specific facts of the case. The consideration of Saluka’s various claims under the F&ET treaty provision is very nuanced: the tribunal neither accepted, nor rejected all of them, but instead considered them methodologically and with considerable sophistication. The fair and equitable treatment clause, together with the nonimpairment and the full security and protection clauses discussed in Section IV.D and Section IV.E is found in Article 3 of the BIT, which provides that: 1.

2.

Each Contracting Party shall ensure fair and equitable treatment to the investments of investors of the other Contracting Party and shall not impair, by unreasonable or discriminatory measures, the operation, management, maintenance, use, enjoyment or disposal thereof by those investors. More particularly, each Contracting Party shall accord to such investments full security and protection which in any case shall not be less than that accorded either to investments of its own investors or to investments of investors of any third States, whichever is more favourable to the investor concerned.56

The tribunal began the discussion of F&ET by assessing the content of the standard, first based on the parties’ submissions, and then based on the tribunal’s own interpretation according to the ordinary meaning of the BIT text, the context of the F&ET provision, and the object and 55

Id., ¶ 244.

Art. 3(1) and 3(2), Czech-Dutch BIT, reproduced at ¶ 280 of the award (emphases added). 56


purpose of the treaty. After considering a number of arguments, it concluded that the F&ET standard here is an “autonomous Treaty standard” that should not be tied to the customary international standard.57 Because the approach taken to reach this conclusion is illustrative of the award’s high standard of reasoning, it will be presented in some detail below. The tribunal first noted that the parties agreed with the Mondev award that “[a] judgment of what is fair and equitable cannot be reached in the abstract; it must depend on the facts of the particular case.”58 The claimant, however, argued that Article 3 of the BIT does not limit an investor’s recourse to protection only against conduct that is egregiously unfair, but is meant to ensure, in the language of Pope & Talbot v. Canada, “the kind of hospitable climate that would insulate [it] from political risks or incidents of unfair treatment.”59 Opposed to this view of F&ET as a positive duty, the respondent argued that the standard is rather a minimum standard of conduct which, in the language of Genin, is violated by “acts showing a willful neglect of duty, an insufficiency of action falling far below international standards, or even subjective bad faith.”60 The Saluka tribunal did not adopt or blindly rely on either of these formulations, and instead noted that the difference suggested by the parties “may well be more apparent than real,” and that “an in-depth analysis [of the various thresholds] may well demonstrate that they could be explained by the contextual and factual differences of the cases to which the standards have been applied.”61 The tribunal then noted that because the Czech-Dutch BIT does not make a reference to the customary international standard, the inquiry should be under an autonomous and narrowly tailored BIT standard. As such, it would not share the difficulties that may arise under treaties such as NAFTA, which, even though aiming to promote investment, expressly

57

Id., ¶¶ 294, 309.

Mondev International Ltd v. United States of America, ICSID Case No. ARB (AF)/99/2, Final Award, ¶ 118 (Oct. 11, 2002) (quoted by the Saluka tribunal in Award, supra note 2, at ¶ 285). 58

59 Pope & Talbot Inc. v. The Government of Canada, NAFTA, Award on the Merits of Phase 2 at ¶ 116 (Apr. 10, 2001) (quoted by the Saluka tribunal in Award, supra note 2, at ¶ 286). 60 Alex Genin, Eastern Credit Limited, Inc. and A.S. Baltoil v. The Republic of Estonia, ICSID Case No. ARB/99/2 at ¶ 367 (June 25, 2001) (quoted by the Saluka tribunal in Award, supra note 2, at ¶ 289). 61



tie the F&ET standard to the customary minimum standard.62 The tribunal effectively rejected the Czech Republic’s contention that the minimum standard is incorporated implicitly in Article 3(1) of the BIT, by examining the Genin case and showing that it does not support the proposition which the respondent sought to derive from it.63 After looking at the case law and the parties’ submissions to determine what the F&ET standard is not, the tribunal began to narrow-tailor its own approach. It adopted the mode of treaty interpretation suggested by the Vienna Convention64 and first considered the ordinary meaning of the BIT’s provisions and the immediate context in which the F&ET language is used. Unsurprisingly, this inquiry proved unfruitful due to the fairly subjective nature of the terms used in the BIT. Next, the tribunal considered the object and purpose of the BIT from the preamble and aptly noted that it presents: a more subtle and balanced statement of the Treaty’s aims than is sometimes appreciated. The protection of foreign investments is not the sole aim of the Treaty, but rather a necessary element alongside the overall aim of encouraging foreign investment and extending and intensifying the parties’ economic relations. That in turn calls for a balanced approach to the interpretation of the Treaty’s substantive provisions for the protection of investments, since an interpretation which exaggerates the protection to be accorded to foreign investments may serve to dissuade host States from admitting foreign investments and so undermine the overall aim of extending and intensifying the parties’ mutual economic relations.65 In line with this interpretation, the tribunal considered but ultimately refused to follow some of the recent, much-criticized arbitral awards that have read F&ET to require, in effect, a freezing of the host country’s regulatory apparatus just as the investor found it upon entry.66 62

Id., ¶ 294.

63

Id., ¶ 295.

Article 31(1) of the Vienna Convention requires that a treaty be interpreted “in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.” See Vienna Convention on the Law of Treaties, 1155 U.N.T.S. 331 (1969). 64

65


66

This over-emphasis on investor expectations and investor reliance can be seen


The tribunal aptly noted that to do so would impose an “inappropriate and unrealistic” obligation upon the host state and that F&ET “cannot exclusively be determined by foreign investors’ subjective motivations and considerations.” To be protected, those should “rise to the level of legitimacy and reasonableness in light of the circumstances.”67 After this carefully reasoned investigation, the tribunal reached its conclusion on F&ET, which has already been cited with approval by commentators,68 and is therefore worth quoting in full: The “fair and equitable treatment” standard in Article 3.1 of the Treaty is an autonomous Treaty standard and must be interpreted, [sic] in light of the object and purpose of the Treaty, so as to avoid conduct [. . .] that clearly provides disincentives to foreign investors. The Czech Republic, without undermining its legitimate right to take measures for the protection of the public interest, has therefore assumed an obligation to treat a foreign investor’s investment in a way that does not frustrate the investor’s underlying legitimate and reasonable expectations. A foreign investor whose interests are protected under the Treaty is entitled to expect that the Czech Republic will not act in a way that is manifestly inconsistent, non-transparent, unreasonable (i.e., unrelated to some rational policy), or discriminatory (i.e., based on unjustifiable distinctions). In applying this standard, the Tribunal will have due regard to all relevant circumstances.69 The subsequent application of this standard was just as careful as its derivation. For example, in considering whether the Czech Republic’s in Waste Management v. Mexico (ICSID Case No. ARB(AF)/00/3) (“In applying [the F&ET] standard it is relevant that the treatment is in breach of representations made by the host State which were reasonably relied on by the claimant.”) and Occidental Exploration and Production Co. v. Ecuador, LCIA Case No. UN3467 (“The stability of the legal and business framework is thus an essential element of fair and equitable treatment.”). For a criticism of this approach, see Stuart G. Gross, Inordinate Chill: BITs, Non-NAFTA MITs, and Host-State Regulatory Freedom—An Indonesian Case Study, 24 MICH. J. INT’L L. 893 (2003). 67

Award, supra note 2, at ¶ 304 (emphasis in the original).

See, e.g., Ethan Shenkman, Investor-State Arbitration: An Increasingly Powerful Tool for Resolving Disputes with Sovereigns, in THE INTERNATIONAL COMPARATIVE LEGAL GUIDE TO INTERNATIONAL ARBITRATION (2006), available at http://www.iclg.co.uk. 68

69



response to the bad debt problems in the banking sector had been discriminatory and had therefore violated the F&ET standard, the tribunal: (1) established the comparable position of the big four banks regarding the bad debt problem, (2) detailed the differential treatment of IPB with regard to state assistance, and (3) examined thoroughly whether there were reasonable justifications for the Czech Republic’s actions. Only after following this logical chain and supporting each finding with facts from the record did the tribunal conclude that the Czech Republic had violated the standard through its discriminatory response to the bad debt problem.70 Similarly, in evaluating the claim that the Czech Republic had failed to ensure a predictable and transparent framework for investment, the tribunal set out the test (frustration of Saluka’s “legitimate expectations regarding the treatment of IPB without reasonable justification”71) and then considered each of the prongs that could sustain it. Through references to the facts, the tribunal established: (1) that Nomura had no basis for expecting that there would be no future change in the government’s policy towards the bank sector’s bad loan problem, or in state aid policies; (2) that Nomura was not reasonable in expecting that the government would not introduce a more rigid system of prudential regulation in the banking sector that could affect its IPB investment adversely; and (3) that Nomura should have been aware of the longstanding shortcomings of the Czech legal system with regard to the protection of creditors and could not have reasonably expected these shortcomings to be fixed quickly. Because the tribunal found each of Saluka and Nomura’s expectations to have been unreasonable, it did not need to consider whether there were any reasonable justifications for the government’s actions. It concluded that the claim that the Czech Republic had violated the F&ET standard by failing to ensure a predictable and transparent framework for investment was without any merit.72 The tribunal discussed Saluka’s next F&ET claim—that the Czech Republic had refused to negotiate in good faith—over 71 lengthy paragraphs.73 Again, the analysis began with a clear statement of the appro-

70

Id., ¶¶ 311–347.

71

Id., ¶ 348 (emphasis added).

72

Id., ¶¶ 348–360.

73

Id., ¶¶ 361–432.


priate standard, which was finely tuned to the facts of this dispute and did not merely cite a set of general propositions: [T]he Czech Republic’s conduct was unfair and inequitable if it unreasonably frustrated IPB’s and its shareholders’ good faith efforts to resolve the bank’s crisis. A host State’s government is not under an obligation to accept whatever proposal an investor makes in order to overcome a critical financial situation like that faced by IPB. Neither is a host State under an obligation to give preference to an investor’s proposal over similar proposals from other parties. An investor is, however, entitled to expect that the host State takes seriously a proposal that has sufficient potential to solve the problem and deal with it in an objective, transparent, unbiased and even-handed way.74 In applying this test, the tribunal carefully detailed (1) the Czech Republic’s discriminatory and one-sided assistance to CSOB in its acquisition of IPB and (2) the government’s repeated refusals to negotiate with Nomura during the crisis. The tribunal then found: (1) that the Czech Republic was culpable for a lack of even-handedness because it failed to deal with the IPB and Nomura, on the one side, and CSOB, on the other, in an unbiased and even-handed way; (2) that the Czech Republic had displayed a lack of consistency in its dealings with Nomura which made it “difficult or even impossible” for Nomura to accommodate the government’s position in its successive proposals; (3) that the Czech Republic’s communications with Nomura and the IPB lacked sufficient transparency to allow them to understand the government’s terms and conditions for an acceptable solution; and (4) that the Czech Republic had unreasonably refused adequate and direct communication to Nomura at a time when active engagement was crucial to finding a solution.75 All of this led the tribunal to decide that the Czech Republic had not complied with the F&ET standard and had breached Article 3 of the BIT. The analysis of this claim was sophisticated, consistent, and coherent, and the conclusion followed logically from the reasons presented. The tribunal did not find a violation under Saluka’s last two claims based on the F&ET standard. This again showed the tribunal’s thorough understanding and nuanced treatment of the circumstances surround-

74

Id., ¶ 363.

75

Id., ¶¶ 407–432.


ing the failed IPB privatization. Saluka had argued that the provision of financial assistance to IPB after its acquisition by CSOB was in violation of the F&ET standard because it had been unlawful under Czech and EU law. The tribunal examined the case law and noted that “unlawfulness of a host State’s measures under its own legislation or under another international agreement [. . .] is neither necessary nor sufficient for a breach of [the F&ET standard].”76 The tribunal also found that the alleged unlawfulness should have been the proximate cause of some harm to the investor, and that no such harm to Saluka could be gleaned from the factual record, since IPB and CSOB received state aid only after Saluka had lost control of the bank.77 The tribunal used similar logic to dismiss the final F&ET claim—that the Czech Republic was liable for CSOB’s unjust enrichment at the expense of Saluka. The award cited the Iran-United States Claims Tribunal to clarify the content of the unjust enrichment standard as a general principle of international law.78 Quite notably, however, it did not make the mistake of inserting this international law standard into the autonomous, treaty-based F&ET standard. The relationship between the F&ET standard and the general principle of unjust enrichment is contested, and to equate them here would have required much argumentation. Instead, the tribunal prefaced its discussion of unjust enrichment by noting that it would assume, but not decide that unjust enrichment is part of F&ET.79 The tribunal then found that, as a matter of principle, neither the Czech Republic, nor CSOB could be held liable for CSOB’s alleged enrichment, because the Czech Republic did not control CSOB, and because CSOB was not a party to the BIT or the arbitration. Furthermore, based on the facts, the Czech Republic’s actions, while still in violation of the F&ET standard on other grounds, did not satisfy the proximate cause requirement here and did not sufficiently cause or enable CSOB’s unjust enrichment at the expense of Saluka.80 Since there 76

Id., ¶ 442.

77

Id., ¶¶ 445–447.

78 See Benjamin Isaiah v. Bank Mellat, 2 IRAN-U.S. CL. TRIB. REP. 232, at 236–37 (Mar. 30, 1983). (“There must have been an enrichment of one party to the detriment of the other, and both must arise as a consequence of the same act or event. There must be no justification for the enrichment, and no contractual or other remedy available to the injured party whereby he might seek compensation from the party enriched.”) 79


80

Id., ¶¶ 448–456.


was no finding of unjust enrichment, the tribunal did not have to wrestle with the difficult question of whether the principle of unjust enrichment is covered by the Treaty’s F&ET provision. Just like the section on jurisdiction, the award’s treatment of fair and equitable treatment is outstanding. The tribunal considered multiple sources and took extra care to determine the appropriate F&ET standard for the Czech-Dutch BIT. It then addressed each of Saluka’s various claims under F&ET comprehensively, first setting out the legal test to be used for a given claim, and then evaluating the claim based on the factual record. The treatment of the claims was careful and nuanced: the very same set of basic facts pointed to a violation of the F&ET requirement under some claims (discriminatory response to the bad debt problem, persistent refusal to negotiate in good faith) but did not rise to a violation of the F&ET requirement under others (failure to ensure predictable and transparent framework for investment, unlawful provision of financial assistance to CSOB, unjust enrichment of CSOB at the expense of Saluka). D.

Decision on Non-Impairment

The legal basis of the argument that the Czech Republic had violated its treaty obligations by failing to “not impair, by unreasonable or discriminatory measures, the operation, management, maintenance, use, enjoyment or disposal [of Saluka’s investment]” is found in Article 3(1) of the Czech-Dutch BIT, reproduced in Section IV.C. Non-impairment must surely have been a novelty, if not a challenge, for the tribunal to address, because the claim is not a standard part of investment law awards81 and because the content of the non-impairment standard can overlap, sometimes in imperfect and uncertain ways, with the content of the F&ET and expropriation standards. The tribunal, however, provided a thorough treatment of the claim and once again reached a lucid and well-supported decision. Given the scarcity of authorities on non-impairment, the tribunal reasoned only from the ordinary meaning of the treaty to derive a standard that might well become the benchmark for non-impairment in 81 A digest annotating publicly available investment treaty decisions and awards as of September 2006 listed only one other decision mentioning a non-impairment claim (in that case, actually termed “impairment”)— Occidental Exploration and Production Co. v. Ecuador (UNCITRAL, Final Award (July 1, 2004)). See Devashish Krishan & Ania Farren, Digest of Investment Treaty Decisions and Awards, 31 Y.B. COMM. ARB. 199–317 (2006).


investment law jurisprudence. The tribunal read “impairment” to mean “any negative impact or effect caused by ‘measures’ [i.e., actions or omissions] by the Czech Republic.”82 The treaty only covers measures that are “unreasonable or discriminatory,”83 and, accordingly, the tribunal noted that the standards of reasonableness and non-discrimination have no different meaning here than they do in the context of F&ET. The tribunal also determined that the standard of conduct that would violate the nonimpairment requirement does not differ substantially from the F&ET standard. It noted, however, that non-impairment looks almost exclusively to the specific effects of a violation, while the F&ET standard also has a conduct-based component.84 This subtle distinction renders the overlap incomplete: it is still possible for a pattern of conduct that does not violate the F&ET standard to be found in violation of the non-impairment standard, based on the conduct’s effect upon the investor. The tribunal applied its non-impairment test to three sets of facts related to the failed IPB privatization. The first set of facts, already discussed above, was the one supporting the F&ET violation. After brief reconsideration, the tribunal found that this conduct also violated the Czech Republic’s non-impairment obligation. The second set of facts related to Saluka’s deprivation (or expropriation) claim,85 and the tribunal decided that, just as that pattern of conduct did not support a finding of deprivation, it also did not support a finding of non-impairment.86 The third set of facts had not been considered before and, presumably, would not have led to a finding of deprivation or a F&ET violation. The claim was that government officials leaked information to the media and helped trigger the second run on the IPB, thereby breaching the non-impairment obligation. The tribunal found some evidence of such conduct, but was still careful to set out the appropriate legal test: “[t]he crucial question [. . .] relates to causation: was the publication of the information [. . .] a conditio sine qua non for IPB’s forced administra82


Art. 3(1), Czech-Dutch BIT. For the full text, see supra note 56 and accompanying text. 83

84


Chronologically, the award discusses the deprivation claim before the nonimpairment claim, but for reasons which will become obvious, this chapter delays analysis of deprivation until Section V.A. 85

86



tion?”87 After considering the available evidence, it found that even though the public was already aware of the IPB’s troubles, government leaks precipitated the bank run: once forced administration was discussed publicly, it became a self-fulfilling prophesy. The government’s actions were also unreasonable and discriminatory. This led the tribunal to conclude that the Czech Republic had violated its non-impairment obligation under Article 3(1) of the BIT.88 E.

Decision on Full Security and Protection

Similar to non-impairment, the discussion of full security and protection (FS&P) was fairly short because the tribunal had already considered the majority of the facts and claims related to the dispute and could incorporate them by reference as needed. This did not result in a lack of consideration of the appropriate legal standards, however. The tribunal referred to several arbitral awards and commentators in order to clarify the obligation to “accord [foreign] investments full security and protection” found in Article 3(2) of the BIT. It explained that the standard “obliges the host State to adopt all reasonable measures to protect assets and property from threats or attacks which may target particularly foreigners or certain groups of foreigners.” However, the FS&P standard “does not imply strict liability of the host State” and is “not meant to cover just any kind of impairment of an investor’s investment, but to protect more specifically the physical integrity of an investment against interference by use of force.”89 Finally, the tribunal noted that a situation involving “civil strife and physical violence” would “essentially” be necessary for the successful invocation of the FS&P standard.90 The tribunal did not find the circumstances surrounding the IPB privatization to involve civil strife and physical violence. Since such contingencies were declared “essential” prerequisites for a breach of FS&P, the tribunal could have legitimately ended the inquiry there. Instead, it proceeded and applied the legal standard to each of Saluka’s claims. In a sequential and well-reasoned fashion, the tribunal determined that even if the initial threshold to beginning a FS&P inquiry were met, none of Saluka’s three arguments established a violation of the standard. The sus87

Id., ¶ 480.

88

Id., ¶ 481.

89

Id., ¶ 484.

90

Id., ¶ 483.


pension of trading of IPB shares, however unfortunate for Saluka, could not be shown to be “totally devoid of legitimate concerns relating to the securities market.” The prohibition of transfers of Saluka’s shares, and the police searches of Nomura’s headquarters, while disruptive, were being addressed by domestic courts (with relief already granted to Nomura for the police searches), and did not rise to a denial of justice that would signal a violation of the Czech Republic’s full security and protection obligation.91 V.

THE TRIBUNAL’S SHORTCOMINGS

Despite the thorough and sophisticated treatment of the Czech Republic’s objections to jurisdiction and Saluka’s arguments under fair and equitable treatment, non-impairment, and full security and protection, the patterns of reasoning in a few other sections of the award do not rise to the same high standard. The instances of adequate or even exemplary reasoning outnumber those of less-than-adequate reasoning, and this chapter finds that—on the whole—the Saluka award satisfied the reasons requirement under Article 32(3) of the UNCITRAL Rules. The purpose of the discussion that follows is not to challenge the soundness of the tribunal’s overall reasoning or the eventual decision, but to highlight two areas where the award could have been improved. These “glitches” in the analysis, while far from trivial, do not amount to a miscarriage of justice that would threaten the validity of the award. A.

Attenuated and Flawed Reasoning on Expropriation

The tribunal’s discussion of the deprivation claim against the Czech Republic exhibits certain puzzling traits. First, the tribunal was neither clear nor consistent in the legal standard it used to evaluate Saluka’s allegations that it had been unlawfully deprived of the value of its IPB shares. Second, the tribunal summarily dismissed—without evaluation and without any explanation—several of Saluka’s deprivation claims which could have been viable at least in theory. The legal basis for the deprivation claims is Article 5 of the CzechDutch BIT, which reads as follows: Neither Contracting Party shall take any measures depriving, directly or indirectly, investors of the other Contracting Party 91

Id., ¶¶ 485–496.


of their investments unless the following conditions are complied with: a. the measures are taken in the public interest and under due process of law; b. the measures are not discriminatory; c. the measures are accompanied by provision for the payment of just compensation. Such compensation shall represent the genuine value of the investments affected and shall, in order to be effective for the claimants, be paid and made transferable, without undue delay, to the country designated by the claimants concerned and in any freely convertible currency accepted by the claimants.92 Two general observations about Article 5 are in order. First, it speaks only of “deprivation” and does not mention “expropriation,” but—in line with the practice adopted by the tribunal—the two terms will be used interchangeably here.93 There could be no linguistic disagreement based on nuances in the meaning of the Czech and Dutch terms for “deprivation,” since the BIT sets English as the controlling language. Second, on its face, the structure of the deprivation provision is straightforward: a direct or indirect deprivation of investment would not be considered unlawful only if all three of the enumerated conditions are met. Nonetheless, the tribunal did not adopt this seemingly self-contained framework and instead launched into a precarious search for a standard that interprets the deprivation provision in light of customary international law. 1.

Unclear and Contradictory Legal Standards

The tribunal began the discussion of expropriation with a succinct overview of Article 5 and the parties’ main arguments. Saluka emphasized that (1) the Czech Republic’s actions caused Saluka to be deprived of the value of its IPB shares and (2) that the Czech Republic’s actions did not satisfy any of the exculpatory conditions in Article 5(a)–(c). For 92

Art. 5, Czech-Dutch BIT, reproduced at Award, supra note 2, ¶ 245.

A similar claim of “deprivation” in lieu of “expropriation” appears in Eureko v. Poland (ad hoc, Netherlands-Poland BIT, Partial Award, Aug. 19, 2005). It appears that the use of “deprivation” is specific to Dutch BITs. 93


example, Saluka contended that by rejecting its proposal, the Czech Republic had not acted in the public interest, thereby breaching Article 5(a) of the BIT. It also claimed that the Forced Administrator of the IPB, selected by the Czech Republic, never exercised truly independent judgment, that this rendered his actions discriminatory and in violation of due process, and thereby in breach of Article 5(a) and Article 5(b).94 The Czech Republic, on the other hand, contended that the forced administration entailed “permissible regulatory actions which cannot be considered expropriatory.”95 While the award does not specify this, it appears that the Czech Republic disputed the first prong of Saluka’s claim, i.e., that a taking had occurred, and then argued that it is not necessary to inquire into whether the action was unlawful (and tantamount to expropriation), or lawful (either through compliance with Articles 5(a)–(c), or through a regulatory powers exception). After the brief presentation of the parties’ positions, the tribunal looked into the appropriate legal standard in a sub-section entitled “The Law.” It quickly noted that the BIT’s deprivation provision is drafted very broadly and does not contain any exception for the exercise of regulatory power. However, in using the concept of deprivation, Article 5 imports into the Treaty the customary international law notion that a deprivation can be justified if it results from the exercise of regulatory actions aimed at the maintenance of public order.96 The tribunal went on to state that “[i]t is now established in international law that States are not liable to pay compensation to a foreign investor when, in the normal exercise of their regulatory powers, they adopt in a non-discriminatory manner bona fide regulations that are aimed at the general welfare.”97 No case authority or other support was provided for either of these assertions. The tribunal then evoked the 1961 Harvard Draft Convention on the International Responsibility of States for Injuries to Aliens to suggest that an uncompensated taking could be excused if it is, inter alia, “incidental 94


95

Id., ¶ 250 (internal quotation marks omitted).

96

Id., ¶ 254.

97

Id., ¶ 255.


to the normal operation of the laws of the State.”98 The tribunal also noted that the 1967 OECD Draft Convention on the Protection of Foreign Property provides that “measures taken in the pursuit of a state’s political, social or economic ends do not constitute compensable expropriation.”99 (Both of these authorities are draft conventions.) Next, the tribunal quoted one of the comments to the Restatement (Third) of the U.S. Law of Foreign Relations, which includes in its list of non-compensable regulatory measures “other action of the kind that is commonly accepted as within the police power of the State.”100 Curiously, the tribunal omitted the caveats that immediately follow: “. . . if it [the action] is not discriminatory, [. . .] and is not designed to cause the alien to abandon the property to the state or sell it at a distress price.”101 After citing these three authorities and referencing several cases in an endnote, the tribunal concluded that the notion of deprivation should be understood in the meaning it has acquired in customary international law (CIL).102 It is useful to pause and consider the pattern of reasoning employed here. By deciding to adopt one particular view of CIL, the tribunal severely curtailed the probability that a finding of expropriation would be reached. And yet, the tribunal did not explain sufficiently why the deprivation inquiry required a reference to CIL. The parallel between deprivation (Article 5 of the BIT) and fair and equitable treatment (Article 3 of the BIT) is perfect: in both cases, the treaty provision is, in the words of the tribunal, “drafted very broadly” and “vague,”103 and in both cases there is a customary international law standard. If an autonomous treaty standard was sufficient to judge the fact-specific inquiry into F&ET, why isn’t an autonomous treaty standard along the same lines sufficient to carry out the fact-specific inquiry into deprivation?

98

Id., ¶ 256.

99

Id., ¶ 259 (internal quotation marks omitted).

100

Id., ¶ 260.

101 RESTATEMENT (THIRD) OF THE FOREIGN RELATIONS LAW OF THE UNITED STATES § 712, cmt. (g). This omission is notable in view of the tribunal’s finding, already discussed, that the Czech Republic’s response to the banking crisis had been “discriminatory” and in violation of the fair and equitable treatment standard. Still, the content of the two discrimination tests is not entirely identical: the BIT targets mostly economic discrimination, whereas the Restatement is aimed at more invidious discrimination. 102


103

Id., ¶¶ 254 (on art. 5), 297 (on art. 3).


The line of reasoning used to identify the CIL standard is also logically suspect. The tribunal provided an endnote citation to cases from ICSID, NAFTA, and the Iran-United States Claims Tribunal, and discussed three secondary sources, but it did not make clear why it thought these carried sufficient weight to justify an adoption of a CIL standard. Had the tribunal presented the authorities in some detail, it would have become apparent that they do not in fact point to one and the same CIL standard. Moreover, commentators have warned against using such an approach by noting that “[i]nsights from decisions handed down by disparate fora cannot [. . .] be thrown into a single stew for subsequent analysis, [because] there are important distinctions to be made between each source of precedent.”104 Furthermore, the tribunal adopted the view that uncompensated takings resulting out of bona fide regulations aimed at the general welfare could be justified under CIL without even acknowledging the splits in the case law and in scholarly opinion. While it might be true that this view has almost acquired the status of jurisprudence constante in recent years,105 there are numerous valid authorities and commentators who disagree with it. For example, some argue that only actions resulting from the state’s police powers, strictly construed, should be excused from a finding of unlawful expropriation, while others argue that “adequate compensation” is always required when there is expropriation, but leave flexible the question of how it should be calculated.106 The claim here is not that the tribunal made a legal mistake and should have followed any of these positions. An arbitral tribunal is, to be sure, fully empowered to choose the CIL standard it will apply. However, for that choice to meet the reasons requirement, it should be made after adequate consideration and discussion of the other standards that might be available. The less uniformity there is between the various standards, the more time should be spent motivating a tribunal’s eventual choice. 104 Jan Paulsson & Zachary Douglas, Indirect Expropriation in Investment Treaty Arbitrations, in ARBITRATING FOREIGN INVESTMENT DISPUTES: PROCEDURAL AND SUBSTANTIVE LEGAL ASPECTS at 147 (Norbert Horn ed., 2004). 105 See, e.g., S.D. Myers, Inc. v. Canada, 40 I.L.M. 1408, (2001); Lauder (USA) v. Czech Republic (2002). 106 For an overview, see Rudolf Dolzer, Indirect Expropriations: New Developments?, 11 N.Y.U. ENVTL. L.J. 64 (2002); L. Yves Fortier & Stephen L. Drymer, Indirect Expropriation in the Law of International Investment: I Know It When I See It or Caveat Investor, 19 ICSID REV.—FOREIGN INV. L.J. 293 (2004). See also Michael Reisman & Robert D. Sloane, Indirect Takings and Its Valuation in the BIT Generation, 74 BRIT. Y.B. INT’L L. 115 (2003).


Finally, the tribunal did not explain why, in setting out the legal test, it skipped the first two steps in a traditional deprivation inquiry. Its exclusive focus here was on how to determine whether a set of actions had “crossed the line” from permissible regulation into expropriation. The test for expropriation, however, does not start with an inquiry into the nature of the action. An action that might amount to expropriation is first judged by its impact on the investor (Did the investor suffer a taking of property?), and not on its face (Was the action somehow impermissible or unlawful?). If an investor is found to have suffered a taking of property, the tribunal needs to inquire into whether the state’s action proximately caused the taking. Only if this condition is also met does a tribunal have to examine the permissibility or lawfulness of the action.107 Since the Saluka tribunal presented only the last condition in this subpart of the award, one of the following conclusions seems warranted: 1.

2.

The tribunal had already assumed that the first two conditions (effect on investor, proximate cause) had been met. If these conditions are not included in the legal test, however, the reasoning on expropriation is rendered inadequate.108 The tribunal’s logic was flawed. It confused impermissible regulation, which is only a necessary condition in the determination of expropriation, for a sufficient condition.

The tribunal picked up the discussion of expropriation in the “Analysis and Findings” sub-section, which follows directly after the “The Law” sub-section just analyzed. At the outset, the tribunal tried to restate its test for deprivation as follows: Pursuant to Article 5 of the Treaty, the Czech Republic was prohibited from taking any measures depriving, directly or indirectly, Saluka of its investment in IPB unless one or more of the cumulative conditions set out in that Article were complied with. If the Tribunal finds that the Czech Republic has adopted such measures without having complied with one or more of these conditions, the

107 Failure to follow this approach, though clearly erroneous, is not uncommon. See Paulsson & Douglas, supra note 104, at 148–50. 108 The tribunal does state, without adducing proof, that Saluka “ha[d] been deprived of its investment in IPB” in the following sub-section (Award, supra note 2, at ¶ 267). The point here, however, is that the deprivation inquiry needs to be a clear part of the legal test.


conclusion will inevitably follow that the Respondent has breached Article 5 of the Treaty.109 Instead of a restatement, however, this is a new and logically flawed test. The first sentence suggests that fulfilling “one or more” of the conditions of Article 5 would be sufficient to bring the Czech Republic’s actions in compliance with its BIT obligation. This is incorrect under any conceivable interpretation of Article 5. The second sentence then correctly restates the actual content of Article 5, which requires compliance with all three conditions. The two sentences of the test are in logical contradiction with one another and cannot be reconciled. Even without this oversight, this is an entirely different test from the one the tribunal established in the previous sub-section of the award. Here, Article 5 is interpreted as self-contained: there is no reference to the CIL standard, and certainly no possibility for a “bona fide regulatory action” or even a “police powers” exception. The tribunal went to considerable lengths to reject such an autonomous interpretation of Article 5 in the immediately preceding paragraphs, but it seems to endorse it here. The two sub-sections of the award also stand in logical contradiction to one another and cannot be reconciled. The only piece of evidence the tribunal references during its consideration of deprivation is the CNB’s Decision from June 16, 2000, imposing forced administration upon the IPB. That decision is nevertheless reproduced in full, taking up over four pages in the award.110 The CNB decision motivated the regulatory measure according to the various provisions of Czech law and appears well-reasoned. Without examining any specific points in it, the tribunal concluded: The CNB’s decision is, in the opinion of the Tribunal, a lawful and permissible regulatory action by the Czech Republic aimed at the general welfare of the State, and does not fall within the ambit of any of the exceptions to the permissibility of regulatory action which are recognised by customary international law. Accordingly, the CNB’s decision did not, [sic] fall within the notion of a “deprivation” referred to in Article 5 of the Treaty, and thus did

109

Award, supra note 2, at ¶ 266 (emphases added).

110

Id., ¶ 270 (pp. 54–58).


not involve a breach of the Respondent’s obligations under that Article.111 This conclusion stands in logical contradiction with several previous statements and findings of the tribunal. First, it is unclear why the tribunal defers to an examination of the regulatory action only under Czech law, when in the previous sub-section it framed the legal test as requiring “permissible” and “commonly acceptable” actions under customary international law.112 The presentation of the “non-expropriatory jurisprudence” under CIL, albeit one-sided, was predicated on the very notion that there is indeed a minimum international standard that governs the lawfulness of regulatory actions that result in a taking. This international standard is entirely missing from the tribunal’s analysis here. The implication seems to be that, regardless of the legality of a regulatory action under CIL, such an action does not constitute unlawful expropriation so long as it complies with the host state’s own laws.113 If the tribunal meant what it said, this would be a far-reaching and illogical deviation from the existing case law. The quoted paragraph also contains yet another (third) restatement of the tribunal’s test for unlawful expropriation. By referring to “exceptions to the permissibility of regulatory action,” the tribunal adopts a standard whereby all regulatory actions, including those resulting in a taking, are permissible unless they fall within certain exceptions under CIL. Yet, under the CIL view announced earlier (and also based on the plain meaning of Article 5), regulatory actions resulting in a taking are by default impermissible unless they comply with certain narrowly defined conditions. This error in logic and the third consecutive alteration of the applicable standard simply do not allow the reader to ascertain which legal test was used to reach the ultimate decision. The second sentence in the quoted statement is problematic even further, because it equates “deprivation” with “unlawful deprivation.” In so doing, the tribunal contradicts itself and summarily whites out the BIT 111


112

Id., ¶¶ 261, 263.

113 This oversight is also puzzling given that the tribunal correctly separated domestic from international standards with respect to the unlawfulness of a state action in the section on fair and equitable treatment. There, it noted “[t]he unlawfulness of a host State’s measures under its own legislation [. . .] is neither necessary nor sufficient for a breach of Article 3.1 of the Treaty” (Award, supra note 2, at ¶ 442).


provisions under Article 5(a)–(c). The tribunal had already determined that deprivation took place: “There can be no doubt, and the Tribunal so finds, that Saluka has been deprived of its investment in IPB as a result of the imposition of the forced administration.” Yet, here it concludes the opposite: that the state’s actions “do not fall within the notion of a ‘deprivation’ referred to in Article 5 of the Treaty.”114 To assume that this is a problem of semantics rather than faulty reasoning would do nothing to redeem the sentence in question. The tribunal concludes that there is no breach of Article 5 because there is no deprivation. This is equivalent to the following logical statement: If no “deprivation,” then no “breach.” The logic of the treaty, however, is different. To find a breach, the tribunal needs to find deprivation, which does not meet one or more of the conditions under Articles 5(a)–(c), or: If “deprivation” & no “Art. 5(a)–(c),” then “breach.” The two statements are simply not logical equivalents. In reasoning under the first instead of the second, the tribunal commits the “necessary vs. sufficient condition” fallacy and entirely disregards the three cumulative exculpatory conditions of Articles 5(a)–(c). In sum, both the derivation and the application of the legal test used in the analysis of expropriation contain significant logical lacunae that undermine the coherence of this part of the award. The tribunal used inadequate authorities, failed to justify its reliance on one particular view of the CIL test when a split was present, employed faulty reasoning, and seemed to apply three contradictory legal standards to decide the same question in the same sub-section. 2.

Plausible Claims Not Addressed

Even though the tribunal presented all of Saluka’s claims under Article 5 of the BIT at the beginning of the section on deprivation, it subsequently considered only some of them.115 After finding that the Czech Republic was not liable for unlawful expropriation based on the imposition of forced administration on IPB, the tribunal noted that it would: not consider the Claimant’s allegations that the Czech Republic was an accessory to CSOB’s alleged plan to take over IPB, that 114

Award, supra note 2, at ¶ 267 and ¶ 275.

There are also some unexplained differences in the presentation of Saluka’s claims at the beginning (id., ¶ 249) and at the end (id., ¶ 278) of the section on deprivation. 115


the Forced Administrator did not exercise truly independent judgment or that the Czech Government discriminated against IPB by granting State aid to Saluka’s competitors. [T]hese allegations, even if proven, would not rise to the level of a breach of Article 5. They will in any event be considered in the next Chapter [. . .] that addresses [. . .] Article 3 of the Treaty.116 There is a manifest absence of reasoning in this statement. Just because a claim will be considered under Article 3 of the treaty does not excuse the tribunal from having to consider it under Article 5. The legal standard for unlawful deprivation (Article 5), and the legal standards for fair and equitable treatment, non-impairment, and full security and protection (Article 3) contain some of the same elements, but they are by no means identical or even linked in any formal way. A claim that does not prevail under Article 3 (sometimes considered an easier standard to meet), could nevertheless be successful under Article 5. Under the approach adopted by this tribunal, such a claim would be overlooked. As already discussed in Section II, the Saluka arbitration was conducted under the UNCITRAL Rules that, unlike the ICSID Convention, do not contain a formal requirement that every question submitted by the parties should be addressed by the tribunal.117 But this does not obviate the problem. It is difficult to conclude that the tribunal here fulfilled the requirement to state reasons for its decision, a formal element of the UNCITRAL Rules, if it left doubts that Saluka could have prevailed on the unaddressed claims. At a minimum, the tribunal should have (1) stated the legal threshold these three claims must meet in order to rise to the level of unlawful expropriation and (2) explained why they do not meet it. Finally, even if Saluka’s claims could not satisfy the requirement for direct deprivation, they could have been sufficient for a finding of indirect deprivation. For example, the tribunal in Metalclad v. Mexico found indirect expropriation based on two elements also present in the Saluka case—specific undertakings or representations on the part of the host state, and legitimate reliance or expectations on the part of the investor.118 It is 116

Id., ¶ 278.

117

See supra note 5 and accompanying text.

Metalclad Corp. v. United Mexican States, ICSID Case No. ARB(AF)/97/1 (Aug. 30, 2000). 118


uncertain whether a similar conclusion could be reached in Saluka. Nevertheless, without a discussion of indirect deprivation—an express source of liability under Article 5—it is difficult to conclude that this section of the award exhibits clear and manifest reasoning. B.

Insufficient Discussion of the “Put Option”

Any possible criticisms of the award outside of its discussion of deprivation are significantly less serious. The only one that casts a shadow over the adequacy of reasoning is the insufficient consideration given by the tribunal to the put option negotiated and exercised by Nomura. As explained in Section III.A, Nomura bought IPB’s shares in the Pilsner Urquell brewery through a complex series of transactions, combined the Pilsner shares with the Radegast ones it already owned, and sold the resulting entity to a major international brewing company at a huge profit. The put option is notable because it allowed Nomura to pay for its purchase of other assets with IPB shares and because it was eventually exercised to pay for Nomura’s acquisition of Pilsner Urquell shares. The Pilsner Urquell transaction was structured as a deferred payment: Nomura bought these Pilsner shares before the forced administration of the IPB (i.e., when the IPB shares still had value), but it paid for the Pilsner shares after the forced administration of the IPB, when the IBP shares used as means of payment were entirely worthless. Thus, Nomura acquired a valuable asset from IPB and then paid CSOB for it with something that had no value. This transaction gave rise to an arbitration claim that was decided in favor of Nomura in Torkmain Investments Ltd et al. v. Pembridge Investments BV et al.119 The huge profit Nomura made from the sale of an IPB asset, while simultaneously refusing to enter into IPB as a strategic investor, was also the source of much public attention both before and after the forced administration of the bank. There is evidence that this transaction “was the last drop and the main officially articulated reason why the [Czech] government strictly refused to provide any financial assistance to IPB.”120 Even though the award contains some discussion of the put option, the consideration given seems insufficient in light of its importance to the dispute between Nomura/Saluka and the Czech Republic. But was the put option a part of the Czech Republic’s defense? The tribunal 119 120

See supra note 40 and accompanying text. Balas, supra note 24, at 374.


indeed noted that the Czech Republic referred to the legality and the propriety of the put option as part of its arguments under jurisdiction and under fair and equitable treatment.121 The submissions of the parties in Saluka are not publicly available, so it is not possible to ascertain how central the put option was to the Czech Republic’s defense. However, there is reason to believe that it was quite important. In its discussion of the parties’ submissions, the tribunal suggested that that the respondent’s notice to dismiss and its countermemorial contained similar arguments.122 The decision on jurisdiction over the Czech Republic’s counterclaims indicates that the put option was central to the Czech Republic’s counterclaims against Saluka,123 which in turn suggests that the put option must have been similarly important to the respondent’s defense in the current arbitration.124 Despite this significance, the tribunal’s discussion of the put option was in effect limited to the following pronouncement under fair and equitable treatment: So far as concerns any alleged illegality involved in the creation or operation of the Put Option, the Tribunal notes, and sees no reason to dissent from, the decision of the tribunal in the first arbitration under the Put Option agreement in Torkmain Investments Ltd et al. v. Pembridge Investments BV et al., in its second interim award, that the Put Option agreement was valid, as was the Put Option itself. Moreover, the Tribunal notes that, in the second such arbitration, it was accepted by CSOB (apparently acting on behalf of the Czech Republic) that those two matters were res judicata as a matter of Czech law.125 Even though this statement does not entirely disregard claims related to the put option, it is not an example of clear and manifest reasoning. 121

Award, supra note 2, at ¶¶ 183(a), 184(d), 212, 231, 233.

122

Id., ¶¶ 12–21.

See Decision on Jurisdiction, supra note 33. The separately-issued decision on jurisdiction over the Czech Republic’s counterclaim should not be confused with the tribunal’s decision on jurisdiction in this award, discussed in Section III.B. 123

124 The question of the put option was not discussed on the merits in the decision on jurisdiction over the Czech Republic’s counterclaim and so it could not have been res judicata. 125



That CSOB accepted the legality of the put option under Czech law is not sufficient to remove questions about the put option’s compliance with international bonos mores.126 CSOB, a private party, also cannot formally bind the Czech Republic, a sovereign, through its alleged acquiescence to the legality of the put option. CSOB might have conceded this point for a variety of reasons, including strategic litigation reasons.127 If the Czech Republic raised questions about the put option before the Saluka tribunal, the tribunal should have addressed those questions in a reasoned manner, independent of third parties’ prior positions on the matter. In a similar vein, it would also have been helpful, if the tribunal had explained what led it to “see no reason to dissent from” the decision in the separate put option arbitration that was made by a different tribunal. For the outside observer, the entire Torkmain v. Pembridge arbitration is a black box, since nothing in connection with it has been made publicly available. Reliance on such non-public materials here is problematic in view of the fact that arbitral awards are not merely for the consumption of the parties to the dispute, but are frequently relied upon by subsequent awards, and thus have a substantial impact on the development of international investment law jurisprudence. VI. THE BENEFITS OF THE WELL-REASONED AWARD IN SALUKA After scrutinizing and critiquing the patterns of reasoning employed by the Saluka tribunal, it is useful to return to the history of the underlying dispute. Developments following a Swiss federal court’s refusal to set aside the award provide an interesting post-mortem of the Saluka arbitration and make it a compelling case study of the benefits that wellreasoned arbitral decisions can confer. This chapter found that the Saluka tribunal drafted an award that is for the most part coherent, cogent and persuasive, and that fulfills the reasons requirement under Article 32(3) of the UNCITRAL Rules. As 126 It is also not clear whether the tribunal’s statement about CSOB’s position is even correct. When the award was released, CSOB was still engaged in domestic court litigation over the legality of the Pilsner Urquell transaction of which the put option was an integral part. See Czech News Agency, supra note 39. 127 That the Czech Republic and CSOB are not behaving as a single actor in this arbitration and indeed have separate interests is confirmed by the fact that CSOB had still not joined the settlement between Nomura and the Czech Republic several weeks after it was negotiated. See Section VI.


already noted, observers have also found that it allows them to “read in between the lines.”128 The award most likely gives the parties who are insiders to the dispute even more information about the tribunal’s disposition with regard to the range of damages that might be awarded at the next stage. Such specific information is beneficial, because it allows the parties to limit or resolve the dispute in an amicable way, without continuing to bear the uncertainties and the expense involved in openended arbitration proceedings. This is exactly what happened in the Saluka case. Even though the parties had been trying to reach a settlement at least since the autumn of 2005,129 it was only after the present award was cleared by a Swiss court in September 2006 that these negotiations met with some success. On November 30, 2006, Nomura and the Czech Republic announced a preliminary settlement that placed a $332 million (U.S.) cap on the amount of damages the tribunal can award to Saluka in the final stage of the arbitration. In return, the Czech Republic agreed to drop its counterarbitration against Nomura, in which a decision had been expected shortly.130 The negotiated damages cap is notable in light of the $1.9 billion in damages Saluka had demanded from the Czech Republic, and the $5.7 billion in damages the Czech Republic had demanded from Nomura in the parallel arbitration.131 Nomura expected that this agreement would lead to a settlement of its dispute with CSOB as well, thereby putting an end to the seven-year chain of domestic and international litigation over the failed privatization of the IPB. The final terms of the settlement between the Czech Republic and Nomura were announced on June 20, 2007. The government agreed to pay around $300 million, or 10 percent of the difference between the price of IPB in 2000 when it was put under forced administration and taken over by CSOB, and the costs of rescuing IPB’s assets.132 Balas, supra note 24, at 372. 129 See Frantis ek Bouc, IPB Bank Dispute Nears End, PRAGUE POST, Oct. 26, 2005; available at http://www.praguepost.com/articles/2005/10/26/ipb-bank-dispute-nearsend.php. 128

130

Vis-Dunbar, supra note 35.

Central Europe Banking & Finance Weekly, Czech State, Japan’s Nomura Agree to Drop Multibillion Euro Counterclaims over IPB bank Collapse” INTERFAX NEWS AGENCY, Dec. 2, 2006, available through Lexis. 131

132 Czech News Agency, Czech Republic to Pay Some Kc6bn for Settlement with Nomura, GLOBAL NEWS WIRE—EUROPE INTELLIGENCE WIRE, June 20, 2007, available through Lexis.


The amicable conclusion of the dispute between Nomura and the Czech Republic stands as a counterpoint to post-arbitration developments in disputes where the tribunal failed to comply with the reasons requirement. The ICSID case of Wena Hotels Ltd. v. Arab Republic of Egypt, discussed in detail in Chapter 8, serves as a good example. The lack of clear and manifest reasoning in the original award and in the ad hoc committee’s review of the award led to much further litigation in domestic courts by parties related to the dispute, and, some seven years after the beginning of the arbitration, to an application for interpretation of the original award. The prevailing party, Wena, was forced to relitigate the case for years and in multiple fora because, in failing to state reasons, the original tribunal created the impression that it had actually not decided a number of important subsidiary questions.133 In contrast, the Saluka tribunal foreclosed such a nebulous outcome in this case by providing wellelaborated reasons at each crucial stage in the award, and by eventually advancing a settlement that includes and binds all parties to the dispute. VII. CONCLUSION Because of the lack of precise standards for what constitutes adequate reasoning, any such assessment is necessarily subjective. To analyze the Saluka award, this chapter considered its internal logic, coherence, consistency, the discussion of applicable law, the presentation of background information, and the logical links between the facts, the arguments accepted by the tribunal, and the decision reached. Evaluated on these criteria, the Saluka award stands in contrast to the majority of the decisions discussed in this volume. The award also contains an interesting contrast within itself: it highlights the difference between adequate and inadequate reasoning on legal claims that are based on the very same set of facts. The discussion on expropriation, unreasoned in its own right, seems even more inadequate when compared to the tribunal’s masterful treatment of fair and equitable treatment, non-impairment, full security and protection, and jurisdiction. The analysis of the latter promotes a thorough understanding of the dispute’s complexity and builds confidence in the ultimate decision on the merits; the analysis of the former confuses and prompts the reader to at least question parts of the eventual outcome. Virtually all of the claims that arise in investment arbitrations require a highly contextual analysis that, in turn, requires clear and manifest rea133 See Julie Maupin, The Award in Wena Hotels Limited v. Arab Republic of Egypt, supra Chapter 8, particularly Section V.


soning in order to be persuasive. As Saluka shows, it is not possible to apply a general legal test without narrow-tailoring it to the specific circumstances of the dispute, the treaty regime, and the legal context. The tribunal in Saluka provided excellent examples of how an award should be drafted and what it should contain. Such an approach predicated upon adequate reasoning, even though more elaborate, is central to the integrity of the international investment arbitration system. It contributes to the finality of the dispute, creates models of reasoning that can be adopted by later tribunals, advances international investment law, and provides host states with much-needed, detailed guidance about the specific treatment they are required to accord to foreign investors.

CHAPTER 6

THE DECISION ON JURISDICTION IN TOKIOS TOKELES V. UKRAINE Nartnirun Junngam*

I.

INTRODUCTION

Nationality, usually defined as the quality of being a subject of a certain state,1 has been one of the legal concepts that plays a prominent role in both the domestic and international arenas. It not only provides a basis for state jurisdiction, but also creates a link between nationals, including individuals and corporations, and international law. For example, pursuant to traditional international law, if a state’s nationals were mistreated abroad, their home state exclusively would have the right to ask for remedy through diplomatic protection.2 Nationals themselves had no such a right. The Nottebohm3 and Barcelona Traction4 cases serve as examples of this principle. Similarly, in the context of international trade law, no nationals of one state can bring a claim directly against other states, even if their private and economic interests are hurt. As two commentators have noted, “[v]iolations of trade law, even though they strike at the economic interests of private parties, are matters resolved directly and solely by states.”5

* J.S.D. Candidate, Yale Law School, LL.B. (Honors) (Thammasat), Barrister-atLaw (Thai Bar Association), LL.M. (Yale). I am grateful to Professors W. Michael Reisman and Guillermo Aguilar Alvarez, who read and commented on an early draft of this article. I remain responsible for any errors and omissions. 1 LASSA OPPENHEIM, OPPENHEIM’S INTERNATIONAL LAW 851 (Robert Jennings & Arthur Watts eds., Longman 1992) (1905). 2

Id. at 849.

3

Nottebohm case (Liechtenstein. v. Guatemala.), 1955 I.C.J. 4 (Apr. 6).

Barcelona Traction, Light and Power Company, Limited (Belg. v. Spain), 1970 I.C.J. 3 (Feb. 5). 4

5 Jeswald W. Salacuse & Nicholas P. Sullivan, Do BITs Really Work?: An Evaluation of Bilateral Investment Treaties and Their Grand Bargain, 46 HARV. INT’L L.J. 67, 88 (2005).

191


In 1965, the Convention on the Settlement of Investment Disputes between States and Nationals of Other States6 created the first forum for bringing and adjudicating foreign investment disputes between nationals of one state and another sovereign state,7 the International Center for Settlement of Investment Disputes (ICSID). Again, the nationality of claimants is one of the decisive elements for determining whether a claim can be arbitrated under an international regime. Only nationals of the contracting states can benefit from the ICSID’s mechanism. It seems that the role of nationality in the Convention is more important than it is in public international law and international trade law. As lex specialis allowing nationals to bring a claim directly against a sovereign state without their national state’s filter, it should be applied carefully and only with the sufficient line of reasoning. Nationality as a part of jurisdiction ratione personae of ICSID’s tribunals has often been challenged by host states, arguing that a claimant is not an investor of another contracting state, and asking a tribunal to deny jurisdiction over the case. The dispute in Tokios Tokeles v. Ukraine 8 is among the cases in which the tribunal had confirmed its jurisdiction, despite such objection. The Tokeles tribunal issued two procedural orders and ultimately rendered an award of July 26, 2007, in favor of Ukraine. The critical part of the decision on jurisdiction and the dissenting opinion both focused on the nationality of the investor. While the majority seemed to adopt a passive approach when dealing with the issue, the dissident took an overly active role. This case is problematic, not only because after the decision on jurisdiction was issued, the dissenting chair of the tribunal resigned, but also because of the quality of its reasoning. The analytical purpose here is to consider the instances of inadequacy or manifest absence of reasoning in the case. The first part of the article will present the relevant facts, the pertinent legal provisions, the problematic issues, and the controlling and dissenting opinions. The second part will focus exclusively on the patterns of reasoning employed by the tribunal. 6 Convention on the Settlement of Investment Disputes between States and Nationals of Other States, ICSID (World Bank), art. 1, opened for signature Mar. 18, 1965, 575 U.N.T.S. 159 [hereinafter ICSID Convention or Convention]. 7 R. DOAK BISHOP COMMENTARY 5 (2005).

ET AL.,

FOREIGN INVESTMENT DISPUTES: CASES, MATERIAL

AND

8 Tokios Tokeles v. Ukraine, Decision on Jurisdiction, ICSID Case No. ARB/ 02/18 (Apr. 29, 2004), available at http://www.worldbank.org/icsid/cases/awards. htm [(hereinafter Tokeles].

The Decision on Jurisdiction in Tokios Tokeles v. Ukraine • 193

II.

FACTS OF THE CASE

The claimant, Tokeles, was a juridical person established under the laws of Lithuania. First founded as a cooperative in 1989, it had been registered as a closed joint-stock company since 1991. Almost all of its shares (99 percent) were owned by the nationals of Ukraine. In addition, twothirds of its management was also run by Ukrainians. The claimant was engaged primarily in the business of advertising, publishing, and printing in Lithuania and outside its borders.9 In 1994, Tokeles established Taki spravy, a wholly owned subsidiary, in conformity with the laws of Ukraine. Taki spravy also did the same business in Ukraine and outside its borders. Tokeles made an initial investment and then continuously reinvested in Taki spravy in the period 1994–2002.10 Tokeles alleged that governmental authorities in Ukraine breached their obligations under the bilateral investment treaty between Ukraine and Lithuania (BIT). It contended that the respondent engaged in a series of unreasonable and unjustified actions against Taki spravy that affected its investment adversely.11 Tokeles asserted that it tried to seek recourse from local remedy. However, it was unsuccessful, and the alleged governmental actions continued.12 Tokeles and Taki spravy subsequently filed a Request for Arbitration to the ICSID. However, the ICSID later notified Tokeles and Taki spravy that Ukraine and Lithuania had not agreed that Taki spravy should be treated as a national of Lithuania under Article 25(2)(b) of the ICSID Convention and Article 1(2)(c) of the BIT. Thus, Tokeles removed Taki spravy as a requesting party.13 III. PROBLEMATIC ISSUES: OBJECTION TO JURISDICTION RATIONE PERSONAE At the beginning of the proceedings, the respondent raised a number of jurisdictional objections and requested the tribunal to consider these objections as a preliminary matter. Despite the claimant’s objections to such an approach, the tribunal agreed to bifurcate the proceedings.14 The respondent contended that although Tokeles was established 9

Id. at ¶ 1.

10

Id. at ¶ 2.

11

Id. at ¶ 3.

12

Id. at ¶ 4.

13

Id. at ¶¶ 5–9.

14

Id. at ¶ 11.


under the laws of Lithuania, it was not a “genuine entity of Lithuania.” This was due to the undisputed fact that the nationals of Ukraine owned 99 percent of the outstanding shares of Tokeles and comprised two-thirds of the management. In addition, the respondent argued that Tokeles had no substantial business activities in Lithuania and maintained its siège social (social seat), or administrative headquarters, in Ukraine. Therefore, the respondent argued that the claimant was “a Ukrainian investor in Lithuania,” not “a Lithuanian investor in Ukraine,” if judged in terms of economic substance.15 The respondent argued that if the tribunal affirmed its jurisdiction over this case, it would be tantamount to allowing Ukrainian nationals to pursue international arbitration against their own national state. Doing so would be inconsistent with the object and purpose of the ICSID Convention. Based upon this contention, the respondent asked the tribunal to pierce the corporate veil and disregard the claimant’s country of incorporation. The respondent proposed that the tribunal determine the claimant’s nationality according to the nationality of its predominant shareholders and managers in conformity with the control test.16 IV. RELEVANT LEGAL PROVISIONS Article 25 of the ICSID Convention sets forth the objective criterion for jurisdiction: (1) The jurisdiction of the Centre shall extend to any legal dispute arising directly out of an investment, between a Contracting State . . . and a national of another Contracting State, which the parties to the dispute consent in writing to submit to the Centre. When the parties have given their consent, no party may withdraw its consent unilaterally. (2) National of another Contracting State means: .... (b) any juridical person which had the nationality of a Contracting State other than the State party to the dispute on the date on which the parties consented to submit such dispute 15

Id. at ¶ 21.

16

Id. at ¶ 22.


to conciliation or arbitration and any juridical person which had the nationality of the Contracting State party to the dispute on that date and which, because of foreign control, the parties have agreed should be treated as a national of another Contracting State for the purposes of this Convention. . . .17 Article 1(2) of the BIT between Ukraine and Lithuania defines investors as: (a) in respect of Ukraine: — natural persons who are nationals of Ukraine according to Ukrainian laws; — any entity established in the territory of Ukraine in conformity with its laws and regulations; (b) in respect of Lithuania: — natural persons who are nationals of the Republic of Lithuania according to Lithuanian laws; — any entity established in the territory of the Republic of Lithuania in conformity with its laws and regulations; (c) in respect of either Contracting Party—any entity or organization established under the law of any third State which is, directly or indirectly, controlled by nationals of that Contracting Party or by entities having their seat in the territory of that Contracting Party; it being understood that control requires a substantial part in the ownership.18 V.

HOLDING

In light of jurisdiction ratione personae, the tribunal recognized that the Convention did not set forth any method for determining the nationality of juridical entities, leaving this task to the reasonable discretion of the contracting parties.19 Therefore, in order to decide whether the claimant had met the jurisdiction ratione personae requirement, the tribunal had to consider first the nationality laws of each party and the def17

ICSID Convention, supra note 6, art. 25.

18 Agreement between the Government of Ukraine and the Government of the Republic of Lithuania for the Promotion and Reciprocal Protection of Investments art. 1(2) (Feb. 8, 1994). 19

Tokeles, supra note 8, at ¶ 24.


inition of corporate nationality contained in the agreement between the contracting parties, that is, the BIT.20 The tribunal referred to Professor Schreuer’s opinion that the contracting parties enjoy broad discretion to define corporate nationality, meaning that definitions of corporate nationality in national legislations or in treaties providing for ICSID jurisdiction will be controlling for determining whether the nationality requirement of Article 25(2)(b) is met.21 The tribunal asserted that it relied on Article 31 of the Vienna Convention on the Law of Treaties22 in order to interpret the BIT and the ICSID Convention.23 It began considering the nationality of the claimant by referring to the definition provided in the BIT. It found that, with respect to the ordinary meaning of the terms “entity” (a thing that has a real existence) and “establish” (to set up on a permanent or secure basis, bring into being, found [a business]), the claimant would have the nationality of Lithuania, if it had a real legal existence founded on a secure basis in the territory of Lithuania in conformity with its laws and regulations. No additional requirements were needed.24 Based on the foregoing condition, the tribunal held that: (1) the claimant was an entity of real legal existence founded on a secure basis in the territory of Lithuania and (2) the registration of Tokeles by the Lithuanian government indicated that it was founded in conformity with the laws and regulations of that country. Therefore, regardless of the economic criteria indicating that the claimant was owned and controlled predominantly by the nationals of Ukraine, the claimant was an investor of Lithuania according to the ordinary meaning of Article 1(2)(b) of the BIT.25 However, the respondent further argued that the “substantial part in ownership” requirement of Article 1(2)(c) of the BIT should be used to deny the benefits of the BIT under Article 1(2)(b). In its view, the control test in Article 1(2)(c) which applied to entities established in third countries could be transposed to limit the scope of the BIT in this (bilat20

Id. at ¶ 25.

21

Id. at ¶ 26.

Vienna Convention on the Law of Treaties, art. 31, UN Doc. A/Conf.39/27; 1155 U.N.T.S. 331, 8 I.L.M. 679 (1969), 63 AM. J. INT’L L. 875 (1969). 22

23


24

Id. at ¶ 28.

25

Id. at ¶ 29.


eral) case. The tribunal did not find this argument persuasive. It explained that if the contracting parties had intended such criterion to apply to entities legally established in Ukraine or Lithuania, they would have included the criterion in Article 1(2)(a) or (b), just as they did in Article 1(2)(c). Therefore, the investor nationality test in Article 1(2)(c) applied only to entities established under the law of a third state provided that certain conditions were met.26 In the tribunal’s view, in the absence of a denial of benefits provision in the BIT, it was irrelevant whether the claimant conducted its substantial business activity in Lithuania. The only relevant consideration was whether the claimant was established under the laws of Lithuania.27 Furthermore, it declined to look beyond (or through) the claimant to its shareholders or other juridical entities that might have an interest in the claim. The tribunal found support in Amco Asia Corp. v. Indonesia,28 which held that the concept of nationality in the ICSID Convention was based upon the law under which the juridical person had been incorporated, the place of incorporation, or the place of the social seat.29 Thereafter, the tribunal reaffirmed the consistency of Article 1(2) of the BIT with Article 25(2)(b) of the Convention. While the former relied only upon the state of incorporation rule, the first clause of the latter provided no explicit criteria for determining the nationality of a juridical person. Although the BIT did not refer to the state of seat rule, the tribunal found that if applied, it would bring the claimant the same nationality. At this point, the tribunal considered that the BIT laid down the reasonable and effective criterion for determining nationality consistent with the Convention; hence, there was no reason to reject the will of the parties.30 The next step the tribunal took was to explain under which circumstance the control test would be adopted for the purpose of ICSID jurisdiction. From its perspective, the control test was strictly applied to the sole context described in the second clause of Article 25(2)(b); that is, 26

Id. at ¶ 30.

27

Id. at ¶ 38.

28 Amco Asia Corp. v. Indonesia, Decision on Jurisdiction, ICSID Case No. ARB/81/1 (Sept. 25, 1983), 1 ICSID REP. 377, 405, 565 (1993). 29

Tokeles, supra note 8, at ¶¶ 40–41.

30

Id. at ¶¶ 42–43.


when contracting parties agree to treat a national of the host state as a national of another contracting party because of foreign control.31 The tribunal cited Broches’ opinion that the purpose of the control test in the second clause of Article 25(2)(b) is to expand the jurisdiction of ICSID.32 Hence, the tribunal confirmed that in the context of ICSID jurisprudence, the second clause of Article 25(2)(b) should not be used to determine the nationality of juridical entities in the absence of an agreement between the parties. Thus, when the claimant was not a national of the host state nor had the parties agreed to treat the claimant as a national of a state other than its state of incorporation, the control test had nothing more to say.33 In light of the respondent’s argument concerning the equitable doctrine of veil piercing, (“lifting the corporate veil” or “disregarding the legal entity”34 by ignoring the claimant’s state of incorporation and determining its nationality according to the nationality of its predominant shareholders),35 the tribunal considered whether such a doctrine could override the terms of the BIT and cause it to deny jurisdiction. As far as the doctrine was concerned, the tribunal, as others tribunals had always done, turned to Barcelona Traction as a seminal case. In that case, the International Court of Justice (ICJ) held that the principle of veil piercing was internationally and domestically admissible only in certain circumstances and for certain purposes, for example, to prevent misuse of the privileges of legal personality (as in a certain case of fraud or malfeasance), to protect third persons such as creditors or purchasers, or to prevent the evasion of legal requirements or of obligations. The tribunal, however, found that the respondent had not made a prima facie showing that the claimant had engaged in any such conduct. It had not shown or even suggested that the claimant had used its status as a Lithuanian juridical entity to perpetrate fraud or engage in malfeasance exemplified by the Barcelona Traction case.36 Therefore, the tribunal denied piercing the corporate veil.37 31

Id. at ¶ 45.

32

Id. at ¶ 25.

33

Id. at ¶¶ 46–50.

W. MICHAEL REISMAN ET AL., INTERNATIONAL LAW IN CONTEMPORARY PERSPECTIVE 450 (2d ed. 2004). 34

35


36

Id. at ¶ 55.

37

Id. at ¶¶ 55, 56.


In conclusion, the tribunal affirmed that economic criteria were irrelevant since the parties had specifically identified the country of legal establishment as the criterion to be applied and had not chosen to subordinate their consent to ICSID arbitration to any other criteria. Thus, the tribunal was obliged to respect the parties’ agreement.38 Furthermore, when the respondent further argued that the claimant had not made an investment in Ukraine as defined by the BIT, arguing that the claimant had not proved that the capital originated from outside Ukraine,39 the tribunal found that the source of capital was also irrelevant in this case. In the absence of a capital source requirement to the definition of investment, the tribunal denied imposing the restriction proposed by the respondent on the scope of covered investment.40 VI. DISSENTING OPINION The tribunal chair disagreed with his colleagues’ analysis and found that their approach “was at odds with the object and purpose of the Convention and might jeopardize the institution’s future.”41 The chair strongly opined that the source of capital was not only relevant but also decisive. Focusing on the undisputed fact that the nationals of Ukraine owned 99 percent of the shares of Tokeles and comprised two-thirds of the management, he argued that while the dispute was superficially between Ukraine and a national of Lithuania, it was indeed a dispute between Ukraine and its nationals.42 On the basis of the foregoing considerations, he disagreed with the majority of the tribunal in three aspects: (1) the relationship between the BIT and the Convention; (2) the importance of the economic criteria in determining ICSID’s jurisdiction; and (3) the apprpach adopted by the majority of the tribunal in interpreting the second clause of Article 25(2)(b). First, considering the relationship between the provisions relating to ICSID jurisdiction contained in the BIT and the Convention, it was Article 25 of the Convention not Article 1(2) of the BIT that should be considered first. Only if the tribunal had reached the conclusion that it

38

Id. at ¶ 63.

39

Id. at ¶ 72.

40

Id. at ¶ 77.

Tokios Tokeles v. Ukraine, Dissenting Opinion, ICSID Case No. ARB/02/18, ¶ 1 (Apr. 29, 2004), available at http://www.worldbank.org/icsid/cases/awards.htm. 41

42

Id. at ¶ 10.


had jurisdiction under the Convention would it have had to examine the jurisdictional requirements under the BIT.43 This meant that the BIT could not extend the jurisdiction on the tribunal beyond the jurisdiction bestowed upon it by the Convention. The contracting parties to the BIT were free to limit the jurisdiction of ICSID (compared to the scope set forth in the Convention), but they could not extend the jurisdiction of ICSID tribunals beyond the provisions of the Convention.44 Second, the chair of the tribunal believed that the economic criteria should have been critical to the tribunal’s decision on jurisdiction. He found that the origin of capital was not only relevant but also decisive. The tribunal’s lack of consideration of this issue was tantamount to giving no effect to the letter and spirit as well as the object and purpose of the Convention.45 Third, the majority of the tribunal and the dissenting chair disagreed on how broadly Article 25(2)(b) should be interpreted.46 The dissident argued that the object and purpose of such a provision was to have the reality of foreign investment prevailed over its legally domestic character when an investment was made through the channel of a domestic corporation. In other words, its object and purpose was to give effect to the genuinely international character of an apparently national investment and to prevent a genuinely foreign investment from being deprived of the protection of the ICSID mechanism because of its domestic legal structure. By the same token, giving effect to the economic reality over the legal structure would have led the tribunal to decide that the investment made in Ukraine by the Ukrainian nationals with Ukrainian capital, even through a Lithuanian corporation, could not benefit from the ICSID mechanism. Doing so was not extending the control test but giving effect to a provision whose rationale was to grant the protection of the ICSID procedures only to genuine international investments.47 He noted that according to a drafting history of the Convention, the ICSID arbitration mechanism was meant for international investment disputes arising between states and foreign investors, not for investment dis43

Id. at ¶¶ 14 and 29.

44

Id. at ¶ 13.

45

Id. at ¶ 20.

46

Id. at ¶ 22.

47

Id. at ¶ 23.


putes between states and their own nationals.48 By applying the economic approach, he concluded that the tribunal had no jurisdiction under Article 25(2)(b) of the Convention. Therefore, there was no need to consider the jurisdiction under Article 1(2) of the BIT.49 VII. ABSENCE OR INADEQUACY OF REASONING In the author’s view, although the decision on jurisdiction and the dissenting opinion of this case seem to meet the minimum standard of reasoning on the whole, certain aspects exhibit either an absence or inadequacy of reasoning. A.

The Determination of Tokeles’s Nationality

The tribunal correctly determined its jurisdiction ratione personae in accordance with the applicable law (lex lata). However, it would have been preferable if the tribunal had provided more reasons why the criterion for determining nationality contained in the BIT should be honored; whereas, the acquisition of nationality falls initially under domestic law of states. The tribunal should have reasoned why the definition of nationality contained in the BIT was consistent with international law, including the ICSID Convention, by examining the question of how a prima facie domestic matter (nationality) could have an effect on international consequence? The tribunal would have received more cogency if it had approached the question of nationality in accordance with these steps: 1.

2.

Recognize that Article 25 of the Convention sets forth the jurisdiction of ICSID tribunals, but does not provide the specific criteria for determining the nationality of a juridical person. Note that tribunals have to consider traditional international law that also fails to declare a single decisive rule relating to the acquisition of corporate nationality.50 International law only refers to domestic laws of each state concerned,

48

Id. at ¶¶ 5, 19.

49

Id. at ¶ 29.

MOSHE HIRSCH, THE ARBITRATION MECHANISM OF THE INTERNATIONAL CENTRE FOR SETTLEMENT OF THE INVESTMENT DISPUTES 81(1992).

50

THE


3.

4.

5.

51

accepting that the issue of nationality is a prima facie domestic matter.51 Note that according to state practice, there have been three criteria applicable for determining the nationality of a juristic person: (a) the state of incorporation rule,52 (b) the state of seat rule,53 and (c) the control test,54 which can be adopted separately or cumulatively.55 Note that in the domestic arena, the application of the different tests does not seem to cause problems since the juridical person in question will fall within the jurisdiction of its national state, whichever test is applied. At the international level, however, conflicts arise when a state claims the nationality granted to its nationals against another state or when a juridical person with multiple nationality attempts to be a beneficiary of more than one nationality. Explain that in order to avoid ambiguity and controversy, each state has a right to adopt any reasonably specific criteria to define whom it considers as its nationals in domestic

IAN BROWNLIE, PRINCIPLES OF PUBLIC INTERNATIONAL LAW 373 (6th ed. 2003).

The state of incorporation rule provides that a juridical person will have the nationality of the place of its incorporation, that is, the state, in accordance with its laws, where the juridical person is incorporated. This test is adopted by AngloAmerican common law countries. 52

53 The state of seat rule (siège social) provides that a juridical person will have the nationality of the place of its seat, that is, the state where its center of management is located. This test is applied by civil law countries. Remarkably, the state of incorporation rule and the state of seat rule have a close connection because a corporation’s statutory seat is usually located in the state of incorporation. However, if the statutory seat of the company does not correspond to the actual location of corporate administration, the “real” seat (siège social rèel) may be applied to determine the nationality of that juridical person. 54 The control test or the state of nationality of controlling shareholders was initially applied during wartime for distinguishing the enemy property. This test had merely relied on share ownership. Then, it was expanded to cover the nationality of principal corporate officers, the exercise of controlling influence through means other than shareholding and the presence of substantial though not necessarily controlling interest. In the post-war period, the test has remained in many countries in measures concerning the economic regulation of foreign trade and investment for political or economic purposes. 55 A. A. Fatouros, National Legal Persons in International Law, in 3 R. BERNHARDT ED., ENCYCLOPEDIA OF PUBLIC INTERNATIONAL LAW 495–96 (1992).


6.

legislation and then internationalize such criteria by entering into international agreements including nationality provisions with other states. Find that when such criteria are reasonable, tribunals should respect the will of states and adopt it for determining their jurisdiction.

In other words, the tribunal should have elucidated that it honored the definition of nationality in the BIT because it was the will of the parties, which was also consonant with general state practice. It could have done so by drawing a bridge between acquisitions of nationality and its consequences, based upon the notion that “although the grant of nationality is for each state to decide for itself in accordance with its own laws, the consequences as against states of this unilateral act occur on international plane and are to be determined by international law.”56 This should have been clearly set out unless the tribunal had assumed that the parties agreed on this point as a matter of international law. In addition, it is unconvincing to decide the case by way of reference, citing the principle accepted in a previous award by an altogether different tribunal.57 Doing so without further rationalization is tantamount to the acceptance of the doctrine of precedent or stare decisis, which is not recognized in international law.58 For instance, with the ICJ decision in Barcelona Traction that “the traditional rule attributes to the right of diplomatic protection of a corporate entity to the States under the law of which it is incorporated and it has its registered office,”59 the tribunal easily confirmed the consistency of the definition of corporate nationality in the BIT with the ICSID Convention and the approach in international law. Taking a teleological approach into consideration, the ICJ’s goal is aimed at maintaining international peace and security, whereas the purpose of the ICSID is narrower—to facilitate foreign investment. The question is why the tribunal accepted the principle applied to the state-state dispute resolution mechanism in the context of diplomatic protection under customary international law to decide the dispute concerning the protection of the rights of investors under the BIT relating to the protection of investments before the investor-state dispute resolution mechanism. There was no clear answer to this question. 56

OPPENHEIM, supra note 1, at 853.

57

Tokeles, supra note 8, at ¶¶ 40 and 42.

58

MALCOLM N. SHAW, INTERNATIONAL LAW 103 (5th ed. 2003).

59



B.

Who Should Be the First Custodian of Public Policy?

In the present international investment climate, situations where nationals of one state establish and own a company in another state are commonplace. Such a company may make an investment inside or outside its state of incorporation including the national state of its founders. This situation is not beyond the expectation of states. The tribunal correctly pointed out that host states realize the significance of this circumstance and are well prepared to deal with it, at least with respect to investment protection and benefit. They do so by signing investment treaties with the provisions concerning the status of such a company in order to restrict or expand its access to ICSID tribunals. For example, the BIT between the United States and Argentina contains the so-called “denial of benefits” provision which states that “[e]ach Party reserves the right to deny to any company of the other Party the advantages of this Treaty if (a) nationals of any third country, or nationals of such Party, control such company and the company has no substantial business activities in the territory of the other Party.”60 As for the BIT between Ukraine and Lithuania, it contains the contrary provision defining investors to include “any entity or organization established under the law of any third State which is, directly or indirectly, controlled by nationals of that Contracting Party or by entities having their seat in the territory of that Contracting Party. . . .”61 These provisions explicitly reaffirm that states are aware of the variation in forms of business structure and take them into account when considering whether they should be regarded as investors benefited by the BIT and the Convention. These restrictive and expansive provisions cannot be applied interchangeably. Therefore, in the absence of any of these specific provisions, it should be assumed that the contracting parties consider them irrelevant. The tribunal did not adopt this line of reasoning to consider the consequential effect of its decision on the public policy of Ukraine. After deciding that Tokeles was a Lithuanian investor, the tribunal should have 60 Treaty between the United States of America and the Argentina Republic Concerning the Reciprocal Encouragement and Protection of Investment art. 1(2) (Nov. 14, 1991), available at http://www.unctad.org/sections/dite/iia/docs/bits/ argentina_us.pdf. 61 Agreement between the Government of Ukraine and the Government of the Republic of Lithuania for the Promotion and Reciprocal Protection of Investments, art. 1(2)(c) (Feb. 8, 1994).


reasoned in response to the fact that 99 percent of Tokeles’ shares and two-thirds of its management were in the hands of the nationals of the host state, Ukraine. This circumstance raises concerns about public interest or public policy of the host state. For instance, in the situation where nationals of state A incorporate a company in state B and run it with capital they have earned from illegal business in state A, such a company will have a standing to bring a claim to the ICSID if it makes an investment in state A, a national state of its creators, and an investment dispute arises. As a result, it appears to suggest that misfeasors who are a criminal in the view of state A have the right to bring the claim against state A through the channel of their foreign entity. In response to such an entirely plausible hypothesis, the tribunal should have stated clearly that it was the tribunal itself or the contracting states that had to be concerned first about a consequential effect on public policy. It might have applied the same approach—the absence of the express intent of the contracting parties on the issue suggested that they considered it irrelevant and without any possible effects on public policy. In doing so, the tribunal could have concluded that it had no duty to fill imaginary gaps not created by the parties. This kind of reasoning, irrespective of its controversial accuracy, is not completely unknown. Judges Spender and Fitzmaurice stated clearly in 1962 that international law could best fulfill its goal by separating itself from social policy. Considerations of a non-judicial, social, or humanitarian character were political matters that should not be allowed to deflect judges from their duty to reach a conclusion on a strictly legal basis.62 C.

Barcelona Traction: A De Facto Precedent or an Anachronism with Respect to this Case?

The reasons provided by the tribunal to deny piercing Tokeles’ corporate veil did not shed any light on this question. The tribunal relied upon Barcelona Traction and ruled that there existed no grounds to have the corporate veil pierced.63 It did not even clarify why Barcelona Traction, a case decided in the context of diplomatic protection, should play any role in Tokeles. Briefly, the ICJ held that Belgium, a national state of the majority of shareholders of a company established in Canada, could not exercise the right of diplomatic protection against Spain for the damage 62 South West Africa, (Liber. v. S. Afr.), 1962 I.C.J. 319, 466 (Dec. 21) (joint dissenting opinion of Judges Spender and Fitzmaurice). 63

Tokeles, supra note 8, at ¶¶ 53–56.


done to the company. It is without doubt that the circumstances in Barcelona Traction and in Tokeles are different. Whereas Barcelona Traction was primarily concerned with the unsuccessful protection of shareholders by means of diplomatic protection, the present case focused on the protection of the company, as an investment through the BIT. Even in the context of shareholders’ protection, some scholars, Professor Schreuer for example, are of the view that Barcelona Traction is no longer an authority for the general proposition that shareholders enjoy no protection under international law.64 This is because the ICJ in ELSI allowed the United States to bring a claim against Italy on behalf of the U.S. shareholders of a wholly owned Italian company.65 Comparing Barcelona Traction and ELSI with Tokeles, it was ELSI and Tokeles that seemed to arise from a similar basis. Both of them were based on the treaties: the Treaty of Friendship, Commerce, and Navigation (FCN) and the Lithuania-Ukraine BIT, respectively. Hence, it is unclear why the Tribunal did not discuss ELSI instead. In a more recent arbitral award, the applicability of Barcelona Traction to cases of foreign investment protection was explicitly denied. The GAMI tribunal articulated that it “does not accept that Barcelona Traction established a rule that must be extended beyond the issue of the right of espousal by diplomatic protection.”66 Furthermore, one commentator has gone so far to suggest that since the goals of the ICJ and ICSID are different, ICSID tribunals do not need to follow ICJ decisions. To catch up with the velocity of changes in international investment and the variety of business organization, ICSID tribunals should determine the corporate nationality in accordance with the nationality of its shareholders and managers.67 D.

Was the Date of Tokeles’s Establishment an Issue in this Case?

Apart from failing to provide any reason why Barcelona Traction was relevant to this case, the way in which the principle of veil piercing was 64 See, e.g., Christoph Schreuer, Shareholder Protection in International Investment Law, at 3 (2005), available at http://www.univie.ac.at/intlaw/pdf/csunpublpaper_2.pdf. 65

Elettronica Sicula S.p.A. (ELSI) (U.S. v. Italy), 1989 I.C.J. 15 (July 20).

GAMI Investments, Inc. v. Mexico, Final Award, at ¶ 30 (Nov. 14, 2004), available at http://ita.law.uvic.ca/alphabetical_list.htm. 66

67

CUSE

Katherine E. Lyons, Piercing the Corporate Veil in the International Arena, 33 SYRAJ. INT’L L. & COM. 523, 545 (2006).


applied lacked cogency. The tribunal stated that “the Claimant manifestly did not create Tokeles for the purpose of gaining access to ICSID arbitration under the BIT against Ukraine, as the enterprise was founded six years before the BIT between Ukraine and Lithuania entered into force.”68 By way of reverse engineering, did the tribunal mean that if Tokeles had been incorporated after the BIT’s entry into force for the purpose of obtaining access to ICSID, it would have tended to pierce the veil, ignoring the state of incorporation and looking through the nationality of the shareholders to determine the nationality of Tokeles? It cannot be the case that the mere creation of a company in order to obtain access to ICSID is tantamount to an abuse or misuse of the privilege of legal personality. Indeed, it is very normal for companies to be incorporated outside a national state of its founders in order to gain benefits or to comply with the host states’ laws and regulations. Investors can employ any legal or political strategy in doing business. Therefore, there seems to be no reason to prohibit investors from planning for dispute settlement resolution including that under the ICSID regime in the absence of the explicit will to the contrary of the states concerned. Companies may benefit from the ICSID Convention regardless of whether they are preexistent or created for the purpose of the Convention unless otherwise stated in the relevant BIT. Nevertheless, it has been suggested that tribunals should consider the equitable factors to delimit companies’ access to the international dispute settlement mechanism. The first factor is timing. Tribunals should consider when a company has acquired the nationality that will allow access to investment arbitration. Second, they should focus on whether the company simultaneously claims benefits as a national of the state it is bringing a claim against as a foreign investor.69 With respect to the first factor, no panel, at least in cases brought under the North American Free Trade Agreement (NAFTA), has done so.70 Considering ICSID jurisprudence, the tribunal in Tunari ruled that the mere changing of corporate structure of the claimant did not bar the jurisdiction of the tribunal although such restructuring happened after 68


Judith Wallace, Corporate Nationality, Investment Protection Agreements, and Challenges to Domestic Natural Resource Law: The Implications of Gold’s NAFTA Chapter 11 Claim, 17 GEO. INT’L ENVTL. L. REV. 365, 388 (2004–2005). 69

70

Id.


the claimant entered into an investment contract and after the entry into force of the BIT. Though the respondent argued that the tribunal should take into account the timing and the intent of the restructuring of the claimant’s ultimate controlling company in order to be able to benefit from the BIT, the tribunal held that the claimant who did not have access to ICSID arbitration before its restructuring could rely on the nationality of its newly established intermediate controlling company to meet the jurisdictional requirement in the BIT and the Convention.71 E.

The Standing for Economic Criteria

The line of reasoning of the dissenting arbitrator’s opinion is based primarily upon a particular conception of public international law. His point of view seems reflect lex ferenda, trying hard to protect the state by imposing upon the investor another criterion not found in the BIT. The view that the source of capital was not only relevant but also critical for determining the international character of an investment is impractical and has no support from the texts of the relevant treaties and from the concept of international investment. Generally speaking, what constitutes an investment of international character varies. It includes the nationality of investors and the place of investment. As an illustration, if a national of state A makes an investment in state B with capital he borrows from a financial institution in state B, it will appear that his investment still falls within the category of international investment. By contrast, if a national of state A makes an investment in his national state with capital from a financial institution abroad, it cannot be the case that he is considered a foreign investor. If the source of capital test is applied, the character of the investment in these examples will be unpredictable. This is because it is plainly unfeasible to trace the exact route of capital in the present world of free trade and high capital mobility.72 We can observe the structure of a company to locate its intermediate and ultimate controlling companies or individuals, but we cannot search for the original, intermediate or ultimate source of capital. To engage in such a pursuit as part of making legal determinations will be the same as trying to determine the starting and ending points of a circle.

71 Aguas del Tunari, S.A. v. Republic of Bolivia, Decision on Respondent’s Objection to Jurisdiction, ICSID Case No. ARB/02/3, ¶¶ 156–181, 206–264 (Oct. 21, 2005), available at http://ita.law.uvic.ca/alphabetical_list.htm. 72 Robert Wisner & Nick Gallus, Nationality Requirements in Investor-State Arbitration, 5 J. WORLD INV. & TRADE 927, 944 (2004).


Moreover, under some definitions of foreign direct investment (as only one kind of international investment), it is clear that the source of capital is irrelevant. First, Professors Trebilcock and Howse define foreign direct investment as “ownership and (normally) control of business or part of a business in another country.”73 Second, from the perspective of the World Trade Organization Secretariat, foreign direct investment occurs “when an investor based in one country (the home country) acquires an asset in another country (the host country) with the intent to manage the asset.”74 According to these definitions, the decisive elements of international investment are investors who make money flow and the place where their investment is made. Based upon the concept that the evolution of international investment law moved toward the idea that the responsibility of the state would arise if it did not treat aliens in accordance with a minimum standard of treatment, it is the nationality of investors that pointed out their status as aliens and international character of their investment.75 In the view of Gus Van Harten and Martin Loughlin, “[a]n investor can become foreign by establishing a holding company or by paper transfer of assets among entity within its corporate structure, without any commitment of new capital to the host economy.”76 Concurring with this view, ICSID jurisprudence takes the proposition that an investment can exist without a physical transfer of funds.77 Economic criteria may prevail over legal ones only in cases where the contracting parties state so explicitly. The dissenting arbitrator was right that the purpose of the second clause of Article 25(2)(b) was to have the reality of foreign investment prevailed over its legally domestic character. 73 MICHAEL J. TREBILCOCK & ROBERT HOWSE, THE REGULATION OF INTERNATIONAL TRADE 439 (3d ed. 2005). 74 World Trade Organization Secretariat, Trade and Foreign Direct Investment, PRESS/57, at 6 (Oct. 9, 1996). 75

M. SORNARAJAH, THE INTERNATIONAL LAW

ON

FOREIGN INVESTMENT 9 (2d ed.

2004). 76 Gus Van Harten & Martin Loughlin, Investment Treaty Arbitration as a Species of Global Administrative Law, 17 EUR. J. INT’L L. 121, 139 (2006). 77 See, e.g., Fedax N.V. v. the Republic of Venezuela, Decision on Jurisdiction, ICSID Case No. ARB/96/3, ¶ 41, (July 11, 1997), available at http://ita.law.uvic. ca/alphabetical_list.htm; Ceskoslovenska Obchodni Banka, A.S. v. the Slovak Republic, Decision on Jurisdiction, ICSID Case No. ARB/97/4, ¶ 78 (May 24, 1999), available at http://ita.law.uvic.ca/alphabetical_list.htm.


However, it appears that, for whatever reason, he did not consider the entire provision of the ICSID Convention, which states clearly that an agreement between the parties is required in order to make an economic reality prevail. VIII. CONCLUSION Regardless of the authorities under which Tokeles is decided, sufficient reasoning is indispensable. De facto precedents cannot be relied upon as per se principles that obviate the need for adequate reasoning. Moreover, it should be kept in mind that one decision cannot be a precedent unless its reasoning is persuasively articulated. Even a well-reasoned precedent is not automatically applicable unless the circumstances of the cases are comparable. On the basis of the aforementioned notion, it seems that the decision on jurisdiction meets the minimum reasoning requirement except in the part dealing with the doctrine of piercing the corporate veil. Still, the tribunal should have considered and discussed the issue of public policy. This decision could have avoided the accusation of overly literal interpretation if the tribunal (whatever its views on the public policy objections were) had addressed these concerns and pointed to possible ways to reconcile them with the decision on jurisdiction.

CHAPTER 7

THE FINAL AWARD IN OCCIDENTAL V. ECUADOR Aloysius P. Llamzon*

In 2004, an UNCITRAL arbitral tribunal unanimously ruled that the Republic of Ecuador violated key provisions of the U.S.-Ecuador Bilateral Investment Treaty (BIT)1 and awarded over $75 million to Occidental Exploration and Production Company (Occidental), a U.S. investor.2 While the award itself is relatively unremarkable in monetary terms, Occidental’s implications for general investment arbitration are stunning, especially given its potential as a source of precedent. Different observers will have varied reasons to find controversy in Occidental: some will find disturbing its expansion of investor protections under international investment law for being too intrusive upon the regulatory authority of host states. Others will note the decision reached by a subsequent investment tribunal in EnCana v. Ecuador,3 which considered an analogous factual record and adopted almost diametrically opposed findings.4 This * LL.M. 2006; J.S.D. candidate, Yale Law School; A.B., J.D., Ateneo de Manila University. I am grateful to Professors W. Michael Reisman and Guillermo Aguilar Alvarez for their guidance in this chapter and beyond. I am also obliged to my contemporaries in the 2005–06 seminar on international investment law at Yale Law School, and to George S. Georgiev and Rocío Digón. 1 Treaty Between the United States of America and the Republic of Ecuador Concerning the Encouragement and Reciprocal Protection of Investment (U.S.Ecuador), signed Aug. 27, 1993, entered into force May 11, 1997, available at http://www. sice.oas.org/bitse.asp. [hereinafter BIT]. 2 Occidental Expl. and Prod. Co. v. Republic of Ecuador, LCIA Case No. UN 3467, Final Award (July 1, 2004) [hereinafter Occidental]. 3 EnCana Corp. v. The Republic of Ecuador, LCIA Case No. UN 3481, Award (Feb. 3, 2006) [hereinafter EnCana]. 4 The EnCana tribunal found, inter alia, that the Canada-Ecuador BIT prohibited the arbitrability of tax measures except on two grounds, neither of which were applicable under the facts. On the issue of expropriation, the EnCana tribunal assumed (without directly ruling), in contrast to Occidental, that tax refunds may be considered investments subject to direct expropriation, although the facts therein ultimately did not warrant such a determination. (Part of the reason for the divergence of opinion

211


notwithstanding, the core rulings of the Occidental tribunal were upheld on judicial review by the U.K. High Court.5 The most unsettling aspect of the award, however, lies in what is left unsaid and unsubstantiated rather than in its legal and policy findings. The paucity of reasoning is most evident in the substantiation of the award’s controlling prescriptions, with the most contentious findings suffering from the least articulation. Accordingly, this chapter aims not so much to criticize the outcome of the Occidental arbitration as to critically appraise the soundness of the reasoning employed by the tribunal to justify its conclusions. By unpacking the major fault lines of the award, the more glaring lapses in the tribunal’s reasoning will be exposed. The chapter proceeds as follows: Section I provides a factual and procedural overview to the dispute. Section II considers three aspects of the tribunal’s reasoning on the jurisdictional issues presented. Section III proceeds to focus on two aspects of the tribunal’s reasoning on the substantive issues. A brief conclusion follows. I.

BACKGROUND TO THE DISPUTE

The core dispute arose out of differing interpretations by Occidental and Ecuador of Ecuador’s value-added tax (VAT) law. Occidental argued that it had to pay output VAT that it was neither contractually nor statutorily liable for. A.

Factual Background

The factual antecedents are uncomplicated: In 1999, Occidental concluded a “participation contract” with the state-owned oil company Petroecuador to undertake the exploration and production of oil in the

between the EnCana and Occidental tribunals may be ascribed to the differences in language contained in the respective BITs in question). 5 See The Republic of Ecuador v. Occidental Expl. & Prod. Co., Case No. 04/656 (Mar. 2, 2006), [2006] EWHC 345 (Comm). While the Court (Aikens, J.) held that the dispute involved a “matter of taxation” and would thus have ordinarily been excluded from the application of the BIT, it found that the value-added tax (VAT) payments “had reference to the performance of the obligations of the Contract” between Occidental and the Republic. The Court found fault in some of the tribunal’s reasoning, but ultimately affirmed that the arbitral tribunal acted within its jurisdiction. In addition, the Court also dismissed Ecuador’s alternative argument, ruling that even if the tribunal had exceeded its powers, the Republic suffered no “substantial injustice.”

The Final Award in Occidental v. Ecuador • 213

Ecuadorian Amazon Region.6 In the course of its exploitation activities, Occidental paid VAT on various goods and services. From 1999 to mid2000, the company was able to claim a refund of these taxes.7 Beginning in 2001, however, Occidental’s applications for tax reimbursement were denied by Ecuador’s tax collection agency, the Servicio de Rentas Internas (SRI). Previous resolutions granting reimbursements were revoked shortly thereafter.8 The SRI’s primary rationale behind the policy change was that the “Factor X” formula under Section 8.1 of the participation contract, which represented the percentage of oil production that would be allocated to Occidental,9 was already calculated in a manner that included VAT reimbursements. Ecuador also disputed Occidental’s contention that its tax legislation provided for a right to tax credits for all VAT paid in local acquisition, use, and importation of goods and services.10 The SRI’s reversal of policy vis-à-vis Occidental was not unique, as VAT refund requests of other oil companies doing business in Ecuador, both foreign and local (and including Petroecuador itself), were similarly rejected.11 B.

Procedural Background

Following domestic law, Occidental filed four lawsuits in Ecuadorian tax courts prior to the expiry of the 20-day prescriptive period for filing actions contesting adverse SRI resolutions. It then decided not to continue submitting VAT refund applications with the SRI, believing this to be futile.12 With decisions on the matter still unresolved in national courts,13 Occidental commenced arbitration proceedings under the U.S.Ecuador BIT on November 11, 2002, claiming that Ecuador, through the SRI, had breached BIT guarantees protecting its investment.14 6

Occidental, supra note 2, at ¶ 27.

7

Id., ¶ 32.

8 9 10

Id. Id., ¶ 28. Id., ¶¶ 31, 34.

See, e.g., EnCana, supra note 3; Occidental, supra note 2, at ¶ 35 (noting that discussions were held between Ecuadorian agencies and “other companies similarly affected”). 11

12


Indeed, at the time of the award, “[d]ecisions on the matter (were) still pending before the courts (of Ecuador).” Id., ¶ 4. 13

14

Id., ¶ 6.


In particular, Occidental invoked four BIT protections: the Republic had, in Occidental’s view, breached the obligations (1) of fair and equitable treatment; (2) of treatment not less favorable than that accorded to Ecuadorian exporters; (3) not to impair by arbitrary or discriminatory measures the management, use, and enjoyment of Occidental’s investment; and (4) not to expropriate directly or indirectly all or part of that investment.15 The Republic refuted each of these assertions and added a number of jurisdictional objections: (1) that by virtue of the “fork in the road” provision in the BIT, Occidental’s suits before Ecuadorian courts constitutes an irrevocable choice that precludes the submission of the dispute to arbitration; (2) Article X, which places most matters of taxation outside the ambit of investor protection provided under the BIT, precludes the tribunal from hearing the claim; and (3) the lack of any direct or indirect expropriation renders Occidental’s claims inadmissible.16 The sections that follow shall examine, in turn, the tribunal’s reasoning in relation to the jurisdictional and substantive issues present, focusing only on those instances where the lack of adequate reasoning was most apparent. II.

REASONING ON JURISDICTION

A.

Form Obscuring Reasoning

A preliminary observation bears mention at this point. Throughout the award, the structure of argumentation took the form of an outline, beginning with a listing of all the issues presented by the parties on jurisdiction. Each claim was then broken down to two parts—the first recalled the arguments raised by the parties in support or resistance of the particular claim or affirmative defense at issue. This would then be followed by a discussion of the tribunal’s findings. The merits were also presented in the same manner. While this may have resulted in a neat and ostensibly comprehensive consideration of the parties’ arguments, the approach taken did seem at times artificial and arguably facilitated the abdication of the tribunal’s duty to discuss issues in a more logical, methodical, and reasoned manner. To take one example, one may assume (as the EnCana tribunal did) that a threshold issue should have been whether VAT refunds were 15

Id., ¶ 36.

16

Id., ¶ 37.


“investments” within the meaning of the BIT in the first place. It seemed highly curious that the tribunal’s finding that the tax refunds at issue did not amount to an “investment”17 was not considered a fundamental, jurisdictional question that rendered the rest of the award functionally unnecessary, as the absence of an “investment” within the meaning of the BIT would preclude resort to the tribunal’s jurisdiction. Instead, the finding that a lack of investment existed was imbedded within the discussion on expropriation, and even then, was treated as only one among many issues. The award thus seemed to be drafted almost entirely as a reaction to the parties’ arguments, rather than a decision whose reasoning depended largely on the tribunal’s initiative. B.

Treatment of Expropriation as a Jurisdictional Issue

Perhaps the most glaring lack of reasoning in the entire award lies in the tribunal’s findings on expropriation. In virtually all foreign investment jurisprudence, expropriation is treated as a matter of substantive law and discussed in the merits phase. Occidental acknowledged this at the outset: “[a] claim of expropriation should normally be considered in the context of the merits of a case.”18 Startlingly, the next sentence then reads: “[h]owever, it is so evident that there is no expropriation in this case that the Tribunal will deal with this claim as a question of admissibility.19 No authority was given and no reason proffered to justify this departure from existing law, apart from the supposed “evident” lack of merit to the claim of expropriation. To reasonably conclude that expropriation is absent in any given case, however, delving into the parties’ substantive arguments, evidence, and their relation to contemporary standards of international investment law is unavoidable. Indeed, the tribunal did precisely that in paragraphs 81 to 89 of the award. Thus, if the tribunal’s intention was to expeditiously dismiss the charge in quasidemurrer fashion for self-evident absurdity, the subsequent discussion plainly negated whatever perceived benefit the jurisdictional treatment of expropriation may have had. That the tribunal advanced no good explanation for this very substantial break from established practice sug17 Id., ¶ 86 (“The Tribunal, however, is not persuaded by the Claimant’s arguments that in this case there has been an expropriation. It is not tenable to argue that there can be ‘no doubt that under the Treaty the Refund Claim is an investment per se.’ However broad the definition of investment might be under the Treaty it would be quite extraordinary for a company to invest in a refund claim.”). 18

Id., ¶ 80.

19

Id.


gests careless legal craftsmanship. Indeed, the U.K. High Court was critical of the Occidental tribunal’s rather arbitrary reassignment of expropriation from substantive law to a matter of jurisdiction.20 While moving expropriation from an issue on the merits to one of jurisdiction may seem innocuous, the subsequent proceeding initiated by the Republic before English courts demonstrate that this is far from so. Treating expropriation as a jurisdictional matter gave the Republic some leeway to argue that the award’s findings on expropriation were not res judicata and thus non-binding upon courts. Although the Court rejected that argument, having found that the tribunal’s expropriation decision was made on the merits,21 nothing precludes future parties from resorting to Occidental as authority for making this rather specious argument. The other bothersome aspect of the tribunal’s reasoning on expropriation relates to its disregard of the argument that tax refunds are part of Occidental’s investment and thus protected under the BIT.22 The treaty’s definition of “investment” includes “intangible property, including rights, a claim to money associated with an investment and any right conferred by law.”23 Despite the definition’s plain meaning and an acknowledgment that “[t]axes can result in expropriation,”24 however, the tribunal was “not persuaded . . . (that) there has been an expropriation.”25 To the tribunal, “[h]owever broad the definition of investment might be under the Treaty it would be quite extraordinary for a company to invest in a refund claim.”26 20

Ecuador v. Occidental, supra note 5, at ¶ 136. It is, in my view, equally clear that the Tribunal was dealing with the merits of the expropriation claim in paragraphs 78–92 of the award. The phraseology of paragraphs 86 to 89 make this plain. The Tribunal decided that, because it was “so evident that there is no expropriation in this case,” therefore it should deal with this claim as a matter of “admissibility.” English lawyers may find that a curious word to use in the circumstances. But that does not matter. In my view it is evident, looking at the substance of paragraphs 78–92 of the award, that the Tribunal was making an award on the merits of the expropriation claim. It did not decide it had no jurisdiction to entertain the claim.

21

Id.

22

See Occidental, supra note 2, at ¶ 80.

23

Id., citing BIT, supra note 1, art. I(1)(i),(iii), and (v) (emphasis added).

24

Id., ¶ 85.

25

Id.

26

Id., ¶ 86.


The reasoning employed here is largely indefensible. As reprinted above, the text of the BIT itself includes within the definition of investment a “claim to money associated with an investment.” Moreover, investor-state tribunals applying international investment law have increasingly recognized both the expanded scope of the term “investment”27 and acts that reduce the economic potential of a given investment as a mode of indirect expropriation.28 To follow the tribunal’s reasoning to its logical conclusion, no redress would be available for relatively “small” multimillion dollar claims when the total investment of the investor runs in the billions of dollars. Common experience suggests, however, that relatively small deprivations can easily comprise most or all the profits of a given concern, thereby defeating a project’s raison d’être. In the face of all these putative sources of controlling law, Occidental’s inability or unwillingness to characterize the VAT refunds as an “investment” for dispute resolution purposes was simply not articulated. Overall, one is left with the impression that the lack of ratiocination found here resulted from the tribunal’s dismissive stance towards the expropriation claim in general under the facts. Regardless of what one may think about tax refunds as subject to expropriation claims, the conclusion that no expropriation existed here was by no means self-evident, and the tribunal would have been better served by careful reasoning on the merits, instead of summary dismissal. It also bears stressing that the foreclosure of expropriation as a possible violation by the Republic led the tribunal to frame the substantive issues involved largely within the ambit of the fair and equitable treatment calculus. As discussed in Section III, this in turn led to further problems in reasoning.

27 See, e.g., Fedax NV v. Republic of Venezuela, Decision on Objections to Jurisdiction (July 11, 1997), 5 ICSID REP. 183 (2002), wherein the phrase “titles to money, to other assets or to any performance having an economic value” in the Netherlands-Venezuela BIT was interpreted broadly enough to include within the ambit of “investment” promissory notes issued by the Venezuelan government that were purchased by third parties. (Interestingly, Professor Vicuna was also the President of the Fedax tribunal). 28 See, e.g., Metalclad Corp. v. The United Mexican States, Award, at ¶ 103, 5 ICSID REP. 209 (2000) (“[E]xpropriation under NAFTA includes not only open, deliberate and acknowledged takings of property, such as outright seizure or formal or obligatory transfer of title in favor of the host State, but also covert or incidental interference with the use of property which has the effect of depriving the owner, in whole or in significant part, of the use or reasonably-to-be-expected economic benefit of property even if not necessarily to the obvious benefit of the host State.”).


C.

The “Fork-in-the-Road”

The Republic’s first jurisdictional objection was that Occidental had triggered the “fork-in-the-road” proscription of the BIT through its prior resort to Ecuadorian national courts. Citing a string of ICSID jurisprudence that differentiated between “contract-based claims and treatybased claims,”29 the tribunal ultimately rejected the Republic’s argument, having found that, “[t]o the extent that the nature of the dispute submitted to arbitration is principally, albeit not exclusively, treaty-based, the jurisdiction of the arbitral tribunal is correctly invoked.”30 For all its extensive (and entirely uncontroversial) discussion of prevailing jurisprudence, a number of critical points relating to the “fork-inthe-road,” which appeared to have had strong effect on the decision outcome, were given insufficient treatment. The most critical concerned the issue of “characterization,”31 that is, that a claimant’s mere couching of claims in terms of BIT violations is sufficient to overcome jurisdictional challenges, regardless of whether such claims may also be viewed as relating to the underlying contractual relationship with the host state. The tribunal seemed to approve of the principle almost at the outset of its reasoning on this issue: “[t]he characterization of the dispute by the Claimant probably would suffice alone for the Tribunal to reach a determination on jurisdiction.”32 Instead of discussing the legal or policy issues behind this idea, however, the tribunal almost immediately moved on to other issues.33 29 Id., ¶ 51 (citing the Lauder, Genin, Aguas del Aconquija, CMS, Azurix, and Vivendi (ad hoc) awards). 30

Id., ¶ 57.

31

Id., ¶ 47.

Id. The tribunal supported this view by quoting from the Azurix and SGS v. Pakistan decisions. 32

33 See id., ¶ 48 (“But the fact is that this dispute, its contractual aspects aside, involves a number of issues.”).

More peripheral issues were given greater analysis, however. In paragraph 60 of the award, for example, the tribunal denied the applicability of the “fork-in-the-road” defense because, in their view, “its very definition assumes that the investor has made a choice between alternative avenues. This in turn requires that the choice be made entirely free and not under any form of duress.” Id., ¶ 60. Having found that the 20day prescriptive period of Ecuadorian tax law amounted to such duress, “[t]he Tribunal is of the view that in this case the investor did not have a real choice. Even if it took the matter instantly to arbitration . . . the protection of its right to object to the adverse decision of the SRI would have been considered forfeited before local courts.” Id.,


Overall, Occidental falls within a trend in investment jurisprudence that seems in fact (rhetoric to the contrary notwithstanding)34 to have reduced “fork in the road” provisions to dead letter, as the mere “characterization”35—that is, couching a claim in terms of BIT violations— would for all practical intents be sufficient to overcome jurisdictional challenges. This may indeed be a wise thing to do as a policy matter, as the flipside of the argument—that every BIT dispute can be characterized as a breach of contract, thus triggering the fork whenever any kind of domestic remedy is sought—would have been the clear and unfortunate implication of sustaining the Republic’s jurisdictional objection on this ground. However, if, as one may suspect, carefully calibrated policy determinations factored into their findings, the tribunal should have articulated their decision-making process explicitly instead of leaving readers guessing. By paying lip service to existing jurisprudence and refraining from engaging in policy determinations, the tribunal signaled its unwillingness to depart from established norms. One wishes, however, that such adherence be less mechanized and more tightly reasoned. D.

Taxation Exception

The Republic’s basic argument with respect to taxation was rooted in the plain text of Article X of the BIT, which provides: 1. With respect to its tax policies, each Party should strive to accord fairness and equity in the treatment of investment of nationals and companies of the other Party. 2. Nevertheless, the provisions of this Treaty, and in particular Articles VI and VII, shall apply to matters of taxation only with respect to the following:

¶ 61 (emphasis added). While this may be true, the conclusions drawn seem poorly reasoned. No explanation was proffered as to why there was a legitimate “lack of choice” here, since the logical conclusion of the tribunal’s findings would have been that precisely because BIT claims were entirely separate from domestic causes of action, the preclusion of certain domestic contractual remedies would in no way have impacted the BIT claim itself. 34 Id., ¶ 50 (“This finding of the Tribunal cannot be taken to mean that the death knell has sounded for ‘fork in the road’ provisions of bilateral investment treaties, as the Respondent has argued.”). 35

Id., ¶ 47.


(a) expropriation, pursuant to Article III; (b) transfers, pursuant to Article IV; or (c) the observance and enforcement of terms of an investment agreement or authorization as referred to in Article VI(1)(a) or (b), to the extent they are not subject to the dispute settlement provisions of a Convention for the avoidance of double taxation between the two Parties, or have been raised under such settlement provisions and are not resolved within a reasonable period of time.36 The Republic’s view was that all VAT-related concerns were clearly “matters of taxation” excluded from arbitrability under Article X. Moreover, under Article II of the BIT, which covered the principles of no less favorable treatment, fair and equitable treatment, and the proscription against arbitrary and discriminatory treatment, such concerns were outside the tribunal’s jurisdiction following the second paragraph of Article X.37 The tribunal rejected this argument in its entirety. It first professed to “agree with both parties in that the proper interpretation of Article X must not result in rendering it meaningless.”38 Curiously, however, it then ascribed meaningless-ness to the Republic’s argument that “all matters of taxation are exempted from dispute settlement under the Treaty, with the exception of the specific categories mentioned in Article X,”39 finding that such an interpretation would constrain the BIT “to a quite marginal application.”40 One doubts immediately how a category as broad and developed as expropriation (Article X(2)(a)) can be described as a “marginal” exception. The full weight of the tribunal’s approach towards Article X is found in subsequent paragraphs. Despite the clear non-binding language found

36 Reprinted in id., ¶ 64 (“Is it possible to obtain a direct citation to the treaty, rather than reference the reprinted text from the award? This is obviously not high priority.”). 37

Id., ¶ 65.

38

Id., ¶ 68.

39

Id.

40

Id.


in Article X(1)—each party “should strive to accord fairness and equity” in its treatment of the other’s investors and investments—the tribunal chose to ascribe binding and controlling consequence to the provision. It found, rather euphemistically, that the mention of fairness and equity in the treatment of foreign investment by the host state “is not devoid of legal significance.”41 It then reasoned: It imposes an obligation on the host State that is not different from the obligation of fair and equitable treatment embodied in Article II, even through admittedly the language of Article X is less mandatory. This legal effect is not derogated from by the “nevertheless” proviso with which paragraph 2 opens, as this expression cannot be read to mean that in respect of tax policies the host State could pursue an unfair or inequitable treatment. It only means that such obligation is concerned with the three categories of tax matters therein listed, that is, expropriation, transfers and the observance and enforcement of an investment agreement or authorization.42 Construing Article X(1) in this manner seems both plainly unconvincing and inadequately reasoned. The practical effect of subjecting all tax measures adopted by the Republic to fair and equitable scrutiny is an evisceration of Article X’s efficacy as a control on an investor’s arbitration option when matters of taxation are concerned. Because of Article X’s plain meaning and the serious consequences such a divergent interpretation would engender, the tribunal’s obligation, at a minimum, was to explain in clear terms why the provision should be understood in such a manner. Yet, as the text above suggests, the tribunal did not find that this issue required extensive legal analysis or policy weighing. While acknowledging that Article X(1) was couched in “less mandatory” terms, and that paragraph 2 contained “nevertheless” language to convey the exclusiveness of the grounds enumerated therein, the tribunal seemed content to disregard the text by tersely concluding that Article X “cannot be read to mean that in respect of tax policies the host State could pursue an unfair or inequitable treatment.” Neither authority nor further reasoning was proffered, and this portion of the award is thus open to the criticism that its ultimate import, if followed, would render Article X meaningless. 41

Id., ¶ 70.

42

Id. (emphasis added).


What seems to compound the lack of cohesiveness in this part of the award is the tribunal’s “second consideration”43 for finding jurisdiction despite Article X. After briefly recalling the three exceptions found in Article X(2), the tribunal devoted the rest of the part to examining the applicability of Article X(2)(c), that is, “whether the observance and enforcement of the terms of an investment agreement concerning matters of taxation is at issue in this dispute.”44 Here, the tribunal seems to have contradicted its earlier finding denying application of the “fork in the road” provision of the BIT due to the treaty-based nature of the dispute before the tribunal. Finding that Article X(2)(c) operated in this case, the tribunal concluded that “[e]ven if the Claimant has not characterized the dispute as one concerning the Contract, the objective fact is that the Contract is central to the dispute.”45 The apparent contradiction is evident: the tribunal was, at least by its language in paragraph 73, willing to accept the primarily contractual nature of the dispute for purposes of Article X while simultaneously denying the “central” importance of the contract when it addressed the in “fork in the road” arguments. There may have been cogent reasons for the divergence; justification was needed, however. More serious from a reasoning perspective was that this “second consideration” was in fact the tribunal’s ultimate basis for disregarding Article X. In paragraph 77, the tribunal stated: because of the relationship of the dispute with the observance and enforcement of the investment contract involved in this case, [the Tribunal] has jurisdiction to consider the dispute in connection with the merits insofar as a tax matter covered by Article X may be concerned, without prejudice to the fact that jurisdiction can also be affirmed on other grounds as respects Article X as explained above.46 Thus, while it initially resorted to pure statutory construction, the tribunal appeared to have backpedaled in the end, opting instead to use the “investment agreement” exception under Article X(2) as its primary basis for finding jurisdiction. This is understandable in one sense, as this 43

Id., ¶ 71.

44

Id.

45

Id., ¶ 73.

46



basis for finding jurisdiction rests on much more solid, treaty-consistent grounds.47 Nevertheless, as a matter of legal reasoning, the tribunal may be taken to task for hedging its reasoning so haphazardly. If the “second consideration” was indeed its primary basis for affirming jurisdiction, this portion of the award should have been reasoned that way. If it realized that its argument was weak, as it appears it did by reducing it to an “other ground” in paragraph 77, the tribunal should not have made this argument at all. III. REASONING ON THE SUBSTANTIVE ISSUES While the claimant failed on its jurisdictional objections, it achieved near-complete success on the merits. The tribunal found Ecuador responsible for numerous substantive violations of the BIT, including the guarantees of non-impairment through arbitrary and discriminatory measures,48 national treatment,49 fair and equitable treatment,50 and full protection and security.51 Of these, the first two issues are the most controversial from a reasoning perspective. A.

National Treatment/Most-Favored-Nation Treatment

If the award’s most glaring instance of unreason was its treatment of expropriation, its most dangerous implications lie in the findings on national treatment and most-favored-nation treatment. Article II(1) of the BIT “establishes the obligation to treat investments and associated activities ‘on a basis no less favorable than that accorded in like situations to investment or associated activities of its own nationals or companies, or of nationals or companies of any third country, whichever is the more favorable.’”52 The national treatment issue centered on the interpretation of the phrase “in like situations.” Given the fact that the VAT refund requests of all oil companies producing, operating in, and export47 Indeed, the High Court of the United Kingdom subsequently confirmed that the Occidental tribunal had acted within its jurisdiction primarily on the strength of Article X(2)(C). See supra note 5 and accompanying text. 48

BIT, supra note 1, at art. II(3)(b), reprinted in Occidental, supra note 2, at ¶ 159.

49

Id., art. II(1).

50

Id., art. II(3)(a).

51

Id.

52

Id., ¶ 167.


ing from Ecuador, whether domestic or local, state-owned or private (and indeed, even Petroecuador itself), were all similarly denied by the SRI,53 Occidental was constrained to argue that the Republic’s national treatment obligation under the BIT was to accord it the same VAT treatment as industries entirely unrelated to oil, “particularly flowers, mining and seafood products,”54 which were apparently granted output VAT refunds by the SRI. Occidental would have the phrase “in like situations” interpreted as referring simply to all companies “engaged in exports even if encompassing different sectors.”55 Unsurprisingly, the Republic’s interpretation of “in like situations” was very different, and was based on policy grounds: [it] can only mean that all companies in the same sector are to be treated alike and this happens in respect of all oil producers. The comparison . . . cannot be extended to other sectors because the whole purpose of the VAT refund policy is to ensure that the conditions of competition are not changed, a scrutiny that is relevant only in the same sector.56 In a tersely worded paragraph that belies the revolutionary nature of its findings, the tribunal summarily concluded that “the Claimant is right and its arguments are convincing.”57 “In like situations,” according to the tribunal, “cannot be interpreted in the narrow sense advanced by Ecuador as the purpose of national treatment is to protect investors as compared to local producers, and this cannot be done by addressing exclusively the sector in which that particular activity is undertaken.”58 The tribunal effectively identified the aggregate of exporters in Ecuador, not particular industries, as the relevant set against which charges of discrimination were to be evaluated. No reasoning was advanced to support the conclusion that the Republic’s understanding was “narrow,” nor as to why such a generous interpretation of “in like situations” was proper. Moreover, it was utterly non-sequitur for the tribunal to have argued that protecting foreign investors vis-à-vis domestic companies “cannot be done by addressing 53

Id., ¶ 172.

54

Id., ¶ 168.

55

Id.

56

Id., ¶ 171.

57

Id., ¶ 173.

58

Id.


exclusively the sector in which that particular activity was undertaken,” since in fact, each industry is subject to different economic environments, with different export opportunities and different capital requirements, requiring different incentives and tax regimes. Even absent the policy implications of the decision, however, the reasoning of the award was severely wanting on its own terms: the Republic’s policy argument that national treatment’s essence involved conditions of competition, which could only be made on an industry basis, was never addressed. Indeed, instead of policy or jurisprudential ratiocinations to support its interpretation, the tribunal’s primary syllogistic device was to erect a straw man in the form of the World Trade Organization. In the next paragraph, it compared the BITs “in like situations” language with the “like products” concept of national treatment under General Agreement on Tariffs and Trade/World Trade Organization (GATT/WTO) law. Acknowledging that “like products” under the GATT relate to “the concept of directly competitive or sustainable products,”59 the tribunal stated: In fact, the purpose of national treatment in this dispute is the opposite of that under the GATT/WTO, namely it is to avoid exporters being placed at a disadvantage in foreign markets because of the indirect taxes paid in the country of origin, while in the GATT/WTO the purpose is to avoid imported products being affected by a distortion of competition with similar domestic products because of taxes and other regulations in the country of destination.60 Thus, the tribunal believed that the GATT’s conception of national treatment, which “is to avoid imported products being affected by a distortion of competition with similar domestic products because of taxes and other regulations in the country of destination,” is not analogous to the purpose of national treatment in the BIT context, which is to “avoid exporters being placed at a disadvantage in foreign markets because of the indirect taxes paid in the country of origin.”61 In an attempt to further widen the policy distinction, the succeeding paragraph stated that in the situation of the BIT and this dispute, “no exporter ought to be put in a disadvantageous position as compared to other exporters,” whereas in the case of 59

Id., ¶ 174.

60

Id., ¶ 175.

61

Id.


the GATT/WTO, “the comparison needs to be made with the treatment of the ‘like’ product and not generally.”62 Yet again, no authority whatsoever was cited to substantiate the supposed divergence between international investment and international trade law on national treatment, giving the tribunal’s policy extrapolations an entirely speculative appearance. Even assuming that this is more than a hairline distinction, however, the largely unconvincing reasoning employed is cause for great concern. One can only speak of “exporters being placed at a disadvantage in foreign markets because of indirect taxes paid in the country of origin” when particular exported goods are compared to exports from other importing countries and disadvantages are in fact detected. This, in turn, can only be performed when comparisons between different exporting countries are made concerning similar products. Moreover, the reader is left to guess as to why the BIT should be interpreted to constrain the sovereign right of states to enact differential tax policies in support of different growth strategies across dissimilar industries, while the GATT/WTO should be viewed as responsive and sensitive to these needs, as both branches of law are based essentially on the consent of the same states having the same developmental goals. The final argument advanced in support of the tribunal’s conclusion again concerned the GATT/WTO versus BIT distinction. This time, the argument was strictly textual in nature: “In any event, the reference to ‘in like situations’ used in the Treaty seems to be different from that to ‘like products’ in the GATT/WTO. The ‘situation’ can relate to all exporters that share such condition, while the ‘product’ necessarily relates to competitive and substitutable products.”63 It was thus found that “situations” were different from “products” in that the former must have referred to the aggregate of all exporters within Ecuador because it was a more expansive term than “products.” It is difficult to understand, however, why it cannot be argued that the word “situations” was the language employed by the BIT simply because investments and investors are neither commodities nor “products.” Moreover, it was simply unconvincing to base almost the entire argument on national treatment on a differentiation between the GATT/WTO system and the BIT and not on a real discernment of the policy objectives that animate national treatment under contemporary international investment law and policy. 62

Id., ¶ 176.

63

Id.


Finally, if adherence to text is the order of the day, one wonders why plain textual interpretation of Article II(1) of the BIT, following its ordinary meaning, was not employed by the tribunal. To ascribe to Article II(1) the meaning given to it by the tribunal renders the phrase “in like situations” altogether meaningless. If the provision had read: “treat investment, and activities associated therewith, on a basis no less favorable than that accorded to investment or associated activities of its own nationals,” it may have been plausible to argue that the BIT demanded blanket national treatment for all exporters, local or foreign, regardless of industry. However, the provision states “no less favorable than that accorded in like situations to investment or associated activities of its own nationals.” Thus, the BITs national treatment protection of foreign investment is qualified by the phrase “in like situations,” which, it is submitted, can be sensibly interpreted only by acknowledging the right of the host state to treat different industries (usually composed of nationals and non-nationals alike) in a differential manner to achieve certain goals. This chapter’s primary source of concern with the award’s national treatment argumentation is its unconvincing reasoning, which was made with an almost complete lack of authority or argumentation to justify its findings. It bears mentioning, however, that the substantive implications of the tribunal’s findings are at least equally invidious. The award’s implication if writ large would mean that all investors conducting export activities within a given host state are subject to equal tax treatment regardless of the nature of their respective industries. Following this horizontal rule, host states would be forced to accord the same tax incentives given to large-scale, capital-intensive industries to even the most incidental foreign investor, so long as it is engaged in export activities. Similarly, highly profitable extractive industries exporting their products would be given the same tax incentives that small exporters would be entitled to, if the host state had earlier decided to incentivize the latter. The tribunal did not point to a single state’s practice (nor could it), nor to any national or international judgment that had adopted this interpretation of national treatment. Governments would now be forced to adopt an “all-or-nothing” approach to the regulation of foreign investment, and differential and incremental policy approaches entirely justifiable under domestic and indeed in general international law (particularly the international law on trade) would nevertheless be a violation of international investment obligations, broadly conceived (virtually all bilateral and multilateral investment treaties will contain national treatment guarantees). Indeed, the tribunal seemed unaware or unwilling to acknowledge that


in Pope & Talbot, a prior investment tribunal had categorically recognized that differential treatment did not in itself amount to discrimination.64 One final observation on this part of the award merits discussion. The tribunal would arguably have been better served by relying on the mostfavored-nation principle to strengthen its findings on national treatment, as the phrase “in like situations” was not present in the Spain-Ecuador and Argentina-Ecuador BITs.65 But in the penultimate paragraph of this part of the award, the tribunal dealt with the issue summarily: This finding [of breach of national treatment] makes it unnecessary for the Tribunal to examine whether there were in addition most-favorednation-treatment obligations involved. In view of the fact that the parties have discussed in detail the meaning of Maffezini in this context, the Tribunal believes it appropriate to clarify that that case is not really pertinent to the present dispute as it dealt with the most-favored-nation-treatment only insofar as procedural rights of the claimant there were involved, not substantive treatment as is the case here.66 In one short sentence, the tribunal conflated national treatment with most-favored-nation treatment and stated that because they found that a breach of Ecuador’s national treatment obligation towards Occidental had occurred, most-favored-nation treatment discussions were “unnecessary.” The two grounds are of course of a different nature, and if, as was done in its jurisdictional arguments, the tribunal wanted to support its reasoning with multiple plausible reasons, this area would have been indispensable to buttress the overall argument. Yet it chose not to do so. B.

Impairment Through Arbitrary Measures

Article II(3)(b) of the BIT provides that “[n]either party shall in any way impair by arbitrary or discriminatory measures the management, operation, maintenance, use, enjoyment, acquisition, expansion, or dis64 Pope & Talbot Inc. v. The Government of Canada, Award on the Merits of Phase 2, at ¶¶ 73–104 (Apr. 10, 2001). Interpreting the “in like circumstances” provision under Article 1102 of the North American Free Trade Agreement (NAFTA), the tribunal found that differential treatment given to different softwood lumber companies within different regions was not a violation of national treatment. 65


66

Id., ¶ 178.


posal of investments.” The Republic argued that the claim merely duplicated other claims by Occidental concerning discrimination and fair and equitable treatment. More importantly, the Republic denied in any event that this guarantee was frustrated by the SRI actuations.67 In response, the tribunal found unmeritorious the argument that the management, operation, maintenance, use, enjoyment, acquisition, expansion, or disposal of the investment was impaired, by simply stating: “[i]n fact, it is quite evident from the record that the Claimant continues to exercise all these rights in a manner which is fully compatible with the rights to property.”68 Again, however, the strength of that finding rested largely on what the precise meaning of “investment” was given the context, because if (as was found in EnCana)69 the tax refund was considered an investment in itself, impairment within the tribunal’s contemplation probably occurred. Because the tribunal summarily found in its discussion on expropriation that tax refunds did not constitute an “investment,” it was at least consistent on this broad point. Nevertheless, the tribunal’s lack of reasoning here is evident, as the scope of the term was once again neither clarified nor accorded proper treatment. The primary focus of the impairment discussion was the issue of arbitrariness. The tribunal quoted with approval the Lauder tribunal’s definition of “arbitrary,” which in turn was copied from Black’s Law Dictionary.70 Resort to a general legal dictionary’s definition of the term is at least curious and was probably improper here, as “arbitrariness” may have a meaning that is unique in international investment law or the BIT. This is not to say, of course, that Black’s definition itself was inconsistent with prevailing law; it is poor draftsmanship, however, to overtly rely on a legal dictionary as an authoritative pronouncement of international investment law. More difficult to comprehend is why, after having introduced the Black’s Law Dictionary definition of “arbitrary,” the tribunal then abandoned it in the next paragraph and abruptly employed its own appraisal of the term. It acknowledged that the SRI decisions “do not appear to have been founded on prejudice or preference rather than on reason or 67


68

Id., ¶ 161.

69

See supra note 49 and accompanying text.

70



fact”71 and that the SRI had no intention to be arbitrary. Nevertheless, the tribunal found that “some form of arbitrariness” resulted because of the “very confusion and lack of clarity” of the Republic’s tax regime, which the SRI had the “task of bringing some resemblance of order.”72 Given the apparently innovative nature of the tribunal’s ruling that “unintentional arbitrariness” exists and that legal uncertainty as a result of good faith reinterpretations could potentially open a host state to allegations of arbitrariness, one would expect the tribunal to have reasoned carefully the serious implications this finding would have on international investment law. No guidance was given at all, however, nor was any real reasoning proffered. IV. CONCLUSION Because investment tribunals lack effective enforcement power, reliance is placed in large measure on the investor and, more importantly, the host state’s acceptance of the award in esse, which can only occur when the reasoning that formed the basis for the decision outcome is objectively persuasive. In a real sense, therefore, the award in investor-state dispute resolution is a pedagogical tool that reinforces the legitimacy of the system by affording both investors and host states an understanding of why they should comply with international investment law. Viewed in this light, Occidental represents a low water mark in investorstate arbitral jurisprudence, not so much for any perverse conclusions made (although some such as the issues of national treatment and expropriation are particularly disquieting), but for the remarkably unconvincing manner through which those judgments were reached. The number of decision pathologies discussed above may lead cynics to consider this award animated less by law than by exogenous policy considerations, the content and weight of which we are left to guess at. This of course may in fact be untrue; unless the Tribunal articulates its reasoning adequately, however, such conjectures can never be fully exorcised.

71

Id., ¶ 163.

72

Id.

CHAPTER 8

THE AWARDS IN WENA HOTELS LIMITED V. ARAB REPUBLIC OF EGYPT Julie Maupin*

I.

INTRODUCTION

This chapter examines the case of Wena v. Egypt1 in light of the ICSID Convention’s Article 48 “reasons requirement.”2 It argues that the Wena arbitral tribunal failed to honor Article 48’s requirement that “[t]he award shall deal with every question submitted to the tribunal, and shall state the reasons upon which it is based.” In both the jurisdictional and merits phases, the tribunal declined to provide reasons for critical decisions affecting the outcome of the case. These avoidable omissions contravened the Convention’s reasons requirement, thereby undermining the legitimacy of the award. The chapter then examines the ad hoc committee’s review of the Wena tribunal’s award. It finds that the committee also failed to discharge its obligations with respect to the reasons requirement. Although the committee should have annulled the award for lack of reasons under Article 52 of the Convention, the committee instead attempted to “cure” the defective award by supplying the missing reasons on its own. This exceeded the committee’s powers under the Convention. Finally, the chapter demonstrates how the reasoning failures of both the tribunal and the ad hoc committee hampered a later request for interpretation of the award and led to unsatisfactory outcomes for both of the parties.

* M.A. Economics, Yale; J.D., Yale Law School. This chapter was written as part of the Reasons Project, directed by Professors Michael Reisman and Guillermo Aguilar Alvarez, Yale Law School. 1 Wena Hotels Ltd. v. Arab Republic of Egypt, ICSID Case No. ARB/98/4, Proceeding on the Merits (Dec. 8, 2000), 41 I.L.M. 896 (2002) [hereinafter Wena— Merits]. 2 Convention on the Settlement of Investment Disputes Between States and Nationals of Other States, ICSID (World Bank), opened for signature Mar. 18, 1965, 575 U.N.T.S. 159, art. 48 [hereinafter ICSID Convention].

231


If accepted as general practice, the Wena tribunal’s lack of reasoning and the ad hoc committee’s assumption of unauthorized powers threaten to undermine the legitimacy of international investment law. The chapter concludes by offering suggestions for how the Convention’s reasons requirement might be applied more effectively to prevent this result. II.

THE FACTUAL BACKGROUND OF THE DISPUTE

The dispute in Wena arose out of two long-term lease agreements entered into by Wena, a U.K. registered company, and the Egyptian Hotels Company (EHC), a state-owned enterprise. The agreements concerned the development and management of two hotels in Luxor and Cairo, Egypt. Under the Luxor Hotel agreement, the investor was obliged to “operate and manage the ‘Hotel’ exclusively for [its] account through the original or extended period of the ‘Lease,’ to develop and raise the operating efficiency and standard of the ‘Hotel’ to an upgraded four star hotel according to the specifications of the Egyptian Ministry of Tourism or upgratly [sic] it to a five star hotel if [Wena] so elects.”3 The 25-year agreement concerning the Nile Hotel, in Cairo, contained similar provisions. Shortly after Wena took over the two hotels, disputes arose between the parties concerning their respective obligations under the agreements. Wena complained that the condition of the two hotels was “far below that stipulated in the lease,”4 and EHC claimed that Wena was violating its obligations by failing to pay its rent. Some of these early disputes were resolved through arbitral proceedings brought by Wena in Egypt against EHC. However, relations between the parties continued to deteriorate. On April 1, 1991, “large crowds attacked the Luxor and Nile Hotels and the staff and guests were forcibly evicted. Both parties agree[d] that EHC participated in these attacks and subsequently took control of the hotels.”5 Evidence presented to the tribunal during the merits phase indicated that high ranking officials in the Egyptian government, including the Egyptian Minister of Tourism—who exercised effective control over 3 Wena Hotels Ltd. v. Arab Republic of Egypt, ICSID Case No. ARB/98/4, Proceeding on Jurisdiction, (May 25, 1999), 41 I.L.M. 881, 884 (2002) [hereinafter Wena—Jurisdiction]. 4

Id.

5

Id. at 885.

The Awards in Wena Hotels Limited v. Arab Republic of Egypt • 233

EHC and its board of directors—were made aware of the intended seizures in advance, did nothing to prevent them, and afterwards failed to act to restore the hotels to Wena in a timely fashion.6 Domestic Egyptian legal proceedings declared EHC’s seizure of the hotels unlawful and ordered EHC to return them to Wena. When EHC finally complied, after significant delays, the hotels were returned in a severely damaged state, having been looted during the seizures. In addition, Wena’s operating licenses were suspended for alleged fire code violations, preventing Wena from resuming normal hotel operations. Following numerous rounds of unsuccessful negotiations between the parties, Wena brought a claim before an ICSID tribunal pursuant to the Agreement for the Protection and Promotion of Investments (IPPA) between the United Kingdom and Egypt.7 Wena claimed violations of the IPPA’s expropriation provision,8 as well as its provision on fair and equitable treatment and full protection and security.9 Several contentious issues plagued the proceedings at both the jurisdictional and merits phases of the proceedings. First, although Wena was incorporated in the United Kingdom, a majority of its shares were owned by an Egyptian national, Mr. El-Farargy, who obtained British citizenship in 1987, only two years prior to the commencement of Wena’s agreements with EHC. This raised the question of whether Wena could be treated as a U.K. company for purposes of the IPPA. In its decision on jurisdiction, the tribunal held that Wena did qualify for IPPA protection as a U.K. company, despite vigorous objections from Egypt.10 Second, the case involved troubling allegations of corruption. Egypt claimed that the director of the EHC had accepted money from Wena, under a consulting contract, in exchange for the director’s efforts to improperly influence the Egyptian government in awarding the contracts to Wena. The tribunal, however, declined to consider the legal ramifica6

See Wena—Merits, supra note 1, at ¶¶ 28–62 .

Agreement between the Arab Republic of Egypt and the United Kingdom of Great Britain and Northern Ireland for the Promotion and Protection of Investments, 1976 Gr. Brit. T.S. No. 97 (Cmnd. 6141) [hereinafter IPPA]. 7

8

Id. at art. 5(1).

9

Id. at art. 2(2).

10

Wena—Jursidiction, supra note 3, at 886–89.


tions of these corruption allegations at the merits phase. It found that Egypt had failed to provide sufficient proof of the corruption and, in any case, had failed to prosecute any of the people involved, thus suggesting a tacit complicity on the part of the government and estopping it from raising corruption as a bar to Wena’s claim.11 Third, Wena’s evidence concerning the amount of its damages was highly contested. The tribunal determined on the merits that Wena should be awarded damages in the amount of its actual investment.12 However, it declined to determine which damages had been suffered by Wena directly, as opposed to its affiliates—some of which may not have been eligible for recovery under the IPPA.13 The tribunal also failed to precisely determine the actual amount of Wena’s investment, relying instead upon the investor’s claim of U.S. $8.8 million as a fair approximation of the true amount.14 And although more than half of the award consisted of accrued interest (U.S. $11.4 million),15 the tribunal failed to make any findings concerning the dates of Wena’s actual investments. It appears to have calculated accrued interest on the entire amount of Wena’s claimed investments as of the date of the illegal seizure of the hotels, notwithstanding suggestions that a substantial portion of Wena’s investments were made well after that date.16 It is against this factual backdrop that the sufficiency of the reasoning employed in this case is analyzed. As Sections III and IV will illustrate, the inadequacy of the tribunal’s reasoning at the jurisdiction and merits 11 Wena—Merits, supra note 1, at ¶¶ 70–74, 111–117. Egypt raised the corruption allegations in order to argue that the contracts between Wena and EHC were never legal, and therefore Wena could not claim any rights under the IPPA. Interestingly, the tribunal seemed to agree that the allegations, if proven, would be grounds for dismissal of the claim. 12

Id. at ¶ 125.

13

Id. at ¶ 126.

Id. at ¶ 127. The tribunal did reduce Wena’s claimed damages by two amounts. It subtracted $435,570 for a domestic Egyptian arbitral award which Wena had already recovered, and it subtracted $322,000 which Egypt had demonstrated probably constituted “double counting in certain instances.” The implications of the tribunal’s decision on this point will be discussed in Section IV. 14

15

Id. at ¶ 130.

16 Wena Hotels Limited v. Arab Republic of Egypt, ICSID Case No. ARB/98/4, Annulment Proceeding, ¶¶ 98–99 (Jan. 28, 2002), 41 I.L.M. 933 (2002) [hereinafter Wena—Annulment].


phases led to an appeal by Egypt for the annulment of the award. Section V demonstrates that the ad hoc committee’s annulment review suffered from similar deficiencies of reasoning. Section VI briefly discusses how the reasoning deficiencies led to further disputes between the parties, necessitating a further interpretation of the award by a new tribunal nearly five years later. Section VII concludes with some suggestions on how the ICSID Convention’s reasons requirement might be applied in a way that corrects these deficiencies in future cases and lends greater legitimacy to international investment law as a whole. III. UNREASONED FINDINGS OF THE TRIBUNAL AT THE JURISDICTIONAL PHASE Egypt initially lodged four objections to the tribunal’s jurisdiction to hear the dispute. It later withdrew two of these, however, leaving the tribunal with only two objections to consider:17 (1) that Egypt “ha[d] not agreed to arbitrate with the Claimant as it is, by virtue of ownership, to be treated as an Egyptian company” and (2) that there was “no legal dispute between the Claimant and the Respondent.” The latter objection was based upon Egypt’s contention that Wena’s dispute was not with Egypt but with EHC, “with whom Wena entered its original lease agreements and whose employees allegedly attacked the two hotels.”18 The tribunal easily dispensed with this objection in six concise and clearly reasoned paragraphs. It cited Wena’s allegation that Egyptian officials “assist[ed] in, or at least fail[ed] to prevent, the expropriation of Wena’s assets” and found that Wena had “presented at least some evidence that suggests Egypt’s possible culpability.”19 On this basis, the tribunal held that Wena had made out a prima facie claim “sufficient to invoke jurisdiction under the IPPA and the ICSID Convention.”20 In so holding, it noted that it remained “Wena’s burden to prove that Egypt is, indeed, responsible for the conduct it has alleged.”21 The tribunal’s finding on this point was thus well supported and clearly stated. The tribunal’s approach to Egypt’s first jurisdictional objection, however, proved much less satisfactory. The question of whether Wena should 17

Wena—Jurisdiction, supra note 3 at 885–86.

18

Id. at 890.

19

Id. at 891.

20

Id.

21

Id.


be treated as a British company for purposes of the arbitration revolved around a mix of questions both legal and factual.22 The legal questions concerned the proper interpretation of Article 8(1) of the IPPA. The first sentence of Article 8(1) allows nationals or companies of each of the IPPA’s contracting parties to submit investment claims against the other contracting party to ICSID arbitration. The second sentence of Article 8(1) states: “a company of one Contracting Party in which before such a dispute arises a majority of shares are owned by nationals or companies of the other Contracting Party shall in accordance with Article 25(2)(b) of the Convention be treated for the purposes of the Convention as a company of the other Contracting Party.” Inserting the identities of the parties into this second sentence, Article 8(1) would read roughly as follows: “a British company (such as Wena), in which before the dispute arises a majority of shares are owned by Egyptian nationals or companies (such as Mr. Farargy) shall in accordance with Article 25(2)(b) of the Convention be treated for purposes of the Convention as an Egyptian company.” This preliminary plain language reading of the IPPA would seem to deny the tribunal jurisdiction over the dispute if in fact Mr. Farargy was to be considered an Egyptian national, as Egypt alleged. If, on the other hand, Mr. Farargy was to be considered solely a British national (as a result of his adoption of British citizenship), then the tribunal would clearly have jurisdiction over the dispute. The tribunal, however, declined to address Mr. Farargy’s citizenship at all. Instead, it embarked upon a detailed analysis of the meaning of the second sentence of Article 8(1) of the IPPA in conjunction with Article 25(2)(b) of the Convention. After performing a thorough analysis of these two legal texts, the tribunal agreed with Wena’s position that the combined effect of the texts was to “extend jurisdiction in cases where a company of the host State [Egypt] is controlled by nationals or companies of the non-host State [Britain]”23 rather than to (as Egypt had argued) “reverse the consent [to arbitration] given in the first sentence of Article 8(1) when a Contracting State is a party to a dispute with a 22 The factual questions, discussed later in this section, are: (1) whether a majority of Wena’s shares were owned by Mr. Farargy, (2) whether Mr. Farargy remained an Egyptian national despite his adoption of British citizenship in 1987, and (3) if so, whether Mr. Farargy’s majority ownership and Egyptian nationality preclude Wena from receiving treatment as a U.K. company under the IPPA. See id. at 886–87. 23

Id. at 886, 888.


juridical person of the other Contracting State.”24 Standing on its own, this interpretation appears well reasoned. The tribunal pointed out that it comports with a plain language reading of the text and also squares with the extensive academic literature on the subject.25 But this conclusion alone could not establish the tribunal’s jurisdictional competence in the matter. To comply with the ICSID Convention’s reasons requirement, the tribunal needed to explain not only its reasons for rejecting Egypt’s restrictive interpretation of the second sentence of Article 8(1) of the IPPA, but also its reasons for finding jurisdiction under the first sentence of Article 8(1). In other words, the tribunal needed to determine whether Article 8(1) also confers jurisdiction in cases where a company of the non-host state (Britain) is controlled by nationals or companies of the host-state (Egypt). Unfortunately, the tribunal failed to address this question. Instead it held that, because it rejected Egypt’s proposed interpretation of the second sentence of Article 8(1) of the IPPA, it did not need to consider the citizenship issues pertaining to its jurisdiction under the first sentence of Article 8(1). The latter does not follow from the former.26 Under the first sentence of Article 8(1), the tribunal could only exercise jurisdiction over disputes between one contracting party and nationals or companies of the other contracting party. Egypt was clearly a contracting party. However, the tribunal’s discussion does not make clear whether Wena could properly be considered a “national or company of the other Contracting Party,” or whether—owing to Mr. Farargy’s majority ownership—Wena did not qualify as a U.K. company under the IPPA. The tribunal appears to have deemed irrelevant: (1) whether a majority 24

Id. at 889.

To its credit, in the absence of any conclusive evidence on the negotiating history of the IPPA, the tribunal embarked upon a brief survey of the academic literature pertaining to the interpretation of such bilateral investment provisions. The tribunal noted that most academics have interpreted the provisions as “expand[ing] ICSID jurisdiction by permitting parties to a dispute to stipulate that a subsidiary of a ‘national of another contracting state’ which is incorporated in the host state (and therefore arguably a ‘local national’) will be treated as itself a national of another contracting state.” Id. at 888 (emphasis in original) (internal citations omitted). 25

26 Though the tribunal’s analysis adequately established that the second sentence of Article 8(1) was irrelevant to a determination of jurisdiction in the Wena dispute, it did not establish upon what grounds jurisdiction did exist—a basic preliminary step for any tribunal. The Wena tribunal thus vastly over-reached its authority in assuming jurisdiction without undertaking the appropriate factual inquiries.


of Wena’s shares were owned by Mr. Farargy, (2) whether Mr. Farargy remained an Egyptian national despite his adoption of British citizenship,27 and (3) if so, whether Mr. Farargy’s majority ownership and continued Egyptian nationality could preclude Wena from receiving treatment as a U.K. company under the IPPA. Egypt had put all of these questions before the tribunal, yet the tribunal’s award does not explain why it deemed them irrelevant. This oversight by the tribunal has important implications. By declining to provide reasons for its failure to address key questions concerning Mr. Farargy’s citizenship, the tribunal inadvertently opened the door for Egyptian citizens to bring ICSID arbitration claims against the government of Egypt simply by incorporating in Britain prior to making any investments. Egypt cannot be presumed to have consented to such claims in concluding the IPPA. Indeed, the purpose of bilateral investment treaties (BITs) is not to render states liable to binding arbitration by their own citizens but to provide an impartial forum for foreign investors’ claims. None of the traditional ills that BITs are designed to prevent are present in the case of nationals investing in their own states.28 To be sure, the factual pattern present in the Wena case presented a complex challenge. The IPPA’s definition of “nationals” is silent on the question of dual-citizens.29 The tribunal would no doubt have faced a difficult task in discerning how to correctly address the matter of Mr. Farargy’s potential dual nationality.30 But the prospective difficulty of an 27 This question can only be answered by reference to Egyptian law. The tribunal’s decision not to apply Egyptian law was problematic in this regard. 28 Traditional justifications for BITs cite the foreign investor’s relative unfamiliarity with local customs, business environment, and legal norms; inability to access the local political process; prejudicial disadvantage in local courts and tribunals; etc. See Thunderbird v. Mexico, NAFTA arbitration under the UNCITRAL rules, dissenting opinion, ¶ 12 (Jan. 26, 2006). See also Tokios Tokeles v. Ukraine, ICSID Case No. ARB(AF)/00/2 (May 19, 2003); Soufraki v. United Arab Emirates, ICSID Case No. ARB/02/07 (July 7, 2004). Of course, others would argue that the prevention of capital flight is also a legitimate goal for BITs, and therefore domestic investors who incorporate abroad should be extended the same protections as foreign investors in order to encourage them to reinvest in the home country. However, this is a newer, and still quite controversial, view of the role of BITs. To date, most sovereign governments have preferred to address the problem of capital flight via domestic measures, rather than by allowing their own citizens to bring arbitration disputes against them in the international arena. 29

IPPA, supra note 7, at art. 1(c).

30

To properly determine nationality, the tribunal would have had to apply


issue does not release an arbitral tribunal from its duty to consider a putative claimant’s nationality and to provide sufficiently explained reasons for its conclusions on questions of nationality as a pre-requisite to any finding of jurisdiction. In the end, by making Egypt potentially vulnerable to ICSID claims by its own citizens, the Wena tribunal’s willful avoidance of this issue may create more problems than it solved. IV. UNREASONED FINDINGS OF THE TRIBUNAL AT THE MERITS PHASE The Wena tribunal’s decision on the merits contains a lengthy and detailed presentation of the facts adduced by both parties.31 Referring back to these facts, the tribunal conducted a thorough analysis of each of the investor’s legal claims in Part III of the award. It found that Egypt had indeed breached its obligations under the IPPA, first by failing to accord Wena’s investments “fair and equitable treatment” and “full protection and security,”32 and second by failing to provide Wena with “‘prompt, adequate and effective compensation’ for the losses it suffered as a result of the seizures of the Luxor and Nile Hotel.”33 The tribunal’s reasoning throughout this portion of the award, while not perfect, appears adequate on the whole. The bases for the tribunal’s legal conclusions are clearly stated in a manner which allows an informed reader to readily follow its logic.34 The opening paragraphs of the tribunal’s analysis on damages likewise signaled that the tribunal would remain true to the Convention’s “reasons requirement.” Part IV of the award began by quoting Article 5 Egyptian law, as that is the only source of law having the authority to confer or deny Egyptian nationality. If, as Wena claimed, Mr. Farargy lost his Egyptian nationality upon obtaining British citizenship, then there would be no bar to the tribunal’s jurisdiction. If, however, as Egypt claimed, Mr. Farargy retained his Egyptian nationality after adopting British citizenship, then the tribunal would have had to consider whether and how Mr. Farargy’s dual nationality may have affected Wena’s status as a “U.K. company” under the IPPA. 31

Wena—Merits, supra note 1, at ¶¶ 15–74.

32

Id. at ¶ 95.

33

Id. at ¶ 101.

34 One exception may be the tribunal’s treatment of the choice of law question. The tribunal’s decision to apply only international law (rather than Egyptian law) to certain portions of the dispute was problematic and insufficiently reasoned. For example, it is unclear why the tribunal did not apply Egyptian law in order to determine the legal effect of the previously concluded Egyptian arbitration involving one of Wena’s two hotel contracts. However, an in-depth analysis of choice of law issues is beyond the scope of this chapter.


of the IPPA: “in the event of an expropriation, the private investor shall be entitled to ‘prompt, adequate, and effective compensation’ and ‘such compensations shall amount to the market value of the investment immediately before the expropriation.’”35 The tribunal acknowledged the discounted cash flow (DCF) method as the usual means by which tribunals have satisfied the compensation requirement. However, it found the DCF method inappropriate in Wena’s case, because such an award would have been too speculative. The tribunal quoted Metalclad v. Mexico36 and SPP v. Egypt 37 in support of this proposition. It also provided four cogent reasons why Wena’s claimed damages were too speculative to base the award on the DCF method:38 1.

2.

3.

4.

35

Wena’s operating history prior to the expropriation was too short to establish a “solid base on which to found any profit calculations”; Wena had not yet completed its renovations on one of the hotels (and therefore could not determine how large a revenue increase the renovations would bring); There was some doubt as to whether Wena “had sufficient finances to fund its renovation and operation of the hotels”; and The huge disparity between Wena’s requested DCF damages (GB£ 45.7 million) and its stated investment in the two hotels (U.S. $8.8 million) cast doubts upon the validity of its DCF calculations.

Wena—Merits, supra note 1, at ¶ 118 (quoting IPPA, supra note 7, at art. 5).

36 The portion of the Metalclad decision quoted by the tribunal reads: “Normally, the fair market value of a going concern which has a history of profitable operation may be based on an estimate of future profits subject to a discounted cash flow analysis. However, where the enterprise has not operated for a sufficiently long time to establish a performance record or where it has failed to make a profit, future profits cannot be used to determine going concern or fair market value.” Id. at ¶ 123 (quoting Metalclad Corp. v. United Mexican States, ICSID Case No. ARB(AF)/97/1 at ¶¶ 119–120 (2000)). 37 The tribunal stated: “Similarly, the ICC panel in the SPP (Middle East) v. Egypt arbitration case declined to accept a discounted cash flow projection because, inter alia, by the state of cancellation the great majority of the work had still to be done, and the calculation put forward by the Claimants produce[d] a disparity between the amount of the investment made by the Claimants and the supposed value of the investment as calculated by the DCF analysis.” Id. (internal quotation marks omitted). See also SPP (Middle East) Ltd v. Arab Republic of Egypt, 22 I.L.M. 752, 782 (1983), at ¶ 65.5. 38

Wena—Merits, supra note 1, at ¶ 124.


In light of these factors, the tribunal opted to determine “the market value of the investment immediately before the expropriation . . . by reference to Wena’s actual investments in the two hotels.”39 Thus far, the damages section of the award comported with the ICSID Convention’s Article 48 requirement that “[t]he award shall . . . state the reasons upon which it is based.”40 The situation changed, however, once the tribunal turned its attention to this alternate method of damage calculations. The remainder of the award suffers from two fundamental failures of reasoning. First, the tribunal failed to determine the actual amount of the investments made and the portion of those investments directly attributable to Wena. It instead appears to have accepted Wena’s claimed sum on its face. Second, the tribunal failed to justify its interest award. It awarded interest at a rate of 9 percent, compounded quarterly—without providing a reasoned explanation for these choices—and neglected to make any findings concerning the dates on which Wena’s investments were made or the dates on which they were expropriated. As a result, the tribunal’s interest calculations, which amounted to over 60 percent of the total award, were based on either a wholly arbitrary or factually erroneous date. Taken together, these abdications of the tribunal’s responsibilities combined to create an impermissibly unreasoned damages award. I examine each of them in turn. A.

Precisely How Much Was Invested? And by Whom?

Where a tribunal decides to base an award upon the “actual investment method,” a determination of how much was invested and by whom is obviously central to any proper calculation of damages. The claimant bears the burden of proving its actual damages. Merely alleging damages will not suffice. The tribunal in Feldman v. Mexico explained this burden of proof by reference to the “international law standard” exposited by the World Trade Organization (WTO) Appellate Body: “various international tribunals, including the International Court of Justice, have generally and consistently accepted and applied the rule that the party who asserts a fact, whether the claimant or respondent, is responsible for providing proof thereof.”41

39

Id. at ¶ 125.

40

ICSID Convention, supra note 2, at art. 48(3).

Marvin Feldman v. Mexico, ICSID Case No. ARB/(AF)/99/1, Award, ¶ 177 (Dec. 16, 2002) (quoting United States—Measures Affecting Imports of Woven Wool Shirts and Blouses from India, adopted May 23, 1997, WT/DS33/AB/R, at 14). 41


Scholarly writings concur with this formulation. Durward Sandifer, for example, writes, “[t]he broad basic rule of burden of proof adopted, in general, by international tribunals resembles the civil law rule and may be simply stated: that the burden of proof rests upon him who asserts the affirmative of a proposition that if not substantiated will result in a decision adverse to his contention.”42 Charles Brower similarly concurs: The lack of standard international rules of evidence, and the fact that international tribunals are liberal in their approach to the admission of evidence in no way goes as far as to waive the burden resting upon a claimant to prove his case. In international arbitral proceedings a party making an allegation of fact has an obligation to demonstrate that fact with sufficient evidence. This principle derives from the Roman law rule of burden of proof expressed through certain maxims such as ei qui affirmat non ei qui negat incumbit probatio (onus of proof is on he who affirms, and not on he who denies) and actori incumbit probatio (the claimant carries the burden of proof).43 International law thus clearly places the burden of proving actual damages upon Wena in this case. Yet the tribunal appears to have largely accepted Wena’s alleged damages without stating how Wena actually established them. The tribunal’s finding on actual investment damages consists of a single paragraph: On the basis of investment, Claimant states its loss as US$8,819,446.93. However, the panel in pursuing an objection raised by the Respondent that there were certain elements of double counting, decided that the gross figure should be diminished by US$322,000.00 to eliminate probably double counting in certain instances. Beyond that, however, the panel was not persuaded by Respondent’s evidence that there were significant other instances of double counting. Thus, the figure of US$8,819,466.93 should be diminished by US$322,000.00 leaving a total of US$8,497,466.93, which the Tribunal judges to be the approximate total of Wena’s investment. From this, the Tribunal agreed that $435,570.38

42

DURWARD V. SANDIFER, EVIDENCE BEFORE INTERNATIONAL TRIBUNALS (rev. ed.

1975). 43 Charles N. Brower, Evidence Before International Tribunals: The Need for Some Standard Rules, 28 INT’L LAW. 47, 49 (1994) (internal citations omitted).


should be deducted for the amount received already by Claimant as a result of the Egyptian arbitration award.”44 This paragraph is remarkable in several respects. First, the tribunal simply accepted the claimant’s alleged damages as a starting point, without citing any evidence by which the claimant had proven this amount to the tribunal.45 Next, the tribunal appears to have placed the burden upon the respondent to disprove Wena’s claim rather than placing the burden upon Wena to establish every element of its claim. And most surprisingly, the tribunal awarded damages in the amount of the “approximate total” for Wena’s investment, less certain elements of “probably” double counting. These phrases stand in stark contrast to the general international law standard concerning the burden of proof discussed above. Damages that are “approximate” cannot possibly have been “proven” by the claimant. By its use of such language, the tribunal implicitly acknowledged its own failure to apply the accepted international burden of proof standard, yet it provided no justification for the departure. The tribunal compounded this failure by neglecting to determine whether Wena itself had suffered the entire amount of the actual damages claimed. Egypt contended that “much of the investment came from affiliates of Wena rather than from Wena.”46 The tribunal summarily dismissed this contention, saying it was “not persuaded by the relevance of the Respondent’s contention.”47 It justified this dismissal as follows: [T]he panel takes the view that whether the investments were made by Wena or by one of its affiliates, as long as those investments went into the Egyptian hotel venture, they should be recognized as appropriate investments. The panel was persuaded from the testimony it received that it is a widely established practice for hotel enterprises to adopt allocation measures, which spread the profits form [sic] the group operations into various jurisdictions where there are tax advantages to the group as a whole.48 44

Wena—Merits, supra note 1, at ¶ 127 (emphasis added).

The tribunal could have, for example, referred to bank records, copies of contracts, loans, or lease agreements, or other such financial records as evidence in support of its findings. 45

46


47

Id.

48

Id.


Here again, the tribunal missed the critical inquiry. The question is not whether Wena’s profit-sharing arrangements with its affiliates conformed to “widely established practice” but whether the affiliates, as investors, qualified for recovery under the IPPA. This is an essential jurisdictional question. Without first finding that the affiliates qualified for protection under the IPPA—either directly or via their contractual relationships with Wena—the tribunal, in awarding to Wena damages based upon the actual investments of Wena’s affiliates, manifestly exceeded its powers.49 One observation concerning the tribunal’s failure to clearly establish the actual amount and source of the investments bears mentioning: the source of the reasoning failures is not obvious from reading the award. Did the failures arise on account of a lack of evidence presented to the tribunal? If so, why did the tribunal not exercise its power, pursuant to Article 43 of the ICSID Convention, to compel further evidence? Alternatively, did the tribunal have before it adequate evidence with which to justify its findings but simply neglect to state it? Or did the tribunal fundamentally fail to discern the crucial legal and evidential questions of the case? Whatever the explanation, the tribunal’s lack of reasoning appears to violate the ICSID Convention’s requirement that “[t]he award shall deal with every question submitted to the Tribunal, and shall state the reasons upon which it is based.”50 B.

What Constitutes a Reasonable Interest Award?

Having set the amount of damages at $8.5 million, the tribunal next turned its attention to the question of interest. After noting that “the IPPA, the lease agreements, and the ICSID Convention and Rules are all silent on the subject of interest,”51 the panel adopted a 9 percent rate of interest52 and determined that it should be compounded quar49 This should constitute grounds for annulment under Article 52 of the ICSID Convention. See discussion in Section V. 50

ICSID Convention, supra note 2, at art. 48(3).

51


Footnote 289 of the award notes that long-term government bonds in Egypt were yielding 10 percent. However, the panel does not mention this in its decision and provides no explanation for why it adopted a 9 percent (rather than 10 percent) interest rate. Id. at n.289. Also, as Guillermo Aguilar Alverez has pointed out, it is unclear why a tribunal would use the rate on Egyptian long-term bonds as the point of departure for an award denominated in U.S. dollars (rather than local currency). Most tri52


terly. To its credit, the panel engaged in a paragraph-long explanation of its election of compound (as opposed to simple) interest. This paragraph is concise, well reasoned, and supports the general proposition that international tribunals enjoy the discretion to award compound interest where appropriate. The paragraph does not, however, explain why the adoption of compound interest, and specifically quarterly compounding, is appropriate in this particular case.53 In fact, the tribunal, in paragraph 129, quite transparently telegraphed its motivation for limiting its discussion to the general merits of compound interest, stating, “[t]his tribunal believes that an award of compound (as opposed to simple) interest is generally appropriate in most modern, commercial arbitrations.” The arbitrators thus explicitly based Wena’s interest award upon their personal views concerning what is appropriate in modern arbitrations. Unfortunately, ICSID arbitrators, while enjoying wide discretion in many matters, are bound to apply the governing law of the dispute. And while the IPPA, the lease agreements, and the ICSID Convention and rules may all be silent on the subject of interest, international law (which also governs the IPPA) is not. The Iran-U.S. Claims tribunal had occasion to review international practice concerning interest awards in McCullough. It concluded: [N]o uniform rule of law relating to interest has emerged from the practice in transnational arbitration. [. . .] The absence of a uniform rule does not, however, imply the absence of general principles. On the contrary, two principles or guidelines, of general import, albeit of delicate implementation, can be deduced from the international practice briefly described above. The first principle is that under normal circumstances, and especially in commercial cases, interest is allocated on the amounts awarded as damages in order to compensate for the delay with which the payment to the successful party is made. [. . .] The second principle is that the rate of interest must be reasonable, taking due account of all perbunals would refer to London Interbank Offered Rate (LIBOR) or another similar prime rate in awarding interest on U.S. dollar-denominated damages. 53 One of the panel members, Professor Wallace, concurred in the decision but did not agree that compounding should be quarterly. However, like the majority of the panel, he gave no reasons for his conclusion. Id. at 920, Statement of Professor Don Wallace Jr.


tinent circumstances, which the Tribunal is entitled to consider by virtue of the discretion it is empowered to exercise in this field.54 James Crawford, in surveying the question of compound interest in his role as Special Rapporteur for the International Law Commission (ILC) found that “the general view of courts and tribunals has been against the award of compound interest, and this is true even of those tribunals which hold claimants to be normally entitled to compensatory interest.”55 However, he summarized his report by noting: “although compound interest is not generally awarded under international law or by international tribunals, special circumstances may arise which justify some element of compounding as an aspect of full reparation.”56 The ILC appears to have incorporated Crawford’s conclusions into the Articles on State Responsibility. Article 38 now provides: “interest . . . shall be payable when necessary in order to ensure full reparation. The interest rate and mode of calculation shall be set so as to achieve this result.”57 The above discussion suggests that the Wena tribunal was indeed within its powers in choosing to award compound interest. However, in doing so, it should have explicitly identified the factors that made the award of compound interest “necessary to ensure full reparation” to Wena in this specific case. The commentary of Professors Bishop, Crawford, and Reisman suggests four such possible factors: (1) the tribunal views the government’s conduct as egregious, (2) the tribunal has been conservative in determining actual damages, (3) the claimant has obtained loans for the project in the host country that require it to pay compound interest, or (4) the government-investor contract calls for compound interest.58 Factors (1) and (2) would seem to apply directly to 54 McCullough & Company, Inc. v. Ministry of Post, Telegraph and Telephone, Award No. 225-89-3 (Chamber Three) (Apr. 22, 1986), ¶¶ 97–99, 11 IRAN-U.S. CL. TRIB. REP 3 (emphasis added) [hereinafter McCullough]. 55 James Crawford, Special Rapporteur, Third Report on State Responsibility, International Law Commission 52d Sess., at 22, U.N. Doc. A/CN.4/507, ¶ 208 (2000). 56


Responsibility of States for Internationally Wrongful Acts, adopted by the International Law Commission in 2001 at its 53d session, annex to General Assembly Resolution 56/83 (Dec. 12, 2001), available at http://www.un.org/law/ilc, art. 38 (emphasis added) [hereinafter Articles on State Responsibility]. 57

58 R. DOAK BISHOP, JAMES CRAWFORD & W. MICHAEL REISMAN, FOREIGN INVESTMENT DISPUTES: CASES, MATERIALS AND COMMENTARY 1386 (2005). While identifying these factors as the usual basis for compound interest, the authors also note that “[r]ecent cases [including Wena] may indicate a trend toward awarding compound interest generally.”


the facts of the Wena dispute. The tribunal could have reached the same result by providing justifications in accordance with international law. But by opting instead to base its interest award upon its own view of what is “generally appropriate,” the tribunal failed to justify its decision at all. It thereby again ran afoul of the ICSID Convention’s Article 48 reasons requirement. It also exceeded its authority under the Convention by assuming for itself the power to make new international law, consistent with its own vision, rather than with the “general principles of law recognized by civilized nations”59 that constitute customary international law. C.

What Are the Relevant Dates Underpinning the Tribunal’s Award?

In addition to its failure to provide justification for the amount and type of interest awarded, the Wena tribunal was remiss in its duty to make key factual determinations pertaining to the amount of the interest award. After explaining its view on compound interest, the tribunal summarily concluded: “Thus, the total, with interest through December 1, 2000 (US$11,431,386.88) is US$19,493,283.43.”60 No other discussion concerning the amount of the interest appears in the award. The most baffling feature of this statement is the complete absence of any dates or other clues that might indicate how the tribunal arrived at this amount. Although the tribunal purported to base its award on the “actual investment method,” at no time did it make any findings concerning: 1. 2. 3.

the dates of Wena’s actual investments, the dates of the expropriations or other breaches of the IPPA that gave rise to Egypt’s liability, or the dates upon which interest accrual for those liabilities should attach.

A casual reader of the award therefore cannot reconstruct how the tribunal came up with its $11.4 million interest award. Indeed, neither of the parties could hope to decipher this from the text of the award. Considering that the interest amounted to some 60 percent of the total award, it is perhaps not surprising that Egypt objected to this omission of reasons in its annulment appeal (discussed in Section V).61 Did the tri59

Statute of the International Court of Justice art. 38.1(c).

60


61

Wena—Annulment, supra note 16, at ¶¶ 87, 90.


bunal start its interest accrual on the date of the seizure of the hotels (April 1, 1991)? Did it adjust for the fact that different investments were made at different times? Did it suspend the interest accrual during the period in which Wena actually reoccupied one of the hotels (1992–1997)? Did it take account of the fact that Wena had already recouped some of its losses for the Nile Hotel in an Egyptian arbitration award in 1997? One simply cannot tell. Again, it is not obvious why the tribunal failed to provide any basis for its interest award. International law affords arbitral tribunals ample discretion in determining the date from which interest should accrue. This discretion is no more circumscribed than the discretion they enjoy vis-à-vis rates and types of interest. As the McCullough tribunal (discussed above) found in its first “general principle” on this subject: [I]nterest is allocated on the amounts awarded as damages in order to compensate for the delay with which the payment to the successful party is made. This delay, however, varies in relation to the date determined to be the time when the obligation to pay arose. This date can be the date when the underlying damage occurred, the date when the debt was liquidated, the date of a formal notice to pay, the date of the beginning of the arbitral proceedings, the date of the award or of the judgment determining the amount due, or the date when the judicial or arbitral decision reasonably should have been executed.62 The Wena tribunal could have justified its interest award by reference to any one of these possible dates, subject only to the requirement that it provide its reasons for selecting any particular date. The tribunal’s failure to do so constitutes a blatant violation of the ICSID Convention’s reasons requirement. V.

FAILURE OF THE AD HOC COMMITTEE TO PERFORM ITS PROPER FUNCTION DURING THE ANNULMENT PROCEEDING

After the Wena tribunal’s gross failures of reasoning in several key aspects of the case, the proceedings before the ad hoc committee (AHC) provided an important opportunity for an ICSID panel to reimpose some much needed discipline within the arbitral community. The Wena committee unfortunately declined to seize that opportunity. In fact, it may 62

McCullough, supra note 54, at ¶ 98.


have actually made the problem worse. The AHC’s major shortcomings in reviewing the panel award were twofold. First, it declined to apply the minimal requirements of the ICSID Convention’s reasons requirement under international law. Second, it erroneously construed Articles 48 and 52 of the ICSID Convention as requiring only “implicit” reasons for the award, and then exceeded its powers under Article 52 by supplying its own reasons where the panel’s reasons proved lacking.63 A.

The AHC Failed to Apply the Minimal Reasons Requirements of the ICSID Convention Under International Law

The AHC began its analysis of Egypt’s reasons-based objections to the award by referencing the texts of Articles 48 and 52 of the ICSID Convention. Article 48(3) states: “[t]he award shall deal with every question submitted to the Tribunal, and shall state the reasons upon which it is based.” Article 52(1), in turn, provides that “[e]ither party may request annulment of the award by an application in writing addressed to the Secretary-General on one or more of the following grounds: . . . (e) that the award has failed to state the reasons on which it is based.” Having acknowledged the relevant texts, the AHC then reviewed how this reasoning standard had been applied by previous ad hoc committees. It specifically highlighted the interpretation of the ad hoc committee in MINE: the requirement to state reasons is satisfied as long as the award enables one to follow how the tribunal proceeded from Point A to Point B, and eventually to its conclusion, even if it made an error of fact or law. The minimum requirement is in particular not satisfied by either contradictory or frivolous reasons. [. . .] [T]he requirement that an award has to be motivated implies that it must enable the reader to follow the reasoning of the Tribunal on points of fact and law. It implies that, and only that. The adequacy of the reasoning is not an appropriate standard of review under paragraph 1(e), because it almost inevitably draws an ad hoc Committee into an examination of the substance of the tribunal’s decision, in disregard of the exclusion of the remedy of appeal by Article 53 of the Convention.64 63 The analysis in this section again focuses on the damages-related aspects of the dispute, although other parts of the tribunal’s reasoning were also challenged before the AHC. 64

Wena—Annulment, supra note 16, at ¶ 77 (quoting MINE). See Maritime


The Wena AHC also noted in passing that other ad hoc committees had referred to “reasons that are ‘sufficiently relevant,’ that is, ‘reasonably sustainable and capable of providing a basis for the decision,’ or demonstrating a ‘reasonable connection between the basis invoked by the tribunal and the conclusions reached by it.’”65 These interpretations suggest at least some minimal requirement concerning the sufficiency of a tribunal’s reasoning. Indeed, even the MINE committee quoted above, while disavowing any mandate to review the “adequacy” of a tribunal’s reasoning, admitted that “contradictory or frivolous” reasons would not satisfy the Convention’s reasons requirement. Thus, while ad hoc committees are not empowered to sit as appellate bodies, they are required to ensure that a tribunal’s reasoning exhibits some basic level of sufficiency, such that an award may be justifiably “based” upon that reasoning. The Wena AHC, however, while appearing to acknowledge these minimum standards, declined to apply them in its review of the Wena tribunal’s award. This became evident in the AHC’s consideration of Egypt’s complaint that “on the face of the Award, it cannot understand the legal basis on which Wena’s claim for payment of damages has been made and how it has been quantified.”66 As demonstrated in Section IV, Egypt’s complaint on this count was perfectly justified. Yet the AHC did nothing more than recite several statements from the five paragraphs comprising the tribunal’s damages award. It did not evaluate whether any of these statements were “sufficiently relevant,” “reasonably sustainable and capable of providing a basis for the decision,” or demonstrated a “reasonable connection between the basis invoked by the tribunal and the conclusions reached by it.” Instead, it merely stated that “[t]he reasons given thus allow [sic] to understand that the Tribunal accepted Wena’s evidence on the losses it claimed to have suffered, to the exception of a correction related to possible double counting.”67 This statement becomes all the more surprising when one considers that simply accepting one party’s evidence—without so much as stating why—could International Nominees Establishment v. Republic of Guinea, Decision on Annulment, ¶¶ 5.08–5.09 (Dec. 22, 1989), 4 ICSID REP. 79, 88 [hereinafter MINE]. 65 Wena—Annulment, supra note 16, at ¶ 78 (quoting Klöckner I and Amco I). See Klockner v. Cameroun, Decision on Annulment, ¶ 120 (May 3, 1985)2 ICSID REP. 95, 139 [hereinafter Klockner I]; Amco Asia Corp. and Others v. Republic of Indonesia, Decision on Annulment, ¶ 43 (May 16, 1986), 1 ICSID REP. 509, 520 [hereinafter Amco I]. 66

Wena—Annulment, supra note 16, at ¶ 87.

67

Id. at ¶ 92.


surely constitute a “frivolous reason” under even the most lax of burden of proof norms, let alone the generally accepted standard outlined above.68 The AHC thus condoned the tribunal’s complete failure to provide even minimally sufficient reasons for its award under international law. In fact, the AHC seemed to grant carte blanche to future tribunals to do the same with the following general pronouncement: With respect to determination of the quantum of damages awarded, it may be recalled that the notion of “prompt, adequate and effective compensation” confers to the Tribunal a certain margin of discretion, within which, by its nature, few reasons more than a reference to the Tribunal’s estimation can be given, together with statements on the relevance and evaluation of the supporting evidence.”69 This suggests tribunals can pass Article 52(1)(e)’s muster merely by making broad, conclusory statements concerning their evaluation of the evidence. After all, this is precisely what the Wena tribunal did with its repeated use of such phrases as “the Tribunal is not persuaded . . . ,”70 “the panel was persuaded,”71 “the panel [. . .] decided,”72 “the Tribunal judges to be the approximate total,”73 “[t]he Panel is of the view,”74 “this Tribunal also has determined,”75 and finally, “[t]his Tribunal believes.76 Why state reasons when one need only state conclusions? By condoning and even encouraging such practices, the Wena AHC undermined the very purpose of the ICSID Convention’s reasons requirement.

68

See discussion in Section IV.A.

69

Wena—Annulment, supra note 16, at ¶ 91 (emphasis added).

70

Wena—Merits, supra note 1, at ¶¶ 126, 127.

71

Id. at ¶ 126.

72

Id. at ¶ 127.

73

Id. at ¶ 127.

74

Id. at ¶ 128. See also id. at ¶ 126 (“the panel takes the view”).

75

Id. at ¶ 129.

Id. at ¶ 129. Remarkably, all of these statements appear within the span of four short paragraphs. 76


B.

”Implicit” or “Inferred” Reasons Cannot Satisfy the ICSID Convention’s Reasons Requirement, and Ad Hoc Committees Lack Jurisdiction to Cure Defective Awards by Supplying Their Own Reasons

The previous section showed how the Wena AHC avoided annulling the tribunal’s damages award by evading the minimal reasoning requirements of Articles 48 and 52 of the Convention. The AHC compounded this error in its review of the tribunal’s interest award by means of an entirely different method. Rather than simply declining to review the existence and minimal sufficiency of the tribunal’s reasons, the AHC attempted to rescue the interest portion of the award by implying reasons instead. The AHC set up this tactic in its initial analysis of the reasons requirement, stating: Neither Article 48(3) nor Article 52(1)(e) specify the manner in which the Tribunal’s reasons are to be stated. The object of both provisions is to ensure that the Parties will be able to understand the Tribunal’s reasoning. This goal does not require that each reason be stated expressly. The Tribunal’s reasons may be implicit in the considerations and conclusions contained in the award, provided they can be reasonably inferred from the terms of the decision.77 What is most remarkable about this interpretation is its departure from the plain language of Articles 48 and 52, which both explicitly require an award to “state”—not “imply”—the reasons upon which it is based. Such textual manipulation is troubling in and of itself, but the tremendous import of this departure becomes strikingly clear two paragraphs later, where the AHC states: If the award does not meet the minimal requirement as to the reasons given by the Tribunal, it does not necessarily need to be resubmitted to a new Tribunal. If the ad hoc Committee so concludes, on the basis of the knowledge it has received upon the dispute, the reasons supporting the Tribunal’s conclusions can be explained by the ad hoc Committee itself. (Emphasis added.) However, nothing in the ICSID Convention confers upon ad hoc committees the power to cure defective awards by supplying missing reasons. On the contrary, Article 52 makes annulment the sole remedy for awards 77

Wena—Annulment, supra note 16, at ¶ 81 (emphasis added).


that fail to state the reasons upon which they are based. While Article 50 allows disputing parties to seek clarification of poorly reasoned awards by submitting requests for clarification or correction to the original tribunal, Article 50 does not apply to ad hoc committees.78 The Wena AHC’s assumption of this power therefore amounts to a shocking overreach of its authority.79 The motivation behind the Wena AHC’s reinterpretation of the Convention’s reasons requirement and its self-assumption of the power to cure defective reasoning becomes evident in Part IV.D of the Annulment Proceeding. In that part, Egypt complained that the tribunal had “failed to give reasons for its decision to adopt a rate of interest at 9% and for its determination of the date from which interest accrues.”80 Egypt also argued “that if the relevant date would appear to be April 1, 1991, it would be wrong, because a substantial portion of Wena’s investment had been invested long after that date.”81 In addressing these complaints, the AHC could not simply quote and decline to evaluate the relevant passages of the tribunal’s decision, as it did in its review of the damages award. There simply are no such passages to quote. Thus, in

78 Article 52(4) makes clear that only “the provisions of Articles 41–45, 48, 49, 53, and 54, and of Chapters VI and VII shall apply mutatis mutandis to proceedings before the Committee.” ICSID Convention, supra note 1, at art. 52(4). 79 Numerous policy considerations also counsel against allowing ad hoc committees to cure defective awards with their own reasoning. Ad hoc committees do not enjoy the benefit of all of the direct testimony and evidence presented to tribunals during the jurisdictional and merits phases. As such, they lack the knowledge necessary to supply sound reasons. If the parties themselves cannot understand the tribunal’s award, despite having been privy to all of the relevant proceedings, there is no reason to believe an ad hoc committee should do any better. Moreover, in order to supply missing reasons, an ad hoc committee would have to review the entire record of the dispute, evaluate the evidence, and make its own independent findings of fact and of law. But as the Wena AHC noted, Article 53 of the Convention explicitly prohibits this type of de novo appellate review. And since parties cannot fully retry their case before an ad hoc committee, such a review could raise many serious due process concerns as well. Guillermo Aguilar Alvarez would add one additional policy consideration: access to this type of appellate style judicial review (e.g., the British Columbia courts in Metalclad, supra note 36, or the English courts in Occidental Exploration and Production Company v. Republic of Ecuador (LCIA Case No. UN 3467, Final Award (July 1, 2004)) may eliminate any incentive for non-members to ratify the ICSID Convention, thereby undermining its prospects for universal proliferation. 80

Wena—Annulment, supra note 16, at ¶ 94.

81

Id. at ¶ 99.


order to avoid overturning the award for lack of reasons, the AHC chose to supply them itself. It began by addressing the tribunal’s award of a 9 percent interest rate. After noting the tribunal’s footnote reference to the 10 percent interest rate on long-term Egyptian government bonds, the AHC justified the tribunal’s decision to grant interest “close to but still below such bonds,”82 stating: “[i]t must be assumed that [the tribunal] took such decision in order to award damages corresponding to an ‘adequate and effective compensation’ as provided for in Article 5 of the IPPA. The reasons underlying the Tribunal’s decision in this respect are thus stated sufficiently.” But how can reasons be “stated sufficiently” when they “must be assumed?” If a vague reference to some general provision of a BIT can cure a tribunal’s utter failure to state reasons for its decision, then the ICSID Convention’s reasons requirement holds no meaning. The AHC’s logic in this regard only deteriorated when it turned its attention to the tribunal’s failure to make any factual findings concerning the dates upon which the award’s damage calculations were based. The AHC began by admitting “[i]t is true that no such date is specified expressly in the Tribunal’s decision.”83 It then attempted to avoid the issue by blaming Egypt for failing to undertake its own calculations, thereby suggesting that Egypt should have inferred the relevant dates from the tribunal’s award. On this basis, the AHC held, “this Committee need not inquire on its own initiative whether the Tribunal’s calculation is based on April 1, 1991 as the ‘dies a quo,’ as this appears implicitly from the Tribunal’s statement with respect to the day when the expropriation of Wena’s rights occurred.” This explanation does not appear to have convinced even the AHC itself, however. In the next sentence, it continued:84 Although this is outside the scope of examination as required in a proceeding under Article 52(1),85 the Committee has anyhow made its own calculation. In this respect, the Committee con82

Id. at ¶ 96.

83

Id. at ¶ 98.

84

Id. (emphasis added).

Interestingly, the AHC apparently believed that this calculation was outside the scope of Article 52 review because Egypt itself had not made its own calculation and submitted it to the AHC as evidence of the tribunal’s lack of reasoning. In fact, the calculation is outside the scope of Article 52 because the AHC was not authorized to supply missing dates omitted from the original award. 85


cludes that, when taking into account the payment of the amount awarded in the Nile Hotel arbitration on June 14, 1997 and the amounts respectively owed to Wena, before and after this date, together with the respective amounts of interest accrued as stated in the Award, the resulting amount is very close to the total amount as set forth in the Award. The Committee is satisfied that April 1, 1991 is discernible from the Award as the “dies a quo.” In this excerpt, the Wena AHC made extensive use of its newly selfconferred power to supply missing reasons. It “discerned” a date that did not actually appear in the award and attempted to infer all of the other relevant factors that should have gone into the tribunal’s calculation of the interest amount.86 It then combined all of these implicit factors in some unspecified manner in order to arrive at its own estimation of the total. Even after performing these ill-advised mental gymnastics, the AHC was only able to get “very close” to the amount actually reflected in the award. This result might be comical if it were not so egregious. One is left with the distinct impression that the AHC was itself unable to duplicate the tribunal’s calculations. Certainly one cannot follow “how the tribunal proceeded from Point A to Point B, and eventually to its conclusion,”87 as the reasons requirement demands. But desiring, for some unknown reason, to leave the award intact, the AHC pronounced the calculations “close enough.” The AHC’s approach to the interest award, like its approach to the damages award, thus displayed fundamental defects. Faced with an overt lack of reasoning in the tribunal’s award, the AHC should have overturned the award and allowed the parties to refer the matter to a new tribunal. This would have fulfilled the AHC’s obligations pursuant to its express powers under Article 52 of the ICSID Convention. By instead supplying its own reasoning to uphold the award, the AHC undermined the ICSID Convention’s Article 48 reasons requirement and directly contravened its Article 52 ad hoc committee provisions.

86 It bears mentioning that the AHC’s calculation here suffers from the same defects as the original award. Although it mentions the factors that should have been taken into account by the tribunal, it fails to specify precisely how they were taken into account. It also declines to outline the AHC’s mathematical calculations and even declines to state the actual figure arrived at. 87

Wena—Annulment, supra note 16, at ¶ 77 (quoting MINE, supra note 64).


VI. WENA’S APPLICATION FOR INTERPRETATION OF THE AWARD: HOW THE REASONING FAILURES OF THE ORIGINAL TRIBUNAL AND THE AD HOC COMMITTEE LED TO FURTHER DISPUTES BETWEEN THE PARTIES The subsequent history of the Wena case demonstrates that a tribunal’s failure to comply with the ICSID Convention’s reasons requirement—especially when combined with an ad hoc committee’s failure to uphold that requirement by annulling an unreasoned decision—can lead to further disputes between the parties. After the Wena AHC declined to annul the original tribunal’s award, Egypt paid to Wena the full amount of damages levied against Egypt in that award. The award was thus fully satisfied from a monetary standpoint. Yet because the original tribunal did not explain its damage calculations in the award, the legal proceedings between the parties did not come to a close with the satisfaction of the judgment. Instead, a state-owned entity known as EGOTH88 (the successor in interest to the Egyptian Hotels Company89) proceeded to press a series of domestic Egyptian legal claims against Wena in an attempt to recover unpaid rents owing under the breach of one of the hotel lease agreements.90 In pressing these claims, EGOTH appears to have interpreted the tribunal’s failure to explain crucial decisions underlying its damage award (such as why the tribunal did not reduce Wena’s award by the amount of Wena’s unpaid rents under the leases) as an indication that the tribunal did not decide these questions at all. If correct as a matter of law, then EGOTH’s lease-based claims against Wena would not be barred by the doctrine of res judicata, leaving EGOTH free to pursue recovery of the unpaid rents in accordance with the lease agreement. Wena, on the other hand, viewed EGOTH’s ongoing legal claims against it as a form of 88

Egyptian General Company for Tourism and Hotels.

Note that Wena had originally brought its claim against Egypt on the basis of a series of actions taken by the EHC, a state-owned entity that ejected Wena from the hotels in 1991 and that later (after possession was restored to Wena in 1992) prevented Wena from operating the hotels in a normal fashion by denying Wena the necessary licenses over a period of several years. Since the EHC was fully owned and controlled by the Egyptian government, the original tribunal attributed its actions to Egypt. See discussion in Section II. EGOTH, which is also a state-owned entity, later became the successor in interest to the EHC. 89

90 For details of EGOTH’s legal claims against Wena, see Wena Hotels Limited v. Arab Republic of Egypt, ICSID Case No. ARB/98/4, Decision on the Application by Wena Hotels Ltd. for Interpretation of the Arbitral Award dated Dec. 8, 2000 (Oct. 31, 2005) [hereinafter Wena—Interpretation].


unjustified harassment.91 It argued that the tribunal’s expropriation finding meant that Egypt had “totally and permanently deprived”92 Wena of its “fundamental rights of ownership”93 and that, as such, Wena could not possibly have any further outstanding liabilities toward EGOTH under the lease. In an attempt to stop EGOTH from continuing to press leasebased legal actions against it, Wena filed an application for interpretation of the original award. Specifically, Wena asked the interpreting tribunal94 to clarify whether the original tribunal’s finding of expropriation “constituted a total, permanent deprivation of Wena’s rights in the Luxor Lease, such as to preclude subsequent legal actions by Egypt that presume the contrary.”95 At the oral hearing, Wena also asked the tribunal to reaffirm that the date of expropriation was April 1, 1991.96 It further asked the tribunal to interpret the original expropriation finding in such a way as to declare that EGOTH, like Egypt, should be barred from taking any actions contrary to the original award.97 Egypt opposed Wena’s application for interpretation on two grounds. First, it argued that the judgment against Egypt had already been fully satisfied by Egypt’s payment of the damage award. Thus, in its view, no legitimate dispute over interpretation of the award remained to be resolved. And second, Egypt maintained that Wena’s request for an interpretation with respect to EGOTH’s actions exceeded the scope of a permissible interpretation under the ICSID Convention, since EGOTH was not a party to the original arbitration, and its disputed actions did not occur until after the full satisfaction of the original judgment.

91 See id. at Part II.A. In addition to complaining of the expense and burden of defending itself against ongoing legal claims, Wena complained that EGOTH’s pursuit of the claims was damaging its reputation and costing it business opportunities both in Egypt and abroad. 92

Id. at ¶ 31.

93

Id.

Apparently it was not possible to reconstitute the original tribunal for purposes of deciding on the application for interpretation. Thus, in accordance with the ICSID Convention, a new tribunal was constituted to review and interpret the original award. Id. at ¶ 8. 94

95

Id. at ¶ 33 (emphasis omitted).

Id. at ¶ 34. It asked for this clarification because some of EGOTH’s legal claims sought to recover unpaid rents arising out of periods after that date. 96

97

Id. at ¶ 96.


The interpreting tribunal admirably worked through each party’s claims in the proceedings surrounding Wena’s application for interpretation of the award. In making its findings, it performed an extensive review of the appropriateness and limits of applications for interpretation under the ICSID Convention and under international law.98 It then carefully and conscientiously applied the principles it elucidated in this review to each of the claims raised by the parties.99 In doing so, the interpreting tribunal appears to have done its best to comply with the ICSID Convention’s Article 48 reasons requirement. But the interpreting tribunal was nevertheless limited in its interpretations by the absence of reasons in the original award. For example, in affirming the date of expropriation to be April 1, 1991, the interpreting tribunal had to rely upon the ad hoc committee’s inference of this date, since no such date was given in the original award.100 Likewise, the interpreting tribunal was forced to make inferences based upon the original award’s amount and type of damages in order to clarify that the award should be interpreted to mean that Egypt’s expropriation had “totally and permanently deprived Wena of its fundamental rights of ownership.”101 The original award did not provide explicit reasons for either of these interpretations; yet because the AHC left the award intact, the interpreting tribunal was left with no choice but to issue interpretations as best it could under the circumstances given a dearth of explicit reasons. The stark impact of the original tribunal’s and AHC’s omissions of reasoning came into sharp relief with the interpreting tribunal’s pronouncement concerning the most controversial portion of Wena’s interpretation request—whether or not the award’s expropriation finding should be deemed to preclude EGOTH (as distinct from Egypt) from pursuing any further lease-based legal claims against Wena. Here, the interpreting tribunal’s hands were tied. The original award dealt solely with Egypt’s liabilities under the IPPA. The original tribunal did not address Wena’s rent liabilities under the lease. It made no mention of them either in its expropriation findings or in its damage calculations. The AHC’s attempt to “cure” the defective award similarly neglected to 98

Id. at ¶¶ 73–92.

99

Id. at ¶¶ 93–131.

100

Id. at ¶ 124 n.108.

101

Id. at ¶¶ 120–123.


determine whether the tribunal had addressed Wena’s potential liabilities under the lease. Thus, the interpreting tribunal was forced to conclude that “the Original Tribunal did not decide on the consequences of the expropriation on the legal relationships between Wena and third parties.”102 On this basis, it denied Wena’s request to interpret the award as binding upon EGOTH. It is debatable whether this outcome was technically correct given the interpreting tribunal’s limited powers under the ICSID Convention. Convincing arguments can be made on both sides. But even if technically correct, the outcome was clearly unjust toward Wena. Having prevailed in its IPPA-based expropriation claims against Egypt, there is no reason why Wena should have been forced to endure subsequent litigation by EGOTH arising out of the very same dispute. On the other hand, the original award was arguably unjust toward Egypt in that it awarded full damages to Wena without ever addressing whether those damages should be reduced by the amount of the rents allegedly owed to Egypt under the lease. In short, the original tribunal’s failure to comply with the ICSID Convention’s reasons requirement, combined with the AHC’s failure to uphold that requirement, led to a situation in which neither party received a just and final resolution of its claims. Such outcomes threaten to undermine the faith of both governments and investors in international investment arbitration. VII. CONCLUSION BITs are consent-based instruments, and the arbitral tribunals tasked with applying their norms are tribunals of limited jurisdiction. Tribunals should therefore be very circumspect in wielding their authority. The ICSID Convention’s Article 48 reasons requirement is designed to encourage this circumspection. It requires tribunals to answer each and every question put before them and demands that they state the reasons behind their decisions. Tribunals should hold this requirement in esteem, not merely because it provides clarity to the parties, but because it also protects tribunals from overstepping their bounds, to the detriment of international investment law’s general legitimacy.103 102

Id. at ¶ 128.

For a more thorough discussion of the ICSID Convention’s reasons requirement and suggested methods of implementing it successfully, see W. MICHAEL REISMAN, SYSTEMS OF CONTROL IN INTERNATIONAL ADJUDICATION AND ARBITRATION: BREAKDOWN AND REPAIR (1992). 103


As the Wena case has shown, a conscious and studied appreciation for the reasons requirement can prevent tribunals from inadvertently expanding an investment agreement’s coverage to claimants not clearly within the ambit of the contracting parties’ original understanding. It can guide their deliberations and ensure that they answer all of the relevant questions put before them—both factual and legal. It can protect individual arbitral panels from accusations of bias or carelessness. And most importantly, it can assure parties to investment disputes—whether governments or private claimants—of more fair and reasonable outcomes. In order to accomplish these laudable goals, the reasons requirement must be taken at its words. Tribunals must be willing to actually state, not merely imply, all of the reasons upon which their conclusions are based. Conclusory statements cannot suffice. For their part, ad hoc committees must be willing to give teeth to the reasons requirement by ensuring that the minimum requirements of international law are met. Namely, to survive scrutiny, reasons must be clearly stated, sufficiently relevant, reasonably sustainable, and capable of providing a basis for the award. Defects in reasoning must not be cured by reviewing committees, but should instead be overturned in accordance with the Convention. If both tribunals and ad hoc committees remain true to the Convention’s reasons requirement in this fashion, then both will serve to reinforce the legitimacy of international investment law, rather than to undermine it.

CHAPTER 9

THE AWARD IN THUNDERBIRD V. MEXICO Santiago Montt*

I.

INTRODUCTION: THE NON-COURTS OF APPEALS DOCTRINE AS A SOURCE OF OBFUSCATION

International investment law tribunals (IIL tribunals)—including both those convened under bilateral investment treaties (BITs) and under the investment chapters of free trade agreements (FTAs)—frequently repeat that they are not courts of appeals.1 International Thunderbird Gaming Corporation v. Mexico, by affirming that “it is not the Tribunal’s function to act as a court of appeal or review,”2 joins therefore a relatively long progeny of cases referring to this “doctrine.” It is argued in this chapter that the non-court of appeals doctrine is usually—and certainly in the Thunderbird decision—a source of obfuscation in IIL adjudication. In its most trivial sense, it is doubtful—though open to discussion —whether BITs permit judicial review as a remedy, that is, the nullification of state measures3 (the North American Free Trade Agreement (NAFTA) certainly does not allow it4). But that does not mean that judicial * LLB, Universidad de Chile. LLM. and JSD, Yale Law School. Associate Attorney, Weil, Gotshal & Manges LLP. 1 See, e.g., ADF Group Inc. v. United States, ICSID Case No. ARB(AF)/00/1, Final Award, ¶ 190 and n.182 (Jan. 9, 2003); CME Czech Republic B.V. v Czech Republic (UNCITRAL ad hoc case), Final Award, ¶ 467 (Mar. 14, 2003); Tecnicas Medioambientales TECMED, S.A. v. Mexico, ICSID Case No. ARB(AF)/00/2, Award, ¶ 119 (May 29, 2003); Encana v. Ecuador (UNCITRAL ad hoc case, Canada/Ecuador BIT), Final Award, ¶ 145 (Feb. 3, 2006) [hereinafter Encana]. 2 Int’l Thunderbird Gaming Corp. v. Mexico, UNCITRAL (NAFTA, Canada/ Mexico), Award, ¶ 125 (Jan. 26, 2006), available at http://www.naftaclaims.com/ Disputes/Mexico/Thunderbird/Thunderbird_Award.pdf [hereinafter Thunderbird Award]. 3 In Occidental Expl. and Prod. Co. v. Ecuador, LCIA Case No. UN 3467 (U.S./ Ecuador BIT), Award, ¶ 202 (July 1, 2004), the tribunal not only issued a damage award, but also declared an Ecuadorian tax decree null and void. 4 See NAFTA, Dec. 17, 1992, U.S.—Can.—Mex., art 1135.1, 32 I.L.M., 289, 646 (entered into force Jan. 1, 1994).

261


review is not present in any way. As Mashaw generally points out, plaintiffs who seek damages “invariably question the legality [in the broad sense, including arbitrariness] of administrative conduct. To that degree, suits against the government and its officials sounding essentially in tort, contract, or property also invite judicial review of administrative action.”5 In the context of the implied judicial review of BIT/FTA damages action, the obfuscation of the non-court of appeals doctrine begins with its many possible meanings. This ambiguity includes at least the following dimensions: first, whether it covers questions of law, questions of fact, or both of them; second, whether it requires the tribunal to remand the relevant issues of domestic law to local courts; third, in the case those issues are not remanded to domestic courts, whether it implies a deferential standard of review of the application of domestic law by domestic administrative agencies, including the question of how much deference deserves the latter. In the specific dimension of questions of law, the obfuscation derives from using the doctrine for denying any relevance of domestic law in IIL adjudication. This standard version of the doctrine argues that when reviewing administrative measures (not judicial decisions6) according to treaty standards, IIL tribunals should decide investment disputes applying only international law and avoiding all issues of domestic public law, particularly when the applicable law defined in the treaty is only international law. 5 JERRY L. MASHAW ET AL., ADMINISTRATIVE LAW: THE AMERICAN PUBLIC LAW SYSTEM: CASES AND MATERIALS 780 (2003). In the United States, the Supreme Court has considered that the implicit judicial review present in tort cases against the government is precisely the reason that Congress chose to ban them. See United States v. S.A. Empresa de Viacao Aerea Rio Grandense (Varig Airlines) et al., 467 U.S. 797, 814 (1984) (“[T]his emphasis upon protection for regulatory activities suggest an underlying basis for the inclusion of an exception [i.e. immunity] for the discretionary functions in the Act: Congress wished to prevent judicial ‘second-guessing’ of legislation and administrative decision grounded in social, economic, and political policy through the medium of an action in tort.”). 6 In contrast to administrative agencies, domestic courts are typically the final arbiters of the interpretation of domestic law. Consequently, there are more reasons to accept (and defer to) their decisions on those issues. See, e.g., Azinian v. Mexico, ICSID Case No. ARB(AF)/97/2 (NAFTA), Award, ¶ 99 (Nov. 1, 1999); Mondev International Ltd. v. United States, ICSID Case No. ARB(AF)/99/2 (NAFTA), Award, ¶ 126 (Oct. 11, 2002); Loewen v. United States, ICSID Case No. ARB(AF)98/3 (NAFTA), Award, ¶ 242 (June 26, 2003); Waste Management, Inc. v. Mexico, ICSID Case No. ARB(AF)/00/3 (NAFTA), Final Award, ¶ 48 (Apr. 30, 2004) [hereinafter Waste Management II].

The Award in Thunderbird v. Mexico • 263

However, as claimed in this chapter, this is not only legally incorrect but also impossible in IIL adjudication. Because the regulatory state is almost everywhere in the form of Rechtsstaat or using MacCormick’s translation, a “state-law,” international law cannot overlook that “the question [of] what exercises of executive power are valid is a question of law; the political power of the executive is restrained under the authority of law.”7 And the law that contained the most detailed program of public action is, beyond any doubt, domestic public law. Therefore, the conduct of the regulatory state according to that law cannot be inconsequential, not at least when the domestic legal system does not fall below international legal standards. This means that, unless IIL tribunals are ready either to remand domestic law issues to local courts—as two IIL tribunals presided by James Crawford have suggested recently8—or to blindly accept the position of the defendant state on those issues, it is impossible to avoid its application. More precisely, what is exactly unavoidable is the review of previous applications of domestic law by administrative agencies, revision that of course must be done, to some degree, applying that same law. In Jenk’s words, there is no “frank” way to resolve the dispute without undertaking that review.9 High costs are to be paid in terms of the award’s reasoning when following this standard obfuscating doctrine. In the best scenario, it leaves the reader without any clue about how the tribunal resolved some of the key issues for finding (or not) a treaty violation. In the worst scenario, particularly when coupled with the idea of municipal law as fact, it serves tribunals to do precisely the opposite of what they are announcing to do, that is, to act as court of appeals applying full de novo reviewing tests. Thunderbird is a case of particular interest because it shows the extent to which domestic law issues become intertwined in IIL adjudication, particularly in a setting in which the investor was fully deprived, and still no expropriation was found. It also proves the high degree of obfuscation 7 NEIL MACCORMICK, QUESTIONING SOVEREIGNTY. LAW, STATE, AND NATION IN THE EUROPEAN COMMONWEALTH 9 (1999). 8

See, e.g., Waste Management II, supra note 6 and Encana, supra note 1.

See Wilfred Jenks, The Interpretation and Application of Municipal Law by the Permanent Court of International Justice, 19 BRIT. Y.B. INT’L L. 67, 99–100 (1938) (“[I]t is desirable that there should be a franker and fuller recognition than there has been that in a variety of types of cases the functions of the Court necessarily include the interpretation and application of municipal law.”). 9


and confusion that the non-court of appeals introduces in the reasoning of those awards that adhere to it, even of those awards that are generally well-reasoned and well-decided as is the case of the Thunderbird decision. Concretely, as will be shown below, the non-court of appeals doctrine led the tribunal into three reasoning flaws: first, the tribunal “reduced” a legal dispute into a factual one; second, the factual dispute, as depicted by the tribunal, did not play any meaningful role in the conclusions later adopted in the award; and, third, the implicit resolution of the legal controversy can still be found in the interstices of the award, leaving—as indicated—the reader questioning how the relevant issues of Mexican law were dealt with.” II.

FACTUAL BACKGROUND OF THE THUNDERBIRD CASE

Thunderbird Gaming Corporation, a Canadian investor, brought a NAFTA Article 1117 claim against the Mexican government, on behalf of six subsidiaries incorporated in Mexico (subsidiaries referred to as EDM). Arguing violations of NAFTA Articles 1102, 1105, 1110, it sought compensation for damages in an amount of at least $100 million.10 Thunderbird and its vehicles were engaged in the business of operating gaming facilities. Thunderbird planned and established several facilities offering so-called “skill machines” to the general public. Those machines are depicted by the plaintiff “as differing from slot machines in that the skill machine player is able to start and stop the activity at play, to make decisions. . . and to effect, through his skill and dexterity, the outcome of the game.”11 Those elements distinguish them from “slot machines,” where “the player simply pulls the handle and waits to see if he has won anything.”12 The main obstacle to Thunderbird’s plan was Article 1 of the Mexican Federal Law on Games and Raffles of 1947 (FLGR). This article contains a general prohibition worded in the following terms: “games of chance and games involving betting are prohibited in the whole national territory.”13 Given this general and abstract prohibition, in order to establish 10

Thunderbird Award, supra note 2, at ¶¶ 6–8.

Thunderbird, Particularized Statement of Claim of Claimant, at 4, available at http://naftaclaims.com/Disputes/Mexico/Thunderbird/PSOC.pdf. 11

12

Id.

Mexico, Response to a Complaint, ¶ 30, at 7, available at http://naftaclaims. com/Disputes/Mexico/Thunderbird/PSOD.pdf. 13


a legal basis to operate in Mexico, Thunderbird tried to convince the Secretary in charge of regulating gaming and gambling (SEGOB, the Regulator14) to distinguish between “slot machines”—the prohibited ones—whose outcomes depend purely on chance, and “skill machines”— allegedly legal in nature—whose outcomes depend partially on skill. However, as affirmed by Mexico in the case, SEGOB’s regulatory position at the time Thunderbird started its business in Mexico was that machines similar or identical to investor’s skill machines were covered by the prohibition contained in Article 1 of the FLGR. The explanation was that they involved both chance and betting, any of which, standing alone, would have been sufficient to prohibit Thunderbird’s “skill machines.” In fact, SEGOB was in the process of prosecuting a Mexican gaming businessman for the operation of comparable machines, and Thunderbird knew it before making its investment. The essential factual and policy issue at stake in this case was that there are ultimately no human games in which chance does not play any relevant role. Indeed, all games that are not commonly recognized as “chance games,” including sports, involve to a greater or lesser extent some level of chance. Consequentially, legislators and regulators in any jurisdiction must draw a line somewhere between games of pure chance and “ideal” games of pure skill. Hence the term “games of chance” and, to a certain extent, “games involving betting” in FLGR are open-textured concepts that confer SEGOB discretionary powers to draw the precise regulatory line within the permissible interpretative area defined by the general principle contained in the Statute. Thunderbird intended to exploit precisely this uncertainty of the law in its favor, by trying to convince the Regulator that its skill machines could be found legal given the regulatory discretion of SEGOB. In other words, Thunderbird never acknowledged that its machines were illegal under the dominant interpretation of FLGR.15 In its opinion, the application of FLGR with respect to skill machines was uncertain, permitting the Regulator, inside its zone of discretion, to accept the legality of the

14 According to the Article 3 of FLGR, SEGOB is authorized “to regulate, authorize, control and oversee games when they involve betting of any kind, as well as raffles, with the exception of the National Lottery, which will be governed by its own law.” 15 See, e.g., Thunderbird, Post-Hearing Brief of Claimant, at 53, available at http://naftaclaims.com/Disputes/Mexico/Thunderbird/Thunderbird%20Post%20He aring%20Brief.pdf. See Section V.


investment.16 Furthermore, under Mexican constitutional law, uncertainties in the interpretation of the law must not be read as destroying the default position of “liberty.”17 Given this allege uncertainty, and before making its investment, Thunderbird decided to request—by way of a Solicitud—a Legal Opinion from the Regulator—the Oficio—in order to force the Regulator to take a position and to receive special assurances in relation to the legality of its machines under Mexican law. But, when requesting the Legal Opinion from SEGOB, Thunderbird went against the facts and described its machines as video games not involving chance at all.18 By letter of August 15, 2000, SEGOB issued a formal response to Thunderbird. This response stated that the machines were lawful if two conditions were met: (1) Thunderbird’s own factual description of the machines’ nature and functionality, concretely, the specific assertion that the machines did not involve the “the intervention of luck or gambling, but rather the user’s ability and skillfulness,”19 was truthful; and (2) that in operating such machines, “there shall be no intervention of luck or gambling.”20 However, SEGOB also noted, contradictorily, that Mexican law prohibited slot machines, meaning those machines in which “the principal factor of the operation is luck or gambling and not the user’s ability of skillfulness.”21 After the issuance of this Oficio, Thunderbird considerably expanded its business operations in Mexico (it had started its investment before 16 See, e.g., id. at 20 (“the Ley Federal de Juegos y Sorteos [FLGR] vested Mexican officials with the discretion necessary to permit the use of equipment involving chance and betting.”). 17

Id. at 52.

Thunderbird Award, supra note 2, at ¶ 50 (emphasis added). The relevant part of Thunderbird’s Solicitud, see id., deserves to be quoted in full: 18

The video game machines for games of skills and ability, which the entity I represent commercially exploits, are devices for recreation which have been designed for the enjoyment and entertainment of its users. In these games, chance and wagering or betting is not involved, but the skills and abilities of the user who has to align different symbols on the machine screen by touching the screen or pushing buttons in order to stop the wanted symbol from several other symbols which spin in a sequential manner in each of the lanes or squares of each video game. The user has to align symbols in an optimum combination to receive a ticket with points which can be traded for goods or services; as this is already done at different locations in the country. 19

Id. at ¶ 55, quoting SEGOB’s Oficio, ¶ 5.

20

Id.

21

Id., quoting SEGOB’s Oficio, ¶ 6.


that date). However, in December 2000, Vicente Fox became the President of Mexico. His administration, controlled by a political party different from the one issuing the Oficio, acted on a conservative impulse, which commanded that any “irregularity” under the gambling law would not be permitted.22 The newly staffed SEGOB opened an administrative procedure against Thunderbird, with the purpose of determining whether or not the “skill machines” were forbidden by FLGR. Sticking to the FLGR general prohibition, the Regulator concluded that the machines operated by the investors were slot machines prohibited by FLGR,23 specifically affirming that they involved games of chance.24 SEGOB proceeded then to issue an administrative order closing all investors’ subsidiary facilities and seizing all investors’ machines as forbidden gambling equipment. After witnessing the full destruction of its investment, Thunderbird initiated NAFTA Chapter 11 arbitration against Mexico. III. REDUCING THE DISPUTE TO FACTS: THE SCOPE OF THE NON-COURT OF APPEALS DOCTRINE IN THE TRIBUNAL’S DECISION Thunderbird brought its NAFTA claim against Mexico, arguing that its activities were approved and that Mexico later reversed course and reneged upon its prior approval. The investor argued that the Oficio provided “negative clearance” by reference to the general prohibition of the Statute.25 From the perspective of this chapter, Thunderbird’s position included two relevant allegations. First, a detrimental reliance argument, according to which the assurance given by SEGOB in the Oficio created legitimate expectations in the investor;26 second, an abuse of power allegation, according to which the new administration of SEGOB concluded that the machines were “slot machines” in an arbitrary and abusive way.27 As a way of introduction, it is worth noting the general legal context of this case. Customary international law provides that non-discrimina22 See the description provided by the dissenting arbitrator: Thomas W. Wälde, Partial Separate Opinion, ¶¶ 18, 77, and 102–110 (see, in particular, n.144), available at http://www.naftaclaims.com/Disputes/Mexico/Thunderbird/Thunderbird _Dissent.pdf [hereinafter Wälde, Separate Opinion]. 23

See Mexico, Response to a Complaint, supra note 13, ¶ 65.

24

See infra notes 34 and 35.

See Thunderbird, Post-Hearing Brief of Claimant, supra note 15, ¶ 7.3, at 26. See also Thunderbird Award, supra note 2, at ¶ 139. 25

26

See Thunderbird Award, supra note 2, at ¶¶ 137–140.

27

Id. at ¶¶ 185–186.


tory forfeiture and prohibition of illegal activities in accordance with domestic law does not constitute a compensable expropriation.28 Moreover, it is even possible to conclude that an investment or vested rights did not exist from the very first place. Dicta in this sense may be found in the case law of the Iran-U.S. Claims Tribunal,29 and also in the case law of the European Court of Human Rights.30 A detailed explanation about why this situation does not amount to a violation of international law is also provided by Sohn & Baxter’s Draft Convention on the International Responsibility of States for Injuries to Aliens.31 In this case, the tribunal started its analysis by recognizing that Mexico had “a wide regulatory ‘space’ for regulation; in the regulation of the gambling industry, governments have a particularly wide scope of regulation reflecting national views on public morals. Mexico can permit 28 See IAN BROWNLIE, PRINCIPLES OF INTERNATIONAL LAW 511–12 (6th ed. 2003) (“Jurists supporting the compensation rule recognize the existence of exceptions [to the rule of expropriations], the most widely accepted of which are as follows: under treaty provisions; as a legitimate exercise of police powers, including measures of defence against external threats; confiscations as a penalty for crimes; seizure by way of taxation or other fiscal measures; loss caused indirectly by health and planning legislation and the concomitant restrictions on the use of property; the destruction of property of neutrals as a consequence of military operations, and the taking of enemy property as part payment of reparation for the consequences of an illegal war.”). See also AMERICAN LAW INSTITUTE, RESTATEMENT (THIRD) OF FOREIGN RELATIONS LAW OF THE UNITED STATES, § 712, cmt. g (“A state is not responsible for loss of property or for other economic disadvantage resulting from. . . forfeiture for crime.”). A more complete study of this issue can be found in Andrew Newcombe, The Boundaries of Regulatory Expropriation in International Law, 20 ICSID REV.—FOR. INV. L.J. 1 (2005). 29 See SEDCO Inc. et al. v. National Iranian Oil Company et al., 9 C.T.R. 248, 275 (1985). 30 See AGOSI v. The United Kingdom, Case No. 9118/80, [1986] ECHR 13, ¶ 51 (Oct. 24, 1986) (“The forfeiture of the coins [which applicant AGOSI had sought to import into the United Kingdom against the law] did, of course, involve a deprivation of property, but in the circumstances the deprivation formed a constituent element of the procedure for the control of the use in the United Kingdom of gold coins such as Kruegerrands [finding therefore that the claimant is not entitled to compensation].”). 31 Louis B. Sohn & R.R. Baxter, Responsibility of States for Injuries to the Economic Interests of Aliens, 55 AM. J. INT´L L. 545, 551 (1961) (Art. 9.2: “A destruction of the property of an alien resulting from the judgment of a competent tribunal or from the action of the competent authorities of the State in the maintenance of the public order, health, or morality shall not be considered wrongful, provided there has not been: (a) a clear and discriminatory violation of the law of the State concerned; (b) a violation of any provision of Article 6 to 8 of this Convention; (c) an unreasonable departure from the principles of justice recognized by the principle legal systems of the world; or; (d) an abuse of the powers specified in this paragraph for the purpose of depriving an alien of his property.” (Emphasis added.)).


or prohibit any form of gambling as far as the NAFTA is concerned.”32 Leaving aside the ratione temporis issues arising from the fact that the FLGR was passed in 1947, the tribunal recognized that the Mexican legal system in the field of gambling was not below international minimum standards (IMS) embedded in NAFTA Articles 1105 and 1110. In consequence, the tribunal accepted that “gambling is an illegal activity” in Mexico, and therefore, that FLGR was not, as such, a violation of NAFTA.33 But of course, the case did not end there. Plaintiff was not claiming against FLGR, but against SEGOB’s arbitrary application of that statute. In other words, the application of a domestic statute, not in itself below IMS, was in this case the alleged cause of the breach of international law. For full clarity of this key dimension of the case, two steps—which will be refined later in this chapter—must be clearly distinguished: Step 1/question of fact: SEGOB found that Thunderbird’s machines involved chance;34 Step 2/question of law: Given that finding of fact, and given the Regulator’s understanding of FGLR in the sense that any involvement of chance renders gaming into gambling, SEGOB characterized Thunderbird’s machines as slot machines and, more generally, as illegal gambling equipment.35 Undoubtedly, Step 2 was the crux of the case; as Thunderbird affirmed in its pleadings, “the arbitrary application, or misapplication, of Mexican law to the EDM operations is the heart of this dispute.”36 So, how did the Thunderbird tribunal approach these issues after recognizing defendant’s wide regulatory space? As mentioned, the tribunal found shelter in the non-courts of appeals doctrine. But what was the 32

Thunderbird Award, supra note 2, at ¶ 127.

33

Id. at ¶ 124.

Claimant quotes the relevant fact finding. See Thunderbird, Particularized Statement of Claim of Claimant, supra note 11, at 95: “Therefore, we [SEGOB] are in the presence of games of chance, in which fortune, the unpredictable, the indeterminate, coincidence and luck are the elements that operate the machine in question.” 34

35 Claimant also quotes SEGOB’s legal conclusion. See id., 97: “It is concluded [by SEGOB] that the cash slot machines operated by Entertainmens (sic) de Mexico S. de R.L. de C.V., in the national territory, fall within the prohibition of the Federal Law on Gaming and Lottery.” 36 Thunderbird, Statement of Reply of Claimant, at 32, available at http://naftaclaims.com/Disputes/Mexico/Thunderbird/Thunderbird%20-%20Reply% 20Memorial.pdf.


exact understanding of the non-court of appeals doctrine followed by the tribunal? Did the doctrine cover Step 1, Step 2 or both of them? If Step 2 proved to have any bearing on the case, did the tribunal consider remanding those issues to local courts? And, if the tribunal were not ready to remand, should it be deferent towards SEGOB’s findings of fact and law? How much deference should it give? Even though the tribunal did not explicitly answer any of these questions, the award’s reasoning shows that a position was adopted on this matter. The tribunal defined its mission in terms of examining whether SEGOB’s measure “were consistent with Mexico’s obligation under Chapter Eleven of NAFTA.”37 In contrast, the application of Mexican Law was held to be out of the scope of the tribunal’s duties: The Tribunal’s role in this arbitration is not to determine whether the EDM machines were prohibited gambling equipment under the Ley Federal de Juegos y Sorteos [FLGR], as acknowledged by both Parties. It is not the Tribunal’s function to act as a court of appeal or review in relation to the Mexican judicial system regarding the subject matter of the present claims, or in relation to the SEGOB administrative proceedings for that matter.38 At the same time, the ttribunal did not mention the possibility of remanding domestic law issues to local courts. It simply held that Step 2 was covered by the non-court of appeals doctrine and therefore outside the scope of the tribunal’s functions. Besides, the tribunal took this stance without recognizing in any way that Step 2—an issue highly debated between the parties—could be of any obstacle for the process of IIL adjudication. In contrast, the tribunal appears to have reached the opposite conclusion with respect to Step 1. An IIL tribunal can therefore act as a court of appeals with respect to findings of fact made by domestic regulators, or at least, that seems to be implied in the award’s reasoning. In fact, the tribunal took a decision over the involvement of chance in the operation of Thunderbird’s machines. Curiously, it did it only after excusing itself from so doing: “the Tribunal does not need to enter into a detailed technical dis-

37

Thunderbird Award, supra note 2, at ¶126.

38

Id. at ¶ 125.


cussion as to the precise nature and functionality of the machines . . . [because] it is not up to the Tribunal to determine the legality of the machines under the Ley Federal de Juegos y Sorteos.”39 The tribunal then concluded that the machines did involve a considerable degree of chance: Thunderbird has submitted evidence in support of its contention that the EDM machines are “skill” machines, whereas Mexico has provided evidence establishing, according to Mexico, that the machines in question are “tragamonedas” [slot machines] prohibited under Mexican Law.40 . . . The Tribunal notes that the machines operated by EDM are equipped with computerized random generators and that it is possible to set the level of payouts, and thus the odds for winning . . . The Tribunal notes further that the machine’s percentage of payout is not visible or otherwise known to the player. The Tribunal infers that the operation of these video games machines with a built-in and modifiable random number generator involves a considerable degree of chance, and that by adjusting the payout rate the machine operator can manipulate the odds for winning regardless of the skill of the player.41 Why did the tribunal decide to limit the non-court of appeals doctrine to questions of law only? The award does not give any explanation in this regard. Then, if the tribunal decided to review questions of fact, which standard was applied to reviewing such findings? The award is silent on this issue too. A careful analysis shows two possible answers. On the one hand, by presenting the issue as “findings of fact” and not as a “review of previous findings of fact” the tribunal seems to be reviewing de novo. But, on the other hand, the tribunal presented its review as a nondetailed technical discussion, a consequence of the fact that its mission was not “to determine the legality of the machines under the Ley Federal de Juegos y Sorteos.”42 Although this can be interpreted as a deferential review, the reader is ultimately left with no clue in this important matter. But this is not all. The award’s use of this partial form of the noncourt of appeals doctrine—which applies only to findings of law and not to findings of fact—should be criticized also under another and more 39


40

Id. at ¶ 134.

41

Id. at ¶ 136 (emphasis added) (internal citations omitted).

42

Id. at ¶ 135.


important account. In this case, the tribunal not only decided to review a previous administrative agency’s finding of fact, but also to solely make a finding of fact, that is, to keep the controversy only as Step 1. The tribunal did not hold, anywhere in the award, that Thunderbird’s machines were slot machines and therefore illegal gambling machines under Mexican law. Thus, the tribunal strategically coupled the partial non-court of appeals doctrine with a “reduction” of the controversy from Step 2 to Step 1. By doing this the tribunal did not resolve the questions the parties were posing before it. The controversy was undoubtedly a Step 2 dispute; even the tribunal depicted it in those terms: “Thunderbird has submitted evidence in support of its contention that the EDM machines are ‘skill’ machines, whereas Mexico has provided evidence establishing, according to Mexico, that the machines in question are ‘tragamonedas’ [slot machines] prohibited under Mexican Law.”43 Moreover, there was not even a controversy between the parties at the Step 1 level. That is, the tribunal mischaracterized the case by reducing it to Step 1. In its main pleadings, Thunderbird did not claim that its machines did not involve chance; it affirmed that chance was not the preponderant or principal factor. Again, the key question was of course whether the level of chance involved in the operation of the machines was sufficient to properly justify a legitimate characterization of those machines as illegal gambling equipment under FLGR. Beyond these initial critiques, the next section will analyze the internal coherence of the award’s reasoning, particularly the consequences of reducing a Step 2 controversy to a Step 1 controversy. IV. LOOKING FOR THE TRIBUNAL’S REASONING AND INTERNAL COHERENCE: THE DECISION ON THE LEGITIMATE EXPECTATIONS/FAIR AND EQUITABLE TREATMENT CLAIM Following the tribunal’s analysis, we already know, as a matter of fact, that the operation of Thunderbird’s machines involved a “considerable degree” of chance. Now, what can the tribunal conclude from those facts? In other words, why did the tribunal enter into such considerations about the nature and functionality of the investor’s machines?

43

Id. at ¶ 134 (square brackets in the original).


The obvious answer would be to assess the alleged arbitrariness of SEGOB’s legal characterization of Thunderbird’s machines as illegal gambling equipment (Step 2). During the arbitral proceedings, both parties made affirmations in this sense.44 The connection between the pure facts and domestic law was indeed particularly stressed by the Mexican expert witness: [Question (cross-examination)]. Professor Rose, . . . is it your opinion and general position that the introduction of a skill-stop feature into a gaming device does not make it a lawful gaming device? [Answer]. Whether something is lawful or not depends on the law of the jurisdiction, so there is no way to know unless you know the law of the jurisdiction.45 However, by adopting the non-court of appeals doctrine, the tribunal appears to be expressing its unwillingness to review any possible arbitrariness with respect to findings of law. Instead, it presented the purposes of its findings of fact in the following terms: “Both Parties have argued that the functionality of the EDM machines is relevant to certain issues pending before the Tribunal. . . . The Tribunal agrees that the nature and functionality of the machines may be relevant in considering certain issues of this arbitration.”46 But, what are those certain issues? The tribunal, surprisingly, does not provide any further explanation. It merely recited the position of the par-

44 See Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 17 (“[T]he functionality of the machines is relevant . . . in assessing Mexico’s justification, or lack thereof, for seizing Thunderbird’s investment enterprises.”) See also Mexico, Escrito de Mexico Posterior a la Audiencia, ¶ 138, available at http://naftaclaims. com/Disputes/Mexico/Thunderbird/Thunderbird%20-%20Mexico%20%20Post%20hearing%20Submission.pdf (“El funcionamiento de las máquinas puede ser relevante para establecer si la determinación de la SEGOB de que EDM realizaba actividades ilícitas por tratarse de máquinas de juegos prohibidos, y la consecuencia aplicación de las sanciones previstas en la ley fue realmente idiosincrática, aberrante o arbitraria, si fue ‘tan injusta o arbitraria que el trato equivale a un nivel que resulta inaceptable desde la perspectiva internacional’ [citing S.D. Myers], de modo que constituye una violación [del] nivel mínimo de trato que requiere el derecho internacional.”). 45 Prof. Rose, Transcripts of Hearing—Day Three, at 733–34, available at http:// naftaclaims.com/Disputes/Mexico/Thunderbird/Thunderbird%20%20Hearing%20Transcript%20-%20Day%2003.pdf (emphasis added). 46

Id. at ¶¶ 134–135 (emphasis added).


ties in this respect, without expressing any independent consideration. And the tribunal could not rely on the parties for this answer, because they were not pointing to the same certain issues. This divergence was a direct consequence of the fact that both parties were arguing in favor of very different versions of the non-court of appeals doctrine. Indeed, Mexico presented the non-courts of appeals doctrine in order to immunize SEGOB’s determination of Thunderbird’s machines as slot machines, that is, with the purpose of receiving full deference from the tribunal.47 Of course, Thunderbird disagreed with that interpretation, observing the disastrous consequences that such an extreme deferential stance would have for IIL adjudication: [Mexico] seeks to (1) recast the issues before the Tribunal so as to avoid any scrutiny of its arbitrary conduct towards Thunderbird and (2) make it impossible for claimant, much less any other NAFTA claimant, to enforce treaty obligations of ‘fair and equitable treatment’ undertaken by the parties under the NAFTA. Simply put, Mexico wishes to ‘stack the deck’ against Thunderbird in clear contravention of the obligations it accepted under the NAFTA. Thunderbird agrees that it is not for this Tribunal to analyze and independently determine whether the EDM skill machines were legal or illegal under Mexican law. That would not be the proper function of a NAFTA Tribunal under these circumstances. However, the Tribunal also cannot do what Mexico asks it to do—simply defer to determinations made by Mexico’s so-called ‘competent authority’ without consideration of the facts and circumstances underlying those determinations. Mexico’s entire defense is based upon this premise. It advocates complete deference to its administrative and judicial processes, but ignores the substantial evidence establishing the arbitrary and discriminatory means by which such processes were used against Thunderbird and its EDM enterprises.48 Thunderbird’s own position with respect to the non-court of appeals doctrine requires a detailed analysis, one that will be provided below. For purposes of this section, it is enough to note claimant’s rejection of Mexico’s full deferential thesis. 47 Id. at ¶ 133 (“Mexico recalls that the Tribunal does not have jurisdiction to determine . . . whether the machines in question were prohibited under Mexican law.”). 48

See Thunderbird, Statement of Reply of Claimant, supra note 36, at 4.


At the same time, it is also crucial to observe Thunderbird’s legitimate expectations argument—clearly its most important one—and its connection with the non-court of appeals doctrine. The investor argued that it had received assurances, creating legitimate expectations on its side. Its argument had two prongs: first, that according to the Oficio its machines were lawful as a matter of Mexican law; second, that, in any case, the Oficio had defined an interpretation of FGLR, embodying the “principal or preponderant factor” test.49 So, by not assessing its machines according to the latter, and by not concluding that they were skill machines according to the latter, SEGOB acted arbitrarily.50 Thunderbird tried here to present its case as a change of policy, and for that purpose it invoked a different version of the non-court of appeals doctrine, whose main objective was to impede the arbitral tribunal to “deduce” or “create” its own new test.51 The alleged change occurred when comparing the preponderant test defined in the Oficio and the ex post no-chance test applied by SEGOB when closing the investor’s facilities.52 Besides, the change in policy was not effected in order to return to an “original” no-chance policy; in other words, according to Thunderbird, SEGOB issued the Oficio not contra legem, but praeter legem, that is, in

49

See Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 17.

50 See Thunderbird, Statement of Reply of Claimant, supra note 36, at 12 (“There is no evidence that Thunderbird and its EDM entities operated machines in which the ‘principal factor of the operation was luck or gambling and not the user’s ability of skillfulness.’ There is no evidence that Thunderbird and its EDM entities operated machines in any manner violative of the directives of the August 15, 2000 opinion letter.”). See also, Thunderbird Award, supra note 2, at ¶ 131 (According to Thunderbird, the only criteria relevant for the Tribunal to determine the functionality of the machines is the standard set forth in the Oficio, that is, whether “the principal factor of the operation is luck or gambling and not the user’s ability of skilfulness.” Thunderbird contends that the machines in question met the applicable criteria and were as a matter fact established by the evidence before the Tribunal, to be “skill machines” operated in accordance with the Oficio.”). 51 Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 15 (“This Tribunal has no role in relation to the Mexican judicial system regarding the subject matter of this case. It does not stand as a domestic court of appeal.”). 52 Thunderbird, Particularized Statement of Claim of Claimant, supra note 11, at 85 (“Fourteen months after it’s issuance of the opinion letter, Gobernacion unilaterally reversed its position and forcibly closed the facilities at Matamoros, Nuevo Laredo and Reynosa. In doing so, Mexico destroyed investment enterprises, created in reliance upon the word of the Mexican government, worth tens of millions of dollars.”).


the context of the lack of legal determinacy of Mexican law. In its closing arguments, Thunderbird summarized its position: Accordingly, whether the machines were in compliance with the rudimentary prohibition set out in that same legislation cannot be at issue in this case. The question of functionality is material to whether the EDMs were entitled to reasonably rely upon the SEGOB letter of August 15, 2000 and whether other domestic operators received more favorable treatment by being permitted to operate similar machines while the EDMs were not. The relevant “standards” are those found in NAFTA Articles 1102, 1105 and 1110, and in the applicable rules of international law. That Mexico has failed to provide a more transparent and functional set of domestic standards for gaming machines merely set the stage for how and why the relevant NAFTA standards were breached.53 The question now is whether the tribunal’s strategy of reducing a Step 2 controversy to Step 1, of applying a partial non-court of appeals doctrine, and of resolving all issues as exclusively matters of international law, was coherent and well justified given the parties’ claims and defenses with respect to the legitimate expectations claim. The tribunal concluded in this respect that “the Oficio [did] not [generate] legitimate expectations upon which EDM could reasonably rely in operating its machines in Mexico”54 and consequently that there was no breach of NAFTA. With respect to the first prong of the legitimate expectations argument, the tribunal dismissed the claimant’s allegations. It based this rejection on the fact that the investor had provided “incomplete and inaccurate” information to SEGOB in the Solicitud,55 particularly when affirming that the operations of its machines did not involve chance.56 The tribunal concluded that, The Solicitud is not a proper disclosure and that it puts the reader on the wrong track. The Solicitud creates the appearance that the machines described are video arcade games, designed 53

Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 20.

54


55

Id. at ¶ 151.

56

Id. at ¶¶ 152–153.


solely for entertainment purposes. . .57 Thunderbird was the moving party presenting a “Solicitud” to the Mexican administration; one would therefore expect that the moving party supply adequate information and make a proper disclosure. In the Tribunal’s view, the Solicitud did not give the full picture, even for an informed reader.58 The tribunal also held that the content of the Oficio was also not sufficiently unambiguous and unqualified so as to give place to legitimate expectations. The tribunal noted that any possible assurances contained in the Oficio were expressly conditioned and qualified by SEGOB as depending on the facts indicated by the claimant in the Solicitud.59 So, if statements contained in the latter were not true, then it was not possible for the Oficio to raise any expectation on which the claimants could have relied. In defense of the tribunal’s reasoning, one could argue that this was the main purpose of assessing the nature of functionality of the machines. The certain issues would be, if this is correct, to check whether the investor had been honest with the Regulator, and, therefore, whether its expectations deserved to receive NAFTA Chapter 11 protection. However, if this was all, then characterizing the machines would have been completely unnecessary. As mentioned before, in the arbitration, the investor, by arguing that its machines were predominantly of skill, admitted that their operation involved chance. Hence, we have to conclude that this was not sufficient justification for reducing the controversy to a Step 1 and for applying the partial non-court of appeals doctrine. With respect to the second prong of the legitimate expectations argument, the tribunal’s reasons were completely unconvincing. Here, the claimant’s interpretation of the Oficio followed this logic: (1) the Oficio says that the FLGR prohibits slot machines; (2) then it defines slot machines as those involving chance as the preponderant factor; (3) therefore, machines whose operation involves predominantly skill, such as those operated by Thunderbird, are not illegal slot machines under Mexican law; (4) consequently, SEGOB breached the international minimum standards of arbitrary and irrational behavior when it changed its 57

Id. at ¶ 155.

58

Id. at ¶ 159.

59

Id. at ¶ 161.


policy and concluded that Thunderbird’s machines were slot machines without characterizing them in accordance to the predominant factor test.60 If the tribunal would have agreed with the claimant in the sense that the Oficio “defined a standard for the operation of skill machines”61—the new standard being “predominant chance”—then the Step 1 factual determination would have been necessary.62 That is, it would have had to make the relevant finding of fact to check whether Thunderbird’s machines passed the predominant chance test. However, the tribunal did not accept the new test and therefore did not agree to assess the machines in accordance with it: According to SEGOB, such machines [slot machines] are devices in which the ‘principal factor’ (‘factor preponderante’ in the original Spanish text) of operation is luck or gambling. SEGOB’s description of slot machines cannot be interpreted a contrario as describing a standard for skill machines, according to which machines in which skill is the ‘factor preponderante’ cannot be treated as gambling equipment. . . . Furthermore, Thunderbird was clearly cautioned in ¶ 7 of the Oficio, “in the machines that your representative operates there shall be no intervention of luck or gambling.”63 It is difficult not to agree with the dissenting arbitrator in his critique of the majority in this aspect. He manifested having “trouble with the tri60 It is interesting to note that the dissenting arbitrator considered that this interpretation was “not implausible.” See Wälde-Separate Opinion, supra note 22, at ¶ 79. This favorable interpretation is described in the following terms: “The main operative message of the letter [Oficio] is: Yes—go ahead with the machines if they are as you qualified them, but bear in mind that machines which involve ‘predominantly’ chance are not allowed” (emphasis added). Id., ¶ 62: “Thunderbird’s proposed interpretation— that machines involving a substantial amount of skill (in addition to chance) could be considered legal—was not implausible.” Id., ¶ 83: “The ‘Oficio’ can therefore be read as suggesting to Thunderbird that it should not qualify its machines as ‘only skill’ (reflecting the label of ‘skill machines’ used for presentation purposes), but as ‘predominantly skill’.” 61

Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 17.

This was indeed Thunderbird’s argument. See Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 20: “The question of functionality is material to whether the EDMs were entitled to reasonably rely upon the SEGOB letter of August 15, 2000 ” (emphasis added). 62

63



bunal’s rejection.”64 Indeed, as he explains, “the use of the ‘predominant’ criterion inevitably leads to the conclusion that if an operation that is ‘predominantly chance’ is forbidden, then an operation that is ‘predominantly’ skill is allowed. Predominant means ‘more than 50%.’ There is a zero-sum relationship between skill and chance.”65 That observation does not only sound correct as a matter of common sense, but it was also backed up by the expert witness presented by Mexico.66 Yet the key point to stress here is that this second prong of the argument could not be one of the certain issues and also could not be dismissed under the Step 1 dimension of the case. Even if the tribunal concluded that the operation of the machines involved a considerable degree of chance, it did not go as far as to hold that chance was the predominant or principal factor in their operation. And more important, instead of applying the predominant chance test plus Step 1, the tribunal simply rejected the test, making Step 1 lose any relevant role when rejecting the legitimate expectations/fair and equitable treatment claim. As will be explored in the last section of this chapter, a proper discussion of the meaning and scope of the Oficio would have led again to issues of domestic law. Yet, instead of dealing with them, the tribunal avoided all difficult questions of domestic law, opting for denying any controversy over the content of the Oficio. As a matter of reasoning, it should be noted that it is not a proper technique to deny the existence of an ambiguity when it is patently clear that there is one. The Oficio is evidently contradictory when defining slot machines according to the preponderant test and then affirming that there shall be no intervention of luck or gambling at all. So, if Step 1 and the certain issues were all related with the claimant’s legitimate expectations argument, then we must conclude that the award’s internal coherence does not pass muster. Summarizing, the tribunal reduced a Step 2 controversy into a Step 1 controversy, mischaracterizing the parties’ arguments. Then, it found a Step 1 conclusion —the considerable degree of chance—that was non-disputed among the

64

Wälde, Separate Opinion, supra note 22, at ¶ 81.

65

Id. at ¶ 80.

66 Prof. Rose, Transcripts of Hearing—Day Three, supra note 45, at 733–34: “there is the difference between whether it is a game of skill or a game of chance, so if it’s predominantly skill, it’s not gambling.”


parties. And finally, it made a factual determination that ended up not having any relevant role in the decision on the merits, unless of course that factual determination was playing a hidden function. That hidden function is the main concern of the next section. V.

GUESSING THE EXPROPRIATION CLAIM DISMISSAL BASIS: FULL DEFERENCE TOWARDS THE REGULATOR’S FINDINGS OF LAW, JUDICIAL REVIEW OR MUNICIPAL LAW AS FACTS?

As a matter of Mexican law, Thunderbird’s case was undoubtedly very weak. Indeed, its national treatment claim was structured to point to a national investor that had been constantly prosecuted by the Regulator. Therefore, it is not surprising that Thunderbird wanted to avoid reviewing the domestic unlawfulness characterization of its machines according to SEGOB’s interpretation of the general prohibition contained in FLGR. Yet the investor never conceded that its machines were illegal or that it had conducted an illegal investment in Mexico. In contrast, and in an autonomous way with respect to the main legitimate expectations argument, it relied on a subtle argument of Mexican law that deserves closer attention. Thunderbird argued that, before the Oficio, Mexican law was confusing and unclear, in other words, that FLGR had a kind of lacunae with respect to skill machines.67 The Regulator then, by adopting the zero-chance conservative standard, filled the lacunae in FLGR, but it did so after the investor had made its investment.68 Here lies the fundamental dimension of Thunderbird’s particular version of the non-court of appeals doctrine: On the one hand, given this lacunae, the tribunal cannot interpret de novo the general prohibition contained in FLGR to assess whether the skill machines where illegal in Mexico at the time the investor made the investment (i.e., to create a new test). The only thing the tribunal can do without acting as a court of

67 See, e.g., Thunderbird, Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 52–53 (“In this case, it is clear that the state of Mexican law concerning the operation of skill game machines is, at best, rudimentary and unclear. It is undisputed that Mexican law does not define or even address skill machines in the context of its constitutional prohibition against gambling. . . [T]he gaming industry in Mexico was not a good example of a regulated industry.”). 68 Thunderbird, Statement of Reply of Claimant, supra note 36, at 44 (“Mexico cannot escape liability by designating an investment activity ‘illegal’ after the fact.”).


appeals is just to recognize the existence of a lacunae or legal indeterminacy. On the other hand, given the general principle of freedom of enterprise recognized in Mexican Constitution,69 it is Mexico’s burden to prove in the arbitration that there was an actual prohibition of skill machines at the time the investor made the investment.70 Despite the fact that SEGOB completely destroyed Thunderbird’s investment—a full deprivation in the indirect expropriations’ nomenclature—the tribunal rejected the NAFTA Article 1110 claim in one seven-line paragraph of the award. The tribunal simply stated that “[c]ompensation is not owed for regulatory takings where it can be established that the investor or investment never enjoyed a vested right in the business activity that was subsequently prohibited.”71 Here is where the tribunal confronted the crux of the case, that is, the Step 2 dimension of the case (SEGOB’s findings of law). Notwithstanding the fact that the tribunal had been constantly eluding this task throughout the award, it finally held that the claimant did not have vested rights when investing in Mexico. But, why didn’t the investor have vested rights? Surprisingly, the decision does not provide any explicit reason. A thorough reading of the award reveals two paragraphs that may answer this question: In any event, even if Thunderbird had established without doubt that Mexico’s line of conduct with respect to gambling operations was not uniform and consistent, one cannot overlook the fact that gambling is illegal in Mexico. In the Tribunal’s view, it would be inappropriate for a NAFTA tribunal to allow a party to rely on Article 69 Thunderbird, Post-Hearing Brief of Claimant, supra note 15, at 52 (“Because the NAFTA Parties maintain legal systems that support mixed-market economies, vested rights are to be presumed from the default position of ‘liberty’ enjoyed by all legal persons in each territory. In other words . . . it would be expected that a business activity not otherwise clearly prohibited would be entitled to protection from taking. In this case, it is clear that the state of Mexican law concerning the operation of skill game machines is, at best, rudimentary and unclear. It is undisputed that Mexican law does not define or even address skill machines in the context of its constitution prohibition against gambling.”). 70 Thunderbird, Statement of Reply of Claimant, supra note 36, at 3 (“It is Mexico that claims the skill machines were illegal. It is Mexico that claims such illegality warranted the confiscatory actions its (sic) took against the EDM enterprises. It is Mexico that bears the burden to justify its seizure of the EDM facilities.”). See id. at 62–63. 71

Thunderbird Award, supra note 2, at ¶ 208 (emphasis added).


1102 of the NAFTA to vindicate equality of non-enforcement within the sphere of an activity that a Contracting Party deems illicit.72 Thunderbird knew when it chose to invest in gaming activities in Mexico that gambling was an illegal activity under Mexican law. By Thunderbird’s own admission, it also knew that operators of similar machines (Guardia) had encountered legal resistance from SEGOB. Hence, Thunderbird must be deemed to have been aware of the potential risk of closure of its own gaming and it should have exercised particular caution in pursuing its business venture in Mexico.73 The tribunal seems to be saying that the lack of vested rights was due to the fact that these activities were either illegal or highly-probable-to-befound (later)-illegal-by-the-Regulator (hereinafter quasi-illegal) at the time of the investment’s inception in Mexico. Any other reading of the award would mean that the tribunal completely eviscerated the concept of regulatory/indirect takings. If any originally legal investment at the time it is made can be later prohibited by the state without having to pay compensation, then NAFTA Article 1110 and the BITs’ equivalent articles embodying the concept of indirect expropriation would be deprived of all rational content. The tribunal should therefore be criticized here both for what is trying to do and for what it is really doing. On the one hand, the tribunal is trying to persuade the reader that it is adjudicating the case based on the rule that gambling was an illegal activity in Mexico (though, as previously noted, the tribunal carefully avoided characterizing Thunderbird’s machine as gambling/illegal equipment). On the other hand, the tribunal did effectively make a finding of domestic law: it held that the investor did not enjoy a vested right because the investments were either illegal or quasi-illegal from the beginning. All this obfuscation finds its source in the non-court of appeals doctrine, which impedes the tribunal from openly disclosing why it considers that the investor did not enjoy vested rights over its gaming operations. Yet, as already mentioned in the introduction, if an IIL tribunal is not ready to remand the domestic law issues to local courts or to

72

Id. at ¶ 183 (when dealing with national treatment) (emphasis added).

73

Id. at ¶ 164 (when dealing with legitimate expectations) (emphasis added).


blindly accept what the defendant state affirmed over domestic law, it is impossible to avoid reviewing previous applications of domestic law, operation that requires, in a way, and to some degree, to apply that law. One is left wondering how the tribunal moved from the mere fact that Thunderbird’s machines operations involved chance to the legal conclusion that Thunderbird’s gaming activities were illegal/quasi-illegal in Mexico at the time the investment was made. From a methodological perspective, there are three ways the tribunal could have concluded that Thunderbird lacked vested rights and that therefore SEGOB’s measures destroying claimant’s investment were not arbitrary and in breach of NAFTA. The award lacks any clues to help the reader decipher which approach was used. The first possible approach is a radical version of the non-court of appeals doctrine—to blindly trust the defendant state and administrative agencies on issues of domestic law. In this case, it would mean showing full deference to the following issues of domestic law: (1) to SEGOB’s characterization of skill machines as illegal equipment according to domestic law; (2) to the government’s affirmation that there has been no change in gambling policy, that is, that SEGOB’s interpretation of FLGR is either the unique possible reading of the statute or the constant and consistent reading of it in the past. That was in fact the argument presented by Mexico, and that is what the reader is originally inclined to believe from the tribunal’s enunciation and adoption of the non-courts of appeals doctrine.74 But, in its reasoning, the tribunal did not pay any explicit attention and deference to the administrative agency determinations or to the background domestic judicial and administrative jurisprudence. And, from the perspective of the merits, it is difficult not to agree with the claimant when commenting on how damaging the general adoption of this version of the non-court of appeals doctrine would be on a broader scale for investors in IIL adjudication.75 As such, it is not clear at all that the tribunal followed this first alternative. A second approach would be to openly conduct domestic law judicial review but in restricted terms. The level of deference would range 74

Id. at ¶ 125.

See Thunderbird, Statement of Reply of Claimant, supra note 36, at 4 (for full quotation, see supra footnote 48). 75


from the one that sometimes was applied in the past in denial of justice cases—that is, the manifestly unjust application of domestic law test of the Martini case76—to the application of the same level of deference that domestic courts display towards administrative agencies. Indeed, under the authority of the Serbian Loans case, it can be argued that, if required, an IIL tribunal has to apply domestic law as local courts do it, including the level of deference observed toward the executive branch and its agencies.77 And, in this sense, it should not be forgotten that, from a comparative perspective, local courts rarely function as full court of appeals with respect to administrative agencies’ decisions.78 But of course, the formal adoption of the non-court of appeals doctrine—if it clearly means something—is that the tribunal did not accept this second alternative. Note, however, its superiority over its competitors: it is a much clearer and more honest resolution of the central issue in this dispute. This alternative offers open judicial review as a matter of law and not the excessively deferrential and paradoxical outcomes that follow, respectively, from the other two approaches.

76 See, e.g., The Martini Case (Italy v. Venez.), Award (Geneva, May 3, 1930), reprinted in 25 AM. J. INT’L L. 554, 571–72 (1931) (emphasis added) (“The question [under dispute] . . . belongs to those which, in any country, may easily give rise to different interpretation, in the absence either of detailed and precise legislative or contractual provisions, or of well-established jurisprudence. It has not been proven that the interpretation of the Venezuelan Law in the matter, given by the Court of Caracas, was erroneous, much less that it was manifestly unjust.”). 77 This is required by the Case Concerning the Payment of Various Serbian Loans Issued in France (1929), Series A 20, at 46–47 (“For the Court itself to undertake its own construction of municipal law, leaving on one side existing judicial decisions, with the ensuing danger of contradicting the construction which has been placed on such law by the highest national tribunal and which, in its results, seems to the Court reasonable, would not be in conformity with the task for which the Court has been established and would not be compatible with the principles governing the selection of its members. It would be a most delicate matter to do so, especially in cases concerning public policy—a conception the definition of which in any particular country is largely dependent on the opinion prevailing at any given time in such country itself—and in cases where no relevant provisions directly relate to the question at issue. It is French legislation, as applied in France, which really constitutes French law, and if that law does not prevent the fulfillment of the obligations in France in accordance with the stipulations made in the contract, the fact that the terms of legislative provisions are capable of a different construction is irrelevant.”). 78 For example, this is the case of the United States and its well-known Chevron doctrine. See Chevron U.S.A., Inc. v. Natural Res. Defense Council, Inc., 467 U.S. 837 (1984).


The third approach is the opposite of the first, and it finds its basis in a well-known but controversial principle of general international law (also a source of obfuscation): domestic law as facts.79 It should be noted that the relation of this third way with the non-court of appeals doctrine is somewhat paradoxical: On the one hand, domestic law as fact represents the complete negation of the non-court of appeals doctrine. But, on the other hand, it constitutes a means of intruding into domestic law without explicitly recognizing that in doing so the IIL tribunal is acting, to some degree, as a court of appeals. Curiously, in the Thunderbird case, the tribunal also seems to have flirted with this third approach. Early in the award, almost immediately after espousing the non-court of appeals doctrine, the tribunal held that “the international law disciplines of Articles 1102, 1105, and 1110 in particular only assess whether Mexican regulatory and administrative conduct breach these specific disciplines. The perspective is of an international law obligation examining national conduct as ‘fact.’ ” 80 One alternative to reaching the illegal/quasi-illegal conclusion under this third approach would be based on the idea of Thunderbird failing to meet its burdens. That is, that Thunderbird failed to provide sufficient evidence in the sense that SEGOB’s characterization of skill machines as slot machines was an ex post facto decision that lacked sufficient rationality in the light of Mexico’s previous application and FLGR’s practice. The tribunal made some tangential observations in this sense81 but not sufficiently strong as to indicate that this was the decisive approach. As the U.S. district court put it in its recent decision denying Thunderbird’s

79 Case Concerning Certain German Interests in Polish Upper Silesia, 1926 P.C.I.J. (ser. A) No. 7, at 19 (“From the standpoint of International Law and of the Court, which is its organ, municipal laws are merely facts which express the will of States, in the same manner as do legal decision or administrative measures.”). See generally BROWNLIE, supra note 28, at 33–40 and JENKS, supra note 9) (both discussing and criticizing the premise of municipal laws as “facts” before international tribunals). 80


81 Id. at ¶ 183 (“In any event, even if Thunderbird had established without doubt that Mexico’s line of conduct with respect to gambling operations was not uniform and consistent, one cannot overlook the fact that gambling is illegal in Mexico”) (emphasis added). See also id. at ¶ 197 (“[T]he Tribunal cannot find sufficient evidence on the record establishing that the SEGOB proceedings were arbitrary or unfair, let alone so manifestly arbitrary or unfair as to violate the minimum standard of treatment.”). (The latter refers more to procedural than to substantive issues.)


action to set aside the award, “it appears that the tribunal concluded that Thunderbird had not met its initial burden.”82 Another alternative would simply permit the tribunal to apply the typically broad power on fact-finding in international commercial arbitration. That would permit the tribunal to approach the FLGR and its indeterminacy as facts and to conclude, de novo, that a considerable degree of chance converts skill machines into illegal gambling equipment in Mexico. Though there seems to be no explicit basis for affirming that the tribunal adopted this final option, a careful analysis of the award may lead to the opposite conclusion. Indeed, the lack of internal coherence explored in Section IV constitutes evidence in this regard. As observed before, the tribunal opted first to make a finding of fact that was completely unnecessary—the degree of chance present in Thunderbird’s machines. That unnecessary factual finding becomes highly suspicious when, as a matter of Step 2/Mexican law, the tribunal implicitly reaches the conclusion previously adopted as a Step 1 factual matter. In other words, by jumping from the “a considerable degree of chance” fact-finding to the conclusion that Thunderbird lacked vested rights, and by not having demonstrated any explicit reference to Mexican law as applied by judicial or administrative agencies, it appears that the tribunal interpreted the Statute by itself to conclude that skill machines were prohibited gambling equipment under Mexican law. And it did it under the spirit of fact-finding, that is, a mere Step 1 controversy. If this conclusion is correct, we would encounter here the great paradox noted before: by adopting the non-courts of appeals doctrine plus municipal law as facts, the outcome reached by the tribunal would have been precisely the opposite of what it purported to be in the first place. Here, the Thunderbird tribunal would have ended up making crucial findings of law and policy—the degree of chance that makes an activity illegal under Mexican law—completely de novo.83

82 See In re Arbitration between Int’l Thunderbird Gaming Corp., v. United Mexican States, 473 F. Supp. 2d 80, 84 (D.D.C. 2007). The court then added that “Because Thunderbird’s own evidence failed to show a violation of international law, additional rebuttal evidence was therefore unnecessary. In any event, for this court to disturb the award, it would have to be plainly manifest that the tribunal both (1) determined that Thunderbird had met its prima facie burden and (2) refused to require Mexico to overcome the resulting presumption of a violation of international law.” 83

It is highly interesting to note that Thunderbird brought its action seeking to


In sum, the award should be criticized for the high degree of obfuscation introduced with the adoption of the non-court of appeals doctrine. The reader is left wondering how the tribunal reached one of the most important conclusions of the decision—the investor’s lack of vested rights at the time the investment was made. But also, the explicit adoption of the non-court of appeals doctrine left the tribunal facing the two worst alternatives to reach that conclusion: either the tribunal adopted the full deferential non-court of appeals doctrine, or the tribunal adopted the very low deferential test deriving from the idea of municipal law as facts. VI. CONFIRMING THAT THE TRIBUNAL ACTED AS A COURT OF APPEALS— PRAETER LEGEM AND CONTRA LEGEM ASSURANCES AND THE DIFFERENT PROTECTION THEY DESERVE Section IV showed that the award’s dismissal of the second prong of Thunderbird’s legitimate expectations argument—the one affirming that the Oficio established a new test for assessing slot machines, i.e., the preponderant chance test—was highly unpersuasive. This section suggests that a somewhat hidden reason may have weighed more in the tribunal’s decision—that the interpretation of the Oficio advanced by the claimant was contra legem and therefore not subject to the protection of NAFTA. This interpretation would confirm that the tribunal acted as a court of appeals. Indeed, the tribunal’s conclusions with respect to the absence of vested rights and the illegal/quasi-illegal nature of the machines at the time the investment was made—all embedded deeply in the award—require us to take a second look at the way in which Thunderbird’s legitimate expectations arguments were rejected. What really seems to be the case here is that the tribunal concluded that the claimant’s interpretation of the ambiguous Oficio could not stand in the way of SEGOB’s legitimate powers to close Thunderbird’s illegal/quasi-illegal facilities. In other words, a weak and ambiguous Oficio should not be interpreted against the accepted or quasiaccepted understanding of Mexican law. It is important to recall here the tribunal’s words when rejecting the legitimate expectation claim: “Thunderbird knew when it chose to invest in gaming activities in Mexico that gambling was an illegal activity under Mexican law.”84 vacate and set aside the award before the U.S. district court arguing, among other reasons, that “the tribunal exceeded its authority by determining that Thunderbird’s machines were illegal under Mexican law.” (id. at 85). The court did not want to scratch the surface; in this respect, it was satisfied by the tribunal’s express declaration that it was not a court of appeals: “The tribunal made no such determination. Indeed, it was careful to make this point clear. Award ¶¶ 125–27.” (id.). 84



Hence, the bottom line of the tribunal’s undisclosed reasoning seems to be that if the illegality or quasi-illegality existed from the very beginning, then the subsequent ambiguous Oficio must be interpreted in accordance with domestic law, rejecting a possible illegal interpretation of the Oficio. Or, in other words, that ambiguous assurances do not confer legitimate expectations if contra legem or contra-the-established-interpretationof-the-law-in-gray-areas. If this was the case, note in consequence with how much obfuscation the tribunal resolved one of the crucial aspects of the case. Consider moreover that this is one of the points where the dissenting arbitrator reached precisely the opposite conclusion, demonstrating that this must have been part of the deliberative process. In perfectly open and clear terms, Wälde expressly considered that legitimate expectations arising from reasonable readings of ambiguous Oficios, if not contra legem,85 trump dominant previous interpretations of domestic law: The good-faith and legitimate expectations [based on the favorable interpretation of the Oficio] control, for the relationship between the parties (e.g. Mexico and Thunderbird), the way the Mexican gambling law has to be interpreted. Governments can not, against a determination that under the international law-based “fair and equitable treatment” principle a legitimate expectation of a specific interpretation has emerged, invoke a dominant contrary interpretation under domestic law. The implication of this analysis is that the principle of “legitimate expectation” under Art. 1105 of the NAFTA overrides any dominant interpretation of applicable Mexican law on the legality of the operation at issue if SEGOB can be considered to have given—reasonably and legitimately—such an assurance.86 From a policy perspective, the tribunal missed a good opportunity to elaborate on one of the core issues of assurances and their relation with domestic law—whether representations made in Oficios, contra legem, should receive the protection of IIL tribunals. In any case, the outcome in this case seems to have been the correct one: expectations, to be pro85 Wälde, Separate Opinion, supra note 22, at ¶ 93 (“The ‘Oficio’ was also within the competence of the government officials who sign it. Interpretative and similar official assurances and representation must be ‘legitimate,’ i.e., they must be issued by competent officials and not, at least from the due-diligence horizon of the recipient, be against the law.”) (emphasis added). 86



tected, must be legitimate, and expectations contra legem are not legitimate almost by definition. And the importance of this rule cannot be minimized: if expectations are permitted contra legem, public officials could derogate constitutions and statutes using mere Oficios. By altering this basic rule, international investment law would indeed destroy the rule of law and impose the empire of corruption in the developing world.87 By not addressing the issue of lawfulness or unlawfulness of the interpretation of the ambiguous Oficio, the tribunal resolved this issue on grounds not clearly and entirely expressed in the award. In fact, there are essentially two alternatives with respect to this issue: either the domestic lawfulness of assurances does not matter in IIL adjudication, meaning legitimate expectations can be contra legem, or domestic lawfulness does matter, meaning legitimate expectations cannot be contra legem. Which alternative did the Thunderbird tribunal pursue? Under a cursory reading of the award, it can be argued that the explicit adoption of the standard non-court of appeals doctrine reflected the tribunal’s acceptance that expectations can be contra legem. But a detailed analysis, as the one made before in this section, proves precisely the opposite; that is, that the tribunal seems to have considered that at least ambiguous assurances do not confer legitimate expectations if contra legem or contra-theestablished-interpretation-of-the-law-in-gray-areas. VII. CONCLUDING REMARKS The tribunal ultimately dismissed Thunderbird’s claims—correctly from the perspective of the outcome—on two substantive accounts: first, the machines were illegal/quasi-illegal under domestic law, and therefore SEGOB’s behavior could not be qualified as arbitrary and constituting an abuse of right; second, the Oficio—because of how it was obtained and what it said—did not have the sufficient strength to overcome that basic illegality/quasi-illegality and give rise to expectations on which the investor could have relied.

87 Vaughan Lowe, Regulation or Expropriation, 55 CURRENT LEGAL PROBS. 447, 465 (2002) (“If the State’s own law does not permit the government to make such commitments, such a decision may easily appear to put the investor above the law, and to entail the conclusion that individuals within the government have given favours, beyond the limits of what the law allows, to the investor.”).


This proves the assertion made at the beginning of this chapter: it is not only incorrect but also impossible to escape from reviewing, to some degree, the application of domestic law. Here, notwithstanding the fact that an alleged arbitrary behavior with a component of domestic illegality ended up being one of the key issues to be resolved, the tribunal did not clearly express how it concluded that the investor lacked vested rights and how its investments were non-arbitrarily seized by the Regulator. The tribunal preferred to formally avoid this issue and adhered to the standard version of the non-court of appeals doctrine coupled with the idea of municipal law as fact. One final observation should be made. The critique presented here does not intend to defend the old Latin American idea of domestic law as the maximum protection that an investor can claim. The claim is much more modest. If states possess, as the tribunal accepted here, “a wide regulatory ‘space’ for regulation,”88 then IIL tribunals must generally recognize that that space is covered by domestic law, and consequently, that states should be bound by those same rules. In other words, if regulation does not fall below international minimum standards—a key condition that IIL tribunals must also control—then the question of whether the state misapplied its own rules when dealing with foreign investors necessarily has to be a relevant one in finding arbitrariness and, hence, a BIT violation (under either expropriation or fair and equitable treatment clauses). The BIT generation forces us to reconceive the relationship between international law and domestic law or, more exactly, the extent to which international law requires, by renvoi, reviewing administrative agencies’ previous application of domestic law. The challenge now is to build a framework that allows us to fully understand the proper place of domestic law in international investment adjudication.89 The regulatory state continues to be a Rechtsstaat, that is, a state commanded by domestic law. What is an expropriation or unfair and inequitable under international law cannot overlook such a main piece of the institutional structure of the world order.

88


A seminal approach in this direction can be found in Zachary Douglas, The Hybrid Foundation of Investment Treaty Arbitration, 74 BRIT. Y.B. INT’L L. 151 (2004). 89

CHAPTER 10

THE FINAL AWARD IN LOEWEN V. UNITED STATES Dirk Pulkowski*

I.

INTRODUCTION

In two recent complaints against the United States, Mondev v. United States1 and Loewen v. United States,2 ICSID tribunals had to confront a sensitive issue—the proper administration of justice in the domestic courts of a North American Free Trade Agreement (NAFTA) party. While denial of justice is a central component of the classic international law on the protection of aliens, the two disputes were the first to raise a comparable issue within the context of NAFTA. Under what conditions can investors use NAFTA claims as vehicles for obtaining compensation for perceived miscarriages of justice? In the Loewen case, a Canadian investor alleged nationality-based, race-based, and class-based discrimination in jury proceedings before a Mississippi trial court. In the investor’s view, the “introduction of extensive anti-Canadian and pro-American testimony and counsel comments”3 during the Mississippi trial amounted to a violation of the obligation of national treatment pursuant to NAFTA Article 1102. More generally, the “improper administration of civil and criminal justice as regards an alien”4 constituted both a procedural and a substantive denial of justice * Attorney, International trade and arbitration group, Sidley Austin LLP, Brussels. Doctoral candidate in international law, Ludwig-Maximilians-Universität (München); LL.M., Yale Law School (New Haven); Ass. jur. (München). Email: dirk.pulkowski@ aya.yale.edu. 1 Mondev Int’l Ltd. v. United States of America, ICSID Case No. ARB(AF)/99/2), Award (Oct. 11, 2002), 42 I.L.M. 85 (2003), 6 ICSID REP. 192 (2004), 125 I.L.R. 110 (2004); see also the discussion in Chapter 2 of this volume. 2 The Loewen Group, Inc. and Raymond L. Loewen v. United States of America, ICSID Case No. ARB(AF)/98/3, Award (June 26, 2003), 42 I.L.M. 811 (2003), 7 ICSID REP. 442 (2005). 3

Notice of Claim, at ¶ 139 (Oct. 30, 1998), available at http://www.naftaclaims.com.

4

Id. at ¶ 156.

291


in violation of the minimum standard of treatment pursuant to NAFTA Article 1105. The outcome is a series of highly unusual decisions5—a decision on jurisdiction, a final award, and a supplementary decision—in which the Loewen tribunal ultimately dismisses most of the claims on jurisdictional grounds, while ostensibly dismissing one claim on the merits. The decisions have stirred a lively controversy in the investment law community, which has mainly focused on two issues. First, the tribunal has been criticized for its extensive use of obiter dicta. While the detailed discussion of the Mississippi trial over a hundred paragraphs is at the core of the final award, the tribunal ultimately makes no holdings on the substance of Loewen’s denial of justice claim. The emphatic condemnation of the trial as “a miscarriage of justice amounting to a manifest injustice as that expression is understood in international law”6 is a mere obiter dictum.7 Second, serious doubts have been raised as to whether the tri5 Loewen Group, Inc. and Raymond L. Loewen v. United States, Case No. ARB(AF)/98/3, Decision on Hearing of Respondent’s Objection to Competence and Jurisdiction (Jan. 5, 2001) [hereinafter Decision on Jurisdiction]; Loewen Group, Inc. and Raymond L. Loewen v. United States, Award (June 26, 2003) [hereinafter Final Award]; Loewen Group, Inc. and Raymond L. Loewen v. United States, Decision on Respondent’s Request for a Supplementary Decision (Sept. 6, 2004) [hereinafter Supplementary Decision]. Other, purely procedural decisions are not discussed. All awards are available at http://www.naftaclaims.com. 6

Final Award, supra note 5, at ¶ 54.

Generally, obiter dicta allow courts and tribunals to clarify the policies that animate a particular legal concept. Yet the Loewen tribunal retreats to the broad statement that Loewen’s trial was flawed “by any standard of evaluation” (Final Award, supra note 5, at ¶ 119); it thus effectively sidesteps a conscious articulation of the particular standards of NAFTA and general international law. As Don Wallace put it, “the Loewen facts . . . gave the Tribunal a chance to place some solid content in the words and to draw lines. Regrettably, the Tribunal used the words, acknowledged the content, but shrunk ultimately from drawing the lines.” D. Wallace, Fair and Equitable Treatment and Denial of Justice: Loewen v. US and Chattin v. Mexico, in T. WEILER ED., INTERNATIONAL INVESTMENT LAW AND ARBITRATION: LEADING CASES FROM THE ICSID, NAFTA, BILATERAL TREATIES AND CUSTOMARY INTERNATIONAL LAW ch. 18 (2005). Wallace’s (in my view, warranted) skepticism stands in contrast to the more positive view expressed by R.D. Ahdieh, Between Dialogue and Decree: International Review of National Courts, 79 N.Y.U. L. REV. 2029, 2113 (2004): “The Final Award in Loewen might thus be conceived as an outline of the proposed parameters of a denial of justice standard, specifically designed to provoke reaction and response from—and hence a dialogue with—the national courts of the United States, as well as Canada and Mexico. Further elaboration in subsequent Chapter 11 litigation might be expected to follow. Initiation of the 7

The Final Award in Loewen v. United States • 293

bunal has properly applied the customary international law on denial of justice8 and whether it has convincingly incorporated this customary law concept into the malleable NAFTA standard of fair and equitable treatment.9 In light of these perceived shortcomings, one commentator has concluded that this “leading case” on denial of justice, ironically, should be considered “itself a miscarriage and denial of justice.”10 This chapter reveals the extent to which deficient reasoning has contributed to the problematic nature of the decisions. To that end, the focus is on the quality of the reasoning in support of the tribunal’s holdings (“orders”) on questions of jurisdiction and merits.11 Apart from the confusing structure of the final award,12 severe flaws in the tribunal’s dialectical exchange, however, necessitated some preliminary elaboration of the relevant standard, as was accomplished in the “dicta” of the Final Award.” 8 One commentator criticized the award as “inconsistent with the current understanding of denial of justice,” since it erroneously postulated special rules for acts by the judiciary. B.K. Gathright, A Step in the Wrong Direction: The Loewen Finality Requirement and the Local Remedies Rule in NAFTA Chapter Eleven, 54 EMORY L.J. 1093, 1116 (2005). Another commentator questioned the appropriateness of the tribunal’s strict standard of review of the Mississippi proceedings. Since “losing defendants could make similar arguments in any number of cases,” “the bar for finding such a denial [of justice] should be set much higher.” E.A. Young, Institutional Settlement in a Globalizing Judicial System, 54 DUKE L.J. 1143, 1177, 1257 (2005). 9 One observer was puzzled that the award “leaves the reader not knowing which particular parts of the trial failed to comport with the international minimum standard, and fails to give guidance as to what acts in particular might be sufficient to trigger international liability” pursuant to Article 1105 of NAFTA. A.K. Bjorklund, Reconciling State Sovereignty and Investor Protection in Denial of Justice Claims, 45 VA. J. INT’L L. 809, 870 (2005). 10

Wallace, supra note 7.

For this purpose, the extensive obiter dictum scrutinizing the Mississippi trial, which has been the focal point of much scholarly debate, will be set aside. 11

12 The award is divided into 31 sections. Often, the relevance of one section for the discussion in another section remains obscure. For example, in Section XVI the tribunal discusses the “Respondent’s Case that Excessive Verdict Was Caused by Loewen’s Flawed Trial Strategy.” Two sections later, in Section XVIII, the tribunal turns to the discussion of “NAFTA Article 1105.” Yet, logically, Loewen’s trial strategy is relevant only as one out of several considerations for ascertaining whether the international minimum standard pursuant to Article 1105 has been violated. The structure of the award does not reflect this “internal—logical—arrangement” of the underlying legal principles. Myres McDougal, Perspectives for an International Law of Human Dignity, 1959 AM. SOC’Y INT’L L. PROCS. 107, 121.


argumentation render the task of “reverse-engineering”13 the decision difficult, if not impossible. The tribunal’s reasoning on jurisdictional matters is coherent but scanty. The tribunal fails to discuss important arguments explicitly made by the claimants and, at times, provides no justification for important policy choices. To the extent that the awards contain holdings on the merits, the reasoning is clearly deficient. Strikingly, the tribunal fails to specifically address one of Loewen’s claims altogether. At times, the tribunal’s argumentation is incoherent (when the tribunal confuses several levels of analysis with respect to an alleged “substantive” local remedies requirement), at times outright contradictory (when the tribunal enunciates a legal standard which it then declines to apply). II.

THE LOEWEN GROUP INC. V. UNITED STATES

A.

Facts and Procedural History of the Case

The dispute concerns NAFTA claims brought by Raymond Loewen and the Loewen Group Inc., respectively, against the United States. Founded in 1969 by Raymond Loewen, a Canadian national, the Loewen Group Inc. (Loewen) became one of the largest operators of funeral homes in North America. As part of the Group’s expansion strategy, the Group purchased a local funeral home business in Jackson, Mississippi, in 1990. Subsequently, a contractual dispute arose between Loewen and Jeremiah O’Keefe, a regional competitor. After a failed settlement, O’Keefe filed a complaint before Mississippi courts in which he alleged breach of a settlement agreement, common law fraud, and violations of anti-trust law. The actual damages sought were below $6 million (with an additional $30 million in lost future revenue, the recoverability of which under Mississippi law is doubtful). Moreover, O’Keefe requested an award of punitive damages.14 During the jury trial, O’Keefe’s attorney made repeated reference to Loewen’s foreign nationality in order to discredit him in the eyes of the jurors, while emphasizing O’Keefe’s Mississippi roots. In addition, he alleged that Loewen—in contrast to O’Keefe—was essentially catering to “white people.” Moreover, Raymond Loewen was caricatured as the owner of a large, foreign corporation, whereas O’Keefe was portrayed as a local family businessman. In most instances, Loewen’s counsel did not 13 See W.M. Reisman & G. Aguilar Alvarez, How Well Are Investment Awards Reasoned?, supra Chapter 1, for a discussion of “reverse-engineerability” as an ideal of legal reasoning. 14

Final Award, supra note 5, at ¶¶ 3–4.


object to such nationality-based, race-based, or class-based comments. Nor did Judge James E. Graves, who presided over the trial, intervene of his own motion. He also rejected a request by Loewen’s counsel to submit specific instructions to the jury that considerations of nationality, race, or class should not influence their decision.15 Initially, a jury verdict of $260 million was handed down (in which the sum total was erroneously calculated as $300 million). However, this initial verdict was flawed for procedural reasons. The jury had awarded compensatory damages and punitive damages in one and the same verdict, instead of following a bifurcated procedure as required by Mississippi state law. Following a motion for mistrial by Loewen, the jury returned a second, final verdict on punitive damages at a higher amount. The punitive damages verdict totaled $400 million. The jury upheld the previous verdict as far as compensatory damages of $100 million were concerned, without, however, specifying how that sum was related to any actual or potential loss on the part of O’Keefe. Final judgment was entered in November 1995.16 On November 27, 1995, Loewen filed an appeal of the trial court judgment with the Mississippi Supreme Court. Under Mississippi law, a party may pursue an appeal without posting a bond. However, a judgment remains enforceable while the appeal is pending, unless a bond equal to 125 percent of the verdict appealed is posted as security. Loewen found it difficult to post a bond of $625 million and filed a motion asking the trial court to stay enforcement of its judgment of $125 million (equal to 125 percent of the compensatory damages component). The trial court denied the motion to relax the bonding requirement. On appeal, the Supreme Court of Mississippi dismissed the defendants’ motion for stay of execution on January 24, 1996. Loewen decided not to pursue further remedies (application for certiorari) before the U.S. Supreme Court against that decision. Instead, Loewen agreed to settle the case for $175 million.17 In 1998, the Loewen Group Inc. and Raymond Loewen initiated investment arbitration proceedings against the United States under Chapter 11 of NAFTA. Each claimant filed two claims: Loewen Group Inc. filed claims on its own behalf and on behalf of its subsidiary, Loewen 15

Id. at ¶¶ 83–87.

16

Id. at ¶ 173.

17

Id. at ¶¶ 172–204.


Group International Inc. Raymond Loewen pursued two concurrent claims strategies. On the one hand, he filed a claim on behalf of the Loewen Group Inc. as its controlling shareholder. On the other hand, he filed a claim in his own name as a foreign investor. Meanwhile, the Loewen Group had experienced serious financial difficulties. In June 1999, Loewen filed for bankruptcy protection in the United States and insolvency protection in Canada. In the course of bankruptcy reorganization, Loewen underwent substantive restructuring: the Loewen Group Inc. continued to exist as a corporate entity under the laws of British Columbia; yet all substantial assets, with the exception of the NAFTA claim, were transferred to a newly founded company incorporated in the United States, the Alderwoods Group. The bankruptcy reorganization was concluded in January 2002.18 B.

The Tribunal’s Decisions

In a decision on jurisdiction of June 2, 2000, the tribunal disposed of one jurisdictional objection raised by the respondent. The respondent had argued that judicial acts in litigation between private parties do not constitute “measures” pursuant to Article 1101 of NAFTA. The tribunal rejected the respondent’s restrictive reading of NAFTA. It pointed out that “[t]he text, context and purpose of Chapter Eleven combine to support a liberal rather than a restricted interpretation”19 of the word “measure.” Under NAFTA, a host state is under the obligation to extend protection to foreign investors from private parties if they act through judicial organs of the host state. In a lengthy final award of June 26, 2003, which to a great part consists of obiter dicta, the tribunal then dismissed all claims in their entirety. With respect to the Loewen Group’s claims, the tribunal declined jurisdiction on the ground that the collection of any damages would now ulti18 The tribunal, in paragraph 220 of the Final Award, mistakenly states that Nafcanco was assigned the NAFTA claim. As the claimants and the respondent concurrently state in their submissions, the Loewen Group Inc. continued to “own” the claim. Nafcanco was merely involved in administering the claim and collecting the potential damages. See U.S. Memorial on Matters of Jurisdiction and Competence Arising From the Restructuring of the Loewen Group, Inc., at 4–6 (Mar. 1, 2002); Countermemorial of the Loewen Group, Inc. on Matters of Jurisdiction and Competence, at ¶¶ 3–6 (Mar. 29, 2002) (accepting the U.S. factual account), all available at http://www.naftaclaims.com. 19



mately benefit an American corporation. Hence, the claimant was no longer a foreign national as required by NAFTA. With respect to Raymond Loewen’s claim on behalf of the Loewen Group, the tribunal held that Raymond Loewen failed to demonstrate direct or indirect ownership or control of the Loewen Group, as required under Article 1117, at the time when the claim was submitted to arbitration. The tribunal failed to specifically address Raymond Loewen’s claim in his own name under Article 1116. In the third paragraph of the dispositif, all claims are summarily “dismissed in their entirety.”20 Following an unusual request for clarification by the respondent, the United States, the tribunal issued a third, supplementary decision on September 6, 2004. In that decision, the tribunal clarified that Loewen’s Article 1116 claim was equally dismissed. Since no jurisdictional objection had been raised with respect to Loewen’s personal claim, this claim was dismissed on the merits. The tribunal noted that the reasons for the dismissal could be inferred from paragraphs 213 to 216 of the final award. In these paragraphs, the tribunal found that, as a matter of substantive law, judicial measures cannot be challenged in international arbitration unless the claimant has pursued reasonably available and adequate local remedies. In the tribunal’s view, the Loewen Group Inc.’s decision not to apply for certiorari to the U.S. Supreme Court, but to instead settle the case, precluded a violation of NAFTA. The bulk of the reasoning in the final award, however, is concerned with a different question. In an obiter dictum comprising almost a hundred paragraphs, the tribunal provided its reasons as to why the judgment of the Mississippi trial court constituted a miscarriage of justice. The judgment would have amounted to a “manifest injustice,”21 satisfying the conditions of the customary international law of denial of justice were it not for Loewen’s failure to exhaust all available remedies in the domestic U.S courts. III. COGENCY OF THE DECISIONS ON MATTERS OF JURISDICTION A.

The Mississippi Trial Constitutes a “Measure”

In its decision on jurisdiction, the tribunal made a determination on a minor objection to jurisdiction. Respondent argued that judicial acts in 20 In the following, the “orders” inserted in the Final Award between paragraphs 240 and 241 will be referred to as the “dispositif.” 21



litigation between private parties did not constitute “measures” within the meaning of NAFTA. The tribunal rejected the proposed narrow reading of the term measures based on a convincing systematic (contextual) interpretation of NAFTA Article 1101(1). That article delimitates the scope of Chapter 11 to “measures adopted or maintained by a party.” The tribunal correctly relates Article 1101 to the general provision of Article 201, which defines the terms measures authoritatively for NAFTA. After concluding that judicial acts are prima facie covered by the definition, the tribunal confirms its finding by reference to other NAFTA provisions containing the term “measures.”22 Thus far, the reasoning is impeccable. Less convincingly, the tribunal then directs the reader to instances outside the NAFTA context in which international courts and tribunals have interpreted the term “measure” as including judicial acts. This move is doubtful as a matter of treaty interpretation. It is unclear why, for example, the judgment in Regina v. Pierre Bouchereau by the European Court of Justice (ECJ) provides a legitimate source of authority for interpreting terms of art contained in the NAFTA.23 Given the highly specific, often technical character of European Community (EC) law, the ECJ’s interpretation of “measure” can hardly be said to give expression to “the ordinary meaning” of that term as required by Article 31(1) of the Vienna Convention on the Law of Treaties. On the other hand, to be relevant as “context” according to Article 31(2) of the Vienna Convention,24 a provision must be either part of the same legal instrument or adopted in connection with the conclusion of the legal instrument to be interpreted. Neither of these conditions is satisfied with respect to the relationship between the EC Treaty and NAFTA. The reference to rulings outside the NAFTA context would seem to detract from, rather than reinforce, the soundness of the tribunal’s reasoning. 22 Specifically, NAFTA arts. 1019, 1716, and 1701. See Decision on Jurisdiction, supra note 5, at ¶¶ 39–44. 23

Decision on Jurisdiction, supra note 5, at ¶ 45.

Article 31(2) of the Vienna Convention on the Law of Treaties, 1155 U.N.T.S. 331 (May 23, 1969) provides: 24

The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes: (a) any agreement relating to the treaty which was made between all the parties in connection with the conclusion of the treaty; (b) any instrument which was made by one or more parties in connection with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty.


B.

Consequences of Loewen’s Bankruptcy-Induced Reorganization

In the final award, the tribunal—now in a modified composition— upheld two separate objections to jurisdiction raised by the United States. With respect to the first claimant, the Loewen Group Inc., the tribunal declined jurisdiction since, “in consequence of [the Loewen Group International’s] assignment of those claims to a Canadian corporation owned and controlled by a United States corporation,”25 Loewen’s claims ceased to exist. The tribunal’s decision expressly disposed of both claims raised by Loewen, the Group’s own claim pursuant to Article 1116 and the claim on behalf of its subsidiary, Loewen Group International Inc., pursuant to Article 1117. With respect to the second claimant, Raymond Loewen, the tribunal declined jurisdiction over the claim brought on behalf of the Loewen Group, because “it was not shown that [Raymond Loewen] owned or controlled directly or indirectly [the Loewen Group International] when the claims were submitted to arbitration or after [the Group] was reorganized under Chapter 11 of the United States Bankruptcy Code.”26 The latter objection relates to specific features of U.S. bankruptcy law. It turns on a question of evidence that cannot be verified in the context of this chapter. As the tribunal explains, Raymond Loewen failed to adduce evidence that he continued to hold stock in the Loewen Group International at the relevant point in time, although the respondent had specifically alleged that Loewen no longer had control over his stock.27 More significant is the tribunal’s treatment of the first U.S. objection to jurisdiction—the contention that a change of nationality in the course of bankruptcy proceedings inevitably leads to a loss of NAFTA claims.28 Analytically, three questions should be distinguished (which the tribunal’s analysis tends to blend together). First, do changes of nationality subsequent to the initiation of NAFTA arbitration impact the position of investors? Second, if the answer is in the affirmative, does Loewen qualify as a “continuing national,” although it has transferred all substantial assets—with the exception of the NAFTA claims—to a U.S. entity? Third, is the United States estopped from invoking continuous nationality since the change of nationality was caused by its own allegedly unlawful conduct?

25


26

Id.

27

Id. at ¶ 239.

28

Id. at ¶ 232.


According to the requirement of continuous nationality (which most commentators have held to apply in the context of NAFTA29), the claimant must be a national of another NAFTA contracting party from the time of the unlawful act (dies a quo) until a critical dies ad quem. Claimants contended that under NAFTA, and similar to the International Center for the Settlement of Investment Disputes Washington Convention,30 the critical date was the day of the presentation of the claim. Since Loewen’s bankruptcy proceedings were only instituted subsequent to that date, the claimants argued that the requirement of continuous nationality was met. The tribunal, however, did not agree. In its view, “there must be continuous national identity from the date of the events giving rise to the claim, which is known as the dies a quo, through the date of the resolution of the claim, which date is known as the dies ad quem.”31 Since, at the time of the award, a Canadian entity no longer existed, Loewen’s “vested claim, already subject of valid proceedings, simply ceases to exist.”32 Thus far, the tribunal’s conclusion appears to be supported by coherent reasoning. The tribunal correctly notes that the text of NAFTA does not specify whether the dies ad quem should be the date of the presentation or the resolution of the claim. Hence, two divergent interpretations could be envisioned, the tribunal’s argument goes: On the one hand, a dynamic interpretation in light of NAFTA’s object and purpose would favor the date of the presentation of the claim. Under many bilateral investment treaties (BITs), as well as in Article 25 of the ICSID Washington Convention, the day of the presentation of the claim is considered the “critical date” so as to avoid burdening the investor with uncertainties arising subsequent to the initiation of arbitration. On the other hand, a more conservative reading of NAFTA would emphasize that, in the law of diplomatic protection, continuous nationality is required until the resolution of the dispute. Unless a treaty instrument contains a clear indication to the contrary, this “traditional” rule of continuous

29 See the detailed discussion in M. Mendelson, The Continuous Nationality Rule, in WEILER, supra note 7, at ch. 4. For a critical view, see F. ORREGO VICUÑA, INTERNATIONAL DISPUTE SETTLEMENT IN AN EVOLVING GLOBAL SOCIETY: CONSTITUTIONALIZATION, ACCESSIBILITY, PRIVATIZATION 37 (2004). 30 See Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (Washington Convention), 575 U.N.T.S. 159, 17 U.S.T. 1270, art. 25 (registration of the claim). 31


32

Id. at ¶ 232.


nationality must be deemed to apply even in a foreign-investment context: In dubio mitius. The tribunal opts for the conservative reading. Yet the seemingly coherent reasoning has a major flaw: The content of the traditional rule of continuous nationality in the area of diplomatic protection is not half as clear as the tribunal suggests. As if it were referring to a generally accepted fact, the tribunal explains that, “any change in nationality of the claimant,” up to the resolution of the dispute, “defeated the only reason for the negotiations to continue. The claiming government no longer had a citizen to protect.”33 Contrast the tribunal’s reasoning with what is arguably the most authoritative restatement of the contemporary law on diplomatic protection, the draft articles on diplomatic protection adopted by the U.N. International Law Commission (ILC)34 in 2006. The ILC’s codification plainly contradicts the tribunal’s interpretation of the continuous nationality rule. Article 5(1) reads: A State is entitled to exercise diplomatic protection in respect of a person who was a national of that State continuously from the date of injury to the date of the official presentation of the claim. Continuity is presumed if that nationality existed at both these dates.35 In the commentary, authored under the intellectual leadership of Professor John Dugard, the ILC explains that “[t]he phrase ‘presentation of the claim’ is that most frequently used in treaties, judicial decisions and doctrine to indicate the outer date or dies ad quem required for the exercise of diplomatic protection.”36 Its rationale is to protect the individual claimant from hardship, since “many years may pass between the presen33

Id. at ¶ 229.

According to its Statute, the ILC has 34 members from different states of the world, all of who “shall be persons of recognized competence in international law.” Adopted by the General Assembly in Resolution 174(II) (Nov. 21, 1947), as amended by Resolutions 485(V) (Dec. 12, 1950), 984(X) (Dec. 3, 1955), 985(X) (Dec. 3, 1955), and 36/39 (Nov. 18, 1981). 34

35 Draft Articles on Diplomatic Protection with commentaries, International Law Commission, Report on the work of its fifty-eighth session, A/61/10, at 17 (May 1 to June 9 and July 3–11, 2006), (emphasis added). Enumerating the cornerstone principles of diplomatic protection in the introduction to the report, the ILC makes reference to “the rule of continuous nationality which requires a State to prove that the injured national remained its national after the injury itself and up to the date of the presentation of the claim.” Id. at 25. 36

Id. at 37.


tation of the claim and its final resolution and it could be unfair to penalize the individual for changing nationality . . . during this period.”37 While the text adopted in 2006 was not yet available to the tribunal, John Dugard’s First Report of March 2000 clearly foreshadowed the stance that the ILC would take. Dugard consistently presented the “presentation of the claim” as the strictest conceivable standard, thus excluding the date of the resolution of the dispute as the relevant dies ad quem.38 The tribunal was more than familiar with this scholarly controversy. A brief look at the submissions shows that the proper interpretation of the law of diplomatic protection was a matter of intense contestation during the proceedings. In their countermemorials, the claimants presented extensive evidence of state practice suggesting that the presentation of the claim constitutes the relevant dies ad quem in the area of diplomatic protection. According to sources quoted by claimants, the British government maintained that a person must be “a national of the State which is presenting the claim both at the time when the injury occurred and continuously thereafter up to the date of formal presentation of the claim.”39 Similarly, according to a communication by the U.S. Department of State, the United States considered itself entitled to espouse claims “on behalf of a national . . . who had that status at the time the claim arose and continuously thereafter to the date of the presentation of the claim.”40 Backed by expert opinions and reference to several renowned international law scholars, the claimants argued at length that the contemporary law on diplomatic protection recognized the “date of presentation rule” in lieu of the “date of award rule.” The claimants were entitled to an argumentative engagement with their submissions. Such engagement would have involved confronting their evidence concerning the relevant dies ad quem and explaining why it is not to the point, or why it is normatively unconvincing. Stating instead that “[t]here is only limited dispute as to the history of the 37

Id. at 38.

38 John R. Dugard, First report on diplomatic protection, A/CN.4/506 and ADD.1 (Mar. 7, 2000) 39 Investor’s Countermemorial on Jurisdiction Concerning Loewen Corporate Restructuring (Raymond Loewen), at ¶ 53 (Mar. 19, 2002), available at http://www. naftaclaims.com. 40 Investor’s Countermemorial on Jurisdiction Concerning Loewen Corporate Restructuring (Loewen Corp.), at ¶ 86 (Mar. 29, 2002), available at http://www.naftaclaims.com.


requirement of continuous nationality to the end of any international proceeding,”41 the tribunal effectively disregards 14 paragraphs of legal argument to the contrary in Loewen’s countermemorial.42 The apparent coherence of the tribunal’s reasoning on continuous nationality comes at the price of concealing a relevant scholarly controversy altogether. C.

Continuous Nationality Subsequent to a Transfer of Substantial Assets to a U.S. Company?

In the course of bankruptcy reorganization, Loewen’s substantial assets were transferred to a U.S. entity. To ensure an ongoing diversity of nationality between claimant and respondent (as required under NAFTA Chapter 11), the Loewen Group Inc. continued to exist as an entity under the laws of British Columbia and continued to own the Group’s NAFTA claims. In addition, another Canadian company, Nafcanco, was established to direct the prosecution of those claims. During the ICSID arbitration, the United States disputed that either the “remainder” of Loewen or Nafcanco qualified as “continuing nationals” for purposes of the ICSID arbitration.43 The tribunal agreed with the U.S. objection. A “naked entity,” even if it were properly incorporated in Canada, could not be a claimant under NAFTA. It is worth quoting the tribunal’s argumentation on the relevant point at length: All of the assets and business of [Loewen] have been reorganized under the mantle of an American corporation. All of the benefits of any award would clearly inure to the American corporation. Such a naked entity as Nafcanco, even with its catchy name, cannot qualify as a continuing national for the purposes of this proceeding. Claimants also urge that [Loewen] remains in existence, since its charter remains in existence. The Tribunal is being asked to look at form rather than substance to resolve a complicated claim under an international treaty. Even if 41


42 The Loewen Corporation’s argument on this point extends over 14 paragraphs, Raymond Loewen’s argument extends over ten paragraphs. Both countermemorials contain numerous citations that appear to support claimants’ reading of the continuous nationality rule. 43 U.S. Memorial on Jurisdiction Concerning Loewen Corporate Restructuring, at 13–34 (Mar. 1, 2002).


[Loewen] has some kind of ethereal existence, it sought to place any remaining NAFTA marbles in the Nafcanco ring. Claimants insist that Respondent is asking the Tribunal to “pierce” the corporate veil of Nafcanco and point out the legal complications involved in such a piercing. The Tribunal sees no need to enter into that thicket. The question is whether there is any remaining Canadian entity capable of pursuing the NAFTA claim.44 The above paragraph constitutes the entirety of the tribunal’s reasoning. The tribunal certainly gets to the heart of the matter when it wonders whether it should “look at form rather than substance.” It implicitly answers the question in the negative. All benefits of a damages award would “inure to the American corporation,” and there is no “remaining Canadian entity capable of pursuing the NAFTA claim.” It may be inferred that, in the tribunal’s view, two conditions must be met to properly be a claimant under NAFTA. First, the ultimate beneficiary of an award must be a foreign national; and, second, the entity that brings the claim must have an independent capacity to act. The tribunal fails to indicate any reason for its exclusive preference for substance over form. It certainly cannot claim that the obviousness of its conclusions would make reasons dispensable. Contrary to the tribunal’s theory, the prevailing view in foreign investment arbitration regards as decisive the formal criterion of incorporation. In the case of Tokios Tokeles v. Ukraine,45 the claimant was a Lithuanian company; yet 99 percent of the shares were held by Ukrainian nationals. The Ukraine requested that the tribunal pierce the corporate veil to determine the nationality of the claimant in accordance with that of its predominant shareholders or the seat of management. In its award on jurisdiction, the Tokios Tokeles tribunal demonstrated an exemplary awareness of the need for a well-reasoned choice between the formal criterion of incorporation and the substantive criteria of ultimate ownership and management. After 50 paragraphs of detailed discussion, the tribunal found that the formal criterion of incorporation better corresponded to the wording of the Lithuanian-Ukrainian BIT. Moreover, a preference for form over substance was considered normatively justified, since it

44


Tokios Tokelés v. Ukraine, ICSID Case No. ARB/02/18, Decision on Jurisdiction (Apr. 29, 2004). 45


fulfills the parties’ expectations, increases the predictability of dispute settlement procedures, and enables investors to structure their investments to enjoy the legal protections afforded under the Treaty. We decline to look beyond (or through) the Claimant to its shareholders or other juridical entities that may have an interest in the claim.46 As demonstrated by the dissenting opinion of Prosper Weil, president of the Tokios Tokeles tribunal,47 the opposite conclusion—a preference for substance over form, for the siège social over the place of incorporation—can also be supported by strong and coherent reasons. In his dissent, Weil argued that the object and purpose of BITs and the ICSID Washington Convention was to encourage the influx of foreign capital. This purpose was defeated, in his view, when a foreign company merely served as a shell through which domestic investors channeled their capital. Similarly, in its draft articles on diplomatic protection, the ILC has indicated a willingness to adopt a substantive test of nationality— albeit only in exceptional circumstances.48 The comparison with the extensive reasoning on the same point in the Tokios Tokeles award strikingly demonstrates the shortcomings of Loewen. As the Tokios Tokeles tribunal aptly held, [a]lthough the Loewen tribunal denied that it pierced the claimant’s corporate veil, in reality, the tribunal did exactly that. Indeed, the tribunal could not have concluded that the nation46

Id., at ¶ 40.

Tokios Tokelés v. Ukraine, ICSID Case No. ARB/02/18, Dissenting Opinion (Apr. 29, 2004). 47

48 According to Article 9, entitled “State of nationality of a corporation,” reference must be had to the place of incorporation. However, if a company has no substantial business activity in that state, the nationality of the corporation may be determined by substantive criteria. The article reads:

For the purposes of the diplomatic protection of a corporation, the State of nationality means the State under whose law the corporation was incorporated. However, when the corporation is controlled by nationals of another State or States and has no substantial business activities in the State of incorporation, and the seat of management and the financial control of the corporation are both located in another State, that State shall be regarded as the State of nationality. Draft Articles on Diplomatic Protection with commentaries, supra note 35, at 52.


ality of the claimant had changed from Canadian to U.S. origin without piercing the claimant’s corporate veil. Although one may debate whether veil piercing was justified in that case, the Loewen decision does not clarify the jurisprudence of veil piercing because the tribunal did not admit to, much less explain its reasons for, piercing the claimant’s corporate veil.”49 The Loewen case turned on the proper methodology for determining corporate nationality just as much as the Tokios Tokeles case. Yet the Loewen tribunal first denied the relevance of the controversy of form versus substance and then adopted a substantive test without providing reasons for its preference. Prosper Weil’s dissent in Tokios Tokeles can serve as a blueprint of what the tribunal could have accomplished in arguing for a substantive test of nationality. D.

Is the United States Estopped from Invoking Continuous Nationality?

Finally, the tribunal’s treatment of the continuous nationality requirement leaves a key legal question open: are tribunals entitled to apply the continuous nationality rule even to cases in which nationality has been lost as a consequence of the alleged wrongful act? One of the claimants’ central arguments50 was that the Loewen Group was driven into bankruptcy as a consequence of the United States’ unlawful conduct. Loewen specifically urged the tribunal that it must be recognized that [Loewen]’s loss of its investment was largely a result of the NAFTA violations now complained of and may therefore be considered an indirect expropriation; at a minimum, it will add to Loewen’s Article 1116 damages claim.51 It is certainly true that the claimants’ argument is based on the assertion of a potentially problematic chain of causation: the bankruptcy-related change of nationality was forced upon the company due to an excessively high settlement, which, in turn, was due to a denial of justice by a U.S. court. Yet, it was the task of the tribunal to consider whether, as a factual

49

Tokios Tokelés v. Ukraine, supra note 45, at ¶ 65.

In fairness to the tribunal, it should be noted that claimants, too, did not emphasize this point as strongly as they could have. 50

51 Countermemorial of the Loewen Group, Inc. on Matters of Jurisdiction and Competence, at ¶ 138 (Mar. 29, 2002), available at http://www.naftaclaims.com.


matter, the denial of justice really constituted the root cause of Loewen’s bankruptcy and whether, as a legal matter, the United States was not consequently estopped from relying on the change of nationality brought about by its unlawful conduct. The tribunal failed to address both the factual and legal questions raised, although it was manifestly aware of the possibility that the bankruptcy-induced change of nationality was related to the financial strain of a $150 million settlement. In lieu of a reasoned explanation, the tribunal invokes a dubious distinction between domestic “chancery courts” and international tribunals: [Loewen] urges some equitable consideration be given because it was the underlying Mississippi litigation which brought about the need for it to file bankruptcy in the first place. We have already rehearsed our view of the inequities that befell [Loewen] in that litigation, and a chancery court would certainly take such claims into account in assessing damages. But this is an international tribunal whose jurisdiction stems from and is limited to the words of the NAFTA treaty.52 Relieved of the burden to further inquire into the circumstances of the bankruptcy, the tribunal concluded that “[w]hatever the reasons for [Loewen]’s decision to follow the bankruptcy route it chose, the consequences broke the chain of nationality that the Treaty requires.”53 The tribunal’s point of departure is of course correct: A claim under NAFTA requires a diversity of nationality between the claimant and the respondent on the dies ad quem. Yet the tribunal’s subsequent reasoning implies that, other than domestic legal orders, international law provides no tools for remedying inequities caused by illicit conduct on the part of the respondent. A brief look at blue-chip textbooks, decisions by the International Court of Justice (ICJ), or arbitral awards should have cast doubts on the soundness of the tribunal’s distinction between domestic proceedings and international adjudication. Judge Reed of the ICJ affirmed that “in any proceedings which recognized the principles of justice,” no government would be allowed to raise an objection that would

52


53

Id.


“let such a government profit from its own wrong.”54 In the ICJ’s Temple of Preah Vihear case, Vice President Alfaro provided a useful summary of maxims applicable in international law.55 In particular, Alfaro pointed to the maxim nullus commodum capere de sua injuria propria pursuant to which a “State must not be allowed to benefit by its inconsistency when it is through its own wrong or illegal act that the other party has been deprived of its right or prevented from exercising it.”56 This maxim is also firmly anchored in the practice of international arbitral tribunals. Classically, in the Betsey case, Commissioner Pinkney called it “the most exceptionable of all principles, that he who does wrong shall be at liberty to plead his own illegal conduct on other occasions as a partial excuse.”57 The validity of the maxim was affirmed, inter alia, by the umpire in the Montijo case,58 the Permanent Court of International Justice in the Chorzów Factory case,59 and the Iran-U.S. Claims Tribunal.60 Not surpris54 Interpretation of Peace Treaties (2d Phase), Diss. op. Judge Reed, 1950 I.C.J. 221, at 244. 55

According to Alfaro, a State must not be permitted to benefit by its own inconsistency to the prejudice of another State (nemo potest mutare consilium suum in alterius injuriam). A fortiori, the State must not be allowed to benefit by its inconsistency when it is through its own wrong or illegal act that the other party has been deprived of its right or prevented from exercising it. (Nullus commodum capere de sua injuria propria.) Finally, the legal effect of the principle is always the same: the party which by its recognition, its representation, its declaration, its conduct or its silence has maintained an attitude manifestly contrary to the right it is claiming before an international tribunal is precluded from claiming that right (venire contra factum proprium non valet).

Case concerning the Temple of Preah Vihear (Cambodia v. Thailand), Sep. op. Alfaro, 1962 I.C.J. 39, at 40. 56

Id.

4 J.B. MOORE, INTERNATIONAL ADJUDICATION ANCIENT AND MODERN: HISTORY AND DOCUMENTS, 179, at 277 (1931). 57

“No one can be allowed to take advantage of his own wrong,” 2 J.B. MOORE, HISDIGEST OF THE INTERNATIONAL ARBITRATIONS TO WHICH THE UNITED STATES HAS BEEN A PARTY 1421 at 1437 (1898). 58

TORY AND

59

Chorzów Factory, 1927 P.C.I.J. (ser. A) no. 9, 31.

60 “It is a well recognized principle in many municipal systems and in international law that no one should be allowed to reap advantages from their own wrong, Nullus Commodum Capere De Sua Injuria Propria.” Harza Engineering Company v. The Islamic Republic of Iran, 1 IRAN-U.S. CL. TRIB. REP. 499 (1982). See also Tippetts, Abbett, McCarty, Stratton v. TAMS-AFFA, 6 IRAN-U.S. CL. TRIB. REP. 219, at 228 (1984).


ingly, leading treatises in international law have counted the maxim among the general principles of international law within the meaning of Article 38 of the ICJ Statute.61 Hence, the formal distinction between “chancery courts” and international tribunals does not hold up. On its face, the Loewen case would suggest an application of nullus commodum capere de sua injuria propria— provided, of course, that the claimants’ factual allegations prove right. There may have been good reasons not to apply the maxim; yet the award offers no reasoned explanation. The tribunal’s failure to discuss the legal consequences of the claimants’ contention that the bankruptcy was related to the alleged mistrial constitutes a serious omission at a critical point of the award. IV. COGENCY OF THE DECISION ON THE MERITS A.

The “Forgotten Claim”: Raymond Loewen’s Article 1116 Claim

The decision on the merits fares no better than the jurisdictional decisions. The most stunning omission of the final award is the tribunal’s apparent failure to expressly consider and make a determination on Raymond Loewen’s personal claim pursuant to Article 1116. As the tribunal explains in its supplementary decision, this claim was impliedly dismissed on the merits in the final award. The contention that the final award was to dispose of Raymond Loewen’s personal claim on the merits appears doubtful for several reasons. First, the final award is ostensibly an award limited to questions of jurisdiction. The introductory paragraph of the award suggests that all of claimants’ NAFTA claims were dismissed on jurisdictional grounds: Ultimately [the case] turns on a question of jurisdiction . . . This question was raised by Respondent’s motion to dismiss for lack of jurisdiction filed after the oral hearing on the merits. In this Award we uphold the motion and dismiss Claimants’ NAFTA claims.62

61 See, e.g., A. VERDROSS & B. SIMMA, UNIVERSELLES VÖLKERRECHT: THEORIE UND PRAXIS § 614 (3d ed. 1983); G. Fitzmaurice, The General Principles of International Law, Considered from the Standpoint of the Rule of Law, 92 R.D.I.C. 117 (1957), both with further references. 62

Final Award, supra note 5, at ¶ 1 (emphasis added).


In introductory paragraph 2 of the award, the tribunal makes it quite unequivocal that any discussion on the merits contained in the award is obiter dictum, included only “[a]s our consideration of the merits of the case was well advanced when Respondent filed this motion to dismiss”63 for lack of jurisdiction. Second, the dispositif does not contain any express reference to Raymond Loewen’s personal claim. This is peculiar since the tribunal otherwise provides, in one sentence, the key rationale for each claim’s dismissal. In the paragraph containing the order with respect to Raymond Loewen’s claim on behalf of the Loewen Group Inc., for example, the tribunal decides [t]hat it lacks jurisdiction to determine Raymond L. Loewen’s claims under NAFTA concerning decisions of the United States courts on the ground that it was not shown that he owned or controlled directly or indirectly [the Loewen Group Inc.] when the claims were submitted to arbitration or after [Loewen] was reorganized under Chapter 11 of the United States Bankruptcy Code.64 No corresponding determination was reached in the dispositif with respect to Raymond Loewen’s personal claim. The suspicion that the tribunal has in fact “forgotten” to reach a decision on that claim is further nurtured by the tribunal’s failure to mention this claim at various other critical paragraphs in the award.65 Paragraph 9 is revealing: the tribunal notes that the Loewen Group Inc. has submitted two concurrent claims, namely pursuant to Articles 1116 and 1117. It then goes on to say that Raymond Loewen has filed “claims as ‘the investor of a party’ on behalf of TLGI under NAFTA, Article 1117,” thus completely omitting the concurrent personal claim pursuant to Article 1116.

63

Id. at ¶ 2.

64

Id., dispositif ( “orders”).

This observation is set out in detail in Raymond Loewen’s application for an annulment of the award at the D.C. District Court, Raymond L. Loewen v. United States, Notice of Petition to Vacate, Case No. 1:04CV02151, 12/13/2004, 7 et seq., available at http://www.naftaclaims.com. 65


On its face, the final award was meant to be limited to jurisdictional questions. Nonetheless, in its supplementary decision following a request for clarification by the respondent, the tribunal denied its omission. The tribunal argued that Raymond Loewen’s personal claim, while not expressly mentioned, was covered by its holding that “[The Loewen Group Inc.]’s claims and Raymond L. Loewen’s are hereby dismissed in their entirety.”66 The rationale for this dismissal, according to the tribunal, can be inferred from paragraphs 213 to 216, which deal with Loewen’s decision to settle the case instead of pursuing additional remedies. Problematically, however, this section does not contain any express reference to Mr. Loewen’s personal claim, although this claim is the only one that turns on the considerations laid out. A reader of the award is, thus, likely to read the section in light of the introductory paragraphs, which refer only to the Loewen Group’s claims and Raymond Loewen’s claim on behalf of the Group. In short, the reader is left with two alternative interpretations of the final award, both equally unsatisfactory. If the tribunal has in fact “forgotten” Loewen’s personal claim, then the final award is internally coherent but incomplete on a major point. Alternatively, the final award can be read, as the tribunal contends in the supplementary decision, as implying a determination on all claims, including Raymond Loewen’s personal claim. Then, however, the reasoning with respect to that claim is blatantly inadequate. The international community of interpreters cannot be expected to forage for potential findings on the merits in sections that, on their face, appear to constitute mere obiter dicta to a jurisdictional decision. B.

The Distinction Between Finality and Exhaustion of Local Remedies

If Raymond Loewen’s personal claim was in fact dismissed on the merits (as the tribunal contends), a closer examination of the reasoning supporting the dismissal is at order. In a nutshell, the tribunal argues that Loewen deprived itself of the possibility of pursuing a denial of justice claim at the international level by prematurely settling the case, that is, at a moment when judicial remedies were still available. Under NAFTA, whether judicial remedies must be exhausted is customarily considered a question of the interpretation of the Agreement’s waiver provision. Article 1121 requires investors to “waive their right to 66

Final Award, supra note 5, dispositif.


initiate or continue before any administrative tribunal or court under the law of any Party . . . any proceedings with respect to the measure of the disputing Party that is alleged to be a breach” of NAFTA. The question is whether this provision should be read as an implicit abrogation of the exhaustion of local remedies rule under customary international law.67 The tribunal does not address the exhaustion of local remedies under the customary heading of Article 1121. Instead, the tribunal argues that, irrespective of the interpretation of the waiver provision, original judicial wrongs must always be challenged in domestic courts as a matter of substantive law. In the tribunal’s view, it is a substantive “principle [of international law] that a court decision which can be challenged through the judicial process does not amount to a denial of justice.”68 The tribunal’s argument is based on a distinction between the exhaustion of local remedies rule and what could be called the “finality requirement.”69 The exhaustion of local remedies rule is defined as operating as a bar to jurisdiction. If it is not met, a claimant cannot “raise the complaint at the level of international law.”70 The finality requirement, on the other hand, operates as a special substantive bar to a finding of unlawfulness, applying exclusively to judicial acts. Finality is an “essential condition of a State being held responsible for a judicial decision.”71 The tribunal thus establishes a theory of two-tiered responsibility. As far as original administrative wrongs are concerned, the exhaustion of judicial remedies is only a question of jurisdiction. As far as original judicial wrongs are concerned, an unsuccessful attempt to remedy the wrong in domestic judicial proceedings is, in addition, a pre-condition for substantive responsibility. The validity of the distinction between the jurisdictional exhaustion of local remedies rule and the substantive finality requirement is questionable. Three analytic steps are not differentiated as clearly as they should be:

67

See infra notes 72 and 73 for an overview of the debate.

68


69 This term was coined by Bradford K. Gathright in his comment on Loewen, supra note 8. 70


71

Id. at ¶ 153.


1.

2.

3.

As a matter of substantive law, what kind of state conduct falls under the disciplines of NAFTA Article 1105? If the “minimum standard of treatment” is considered to embody the customary international law on denial of justice, can original wrongs by the judiciary constitute a violation of that standard? A second question concerns the interpretation of the law of state responsibility. Is the exhaustion of local remedies rule a procedural requirement or a substantive one? In other words, does international responsibility arise immediately when an alien is not treated in conformity with the substantive rules of international law; or is the exhaustion of local remedies a pre-condition of the unlawfulness of state conduct? The prevailing view treats the local remedies rules as a procedural requirement.72 Assuming that the exhaustion of local remedies rule is procedural in character, does the rule prevent the tribunal from exercising jurisdiction over Loewen’s claim? The crucial question at this level of analysis is whether Article 1121 of NAFTA implies a waiver of the requirement to pursue local remedies. The Feldman, Waste Management, and Metalclad tribunals73 and a number of scholars74 have answered this question in the affirmative.

72 For an overview of this classic debate, see VERDROSS & SIMMA, supra note 61, at § 1306. The ILC appears to side with the first view according to which the exhaustion of local remedies vel non does not affect the qualification of conduct as unlawful. Article 44 of the Articles on State Responsibility, entitled Admissibility of Claims, characterizes the local remedies rule as a bar for the invocation of state responsibility. The provision reads: “The responsibility of a State may not be invoked if: . . . The claim is one to which the rule of exhaustion of local remedies applies and any available and effective local remedy has not been exhausted.” International Law Commission, Report on the Work of Its Fifty-third Session, Official Records of the General Assembly, Fifty-Sixth Sess., Supp. No. 10 (A/56/10), at 58. 73 Feldman v. Mexico, ICSID Case No. ARB (AF)/99/1, ¶ 73, 42 I.L.M. 625, 639 (2003); Waste Management, Inc. v. Mexico, ICSID Case No. ARB(AF)/00/3, Preliminary Objection, at ¶ 30, 41 I.L.M. 1315, 1321 (2002); Metalclad Corp. v. Mexico, ICSID Case No. ARB (AF)/97/1, at ¶ 97 n.4, 40 I.L.M. 36, 49 n.4 (2001). 74 W.S. Dodge, National Courts and International Arbitration: Exhaustion of Remedies and Res Judicata Under Chapter Eleven of NAFTA, 23 HASTINGS INT’L & COMP. L. REV. 357, 373–76 (2000) Gathright, supra note 8, at 1134; Bjorklund, supra note 9; Wallace, supra note 7.


The tribunal attempts to sidestep the third question completely, and blends the first and second question through the judicial invention of the finality requirement. This reasoning is both unnecessarily complicated and problematic from a legal point of view. First, it is logically faulty for the tribunal to make a determination on the merits while leaving open the question of jurisdiction. By characterizing the finality requirement as distinct from the local remedies rule, the tribunal purports to avoid pronouncing itself on the question as to whether Article 1121 should be read as a waiver of the exhaustion of local remedies rule. Yet, if the rule did apply, it would bar the tribunal from exercising jurisdiction in the first place. The interpretation of Article 1121 logically precedes any merits decision. Second, the distinction between judicial finality and the exhaustion of local remedies is normatively unconvincing. Since the purpose of the tribunal’s finality requirement is identical to that of the classic exhaustion of local remedies, it is difficult to see why the finality requirement had to be introduced in the first place. In the tribunal’s view, the requirement of judicial finality must be deemed to apply because it would be very strange if a State were to be confronted with liability for a breach of international law committed by its magistrate or low-ranking judicial officer when domestic avenues of appeal are not pursued, let alone exhausted. . . . [I]t would encourage resort to NAFTA tribunals rather than resort to the appellate courts and review processes of the host State, an outcome which would seem surprising, having regard to the sophisticated legal systems of the NAFTA Parties. Such an outcome would have the effect of making a State potentially liable for NAFTA violations when domestic appeal or review, if pursued, might have avoided any liability on the part of the State.75 This justification of the tribunal’s finality requirement reads like a textbook rehearsal of the rationale of the exhaustion of local remedies rule. The local remedies rule “intends to give a state that violates international law the possibility to remedy the injustice with her own means, before she can be held accountable under international law,”76 as Alfred Verdross and Bruno Simma have noted. 75


76

VERDROSS & SIMMA, supra note 61, at § 1306: “Da die in Rede stehende Norm


Thus the tribunal’s attempt to adduce policy reasons for the finality requirement misses the point: no one will deny that, in the abstract, a requirement to exhaust local remedies may be supported by sound policy reasons. The crucial legal question is whether the NAFTA parties intended to dispense with the exhaustion of local remedies requirement. If, by virtue of Article 1121, investors are permitted to directly challenge executive measures at the international level, it is difficult to see why this should be any different for original judicial wrongs. Either way, the dispute is deliberately taken away from the “sophisticated legal systems of the NAFTA parties” in the interest of prompt and unbiased adjudication at the international level. Third, the tribunal’s reasoning is difficult to square with the modern doctrine of denial of justice. Under the tribunal’s theory, original judicial wrongs cannot trigger international responsibility unless they are confirmed by a last-instance court. As a consequence of the finality requirement, original judicial wrongs are less likely to be subject to NAFTA proceedings than executive wrongs. Such a distinction is reminiscent of the classic conception of denial of justice. Formerly, international claims were restricted to cases in which the judiciary proved unwilling or unable to adequately remedy an antecedent executive wrong.77 The question whether substantive responsibility for state conduct arises was conflated with the question whether state responsibility could be invoked. By contrast, the modern conception of denial of justice does not discriminate between judicial and executive conduct.78 The modern consensus is that both administrative action and lower-level court action may constitute a denial of justice in violation of international law.79 Whether this denial of dem Völkerrechtsverletzer die Möglichkeit geben will, mit eigenen Mitteln das Unrecht zu beheben, bevor er [völkerrechtlich] belangt werden kann.” “Because the rule in question intends to provide the violator of international law with the opportunity to remedy the wrong by her own means, before she can be held accountable under international law” (translation by the author). 77 C. Eagleton, Denial of Justice in International Law, 22 AM. J. INT’L L. 538, at 542 (1928). 78 See, for a good overview, Gathright, supra note 8, at 1109–14. There remains some dispute as to the extent to which “manifestly unjust judgments” (as opposed to procedural wrongs) are challengeable at the international level; see E.M. Borchard, The Law of Responsibility of States for Damage Done in Their Territory to the Person or Property of Foreigners, 23 AM. J. INT’L L., SUPP.: CODIFICATION OF INT’L L. 131, at 134 (1929). 79

See, for a concurring view, Bjorklund, supra note 9, at 858: From a policy perspective, however, it is difficult to distinguish the desirability of requiring a decision of the highest body within a court system


justice is immediately challengeable at the international level, by contrast, is customarily considered a question of the applicability of the local remedies rule. C.

Loewen Was Not Entitled to Seek Settlement, Since Reasonable Local Remedies Were Yet to Be Exhausted

Irrespective of whether the issue is framed in terms of the exhaustion of local remedies rule or in terms of judicial finality, the question remains what kind of judicial remedies need to be exhausted. This debate is an unresolved “classic” in the international law of state responsibility. In the Finnish Shipowners case, Britain contended that every available remedy had to be exhausted. Justice Bagge recognized that this would be asking too much indulgence of an alien subjected to unlawful conduct, holding that “it appears hard to lay on the private individual the burden of incurring loss of money and time by going through the courts, only to exhaust what to him—at least for the time being—must be only a very unsatisfactory remedy.”80 Three decades later, in the Ambatielos case, the arbitral commission applied a stricter standard: “The defendant State has the right to demand that full advantage shall have been taken of all local remedies.”81 Until the present day, the line between required and unreasonable remedies has remained blurred.82 from requiring a final decision from the highest official in an administrative system. Thus, if a lower-level official denies a request for a permit, why not require an applicant to appeal to the official’s superiors for a different decision, or to an administrative body with supervisory or appellate oversight? When the Loewen tribunal opined that only a final act of a judicial system may give rise to a NAFTA claim, the logical extension of that concept to requiring ‘appeals’ within different hierarchical structures may lead to claims by states party to NAFTA that eviscerate any waiver of the local remedies rule. 80

Finnish Shipowners case, 3 R.I.A.A. 1479, 1497 (1934).

Ambatielos (Greece v. UK), Award of the Commission of Arbitration, 23 I.L.R. 306, 334 (1956). 81

82 However, it may be expected that the 2006 codification of the draft articles on diplomatic protection will serve as a focal-point for future tribunals, upon which a jurisprudence constante can be built. Article 14(2) defines as local remedies “legal remedies which are open to an injured person before the judicial or administrative courts or bodies, whether ordinary or special, of the State alleged to be responsible for causing the injury.” Article 15 spells out the following exceptions. There is no obligation to exhaust remedies when

(a) There are no reasonably available local remedies to provide effective redress, or the local remedies provide no reasonable possibility of such redress; (b) There is undue delay in the remedial process which is attrib-


The Loewen ruling offers little in terms of policy clarification. In casu, the crucial question was whether Loewen prematurely agreed to a settlement with O’Keefe, even though exhaustible remedies were still available. The tribunal adopts Sohn and Baxter’s proposed benchmark of whether remedies are “reasonably available.”83 Loewen’s trial strategy failed to meet this standard: while the company’s decision to settle the case with O’Keefe may have been a reasonable business strategy,84 it was not clear that “the settlement agreement was the only course which Loewen could reasonably be expected to take.”85 Alternatively, Loewen could have filed an application for certiorari with the U.S. Supreme Court. Loewen failed to demonstrate why such an application would have been unpromising. The reasonable availability of remedies, the tribunal elaborates, must be determined “in the light of [the investor’s] situation, including its financial and economic circumstances as a foreign investor, as they are affected by any conditions relating to the exercise of any local remedy.”86 utable to the State alleged to be responsible; (c) There was no relevant connection between the injured person and the State alleged to be responsible at the date of injury; (d) The injured person is manifestly precluded from pursuing local remedies; or (e) The State alleged to be responsible has waived the requirement that local remedies be exhausted. Draft Articles on Diplomatic Protection with commentaries, supra note 35, at 20. 83 The genesis of this benchmark is curious: On the one hand, the tribunal appears to follow Professor Greenwood’s characterization of “the principle that a court decision which can be challenged through the judicial process does not amount to a denial of justice” (Final Award, supra note 5, at ¶ 151) as a principle of substantive law, distinct from the procedural requirement of exhaustion of local remedies. On the other hand, the tribunal borrows the test of reasonable availability precisely from a classic source dealing with exhaustion of local remedies: Sohn & Baxter, “Convention on the International Responsibility of States for Injuries to Aliens,” 12th Draft (1961) 168, commentary on Article 19, sub-paragraph 2(b)). A word of explanation as to why the content of the substantive requirement to pursue available remedies in a denial of justice context should be identical to the procedural local remedies rule would have been helpful. 84 See Final Award, supra note 5, at ¶ 213: “Entry into the settlement agreement no doubt reflected a business judgment by Loewen that, of the various options then open, settlement was the most attractive, in all probability because it provided certainty. Other alternatives involved financial consequences which would not have been easy to predict.” 85


86

Id. at ¶ 169.


This second-level definition adds little flesh to the vague standard of reasonable availability. The only substantial clarification is the alleged relevance of “economic circumstances.”87 Yet the tribunal never actually applies its own test in a coherent way. At no point in the decision are the “financial and economic circumstances” of Loewen “as they are affected by any conditions relating to the exercise of any local remedy” examined. Instead of rationalizing where to draw the line between required and unreasonable remedies, the tribunal attempts to sidestep such clarification through recourse to rules on the burden of proof. The tribunal finds that Loewen failed to present evidence disclosing its reasons for entering into the settlement agreement in preference to pursuing other options, in particular the Supreme Court option which it had under active consideration and preparation until the settlement agreement was reached.88 Since the Supreme Court option constituted, at least, a reasonable alternative, the tribunal concludes that “Loewen failed to pursue its domestic remedies.”89 Such an attempt to resolve a legal (as opposed to a factual) controversy at the level of burden of proof is logically faulty. The application of rules on burden of proof presumes that the legal benchmark is already known. A determination that the claimant has failed to present evidence on a particular point implies a theory as to

87 As a matter of substantive law, precisely this point is doubtful. Does the tribunal’s reasoning imply that investors in financial difficulty are subject to a more lenient requirement of exhausting available remedies than wealthy investors? As a consequence, a finding of a denial of justice would be more likely with respect to financially impotent investors than with respect to potent investors. The tribunal fails to present any argument as to why a host state should be burdened with economic difficulties of the investor when it comes to judicial remedies. Generally, theories of differential treatment of investors according to their financial strength are alien to investment law. 88 Final Award, supra note 5, at ¶ 215. The reasoning contained in this paragraph is contested by the claimants on factual grounds. In its Notice of Petition to Vacate (see supra note 65), counsel for Raymond Loewen argues that, contrary to the tribunal’s contention, the claimants presented ample evidence, including expert testimonies, as to why Loewen decided to settle the case. The tribunal had simply missed all this evidence. Since this point concerns the factual basis of the award rather than the cogency of its reasoning, this point is not treated in-depth in the present article. 89



which facts the claimant should have proven. Ostensibly leaving open the question of the precise meaning of reasonable availability, the tribunal in reality introduces a rather strict interpretation through the back door. In choosing between settlement and formal appeals, an investor has no discretion. Settlement is out of the picture unless it is “the only course which [the investor] could reasonably be expected to take.”90 The tribunal implies an absolute preference for judicial remedies over mutual settlements without explicating its rationale. D.

Who Determines the “Reasonable Availability” of Local Remedies?

As a consequence of its failure to clarify what constitutes a reasonably available remedy, the tribunal sidesteps the second, essential policy question. Who determines whether remedies are reasonably available? Three alternative solutions could be envisaged: According to what could be called the “objective approach,” a tribunal decides, on the basis of the evidence before it, whether remedies were likely available to the claimant. This approach would have required the Loewen tribunal to determine whether the expert testimony by Professor Days or Professor Tribe concerning the availability of review before the U.S. Supreme Court was more convincing. The obvious problem of this approach is that tribunals would arrogate themselves a superior competence on questions of municipal law, which, in most cases, they will not possess. To avoid playing the ultimate arbiter on questions of domestic law, tribunals are likely to acknowledge the pertinence of both expert witnesses and decide against the investor according to rules on the burden of proof. Despite these problems, this first, objective approach has been the prevailing one in international adjudication.91 Alternatively, a purely “subjective” test could be envisioned. According to this approach, it is sufficient for the investor to demonstrate that she, acting on the basis of professional legal advice, was convinced that no remedies were reasonably available. The Loewen tribunal would have to hear evidence from Loewen’s key decisionmakers and legal counsel. 90

Id. at ¶¶ 215 and 216.

91 See, for example, Judge Bagge’s emphasis on a “certain strictness” in applying the local remedies rule in the Finnish Shipowners case. According to Bagge, claimants were required to show that remedies were “obviously futile” in a particular legal system. (Finnish Shipowners case, supra note 80, at 1504) Bagge then went on to explore in great detail whether the Admiralty Board’s decision that formed the object of the dispute contained any appealable points of law.


As a third option, a tribunal could apply the objective test but shift the burden of proof to the host state. According to this approach, a tribunal must ascertain whether remedies were reasonably available. However, if expert witnesses disagree on that point, the investor is entitled to assume that this is not the case. The policy rationale of this latter approach is that, when legal experts are unable to predict whether remedies are available, a diligent businessperson must be entitled to discontinue costly legal proceedings and pursue a “safe” settlement option instead. The Loewen award fails to present a conscious and reasoned choice between these alternatives. The tribunal seems to waver between the first and the second approach. At the outset, the benchmark of reasonable availability is presented as an objective standard. The (brief and inconclusive) consideration of the expert testimonies of Professors Days and Tribe points in that same direction: both professors testify on the question as to whether litigants in situations similar to Loewen’s can generally—“objectively”—expect the Supreme Court to grant certiorari. On the other hand, the tribunal regards as “the central difficulty in Loewen’s case” the fact that “Loewen failed to present evidence disclosing its reasons for entering into the settlement agreement.”92 Such reasoning is at odds with an objective standard. If the objective standard is applied consistently, the subjective reasons that led Loewen to prefer settlement to judicial remedies are irrelevant. What counts is whether Loewen was objectively entitled to pursue the settlement option without losing its denial of justice claim. The tribunal’s reasoning involves a contradiction: although it enunciates an objective standard, the tribunal ultimately measures Loewen’s conduct against subjective criteria. V.

CONCLUSION

None of the preceding discussion is intended to suggest that the Loewen tribunal reached the wrong conclusion. Rather, the purpose was to demonstrate that, quite aside from the outcome, inadequate reasoning has added substantially to the widely felt dissatisfaction among the multiple constituencies of investment arbitration. Claimants are unlikely to feel that they had “their day in court,”93 because important aspects of 92

Final Award, supra note 5, at ¶ 215 (emphasis added).

93 See, on this aspect of “a reasoned opinion” to assure “individual litigants that their day in court was meaningful,” L.R. Helfer & A.-M. Slaughter, Toward a Theory of Effective Supranational Adjudication, 107 YALE L.J. 273, at 322 (1997).


their argument on jurisdiction and the merits were ignored by the tribunal. For the respondent, the award failed to put an end to the legal proceedings. The lacunae and contradictions in the tribunal’s reasoning made the award easily susceptible to a challenge (“petition to vacate”) before a D.C. District Court, in which the United States ultimately prevailed on grounds of a statutory limitation.94 More generally, governments and other investors fail to obtain guidance for future conduct. They will look in vain for a clear indication as to how the concepts of continuous nationality, the exhaustion of local remedies, or denial of justice should be interpreted in the NAFTA context. In its “final word,” the tribunal guards itself against the “natural instinct, when someone observes a miscarriage of justice, to step in and try to put it right.” Instead, the tribunal notes, it was tasked to rigorously “observe the principles that we have been appointed to apply, and stay our hands.”95 Fiat jus trumps fiat justitia, ruat cœlum. Let the law be done rather than justice though the heavens fall, one could say.96 It is peculiar to read the tribunal’s credo of moderation at the end of a lengthy award that combines passionate obiter dicta with scanty reasoning on the rationes decidendi. In half of the award, concerning the perceived denial of justice by the Mississippi court, the quest for justitia seems to be driving the tribunal; yet no findings are made. In the other half, the tribunal makes holdings on questions of jurisdiction and the merits; yet significant parts of its argumentation are incomplete, incoherent, and even contradictory. It is doubtful whether the Loewen award has, thus, done a great service to either jus or justitia.

94 On October 31, 2005, the motion was denied and the petition dismissed, because the district judge considered it to be time-barred under 9 U.S.C. Section 12. 95


Fiat justitia, ruat cœlum means let justice be done, though the heavens fall. The Latin maxim was prominently employed by Earl William Murray in the 1772 James Somerset case, which led to the abolition of slavery in England. J. Oldham, New Light on Mansfield and Slavery, 27 J. BRIT. STUD. 45 (1988). 96

CHAPTER 11

THE AWARD IN PETROBART LIMITED V. KYRGYZ REPUBLIC Galina Zukova*

I.

INTRODUCTION

The primary task in this chapter is to appraise whether the reasoning requirement was met by the award in Petrobart Limited v. Kyrgyz Republic arbitration,1 which was conducted under the Arbitration Institute of the Stockholm Chamber of Commerce (the SCC Institute). The dispute was brought and decided on the basis of the Energy Charter Treaty2 (the ECT or Treaty). Although there is no rule in the text of the ECT similar to that of Article 48 of the International Center for the Settlement of Investment Disputes (ICSID) Convention3 or Article 32 of the U.N. Commission on International Trade Law (UNCITRAL) Arbitration Rules,4 both of which set the basic principle of stating the reasons upon which the award is based, the principle as such is not foreign to the ECT. Thus, Annex D to the Treaty, which deals with interim provisions for trade dispute settlement, reads: “The final report shall deal with every substantial issue raised before the panel and necessary to the resolution of the

* Visiting Scholar, Yale Centre for International and Area Studies. University of Latvia. LL.B. University of Latvia, LL.M. Universtiy of Exeter, Ph.D. European University Institute. 1 Petrobart Ltd v. the Kyrgyz Republic, Arbitral Award, Arbitration No. 126/2003 (Mar. 29, 2005), rendered by the SCC Arbitration Institute, available at http://www. investmentclaims.com and 2005(3) STOCKHOLM INT’L ARB. REV. 45–100 [hereinafter Award]. 2

Energy Charter Treaty, available at http://www.encharter.org.

3 International Center for the Settlement of Investment Disputes (ICSID) Convention, available at http://icsid.worldbank.org/ICSID/Index.jsp. 4 UNCITRAL Arbitration Rules, available at http://www.uncitral.org/pdf/english/texts/arbitration/arb-rules/arb-rules.pdf. On this point, see also W. Michael Reisman & Guillermo Aguilar Alvarez, How Well Are Investment Awards Reasoned?, supra Chapter 1.

323


dispute and shall state the reasons for the panel’s conclusions.”5 Even if this provision, which strongly resembles the language of Article 48(3) of the ICSID Convention, is designed for the purposes of dealing with interim measures within the realm of trade disputes, there is no reason to suggest that lower standards would be applicable to awards in investment cases under the ECT. More important and pertinent in the present context are the 1999 Rules of the SCC Arbitration Institute, the arbitration forum in the Petrobart case. The rules impose an obligation on appointed arbitration tribunals to provide reasons for their award: “The Award . . . shall state the date on which it was rendered, contain an order or a declaration, as well as the reasons for it” (Article 32(1), emphasis added).6 Although this clause is not as specific as the relevant provisions of the ICSID Convention or the UNCITRAL Arbitration Rules, it imposes an unconditional obligation on established arbitral tribunals to state to the parties the reasons for their final determinations. It is in light of this provision (Article 32(1) of the SCC Arbitration Institute Rules) that the following discussion will proceed. Two important considerations regarding the Petrobart case are of major significance in the context of this study. First, so far, there have been only a few cases where the ECT was at stake. Only in two (out of around 12) cases was an award actually delivered where investors relied on the ECT in protection of their rights,7 namely, Nykomb v. Latvia8 and the Petrobart v. Kyrgyzstan case analyzed herein. In Plama v. Bulgaria9 the decision on jurisdiction came in 2005, with no award on the merits being pronounced so far.10 From this 5

Energy Charter Treaty, supra note 2, Point 4(a), ¶ 3 (emphasis added).

New SCC Arbitration Rules, which entered into force on January 1, 2007, provide, at Article 36(1), that the “Arbitral Tribunal . . . unless otherwise agreed by the parties, shall state the reasons upon which the award is based.” This new version more closely reflects the language of the ICSID Convention and of the UNCITRAL Arbitration Rules. 6

7 According to data provided by the Energy Charter Secretariat, see http://www. encharter.org/fileadmin/user_upload/document/Disputes_ECT_table_01.pdf. 8 Nykomb Synergetics Technology Holding AB v. Republic of Latvia, Arbitral Award, Arbitration No. 118/2001 (Dec. 16, 2003), rendered by the SCC Arbitration Institute [hereinafter Nykomb]. 9 Plama Consortium Ltd. v. Republic of Bulgaria, ICSID Case No. ARB/03/24, Decision on Jurisdiction (Feb. 8, 2005) [hereinafter Plama]. 10 The dispute in AES Summit Generation v. Republic of Hungary, ICSID Case No. ARB/01/04 was settled amicably between the parties. In Alstom Power Italia SpA

The Award in Petrobart Limited v. Kyrgyz Republic • 325

perspective, every new award represents a building block in the foundation of ECT jurisprudence. Second, the Petrobart dispute arose out of a purely commercial claim whose basis was the failure of a state enterprise to execute payment in conformity with the terms of a sale of goods contract with the putative investor, that is, Petrobart. The arbitral tribunal found, and the Sveå Court of Appeal confirmed in its decision,11 that the right to sell energy products and the right to payment for the commodity delivered formed an investment pursuant to the ECT, and, therefore, there was an investment dispute to be adjudicated. The tribunal radically departed from the “orthodox” notion of investment, which does not embrace single sales. Although it is undisputed that the wording of the ECT, as the tribunal itself admitted, presents “certain ambiguities,”12 thus providing fertile soil for possible conflicting interpretations, it is still highly questionable whether the ECT accommodates the broadest possible reading of the investment definition. Yet, the tribunal opted in favor of this latter approach. Did it venture too far? Are there good reasons in the text of the award to support the tribunal’s holding? How adequate were the reasons that the tribunal advanced? These questions form the core elements of the inquiry in this contribution. This chapter begins by offering factual information about the dispute and continues with an analysis of the award’s section on the existence of an investment. I conclude with an evaluation of the adequacy of the tribunal’s reasoning. This analysis reflects the structure of the award in which the same elements (the contract and the judgment by the national court, a claim for money, and the right to pursue an economic activity in the energy sector) are reviewed against their likelihood of being recognized as investments within the meaning of the ECT. It is and Alstom SpA v. Republic of Mongolia, ICSID Case No. ARB/04/10, the proceedings were discontinued at the parties’ request. A number of cases have been brought against the Russian Federation by Yukos shareholders under UNCITRAL arbitration (Group Menatep v. Russian Federation). In this context, it must be noted that in the framework of the Russian-Ukrainian dispute over gas supply, which arose at the end of 2005, the Ukrainian side referred to the possibility of bringing a claim against Russia to the SCC Institute based on the ECT. However, the dispute was settled between the parties without recourse to arbitration. 11 Judgment of January 19, 2007, of the Sveå Court of Appeal, Case No. T 520805, unofficial translation available at http://www.sccinstitute.com/_upload/shared_files/ newsletter/images/svea_court_petrobart.pdf. 12

Award, supra note 1, at 72.


argued that the arbitral tribunal delivered its final conclusion, that is, that the right to sell energy products amounts to an asset and thus an investment at the expense of providing supporting reasons. The chapter also offers a quick insight into the decision of the Sveå Court of Appeal, which confirmed the arbitral tribunal’s determination on the investment made. The chapter then proceeds with an analysis of the sales transaction in light of the universal characteristics of an investment—an issue that the tribunal left beyond its purview. This part of the chapter suggests that the commercial deal lacked most investment characteristics, whereas the award did not provide arguments that would categorically dispel the assumption that the ECT does not incorporate common sales deals into the investment definition. The final part of the chapter provides a set of conclusions. II.

FACTUAL BACKGROUND OF THE CASE

On February 23, 1998, Petrobart Limited (Petrobart), a company registered in Gibraltar, and the state joint stock company Kyrgyzgazmunaizat (KGM), charged with procurement and delivery of gas to Kyrgyz consumers, concluded a one-year goods supply contract (the Contract). According to the terms of the Contract, Petrobart undertook to supply stable gas condensate to KGM at a fixed price on a monthly basis. The dispute arose out of KGM’s failure to pay for three out of five executed deliveries for a total amount of $1.5 million. In November 1998, Petrobart initiated legal proceedings before the Bishkek City Court of Arbitration, which ruled in favor of Petrobart. Enforcement of the court decision was stayed after intervention by the Kyrgyz Vice-Premier. During this stay, the assets of the company were transferred for a nominal consideration to three state-owned companies, without transferring the company’s liabilities to its creditors. As a result of the company’s break-up, KGM was declared bankrupt. After unsuccessful attempts to obtain redress through the domestic courts, Petrobart initiated an investment arbitration based on the 1997 Kyrgyz Foreign Investment Law. However, both this UNICTRAL arbitration and the proceedings before the Bishkek Court of Arbitration (initiated by the Kyrgyz government) found that Petrobart had made no investment within the meaning of the above-mentioned law. Petrobart’s Request for Arbitration on the basis of the ECT was lodged at the SCC Institute on September 1, 2003. In its submission, Petrobart raised six claims altogether.13 Most importantly, Petrobart asked 13

Id. at 18.


the arbitral tribunal to declare that the latter had jurisdiction to entertain the claim and to order the Kyrgyz Republic to pay compensatory damages to Petrobart in the amount of $1.5 million, together with interest and compensation for future loss of profit (lucrum cessans). On the merits, Petrobart complained about a number of ECT violations by the Kyrgyz Republic, including a failure to accord fair and equitable treatment, and illegal expropriation, under Articles 10(1), 10(12), 13(1), and 22(1) of the ECT, respectively.14 The arbitral tribunal had to deal with a number of controversial issues, many of which were raised for the first time in the history of ECT arbitration. Thus, the tribunal found in favor of the Treaty’s application to Gibraltar, though on a provisional basis only.15 The tribunal had to address the issue of ECT Article 17(1)’s applicability—the right of the contracting party (i.e., the Kyrgyz Republic) to deny an investor the advantages of Part III (Investment Protection and Promotion) of the Treaty. The same question was dealt with by another arbitral tribunal in the Plama case.16 In both cases, the tribunals ruled that the conditions for application of Article 17(1) were not met.17 Another respondent’s contention, that the tribunal lacked jurisdiction due to the fact that the matter was already decided by an UNCITRAL arbitration and the case at the Bishkek Court of Arbitration (res judicata), was dismissed since the legal bases of claims in these two cases differed from the present case.18 Although it would be interesting to go into greater depth on all these issues, this contribution will concentrate only on one—the notion of investment as seen by the tribunal, and, as stressed earlier, an evaluation of the adequacy of the tribunal’s reasoning in this portion of the award. III. THE NOTION OF “INVESTMENT” UNDER THE ECT A.

Pertinent Norms and Preliminary Observations

It is a trite observation that the definition of the term “investment” is still developing and its scope differs from treaty to treaty. The Petrobart tribunal begins its analysis by acknowledging the constantly evolving nature of the term “investment:” 14

Id. at 73.

15

Id. at 60–63.

16

Plama, supra note 9, at ¶¶ 143–179.

17


18

Id. at 63–68.


There is no uniform definition of the term investment, but the meaning of this term varies [. . .]. While in ordinary language investment is often understood as being capital or property used as a financial basis for a company or a business activity with the aim to produce revenue or income, wider definitions are frequently found in treaties on the protection of investments, whether bilateral (BITs) or multilateral (MITs).19 In fact, in comparison to most investment protection treaties, the ECT grants the term a very broad scope. Article 1(6) of the Treaty reads: “Investment” means every kind of asset, owned or controlled directly or indirectly by an Investor and includes: (a) tangible and intangible, and movable and immovable, property, and any property rights such as leases, mortgages, liens, and pledges; (b) a company or business enterprise, or shares, stock, or other forms of equity participation in a company or business enterprise, and bonds and other debt of a company or business enterprise; (c) claims to money and claims to performance pursuant to contract having an economic value and associated with an Investment; (d) Intellectual Property; (e) Returns; (f) any right conferred by law or contract or by virtue of any licences and permits granted pursuant to law to undertake any Economic Activity in the Energy Sector. [. . .] “Investment” refers to any investment associated with an Economic Activity in the Energy Sector. In the Petrobart case, the claimant’s submission regarding the term investment rested on three assumptions: (1) Petrobart’s contract with the KGM, (2) claims for money, and (3) the domestic court’s conclusion that the sales represented investments within the meaning of the ECT.20 Accordingly, the arbitral tribunal concentrated its analysis on points

19

Id. at 69.

20

Id. at 24.


(c) and (f) of ECT Article 1(6), which recognize claims for money pursuant to contracts and rights provided by contracts, as investments. The tribunal concluded that Petrobart made an investment in the form of a “right conferred by contract to undertake an economic activity concerning the sale of gas condensate [. . .]. This must also include the right to be paid for such a sale.”21 This conclusion is the bottom line of the tribunal’s determination on Petrobart’s investment. All the tribunal had to do at this point was to adduce sufficient evidence that single sales, including the right to be paid for them, fall within the investment notion under the ECT. The tribunal had to provide such evidence because normally the ordinary commercial transaction and claims arising from it, are denied the qualification of an investment. Does the ECT do away with the differences between commercial and investment disputes in the realm of energy trade and investment, as was the case until now? Or does the traditional reading, which recognizes the inter-connectedness of commercial and investment claims, but admits that their origins are not the same,22 still hold true for the ECT? The arbitral tribunal’s answer suggests that the Treaty marks a new era of development of international investment law, embracing the protection of all proprietary rights of both foreign merchants and foreign

21

Id. at 72.

“[A]n investment dispute is not necessarily distinct from a commercial dispute; it is merely that derives from an investment,” R. DOAK BISHOP, JAMES CRAWFORD & W. MICHAEL REISMAN, FOREIGN INVESTMENT DISPUTES: CASES, MATERIALS AND COMMENTARY 9 (2005). On division between investment and trade in goods in the context of international dispute resolution, see Observations by Georgios Petrochilos and Noah Rubins on the Petrobart award, 2005(3) STOCKHOLM INT’L ARB. REV. 108, ¶ 20. See also CME Czech Republic B.V. v. Czech Republic, UNCITRAL Arbitration Proceedings, Partial Award, (Sept. 13, 2001): 22

¶ 402. [. . .] Commercial disputes and proceedings between private parties, though one party be the investor and/or his joint venture company, do not per se exclude the existence of an investment dispute under the Treaty. ¶ 403. The investment dispute under the Treaty and the commercial dispute between the investors’ joint venture company in the Czech Republic and its shareholders and/or business partners must be distinguished [. . .]. ¶ 404. The private commercial disputes in question are different in respect to the parties, certain basic facts and underlying legal rights and obligations.


investors in the investment definition. Unfortunately, the award lacks any reasoning to this end. It is correct that in contrast to various investment treaties, which explicitly exclude claims for payment arising from sale contracts from the scope of the investment term,23 the ECT leaves enough room to enlist claims of such a kind amidst assets enumerated in ECT Article 23 For example, the North American Free Trade Agreement (NAFTA) and the Free Trade Area of the Americas (FTAA) unambiguously exclude claims for money arising from sale of goods contracts from the “investment” definition. Thus, Article 1139 of NAFTA reads as follows:

But investment does not mean (i) claims to money that arise solely from: (i) commercial contracts for the sale of goods or services by a national or enterprise in the territory of one Party to an enterprise in the territory of another Party.” Article 1 of Chapter XVII “Investments” under the draft FTAA reads: but investment does not mean: . . . j) claims to money that arise solely from: i) commercial contracts for the sale of goods or services by a national or enterprise in the territory of a Party to an enterprise in the territory of another Party.” Http://www.ftaa-alca.org/FTAADraft03/Worddocs03/ChapterXVII_e.doc. The Draft Multilateral Agreement on Investment, prepared by the Organization for Economic Cooperation and Development (OECD) and never adopted, did not recognize claims arising from simple business transactions as investments: see the Commentary to the Consolidated text of the OECD Draft Multilateral Agreement on Investment (MAI) at ¶ 11: “Claims to money may also arise as a result of a sale of goods or services. These claims are not generally considered as investments,” available at http://www1.oecd.org/daf/mai/pdf/ng/ng988r1e.pdf. See also US Model bilateral investment treaty (BIT), footnote 1: “Some forms of debt, such as bonds, debentures, and long-term notes, are more likely to have the characteristics of an investment, while other forms of debt, such as claims to payment that are immediately due and result from the sale of goods or services, are less likely to have such characteristics,” at http://www.ustr.gov/assets/Trade_Sectors/Investment/Model_BIT/asset_upload_file8 47_6897.pdf. It is usually understood that under the ICSID Convention on the Settlement of Investment Disputes, sales contracts do not fall under the “investment” definition; see C.F. Amerasinghe, The Jurisdiction of the International Centre for the Settlement of Investment Disputes,” 19 INDIAN J. INT’L L. 177–81 (1979); P.C. Szasz, A Practical Guide to the Convention on Settlement of Investment Disputes,” 1 CORNELL INT’L L.J. 1 (1968); C.H. SCHREUER, THE ICSID CONVENTION: THE COMMENTARY (2001), all cited in Bishop et al., supra note 22, at 329–30 and 345; Petrochilos & Rubins, supra note 22, at 110, ¶ 22; S. Jagusch, S. & A. Sinclair, The Limits of Protection for Investments and Investors under the Energy Charter Treaty, in C. RIBEIRO ED., INVESTMENT ARBITRATION AND THE ENERGY CHARTER TREATY, 80 (2006).


1(6). In fact, a number of authors lean towards inclusion of contractual claims in the definition of “investments” for the purpose of the ECT. Wälde, for instance, argues in favor of an extensive definition of investment, comprising proprietary rights of any sort. This means that the ECT covers not only contractual claims to performance when the contract is part of a foreign direct investment scheme—the traditional notion, but also simple contract claims, e.g. non-payment on a sales contract or on a judgment based on the sale contract.24 However, this statement by Wälde contradicts an argument he made nearly a decade ago: expansion [of the notion of protected “property”] has not yet reached merely contractual claims against government or private entities, arising out of short-term commercial transactions without relation to a more lasting, more substantial, “investment” project.25 At first glance, this change of opinion is supported by ECT case law, as meager as it stands. In its holding in the Nykomb case, which involved construction of a cogeneration plant in Latvia by a small Swedish firm, the SCC tribunal, without any specification, considered that the purchase of an electric power contract fell “clearly” within the “investment” definition of the ECT.26 In the Plama case, where the Bulgarian government was 24 Th. W. Wälde, Energy Charter Treaty-Based Investment Arbitration. Controversial Issues, 5(3) J. WORLD INV. & TRADE 409–10 (2004). See also Petrochilos & Rubins, supra note 22, at 107–15; T. Weiler, & Th. W. Wälde, Investment Arbitration under the Energy Charter Treaty in the Light of New NAFTA Precedents: Towards a Global Code of Conduct for Economic Regulation, 1(1) TRANSNAT’L DISPUTE MGMT., (Feb. 2004), available at http://www.transnational-dispute-management.com/samples/freearticles/tv1-1-article_51.htm. 25 Th. W. Wälde, International Investment Under the 1994 Energy Charter Treaty, in THOMAS W. WÄLDE ED., THE ENERGY CHARTER TREATY: AN EAST-WEST GATEWAY FOR INVESTMENT AND TRADE 271, footnotes omitted (1996). 26

Nykomb, supra note 8, at 4.3.3(d): the Arbitral Tribunal cannot regard the purchase [of electric power] contract as purely commercial, nor can the action to refuse payment of the double tariff under the contract be considered as purely commercial. As for the objection that the purchase contract is not an investment contract within the meaning of the Treaty, it suffices to note that such a contract clearly falls within the definition of investments in Article 1 of the Treaty.


accused of blocking investor-owned refinery operations, the ICSID tribunal wrote: “The definition of “Investment” under Article 1(6) refers to the Investor’s investment [. . .]; and as already noted above, the definition is broad, extending to “any right conferred by law or contract.” That definition would be satisfied by a contractual or property right even if it were defeasible.”27 However, in contrast to Petrobart’s situation, in both the Nykomb and Plama cases the investors’ claims did not rest exclusively on single sale transactions but were big investment undertakings: in the first case Windau, the Latvian company, which carried out construction of the cogeneration plant, was a subsidiary owned by the Swedish investor; in the second case, a Cypriot investor purchased capital in a Bulgarian jointstock company, the owner of the local oil refinery. Thus, the jurisprudence of the ECT (similarly to the Fedax N.V., Salini Costruttori, and SGS Société Générale de Surveillance S.A. cases, referred to by the tribunal and discussed later in this chapter (see Section III.B.2)) suggests that claims to payment arising from commercial contracts have to be the “accessories” of a broader investment. This requirement of association of contractual claims with the broader investment is rooted in the last paragraph of ECT Article 1(6), which provides that “Investment” refers to investments associated with economic activity in the energy sector. Similarly, Article 1(6)(c) sets the rule for a contractual claim to payment to be associated with an investment. As the following discussion suggests, the Petrobart tribunal ignored the requirement of an “association” while pronouncing on the “claims to money” and “rights conferred by . . . contract” clauses. Arguably, this requirement was of major significance, and the tribunal’s failure to take it into consideration led to incomplete and unconvincing argumentation. As a last remark in this section, it should be said that for an investment claim to succeed, it has to satisfy two mandatory conditions: the existence of both an investment and an investor. Indeed, before going into analysis on the investment made, the tribunal had to secure that Petrobart was an investor that could avail itself of the possibility of recourse to ECT Article 26.28 In accordance with ECT Article 1(7)(a)(ii), the term “Investor” means “a company or other organization organized in accordance with the law applicable in that Contracting Party.” The Treaty does not prescribe any other additional conditions which the 27

Plama, supra note 9, at ¶ 128.

Article 26 ECT is about settlement of disputes between an investor and a contracting party. 28


investor-to-be has to meet in order to benefit from investor status. As a result, the tribunal confirmed that Petrobart was an investor for the purpose of the ECT: Petrobart is a company organized in accordance with the law of Gibraltar. The Arbitral Tribunal has found above that the Treaty should be considered to continue to apply provisionally to Gibraltar, and the law of Gibraltar must in this context be regarded as being part of the law of one of the Contracting Parties, i.e. the United Kingdom. Petrobart therefore satisfies the condition of being an investor under Article 1(7).29 This tribunal’s finding is well reasoned and non-controversial, and does not call for any additional inquiry. Unfortunately, the same cannot be said about the determination on the investment made. B.

The Tribunal’s Approach

In its submission Petrobart argued that its contract with KGM, the claims for money, and the domestic court’s judgment in favor of Petrobart suggested the existence of an investment. Unable to concur with the claimant that the contract and the judgment might in themselves qualify as an investment, the tribunal found it suitable to paraphrase Petrobart’s question in the following manner: whether the rights provided for in the Contract and confirmed in the judgment constituted assets and were therefore an investment in the meaning of the Treaty. In other words, the question is whether Petrobart’s right under the Contract to payment for delivered goods was an asset and constituted an investment under the Treaty30 (emphasis added). The tribunal thus drastically changed the question initially posed by the claimant, in a manner that allowed it to construct the rest of the analysis. However, in doing so, the tribunal failed to explain the reasons for such a decision, nor did it reply to the original question raised by the claimant, at least not explicitly. The tibunal’s subsequent reasoning in the award implicitly upholds the view that the tribunal denied all three variables

29


30

Id. at 71.


(including claims for money) and the qualification of an asset and thus investment. 1.

Contract and Judgment as an Investment

In the eyes of the tribunal, a correct legal analysis precluded qualifying the contract and the judgment as assets (chapeau of ECT Article 1(6)), since these are merely bearers of legal rights: In the Arbitral Tribunal’s view, a correct legal analysis leads to the conclusion that the Contract and the judgment are not in themselves assets but merely legal documents or instruments which are bearers of legal rights, and these legal rights, depending on their character, may or may not be considered as assets31 (emphasis added). Unfortunately, what one cannot find in the award is the “legal analysis” that the tribunal alluded to. The reader is also left with no option but to trust the tribunal’s word that the legal analysis the tribunal undertook was indeed the “correct” one. Finally, there is no indication as to how the analysis “led” to the conclusion that the contract and judgment are not assets in themselves. Although it would be difficult to disagree with the tribunal on this issue and to argue that the tribunal’s conclusion makes little sense, the question still remains: why refer to a correct legal analysis without giving some indications as to the main features of that analysis? What is unequivocal from the language of the award, though, is that the tribunal denied admitting the contract and judgment as constituting an investment. One more phrase in the above-reproduced paragraph is worth considering: According to the tribunal, legal rights stipulated by contracts may be treated as assets “depending on their character.” The fact that the tribunal embarked on a discussion whether the right to payment for goods delivered amounts to an asset and thus an investment, pre-supposes that the tribunal believed the right to payment to be of such a particular character. The nuances of “character” are not, however, explained in the award. Would, for example, the right to performance of contract be treated differently from the right to payment? The award is silent on this point.

31

Id.


2.

Claims to Money as an Investment

As far as claims to money are concerned, ECT Article 1(6)(c) is of primary focus. The clause introduces “claims to money and claims to performance pursuant to contract having an economic value and associated with an Investment” into the investment definition. The tribunal rightly emphasized, without prejudice to any of the possible solutions, that the concept’s scope is a broad one, depending on the terms of the particular treaty, and that “claims to money may constitute investments even if they are not part of a long-term business engagement in another country.”32 To this end, the Petrobart tribunal provided some references to existing bilateral investment treaties and case law where “titles to money,” “any contractual obligation having an economic value,” and “any right of an economic nature conferred by law or by contract” were interpreted by the ICSID Tribunals as falling under the “investment” notion. Thus, the arbitral tribunal in its support of the broadest possible reading of the “claims to money” concept, referred to the Fedax N.V. case33 (involving interpretation of the Netherlands-Venezuelan BIT), where the ICSID tribunal ruled that promissory notes issued by the Venezuelan government formed an investment. The next case, which the Petrobart tribunal adduced in support of its position, was Salini Costruttori v. Morocco,34 where, on the basis of the BIT between Morocco and Italy, the ICSID tribunal held that the term “investment” also includes “any contractual benefit having an economic value.” The Petrobart tribunal’s final reference to the case law was SGS Société Générale de Surveillance S.A. v. the Islamic Republic of Pakistan,35 in which the tribunal found that claims for payment by the Pakistani government for customs inspections services provided by the Swiss investor fall within the notion of “claims to money” under the Pakistan-Swiss BIT.

32

Id.

33 Fedax N.V. v. the Republic of Venezuela, ICSID Case No. ARB/96/3, Decision on Jurisdiction (July 11, 1997) [hereinafter Fedax N.V.]. 34 Salini Costruttori S.p.A. and Italstrade S.p.A. v. Kingdom of Morocco, ICSID Case No. ARB/00/4, Decision on Jurisdiction (July 23, 2001) [hereinafter Salini Costruttori]. 35 SGS Société Générale de Surveillance S.A. v. the Islamic Republic of Pakistan, ICSID Case No. ARB/01/13, Decision on Objections to Jurisdiction (Aug. 6, 2003) [hereinafter SGS Société Générale de Surveillance S.A.].


However, taken out of context, all three references appear to be grossly misleading. Thus, in the Fedax N.V. case, for example, the ICSID tribunal particularly stressed the “significant relationship between the transaction [acquisition of notes] and the development of the host state.”36 This tribunal also considered the difference between a commercial transaction and a transaction leading to an investment, concluding that “the transactions involved in this case are not ordinary commercial transactions and indeed involve a fundamental public policy interest.”37 We find no allusion to these critical paragraphs in the Petrobart award: the tribunal did not examine the question of the Petrobart-KGM transaction in light of its importance to the development of the Kyrgyz Republic or the impact it could have on that country’s public policy. In the same vein, concerning the reference to the Salini Costruttori award, the Petrobart tribunal failed to mention assessment of the parties’ contract for construction of a highway in Morocco in light of the requirements of (1) investor contributions, (2) duration of performance, (3) participation in risk, and, similarly to the argument advanced in the Fedax N.V. award, (4) the investment’s contribution to the economic development of the host state, whereas all of the enumerated criteria were satisfied by the Salini investment.38 Lastly, in SGS Société Générale de Surveillance S.A. the ICSID tribunal particularly accentuated the fact that the investor’s expenditures in the host state “involved the injection of funds into the territory of Pakistan for the carrying out of SGS’s [investor] engagements under the PSI Agreement”39 (emphasis added). Even though the circumstances in the SGS Société Générale de Surveillance S.A. and Petrobart present certain similarities, any analogous reasoning is missing in the Petrobart award. As a result, the conclusion has to be that the Petrobart tribunal’s fractional citations from all three above-mentioned awards are disingenuous, which in turn attests to the tribunal’s disregard for the rest of these highly relevant and decisive argumentations. This last criticism does not impinge on the fact that the tribunal conveyed enough evidence corroborating its earlier statement that in the 36

Fedax N.V., supra note 33, at ¶ 43.

37

Id. at ¶ 42.

38

Salini Costruttori, supra note 34, at ¶¶ 53–58.

SGS Société Générale de Surveillance S.A., supra note 35, at ¶ 136. The separate opinion of March 14, 2003 of Ian Brownlie in the CME v. Czech Republic case echoes this finding: “the nature of an investment as a form of expenditure or transfer of funds for the precise purpose of obtaining a return,” id. at ¶ 34. 39


current state of development of international investment law the notion “claims to money” is a broad one. The tribunal then turned to the question of the scope of this term under the ECT. At this stage, the tribunal encountered some difficulties in proceeding with analysis based on Article 1(6)(c). These difficulties arose due to the fact that—in the opinion of the tribunal—the definition appears to be a circular one. In its complicated semantic analysis, the tribunal wrote: the Arbitral Tribunal considers that the wording “claims to money and claims to performance pursuant to contract having an economic value and associated with an Investment” presents certain ambiguities. In particular, it is not entirely clear whether the words “pursuant to contract having an economic value and associated with an Investment” or parts of these words—“having an economic value and associated with an Investment” or “associated with an Investment”—relate only to “claims to performance” or also to “claims for money.” If we assume that at least the terms “associated with an Investment” also relate to “claims for money,” we are faced with the logical problem that the term “Investment” is not only the term to be defined but is also used as one of the terms by which “Investment” is defined. This means that the definition is in reality a circular one which raises a logical problem and creates some doubt about the correct interpretation.40 Hence, the tribunal opted against testing this ground further, declining to pronounce on “claims to money” as an asset and thus an investment.41 The tribunal declined to admit claims to money as an investment on the basis of “some doubt about the[ir] correct interpretation.” Thus, not being entirely convinced about which path of argumentation to pursue, the tribunal chose to leave the issue open to later tribunals rather than deciding this issue incorrectly. It is nonetheless disputable whether the tribunal tested all the grounds and explored all the possibilities which were available to it. Indeed, there are good reasons to question the tribunal’s determination on the “circularity” of the definition. As initially explained, the tribunal completely ignored the word “association” in this part of the 40


41 For more on this, see B. Poulain, Petrobart vs. Kyrgyz Republic—A Few Reservations Regarding the Tribunal’s Constructions of the Material, Temporal and Spatial Application of the Treaty, 3(3) OGEL ¶¶ 10–12 (Oct. 2005).


investment definition, instead emphasizing the double appearance of the word “investment” (investment is claims to money associated with an investment). There are two separate terms: (1) investment and (2) claim to money, where the latter directly relates to the former, but not a synonym of it. This means that “investment” and “claims to money” terms are not interchangeable, but one (claims to money) is inherently linked—or associated—to the other (investment), the broader investment package.42 All the cases that the tribunal referred to (Fedax N.V., Salini Costruttori, SGS Société Générale de Surveillance S.A.) and the two cases decided based on the ECT (Nykomb and Plama, see the discussion above (Section III.A)) speak in favor of the interpretation hereby proposed: in all of them the investors based their claims not on a single transaction, but on broad investment undertakings. This understanding is shared by Poulain, who advocates a very cautious reading of the investment and claims to money correlation: From our prospective . . . for application of the ECT, if the definition of “investment” effectively covers “claims and rights on services” (which, as such, do not constitute investment), it is only to the extent that they are the accessories of an “Investment.”43 All in all, the paragraph on the circularity of the “claims to money” definition leaves the reader with an impression that the tribunal was unwilling to go into an in-depth analysis of this clause, instead preferring to abandon the discussion where it had only started. The tribunal did not find it appropriate either to dispel “some doubt about the correct interpretation,” or to elucidate those doubts. 3.

Right to Pursue an Economic Activity as an Investment

It becomes plain from the discussion so far that the arbitral tribunal declined to admit all three elements put forward by the claimant as assets under ECT Article 1(6)(c). The tribunal declined to declare the contract and judgment of the domestic court as assets on the premise of a “correct legal analysis;” the reasoning on this point is missing from the body of the award. Later on, the tribunal declined to proceed with an analy42 See also Poulain, id. at ¶ 11. For the opposite view, see Petrochilos and Rubins, according to whom in the absence of travaux préparatoires and other supplementary interpretation tools, the tribunal was “obliged to read the Treaty’s definitions literally” (emphasis added), Petrochilos & Rubins, supra note 22, at 113, ¶ 26. 43

Poulain, supra note 41, at ¶ 11.


sis in light of the “claims for money” clause because of “some doubt about the[ir] correct interpretation.” At this juncture, the tribunal came up with its own solution on how to address Petrobart’s claim to payment: However, in this case further guidance can be sought in Article 1(6)(f) which provides that as an asset constituting an investment shall also be counted “any right conferred by law or contract [. . .] to undertake any Economic Activity in the Energy Sector.”44 Article ECT 1(6)(f) has to be read together with 1(5), which defines “economic activity in the energy sector” as an “economic activity concerning . . . distribution, trade, marketing, or sale of Energy Materials and Products.” Being satisfied that gas condensate (the commodity in dispute) is an energy product, the tribunal found that a right conferred by contract to undertake an economic activity concerning the sale of gas condensate is an investment according to the Treaty. This must also include the right to be paid for such a sale.45 The tribunal inferred that a right pursuant to the contract to undertake an economic activity concerning the sale of gas condensate constituted an investment which is protected by the ECT. Few considerations deserve attention in connection with this conclusion by the tribunal. First of all, at no point of time did the claimant assume that its contract with the KGM conferred on it a right, especially such a right as undertaking economic activity in the energy sector, which would qualify as an asset and thus an investment. The tribunal on its own initiative paraphrased the claimant’s initial claim. This enabled it to ponder whether the right conferred by the contract (to undertake an economic activity) amounted to an asset. As stressed earlier, the tribunal did not explain the reasons for this move. Further, the award is exhaustive on the question whether gas condensate falls within the category of “energy materials and products,” to which the ECT applies. It is thus surprising that the tribunal did not find it appropriate to elaborate on what “distribution,” “trade,” or “sale” of

44


45

Id.


energy materials and products (as some types of economic activity in the energy sector) mean under the ECT: It is not contested that the gas condensate which Petrobart sold in the Contract is to be regarded as Energy Materials and Products. It may be added that gas condensate is not one of the exceptions in Annex NI. This oversight is even more unexpected if we have a look at the Understanding to the Final Act of the ECT Conference.46 The Understanding lists some examples of Economic Activities in the Energy Sector.47 These include “with regard to Article 1(5) ECT marketing and sale of, and trade in Energy Materials and Products, e.g., retail sales of gasoline” (emphasis added).48 Such a retail sale presumably can only be carried out if an investor is established in one or another form (e.g., a gas station) in the territory of the host state. This understanding is also supported by the last paragraph in the definition of an investment in Article 1(6), which says that “investment” refers to any investment associated with an Economic Activity in the Energy Sector.” If one “translates” the example above about gas stations in light of the last paragraph of Article 1(6), 46

See also Poulain, supra note 41, at 15, ¶ 19.

47

Understanding to the Final Act reads: With respect to Article 1(5): (a) It is understood that the Treaty confers no rights to engage in economic activities other than Economic Activities in the Energy Sector. (b) The following activities are illustrative of Economic Activity in the Energy Sector: (i) prospecting and exploration for, and extraction of, e.g., oil, gas, coal and uranium; (ii) construction and operation of power generation facilities, including those powered by wind and other renewable energy sources; (iii) land transportation, distribution, storage and supply of Energy Materials and Products, e.g., by way of transmission and distribution grids and pipelines or dedicated rail lines, and construction of facilities for such, including the laying of oil, gas, and coal-slurry pipelines; (iv) removal and disposal of wastes from energy related facilities such as power stations, including radioactive wastes from nuclear power stations; (v) decommissioning of energy related facilities, including oil rigs, oil refineries and power generating plants; (vi) marketing and sale of, and trade in Energy Materials and Products, e.g., retail sales of gasoline; and (vii) research, consulting, planning, management and design activities related to the activities mentioned above, including those aimed at Improving Energy Efficiency.

48 According to Petrochilos and Rubins, any contract relating to trade, marketing, and sale of energy products falls automatically within the “investment” definition under the ECT; see Petrochilos & Rubins, supra note 22, at 112, ¶ 26.


this would mean that an investor’s contractual claims, for example, to sell gas condensate and to be paid for such a sale, must be associated with that investor’s investment (such as, e.g., purchase of property or shares) in the business of operating gas stations. In contrast, if one paraphrases the last paragraph of ECT Article 1(6) in light of the factual circumstances in the Petrobart case in line with the tribunal’s reasoning, then an impasse occurs: Petrobart’s contractual right to sell gas condensate refers to an investment associated with the sale of gas condensate (as an economic activity in the energy sector). Unfortunately, there was nothing with which Petrobart’s rights under the contract could be associated, since in the Petrobart dispute there was—no and could be no—other investor link to the investment made in the Kyrgyz Republic. Without denying the fact that the drafting of ECT Article 1(6) is confusing (to say the least), it nonetheless requires a link between (1) the economic activity alluded to and (2) assets “owned and controlled” by an investor (chapeau to Article 1(6)). A sale transaction per se, as was the case with Petrobart’s Contract with KGM, without a link to a broader investment scheme, clearly does not demonstrate these characteristics. As such, a reading of the “investment” definition in ECT Article 1(6) by the tribunal pre-supposes a separate view of all notions of the assets listed (tangible and intangible assets, property, and rights), without linking them to investment. This view is confirmed by the fact that the tribunal sees claims to money and the pursuit of an economic activity as independent variables in an investment rather than as part of a broader investment package. Arguably, the text of the Treaty does not entirely support this conclusion. It is highly frustrating that the tribunal reached its determinations without considering those Treaty clauses that appear to be of primary importance for a well-reasoned holding on the issue, even though they offer a range of potentially contradictory interpretations. This reluctance—intentional or not—on the part of the tribunal to address the issue further contributes to doubts about the persuasiveness of the tribunal’s final holdings. C.

The Sveå Court of Appeal’s Approach

Dissatisfied with the award by the SCC arbitral tribunal, the Kyrgyz government appealed the decision to the Sveå Court of Appeal, including the part on determination of an investment. This appeal court upheld the decision of the arbitral tribunal. The part of the judgment dealing with the Kyrgyz claim that by Petrobart had made no investment in the territory of the Kyrgyz Republic reads as follows (this part of the judgment is reproduced in its entirety):


It follows from the investigation that the term investment can have different meaning in different international contexts. The meaning of the term as defined in the ECT has a wide scope of application. It is evident from the testimony given by Adnan Amkhans that when the ECT was negotiated it was an explicit intention to give investment a broad application. The Court of Appeal finds, as well as the Arbitral Tribunal, that the definitions in Article 1(6) and 1(5) in the ECT must be interpreted as including the investment made by Petrobart. Thus, it can not be claimed that the arbitration award is not covered by a valid arbitration agreement between the parties.49 The Sveå Court of Appeal made its final determination based on a testimony by Mr. Adnan Amkhan, the former Head of Legal Affairs of the Energy Charter Secretariat, according to whom the explicit intent of the drafters of the ECT was to give the broadest possible coverage to the term “investment.” However, the question remains whether the intention of the ECT drafters, as reproduced in the testimony (as stressed earlier, the ECT has no travaux préparatoires), takes priority over the objective definition of the term “investment.” IV. PETROBART’S TRANSACTION IN LIGHT OF GENERAL CHARACTERISTICS OF INVESTMENT This section tries to appraise whether Petrobart’s investment, as established by the arbitral tribunal, satisfies the general characteristics of an investment. According to the tribunal, a right conferred by contract to undertake an economic activity concerning the sale of gas condensate, including the right for payment for such a sale, was an investment pursuant to the Treaty. Do these rights meet such main features of investment as (1) certain temporal duration, (2) substantial (financial) commitment by the investor, (3) expectation of profit, (4) undertaking of risk by the investor.50 A fifth and non-compulsory attribute is the “con49

Sveå Court of Appeal judgment, supra note 11.

Bishop et al., supra note 22, at 9. For definitions of investment see, for example, Dolzer: 50

In economics, it is often assumed that an investment involves the transfer of funds, to a longer-term project, for the purpose of regular income, the participation of the person transferring the funds, at least to some extent, in the management of the project, and a business risk. These ele-


tribution of the investment to the development of the state, by building or enhancing its infrastructure or its economy.”51 It should be noted at the outset that no analysis along these lines appears in the arbitral tribunal’s award. As illustrated by the previous discussion, the existence of an investment was a crucial point in the Petrobart dispute. From that standpoint, it is surprising that it was of no concern to the tribunal to consider whether the investment made satisfied basic investment characteristics. The tribunal’s reluctance to provide an appropriate analysis, which would be no less than innovative and even revolutionary, is highly regrettable. Although there was no obligation on the tribunal to do so, given the highly controversial nature of the assumed investment and the scarcity of ECT jurisprudence in general, the tribunal would only benefit from an in-depth revision of the reputed investment in line with the “classic” characteristics of an investment. This review would have secured more credibility for the award, independently of the outcome reached. By means of such an analysis, the tribunal could have availed itself of the opportunity to dismantle the respondent’s contention and, more generally, the doubts of the legal community that a right inferred by private contract to sell energy products, without any supple-

ments distinguish foreign direct investment from a portfolio investment (no element of personal management), from an ordinary transaction for purposes of a sale of a good or a service (no management, no continuous flow of income), and from a short-term financial transaction. R. Dolzer, The Notion of Investment in Recent Practice, in STEVE CHARNOVITZ, DEBRA P. STEGER & PETER VAN DEN BOSSCHE EDS., LAW IN THE SERVICE OF HUMAN DIGNITY: ESSAYS IN HONOUR OF FLORENTINO FELICIANO 263(2005). See also Sornarajah: “Foreign investment involves the transfer of tangible or intangible assets from one country into another for the purpose of their use in that country to generate wealth under the total or partial control of the owner of the assets.” M. SORNARAJAH, THE INTERNATIONAL LAW ON FOREIGN INVESTMENT, 7 (2d ed. 2004). Finally, see the OECD, which offers a somewhat similar reading of what constitutes an investment: Foreign direct investment reflects the objective of obtaining a lasting interest by a resident entity in one economy (“direct investor”) in an entity resident in an economy other than that of the investor (“direct investment enterprise”). The lasting interest implies the existence of a long-term relationship between the direct investor and the enterprise and a significant degree of influence on the management of the enterprise. OECD BENCHMARK DEFINITION OF FOREIGN DIRECT INVESTMENT 7–8 (3d ed. 1999). 51

Bishop et al., supra note 22, at 9. See also M.J. TREBILCOCK & R. HOWSE, THE REGTRADE 441 (3d ed. 2005).

ULATION OF INTERNATIONAL


mentary link to the host state or broader investment scheme as such, may indeed amount to an investment within the meaning of the ECT. Certainly, there are compelling reasons to believe that a single commercial transaction between Petrobart and KGM arguably lacked the basic characteristics of an investment, and the following paragraphs proceed with an inquiry into the nature of the Petrobart/KGM sales agreement in light of each earlier-outlined attribute of an investment. As emphasized on numerous occasions, what we see in the given case is an ordinary commercial transaction. This was a contract to supply goods concluded for a period of one year only, with the remote possibility of being extended in the future. Thus, one cannot really speak about longterm commitment on the part of an investor in the given circumstances. The fact that the sale took the form of multiple deliveries rather than one single release does not impinge on the earlier determination. There was no commitment of capital or transfer of funds by Petrobart. This understanding is also shared by the tribunal, which ruled that “the Contract did not involve any transfer of money or property as capital in a business in the Kyrgyz Republic but was a sales contract.”52 Rather, in line with the terms of the contract, it was entitled to payment for deliveries of the commodity. The argument that Petrobart had first to purchase gas condensate from the Uzbek company, Uzneftegazdobicha, thus presumably advancing sums of money, is of no relevance in the present context. Indisputably, a commercial risk was involved, but this aspect is common to all commercial and investment operations. Similarly, there was no expectation of profit, but rather a right to receive payment for goods delivered. Additionally, absence of the alleged investor’s attachment to the host state also speaks in favor of the transaction’s qualifying as a mere commercial dealing. This argument can be split into two parts.53 First, from 52


“In international investment two polar strategies can be discerned. In one, the savers of one economy invest on a net basis in assets of another economy, typically through international loans, bonds, and portfolio investment. In the other, a firm opens a branch or subsidiary in another nation.” JOHN H. BARTON, INTERNATIONAL TRADE AND INVESTMENT: REGULATING INTERNATIONAL BUSINESS 855 (1986). See also OECD BENCHMARK DEFINITION . . . , supra note 50, at 9: 53

In the process of globalisation of economic activities, cross border transactions are carried out that at first glance may be regarded as foreign direct investment when in fact they do not meet the criteria. For exam-


the case materials it follows that there was no acquisition (be that KGM or any other enterprise) or construction of physical capital by Petrobart in the territory of the Kyrgyz Republic, to which the transaction here examined would be directly related (or “associated,” as the last paragraph pf Article 1(6) puts is, see discussion above (see Section III.B.2)). Second, the investor’s presence in the target market is demonstrated if it can be shown that it operates in the host state through a subsidiary or associate54 (as was the case with the Nykomb and Plama cases). It is not disputed that there was no such representation in the given circumstances: the claimant had offices in Gibraltar, London, Almaty, and Moscow, but not on the territory of the Kyrgyz Republic. Equally important, the historical context of the notion of “protection of investment” emphasizes the fact that states grant protection to aliens and their property on the territory of the host state.55 Indeed, ECT Article 26 on the settlement of investment disputes is drafted bearing this rationale in mind. This clause spells out the settlement of disputes between a contracting party and an investor of another contracting party relating to an investment of the latter [made] in the Area of the former, that is, the territory under its sovereignty (ECT Article 1(10)). Furthermore, ECT Article 1(8) ECT gives a definition of “Make Investments” or “Making of Investments,” which signify “establishing new Investments, acquiring all or part of existing Investments or moving into different fields of Investment activity” (emphasis added). Or, as Wälde stated, An investment would, therefore, be seen as “made” when . . . a right is validly acquired, something that might take years, before ple: . . . e) Other cases might include foreign sales and representative offices . . . . Such offices or activities can be treated as direct investment only if they meet the requirements of residence and the attribution of production in an economy. 54 See, for example, the OECD proposed definition of an investor: “A foreign direct investor is . . . an incorporated or unincorporated public or private enterprise, . . . which has a direct investment enterprise—that is, a subsidiary, associate or branch—operating in a country other than the country or countries of residence of the foreign direct investor or investors,” in OECD Benchmark Definition . . . , supra note 50, at p. 8. 55 “Historically, [foreign investment] law has been built up as a part of the area of the diplomatic protection of citizens abroad and of state responsibility for injuries to aliens . . . The definition of foreign investment must be rooted in this historical sense and not be extended beyond the meaning attributed to it in state practice and the precise words used in the treaties,” Sornarajah, supra note 50, at 18.


a commercial project is established. From the moment the right is established (be it of a preliminary character to be followed by other “rights,” e.g., an exploration right to be followed by a production licence or production-sharing contract . . .), an “investment” is made and this investment . . . is protected.56 It is thus striking that notwithstanding the crucial importance of the earlier mentioned ECT Articles 1(8), 1(10), and 26(1) for determining whether an investment took place in the circumstances here described, the tribunal failed to make use of them in order to sustain its position. All this pre-supposes a long-term commitment of capital by the investor. However, the previous discussion makes it plain that Petrobart did not meet any of these conditions. A private law commercial contract was the only basis for Petrobart’s claim, and this was a valid cause for a commercial claim. However, as the discussion above shows (see Section III.A), it is not clear whether the ECT embraces pure contractual claims in its scope. The interpretation proposed by the Petrobart tribunal suggests that this is the case. At the same time, as was demonstrated, in reaching this conclusion the tribunal did not provide plausible reasons in support of this contention. As a result, on one side a strong case was made proving that Petrobart’s transaction failed to meet the traditional characteristics of an investment, while on another side a weak case suggests that contractual claims of Petrobart’s type fall within the investment notion under the ECT. V.

CONCLUSION

The crucial point in the Petrobart award was the determination on investment. The award is remarkable in that the arbitral tribunal held rights embedded in a private law sale of goods contract, which lacked any association with a broader investment, as an asset and thus as an investment under the ECT. In that sense, this award continues the tradition of previous decisions involving interpretation of the ECT and further strengthens the position of foreign investors vis-à-vis contracting states.57 Yet, in its search to accord maximum protection to the alleged investor, 56

Th. W. Wälde, supra note 25, at 280.

57

In this regard, see also the prophetic words of Wälde in 1996: well-established international energy companies are unlikely to carry out major investment in high-risk countries (such as all of the CIS) without a carefully crafted international arbitration clause. However, unexperi-


the tribunal did not respect its duty to provide the parties to the dispute with cogent reasons for its major and far-reaching conclusions. The point of departure for the present discussion was that no party should be denied reasoning. However, in its determination on the investment made, the tribunal first declined to qualify all three variables put forward by the claimant as an investment on the basis of “correct legal analysis,” the text of which is lacking in the body of the award and “some doubt about the[ir] correct interpretation.” Instead, the tribunal proposed its own approach on how to address the issue and ruled that a contracting right to undertake economic activity in the form of sale of energy products, constitutes an asset. We find no convincing arguments in the body of the award in favor of the tribunal’s belief that a purely commercial claim arising from the Petrobart-type transaction qualifies as an asset and thus an investment for the purposes of the ECT. The tribunal significantly stretched the outer limits of the term “investment,” so that this new interpretation stands in sharp contrast with today’s mainstream interpretation. As with any significant departure from widely accepted and followed practice, a new proposed approach requires good reasons to make others believe its accuracy. Without disagreeing that rights conferred by sales contracts may give rise to investment claims, the award contains no compelling arguments to sustain the tribunal’s holding. As a result, on the one hand, the tribunal opened the door to bringing commercial claims under the auspices of the ECT by advocating an extremely liberal reading of the notion “investment.” On the other hand, given the lack of cogent arguments to support the tribunal’s determinations, it is difficult to predict whether this approach will be of any precedential value for later tribunals. All in all, the failure of the tribunal to provide adequate reasoning for its determinations calls the credibility and authority of the final award into question.

enced “junior” companies, the adventurous promoters one now finds very often in the former Soviet Union, are keener to get some form of contract or title at once than to apply time, resources and expertise to negotiate a satisfactory investment agreement. This type of promoterinvestor benefits from the Treaty: it obtains the arbitral right—which greatly bolsters whatever title/contract obtained—without having to bargain of it. Th. W. Wälde, Investment Arbitration under the Energy Charter Treaty—From Dispute Settlement to Treaty Implementation, 12(4) ARB. INT’L 439 (1996).

TABLE OF LEGAL INSTRUMENTS

MULTILATERAL INSTRUMENTS Council of Europe European Convention on Human Rights Art. 6(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Energy Charter Conference Energy Charter Treaty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323–47 Ann. D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323–24 Art. 1(5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339–40, 342 Art. 1(6) . . . . . . . . . . . . . . . . . 328–29, 331–32, 334, 340–42, 345 Art. 1(6)(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332, 335, 337–38 Art. 1(6)(f) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339 Art. 1(7) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332–33 Art. 1(8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346 Art. 1(10) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345–46 Art. 10(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Art. 10(12) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Art. 13(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Art. 17(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Art. 22(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Art. 26 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332 Art. 26(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346 Pt III . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 European Community Treaty establishing the European Community . . . . . . . . . . . . . . . . . . 298 Art. 190 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 International Center for the Settlement of Investment Disputes (ICSID) Additional Facility Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Art. 52(2)(i). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Art. 53(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12–13 Arbitration Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Convention on the Settlement of Investment Disputes between States and Nationals of Other States (Washington Convention) . . . 118, 147, 151–52, 159–60, 184, 202–03, 207–10, 257–58, 260, 305 349


Art. 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 Art. 25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194–95, 201, 300 Art. 25(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Art. 25(2)(b) . . . . . . . . . . . . . . . . . . . . . . . 24, 193–201, 209, 236 Art. 41(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Art. 42 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Art. 42(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–4, 6–7 Art. 43 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 Art. 48 117, 231, 235, 237, 241, 247, 249, 252, 255, 258–59, 323 Art. 48(3). . . . . . . . . . . . . . . 2–4, 6, 117, 148, 244, 249, 252, 324 Art. 50 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Art. 52 . . . . . . . . . . . . 4–10, 14, 16, 20, 25–26, 231, 249, 252–55 Art. 52(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249, 254–55 Art. 52(1)(e) . . . . . . 3–4, 9, 11, 14–18, 20, 145, 151–52, 251–52 Art. 52(4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Art. 53 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 249 Regulations and Rules Art. 47(i) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 International Court of Justice Statute Art. 38(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Art. 38(1)(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 International Law Commission Articles on State Responsibility Art. 25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25–26, 142 Art. 27. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142–45 Art. 27(b) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 Art. 38 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 Art. 44 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 Draft Articles on Diplomatic Protection Art. 5(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Art. 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Art. 14(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 Art. 15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316–17 Draft Code on International Commercial Arbitration. . . . . . . 3 Statute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Vienna Convention on the Law of Treaties Art. 28 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Art. 31 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6, 121, 196

Table of Legal Instruments • 351

Art. 31(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167, 298 Art. 31(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 NAFTA North American Free Trade Agreement . . . . . . . . . . 31, 33–60, 63–116, 122, 127, 134, 160, 166, 261, 264–91 Art. 1019 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Art. 1101 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296 Art. 1101(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Art. 1102 . . . . . . . . . . . . . 35, 37–38, 58, 63, 67, 77, 101–02, 112, 228, 264, 276, 281–82, 285, 291 Art. 1102(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Art. 1102(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58, 100 Art. 1102(3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Art. 1105 . . . . . . . . . . . . . . . . . . 11, 38, 40, 45, 48–9, 60, 67, 264, 269, 276, 285, 288, 292–93, 313 Art. 1105(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38–41, 45–46 Art. 1110 . . . . . . . . . . . . . . . . 38, 54–56, 60, 67, 77, 85, 87–8, 91, 98, 264, 269, 276, 281–82, 285 Art. 1110(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Art. 1110(1)(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Art. 1116 . . . . . . . . . . . . . . 33–34, 37, 54, 297, 299, 306, 309–11 Art. 1116(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Art. 1117 . . . . . . . . . . . . . . 33–34, 37, 67, 69, 264, 297, 299, 310 Art. 1117(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68, 72–74 Art. 1120 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Art. 1121. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311–15 Art. 1121(2)(b) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79, 115 Art. 1131 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40, 78 Art. 1131(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Art. 1136(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Art. 1139 . . . . . . . . . . . . . . . . . . . . . 54, 80–2, 84–85, 88, 94, 330 Art. 1701 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Art. 1716 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Art. 2103(6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Art. 2105 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104–07, 112 Ch. 11 . . . . . . . . . . . . . . . 37–38, 50, 54, 56–57, 85, 97, 207, 267, 270, 277, 295–96, 303 OECD Draft Convention on the Protection of Foreign Property . . . . . . . . . 178


Stockholm Chamber of Commerce Arbitration Institute 1999 Rules Art. 32(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 2007 Rules Art. 36(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 UN Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York Convention) Art. V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18–19 General Assembly Resolution on Permanent Sovereignty Over Natural Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 UNCITRAL Arbitration Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151–52, 159, 184 Art. 32 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 Art. 32(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149, 153 Art. 32(3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2, 175, 187–88 Model Law on International Commercial Arbitration. . . . . . . . 152, 159 Art. 31(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 WTO Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) Pt III . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Other Draft Convention on the International Responsibility of States for Injuries to Aliens (Harvard Draft Convention) . . . . . . . . . . . 39, 47, 134, 177–78, 268, 317 BILATERAL TREATIES Argentina-Ecuador Bilateral Investment Treaty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 Czech Republic-Netherlands Bilateral Investment Treaty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149–90 Art. 1(a) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 Art. 1(a)(ii) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 Art. 1(b)(ii) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Art. 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162


Art. 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165–66, 170, 178, 184 Art. 3(1) . . . . . . . . . . . . . . . . . . . . . . . . 165, 167–68, 172–74, 182 Art. 3(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165, 174 Art. 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175–78, 180–85 Art. 5(a)–(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177, 182–83 Art. 8(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162–63 Art. 8(6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160–61 France-Italy Treaty of Peace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Pakistan-Switzerland Bilateral Investment Treaty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335 Spain-Ecuador Bilateral Investment Treaty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 Ukraine-Lithuania Bilateral Investment Treaty. . . . . . . . . . . . . . . . . . . . . . . 193–210, 304–05 Art. 1(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195, 197, 199–201 Art. 1(2)(a) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Art. 1(2)(b) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196–97 Art. 1(2)(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193, 195–7, 204 United Kingdom-Egypt Bilateral Investment Treaty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233–60 Art. 1(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 Art. 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239–40, 254 Art. 8(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236–38 United States-Argentina Bilateral Investment Treaty . . . . . . . . . . . . . . . 19–20, 22, 24, 26, 118–48 Art. II(2)(b). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123–24 Art. II(2)(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21–24 Art. IV(3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141–42 Art. XI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25–26, 139–45 Supplementary Agreement Art. I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 United States-Ecuador Bilateral Investment Treaty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211–36 Art. I(i), (iii) and (v) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 Art. II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221


Art. II(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223–24, 227 Art. II(3)(a) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Art. II(3)(b). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228–29 Art. III . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Art. IV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Art. VI(1)(a) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Art. X . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214, 219–22 Art. X(1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Art. X(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221–23 Art. X(2)(a) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Art. X(2)(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 NATIONAL LEGISLATION Argentina Convertibility Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Decree 1189/92 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Emergency Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118, 120 Gas Law120–21, 146 Resolution 38/02 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 State Reform Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Germany Civil Code para. 203 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Greece Civil Code Art. 255 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Mexico Federal Law on Games and Raffles . . . . . . . . . . . . . . . . . . . . . . . . 264–96 Art. 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264–65 Art. 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 IEPS law 64–67, 73–74, 76–78, 81, 85, 88–89, 91–93, 95, 100, 102, 104, 111 Art. 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Art. 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . 65–66, 68, 90, 93–7, 102 Peru Decree 258. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17–18 Decree 259. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17–18


Switzerland Federal Code of Private International Law (CPIL) Art. 180(2)(e) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152–53 Art. 189(2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 United States Civil Rights Act (Federal) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47–48 Restatement (Third) of Foreign Relations Law of the United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88–89, 97, 134, 178, 268 Tort Claims Act (Federal). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Tort Claims Act (Massachusetts) § 10(c) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36, 45, 47–48 United States Bankruptcy Code Ch. 11. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299, 310

TABLE OF CASES (For entries with numerous references, the main discussion is indicated in boldface type.)

Aconquija, see Compañía de Aguas del Aconquija, S.A. & Compagnie Générale des Eaux (Vivendi) v. Argentina ADF Group Inc. v. United States . . . . . . . . . . . . . . . . . . . . . . . . . . 34, 261 AES Summit Generation v. Republic of Hungary. . . . . . . . . . . . . . . . 324 AGOSI v. United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 Aguas del Tunari, S.A. v. Republic of Bolivia . . . . . . . . . . . . . . . . . . . 208 Alstom Power Italia SpA and Alstom SpA v. Republic of Mongolia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324–25 Ambatielos (Greece v. UK). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 Amco Asia Corp. and Others v. The Republic of Indonesia . . . . . . . . . . . . . . . . . . . . . . . . 10–12, 16, 40, 118, 197, 250 Arbitral Award Made by the King of Spain on 23 December 1906, see Honduras v. Nicaragua Ashingdane v. United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Azinian v. Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40, 50, 87, 262 Azurix Corp. v. Argentina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145, 218 Barcelona Traction, Light and Power Company, Limited (Belgium v. Spain) . . . . . . . . . . . . 43, 191, 198, 203, 205–06 Biloune and Marine Drive Complex Ltd v Ghana Investments Centre and Government of Ghana . . . . . . . . . . . . . . . . . . . . . . . . . 54 Carlsen and Others v. Council of the European Union. . . . . . . . . . . . 28 Ceskoslovenska Obchodni Banka, A.S. v. Slovak Republic. . . . . . . . . 209 Chevron U.S.A., Inc. v. Natural Res. Defense Council, Inc.. . . . . . . . 284 Chorzów Factory Case (Germany v. Poland). . . . . . . . . . . . . . . . . . . . 308 CME Czech Republic B.V. (Netherlands) v. Czech Republic . . . . . . . . . . . . . . . 36, 53, 56, 60, 131, 261, 329, 336 CMS Gas Transmission Co. v. Argentina . . . . . . . . . . . . . 16–17, 19–26, 119, 122, 129, 145–46, 218 Compañía de Aguas del Aconquija, S.A. & Compagnie Générale des Eaux (Vivendi) v. Argentina. . . . . . . . 18, 20, 145, 218 Ecuador v. Occidental Expl. & Prod. Co. . . . . . . . . . . . . . . . . . . . . . . 212 Elettronica Sicula S.p.A. (ELSI) (United States v. Italy) . . . . . . . . . . . . . . . . . 40, 44, 124–26, 128, 206 ELSI, see Elettronica Sicula S.p.A. (ELSI) (United States v. Italy) EnCana Corp. v. Ecuador . . . . . . . . . . . . . . . . . . . . 211–14, 229, 261, 263 357


Ethyl Corp. v. Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Eureko v. Poland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Fedax NV v. Republic of Venezuela . . . . . . . . 209, 217, 332, 335–36, 338 Feldman v. Mexico . . . . . . . . . . . . . . . . . . 54, 63–115, 127, 138, 241, 313 Fields of Texas, Inc. v. Islamic Republic of Iran . . . . . . . . . . . . . . . . . . 51 Finnish Shipowners’ Claim in respect of use of Finnish vessels during the War (Finland v. United Kingdom) . . . . . 316, 319 Foremost Tehran, Inc. v. Islamic Republic of Iran . . . . . . . . . . . . . . . . 56 Gabcíkovo-Nagymaros Project (Hungary/Slovakia). . . . . . . . . . . . . . 144 GAMI Investments, Inc. v. Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 García and García (Mexico) v. United States . . . . . . . . . . . . . . . . . . . . 59 Garrison Case (United States v. Mexico). . . . . . . . . . . . . . . . . . . . . . . . 44 Genin, Eastern Credit Ltd., Inc. and A. S. Baltoil v. Republic of Estonia . . . . . . . . . . . . . . . . . . . . . . . . . 135, 166–67, 218 German Interests in Polish Upper Silesia . . . . . . . . . . . . . . . . . . . . . . 285 Harza Engineering Company v. Islamic Republic of Iran . . . . . . . . . 308 Honduras v. Nicaragua . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Houston Indus. Energy, Inc. and Others v. Argentina . . . . . . . . . . . . 119 Hurney v. Boston Redev. Auth.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Industria Nacional de Alimentos, S.A. and Indalsa Perú, S.A. v. Republic of Peru . . . . . . . . . . . . . . . . . . . 16–18 Interhandel (Switzerland v. United States) . . . . . . . . . . . . . . . . . . . . . . 47 Int’l Thunderbird Gaming Corp. v. Mexico . . . . . . . . . . 41, 238, 261–90 Isaiah v. Bank Mellat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Jalapa Railroad and Power Co. (United States v. Mexico) . . . . . . . . . . 43 Klöckner Industrie-Anlagen GmbH and Others v. Republic of Cameroon . . . . . . . . . . . . . . . . 4–12, 16, 20, 27–28, 250 LeBlanc v. Molloy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Lewis v. Boston Redev. Auth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 LG&E Energy Corp., LG&E Capital Corp. and LG&E Int’l Inc. v. Argentina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117–48 Loewen Group, Inc. & Raymond L. Loewen v. United States . . . . . . . . . . . . . . . . . . . . . . . . 34, 44, 59, 262, 291–321 Lucchetti, see Industria Nacional de Alimentos, S.A. and Indalsa Perú, S.A. v. Republic of Peru Maffezzini v. Kingdom of Spain . . . . . . . . . . . . . . . . . . . . . . . . 40–41, 228 Martini (Italy v. Venezuela) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43, 284 Marvin Feldman, see Feldman v. Mexico Matthews v. Ministry of Defense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Mayer v. Boston Metropolitan Airport. . . . . . . . . . . . . . . . . . . . . . . . . . 43 McCullough & Company, Inc. v. Ministry of Post, Telegraph and Telephone . . . . . . . . . . . . . . . . . . . . . . . . 245–46, 248

Table of Cases • 359

McElhinner v. Ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Metalclad Corp. v. Mexico (see also Mexico v. Metalclad) 12–13, 40, 54–56, 99, 137–38, 184, 217, 240, 253, 313 Methanex Corp. v. United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Mexico v. Metalclad (see also Metalclad Corp.) . . . . . . . . . . . . . . . . 12–13 MINE v. Guinea . . . . . 11–12, 14, 16, 18, 20, 117, 120, 147, 249–50, 255 Mitchell v. Democratic Republic of Congo . . . . . . . . . . . . 13, 15–16, 148 Mondev International Ltd. v. United States . . . . . . 33–61, 166, 262, 291 Moran v. Bench . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 MTD v. Chile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 National Property Fund of the Czech Republic and Czech Republic v. Nomura Principal Investment plc (Japan) . . . . 153, 159 Neer, see United States (L.F. Neer) v. Mexico Noble Ventures, Inc. v. Romania . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Nomura Principal Investment Plc and Nomura International Plc v. Commission . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Nottebohm (Liechtenstein. v. Guatemala.). . . . . . . . . . . . . . . . . . . . . 191 Nykomb Synergetics Technology Holding AB v. Republic of Latvia . . . . . . . . . . . . . . . . . . . . . . 324, 331–32, 338, 345 Occidental Expl. & Prod. Co. v. Ecuador . . . . . . . . . . . . . . . 122, 126–27, 168, 172, 211–30, 253, 261 Petrobart Limited v. Kyrgyz Republic . . . . . . . . . . . . . . . . . . . . . . 323–47 Plama Consortium Ltd. v. Republic of Bulgaria. . . . . . . . . . . . . . . . . . . . . . . . . . . 324, 327, 331–32, 338, 345 Pope & Talbot, Inc. v. Canada 33–34, 40, 44, 99, 127, 131, 134, 166, 228 Poznik v. Massachusetts Med. Prof’l Ins. Assoc. . . . . . . . . . . . . . . . . . . 45 Raso v. Lago . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Regina v. Bouchereau . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Rihani Claim (United States-Mexico) . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Russian Indemnity Case (Russia v. Turkey) . . . . . . . . . . . . . . . . . . . . . 141 Salini Costruttori S.p.A. and Italstrade S.p.A. v. Kingdom of Morocco. . . . . . . . . . . . . . . . . . . . . . . . 332, 335–36, 338 Saluka Investments BV (Netherlands) v. Czech Republic . . . . 136, 149–90 S.D. Myers, Inc. v. Government of Canada . . . . . . . . . . . . . . . . . 33–34, 40, 56, 58, 99, 127–28, 179, 273 SEDCO Inc. et al. v. National Iranian Oil Company et al. . . . . . . . . . 268 Sempra Energy Int’l v. Argentina. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Serbian Loans (Case concerning payment of various Serbian Loans issued in France) . . . . . . . . . . . . . . . . . . . . . . . . . . 284 SGS Société Générale de Surveillance S.A v. Islamic Republic of Pakistan . . . . . . . . . . . . . . . . . . . . . . . . 332, 335–36, 338


Soufraki v. United Arab Emirates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 South West Africa (Liberia v. S. Africa) . . . . . . . . . . . . . . . . . . . . . . . . 205 SPP (Middle East) Ltd v. Arab Republic of Egypt. . . . . . . . . . . . . . . . 240 Starrett Housing v. Iran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56, 134 Sunday Times v. United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Tecmed, see Técnicas Medioambientales Tecmed, S.A. v. Mexico Técnicas Medioambientales Tecmed, S.A. v. Mexico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41, 122, 132–34, 261 Temple of Preah Vihear (Cambodia v. Thailand) . . . . . . . . . . . . . . . 308 Thunderbird, see Int’l Thunderbird Gaming Corp. v. Mexico Tippetts, Abbett, McCarty, Stratton v. TAMS-AFFA. . . . . . . . . . . . 54, 308 Tokios Tokelés v. Ukraine . . . . . . . . . . . . . . . . . . . . 191–210, 238, 304–06 Torkmain Investments Ltd et al. v. Pembridge Investments BV et al. . . . . . . . . . . . . . . . . . . . . . . . . 155, 159, 185–87 TP and KM v. United Kingdom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Tunari, see Aguas del Tunari, S.A. v. Republic of Bolivia United States (L.F. Neer) v. Mexico. . . . . . . . . . . . . . . . . . . . . . . . . 39, 41 Vivendi, see Compañía de Aguas del Aconquija, S.A. & Compagnie Générale des Eaux (Vivendi) v. Argentina Waste Management, Inc. v. Mexico . . . . . 50, 122, 135, 168, 262–63, 313 Wena Hotels Limited v. Arab Republic of Egypt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20–21, 23, 189, 231–60

INDEX (For entries with numerous references, the main discussion is indicated in boldface type.)

acceptance of rulings, 230, 256–60, 320–21, 341, see also legitimacy of awards ad hoc committees, 4–6, 10–11, 13–16, 18–22, 25 curing of defective awards, 237, 252–55 exceeding powers, 231, 237, 249, 252–55 MINE case, 117–18 and reasons requirement, 249–49 Wena case, 231–32, 235, 248–56, 258, 260 adequacy of reasoning, 1, 11, 25, see also inadequacy of reasoning definition of adequate reasons, 159 LG&E case, 117, 121, 145 Mondev case, 34 Petrobart case, 325, 327 Saluka case, 150, 185 Wena case, 249–50 administrative agencies, 262–63, 272, 283–84, 286, 290 admissibility, 8, 215–16 aggregate argumentation, 64 AHC, see ad hoc committees Ahdieh, R., 59 ambiguous assurances, 288–89 annulment proceedings, 3, 5–6, 8–14, 17–20, 24, see also ad hoc Committees difference from appeal procedure, 14, see also non-court of appeals doctrine limited function, 20 MINE case, 117–18 Mitchell case, 148 and reasons requirement, 159

Wena case, 231, 234–35, 234–35, 244, 247–53, 252–55 applicable law, 3, 16 Feldman case, 85, 88–89, 97 Mondev case, 40 Saluka case, 159–61, 189 Thunderbird case, 262 Tokios Tokeles case, 201 arbitrariness, see also non-impairment definition, 229–30 Feldman case, 63, 96–97 LG&E case, 117, 119, 123, 126, 129–30 Occidental case, 214, 228–30 Thunderbird case, 262, 269, 273–74, 277, 288–90 Argentina, 17 CMS case, 19–27, 145–46 LG&E case, 117–48 state of necessity, 119, 139–46 audit failure, discrimination, 104–07 authority of awards, see acceptance of rulings; legitimacy of awards autonomous treaty standards, 166, 168, 178 balancing tests, 135–37 banking sector, 149–90 bankruptcy, 36, 157, 296, 299, 303, 306–07, 309 “basic principle,” 6–8 benefits of a well-reasoned award, 187–89 Berman, Sir Franklin, 18–19 Bishop, R.D., 246 bona fide general taxation, see general taxation

361

362 • The Reasons Requirement in International Investment Arbitration bona fide regulatory action, 89, 177, 179, 181, see also regulation Boston Redevelopment Authority (BRA), 35 Brower, C., 242 Brownlie, I., 41, 124 burden of proof amount of investment, 241–44 discrimination, 104, 106–07, 109–15 Feldman case, 64, 95, 104, 106–7, 110–15 LG&E case, 141 Loewen case, 318–20 Mondev case, 47 and prima facie case, 106–07, 110–14, 235 Thunderbird case, 281, 285–86 Wena case, 235, 241–43, 251 burden of review, 141, 307 business problems and expropriation, 87–88 Cameroon, 4–10 Campeau, 43, 53 Canada Barcelona Traction case, 205 domestic courts and reasons requirement, 12–13 EnCana case, 211 Feldman case, 99, 106 Loewen case, 291–92, 296, 303–04 Mondev case, 33–35, 40–41, 55–56, 58 Pope & Talbot case, 166, 228 S.D.Myers case, 127–28, 166, 179 Thunderbird case, 261, 264 capital source and investor status, 199 case law, departure from, 42–43, see also jurisprudence; precedent causation, Mondev case, 33 CEMSA, 63–115 certiorari, 37–38, 50, 295, 297, 317 chance, 264–67, 269–72, 276–79, 286, 292 characterization, 218–19 cigarette production, 64–115

circular reasoning, 139, 143, 337–38 citizenship, see nationality claims to money, as investment, 335–38 coherence of reasoning, see internal coherence commercial disputes and investment disputes, Petrobart case, 328–32 common law adjudication, limits of, 43 compensation, 16, see also damages discounted cash flow (DCF) method, 240 LG&E case, 120–21, 132–33, 135, 138, 140, 142–45, 147 market value before expropriation, 241–44 Mondev case, 35, 38, 50, 53, 55–57, 60–61 Saluka case, 150, 176–77 valuation, see valuation of compensation Wena case, 239, 251–55 compounding of interest, 244–47 confusion among commentators, 129–30 Congo, Democratic Republic, 13–16 consensual obligations, 23–24 consent, see acceptance of rulings; legitimacy of awards consistency with earlier arbitrations, 99 continuing/continuous nationality, 299–309, 321 contra legem assurances, 287–89 contradiction of reasons, 5, 8, 14, 18, see also inadequacy of reasoning; inconsistencies; internal coherence; unreasoned findings Feldman case, 80, 89, 91–92, 94, 96–97, 111 Klöckner case, 8 LG&E case, 123, 128–29, 131–32, 142, 145–47 Loewen case, 320–21

Index • 363 Lucchetti case, 17–18 Occidental case, 222 Petrobart case, 341 Saluka case, 176–83 contradictory reasons, see contradiction of reasons control test, 194, 196–98, 202 controlling language, 176 corporate nationality, see nationality corporate veil, 163, 194, 198, 205–6, 210, 304–06 corruption, 233–34 court of appeal, tribunal acting as, 287–89, see also non-court of appeals doctrine Crawford, J., 246, 263 credibility of rulings, see acceptance of rulings; legitimacy of awards creeping expropriation, see indirect expropriation CSOB, 156–59, 170–72, 170–72, 183, 185–88 customary international law denial of justice, 292–93 and expropriation, 177–79, 181–82 fair and equitable treatment, 166–67 Feldman case, 79, 87, 91 and immunity, 45–50 LG&E case, 124, 138 local remedies, see local remedies, exhaustion Loewen case, 312–13 minimum standard of treatment, 40–41 Mondev case, 33, 38–41, 57 non-discrimination principle, 124 Saluka case, 165 Tokios Tokeles case, 203 Wena case, 247 Czech National Bank (CNB), 153, 155–56, 155–57, 162–63, 181–82 Czech Republic CME case, 53, 56, 131, 329 Saluka case, 149–90

damages, 1, 12, see also compensation Feldman case, 77, 79 LG&E case, 119, 147 Loewen case, 295 Mondev case, 33, 50–51 Saluka case, 165, 188 Wena case, 234, 241–48, 251–55, 258–59 dates Loewen case, 300–303 Thunderbird case, 269 Tokios Tokeles case, 206–08 Wena case, 247–48, 254–55, see also dies a quo de minimis defects, 10 de Visscher, Ch., 49–50 declaratory judgments, 67, 79–80, 115 declaratory relief, authority of tribunal, 81–84 defective awards, 231, 252–53, 258 deference to domestic courts, see also non-court of appeals doctrine Feldman case, 67, 79–84, 90, 92–95, 106 Mondev case, 49, 59 Thunderbird case, 262, 270, 274, 280–87 denial of benefits provisions, 197 denial of justice customary international law, 292–93 Feldman case, 83–85, 93 Loewen case, 291–92, 297, 306–07, 311–21 Mondev case, 38–50 Mondev case, 59 Saluka case, 175 deprivation claims, see direct expropriation; expropriation; indirect expropriation dies a quo, 254–55, 300, see also dates dies ad quem, 300–302, 307, see also dates differential treatment, 127, 169, 228, 318, see also discrimination

364 • The Reasons Requirement in International Investment Arbitration diplomatic protection Loewen case, 301–02, 305, 316–17 Petrobart case, 345 Tokios Tokeles case, 191, 202–3, 205–06 direct expropriation, see also expropriation; indirect expropriation Feldman case, 86, 98 LG&E case, 133–34 discounted cash flow (DCF) method, 240 discouragement of lawsuits by respondents, 68, 71 discretion, 16 Feldman case, 95, 107–08 LG&E case, 123 Loewen case, 319 Mondev case, 34, 44 Thunderbird case, 265–66 Tokios Tokeles case, 195–96 Wena case, 245–46, 248, 251 discrimination, see also national treatment audit failure, 104–07 burden of proof, 104, 109–14 definition, 124–25 denial of export registration, 107–08 denial/recouping of rebates, 108–14 and domestic companies in “like circumstances/situations”, 102–03, 108–11, 223–27 ELSI standard, 124–25 Feldman case, 63–64, 66–68, 76, 89, 93, 96–97, 100–114, 127 and intent, 124, 128–29 LG&E case, 117, 119, 123–30 Loewen case, 291–321 Maffezzini case, 228 Mondev case, 38, 49–50, 57–58 non-discrimination principle in customary international law, 124 Occidental case, 126–27, 214, 223–28 Pope & Talbot case, 127, 227–28 Saluka case, 168–69, 170, 172–74, 178

S.D. Myers case, 127–28 Thunderbird case, 267–68, 280 dissenting arbitrators, 5 Feldman case, 67, 100–01 Loewen case, 305 Thunderbird case, 267, 278, 288 Tokios Tokeles case, 192, 199, 201, 209 Wena case, 238 doctrine, 8, 18, 301 Dolzer, R., 60, 136 domestic companies in “like circumstances/situations”, 67, 100–03, 108–11, 127–28, 223–28 domestic courts, 12, see also deference to domestic courts; local remedies; non-court of appeals doctrine confusion of rulings, 82–83 controlling decisions, 82 deference warranted, 79–80, 82–84 examination of compliance with decisions, 83–84 exhaustion of remedies, see local remedies, exhaustion and expropriation claims, 79–81 Feldman case, 67, 78–80, 82–84, 90 finality of rulings, 82, 311–16 Loewen case, 291–321 Mondev case, 33–35 Occidental case, 213, 218–19 ongoing proceedings, 67, 78, 82, 90 Petrobart case, 325–26 and reasons requirement, 12–13 review by arbitral tribunal, 35–63, 291–321 Saluka case, 152, 175, 187, 189 Thunderbird case, 262–63, 270, 284, 291–92 waiving of rights to proceed in, 79–81 domestic law, 7 as facts, 285 Feldman case, 78, 80–81, 83–84 Loewen case, 319 Mondev case, 36, 39

Index • 365 Occidental case, 213 Saluka case, 152 Thunderbird case, 261–64, 268, 273, 279–90 Tokios Tokeles case, 201 double counting, 234, 242–43, 250 Drymer, S.L., 134 dual citizenship, see nationality Dugard, J., 301–02 economic criteria, 196, 199–200, 209 Ecuador, 126, 168, 172, 261 Occidental case, 211–30 Egypt, 231–60 Egyptian Hotels Company (EHC), 232–33, 235 ENARGAS, 120 energy sector, see gas industry; oil industry equitable estoppel, see estoppel errors, 12, 17, 20, 26 Feldman case, 110 LG&E case, 140, 146–47 Mondev case, 43–44 Saluka case, 149, 182 Tokios Tokeles case, 191 Wena case, 249, 252 estoppel Feldman case, 68–78, 85, 114 Loewen case, 299, 306–09 and state practice, 69–71 and time limitation, 68–78 Wena case, 234 evidence Feldman case, 74–75, 95–96, 101–03, 105–12 LG&E case, 124–25 Loewen case, 299, 302, 318–19 Mondev case, 58–59 Saluka case, 173–74 Thunderbird case, 271–72, 274–75, 285–86 Wena case, 242–44, 251, 253–54 exceptional circumstances and time barring, 72–73

excess of power, see manifest excess of powers exemption from liability, 140, 143–44 exhaustion of local remedies, see local remedies, exhaustion expert testimonies, 318–20 export rebates, 65–66, 69, 74, 76–77, 113 expropriation creeping, see indirect expropriation and customary international law, 177–79, 181–82 direct, see direct expropriation domestic court rulings on possession of rights, 79–81, 92–95 Feldman case, see indirect expropriation, Feldman case foreclosure as issue, 217 formal, see formal expropriation as jurisdictional issue, 215–17 LG&E case, 130, 132–35, 137–38 Mondev case, 38–39, 50–51, 54–57 Occidental case, 214–17, 220–21, 229–30 and regulation, 63–115, 211–30 Saluka case, 175–80, 175–85 as taking of property, 99–100 Thunderbird case, 280–87 and transfer of ownership, 94, 99–100 Wena case, 233, 239–40, 257–59 F&ET, see fair and equitable treatment failure to address claims Loewen case, 309–11 Saluka case, 183–85 failures to state reasons, 8–10, 17, see also missing reasons; unreasoned findings CMS case, 24–25 Klöckner case, 5, 9–10 Mitchell case, 13–15 Wena case, 20

366 • The Reasons Requirement in International Investment Arbitration fair and equitable treatment, 29, 31, see also legitimate expectations CMS case, 19–20 Feldman case, 95 Genin case, 166–67 and indirect expropriation, 135–36 and international law, 166–67 LG&E case, 117, 119–23, 129, 135–36, 146–47 Loewen case, 292–93 Mondev case, 33 Mondev case, 166 Occidental case, 214, 220–21 Pope & Talbot case, 166 proximate cause of harm, 171 Saluka case, 149–50, 165–72, 178, 186 and state of necessity, 146–47 Thunderbird case, 272–80 unjust enrichment, 171–72 Wena case, 239 false premises, 64 Federal Trade Commission (FTC) (United States) and minimum standard of treatment, 40–41 finality of rulings, 50, 82, 190, 311–16, see also res judicata findings of fact and non-court of appeals doctrine, 271–72 findings of law and non-court of appeals doctrine, 271–72, 280–87 forced administration, 150, 156–57, 159, 174, 177, 181, 183–85, 188 forgotten claims, see failure to address claims “fork in the road” provisions, 49, 214, 218–19, 222 form obscuring reasoning, 214–15 formal and authorized recognition, 72–74 formal expropriation and indirect expropriation, 53, 57, see also direct expropriation; expropriation; indirect expropriation

Fortier, L.Y., 134 forum shopping, 164 fraud, 294 full deprivation, 280–87 full security and protection LG&E case, 136 Saluka case, 149–51, 165, 174–75, 184, 189 Wena case, 239 funeral homes, 291–321 gambling industry, 261–90 gas industry, 117–48, 326–47 general legislation, circumscription of legal recourse, 49 general principles of law, 7, 16, 70–71, 171, 245, 247–48, 265, 281, see also principles of law general taxation, 89, 92, see also tax measures good faith LG&E case, 121 Mondev case, 52 Saluka case, 164, 167, 169–70, 172 Wena case, 259 government liability, see immunity “gray market” exports and expropriation, 87, 90–91 Guinea, 11–12 Hayward Parcel, 51–52, 55 hotel trade, 231–60 ICSID, see International Center for the Settlement of Investment Disputes (ICSID) immunity, 37, 45–50, 59–61, 262 impairment, see arbitrariness; nonimpairment implicit reasons, 249, 252–55 implied judicial review, see annulment proceedings; non-court of appeals doctrine

Index • 367 inadequacy of reasoning, see also contradiction of reasons; failures to state reasons; incoherent reasoning; inconsistencies; missing reasons; unreasoned findings LG&E case, 130, 133, 143, 147–48 Loewen case, 320 Mondev case, 34 Petrobart case, 346–47 Saluka case, 189 Tokios Tokeles case, 201–10 Wena case, 234–35 incoherent reasoning, 118, 132, 139, 146, see also internal coherence inconsistencies, see also contradiction of reasons; inadequacy of reasoning; internal coherence; unreasoned findings Feldman case, 72, 75, 77, 90 LG&E case, 117–18, 123–30, 137–38, 146–48 indirect deprivation, 176, 184–85 indirect expropriation and business problems, 87–88 CME case, 53 definition, 131–34 and direct expropriation, 133–35 failure to answer question, 87–106 and fair and equitable treatment, 135–36 Feldman case, 63, 67, 77, 85–85, 93, 98–99, 137–38 and formal expropriation, 53 and “gray market” exports, 87, 90–91 LG&E case, 117 LG&E case, 117, 119 LG&E case, 119, 122, 130–39 LG&E case, 130–39 Metalclad case, 137–38 Mondev case, 35, 50–57, 60–61 Occidental case, 214, 217 Saluka case, 179, 184 and tax measures, 86, 98–99

Tecmed case, 132–33 Thunderbird case, 280–81 time barring, 54–57 Indonesia, 118, 197 AMCO case, 10–11 inequitable treatment, see fair and equitable treatment inferred reasons, see implicit reasons intentional interference with contractual relations, 47 intentional tort, 45–47 interest awards, 241, 244–48, 252–55 interference, 22, 36, 46, 55, 99, 130, 132–34, 174 internal coherence, 14, 16, see also contradiction of reasons; inadequacy of reasoning; incoherent reasoning; inconsistencies LG&E case, 117–18, 147 Loewen case, 303, 309–11 Lucchetti case, 17–19 Saluka case, 149–50, 159, 183, 189 Thunderbird case, 272–80, 286 International Center for the Settlement of Investment Disputes (ICSID), 2–4, see also Table of Legal Instruments practice, 4–27 international law, see also customary international law and domestic law, 83–84 international responsibility, 45, 51, 313, 315 international torts, immunity, 45–50 investment, see also expropriation; indirect expropriation amount burden of proof, 241–44 Feldman case, 241 Wena case, 234, 241–44 claims to money as, 335–38 contract as, 334 definition Feldman case, 85–86 Occidental case, 214–17

368 • The Reasons Requirement in International Investment Arbitration investment (continued) Petrobart case, 325–26, 327–28, 330–32, 335, 340–42 Tokios Tokeles case, 199 general characteristics, 342–46 judgment as, 334 right to pursue economic activity as, 338–41 and standing Feldman case, 94, 100, 115 Occidental case, 214–16 Petrobart case, 325–47 Saluka case, 162–63 Thunderbird case, 268, 281–82 investor status economic criteria, 199–200 and standing Petrobart case, 332–33, 344–46 Saluka case, 162–64 Tokios Tokeles case, 191–210 Wena case, 233, 235–36, 243–44 invoices, 66, 76–77, 90, 93, 95–96, 109, 112 IPB, 149–90 Iran, 54–57, 134, 308 irrelevant reasons, 87, 98–100 judicial finality, see finality of rulings judicial review, 212, 253, 261–62, 280, 283, see also annulment proceedings; non-court of appeals doctrine jurisdiction, 18–19, 21 expropriation as issue, 215–17 Feldman case, 67–84, 94 “fork in the road” provisions, 218–19 LG&E case, 118–19 Loewen case, 292–93, 296–309, 312–14 Occidental case, 214–23 ratione personae, 192–96, 201 reasons requirement, 18–19 Saluka case, 149–50, 158–59, 161–65, 186

tax measures, 219–23 Tokios Tokeles case, 191–210 Wena case, 231–39 jurisprudence, 8, 10, 27, see also case law; precedent LG&E case, 127 Loewen case, 306 Mondev case, 42, 60 Occidental case, 218–19 Petrobart case, 332, 335 Thunderbird case, 284 KGM, 326, 328, 333, 339, 341, 344–45 Kyrgyz Republic, 323–47 Lafayette Place Associates (LPA), 35–63 Lauterpacht, E., 48 law of Contracting State, requirement to apply, 6–7, see also domestic law lease agreements, 36, 52–53, 232, 235, 243–45, 256–59, 328 legal purity, 6 legal tests, 172–73, 180, 182–83, 190 legitimacy of awards, 231, 346–47, see also acceptance of rulings legitimate expectations, 23, see also fair and equitable treatment Feldman case, 96 Mondev case, 35, 38, 50 Saluka case, 169 Thunderbird case, 267, 272–80, 282, 287–89 liability of public bodies, see immunity licenses, 10, 21–22 Feldman case, 75 LG&E case, 119–20, 123, 137, 146–47 Saluka case, 154, 156 Wena case, 233, 256 like circumstances/situations, see domestic companies in “like circumstances/situations”

Index • 369 limitations, time, see time barring limited jurisdiction, 26, 259, see also jurisdiction Lithuania, 193–200, 204, 206–07, 304 local courts, see domestic courts local remedies exhaustion Feldman case, 68, 78–81, 114–15 general derogation from customary rule, 79–80 Loewen case, 293–94, 311–12, 317, 321 Mondev case, 37, 39, 47, 50 Feldman case, 81–84, 114 Loewen case, 297, 311–19, 321 Mondev case, 39, 47 and NAFTA Article 1121(2)(b), 79–80 Occidental case, 218–19 reasonable, 316–20 Thunderbird case, 270–71 Tokios Tokeles case, 193 Loughlin, M., 209 LPA, see Lafayette Place Associates (LPA) MacCormick, N., 263 McDougal, M.S., 30 manifest excess of powers, 5, 13, 15, 20, 25–26 manifest injustice, 44–45, 292, 297 market value, 52, 240–44 Mashaw, J.L., 262 measures, Loewen case, 296–99, see also tax measures merits, 14–16, 22 Loewen case, 292–94, 309–21 Mondev case, 38, 49, 51, 58 Occidental case, 214–17, 222–23 reasons requirement, 18–20 Saluka case, 150 Wena case, 231–34, 239–48 Mexico Feldman case, 63–115 Metalclad case, 12–13

Thunderbird case, 261–90 MINE standard, reasons requirement, 13–14, 18, 20 minimal requirements, 20, 249–50, 252 minimum standard of treatment customary international law v international law, 40–41 Feldman case, 67 Loewen case, 292, 313 Mondev case, 33, 44, 46 Thunderbird case, 285 Tokios Tokeles case, 209 miscarriages of justice, 34, 175, 291–93, 297, 321 missing reasons, 231, 252–53, 255, see also failures to state reasons; unreasoned findings mistrial, 295, 309 Morocco, 335–36 most-favored nation treatment, 223–28, see also discrimination; national treatment municipal law, see domestic law naked entities, 303–04 national courts, see domestic courts national law, see domestic law National Property Fund (NPF), 153–54 national security, see state of necessity national treatment, see also discrimination Feldman case, 63, 67, 77, 100–02, 113 LG&E case, 124, 126, 136 Mondev case, 38 Thunderbird case, 280, 282, 291 Wena case, 223–28, 230 nationality Barcelona Traction case, 191, 198–99, 205–06 control test, 194, 196–98, 202 and date of establishment, 206–08 definition, 201, 203 economic criteria, 208–10 LG&E case, 125, 127

370 • The Reasons Requirement in International Investment Arbitration nationality (continued) Loewen case, 294–95, 299–309 Mondev case, 50 Nottebohm case, 191 state of incorporation rule, 202 state of seat rule, 202 Tokios Tokeles case, 191–210, 304–06 Wena case, 235–39 necessary evidence, 106–7, see also evidence necessity, see state of necessity Netherlands, 131, 149–64, 217, 335 neutralization, 55, 131–32 Nomura, 149–90 non-court of appeals doctrine, 261–64, 267–72, 274–75, 280–87, see also annulment proceedings; court of appeal non-discrimination, see discrimination; national treatment non-impairment, see also arbitrariness Occidental case, 214, 223, 228–30 Saluka case, 149–51, 172–75, 184, 189 non-submission of documents, 104–06 non-transparency, 63, 97 obiter dicta AMCO case, 10 Loewen case, 292–93, 296–97, 310–11, 321 obligation to state reasons, see reasons requirement oil industry, 211–30 O’Keefe, J., 294–95, 317 original judicial wrongs, 312, 315 other representations, promises or actions, 71–72 ownership Feldman case, 85, 88, 98–99 LG&E case, 131–32, 138 Loewen case, 297 Saluka case, 155 Tokios Tokeles case, 195, 202, 209 Wena case, 235, 257–58 partial non-court of appeals doctrine, 272, 276–77

participation contracts, 212–13 Paulsson, J., 48–49, 130, 132 period of limitation, see time barring personal claims, 297, 309–11 Peru, 16–19 physical taking of tangible property, 99–100 Poblano Group, 102–14 police powers, 89, 133, 178–79, 181, 268 positive law, 7–8 Poulain, B., 338 praeter legem assurances, 275, 287–89 precedent, 43, 61, 145, 179, 203, 210–11, see also case law; jurisprudence prevention of lawsuits by respondents, 71 prima facie case and burden of proof, 106–07, 110–14, 235 principal factors, 266, 272, 275, 278–79 principe de base, see “basic principle” principles, general, see general principles of law principles of law and rules of law, 6–7 private parties, 29 Loewen case, 296, 298 Mondev case, 37, 43, 46 Petrobart case, 329 Saluka case, 187 Tokios Tokeles case, 191 privatization, 21, 120, 126, 149–50, 153–54, 156, 188 proof norms, see burden of proof proximate cause of harm, 171 public interest, 138, 168, 176–77, 205 public order, 30, 139–40, 146, 177, 268 public policy, 152–53, 204–05, 210, 284, 336 publishing sector, 191–210 punitive damages, 294–95, see also damages put option, Saluka case, 155, 159, 185–87

Index • 371 quality of investment award reasoning, 1–32, see also adequacy of reasoning; contradiction of reasons; failures to state reasons; inadequacy of reasoning; incoherent reasoning; inc onsistencies; internal coherence; missing reasons; successful reasoning; unconvincing argumentation; unreasoned findings ratione temporis issues, see dates reasonable availability of local remedies, see local remedies, reasonable reasonable availability test, 317–20 reasons requirement, 2–32, see also detailed topics receipts issue, 65–66, 68–69, 74, 76–77, 81–85, 92–93, 95–96, 100–02, 113 recognition of claims by respondent and time barring, 72–73 redress in domestic law absence of, 39 regulation and expropriation Feldman case, 63–115 Occidental case, 211–30 Saluka case, 177–81 Thunderbird case, 261–90 and indirect expropriation, 134–35 regulatory authorities, immunity, 45–50 Reisman, W.M., 246 remedies, local, see local remedies representations, 68, 71–72, 137, 168, 184, 288, 308, 345 requirement to state reasons, see reasons requirement res judicata, 192, 216, 262, 327, see also finality of rulings right reasons, 24–25 right to pursue economic activity, as investment, 338–41 rules of law and principles of law, 6–7

rules of treaty interpretation, see treaty interpretation Sabahi, B., 128 Sandifer, D., 242 Scelle, G., 3 seat of management, 304–05 SEGOB, 265–89 shareholders, 21, 24 Feldman case, 98 LG&E case, 125 Loewen case, 296, 304–05 Petrobart case, 329 Saluka case, 156, 170 Tokios Tokeles case, 194, 197–98, 202, 205–07 SHCP, 65–113 shelf companies, 86 “shock the conscience” standard, see manifest injustice Simma, B., 314 skill machines, 264–86 slot machines, 264–87 sole effect doctrine, 54 sources, failure to cite, 69–74 sovereign immunity, see immunity special purpose entities, 154–55 SRI, 127, 213, 218, 224, 229–30 standards, see also legal tests appropriate, 27–29 denial of justice, 40–42 inconsistent, 17 Klöckner case, 5–6, 9–10, 28 manifest excess of power, 20 manifest injustice, 44–45 MINE standard, 13–14, 18, 20 reasons requirement, 4–18, 20 Wena case, 20–21, 23 standing, see investment and standing; investor status and standing stare decisis, see case law; jurisprudence; precedent state of necessity, 25–26 and Articles on State Responsibility, 142–45 CMS case, 25, 145–46 and exemption from liability, 140–45

372 • The Reasons Requirement in International Investment Arbitration state of necessity (continued) and fair and equitable treatment, 146–47 Gabcíkovo-Nagymaros case, 144 and international law, 142–45 LG&E case, 117, 119, 129, 139–48 Mondev case, 47 and national security, 25 state of seat rule, 197, 202 state organs, 60, 72–75, 77 state practice and estoppel, 69–71 statement of reasons, requirement for, see reasons requirement statutes of limitations, see time barring statutory immunity, see immunity subjective tests, 40, 319 subsidiaries Loewen case, 295, 299 Petrobart case, 332, 344–45 Saluka case, 163 Thunderbird case, 264 Tokios Tokeles case, 193 Wena case, 237 substantive law, matters of, 215–16, 297, 312–13, 317–18 successful reasoning, 159–75 sufficiency of reasons, see adequacy of reasoning; quality of investment award reasoning sufficiently pertinent/relevant reasons AMCO case, 11 Klöckner case, 9 Wena case, 250–51 tangible property and expropriation, 99–100 tax measures, see also general taxation EnCana case, 211, 229 Feldman case, 63–115 and indirect expropriation, 86 Occidental case, 211–30 tax rebates, see tax measures time barring, 68–72, 74, 321 agreement not to raise as defence, 71 estoppel, 68–78

exceptions to in NAFTA, 72 indirect expropriation claim, 54–57, 59–61 Mondev case, 59–61 and recognition of claims by respondent, 72–73 and sole effect doctrine, 54–57 and state behavior, 72 time limitations, see time barring tolling, see estoppel tort, 46–47, 49, 262 transfer of ownership and expropriation, 94, 99–100 transparency, lack of, see non-transparency treaty interpretation, 6, 18–19, 167, 298 treaty shopping, 164 Tysoe, Justice, 13 Ukraine, 191–210, 238, 304–05 umbrella clauses, 20–25, 119, 129, 145 unclear facts, 64 unconvincing argumentation, 203, 221, 226, 277, 302, 314, 332 unfair and inequitable treatment, see fair and equitable treatment uniform, consistent and effective behavior, 73–75 unintentional arbitrariness, 230 United Kingdom, 44, 47, 223, 233, 268, 333 United States Loewen case, 291–321 Mondev case, 33–63 unjust enrichment, 171–72 unlawful conduct, 299, 306–07, 316 unlawful expropriation, see direct expropriation; expropriation; indirect expropriation unreasonable treatment, see fair and equitable treatment unreasoned findings, see also failures to state reasons; inadequacy of reasoning; missing reasons Tokios Tokeles case, 201–10

Index • 373 Wena case, 235–48 urban renewal, 35–63 valuation of compensation, see also compensation; damages; interest awards Mitchell case, 13, 15–16 Wena case, 239–44, 251, 254 Van Harten, G., 209 VAT refunds, 211–30 veil piercing, see corporate veil Verdross, A., 314 Wälde, Th. W., 267, 331 Wena standard, reasons requirement, 20–21, 23

The reasons requirement in international investment arbitration: critical case studies

Case Studies in Pediatric Critical Care

Investment Arbitration and the Law of Countermeasures

Arbitration Law in America: A Critical Assessment

Arbitration Law in America: A Critical Assessment

State Liability in Investment Treaty Arbitration: Global Constitutional and Administrative Law in the BIT Generation (Studies in International Law)

International Arbitration: Law and Practice

Due Process in International Commercial Arbitration

The Reasons for Underdevelopment: The Case of Decolonisation in Somaliland

International Competitiveness, Investment and Finance: A Case Study of India (Routledge Studies in Development Economics, 33)

Studies in Critical Philosophy

Investment Promotion Provisions in International Investment Agreements (Unctad Series on International Investment Policies for Development)

International Investment Perspectives

International Investment Perspectives 2004

International Investment Perspectives 2003

Liability Insurance in International Arbitration: The Bermuda Form

Capital Investment for Health: Case Studies from Europe

Case studies in engineering design

Case studies in strategic bombardment

Case Studies in Nurse Anesthesia

Case Studies in Infectious Disease

Case Studies In Sport Communication

Case Studies in Forensic Epidemiology

Case studies in environmental archaeology

Binding Arbitration

Case Studies in Emergency Medicine

Case Studies in Occupational Epidemiology

Case Studies in Ecotourism (Tourism)

Case Studies in Geriatric Medicine

Case Studies in Lipid Management

Case Studies in Neurological Rehabilitation

The reasons requirement in international investment arbitration: critical case studies

Case Studies in Pediatric Critical Care

Investment Arbitration and the Law of Countermeasures

Arbitration Law in America: A Critical Assessment

Arbitration Law in America: A Critical Assessment

State Liability in Investment Treaty Arbitration: Global Constitutional and Administrative Law in the BIT Generation (Studies in International Law)

International Arbitration: Law and Practice

Due Process in International Commercial Arbitration

The Reasons for Underdevelopment: The Case of Decolonisation in Somaliland

International Competitiveness, Investment and Finance: A Case Study of India (Routledge Studies in Development Economics, 33)

Studies in Critical Philosophy

Investment Promotion Provisions in International Investment Agreements (Unctad Series on International Investment Policies for Development)

International Investment Perspectives

International Investment Perspectives 2004

International Investment Perspectives 2003

Liability Insurance in International Arbitration: The Bermuda Form

Capital Investment for Health: Case Studies from Europe

Case studies in engineering design

Case studies in strategic bombardment

Case Studies in Nurse Anesthesia

Case Studies in Infectious Disease

Case Studies In Sport Communication

Case Studies in Forensic Epidemiology

Case studies in environmental archaeology

Binding Arbitration

Case Studies in Emergency Medicine

Case Studies in Occupational Epidemiology

Case Studies in Ecotourism (Tourism)

Case Studies in Geriatric Medicine

Case Studies in Lipid Management

Case Studies in Neurological Rehabilitation

Recommend Documents