International Investment Law: An Analysis of the Major Decisions 9781509929078, 9781509929047, 9781509929061

Written by leading experts in the field, this collection offers a critical and comparative analysis of the existing case

297 66 9MB

English Pages [644] Year 2022

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

International Investment Law: An Analysis of the Major Decisions
 9781509929078, 9781509929047, 9781509929061

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

List of Contributors Henok Asmelash is a Lecturer at the Birmingham Law School. Danae Azaria is an Associate Professor at the Faculty of Laws of the University College London (UCL). George A Bermann is Professor of Law at Columbia Law School and Director of the Center for International Commercial and Investment Arbitration. Andrea K Bjorklund is a Full Professor and the L Yves Fortier Chair in International Arbitration and International Commercial Law at McGill University. Gillian Cahill is a practising Barrister at law. Ivan Čavdarević is a Research Fellow at the Max Planck Institute Luxembourg for Procedural Law. Eric De Brabandere is Professor of International Dispute Settlement Law at Leiden University and Member of the Brussels Bar (Partner with DMDB Law). Diane A Desierto is Professor of Law and Global Affairs, Faculty Director of the LLM in International Human Rights Law, at Notre Dame Law School, with a joint appointment as full Professor at the Keough School of Global Affairs at the University of Notre Dame, where she is also Faculty Fellow in five of the University’s institutes (Klau Center for Civil and Human Rights, Kellogg Institute of International Studies, Liu Institute of Asia and Asian Studies, Pulte Institute of Global Development, Nanovic Institute of European Studies). She is concurrently Professor of International Law and Human Rights at the Philippines Judicial Academy of the Supreme Court of the Philippines, and Chair-Rapporteur of the United Nations Expert Drafting Group on the Right to Development. Edouard Fromageau is a Lecturer in International Economic Law at the University of Aberdeen. Guillermo J Garcia Sanchez is an Associate Professor at Texas A&M University School of Law. Javier Garcia Olmedo is a Postdoctoral Researcher at the University of Luxembourg and a Lecturer, for the Distance Learning LLM in International Dispute Resolution, at Queen Mary University of London. Chiara Giorgetti is Professor of Law at the Richmond Law School. Shotaro Hamamoto is Professor of the law of international organisations at Kyoto University. Dominique Hascher is a Judge on the Supreme Judicial Court of France. Cristina Hoss is a Legal Officer/Secretary of the Court at the International Court of Justice. She holds a PhD from Panthéon-Assas University (Paris II) and was a Legal Adviser at the Iran-US Claims Tribunal from 2015 to 2017. Robert Howse is Professor of International Law at New York University Law School.

x

List of Contributors

Ridhi Kabra is an associate at Three Crowns LLP, London. Ursula Kriebaum is Professor of International Law at the Department of European, International and Comparative Law at the Law Faculty of the University of Vienna. Evelyne Lagrange is a Law Professor at the Sorbonne Law School (University Paris I (Panthéon-Sorbonne)). Frank Latty is a Professor of International Law at Université Paris Nanterre. He is a Member and former director of the Nanterre Center of International Law (CEDIN). He is the President of the French Branch of the International Law Association, and an Arbitrator (member of the Court of Arbitration for Sport) and Conciliator (designated by France to the ICSID Panel (art 13 of the ICSID Convention)). Paul Jean Le Cannu is Team Leader/Legal Counsel at the International Centre for Settlement of Investment Disputes (ICSID) Washington DC, USA. Irmgard Marboe is Professor of International Law at the Department of European, International and Comparative Law at the Law Faculty of the University of Vienna. Eleni Methymaki is a DPhil candidate in Law at the University of Oxford. Makane Moïse Mbengue is Professor of Public International Law at the University of Geneva and Affiliated Professor at Science Po Paris (School of Law). Niall Moran is an Assistant Professor in Law at Dublin City University. Daniel Müller is Avocat à la Cour, Partner at FAR Avocats in Paris, and an Associate Researcher of the Centre de droit international de Nanterre (CEDIN). Joshua Paine is a senior Lecturer in Law at the University of Bristol. August Reinisch is Professor of International and European Law at the University of Vienna and a member of the International Law Commission. Elise Ruggeri Abonnat is legal consultant at the French Ministry for Europe and foreign affairs and PhD candidate at the University of Geneva and the University of Paris Panthéon Assas. Giorgio Sacerdoti is Emeritus Professor of International Law at Bocconi University in Milan, an international arbitrator and former chairman of the Appellate Body of the WTO. Benjamin Samson is a PhD Candidate at University Paris Nanterre. Daniel Sarmiento is a Professor of Administrative and European Law at the Complutense University of Madrid. Stefanie Schacherer is a Post-doctoral Fellow at the National University of Singapore, Centre for International Law. Stephan W Schill is Professor of International and Economic Law and Governance at the University of Amsterdam, Faculty of Law. Christoph Schreuer is a graduate of the Universities of Vienna, Cambridge and Yale. He was professor of law at the Universities of Salzburg and Vienna as well as Johns Hopkins University. He is currently Of Counsel with the law firm Zeiler Floyd Zadkovich in Vienna.

List of Contributors

xi

Alexandre Senegacnik holds a PhD in Law. He is an Affiliated Researcher at Sciences Po Law School (Paris). Marina Sim is a doctoral student at Paris Nanterre University. Bruno Simma is a Professor of Law at the University of Michigan, and a Member of the Iran-United States Claims Tribunal in The Hague (Since December 2012). From 2003–2012 he served as Judge at the International Court of Justice. Edoardo Stoppioni is Professor at the Law Faculty of the University of Strasbourg. Christian J Tams is Professor of International Law at the University of Glasgow and an academic member of Matrix Chambers, London. Luisa F Torres is Legal Counsel at the International Centre for Settlement of Investment Disputes (ICSID) Washington DC, USA. Johannes Tropper is Researcher and Assistant Lecturer at the Department of European, International and Comparative Law at the Law Faculty of the University of Vienna. Güneş Ünüvar is a Senior Research Fellow at the Max Planck Institute Luxembourg for Procedural Law. Alexandra van der Meulen is Avocat à la Cour in Paris, member of the Bar of the State of New York and counsel at Freshfields Bruckhaus Deringer LLP in Paris.

1 Introduction HÉLÈNE RUIZ FABRI AND EDOARDO STOPPIONI

A

CCORDING TO MICHEL Foucault, in order to study a discursive mass that can be described as ‘thought’, ‘il faudra bien un jour essayer de définir les formes et les catégories fondamentales’1 of this type of thinking. This book aims to define the forms and fundamental categories of the legal thinking that have emerged in a particularly active international law field that is now commonly referred to as ‘international investment law’. More specifically, it attempts to provide an overview of international investment law by offering a picture of how different jurisdictions have structured its substance. The chosen method is the analysis of some landmark decisions to provide material that can be examined from a historical and comparative perspective. The volume has therefore brought together around forty international law scholars from many jurisdictions and legal traditions to reconsider the history and contemporary relevance of international investment dispute settlement. A few basic research questions underlie the project. Is it possible today to have a vision of international investment law reduced to investment arbitration, as the vast majority of the literature on this subject tends to show? Can we speak of one international investment adjudication, or are there several dispute settlement visions in this field? What has led us to the current way of reading the phenomenon of investment in international law, and how has it influenced the procedure for settling the related disputes?

I. MAJOR OR LANDMARK DECISIONS

In some civil law countries, legal disciplines with a strong judicial fabric have been studied by collecting ‘grands arrêts’. The etymology of the French ‘grand’ is debated. Some linguists (Lotner and Walter) find its origin in the Latin adjective gravis: this designates a profound object, in our case, the profound meaning of a decision. Other linguists (Georges) go back to the Sanskrit word kra, referring to the action of making or producing. Thus, the ‘grand’ decision would be one crafting a meaningful legal framework. At the crossroads of these two ideas, in their preface to the first edition of the well-known Grands arrêts de

1 ‘at some point we will have to try to define the fundamental forms and categories’. M Foucault, ‘La pensée du dehors’ [1966]) Critique 229, included in Dits et écrits, vol 1, text 38.

2 Hélène Ruiz Fabri and Edoardo Stoppioni la jurisprudence administrative, René Cassin and Marcel Waline defined a grand arrêt as bringing ‘un progrès, une évolution ou un revirement durable de la jurisprudence sur un point important’.2 The identification of landmark decisions is inherent in the common law tradition, in accordance with the rule of precedent. The term ‘landmark decision’ is revealing. The adjective landmark comes from a Proto-Germanic adjective (landamarkō) denoting something used as a point of orientation to locate other structures, or a feature that marks land boundaries. In the sixteenth century, it designated an object visible from the sea, used as a guide for travellers (a marker of the land – Londes-mark). In this sense, a landmark case has a double meaning. On the one hand, it is a decision that draws the boundaries of a particular legal regime or concept. It clarifies how a certain rule, or set of rules, identifies itself or follows a logical sequence. The decision stands out for its mapping, exploratory endeavour. It is the meaning still adopted by the German expression Grundrechtsprechungen, a fundamental legal decision, which sets the stage for reflection. On the other hand, ‘landmark’ refers to a decision that is not only a sound description of the contours of a regime but also an illuminating guide through the nuances of legal uncertainty. Here it stands for a bright and shining assessment of legal content, its clarity or limpidity. Whether it is an in-depth analysis of the law, a creative endeavour, a judicial object that maps a field or a legal beacon for navigating between certain rules, we are dealing with a groundbreaking decision. Displaying great wisdom, Guy Braibant suggested that ‘chacun peut choisir ses grands arrêts: il n ’y a pas de definition certaine et juridique du grand arrêt.’3 While books on case law contain the ‘key cases’,4 or the ‘most influential’5 or those that have the greatest ‘societal impact’,6 the fundamental idea behind this book is to illustrate the diversity of the jurisprudential fabric of international investment law, rather than to isolate or classify the decisions that should necessarily be considered as major ones.

II. REASSESSING THE DIVERSITY OF INTERNATIONAL INVESTMENT JUDICIAL LAW-MAKING

It has been argued that ‘most of the literature today represents mainstream international investment law, which focuses on how the law is applied in investment dispute settlement and which is the product of specific professional perspectives and biases’.7 Much to the contrary, the purpose of this volume is to demonstrate that we need to broaden our vision of international investment law. This seems essential if one is to understand that the forces driving the

2 ‘a progress, an evolution or a lasting reversal of jurisprudence on an important point’. R Cassin and M Waline, ‘Préface de la première édition’ in M Long and others (eds), Les grands arrêts de la jurisprudence administrative (Paris, Dalloz, 2020) vii. 3 ‘each person can choose his or her great judgments: there is no certain legal definition of a great judgment’. G Braibant, ‘Qu’est-ce qu’un grand arrêt?’ (2006) 26 Actualité juridique, droit administratif 1428. 4 JR Vile, Essential Supreme Court decisions: summaries of leading cases in US constitutional law (Lanham, Rowman & Littlefield Publishers, 2010) xxii. 5 GR Hartman, RM Mersky and CL Tate, Landmark Supreme Court cases: the most influential decisions of the Supreme Court of the United States (New York, Infobase Publishing, 2014). 6 DE Lively and DS Broyles, Contemporary Supreme Court Cases: Landmark Decisions since Roe v Wade: Landmark Decisions since Roe v Wade, 2nd edn (Santa Barbara, ABC-CLIO, 2016) xi. 7 SW Schill, ‘W (h) ither fragmentation? On the Literature and Sociology of International Investment Law’ (2011) 22(3) European Journal of International Law 875, 877.

Introduction 3 field are based on a specific view of the role of international law and that its understanding varies continuously, depending on the context in which investment law is applied. Ubi maior minor cessat: the book has sought to select some significant cases that deserve to be considered as entry points for particular topics, as fundamental starting points for tracing the lines of intellectual construction of a field. Our choices cast a shadow on other decisions, which is inevitable. The choices have attempted to make visible the key judicial moments in the construction of international investment law: through selection and renvois, we have tried to sketch a framework for illuminating the complexity of international investment law lato sensu, its ramifications and reversals. In so doing, the book has embraced different methodological dimensions. The first dimension is historical. This volume generally shows that international investment law has a longer history than is generally considered and that this history is fundamental to understanding its development. However, one of the most glaring features of current international economic law doctrine is the tendency to avoid considering the impact of history on the development and formation of the discipline and to focus on the technical functioning of its instruments, without considering their effect on individuals. This book aims to tackle this lack of self-reflection regarding international investment law, which represses its historical origins and impact assessment.8 Fundamental tendencies can be drawn from the first part, which places international investment adjudication in a historical context. This history is longer than is generally admitted: it includes a prehistory, followed by a significant initial shift from contract-based litigation to the internationalisation of litigation, and then a second shift that has given rise to various movements. Internally in arbitration, the construction of sagas has pushed the system to evolve and develop (be it the early North American Free Trade Agreement (NAFTA) cases, the Argentinian crisis, the Russian cases or the recent Spanish renewable energy episode) and to move from litigation traditionally dominated by expropriation to the pre-eminence of fair and equitable treatment cases in the most recent practice. Externally, the initial move towards judicialisation provoked an intense response from courts and tribunals that are not strictly investment-oriented, which has led to reflection on alternative forms of dispute settlement. The second dimension of the work is comparative. This volume takes the fragmentation of international law seriously. It shows that international investment law is not only made within the framework of investment arbitration and at the International Court of Justice (ICJ) but also by other international courts and tribunals, all of which contribute to the development of the content of international investment law. The first type of comparison is endogenous and aims to show the complexity of the existing case law in international investment arbitration, where different tribunals have opted for opposed solutions concerning the same issues. Indeed, the flourishing of dissenting opinions shows the substantial diversity of views that inhabit the field. Very fundamental and structural issues are still the subject of vigorous debate among arbitrators and scholars: examples of this include the concept of investment (torn between the Phoenix approach or the more restrictive Saba Fakes analysis), the scope of state consent

8 See R Lima Sakr, ‘Beyond History and Boundaries: Rethinking the Past in the Present of International Economic Law’ (2018) 9 LSE Legal Studies Working Paper. The author shows how history and boundaries interact to produce a ‘traditional’ view of international economic law. This interaction plays a central role in structuring how international lawyers assert the authority and legitimacy of IEL in global economic governance. He then argues that the commitments of the traditional approach to Anglocentrism and Modernism limit lawyers’ ability to understand and solve present-day issues since it produces lessons only in support of the dominant programmes, norms and ideas under contestation. Consequently, it constrains, rather than empowers, lawyers’ imagination.

4 Hélène Ruiz Fabri and Edoardo Stoppioni in investment arbitration (the Most-Favoured-Nation clauses saga shows how public international lawyers and commercial arbitrators have interpreted the question in opposite ways, as expressed in the Maffezzini vs Plama debate), the content of fair and equitable treatment (as evidenced by the dissents in the Spanish saga). The second type of comparison is exogenous and relates to the diversity of perspectives on international investment adopted by different courts and tribunals, whether they are international courts or tribunals other than investment tribunals (human rights courts, international administrative tribunals or regional economic integration courts) or national judges. To give but one example among others, the role of human rights can be seen as paradigmatic. The opening of investment arbitration to exogenous sources of international law has met with some reluctance to incorporate these axiologically different norms; thus, investment arbitration has been accused of being blind to non-economic issues. This criticism has led to some developments with different attempts at integration. The Saur case is a clear example, where arbitrators state that human rights in general, and the right to water in particular, constitute one of the various sources of law that the tribunal shall have to take into consideration in order to solve that dispute, as these rights enjoy a constitutional status in the Argentine legal system and, what is more, they form part of the general principles of international law.9

The arbitral tribunal’s task is to ‘counterbalance’ investor protection and human rights ‘while analysing Saur’s substantive allegations’.10 Another example is the ICSID award in the Urbaser case, where the tribunal laid the foundations for investor responsibility in international law by allowing counterclaims. It identified an obligation on private parties not to engage in activity aimed at destroying the enjoyment of human rights by examining human rights documents – including the Universal Declaration of Human Rights, the International Covenant on Economic, Social and Cultural Rights, the Ruggie Guiding Principles on Business and Human Rights, and UN General Assembly resolutions.11 On the other hand, human rights courts have sometimes analysed investment issues from very different perspectives. The Inter-American Court of Human Rights has taken a particular philosophical approach regarding the relations between international investment law instruments and the Inter-American Convention: the Court has not been furnished with the aforementioned treaty between Germany and Paraguay, but, according to the State, said convention allows for capital investments made by a contracting party to be condemned or nationalized for a ‘public purpose or interest’, which could justify land restitution to indigenous people. Moreover, the Court considers that the enforcement of bilateral commercial treaties negates vindication of non-compliance with state obligations under the American Convention; on the contrary, their enforcement should always be compatible with the American Convention, which is a multilateral treaty on human rights that stands in a class of its own and that generates rights for individual human beings and does not depend entirely on reciprocity among States.12

9 Saur International SA v the Republic of Argentina, decision on liability (6 June 2012) ICSID case No ARB/04/4, §330 (our translation). 10 ibid §332 (our translation). 11 Urbaser SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v The Argentine Republic, Award (8 December 2016) ICSID Case No ARB/07/26, para 1195. 12 IACtHR, Sawhoyamaxa v Paraguay, Decision, 29 March 2006 (merits, reparation and costs), IACtHR Series C No 146, §140.

Introduction 5 The Convention is not a classic instrument of international law based on reciprocity and prevails over BITs in the inter-American legal order. Similarly, the African system for the protection of human and people’s rights has denounced Nigeria’s action in favour of Shell’s alleged pollution of the forests where the Ogoni population traditionally lived13 or Kenya’s behaviour in supporting multinational companies’ investments by expelling the Ogiek population from their territories without previously consulting them.14 This diversity presents a broader type of international investment law than what is generally stated. Such an approach also forces us to embrace the complexity of a subject that goes well beyond the boundaries of arbitration. This burgeoning case law will have to be closely observed and will be the object of more specific future research. Third, the book also takes the role of national judges seriously. Breaking with the notion that the national judge cannot be the natural forum for international investment disputes – a widespread idea in investment arbitration – the book brings together in one volume landmark decisions from different national courts around the world that have dealt with Bilateral Investment Treaties or international investment protection instruments. The well-known paragraph 292 of the Court of Justice’s Opinion 2/15 made it clear that national jurisdictions are the natural forum (juge naturel) for legal issues related to international investments.15 This stance provides an incentive to examine and understand the different approaches of local courts to investment instruments. Their diversity is impossible to crystallise in a single volume. However, the panorama drawn from selected examples of jurisdictions around the world shows how recurring topics (such as constitutional challenges and state immunity) can trigger very diverse responses.

III. BOOK STRUCTURE

Part 1 presents the history and background of the various dispute settlement mechanisms that have shaped modern international investment law. Although some of these cases are little known and studied, their inclusion is in itself an important asset. Moreover, subjecting them to critical assessment allows for a better understanding of the multifaceted origins of the discipline. Part 2 deals with the fundamental standards, clauses and concepts on which investment arbitration is based. Starting from the most well-known and/or seminal cases that have addressed a topic, the authors go beyond commenting on awards to show the judicial strategies adopted to move litigation forward. Part 3 provides an overview of how courts and tribunals other than arbitral tribunals have dealt with international investment issues. The main objective is to show how investment arbitration has adopted an axiological reading of the discipline. Such an axiological vision may

13 African Commission of Human and People’s Rights, Social and Economic Rights Action Center for Economic and Social Rights vs Nigeria, 2001. 14 African Court of Human and People’s Rights, Commission v Kenya, 26 May 2017, application 006/2012. 15 Opinion 2/15 Accord de libre-échange avec Singapour [2017] ECLI:EU:C:2017:376, para 292: ‘such a regime, which removes disputes from the jurisdiction of the courts of the Member States, cannot be of a purely ancillary nature within the meaning of the case law recalled in paragraph 276 of this opinion and cannot, therefore, be established without the Member States’ consent.’

6 Hélène Ruiz Fabri and Edoardo Stoppioni conflict with other areas, such as human rights, the EU legal order, or the specific needs of other dispute settlement mechanisms. Part 4 provides an overview of how different national legal systems have dealt with international investment law. The questions range from the invocability of a BIT in the national legal order to the constitutionality of a BIT, the impact of sovereign immunity, and the role of the national judge.

2 Landmark Decisions for a Pre-History of International Investment Law EDOARDO STOPPIONI*

T

HIS CHAPTER AIMS at dealing with the ante litteram forms of international investment protection before investment arbitration appeared. The breadth of the analysis is highly selective, as it aims at showing the macro-tendencies that emerged in the history of investment law between Ancient Greece and the beginning of the twentieth century. Focusing on some topical cases, generally investigated by historians and unknown to lawyers, it tries to show that the history of international investment law should be seen as starting much earlier than conventional dates suggest: be that in 1959 when the first Bilateral Investment Treaty (BIT) was signed, or 1999 when the award in the International Centre for Settlement of Investment Disputes (ICSID) case AAPL was delivered.1 Going back in time allows one to adopt an unusual perspective. There is indeed an ongoing debate on how to write about the history of international investment law. While mainstream scholars defend the need to focus on a Dogmengeschichte , on giving account of the evolution of the articulation of the sources of investment law through time, 2 critical scholars have underlined the importance of reflecting on the colonial roots of the discipline and on their impact on the functioning of the current system. 3 This chapter tries to combine both approaches in order to broaden our horizons regarding the development of an international investment regulation and dispute settlement. It seeks to provide an impressionistic sense of the milestones along the road to a corpus of international investment case law, while providing a critical assessment of their functioning.

* Edoardo Stoppioni is Professor of Public Law at the Law Faculty of the University of Strasbourg. 1 See E Lagrange, ‘SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?’, ch 6 in this volume. 2 J Kammerhofer, ‘The Challenges of History in International Investment Law: A View from Legal Theory’ in SW Schill, CJ Tams and R Hofmann (eds), International Investment Law and History (Cheltenham, Edward Elgar, 2018) 164. 3 K Miles, The origins of international investment law: empire, environment and the safeguarding of capital (Cambridge, Cambridge University Press, 2013); D Schneiderman, ‘The Coloniality of Investment Law’ (November 2019) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3392034.

10 Edoardo Stoppioni I. INTERNATIONAL INVESTMENT CASES IN ANCIENT GREECE

Taking the problem of Eurocentrism in the historiography of international law seriously4 makes it necessary to recall that the vast majority of existing literature locates the starting point of ‘international arbitration’ in Ancient Greece. Nevertheless, epigraphy has shown that this form of dispute settlement was already quite widespread in Egypt and Mesopotamia in 3000 BC, where it was used in the mutual relations between kingdoms or cities.5 Similarly, the Hittites believed in a customary rule prohibiting war among the kings of the same Empire and this practice was probably followed in Asia Minor.6 Two fundamental factors merit putting greater emphasis on Greece. The first is pragmatic in nature: it is the greater accessibility to archaeological sources relating to this period. The second relates to what Max Huber underlined as being striking in the Greek experience: whereas the mosaic of sovereign entities inhabiting the same small part of the continent might have led to anarchy, these cities built similarity into their legal systems, strove for mutual facilitation and for trade exchanges, and therefore contributed to establishing the idea of international legal order in a way that would not be replicated until the twelfth century.7 What Max Huber underlines while speaking of Greek cities’ legal systems tending towards convergence is a progressive phenomenon of κοινός νόμος τῶν Ἐλλήνων (a law common to the Greeks), a common legal tradition which is as striking to lawyers as the κοινὴ διάλεκτος is to linguists. Even more interesting for international lawyers is the fact that historians defended the viewpoint that it is the law of aliens which allowed the progressive emergence of an intermunicipal law,8 just as the practice of the minimum standard of the treatment of aliens would be at the inception of a development of the law on international responsibility. The protection of aliens was a fundamental issue in Greek law. In his Ways and Means (355 BC), Xenophon discussed the advantages of Athens as a centre of commercial enterprise and clearly stated that: supposing prizes were offered to the magistrates in charge of the market for equitable and speedy settlements of points in dispute to enable anyone so wishing to proceed on his voyage without hindrance, the result would be that far more traders would trade with us and with greater satisfaction.

Rapidly, international law was used as a means to protect aliens’ property rights. As part of this protection, access to tribunals for foreign merchants became a fundamental issue. A particular type of treaties, concluded from the fifth century, judicialised the relationship between the foreign merchant and the Host city: they are known as symbola or symbolai treaties.9 A typical example is the treaty between Chaleïon and Oiantheïa of 286 BC, the marble of which is kept at the British Museum.

4 M Koskenniemi, ‘Histories of international law: significance and problems for a critical view’ (2013) 27 Temple International and Comparative Law Journal 215. 5 MN Tod, International arbitration among the Greeks (Oxford, Clarendon Press, 1913) 170. 6 V Korosec, Hethitische Staatsverträge (Leipzig, Leipziger Bechtswissenschaftliche Studien, Heft 60, 1931). 7 M Huber, ‘Die soziologischen Grundlagen des Völkerrechts’ (1910) Archiv für Rechts- und Wirtschaftsphilosophie 21, 26: ‘Gleichartige Kultur und gleichartiges Recht, die leichten Seeverbindungen und die Tendenz der Seestaaten zur Anknüpfung von Handelsverbindungen brachten die internationale Rechtsordnung auf eine Stufe, welche (…) von den westeuropäischen Völkern frühestens wieder im XII. Jahrhundert an Italien und im Norden im XIV. oder XV. Jahrhundert erreicht wurde’. 8 M De Taube, ‘Les origines de l’arbitrage international. Antiquité et Moyen âge’ (1932) 42 Recueil des Cours de l’Académie de Droit International 31. 9 For an analysis of intermunicipal treaties in this domain, see FM Hitzig and F Regelsberger, Altgriechische Staatsverträge über Rechtshilfe (Zurich, Orell Füssli, 1907).

Landmark Decisions for a Pre-History of International Investment Law

11

These treaties were qualified in German literature as Rechtshilfvertäge, as they opened the tribunals of the city to foreign claimants to ensure a rapid and equitable administration of justice. In principle, foreign merchants had no access to local courts, but several mechanisms started to provide various dispute settlement possibilities. Historiographies generally depict an evolution. In the archaic era, the foreigner had no procedural rights. Progressively some non-judicial institutions were created, which allowed the representation of a foreigner by entitled consuls (proxenia). Then by way of treaties, some cities allowed mutual access of their nationals to local courts. In a late phase and only in Athens, foreigners were able to bring their commercial claims before local courts, irrespective of any treaty protection (according to the system of dikai emporikai).10 The practice concerning foreigners in local courts evolved along ‘private international law’ lines of reasoning.11 As far as inter-city arbitrations are concerned, historians have shown that the vast majority of these dealt with boundary disputes and that litigation on ‘investment’related situations was rare, save for one notable exception:12 the Loan Arbitration opposing Kos v Kalymna (300-286 BC). Around 360 BC, two bankers (Pausimachos and Hippokrates), citizens of Kos, had lent the state of Kalymna some money. The descendants of the original creditors claimed that the debt was outstanding whereas the Kalymnians claimed that the debt had been settled. Because of the significant amount of the debt, the state of nationality of the claimants took up the cause of the injured bankers. An arbitration was established, with the state of Knidos acting as an arbitrator. It is disputed among historians whether the arbitration was established on a voluntary ad hoc basis or following a compulsory clause in a symbolon treaty. The Knidians established a tribunal composed of 204 individuals. The arbitration started with a round of written memorials (grafe and antigrafe) before the Tribunal heard witnesses, who were cross-examined. Thereafter, the Knidians called for a vote which found in favour of the defendant. No reasons were given in the decision. This can be considered a landmark case in that it shows ‘how a dispute involving private citizens of a state could escalate into or be adopted as a public concern’ in a financial matter.13 It is important to clarify why the Roman period was not included in this analysis. The general assumption in the historiography of international adjudication is that Rome did not develop international arbitration because of its aspiration to world domination: as there was no other legal subject considered as an equal, no international dispute could be solved by law.14 As far as the Roman political vision was concerned, peace was not achieved through law or the legal settlement of disputes, but through domination. The political posture of Rome aimed to make ‘international law’ meaningless, as its objective was the formation of a single state exerting dominion over the whole world.15

10 For an excellent analysis, see P Gautier, Symbola – Les étrangers et la justice dans les cités grecques (Nancy, Annales de l’Est, 1972). 11 P Vinogradoff, Historical types of international law (Leyden, Bibliotheca Visseriana, 1923). 12 SH Ager, Inter-State Arbitration in the Greek World (Berkeley, University of California Press, 1997) 76–83: ‘evidence for international arbitral settlements of conflicts over debt, or other primarily financial matters, is actually relatively scarce, especially when we compare it with the overwhelming evidence for such issues as boundary disputes’. 13 ibid 82. 14 N Politis, La justice internationale (Paris, Hachette, 1924) 26–27: ‘se considérant comme arbitre du monde, elle acceptait d’être juge, non justiciable’. 15 De Taube, above (n 8) 50.

12 Edoardo Stoppioni However, as private arbitration was extremely widespread in Rome, lawyers in the Middle Ages used this experience and legal practice to develop a theory of arbitration.16 We will therefore focus on this medieval period.

II. INTERNATIONAL INVESTMENT CASES DURING THE MIDDLE AGES

Historiography has long dispelled two opposite chimeras regarding the role of arbitration during the Middle Ages. The first relied on its supposed marginal role (arbitrations remained ‘essais accidentels’ or ‘curiosités historiques’17); the second magnified its fundamental role in Roman law which explained its survival through the dark ages of barbarian kingdoms.18 Three different layers contributed to the emergence of robust arbitration practice at that time: (1) the important role of ecclesiastic arbitration for the Catholic Church to remove the power of deciding disputes from secular power; (2) the arbitral practice of Italian cities in the north,19 Swiss cantons20 and German principalities;21 and (3) the suitability of arbitration for feudal power structures. Functional equivalents to what we consider today as international investment dispute settlement mechanisms were present in two different and opposite tendencies. The first is that inter-state arbitration had a strong private component. The second is that private arbitration had an important impact on the international legal field, notably because of its extraterritorial repercussions.22 In both cases, it would be artificial to speak of the existence of a foggy ‘public/private law divide’. Nevertheless, historical works point out that medieval arbitration revealed great confusion between the public and the private spheres.23 Concerning inter-state arbitrations, the private factor behind the disputes has various declinations. For instance, as early as 1147 there are examples of treaties establishing arbitration mechanisms to decide on war reparations. But what is striking is that, in most arbitrations between cities, cantons or communalities (‘half sovereign entities’), private interests that a government espoused were at the core of the disputes. We can speculate that arbitration was conceived as a sort of insubordination, to escape the judicial system of the Empire. As far as purely private arbitration is concerned, a distinctive feature of this period is the desire of merchants to develop alternative dispute settlement mechanisms within guilds, trade fairs or other sorts of commercial organisations.24 The jurisdictional function of consuls emerged in this context. These consuls were appointed to ensure three services for their co-national merchants: information, protection and adjudication.25 In their exercise of adjudication, consuls began to apply customary maritime rules, which had been progressively codified in various instruments (most notably the Rôle d’Oléron), and initiated a practice of 16 ibid

55. Dreyfus, L’arbitrage international (Paris, Calmann Lévy, 1894) 26. 18 M Revon, L’arbitrage international: son passé, son présent, son avenir (Paris, Librairie nouvelle de droit et de jurisprudence, 1892) 55. 19 P Frey, Das öffentlich-rechtliche Schiedsgericht in Oberitalien im XII. und XIII. Jahrhundert (Lucerne, E Haag, 1928). 20 E Usteri, Das öffentlich-rechtliche Schiedsgericht in der schweizerischen Eidgenossenschaft des Xlll und XV Jahrhunderts (Zurich, Orell Füssli, 1925). 21 M Novacovitch, Les compromis et les arbitrages internationaux du XII au XV siècle (Paris, Pedone, 1908). 22 De Taube, above (n 8) 59. 23 A Pierantoni, Gli arbitrati internazionali ed il Trattato di Washington (Naples, Fratelli de Angelis, 1872). 24 G Born, International Commercial Arbitration (Aalphen aan Rijn, Kluwer, 2009) 27. 25 A Bartolomei, ‘Débats historiographiques et enjeux scientifiques autour de l’utilité commerciale des consuls’ (2016) 93 Cahiers de la Méditerranée 49. 17 F

Landmark Decisions for a Pre-History of International Investment Law

13

judicial dialogue whereby consular decisions in Catalonia, Provence in France and Italian maritime cities tended to converge.26 Two cases, relating to what we could consider today as ‘investment’ cases, substantiate these fundamental trends. The first is the fundamentally private nature of the disputes involving cities or commonalities. The Nau dels Barquers case before the Consulat del Mar of Barcelona (1417–1422) is a particularly interesting example.27 The Kingdom of Aragon and Genoa were at the time in open conflict for military and economic predominance in the Mediterranean, having provoked several wars and the declaration of commercial cross-countermeasures. In 1416, in the framework of a policy of countermeasures, Pietro Re (a Genoese merchant and consul of the Catalans in Famagusta, Cyprus – a Genoese enclave) seized the cargo of the Nau dels Barquers, a Catalan ship belonging to Bartomeu Amar and Joan Gros. As a condition to signing any peace treaty, the Catalans insisted on the reparation of the Nau dels Barquers case. The 1417 Peace Treaty was therefore signed with a clause of reparation for the damages caused to Amar and Gros’ ship. The determination of the reparation was entrusted to the Consulate of the Sea in Barcelona. The proceedings started on 1 July 1417 and ended only in 1422. The Consulat del Mar invited all those who had suffered any damage to register their complaints and provide the necessary evidence at the Consulate. The task of the eight consuls appointed (prohoms) was to share equitably among the victims the amount of money that Genoa had surrendered in conformity with the Treaty of Peace. Their task was to compare the evidence of the 62 victims with the manifest (the register of the ship) and the albarans (contracts and declarations related to the cargo). The eight consuls were then reduced to a commission of three arbitrators who had to decide whether to act per viam juris or per amicabilem compositionem. Their decisions could have been appealed before the appellate mechanism of the Consulate or even before the King’s judicial system. The payment of the award had to be made via the city’s bank system (taula de canvi). Whilst history records that of the 8,000 florins, 4,250 were taken as procedural costs, the historical data concerning the end of the story are less clear. The second case includes a clear statement of the reasons why merchants wanted to have recourse to some form of arbitration and to consular jurisdiction, rather than to local courts. The case opposing Honoré de Cuges and Joseph Vaccon of Pisa in 1615 is an excellent example thereof.28 De Cuges introduced a ‘procedural supplication’ to the Grand Duke of Tuscany to allow him to bring before the French Consul in Livorno, and not before local courts, his claims concerning a company that he had established together with his son-in-law Vaccon to trade goods in Egypt’s harbours. This request relied on the fact that the adjudicator needed to have a good command of the French language as the accounting ledgers were in French and both parties were French nationals.29 Moreover, absolute discretion was needed as the situation was 26 A Xerri Salamone, ‘Consuetudini e tradizione nella formazione del diritto marittimo uniforme’ (2009) 108 Trasporti: Diritto Economia Politica 29, 36–37. 27 See the archival work by E Maccioni, ‘Una rappresaglia contro mercanti genovesi gestita dal consolato del mare di Barcellona’ in E Maccioni and S Tognetti (eds), Tribunali di mercanti e giustizia mercantile nel tardo medioevo (Florence, Leo S Olschki, 2016). 28 See the archival work of G Calafat, ‘La juridiction des consuls français en Méditerranée: litiges marchands, arbitrages et circulations des procès (Livourne et Tunis au XVIIe siècle)’ in A Bartolomei, G Calafat and M Grenet (eds), De l’utilité commerciale des consuls. L’institution consulaire et les marchands dans le monde méditerranéen (XVIIe-XXe siècle) (Rome, École française de Rome, 2017). 29 Archivio di Stato di Pisa (ASP), Consoli del Mare, ‘Suppliche’, 971, n° 194: ‘Il Mag.co Honorato de Cuges, franzese et esponendo disse et dice come non havendo possuto saldare li conti con Giuseppe Vaccone, suo genero pur franzese, supplicò à SAS che li facessi gratia tal conti fussino saldati dal Consolo dei Franzesi con tre, o quattro altri mercanti franzesi, massime che essi Cuges et Vaccone, oltre all’essere paesani, erono ancor’ parenti’.

14 Edoardo Stoppioni sensitive because the parties were relatives. The Grand Duke rejected this supplication and designated the Consulate of Pisa as competent to adjudicate the significant miscalculations in the accounts. The case demonstrated that merchants considered foreign consuls as possible arbitrators for technical disputes. Historians showed that consuls were particularly active in this regard in Tuscany and in the harbours of the Ottoman Empire but that local courts mainly allowed this kind of supplication for small claims or less problematic disputes.30 Many lessons can be learnt from this case. The requests substantiate the main reasons why merchants started to favour arbitration over local dispute settlement mechanisms of the host state: rapidity, particular expertise or linguistic skills, secrecy. Mistrust of national courts seems to have already been an important factor in the reasoning of medieval merchants.

III. INTERNATIONAL INVESTMENT CASES IN MODERN TIMES BEFORE INVESTMENT ARBITRATION

The period between the eighteenth and the twentieth centuries witnessed the emergence of a new dispute settlement phenomenon: the so-called ‘mixed commissions’.31 The starting point conventionally taken to study this phenomenon is the Jay Treaty of 1794, concluded between the US and Great Britain, which established three mixed commissions to adjudicate the disputes arising between the two parties. Following this example, many mixed commissions were founded, mainly between a European state and a Latin American state but later also between European states. Interestingly enough, Asian and African states did not follow this path. The burgeoning of mixed commissions can be traced back to some notable historical facts that gave rise to situations requiring different sorts of dispute settlement: the wars for US independence, the Napoleonic wars (1803–1815), the First and Second French interventions in Mexico (1838 and 1868), Chile’s war with Peru and Bolivia, internal conflicts in Peru at the beginning of the twentieth century, the First World War and the Mexican revolution in the 1920s.32 Because of the diverse nature of the disputes arising from these events, the notion of ‘mixed commissions’ is extremely broad, encompassing very different kinds of experiences. One possible taxonomy consists in distinguishing mixed boundaries commissions, mixed diplomatic commissions (relating to fact-finding or conciliation) and mixed arbitral commissions or mixed claims commission (‘settling individual disputes on financial matters between nationals of different States’33). As far as the contribution to the development of international investment law is concerned, mixed claims commissions are the focus of this analysis. Indeed, they dealt with state responsibility for injuries caused to aliens. More generally, their practice contributed to the reflection on the emergence of state responsibility, as the works of Francisco V. García Amador, the first ILC special rapporteur on the topic, showed.34 Indeed, he considered that after the experience

30 G Calafat, ‘Expertises et tribunaux de commerce. Procédures et réputation à Livourne au XVIIIe siècle’ (2011) 14 Hypothèses 141. 31 L Boisson de Chazournes and D Campanelli, ‘Mixed commissions’ in R Wolfrum’ (ed), Max Planck Encyclopedia of Public International Law (Oxford, Oxford University Press, 2008) [hereinafter: MPEPIL] 2. 32 R Dolzer, ‘Mixed claims commissions’ in MPEPIL, 7. 33 Boisson de Chazournes and Campanelli, above (n 31) 1 and 8–14. 34 D Müller, ‘The Work of García Amador on State Responsibility for Injury Caused to Aliens’ in J Crawford, A Pellet and S Olleson (eds), The Law of State Responsibility (Oxford, Oxford University Press, 2010).

Landmark Decisions for a Pre-History of International Investment Law

15

of mixed commissions, state responsibility for injuries to aliens was mature enough to start theorising state responsibility on this basis, even though his intention of codifying primary obligations arising from their practice subsequently failed.35 A critical remark is necessary from the very outset. The body of literature, and often of cases, stemming from this practice, is riddled with strongly colonial discourse, which needs to be deconstructed. Indeed, this discourse is normalised in the doctrinal works of scholars like Jessup who long studied this segment of the history of international law. To him, it was perfectly natural to find the rationality for the emergence of this branch of international law in the ‘inevitable’ need of expansion of developed states: states which had achieved a large measure of local industrial and financial development should seek outlets for the investment of surplus funds and for the energies of their ranchers, bankers, mining engineers, railroad builders, constructors of ports, and other trained personnel. In many instances those energies found an outlet in colonies.36

Jessup is not even cynical when he uncritically remarked that the works of García Amador showed how mixed commissions were the results of asymmetric relations between contracting states, opposing a strong and a weak party.37 In that regard, he considered it totally normal and necessary that ‘history of the development of the international law on the responsibility of states for injuries to aliens is thus an aspect of the history of imperialism’.38 When modern scholars are in denial concerning the colonial origins of international investment law, they would be wise to revert to those pages. Studying the legal connection of investment strategies with these outlets requires bearing in mind the fact that western states viewed a myriad of host states as being ‘open for business’, which makes it ‘impossible to understand mixed claims commissions outside the peculiar history of imperialism’.39 For this reason, we shall not try in the limited space of this chapter to explain what the leading cases were within this abundant practice. We shall not try to argue if the Alabama case (1872) is really the starting point of the reflection on a certain philosophy of reparation or if the Lusitania case (1923) is really the milestone in the reflection on the role of the basic treaty in the settlement of the dispute. Instead we shall select a few cases highlighting some of the fundamental features of this dispute settlement mechanism. At least four fundamental features need to be emphasised, both from a procedural and substantive perspective. From a procedural point of view, the first characteristic feature is the reason that an international dispute settlement mechanism exists: mistrust of local courts. A very clear example of this trend is the Don Pacifico case, in relation to which Lord Palmerston clearly stated that: there may be cases in which no confidence can be placed in the tribunals, those tribunals being, from their composition and nature, not of a character to inspire any hope of obtaining justice from them. It has been said: ‘We do not apply this rule to countries whose governments are arbitrary or despotic, because there the tribunals are under the control of the government, and justice cannot be had’.40 35 FV García Amador, ‘La Responsabilité de l’État (premier rapport)’ (1956) II Annuaire CDI, A/CN.4/96, 173 notably 221. 36 PC Jessup, ‘Responsibility of States for Injuries to Individuals’ (1946) 46 Columbia Law Review 903, 905. 37 PC Jessup, A Modern Law of Nations (New York, MacMillan, 1958) 96. 38 Jessup, ‘Responsibility of States’, above (n 36) 906. 39 F Mégret, ‘Mixed Claim Commissions and the Once Centrality of the Protection of Aliens’ in I de la Rasilla and JE Viñuales (eds), Experiments in International Adjudication: Historical Accounts (Cambridge, Cambridge University Press, 2019) 128. 40 HJ Temple Palmerston, Speech in the House of Commons, on the 25h of June, 1850, on Mr. Roebuck’s Motion on the Foreign Policy of the Government (J Ollivier, 1850) 7, quoted in Mégret, above (n 39) 131.

16 Edoardo Stoppioni This belief in the ‘bad quality’ of national courts of the host state is still present in the ongoing discourse supporting the need of investment arbitration. It is sometimes diluted within the idea that national judges cannot be impartial and that international mixed arbitration is a more neutral forum for disputes. This dogma needs to be challenged, for example by reverting to some of Carlos Calvo’s ideas in that respect. That an investor wishing to enjoy the economic privileges offered by the host state also accepts the respect for local justice as being part of the deal, could be considered as a necessary corollary of the respect of sovereignty. Moreover, the asymmetric privileges conceded to foreign investors compared to national ones are structurally difficult to justify.41 The second feature is related to the procedural status of the individual in the international legal order. All the debates on the legal personality of the individual in the international legal order have led to a vicious circle in international legal theory: the individual was not a subject of international law as it did not have international legal rights or duties, including standing; conversely, the individual did not have international locus standi as he was not a subject of international law.42 This approach is clearly mirrored in the 1931 Dickson Carwheel decision of the US-Mexico mixed commission, which considered in general terms that, as the relation of responsibility in international law was an inter-state one, no legal link existed with the individual who remained a subaltern object of the dispute and of the international legal order, and in no way a subject thereof.43 This stance remained until very late in the practice of mixed commissions. The 1950 French pleadings in the Ottoz case put it particularly clearly: Il est exact que, à la différence des tribunaux arbitraux mixtes, la Commission de Conciliation ne peut trancher que des litiges entre Etats; mais les Gouvernements agissent dans l’intérêt – et pour assurer le respect des droits – des ressortissants de leurs pays; c’est surtout et presque exclusivement sur le plan de la procédure que le litige demeure strictement interétatique; sur le fond du droit, la Commission […] est appelée à reconnaître ou à nier l’existence non pas seulement d’une obligation de l’Etat italien, mais d’un droit subjectif d’un ressortissant d’une des Nations Unies.44

This reveals a particularly interesting differentiation between mixed commissions’ practice, where the individual remained a subaltern, and mixed arbitral tribunals, which were considered to grant the individual autonomous standing. The contribution of the latter to the international procedural dignity of the individual deserves more attention, as this case law has been rarely studied by legal doctrine.45 In any event, investment arbitration has recently

41 F Tamburini, ‘Historia y destino de la ‘doctrina calvo’:¿ actualidad u obsolescencia del pensamiento de Carlos Calvo?’ (2002) 24 Revista de estudios histórico–jurídicos 81. 42 E Stoppioni, ‘L’accès direct de l’individu à la juridiction internationale: une comparaison entre l’arbitrage d’investissement et le contentieux de la Cour européenne des droits de l’homme’ in H Ruiz Fabri (ed), International Law and Litigation (Baden Baden, Nomos, 2019). 43 US-Mexico Mixed Commission, Dickson Car Wheel Company (USA) v United Mexican States, July 1931, IV RIAA, 669. 44 Franco-Italian Mixed Commission, Différend Ottoz, Decision No 85, 18 September 1950, 13 RIAA 232–242, 236 (‘It is true that, unlike mixed arbitral tribunals, the Conciliation Commission can only settle disputes between States; but Governments act in the interests – and to ensure respect for the rights – of their nationals; it is above all and almost exclusively at the procedural level that the dispute remains strictly inter-State; at the substantive level of law, the Commission […] is called upon to recognise or deny the existence not only of an obligation of the Italian State, but of a subjective right of a national of one of the United Nations’). 45 See notably R Bluhdorn, ‘Le fonctionnement et la jurisprudence des Tribunaux Arbitraux Mixtes créés par les Traités de Paix’ (1932) 41 Recueil des Cours de l’Académie de Droit International de La Haye 137.

Landmark Decisions for a Pre-History of International Investment Law

17

consecrated this progressive evolution. Thus, the Urbaser award stated: ‘In light of this more recent development, it can no longer be admitted that companies operating internationally are immune from becoming subjects of international law.’46 Concerning substantive matters, two fundamental aspects of the way in which commissioners have dealt with international law rules applying to financial disputes deserve to be highlighted. A real discussion, especially visible in the works of the Germany-Venezuela mixed commission, existed on the role that international law had to play in this dispute settlement mechanism and its binding effect. Umpire Duffield’s opinion in the Fischbach and Friedericy cases, quoted in many other cases, stated: International law is not law in its usually defined sense. It is not a rule of conduct prescribed by a sovereign power. It is merely a body of rules established in custom or by treaty by which the intercourse between civilized nations is governed. Its principles are ascertained by the agreement of independent nations upon rules which they consider just and fair in regulating their dealings with each other in peace and in war. They reach this agreement by comparing the opinions of text writers and in precedents in modern times, and these ultimately appeal to the principles of natural reason and morality and common sense. It therefore rests solely upon agreement. Obedience to it is voluntary only and can not be enforced by a common sovereign power. Any nation has the power and the right to dissent from a rule or principle of international law, even though it is accepted by all the other nations. Its obedience to the rule can only be compelled by an appeal to its reason and love of justice or by the superior force of the particular nation or nations whose interests are involved.47

According to the commissioners, international law is not a law nor a rule which can be imposed upon a state whose opinion differs from general international law. This in itself will not be disputed, but since Germany and Venezuela have agreed to adjust the claims of the subjects of Germany by means of arbitration, it is the duty of this Commission to apply to the different cases the law of nations, such as the Commission and not Venezuela understands it to be.48

The idea of a general international law obligation to protect aliens becomes well-established, even though commissioners keep on recalling that it is reinforced by unilateral acts of states inviting and accepting foreign investors. Commissioner Goetsch in the Fulda case is particularly clear in that regard: it ought to be considered as an obligation that international law imposes upon all civilized nations, to offer protection to foreigners – an obligation from which Venezuela can not escape. All the less, since by the law of May 14, 1869, she has invited foreigners to embark their capital and skill in Venezuelan commerce.49

The debate among Commissioners on the legal nature of international law is not solved theoretically, nevertheless in a more pragmatic way they agree that Venezuela has accepted arbitration and some investment protection obligations.

46 Urbaser SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v Republic of Argentina, ICSID case No ARB/07/26, award of 8 December 2016, para 1195. 47 Germany-Venezuela mixed claims commission, Fischbach and Friedericy cases, 1903, RIAA, 397. 48 ibid 452. 49 Germany-Venezuela mixed claims commission, Fulda case, 385.

18 Edoardo Stoppioni In this context, a central problem was the definition of a standard of treatment. The fundamental cases in that regard are clearly the Neer and the Roberts cases which gave rise to the idea that [T]he treatment of an alien, in order to constitute an international delinquency, should amount to an outrage, to bad faith, to wilful neglect of duty, or to an insufficiency of government action so far short of international standards that every reasonable and impartial man would readily recognise its insufficiency.50

Despite the well-known debates on their contemporary relevance,51 even tribunals that disagree keep on using these cases as a starting point.52 In a historical perspective, we have to focus on the discursive shift from the equal treatment theory to the minimum standard theory that the 1926 decisions implied, more than on their contribution to the modern theory of fair and equitable treatment. Latin American scholars, in line with Calvo, considered that the only thing a state owed to aliens was equal treatment as compared to its own nationals. 1926 is a fundamental turning point in the case law.53 As Mégret pointed out, ‘increased emphasis on a “minimum standard” rather than “equal treatment” meant that it mattered little whether the host state would not have granted such rights to its own citizens. That was not at issue and was therefore irrelevant’.54 The second fundamental issue is systemic in nature, as it concerned the articulation of the basic treaty with the system of public international law. We believe that at that time the legal community was faced with a sort of ante litteram debate on the ‘fragmentation’ of international law. This is particularly evident in the case law of the US-Germany mixed commission. The American arbitrator called for a strict interpretation of the Berlin Treaty, considered as a lex specialis that departed from the general rules on state responsibility.55 Much to the contrary, the German member argued that the treaty had to be interpreted in light of the general framework of international responsibility and that Germany could be deemed responsible only for internationally wrongful acts, so no reparation could be required for lawful acts.56 This last stance is supported by the idea of a systemic interpretation in light 50 USA-Mexico

Commission, LFH Neer and Pauline E Neer v Mexico, 1926, para 4. SW Schill, ‘Landmark Cases on Fair and Equitable Treatment: Empowering and Controlling Arbitrators as Law-Makers’, ch 22 in this volume. 52 Ex multis see OAO Tatneft v Ukraine (2014) PCA Case No 2008–8, para 392: ‘The Tribunal is also mindful of the discussion about whether denial of justice is an expression of the customary law ‘international minimum standard’ and how this customary standard relates to present day treaty standards of protection. What is certain is that the ‘international minimum standard’ has not been frozen at the time when it was first formulated in the Neer case in the 1920s (…). It is today accepted that customary law has evolved with time in this respect and that its own standard of protection is not necessarily different from the widespread treaty protection available at present.’ 53 In the same sense, Mégret indeed uses Teodoro García and MA Garza (United Mexican States) v United States of America, 3 December 1926. 54 Mégret, above (n 39) 141. 55 US-Germany Commission, Décision administrative n VI, RIAA, vol VII, 155–156: ‘The German Commissioner holds that neither the terms of the Treaty of Berlin nor the rules of international law impose any liability upon Germany to make compensation in such cases. The American Commissioner, on the other hand, holds that by the terms of the Treaty of Berlin, as interpreted by this Commission, Germany is clearly obligated to make compensation for losses suffered by American nationals in such cases, estimated in accordance with the rules laid down by this Commission in the Lusitania Opinion.’ The American arbitrator stated: ‘Inasmuch, therefore, as these claims come within the terms of the Treaty of Berlin, it is unnecessary to consider whether or not Germany would be liable for them under any principles of international law independently of that Treaty, because Germany’s liability under that Treaty is not limited to claims which can be supported by international law independently of that Treaty.’ 56 US-Germany Commission, Provident Mutual Life Insurance Company And Others v Germany, 18 September 1924, RIAA, vol VII, 121–140, opinion of Kiesselbach: ‘The contention of American counsel that through the death of a person insured with an – American – insurance company such company sustained a loss, and that such fact 51 See

Landmark Decisions for a Pre-History of International Investment Law

19

of general international law, timidly set out in the Lusitania case57 but originating in a more ancient litigation.58 Indeed, the umpire privileged the American position59 and this solution started to be followed in various commissions’ practice. There is a discursive spectrum in that regard, mostly explored in Mexican commissions. On one side of the spectrum, the basic treaty was considered the only instrument to take into consideration and no general rule played a role. The Kate Hoff decision considered more generally that ‘Responsibility in the cases coming before the American-German Commission was determined not in accordance with rules and principles of international law but under treaty stipulations’.60 On the other side of the spectrum, the Pinson decision established a clear principle of systemic interpretation according to which: ‘Toute convention internationale doit être réputée s’en référer tacitement au droit international commun, pour toutes les questions qu’elle ne résout pas elle-même en termes exprès et d’une façon différente.’61 This position was confirmed in the Jean-Baptiste Caire decision, which considered that treaty provisions had to be read ‘à la lueur des règles et principes généraux du droit international positif, conventionnel ou coutumier’, as customary international law had a gap-filling function.62 The Italo-French mixed commission was also an excellent example in that regard. Indeed, after a first line of cases that strictly followed the idea of lex specialis,63 the practice of the commission evolved towards giving greater importance to general rules as the normative

suffices to make Germany liable, makes it necessary to come to a clear and full understanding of the basic principles of the Treaty and, as far as it is governed by international law, to an understanding of the applicable principles of such law (…). The obligations laid upon Germany under the Treaty go so far beyond what would be justified under international law, and are so heavy, that, as already shown, they cannot be enlarged by ignoring the rules established under the Treaty. Its wording does not only fix Germany’s liability but also fixes the limitations upon it.’ 57 USA-Germany Commission, Affaire du Lusitania, 1 November 1923, RIAA, vol VII, 17–32: ‘Treaty provisions must be so construed as to best conform to accepted principles of international law rather than in derogation of them.’ 58 UK-Venezuela Commission, Arao Mines (Limited) Case, Opinion of Umpire Plumley, RIAA, vol XIX, 386–387. 59 USA-Germany Commission, United States, Garland Steamship Corporation and Others v Germany, 25 March 1924, RIAA, vol VII, 75–101: ‘The Commission is not here concerned with the quality of the act causing the damage. The terms of the Treaty fix and limit Germany’s obligations to pay, and the Commission is not concerned with enquiring whether the act for which she has accepted responsibility was legal or illegal as measured by rules of international law. It is probable that a large percentage of the financial obligations imposed by said paragraph 9 would not arise under the rules of international law but are terms imposed by the victor as one of the conditions of peace.’ See also Robert Da Vie Trudgett c Allemagne, 31 August 1926, RIAA, vol VIII, 15: ‘It will not be profitable here to consider the legality or illegality as tested by rules of international law of such capture, confinement, and detention. As this Commission has frequently held, Germany’s liability in claims presented here is determined not by rules of international law but by the terms of the Treaty of Berlin irrespective of the legality or illegality of the act complained of ’. See also, Harriss, Irby & Vose v Germany, 31 August 1926, RIAA, vol VIII, 20: ‘As this Commission has frequently held, Germany’s liability is determined by the provisions of the Treaty of Berlin rather than by the legality or illegality of her acts as measured by rules of international law.’ 60 US-Mexico Commission, Kate A. Hoff, Administratrix Of The Estate Of Samuel B Allison, Deceased (USA) v Mexico, 12 April 1929, RIAA, vol IV, 444 ss. 61 France-Mexico Commission, Georges Pinson (France) v Mexico, 19 October 1928, RIAA, vol V, 422 (‘Any international convention shall be deemed to refer tacitly to ordinary international law for all matters which it does not itself resolve in express terms and in a different manner’). 62 France-Mexico Commission, Estate of Jean-Baptiste Caire (France) v Mexico, 7 June 1929, RIAA, vol V, 528: ‘Ainsi que j’ai déjà eu l’occasion de le constater en termes généraux au §12 de la sentence No 1 (G Pinson), les questions de responsabilité internationale visées ci-dessus doivent être résolues, sous le coup de la convention, à la lueur des règles et principes généraux du droit international positif, conventionnel ou coutumier (…). Si, (…) il n’existe aucune disposition de traité qui régisse le cas (…), les controverses indiquées ci-dessus ne peuvent être tranchées que sur la seule base du droit commun coutumier.’ 63 US-Italy Commission, Mergé, 10 June 1955, RIAA, vol XIV, 240: ‘The clauses of the Treaty must be strictly followed, even when they constitute a derogation from the general rules of international law’.

20 Edoardo Stoppioni environment of the treaty. Arbitrators started to consider that to interpret a treaty meant having systematic recourse to the customary rules of interpretation,64 a position which is well-established in contemporary practice.

IV. CONCLUSION

The main thesis of this chapter has two facets. The first is methodological: there cannot be a Dogmengeschichte of international investment law without a critical assessment of its original imperial and colonial mindset. The second challenges the most common discourse on the topic: the story of international investment law started long before the first BIT or the first ICSID awards. A certain arbitral practice dealing with international financial matters probably already existed in Mesopotamia, but we lack the evidence to validate this supposition. There is however proof of the existence of international protection of foreign property in Ancient Greece. This protection relied either on the foreigner’s access to local courts via a symbolon treaty or was, more rarely, found at the inter-state level. During the Middle Ages, the arbitral practice of Northern Italian cities, Swiss cantons and German principalities revealed a clear intertwining of public and private law. Consular jurisdiction gave rise to a transnational merchants’ dispute settlement practice, based on the need for a technical, rapid and discreet mechanism, by derogation to local courts. In the same spirit, starting with the Jay Treaty, mixed claims commissions were established to settle state responsibility claims concerning injuries borne by aliens. A colonial philosophy is evident in their practice. In order to allow international protection of western investors, notably in Latin America, their practice strongly contributed to the development of international law and international dispute settlement. It advanced the development of the procedural dignity of the individual in international law due to a general mistrust of the host state’s national courts. Similarly, mixed claims commissions developed a regime of state responsibility based on customary minimum standards rules, implying a discursive shift from the theory of equality with local investors. This impressionistic tour d’horizon of the history of international investment law has attempted to show how important it is to be aware of the roots of its developments, which are beautifully analysed in the following chapters of this book.

64 Ex multis US-Italy Commission, Flegenheimer, 20 September 1958, RIAA, vol XIV, 366 and Fubini, 12 December 1959, RIAA, vol XIV, 425.

3 Ad hoc Investment Arbitration Based on State Contracts: From Lena Goldfields to the Libyan Oil Arbitrations DANIEL MÜLLER*

T

RADITIONAL INTERNATIONAL CLAIMS mechanisms, which have nowadays ‘somewhat faded’1 but are still not entirely outdated,2 suffer from major shortcomings that render the efficient protection of foreign investors before international courts and tribunals a difficult endeavour. Exhaustion of local remedies, the necessary endorsement by a state of a claim of one of its nationals, the discretionary nature of diplomatic protection3 and the role of the investor in the claims resolution process and its aftermath have long been considered major obstacles to a more efficient administration of justice. Alternatives to traditional diplomatic protection evolved well before the International Centre for Settlement of Investment Disputes (ICSID) Convention and investment treaties. Recourse to arbitration as a mechanism to resolve disputes between an investor and a state can be traced back to the proceedings between the Compagnie universelle du canal maritime de Suez and Egypt concerning issues of compensation for regulatory changes and the use of corvée work, transfer of land and other parts of the Suez Canal project. The Company and the ruler of Egypt agreed to submit their dispute to French Emperor Napoleon III; the arbitral commission rendered its award in 1864.4

* Daniel Müller is a Member of the Paris Bar, Partner at FAR Avocats, and an Associate Researcher of the Centre de droit international de Nanterre (CEDIN). 1 See Ahmadou Sadio Diallo (Republic of Guinea v Democratic Republic of the Congo), Preliminary Objections, Judgment, ICJ Reports 2007, 614, para 88; EDF International SA, SAUR International SA and León Participaciones Argentinas SA v Argentine Republic, Decision on Annulment (5 February 2016) ICSID Case No ARB/03/23, para 256; Enrique and Jorge Heemsen v Bolivarian Republic of Venezuela, Award on Jurisdiction (29 October 2019) PCA Case No 2017–18, para 249. 2 Even today, diplomatic protection is still used, including by investors and their states of nationality. See, eg the arbitration between Italy and Cuba on the basis of Art 10 of the 1993 Italy-Cuba BIT (Italian Republic and Republic of Cuba, ad hoc arbitration, Interim Award, 15 March 2005). 3 For a recent illustration, see the dispute between Swissbourgh Diamond Mines (PTY) Ltd and the Kingdom of Lesotho where the investor unsuccessfully sought diplomatic protection by the Government of South Africa. Swissbourgh Diamond Mines (PTY) Ltd and others v Kingdom of Lesotho, Judgment, Court of Appeal of the Republic of Singapore, 27 November 2018, [2018] SGCA 81, para 24. 4 For a detailed account of the factual and legal circumstances of this arbitration, the proceedings and the award, see JW Yackee, ‘The First Investor-State Arbitration? The Suez Canal Dispute of 1864 and some Reflections on the Historiography of International Investment Law’ in SW Schill, CJ Tams and R Hofmann (eds), International

22 Daniel Müller Arbitration has also become an important feature of concession agreements. In particular in the extractive industries, involving long-term concessions that are coupled more often than not with substantial up-front investment in exploration or exploitation activities, companies were keen to attract the protection of their investment from state interference prior to the development of modern international investment law. Legal scholars and arbitral tribunals have enthusiastically discussed the legal nature and the specific legal regime of what used to be called ‘State contracts’ or ‘contrats d’État’.5 Even if these discussions attract much less interest and have been largely overtaken by the modern rules and realities of investment protection,6 arbitral decisions based on such specific contractual relationships between states and investors have greatly contributed to the development of the protection of international investments as part of international law.

I. THE LENA GOLDFIELDS ARBITRATION

The Lena Goldfields arbitration7 is one of the first transnational arbitrations between an investor and a state, the USSR, under a concession contract.8 In 1950, Nussbaum referred to the decision as being ‘one of the most remarkable occurrences in the field of arbitration’.9 Nevertheless it remains largely ignored in modern case law10 and, from time to time, is said to have ‘probably contributed nothing but a great deal of confusion’.11 Investment Law and History (Cheltenham, E Elgar Publishing, 2018) 70–101; JW Yackee, ‘The First Investor-State Arbitration: The Suez Canal Company v Egypt’ (1864) (2016) 17 Journal of World Investment & Trade 401–462. 5 See for instance, FA Mann, ‘State Contracts and International Arbitration’ (1967) 42 British Yearbook of International Law 1; JF Lalive, ‘Contracts between a State or a State Agency and a Foreign Company’ (1964) 13 International & Comparative Law Quarterly 987–2021; P Weil, ‘Problèmes relatifs aux contrats passes entre un État et un particulier’ (1969) 128 Recueil des Cours de l’Académie de Droit International 95–240; J-F Lalive, ‘Contrats entre États ou entreprises étatiques et personnes privées: Développements récents’ (1983) 181 Recueil des Cours de l’Académie de Droit International 9–283; Hague Academy of International Law. Center for Studies and Research. Transnational Arbitration and State Contracts (Dordrecht, Nijhoff, 1988); L Lankarani El-Zein, Les contracts d’États à l’épreuve du droit international (Brussels, Bruylant, 2001); GR Delaume, ‘The Proper Law of State Contracts Revisited’ (1997) 12 ICSID Review – Foreign Investment Law Journal 1–28; C Leben, ‘La théorie du contrat d’État et l’évolution du droit international des investissements’ (2003) 302 Recueil des Cours de l’Académie de Droit International 197–386; J-M Jacquet, ‘Voici venu le temps des traités ! Quelques réflexions sur l’évolution du droit des contrats d’État’ in MG Kohen (ed), Promoting justice, human rights and conflict resolution through international law: liber amicorum Lucius Caflisch (Leiden, Martinus Nijhoff Publishers, 2007) 1121–1131; M Abdel Raouf, L’arbitre international et les contrats d’État (Saarbrücken, Éditions universitaires européennes, 2017). 6 The ICSID Convention was mainly based on the idea of a direct contractual relationship between an investor and a host state. However, against the development of bilateral and multilateral investment treaties, the importance of arbitrations brought on the basis of a contract between an investor and a host state has significantly reduced. According to statistics published by ICSID, only 15% of all cases registered by ICSID have been based on such contracts. The ICSID Caseload – Statistics, Issue 2021–2, 11. 7 Lena Goldfields Co Ltd und Regierung der USSR, Award (2 September 1930) (hereafter Lena Goldfields, Award), German original available at www.trans-lex.org; English translation available in A Nussbaum, ‘Arbitration Between the Lena Goldfields Ltd and the Soviet Government’ (1950) 36(1) Cornell Law Review 42–53. 8 For a number of other early cases based on concession contracts between investors and States, see JG Wetter and SM Schwebel, ‘Some Little-Known Cases on Concessions’ (1964) 40 British Yearbook of International Law 183. 9 Nussbaum, above (n 7) 31. 10 Only a few recent arbitral tribunals and international tribunals have referred to the Lena Goldfields arbitration. See, eg, Biwater Gauff (Tanzania) Limited v United Republic of Tanzania, Award (24 July 2008) ICSID Case No ARB/05/22, para 505; LG&E Energy Corp, LG&E Capital Corp and LG&E International Inc v Argentine Republic, Award (25 July 2007) ICSID Case No ARB/02/1, para 91 (fn 36); Saluka Investments BV v The Czech Republic, Partial Award (7 March 2006) PCA Case No 2001–04, para 449; Methanex Corporation v United States of America, Decision of the Tribunal on Petitions from Third Persons to Intervene as ‘amici curiae’ (15 January 2001) para 44; Liberian Eastern Timber Corporation v of Liberia, Award (31 March 1986) ICSID Case No ARB/83/2, para 95; Amco Asia Corporation and others v Republic of Indonesia, Award (20 November 1984) ICSID Case No ARB/81/1,

Ad hoc Investment Arbitration Based on State Contracts 23 A. Lena Goldfields and the Concession Agreement Lena Goldfields was a company established in 1908 by British, Russian, French, German and American shareholders. Based in London, it operated several goldfields in East Siberia in the vicinity of the Lena river. The profitable activities of Lena Goldfields were severely affected by the 1912 Lena massacre, when striking goldfield workers were killed by soldiers of the Imperial Russian Army, as well as by World War I. In 1918, the Soviet Government nationalised all mining properties. When the new economic policy was put into place in the 1920s by the Soviet Government,12 Lena Goldfields started negotiations with the Soviet authorities for a new concession agreement. In 1925, they entered into a long-term Concession agreement, granting the company exclusive rights to prospect and mine gold, iron, copper, lead, zinc and other precious metals in the Urals and large parts of Siberia, along with other related rights to land and to conduct economic activity, to import, to export, etc.13 Under the contract, the government assumed responsibility for all losses caused by a breach of the Concession agreement or by illegal conduct of the local or central authorities14 and undertook to provide the protection necessary to guarantee the safety of the property of Lena Goldfields and its metal production.15 The company accepted to be subject to the existing and future legislation, unless special conditions were provided under the agreement; the Russian authorities agreed not to alter the conditions of the agreement unilaterally. In other words, Lena Goldfields’ legal position was fully protected and the provisions of the agreement were aimed at preventing ‘the mutual rights and obligations of the parties under the contract [from] being altered by any act of the Government, legislative, executive, or fiscal, or by any action of local authorities or trade unions’.16 Article 90 of the Concession agreement included provisions for the resolution of disputes between Lena Goldfields and the Soviet authorities through arbitration.17 It was later established that ‘Lena would not have entered into the Concession Agreement at all but for the presence in the contract of this arbitration clause’.18 Despite some shortcomings in the implementation of the Concession agreement by the Soviet authorities, Lena Goldfields was able to generate profits from its activities in the USSR. However, in 1929, the economic policy of the USSR changed drastically with the instauration of the ‘Five-Year Plan’ and the submission of the political and economic life in the USSR to

para 267. It is still invoked from time to time in the parties’ written pleadings. See, eg, President Allende Foundation, Victor Pey Casado and Coral Pey Grebe v Republic of Chile, Claimants’ Memorial on Jurisdiction and Merits (6 January 2018) PCA Case No 2017–30, para 477; Certain Property (Liechtenstein v Germany), ICJ, Memorial of Liechtenstein, 28 March 2002, para 6.29; Marvin Roy Feldman Karpa v United Mexican States, Claimant’s Memorial (30 March 2001) ICSID Case No ARB(AF)/99/1, 70. 11 C Schreuer, ‘Unjust Enrichment in International Law’ (1974) 22(2) American Journal of Comparative Law 289; B Juratowitch and J Shaerf, ‘Unjust Enrichment as a Primary Rule of International Law’ in M Andenæs and others (eds), General Principles and the Coherence of International Law (Leiden, Boston, Brill/Nijhoff, 2019) 244. See also, Azurix Corp v The Argentine Republic, Award (14 July 2006) ICSID Case No ARB/01/12, para 435. 12 See Lena Goldfields, Award, para 19. 13 ibid paras 16 and 18(c). 14 ibid para 18(e). 15 ibid para 18(g). 16 ibid para 18(i). 17 Translation of Article 90 of the Concession agreement, in VV Veeder, ‘The Lena Goldfields Arbitration: The Historical Roots of Three Ideas’ (1998) 47(4) International & Comparative Law Quarterly 790–792 (extracts). A German version of the arbitration clause is reproduced in the German text of the Award, para 5. 18 Lena Goldfields, Award, para 6.

24 Daniel Müller communist principles. This policy change had far-reaching consequences for Lena Goldfields and its Concession. As explained in the Award: The Five-Year Plan thus put Lena into a position in the Communist State where it became peculiarly exposed to hostile criticism. The official Soviet Press has been filled with such attacks in an increasing degree during the last 12 months. As an equally inevitable consequence Lena has been regarded as a capitalist outcast by the Communist public of the U.S.S.R. This complete reversal in 1929 of the official policy of 1925 towards Lena necessarily meant, when measured in terms of contractual [provisions], the breach by the Government of many of the [express and] fundamental provisions […] of the Concession Agreement.19

B. The Arbitration Proceedings and their Outcome On 12 February 1930, Lena Goldfields began arbitration proceedings against the USSR under the Concession agreement, claiming £13 million as compensation for the undue difficulties and interference created by the government, rendering the performance of the Concession agreement practically impossible and preventing Lena Goldfields from enjoying the privileges and benefits granted under the Concession agreement.20 The Soviet Government formulated several unsubstantiated counter-claims in respect of alleged breaches of the Concession agreement by Lena Goldfields. The Award was rendered on 2 September 1930 by the arbitrator appointed by Lena Goldfields, Sir Leslie Scott, and the presiding arbitrator, Dr Otto Stutzer, a professor of mining at the Freiberg University of Mining in Germany. Although the USSR had participated in the constitution of the arbitral Tribunal and had appointed Dr Semyon Borisovich Chlenov as its arbitrator, it withdrew from the proceedings in May 1930. The government did not participate in the further proceedings and instructed the arbitrator it had appointed to do the same. The position of the Respondent was based on the allegation that, by refusing to assume any further responsibilities under the Concession agreement, Lena Goldfields had terminated the agreement and that therefore the arbitral Tribunal had ceased to function.21 The Tribunal however continued its work despite the Soviet objection, the absence of the arbitrator appointed by the Respondent, and the absence of any representative for the Respondent appearing before the truncated Tribunal. Indeed, the possibility for the Tribunal to continue its work was specifically addressed in the arbitration clause.22 The Tribunal also considered that the Concession agreement was still operative and that therefore its jurisdiction remained unaffected.23 The Tribunal’s decision on the Soviet objection constituted a reaffirmation of the principle of Kompetenz-Kompetenz, which had been applied by other international tribunals

19 ibid

para 19. para 8. 21 ibid para 10. It has been suggested that, in the opinion of the Soviet authorities, the issue of the termination of the Concession agreement had not been properly submitted to the arbitration proceedings and that, by entertaining this issue, the Tribunal was deemed as acting ultra vires. Veeder, above (n 17) 782–785. 22 Article 90 of the Concession agreement, in Veeder, above (n 17) 791 (‘If, upon receipt of the summons from the [presiding arbitrator] appointing the date and place of the first session, one of the parties, in the absence of insuperable obstacles, fails to send its arbitrator to the arbitration court or the arbitrator avoids participating in the arbitration court, then the matter in dispute shall, at the request of the other party, be settled by the [presiding arbitrator] and the other member of the court, such settlement to be valid only if unanimous.’) 23 Lena Goldfields, Award, para 11. 20 ibid

Ad hoc Investment Arbitration Based on State Contracts 25 before 1930.24 It is now well established that ‘an international tribunal has the right to decide as to its own jurisdiction and has the power to interpret for this purpose the instruments which govern that jurisdiction’.25 Indeed, as rightly pointed out, ‘[w]ithout a principle of Kompetenz-Kompetenz, any form of third party decision in international law could be paralyzed by a party which challenged jurisdiction’.26 More remarkable are the Tribunal’s findings in respect of the law it had to apply. The Concession agreement did not determine the applicable law. It merely referred generally to goodwill, good faith and the reasonable interpretation of its terms.27 In the Award, counsel for Lena Goldfields was recorded as submitting that in respect of the performance of the Concession agreement inside the USSR, Russian law was the proper law of the contract.28 This seems to be in line with Lena Goldfields accepting under the terms of the Concession agreement to be submitted to Russian law except as otherwise provided in the agreement.29 This position is also in line with the judgment of the Permanent Court of International Justice in the Serbian Loans case.30 However, counsel for Lena Goldfields also submitted that ‘for other purposes’, ie, for issues that do not concern the performance of the Concession agreement inside the USSR: the general principles of law such as those recognized by Article 38 of the Statute of the Permanent Court of International Justice [in] the Hague should be regarded as ‘the proper law of the contract’ …31

This proposition was based on the fact that the Concession agreement had also been signed by the Deputy Commissioner for Foreign Affairs of the USSR and that certain provisions of the agreement contemplated the application of international rather than national principles of law. The Arbitral Tribunal agreed with this proposition in what was later said to be ‘laconic words’:32 ‘In so far as any difference of interpretation [arises,] the Court holds that this contention is correct.’33 It might indeed be true that the ‘tribunal expressed itself obscurely on issues of critical importance’,34 and that the tribunal ‘did not fully consider the enormity of what [it was] deciding’35 and whether it was indeed necessary to make such a decision.36 However, rather than internationalising the Concession agreement and applying general

24 See,

eg, Affaire du Guano (Chili/France), Judgment (20 October 1900) RIAA, vol XV, 100. (Liechtenstein v Guatemala), Preliminary Objections, Judgment, ICJ Reports 1953, 119; Prosecutor v Duško Tadić a/k/a ‘Dule’, ICTY Case No IT-94-1-AR72, Appeals Chamber, Decision (2 October 1995) para 18; Quiborax SA, Non-Metallic Minerals SA v Plurinational State of Bolivia, Decision on Jurisdiction (27 September 2012) ICSID Case No ARB/06/2, para 63; Blue Bank International & Trust (Barbados) Ltd v Bolivarian Republic of Venezuela, Award (26 April 2017) ICSID Case No ARB/12/2, para 102. 26 Abyei Arbitration (The Government of Sudan/The Sudan People’s Liberation Army/Movement), Final Award, 22 July 2009, RIAA, vol XXX, 330, para 499. See also Arbitration Between the Republic of Croatia and the Republic of Slovenia, Partial Award (30 June 2016) PCA Case No 2012–04, para 149. 27 Lena Goldfields, Award, para 6. 28 ibid para 22. 29 See n 16 above and accompanying text. 30 Serbian Loans, Judgment (1929) PCIJ, Series A, No 20, 41 (‘Any contract which is not a contract between States in their capacity as subjects of international law is based on the municipal law of some country’). See also, Convial Callao SA and CCI – Compañía de Concesiones de Infraestructura SA v Republic of Peru, Final Award (21 May 2013) ICSID Case No ARB/10/2, para 502. 31 Lena Goldfields, Award, para 22. 32 Veeder, above (n 17) 767. 33 Lena Goldfields, Award, para 22. 34 Veeder, above (n 17) 767. 35 ibid 768. 36 FA Mann, ‘The Proper Law of Contracts Concluded by International Persons’ (1959) 35 British Yearbook of International Law 56. 25 Nottebohm

26 Daniel Müller principles of law to the interpretation of this agreement,37 the Tribunal reasoned on the basis of general principles of law only in respect of Lena Goldfields’ claim under the principle of unjust enrichment, referring to European Continental law, including Soviet, French, and German law, but also to Scottish law and, to some extent, English law.38 In other words, despite the doubtful formulation in the Award, the Lena Goldfields Arbitration remains a milestone in the internationalisation of the legal relationship between an investor and the host state of the investment in the sense that a Respondent state cannot only rely on its municipal law, but must admit the application of international law principles in its relationship with a foreign investor. The decision of the Arbitral Tribunal in the Lena Goldfields case, despite the remaining uncertainties, criticisms and doubts, constitutes one of the milestones for the development of Article 42 of the ICSID Convention. The Tribunal awarded Lena Goldfields the sum of £12,965,000 plus interest at 12 per cent in accordance with the provisions of the Concession agreement.39 The valuation undertaken by the Arbitral Tribunal is a classic example of determining the net present value of a company, taking into account future profits and expenses. The Tribunal explained: The problem before the Court is to arrive at the present value if paid in cash now, of future profits which the company would have made and which the Government now can make …. The problem is, therefore, similar to that of ascertaining a fair purchase price for a going concern.40

To achieve this goal, the Tribunal applied well-known valuation principles of mining properties that to some extent are still used today, weighing reserves and resources of ore, the future price of ore and metal, the future costs for mining, and the time of the exploitation of the ore.41 The Lena Goldfields case is ground-breaking, in particular because the Tribunal clearly showed the path to the application of international law (ie, general principles of law) rather than of municipal law. However, for Lena Goldfields the outcome of the arbitration was less satisfactory. The USSR refused to honour the Award, and there was no legal remedy available to the company. In 1935 a settlement was reached but this was not honoured either.42

II. EARLY ARBITRATIONS CONCERNING THE SCOPE, INTERPRETATION AND APPLICATION OF PETROLEUM CONCESSIONS

The idea of internationalising the legal relationship between an investor and the host state of the investment has been repeatedly applied since the Lena Goldfields Arbitration. In the late 1950s, there were a series of Middle Eastern arbitrations relating to the oil industry that

37 The Tribunal noted however that the breach of the agreement by the government justified a claim for damages under common principles of law. Lena Goldfields, Award, para 25. Nevertheless, it preferred to base its decision on unjust enrichment considering that the result in terms of monetary compensation would have been the same. ibid. 38 Lena Goldfields, Award, para 23. 39 ibid paras 26 and 31. 40 ibid para 26. 41 ibid. Unfortunately, this part of the Award of the Arbitral Tribunal has not been translated by A Nussbaum. It is however contained in the German version of the Award. 42 For a detailed account of the ‘implementation’ of the Lena Goldfields Award and the compensation of the company, see Veeder, above (n 17) 786–789; and SN Lebedev and NG Doronina, ‘Arbitrating Russian Concession Contracts: The Lena Goldfields Case’ in U Franke and others (eds), Arbitrating for Peace: How Arbitration Made a Difference (Alphen aan den Rijn, Kluwer Law International, 2016) 51.

Ad hoc Investment Arbitration Based on State Contracts 27 were equally important milestones on the way to the development of international investment arbitration. In all of these arbitrations, the arbitral tribunals refused to base their reasoning on local law only, internationalising the contractual and legal relationship between the petroleum companies and the host state.

A. The Use of International Law as a Substitute for or a Corrective of Insufficiently Developed Municipal Law: Petroleum Development v Abu Dhabi43 and Saudi Arabia and ARAMCO44 In 1939, Petroleum Development (Trucial Coast) Ltd, a UK company, entered into an oil concession agreement with the Sheikh of Abu Dhabi, which was then, in the terms of the award, ‘a large, primitive, poor, thinly populated country, whose revenue, until oil was discovered, depended mainly on pearling’.45 Abu Dhabi, like other principalities on the Trucial Coast, was at that time still a British Protectorate. Against the background of the development of the continental shelf doctrine, and in particular the Truman Declaration of 1945, the Sheikh of Abu Dhabi proclaimed exclusive rights and jurisdiction of Abu Dhabi over the seabed.46 A dispute arose between Petroleum Development and the Sheikh concerning the question of whether the 1939 Concession agreement included the exclusive right for Petroleum Development to exploit the seabed of Abu Dhabi’s territorial sea and the continental shelf. The company started arbitration in accordance with the arbitration clause included in the 1939 Concession agreement. Lord Asquith of Bishopstone, a British barrister and judge, was appointed as umpire in charge of rendering a binding award on the proper interpretation of the territorial scope of the agreement. He held that the seabed of the territorial sea of Abu Dhabi and its islands was included in the territorial scope of the Concession, but that the continental shelf adjacent to the mainland and the islands of Abu Dhabi was not. In order to reach his conclusion, the umpire had to determine the ‘proper law’ applicable to the construction of the Concession agreement. Like the Concession agreement in the Lena Goldfields case, the 1939 Concession agreement included a rather ambiguous provision on applicable law, Article 17 provided that: The Ruler and the Company both declare that they base their work in this Agreement on goodwill and sincerity of belief and on the interpretation of this Agreement in a fashion consistent with reason.47

The umpire noted that the agreement was concluded in Abu Dhabi and had to be performed in that country. Therefore, ‘if any municipal system of law were applicable, it would prima facie

43 Petroleum Development (Trucial Coast) Ltd v The Sheikh of Abu Dhabi, Award (September 1951) reproduced in 1 International & Comparative Law Quarterly, 1952, 247–261 (hereafter ‘Petroleum Development, Award’). A report of the Award is included in (1951) 18 International Law Reports 144–161. 44 Saudi Arabia and Arabian American Oil Company (ARAMCO), Award (23 August 1958) reproduced partly in 27 International Law Reports 117–229 (hereafter ‘ARAMCO, Award’). 45 Petroleum Development, Award, 247. 46 ibid 255. 47 ibid 250.

28 Daniel Müller be that of Abu Dhabi’.48 Nevertheless, Lord Asquith refused to apply the law of Abu Dhabi for reasons that by today’s standards would seem appalling: But no such law [i.e., the municipal law of Abu Dhabi] can reasonably be said to exist. The Sheikh administers a pure discretionary justice with the assistance of the Koran; and it would be fanciful to suggest that in this very primitive region there is any settled body of legal principles applicable to the construction of modern commercial instruments.49

The umpire also rejected the idea of applying any specific municipal law. He opined that: The terms of that clause invite, indeed prescribe, the application of principles rooted in the good sense and common practice of the generality of civilised nations – a sort of ‘modern law of nature.’50

The reasoning of the umpire remains open to criticism. Of course, it needs to be put into the perspective of the then still prevailing western perception of the world. Although the reference to ‘general principles of law recognized by civilized nations’ in Article 38(1)(c) of the Statute of the International Court of Justice is certainly outdated,51 it has been suggested that it is limited ‘to consideration of municipal systems [that] are sufficiently developed to reveal the extent to which they share common underlying principles’.52 This slightly more nuanced approach seems to have been adopted by the sole arbitrator in Ruler of Qatar and International Marine Oil Company, Ltd.53 Sir Alfred Bucknill, who was a judge in England, rejected the application of Islamic law to the Concession agreement between the Ruler of Qatar and the company after having considered the evidence to the effect that the law applicable in Qatar did not contain legal principles applicable to the construction of modern commercial instruments and that, if Islamic law had been the proper law of the contract, the agreement would have been invalid.54 In other words, it could not have been the intention of the parties to the Concession to subject their relationship to Islamic law applicable in Qatar, but rather to ‘the principles of justice, equity and good conscience’.55 This more nuanced approach to the issue of the law applicable to a petroleum concession agreement was embraced by the arbitral tribunal in Saudi Arabia and Arabian American Oil Company (ARAMCO). A dispute arose between ARAMCO and Saudi Arabia in respect of the scope of the rights conferred under a concession agreement entered into in 1933. This dispute was triggered by the conclusion of an agreement between Saudi Arabia and Saudi Arabian Maritime Tankers Ltd, granting the latter company a right of priority for the transport of Saudi oil. Saudi Arabia and ARAMCO submitted the dispute to arbitration pursuant to a special agreement signed on 23 February 1955.56 The Tribunal conducted a thorough and detailed analysis of the law applicable to the Concession agreement and its effects, including in respect of the private international law principles to be applied. It noted that obviously ‘no contract can exist in vacuo, i.e., without being based on a legal system’. The Tribunal further added that, as a contract concluded 48 ibid. 49 ibid

250–251. 251. 51 A Pellet and D Müller, ‘Article 38’ in A Zimmermann and CJ Tams (eds), The Statute of the International Court of Justice: A Commentary, 3rd edn (Oxford, Oxford University Press, 2019) 927; A de Nanteuil, Droit international de l’investissement (Paris, Pedone, 2020) 43. 52 H Thirlway, ‘The Law and Procedure of the International Court of Justice: 1960–1989: Part Two’ (1991) 61 British Yearbook of International Law 124. 53 Award (June 1953) reported in 20 International Law Reports 534. 54 ibid 544–545. 55 ibid 545. 56 The Arbitration Agreement is published in (2005) 27 International Law Reports 117–233. 50 ibid

Ad hoc Investment Arbitration Based on State Contracts 29 between a state and a private company, the applicable law cannot be public international law, but only the municipal law of some country.57 Applying general principles of conflict of laws, the Tribunal concluded that it was the ‘law in force in Saudi Arabia [that] should also be applied to the content of the Concession’,58 subject to the specific provisions of the Concession agreement – being the ‘fundamental law of the Parties’ – filling the gaps of Saudi Arabian law on oil concession agreements.59 The application of Saudi law was however subject to an important caveat: Matters pertaining to private law are, in principle, governed by the law of Saudi Arabia but with the one important reservation. That law must, in case of need, be interpreted or supplemented by the general principles of law, by the custom and practice in the oil business and by notions of pure jurisprudence, in particular whenever certain private rights – which must inevitably be recognised to the concessionaire if the Concession is not to be deprived of its substance – would not be secured in an unquestionable manner by the law in force in Saudi Arabia.60

This formulation is reminiscent of Article 42(1) of the ICSID Convention and the function of international law as a supplement or correction to municipal law.61 On this basis and having interpreted the Concession agreement, the Tribunal held that ARAMCO enjoyed exclusive rights in respect of transport of Saudi oil. Considering the compatibility between these exclusive rights and the competing agreement entered into by the state, the Tribunal held: In its capacity as first concessionaire, Aramco enjoys indeed exclusive rights which have the character of acquired or vested rights and which cannot be taken away from it by the Government by means of a contract concluded with a second concessionaire, even if that contract were equal to its own contract from a legal point of view. The principle of respect for acquired rights is one of the fundamental principles both of public international law and of the municipal law of most civilized States. … In the Hanbali school of Islamic law, respect for previously acquired private rights, and especially for contractual rights, is a principle just as fundamental as it is in the other legal systems of civilized States.62

B. The Implicit Rejection of Any Municipal Legal System in Concession Agreements: Sapphire International v NIOC63 The case between Sapphire International Petroleums, a Canadian company, and the National Iranian Oil Company (NIOC) concerning the revocation and breach of an oil concession agreement constituted a further step towards the internationalisation of investment contracts and the rejection of the application of a specific municipal legal system.

57 ARAMCO,

Award, 165. 167. 59 ibid 168. 60 ibid 169. 61 C Schreuer and others, The ICSID Convention: A Commentary, 2nd edn (Cambridge, Cambridge University Press, 2009) 617–630. 62 ARAMCO, Award, 205. 63 Sapphire International Petroleums Ltd v National Iranian Oil Company, Award (15 March 1963) reproduced in (2012) 35 International Law Reports 136–192 (hereafter ‘Sapphire, Award’). Reports of the Award are also contained in Lalive, above (n 5) 1002–1006 and JF Lalive, ‘Un récent arbitrage suisse entre un organisme d’État et une société privée étrangère’ (1962) 19 Annuaire Suisse de Droit International 273–302. 58 ibid

30 Daniel Müller Sapphire and NIOC entered into the Concession agreement on 16 June 1958. Under the Agreement, the parties set up a specific joint-venture, the Iranian Canada Oil Company (IRCAN) for the implementation of the Agreement. The Parties agreed that they would carry out their obligations under the Agreement in a spirit of good faith and reciprocal good will. Difficulties soon arose and Sapphire found itself unable to proceed with the work given NIOC’s persistent attitude of non-cooperation and obstruction. Sapphire decided to go to arbitration and to claim reimbursement of the contractual penalty retained by NIOC, as well as compensation for its losses. The sole arbitrator, Pierre Cavin, a Swiss Federal judge, took a drastic view in respect of the law applicable to the Concession contract.64 Although he recognised that general rules of private international law (as opposed to only private international law rules of the lex fori) pointed to the application of Iranian law, he considered that the nature and the content of the Concession agreement advocated not retaining the municipal law of NIOC for the interpretation and application of the agreement. One of the forceful considerations put forward concerned the effective protection of Sapphire’s investments: Under the present agreement, the foreign company was bringing financial and technical assistance to Iran, which involved it in investments, responsibilities, and considerable risks. It therefore seems natural that they should be protected against any legislative changes which might alter the character of the contract, and that they should be assured of some legal security. This could not be guaranteed to them by the outright application of Iranian law, which it is within the power of the Iranian State to change.65

Cavin further relied on the ‘applicable law’ provision of the agreement that pointed to the principles of good will, good faith and to the spirit and letter of the agreement. In his view, ‘such a clause is scarcely compatible with the strict application of the internal law of a particular country’.66 This does not imply, however, that the Concession agreement is not submitted to any law: [A] reference to rules of good faith, together with the absence of any reference to a national system of law, leads the judge to determine, according to the spirit of the agreement, what meaning he can reasonably give to a provision of the agreement which is in dispute. It is therefore perfectly legitimate to find in such a clause evidence of the intention of the parties not to apply the strict rules of a particular system but, rather, to rely upon the rules of law, based upon reason, which are common to civilised nations. These rules are enshrined in Article 38 of the Statute of the International Court of Justice as a source of law, and numerous decisions of international tribunals have made use of them and clarified them. Their application is particularly justified in the present contract, which was concluded between a State organ and a foreign company, and depends upon public law in certain of its aspects. This contract has therefore a quasi-international character which releases it from the sovereignty of a particular legal system, and it differs fundamentally from an ordinary commercial contract.67

Therefore the nature of the agreement and the reference to good will and good faith, as well as a reference to international law in relation to force majeure68 were sufficient to exclude the

64 See also CM Spofford, ‘Third party judgment and international economic transactions’ (1964) 113 Collected courses 202–205. 65 Sapphire, Award, 171. 66 ibid 173. 67 ibid. 68 ibid 173–174.

Ad hoc Investment Arbitration Based on State Contracts 31 application of Iranian law and to submit the agreement in its entirety to general principles of law. The arbitrator added that this did not imply deciding the dispute in equity, but solely on the basis of ‘the rules of positive law, common to civilised nations’.69 Faithful to these undertakings, Cavin established that it is a general rule of private law to be found in positive systems of law that the failure of a party to honour its obligations releases the other party from its obligations and gives rise to compensation of loss suffered.70 He also established that, as a general principle, damages have to place a party in the same position that it would have been in if the contract would have been performed as envisaged, including loss suffered (damnum emergens) and lost profits (lucrum cessans).71 All these general principles remain relevant in modern, truly international investment law irrespective of whether or not the investment is made in the form of a concession agreement.

III. ARBITRATIONS CONCERNING THE NATIONALISATION OF OIL CONCESSIONS IN LIBYA AND KUWAIT

In the late 1970s and early 1980s a relatively small number of arbitrations between oil concessionaires and states concerning oil concession agreements had an important impact on the development of international investment law. Of course, in the 1970s, the idea of the protection of investors through specific remedies in an international environment had already crystallised in the establishment of ICSID. However, the emergence of a great number of newly independent states with aspirations to create a new international economic order gave rise to considerable uncertainties for the protection of private investments, in particular in respect of the exploration and exploitation of natural resources.72 One of the principles of this new international economic order was the full permanent sovereignty of every state over its natural resources and all economic activities, including the right to nationalisation.73 The three Libyan arbitrations concerned BP,74 Texaco-Calasiatic,75 and LIAMCO,76 on the one hand, and the Libyan Government, on the other.77 All three companies had entered into concession agreements in the late 1950s and 1960s. After the 1969 coup d’état, the Libyan authorities nationalised the rights and properties of BP in 1971, and the rights of Texaco-Calasiatic and LIAMCO through two nationalisation decrees adopted in 1973 and 1974. All three companies started arbitration proceedings invoking similar arbitration clauses 69 ibid

175. 182–184. 71 ibid 185–186. 72 See UNGA, Declaration on the Establishment of a New International Economic Order (1 May 1974) UN Doc A/Res/3201 (S–VI). 73 ibid para 4(e). 74 BP Exploration Company (Libya) Limited v Government of the Libyan Arab Republic, Award (10 October 1973) reproduced in (2000) 53 International Law Reports 300–357 (hereafter ‘BP, Award’). A second award was rendered on 1 August 1974 rejecting the Claimant’s request to reopen the proceedings. See (2000) 53 International Law Reports 375–388. 75 Texaco Overseas Petroleum Company and California Asiatic Oil Company v The Government of the Libyan Arab Republic, Award on the Merits (19 January 1977) reproduced in (2000) 53 International Law Reports 420–511 (hereafter ‘Texaco-Calasiatic, Award on the Merits’). The original French version of the Award on the Merits was published in [1977] Clunet 350–389. The sole arbitrator had also rendered a preliminary award dealing with certain allegations of jurisdiction. This award is reproduced in (2000) 53 International Law Reports 393–419. 76 Libyan American Oil Company (LIAMCO) v Government of the Libyan Arab Republic, Award (12 April 1977) reproduced in (2009) 62 International Law Reports 145–219. 77 For a comprehensive review of these arbitrations, see C Greenwood, ‘State Contracts in International Law: The Libyan Oil Arbitrations’ (1982) 53 British Yearbook of International Law 27–81. 70 ibid

32 Daniel Müller in their respective concession agreements. Libya refused to engage in the proceedings. Therefore the companies requested the president of the International Court of Justice to appoint sole arbitrators in accordance with the provisions in the concession agreements; he appointed judge G Lagergren from Sweden, Professor R J Dupuy from France, and Dr S Mahmassani from Lebanon, respectively. The arbitration between Kuwait and Aminoil78 concerned a concession agreement concluded in 1948 for the exploration and exploitation of oil resources in the then Kuwait-Saudi Arabia Neutral Zone. In September 1977, Kuwait terminated the Concession agreement and nationalised Aminoil’s assets. The company started arbitration proceedings under the agreement. These proceedings where discontinued when Kuwait and Aminoil concluded an agreement establishing a specific ad hoc arbitration to determine, mainly, the amount of compensation due for the termination and the nationalisation. The Arbitral Tribunal was composed of Professor Paul Reuter from France, Professor Dr Hamed Sultan from Egypt and Sir Gerald Fitzmaurice, a British barrister and judge. All four awards confirm that, given the nature of the concession agreements and their specific object, they were not submitted to a specific national law, even less so the law of the host States, Libya or Kuwait. However, the awards differ to a large extent on the question of the law that had to be applied to the concession agreements, even though they are largely aligned on the legal outcome.

A. The Confirmation of the De-nationalisation of Concession Agreements The arbitrators seemed to agree that the respective concession agreements were not submitted to Libyan or Kuwaiti law, or indeed any other municipal law, exclusively. The three concession agreements entered into by Libya contained an identical provision on applicable law that was part of a standardised contract under the Libyan petroleum legislation: This Concession shall be governed by and interpreted in accordance with the principles of law of Libya common to the principles of international law and in the absence of such common principles then by and in accordance with general principles of law, including such of those principles as may have been applied by international tribunals.79

Lagergren considered that although this provision generates practical difficulties in its implementation, ‘it offers guidance in a negative sense by excluding the relevance of any single municipal system of law’.80 Dupuy also accepted that Libyan law alone could not determine the legal value and status of the concession; however, he did not reason in terms of the applicable law provision, but by reference to the nature of the agreement: [T]he Deeds of Concession in dispute are not controlled by Libyan law or, more exactly, are not controlled by Libyan law alone. It is incontestable that these contracts were international contracts, both in the economic sense because they involve the interests of international trade and in the strict legal sense because they include factors connecting them to different States …81

78 Government of Kuwait and American Independent Oil Company (Aminoil), Award (24 March 1982) reproduced in (2013) 66 International Law Reports 529–627 (hereafter ‘Aminoil, Award’). 79 See BP, Award, 303. 80 ibid. 81 Texaco-Calasiatic, Award on the Merits, 441, para 22.

Ad hoc Investment Arbitration Based on State Contracts 33 In the Aminoil case, the Tribunal had been specifically entrusted by the parties to determine the law governing the substantive issues ‘having regard to the quality of the Parties, the transnational character of their relations and the principles of law and practice prevailing in the modern world’.82 The Tribunal considered that it had to take into account the law chosen by the parties to the Concession agreement which referred to the principles common in the laws of Kuwait and the State of New York, or, in the absence of such common principles, ‘general principles of law normally recognised by civilised States in general’. Comparing this provision with those in similar concession agreements entered into by Kuwait, the Tribunal concluded that ‘it must have been the general principles of law that were chiefly present in the minds of the Government of Kuwait and its associates’.83 Finally the Tribunal noted: Article III, 2, with good reason, makes it clear that Kuwait is a sovereign State entrusted with the interests of a national community, the law of which constitutes an essential part of intracommunity relations within the State. At the same time, by referring to the transnational character of relations with the concessionaire, and to the general principles of law, this Article brings out the wealth and fertility of the set of legal rules that the Tribunal is called upon to apply.84

The aim to internationalise the contractual relationship and to exclude the sole application of the law of the State of the investment is (no longer) limited to supplementing an otherwise unreliable or underdeveloped system of law.85 Referring to previous arbitral decisions on State contracts, Professor Dupuy explained: It should be noted that the invocation of the general principles of law does not occur only when the municipal law of the contracting State is not suited to petroleum problems. Thus, for example, the Iranian law is without doubt particularly well suited for oil concessions but this does not prevent the contracts executed by Iran from referring very often to these general principles. The recourse to general principles is to be explained not only by the lack of adequate legislation in the State considered (which might have been the case, at one time, in certain oil Emirates). It is also justified by the need for the private contracting party to be protected against unilateral and abrupt modifications of the legislation in the contracting State: it plays, therefore, an important role in the contractual equilibrium intended by the parties.86

In other words, only the internationalisation of the contractual relationship was considered apt to provide sufficient protection to the private contracting party against the illegitimate use by the State party of its municipal law – over which it remains the master – to the detriment of the concessionaire’s rights.87 To use the words of Professor Pierre Mayer, a peculiar feature of state contracts is to ‘neutraliser le pouvoir normatif de l’État’.88 This still constitutes one of the raisons d’être of modern international investment law: protecting the investor against unjustified State interference and creating an internationally protected and predictable legal regime.

82 Arbitration

Agreement, Article III(2), in Aminoil, Award, 561.

83 ibid. 84 ibid. 85 See

n 49 above and accompanying text. Award on the Merits, 453–454, para 42. also Sapphire, Award, 176 (‘It is in the interest of both parties to such agreements that any disputes between them should be settled according to the general principles universally recognized and should not be subject to the particular rules of national laws, which are very often unsuitable for solving problems concerning the rights of the State where the contract is being carried out, and which are always subject to changes by this State and are often unknown or not full known to one of the contracting parties.’). 88 P Mayer, ‘La neutralisation du pouvoir normatif de l’État en matière de contrat d’État’ (1986) 113 Clunet 5–78. 86 Texaco-Calasiatic, 87 See

34 Daniel Müller B. The Uncertainties Concerning the (System of) Law Applicable to the Concession Contracts The arbitrators in the Libyan arbitrations however differed quite substantially on the question of what system of law should be applied to the concession agreements, whether or not Libya breached those agreements and with what consequences.89 Professor Dupuy took the view that ‘the Deeds of Concession in dispute are within the domain of international law and that this law empowered the parties to choose the law which was to govern their contractual relations’.90 He then continued to validate the choice of law operated by the Parties in the concession agreement and added that ‘the reference which is made mainly to the principles of international law and, secondarily, to the general principles of law must have as a consequence the application of international law to the legal relations between the parties’.91 However, with the notable exception of the basic principle of pacta sunt servanda, it remains difficult to extract from the award – or from public international law as it then was, and probably still is – the content of this ‘international law of contracts’.92 Dupuy also tried to anticipate any criticism of his decision by explaining that ‘to say that international law governs contractual relations between a State and a foreign private party neither means that the latter is assimilated to a State nor that the contract entered into with it is assimilated to a treaty’.93 Moreover, he also analysed both Libyan law and international law, in particular in respect of the question of restitutio in integrum.94 Lagergren relied more closely on the choice of law of the parties and refused to apply only public international law. In particular, he refused to accept that ‘general principles of law’ are the same thing as ‘public international law’.95 His award also shows the difficulties created by comparing public international law solutions to questions of private law relationships and contract law.96 Mahmassani pointed out that Libyan law included common rules and principles with international law and that these rules are therefore in harmony.97 A similar position was taken by the Aminoil Tribunal, which pointed out that international law and general principles of law were part of the law of Kuwait.98 However, in all four cases the tribunals accepted that the nationalisations of the contracts gave rise to full compensation of the value of the concessions,99 even if they differed markedly on the legal bases of these conclusions. Lagergren and Dupuy considered that Libya had

89 See also B Stern, ‘Trois arbitrages, un même problème, trois solutions: les nationalisations pétrolières libyennes devant l’arbitrage international’ [1980] Revue de l’arbitrage 1–43. 90 Texaco-Calasiatic, Award on the Merits, 450, para 35. 91 ibid 453, para 41. 92 J Verhoeven, ‘Droit international des contracts et droit des gens’ [1978–1979] Revue belge de droit international 229. 93 Texaco-Calasiatic, Award on the Merits, 457, para 46. 94 ibid 495–509. 95 BP, Award, 329. 96 ibid 346 (the arbitrator accepting that the comparison of the ‘principle of the continuing validity of the agreement rests only on a basis of extreme generality and has never been fully considered in the context of facts such as those which are at issue here where one party is a sovereign State’). 97 LIAMCO, Award, 175. 98 Aminoil, Award, 561–562. 99 Only Professor Dupuy accepted and ordered restitutio in integrum and the re-establishment of the contractual relationship, which of course did not happen.

Ad hoc Investment Arbitration Based on State Contracts 35 breached the concession agreements.100 Mahmassani and the Aminoil Tribunal reasoned in terms of lawful nationalisation.101

IV. CONCLUSION

These arbitrations concerning state contracts, and in particular concession agreements, have certainly shaped the discussion around a necessarily internationalised régime for the protection of foreign investments. The reason seems obvious. Only a non-national legal régime can provide efficient protection to foreign investors and their investments as long as states can freely adopt and modify their municipal laws, including to the detriment of investors. Internationalised protection is therefore a guarantee against the overarching sovereignty of states. These arbitrations have also shown that an internationalised régime does not necessarily imply the application of public international law to investment contracts. This was not and is not the purpose of general international law. As Lord McNair put it in 1957: Although it cannot be said that this type of contract is governed by public international law, it is public international law which has given birth to the legal system gradually being recognized as appropriate to it, for Article 38, paragraph I (c), of the Statute of the International Court of Justice which forms the title of this article, merely places on record one of the main sources of the rules of public international law. Indeed, it describes the inexhaustible reservoir of legal principles from which tribunals can enrich and develop public international law. In other words, it is submitted that the legal system appropriate to the type of contract under consideration is not public international law but shares with public international law a common source of recruitment and inspiration, namely, ‘the general principles of law recognized by civilized nations’.102

If these arbitrations have added anything to the substantive law of foreign investments and the protection of investment contracts, it is probably the proposition that a state cannot use its sovereign powers to free itself from its obligations and undertakings easily. Even if a state always remains at liberty to repudiate its contractual engagements, it must compensate the investor; this is an important condition. This solution was and still is dictated by international investment law and constitutes the forceful protection of investors then and now. The so-called ‘umbrella clauses’ regularly included in bilateral and multilateral investment instruments constitute an expression of this principle in modern international investment law.

100 BP,

Award, 329; Texaco-Calasiatic, Award on the Merits, 511. Award, 197; Aminoil, Award, 591. McNair, ‘The General Principles of Law Recognized by Civilized Nations’ (1957) 33 British Yearbook of International Law 6. 101 LIAMCO, 102 A

4 The World Court’s Influence on Contemporary Investment Law CHRISTIAN J TAMS AND ELENI METHYMAKI*

I. INTRODUCTION

T

HE WORLD COURT is not an investment court. Over the course of the last century, it has addressed inter-state disputes that reflect the breadth and diversity of international law, from armed activities in the Congo and Nicaragua to questions concerning the guardianship of infants. Only a fraction of its proceedings concern disputes about state interference with foreign investments. And yet, the decisions of the Permanent Court of International Justice (PCIJ) and the International Court of Justice (ICJ) feature prominently in investment jurisprudence: according to Fauchald’s empirical study, one in two awards refers to it in one way or another.1 Alain Pellet agrees, noting that ICSID tribunals ‘systematically refer to the Court’s jurisprudence’ and ‘show a particular deference to it’.2 Many of these references reflect the embeddedness of international investment law (IIL) and arbitration in general international law: PCIJ and ICJ statements on principles of interpretation are widely cited, as is their case law on procedural issues common to all dispute settlement regimes, on the legal effects of interim measures, or on meta-questions such as the requirements for the identification of customary international law.3 Our discussion in the following leaves these aspects to one side. It adopts a narrower focus, asking how the World Court has contributed to IIL as a special regime of contemporary international law. We identify three main contributions. First, the World Court’s jurisprudence has accentuated limits to investment protection under general international law and Friendship Commerce and Navigation (FCN) treaties. This may have been one factor favouring the emergence of a special regime of international investment agreements (IIAs). Second, occasional ICJ and PCIJ pronouncements offer clues as to the interpretation of select IIA provisions.

* Christian J Tams is Professor of International Law at the University of Glasgow and an academic member of Matrix Chambers, London; Eleni Methymaki is a DPhil in Law candidate at the University of Oxford and a Research Associate at the Institute of Sustainability Governance, Leuphana Universität Lüneburg. 1 OK Fauchald, ‘The Legal Reasoning of ICSID Tribunals – An Empirical Analysis’ (2008) 19 European Journal of International Law 301, 340 (covering 98 decisions rendered between 1998–2006). 2 A Pellet, ‘The Case Law of the ICJ in Investment Arbitration’ (2013) 28 ICSID Review 223, 230 and 239. 3 eg Saluka Investments BV v Czech Republic, PCA, Partial Award (17 March 2006) para 254; Hochtief AG v Republic of Argentina, Decision on Jurisdiction (24 October 2011) ICSID Case No ARB/07/31, para 95; United Parcel Service of America Inc v Canada, UNCITRAL, Award on Jurisdiction (22 November 2002) para 84; and further Pellet, above (n 2) 231–32.

38 Christian J Tams and Eleni Methymaki Investment tribunals rely on such pronouncements as additional authority; however, the World Court’s influence remains limited. Third, pronouncements by the PCIJ laid down general principles of a regime on reparation, which the International Law Commission (ILC) later consolidated. In the absence of express provisions in investment treaties, this general regime has shaped investment jurisprudence on damages.

II. THE WORLD COURT AND THE EMERGENCE OF CONTEMPORARY INVESTMENT LAW

The World Court’s first contribution concerns the architecture of contemporary IIL, defined by a large number of IIAs and a dedicated system of investor-state dispute settlement. The reasons leading to the emergence of this special regime are manifold. In this process, the World Court’s jurisprudence may have played some role. It emphasised limits to investment protection based on general international law and traditional instruments of protection – which suggested that effective protection would better be based on a special regime. The World Court’s jurisprudence, as described in these introductory lines, is not a systematic body of case law. It comprises a handful of decisions in diverse investment disputes. These reached the Court via the ordinary jurisdictional channels – ie based on special agreements, compromissory clauses or optional clause declarations.4 They reached it as inter-state cases, brought by the investor’s state of nationality by way of diplomatic protection.5 Compared to contemporary investor-state dispute settlement, this process (itself shaped by the World Court’s jurisprudence6) was cumbersome. But occasionally, home states did bring proceedings on behalf of their investors. The ICJ’s record in dealing with such disputes was not encouraging to investors – according to FA Mann, it was ‘particularly chilling’.7 Two aspects stand out: the cautious handling of claims based on internationalised contracts and the Court’s restrictive approach to shareholder rights.

A. A Cautious Approach to Internationalised Contracts The first aspect takes us into a by-gone era, of investment protection via internationalised contracts. For nearly half a century, claims that investment contracts should be governed by international law were a central feature of debates. In the pre-BIT era, this seemed an ingenuous strategy to ringfence contractual rights against host state interference.8 In the Lena Goldfields arbitration, this strategy paid off, as an arbitral tribunal considered a private

4 Statute

of the International Court of Justice, 33 UNTS 993, Arts 36 and 37 (ICJ Statute). Art 34. 6 See notably Mavrommatis Palestine Concessions (Greece v UK) (Jurisdiction) PCIJ Ser A No 2. 7 FA Mann, ‘Foreign Investment in the International Court of Justice: The ELSI Case’ (1992) 86 American Journal of International Law 92. 8 For insightful contributions see, eg M Kamto, ‘La notion de contrat d’Etat: une contribution au débat’ [2003] Revue de l’Arbitrage 719; E Paasivirta, Participation of States in International Contracts and Arbitral Settlement of Disputes (Helsinki, Lakiniesliiton Kustannus, 1990); J Ho, State Responsibility for Breaches of Investment Contracts (Cambridge, Cambridge University Press, 2019) offers a comprehensive account. According to Sornarajah, the internationalisation of contracts could be seen as ‘a continuation of … the structures of extractive colonialism’, see M Sornarajah, ‘The Battle Continues: Rebuilding Empire through Internationalization of State Contracts’ in J von Bernstorff and P Dann (eds), The Battle for International Law: South-North Perspectives on the Decolonization Era (Oxford, Oxford University Press, 2019) 175, generally and 190. 5 ibid

The World Court’s Influence on Contemporary Investment Law 39 concession agreement to be governed partly by international law: ‘a gigantic first step …, almost equivalent to the caveman’s discovery of fire’.9 In the subsequent evolution of the internationalisation movement – from the gold fields of Siberia via oil-rich Libya into today’s ‘noble ventures’ of umbrella clause litigation before ICSID tribunals10 – the ICJ did not play the lead role, but made one crucial pronouncement. This came in the 1951–1952 Anglo-Iranian Oil Co. case, brought by the UK in response to Iran’s unilateral termination of a pre-war oil concession. Britain relied on Iran’s optional clause declaration, submitted in 1932 and limited to disputes about treaties and conventions ‘subsequent to the ratification of this declaration’.11 According to the UK, the 1933 concession was such a subsequent treaty: it was … at once a concessionary contract between the Iranian Government and the Company and a treaty between the two Governments. … [I]t must be considered to be within the meaning of the term ‘treaties or conventions’ contained in the Iranian Declaration.12

This was, as James Crawford has observed, ‘a deliberate attempt to internationalize development contracts in the context of the oil industry’ and to bring them within the ICJ’s jurisdiction.13 However, the ICJ ‘would have none of it’.14 Adopting a ‘formalist interpretive approach’,15 it viewed the concession as nothing more than a concessionary contract between a government and a foreign corporation. The United Kingdom Government is not a party to the contract …. Under the contract the Iranian Government cannot claim from the United Kingdom Government any rights which it may claim from the Company, nor can it be called upon to perform towards the United Kingdom Government any obligations which it is bound to perform towards the Company.16

The upshot of this, as noted by Doreen Lustig, was that ‘corporations, even if significantly owned by states, … were excluded from the (formal and informal) interference of international legal institutions’ operating in the state-to-state world.17 This did not mark the end of the internationalisation movement. But it foreclosed a potential jurisdictional variant, and it entrenched the divide between (internationalised) contracts and treaties. Since 1952, that point has not been re-argued. The ICJ’s decision in Anglo-Iranian Oil Co. contained the ‘fire’ discovered in Lena Goldfields.

9 See VV Veeder, ‘The Lena Goldfields Arbitration: The Historical Roots of Three Ideas’ (1998) 47 International & Comparative Law Quarterly 747, 772. See also D Müller, ‘Ad hoc Investment Arbitration Based on State Contracts: From Lena Goldfields to the Libyan Oil Arbitrations’, ch 3 in this volume. 10 Jean Ho distinguishes between ‘panoptic internationalisation’ (drawing on the wider context) and the more recent form of ‘umbrella clause internationalisation’ (drawing on a treaty provision): Ho, above (n 8) 181. Our discussion remains ‘panoptic’. 11 See Anglo-Iranian Oil Co (UK v Iran) (Jurisdiction) [1952] ICJ Rep 93, 103–07 (hereinafter Anglo-Iranian Oil Co) and section III(A) below. For a comprehensive account see D Lustig, Veiled Power: International Law and the Private Corporation, 1886–1981 (Oxford, Oxford University Press, 2020) 144–78. 12 Anglo-Iranian Oil Co, above (n 11) 111–12. 13 J Crawford, ‘International Protection of Foreign Direct Investments: Between Clinical Isolation and Systemic Integration’ in R Hofmann and CJ Tams (eds), International Investment Law and General International Law (Baden-Baden, Nomos, 2011) 19. This attempt was significantly bolstered by the fact that the British Government had been closely involved in the negotiation of the 1933 concession and was the largest shareholder in the company. 14 ibid. 15 Lustig, above (n 11) 169. 16 Anglo-Iranian Oil Co, above (n 11) 112. 17 Lustig, above (n 11) 177.

40 Christian J Tams and Eleni Methymaki B. Strict Limitations on Shareholder Claims under General International Law The second limitation is more obvious. It concerns the capacity of states to exercise diplomatic protection on behalf of investors. As noted above, in proceedings before the World Court, nationality is the nexus that transforms a wrong done to an individual or legal person into an injury of the state. Looked at from the perspective of IIL, one aspect of the World Court’s jurisprudence spelling out the requirements of this nexus stands out. Through a series of decisions, the Court has clarified that under general international law, only a particular type of commercial interest can be taken up in diplomatic protection claims. More specifically, the general regime protects interests of a corporation or company, but only exceptionally interests in a corporation or company, including those of shareholders. As the ILC would later note: The most fundamental principle of the diplomatic protection of corporations is that a corporation is to be protected by the State of nationality of the corporation and not by the State or States of nationality of the shareholders.18

i. The ICJ’s Pronouncements on Shareholder Protection This ‘fundamental principle’ was articulated in the ICJ’s 1970 Barcelona Traction judgment.19 In it, the Court rejected claims, by Belgium, in respect of harm done to a Canadian company controlled by Belgian shareholders. This seemingly straightforward proposition (a state cannot bring claims on behalf of foreign firms) was occasioned by a dramatic dispute: a Spanish businessman, with the help of Spanish authorities, succeeded in taking over a foreign-owned utility company.20 The judgment, rendered after decades of pleadings on the merits, turned on the technical question of Belgium’s (lack of) standing, which the Court rejected in a three-step reasoning.21 First, drawing support from the approach in national legal systems,22 the ICJ insisted on the distinction between a corporation and its shareholders: The former ‘enjoys an independent existence … [T]he interests of the shareholders are both separable and indeed separated from those of the company’.23 Second, the ICJ clarified that under international law, a company possessed the nationality of the state in which it was incorporated, provided there was some tangible connection.24 On that basis, there were two relevant nationalities: that of the company (Canadian), and that of its controlling shareholders (Belgian).

18 ILC,

Draft Articles on Diplomatic Protection (2006) II(2) YbILC 39, para 1 (Art 11). Traction, Light and Power Company, Limited (Belgium v Spain) (New Application: 1962) (Judgment) [1970] ICJ Rep 3 (hereinafter Barcelona Traction). As with the Anglo-Iranian Oil Co case, Lustig’s account is instructive, see Lustig, above (n 11) 201–5. 20 For details see J Brooks, ‘Annals of Finance: Privateer’ [1979] The New Yorker 42. 21 The following builds on CJ Tams and A Tzanakopoulos, ‘Barcelona Traction at 40: The ICJ as an Agent of Legal Development’ (2010) 23 Leiden Journal of International Law 781. 22 Barcelona Traction, above (n 19) para 38: as ‘international law had not established its own rules’, it was ‘called upon to recognize’ the domestic distinction between corporation and shareholder. 23 ibid para 45. 24 The Court did not require a ‘genuine link’ but spoke of a ‘close and permanent connection’ between company and state: see Barcelona Traction, above (n 19) para 71; cf Nottebohm (Liechtenstein v Guatemala) (Second Phase) [1955] ICJ Rep 4, 23. 19 Barcelona

The World Court’s Influence on Contemporary Investment Law 41 Third, crucially, harm suffered by a company and its shareholders was to be exclusively allocated to the former: ‘[A]lthough two entities may have suffered from the same wrong, it is only one entity whose rights have been infringed’.25 Moreover, in case of ‘an unlawful act committed against a company representing foreign capital, the general rule of international law authorizes the national State of the company alone to make a claim’.26 This third step in particular simplified the claims process. It meant that interests in a company were absorbed into those of the company. The Court justified this, inter alia, by warning that ‘opening the door to competing diplomatic claims, could create an atmosphere of confusion and insecurity in international economic relations’.27 Yet under the Court’s ‘rigid’28 approach, the right to bring claims was ‘divorced’ from the underlying interests.29 What remained, for shareholders (and their states of nationality) to pursue were limited rights directly vested in them as shareholders, among them ‘the right to any declared dividend, the right to attend and vote at general meetings [and] the right to share in residual assets of the company upon liquidation’30 – rather meagre substitutes. All this, to reiterate, reflected the position under general international law (as seen by the ICJ). Voluntary agreements could go further, and increasingly did: the Court noted that, ‘whether in the form of multilateral or bilateral treaties between States, or in that of agreements between States and companies, there has since the Second World War been considerable development in the protection of foreign investments’ (as well as in regional human rights law).31 However, ‘the [conventional] law on the subject has been formed in a period characterized by an intense conflict of systems and interests’32 and did not easily trickle down into general international law. Half a century onwards, the ‘fundamental principle’33 remains central. The ICJ affirmed it in the Diallo case, in which Guinea espoused claims of a Guinean businessman imprisoned in the Democratic Republic of the Congo (DRC).34 While Guinea could assert violations of Mr Diallo’s individual rights, the key question was whether it could also defend his business interest as sole shareholder of two companies incorporated in the DRC. Unlike in Barcelona Traction, the claims in Diallo were brought against the state of incorporation (DRC), which had allegedly violated rights of a national company, and Guinea argued that it could exercise diplomatic protection by substitution. In its judgment, the Court rejected Guinea’s claim. First, as per Barcelona Traction, company rights were to be taken up by the company’s state of nationality, not by that of the shareholder. Second, this was true even in claims directed against the state of nationality: there was insufficient evidence ‘at least at the present time … [of] an exception in customary international law allowing for protection by substitution’.35

25 Barcelona

Traction, above (n 19) para 44. para 88. para 49. 28 RB Lillich, ‘Two Perspectives on the Barcelona Traction Case: The Rigidity of Barcelona’ (1971) 65 American Journal of International Law 522, 523. 29 See Sep Op Fitzmaurice, Barcelona Traction, above (n 19) 77–78. 30 Barcelona Traction, above (n 19) paras 46–47. 31 ibid paras 90 and 91. 32 ibid para 89. 33 ILC, Draft Articles on Diplomatic Protection, above (n 18). 34 Ahmadou Sadio Diallo (Guinea v DRC) (Preliminary Objections) [2007] ICJ Rep 582 (hereinafter Diallo). 35 ibid paras 89–90. 26 ibid 27 ibid

42 Christian J Tams and Eleni Methymaki The fact that, in between Barcelona Traction and Diallo, contemporary IIL had blossomed, left the Court unperturbed: the fact that various international agreements … have established special legal regimes governing investment protection, … is not sufficient to show that there has been a change in the customary rules of diplomatic protection; it could equally show the contrary.36

Diallo thus marked a divide between the general regime of claims premised on injury to companies, and the ‘special legal regimes governing investment protection’.37 ii. Their Influence on Contemporary Investment Law The combined impact of the Barcelona Traction and Diallo judgments has been significant. Both reflect a restrictive approach to the protection of commercial interests under general international law, which today is widely accepted. But this acceptance has come at a price: the general regime has been hollowed out by the decade-long practice of agreeing on ‘special legal regimes governing investment protection’.38 The World Court’s jurisprudence provides no more than a faint echo of this – but an echo there is: in the ELSI case, brought on the basis of the 1948 US-Italian FCN Treaty, a chamber of the Court recognised the US right to act on behalf of a US company in respect of injury sustained by its Italian subsidiary.39 While the US claims failed on the merits, one cannot help but notice how little space the chamber devoted to the question of standing. The most plausible reason for this is that the chamber considered the FCN Treaty to recognise a broader range of shareholder rights than general international law: seen in this light, ELSI illustrates how particular treaties protecting commercial interests of foreigners can opt out of the general regime set out in Barcelona Traction and Diallo.40 The real hollowing-out has taken place elsewhere, though: it is a result of the ‘BIT revolution’ in IIL. A quick glance at the definitional clauses of standard IIAs is sufficient to realise how decisively contemporary IIL has moved away from the general approach to corporate nationality. Shares in a company are inevitably included among the assets protected as investments;41 and the heated discussions about outer limits of protected interests are a world away from Belgium’s and Spain’s arguments about standing in Barcelona Traction. Half a century onwards, the special legal regimes of investment protection have quite obviously left behind anything remotely resembling the ICJ’s ‘rigidity’42 in construing the general law of

36 ibid para 90. For the earlier debate see, eg, A Al Faruque, ‘Creating Customary International Law Through Bilateral Investment Treaties. A Critical Appraisal’ (2004) 44 Indian Journal of International Law 292; SM Schwebel, ‘Investor-State Disputes and the Development of International law – The Influence of Bilateral Investment Treaties on Customary International Law’ (2004) 98 American Society of International Law Proc 27. 37 Diallo, above (n 34) para 90. The Court followed up on this divide (in terms of a substantive rule this time) more recently in Obligation to Negotiate Access to the Pacific Ocean (Bolivia v Chile) [2018] ICJ Rep 507, para 162. 38 Diallo, above (n 34) para 90. 39 Elettronica Sicula SpA (USA v Italy) (Judgment) [1989] ICJ Reports 15 (ELSI). As judge Oda noted in his Separate Opinion (ibid 83) the distinction between shareholder rights and rights of the company – ‘so clearly expounded’ in Barcelona Traction – required fuller analysis. 40 For pertinent comment see, eg, Z Douglas, The International Law of Investment Claims (Cambridge, Cambridge University Press, 2009) 408–13. 41 Pars pro toto, see Art 1(1)(b) of the 2008 German Model BIT; Art 1 of the 2014 Canada Model FIPA (‘investment’); Art 1(b)(ii) of the 2003 Indian Model BIT; Art 1(2)(a) of the 2000 Turkey Model BIT. 42 Lillich, above (n 28).

The World Court’s Influence on Contemporary Investment Law 43 claims. At the same time, it is difficult to avoid the impression that investment jurisprudence, ‘by opening the door to competing [investment] claims’ by a wide range of stakeholders, has indeed ‘create[d] an atmosphere of confusion and insecurity in international economic relations’.43 Whatever the difference, courts and tribunals administering the two separate regimes are aware of the gap. Just as the ICJ, in Diallo, felt safe to treat the special regime of investment protection as an aliud, so investment tribunals have not felt constrained by general rules of nationality. The number of awards that briefly mentions Barcelona Traction, only to move on from it immediately, is legion. The statement in Camuzzi reflects a common approach: ‘whatever may have been the merits of Barcelona Traction, that case was concerned solely with the diplomatic protection of nationals by their State, while the case here disputed concerns the contemporary concept of direct access for investors to dispute resolution by means of arbitration between investors and the State’.44 Similarly, in Teinver, the Tribunal found ‘Barcelona Traction’s discussion of shareholder rights … inapposite to the circumstances of the present case … as the [ICJ’s] decision was made in the absence of the specific framework of a BIT’.45 The net result, aptly described by Pellet, is that ‘with regard to the protection of shareholders, ICSID tribunals unanimously greet the jurisprudence of the ICJ, … but … then hide behind the lex specialis dogma’.46 Fifty years after Barcelona Traction, and 30 years after the beginning of treaty-based investment arbitration, two separate regimes have emerged, and their respective starting points for determining corporate nationality are worlds apart.

C. Interim Conclusions The preceding sections cover two very different questions, but in one respect, point in the same direction. They show that the World Court, in a number of decisions, was not favourably disposed towards claims brought on behalf of foreign investors. In Anglo-Iranian Oil Co., the Court gave short shrift to the UK’s attempt to portray an internationalised contract claim as a treaty dispute: an attempt to bring contractual rights within the World Court’s jurisdictional remit ended before it had really begun. In Barcelona Traction and Diallo, the Court admitted diplomatic protection claims in defence of commercial interests only within narrow limits, notably dismissing the right of shareholders (and their states of nationality) to present claims for harm done to the company. In what respect do these holdings matter to contemporary IIL? As we noted at the outset, their effect is elusive: we do not seek to argue that there is a direct link between the ICJ’s refusal to accept claims in a handful of disputes, and the emergence of IIL in its contemporary variation. Such a claim would require a lot more empirical work. But we would advance a more modest claim, which echoes Mann’s statement cited above: in the pre-IIA period, outside instances of contract-based arbitration, ‘international litigation arising from disputes of a commercial character [was] … far from encouraging to investors’.47 The ICJ’s decisions

43 Barcelona 44 Camuzzi

Traction, above (n 19) para 49. International SA v Argentina, Decision on Jurisdiction (11 May 2005) ICSID Case No ARB/03/2,

paras 141–42. 45 Teinver et al v Argentina, Decision on Jurisdiction (21 December 2012) ICSID Case No ARB/09/1, para 218. 46 Pellet, above (n 2) 234. 47 Mann, above (n 7) 92 (Mann adds ‘unfortunately’).

44 Christian J Tams and Eleni Methymaki ‘highlighted some of the procedural and substantive inadequacies with the diplomatic protection model in safeguarding shareholder interests’.48 This did not mean that IIL had to enter its BIT era: its ‘incremental, partly accidental emergence’ reflects a curious mix of ‘Rational Design [and] Accidental Evolution’.49 But the experience of occasional World Court litigation may have prompted those interested in enhancing investment protection to pursue new paths. The new path eventually chosen has certainly led away from Lena Goldfield contract-based arguments: ‘The BIT phenomenon involves … a rejection of the internationalization of contracts’.50 As far as shareholder rights are concerned, the ICJ’s restrictive approach may well have been an incentive to seek solutions via special treaties: in Lustig’s phrase, ‘the Court [in Barcelona Traction] effectively left investors with little choice but to look elsewhere for protection and seek remedy in alternative forums’.51 This certainly was the solution favoured by Mann who (writing after ELSI) placed his faith in a new generation of treaties: ‘Probably the most important lesson is that the treaties have to be so drafted as to entitle the foreign investor to make claims in respect of acts suffered by a domestic company substantially owned by him’.52 For better or worse, this call has been heeded. The number of treaties so entitling foreign investors – around 300 at the time of the ELSI judgment – has increased almost ten-fold since then, and BITs and other IIAs have come to dominate contemporary IIL. Very few of them envisage recourse to the ICJ. So, does the World Court’s jurisprudence hold any lesson for their interpretation? To this question we now turn.

III. THE ICJ AND TREATY STANDARDS

In assessing the ICJ’s influence on standards contained in IIAs, we leave the ICJ’s ‘home turf ’. IIAs make up a special regime of investment protection, based on formally separate but closely related treaties. This special regime was meant to overcome the perceived limitations of traditional international law partly shaped by the ICJ. In this realm of IIAs, the ICJ is not accorded any formal role – and as Diallo indicates, the Court is fully aware of this. Instead, IIAs rely on domestic litigation and international arbitration, which has become the main forum of clarification and legal development. All of this points to a limited role for the World Court in relation to treaty standards. It is limited not because investment tribunals or other law-interpreters do not trust the World Court – but for simple reasons of ‘supply and demand’. Especially for protection standards that are peculiar to IIAs, ‘demand’ is virtually non-existent: the more we move into the depths of written IIL, the more detailed the regulation, the denser the investment jurisprudence, the less need and room there is for guidance from the World Court. ‘Supply’ is equally scarce: a century of World Court adjudication yields little on, say, the asset-based definition of an ‘investment’ and does not clarify whether, in the absence of specific representations, an investor can legitimately rely on the stability of the host state’s regulatory framework. 48 A

Newcombe and L Paradell, Law and Practice of Investment Treaties (Alphen aan den Rijn, Kluwer, 2009) 39. Pauwelyn, ‘Rational Design or Accidental Evolution? The Emergence of International Investment Law’ in Z Douglas, J Pauwelyn and H Vinuales (eds), The Foundations of International Investment Law (Oxford, Oxford University Press, 2014) 42 and 11. 50 Crawford, above (n 13) 19. 51 Lustig, above (n 11) 214. See further Newcombe and Paradell, above (n 48) 38: the ‘uncertainties and inadequacies [of ICJ decisions] may have provided compelling rationales for the development of IIAs’. See also S El Boudouhi, ‘Barcelona Traction Re-Imagined: The ICJ as a World Court for Foreign Investment Cases?’ in I Venzke and KJ Heller (eds), Contingency in International Law (Oxford, Oxford University Press, 2021) 406. 52 Mann, above (n 7) 100. 49 J

The World Court’s Influence on Contemporary Investment Law 45 Pellet’s overall assessment reflects these considerations: ‘generally speaking’, he notes that investment tribunals are more likely to refer to the World Court when addressing ‘procedural issues … or questions of general international law rather than when dealing with investment protection standards’.53 Yet matters are not absolute. At least some IIA provisions do refer to general international law or have put into written form standards that pre-exist the BIT era; and many of them are formulated in an open-ended manner. In construing such provisions, occasional pronouncements by the World Court can inform the analysis. The subsequent sections illustrate this process by identifying two representative World Court contributions.54

A. Most-favoured Nation Clauses and Dispute Settlement Debates about the scope of MFN clauses provide a first example. Whether the MFN clause of an investment treaty can be invoked to broaden access to dispute settlement is much discussed among investment lawyers and tribunals.55 These debates often rely on treaty-specific and investment-specific arguments – including the wording of clauses and the context in which they appear. However, both sides in the debate have anchored their position to a broader normative context and as part of that, have considered three ICJ judgments: Ambatielos, US Nationals in Morocco and Anglo-Iranian Oil Co. i. ICJ Pronouncements Anglo-Iranian Oil Co. has been mentioned already, insofar as it reflected the ICJ’s cautious take on internationalised contracts and the problems arising from the temporal limitation of Iran’s 1932 optional clause declaration.56 In addition to the 1933 oil concession, the UK also considered Iran to have violated bilateral treaties concluded in 1857 and 1903 – but these obviously predated Iran’s declaration. The UK sought to bypass the temporal limitation by relying on two other treaties concluded by Iran with Denmark and Turkey in 1934 and 1937 respectively, coupled with the broadly worded MFN clauses contained in its own treaties of 1857 and 1903.57 The Court was not convinced. It held that, in applying Iran’s optional clause, the focus had to be on the treaty containing the MFN clause – this was the ‘basic treaty’, and it had to postdate the optional clause declaration.58 In addition, the Court rejected the British argument that if Denmark could bring disputes before the Court by virtue of its treaty with Iran and the UK could not, then the UK would not enjoy MFN status: ‘the most-favoured-nation clause in

53 Pellet,

above (n 2) 239–40.

54 Our discussion looks at World Court pronouncements that have left a tangible legacy in investment jurisprudence;

it is illustrative rather than exhaustive. In particular, the following discussion does not explore the fate of important PCIJ pronouncements on the scope of expropriation, among them the clarification that an interference with ‘favourable business conditions and goodwill’ did not amount to an expropriation (Oscar Chinn (UK v Belgium) PCIJ Series A/B No 63, 88), but that intangible rights were in principle protected (Certain Polish Interests in Polish Upper Silesia (Germany v Poland) (Merits) PCIJ Ser A No 7, 44). 55 See, eg, Z Douglas, ‘The MFN Clause in Investment Arbitration: Treaty Interpretation off the Rails’ (2011) 2 Journal of International Dispute Settlement 97; S Schill, ‘Allocating Adjudicatory Authority: Most-Favoured Nation Clauses as a Basis of Jurisdiction – A Reply to Zachary Douglas’, ibid 355. For a recent account see E Stoppioni, ‘Jurisdictional Impact of Most Favoured Nation Clause’ (January 2017) MPEiPro. 56 See section II(A). 57 Anglo-Iranian Oil Co, above (n 11) 108–9. 58 ibid 109–10.

46 Christian J Tams and Eleni Methymaki the Treaties of 1857 and 1903 between Iran and the United Kingdom has no relation whatever to jurisdictional matters’.59 This would remain the only express ICJ pronouncement on MFN treatment and the establishment of jurisdiction. The two other cases looked at MFN clauses from the perspective of substantive rights. The US Nationals in Morocco case concerned the scope of US consular jurisdiction over its citizens and protégés in the French Zone of Morocco. The US claimed that through the MFN clauses in its 1836 Treaty with Morocco it was entitled to enjoy comprehensive consular jurisdiction, as Morocco had granted such to Great Britain and Spain in later treaties. This argument was bound to fail. At the time of the dispute, Great Britain and Spain had already renounced their treaty rights – so, the US claim would have privileged it over Great Britain and Spain, whereas MFN clauses aimed at ‘fundamental equality without discrimination’, not preferential treatment.60 And in that, they could not operate as ‘a form of drafting by reference’.61 The ICJ’s pronouncement in the Ambatielos62 case was more limited still. As the Court noted, Greece and the UK disagreed on the scope of a clause promising MFN treatment in ‘all matters relating to commerce and navigation’: Greece relied on the clause to invoke provisions in treaties between the UK and third states relating to the administration of justice, whilst the UK claimed the clause could not be extended beyond its express subject-matter.63 As in US Nationals, the MFN clause was invoked to ‘import’ into the basic treaty substantive rights. Unlike in US Nationals, the ICJ in Ambatielos left the matter open, holding that the dispute was to be decided by a Commission of Arbitration. However, in their joint dissenting opinion, judges McNair, Basdevant, Klaestad and Read upheld the British view.64 (By contrast, the Commission of Arbitration, held that a limited MFN clause could cover administration of justice, as the issues could be viewed as belonging to the same general matter [the ejusdem generis rule]. However, that did not help Greece much: insofar as the UK had granted other nations a substantive right to treat their nationals fairly, it had merely agreed to apply in relation to them its national laws.65) ii. Their Reception in Investment Jurisprudence What has been the impact of this ‘MFN jurisprudence’ on investment arbitration? Two things are clear: First, the ICJ’s decisions have been referred to by arbitral tribunals, as part of a ‘general theory’ of MFN clauses.66 And second, such guidance has not resulted in a consistent jurisprudence; instead, tribunals continue to disagree on the proper reading of MFN clauses. Curiously, both sides in that debate have sought to anchor their findings in the ICJ’s jurisprudence. This has involved a fair share of misunderstandings, as the subsequent analysis of two landmark cases, Mafezzini v Spain and Plama v Bulgaria, illustrates.67 59 ibid

110. of Nationals of the United States of America in Morocco (France v USA) (Judgment) [1952] ICJ Rep 176, 192 (hereinafter US Nationals in Morocco). 61 ibid 191–92. 62 Ambatielos (Greece v UK) (Merits) [1953] ICJ Rep 10. 63 ibid 21–22. 64 ibid, Diss Op McNair, Basdevant, Klaestad and Read, 34. 65 The Ambatielos Claim (Greece v UK) RIAA XII (6 March 1956) 83, 106–9. 66 Wintershall v Argentina, Award (8 December 2008) ICSID Case No ARB/04/14, paras 96–106. ILC, Second Report on the most-favoured-nation clause, by Mr Endre Ustor, Special Rapporteur (9 March/18 May 1970) UN Doc A/CN.4/228 and Add.1, paras 6–7. 67 For a fuller account see Stoppioni, above (n 55) MN 7–8. 60 Rights

The World Court’s Influence on Contemporary Investment Law 47 Maffezini was the first investment case to accept that MFN clauses applied to conditions attached to the tribunal’s jurisdiction, and it did so even though the MFN clause was not self-standing, but was included in the provision guaranteeing fair and equitable treatment. Perhaps surprisingly, the Maffezini tribunal supported its approach by reference to the ICJ’s case law. It noted (correctly) that the MFN debate was ‘familiar to international lawyers and scholars’.68 Drawing on the Anglo-Iranian Oil Co. case, it used the concept of the ‘basic treaty’ to define the issues to which the MFN clause applies – and which, in the case of a BIT, comprised substantive rights and access to arbitration.69 Did this mean that the MFN clause applied to both, even though it was contained in one of the provisions guaranteeing substantive rights? The tribunal asserted that the ICJ had ‘indirectly … touched upon’ the question when considering the scope of US consular jurisdiction in US Nationals in Morocco. It did not mention the fact that consular jurisdiction as discussed in US Nationals in Morocco was a substantive right accorded to the home state, unrelated to dispute settlement. It did not mention either that in Anglo Iranian Oil Co., the ICJ had considered the MFN clause to ‘ha[ve] no relation whatever to jurisdictional matters’;70 and ignored the joint dissent of judges McNair, Basdevant, Klaestad and Read in the ICJ’s Ambatielos case. Instead, the Maffezini tribunal relied on the Commission of Arbitration’s report in Ambatielos – and from this, derived the very broad proposition that ‘the protection of the rights of persons engaged in commerce and navigation by means of dispute settlement provisions embraces the overall treatment of traders covered by the clause’.71 This rather selective engagement with the case law set the stage: ‘It is in the light of this background that the operation of the most favored nation clause in bilateral investment treaties must now be considered.’72 The rest, as they say, is history. If Maffezini heavily relied on (a selective reading of) the ICJ’s case law, the award in Plama73 was mostly tailored to the specifics of investment treaty interpretation. To this, the Tribunal added its own, more balanced, analysis of the ICJ cases: it did not bypass the ICJ’s most obvious dictum (viz that a MFN clause ‘ha[d] no relation whatever to jurisdictional matters’ in Anglo-Iranian Oil Co.74), and noted that ‘in the view of the International Court of Justice, a MFN provision does not operate as an automatic incorporation by reference’.75 The more equivocal approach in Ambatielos was acknowledged,76 but the Tribunal emphasised that the case concerned the ‘administration of justice, … in the sense of denial of justice in the domestic courts’ – not a jurisdictional issue.77 This analysis suggested that the earlier precedents ‘do not provide a conclusive answer to the question’78 – a cautious assessment, but sufficient to

68 Emilio Agustín Maffezini v Spain, Decision on Jurisdiction (25 January 2000) ICSID Case No ARB/97/7, para 43 (hereinafter Maffezini). 69 ibid para 45. In Garanti Koza LLP, dissenting Arbitrator Boisson de Chazournes reached the opposite result: Garanti Koza LLP v Turkmenistan, Decision on Jurisdiction (3 July 2013) Dissenting Opinion, ICSID Case No ARB/11/20, para 67 and fn 64. 70 Anglo-Iranian Oil Co, above (n 11) 110 [emphasis added]. 71 Maffezini, above (n 68) para 50. 72 ibid para 51. 73 Plama Consortium Ltd v Bulgaria, Decision on Jurisdiction (8 February 2005) ICSID Case No ARB/03/24 (hereinafter Plama). See already, equally briefly, Salini Construttori SpA and Italstrade SpA v Jordan, Decision on Jurisdiction (9 November 2004) ICSID Case No ARB/02/13, para 106. 74 Plama, above (n 73) para 214. 75 ibid para 213. 76 ibid para 215. 77 ibid. 78 ibid para 217.

48 Christian J Tams and Eleni Methymaki confirm the Tribunal’s holding that MFN clauses would only extend to questions of dispute settlement if the treaty clearly and unambiguously said so. iii. The Influence of the Court’s Jurisprudence In later awards, the presence of the ICJ pronouncements has faded, but not completely disappeared; perhaps one could say that, once investment tribunals had addressed the matter, demand for an ICJ pronouncement decreased. In the subsequent debate, there is limited room in between the Maffezini and Plama camps: while the wording of MFN clauses may permit nuance,79 on the general question – do MFN clauses, as a general matter, cover questions of dispute settlement? – tribunals are typically divided between the two approaches. For present purposes, the difference between the two decisions is not as sharp; in fact, in their respective treatment of the ICJ’s jurisprudence, there is common ground: the two decisions illustrate the power and limits of ICJ pronouncements as an ‘anchor’ for investment decisions on treaty standards. Three points stand out: (i)

Both tribunals were clearly keen to anchor their decisions in the wider discourse and to assert (convincingly or not) that their outcomes fit within the ‘general theory’ of the MFN clause. What is more, both intuitively accepted the ICJ’s jurisprudence as a relevant factor in the development of this general theory. In that sense, Plama, Maffezini and their epigones reflect the influence of the ICJ on the interpretation of investment treaty clauses that draw on pre-existent standards. (ii) Neither Maffezini nor Plama blindly followed the ICJ: its decisions were useful, perhaps persuasive, pointers, but operated within the regular framework of treaty interpretation. That framework (vague and general though it may be) controls and determines how much room there is to look to a general theory shaped in non-investment case law. What is more, neither tribunal considered the ICJ to have a monopoly on the development of the general theory: the Ambatielos report mattered as well, and for the Maffezini tribunal it mattered more. (iii) Finally, both tribunals referred to the ICJ’s case law as they read it. The same cases were read differently, nudging the Maffezini tribunal towards an expansive reading, while confirming the restrictive reading in Plama. This suggests that when anchoring their decisions in the Court’s jurisprudence, investment tribunals choose, select – and occasionally misconstrue. The ICJ can do little about that: it pronounces on legal issues, but does not have ‘the last word’.80 It is investment tribunals that do the anchoring.

B. Defining ‘Arbitrary Conduct’ Our second illustration confirms, and to some extent nuances, these points. It concerns the attempt, by an ICJ Chamber, to give meaning to the concept of ‘arbitrariness’, which has been used as an anchor by a significant number of arbitral awards.

79 See ICS Inspection and Control Services Limited (UK) v The Republic of Argentina, Decision on Jurisdiction (17 December 2009) UNCITRAL, PCA Case No 2010–9, para 275. 80 A Pellet and D Müller, ‘Article 38’ in A Zimmermann, CJ Tams, K Oellers-Frahm and C Tomuschat (eds), The Statute of the International Court of Justice. A Commentary, 3rd edn (Oxford, Oxford University Press, 2019) MN 336.

The World Court’s Influence on Contemporary Investment Law 49 i. The Chamber’s Pronouncement The relevant case is ELSI, one of the few disputes brought before the ICJ on the basis of an FCN treaty. Under Article 1 of the Supplementary Agreement to the US-Italian FCN, the ‘corporations … of either High Contracting Party shall not be subject to arbitrary or discriminatory measures within the territories of the other High Contracting Party’.81 The US argued that the requisition of ELSI’s factory by the Mayor of Palermo constituted such arbitrary and discriminatory measure injuring the US corporation. Having rejected the claim based on discrimination,82 the Chamber examined whether the requisition was arbitrary – and made two pertinent pronouncements. First, it distinguished between arbitrariness under international law and unlawfulness under domestic law: [T]he fact that an act of a public authority may have been unlawful in municipal law does not necessarily mean that that act was unlawful in international law … by itself, and without more, unlawfulness cannot be said to amount to arbitrariness.83

Second, the Chamber offered a definition of arbitrariness: Arbitrariness is not so much something opposed to a rule of law, as something opposed to the rule of law … It is a willful disregard of due process of law, an act which shocks or at least surprises, a sense of juridical propriety.84

Assessed against that standard, the requisition order – made by the competent authority, and subject to appeal before the local court – was not deemed arbitrary.85 ii. Reception in Investment Jurisprudence These relatively succinct pronouncements by the ICJ Chamber on arbitrariness have fallen on fertile ground, mainly because many treaty standards, in one way or another, preclude arbitrary conduct.86 Some treaties, like the US-Italian treaty, rule out arbitrary measures; in others, arbitrariness is a feature of other standards, such as FET, expropriation or the rule against denial of justice. Over the course of the past three decades, the succinct definition put forward by the ICJ’s Chamber has been relied on in all these contexts. It became an instant success when, less than a year after the ELSI judgment, the Tribunal in the resubmitted Amco Asia v Indonesia case relied on it. In its view, the Chamber’s distinction between unlawfulness under domestic law and arbitrariness under international law was ‘equally germane’ to the question of denial of justice – so that the Claimant had to show a ‘wilful disregard of due process of law’.87 81 ELSI,

above (n 39) para 120. para 122. 83 ibid para 124. 84 ibid para 128. 85 ibid para 129. 86 Our discussion in the following concerns ELSI’s pronouncements on the meaning of ‘arbitrariness’. We do not address the Court’s prior determination that ‘unlawfulness under domestic law does not necessarily mean unlawfulness under international law’ (and, conversely, ‘lawfulness under domestic law does not guarantee lawfulness in international law’), on which investment tribunals also often rely but which reflects a rule of general international law – namely international law’s autonomy from domestic law (codified in Art 3 of the Articles on the Responsibility of States for Internationally Wrongful Acts: ILC, ‘Articles on the Responsibility of States for Internationally Wrongful Acts with commentaries’ (2001) II(2) Yearbook of the International Law Commission 31, 37 (Art 3 commentary) (hereinafter ‘ARSIWA Commentary’)) – and does not pertain to investment law as a special regime. 87 Amco Asia v Indonesia , Award in Resubmitted Proceedings (31 March 1990) ICSID Case No ARB/81/1, paras 136–37 (Amco Asia). 82 ibid

50 Christian J Tams and Eleni Methymaki Since then, reference to ELSI has been a constant feature in investment awards dealing with similar issues, and the Chamber’s pronouncement has ‘contributed to the clarification of the concept of arbitrariness in international law’.88 Such subsequent tribunals have engaged with ELSI in different ways, illustrating its ‘anchoring’ function. Dealing with a self-standing treaty standard which prohibits arbitrary and/or discriminatory measures or treatment, some tribunals use ELSI without adding much of their own analysis. Noble Ventures v Romania is a useful illustration: the Tribunal simply acknowledged that the BIT did not contain a definition of what is ‘arbitrary’ and then turned to ELSI: ‘Regarding arbitrariness, reference can … be made to the decision of the ICJ in the ELSI case.’89 It then reiterated and applied ELSI’s definition, finding no violation. LG&E and El Paso similarly considered ELSI’s definition to reflect ‘arbitrariness’ ‘under international law’.90 Other cases have invoked the ELSI definition, but added further elements. Genin is an obvious example: the Tribunal focused on part of the definition (‘a wilful disregard of due process of law’ and offence to a ‘sense of juridical propriety’), but insisted that in order to be arbitrary, a governmental measure had to be taken in bad faith.91 In the context of NAFTA, ELSI has also been used ‘creatively’. Once the NAFTA Free Trade Commission had clarified that Article 1105 NAFTA ‘d[id] not require treatment in addition to or beyond that which is required by the customary international law minimum standard of treatment of aliens’,92 the Tribunal in Pope&Talbot put forward the ELSI definition as a reflection of that minimum standard. It did so in an attempt to overcome the exacting standard enunciated in the 1926 Neer claim (requiring bad faith and an outrage), suggesting that ELSI ‘permits a bit less injury to the psyche of the observer, who need no longer be outraged, but only surprised by what the government has done’.93 But the Tribunal’s assertation that in ELSI, ‘the International Court of Justice has moved away from the Neer formulation’94 remained unsubstantiated. Both Genin and Pope&Talbot illustrate the risk of taking nebulous standards out of context. In the subsequent jurisprudence, these more creative uses of the ELSI pronouncement no longer seem so prominent. In later awards under NAFTA, tribunals have continued to refer to the ELSI standard, but no longer sought to use it to downplay the Neer standard.95 As noted in Cargill, the Chamber’s pronouncement ‘though not based on the NAFTA, has been accepted by at least two of the State Parties to the NAFTA as the “best expression” of arbitrariness’.96 Awards rendered outside NAFTA point in the same direction: to illustrate, in Azurix, the

88 P

Tomka, ‘Elettronica Sicula Case’ (February 2007) MPEPIL, para 23. Ventures Inc v Romania, Award (12 October 2005) ICSID Case No ARB/01/11, paras 175–79. v Argentina, Decision on Liability (3 October 2006) ICSID Case No ARB/02/1, paras 156–57 (hereinafter LG&E); El Paso Energy International Company v Argentina, Award (31 October 2011) ICSID Case No ARB/03/15, para 319. 91 Alex Genin v Republic of Estonia, Award (25 June 2001), ICSID Case No ARB/99/2, para 371, referencing Amco Asia, above (n 84) and its reliance on ELSI, above (n 36). 92 Notes of Interpretation of Certain Chapter 11 Provisions (NAFTA Free Trade Commission, 31 July 2001), https://perma.cc/V7YX-D832. 93 Pope&Talbot v Canada, Award in Respect of Damages (31 May 2002) UNCITRAL, para 64. 94 ibid para 63. 95 Mondev v USA, Award (11 February 2002) ICSID Case No ARB(AF)/99/2, para 127; Loewen and Loewen v USA, Award (26 June 2003) ICSID Case No ARB(AF)98/3, paras 131–32; ADF Group v USA, Award (9 January 2003) ICSID Case No ARB(AF)/00/1, para 190; Cargill v Mexico, Award (18 September 2009) ICSID Case No ARB(AF)/05/2, para 291 (hereinafter Cargill). 96 Cargill, above (n 95) para 291. 89 Noble

90 LG&E

The World Court’s Influence on Contemporary Investment Law 51 Tribunal noted that in interpreting the term ‘arbitrary’, ‘[t]he findings of other tribunals, and in particular the ICJ, should be helpful’.97 It then found that ELSI’s definition was close to the ordinary meaning of the word.98 iii. The Impact of the Court’s Pronouncements What does this jurisprudence tell us about the impact of the Court on IIL? In two basic respects, it reinforces points made in relation to our first illustration, viz the ICJ’s jurisprudence on MFN clauses: (i) ICJ pronouncements have no binding authority, but (ii) they are attractive anchor points for investment tribunals who consider that a reference to ICJ jurisprudence will lend credibility to their awards. This may not fully account for the success of the ELSI pronouncements. Three further considerations seem at play. The first takes us back to our introductory considerations about supply and demand. Regularly working with ‘arbitrariness’ as a concept but never clearly articulating what it meant, IIL clearly had demand for a definition, and the ICJ’s jurisprudence supplied it. The second goes to the relative openness of the ELSI pronouncement. In defining arbitrariness, the ICJ Chamber articulated an intuitively plausible standard that gave some meaning to a vague treaty clause; simultaneously, it remained sufficiently vague: if the ELSI case has ‘contributed to the clarification of the concept of arbitrariness in international law’,99 then not the least this is because the Chamber’s pronouncement remains malleable. And finally, the fate of ELSI illustrates a particular quality of ICJ pronouncements. Precisely because the ICJ is not a specialised court, but is considered a guardian of generalist international law, its jurisprudence is taken to reflect the state of general international law. The experience with ELSI suggests that this particular quality – which has been described as a mainstreaming of special areas of international law100 – is shared even by tribunals/arbitrators that seek to ascertain the meaning of a treaty clause prohibiting arbitrary measures. And so, this second example also illustrates that, while its effect on the interpretation of investment treaty standards is limited, ICJ pronouncements occasionally have a significant influence.

IV. REMEDIES

Remedies are a defining feature of investment disputes.101 Cases turn on remedies, and not just in the trite sense that these embody the outcome of a tribunal’s decision. The dominant remedy sought in investment arbitration – damages – is frequently addressed in a dedicated phase of the proceedings. It is governed by highly technical principles of valuation and calculation, and remains, to the uninitiated, ‘guère moins spéculatifs et tout aussi obscurs que les prophéties de Nostradamus’.102 What is more, the amount of compensation awarded in some disputes

97 Azurix

Corp v Argentina, Award (14 July 2006) ICSID Case No ARB/01/12, paras 391–92.

98 ibid. 99 Tomka,

above (n 88). eg, B Simma, ‘Mainstreaming Human Rights: The Contribution of the International Court of Justice’ (2012) 3 Journal of International Dispute Settlement 3, 7. 101 I Marboe, Calculation of Compensation and Damages in International Investment Law, 2nd edn (Oxford, Oxford University Press, 2017) 2. 102 I Seidl-Hohenveldern, ‘L’évaluation des dommages dans les arbitrages transnationaux’ (1987) 33 Annuaire Français de Droit International Année 7, 24. 100 See,

52 Christian J Tams and Eleni Methymaki has fuelled concerns about IIL’s (real and perceived) investor bias and has prompted calls for a ‘rethinking [of] the quantification of damages in international investment arbitration’.103 All this is far removed from the practice of the World Court, which has rarely awarded compensation.104 And yet, it is with respect to reparation that the World Court has most clearly contributed to contemporary IIL. The reason for this is simple: IIAs tend to say little on questions of remedies. Many of them prescribe a standard of compensation (often requiring compensation to be full, prompt and effective), but this is formulated as a condition for expropriation to be lawful, not as a form of reparation for treaty breaches.105 Investment law, therefore, has a lot of demand for guidance.106 And despite the limited number of damages awards rendered by the World Court, its jurisprudence supplies general principles. Consolidated in the ILC’s work on state responsibility, these general principles have become a point of reference for investment tribunals who seek to anchor their decisions in a general law of remedies and ascertain the rules applicable when awarding reparation.

A. General Principles of Reparation: The PCIJ’s Pronouncements The World Court’s key statements on reparation were made almost a century ago, in the Factory at Chorzów case, which concerned Poland’s takeover of a German-owned nitrate factory in Upper Silesia.107 Having found the taking to be in breach of the Geneva (Upper Silesia) Convention of 1922,108 the PCIJ was concerned with Germany’s claim for reparation. In this respect, it made three pronouncements, which have shaped subsequent debates. First, reparation was a natural consequence of an illegal act: [I]t is a principle of international law, and even a general conception of law, that any breach of an engagement involves an obligation to make reparation … reparation is the indispensable complement of a failure to apply a convention, and there is no necessity for this to be stated in the convention itself.109

Second, reparation aimed to establish the status quo sine: [It] must, as far as possible, wipe out all the consequences of the illegal act and reestablish the situation which would, in all probability, have existed if that act had not been committed. Restitution in kind, or, if this is not possible, payment of a sum corresponding to the value which a restitution in kind would bear; the award, if need be, of damages for loss sustained which would not be covered by

103 T Marzal, ‘We Need to Talk about Valuation in ISDS’ (Verfassungsblog, 5 March 2020), https://perma. cc/49YJ-YVWV. 104 The ICJ has only done so in Corfu Channel (UK v Albania) (Compensation) [1949] ICJ Rep 244; Diallo, above (n 34) (Compensation) [2012] ICJ Rep 324; Certain Activities Carried Out by Nicaragua in the Border Area (Costa Rica v Nicaragua) (Compensation) [2018] ICJ Rep 15; and is in the process of doing so in the case concerning Armed Activities on the Territory of the Congo (DRC v Uganda) (Order of 8 September 2020) [2020] ICJ Rep 264. 105 See section IV(B). But see for a discussion on this Toni Marzal, ‘Quantum (In)Justice: Rethinking the Calculation of Compensation and Damages in ISDS’ (2021) 22 Journal of World Investment and Trade 249. 106 IIL is not exceptional, treaties in many fields omit to spell out remedies. As Chester Brown notes ‘[t]he absence of clear provisions … has led international courts to consider the practice of other international tribunals in the awarding of remedies’: C Brown, A Common Law of International Adjudication (Oxford, Oxford University Press, 2007) 186. 107 Case Concerning the Factory at Chorzów (Germany v Poland) (Merits) PCIJ Ser A No 17 (hereinafter Factory at Chorzów). 108 Certain German Interests, above (n 54). 109 Factory at Chorzów, above (n 107) 29. See also Factory at Chorzów (Jurisdiction) PCIJ Ser A No 9, 21.

The World Court’s Influence on Contemporary Investment Law 53 restitution in kind or payment in place of it – such are the principles which should serve to determine the amount of compensation due for an act contrary to international law.110

Third, compensation for expropriation followed a different logic: The ‘principles … determin[ing] the amount of compensation due for an act contrary to international law’ were different from those governing the payment of compensation for lawful expropriation.111 In the latter case, the standard of compensation was set to ‘fair compensation’ or the ‘just price’ of what has been lawfully expropriated, consisting ‘of the value of the undertaking at the moment of dispossession plus interest to the day of payment’.112 By contrast, compensation for unlawful takings needed to cover both the value of the property at the time of the indemnification as well as any loss sustained as a result of its illegal seizure.113 These three pronouncements are at the level of principles that do not translate into readily applicable rules. However, as principles, they have found general acceptance. Of particular relevance in this is the ILC’s work on state responsibility, which (in line with the first pronouncement) views the duty to make reparation as a ‘new legal relation[] that arise[s] from the commission … of an internationally wrongful act’,114 and, in Article 31, stipulates that states incurring responsibility are under a ‘duty to make “full reparation” in the Factory at Chorzów sense’.115 The ICJ itself has regularly affirmed the PCIJ’s Chorzów pronouncements, which it described, in its 2004 Wall opinion, as having ‘laid down’ ‘the essential forms of reparation in customary law’.116

B. Their Reception in Investment Arbitration In investment arbitration, the Chorzów Factory judgment is a constant point of reference. As Ursula Kriebaum states, it is ‘probably the PCIJ’s judgment most frequently referred to in investment arbitration’,117 often in conjunction with the ILC’s provision on reparation.118 110 Factory at Chorzów, above (n 107) 47. Our focus in the following is on damages. The second Chorzów pronouncement remains influential also for its assertion of a hierarchy between the two most common modes of reparation: the PCIJ’s preference for restitution over compensation (the latter being available only if restitution ‘is not possible’) does not square easily with the fact that, in Chorzów, Germany had only demanded compensation. However, it has shaped the ILC’s subsequent work on state responsibility, which in Art 36 ARSIWA accepts restitution’s ‘primacy as a matter of legal principle’: see ARSIWA Commentary, above (n 86) 99; and further CG Gray, Judicial Remedies in International Law (Oxford, Clarendon Press, 1987) 95 et seq.; Investment tribunals have not often engaged with the question of hierarchy, but have at times clarified that restitution could in principle be awarded: see, eg Enron Corporation and Ponderosa Assets, L.P. v Argentine Republic, Decision on Jurisdiction (14 January 2004) ICSID Case No ARB/01/3, para 79; and further S Hindelang, ‘Restitution and Compensation – Reconstructing the Relationship in Investment Treaty Law’, in Hofmann and Tams, above (n 13) 161; U Kriebaum, ‘Restitution in International Investment Law’, in Hofmann and Tams, above (n 13) 201. 111 Factory at Chorzów, above (n 107) 46. 112 ibid 46–47. 113 ibid 48. 114 ARSIWA Commentary, above (n 86) para 3(f) (General commentary). 115 ibid 91, para 3 (Art 31). 116 Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory (Advisory Opinion) [2004] ICJ Rep 136, para 152. 117 U Kriebaum, ‘The PCIJ and the Protection of Foreign Investments’ in M Fitzmaurice and CJ Tams (eds), Legacies of the Permanent Court of International Justice (Leiden, Brill, 2013) 161 (then noting that ‘[n]o fewer than fortyeight decision and awards refer to Chorzów’). According Marboe, above (n 101) 417ff between 2008–2017, 30 awards (out of 63 she examined) explicitly referred to Chorzów in calculating compensation. 118 See, eg LG&E, above (n 87) Award on Damages (25 July 2007) para 31, where the Tribunal agreed ‘with the Claimants that the appropriate standard for reparation under international law is ‘full’ reparation as set out by the Permanent Court of International Justice in the Factory at Chorzów case and codified in Article 31 of the International Law Commission Draft Articles on Responsibility of States for Internationally Wrongful Acts’.

54 Christian J Tams and Eleni Methymaki A nuanced analysis suggests that all three pronouncements set out in the preceding section play some role in contemporary IIL. The PCIJ’s first pronouncement – that reparation is the natural consequence of an illegal act, which does not need to be expressly articulated – is the basis of nearly all investment decisions awarding damages: as IIAs do not address the consequences of breaches, reparation can only be ordered if it ‘is the indispensable complement of a failure to apply a convention’.119 That this is so is today accepted as a matter of course, and few tribunals feel the need to spell it out. Echoes of the first Chorzów pronouncement can be heard in the Biwater Gauff award, in which the Tribunal observed that ‘[f]or claims other than expropriation … the BIT does not offer any guidance for evaluating the damages’ – and then immediately continued that ‘this does not mean that compensation is excluded’, but meant that ‘the common starting point is the broad principle articulated in the well known Factory at Chorzów case’.120 The PCIJ’s third pronouncement – distinguishing between compensation owed in the case of expropriation on the one hand, and damages for treaty breaches on the other, as well as between the relevant valuation date in each case121 – has had a more chequered history. To reiterate, many BITs do spell out the measure of compensation owed for an expropriation to be lawful. Perhaps attracted by such express provisions, a number of investment tribunals over time ‘assessed damages for both lawful and unlawful expropriation using the standard of compensation [expressly stated] in the relevant BIT’.122 Awards such as SD Myers or ADC v Hungary put matters straight: they clarified the limited reach of the express treaty standard of compensation: in the words of ADC, ‘[t]he BIT only stipulates the standard of compensation that is payable in the case of a lawful expropriation, and these cannot be used to determine the issue of damages payable in the case of an unlawful expropriation since this would be to conflate compensation for a lawful expropriation with damages for an unlawful expropriation’.123 Outside instances of lawful expropriation, the path towards the general standard as enunciated in Chorzów Factory is clear: ‘the Tribunal is required to apply the default standard contained in customary international law’ – which ‘is set out in the decision of the PCIJ in the Chorzów Factory case’.124 While not all tribunals have followed this approach, the need to distinguish between different methods of calculating damages is (as noted in a recent award) ‘increasingly recognized in international practice’.125 This can be seen as a late affirmation of the distinction foreshadowed in the third Chorzów pronouncement.

119 Factory

at Chorzów, above (n 107) 29. Gauff v Tanzania, Award (24 July 2008) ICISD Case No ARB/05/22, para 776. See also DA Desierto, ‘Expropriation Cases’, ch 21 in this volume. 121 J Branson, ‘Damages in Investment Arbitration – A Revolutionary Remedy or Reward for Rich Corporations at the Expense of the World’s Poor?’ (2016) 3 Journal of Damages in International Arbitration 23 argues that in the PCIJ’s statement regarding the different valuation dates ‘modern tribunals have clung to in justifying the right shift in valuation dates’. 122 B Sabahi and C Guzman, ‘Award on Damages’ (August 2019) MPEIPro, para 7: eg CMS Gas Transmission Company (‘CMS’) v Argentina, Final Award (12 May 2005) ICSID Case No ARB/01/8, para 410. 123 ADC v Hungary, Award (2 October 2006) ICSID Case No ARB/03/16, para 481. 124 ibid para 483–84. See also SD Myers v Canada, UNCITRAL (1976), Partial Award (13 November 2000) paras 307–10. 125 Caratube International Oil Company LLP v Kazakhstan, Award (5 June 2012) ICSID Case No ARB/08/12, para 1082; and further Quiborax v Bolivia, Award (16 September 2015) ICSID Case No ARB/06/2, para 326 (hereinafter Quiborax); Teinver, above (n 41) Award (21 July 2017) paras 1088–92; Siemens AG v Argentina, Award (6 February 2007) ICSID Case No ARB/02/8, paras 352–53; Compañía de Aguas del Aconquija SA and Vivendi Universal SA v Argentina, Award (20 August 2007) ICSID Case No ARB/97/3, paras 8.2.3–8.2.5. Marboe, above (n 101) 407. 120 Biwater

The World Court’s Influence on Contemporary Investment Law 55 This leaves the second Chorzów pronouncement, viz the PCIJ’s robust statement that reparation ‘must as far as possible, wipe out all the consequences of the illegal act and reestablish the situation which would, in all probability, have existed if that act had not been committed’.126 Of the three pronouncements, this has arguably left the most significant legacy. Two aspects stand out. The first concerns the method of calculating damages. Tribunals are required to compare the actual situation to a hypothetical scenario in which the state has acted lawfully.127 This in turn has meant that damages are not limited to the value of an investment at the time of the taking; subsequent increases in value can be recovered, and tribunals have regularly relied on Chorzów Factory to justify their decision to take them into account in their calculation.128 The second legacy is perhaps best described as the ‘spirit’ of the Chorzów Factory judgment.129 While the PCIJ’s pronouncement came with caveats (‘as far as possible’, ‘in all probability’), its general thrust130 was clear: reparation means ‘full reparation’, and it is not merely designed to allay problems, but to ‘wipe out all the consequences’ of the wrongful conduct, including those that a mere restitution in kind would not cover. What is more, in assessing how such consequences were best to be wiped out, the PCIJ commissioned different valuations, all the while emphasising that it had ‘every right’ to determine which would best suit the ‘obvious aim of implementing the general principle … [of] full reparation’.131 This was, in other words, a judgment emphasising the rights of the aggrieved party and the discretion of a court in choosing how best to remedy it.132 And over the course of the past decades, investment tribunals seem to have acted in that spirit. As reparation is meant to be ‘full’, there is no a priori limitation on the damages that can be compensated – ‘all negative consequences of the wrongful act’ are to be remedied.133 The diversification on valuation methods need not be recounted here: in refining methods for establishing the fair market value and notably in embracing the DCF method,134 investment tribunals have ‘dramatically developed’135 the practice of damages, and asserted their right to ‘select the appropriate method [of valuation], basing its decision on the circumstances of each individual case’.136 In the maze of contemporary damages jurisprudence,

126 Factory at Chórzow, above (n 107) 46–47; According to Marboe, above (n 101) 303–4: ‘The investment arbitration cases involving national court and arbitral decisions show that the principle of full reparation under the Chorzów Factory is regularly referred to as the appropriate standard.’ 127 And they have regularly done so on the basis of the Factory at Chorzów, Marboe, above (n 101) 95. 128 See, eg Burlington Resources Inc v Ecuador, Decision on Reconsideration and Award (7 February 2017) ICSID Case No ARB/08/5, para 326; Quiborax, above (n 125) para 370; as well as ConocoPhillips v Venezuela, Award (8 March 2019) ICSID Case No 07/30, para 259 (referencing arguments suggesting a focus on the date of the expropriation before adding: ‘The Chorzów Judgment does not support such an understanding’). 129 WM Reisman/RD Sloane, ‘Indirect Expropriation and its Valuation in the BIT Generation’ (2003) 74 British Yearbook of International Law 115, 135 and 148 have called this the ‘Chorzów Factory’s imperative’. 130 Not everything in the Chorzów Factory judgment pointed in this direction: cf Occidental v Ecuador, Decision on Annulment (2 November 2015) ICSID Case No ARB/06/11, paras 282–96 (drawing on Chorzów Factory to clarify that, in the calculation of damages, ‘injury resulting to third parties should … be excluded’ [para 289]). 131 Quiborax, above (n 125) paras 375–76. 132 See the recognition of such discretion in Rumeli v Kazakhstan, Decision on Annulment (25 March 2010) ICSID Case No ARB/05/16, para 179(5). 133 Joseph Charles Lemire v Ukraine, Award (28 March 2011) ICSID Case No ARB/06/18, para 151. But see the discussion in M Paparinskis, ‘A Case against Crippling Compensation in International Law of State Responsibility’ (2020) 83 Modern Law Review 1246; M Paparinskis, ‘Crippling Compensation in the International Law Commission and Investor-State Arbitration’ (2021) ICSID Review (advance copy). 134 See the helpful account in S Ripinsky and K Williams, Damages in International Investment Law (London, British Institute of International and Comparative Law, 2008) 181–259. See also I Marboe, ‘Reparation Cases: Applicable Principles in International Investment Arbitration’, ch 26 in this volume. 135 Sabahi and Guzman, above (n 122) para 82. 136 Crystallex International Corporation v Venezuela, Award (4 April 2016) ICSID Case No ARB(AF)/11/2, para 886.

56 Christian J Tams and Eleni Methymaki ‘[t]he devil’ (as noted by the Tenaris tribunal) ‘alas, is in the detail’.137 However, the emphasis, in Chorzów Factory, on full reparation and the discretionary use of valuation methods has enabled tribunals to present the dramatic developments as a finetuning that remained true to the PCIJ’s foundational decision and to largely ignore concerns that the Chorzów principles had been ‘imperfect[ly]’ applied.138

C. The Impact of the Court’s Pronouncements The preceding section highlights the influence of a string of interrelated judicial statements made nearly a century ago. Formulated before the consolidation of the modern law of state responsibility, the PCIJ’s Chorzów pronouncements have influenced the general understanding of reparation and the standards of compensation. And while the work of the ILC translated the pronouncements into a series of draft articles, it has not eclipsed the formative judgment, whose ‘vitality [if anything] was energized’ by the ILC’s work.139 Chorzów Factory has radiated into many fields of international law, but nowhere have its effects been felt more clearly than in IIL. There is a quantitative side to this: as Sabahi and Guzman emphasise ‘the majority of modern awards on damages involving the Factory at Chorzów principle have been rendered in investor State Arbitration’.140 But more than that, no other field of international law has had such a demand for guiding statements on reparation – which is central to IIL and yet, curiously, not addressed in investment treaties. From the preceding discussion, it is clear that the different pronouncements serve different functions. The first (describing the duty to make reparation as an ‘indispensable complement of a failure to apply a convention’141) is today taken for granted. The third (distinguishing between compensation for lawful expropriation and damages for unlawful acts) has guided tribunals in their quest to delimit the scope of application of treaty-based compensation standards – a process that is not complete, but well under way. The second (emphasising the need for full reparation) has facilitated the enormous development of the contemporary damages jurisprudence: notably, it has increased the flexibility of tribunals who today can choose between a wide range of valuation methods which are sometimes said to ‘pertain to the realm of economics’,142 but in reality reflect crucial choices of decisionmakers.143 This is the most significant and most controversial legacy of the Chorzów Factory judgment.

137 Tenaris

v Venezuela, Award (29 January 2016) ICSID Case No ARB/11/26, para 521. Desierto, ‘The Outer Limits of Adequate Reparations for Breaches of Non-Expropriation Investment Treaty Provisions’ (2017) 55 Columbia Journal of Transnational Law 395. 139 See BG Group Plc v Argentina, UNCITRAL, Final Award (24 December 2007) para 426. 140 Sabahi and Guzman, above (n 122) para 13. 141 Factory at Chórzow, above (n 107) 29. 142 Sabahi and Guzman, above (n 122) para 42. 143 See Marzal, above (n 105) generally. 138 D

The World Court’s Influence on Contemporary Investment Law 57 V. OUTLOOK

The World Court’s influence on IIL is not easily summarised; and definitely not in the confines of one single chapter. It cuts across many aspects of the field and has varied over time. Our discussion points to a basic division between the IIL of the pre-BIT era, and the contemporary variation dominated by treaty-based arbitration. It suggests that in cases like Anglo-Iranian Oil Co. and Barcelona Traction, the World Court emphasised limits to investment protection under general international law – and that later cases such as ELSI and Diallo did not mark a radical departure. The World Court’s impact on IIL in the contemporary BIT era is of a different character. The dynamic, special regime of investment law relies on express treaty standards; it has (notably in ISDS) a powerful instrument for clarifying and developing the law; and, it envisages no formal role for the World Court. Unsurprisingly, the Court’s role is most clearly felt where the special regime fails to provide express rules: this explains the vitality of the PCIJ’s Chorzów pronouncements, which have helped fill a glaring drafting gap in modern IIAs, viz remedies. In clarifying the scope of written treaty standards, World Court pronouncements have a lesser role. Our discussion shows that, in engaging with PCIJ and ICJ jurisprudence, investment tribunals have a lot of leeway – which occasionally leads to unexpected re-interpretations of the ICJ’s case law. But at the same time, investment tribunals consider it advantageous to anchor their decisions in the PCIJ’s and ICJ’s case law. This continuing appeal reflects the World Court’s special status in the contemporary universe of dispute settlement bodies.

5 The Contribution of the Iran-United States Claims Tribunal BRUNO SIMMA AND CRISTINA HOSS*

I. INTRODUCTION

A. Brief History of the Tribunal

I

N 2019 THE Islamic Republic of Iran celebrated the 40th anniversary of the ‘Islamic Revolution’. The historic circumstances of that revolution are generally known but are still worth recalling in introducing the Iran-United States Claims Tribunal (hereinafter the ‘IUSCT’ or ‘Tribunal’), the arbitral tribunal set up as an instrument of peaceful settlement of disputes in the aftermath of the revolution. Rich in natural resources and geographically close to the Soviet Union, Iran held a strategically important position for western powers during the Cold War. In 1951, the Iranian Government nationalised the Anglo-Iranian Oil Company. The British response played out in three steps. First, it launched ultimately unsuccessful attempts to obtain compensation for its assets before the International Court of Justice (‘ICJ’) and later before the United Nations (‘UN’) Security Council.1 Second, the UK imposed sanctions on Iran. Finally, when these proved ineffective, the UK and United States’ governments decided to organise a coup d’état, known as Operation Ajax, which brought to an end not only the nationalisation policies of the government of Prime Minister Mossadegh, but also Iran’s young democracy. In the aftermath of the coup, Shah Mohammed Reza Pahlavi, who until then had been a figurehead monarch, dismissed the democratically elected Prime Minister and became the controversial and absolute leader of Iran. Following the Shah’s reinstatement, economic relations between Iran and the West, and with the US in particular, intensified again, including delivery by the US of military equipment and support in large infrastructure projects initiated by the Shah. As to military equipment, the Shah’s government purchased it mostly through the US State Department’s Foreign Military

* Professor Bruno Simma is a Member of the Iran-US Claims Tribunal (since December 2012) and Professor of Law at the University of Michigan. From 2003 to 2012 he served as a judge at the International Court of Justice; Dr Cristina Hoss is a Legal Officer at the International Court of Justice (The views expressed herein are those of the authors and do not necessarily reflect those of the International Court of Justice or the United Nations). 1 Anglo-Iranian Oil Co Case (Judgment on jurisdiction) [1952] ICJ Reports 93.

60 Bruno Simma and Cristina Hoss Sales Program administered by the Department of Defense. The political stability and increasing oil revenues attracted investment and economic exchanges between the two countries; by the end of the 1970s several thousand US nationals were working and living in Iran. In the late 1970s, however, the Shah’s notorious brutality against all opposition, political or otherwise, as well as inflation and social tensions, provoked large-scale demonstrations and strikes across the country. Significant parts of the population also opposed the – real or perceived – domination of Iranian political, cultural and economic life by the US. In January 1979, the Shah left Iran to undergo medical treatment in the US, enabling the return of the cleric Ayatollah Chomeini, a leading opposition figure. On 11 February 1979, Chomeini, who had spent 15 years in exile, declared the victory of the Islamic Revolution and established a new government. After the organisation of a referendum in March 1979, the Islamic Republic of Iran was founded. As a consequence of the regime change, economic exchange and foreign commercial activities were severely restricted and the nationalisation of several industries, in particular oil and gas, was initiated. Once it became known that the Shah had sought refuge in the US, protests on the streets, not only in Teheran, intensified until, on 4 November 1979, Iranian students from a militant group called the ‘Muslim Followers of the Imam’s Policy’, forced their way into the US Embassy in Tehran, seizing and detaining the personnel present at the mission. According to the US Government, the Iranian Government did not make any attempt to intervene, instead, Ayatollah Chomeini expressed his support and sympathy for the students’ actions and officially endorsed the occupation of the Embassy by a decree dated 17 November 1979.2 In that decree, he requested the release of ‘the blacks and women if proven that they did not spy’.3 The conditions of release of the other staff members were clearly set out in the decree: the US had to hand over the Shah for trial and return his property to Iran. Fifty-two members of the diplomatic and consular staff remained in detention at the Embassy. On 29 November 1979, the US instituted proceedings before the ICJ and requested the indication of provisional measures, which the Court ordered but which were not carried out.4 Prior to the filing of the request for the indication of provisional measures, the US had started imposing sanctions on Iran and Iranian nationals residing in the US. Those sanctions culminated in Presidential Executive Order No 12.170 of 14 November 1979, blocking the transfer of Iranian governmental assets subject to US jurisdiction. The frozen Iranian assets were valued at a total of US$12 billion.5 US companies and private persons whose properties had been confiscated by the Iranian Government attempted to recover their losses through court proceedings in the US, seeking attachments over frozen Iranian properties.

2 A more detailed account of the events surrounding the hostage taking is to be found in the judgment of the International Court of Justice in the case concerning United States Diplomatic and Consular Staff in Tehran (Judgment) [1980] ICJ Reports 3, 11–15 and 28–35; See also G Lagergren, ‘The formative years of the Iran-United States Claims Tribunal’ (1997) 66 Nordic Journal of International Law, 23–32; H van Houtte and B Concolino, ‘The Iran-United States Claims Tribunal and its contribution to international law’ in G De Baere and J Wouters (eds), The Contribution of International and Supranational Courts to the Rule of Law (Cheltenham, Elgar, 2015) 303–325 (303–309). 3 Quoted in: United States Diplomatic and Consular Staff in Tehran, above (n 2) 13. 4 United States Diplomatic and Consular Staff in Tehran (Order) [1979] ICJ Reports 7. The Islamic Republic of Iran chose not to participate in the proceedings leading to the Court’s judgment on the merits of the case. In that judgment, the Court found that after the endorsement by Ayatollah Chomeini, the acts of the militant students had become attributable to the Iranian State, which was internationally responsible for the wrongful acts. ICJ Reports 1980, 3 (p 35, and operative paragraph on pp 44–46). 5 Lagergren, above (n 2) 23.

The Contribution of the Iran-United States Claims Tribunal 61 On 24 April 1980, one month before the ICJ issued its judgment on the merits of the case, the US initiated a military operation, to rescue its citizens. It had to be aborted for ‘technical reasons’.6 This operation, which the US justified before the Security Council as an act of self-defence pursuant to Article 51 of the UN-Charter, was severely criticised by the ICJ, as being an operation ‘of a kind calculated to undermine respect for the judicial process in international relations’.7 Shortly thereafter, in the fall of 1980, the Governments of the US and Iran entered into negotiations through the Government of Algeria as an intermediary. In a resolution adopted by the Iranian Parliament, the Majlis, four conditions for an agreement were set out: (i) the nonintervention of the US in Iran’s internal affairs, (ii) the return of frozen Iranian assets and the removal of trade sanctions imposed against Iran, (iii) the cancellation of all US private claims and attachments against the assets and properties of Iran, and (iv) the return of the Shah’s wealth to Iran.8 While the US agreed in principle, it also insisted that the claims against Iran pending before US courts could only be removed if an alternative mechanism of settlement was created. On 19 January 1981, the Government of Algeria announced the signature by the Parties of the ‘Algiers Declarations’, consisting of the ‘General Declaration’ and the ‘Claims Settlement Declaration’.9 The General Declaration contains the commitments of the Parties, namely that Iran would release the 52 US citizens it had held captive and that the US would not intervene directly or indirectly, politically or militarily in Iran’s internal affairs, that it would assist Iran in returning the assets of the former Shah, that it would restore, within the framework of the two Declarations, Iran’s financial position in so far as possible to that which existed prior to 14 November 1979 (General Principle A), ie the date of Executive Order No 12.170, inter alia, by insuring the mobility and free transfer of all Iranian assets within US jurisdiction, and that the US would terminate all legal proceedings in US courts involving claims of US persons and institutions against Iran and its state enterprises, nullify attachments and judgments obtained from US courts, prohibit all further litigation based on such claims, and bring about the termination of such claims through binding arbitration (General Principle B). The Claims Settlement Declaration (‘CSD’) established the Iran-United States Claims Tribunal for the purpose of settling, in particular, the legal proceedings pending before US courts as well as official claims between the two Parties arising out of contractual arrangements between them for the purchase of goods and services. The Tribunal was further vested with the jurisdiction over disputes relating to the interpretation and application of the General Declaration.10

6 United

States Diplomatic and Consular Staff in Tehran, above (n 2) 17. 18. 8 The resolution of 2 November 1980 was not adopted as a piece of legislation and was thus never published as such in the Official Gazette. In a later legislative act passed on 14 January 1981 by the Majlis, the text of the resolution is, however, reproduced in full (Official Gazette No 10469, 4 February 1981) and can be found in its English version in: M Mohebi, The International Law Character of the Iran-United States Claims Tribunal (Alphen aan den Rijn, Kluwer, 1999) 389–390. 9 The General Declaration and the Claims Settlement Declaration of 19 January 1981 are reprinted in 1 Iran-USCTR, 3–17. 10 Article II of the Claims Settlement Declaration reads as follows: 7 ibid

‘1. An International Arbitral Tribunal (the Iran-United States Claims Tribunal) is hereby established for the purpose of deciding claims of nationals of the United States against Iran and claims of nationals of Iran against the United States, and any counterclaim which arises out of the same contract, transaction or occurrence that constitutes the subject matter of that national’s claim, if such claims and counterclaims

62 Bruno Simma and Cristina Hoss The Tribunal, composed of three Members appointed by Iran, three Members appointed by the US and three Third Country Members, held its first meeting on 1 July 1981 in The Hague, which had been chosen as the seat of the new institution. Claims of nationals against one of the Parties had to be filed within a year after the adoption of the Declarations. Almost 4,000 such claims were received: 2,884 small claims (amounting to less than US$250,000), 961 large claims (amounts larger than US$250,000), 75 contractual claims between the two governments (so-called ‘B’ cases) and 33 disputes concerning the interpretation and application of the General Declaration (so-called ‘A’ cases), for which the one-year deadline did not apply. Four hundred out of the 2,884 small cases had either been decided, settled or withdrawn when in 1990 the governments agreed to settle almost all of those cases by a lump sum agreement.11 More generally, an important number of claims were solved through settlement between the parties. Thus, by 2016, which is the date of the latest Communiqué published by the Tribunal’s Secretary General, the Tribunal had issued 238 Awards on Agreed Terms and 18 Partial Awards on Agreed Terms.12 For many years the Tribunal mostly operated through three Chambers each composed of three members (one from each group, with the Third Country Member presiding), and only occasionally sat as Full Tribunal. Up to now it has finalised over 3,900 cases by Award, Decision or Order.13 Nowadays, after almost forty years, only the Full Tribunal is sitting and dealing with the large and complex claims in 16 cases (A and B cases) which are left on its docket.

B. Hybrid Nature of the Tribunal: International Tribunal or Ad Hoc Arbitration Against this historical background, it is not surprising that there has been extensive discussion on the institutional nature of the Tribunal.14 It is a tribunal for the resolution of inter-state disputes and thus the peaceful settlement of disputes within the meaning of Article 33 of the Charter of the United Nations and, at the are outstanding on the date of this agreement, whether or not filed with any court, and arise out of debts, contracts (including transactions which are the subject of letters of credit or bank guarantees), expropriations or other measures affecting property rights, excluding claims described in Paragraph 11 of the Declaration of the Government of Algeria of January 19, 1981, and claims arising out of the actions of the United States in response to the conduct described in such paragraph, and excluding claims arising under a binding contract between the parties specifically providing that any disputes thereunder shall be within the sole jurisdiction of the competent Iranian courts in response to the Majlis position. 2. The Tribunal shall also have jurisdiction over official claims of the United States and Iran against each other arising out of contractual arrangements between them for the purchase and sale of goods and services. 3. The Tribunal shall have jurisdiction, as specified in Paragraphs 16–17 of the Declaration of the Government of Algeria of January 19, 1981, over any dispute as to the interpretation or performance of any provision of that declaration.’ 11 See

Award No 483-Claims of less than US$250,000/86/B38/B76/B77-FT, reprinted in 25 Iran-USCTR, 273. ‘Communiqué regarding the work of the Tribunal’ (9 May 2016), https://perma.cc/4U9N-278L. For instance, as was widely reported, on 17 January 2017 the Obama administration settled by cash payment a part of a pending case (B1) relating to claims arising out of military sales. On the controversy surrounding that payment see: David E Sanger, ‘US concedes $400 Million payment to Iran was delayed as Prisoner “leverage”’ (18 August 2016) New York Times, https://perma.cc/7KDA-VBWM. 13 See ‘Communiqué regarding the work of the Tribunal’, above (n 12). 14 See in particular, Ph Fouchard, ‘La nature juridique de l’arbitrage du tribunal des différends irano-américains’ in B Stern, M-F Labouz, Le Tribunal des différends irano américains (1984) 1 Cahiers du Centre de droit international de Nanterre (CEDIN) 27–48; M Mohebi, The International Law Character of the Iran-United States Claims Tribunal (Alphen aan den Rijn, Kluwer, 1999); D Caron, ‘The Nature of the Iran-United States Claims Tribunal and the Evolving Structure of International Dispute Resolution’ (1990) 84 American Journal of International Law 104. 12 See

The Contribution of the Iran-United States Claims Tribunal 63 same time, one of the first arbitral tribunals to allow investors as well as individuals to make claims against a foreign state. As former IUSCT President Karl-Heinz Böckstiegel put it: The Iran-United States Claims Tribunal was the first permanent body permitting investors as well as other companies and individuals of the two States to sue the other State. It has the characteristics of a court in so far as a great number of cases are decided by the same panel within a permanent institution. But on the other hand, it has the characteristics of arbitration in so far as its procedure is governed by the amended UNCITRAL Arbitration Rules and the State parties decide on selection of members of the Tribunal.15

The contribution of the Tribunal’s jurisprudence to international investment law is, as significant as it is, controversial.16 On the one hand, the bulk of its cases is in essence concerned with the law of expropriation and standards of compensation, its decisions and work are accessible to the public and it operates under the original United Nations Commission on International Trade Law (‘UNCITRAL’) Rules. On the other hand, observers have challenged the Tribunal’s purely arbitral nature and its potential to contribute to investor-state arbitration and have called for caution in using the Tribunal’s case law as precedent in investment law.17 A Tribunal operating amidst ongoing political tensions and, at times, open conflict between two states and dealing with alleged violations of international agreements by the governments of those states, might be suspected of being a politicised institution rather than a court of law that can issue decisions with precedential value.18 Moreover, the specific clause on applicable law, Article V of the CSD, incorporated in the Tribunal’s Rules of Procedure, discards the equivalent UNCITRAL Rule (Article 33), giving the Tribunal considerable flexibility not usually available to arbitral tribunals.19 Despite these challenges to the appropriateness of the Tribunal’s contribution to the law of international investment, its decisions have, in point of fact, been used as an authoritative source for arbitral tribunals in investor-state disputes. Albeit extraneous to its original mandate, the Tribunal’s hybrid nature, its almost permanent character,20 its contribution to the interpretation of procedural rules, the accessibility of its rich and comprehensive

15 K-H Böckstiegel, ‘The Iran-United States Claims Tribunal’ in U Franke, A Magnusson and J Dahlquist (eds), Arbitrating for Peace: how arbitration made a difference (Alphen aan den Rijn, Kluwer, 2016) 91–102, 95. 16 As argued by Michel Virally, ‘Je vois en lui très certainement un tribunal arbitral institué par voie d’accord entre deux Etats et destiné à régler un différend entre ces deux Etats, bien qu’il éclate en quelque sorte en une multitude de réclamations individuelles. Le Tribunal ayant été qualifié au point de vue du droit international public, je ne pense pas qu’on puisse lui donner, qu’on puisse même rechercher, une autre qualification. …. J’ai des doutes en ce qui concerne le parallèle qui a été fait avec le C.I.R.D.I Nous sommes en présence de quelque chose qui est fondamentalement différent’. In B Stern and M-F Labouz, ‘Le Tribunal des différends irano américains’ (1984) 1 Cahiers du Centre de droit international de Nanterre (CEDIN) 50. 17 M Sornarajah, The Pursuit of Nationalized Property (Leiden, Martinus Nijhoff Publishers, 1986): ‘the jurisprudential value of the awards … [is] open to doubt on the ground that they were based on an agreement settling a political dispute and that there was an effort made by the Tribunal to approach issues in a manner favouring compromise’ (202), quoted in: Caron, above (n 14) and (n 4). 18 As a recent Data analysis establishes, the jurisprudence of the Iran-US Claims Tribunal does not lend support to the allegation that the Tribunal, by its internal practice and structure, was bound to produce politicised decisions. See D Charlotin, ‘A data Analysis of the Iran-US Claims Tribunal’s Jurisprudence – Lessons for International Dispute Settlement Today’ (2019) 10 Journal of International Dispute Settlement 443. 19 Article V reads as follows: ‘The Tribunal shall decide all cases on the basis of respect for law, applying such choice of law rules and principles of commercial and international law as the Tribunal determines to be applicable, taking into account relevant usages of the trade, contract provisions and changed circumstances.’ 20 At its inception, it was thought that the Tribunal would only be necessary for just a few years. However, as Elihu Lauterpacht explained in 1982, a quick calculation revealed that the Tribunal would need something over 13 years to complete its work: ‘Assume that each of the [major] 900 cases takes altogether no more than ten days of the Tribunal’s

64 Bruno Simma and Cristina Hoss jurisprudence, as well as the paucity of other, relevant, jurisprudence in the early years of investor-state arbitration, have made the Tribunal’s case law an important point of reference in international investment law. Presenting the entirety of the Tribunal’s contribution to the law of international investment cannot be accomplished within the confines of this chapter. Therefore, only those major decisions that have had a lasting impact on investor-state arbitration and that are repeatedly relied upon in parties’ submissions, awards and opinions are presented here. In passing, it can be observed that many of the Tribunal’s leading decisions are in fact based on ICJ jurisprudence or that of its predecessor. As a result, the Tribunal contributed, albeit indirectly, to the consolidation of the Permanent Court of International Justice (‘PCIJ’) and ICJ jurisprudence in the field of the law of international investment. This should not, however, detract from the substantial impact that the case law of the IUSCT has had, particularly in the early years, on the development of international investment law.21 For the sake of clarity and ease of reference, the contribution of the IUSCT to investment law can be divided into two parts: first, its contribution to the interpretation of the UNCITRAL Rules, and second, its contribution to the substantive law of investor-state arbitration. With regard to the former, the fact that the Tribunal was the first arbitral institution to apply the 1976 UNCITRAL Rules (albeit in an adapted version) in published decisions, explains why later arbitrations turned to its case law for guidance in the interpretation of these rules. Regarding the latter, the Tribunal’s decisions regarding substantive law applied in contemporary investor-state arbitration were among the first decisions in this, at that time, emerging area of international law. The contribution of the IUSCT in this respect seems most relevant on questions like the determination of nationality, expropriation, state responsibility for internationally wrongful acts, attribution and compensation in particular. According to a citation analysis from 2008, almost 45 per cent of ICSID awards on the merits and over 20 per cent of ICSID decisions and awards on jurisdiction, cite precedents from the Tribunal’s case law. According to that study, the most commonly cited Tribunal awards relate to the determination of the point in time at which expropriation occurs and to the standard of compensation to be applied in a given situation.22 According to an earlier study, also time – ten days to deal with the interlocutory stages, read the pleadings, cope with preliminary objections, hold prehearing conferences as well as oral hearings on the merits, and then prepare the awards. That means 9,000 working days will be required, or 3,000 for each of the existing chambers. There are some 220 working days in the Tribunal’s effective year, if regard is had to the national holidays of the Netherlands, the United States, and Iran; the Scandinavian inclination toward summer holidays beginning in July; the sacrosanctity of August to the French; and the universal inclination to adopt the Englishman’s week-end. On this basis unless some new techniques are adopted, it will take each chamber something over 13 years to finish its work.’ E Lauterpacht, ‘The Iran-United States Claims Tribunal – An Assessment’ in M Bender (ed), Private Investors Abroad – Problems and Solutions in International Business in 1982 (Plano, Southwestern Legal Foundation International and Comparative Law Center, 1982) 213–228, 215, quoted in R Khan, The Iran-United States Claims Tribunal: Controversies, Cases and Contribution (Leiden, Martinus Nijhoff Publishers, 1990) xii, above (n 8). 21 See, in particular, C Drahozal, ‘The Iran-U.S. Claims Tribunal and Investment Arbitration: A Citation Analysis’ (2008) 3 Transnational Dispute Management. It cannot, however, be denied that some tensions persisted in particular in the Tribunal’s early days. Leaving aside the attack on a Third-Country Member by two Iranian arbitrators, several opinions and statements attached to Tribunal Awards reveal tense relations between some Members, see, eg ‘Statement by Judge Khalilian as to Why it would have been premature to sign the Award’, explaining why he refused to sign the Award of Chamber Two in the case Phillips Petroleum Company v The Islamic Republic of Iran, The National Iranian Oil Company Award No 425-39-2 (29 June 1989) reprinted in 21 Iran-USCTR, 79 (194). 22 See Drahozal, above (n 21) 8, the three IUSCT awards most frequently cited by ICSID arbitrations are: Amoco Int’l Finance Corp and The Islamic Republic of Iran, Partial Award No 310-56-3 (14 July 1987) reprinted in 15 Iran-USCTR, 189; Phillips Petroleum Company Iran and The Islamic Republic of Iran, National Iranian Oil Company, Award No 425-39-2 (29 June 1989) reprinted in 21 Iran-USCTR, 79 and Starrett Housing Corp and The Islamic Republic of Iran, Award No ITL 32-24-1 (19 December 1983) reprinted in 4 Iran-USCTR, 122.

The Contribution of the Iran-United States Claims Tribunal 65 parties in North American Free Trade Agreement (‘NAFTA’) cases made extensive use of the Tribunal’s case law,23 although the NAFTA awards themselves do not appear to make frequent express references to case law from the IUSCT.24

II. CONTRIBUTIONS TO THE INTERPRETATION OF UNCITRAL PROCEDURAL LAW

At the time of its creation, the Tribunal’s procedural framework was solely determined by the CSD. In particular, Article III(2) of that Declaration determined that the Tribunal ‘shall conduct business in accordance with the arbitration rules of the United Nations Commission on International Trade Law (UNICTRAL) except to the extent modified by the Parties or by the Tribunal to ensure that this Agreement can be carried out’. The UNCITRAL Rules designed for ad hoc arbitration had been adopted on 28 April 1976 under the auspices of UNCITRAL,25 and the UN General Assembly recommended their use underlining the importance of a set of rules ‘acceptable in countries with different legal social and economic systems’.26 During the negotiation of the Accords, the US and Iran were thus in the fortunate position of having these multilaterally elaborated rules for ad hoc arbitration at their disposal for inclusion in the CSD. They appeared acceptable to both, since these rules did not have their origins in any particular domestic legal system but constituted a compromise between civil and common law procedure.27 The choice of the UNCITRAL Rules as a basis for the procedural rules of the Tribunal turned out be both bold and ingenious: they were concise and sufficiently flexible for the Tribunal to carry out its work.28 For instance, in view of the numerous challenges against arbitrators, one of the crucial elements in those rules without which the Tribunal would have been unable to proceed very quickly, were the provisions set out in Articles 10–13 of the UNCITRAL Rules.29 While the procedural framework was thus set by the CSD, it fell upon the full Tribunal with the assistance of the Parties’ agents, participating without a right to vote, to make certain adaptations and to determine the procedural rules which it considered necessary to execute its tasks.30

23 CS Gibson and CR Drahozal, ‘Iran-United States Claims Tribunal Precedent in Investor-State Arbitration’ (2006) 23 Journal of International Arbitration 521, 540. 24 See Drahozal, above (n 21) 5. 25 For official documents on the travaux préparatoires of the UNCITRAL Rules see: https://perma.cc/U9WM-Y8V9. 26 General Assembly Resolution 31/98 adopted on 15 December 1976. 27 See P Bellet, Foreword in: ‘Symposium on the Iran-United States Claims Tribunal’ (1984) 16 Law and Policy in International Business 667, 670. 28 See in particular Art 15 of the UNCITRAL Rules ‘… the arbitral tribunal may conduct the arbitration in such manner as it considers appropriate, provided that the parties are treated with equality and that at any stage of the proceedings each party is given a full opportunity of presenting its case’. 29 As judge Pierre Bellet noted: ‘It might even be asked whether the Tribunal would have been able to function at all if the UNCITRAL Rules had not been agreed on by the two Governments. For example, thanks to Articles 10 and 13 of the UNCITRAL Rules, it was easy to put an end to the challenges instituted by the Iranians against Judge Mangard, which were promptly dealt with by decision of the appointing authority, Chief Justice Moons of the Dutch Supreme Court’. See Bellet, above (n 27) 670; see also GH Aldrich, ‘The selection of arbitrators in DD Caron, JR Crook, ‘The Iran-United States Claims Tribunal and the process of international claims resolution’ (2000) American Society of International Law 65, 71. 30 For authoritative commentaries on the adapted version of the UNICTRAL arbitration rules see JJ van Hof, Commentary on the UNCITRAL Arbitration Rules: The application by the Iran-U.S. Claims Tribunal (Alphen aan den Rijn, Kluwer, 1991); M Pellonpää and D Caron, The UNCITRAL Arbitration Rules as interpreted and applied by the Iran-United States Claims Tribunal (Helsinki, Finnish Lawyers’ Publishing, 1994). See also Lagergren, above (n 2) 28.

66 Bruno Simma and Cristina Hoss The final version of the Tribunal’s Rules of Procedure (‘the Rules’), adopted on 3 May 1983, was structured as follows: First, the provisions of the UNCITRAL Rules, second, the text of the modifications made pursuant to Article III(2) of the CSD, and, finally, the explanatory ‘Notes’ accompanying certain provisions and indicating how the Tribunal was to interpret or implement the modified UNCITRAL Rules. Later, the Tribunal issued a consolidated version of its Rules, consisting of the text of the modified Rules, followed by the Tribunal’s Notes, but without showing the original version of the UNCITRAL Rules.31 In the almost 40 years of the Tribunal’s existence, only one amendment to the Tribunal’s Rules has been made, namely the addition of a 5th paragraph to Article 13, the so-called ‘Mosk-Rule’, introduced after the resignation of judge Mosk on 18 November 1983. It deviates from Article 13 of the UNCITRAL Rules in so far as it determines that a resigning member shall continue to serve as a member of the Tribunal in respect of all cases in which he had participated in a hearing on the merits, and for that purpose shall be considered a member of the Tribunal instead of the person replacing him.32 Most significantly, shortly after its creation, the Tribunal opted for the publication of its awards, resulting in a wide dissemination of its practice, which was all the more important in the early days of the application and interpretation of the UNCITRAL Rules. Initially, the decisions and awards were not published. Now, the awards and opinions of members are accessible through the Reports of the IUSCT (Iran-U.S. C.T.R.),33 making the impact of the Tribunal’s case law on international arbitration possible in the first place. While the transcripts of hearings and written pleadings are not published, one finds extensive summaries of the parties’ arguments drawn from their pleadings included in the text of Tribunal’s awards. Two contributions of the Tribunal to the interpretation and application of the UNCITRAL Rules might be singled out: (1) its flexible approach to counter claims; and (2) its admission of third-party statements.

A. Counterclaims With regard to counterclaims,34 the Tribunal modified Article 19(3) of the 1976 UNCITRAL Rules. According to the original provision, a counterclaim must ‘arise out of the same contract or rely on a claim arising out of the same contract for the purpose of a set-off ’. The Tribunal’s Rules added the words ‘if such counterclaim or set-off is allowed under the Claims Settlement Declaration’, thereby expanding the basis of presenting counterclaims. For instance, Article II(1) of the CSD refers to counterclaims ‘arising out of the same contract, transaction or occurrence that constitutes the subject matter of [a] national’s claim’.35 Counterclaims have been very

31 The

Tribunal’s Rules of Procedure are available at https://perma.cc/MD4K-SFAR. rule is not without difficulties, as underlined by van Houtte and Concolino, above (n 2) 315; for an extensive account of certain controversies arising out of the application of the ‘Mosk Rule’, see C Brower and J Brueschke, The Iran-United States Claims Tribunal (Leiden, Martinus Nijhoff Publishers, 1997) 141–151. 33 The Iran-USCT R are published by Grotius Publications under the editorial supervision of the Lauterpacht Centre for International Law of the University of Cambridge. To date 38 volumes have been published. The Tribunal’s high level of activity in the early years is reflected in the fact that by 1996, 27 volumes had already been published, as noted by Lagergren, above (n 2) 30. Initially, the Tribunal’s decisions and awards were not published, as noted by TL Stein, who lamented that ‘[s]cholarship on the Tribunal’s work has so far been hampered by the relative inaccessibility of the Tribunal’s rules, orders and decisions’. See TL Stein, ‘Jurisprudence and Jurists’ Prudence: The Iranian-Forum Clause Decisions of the Iran-U.S. Claims Tribunal’ (1984) 78 American Journal of International Law 1–52, 1, above (n 2). 34 See also AK Björklund, ‘Counterclaims’, ch 17 in this volume. 35 For the Tribunal’s early practice, see van Hof, above (n 30) 127–134. 32 This

The Contribution of the Iran-United States Claims Tribunal 67 frequently submitted before the Tribunal, mainly because under the CSD, claims by the states’ parties against nationals of the other state were not permitted. The filing of counterclaims was therefore the appropriate way to allow the US or Iran to make claims against nationals of the other state. The Tribunal noted, in particular, that: a right of counterclaim is normal for a respondent, but it is admitted only in response to a claim and it does not mean, by analogy that each State is allowed to submit claims against nationals of the other State. It means, a contrario, just the opposite. Certainly also, several specified sorts of claims are expressly excluded by the same paragraph, but such exclusion is in ‘the framework’ of this paragraph, i.e.: concerning claims made by citizens against States. Such specific exclusions do not mean that, outside of the framework, any claim which has not been excluded, should be admitted ….36

Numerous cases of the Tribunal have dealt with the issue of jurisdiction over counterclaims, thereby building a considerable body of case law in the field.37 Arguably, the basis for the Tribunal’s approach to counterclaims appears to be procedural fairness. The respondent state would be placed in a position of disadvantage if its lack of right to bring claims against nationals also included the inherent prohibition to bring counterclaims. In a framework of settling disputes arising from economic transactions and investments, it would be profoundly unfair to exclude the possibility of counterclaims, arising almost systematically from these transactions. It would nonetheless be inapposite to extend the Tribunal’s approach to counterclaims beyond that particular forum. The historical context in which the Tribunal was created as well as the modification of Article 19(3) of the UNCITRAL Rules signify that its approach to counterclaims cannot be easily transposed to investor-state arbitration generally. Nonetheless, the Tribunal’s findings have been frequently referred to by arbitral tribunals operating under the UNCITRAL Rules in complex cases relating to the question of the connection between principal claims and counterclaims.38 In the context of direct claims between the two states parties, the so-called ‘official claims’, Article II(2) of the CSD is silent as to the permissibility of counterclaims.39 The Tribunal, however, found that, in light of the interpretation of the CSD, as well as Article 19(3) of the UNCITRAL Rules, in accordance with Article 31 of the Vienna Convention of the Law of Treaties, Article II(2) of the CSD should be interpreted ‘as providing the Tribunal with jurisdiction to entertain official counterclaims’.40 The decisive factor was the subsequent practice of the Parties in relation to counterclaims, notably the fact that Iran, as the Party contesting the Tribunal’s jurisdiction over counterclaims, had itself already submitted counterclaims in official claims in the past.41 36 See, in particular, the interpretation of Article II of the CSD given by the Tribunal in Iran and the US, Case No A/2 Request for Interpretation: Jurisdiction of the Tribunal with respect to claims by the Islamic Republic of Iran against nationals of the United States of America, Decision No DEC1-A2-FT (13 January 1982) reprinted in 1 Iran-USCTR, 101 (103). 37 See C Antonopoulos, Counterclaims before the International Court of Justice (The Hague, TMC Asser Press/ Springer, 2011) 27. 38 See, eg Saluka Investments BV v The Czech Republic, UNCITRAL, Decision on Jurisdiction over the Czech Republic’s Counterclaim (7 May 2004) 1, para 3 (paras 68–74). 39 Article II(2) of the Claims Settlement Declaration reads as follows:

‘The Tribunal shall also have jurisdiction over official claims of the United States and Iran against each other arising out of contractual arrangements between them for the purchase and sale of goods and services.’ 40 Islamic Republic of Iran and The United States of America, Case No B1 (Counterclaim), Award No ITL 83-B1-FT (9 September 2004), reprinted in 38 Iran-USCTR, 77 (107–126). 41 ibid 126.

68 Bruno Simma and Cristina Hoss B. Participation of amici curiae42 Article 15(1) of the UNCITRAL Rules, which was incorporated without modification in the Tribunal’s Rules, provides for a flexible approach to the organisation of an arbitral procedure. It reads as follows: Subject to these Rules, the arbitral tribunal may conduct the arbitration in such manner as it considers appropriate, provided that the parties are treated with equality and that at any stage of the proceedings each party is given a full opportunity of presenting his case.

As mentioned above, the Tribunal issued so-called ‘Notes’, essentially interpretative guidelines, to some of its Rules, including five Notes to Article 15. In Note 5 to Article 15(1) of its Rules, the Tribunal interprets that Article as permitting the submission of statements from non-parties to a dispute. It states: The arbitral tribunal may, having satisfied itself that the statement of one of the two Governments – or, under special circumstances, any other person – who is not an arbitrating party in a particular case is likely to assist the tribunal in carrying out its task, permit such Government or person to assist the tribunal by presenting oral or written statements.

This Note has been invoked by the US and by Iran in cases to which they were not Parties, but also by other entities, including foreign banks, as noted by the NAFTA arbitral tribunal in its landmark decision on the petitions to intervene as amici curiae in the case of Methanex v United States of America.43 Referring to the IUSCT’s decision in Case No A/15,44 and its acceptance of amici curiae briefs, the Methanex tribunal concluded: For present purposes, the authoritative guide to the exercise of the Iran-US Claim Tribunal’s discretion under Article 15(1) and this award demonstrate that the receipt of written submissions from a non-party third person does not necessarily offend the philosophy of international arbitration involving states and non-state parties.

III. CONTRIBUTIONS TO THE CORPUS IURIS OF INVESTMENT LAW

With regard to the substantive law of international investment, the Tribunal’s most significant contributions concern: (1) the determination of nationality; (2) its own concept of ‘expropriation’; (3) the standard of compensation and (4) the attribution of acts to a state. A. Nationality Regarding the question of nationality of the investor, which determines whether the investor is protected by a treaty or an investment agreement, this determination is generally made by the 42 On the evolution of third parties in international investment arbitration see E de Brabandère, Amicus curiae intervention: from NAFTA to the intra-EU saga, ch 11. 43 Methanex Corporation v United States of America, Decision of the Tribunal on Petitions from Third Persons to intervene as amici curiae (15 January 2001) para 32, https://perma.cc/M4PM-TXPS. The reference to Case No A/18 was again used by the ICSID Tribunal in the case United Parcel Service (UPS) v Canada, Decision of the Tribunal on Petitions for Intervention and Participation as amici curiae (17 October 2001) ICSID Case No UNCT/02/1, para 64 (including an erroneous reference to Volume 2 of the Iran-USCTR in the Methanex Decision, which should actually read Volume 12). 44 Iran and United States, Case No A/15, Award No 63-A/15-FT (20 August 1986) reprinted in 12 Iran-USCTR, 40 (43).

The Contribution of the Iran-United States Claims Tribunal 69 treaty or agreement in question, or, more directly, by the national law of the claimant. For the Tribunal, Article VII of the CSD defines the term ‘national’ as follows: 1. A ‘national’ of Iran or of the United States, as the case may be, means (a) a natural person who is a citizen of Iran or the United States; and (b) a corporation or other legal entity which is organized under the laws of Iran or the United States or any of its states or territories, the District of Columbia or the Commonwealth of Puerto Rico, if, collectively, natural persons who are citizens of such country hold, directly or indirectly, an interest in such corporation or entity equivalent to fifty per cent or more of its capital stock.

The most prominent question dealt with by the Tribunal in this context was the issue of dual nationals. In Case No A/18, the Tribunal had to determine whether it had jurisdiction over claims of persons with dual nationality, that is to say persons who under Iranian law were Iranian, and under US law, were US citizens.45 A large number of claims had been filed by dual nationals, under Article II(1) of the CSD, and were pending before the three Chambers of the Tribunal. Iran had requested, under Article VI(4) of the CSD, that the full Tribunal examine the question of admissibility of claims filed ‘by nationals of Iran against the Government of the Islamic Republic of Iran’. For Iran, the intended function of the Tribunal being the adjudication of international claims on the basis of diplomatic protection, nationality should be determined by reference to its ordinary meaning, which, according to Iran, encompassed persons with exclusive Iranian or US nationality, while it regarded dual nationality as an ‘abnormal’ status. The US had argued that modern international law favoured an interpretation that made the determination of jurisdiction dependent on the dominant and effective nationality of each dual national.46 The Tribunal decided the issue by interpreting the CSD in accordance with Article 31 of the Vienna Convention on the Law of Treaties. The Tribunal had recourse to ‘any relevant rules of international law applicable in the relations between the parties’ used in treaty interpretation, and held that ‘[t]here is a considerable body of law and legal literature, which leads the Tribunal to the conclusion that the applicable rule of international law is that of dominant and effective nationality’.47 Reviewing the ‘considerable body of law and legal literature’ cited by the Tribunal in support of its conclusion, it appears that the Tribunal quickly turned to the ICJ judgment in the Nottebohm case,48 the effects of which, according to the Tribunal, ‘have radiated throughout international law of nationality’.49 The Tribunal went on to indicate that in order to assess the dominant and effective nationality, it would consider ‘all relevant factors, including habitual residence, center of interests, family ties, participation in public life and other evidence of attachment’, ie all the factors already identified by the ICJ.50 The Nottebohm case, as the Tribunal readily acknowledged, did not involve claims against a state of which Nottebohm was a national, but, the Tribunal continued, the case demonstrated the approval by the ICJ of the applicability of the test of real and effective nationality. Whether the reference to the Nottebohm case was suitable for the purposes of determining the admissibility of claims of dual nationals against one of their national states, might be a matter of debate. The fact is that the Tribunal’s Decision in Case No A/18, was generally met with 45 Iran and United States, Case No A/18, Decision No DEC 32-A18-FT (6 April 1984) reprinted in 5 Iran-US-CTR, 251. 46 ibid

254–259. 260. 48 Nottebohm case (Judgment) [1955] ICJ Reports 4, 22. 49 ibid 263. 50 ibid 265. 47 ibid

70 Bruno Simma and Cristina Hoss approval and remains at least as visible in contemporary investment law as the Nottebohm case itself.51 Regarding the no less important issue of nationality of legal persons, according to Article VII of the CSD the answer to this question with regard to companies or other legal entities organised under the laws of Iran or the US depends on whether ‘natural persons who are citizens of such country hold, directly or indirectly, an interest in such corporation or entity equivalent to fifty per cent or more of its capital stock’.52

B. Expropriation The Tribunal’s extensive case law on expropriation is based on Article II of the CSD, according to which it has jurisdiction, inter alia, to decide claims arising out of ‘expropriations or other measures affecting property rights’. The addition of the words ‘or other measures affecting property rights’ has been seen as constituting a lex specialis for the Tribunal, thus limiting the role of Tribunal jurisprudence to the specific context of the resolution of Iran-US claims.53 Further, the reference to ‘other measures affecting property rights’ suggests that the Tribunal would also have jurisdiction over acts which do not constitute formal acts of expropriation but which would qualify as acts of ‘indirect’ or ‘creeping expropriation’.54 It is this definition of the Tribunal’s jurisdiction, which enabled it to leave its mark in the field of the international law on expropriation. For instance, in the Starrett Housing Corp. and The Islamic Republic of Iran Interlocutory Award,55 the Tribunal held that the property interest taken by the Government of Iran comprised not only physical property but also the right to manage and complete construction, deliver the apartments in question and collect the proceeds of any sales. The Tribunal further held that while Iran had not issued any law or decree nationalising or expropriating the claimants, it is recognized in international law that measures taken by the State can interfere with property rights to such an extent that these rights are rendered so useless that they must be deemed to have been expropriated, even though the State does not purport to have expropriated them and the legal title to property formally remains with the original owner.56

In that case, the claimants had contended that the accumulation of various measures taken by the Government of Iran, including the appointment of a temporary manager by a decree of 30 January 1980, caused the claimants’ inability to exercise their rights, resulting in the deprivation of the effective use, control and benefit of their property.

51 See recently Ballantines v Dominican Republic, PCA case No 2016-17, Final Award (3 September 2019). See also L Reed and J Davis, ‘Who is a Protected Investor?’ in M Bungenberg and others (eds) International Investment Law (Munich, CH Beck, 2015) 625–627. For a monograph on Case No A/18, Decision No DEC 32-A18-FT see M Aghahosseini, Claims of dual nationals and the development of customary international law, Issues before the Iran-United States Claims Tribunal (Leiden, Martinus Nijhoff Publishers, 2007). 52 Mohebi, above (n 8) 77. 53 As noted by Veijo Heiskanen, however, in practice the Tribunal has used the ‘other measures affecting property rights’ sparingly and most claims have been dealt with from the angle of expropriation. See V Heiskanen, ‘The doctrine of Indirect Expropriation in light of the practice of the Iran-United States Claims Tribunal’ (2006) 3(5) Transnational Dispute Management 5. 54 See van Houtte and Concolino, above (n 2) 323. 55 Starrett Housing Corp and The Islamic Republic of Iran, Award No ITL 32-24-1 (19 December 1983) reprinted in 4 Iran-USCTR, 122. 56 ibid154.

The Contribution of the Iran-United States Claims Tribunal 71 The Tribunal held that it was the date on which the temporary manager had been appointed that ought to be considered as the date of the taking or, in the words of the Tribunal, that ‘by the end of January 1980 the Government of Iran had interfered with the claimants’ property rights in the Project to an extent that rendered these rights so useless that they must be deemed to have been taken’.57 As to events occurring prior to that date, the Tribunal considered that there was no reason to doubt that those events seriously affected the claimants’ operations of the housing projects. The Tribunal nonetheless considered that investors had to assume a risk of strikes, lockouts, and other disturbances as well as changes in the economic or political situation. The Tribunal held, in particular, that such events, including a revolution, do not, as such, amount to a taking of the claimants’ contractual rights and shares. The date on which compensation was calculated was, for practical reasons, fixed one day after the date of the taking, which in that specific case was 31 January 1980.58 This definition of deprivation was confirmed by the Tribunal in the Tippetts, Abbett, McCarthy, Stratton and TAMS-AFFA Consulting Engineers of Iran case,59 pursuant to which ‘[a] deprivation or taking of property may occur under international law through interference by a state in the use of that property or with the enjoyment of its benefits, even where legal title to the property is not affected.’60 In that case, the contested measure was the Government of Iran’s appointment of a manager who assumed sole control over the TAMS-AFFA entity, a company providing engineering services to Tehran International Airport, initially controlled jointly by a US partnership (TAMS: Tippets, Abbot, McCarthy, Stratton) and an Iranian firm (AFFA: Aziz Farmanfarmaian and Associates). In assessing the deprivation, the Tribunal took the view that the intention of the government in taking certain measures was not a decisive factor for the qualification of those measures as ‘takings’. The Tribunal held that the objective effects of those measures on the owner and the reality of their impact were more important than such intent.61 The Tribunal found that Iran had effectively expropriated the US entity, TAMS, by taking its 50 per cent interest in the joint partnership with AFFA, and found, citing the Chorzów Factory62 and the Norvegian Shipowner Claims case,63 that ‘under international law and general principles of law’, the compensation due to the claimant was ‘the full value of the property of which it was deprived’.64 Whether intended or not, the Tribunal’s liberal approach to expropriation, which has been coined as the ‘effects doctrine’,65 set out in these early cases and confirmed in subsequent

57 ibid

155. 156. 59 Tippetts, Abbett, McCarthy, Stratton and TAMS-AFFA Consulting Engineers of Iran, Award No 141-7-2 (29 June 1984) reprinted in 6 Iran-USCTR, 219. 60 ibid 225–226. 61 ibid; in Phillips Petroleum Company Iran v The Islamic Republic of Iran, the Tribunal confirmed that it did not need to determine the intent of the Government of Iran. It held, however, that ‘where the effects of actions are consistent with a policy to nationalize a whole industry and to that end expropriate particular alien property interests, and are not merely the incidental consequences of an action or policy designed for an unrelated purpose, the conclusion that a taking has occurred is all the more evident’. (Phillips Petroleum Company Iran and The Islamic Republic of Iran, National Iranian Oil Company, Award No 425-39-2 (29 June 1989) reprinted in 21 Iran-US CTR, 79 (115). 62 Case concerning the Factory at Chorzów (Merits) [1928] PCIJ Series A No 17, 47. 63 Norwegian Shipowners’ Claims, I RIAA (1922), 307–346. 64 Tippetts, Abbett, McCarthy, Stratton and TAMS-AFFA Consulting Engineers of Iran, Award No 141-7-2 (29 Jun. 1984) reprinted in 6 Iran-USCTR, 219 (225). 65 Heiskanen, above (n 53) 9; see above n 24 for more references to arbitral awards using a similar approach. 58 ibid

72 Bruno Simma and Cristina Hoss cases, has had a lasting impact on international investment law.66 The Tribunal’s approach in assessing measures affecting property rights appears to presume that those measures were not taken in conformity with the legal requirements of permissible expropriation under international law. As noted by Veijo Heiskanen, the claimants were ‘rarely required to show that the measures complained of were not for public purpose, discriminatory or otherwise irregular und thus unlawful under international law’.67 In the context of mass-claims arising out of the Iranian revolution or similar situations, for instance armed conflicts, this approach might be justifiable. However, it should be used with great caution outside the context of mass-claims, when the illegality of an expropriation must be positively established.68

C. Standards of Compensation As to standards of compensation, the Tribunal has mostly relied on customary international law, which was also due to the sparsity of relevant international case law in the field. The Tribunal’s most cited decision appears to be its Partial Award in the Amoco Int’l Finance Corp. case.69 There the claim arose out of the Khemco Agreement, which had been concluded between Amoco and the Iranian National Petrochemical Company, whereby the parties agreed to form a joint venture company, Khemco, for the purpose of building and operating a plant for the production of sulphur, natural gas liquids and liquefied petroleum gas. According to the claimant, by the end of July 1979, the respondents had ‘totally and unequivocally’ breached and repudiated the Khemco agreement and had expropriated Amoco International’s rights under the agreement, for their own benefit, use and ownership. The respondents did not deny that an expropriation had taken place. The parties agreed that the 1977 civil unrest had affected production at the plant and, as the events of the Islamic Revolution unfolded in 1978, strikes in the oil industry had disrupted production and had hampered operation in the facilities of Khemco. It was also common ground that the expatriate staff members of Khemco decided to leave Iran in December 1978. The parties agreed and the Tribunal concurred that these events amounted to a state of force majeure. Such a force majeure situation did not, however, lead to the termination of the Khemco Agreement. The expropriation of Amoco, as conceded by the respondents, took place by the enactment of the Single Article Act of 8 January 1980 by the Revolutionary Council of Iran nullifying

66 See R Dolzer and C Schreuer, Principles of International Investment Law, Second edition (Oxford, Oxford University Press, 2012) 112–115. But see the Award rendered in the arbitration Methanex Corp v United States of America, Final Award of the Tribunal on Jurisdiction and Merits under ch1 1 of the North American Free Trade Agreement and the UNCITRAL Arbitration Rules, Final Award of the Tribunal, 3 August 2005, https://perma. cc/2R6P-7RRF. The Tribunal therein found that ‘Methanex is correct that an intentionally discriminatory regulation against a foreign investor fulfils a key requirement for establishing expropriation. But as a matter of general international law, a non-discriminatory regulation for a public purpose, which is enacted in accordance with due process and, which affects, inter alios, a foreign investor or investment is not deemed expropriatory and compensable unless specific commitments had been given by the regulating government to the then putative foreign investor contemplating investment that the government would refrain from such regulation.’ (Part IV-Chapter D) 4, para 7) The Tribunal found that the contested regulation was non-discriminatory, and was accomplished with due process and, thus, was not an expropriation. 67 Heiskanen, above (n 53) 16. 68 ibid 17. 69 Amoco Int’l Finance Corp. and The Islamic Republic of Iran, Partial Award No 310-56-3 (14 July 1987) reprinted in 15 Iran-USCTR, 189.

The Contribution of the Iran-United States Claims Tribunal 73 the Khemco Agreement and thus, according to the respondents, lawfully implementing the nationalisation policy adopted in 1951. The issue of contention between the parties was whether the expropriation was lawful or unlawful. First, the Tribunal determined the applicable law. The 1955 Treaty of Amity between the US and Iran70 had already been found to be applicable in earlier decisions of the Tribunal71 and the Chamber in the Amoco case did not see any reasons to depart from this earlier finding. However, for the analysis of the contribution of the Tribunal to international investment law, the more relevant part of that decision relates to the determination of customary international law, which the Tribunal resorted to as lex generalis, ‘in order to fill in possible lacunae of the 1955 Treaty, the Treaty being considered lex specialis’.72 First the Tribunal considered that the right of states to nationalise foreign property for a public purpose ‘is today unanimously accepted’73 and second set out the condition of ‘prompt payment of just compensation’ as an equally accepted rule of customary international law. It is with regard to this second condition, the standard of compensation, that the Tribunal made a lasting contribution to the law of international investment.74 While both parties relied on the Chorzów Factory judgment of the Permanent Court of International Justice75 as a point of reference for the standard of compensation, their interpretation of that judgment diverged considerably. The claimant asserted that in case of unlawful expropriation, compensation should be more than the ‘full equivalent’ to be applied to lawful expropriation, while the respondents argued that the measure of compensation in cases of lawful expropriation should be ‘substantially less’ and ‘be limited to the unjustified enrichment realised by the nationalising state, with no compensation for lost profit’.76 The Tribunal agreed with the parties as to the relevance of the Chorzów Factory case as the ‘most authoritative exposition of the principles applicable in this field’. For the Tribunal, although Chorzów only dealt with the question of reparation of damage resulting from an unlawful expropriation, it was also ‘illuminating in analysing the lawful expropriation’, present in the case at hand.77 The Tribunal thus analysed the findings of the PCIJ on the possible standards of compensation and inferred that the just price of what had been expropriated was to be determined by reference to the ‘value of the undertaking at the moment of dispossession’.78 The Tribunal rejected the discounted cash flow method as incorrectly extending to lost profits (lucrum cessans) in compensating lawful expropriation and opted for a method limited to the net value of the expropriated assets.79 This, in the Tribunal’s view, included paying a going concern value but excluded future profits.80 The Tribunal noted, in particular, that even for the purpose of restitutio – when 70 Treaty

of Amity, Economic Relations and Consular Rights, Signed at Tehran, on 15 August 1955, 284 UNTS, 93. Corporation and Islamic Republic of Iran, Award No 184-161-1 (12 August 1985) reprinted in 8 Iran-US CTR, 373; Phelps Dodge Corp. and Islamic Republic of Iran, Award No 217-99-2 (19 March 1986) reprinted in 10 Iran-USCTR, 121; Sedco Inc and National Iranian Oil Company, Award No ITL 59-129-3 (26 March 1986) reprinted in 10 Iran-USCTR, 180. 72 Amoco Int’l Finance Corp. and The Islamic Republic of Iran, Partial Award No 310-56-3 (14 July 1987) reprinted in 15 Iran-US CTR, 189 (222). 73 ibid. 74 ibid 244–252. 75 Factory at Chorzów, above (n 63) 47. 76 Amoco Int’l Finance Corp and The Islamic Republic of Iran, Partial Award No 310-56-3 (14 July 1987) reprinted in 15 Iran-USCTR, 189 (246). 77 ibid 247. 78 ibid. 79 ibid 258–265. 80 ibid 246–271. 71 INA

74 Bruno Simma and Cristina Hoss assessing the standard of compensation deemed to ‘wipe out the consequences of an illegal act and re-establish the situation which would, in all probability, have existed if this act had not been committed’81 – the PCIJ had taken into consideration lucrum cessans only for a limited and rather short period of time. In particular, that quantification implied no projection into the future but was determined at the date of the judgment. Thus, the Tribunal contributed to the consolidation of the Chorzów Factory’s determination of the standard of compensation. It did not stop there, however. The Tribunal went on to develop sophisticated case law on compensation. It has applied and refined different standards,82 depending on the lawfulness or unlawfulness of the expropriation, and applied different concepts of evaluation such as discounted cash flow, fair market value or going concern value, depending on the circumstances of each case.83

D. Attribution A frequent issue appearing in investment law is the attribution of certain acts to the state. In this field, the Tribunal has contributed to the consolidation of certain rules of attribution applicable beyond international investment law.84 The Tribunal frequently engaged in the examination of questions of attribution when examining the liability of the Government of Iran for certain acts. The obvious case would be that of agents of the state, including ministries, officials, organs whether judicial, legislative or executive which, according the rules on state responsibility, act on behalf of the state. Thus, the enactment of legislation, as seen in Amoco Int’l Finance Corp., was regarded as an act undisputedly attributable to the Government of Iran.85 A similar reasoning was applied in cases where actions emanated directly from ministries of the Government of Iran, as seen in the Tippetts case.86 Given the historical context of its creation, it was to be expected that the Tribunal’s jurisprudence would include findings on the attribution of acts of revolutionary movements to the state, in particular, of acts of the Revolutionary Guards. This issue was addressed in two major decisions. In the case of William L. Pereira Associates,87 the claimant alleged that its property, offices and company car had been confiscated by the Revolutionary Guards, pursuant to a Notice of Confiscation issued on 5 October 1980. The findings on attribution in that case related to facts which had occurred after the official recognition of

81 ibid

247. in particular Phillips Petroleum Company Iran v The Islamic Republic of Iran, the National Iranian Oil Company, IUSCT Case No 39, Award No 425-39-2 (29 June 1989) reprinted in 21 Iran-US CTR, 79. 83 On approaches to valuation in International Investment Law, see I Marboe, ‘Reparation Cases: Applicable Principles in International Investment Arbitration’, ch 26 in this volume. 84 See the codification efforts of the International Law Commission in the field of the Law of State Responsibility and, in particular, Arts 4–11 (Attribution of conduct to a State) of the ILC Articles on State Responsibility with commentaries, Yearbook of the International Law Commission (2001) II, Part Two. See also DD Caron, ‘The basis of responsibility: attribution and other trans-substantive rules’ in RB Lillich and DB Magraw (eds) The Iran-United States Claims Tribunal: Its Contribution to the Law of State Responsibility (Leiden, Brill, 1998) 109 (126–173). 85 Amoco Int’l Finance Corp and The Islamic Republic of Iran, Partial Award No 310-56-3 (14 July 1987) reprinted in 15 Iran-US CTR, 189 (247). 86 Tippetts, Abbett, McCarthy, Stratton and TAMS-AFFA Consulting Engineers of Iran, Award No 141-7-2 (29 June 1984) reprinted in 6 Iran-USCTR, 219. 87 William Perreira Associates Iran and The Islamic Republic of Iran, Award No 116-1-3 (19 March 1984) reprinted in 5 Iran-USCTR, 198. 82 See

The Contribution of the Iran-United States Claims Tribunal 75 the Revolutionary Guards as forming part of the Islamic Republic, by the adoption of a Decree of May 1979, which recognised the Revolutionary Guard, and of Article 150 of the Constitution of the Islamic Republic of Iran, and also by the referendum of 2 December 1979.88 Considering that the Revolutionary Guard had thereby become de jure organs of the state at the time of the confiscation, the Tribunal determined, in one sentence, that ‘[u]nder international law the Government of the Islamic Republic of Iran must be deemed responsible for the acts of the Revolutionary Guards’.89 In the second decision, the Yeager case, the factual context was different.90 Here, the claimant argued that he had been violently expelled from Iran by members of the Revolutionary Guards on 13 February 1979, ie before the incorporation of the Revolutionary Guards in the structure of the Islamic Republic of Iran. The Tribunal examined the question of attribution in some detail, resorting, as an authoritative source, to draft articles on State responsibility that the International Law Commission had elaborated, but at the time not yet adopted. The Tribunal noted: [w]hile there is some doubt as to whether revolutionary ‘Komitehs’ or ‘Guards’ can be considered ‘organs’ of the Government of Iran, since they were not formally recognized during the period relevant to this Case, attributability of acts to the State is not limited to acts of organs formally recognized under internal law. Otherwise a State could avoid responsibility under international law merely by invoking its internal law. It is generally accepted in international law that a State is also responsible for acts of persons, if it is established that those persons were in fact acting on behalf of the State. See ILC-Draft Article 8 (a). An act is attributable even if a person or group of persons was in fact merely exercising elements of governmental authority in the absence of the official authorities and in circumstances, which justified the exercise of those elements of authority. See ILC-Draft Article 8 (b).91

In considering the facts, the Tribunal found sufficient evidence to conclude that, at the time of Yeager’s expulsion, the Revolutionary Guards had acted on behalf of and exercised elements of the governmental authority of the new Government of Iran. It was evident to the Tribunal that the new government, despite occasional complaints about a lack of discipline, stood behind the Revolutionary Guards and that those involved in the Yeager expulsion were acting ‘for’ Iran. The Tribunal further found that the new government had accepted the activity of the Revolutionary Guards in principle, and their role in the maintenance of public security, in particular. In fact, Iran had argued that its inability to discipline the Revolutionary Guards indicated that their actions could not be attributed to the state, because it showed lack of actual control. The Tribunal rejected this submission. It held that the Respondent has failed, in the Tribunal’s view, to offer satisfactory evidence that these ‘Komitehs’ or ‘Guards’ did not act in fact on behalf of the new government, or that they did not exercise elements of government authority. Rather, the evidence suggests that the new government, despite occasional 88 Article

150 of the Constitution of the Islamic Republic of Iran states:

‘The Islamic Revolution Guards Corps, organized in the early days of the triumph of the Revolution, is to be maintained so that it may continue in its role of guarding the Revolution and its achievements. The scope of the duties of this Corps, and its areas of responsibility, in relation to the duties and areas of responsibility of the other armed forces, are to be determined by law, with emphasis on brotherly cooperation and harmony among them.’ 89 William Perreira Associates Iran and The Islamic Republic of Iran, Award No 116-1-3 (19 March 1984) reprinted in 5 Iran-USCTR, 198 (227). 90 Kenneth P Yeager and The Islamic Republic of Iran, Award No 324-10199-1 (2 November 1987) reprinted in 17 Iran-US CTR, 92. 91 ibid 104.

76 Bruno Simma and Cristina Hoss complaints about a lack of discipline, stood behind them. The Tribunal is persuaded, therefore, that the revolutionary ‘Komitehs’ or ‘Guard’ involved in this Case, were acting ‘for’ Iran. Nor has the Respondent established that it could not control the revolutionary ‘Komitehs’ or ‘Guards’ in this operation. Because the new government accepted their activity in principle and their role in the maintenance of public security, calls for more discipline, phrased in general rather than specific terms, do not meet the standard of control required in order to effectively prevent these groups from committing wrongful acts against US nationals.

and concluded that: [u]nder international law Iran cannot, on the one hand, tolerate the exercise of governmental authority by revolutionary ‘Komitehs’ or ‘Guards’ and at the same time deny responsibility for wrongful acts committed by them. On the basis of the evidence in this Case, therefore, the Tribunal finds the acts of the two men who took the Claimant to the Hilton Hotel attributable to Iran.92

Whether this analysis on attribution withstands scrutiny might be subject to debate,93 but its impact has been considerable. The Tribunal’s decision has been referred to time and again by parties, arbitrators and by the International Law Commission itself.94

IV. ANALYSIS AND CONCLUSION

The Tribunal’s contribution to the development of international investment law can hardly be overstated. In the world of international investment disputes, its pioneering work in formulating legal principles and tests to determine issues of nationality, expropriation, attribution and standards of compensation have become important points of reference and guidance. The Tribunal’s findings have been used in numerous investor-state arbitrations, prompting a leading expert to compare the CSD to a ‘retrospective BIT’.95 The reasons for its significant contribution to the development of international investment law can be readily identified. First, there is the timing and the volume of its case law. The Tribunal was one of the earliest institutions to produce a substantial body of case law on the rights of private investors to bring expropriation and related claims against a state, which was years before the ‘explosion’ of investor-state arbitration that we have witnessed in recent years. A second reason can be seen in the Tribunal’s application and interpretation of procedural rules. The Tribunal was one of the first institutions of its kind to apply a (modified) version of the 1976 UNCITRAL Rules of Arbitration, influencing their application by a plethora of other tribunals and arbitral institutions. Third, the Tribunal’s decision to publish its awards not only contributed to transparency but also impact. There is no other body of case law relevant to international investment law which has a longer tradition or

92 ibid

104–105. will find that there is a certain lack of coherence in saying that the government did not exercise sufficient control to prevent the internationally wrongful act from happening and at the same time concluding that the Guards acted ‘for’ the government. 94 See, eg, Chevron and TexPet v Ecuador (II), PCA Case No 2009-23, Second Partial Award on Track II (30 August 2018) para 8.49; ILC Articles on State Responsibility with commentaries, Yearbook of the International Law Commission (2001) II, Part Two, 48 (commentary to Article 8 ‘Conduct directed or controlled by a State’) and 49 (commentary to Article 9 ‘Conduct carried out in the absence or default of the official authorities’). 95 D Caron, ‘Understanding the Claims Settlement Declaration as a Retrospective BIT’ in CR Drahozal, CS Gibson, The Iran-US Claims Tribunal at 25. The Cases everyone needs to know for Investor-State and International Arbitration (Oxford, Oxford University Press, 2007) 375. 93 Some

The Contribution of the Iran-United States Claims Tribunal 77 greater volume. Last but not least, it must be acknowledged that part of the Tribunal’s influence on the law of investment arbitration law is due to the general and legitimate tendency of ‘mimétisme judiciaire’ or ‘judicial borrowing’96 in the field of international dispute settlement and investor-state arbitration. It is based on the common assumption that in referring to other arbitral and judicial authorities, a decision gains in persuasiveness and validity. In addition, there is a sociological or ‘human factor’ that facilitated the Tribunal’s contribution to the development of international investment law. Whilst more difficult to measure, it cannot be ignored. In the early years of investor-state arbitration, only a comparatively small number of lawyers specialised in this area of law. Many of the legal professionals who had worked for or appeared before the Tribunal continued their legal careers in different capacities and settings, such as law firms, ministries, arbitral institutions, universities and, crucially, international claims commissions. It is thus not unreasonable to assume that Tribunal alumni have made a substantial contribution to the dissemination of the Tribunal’s case law in international arbitration at large. Finally, despite the Tribunal’s many contributions to the development of international investment law, it should not be forgotten that its first and foremost purpose was to provide a dispute mechanism to resolve a major international crisis between two states. Even now, at a time when the US and Iran have cut all diplomatic ties and frequently spar in public, the Tribunal continues with its important work. The creation of the IUSCT by the Algiers Accords was a defining moment in public international law and its continued existence is testimony to the success of a unique mechanism of international dispute settlement. Reducing its role to that of a contributor to investor-state arbitration law would not do justice to the Tribunal.

96 A

Pellet, ‘The Case Law of the ICJ in Investment Arbitration’ (2013) ICSID Review (2013) 28(2), 223–240 (240).

6 SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps? EVELYNE LAGRANGE*

I

N THE LATE 1980s, two companies from Hong Kong instituted arbitral proceedings which had far-reaching consequences. One company, Southern Pacific Properties (SPP), contended that the Arab Republic of Egypt had infringed on its rights derived from a state contract and from the investment statutes of the host state when cancelling the construction of tourist resorts, including one in the neighbourhood of the Gizeh pyramids (the Pyramids case) and claimed compensation for the losses incurred. The other company, Asian Agricultural Products Ltd (AAPL), invoking solely a BIT, claimed compensation for the destruction of a shrimp farm during operations by Sri Lankan military forces in the civil war with the Tamil Tigers. Both cases were brought to an ICSID tribunal. The claimants won both cases. The cancellation of SPP’s touristic project so as to protect antiquities was lawful but compensation for the expropriation was due.1 The failure to comply with the obligation to prevent losses of AAPL’s property resulted in compensation.2 Since the Asian Agricultural Products Ltd v Sri Lanka (AAPL)3 case, the investor-state dispute settlement (ISDS) has shifted from the settlement of contract claims, based on an agreement to arbitrate between the host state and a foreign investor, to the settlement of investor-state disputes on the sole, or prevalent, basis of an investment treaty containing a commitment of the host state to go to arbitration to settle disputes over the implementation of treaty provisions. AAPL v Sri Lanka is more often than not depicted as an ‘unprecedented’ case.

* Evelyne Lagrange is a Law Professor at the Sorbonne Law School (University Paris 1 (Panthéon-Sorbonne)) specialising in public international law. 1 Southern Pacific Properties (Middle East) limited v Arab Republic of Egypt, Award on the merits (20 May 1992) ICSID Case No ARB/84/3 (Southern Pacific Properties v Arab Republic of Egypt). 2 For the sake of brevity, the nature of primary obligations, standards of protection, and issues of liability (attribution, assessment of compensation of losses) will not be considered in the notice, despite their being debatable. See: P Rambaud, ‘L’affaire des Pyramides: suite et fin’ [1993] Annuaire français de droit international 567–576; M Kinnear and F Grob, ‘Chapter 12: Asian Agricultural Products Limited v Sri Lanka: Twenty-Five Years Later’ in U Franke et al, Arbitrating for Peace: How Arbitration Made a Difference (Alphen aan den Rijn, Kluwer, 2019). 3 Southern Pacific Properties v Arab Republic of Egypt, award (27 June 1990) published in English in (1991) 6 ICSID Review – Foreign Investment Law Journal 526; translated into French in (1992) 119 Journal du droit international 217 (excerpts).

80 Evelyne Lagrange In fact, the decision in Southern Pacific Properties (Middle East) v Egypt (SPP)4 predated AAPL by two years. Then, for the first time, an arbitral tribunal settled a dispute on the basis of a state’s consent to ICSID arbitration being contained in one of that state’s statutes, in this case Egypt. Despite the fact that the jurisdiction of ICSID tribunals is very rarely based on only a domestic statute, SPP undoubtedly ushered in a new era of non-contractual consent to arbitration. Considering the motives, it might have been the ground-breaking decision in terms of consent to arbitration. Contrary to classical investment or arbitration agreements,5 the consent to arbitration of the host state was in both cases drafted in abstract terms and was necessarily dissociated from the consent given by a foreign investor, who is no more than one of the addressees of these laws and treaties. The expression ‘arbitration without privity’ was successfully coined by J Paulsson6 to depict that novelty, but obscures it by the same token, notably in that it is difficult to translate in some languages.7 An affirmative formulation can be preferred: SPP and AAPL opened up an era of ‘(transnational) unilateral arbitration’.8 This less popular formulation better highlights the far-reaching consequences of the renunciation to privity: arbitration is instituted at the behest of foreign investors and (only) when they deem it proper. Self-evident as it was from the outset that consent might be given in the form of an arbitration clause or an ad hoc agreement concluded between host states and investors, ICSID was not regarded as an attractive forum in the first decades of its existence. It boomed after SPP and AAPL had broken the consubstantial link between arbitration and contract and set aside the requirement of a consent given in writing by the investor prior to the institution of proceedings. Neither treaties of friendship, commerce and navigation, nor early BITs modelled after the German-Pakistani treaty of 1959, foresaw compulsory resort to investor-state arbitration. For reasons apparently foreign to the immediate interests of capital-exporting countries as they perceived it at that time and of their companies,9 some actors campaigned for the conclusion of BITs encompassing the consent to arbitration (or alternatively for the adoption of domestic statutes containing the same promise). The rise in the number of BITs was spectacular,10 while

4 Southern Pacific Properties v Arab Republic of Egypt, decision on jurisdiction (14 April 1988) published in English in (1991) 16 Yearbook of Commercial Arbitration 28 (excerpts); translated into French in (1994) 121 Journal du droit international 220 (excerpts). 5 On state contracts, see D Müller, ‘Ad hoc Investment Arbitration Based on State Contracts: From Lena Goldfields to the Libyan Oil Arbitrations’, ch 3 in this volume. 6 J Paulsson, ‘Arbitration without Privity’ (1995) 10(2) ICSID Review – Foreign Investment Law Journal 232–257. Paulsson acts as a counsel (see below) and also counts as one of the most influential arbitrators, preferentially nominated by claimants. See his position in the social network of arbitrators along with some of the arbitrators mentioned in that chapter in S Puig, ‘Social Capital in the Arbitration Market’ (2014) 25(2) European Journal of International Law 413. 7 In French, an equivalent would be ‘arbitration without contractual relationship’ according to J Werner, ‘The Trade Explosion and Some Likely Effects on International Arbitration’ (1997) 14 Journal of International Arbitration 5. 8 A Prujiner, ‘L’arbitrage unilatéral: un coucou dans le nid de l’arbitrage conventionnel ?’ (2005) 1 Revue de l’arbitrage 63–99. With a tribute to the thesis developed by W Ben Hamida, L’arbitrage transnational unilatéral, Réflexions sur une procedure réservée à l’initiative d’une personne privée contre une personne publique (PhD thesis, University Pantheon-Assas (Paris, 2003) (unpublished). A Prujiner preferred to drop the adjective ‘transnational’. 9 T Saint-John, The Rise of Investor-State Arbitration. Politics, Law and Unintended Consequences (Oxford, Oxford University Press, 2018) ch 6. 10 ‘By the end of 2012, the IIA regime consisted of 3,196 agreements, which included 2,857 BITs and 339 “other IIAs” (…)’, conclusions of new BITs reaching a peak in the years 1994 to 1997 and 2001 (UNCTAD, World Investment Report, 2013).

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

81

the number of states enacting investment statutes increased up to 108.11 The conjunction of the case law and this trend resulted in the changing nature of BITs – not only umbrella agreements but first and foremost instruments affording full procedural and substantive protections of their own – and the prevalence of BIT-based arbitration since then, cases based on domestic statutes representing annually about 10 per cent of all ICSID cases.12 Abundantly quoted and referred to in doctrinal works as milestone decisions, SPP and AAPL enjoy unique importance in diverging political narratives on international investment law. They represent the latest revolutionary steps for supporters of FDI and counterrevolutionary ones for their opponents. For proponents of ISDS, SPP and AAPL have rightly achieved the aim of emancipating investors from the protection of their state of origin and freeing them from the necessity to negotiate an agreement on the arbitral settlement of disputes arising/arisen or likely to arise with the host state. This pattern accomplished an evolution starting at the beginning of the twentieth century with the replacement of gunboat policy through arbitration, the internationalisation of host states’ consent to arbitration, as well as of substantive norms for the protection of foreign investment and the depoliticization of investorstate disputes now subjected to predictable procedures of settlement and applicable rules at all times (in peacetime and not only post-conflict).13 For moderate critics and detractors of ISDS, the tribunals unbalanced the arbitral settlement of disputes and bestowed upon foreign investors undue privileges. In more political terms, SPP and AAPL would have concretised the potentialities of the Washington Convention and marked the beginning of a counter-revolution, after the New International Economic Order failed and the Washington consensus won.14 Only one point draws consensus: The general economy of international investment law has to be considered to properly assess changes that occurred in the 1990s. This will also be the perspective presented in this chapter. SPP and AAPL have been routinely described as precedents (section I) and all along the ideological spectrum as revolutionary ones in terms of consent to ‘arbitration’ essentially (section II). This perspective has contributed to obfuscate no less important issues dealt with at length in the awards, namely the creation of rights for investors by treaty and the interpretation of bilateral investment treaties, with early commentators focusing instead on the determination of the applicable law (section III). Confronting what arbitrators really said on these issues and what they were said to have set in a changing political context helps in both re-evaluating the role of discrete or vocal actors and their proper input in the process of a ‘silent revolution’.15

11 J Hepburn, ‘Domestic Investment Statutes in International Law’ (2018) 112(4) American Journal of International Law 658–659. 12 For updated data, see Cases Registered by ICSID under the ICSID Convention and Additional Facility Rules, https://perma.cc/XF4C-DHEH. 13 C Leben, ‘Droit international des investissements: un survol historique’ in Droit international des investissements et de l’arbitrage international (Paris, Pedone, 2015) 1–74. See also E Stoppioni, ‘Landmark Decisions for a Pre-History of International Investment Law’, ch 2 in this volume. 14 J Linarelli, ME Salomon and M Sornarajah, The Misery of International Law (Oxford, Oxford University Press, 2018) esp 98 ff, 148 ff. Ironically, the Tribunal seized by SPP reminded Egypt that the price to pay for its understandable reluctance towards the once capitular system was an ‘open door policy’ surrounded with guarantees including ‘the promise of neutral or impartial dispute resolution, so as to dispel investors’ concerns about Egypt’s reputedly hostile attitude towards non-domestic arbitration (…)’. ‘Such considerations should explain why Egypt gave advance and general consent to the Centre’s jurisdiction not only in article 8 of Law n°43, but in an increasing number of bilateral treaties’ (SPP, para 107). 15 According to J Pauwelyn in Z Douglas and others (eds), The Foundation of International Investment Law: Bringing Theory into Practice (Oxford, Oxford University Press, 2014) 31.

82 Evelyne Lagrange I. ARBITRATORS (UNINTENTIONALLY) SETTING PRECEDENTS ON CONSENT

In the words of the Report of the Executive Directors of the World Bank on the ICSID Convention, consent is the cornerstone of the arbitral settlement of dispute. In line with this, the much-debated Article 25 thereof16 provides that ‘(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.’ The next sentence locks in that consent: ‘When the parties have given their consent, no party may withdraw its consent unilaterally.’ In other words, once consent is given by both parties, arbitration becomes compulsory. By contrast, this Article is mute on the forms of expression of the consent of both the investor and the host of the investment. Such a scrupulous and respected scholar as Paul Reuter read the ICSID Convention as permitting state consent to be derived from a treaty, not the investor’s consent from the institution of the proceedings for the sake of reciprocity.17 AAPL and SPP filled the blank most liberally. All commonplaces notwithstanding, it is worth re-examining how SPP and AAPL became landmark decisions or even precedents. According to the commonplace definition of a precedent, the term refers to a prior decision which judges, or arbitrators feel bound to follow so as to settle another dispute, drawing on relevant analogies between the cases. That prior decision may have been underpinned by the intent of the author to set a precedent and is best suited to inaugurate a new trend if the ratio decidendi is clear and persuasive. There is nothing of the like in AAPL. It is undisputed that it was the ‘first time’ (to borrow from the very terms of the award) an investor dared to invoke a BIT (concluded on 13 February 1980 between the UK and Sri Lanka) as the sole basis of the jurisdiction of the ICSID tribunal in the absence of any contractual link. A closer look at the award reveals that its drafting downplayed the reach of the innovation in terms of consent to arbitration. It is under the heading ‘Concerning the Applicable Law’ that the Tribunal famously wrote: ‘The present case is the first instance in which the Centre has been seized by an arbitration request exclusively based on a treaty provision and not in implementation of a freely negotiated arbitration agreement directly concluded between the Parties among whom the dispute has arisen’ (§18) and ambiguously hinted at ‘an arbitration case directly instituted in implementation of an international obligation undertaken between two States in favour of their respective nationals …’ (§19). In the preceding paragraphs, the award entails no developed motives on jurisdiction, no reference to earlier cases. It was little discussed in later cases,18 but the solution was reproduced in numerous cases. There is no indication either that the host state would have felt bound to bow to a permissive interpretation of Article 25 Washington Convention and the BIT, or that the arbitrators intended to set a precedent. If such had been the Tribunal’s intent, it could have said obiter that the attitude of Sri Lanka was in line with the proper meaning of the BIT combined

16 AR

Parra, The History of ICSID, 2nd edn (Oxford, Oxford University Press, 2017) chs 4 and 5. étrangers et arbitrage entre Etats et personnes privées. La Convention de la B.I.R.D. du 18 mars 1965 (Paris, Pedone, 1969) 14 ff. 18 See the awards AMT and Antoine Goetz below. In Lanco International Inc v Argentine Republic. ICSID Case No ARB/97/6, Preliminary Decision (8 December 1998) AAPL was mentioned as a first instance but the Tribunal referred more extensively to AMT as the leading case, while developing a more analytical reasoning (para 20 ff). In Ceskoslovenska Obchodni Banka, AS v The Slovak Republic, Decision on Objections to Jurisdiction (24 May 1999) ICSID Case No ARB/97/4, the Tribunal swiftly refers to AAPL but underlines that ‘(…) the protection of (binding) arbitration afforded by international arbitration’ is (…) the main objective of bilateral investment treaties (para 57). 17 Investissements

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

83

with the ICSID Convention. The Tribunal did not engage in any form of legal reasoning and it was only three years later that President El-Kosheri wrote in the ICSID Review that The case (…) set the first precedent for submission to ICSID jurisdiction directly and exclusively on the basis of a treaty text under which the private investor became the beneficiary capable of invoking the treaty without any need to negotiate even an investment contract with the competent authorities of the host State.

He insisted on Sri Lanka’s stance but foresaw the success of that venue in general terms: This ‘new technique opens unlimited opportunities for ICSID arbitration to become the “natural judge” in adjudicating investment disputes with the governmental authorities of a host state (…)’.19 Still, at the time of the proceedings, the Tribunal was essentially concerned with the proper manner to handle the absence of the prior choice of law referred to in Article 42 ICSID Convention: There were obvious reasons for the slight motivation on jurisdiction ratione voluntatis and the poor attention paid by commentators to that part of the award. First and foremost, Sri Lanka did not challenge in any way the jurisdiction of the tribunal which had thus prima facie no serious reason to scrutinise it proprio motu. Second, the context of a military intervention might have weighed on arbitrators. Third, it might be surmised that Sri Lanka’s counsel were not inclined to resist arbitration or advocate a restrictive interpretation of jurisdictional basis.20 Three of its counsel had eloquently prophetised before SPP was issued in 1988: When the day comes (…) that a foreign investor invokes as a basis for ICSID jurisdiction the unilateral promise made by a host state in its investment promotion legislation to submit certain categories of disputes to ICSID arbitration, a future ICSID tribunal may pose in an altered context the same question considered by past ICSID tribunals: did the parties reasonably envision ICSID arbitration. In resolving this question, the future tribunal would do well to consider two of the principles motivating the decisions analyzed in this article: (1) a party’s reasonable reliance on a promise of ICSID arbitration should not be frustrated; and (2) expansion of ICSID jurisdiction consistent with the reasonable expectations of parties should not be resisted as an affront to the sovereignty of the host state.21

Be that as it may, ellipses in the award should have been a good reason for not considering AAPL as a precedent in cases where the defendant challenges the jurisdiction of the Tribunal. What might have happened if the jurisdiction had been thoroughly assessed against applicable norms? The challenges regarding consent would have been twofold. Article 8(1) BIT unequivocally provides that contracting states ‘hereby’ consent to submit ‘any legal dispute’ (with investors of the other contracting state) to the ICSID; no additional expression of consent of the state was envisaged.22 The consent of Sri Lanka was crystal-clear and would have dictated the first part of that fictitious reasoning – unless the invokability of the BIT by the investor was put into doubt (see below). However, Article 25 of the Washington Convention requires the written consent of both parties to the dispute; Article 36(2) foresees that the

19 ICSID,

‘ICSID Arbitration and Developing Countries; (1993) ICSID Review 107. AAPL was represented by Heribert Golsong, a former legal counsel (1980–1982) and Vice-President of the World Bank (1984), Sri Lanka was represented by A Rohan Perera a consultant, diplomat and international lawyer rather open to international investment law, along with William Rand, Robert Hornick and Paul Friedland who had co-authored ‘ICSID Emerging Jurisprudence. The Scope of ICSID’s Jurisdiction’ (1986–87) 19 NYU Journal of International Law and Politics 33–61. 21 Eod. Loc., 61. 22 By contrast with other treaties – D Asiedu-Akrofi, ‘ICSID Arbitral Decision’ (1992) 86(2) American Journal of International Law 371. 20 While

84 Evelyne Lagrange request delivers information on the consent to arbitration while Article 8(3) BIT unequivocally envisions the consent given by the investor and the request addressed to the Secretary General as two distinct acts. In the case at hand, the ICSID Secretary General registered the arbitration request without any caveat or mention potentially casting doubt on jurisdictional issues (§3). This was in line with his attributions as generally understood (preventing flagrant misuse) but did not per se preclude the Tribunal from re-examining jurisdictional issues. The attempt of AAPL did not encounter any form of contestation on the part of Sri Lanka either. Thus, it remained little noticed that the investor had not consented to arbitration in due form prior to the institution of proceedings. The Tribunal could – or should (although it will not be discussed here) – have deemed it necessary to ascertain the scrupulous observation of the black-letter of the aforementioned provisions. After having clarified that both the 1980 BIT and the ICSID Convention require a formal act of consent on the part of the investor in advance, the Tribunal could still have affirmed its jurisdiction by reference to the forum prorogatum technique.23 However, an explicit reference to forum prorogatum had been dropped from the draft convention upon request of experts from developing countries fearing pressures on the host state once the claim was lodged.24 Alternatively, it could have relied on Article 8(3) 1980 BIT leaving it to the company affected to choose in the event of a disagreement on the suitable proceedings. This right of option would possibly have supported the merging of consent and request in accordance with the BIT. Eventually, the Tribunal refrained from any principled assumption, as well as from any reference to SPP, although that case, which dealt mainly with consent by statute also hinted, for the sake of objections handling, at consent by treaties and at an investor’s ‘right to dispute settlement’ ‘conferred by the bilateral treaty’ (§90).25 The fact is that in SPP, Egypt had raised jurisdictional objections on which the Tribunal decided in two decisions.26 SPP claimed that Egypt delivered its consent to ICSID arbitration through Article 8 law n°43, to be construed according to the rules of the VCLT (1969). Egypt responded that Article 2 of that domestic statute required an additional expression of consent in the form of a compromise. Article 8 (English version) was all but crystal-clear.27 The Tribunal reshaped the contention in the latter decision:28 ‘The issue is whether certain unilaterally enacted legislation has created an international obligation under a multilateral treaty’ (§61) or to put it differently, whether it made an ad hoc agreement to arbitrate superfluous or was just declaratory of the willingness of Egypt to recognise ICSID jurisdiction for certain litigations. A substantial part of the decision revolved around the construction of a domestic statute which provision on consent was said to be inadequately translated into English. Arbitrator Al-Mahdi swiftly rebutted the Egyptian objection drawn from the past capitular system (see above) and questioned the arguments and methods of interpretation of the Tribunal.

23 C

Schreuer et al, The ICSID Convention. A Commentary (Cambridge, Cambridge University Press, 2001) 346. above (n 16) 53, 64. 25 The arbitrators were probably cognizant of SPP despite the late publication. 26 The first (27 November 1985) stayed the proceedings until French Courts had resolved ‘the question of whether the parties agreed to submit their dispute to the jurisdiction of the International Chamber of Commerce’. The second one was issued on 14 April 1988. 27 ‘Investment disputes in respect of the implementation of the provisions of this Law shall be settled in a manner to be agreed upon with the investor, or within the framework of the agreements in force between the Arab Republic of Egypt and the investor’s home country, or within the framework of the Convention for the Settlement of Investment Disputes between the State and the nationals of other countries to which Egypt has adhered by virtue of law n°90 of 1971, where it (i.e. the Convention) applies’. 28 The affirmation of the possibility for states to express their consent in a legislative form had already been sketched out in the first decision on jurisdiction (para 50 ff). 24 Parra,

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

85

He took the pain to demonstrate at length that the original version could not per se or compared with the Washington Convention be construed as containing the required consent to the jurisdiction of the Centre in its arbitral capacity and warned that the Centre might ‘become the Juge de droit commun of investment disputes on the assumption of a consent which does not exist in fact or in law’ (§8). Consequently, investment statutes should be construed according to the applicable domestic rules (§12) (see below). The majority took a different stance, underlining that ‘(…) Article 8 is alleged to be a unilateral declaration of acceptance of the Centre’s jurisdiction, subject to reciprocal acceptance by a national of another Contracting State (…)’ (§61), functionally comparable with unilateral declarations of acceptation of the jurisdiction of the ICJ. Having insisted on the required ‘consent in writing’ of both parties, the majority focused on the consent given by Egypt through a domestic statute. It came as no surprise to lawyers familiar with the ICSID Convention that the Tribunal drew on the Report of the Executive Directors accompanying the Convention: (…) a host State might in its investment promotion legislation offer to submit disputes arising out of certain classes of investments to the jurisdiction of the Centre, and the investor might give his consent by accepting the offer in writing.29

The tribunal not only recalled what all States ratifying the Washington Convention were, or should have been, aware of, it also hinted at the potential practice of consenting to arbitration through a BIT (§103). Eventually, the Tribunal found that the Egyptian law constituted an express ‘consent in writing’, ‘where there is no other agreed-upon method of dispute settlement and no applicable bilateral treaty’ (§116). Without prejudice to the specific wording of each domestic law and bilateral treaty, one virtue of the decision is to clearly single out the two ways of expressing state consent according to Article 25 ICSID Convention: through an (ad hoc) agreement or in advance and regardless of the identity of the investor (through law or treaty). In so doing, the Tribunal rightly asserted the possibility of a non-contractual basis of jurisdiction. Despite a long motivation and the aura of the President of the Tribunal, the authority of SPP might in the very short run have been weakened by ambiguities, the dissent of Al-Mahdi and the attempt of Egypt to have the decision annulled.30 Nonetheless, SPP and AAPL were soon put on the same footing. A few years later, tribunals referred to SPP as the precedentsetting award regarding consent by statute.31 Arbitration consent clauses in foreign investment statutes were generally deemed to entail unilateral commitments, modelled after the optional clause declarations (Article 36 ICJ Statute), despite their addressing private persons, not peers. This is evidenced in Tidewater32 or CEMEX.33 Some divergences persisted as to the due methods of interpretation of domestic statutes, as well as a question crucial for both claimants and respondents: Whether consent by statute should be interpreted restrictively or expansively

29 ICSID,

Documents Concerning the Origin and the Formulation of the Convention, vol 1, part 2, 1069)’ (§70). ICSID Secretary General took the initiative to dismiss the claim on the ground of the distinction between a decision and an award (Art 52 Washington Convention). Eventually, Egypt changed the terms of its investment statute so as to require specific additional agreement for ICSID arbitration. 31 Tradex Hellas SA v Albania (24 December 1996) ICSID Case No ARB/94/2, ICSID Review (1999) 186–187. For further references, see Schreuer, above (n 23) 197ff. 32 Tidewater Inc et al v The Bolivarian Republic of Venezuela, Decision on jurisdiction (8 February 2013) ICSID Case No ARB/10/5, para 79ff. 33 Cemex Caracas Investment BV and Cemex Caracas Investment II BV v Bolivarian Republic of Venezuela, Decision on jurisdiction (30 December 2010) ICSID Case no ARB/08/15. 30 The

86 Evelyne Lagrange or neither, but ‘objectively’ as the Tribunal in SPP assumed (§63)? Some tribunals turned to SPP as the relevant precedent for that purpose,34 some turned to domestic law controlled by international law while others pointed out that the majority in SPP (§61) had defined a ‘mix’ of international rules of interpretation of such unilateral acts but refrained from clarifying the part that each should play in the process. The CEMEX tribunal hence took on a clear-cut position in favour of international law: Unilateral acts by which a State consents to ICSID jurisdiction are standing offers made by a sovereign State to foreign investors under the ICSID Convention. (…) whatever may be their form, they must be interpreted according to the ICSID Convention and to the principles of international law governing unilateral declarations of States (§79).

Purposely or not, both tribunals in SPP and AAPL paved the way for arbitration without mutual consent: states unilaterally commit themselves through law or treaty to bring disputes with a growing class of investors to arbitration; investors unilaterally institute such arbitration. The prior investor’s consent fades away unless it is expressly required in the form of a contractual or unilateral instrument. That consent can be dissociated is after all in the nature of state consent by treaty; that the meeting of consents be discovered in the institution of the proceedings rests by contrast on a ‘slightly artificial reconstruction’.35 It was not until the dispute between American Manufacturing and Trading Inc (AMT) and Zaïre,36 that the arbitral case law expounded some reasons for leaving it to the investor to opt for arbitration in advance or through the institution of the proceedings. Still, instead of building on the creation of substantive and procedural investor’s rights by treaty as suggested in prior case law, the Tribunal asked whether the US could ‘impose upon its nationals the passage of consent to ICSID’ (§5.18). This was a most oblique manner to ascertain the consent of both parties in a litigation instituted by the investor. The answer was in the negative, which is more than debatable after all in view of the history of international adjudication.37 But there, it resulted in particular from the BIT provisions that ‘the parties’ ‘should remain masters of the procedure of their choice which they deem appropriate to apply (…)’. Fortunately, the Tribunal found out that the BIT either required an agreement between the parties or, absent such an agreement, provided that ‘the procedure desired by the national or company concerned (should) be followed’ (§5.22). ‘This was very much the case (…)’ (§5.23). The unspoken (and now fully admitted) conclusion of such a reasoning is that the foreign investor is the master of the procedure – unless the BIT otherwise provides. Another award, Antoine Goetz c. Burundi,38 put to the fore more convincing rationales for the dissociated consent to arbitration, with due references to the precedents AAPL, SPP and AMT. The intent of the drafters of the Washington Convention was essential in the reasoning. At §67, the Tribunal didactically wrote: En ce qui concerne la compétence du CIRDI – et, par voie de conséquence, celle du Tribunal arbitral –, elle trouve sa source (…) directement dans le traité international (…). L’exigence fondamentale du consentement des parties, formulée à l’article 25 de la Convention du CIRDI, se trouve ainsi revêtir une expression originale, à laquelle on ne pensait guère au moment de l’élaboration de la

34 PNG Sustainable Development Program Ltd v The Independent State of Papua New Guinea (28 April 2015) ICSID Case No ARB/13/33, para 250ff. 35 A de Nanteuil, Droit international de l’investissement (Paris, Pedone, 2017) 262, 537. 36 AMT v Republic of Zaïre (21 February 1997) ICSID Case No ARB/93/1. 37 See C Santulli, Droit du contentieux international, 2nd edn (Paris, Lextenso, 2015) 201ff. 38 Antoine Goetz et consorts v Repuclic of Burundi (10 February 1999) CIRDI Case No ARB/95/3 in French.

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

87

Convention du CIRDI, à une époque où la pratique des traités bilatéraux d’investissement était encore embryonnaire.

A handful of paragraphs later, the award quotes (in French) paragraph 24 of the Report of the Executive directors: Consent may be given, for example, in a clause included in an investment agreement, providing for the submission to the Centre of future disputes arising out of that agreement, or in a compromise regarding a dispute which has already arisen. Nor does the Convention require that the consent of both parties be expressed in a single instrument (§80).

Still, this excerpt neither excludes nor includes consent given by the state by treaty. ‘Investment agreement’ is not synonymous with investment treaties in that document (see the use of the expression at §10 of the Report), but rather with contracts between the investor and the host state (as the Tribunal rightly suggested §67).39 Thus, paragraph 24 cannot as such support the trend initiated by AAPL and AMT (mentioned at §81) with regards to state consent. Concerned more specifically with investors’ consent, the Tribunal considered that it resulted from the request for arbitration (§81). The Tribunal conveniently omitted to precisely quote the beginning of paragraph 24 of the Report: ‘Consent of the parties must exist when the Centre is seized (Articles 28(3) and 36(3)) but the Convention does not otherwise specify the time at which consent should be given’, the next sentences dealing with the consent of the state. The retranscription of the Tribunal is much more permissive: ‘(…) il faut en plus que les parties au différend (…) aient consenti à la compétence du Centre. La forme écrite mise à part, l’expression du consentement n’est pas soumise à des règles strictes’ (§80). In the end, the Tribunal precisely reiterated that a single act was not required (§80). Such a ‘slightly artificial’ reconstruction of the Report was useful to elevate AAPL and AMT to the rank of precedents and assert jurisdiction in Goetz. In SPP, although the majority referred once to reciprocal consent, it audaciously cut through the consensual basis of arbitral proceedings. The fact that SPP had as early as August 1983 consented to ICSID arbitration according to the terms of law n°43 – as it understood it – was mentioned in the summary of the proceedings (§50). Instead of expanding thereupon, the majority insisted on the right to arbitration created by law n°43 (§§112–114) and retained the right of the party instituting an arbitration proceeding to opt for conciliation or arbitration in terms conflating request, preference for, and consent of the investor to arbitration (§§102–103), all this without paying much attention to the wording of the ICSID Convention itself (as Arbitrator Al-Mahdi pointed out (second part, 4)). In that process, the majority acknowledged across the lines that BITs may confer to investors a right to (arbitral) dispute settlement (§90) and constitute the yardstick for the proper understanding of ISDS (§§102–103). SPP was hence not unduly celebrated as ‘the tipping point’ by J Paulsson who had gained with this company his ‘first ever client’.40

II. COMMENTATORS (ALSO KEY PLAYERS) PIONEERING A REVOLUTION IN ARBITRATION?

Some precedents are revolutionary, others are not. A legal revolution is synonymous either with a change of paradigm or a contra legem decision. Commentators, sometimes acting as 39 See

also Schreuer, above (n 23) 382–387 and 393. 7: The Tipping Point, in Kinnear et al (eds) Building International Investment Law (Kluwer 2015) 85–96.

40 Chapter

88 Evelyne Lagrange arbitrators or embedded lawyers in changing roles, appropriated the aforesaid approach to consent – before and after the late 1980s. Aron Broches played a decisive role, as the World Bank general counsel (1959–1979), as a Vice-President of the World Bank (1972–1979), as the Secretary General of the ICSID (1967–1980), but also as an ‘authorised’, prolific and constantly quoted commentator of the Washington Convention. As early as 1972, he extended the forms of expression of state consent envisaged in 1965 to consent by treaty and contended that the expression of request by the investor might take place through the institution of the proceedings.41 His views triumphed with AAPL and SPP, with the help of influential members of the doctrine who showcased the revolutionary potential of AAPL and SPP in the mid-1990s without much testing of the technical consistency of the reasonings against rules of interpretation of treaties and preparatory works:42 they conceptualised a revolution they indeed became actors of. The one who decisively encouraged practitioners and scholars to consider and approve the potential of both cases was Jan Paulsson. His seminal paper ‘Arbitration without privity’ was seemingly inspired by two treaties going far beyond the bilateral instruments then in force with their array of procedural provisions, namely the NAFTA (1993, Chapter 11, Section B) and the International Energy Charter (1994, Article 26): ‘(…) they may presage a new era of confidence in the drafting, interpretation, and application not only of other such treaties, but also of national laws and bilateral investment treaties (BITs)’.43 No later than 1995, the man who had been a counsel to SPP (along with Broches) proclaimed with an unusually sharp tone the end of a ‘mindset crystallised in the 1970s’ centred on an alleged refusal to see states account for their actions before international tribunals. The harsh critics against Mr Sornarajah’s publications built on the drafting of the latest generation of IPPAs, without much attention paid to the proper interpretation of the Washington Convention.44 Simultaneously, Paulsson delivered his vision of ‘a new world of arbitration’ ‘where the claimant need not have a contractual relationship with the defendant and where the tables could not be turned: the defendant could not have initiated the arbitration, nor is it certain of being able even to bring a counterclaim’.45 All this was seen as a good omen for the progress of the rule of law. In a no less visionary article published the same year, Geneviève Burdeau explained the break with traditional commercial arbitration by interrelated factors: The spectacular increase in the number of States Parties to the ICSID Convention, linked to the quest for World Bank loans; cases like AAPL paving the way for the implementation of a growing number of BITs not explicitly requiring the joint consent of the host state and the foreign investor; the NAFTA. She envisaged the ultimate consequences of these trends more fairly than Paulsson did. First, the state’s consent moves away from the model of the unilateral declaration of acceptation of ICJ jurisdiction since reciprocity is not anymore requested, states expressing generic offers of arbitration and remaining generally unaware of who might seize this opportunity until a request is filed. Second, consenting states create a new venue at the unilateral disposal of foreign investors for the assessment of their own behaviour, against conventional obligations enshrined in the

41 The Convention on the Settlement of Investment Disputes between States and Nationals of Other States (1972) 136 Collected Courses of The Hague Academy 351. 42 Discretely and too late to change the course of the case law, some authors and practitioners expressed doubts: Prujiner, above (n 8); G Guillaume, ‘The Use of Precedent by International Judges and Arbitrators’ (2011) 2(1) Journal of International Dispute Settlement 17. 43 Paulsson, above (n 6) 233 – the correctness of that view will not be debated here. 44 240–241, targeting M Sornarajah, The International Law on Foreign Investment (Cambridge, Cambridge University Press, 1994). 45 Paulsson, above (n 6) 232.

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

89

BITs in general or loose terms. In a nutshell, the reference model is not anymore (commercial) arbitration but the unilateral right of action of individuals before the ECHR.46 Contrasted with the enduring deference towards the official lexic stemming from arbitration, that qualification was rather bold. Both publications contributed to single out SPP and AAPL as precedents and to popularise their revolutionary potential. Further treaty-based requests for compensation were brought from 1997 onwards,47 while the first awards rendered on the basis of NAFTA fuelled hopes for investors and fears about the implementation of regulatory measures taken in the general interest.48 Brigitte Stern, a scholar and the would-be most designated arbitrator ever in investor-state proceedings (predominantly by respondents), helped identify how far the revolution went or could go, were tribunals to depart from the black-letter of applicable instruments. While acknowledging that the solutions retained during 1988–1990 were ‘innovative’ but ‘plausible’ in view of the text of the Washington Convention and accompanying Report (see above), she targeted the application of MFN clauses to consent to ICSID jurisdiction in Maffezzini v Spain49 and expressed doubts on the ‘increasingly remote consent of States’ that left the door largely opened ‘for challenges to perfectly admissible acts of sovereignty’.50 In a dissent joined to ABCI Investment N.V. c. République tunisienne,51 she approved of the solution retained in AMT and considered it transposable to state consent delivered through a statute. Investors’ consent should be certain and explicit, not mechanically derived from the request for approval of their investment (denunciation of the ‘increasingly remote consent of the investor’) and preferentially expressed prior to any proceedings since statutes conveying state consent can be more easily amended or terminated than treaties. All in all, Stern invited tribunals to draw all consequences deriving from the persistent arbitral design in the ICSID Convention in terms of jurisdiction. If the interpretive choices made at the turn of the 1990s were only ‘plausible’, AAPL and SPP were no more than a pair of precedents dissipating doubts on the intent of states consenting to the jurisdiction of ICSID tribunals. The seeds of revolution would then have lain in the unreasonable use arbitrators made of their unprecedented powers and appeals in the following years. Still, the story was not that simple and linear. Under the benefit of a retrospective overview, it appears that arbitration tribunals and enthusiastic commentators retained an interpretation of the Washington Convention which was closer to the hidden intent of the drafters than to the intent of all states involved in its preparation. Reuter had noticed that the

46 G Burdeau, ‘Nouvelles perspectives pour l’arbitrage dans le contentieux économique intéressant les Etats’ (1995) 1 Revue de l’arbitrage, 14 ff. On his part, T Wälde suggested that states had once more found a way of ‘using private agents’ to trigger – at their own initiative and costs – the review of treaty compliance in the ultimate interest of States Parties to IPPA (‘Investment arbitration under the Energy Charter – From Dispute Settlement to Treaty Implementation’ (1996) 12(4) Arbitration International 437). 47 E Gaillard, ‘L’arbitrage sur le fondement des traités de protection des investissements’ (2003) 3 Revue de l’arbitrage 856. 48 See R Howse and G Ünüvar, ‘Sowing the Seeds of an ISDS Legitimacy Crisis? The Notorious First Wave of NAFTA Chapter 11 Awards’, ch 7 in this volume. 49 Emilio Agustin Mafezzini v The Kingdom of Spain, Decision of the Tribunal on objections to Jurisdictions (25 January 2000) ICSID Case No ARB/97/7. 50 B Stern, ‘ICSID Arbitration and the State’s Increasingly Remote Consent: Apropos the Maffezzini Case’ in S Charnovitz (ed), Law in the Service of Human Dignity. Essays in Honour of Florentino Feliciano (Cambridge, Cambridge University Press, 2005) 249, 250, 259. 51 ABCI Investmensts NV v Republic of Tunisia, Decision on jurisdiction (18 February2011) ICSID Case No ARB/04/12.

90 Evelyne Lagrange Convention was purposely silent on many issues, hence leaving it to the interpreters or interested parties to clarify them.52 This is exactly what investigations into the preparatory works, initiatives undertaken by the WB and ICSID top civil servants, reveal. Stern, again, highlighted that from the outset, Broches and his followers kept in mind that ICSID tribunals might have jurisdiction even in the absence of any state commitment specific to one investor, but intentionally played down that possibility so as not to frighten away capital importing countries.53 The survey of Saint-John on the role of some actors with institutional and/or personal stakes in the success of the Washington Convention confirms the leading role of technocrats, not diplomats, in the negotiation process. Broches himself made no mystery of the fact that it had been a World Bank-driven process and that the Report of the 20 Executive Directors fell short of reflecting the diversity of state stances and concerns with the Convention. This did not prevent him from enriching its black-letter content with the very views he had cunningly concealed from reluctant states.54 His ubiquity enabled him to disseminate his own views and forge the orthodox construction of the Washington Convention. Turning to the BITs now, it appears that ‘the initial push for investor-State arbitration clauses came from the ICSID Secretariat and not investors’.55 The ICSID Secretariat (then led by Broches) took the initiative to promote model clauses including investor-state arbitration as early as 1969 and followed up in assisting negotiations between states. Some misunderstandings on the reach and effects of state consent probably existed before SPP and AAPL. Thereafter, many IPPAs were interpreted or drafted in the 1990s-2000s but few entailed caveats correcting the asymmetries introduced by SPP, AAPL and the subsequent arbitral practice. Is this phenomenon per se evidence of the general acceptation by all states of the scheme of ‘arbitration without privity’? From a purely formalistic standpoint, states should have known what they were consenting to through the combination of the Washington Convention and a BIT without requirement of an agreement on arbitration. In fact, there are some indications that some states were persuaded by western diplomats and experts of the WB to ratify such treaties in the perfect nescience of the reach of their provisions. In some cases, the sense for these commitments got lost. Field researches led by Mavluda Sattorova reveal similar stances of decision-makers or civil servants in Uzbekistan, Nigeria, Turkey, Ukraine, Kazakhstan.56 Lack of expertise – then and now – may explain why states emerging from decolonisation or from the collapse of the Socialist Block did not question from the outset or reconsider afterwards the extensive interpretation of the ICSID Convention and/or of the terms of treaties sticking to BITs models unilaterally developed by their economic partners. However, knowledge of ISDS pitfalls and drawbacks gained traction. The OECD negotiations on a Multilateral Agreement on Investment were abandoned in April 1998: their draft provisions on ISDS, in line with antecedent developments to the benefit of investors, unleashed passionate debates in the global public opinion. Nevertheless, only a handful of states defied the dynamics of IPPAs providing for unilateral arbitration, either generally or due to the mutual trust treaty parties could have

52 Above

(n 17). Stern, ‘Le consentement à l’arbitrage CIRDI en matière d’investissement international: que disent les travaux préparatoires?’ in Mélanges en l’honneur de Philippe Kahn. Souveraineté étatique et marchés internationaux à la fin du 20e siècle (Paris, Litec, 2000) 226–233. 54 See his Hague Lectures, ibid, passim. 55 Saint-John, above (n 9), Introduction and Part I. 56 M Sattorova, The Impact of Investment Treaty Law on Host States. Enabling Good Governance (Oxford, Hart, 2018) 61 ff. See also the mindset of Albanian leaders who enthusiastically concluded BITs to attract a flow of foreign investors and belatedly discovered their scope with Tradex’s claim, as described by P Sands, Lawless World. America and the Making and Breaking of Global Rules (London, Allen Lane, 2005) 117 ff. 53 B

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

91

in their respective judicial system. All in all, both the legitimacy and the legal soundness of unilateral arbitration were upheld by a large majority of states. It is now possible to assess more precisely the breadth of the ‘revolutionary steps’. With regards to the expression of state consent, SPP followed the construction of the Washington Convention envisioned as early as the preparatory works whereas AAPL concretised in hindsight the discrete intent of innovative and persistent technocrats at the World Bank and the Centre. In other words, AAPL and SPP were part of a creeping revolution started during the negotiations on the ICSID Convention, actively supported by some or the same stakeholders in the 1990s and then ideologised. AAPL and SPP not only initiated an obvious break with commercial arbitration, but were disruptive of the very notion of arbitration. In this applicant-driven procedure, the core notion of mutual consent begins to disappear. Practically, the good news – notably for its agents and potential arbitrators or counsels – was that the ICSID was not doomed to stay in limbo (25 cases in 25 years was a poor record). The flip side was that sight was lost of what Paulsson – ironically – called for in his seminal paper: ‘The objective is not arbitration that favors the foreigner, but one that simply favors neutrality’.57 Should investor-state arbitration have been taken seriously by tribunals, then the consent of the investor could or should have been envisaged differently. Concretely, a prior binding statement of the investor could have been required as an alternative to consent through state contract or arbitration agreement. This unilateral commitment would have encountered the arbitration pledge resulting from a domestic statute or a treaty producing unilateral effects towards investors of the nationality of the counterparts. This would have been no less ‘plausibly’ compatible with the Washington Convention and many BITs than the solutions under survey. The protection of the investor against a unilateral withdrawal of state consent would have remained unaltered (or improved) but investors would undoubtedly have been subjected too to the risk of a procedure or a counter-claim introduced by the host state, as envisioned in Article 46 of the Washington Convention.58 It took decades until states explicitly recovered the possibility to counter-claim, derived from the ‘dual possibility to initiate an arbitration’ identified in a BIT not fundamentally different from the 1980 UK-Sri Lanka treaty.59 Doctrinally and politically now, the nature and legitimacy of the further use of the notion of arbitration to properly account for treaty – and statute-based arbitration could have been more resolutely questioned. Case law soared so rapidly that the doctrine, concerned with accompanying the movement, did not shy away from using contradictory, if not oxymoronic qualifications, borrowing from all available legal templates and clashing paradigms:60 offer to arbitrate (despite the absence of procedural and substantive reciprocity), recognition of a right of action on behalf of the investor, unilateral acceptation of the jurisdiction of the arbitral tribunal similar to unilateral declaration of acceptation of the ICJ’s jurisdiction (a permanent court settling interstate disputes …), unilateral right of action manifesting the consent to the jurisdiction of the Tribunal (superfluous if a right of action is enjoyed). The offer-acceptance model is in line with the insistence on the ‘consent in writing’ and conducive to the prevalent attachment to arbitration as the proper paradigm to account for ISDS.

57 Paulsson,

above (n 6) 256 – italicised throughout the text. AK Björklund, ‘Counterclaims’, ch 17 in this volume. 59 Urbaser SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v The Argentine Republic, ICSID Case No ARB/07/26, para 1143 ff. 60 A Roberts, ‘Clash of Pradigms: Actors and Analogies Shaping the Investment Treaty System’ (2013) 107(1) American Journal of International Law 45–94. 58 See

92 Evelyne Lagrange However, it is just a matter of replacing ‘arbitration’ with ‘ISDS’ to shed a fresh light on the mechanism, be it based on state consent released through a treaty or a statute. Either the host state has made an offer to settle disputes arising with consenting investors (whose consent is given in advance or when the proceedings are initiated without contest): the mechanism can remain arbitral by nature; the host state may initiate proceedings once the investor has accepted the offer; counter-claims are not excluded. Or the treaty parties have provided for the settlement of disputes upon unilateral request (not consent) of investors under certain conditions of admissibility (jurisdiction ratione voluntatis being certain) and the mechanism resembles that of international courts settling disputes between states and natural or legal persons (the ECHR for instance). Then, state claims against investors and counter-claims are duly barred while the right to appoint arbitrators accorded to both parties to the dispute and the removal of the condition of exhaustion of local remedies now rest on shaky ground. The issue is then clearly that of a delegation of power to international adjudicative bodies. Facing unbalanced proceedings named ‘arbitration’ but severed from some essential proprieties attached to arbitration, laypersons kept improperly denunciating ‘private arbitration’ of investor-state disputes until well in the 2010s, for the purpose of dramatising the challenges to democracy and equality before the law. The violent backlash against this form of ‘arbitration’ led the EU to frame a new model of ISDS enshrined in the CETA (signed on 30 October 2016, provisionally and partially entered into force on 21 September 2017) and to tackle (again) the creation of a multilateral investment court (see Negotiating Directives of 1 March, 2018). Fulltime judges could more legitimately be tasked with the settlement of disputes involving major issues of state policies and the development of international law with a better sense for consistency and predictability. This would be a complete reversal of the view shared by the father of the ICSID Convention, Aron Broches, who believed the ‘political complications’ of appointing judges with fixed terms would prove an ‘insurmountable obstacle’ to the creation of an investment court.61 Alternatively, ISDS could be revised just to keep up with contemporary expectations in terms of conflicts of interest for instance. These options are discussed within the UNCITRAL. The clarification of the conceptual and political basis of ISDS is in progress with a view to harmonising the functions (notably settling compensation claims for damage arising out of the exercise of police powers) and the procedural features (privileging foreign investors) of this branch of international disputes.

III. THE SEEDS OF MODERATION: ARBITRATORS WARY OF INSULATING INVESTMENT LAW

The preparatory works of the ICSID Convention reveal that states expressed diverging views on the jurisdiction ratione materiae of ICSID tribunals. Western states supported the submission of all types of disputes, contractual or not, to arbitration while India, Israel, Brazil and Lebanon distanced themselves from a catch-all provision on jurisdiction (the would-be Article 25). They feared that these tribunals might transform into reviewers of national legislation intended to protect morality, health, national security, workers, etc.62 Officially, the compromise consisted of retaining the formula ‘any legal dispute’ (Article 25.1) while ensuring

61 T Saint-John, ‘Déjà vu? Investment Court Proposals from 1960 and Today’ (EJIL Talk!, 15 May 2018), https:// perma.cc/DYJ9-H7NP. 62 Stern, above (n 50) 237 ff.

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

93

that reluctant states retained the right to submit contractual disputes only (if any) to ICSID jurisdiction (Article 25.4).63 In the aftermath of AAPL, treaty claims burgeoned. Alternatively, or additionally, states confronted with the necessity to boost investment passed domestic investment statutes. SPP and AAPL both brought into the limelight the prevalence of international law over domestic law. In the 2010s, some historically exporting countries started to claim some caveats expressed by some Third World Countries. These concerns could have been mitigated if more attention had been paid to the careful construction of obligations incumbent on a state party to a BIT in AAPL. In the first place, commentators focused on the determination of applicable law. Article 42.1 ICSID Convention reads: 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.64

In cases where there is no contract nor formal arbitration agreement the Tribunal faces, it has to decide how to determine the applicable law. The dispute opposing AAPL to Sri Lanka was completely foreign to the implementation of a state contract and no choice-of-law had been formalised until the request was filed. The 1980 BIT entailed no choice-of-law clause either. Referring to the special circumstances of the case and Article 42 only, the majority in AAPL drew therefrom that since ‘(…) the Parties in dispute (had) had no opportunity to exercise their right to choose in advance the applicable law determining the rules governing the various aspects of their eventual disputes’, ‘(…) the choice-of-law process would normally materialise after the emergence of the dispute, by observing and construing the conduct of the Parties throughout the arbitration proceedings’ (§19). In the case at hand, the majority said, ‘(…) both Parties acted in a manner that demonstrat(ed) their mutual agreement to consider the provisions of the Sri Lanka/U.K. Bilateral Investment treaty as being the primary source of the applicable legal rules’ (§20) and to admit ‘(…) the supplementary role of the recourse – regarding certain issues – to general customary international law, other specific rules rendered applicable in implementation of most-favored-nation as well as to Sri Lankan domestic legal rules’ (§22). S.K.B. Asante expressed his dissent as to that reconstruction of an agreement on the applicable law: The BIT being intergovernmental by nature, the claimants had ‘(…) to demonstrate either that the treaty itself authorised this course of action, or that the parties to the dispute expressly agreed to regard the provisions of the Treaty as the applicable law’.65 In the absence of any agreement, the Tribunal should have turned to domestic law. Article 157 of the Constitution of Sri Lanka provides for the incorporation and superiority of treaties over contradicting domestic laws – save ‘in the interests of national security’. Consistently, the dissenting arbitrator heavily insisted on the misinterpretation of Article 4 BIT (related to compensation due for losses owing to war, armed conflict, revolution, riots, etc ‘no less favourable than that which the latter Contracting Party accords to its own nationals or companies of

63 Ceylan observed that the mere existence of a mechanism for ISDS would offer importing countries a false choice: Consenting to compulsory arbitration without any restriction as to the cause of action or discouraging foreign investors. Once renamed, the Democratic Socialist Republic of Sri Lanka concluded a BIT with the UK providing for the settlement of any legal dispute with nationals of the latter investing on its territory. 64 See F Latty and M Sim, ‘The Applicable Law Saga’, ch 18 in this volume. 65 ICSID Review (1991) 576.

94 Evelyne Lagrange any third State’, except when destruction occurred in combat actions or was required by the necessity of the situation) and arrived at the conclusion that the respondent was not liable.66 The majority might have shied away from these ultimate consequences. Be that as it may, the majority felt uneasy with the absence – by construction – of a choice-of-law agreed by the parties to the dispute. According to Y Banifatemi, (t)here is, however, no such difficulty since the arbitration agreement contained in an investment treaty is deemed to be stipulated by the contracting States for the benefit of their investors. Any agreed mechanism in the arbitration agreement, including the law applicable to the dispute, is therefore deemed to be chosen directly by the parties to the arbitration. This assumption is nothing more than the implementation of the dissociated nature of consent to arbitration in investment treaty arbitration (…).67 (emphasis added)

Is this assumption less fictitious than the ‘mutual agreement’ discovered by the majority? After all, the freedom of the parties to the dispute fades away when treaty parties decide on the applicable law (at least theoretically since many BITs remain silent on that issue). The relative boldness of the Tribunal was counterbalanced by a thorough assessment of the content of customary rules of international law. The nature of the dispute in SPP was not so clear-cut. The claimant sought compensation for breaches of contractual undertakings, but since the Heads of Agreement fell, according to their proper terms, within the scope of application of law n°43 (and others), breaches of the latter would result in breaches of the former. Was this to say that the Tribunal would be bound to apply Egyptian law only? According to the respondent, the answer was in the affirmative: Since there was an implicit agreement between the parties on the applicability of domestic law, international rules formally incorporated in Egyptian law only should be taken into consideration. Conversely, the claimant alleged that absent any agreement, the dispute should be settled by application of general principles of international law. The Tribunal was of the ambivalent opinion that ‘even accepting the Respondent’s view that the Parties have implicitly agreed to apply Egyptian law, such an agreement cannot entirely exclude the direct applicability of international law in certain situation’, especially in case of lacunae or domestic acts contradicting international law.68 Acts of Egyptian officials might have been null and void or inexistent, but ‘cloaked with the mantle of Governmental authority and communicated as such to foreign investors who relied on them in making their investment’, they ‘created expectations protected by established principles of international law’ (§§82–83). In other words, diverse substantive provisions of domestic instruments could also be seen as unilateral acts of the respondent governed by international law, whose addressees are foreign investors only. However, it remains uncertain whether arbitral tribunals applying domestic statutes are all minded to give effect to domestic or international instruments.69 Within a handful of years, states realised that they might face a triptych of claims – classical contract-claims but also treaty and domestic law claims. Certainly, the applicable law derives from choice-of-law provisions or Article 42.1 ICSID Convention, which is distinctive of arbitration, but ‘in the end, international law is controlling’.70 66 His

reasoning gained some support: see E Gaillard, (1992) Journal de Droit International 230–231. Law Applicable in Investment Treaty Arbitration’ in K Yannaca-Small (ed), Arbitration Under International Investment Agreements: A Guide to the Key Issues (Oxford, Oxford University Press, 2010) 193–194. 68 Award on the merits (20 May 1992) paras 80, 84. 69 Hepburn, above (n 11) chs 17 and 22. 70 Santa Elena SA v Costa Rica, Award (17 February2000) ICSID Case No ARB/96/1, para 65; R Rivier, ‘L’articulation entre droit national et droit international devant les tribunaux arbitraux internationaux d’investissement’ in S RobertCuendet (ed), Droit des investissements internationaux. Perspectives croisées (Bruylant, 2017) 413–483. 67 ‘The

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

95

It is self-evident, and recalled in SPP, that domestic statutes can vest private persons with rights. They can also impose obligations on them. Only the ‘internationalisation’ of their content was debatable (see above). By contrast, the issue of the direct applicability of BITs to foreign investors was crystal-clear in 1990 but became unduly intricate in the following years. The solution retained in AAPL as to the applicable law was congruent with the conviction of the majority, that such a BIT could create substantive rights for foreign investors even if neither treaty provision, nor agreement foresaw it expressly – in sharp contrast with Asante’s assertions stressing that the investor is a third-party to the treaty (at 576). Upon closer examination, articles of the BIT apparently differed, some dealing with the treatment to be reserved to foreign investments, others making explicit the rights of foreign investors. Article 2 eloquently defines the object and purpose of the BIT in terms of favourable conditions, fair and equitable treatment, full protection and security for nationals and companies of the other contracting party. The Tribunal retained from the outset that the arbitration case was ‘(…) directly instituted in implementation of an international obligation undertaken between two States in favour of their respective nationals investing within the territory of the other Contracting State’ (§19).71 Such a formulation could suggest that foreign investors are no more than ‘third party beneficiaries’ of IPPA. Consequently, and by analogy with third states benefitting from a treaty, the intended effects of the treaty would depend on the consent given by the third party – in an agreement on the choice of law or through its conduct. The analogy drawn between third states and investors is, formally speaking, more than perplexing. However, it is politically telling: it ‘(…) comports with the fact that the obligations of investment treaties are asymmetric; states owe obligations to investors, but investors have no corresponding obligations to states’.72 Alternatively, the formulations of the Tribunal could satisfy those who were reluctant to recognise that natural or legal persons, like foreign investors, are subjects of international law and lay the emphasis on the object of the treaty – the promotion and protection of investment. To explain how foreign investors could successfully base their claim before arbitral tribunals on such treaties, imaginative arbitrators and scholars drew up the highly fictitious figure of a foreign investor, himself deprived of international capacity, acting on behalf of his home state, despite the absence of any express mandate. This ‘derivative model’ or ‘delegated espousal model’ was for instance upheld in Loewen Group, Inc. & Raymond L. Loewen v USA:73 ‘claimants are permitted for convenience to enforce what are in origin the rights of Party states’ (§233).74 All these circumvolutions could have been avoided, if scholars and arbitrators had taken seriously the framework outlined by Paulsson himself75 or the proper wording of the AAPL award. The claimants had not only tried a procedural ‘coup’. They had also pleaded in favour of a substantive revolution: By abandoning the ‘diplomatic protection’ theory (…), the foreign investor ‘enjoys’ under the ‘Bilateral Investment Treaties’ (BIT’s) a different method of direct protection. According to the Claimant, ‘the right to protection is vested in the holder of the investment with immediate effect upon the simple

71 It is fair to notice that the use of the singular could also mean that the Tribunal had in mind not substantive obligations endorsed by the contracting states but the obligation to go to arbitration upon request of investors of the other contracting state. 72 A Björklund, ‘Private Rights and Public International Law: Why Competition Among International Economic Tribunal is Not Working’ (2007) 59(2) Hastings Law Journal 266. 73 Lowen group, Inc And Raymond L. Lowen v United States of America, Award (26 June 2003) ICSID Case No ARB(AF)/98/3. 74 For a moderate critic and plea for a ‘direct theory of rights under investment treaties’: Z Douglas, ‘The Hybrid Foundation of Investment Treaty Arbitration’ (2003) 74 British Yearbook of International Law 151. 75 Paulsson, above (n 6) 256, confronts the true defendant with the true complainant.

96 Evelyne Lagrange coming into force of the treaty’ (…). Thus, a deliberate choice is reflected to follow a new pattern in matters of protection different from that which prevailed under traditional International Law (§26).

In response, the majority took the stance that ‘(…) the Bilateral Investment Treaty is not a self-contained closed legal system limited to provide for substantive material rules of direct applicability (…)’ (§21); the Treaty creates ‘a legal regime’ ‘for the protection of (…) investors’ (§41). Discretely, the majority had paved the way for the full recognition of BITs as functional equivalents to domestic statutes (like the Egyptian statute at stake in SPP): Both can vest investors with substantive rights. Early commentators did not really highlight this contribution. AAPL brought a remarkable contribution to the interpretation of IPPAs which remained undervalued, probably due to the attention paid to the division of the Tribunal on the relevant rules of interpretation of the 1980 BIT and then, their concrete application. The dissenting arbitrator retained a single rule – generalia specialibus non derogant – while the majority enumerated no less than six distinct rules plus a synthetic rule of construction so as to determine whether the ‘Sri-Lanka-UK Treaty intended, merely, to consolidate the pre-existing rules of international law or, on the contrary, it tended to innovate by imposing on the host state a higher standard of international responsibility’ (§42). The claimant’s contention was that ‘the ordinary meaning of the words “full protection and security” (art. 2 of the BIT) points to the acceptance by the host State of strict or absolute liability’ (§26). Under the heading ‘applicable law’, the majority asserted that the ‘Bilateral Investment Treaty is not a self-contained close legal system limited to provide for substantive material rules of direct applicability, but it has to be envisaged within a wider juridical context (…)’ (§21). As such, this assertion should have inspired successors of the Tribunal and could have immediately elicited consideration for the applicability of general customary international law or other treaties binding host and home states with a view to protecting investors or other rights-holders. This could have been all the more so as, a few steps forward, the majority underscored that ‘recourse to the rules and principles of international law has to be considered a necessary factor providing guidance within the process of treaty interpretation (…)’, with an express reference to Article 31.3.C of the VCLT (§40), generous references to case law and to the ‘worldwide BIT network’ (§49). At first sight, the majority duly paved the way for enlarging the applicability of international law to foreign investment, instead of focusing on a BIT or international investment law only, either in the process of construction of the BIT at play or in the implementation of the applicable law. This implies thoroughly assessing the rules and principles of international law and assuming the risk of a conflict of norms, to be settled.76 Such an approach might strengthen the protection of investors (if effect is given to human rights norms for instance) or mitigate state obligations toward them (if human rights, environmental or health conventions or police powers exceptions are taken into consideration to assert the reasonableness of state measures, for instance). Cases like Urbaser S.A. v Argentina or Philip Morris v Uruguay are illustrative of this potential. Conversely, isolationism combined with teleological interpretation would benefit almost systematically foreign investors covered by an IPPA, possibly at the prejudice of other rights-holders protected by other rules of international law. It is fair to acknowledge that the reasoning in AAPL encouraged commentators to insist more on the limits to the attraction of ‘external norms’ in ISDS than on the reach of the aforementioned formulas in the abstract. These limits may result from the jurisdiction ratione materiae of the Tribunal, from the choice-of-law, or from the drafting of the BIT. Specifically, the majority characterised

76 P Jacob, ‘La place des normes externes dans le contentieux de l’investissement’ in Robert-Cuendet, above (n 70) 607–640.

SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?

97

the ‘wider juridical context’ as integrating ‘other sources’ (than the BIT) ‘through implied incorporation methods or by direct reference to certain supplementary rules, whether of international character or of domestic nature. Such extension of the applicable legal system is clearly reflected in Articles 3(1) and 4 of the Sri Lanka/UK Bilateral Investment Treaty’ (§21). Following up, the majority insisted that ‘both Parties agreed during their respective pleading to invoke primarily the Sri Lanka/U.K. Bilateral Investment Treaty as lex specialis, and to apply, within the limits required, the international or domestic legal relevant rules referred to as a supplementary source by virtue of Articles 3 and 4 of the Treaty itself ’ (§24).

IV. CONCLUSION

The point of origin of ISDS in application of international rights, accruing to investors under international law (general principles of international law, treaties …), dates back at least to the interwar years.77 Still, the combination of a right of action before an arbitral tribunal and a cause of action derived from a BIT made AAPL a milestone in the twentieth century legal revolution that extracted natural and legal private persons from the insecure status of pure domestic subjects of law, or incapable subjects of international law, let alone objects of international rules. This revolution concretised in the conclusion of thousands of IPPAs (not all providing for a right of action, not all being enforceable by domestic or international adjudicators, however), thus putting under stress the capacity of contracting states to enforce policy powers at a reasonable political and economic cost. Over the decades, an exclusive or excessive focus on the lex specialis character of procedural and substantive international investment law resulted in politically unjustifiable distortions in favour of a class of special interests. Some arbitral tribunals undertake to rebalance the relationship between states and foreign investors. Some states or regional organisations revise investment domestic statutes or IPPAs for the same purpose. In so doing, they do not annihilate the liberal revolution envisaged by some in the 1960s and amplified in the aftermath of SPP and AAPL. They just make it compatible again with the asymmetrical relationship that can but exist between a sovereign and investors under its jurisdiction. Protection, not overprotection is due to the latter.

77 See E Stoppioni, ‘Landmark Decisions for a Pre-History of International Investment Law’, ch 2 and D Müller, ‘Ad hoc Investment Arbitration Based on State Contracts: From Lena Goldfields to the Libyan Oil Arbitrations’, ch 3 in this volume.

7 Sowing the Seeds of an ISDS Legitimacy Crisis? The Notorious First Wave of NAFTA Chapter 11 Awards ROBERT HOWSE AND GÜNEŞ ÜNÜVAR*

I. INTRODUCTION AND OVERVIEW

T

HE EARLY CASES decided by arbitral tribunals under the Investment Chapter (Chapter 11) of the North American Free Trade Agreement (NAFTA) are among the original sources of the legitimacy crisis1 of Investor-State Dispute Settlement (ISDS). In several instances (though not all), tribunals found that regulatory measures were violations of either the national treatment (non-discrimination – Article 1102) or expropriation provisions of Chapter 11 (indirect expropriation – regulatory takings – Article 1110). Violations of fair and equitable treatment (‘FET’) (minimum standard of treatment – Article 1105) were also found. This chapter will examine six early NAFTA disputes, spanning the 1990s through the start of the new millennium: Ethyl v Canada,2 Pope & Talbot v Canada,3 Metalclad v Mexico,4 S.D. Myers v Canada,5 Feldman v Mexico,6 and Methanex v United States.7 In so doing, it will refer to parties’ relevant arguments and tribunal elaborations on Article 1105, though focus will be kept on national treatment and expropriation. On the facts, none of the results in these disputes unambiguously support the view that legitimate policy space is at risk from ISDS – even if some academic commentators have sought to present this case law as a basic threat to regulatory democracy.8 In Ethyl and S.D. Myers,

* Robert Howse is Professor of International Law at New York University Law School; Güneş Ünüvar is a Senior Research Fellow at the Max Planck Institute Luxembourg for Procedural Law. Howse discloses that he acted as consultant to the investor’s legal counsel in some of the disputes discussed in this chapter. 1 See J Stiglitz, ‘Regulating Multinational Corporations: Towards Principles of Cross-Border Legal Frameworks in a Globalized World Balancing Rights with Responsibilities’ (2008) 23 American University of International Law Review 451; R Howse, ‘International Investment Law and Arbitration: A Conceptual Framework’ in H Ruiz Fabri (ed), International Law and Litigation (Baden-Baden, Nomos, 2017). 2 Ethyl Corporation v Canada, UNCITRAL, Award on Jurisdiction, 24 June 1998. 3 Pope & Talbot Inc. v Government of Canada, UNCITRAL (2001). 4 Metalclad Corporation v The United Mexican States, ICSID Case No ARB(AF)/97/1. 5 S.D. Myers, Inc. v Government of Canada, UNCITRAL (2002). 6 Marvin Roy Feldman Karpa v United Mexican States, ICSID Case No ARB(AF)/99/1. 7 Methanex Corporation v United States of America, UNCITRAL (2005). 8 See G Van Harten and M Loughlin, ‘Investment Treaty Arbitration as a Species of Global Administrative Law’ (2006) 17(1) European Journal of International Law 121 and D Schneiderman, ‘NAFTA’s Takings Rule: American Constitutionalism Comes to Canada’ (1996) 46 University of Toronto Law Journal 499. To be clear, while we do not

100 Robert Howse and Güneş Ünüvar while health hazards were asserted as a reason for the measures, only international and/or interprovincial trade were banned, not the allegedly harmful substance or activity itself – a classic protectionist fact pattern. In Pope, the investor sought to impugn the way that Canada had implemented that decision, allocating trading rights among different companies within the country, rather than the underlying policy decision of the government to have a managed trade arrangement with the US. In Methanex, the investors’ claim failed where California’s ban on methyl tertiary-butyl ether (MTBE) was a well-justified health and environmental policy. Metalclad, where the Tribunal appeared to find a regulatory taking based entirely on the measure’s economic effects on the investor, nevertheless considered public policy in the FET analysis, finding that the measures in question were not reasonably related to environmental and health objectives. When one appreciates the nature of these claims, the resulting awards do not suggest a complete disregard for legitimate public policy by arbitrators. The awards instead expose arbitrators’ difficulty in applying treaty texts that do not contain explicit exceptions or limitations to protect legitimate public policies. The NAFTA contains carve outs for investment protection for certain kinds of policies and at the outset permitted State Parties to lodge reservations for policies they sought to protect. Beyond that, the NAFTA Investment chapter did not have a general exceptions clause like Article XX of the General Agreement on Tariffs and Trade (GATT). This is significant with respect to national treatment and indirect expropriation. In the case of FET, public policy reasons are usually considered in the assessment of whether measures are arbitrary or unreasonable. In the case of national treatment and indirect expropriation, it would have been open to the arbitrators to refuse to consider policy justifications, once it was found that the measure provided the claimant less favourable treatment than domestic investors in like competitive circumstances. In Metalclad, with respect to indirect expropriation, the Tribunal did not go beyond the consideration of economic effects – assuming its assessment of the policy and regulatory context under FET was such that no police powers argument needed to be entertained. The national treatment rulings are interesting precisely because of arbitrators’ attempts to develop jurisprudential constructs to assess public policy, in the absence of an exceptions provision like GATT Article XX. This leads to the consideration of public policy in examining whether a claimant and competing domestic enterprises are truly in like circumstances. The tribunals differ in determining whether the standard of review applicable to policy measures is reasonableness, proportionality, or necessity. The tribunals sometimes operate from a consciousness that they are inventing Article-XX substitutes, while at least one tribunal repudiated the GATT/World Trade Organisation (WTO) acquis as a starting point for understanding national treatment in NAFTA’s Investment chapter (Methanex). These early national treatment cases reflect ‘cross-judging’ – the influence of one international tribunal or dispute settlement regime on another, which generates imitation and differentiation.9 NAFTA tribunals have, on occasion, accepted any rational policy basis for a general regulatory scheme, precluding scrutiny of whether its detailed application embeds discrimination or protectionism (the Pope national treatment analysis). This demonstrates a flawed approach to cross-judging – borrowing from another regime without fully appreciating its features. In the case of Article XX of the GATT, the preambular provision requires, in

share these authors’ readings of the early NAFTA cases, it does not mean that we differ from them in their general critiques of ISDS. See Howse, above (n 1). 9 R Teitel and R Howse, ‘Cross-judging: tribunalization in a fragmented but interconnected global order’ (2008) 41 New York University Journal of International Law and Politics 959.

The Notorious First Wave of NAFTA Chapter 11 Awards 101 addition to a requisite nexus between the general measure and one of the stated legitimate policy objectives, that the detailed practical application of the scheme not embed unjustifiable or arbitrary discrimination, nor constitute a ‘disguised restriction.’ Even if the results of these early NAFTA cases are consistent with appropriate deference to legitimate public policy, critics had good reason not to be satisfied by these results. The extreme indeterminacy in the handling of public policy, the lack of a treaty-based justificatory structure and the oscillation between standards of review gave rise to a danger of ‘regulatory chill’, whereby uncertainty about how a given tribunal might evaluate public policy justifications would lead risk-averse governments to avoid otherwise socially desirable regulatory changes. It is useful to consider the differences in remedial framework in the WTO and ISDS. In the WTO, if the adjudicator should find that some aspect of the way a regulatory scheme is applied does not satisfy the conditions of Article XX, the appropriate remedy would be for the regulating state to fix the issue going forward; no reparatory damages exist under the WTO system, often requiring the dismantling of entire regulatory programmes, as opposed to surgically excising some discriminatory element. In the ISDS context, the consequences are quite different: a damages award of many millions of dollars that reflects economic harm to the investor has already been made. When the NAFTA was negotiated, ISDS was largely seen as a system that gave corporations from the North an alternative to domestic litigation in settling disputes with countries in the South, the latter supposedly lacking in the rule of law, and thought likely to engage in arbitrary government actions. Canada reacted to the early cases against it as if they were an abuse of the system: a government with a highly developed domestic legal system should not see its own courts bypassed by foreign investors, nor should such investors attack regulations promulgated by advanced democratic governmental systems. This may explain the relative ease in presenting the defending states as engaged in legitimate governmental activity – even if aspects of discrimination or political motives were clearly present.

II. EARLY NAFTA CASES

Early NAFTA investor-state disputes have had exceptional influence on investment law and jurisprudence, with persistent relevance beyond NAFTA.10 Different interpretative approaches regarding the term ‘tantamount to expropriation’ remain relevant after more than two decades. Despite their now infamous opacity, interpretative implications of protection standards in International Investment Agreements (IIAs), such as FET or indirect expropriation, were virtually unknown. These disputes caused a domino effect in the interpretation of such provisions and have been applied to specific cases beyond their trilateral context. This conundrum is apparent in claims lodged against NAFTA Contracting Parties under Articles 1102(2) on national treatment, 1110 on Expropriation, and 1105 on minimum standard of treatment. Generally, the early NAFTA tribunals adhered to the view that, to violate the minimum standard of treatment, government conduct had to entail serious impropriety, lack of due process, or administrative unfairness. The early NAFTA cases do not deploy FET as a means of protecting the investor’s expectations against regulatory change, avoiding the issue of which regulatory changes are legitimate exercises of public policy, and which undermine the legal and regulatory framework the investor expected and arguably relied on. Nevertheless,

10 C

Titi, The Right to Regulate in International Investment Law (Baden-Baden, Nomos/Hart, 2014).

102 Robert Howse and Güneş Ünüvar in Metalclad, the Tribunal found that under Article 1105 some kinds of investor expectations were protected, albeit by specific representations or commitments of the government. The consideration of substantive rationality of Mexican measures and a conception of proportionality came into the analysis. Still, the Tribunal stayed clear of the notion that the minimum standard of treatment establishes a general entitlement of the investor to stabilisation of the legal and regulatory framework.11

A. Ethyl v Canada Ethyl v Canada, a dispute between Ethyl – a US corporation with a Canadian affiliate to which it supplied a gasoline additive, MMT, through production facilities located in the US and Canada – and Canada, was the first case to give rise to alarm in Canada that ISDS might undermine legitimate health or environmental policies. This is anomalous as the merits were never adjudicated in Ethyl and it is thus unclear how an ISDS tribunal would have evaluated the policy dimensions. While the target of the measure theoretically raised environmental or health concerns, the government acted because automakers found that the chemical concerned interfered with the performance of technology in their vehicles. The government, seeking to satisfy automakers without putting Canadian ethyl producers out of business, banned only international and interprovincial trade in the additive, while continuing to allow its presence in gasoline. Prior to enacting the trade ban, the Canadian Government had received advice that such a measure likely violated NAFTA and WTO law,12 while the Canadian environment minister contemporaneously maintained that MMT ‘endangers our children’s health’ and the ‘air we breathe and the water we drink’.13 These statements were without factual and scientific foundation and dissembled the fact that the government’s initial actions had been based on concerns by automakers about technology in cars malfunctioning and not any matter of environment or public health. Indeed, the Canadian health ministry had not come to any explicit conclusion that MMT was harmful to health and therefore that it could not be prohibited under federal health regulation.14

B. S.D. Myers v Canada S.D. Myers (SDMI) v Canada concerned a claim by SDMI, a transformer maintenance company with business activities in remediation and disposal of products contaminated with polychlorinated biphenyl (PCB), a highly toxic chemical used for insulation. As the US market declined in the 1990s, SDMI had taken interest in Canada to extend the usefulness of their facility and business.15 The plan was to ‘obtain PCB waste for treatment by SDMI in its US facility’,16 thereby transferring PCB waste from Canada to the US. Canada, along with many other nations, prohibited the use of PCBs in new products, and ‘effectively banned the export

11 Metalclad

Corporation v The United Mexican States, Award, 30 August 2000, paras 74–101. Corporation v Canada, Award on Jurisdiction, 24 June 1998, paras 9–10. 13 ibid 10. 14 Ethyl Corporation v Canada, Notice of Arbitration, 10 September 1996, paras 6–8. 15 S.D. Myers, Inc. v Canada, UNCITRAL, Partial Award, 13 November 2000, para 92. 16 ibid para 93. 12 Ethyl

The Notorious First Wave of NAFTA Chapter 11 Awards 103 of PCB waste from Canada to all countries other than the USA’, subject to the approval of local authorities in the US.17 Similarly, the US prohibited the import and export of PCB waste for disposal. The subsequent US-Canada Transboundary Agreement envisaged the cross-border transfer of hazardous waste if the cross-border disposal facility is the ‘nearest’. Canada argued that PCBs were outside the scope of this provision and tried to keep the PCB disposal activities within its borders. Canada sustained its position that ‘the handling of PCBs should be done in Canada by Canadians’,18 eventually promulgating a ministerial order and definitively banning the export of PCBs from Canada. SDMI claimed that Canada violated its obligations under Article 1102 (national treatment) and 1110 (expropriation).19 According to SDMI, ‘U.S. waste disposal companies were not permitted to operate in Canada in the same fashion as Canadian PCB waste disposal companies’, an arbitrary and discriminatory limitation.20 SDMI claimed that Canada was aware of the specific effects the ban would have on its investment, while granting better treatment to Canadian-based companies,21 and that the ban was aimed at protecting Canadian businesses vis-à-vis those based in the US and was not based on ‘bona fide health or environmental concerns’. Canada claimed that it ‘merely established a uniform regulatory regime under which all were treated equally’ – as no business (whether domestic or foreign) was allowed to export, there was no discrimination.22 The Tribunal promptly (and rightly) dismissed this argument as ‘one-dimensional’.23 It considered the measure solely from the point of view of how it affected trade, rather than the claimant’s business as an investment, the real issue under Chapter 11. The Tribunal contended that in order to understand what ‘like circumstances’ means in Article 1102, ‘close attention must be paid to the legal context in which the word [like] appears’,24 identifying that this legal context comprises the NAFTA text and its companion North American Agreement on Environmental Cooperation (NAAEC).25 The Tribunal identified three principles: states have the right to establish high levels of environmental protection; states should avoid distorting trade; and environmental protection and economic development are not mutually exclusive and can support one another.26 The Tribunal noted that the assessment of whether like circumstances existed must take into account differential treatment based on public interest concerns. As all parties of NAFTA are OECD members, the Tribunal considered OECD practice and instruments to be relevant to this context, determining that ‘likeness’ requires foreign and domestic firms to operate in the same sector.27 The Tribunal ultimately decided that SDMI was in like circumstances with Canadian competitors, as: [t]hey all were engaged in providing PCB waste remediation services [and] SDMI was in a position to attract customers that might otherwise have gone to the Canadian operators because it could offer more favourable prices and because it had extensive experience and credibility. It was precisely because SDMI was in a position to take business away from its Canadian competitors that Chem-Security

17 ibid

para 100. para 116. 19 ibid paras 129 et seq. 20 ibid para 131. 21 ibid para 133. 22 ibid para 241. 23 ibid para 242. 24 ibid para 244. 25 The North American Agreement on Environmental Cooperation (1994). 26 S.D. Myers, above (n 20) 247. 27 OECD, 1993. 18 ibid

104 Robert Howse and Güneş Ünüvar and Cintec lobbied the Minister of the Environment to ban exports when the U.S. authorities opened the border.28

The Tribunal acknowledged Canada’s policy objectives, including ensuring ‘the economic strength of the Canadian industry’ and maintaining ‘the ability to process PCBs within Canada in the future’29 – contending that Canada had at its disposal means to achieve these goals that would not violate the national treatment obligation in NAFTA.30 The Tribunal applied the necessity test characteristic of GATT/WTO panels and the Appellate Body under Article XX, determining that there were means by which Canada could pursue its legitimate public policy objectives that did not impair the obligations of the NAFTA to the extent that closing the border did.31 At the same time, the Tribunal should have been more skeptical of Canada’s assertion of the need for domestic processing capacity to ensure safe processing of PCB’s, given the US-Canada Transboundary Agreement and the safety and viability of Myers’ cross-border processing facilities. SDMI alleged that the Ministerial Order was tantamount to expropriation in its Article 1110 expropriation claim. The Tribunal’s observation is striking, in light of the direct–indirect expropriation dichotomy that has since emerged: ‘(…) [t]he general body of precedent usually does not treat regulatory action as amounting to expropriation. Regulatory conduct by public authorities is unlikely to be the subject of legitimate complaint under Article 1110 of the NAFTA.’32 In retrospect, this observation is inaccurate. Subsequent arbitral tribunals have discussed potentially expropriatory regulatory measures,33 with this incremental development creating debates concerning the difference between a non-compensable legitimate regulatory action and an indirect expropriation.34 While it is difficult to ascertain to which ‘general body of precedent’ the Tribunal refers, in the absence of any noteworthy precedent stemming from investor-state arbitration at the time, it is likely that the elaborations of the US–Iran Claims Tribunal, which has extensively dealt with expropriation claims against Iran, played a role in this determination.35 The Tribunal associated the act of ‘taking’ with expropriation, determining that a regulatory measure alone would not amount to a taking. The Tribunal suggested another, inaccurately absolute, distinction: contrasting the ‘deprivation of ownership rights’ and a lesser degree of interference through regulatory action. Subsequent treaty-making practice and arbitral case law contradict these projections. While there are regulatory carve-outs in newer IIAs and investment chapters of FTAs, no such sweeping regulatory exception exists to suggest that a regulatory measure cannot be expropriatory

28 S.D.

Myers, above (n 20) para 250 (emphasis added). para 255. 30 ibid. The Tribunal further suggested that ‘the right to source all government requirements and to grant subsidies to the Canadian industry are but two examples of legitimate alternative measures’. 31 Later, however, a WTO panel and the Appellate Body (India Solar) would reject an Article XX argument by India that its measure was justified by the need for an industrial policy to ensure domestic capacity to support an environmental programme. 32 S.D. Myers, above (n 20) para 281. 33 Glamis Gold Ltd. v The United States of America, UNCITRAL, Final Award, 8 June 2009, para 356; Suez, Sociedad General de Aguas de Barcelona S.A. and Vivendi Universal SA v Argentine Republic, ICSID Case No ARB/03/19, Decision on Liability, 30 July 2010, para 132. 34 In light of the sole effects – police powers dichotomy, this early observation is striking. See, U Kriebaum, Regulatory Takings (Leiden, Brill, 2007). 35 CV Heiskanen, ‘The Doctrine of Indirect Expropriation in Light of the Practice of the Iran-United States Claims Tribunal’ (2007) 8(2) Journal of World Investment & Trade 215, 229–31. 29 ibid

The Notorious First Wave of NAFTA Chapter 11 Awards 105 per se. This notwithstanding, the Tribunal has accurately identified expropriatory measures as amounting ‘to a lasting removal of the ability of an owner to make use of its economic rights’.36 As the measure did not result in a transfer of title or directly benefit others, the Tribunal decided that such ‘delay of opportunity’ did not amount to expropriation.37

C. Pope & Talbot v Canada Pope & Talbot v Canada was a dispute between an American lumber company active in British Columbia and Canada, concerning the latter’s implementation of the Softwood Lumber Agreement (SLA), a bilateral agreement between the US and Canada concerning quotas on softwood lumber export to the US. Pursuant to the SLA, Canada would issue licences to, and collect fees on, softwood lumber exported from British Columbia based on fixed thresholds. Pope & Talbot, which had remained below its allocated export quota for the relevant year, claimed that Canada violated, inter alia, NAFTA Articles 1102 and 1110, asserting that the operation of licences and fees under the scheme disadvantaged the Claimant relative to Canadian operators in ‘like circumstances.’ The Tribunal’s analysis of the Article 1102(2) claim focused, in particular, on the standards to be employed in determining whether the Claimant’s investment was denied treatment ‘no less favorable’ than received by Canadian investors, and the investments/investors which should be considered when determining whether the treatment was less favourable.38 As Article 1102(2) notes that ‘[e]ach Party shall accord to investments of investors’ no less favourable treatment, Canada argued that, for an Article 1102 violation, a Party’s failure must extend to multiple investments. It noted that the provision: [W]ould have been worded differently had it intended that only the circumstances of the disputing investment were the relevant comparator. If the Investor were correct, Article 1102(2) would require comparison of the treatment accorded to a disputing investment with the treatment accorded to domestic investments, even to a single domestic investment.39

As semantically noteworthy as this argument may be, the Tribunal rejected this reading of the provision.40 Canada argued ‘a violation of national treatment obligations can be found only if the measure in question disproportionately disadvantages the foreign owned investments or investors’.41 For a measure to disproportionately disadvantage a foreign investor, the Tribunal must examine whether a sufficient number of Canadian investors are accorded the same treatment vis-à-vis Canadian investors accorded more favourable treatment. If the latter group is smaller than the group of investors accorded the same treatment as the foreign investor, there should be no violation of national treatment.42 The Tribunal disagreed with this ‘disproportionate disadvantage’ test. Citing the WTO Beer case,43 the Tribunal contended that ‘[o]nly

36 ibid

283. 287–288. 38 Pope & Talbot Inc. v Canada, UNCITRAL, Award on the Merits Phase 2, 10 April 2001, para 31–32. 39 ibid para 34, citing Canada’s Phase 2 Supplemental Counter Memorial. 40 ibid para 36. The Tribunal asserts this position by drawing an analogy between the treatment of foreign investors on the one hand and the human rights of women and children. Para 37. 41 ibid para 43. 42 ibid para 44. 43 United States – Measures Affecting Alcoholic and Malt Beverages, Report of the Panel Adopted on 19 June 1992 (DS23/R – 39S/206). 37 ibid

106 Robert Howse and Güneş Ünüvar in the simplest and most obvious cases of denial of national treatment could the complainant hope to make a case for recovery’. This would ‘weaken the provisions and objectives of NAFTA’.44 The Tribunal noted that the expression ‘no less favorable’ ‘means equivalent to, not better or worse than, the best treatment accorded to the comparator’.45 From the perspective of cross-judging, this analysis of WTO precedents is of particular interest. Canada invoked inter alia the Bananas WTO ruling,46 which examined the relationship of the treatment of a particular foreign (non-EU) economic actor to the treatment of internal economic actors, as well as the situations of multiple foreign actors. The Tribunal highlighted the difference between WTO dispute settlement, where a State Party’s government brings a claim on behalf of relevant economic actors, and ISDS under NAFTA, where the Claimant brings the case on its own behalf and cannot be expected to address the situation of other foreign investors under the measure in question.47 The Tribunal relied on this institutional difference, and delved deeply into the WTO jurisprudence to show that the WTO adjudicator had not actually adopted the approach of requiring that overall treatment of foreign versus domestic economic actors be disadvantageous in order for there to be a violation of national treatment. The Tribunal grasped that this sort of balancing approach, which would allow a State Party to treat some foreign economic actors worse than like domestic actors if it treated other foreign economic actors better (a kind of balancing approach), was inconsistent with the legal security of market access required for a system of international economic law to function properly, creating relatively stable conditions for secure expectations of market access. Thus, a foreign economic actor could never be sure of its national treatment rights because these could be reduced to the extent that treatment of some other foreign economic actor exceeded that of like domestic firms. Here, the tribunal cited an early GATT precedent, the United States – Section 337 of the Tariff Act of 1930 case, which rejected this sort of balancing approach:48 If this notion were accepted, it would entitle a contracting party to derogate from the less favourable treatment obligation in one case, or indeed in respect of one contracting party, on the ground that it accords more favourable treatment in some other case, or to another contracting party. Such an interpretation would lead to great uncertainty about the conditions of competition between imported and domestic products and thus defeat the purposes of [national treatment].49

The Tribunal identified ‘domestic entities whose treatment should be compared with that accorded [Pope & Talbot]’,50 and contended that the treatment conferred to a foreign investor should be compared ‘with that accorded domestic investments in the same business or economic sector’.51 The Tribunal noted that differences in treatment will not violate Article 1102(2) if ‘they have a reasonable nexus to rational government policies that (1) do not distinguish, on their face or de facto, between foreign-owned and domestic companies, and (2) do not otherwise unduly undermine the investment liberalizing objectives of NAFTA’.52

44 Pope,

Award of the Merits Phase 2, above (n 43) para 72. para 42. 46 European Communities – Regime for the Importation, Sale and Distribution of Bananas, Report of the Panel Adopted on 22 May 1997 (WT/DS27/R/USA). 47 Pope, Award of the Merits Phase 2, above (n 43) paras 44–72. 48 L/6439–36S/345, January 1989. 49 ibid 5.13–5.14. 50 Pope, Award of the Merits Phase 2, above (n 43) para 73. 51 ibid para 78. 52 ibid. 45 ibid

The Notorious First Wave of NAFTA Chapter 11 Awards 107 The Tribunal contended that the differentiation between the ‘covered’ and ‘non-covered’ provinces under the SLA corresponded to a rational government policy, and that ‘the producers in the non-covered provinces were not in like circumstances with those in covered provinces’,53 thus finding no violation of Article 1102(2) based on the treatment of producers in non-covered provinces. The Tribunal dismissed the investor’s argument that there was discrimination in the allocation of quotas among producers in different Canadian provinces. The Tribunal found that the pattern of allocations in question had ‘a reasonable nexus with the rational policy of providing new entrants and it had no elements of discrimination against foreign-owned producers’.54 It dismissed the claim that the Claimant was in like circumstances to new entrants in covered provinces. Finally, the Tribunal considered the nature of measures affecting the softwood lumber producers in British Columbia, where the Claimant’s mills were located. It considered the adverse effects of the difference between the stumpage fees applicable to coastal and inland producers of timber. Pope & Talbot’s inland mills had been disadvantaged but the Tribunal considered this difference to bear ‘a reasonable relationship to a rational choice of remedies aimed at avoiding a threat to the SLA’.55 Thus the Tribunal ultimately decided that there was no violation of Article 1102(2). The Tribunal was satisfied that there was a general rational public policy that justified the main categories in the SLA, a more deferential test than the necessity test deployed under Article XX by WTO panels and the Appellate Body. This reflects the Tribunal’s failure to grapple with the argument that Canada could have achieved an orderly and regionally equitable resolution of a longstanding trade dispute with the US through allocation practices that resulted in less detrimental treatment of the Claimant. Similarly, under Article XX of the GATT, the preambular paragraph requires that the detailed application of a measure does not result in arbitrary or unjustifiable discrimination. In Pope, the investor made a series of detailed allegations concerning discriminatory anomalies in the specific manner that the Claimant was treated by Canadian authorities. The Tribunal appeared satisfied that national treatment was not violated merely due to the general design or structure of the implementing measures of the SLA having a seemingly rational basis. This approach may allow governments to embed protectionism or favouritism in the administration of general regulatory schemes, provided that the scheme has a rational nexus to non-protectionist regulatory objectives and that it does not evidence prima facie nationalitybased discrimination. Ultimately, the Tribunal found a violation of Article 1105. Still, one should contrast the approach of the Pope Tribunal to that of S.D. Myers, which it cites favourably. On expropriation, the Claimant argued that Article 1110 ‘provide[d] the broadest protection for the investments of foreign investors who may suffer harm by being deprived of their fundamental investment rights’.56 According to its reasoning, ‘measure tantamount to expropriation’ covers not outright taking, and ‘even non-discriminatory measures of general application which have the effect of substantially interfering with the investments of investors of NAFTA Parties’.57 Canada argued that expropriation in international law requires an ‘actual interference with fundamental ownership rights’,58 emphasising the degree of severity

53 ibid

para 88. para 93. 55 ibid para 102. 56 Pope & Talbot Inc. v Canada, Interim Award (26 June 2000) para 83. 57 ibid para 84. 58 ibid para 88. 54 ibid

108 Robert Howse and Güneş Ünüvar and asserting that ‘mere interference is not expropriation: rather, a significant degree of deprivation of fundamental rights of ownership is required’.59 According to Canada, states are not ‘required to compensate an investment for any loss sustained by the imposition of a non-discriminatory, regulatory measure’ and that ‘liability is possibly only if the measure is discriminatory’.60 The Tribunal disagreed with the Claimant that the phrase ‘tantamount’ extended the scope of expropriation, noting that the degree of severity inflicted upon investments was decisive in what measures would be deemed compensable. But the Tribunal also disagreed with Canada that a measure being regulatory per se redeemed it from being expropriatory, noting that ‘creeping expropriation could be conducted by regulation, and a blanket exception for regulatory measures would create a gaping loophole in international protections against expropriation’.61 The Tribunal’s assessment, which concludes that the claimed ‘loss of profits’ does not reach the required degree of severity required from an expropriatory measure, is noteworthy: […] the investor remains in control of the Investment, it directs day-to-day operations of the Investment, and no officers or employees of the Investment have been detained by virtue of the [quota regime]. Canada does not supervise the work […], does not take any of the proceeds of company sales […], does not interfere with management or shareholders’ activities […], does not interfere with the appointment of directors or management and does not take any other actions ousting the Investor from full ownership and control of the Investment.62

Canada’s argument does not suggest a ‘blanket exception for regulatory measures’. A measure would have to be ‘regulatory’ and non-discriminatory to be considered within the defending state’s police powers, and therefore excluded from the scope of measures tantamount to expropriation. Thus, if a measure is discriminatory, its regulatory nature would not make it per se a non-compensable act. The Canadian position in Pope appears to have played a pivotal role in the revision of expropriation provisions in Canadian and American Model BITs. Thus, virtually all North American treaties signed since the early 2000s exclude non-discriminatory regulatory measures from the scope of indirect expropriation. This ‘carve out’ method has also guided the EU position in its more recent IIAs.63 The interpretation of Article 1105 in Pope raises difficult systemic questions of treaty law. The Tribunal found a violation of Article 1105, as the minimum standard of treatment was included in the conception of FET, though this in no way limited the standard of treatment, which would have to be articulated in light of, inter alia, the purposes of the NAFTA.64 Canada maintained that the content of FET was confined to the customary international law of the protection of aliens, as articulated in arbitral commission decisions from the early twentieth century, especially Neer v Mexico.65 Thus, to find a violation of the minimum standard of treatment, the government conduct must be egregious and shocking to any reasonable observer. Rejecting this view, the Tribunal looked to developments in international law since

59 ibid. This contention, which does not exist in the NAFTA text of the FTC Notes of Interpretation, found its way into several new-generation FTAs. Cf, Annex 9-A of the EU–Singapore FTA. 60 ibid. 61 ibid para 99. 62 ibid para 100. 63 For example, Comprehensive Economic and Trade Agreement between the EU and Canada (CETA), Art 8.12 on Expropriation and Annex 8-A. 64 Pope, Award of the Merits Phase 2, above (n 43) para 111. 65 LFH Neer and Pauline Neer (USA) v United Mexican States, 15 October 1926, Vol IV, 60–66.

The Notorious First Wave of NAFTA Chapter 11 Awards 109 Neer, finding that ‘[…] covered investors and investments receive the benefits of the fairness elements under ordinary standards applied in the NAFTA countries, without any threshold limitation that the conduct complained of be ‘egregious, ‘outrageous’ or ‘shocking,’ or otherwise extraordinary’.66 The Tribunal found that the fairness standards in question could be additive to international law, even as it had evolved beyond the customary law of diplomatic protection. Reacting hereto, Canada, Mexico and the US issued a Joint Interpretative Statement on the meaning of Article 1105, stipulating that: the ‘international law’ minimum standard of treatment required by Article 1105 was ‘the customary international law minimum standard of treatment of aliens’; FET did not require treatment ‘in addition to or beyond’ that standard; and a breach of another provision of conventional international law could not be a basis for establishing that Article 1105 had been violated.67 This interpretation was issued pursuant to provision of the NAFTA that permits State Parties, acting as the NAFTA ‘Commission’, to render binding authoritative interpretations. One issue was whether, since the Tribunal had not yet completed the damages phase of the proceedings, the Commission’s interpretation should be applied, as it were, retroactively. The Claimant argued that the State Parties abused the power to interpret and amend the NAFTA outside of the agreed amendment procedures. One way the interpretation took on the character of the amendment is that the original NAFTA Article 1105 refers to ‘international law’ generally, not simply to customary international law, as defining the minimum standard. Limiting ‘international law’ to custom and precluding a tribunal from considering other international law sources would violate the customary international law of treaty interpretation as set out in Articles 31 and 32 of the Vienna Convention on the Law of Treaties. The Commission was, in effect, purporting to create a special rule of interpretation for Article 1105, requiring amendment of Article 102(2), if not Article 1105 itself. Conversely, the rejection of the Commission interpretation supported by all NAFTA governments as an impermissible attempt at amendment would have created a political crisis, given the controversy created by the Ethyl and S.D. Myers rulings. The approach of the Pope Tribunal deftly skirted this outcome. While the Tribunal stated that if it had to determine the legal significance of the Commission statement, it would have been inclined to view it as an attempt at amendment, the Tribunal held that it need not make such a determination, because even if the interpretation were fully valid, customary international law itself had evolved to sustain the interpretation of Article 1105 that the Tribunal had already arrived at in the merits phase. Subsequent tribunals have adopted a version of this approach, viewing the interpretation as focusing the application of Article 1105 on customary international law, but not in any way freezing custom at some historical point in the development of the law of diplomatic protection. With rare exceptions,68 tribunals have considered non-customary sources of international law in ascertaining state practice and opinio juris to determine the state of custom. Such tribunals have read the Commission interpretation to avoid the implication of impermissible amendment, albeit while focusing on articulating the contemporary customary standard. This path was already prepared by the Pope Tribunal.

66 ibid

para 118. of Interpretation of Certain Chapter 11 Provisions (NAFTA Free Trade Commission, 31 July 2001), https:// perma.cc/FU9G-HXCB. 68 Glamis Gold, Ltd. v The United States of America, UNCITRAL, Award, 8 June 2009. 67 Notes

110 Robert Howse and Güneş Ünüvar D. Metalclad v Mexico In Metalclad v Mexico, Metalclad, a US company, claimed that Mexico interfered with the establishment and operation of a hazardous landfill area through its administrative organs,69 thus violating NAFTA Article 1110. Metalclad had acquired COTERIN, a state-owned company which possessed permits and authorisations for the operation of a hazardous landfill. However, following its acquisition of COTERIN, the local governor launched a campaign against the landfill’s operation, subsequently halting its operation due to the absence of a construction permit. Despite environmental impact reports in favour of its operation, the campaign against the landfill persisted. During the opening of the landfill, protesters blocked buses carrying guests and workers, and ‘employed tactics of intimidation’. Ultimately, the construction permit application was denied, and after the submission of the dispute to arbitration,70 the landfill area was declared a natural area for the protection of a rare cactus species via an Ecological Decree. The Metalclad Tribunal’s expropriation analysis is an example of an approach to indirect expropriation focused exclusively on economic impact on the investor, without regard to public policy interests: […] expropriation […] includes not only open, deliberate and acknowledged takings of property, such as outright seizure or formal or obligatory transfer of title in favour 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.71

There are two distinct propositions in this statement, though they have often been conflated or confused by commentators. The first is that government actions other than conversion of title, or formal expropriation may constitute compensable takings under NAFTA, based upon the deprivation of ‘reasonably-to-be-expected economic benefit’ to the investor. The Tribunal neither includes nor excludes explicitly measures that are legitimate non-discriminatory exercises of state police powers. In any case, the Metalclad Tribunal did not go on to examine whether the actions of the Mexican authorities might be regarded as legitimate exercises of police powers, therefore not constituting a compensable regulatory taking. In consequence, Metalclad is widely cited for the proposition that the only question concerning whether there is a compensable regulatory taking is the nature of the economic impact on the investor. Considerations of legitimate public policy are irrelevant. But this was not necessarily the intent of the Tribunal. Having found that the actions of the Mexican authorities constituted violations of FET under the NAFTA, the Tribunal may have considered that the elements of unfairness and arbitrariness that it had discerned in coming to this conclusion rendered the actions of the Mexican authorities obviously ineligible for consideration as legitimate exercises of police powers. This does not mean that in all cases the Metalclad Tribunal would have it that economic effect on the investor will be the sole deciding consideration.72 Thus, since an economic benefit to the state is not a necessary condition

69 Mexico is a federation. Therefore, the state authority is different from the national federal government. Metalclad Corporation v The United Mexican States, Award (30 August 2000) para 1. 70 ibid para 59. 71 ibid para 103 (emphasis added). 72 In some Commonwealth countries (eg Canada), compensation is only required for expropriation when there is deprivation and a corresponding appropriation of economic value by the state. This is not the case for investorstate arbitration – hence the Metalclad Tribunal’s conclusion that the taking does not have to be ‘to the obvious

The Notorious First Wave of NAFTA Chapter 11 Awards 111 for the measure qualifying as an expropriation, the exclusive focus is the negative economic impact on the investor. The question of benefit to the state is simply an entirely separate issue from that of police powers. It is true that where the state does benefit, there might be more of a justifiable suspicion that the state is acting opportunistically, to take valuable assets for its own use rather than out of legitimate public policy concerns. However, those suspicions will be most relevant not to regulatory takings, but rather in cases of direct expropriation where the investor’s property is subject to outright conversion by the state. Nevertheless, the tribunal further scrutinised the consistency of Mexico’s actions: By permitting or tolerating the conduct of Guadalcázar [authorities] in relation to Metalclad which the Tribunal has already held amounts to unfair and inequitable treatment […] and by thus participating or acquiescing in the denial of Metalclad of the right to operate the landfill, notwithstanding the fact that the project was fully approved and endorsed by the federal government[…].

This builds upon the FET analysis conducted by the Tribunal, whereby it considered elements such as legitimate expectations and the proportionality of the governmental measures in question. According to the Tribunal, no public interest had been persuasively argued which could justify the frustration of the legitimate expectations of Metalclad regarding the pending operational permits.73 The Tribunal noted that ‘Metalclad was entitled to rely on the representations of federal officials and to believe that it was entitled to continue its construction of the landfill … Metalclad was merely acting prudently and in the full expectation that the permit would be granted.’74 The Tribunal contended that ‘Mexico failed to ensure a transparent and predictable framework for Metalclad’s business planning and investment.’ The crux of the debate on public policy related to the declaration of the landfill area as a natural area for the protection of a rare cactus species.75 While the paragraph may be proof that the Tribunal outrightly rejected considering public purpose of a regulatory measure in any given situation, the facts of the case and the conduct of the government and the municipality hindered a plausible argument which could justify the protection of the rare cactus species. In the award, the Tribunal emphasises the: […] investor’s justified reliance on the government’s representations regarding the permit, the fact that government authorities knew of the construction for more than one year before issuing the stop work order, the fact that permits had not been required for other projects and the fact that no procedure was in place for dealing with building permit applications.76

The Tribunal concluded that whether such misconduct was carried out by the refusal of the permit or by the issuance of an Ecological Decree would not make a difference in justifying the government’s conduct on the grounds of public purpose.77 The Tribunal does not explicitly declare that a genuine measure with the objective of protecting environment would unequivocally absolve Mexico from any responsibility – but neither does it make the assertion that public purpose is of no consequence to the expropriation analysis. It would stretch

benefit of the host state’. See: A Newcombe, ‘Regulatory Expropriation, Investment Protection and International Law: When Is Government Regulation Expropriatory and When Should Compensation Be Paid?’ (LLM Thesis, Graduate Department of Faculty of Law, University of Toronto, 1999), https://perma.cc/DYX9-YGSG. 73 Metalclad, above (n 74) para 108. 74 ibid para 89. 75 ibid para 111. 76 ibid para 108. 77 ibid para 111.

112 Robert Howse and Güneş Ünüvar the meaning for the Tribunal to assert that any governmental measure to protect environment would categorically fail to ‘redeem’ an otherwise expropriatory measure. It appears that its dismissal of the Ecological Decree stems from the spurious nature of the measure itself – not from its purported objective.

E. Feldman v Mexico Feldman v Mexico concerned the application of Mexican tax laws to an American-owned Mexican-incorporated cigarette export company, CEMSA. CEMSA claimed that Mexico refused to apply rebates to taxes paid for cigarette exports and to apply such rebates to prospective exports by the company. Mexico applied a tax for production and sale of cigarettes in the domestic market, which under certain circumstances could be zero. Initially, the tax law had not excluded the extension of rebates to resellers, though a 1991 amendment limited the application to ‘producers and bottlers of the goods’.78 CEMSA alleged that this amendment was the result of lobbying by domestic cigarette producers against the claimant’s entitlement to zero tax, a point refuted by Mexico.79 Following a challenge to the constitutionality of this amendment, the law was reverted to its previous iteration, allowing resellers to benefit from the rebate. This notwithstanding, CEMSA’s export business was affected by product invoicing requirements, with which it could not comply. The law required the tax deductible to be listed separately on invoices, as was only possible for producers. CEMSA could not list these taxes separately, and while paying the rebates, Mexico continued to request CEMSA to comply with the separate itemisation of taxes. Mexico terminated rebates to CEMSA, valued at US$2 million, after about two years, before the law was amended to prohibit rebate payments to resellers and require the registration of cigarette exporters, which was denied to CEMSA. Following on therefrom, CEMSA was audited, and called upon to partially repay the rebates paid to it. This led Feldman, the owner of CEMSA, to pursue a NAFTA Chapter 11 claim. Feldman claimed violations of NAFTA Articles 1102 and 1110. After identifying the distinctions between direct-indirect expropriation, and expropriation and non-expropriatory regulatory measure,80 the Tribunal summarised different prongs of its expropriation analysis as follows: (a) not all business problems are expropriatory; (b) international law does not require a state to permit ‘gray market’ exports of cigarettes; (c) the relevant Mexican law had not granted cigarette resellers the right to export cigarettes; and (d) the investment remains under ‘complete’ control of the investor and is able to conduct business with regard to other goods.81 The Tribunal acknowledged the ‘great difficulties’ CEMSA experienced in dealing with Mexican tax authorities but noted that ‘tax authorities in most countries do not always act in a consistent and predictable way’.82 Furthermore, it added: […] as in most tax regimes, the tax laws are used as instruments of public policy as well as fiscal policy, and certain taxpayers are inevitably favored, with others less favored or even disadvantaged.83

78 ibid

para 10. Roy Feldman Karpa v United Mexican States, ICSID Case No ARB(AF)/99/1, Award (16 December 2002)

79 Marvin

para 10. 80 ibid paras 98–107. 81 ibid para 111. 82 ibid para 113. 83 ibid.

The Notorious First Wave of NAFTA Chapter 11 Awards 113 The Tribunal scrutinised the ‘public purpose’ of the tax rebates scheme, reiterating CEMSA’s claim that prevention of cigarette resellers ‘from exporting cigarettes from Mexico to other countries’ is ‘in conflict with normal Mexican policies that promote exports’. This highlights the adjacent public policy of preventing cigarette smuggling, reportedly ‘a significant problem for Mexico’,84 and retaining a degree of control over who is authorised to export cigarettes. The Tribunal held that the requirement for invoices to separately itemise relevant taxes to be rebated was ‘valid’.85 Eventually, the Tribunal denied that Mexico expropriated CEMSA’s investment. As for the Article 1102 claim on National Treatment, the Tribunal identified four nonexhaustive steps to determine whether this provision was violated: the question of ‘like circumstances’; the existence of de jure or de facto discrimination; the extent of the causal link between the differential treatment and the nationality of the investor; and whether favourable treatment should be extended domestic investors.86 The Tribunal noted that ‘at least three trading companies’ received rebates under the same scheme. The Tribunal identified two Mexican companies in like circumstances, namely Mercados I and Mercados II, owned by the same Mexican corporate group.87 As CEMSA ‘was denied […] rebates during periods when members of [Mercados I and II] were receiving them’, the Claimant was treated ‘in a less favorable manner than domestically owned reseller/exporters of cigarettes’ and the fact that the CEMSA audit was conducted ‘well before any other domestic reseller/exporters’ were examples of de facto discrimination.88 Similarly, the denial of CEMSA’s registration as an export trading company was deemed discriminatory, in light of the existence of domestic grantees in comparable situations. After attributing these acts to the state, the Tribunal examined whether there was nationalitybased discrimination. Acknowledging the difficulties for the Claimant to prove definitively that discrimination occurred because of their nationality,89 the Tribunal noted that such determination must be explicit, and considered the fact that ‘CEMSA was owned by a very outspoken foreigner, who had, prior to the initiation of the audit, filed a NAFTA Chapter 11 claim against [Mexico]’ as an ‘obvious’ indicator of ‘less favourable de facto treatment’.90 The audit having begun right after the filing of the claim and apparent links between the claim and the registration efforts of CEMSA exacerbated the suspicion that the discrimination was indeed about the Claimant’s nationality.91 The approach to the distinction between de facto and de jure discrimination in this decision, like S.D. Myers, is complicated. Clearly the Tribunal considered that there was some evidence of intentional discrimination based on nationality. Nevertheless, it stopped short of basing its dispositif on intentional or de jure discrimination based on nationality. The Feldman Tribunal cites S.D. Myers stating that once de facto discrimination has been established there is a presumptive violation of Article 1102 of NAFTA, and the onus is on the defending State Party to establish a rational nexus of the measure to government policies that are neither de jure nor de facto discriminatory, nor undermining of the trade liberalising objectives of the

84 ibid

para 136.

85 ibid. 86 ibid

para 166. paras 172–173. 88 ibid paras 173 et seq. 89 ibid para 183. 90 ibid para 182. 91 ibid. 87 ibid

114 Robert Howse and Güneş Ünüvar NAFTA.92 In Feldman, the Tribunal seemed satisfied on this basis that there was a violation of Article 1102, as Mexico had not shown the rational nexus to legitimate non-protectionist government policies. Drawing from WTO Appellate Body jurisprudence, the Tribunal shifted the burden of proof to find de jure or intentional discrimination on the basis of sufficient evidence of nationalitybased discrimination, thereby placing an evidentiary burden on Mexico to show a proper motivation for the differential treatment, a burden which Mexico had not met.93 Ultimately, the Tribunal saw two routes to establishing a violation of Article 1102: the establishment as a factual matter of less favourable treatment of the investor relative to a similar domestic investment, where the defending state could not establish a rational nexus to a non-protectionist public policy; or, alternately, the Claimant met an evidentiary threshold for indicators of intentional nationality-based discrimination. This would shift the burden of proof to the defending state to show that, despite the appearance or suspicion of nationality-based discrimination, its measure has a legitimate, non-discriminatory, nationality-neutral motivation. In the end, the Tribunal contended that it did not have to reach to a general conclusion on whether ‘Article 1102 requires treatment equivalent to the best treatment provided to any domestic investors.’94 Thus, Mexico violated its obligation under Article 1102.

F. Methanex v United States Methanex v United States concerned a state-wide ban on MTBE, a chemical substance used for octane enhancement in gasoline,95 of which methanol is a key component. The Claimant, Methanex, did not produce or sell MTBE, though it produced, transported, and marketed methanol, with ‘one third of [its] methanol production […] directed at the fuel sector, principally for use in methanol-based MTBE’.96 The governmental measures challenged included the 1997 California Bill and the 1999 Executive Order. According to the Bill, the California Senate commissioned an ‘assessment of the human health and environmental risks and benefits associated with the use of MTBE’,97 following which, in 1999, the Governor of California, through Executive Order D-5-99, recorded that ‘there was a significant risk to the environment from using MTBE in gasoline in California’.98 Thus MTBE was to be removed from gasoline in accordance with a specified timetable.99 Methanex subsequently lodged a claim against these measures and argued that Articles 1110 and 1102 had been violated. Regarding the expropriation claim under Article 1110, Methanex argued the measures would end its US business, resulting in a substantial taking of the investment, and that the ban was tantamount to expropriation. The US argued that the ban had a public purpose and was necessary due to the health and environmental risks associated with MTBE, primarily due to contamination of drinking water supplies. Methanex rejected the necessity of the ban and

92 ibid

para 184. paras 181, 183 94 ibid para 186. 95 Methanex Corporation v United States of America, UNCITRAL, Partial Award (7 August 2002) Chapter D, 14, para 22. 96 ibid paras 24–25. 97 ibid paras 26–27. 98 ibid para 29. 99 ibid para 30. 93 ibid

The Notorious First Wave of NAFTA Chapter 11 Awards 115 placed the responsibility regarding the drinking water contamination on several Californian and federal acts and laws. The Tribunal’s conclusions provide an elaborate evaluation of the legislation in question and apply a discretionary regulatory exception to what constitutes (indirect) expropriation. According to the Tribunal: […] as a matter of general international law, a non-discriminatory regulation for a public purpose, which is enacted in accordance with due process and, which affects […] a foreign investor or investment is not deemed expropriatory and compensable unless specific commitments had been given by the regulating government to the then putative foreign investor contemplating investment that the government would refrain from such regulation.100

This approach can be found in newer IIAs, in which non-discriminatory, legitimate regulatory measures are ‘carved out’ from the scope of indirect expropriation. Considering the applicable NAFTA Article 1110 and the absence of preceding tribunal elaborations at the time of the Methanex award, the Tribunal’s contention was novel and controversial. The Tribunal, relying on ‘general international law’, drew a line between non-compensable state acts and indirect expropriation. It determined the existence of an exception to the rule based on ‘legitimate expectations’: the rule is applicable, if investors’ legitimate, investment-backed expectations are not frustrated. The Tribunal specifically pointed to a situation where the state promises to refrain from a particular regulation or measure to a foreign investor, prompting the foreign investor to invest, but subsequently fails to live up to its promise at the expense of the investment. However, the Tribunal failed to identify any such promises given to Methanex and underlined the fact that the Claimant was fully aware of the precarious and political nature of health and environmental risks associated with methanol and ethanol investments in the US. The Tribunal rejected the presumption that Methanex invested in the US as a direct result of ‘special representations made to it’,101 and stated that the ban was a non-expropriatory ‘lawful regulation’, ‘made for a public purpose’ that was ‘non-discriminatory and was accomplished with due process’.102 Methanex’s Article 1102 claim centred on the Claimant’s disadvantaged position vis-à-vis domestic ethanol producers, claiming that the US ‘intended to deny foreign methanol producers, including Methanex, the best treatment it has accorded to domestic ethanol investors’.103 According to the Claimant, in assessing whether the national treatment obligation was violated, the Tribunal should assess whether: (a) like circumstances existed; (b) a portion of the domestic ethanol industry received better treatment; and (c) the US should prove that such difference in treatment stemmed from valid environmental goals.104 The US emphasised that the ‘function of the national treatment provision is to address discrimination on the basis of nationality of ownership of an investment’, and that a relevant comparator should be like the foreign investor ‘in all relevant respects, but for nationality of ownership’.105 Methanex posited that a ‘likeness’ of circumstances sufficed to fulfil the requirement, asserting that ‘competition’ was ‘the most accurate and widely recognised test of “likeness”’.106 100 Methanex Corporation v United States of America, UNCITRAL, Final Award (3 August 2005) Part IV Chapter D, 4 (emphases added). 101 ibid Part IV Chapter D 5. 102 ibid. 103 ibid Part II Chapter D 9, para 26. 104 ibid Part IV Chapter B 7, para 13. 105 ibid para 16. 106 ibid Part IV Chapter B 2, para 5. A similar discussion took place in S.D. Myers, where the emphasis was on S.D. Myers’ competitors in Canada. See above.

116 Robert Howse and Güneş Ünüvar As both methanol and ethanol producers compete for the oxygenate market, Methanex asserted that they are competitors, and thus in like circumstances. It supplemented its claim by highlighting that since the MTBE ban, the companies to which Methanex sold methanol shifted to ethanol, indicating their direct substitutability.107 The US argued that: where there is no identical domestically-owned counterpart to the foreign-owned investment […] a tribunal may look farther afield and expand the scope of domestically-owned comparators as long as they are similar enough.108

A significant issue of cross-judging arose due to the Claimant’s insistence that the Tribunal follow the understanding of national treatment developed by WTO panels and the Appellate Body in interpreting Article III:4 of the GATT. Methanex stressed that the approach to likeness in the determination of like products should apply in determining the comparators to be chosen under ‘like circumstances’ in Article 1102, including products that are in competition or substitutable. Methanex submitted an expert opinion from Claus Dieter Ehlermann, a founding Member of the WTO Appellate Body, arguing that, under the WTO approach, US ethanol producers would be ‘like’ Methanex, since the effect of the ban on MTBE was that market share shifted from methanol to ethanol. The Tribunal objected to the view that ‘like products’ under WTO jurisprudence could be simply applied to ‘like circumstances’ in the NAFTA. The Tribunal, citing the Mox Plant case, stated:109 […] the application of international law rules on interpretation of treaties to identical or similar provisions of different treaties may not yield the same results, having regard to, inter alia, differences in the respective contexts, objects and purposes.110

It further held: International law directs this Tribunal, first and foremost, to the text; here, the text and the drafters’ intentions, which it manifests, show that trade provisions were not to be transported to investment provisions. Accordingly, the Tribunal holds that Article 1102 is to be read on its own terms and not as if the words ‘any like, directly competitive or substitutable goods’ appeared in it.111

The ultimate analysis of the Tribunal is compatible with WTO-type considerations of consumer tastes and habits, and end uses in the determination of like products, though the Methanex Tribunal failed to provide an indication of what ‘like circumstances’ meant in Article 1102 in contrast to ‘like products’ in the GATT. The Tribunal observed that there were ‘comparators which are identical’ to Methanex and that ‘it would be […] perverse to ignore identical comparators if they were available and use the comparators that were less “like”’.112 Thus, the Tribunal noted that the ban ‘had precisely the same effect on the American investors and investments as it had on [Methanex]’.113 Referring to the Pope Tribunal’s take on ‘like circumstances’, it noted that the accurate comparators were those in the most ‘like circumstances’.114 The Tribunal was not deviating from the methodology of the WTO Appellate Body, it was rejecting the notion that the universe of ‘likeness’ should not be extended to all directly

107 ibid

para 6. Part IV Chapter B 7, para 15. 109 MOX Plant Case (Ireland v United Kingdom), Order, Request for Provisional Measures, ITLOS Case No 10, ICGJ 343 (ITLOS 2001), 3 December 2001, International Tribunal for the Law of the Sea [ITLOS]. 110 Methanex, above (n 105), Part II Chapter B 7, para 16. 111 ibid Part IV Chapter B 19, para 37. 112 ibid para 17. 113 ibid. 114 ibid para 19. 108 ibid

The Notorious First Wave of NAFTA Chapter 11 Awards 117 competitive or substitutable products. This is what the WTO Appellate Body itself found when faced with interpreting national treatment with respect to taxation measures, where the text (GATT III:2) refers not only to ‘like’ but also ‘directly competitive and substitutable products’: the universe of like products is a smaller sub-universe as it were of ‘directly competitive and substitutable products’.115 Ultimately, the Tribunal failed to find any violation of Article 1102.

III. CONCLUSION

The early NAFTA cases launched a debate about the investment provisions of the Agreement that had never occurred in earnest, particularly their applicability to developed countries with rule of law, proposals developed for altering the obligations in Chapter 11, and in other investment agreements that Canada and the US signed with other countries. Some of these changes would give arbitrators more of a treaty structure on which to base their assessment of public policy in the post-NAFTA period.116 The successor agreement to NAFTA, the US–Canada– Mexico Agreement (USMCA), goes considerably further in restricting the applicability of ISDS, excluding it other than in disputes between the US and Mexico, where an investment contract is in place, and only in the case of certain sectors, such as oil and gas.117 Canada, the defendant in most of the early ISDS cases, has opted out of ISDS entirely in the USMCA. ISDS is available only for direct expropriation and violations of MFN and national treatment.118 In the case of national treatment, there is a vindication of the approach of the arbitrators in the early NAFTA cases – given the explicit incorporation of public policy considerations included in the USMCA treaty text.119 When first wading into the waters of public policy, the early NAFTA tribunals defined the difficulty or challenge as being the lack of a GATT/WTO type of Article XX general exceptions clause. None of the amendments by the NAFTA Parties of their investment agreements has introduced such a clause. The treaty changes reflected in the USMCA seek to guide arbitrators discretio but they do not as such eliminate it. The question of standard of review is not addressed by these changes; instead, they appear to endorse an open-ended balancing by arbitrators, considering policyrelated factors alongside other kinds of considerations. From the perspective of regulatory chill, enormous ranges of discretion and indeterminacy remain. Carving out entire types of obligations from ISDS is a further step that can be seen in the USMCA. The lack of openness of negotiators to a GATT Article XX-type solution is all the more remarkable given that in the WTO the consideration of public policy through the lenses of Article XX has resulted in jurisprudence broadly accepted by the WTO Membership as legitimate and fair. Indeed, it has resulted in an articulation of a standard of review that lies somewhere between reasonableness and strict necessity, applied to disputes, to give two examples, that include highly sensitive controversies about environmental matters and animal welfare.120

115 Panel Report, Japan – Taxes on Alcoholic Beverages, WT/DS8/R, WT/DS10/R, WT/DS11/R, adopted 1 November 1996, as modified by Appellate Body Report WT/DS8/AB/R, WT/DS10/AB/R, WT/DS11/AB/R, DSR 1996. 116 See, eg: Annex B para 4 of the 2012 US Model BIT. 117 See, USMCA Annexes 14-D and 14-E. 118 Article 14-D-3 USMCA. 119 ibid Art 14.4:4. 120 See R Howse, ‘The World Trade Organization 20 Years On: Global Governance by Judiciary’ (2016) 27(1) European Journal of International Law 9.

8 The Argentinian Crisis Arbitrations AUGUST REINISCH AND JOHANNES TROPPER*

I. INTRODUCTION

T

HE SO-CALLED ‘ARGENTINIAN crisis arbitrations’ or ‘Argentina crisis cases’ are investment arbitration proceedings brought by various foreign investors, following the host state’s reaction to cope with its economic and financial crisis in 2000/2001. Before the renewable energy cases against Spain, Italy and the Czech Republic,1 the Argentina cases constituted the largest wave of investment arbitration claims directed against a single country with over 40 publicly known cases mostly brought under the International Centre for Settlement of Investment Disputes (ICSID) Convention, but also partly as United Nations Commission on International Trade Law (UNCITRAL) proceedings.2 These investment cases are an interesting subject for analysis due to their similar factual background. They have also been the cause of much controversy since tribunals had to review state measures that had been enacted to handle a serious economic crisis.3 Moreover, there was some inconsistency in arbitral case law leading to criticism of the investor-state dispute settlement system and its legitimacy.4 On jurisdiction, tribunals assessed inter alia whether general economic policies adopted in times of crisis were beyond their scope of jurisdiction. Another contentious jurisdictional issue emerged in the wake of the crisis: the first mass claim(s) in investment arbitration following Argentina’s default in sovereign debt and the subsequent restructuring of bonds. On the merits, the Argentina cases addressed, inter alia the question whether the challenged Argentine measures constituted violations of fair and equitable treatment or other investment protection standards. Moreover, many Argentinian cases dealt with possible justifications for measures adopted during the crisis under the so-called non-precluded measures or emergency clauses of

* August Reinisch is Professor of International and European Law at the University of Vienna; Johannes Tropper is Researcher and Assistant Lecturer at the University of Vienna. 1 See D Azaria, ‘The European ECT Litigation: Spanish, Italian and Czech Cases’ ch 10 in this volume. 2 For a brief overview of these cases, see I Torterola and DB Gosis, ‘Argentina’ in JC Hamilton, OE Garcia-Bolivar and H Otero (eds), Latin American Investment Protections. Comparative Perspectives on Laws, Treaties and Disputes for Investors, States and Counsel (Leiden, Martinus Nijhoff, 2012) 5, 20–43. 3 G van Harten, Investment Treaty Arbitration and Public Law (Oxford, Oxford University Press, 2007) 3, 89–90; M Sornarajah, The International Law on Foreign Investment, 4th edn (Cambridge, Cambridge University Press, 2017) 542, 551. 4 W Burke-White, ‘The Argentine Financial Crisis: State Liability under BITs and the Legitimacy of the ICSID System’ in M Waibel and others (eds), The Backlash against Investment Arbitration: Perceptions and Reality (Alphen aan den Rijn, Kluwer Law International, 2010) 407, 431–432; D Jones, ‘Investor-State Arbitration in Times of Crisis’ (2013) 25 National Law School of India Review 27, 57.

120 August Reinisch and Johannes Tropper the applicable investment treaties as well as under state of necessity as a circumstance precluding wrongfulness under customary international law. Finally, the cases raise the issue of the extent of the obligation to comply with arbitral awards rendered under the ICSID Convention and Argentina’s objections to the payment of ICSID awards. This chapter will address primarily but not exclusively four investment disputes which arose out of a series of decrees and laws adopted by Argentina in the course of the crisis, and which deserve particular attention because of their conflicting findings on the question of justification: CMS v Argentina, LG&E v Argentina, El Paso v Argentina and Continental Casualty v Argentina. In the context of the restructuring of bonds and mass claims by bondholders the chapter will focus on Abaclat v Argentina, the first ever decision to discuss collective claims in investment arbitration.5

II. BACKGROUND

During the 1980s Argentina was confronted with an economic crisis characterised by hyperinflation, a currency exchange crisis and recession. In the early 1990s, to overcome these economic problems, the Argentinian government undertook large-scale economic reforms to ensure currency stability and attract foreign investors and capital. The reform package included the privatisation of state-owned enterprises and the public utilities sector, in particular energy, gas and water distribution, through long-term concessions and licences. The goal of this reform was to import both investment and the know-how to operate infrastructure in the public utilities sector. To ensure currency stability and help induce foreign investors to enter the market, Argentina also enacted the ‘Convertibility Law’, which pegged the Argentine currency to the US dollar at the exchange rate of one peso to one dollar (‘dollarisation of the peso’). Due to provisions in concession contracts the tariffs in energy, gas and water distributions sectors were calculated in US dollars and for billing purposes expressed in pesos. Furthermore, tariffs were to be adjusted periodically by Argentinian regulatory agencies according to the US consumer and producer price indices. In the late 1990s a severe economic crisis erupted in Argentina, which seriously deteriorated in 2001. Economic chaos was triggered by various internal and external factors, including high levels of public debt and international monetary crises which had spilled over to Argentina. A dramatic increase in inflation, poverty and unemployment accompanied by social unrest, shortages of basic goods and services as well as abrupt changes in government ensued. Argentina struggled with interest and principal payments on its sovereign bond debt owed to foreign creditors and in 2001 was forced to default on its external debt. To alleviate these problems the Argentinian Government first responded by freezing bank deposits and prohibiting the transfer of currency abroad (‘Corralito Decree’). In 2002 the ‘Public Emergency Law’ followed, which abolished the fixed exchange rate with the US dollar and prescribed a conversion of all dollar

5 The Argentina crisis cases touched upon other issues such as counterclaims and the relationship between human rights and investment law. For an analysis of Urbaser SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v The Argentine Republic, Award (8 December 2016) ICSID Case NoARB/07/26 (Andreas Bucher, Pedro J Martínez-Fraga, Campbell McLachlan), which concerned a counterclaim against the investor partially based on the human right to water, see A Bjorklund, ‘Counterclaims: Counterbalancing Asymmetry?’ ch 17 in this volume. Additionally, SAUR International SA v Republic of Argentina, Decision on Jurisdiction and Liability (6 June 2012) ICSID Case No ARB/04/4, (Juan Fernández-Armesto, Bernard Hanotiau, Christian Tomuschat) discussed the relationship between the human right to water and investment law at some length. See further U Kriebaum, ‘The Right to Water Before Investment Tribunals’ (2012) 1 Brill Open Law 16.

The Argentinian Crisis Arbitrations 121 denominated financial instruments and contracts into pesos at a one-to-one rate (‘pesification’). In addition, tariff rates were no longer adjusted in accordance with the US indices. Due to a devaluation of the peso, the ‘pesification’ and the end of tariff adjustments, revenues in the public utilities sector markedly decreased.6 The ‘Public Emergency Law’ also mandated the renegotiation of long-term concessions and licences in the public utilities sector. These governmental measures triggered several investment disputes between Argentina and foreign investors operating in the public utilities sector.7 At the end of 2002 the economic situation gradually started to improve. In 2005 Argentina planned a restructuring of its sovereign bonds and made an exchange offer to bondholders, who had not received payments since 2001. The offer included issuing of new bonds with a much lower nominal value (25–30 per cent of the original value of the bonds). Seventy-six per cent of the affected bondholders accepted the offer, while holdout creditors demanding full repayment challenged the measure in domestic courts in the US, Germany and other countries as well as in investment arbitration proceedings.8

III. JURISDICTIONAL CHALLENGES: SOVEREIGN POWER TO REGULATE THE ECONOMY AND MASS CLAIMS

In all the Argentinian crisis cases the Argentinian Government mounted several objections to jurisdiction arguing, inter alia, that general economic measures emanate from the sovereign power to regulate its economic affairs and are thus not within the jurisdiction of investment tribunals. In the context of restructuring sovereign bonds, Argentina also argued that bonds did not qualify as investments and the ICSID procedural framework did not permit mass claims.

A. General Economic Policies Contemporary debates about investment arbitration often revolve around the delimitation of a state’s regulatory space.9 Major economic and financial crises can expose the inherent

6 In 2002, the peso devalued to a level of three pesos to one US dollar. Since the Public Emergency Law still required tariffs calculated in US dollars to be converted on a one-to-one basis into pesos, investors only received a third of the previous income. See P Di Rosa, ‘The Recent Wave of Arbitrations Against Argentina under Bilateral Investment Treaties: Background and Principal Legal Issues’ (2004) 36 University of Miami Inter-American Law Review 41, 47. 7 Foreign investors typically operated ‘indirectly’ in the public utilities sector through shareholding in or ownership of local subsidiaries. 8 The preceding summary of the factual background of the crisis and the debt restructuring was compiled from CMS Gas Transmission Company v The Republic of Argentina, Award (12 May 2005) ICSID Case No ARB/01/8 (Francisco Orrego Vicuña, Marc Lalonde, Francisco Rezek) paras 53–66; LG&E Energy Corp, LG&E Capital Corp, LG&E International Inc v The Argentine Republic, Decision on Liability (3 October 2006) ICSID Case No ARB/02/1 (Tatiana B de Maekelt, Francisco Rezek, Albert Jan van den Berg) paras 34–71; Continental Casualty Company v The Argentine Republic, Award (5 September 2008) ICSID Case No ARB/03/9 (Giorgio Sacerdoti, VV Veeder, Michell Nader) paras 100–128, 137–159; Enron Corporation Ponderosa Assets, LP v Argentine Republic, Award (22 May 2007) ICSID Case No ARB/01/3 (Francisco Orrego Vicuña, Albert Jan van den Berg, Pierre-Yves Tschanz) paras 41–46, 62–79; Abaclat and Others v Argentine Republic, Decision on Jurisdiction and Admissibility (4 August 2011) ICSID Case No ARB/07/5 (Pierre Tercier, Georges Abi-Saab, Albert Jan van den Berg) paras 43–97; El Paso Energy International Company v The Argentine Republic, Award (31 October 2011) ICSID Case No ARB/03/15 (Lucius Caflisch, Piero Bernardini, Brigitte Stern) paras 51–113. 9 See, eg, J Karl, ‘International Investment Arbitration: A Threat to State Sovereignty’ in W Shan, P Simons and D Singh (eds), Redefining Sovereignty in International Economic Law (Oxford, Hart Publishing, 2008) 225.

122 August Reinisch and Johannes Tropper tension between the exercise of regulatory powers and respect for the rights of foreign investors enshrined in international treaties. This leads to the question whether or under what circumstances administrative and legislative measures taken in times of crisis are beyond the power of review of investment tribunals. As a matter of principle, Argentina argued that general economic and currency policies adopted during a state-wide economic crisis cannot be challenged before arbitral tribunals and investors cannot receive compensation for adverse effects stemming from these measures because they ‘[were] not directed towards investors but affect[ed] the country and its population as a whole’.10 Tribunals acknowledged that ‘questions of general economic policy not directly related to the investment, as opposed to measures specifically addressed to the operations of the business concerned, will normally fall outside [ICSID jurisdiction]’.11 However, the Tribunals asserted that they were competent to examine whether specific measures affecting the Claimant’s investment or measures of general economic policy having a direct bearing on such investment have been adopted in violation of legally binding commitments made to the investor in treaties, legislation or contracts.12

The situation in the Argentinian cases was not atypical, but rather ‘many, if not most, [previous] disputes based on an alleged breach of international standards […] have arisen from general measures taken by host States, that affected directly those investments, without necessarily being specifically aimed at them’.13 Evidently, ‘measures, though “general” in appearance, may have highly specific effects’.14 Ultimately, the Tribunals held that it could not be a priori excluded that the design and implementation of general economic policies adopted in the course of a crisis contravened commitments in investment treaties – but this was a question for the merits and depended on the evidence presented at that stage.15

B. Mass Claims Holdout creditors rejecting the bond exchange offer by Argentina in 2005 initiated the first mass claim in ICSID history with approximately 180,000 Italian bondholders, subsequently reduced to 60,000 claimants16 – Abaclat v Argentina. Argentina objected to the Tribunal’s jurisdiction on various grounds, but the majority of the Tribunal rejected the Argentine

10 CMS Gas Transmission Company v The Republic of Argentina, Decision of the Tribunal on Objections to Jurisdiction (17 July 2003) ICSID Case No ARB/01/8 (Francisco Orrego Vicuña, Marc Lalonde, Francisco Rezek) para 30. 11 ibid para 27; see also ibid para 29 (‘Similarly, these treaties cannot entirely isolate foreign investments from the general economic situation of a country. They […] cannot prevent a country from pursuing its own economic choices. These choices are not under the Centre’s jurisdiction and ICSID tribunals cannot pass judgment on whether such policies are right or wrong. Judgment cannot [sic] only be made in respect of whether the rights of investors have been violated.’). 12 ibid para 33; see also AES Corporation v Argentine Republic, Decision on Jurisdiction (26 April 2005) ICSID Case No ARB/02/17 (Pierre-Marie Dupuy, Karl-Heinz Böckstiegel, Domingo Bello Janeiro) para 57. 13 Continental Casualty Company v The Argentine Republic, Decision on Jurisdiction (22 February 2006) ICSID Case No ARB/03/9 (Giorgio Sacerdoti, VV Veeder, Michell Nader) para 72. 14 El Paso Energy International Company v The Argentine Republic, Decision on Jurisdiction (27 April 2006) ICSID Case No ARB/03/15 (Lucius Caflisch, Brigitte Stern, Piero Bernardini) para 99. 15 Enron Corporation and Ponderosa Assets, LP v Argentine Republic, Decision on Jurisdiction (14 January 2004) ICSID Case No ARB/01/3 (Francisco Orrego Vicuña, Héctor Gros Espiell, Pierre-Yves Tschanz) paras 30–31; see also CMS, above (n 10) para 34. 16 Abaclat, above (n 8) paras 1, 613.

The Argentinian Crisis Arbitrations 123 position. First of all, it concluded that bonds qualified as investment both under the applicable BIT17 and under the ICSID Convention.18 In addition, as far as the mass nature of these claims and jurisdiction was concerned, the majority held that the mass aspect […] relates to the modalities and implementation of the ICSID proceedings and not to the question whether Respondent consented to ICSID arbitration. Therefore, it relates to the question of admissibility and not to the question of jurisdiction.19

Since the ICSID Convention is silent on mass claims, the Tribunal then had to analyse whether such claims would be admissible. The Tribunal queried whether this silence constituted ‘“qualified silence”, meaning an intended silence indicating that it does not allow for something that is not provided, or whether it is to be considered a gap, which was unintended and which the Tribunal has the power to fill’.20 The Tribunal determined that ‘it would be contrary to the purpose of the BIT and to the spirit of ICSID to interpret this silence as a “qualified silence”’21 and thus concluded that a gap existed, which the Tribunal had the power to fill by making the necessary adaptations to manage these mass proceedings.22 The dissenting arbitrator Abi-Saab, however, rejected the majority’s approach on various grounds, finding inter alia that a mere ‘consent to arbitrate’ does not cover the fundamentally different and atypical proceedings of collective mass claims actions, such as the present one (whether we qualify them as collective, class, representative, aggregate or otherwise), and that a ‘secondary consent’ is required for such proceedings to be covered.23

While Abaclat was endorsed by two subsequent cases against Argentina,24 neither of these cases were ultimately decided on the merits because they were settled or discontinued due to a lack of payment of the advances required by ICSID.25 However, the Abaclat decision on jurisdiction and admissibility essentially broadened the standing before investment tribunals to allow for mass investor claims or ‘“multi-party proceedings”’.26,27

17 ibid

paras 356–357. paras 366–367. 19 ibid para 492. 20 ibid para 517. 21 ibid para 519. 22 ibid paras 517, 520, 551. 23 Abaclat and others v Argentine Republic, ICSID Case No ARB/07/5, Dissenting Opinion, Georges Abi-Saab, 28 October 2011, para 177. 24 Ambiente Ufficio SPA and others v Argentine Republic, Decision on Jurisdiction and Admissibility (8 February 2013) ICSID Case No ARB/08/9 (Bruno Simma, Karl-Heinz Böckstiegel, Santiago Torres Bernárdez); Giovanni Alemanni and others v Argentine Republic, Decision on Jurisdiction and Admissibility (17 November 2014) ICSID Case No ARB/07/8 (Franklin Berman, Karl-Heinz Böckstiegel, Christopher Thomas). 25 LE Peterson, ‘Third (and Largest) of Italian Bondholder Claims Against Argentina is Settled; Cases Blazed a Trail, but Arbitrators Continue to Disagree as to Scope for ICSID to be Used for Sovereign Debt Disputes’ (IAReporter, 2 February 2016), www.iareporter.com/articles/third-and-largest-of-italian-bondholder-claims-against-argentina-issettled/; LE Peterson, ‘As Another Argentine Bond Arbitration Collapses at ICSID, Arbitrators Doubt Their Power to Award Costs in These Circumstances’ (IAReporter, 14 December 2015), www.iareporter.com/articles/as-anotherargentine-bond-arbitration-collapses-at-icsid-arbitrators-doubt-their-power-to-award-costs-in-these-circumstances/; C Trevino, ‘One of Three ICSID Argentine Bond Arbitrations Collapses Due to Lack of Funding’ (IAReporter, 2 June 2015), www.iareporter.com/articles/one-of-three-icsid-argentine-bond-arbitrations-founders-due-to-lack-of-funding/. 26 Ambiente, above (n 24) para 122. 27 For details on these mass claims and the restructuring of Argentinian bonds see A Carlevaris and R Digón, ‘The Argentinian Bonds Saga: An International Investment Law Perspective’ in A Tanzi and others (eds), International Investment Law in Latin America/ Derecho Internacional de las Inversiones en América Latina (Leiden, Brill Nijhof, 2016) 605. 18 ibid

124 August Reinisch and Johannes Tropper IV. THE BREACHES OF BILATERAL INVESTMENT TREATIES

Investors usually claimed a whole range of BIT violations stemming from the above-described state measures, inter alia, violations of the expropriation standard, fair and equitable treatment and umbrella clauses.

A. (Indirect) Expropriation Most investors claimed that their investment had been indirectly expropriated but were unsuccessful because no permanent, substantial deprivation of the investments had occurred according to the Tribunals.28 The Tribunal in LG&E emphasised that ‘although the State adopted severe measures that had a certain impact on Claimants’ investment, especially regarding the earnings that the Claimants expected, such measures did not deprive the investors of the right to enjoy their investment.’29 Since the investors had not ‘lost control over their shares in the licensees’ and were still able ‘to direct the day-to-day operations of the licensees’,30 the impact of the Argentine measures was not ‘permanent on the value of the Claimants’ shares, and Claimants’ investment [had] not ceased to exist.’31 Thus, the requirements for expropriation were not fulfilled.

B. Fair and Equitable Treatment Practically all the Argentinian cases addressed potential violations of fair and equitable treatment (FET). Tribunals elaborated in particular that this standard required the stability and predictability of the legal framework and that the changes to the tariff regime and the fixed currency rate fundamentally transformed this legal framework, thereby violating FET.32 A further factor, normally linked to stability and predictability, were the specific contractual guarantees or promises of stability giving rise to legitimate expectations, which were frustrated due to the modifications of the existing normative framework.33 The Tribunal in El Paso considered the concept of legitimate expectations, ‘which derive[d] from the obligation of good faith’ as the ‘touchstone’ of FET.34 The Tribunal clarified the meaning of legitimate expectations by stressing ‘that legitimate expectations cannot be solely the subjective expectations of the investor, but have to correspond to the objective expectations [that] can be deduced from the circumstances and with due regard to the rights of the State’.35 For legitimate expectations to arise, a ‘specific commitment directly made to the investor, on which the latter has relied’ has to exist.36 This specific commitment can be provided ‘in a contract or in a letter of intent, or even through a specific promise in a person-to-person business meeting – and not

28 CMS, above (n 8) paras 263–264; El Paso, above (n 8) paras 269, 278; on expropriation see D Desierto, ‘Expropriation Cases’ ch 21 in this volume. 29 LG&E, above (n 8) para 198. 30 ibid para 199. 31 ibid para 200. 32 CMS, above (n 8) paras 274–277; LG&E, above (n 8) para 125. 33 CMS, above (n 8) para 275; LG&E, above (n 8) para 127; El Paso, above (n 8) paras 364, 374, 379–379, 510–519. 34 El Paso, above (n 8) para 348; 35 ibid para 358. 36 ibid para 376.

The Argentinian Crisis Arbitrations 125 simply general statements in treaties or legislation which, because of their nature of general regulations, can evolve’.37 However, [t]here can be no legitimate expectation for anyone that the legal framework will remain unchanged in the face of an extremely severe economic crisis. No reasonable investor can have such an expectation unless very specific commitments have been made towards it or unless the alteration of the legal framework is total.38

LG&E, on the other hand, had pointed out that Argentina had created ‘an attractive framework of laws and regulations that addressed the specific concerns of foreign investors’, such as the legal guarantee of tariff adjustments, and had through these laws ‘created specific expectations among investors’.39 In almost all cases violations of FET were found, but the content of FET and the question which state acts – contracts, legislation, unilateral statements or promises – could trigger legitimate expectations were subject to debate by tribunals.40

C. Umbrella Clauses In addition, various tribunals found breaches of umbrella clauses of the respective BITs, which typically elevate a commitment undertaken by a state vis-à-vis an investor to the treaty level, but reached different conclusions on the controversial issue whether these provisions extended to unilateral commitments of a host state in the form of domestic legislation or whether they only covered contractual undertakings.41 While the LG&E tribunal expressly found that rights granted to investors under the Argentinian Gas Law and its implementing regulations contained ‘obligations’ in the sense of the applicable umbrella clause,42 which provided that ‘[e]ach party shall observe any obligation it may have entered into with regard to investments’43 the annulment committee in CMS was clear that umbrella clause language referring to obligations ‘entered into’ excluded general legislation.44

V. JUSTIFICATIONS

Whereas almost all tribunals in the Argentina crisis cases found violations of substantive standards of investment treaties, they occasionally sharply disagreed whether such violations could be ‘justified’ and how such justifications could be legally construed. Some tribunals

37 ibid. 38 ibid

para 374. above (n 8) para 133. 40 On FET and legitimate expectations see S Schill, ‘Fair and Equitable Treatment’ ch 22 in this volume. 41 CMS, above (n 8) para 303; LG&E, above (n 8) para 175; CMS Gas Transmission Company v The Argentine Republic, Decision of the ad hoc Committee on the Application for Annulment of the Argentine Republic (25 September 2007) ICSID Case No ARB/01/8 (Gilbert Guillaume, Nabil Elaraby, James Crawford) para 95; Continental, above (n 8) paras 297, 300–303. On the meaning and scope of umbrella clauses, see B Samson, ‘Umbrella Clauses and Contract Claims’ ch 23 in this volume. 42 LG&E, above (n 8) para 175. 43 Article II(2)(c) Argentina-US BIT (1991). 44 CMS, ad hoc Committee, above (n 41) para 95 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 (ie 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.’). 39 LG&E,

126 August Reinisch and Johannes Tropper relied on the emergency exception of the applicable BIT (frequently, Article XI Argentina-US BIT), others on the customary international law rules of necessity. In general, the outcomes were partially inconsistent on the assessment whether the facts gave rise to a state of necessity under customary international law and treaty law, respectively.45

A. Was There a State of Necessity or Emergency Situation? Notwithstanding that state of necessity has long been recognised as a justification, or rather a ground precluding wrongfulness in state responsibility law, there has been little judicial practice on this defence. The Argentina cases have changed this dramatically. In the first Argentina crisis case dealing with the question of the state of necessity – CMS – the Tribunal applied the very restrictive conditions of the necessity defence under customary law codified in Article 25 of the International Law Commission (ILC) Articles on State Responsibility (ARSIWA).46 In doing so, the Tribunal essentially conflated the emergency clause of the Argentina-US BIT with the customary norm on necessity, without explicitly addressing the relationship between these norms.47 In its analysis of the facts, the Tribunal stressed that [t]he need to prevent a major breakdown, with all its social and political implications, might have entailed an essential interest of the State in which case the operation of the state of necessity might have been triggered.48

However, the Tribunal found that Argentina’s measures to tackle the crisis were not the only steps available, citing the ILC commentary that the necessity defence is ‘excluded if there are other (otherwise) lawful means available, even if they may be more costly or less convenient’.49 Furthermore, the Tribunal concluded that ‘government policies and their shortcomings significantly contributed to the crisis and the emergency and while exogenous factors did fuel additional difficulties they do not exempt the Respondent from its responsibility in the matter.’50 As the conditions for invoking a plea of necessity need to be cumulatively satisfied,51

45 See generally A Reinisch, ‘Necessity in International Investment Arbitration – An Unnecessary Split of Opinions in Recent ICSID Cases – Comments on CMS v Argentina and LG&E v Argentina’ (2007) 8 Journal of World Investment & Trade 191; SW Schill, ‘International Investment Law and the Host State’s Power to Handle Economic Crises – Comment on the ICSID Decision in LG&E v Argentina’ (2007) 24 Journal of International Arbitration 265, 277–284; M Waibel, ‘Two Worlds of Necessity in ICSID Arbitration: CMS and LG&E’ (2007) 20 Leiden Journal of International Law 637; L Mola, ‘International Investment Arbitration and Serious Economic Crises: Lessons Learned in the Argentinean Crisis of 2000–2001’ in A Tanzi and others (eds), International Investment Law in Latin America/ Derecho Internacional de las Inversiones en América Latina (Leiden, Brill Nijhof, 2016) 370. 46 Article 25 ARSIWA (‘Articles on Responsibility of State for Internationally Wrongful Acts’, UNGA Res 56/83 (12 December 2001) UN-Doc A/56/10) (‘1. Necessity may not be invoked by a State as a ground for precluding the wrongfulness of an act not in conformity with an international obligation of that State unless the act: (a) Is the only way for the State to safeguard an essential interest against a grave and imminent peril; and (b) Does not seriously impair an essential interest of the State or States towards which the obligation exists, or of the international community as a whole. 2. In any case, necessity may not be invoked by a State as a ground for precluding wrongfulness if: (a) The international obligation in question excludes the possibility of invoking necessity; or (b) The State has contributed to the situation of necessity.’ (emphasis added)). 47 CMS, above (n 8) paras 353–358, 374; see below V.B. 48 CMS, above (n 8) para 319. 49 ibid para 324. With regard to the existence or lack of existence of alternative economic measures presented by the parties, the Tribunal asserted that it is beyond its tasks to identify ‘[w]hich of these policy alternatives would have been better’ (ibid para 323). 50 ibid para 329. 51 ibid paras 330–331.

The Argentinian Crisis Arbitrations 127 the Tribunal concluded that the plea of necessity under customary international law did not justify Argentina’s measures.52 Relying on the emergency clause of the Argentina-US BIT,53 the Tribunal in El Paso also concluded that Argentina’s breaches of the BIT could not be justified. Holding that ‘protection offered by the BIT to the Claimant’s investment [would be] suspended’ if the conditions of the emergency clause were fulfilled, the Tribunal examined the causes of the economic crisis.54 It pointed out that while ‘[t]he host State is generally not responsible for the consequences of a state of emergency’, it will be ‘if it has significantly contributed to that situation’.55 The majority of the Tribunal reached the conclusion that Argentina had substantially contributed to the crisis by ‘[failing] to control several internal factors, in particular the fiscal deficit debt accumulation and labour market rigidity’ and therefore could not successfully invoke the emergency clause.56 Contrary to the view that the economic crisis did not constitute a state of necessity, the Tribunal in LG&E relying on the emergency clause of the Argentina-US BIT rendered a decision finding that ‘from 1 December 2001 until 26 April 2003, Argentina was in a period of crisis during which it was necessary to enact measures to maintain public order and protect its essential security interests’.57 Describing the economic downturn and the corollary social as well as political turmoil in detail,58 it reasoned that this affected an ‘essential security interest’ in accordance with the provision of the BIT.59 The Tribunal emphasised that ‘[a] State may have several responses at its disposal to maintain public order or protect its essential security interests’ considering the Argentine response as ‘a legitimate way of protecting its social and economic system’, hence a measure necessary to protect the essential security interests.60 In the same vein the Tribunal in Continental Casualty held that the economic crisis concerned an essential security interest since the ‘the protection of essential security interests recognized by Art. XI does not require that a “total collapse” of the country or that a “catastrophic situation” has already occurred before responsible national authorities may have recourse to its protection’.61 In order to determine whether the measures would qualify as necessary to protect Argentina’s essential security interests, the Tribunal innovatively interpreted the term ‘necessary’ by reference to WTO case law.62 Accordingly, the contested measures must have 52 The Tribunals in Enron and Sempra essentially followed a very similar approach and came to the same conclusions: Enron, above (n 8) paras 303–313; Sempra Energy International v The Argentine Republic (28 September 2007) ICSID Case No ARB/02/16 (Francisco Orrego Vicuña, Marc Lalonde, Sandra Morelli Rico) paras 344–355. 53 Article XI Argentina-US BIT (1991) (‘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.’ (emphasis added)). 54 El Paso, above (n 8) para 649. 55 ibid para 615. 56 ibid para 656; Arbitrator Stern disagreed on the question whether Argentina had contributed to the crisis arguing that ‘[i]t should not lightly be assumed that a State is responsible for an economic collapse in a liberal market economy, where the invisible hand of the market is more powerful than the hand of the State’. (ibid para 667). 57 LG&E, above (n 8) para 226. 58 ibid paras 230–237. 59 ibid para 238 (‘To conclude that such a severe economic crisis could not constitute an essential security interest is to diminish the havoc that the economy can wreak on the lives of an entire population and the ability of the Government to lead. When a State’s economic foundation is under siege, the severity of the problem can equal that of any military invasion.’). 60 ibid paras 239, 242. The Tribunal also stressed that ‘[a]lthough there may have been a number of ways to draft the economic recovery plan, the evidence before the Tribunal demonstrates that an across-the-board response was necessary, and the tariffs on public utilities had to be addressed’ (ibid para 257). 61 Continental, above (n 8) para 180. 62 ibid para 192 (‘Since the text of Art. XI derives from the parallel model clause of the U.S. FCN treaties and these treaties in turn reflect the formulation of Art. XX of GATT 1947, the Tribunal finds it more appropriate to refer to the

128 August Reinisch and Johannes Tropper ‘contributed materially to the realization of their legitimate aims under Art. XI’,63 which the Tribunal did not doubt, and any alternatives to these measures must not have been reasonably available or must have been ‘impracticable or speculative as to their effects’.64 Again, the Tribunal generally confirmed that there were no reasonable or practical alternative policies prior to or during the crisis,65 holding that at the time when these measures were introduced Argentina did not have a choice ‘to resort generally to alternatives that would have been economically and legally less disruptive’.66 Thus, under the more lenient test of the BIT’s emergency clause Argentina’s measures ‘conformed, by and large, with the conditions required for derogating from its obligations under Art. XI of the BIT’.67 These conflicting legal assessments of the actual situation prevailing in Argentina during the same critical period demonstrated to critics the lack of predictability of investment arbitration.68 However, the underlying economic assessment of the problems in Argentina was submitted as evidence by economic experts, limiting the Tribunals’ task to assessing the value of the evidence and clarifying legal questions of interpretation and application of the respective BIT and customary international law. Against the backdrop of conflicting evidence presented by economic experts during the proceedings,69 the legal assessment of complex economic issues also depended on the strength or weakness of the evidence submitted by the parties in each individual case,70 leaving the possibility of divergent conclusions in different arbitral proceedings. Irrespective of this, the differing answers to purely legal questions, such as the relationship between necessity under customary international and the emergency clause of the Argentina-US BIT,71 is hardly reconcilable with a desire for consistency and predictability.72 Nonetheless, in the absence of a system of binding precedent,73 the risk of incoherent decisions is almost inevitable. A permanent court or tribunal hearing appeals could help guarantee more consistency in case law, even without the existence of a strict rule of binding precedent. Such a permanent court might not be conducive to achieving coherence for individual disputes based on different investment treaties (and rather cause new challenges

GATT and WTO case law which has extensively dealt with the concept and requirements of necessity in the context of economic measures derogating to the obligations contained in GATT, rather than to refer to the requirement of necessity under customary international law.’ (footnote omitted)). 63 ibid para 196. 64 ibid para 198. 65 ibid paras 200–230. 66 ibid para 231. 67 ibid para 233. 68 SD Franck, ‘The Nature and Enforcement of Investor Rights Under Investment Treaties: Do Investment Treaties Have A Bright Future’ (2005) 12 UC Davis Journal of International Law & Policy 47, 63; JE Alvarez and G Topalian, ‘The Paradoxical Argentina Cases’ (2012) 6 World Arbitration & Mediation Review 491, 491. 69 Enron, above (n 8) paras 297; 300–301; 308; Sempra, above (n 52) paras 336, 340, 342; Continental, above (n 8) para 224; El Paso, above (n 8) paras 654–655, 668; CMS, above (n 8) para 323. 70 El Paso, above (n 8) para 650. (‘[The tribunal] will rely on the evidence made available by the Parties in these proceedings, with very limited consideration of prior decisions in other cases involving Argentina since the arguments and evidence placed before each tribunal are not the same in every case.’); see also J Paulsson, ‘The Role of Precedent in Investment Treaty Arbitration’ in K Yannaca-Small (ed), Arbitration under International Investment Agreements: A Guide to Key Issues, 2nd edn (Oxford, Oxford University Press, 2018) 4.19. 71 See below VI.B. 72 G Kaufmann-Kohler, ‘Arbitral Precedent: Dream, Necessity or Excuse?: The 2006 Freshfields Lecture’ (2007) 23 Arbitration International 357, 376. 73 AES Corporation, above (n 12) para 23(d) (‘There is so far no rule of precedent in general international law; nor is there any within the specific ICSID system […].’); however, the Tribunal also held in para 27 that it ‘would nevertheless reject the excessive assertion […] that […] absolutely no consideration might be given to other decisions on jurisdiction or awards delivered by other tribunals in similar cases.’).

The Argentinian Crisis Arbitrations 129 if it does lead to coherence or a system of precedent74). Nonetheless, the large number of Argentina cases based on virtually identical facts and the same BITs,75 revealed a structural weakness of the existing investor-state dispute settlement mechanism and the inconsistencies were partially responsible for triggering a backlash against investment arbitration.76

B. The Relationship between the Emergency Clause and the Necessity Defence under Customary International Law In particular, the relationship between necessity and the emergency clause of the Argentina-US BIT was approached in an inconsistent manner by several tribunals. In CMS the Tribunal devoted much of its attention to the customary international law defence of necessity and neglected to elaborate on the relationship between necessity and the emergency clause of the Argentina-US BIT. In the subsequent CMS annulment proceedings, however, the ad hoc committee held that the CMS tribunal had committed ‘manifest error[s] of law’.77 It emphasised that despite some similarities, Article 25 ARSIWA and the emergency clause were ‘substantively different’ since the requirements for the invocation were not identical.78 Also the two norms operated in a different manner: the provision in the BIT constituted a ‘threshold requirement’, which implied that if its conditions were fulfilled, the substantive obligations of the BIT would become inapplicable.79 Article 25 ARSIWA, on the other hand, excludes the wrongfulness of a breach of an international obligation, in this specific case the obligations under a BIT.80 Thus, the ad hoc committee found that the CMS tribunal had committed an error of law by failing to first consider whether there had been a breach of the BIT and whether such a breach was excluded by the emergency clause. It stated that ‘[o]nly if [the tribunal had] concluded that there was conduct not in conformity with the Treaty would it have had to consider whether Argentina’s responsibility could be precluded in whole or in part under customary international law’.81 The Tribunal, however, had not engaged in such an analysis because it had merely assumed that the BIT provision and Article 25 ARSIWA were ‘on the same footing’.82 Nevertheless, since the ad hoc committee concluded that this mistake solely amounted to a ‘manifest error of law’ and not to a ‘manifest excess of power’ pursuant to Article 52(1)(b) ICSID Convention, the committee refrained from annulling the award in this respect.83 Due to its limited mandate under Article 52 ICSID Convention, it averred that it could not act ‘as a court of appeal’ and thus was in no position to ‘simply substitute its own view of the law and its own appreciation of the facts for those of the Tribunal’.84

74 cf AA Yusuf and G Yusuf, ‘Precedent & Jurisprudence Constante’ in M Kinnear, GR Fischer, J Mínguez Almedia, LF Torres and M Uran Bidegain (eds), Building International Investment Law – The First 50 Years of ICSID (Alphen aan den Rijn, Kluwer Law International, 2016) 71, 80. 75 eg 14 ‘crisis cases’ and another seven non-crisis cases were brought under the Argentina-US BIT. 76 M Waibel and others, The Backlash against Investment Arbitration: Perceptions and Reality (Alphen aan den Rijn, Kluwer Law International, 2010) xxvi. 77 CMS, ad hoc Committee, above (n 41) paras 130, 132. 78 ibid para 130. 79 ibid para 129. 80 ibid paras 129, 134. 81 ibid para 134. 82 ibid para 131. 83 ibid para 136. 84 ibid paras 135–136.

130 August Reinisch and Johannes Tropper The Tribunal in Sempra had also failed to distinguish between Article 25 ARSIWA and Article XI Argentina-US BIT, explicitly arguing that the latter ‘provision is inseparable from the customary law standard insofar as the definition of necessity and the conditions for its operation are concerned, given that it is under customary law that such elements have been defined’.85 The ad hoc committee in the Sempra annulment proceedings went further than the ad hoc committee in CMS and proclaimed that the Tribunal had manifestly exceeded its power by equating Article 25 ARSIWA with the emergency clause and essentially applying it as ‘primary law’ instead of the Article XI of the BIT.86 These conclusions by the ad hoc committees prompted tribunals in subsequent proceedings such as El Paso and Continental Casualty to examine the interrelationship between the two norms more carefully and to clearly distinguish between them in accordance with the findings of the annulment committees.87 Against this backdrop, it could be argued that even the current investor-state dispute settlement system can ensure consistency for purely legal questions. When arbitral tribunals were confronted with the first Argentina crisis cases, many controversial legal issues had not been addressed by investment tribunals.88 Thus, arguably the applicable rules of international investment law had not been sufficiently developed at that time. For a more uniform approach to evolve, a substantial number of cases needed to be brought before arbitral tribunals, which occurred in the Argentina crisis cases.

C. Legal Consequences: Damages and Necessity Several tribunals touched upon the question of whether a successful reliance on necessity or the emergency clause had any bearing upon the issue of damages. As far as a successful invocation of the emergency clause was concerned, it became readily accepted that the consequence would be the inapplicability of some BIT standards and thus no damages could be awarded during the state of emergency;89 or that the emergency clause excluded the wrongfulness of the act and thereby exempted Argentina from liability.90 However, as soon as the emergency period ended and if the contested measures were still in force, the obligation to pay damages would arise again.91 Under customary international law, necessity also constitutes a temporary defence, implying that after the emergency period ends, a state must comply with the respective international obligation.92 Moreover, as suggested by CMS in reliance on Article 27 ARSIWA93 ‘the plea of

85 Sempra,

above (n 52) para 376. Similarly, Enron, above (n 8) para 334. Energy International v The Argentine Republic, Decision on the Argentine Republic’s Application for Annulment of the Award (29 June 2010) ICSID Case No ARB/02/16, (Christer Söderlund, David AO Edward, Andreas J Jacovides) paras 207–209; see also Enron Corporation Ponderosa Assets, LP v Argentine Republic, Decision on the Application for Annulment of the Argentine Republic (30 July 2010) ICSID Case No ARB/01/3 (Gavan Griffith, Patrick L Robinson, Per Tresselt) paras 395, 404–405. 87 Continental, above (n 8) paras 162, 164–167; El Paso, above (n 8) paras 552–554. 88 As of 30 June 2019, ICSID had registered 716 cases under the ICSID Convention or ICSID Additional Facility Rules, but when the first ICSID cases in the Argentina crisis cases were brought under the ICSID Convention in 2001 a mere 78 contract and treaty arbitration cases had been registered with ICSID (figures compiled from: (2019) 2 The ICSID Caseload – Statistics 8, https://perma.cc/XF4C-DHEH. 89 El Paso, above (n 8) para 612. 90 LG&E, above (n 8) para 261. 91 El Paso, above (n 8) para 612; LG&E, above (n 8) paras 261, 263. 92 CMS, above (n 8) paras 380, 382. 93 Article 27 ARSIWA (‘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 86 Sempra

The Argentinian Crisis Arbitrations 131 state of necessity may preclude the wrongfulness of an act, but it does not exclude the duty to compensate’.94 Thus, regardless of the existence of necessity as a circumstance precluding wrongfulness, compensation for measures adopted in times of emergency might be inevitable as a matter of law.95 However, LG&E viewed such an approach more critically emphasising that Article 27 ‘does not specify if compensation is payable during the state of necessity or whether the State should reassume its obligations’.96 Apart from this issue, tribunals that had found that no state of necessity existed, acknowledged that the dire economic situation had to be taken into account at the valuation stage of the proceedings.97 Since the investor’s income would have been negatively affected by the crisis regardless of the response by the Argentinian government, the amount of damages awarded had to reflect this consideration.98

VI. COMPLIANCE

A further common trait in the Argentina crisis cases is the lack of compliance with the awards by Argentina. Despite the fact that most awards were rendered under the ICSID Convention, which had generally been regarded as providing for a strict as well as effective compliance and enforcement mechanism under public international law,99 Argentina contended that it was not automatically obliged to pay ICSID awards.100 Rather, according to Argentina, successful claimants must always, ‘under Article 54 of the ICSID Convention, […] follow the domestic proceedings to enforce an award’.101 Furthermore, on the basis of the so-called ‘Rosatti Doctrine’, named after Argentina’s former Procurador del Tesoro de la Nación and subsequent Minister of Justice, Argentine officials argued that awards rendered by ICSID tribunals should pass a ‘constitutionality test’ to prevent violations of Argentinian constitutional principles if recognition and enforcement was sought in Argentina.102 precluding wrongfulness no longer exists; b) The question of compensation for any material loss caused by the act in question.’). 94 CMS, above (n 8) para 388. According to CMS the amount of compensation should principally be decided by the parties to the dispute and only in the absence of an agreement shall a tribunal rule on the matter (paras 393f); see also Enron, above (n 8) para 345; Sempra, above (n 52) para 394. 95 AO Sykes, ‘Economic Necessity in International Law’ (2015) 109 American Journal of International Law 296, 310; C Binder and P Janig, ‘Investment agreements and financial crises’ in M Krajewski and RT Hoffmann (eds), Research Handbook on Foreign Direct Investment (Cheltenham, Edward Elgar Publishing, 2019) 646, 678. 96 LG&E, above (n 8) para 260. 97 Sempra, above (n 52) para 397. 98 For more details see I Marboe, ‘Reparation Cases’ ch 26 in this volume. 99 See Art 53(1) ICSID Convention (Convention on the Settlement of Investment Disputes between States and Nationals of other States (1965) 575 UNTS 159) (‘The award shall be binding on the parties and shall not be subject to any appeal or to any other remedy except those provided for in this Convention. Each party shall abide by and comply with the terms of the award except to the extent that enforcement shall have been stayed pursuant to the relevant provisions of this Convention.’); Art 54(1) ICSID Convention (‘Each Contracting State shall recognize an award rendered pursuant to this Convention as binding and enforce the pecuniary obligations imposed by that award within its territories as if it were a final judgment of a court in that State.’); A Broches, ‘Awards Rendered Pursuant to the ICSID Convention: Binding Force, Finality, Recognition, Enforcement, Execution’ (1987) 2 ICSID Review – FILJ 287. 100 Enron Corporation and Ponderosa Assets, LP v Argentine Republic, Decision on the Argentine Republic’s Request for a Continued Stay of Enforcement of the Award (7 October 2008) ICSID Case No ARB/01/3 (Gavan Griffith, Patrick L Robinson, Per Tresselt) para 56. 101 Total SA v The Argentine Republic, Decision on Stay of Enforcement of the Award (4 December 2014) ICSID Case No ARB/04/01 (Teresa Cheng, Álvaro Castellanos, Eduardo Zuleta J) para 32. 102 CMS Gas Transmission Company v The Republic of Argentina, Decision on the Argentine Republic’s Request for a Continued Stay of Enforcement of the Award (1 September 2006) ICSID Case No ARB/01/8 (Gilbert Guillaume, James Crawford, Nabil Elaraby) paras 46–47; H Rosatti, ‘Los Tratados bilaterales de Inversión, el arbitraje internacional obligatorio y el sistema constitucional argentine’ (15 October 2003) La Ley.

132 August Reinisch and Johannes Tropper While this approach might have been explicable on the basis of Argentinian constitutional law, it contravened Argentina’s obligations under the ICSID Convention,103 which ‘does not offer any scope for review on the national plane’.104 Moreover, the annulment Committees in the Argentinian cases clarified that the ICSID Convention ‘does not state that a party to an award must use the enforcement machinery established pursuant to this provision as a condition of the award being complied with’.105 The duty to comply with awards delivered by ICSID tribunals is automatic and actual enforcement proceedings pursuant to Article 54 ICSID Convention are only required insofar as a losing state violates the obligation under Article 53 ICSID Convention to ‘abide by and comply with the terms of the award’.106 If a prevailing claimant had to initiate enforcement proceedings under the domestic law of the respondent state ‘as a prerequisite for compliance with the award by the award debtor’, ‘confidence in the ICSID system’ would be ‘inherently undermin[ed]’.107 Despite these rulings Argentina retained a recalcitrant position and refused to comply with awards in the absence of enforcement proceedings in its domestic courts. External pressure, in particular blocking of World Bank loans by the US and other member states of the International Bank for Reconstruction and Development as well as the suspension of Argentina’s preferential trade status under the US’s Generalized System of Preferences legislation, was instrumental in inducing Argentina to honour the awards.108 However, Argentina’s compliance involved settlement agreements with the investors typically leading to haircuts of 25–31 per cent of the original sums awarded.109 Settlement of several of these outstanding awards was essentially motivated by the intention to regain trust from the international community as well as financial institutions such as the International Monetary Fund in order to access international financial markets.110 The experience with Argentina’s uncompromising stance revealed that in spite of the uniform interpretations of the compliance obligation under the ICSID Convention, observance of these commitments by states is not guaranteed and diplomatic protection111

103 Article 53(1) ICSID Convention (‘The award […] shall not be subject to any appeal or to any other remedy except those provided for in this Convention. […]’); OJ Marzorati, ‘Enforcement of Treaty Awards and National Constitutions (The Argentinian Cases)’ (2006) 7 Business Law International 226, 237–239. 104 Sempra Energy International v The Argentine Republic, Decision on the Argentine Republic’s Request for a Continued Stay of Enforcement of the Award (5 March 2009) ICSID Case No ARB/02/16 (Christer Söderlund, David AO Edward, Andreas J Jacovides) para 41 (emphasis in original); see also Total, above (n 101) para 96. 105 Enron, above (n 100) para 62; see also Sempra, above (n 104) para 52. For a detailed discussion of the findings in the Argentinian cases, see SA Alexandrov, ‘Enforcement of ICSID Awards: Article 53 and 54 of the ICSID Convention’ in C Binder and others (eds), International Investment Law for the 21st Century: Essays in Honour of Christoph Schreuer (Oxford, Oxford University Press 2009) 322, 330–335. 106 Alexandrov, above (n 105) 324; C Schreuer and others, The ICSID Convention: A Commentary, 2nd edn (Cambridge, Cambridge University Press, 2009) 1106. 107 Enron, above (n 100) para 68. 108 M Hirsch, ‘Explaining Compliance and Non-Compliance with ICSID Awards: The Argentine Case Study and a Multiple Theoretical Approach’ (2016) 19 Journal of International Economic Law 681, 699–702; CB Rosenberg, ‘The Intersection of International Trade and International Arbitration: The Use of Trade Benefits to Secure Compliance with Arbitral Awards’ (2013) 44 Georgetown Journal of International Law 503, 517. 109 IAReporter, ‘Argentina Announces Another Settlement of Unpaid BIT Awards, Once Again at a Discount’ (IAReporter, 15 May 2016), www.iareporter.com/articles/argentina-announces-another-settlement-of-unpaid-bit-awards-onceagain-at-a-discount/; D Charlotin, ‘Argentina Settles More Arbitral Awards with Foreign Investors’ (IAReporter, 12 January 2018), www.iareporter.com/articles/argentina-settles-more-arbitral-awards-with-foreign-investors/. 110 Hirsch, above (n 108) 700; AC Porzecanski, ‘The Origins of Argentina’s Litigation and Arbitration Saga, 2002–2016’ (2016) 40 Fordham International Law Journal 41, 42. 111 Article 27(1) ICSID Convention (‘No Contracting State shall give diplomatic protection, or bring an international claim, in respect of a dispute which one of its nationals and another Contracting State shall have consented to submit or shall have submitted to arbitration under this Convention, unless such other Contracting State shall have failed to abide by and comply with the award rendered in such dispute.’ (emphasis added)).

The Argentinian Crisis Arbitrations 133 or alternative measures might occasionally play a very prominent role in the otherwise depoliticised investor-State dispute settlement system.112

VII. CONCLUSION

The breadth of the legal issues covered by the Argentinian crisis arbitrations permits the conclusion that ‘legal history was made’113 in these investment disputes. Due to the particular situation involving a major economic and financial crisis, specific legal guarantees to investors in contracts as well as legislation and the relatively limited number of investment arbitration awards rendered prior to the Argentina crisis cases, tribunals had to venture into unchartered legal territory. A plethora of jurisdictional issues and questions of substantive law could be discussed in the context of the Argentina crisis cases, but this would go beyond the scope of this chapter. However, the legal aspects addressed here already reveal that the crisis cases made a tangible contribution to investment law and investment arbitration. The Argentina crisis cases also partially contributed to the backlash against investment arbitration. A common charge was the dearth of coherence in the Tribunals’ reasoning, particularly on the questions of necessity under customary international law and emergency clauses under treaty law. Incoherence leads to unpredictability and de-legitimises investment arbitration. However, as elaborated above, in the de-centralised system of investor-state dispute settlement inconsistent outcomes are sometimes inevitable and arguably it takes considerable time and a substantial number of cases for some form of consistent case law to develop. It should also be noted that despite Argentina’s dissatisfaction with a significant number of unfavourable awards, it has not retreated from the investment arbitration regime. Neither has it terminated any BITs used by the investors to bring claims because of the crisis measures adopted.114 Nor has Argentina denounced the ICSID Convention, unlike other Latin American states.115 Moreover, Argentina has finally settled most outstanding awards, which it had previously refused to honour. Due to the complexities of the factual background and legal issues covered, the Argentina crisis cases will remain an interesting subject of analysis – not only for investment arbitration scholarship, but public international law scholarship in general.

112 U Kriebaum, ‘Evaluating Social Benefits and Costs of Investment Treaties: Depoliticization of Investment Disputes’ (2018) 33 ICSID Review – FILJ 14, 16. 113 Porzecanski, above (n 110) 43. 114 cf https://perma.cc/ZRV8-DUW5. 115 LE Peterson, ‘Venezuelan Exit from ICSID Raises Questions Both Legal and Financial’ (IAReporter, 31 January 2012), www.iareporter.com/articles/venezuelan-exit-from-icsid-raises-questions-both-legal-and-financial/.

9 The Yukos Saga: A Story of Hidden Matryoshka Dolls ALEXANDRE SENEGACNIK*

I. WHAT MAKES THE YUKOS SAGA A SAGA?

T

HE YUKOS SAGA – nonpareil in almost every aspect. Press conferences following the rendering of legal decisions, global news coverage by specialised and mainstream media, thousand-page-long decisions, record amount of damages (claimed and awarded), week-long hearings, decade-long proceedings, unprecedented costs, sterling counsel and stellar adjudicators, diplomatic strife leading to opportune legislative amendments, etc. The series of cases following the rise and fall of OAO Yukos Oil Company (‘Yukos’) are not a saga in the usual meaning of the word, ie a story of heroic achievement, but they collectively offer the blockbuster investment arbitration saga of the beginning of the century. The Yukos saga is, however, perhaps best pictured as a set of Russian Dolls. Matryoshka dolls – not all of which have yet come to light. Strikingly, the spotlight is indeed directed at only one of several procedures, while all are ultimately linked to the same set of underlying facts. The present chapter aims to shed light on the sometimes forgotten ‘smaller dolls’ in the story and to describe the broader contribution of the saga to international investment law. Yukos was a holding company established by the Russian Federation’s Government in the early nineties to own and control several entities specialised in the production of oil. Initially fully state-owned, it was subsequently privatised and emerged following the collapse of the USSR to become the largest oil and gas company in the Russian Federation.1 In May 2002, Yukos became the first Russian company to be ranked among the top ten largest oil and gas companies by market capitalisation worldwide.2

* Alexandre Senegacnik holds a PhD in Law. He is an Affiliated Researcher at Sciences Po Law School (Paris). The author wishes to express his sincere gratitude to Prof. Diego P Fernández Arroyo and to Bruno Sousa Rodrigues for their valuable comments. 1 Yukos Universal Limited (Isle of Man) v Russia, Interim Award on Jurisdiction and Admissibility (30 November 2009) PCA Case No AA 227, para 46. Yukos was incorporated as a joint stock company in 1993. Fully privatised in 1995–1996, it was a vertically integrated group engaging in exploration, production, refining, marketing and distribution of crude oil, natural gas and petroleum products. At its peak in 2003, it had 100,000 employees, six main refineries and a market capitalisation estimated at over US$33 billion. After its 2003 merger with Sibneft, YukosSibneft may have become the fourth largest oil producer worldwide. At the time of respondent’s measures in 2003, Yukos was engaged in merger negotiations with ExxonMobil and Chevron. 2 Yukos Universal v Russia (n 1) para 46.

136 Alexandre Senegacnik All the cases referred to hereafter share as a common denominator a single respondent – the Russian Federation – whose liability is, however, sought by different claimants, ie the Yukos company, the latter’s majority and some of its minority shareholders. In essence, all the different proceedings in the saga revolve around the same series of measures taken by the Russian Federation against Yukos and associated companies, all of which culminated in the bankruptcy of Yukos in 2006. These include substantive retroactive tax audits and the corresponding imposition and calculation of fines for underpayment of taxes. The Russian Federation contended that from 2000 until 2003, Yukos had implemented an illegal and fraudulent tax evasion scheme specifically designed to misuse special low-tax zones within the Russian Federation by setting up numerous sham companies in these zones to enjoy their more favourable tax regimes.3 All the claimants refuted these allegations, arguing that these practices were not in breach of Russian law at the time and that Russian tax legislation was in fact retroactively amended to justify the penalties – the reassessments were in their view instrumentalised in a global strategy to destroy Yukos.4 Additional measures complained of included VAT exactions, the freezing of shares following tax reassessments, the frustration of a possible merger of Yukos with Sibneft and the threatened revocation of licences. The Russian Federation in particular decided the forced sale at auction of Yukos’ main oil production subsidiary Yuganskneftegaz (‘YNG’). YNG was sold to an entity purportedly controlled by the Russian state before being bought by Russian state-owned company Rosneft.5 Yukos was unable to recover from these measures and was struck from the companies’ register in 2007; assets were ultimately nationalised or acquired by Russian state-owned companies such as Rosneft and Gazprom.6 The previous description suggests a rather conventional conflict between, on the one hand, the interests and property rights of the investors, and on the other hand, the state’s sovereign right to regulate and impose taxes.7 A second essential factual aspect of the saga is the confrontational relationship between two central figures: on the one hand, the notorious Russian oligarch Mikhail Khodorkovsky – who was Yukos’ main shareholder and its chief executive officer – and, on the other hand, none other than the president of the Russian Federation himself, Vladimir Putin.8 In 2003, key Yukos officers, including Mr Khodorkovsky, were arrested and charged with forgery, fraud and tax evasion. The Russian Federation has systematically put forward the same defence strategy contending that its actions, including the arrests, were necessary as Yukos had become a criminal enterprise. As a result of these arrests, a number of key Yukos executives fled the Russian Federation. Keeping this second aspect in mind, the Yukos saga is thus perhaps not only a story about the state’s sovereign taxation rights. On this aspect, the different adjudicators involved in the saga will ultimately disagree. In the end, the Yukos cases are adequately described as a saga not merely because of the sums at play and the extravagant practical contours of mammoth proceedings. It is a saga because of the subject matter of the dispute, the dramatic interplay

3 ibid

para 58. para 59. 5 ibid para 66. 6 ibid para 68. 7 W Sadowski, ‘Yukos and Contributory Fault’ (2015) 5 Transnational Dispute Management 6. See also CS Gibson, ‘A Classic Case of Indirect Expropriation’ (2015) 30(2) ICSID Review 311. 8 See C Ockrent, Les oligarques (Paris, Robert Laffont, 2014) 51–83; M Sixsmith, Putin ’s Oil: The Yukos Affair and the Struggle for Russia (Continuum, 2010); R Sakwa, The Quality of Freedom: Khodorkovsky, Putin, and the Yukos Affair (Oxford, Oxford University Press, 2009). 4 ibid

The Yukos Saga: A Story of Hidden Matryoshka Dolls 137 and dialogue of involved adjudicators, and because of all the unanticipated twists which have been continuously delivered on both the legal and political scene for almost two decades.

II. UNBOXING THE HIDDEN DOLLS: THE KNOWN PROCEEDINGS IN THE SAGA

This chapter aims to draw the saga’s contours by identifying the proceedings known to exist. A comprehensive review of all existing matryoshka dolls remains a likely impossible endeavour – many remain confidential to this very day.9 Despite the more recent aspirations towards greater transparency in the field, the saga confirms that confidentiality is still a reality of present-day ISDS.10

III. OAO NEFTYANAYA KOMPANIYA YUKOS V THE RUSSIAN FEDERATION

Even before the launch of the now-infamous ISDS cases, the Yukos company itself submitted a case before the European Court of Human Rights (ECtHR) against the Russian Federation. Yukos complained specifically about irregularities in proceedings concerning its tax liability, the perceived unlawfulness and lack of proportionality of several tax assessments and their subsequent enforcement.11 A rather strict and formal reading of Article 35(2) of the European Convention on Human Rights (ECHR) was upheld by the ECtHR, which found the case admissible as other proceedings brought by the claimant’s shareholders under the ECT and several Bilateral Investment Treaties (BITs) were not ‘substantially the same’ as the case under consideration.12 Ultimately, the ECtHR identified several violations of Yukos’ rights to a fair trial and to its property rights, mainly on account of the speed with which the Russian courts had conducted various proceedings, the handling of enforcement proceedings and because of an unjustified interference by the Russian Constitutional Court. Yet, the ECtHR had ‘little doubt’ that the factual conclusions of the Russian courts in the contested tax assessment proceedings were ‘sound’, neither arbitrary nor manifestly unreasonable.13

9 An application for a formal espousal of unnamed investors’ claims by the US Government is for instance referred to in a 2014 letter addressed by members of the US Congress to then US Secretary of State John Kerry – no further information is available in this regard. See J Hepburn, ‘As crucial decisions loom this year in several Yukos cases, US Government is urged to espouse claims of US shareholders against Russia’ (IAReporter, 27 May 2014), www.iareporter. com/articles/as-crucial-decisions-loom-this-year-in-several-yukos-cases-us-government-is-urged-to-espouse-claimsof-us-shareholders-against-russia. In addition, other proceedings are only indirectly linked to the underlying facts. See the case brought – in their personal capacities – by Mr Khodorkovskiy with business partner Mr Lebedev. Khodorkovskiy and Lebedev v Russia Apps no 11082/06 and 13772/05 (ECtHR, 25 July 2013). See more generally about the trial and imprisonment of Yukos CEOs, R Polanco Lazo and R Mella, ‘Investment arbitration and human rights cases in Latin America’ in Y Radi (ed) Research Handbook on Human Rights and Investment (Cheltenham, Edward Elgar, 2018) 77–78. 10 About the presumption of transparency, see DP Fernández Arroyo, ‘Nothing is for Free: The Prices to Pay for Arbitrabilizing Legal Disputes’ in L Cadiet, B Hess and M Requejo Isidro (eds), Privatizing Dispute Resolution (Baden-Baden, Nomos, 2019) 628–9. 11 See also U Kriebaum, ‘The European Court of Human Rights’ Case Law on International Investment Issues’, ch 28 in this volume. 12 See OAO Neftyanaya Kompaniya Yukos v Russia App no 14902/04 (ECtHR, 20 September 2011) para 524. For a comment on this decision, see F Balcerzak, Investor – State Arbitration and Human Rights (Leiden, Brill, 2017) 80. 13 Yukos v Russia, above (n 12) paras 590, 594. In addition, the Court concluded that the claimant did not receive any adverse treatment with regard to VAT claims as it found that the claimant had failed to prove that it had properly filed such claims. ibid paras 601–602. On this point the first-wave arbitration tribunals would disagree.

138 Alexandre Senegacnik The ECtHR found it ‘natural’ that in the tax sphere, ECHR Contracting States should enjoy a wide margin of appreciation in order to implement their policies – this understanding did not prevent the Court from exercising its power of review.14 The Court underscored that the Russian state’s measures were obviously capable of ‘dealing a fatal blow’ to the claimant’s ability to survive the tax claims.15 It nevertheless concluded that there had been no disguised expropriation of property and no intentional destruction of Yukos itself. Ultimately, the respondent was ordered to compensate for several unlawful penalties16 and for a disproportionately imposed enforcement fee17 – a total of €1.8 billion.

IV. ROSINVESTCO V THE RUSSIAN FEDERATION

In 2006, Yukos minority shareholder RosInvestCo, a UK subsidiary of the New York-based hedge fund, launched arbitration proceedings against the Russian Federation under the UK-USSR BIT. RosInvestCo claimed that Yukos had been expropriated and that no compensation had been provided for the loss of its investments and the subsequent damage suffered to the value of its shares in the company. The Stockholm Chamber of Commerce Tribunal was chaired by Karl-Heinz Böckstiegel sitting with claimant-appointed Lord Steyn and respondentappointed Franklin Berman. The Tribunal upheld its jurisdiction in 2007, finding first that the peculiar wording of Article 8(1) UK-USSR BIT precluded it from deciding upon any issue other than ‘the amount of compensation and the modality of payment’.18 Nevertheless, it found that it had jurisdiction beyond this limitation on the basis of the Most-Favoured-Nation (MFN) Clause in Article 3(2) UK-USSR BIT (taken together with Art 8 Denmark-Russia BIT). The Tribunal concluded that it could decide whether the respondent’s actions had to be considered as expropriations and whether these were lawful. The Tribunal issued in a well-known statement that if MFN provisions apply to substantive protection, then arguably they should apply even more to ‘only’ procedural protection.19 The Tribunal further focused on the BIT parties’ careful drafting and limitation of the effect of the MFN provision – for example for specific issues such as taxation. The absence of any reference to arbitration confirmed its view that the MFN provision also applied to dispute resolution provisions.20 The Tribunal summarily dismissed the respondent’s argument related to a possible need to exhaust local remedies, finding that the BIT merely conferred a right on the investor to pursue claims before national courts, and that, in any event the equivalent principle grounded in customary international law had been excluded by the special regime established for investor-state arbitration.21 On the merits, the Tribunal concluded that the respondent’s measures, taken together, constituted an indirect expropriation. The ‘cumulative effect’ of the respondent’s measures was found to be ‘relevant’.22 The Tribunal showed sympathy for the claimant’s argument

14 ibid

para 648. para 653. 16 ibid para 26. 17 ibid para 35. 18 RosInvestCo UK Ltd v The Russian Federation, Award on Jurisdiction (1 October 2007) SCC Case No V079/2005, para 123. 19 ibid para 132. 20 ibid para 135. 21 ibid paras 154–5. 22 RosInvestCo UK Ltd v The Russian Federation, Final Award (12 September 2010) para 621. 15 ibid

The Yukos Saga: A Story of Hidden Matryoshka Dolls 139 that despite having used nearly identical tax structures, no other Russian oil company had ultimately been subjected to similar measures as Yukos.23 The Tribunal accepted that a state’s normal application of domestic tax law cannot be seen as an expropriatory act – however, this understanding did not prevent it from examining whether the respondent’s acts ultimately amounted to an ‘abuse of tax law to in fact enact an expropriation’.24 The Tribunal found that Yukos had been – for no convincing reason – treated by the respondent quite differently than its competitors and other comparable taxpayers.25 This treatment could hardly be accepted as bona fide.26 It was undisputed that the respondent’s measures resulted in the deprivation of Yukos’ assets and that no compensation had been offered.27 The Tribunal concluded that the expropriation was unlawful. It noted that Yukos did, in some respects, contribute to its own demise, which could have become relevant in consideration of the quantum of any damage due to claimant – the Tribunal however did not reconsider this issue when calculating damages.28 On damages, the Tribunal took into account the ‘speculative’ nature of the claimant’s investment in Yukos shares.29 The harsh language employed by the Tribunal is telling: the claimant’s assessment of damages was found to be ‘divorced from reality’.30 Publicly available information at the relevant time regarding the likely outcome of the various bankruptcy and other court proceedings confirmed that ‘Yukos would cease to exist as a viable company – irrespective of the legality or other aspects of those proceedings.’31 Interestingly, the Tribunal assessed the claimant’s risks and economic interest in the Yukos shares by taking into consideration an internal arrangement between the claimant and another company in the NY Hedge Fund Group.32 The Tribunal ultimately found that the principal amount of damages due by the respondent was (only) US$3.5 million, financially granting the claimant a pyrrhic arbitral win.33

V. QUASAR DE VALORES AND OTHERS V THE RUSSIAN FEDERATION

In March 2007, seven Spanish minority shareholders initiated proceedings against the Russian Federation, alleging that the respondent had unlawfully dispossessed Yukos of its assets and expropriated it from its shareholders. The claimants demanded compensation for the loss of all value of their American Depository Receipts in Yukos on the basis of the Spain-USSR BIT.

23 ibid

para 621. para 628. See, similarly the ECtHR’s approach. 25 ibid para 556. 26 ibid para 567. 27 ibid paras 574 and 580. 28 ibid paras 585 and 635. 29 ibid para 668. 30 ibid para 672. 31 ibid para 667. 32 ibid para 672: “while the Participation Agreements did not stop Claimant from being an investor under the wide definition of the [BIT], their effect was that any risks regarding the investment were transferred to Elliott during the validity of the Participation Agreements. Claimant had no real economic interest of its own in the Yukos shares during the period the Participation Agreements were in force and thus “had nothing to lose”. Therefore, for valuation purposes of damages, the date must be applied where that risk was taken over by Claimant at the time the Participation Agreements were terminated.’ (Emphasis added) 33 The claimant had reportedly been demanding up to US$200 million. LE Peterson, ‘Arbitrators hold Russian Federation liable for expropriation of Yukos shareholdings; modest damages owed to affiliate of prominent U.S. Hedge Fund’ (IAReporter, 19 December 2010), www.iareporter.com/articles/exclusive-arbitrators-hold-russian-federationliable-for-expropriation-of-yukos-shareholdings-modest-damages-owed-to-affiliate-of-prominent-u-s-hedge-fund. 24 ibid

140 Alexandre Senegacnik The proceedings were administered by the Arbitration Institute of the Stockholm Chamber of Commerce. The Tribunal was composed of Toby Landau, Charles Brower and chaired by Jan Paulsson. In 2009, the Tribunal upheld its jurisdiction over the claimants’ expropriation claim under Article 6 Spain-USSR BIT.34 It concluded that it was empowered to decide whether an expropriation had taken place, despite the jurisdictional clause’s narrow scope, ie limited to disputes relating to the amount or method of payment of compensation due following expropriation.35 The tribunal refused to entertain additional claims arising under other provisions of the BIT which the claimants sought to invoke with the Spain-USSR BIT’s MFN clause.36 The majority of the Tribunal recognised that both substantive treatment and access to arbitration fall under the general subject matter of investor protection coming into play with a MFN Clause – the peculiar drafting of the MFN clause in that case however was found to limit its use in this case.37 The Tribunal found that only four claimants qualified as investors under the peculiar standard of the Spain-USSR BIT.38 This explains the case’s change of name.39 When deciding the merits of the case in 2012, the Tribunal first felt the need to discuss the findings of the RosInvestCo Tribunal as well as of the ECtHR, highlighting that it was not bound by either.40 The Tribunal stressed that the reliability of an investment treaty – in its view explicitly designed to induce foreigners to make investments in reliance upon it – should not be ‘diluted’ by notions such as the ‘margin of appreciation’ otherwise applicable to instruments protecting human rights.41 An interesting feature of the Quasar de Valores case relates to the claimants’ disclosure that their prosecution costs were entirely borne by Group Menatep Limited (GML).42 The respondent suggested that GML – who at one time was a majority owner of Yukos’ shares – sought to establish with this case that portfolio investors in Yukos were able to seek compensation under all investment treaties to which the Russian Federation was a party. The Tribunal summarily dismissed the argument, finding the funder’s motives to be irrelevant as between the disputants in the case.43 On the expropriation claim, the Tribunal clarified on multiple accounts that its mandate was not to decide whether there had been an ‘unlawful’ expropriation, but whether there had been an expropriation at all.44 The Tribunal concluded that the measures amounted to more

34 Renta 4 SVSA, Ahorro Corporación Emergentes FI, Ahorro Corporación Eurofondo FI, Rovime Inversiones SICAV SA, Quasar de Valores SICAV SA, Orgor de Valores SICAV SA, GBI 9000 SICAV SA v The Russian Federation, Award on preliminary Objections (20 March 2009) SCC No 24/2007, para 62. 35 Other tribunals reached the opposite conclusion. See I Marboe, ‘Quasar de Valores SICAV SA and others v The Russian Federation – Another Chapter of the Yucos Affair’ (2013) 28(2) ICSID Review 248. 36 Renta 4 SVSA, and others v The Russian Federation, above (n 34) paras 119–20. 37 ibid para 119. The MFN Clause referred to ‘no less favourable fair and equitable treatment’. The majority rejected the idea that access to an international forum would be encompassed in the fair and equitable treatment standard. See ibid para 106 and the Separate Opinion of Charles N Brower (20 March 2009) para 3. 38 The majority concluded that the BIT disqualified corporate entities bereft of legal personality. It speculated as to the possibility that this disqualification may well have been inadvertent but, as long as it did not lead to absurdity it had to be accepted as the expression of the will of the State Parties. See ibid, para 155. 39 Initially referred to as Renta 4 SVSA and others v The Russian Federation. 40 Quasar de Valores SICAV SA, and others v The Russian Federation, Award (20 July 2012) paras 24–5. 41 ibid para 22. 42 ibid para 31. For a discussion of GML’s ‘non-for-profit-funder’ status, see V Sahani, M Smith and C Deniger, ‘Third-Party Financing in Investment Arbitration’ in CL Beharry (ed) Contemporary and Emerging Issues on the Law of Damages and Valuation in International Investment Arbitration (Leiden, Brill Nijhoff, 2018) 49. 43 Quasar de Valores SICAV SA, and others v The Russian Federation, above (n 40) para 33. 44 ibid para 186.

The Yukos Saga: A Story of Hidden Matryoshka Dolls 141 than ordinary and legitimate good faith tax collection – Yukos’ tax delinquency was found to be only a pretext for seizing Yukos assets and transferring them to state-owned Rosneft.45 The Tribunal evidently tried to avoid any discussion about the legality of the expropriation.46 On damages, the Tribunal effected a 23 per cent downward adjustment of the claim to take into account the various challenges raised against the claims. The value of the shares was assessed at the time of the beginning of the claimants’ investment in 2003 to avoid awarding a ‘windfall’47 – ultimately over US$2 million were awarded. The Tribunal once more referred to the RosInvestCo decision awarding ‘only’ US$3.5 million for the purchase price of shares. It found that it faced a different situation as the claimants’ purchases – contrary to the situation in RosInvestCo – came at a time when the respondent’s series of actions against Yukos were still far from having run their course. The Tribunal flagged that its final award did not comprise any element of a penalty for illegal expropriation, as this was beyond its jurisdiction and that the respondent was only ordered to pay for what it took.48 Interestingly, the Tribunal decided that, having benefited from the ‘Good Samaritan’ GML, the claimants should bear the majority of the costs and that the respondent should not reimburse the claimants for their legal costs.49

VI. HULLEY ENTERPRISES LIMITED, YUKOS UNIVERSAL LIMITED AND VETERAN PETROLEUM LIMITED V THE RUSSIAN FEDERATION (THE TRILOGY AWARDS)

Yukos’s majority shareholders Hulley Enterprises Limited (Cyprus), Yukos Universal Limited (Isle of Man), and Veteran Petroleum Limited (Cyprus) initiated arbitration proceedings in February 2005 against the Russian Federation on the basis of the Energy Charter Treaty (ECT) and under the UNCITRAL Arbitration Rules. Composing the Tribunal of this mammoth case proved to be a challenging endeavour.50 The respondent appointed Stephen Schwebel, a former judge of the International Court of Justice (ICJ). The claimant initially appointed Daniel Price (who resigned), subsequently appointed Gabrielle Kaufmann Kohler (who was successfully challenged by the respondent) before appointing Charles Poncet. Yves Fortier was appointed as Chair by the Secretary of the Permanent Court of Arbitration (PCA). A prowess of this Tribunal was unquestionably its ability to render two unanimous trilogies of awards which have become critical precedents.51 They have been the subject of extensive scholarly scrutiny – crowned with a special decision in the ICSID Review.52 Several preliminary objections as to the Tribunal’s jurisdiction were raised, in particular a ‘pure public international law question’,53 ie whether, despite having refused to ratify the

45 ibid

para 177. in this regard the critical view of Marboe, above (n 35) 249–50. 47 Quasar de Valores SICAV SA, and others v The Russian Federation, above (n 40) para 217. 48 ibid para 214. 49 ibid para 225. 50 Technically, one should distinguish three cases and tribunals as each of the claimants maintained separate claims. The resulting six awards are virtually identical and will be referred to hereafter as the two trilogies. Reference will only be made to the Yukos Universal Limited (Isle of Man) v The Russian Federation, PCA Case No AA 227. The two remaining cases are PCA Case No AA 226 and PCA Case No AA 228. 51 References to the final trilogy can already be found in over 40 cases according to the ISLG-Database. 52 A Newcombe, ‘Yukos Universal Limited (Isle of Man) v The Russian Federation: An Introduction to the Agora’ (2015) 30(2) ICSID Review 283–292. The case was commented through six contributions. 53 As recognised by the Tribunal. See Yukos Universal Limited (Isle of Man) v Russian Federation, Interim Award on Jurisdiction and Admissibility (30 November 2009) PCA Case No AA227, para 315. 46 See

142 Alexandre Senegacnik ECT, the Russian Federation was required to arbitrate the dispute because of the provisional application of the ECT provided for in Article 45 ECT.54 The respondent relied on the so-called ‘Limitation Clause’ in Article 45(1) ECT following which each signatory agrees to apply the ECT provisionally ‘to the extent that such provisional application is not inconsistent with its constitution, laws or regulations’. A series of contradictory interpretations has ultimately been put forward in this regard.55 The claimants advocated for an ‘all-or-nothing’ approach where the provisional application of the ECT ultimately depended on whether the principle of provisional application of treaties is consistent with Russian law. The respondents advocated for a ‘piecemeal’ approach where only those provisions of the ECT (including those on dispute resolution) consistent with Russian law were to be provisionally applied. The Tribunal found that by signing the ECT, the Russian Federation had agreed that the ECT as a whole would be applied provisionally pending its entry into force, unless the principle of provisional application itself were inconsistent ‘with its constitution, laws or regulations’.56 The Tribunal experienced little difficulty in finding the principle of provisional application to exist under Russian law. In a remarkable and lengthy obiter-like section, the Tribunal further examined the compatibility of the ECT’s dispute resolution provisions with Russian law. It concluded that even following the respondent’s interpretation, the ECT was provisionally applicable.57 The Tribunal’s decision is a fantastic playground, or rather, jungle of classical interpretive arguments and narratives. Rarely has an international investment tribunal so meticulously anchored its reasoning in the interpretative rules of Articles 31–33 of the 1969 Vienna Convention on the Law of Treaties (VCLT). Paradoxically, the decision at the same time perhaps best illustrates the illusory character of any VCLT-driven quest in finding the ‘correct’ interpretation – the poor drafting of Article 45 ECT cannot so simply be interpreted away.58 A key argument for the Tribunal was that international law and domestic law should ultimately not be allowed to combine, through the deployment of an ‘inconsistency’ or ‘limitation’ clause, to form a hybrid in which domestic law directly controls the content of an international legal obligation. Allowing this would – for the Tribunal – create unacceptable uncertainty in international affairs and could only be accepted on the basis of clear language admitting no other interpretation.59 The Tribunal’s lengthy justification is easily explained by the fact that it rightly anticipated this issue to be discussed later at the enforcement stage. A second objection to jurisdiction relied on the denial-of-benefits clause of Article 17(1) ECT. For the respondent, the claimants were shell companies owned by Russian nationals and had no substantial business activities in their respective countries of incorporation.60 The Tribunal found that Article 17(1) ECT did not deny simpliciter the substantial protection of the ECT but only ‘reserved’ the Contracting States’ right to do so, in which case this right had to be properly exercised.61 A retrospective application of such denial-of-benefits would be inconsistent with the ECT’s purpose.62

54 See on this issue, T Ishikawa, ‘Provisional Application of Treaties at the Crossroads between International and Domestic Law’ (2016) 31(2) ICSID Review 286–288. 55 For a detailed analysis, see T Gazzini, ‘Provisional Application of the ECT in the Yukos Case’ (2015) 30(2) ICSID Review 293–302. 56 Yukos Universal Limited v Russia, above (n 53) para 301 and summarised in para 394. 57 ibid paras 346–92. 58 The parties and the Tribunal carefully stated that they were not authorised to rewrite the treaty. Yet a substantial part of the analysis precisely speculated on alternative formulations which could support one or another interpretation. See for instance ibid para 302. 59 ibid para 315. 60 Yukos Universal Limited v Russia, above (n 53) para 71. 61 ibid para 458. 62 ibid para 456.

The Yukos Saga: A Story of Hidden Matryoshka Dolls 143 The Tribunal concluded that a state could not retroactively exercise its right before the Tribunal, but it did not imply that such decision should have been made before the relevant investment was made.63 This decision prompted a broader discussion as to the need to clarify such treaty wording and perhaps employ mandatory language for future denial-of-benefits clauses.64 The Tribunal experienced little difficulty to qualify the claimants as investors owning or controlling an investment for the purposes of the ECT. It summarily concluded that ongoing Russian court proceedings and applications to the ECtHR failed to trigger the fork-in-the-road provision of Article 26(3)(b)(i) ECT.65 The Tribunal only ruled on the objection raised on the basis of Article 21 ECT in the final awards. Article 21 ECT aims, in line with a general policy confirmed in recent investment treaties, to protect the states’ fiscal sovereignty by excluding claims related to taxation measures from the jurisdictions of arbitral tribunals.66 The Tribunal upheld a teleological interpretation of Article 21(1), ultimately formulating two ‘bona fide’ and ‘futility’ exceptions. Accordingly, the explicit exclusion of taxation could apply only to bona fide taxation, ie actions motivated by the purpose of raising general revenue for the state.67 The Tribunal observed that it was ‘in good company’ as the RosInvestCo and Quasar de Valores tribunals had come to the same conclusion.68 To find otherwise would mean that the mere labelling of a measure as ‘taxation’ could be sufficient to bring such measure within the ambit of Article 21(1) ECT, thereby producing a loophole in the ECT’s protective scope.69 In addition, the Tribunal further concluded that the investor’s obligation pursuant to Article 21(5) ECT – to refer the issue of whether a tax is an expropriation to the relevant Competent Tax Authority – had become futile. The claimants had not referred the dispute to the very agency that was allegedly harassing it. Recent case law appears to support this approach albeit highlighting that the circumstances of the case were exceptional.70 Interestingly, a recent report published by the Energy Charter Secretariat discusses the need to review and amend Article 21 ECT.71 The Tribunal further ruled on the objection related to the claimants’ so-called unclean hands.72 It concluded that there exists no general ‘clean hands’ principle under Article 38(1)(c) ICJ Statute – the respondent sought to invoke such principle as an applicable rule or principle of international law in the meaning of Article 26(6) ECT.73 Nevertheless, and despite the

63 See more generally, L Gastrell and PJ Le Cannu, ‘Procedural Requirements of ‘Denial-of-Benefit’ Clauses in Investment Treaties: A Review of Arbitral Decisions’ (2015) 30(1) ICSID Review 88. See also PJ Le Cannu and FL Torres, ‘Denial of Benefits Clause’, ch 24 in this volume. 64 See for instance the 2011 UNCTAD Series II Note (Scope and Definition), 98–9. 65 Yukos Universal Limited v Russia, above (n 53) para 598. 66 Energy Charter Treaty 2080 UNTS 95 (entered into force16 April 1998), Art 21(1) provides that nothing in the ECT shall create rights or impose obligations with respect to ‘Taxation Measures of the Contracting Parties.’ 67 Yukos Universal Limited (Isle of Man) v The Russian Federation, Final Award (18 July 2014) para 1431. 68 ibid para 1436. 69 ibid para 1433. 70 S Green Martínez, ‘Taxation Measures under the Energy Charter Treaty after the Yukos Awards Articles 21(1) and 21(5) Revisited’ (2019) 34(1) ICSID Review 97–103. See also D Azaria, ‘The Renewable Energy Arbitrations Under the Energy Charter Treaty’, ch 10 in this volume. 71 UE Özgür, ‘Taxation of Foreign Investments under International Law: Article 21 of the Energy Charter Treaty in Context’ (2015) Energy Charter Secretariat Report 17, https://perma.cc/ND84-3JX7. 72 The respondent listed 28 instances of alleged illegal and bad faith conduct related to the acquisition of Yukos and the subsequent consolidation of control over Yukos and its subsidiaries. Yukos Universal Limited (Isle of Man) v The Russian Federation, Final Award (18 July 2014) paras 1273–4. 73 The Tribunal issued a unanimous decision which is even more remarkable considering that arbitrator/judge Schwebel had himself invoked the doctrine in his dissent in the Nicaragua v United States case before the International Court of Justice. See the discussion of the Tribunal’s finding by A Llamzon, ‘The State of the “Unclean Hands” Doctrine in International Investment Law: Yukos as both Omega and Alpha’ (2015) 30(2) ICSID Review 315.

144 Alexandre Senegacnik absence of any specific textual hook, the Tribunal concluded that a good faith and teleological interpretation of the ECT commanded to condition the protection of investments on their legality, or on the good faith of the investor.74 This being said, post-investment illegality was not found to deprive the Tribunal of its jurisdiction or render the claims inadmissible.75 The Tribunal found that the alleged illegalities were in this case ‘insufficiently’ connected with the final transaction by which the claimants’ investment had been made.76 On the (proper) merits, the Tribunal concluded in 2014 that the Russian Federation had breached its obligations under Article 13 ECT as a result of the measures which had an effect equivalent to expropriation.77 The unprecedented factual narrative of the case was analysed in over 300 pages but the Tribunal’s legal reasoning on a ‘classical’ or ‘paradigmatic example’ of creeping expropriation did not exceed five pages.78 The respondent’s series of measures were found to have been orchestrated through the highest levels of the government and by almost all of its organs.79 The award certainly is in line with a recent trend to require tribunals to conduct case-by-case fact-based inquiries.80 A common two-step approach was followed by first determining the existence of an expropriation, before examining its lawfulness. Admittedly, the determination could have gained from a more structured analytical framework.81 The extreme economic impact of the respondent’s measures on Yukos was not disputed – the claimants’ legitimate expectations, by contrast, were. The respondent argued that its measures were legitimate acts of law enforcement in compliance with Russian law and practices as well as standards of other countries.82 The Tribunal recognised that the claimants should have expected a reaction from the authorities because of Yukos’s tax avoidance operations.83 The Tribunal however embarked on a remarkable stylistic exercise, with a long anaphora listing the extremity of all the respondent’s actions which could not legitimately have been expected by the claimants.84 It concluded that the primary objective of the respondent was not to collect taxes but rather to bankrupt Yukos and appropriate its valuable assets.85 The expropriation was found to be unlawful. The Tribunal found it ‘profoundly questionable’ whether the measures were in the public interest of state, hinting that they certainly were in the interest of the largest state-owned oil company.86 While Yukos had been subjected to processes of law, the Russian courts were found to have ‘bent to the will of Russian executive authorities’ to bankrupt the company, assign its assets to a state-owned company, and ‘incarcerate the man who

The Tribunal’s finding in any event echoes the decision of the International Law Commission to decline to identify the clean hands doctrine as a circumstance precluding liability. 74 Yukos Universal Limited v Russia, above (n 67) para 1352. 75 ibid para 1355. 76 ibid para 1370. 77 ibid para 1580. 78 CS Gibson, ‘A Classic Case of Indirect Expropriation’ (2015) 30(2) ICSID Review 303–4. See also DA Desierto, ‘Expropriation Cases’, ch 21 in this volume. 79 Yukos Universal Limited v Russia, above (n 67) para 1584. 80 See for instance Comprehensive Economic and Trade Agreement between Canada, of the one part, and the European Union and its Member States, of the other part (signed 30 October 2016 entered into force preliminary 21 September 2017) [‘CETA’], Annex 8-A (Expropriation), para 2. 81 See above (n 79). 82 Yukos Universal Limited v The Russia, above (n 67) para 763 – See also Counter-Memorial para 673; Transcript, Day 18 at 113–21 (Respondent’s closing). 83 ibid para 1578. 84 ibid para 1578. 85 ibid para 1579. 86 ibid para 1581.

The Yukos Saga: A Story of Hidden Matryoshka Dolls 145 gave signs of becoming a political competitor’.87 The absence of any prompt, adequate and effective compensation was in any event undisputed.88 As to the assessment of damages,89 the Tribunal identified the substantial and irreversible deprivation of the claimant’s main production asset YNG in December 2004 as the expropriation date. The claimants were allowed to select as the valuation date either the 2014 date of the award or the 2004 expropriation date. The Tribunal calculated the two corresponding amounts90 before deciding to award the higher amount of US$66.694 billion.91 This amount was not awarded in full, in light of the claimants’ contributory fault. The Tribunal claimed a wide margin of discretion in apportioning fault.92 It found that only two actions by the claimants – out of the 28 actions or omissions argued by the respondent – had materially contributed to the subsequently suffered prejudice.93 The Tribunal found it ‘fair and reasonable’ to consider that the claimants had thus contributed to the extent of 25 per cent to their final prejudice.94 Selecting a percentage may appear at first sight rather random but appears common in practice.95 The final amount awarded was thus approximately US$50 billion. The Tribunal decided that it could see no reason why the respondent, as the unsuccessful party, should not bear the costs of the arbitration of over €8 million.96 Strikingly, the Tribunal felt the need to justify the ‘unsurprising’ extraordinary amount in legal fees advanced by the claimants.97 It ultimately granted reimbursement of 75 per cent of claimants’ grand total of costs, ie US$60 million.98

VII. THE NEW CONFIDENTIAL ‘SECOND-WAVE’ TRILOGY: LUXTONA LIMITED, YUKOS CAPITAL AND FINANCIAL PERFORMANCE HOLDINGS V THE RUSSIAN FEDERATION

The first case was brought forward in 2013 by Cypriot company Luxtona Limited on the basis of the ECT. It reportedly focuses on certain Yukos shares held by the claimant and whose value has been adversely affected by the various measures taken by the Russian Federation and the consequent bankruptcy process of Yukos.99 The Tribunal is composed of claimantappointed Luca Radicati di Brozolo, respondent-appointed Rodrigo Oreamuno and chaired

87 ibid

para 1583. para 1584. 89 For a detailed analysis of the decision on damages, costs and interests, see I Marboe, ‘Calculation of Damages in the Yukos Award: Highlighting the Valuation Date, Contributory Fault and Interest’ (2015) 30(2) ICSID Review 333–4; MA Abdala and A Rozenberg, ‘Assessing Investor Damage Involving Publicly Traded Companies – with Examples from the Yukos Cases’ in T Carpenter, M Jansen and J Pauwelyn (eds), The Use of Economics in International Trade and Investment Disputes (Cambridge, Cambridge University Press, 2017) 349–70. 90 Yukos Universal Limited v The Russia, above (n 67) paras 1778–9. 91 ibid para 1826. 92 ibid para 1600. The Tribunal referred to Art 39 of the ILC Draft Articles on State Responsibility. 93 ibid para 1634. For these, the claimants ‘should pay a price.’ 94 ibid para 1637. 95 Marboe, above (n 89) 333. 96 Yukos Universal Limited v Russia, above (n 67) paras 1867–1869. 97 It was arguably three times higher than the respondents’ costs, as the claimants bore the burden of proof for their claims under the ECT and produced many fact-witnesses in the Hearing on the Merits. 98 Yukos Universal Limited v Russia, above (n 67) para 1887. 99 LE Peterson, ‘Investigation: as Yukos enforcement grabs headlines, Russia has faced at least 10 new treaty arbitrations since 2012, with others threatened’ (IAReporter, 14 July 2015), www.iareporter.com/articles/investigation-as-yukosenforcement-grabs-headlines-russia-has-faced-at-least-10-new-treaty-arbitrations-since-2012-with-others-threatened. 88 ibid

146 Alexandre Senegacnik by John Crook. Ongoing and related proceedings before Canadian Courts – the Tribunal is seated in Toronto – reveal that the Tribunal upheld its jurisdiction, rejecting the – now standard – preliminary objections related to Articles 17 and 45 ECT while reserving its decision on the nature of the claimant’s investment for a later stage.100 The second case was brought forward by Luxembourg company Yukos Capital, again on the basis of the ECT. It reportedly centred on certain loans and financing transactions which had been made to Yukos prior to its fall and which were not recognised and repaid following the measures taken by the Russian Federation and the resulting bankruptcy process.101 The Tribunal was composed of claimant-appointed William Rowley, respondent-appointed Brigitte Stern and chaired by Campbell McLachlan. Ongoing and related proceedings before Swiss Courts – the Tribunal is seated in Geneva – revealed that several objections put forward by the respondent were rejected. A majority concluded that the ECT applied provisionally and that certain loans qualified as protected investments.102 Arbitrator Stern reportedly issued a dissenting opinion on this matter, however joining the majority to reject the objection based on a possible denial of benefits.103 The Claimant announced in July 2021 that it had prevailed on the merits, with the tribunal finding Russia liable for the expropriation of the investment, as well as a denial of justice in Russian courts, ultimately awarding around US$5 billion in damages, inclusive of interest.104 The third case was brought forward by Dutch entity Financial Performance Holdings BV on the basis of the ECT and reportedly centred on certain loans and financing transactions like the Yukos Capital case.105 The proceedings were discontinued.106

VIII. ISDS’ EFFICIENCY PARADOX BROUGHT TO LIGHT

Having unboxed the saga’s hidden dolls, the time is ripe to reflect on the possible teachings of the saga. A few observations can be formulated before discussing the final paradox which, in the author’s view, the saga brings to light.

100 J Hepburn, ‘Interim award in Luxtona v. Russia arbitration comes to light, offering new reasoning on provisional application of Energy Charter Treaty and Russia’s attempted denial of benefits to this Yukos shareholder’ (IAReporter, 4 January 2018), www.iareporter.com/articles/interim-award-in-luxtona-v-russia-arbitration-comes-tolight-offering-new-reasoning-on-provisional-application-of-energy-charter-treaty-and-russias-attempted-denial-ofbenefits-to-this-yukos. The latest publicly available information in April 2021 suggests that the Tribunal has not yet proceeded to hear the merits of the case while the jurisdictional decision is still challenged in Toronto. 101 See Hepburn, above (n 100). 102 LE Peterson, ‘In second-wave Yukos arbitration, McLachlan and Rowley see Russia as provisionally bound by Energy Charter Treaty’ (IAReporter, 16 February 2017), www.iareporter.com/articles/ in-second-wave-yukos-arbitration-mclachlan-and-rowley-see-russia-as-provisionally-bound-by-energy-charter-treaty. 103 J Hepburn, ‘Russia turns to Canadian and Swiss courts, seeking to set aside a pair of Yukos ‘second-wave’ Energy Charter rulings – but Swiss bid is deemed premature’ (IAReporter, 10 August 2017), www.iareporter.com/articles/ russia-turns-to-canadian-and-swiss-courts-seeking-to-set-aside-a-pair-of-yukos-second-wave-energy-charter-rulingsbut-swiss-bid-is-deemed-premature. 104 D Charlotin, ‘Claimant in second-wave Yukos case against Russia announces $5 Billion Victory’ (IAReporter, 28 July 2021), www-iareporter-com.acces-distant.sciencespo.fr/articles/breaking-claimant-in-second-wave-yukoscase-against-russia-announces-5-billion-victory. 105 See Hepburn, above (n 100). 106 See Financial Performance Holdings BV (The Netherlands) v The Russian Federation, PCA Case No 2015-02, https://perma.cc/S28M-4FKU.

The Yukos Saga: A Story of Hidden Matryoshka Dolls 147 A. Lost in (In)consistency? – Identifying the Saga’s Inconsistent Findings The concurrent jurisdiction of various adjudicators over a similar set of facts raises palpable and understandable concerns in terms of possible inconsistencies.107 This concurrent jurisdiction is possible as minority shareholders can occasionally enjoy the benefit of investment protection.108 The saga provides stimulating food for thought when it comes to keywords such as consistency and harmony in ISDS. The different proceedings have ultimately become key arguments to invoke or to reject in future proceedings. A never-ending re-discussion of the very same arguments – in particular the provisional application of the ECT – adds a tautological flair to the most recent decisions which appear to revisit now historic questions. The bigger picture of the saga can certainly lead to a finding of sympathy for the involved respondent state, as it is practically forced to spend considerable resources to defend itself in all these proceedings. Yet, one can only but conclude that the respondent state is – as in many cases109 – not interested in joining these proceedings. So far, the biggest divide ultimately emerges between the ECtHR and the various arbitral tribunals. For sure, all involved adjudicators to this day unmistakably have recognised the state’s fiscal sovereignty. They also share a common understanding that a state cannot prevent them from exercising their power to review that state’s actions. Doubtlessly though, one can still be puzzled by the remaining differences which can be found between all rendered decisions. This is particularly true for the conclusion as to whether the respondent’s measures resulted (or not) in an indirect or disguised expropriation.110 The ECtHR’s finding – that they did not – appears in this regard unique. This conclusion stands in tension with the ECtHR’s own assessment that some exacted measures were obviously capable of ‘dealing a fatal blow’ to Yukos’s ability to outlive the tax claims.111 Ultimately, the Yukos saga reveals the limited dialogue existing between international adjudicatory bodies. The structural difficulty to deal on the international plane with parallel proceedings – all of which are ultimately linked to the same set of underlying facts – particularly comes to light with the saga.

IX. THE ENFORCEMENT SAGA WITHIN THE SAGA

Clearly, the saga’s first chapter largely tells the story of the claimants’ multiple victories before the ECtHR and several arbitral bodies. Yet, these decisions only mark the beginning of a second chapter: the story of their enforcement. Particularly for arbitration, the saga exposes what could be labelled ISDS’s efficiency paradox. The paradox rests on the apparent tension between the arbitral tribunals’ efficiency (in ultimately awarding damages) and the difficulties experienced when enforcing these decisions (to effectively recover damages). 107 See E Brabandere, ‘Complementarity or Conflict? Contrasting the Yukos-Case Before the European Court of Human Rights and Investment Tribunals’ (2015) 30(2) ICSID Review – Foreign Investment Law Journal 345. See also M Audit, ‘La coexistence de procédures contentieuses’ in C Leben (ed), Droit international des investissements et de l’arbitrage transnational (Paris, Pedone, 2016) 953. 108 The topic has gained attention in the ISDS Academic Forum en marge of the ISDS Reform efforts at UNCITRAL. See J Arato and others, ‘Reforming Shareholders Claims in ISDS’ (2019) 9 Concept Paper, https://perma.cc/S4JG-9DAP. 109 This is notoriously the strategy pursued by the Argentine state for the different proceedings launched by a considerable number of investors in the aftermath of the country’s 1998-2002 economic crisis. See also A Reinisch and J Tropper, ‘The Argentinian Crisis Arbitrations’, ch 8 in this volume. 110 Brabandere, above (n 107) 355. 111 See in contrast Quasar de Valores SICAV SA and others v The Russian Federation, Award (20 July 2012) para 120.

148 Alexandre Senegacnik These difficulties can be explained by the shortcomings of the applicable enforcement and annulment regime. None of the awards rendered in the saga were ICSID awards.112 The rendered arbitral awards fall under the regime otherwise applicable to international commercial arbitration.113 Practically, this enforcement-saga-in-the-saga is fought on two battlefields. First, claimants have to actively seek enforcement in every jurisdiction with seizable assets. In addition, they must also prevail in the annulment procedures initiated before all the courts where the different tribunals were seated, ie The Hague, Geneva, Toronto and Stockholm. Accordingly, a considerable number of domestic courts are adding their voice to the saga, second-guessing the various findings of tribunals with little to no coordination. The annulment proceedings could perhaps impose one point of view. Yet, the Dutch Courts’ domestic disagreement is best proof of the ongoing tumult. Fortunately, a detailed comparative study of the ongoing enforcement proceedings is not indispensable to discuss the efficiency paradox. It can be revealed that the Russian Federation succeeded in setting aside the awards rendered in both the Quasar de Valores114 and RosInvestCo cases115 – in the latter case, the claimants simply gave up on enforcing their pyrrhic arbitral victory to avoid spending any further costs. As for the so-called second-wave proceedings, the Luxtona Tribunal’s jurisdictional award is currently being challenged before Canadian courts.116 In the meantime, the Swiss Supreme Court decided that the Yukos Capital Tribunal’s jurisdictional award could only be challenged once all jurisdictional objections had been decided.117 This final section will focus on the award trilogies rendered in The Hague in favour of Yukos’ majority shareholders. The Russian Federation had to wait for the final awards to be rendered in 2014 before it could request the setting aside of the interim awards rendered five years earlier. Several grounds of annulment were invoked, out of which two merit particular mention. The first is that the Tribunal wrongly upheld its jurisdiction by finding that the ECT applied provisionally. Second, the Russian Federation further argued that several parts of the award had in fact partially been ghost-written by an assistant to the Tribunal, rather than by the Tribunal members themselves. In 2016, the District Court in The Hague found that the Tribunal had wrongly upheld its jurisdiction.118 The Court sided with the respondent’s interpretation of Article 45(1) ECT before further concluding that, contrary to the Tribunal’s finding, the dispute resolution

112 The

enforcement regime set down in the ICSID Convention is thus inapplicable. for non-recognition are likely to be governed by the 1958 New York Convention on the Enforcement and Recognition of Foreign Arbitral Awards 330 UNTS 3 (entered into force 7 June 1959) [‘New York Convention’] while annulment proceedings are on the other hand provided in national legislation. Depending on the applicable rules, the respondent will sometimes have to wait for a final award, before it can request the setting aside of an award. 114 LE Peterson, ‘Swedish Supreme Court leaves in place lower court’s declaration that Yukos arbitrators lacked jurisdiction over Spanish BIT case; set-aside of award now looks inevitable’ (IAReporter, 16 December 2016), www. iareporter.com/articles/swedish-supreme-court-leaves-in-place-lower-courts-declaration-that-yukos-arbitratorslacked-jurisdiction-over-spanish-bit-case-set-aside-of-award-now-looks-inevitable. 115 In November 2011, the Stockholm District Court rendered a default judgment holding that the Tribunal wrongly upheld its jurisdiction. The Svea Court of Appeal eventually set aside the final award in September 2013. See LE Peterson, ‘ANALYSIS: As domestic courts second-guess arbitral verdicts in past Yukos cases, three more tribunals are currently assessing whether they have jurisdiction over Russia under ECT’ (IAReporter, 25 April 2016), www. iareporter.com/articles/analysis-as-domestic-courts-second-guess-arbitral-verdicts-in-past-yukos-cases-three-moretribunals-are-currently-assessing-whether-they-have-jurisdiction-over-russia-under-ect. 116 The Russia Federation v Luxtona Limited, ONSC 4503, Court File CV-17-11772, 13 December 2019, para 70. 117 Federal Supreme Court of Switzerland, Decision No 4A_98/2017, 20 July 2017. 118 Hague District Court, C/09/477160 / HA ZA 15-1, C/09/477162 / HA ZA 15-2 and C/09/481619 / HA ZA 15-112, Judgment of 20 April 2016. 113 Grounds

The Yukos Saga: A Story of Hidden Matryoshka Dolls 149 provisions of the ECT were not compatible with Russian law.119 As a result, the interim awards rendered seven years earlier were quashed. In 2020, the Hague Court of Appeal reversed the lower court’s decision and effectively reinstated the awards. Strikingly, a new interpretation – in the meantime advanced by the claimants – of Article 45(1) ECT was upheld to confirm the provisional application of the ECT.120 Additional objections raised by the respondent were this time all reviewed and rejected.121 The fact that the ECtHR concluded that there had been no violation of the ECHR in the imposition of several additional tax assessments was not sufficient for the Court of Appeal to conclude that the Tribunal had erred in finding that the taxation measures were not bona fide.122 Lastly, the contribution of the Tribunal’s assistant in the making of the awards was found not be a matter of concern, despite the various stylometric reports filed by the respondent. The Court found that the respondent was unable to prove that the assistant had ultimately participated in the decision-making process. The submission of draft texts written under the responsibility of the arbitrators and accepted by them could not justify the conclusion that the Tribunal has (seriously) violated its mandate.123 Despite their reinstatement, the two award trilogies now await a final decision to be rendered by the Dutch Supreme Court.124 In the meantime, enforcement actions have been initiated by the majority shareholders in various jurisdictions – notably France, Belgium and the US – with to this day little success in recovering the record amount of awarded damages.125 In the majority of these jurisdictions, the grounds for enforcement are those of the 1958 New York Convention. The claimants have to seek enforcement before each of these courts, which all retain their particularities, be it in the interpretation of the Convention,126 or on other matters which remain governed by domestic law, in particular questions of state immunity.127

X. THE SAGA’S TAKEAWAYS FOR FUTURE REFORM

The different interpretations of the ECT offered within the saga have certainly gained the attention of investment treaty drafters. The EU Commission’s 2020 proposal of amendments 119 ibid paras 5.74–5.94. The District Court found that Russian law prohibited arbitration of public disputes and concluded that a specific approval of the Russian Parliament would have been required for the respondent to consent to the arbitration. 120 Hague Court of Appeal, 200.197.079/01 (Judgment, 18 February 2020). The Court of Appeal concluded that the clause prevented only the provisional application of specific provisions of the ECT, when the provisional application of these provisions was prohibited under Russian law (para 4.5.33). The Court found that the prohibition of arbitration of public law disputes – relied upon by the lower court – was not relevant since, in its view, investment disputes should be approached as civil disputes (para 4.7.35). 121 The District Court only focused on the provisional application of the ECT. 122 Hague Court of Appeal, above (n 120) para 5.2.28: ‘Not only did the ECtHR assess a different question than that before the Court of Appeal, but the ECtHR also used an assessment framework that allows the State in question a wide margin of discretion.’ (Unofficial translation) 123 ibid para 6.6.14.1. 124 In November 2021, the Dutch Supreme Court provided another unexpected twist in the saga by rejecting most of Russia’s arguments for overturning the Hague Court of Appeal decision. The lower court’s decision was nevertheless quashed as the Supreme Court upheld Russia’s objection that it was wrongly denied the right to challenge the award on account of alleged fraud. The matter was referred to the Amsterdam Court of Appeal. See Dutch Supreme Court, ECLI:NL:HR:2021:1645 (Judgment, 5 November 2021). 125 Most are available at www.italaw.com/cases/1175. 126 See generally GA Bermann (ed), Recognition and Enforcement of Foreign Arbitral Award: The Interpretation and Application of the New York Convention by National Courts (New York, Springer, 2017) 30–71. 127 See on this, J Fouret and P Daureu, ‘Enforcement of the Yukos Awards: A Second Noga Saga or a New Sedelmayer Fight?’ (2015) 30(2) ICSID Review 336.

150 Alexandre Senegacnik for the ECT is just one specimen among many.128 The extent to which the saga will ultimately shape the text of a possibly amended ECT will certainly be an interesting subject for future scrutiny. On the procedural plane, the saga amply highlights the chaotic control of ISDS awards and the general difficulty of ensuring their enforcement. The general policy decision in favour of the finality of ISDS awards is currently the subject of a heated debate in the arbitration community. More control is to be expected: the UNCITRAL Working Group III working on a possible ISDS reform is already considering, inter alia, the possible benefits of creating a stand-alone or jurisdictional appellate body.129 In this regard, the saga allows the formulation of three relevant observations. First, many domestic courts are carefully awaiting the final decision of the Dutch courts as to the possible annulment of the two award trilogies. Article V(1)(e) of the New York Convention notoriously authorises courts where enforcement is sought to refuse recognition of annulled awards. Such refusal is, however, not automatic. A few jurisdictions – at the forefront of which is France – have on occasion recognised awards despite their annulment.130 As such, the saga confirms that the existing framework might not be optimal to ensure effective and coordinated control.131 Second, domestic courts will often not hesitate to review de novo the jurisdiction of arbitral tribunals. Concretely: arbitrated arguments will often be (re)litigated. Other than mere implications of costs and delay, such de novo consideration of arbitral findings is a cause for great concern. This is particularly evident for the contentious issue of the provisional application of the ECT. In 2017, the Paris Court of Appeal considered the possibility of referring to the Court of Justice of the European Union a question on the interpretation of the ECT.132 The claimants ultimately decided to withdraw their enforcement bids in France.133 Since then, the French Court has (in a different dispute) referred a question on the interpretation of the ECT to the Court in Luxembourg.134 In 2020, the Russian Federation requested the Dutch Supreme Court to refer certain questions of interpretations of the ECT to the Court in Luxembourg.135 The resulting political and legal difficulties are obvious. Third, there remains a final and non-negligible hurdle even once enforcement is granted. The claimants still have to identify seizable assets136 and carefully consider the relevant and 128 See for instance the proposed denial-of-benefits clause – the proposal is available at https://perma. cc/4BR2-Q5LP. See also. See J Hepburn, ‘Analysis: Energy Charter Treaty Secretariat release wide-ranging proposals by governments for treaty amendments’ (IAReporter, 11 October 2019), www.iareporter.com/articles/ analysis-energy-charter-treaty-secretariat-release-wide-ranging-proposals-by-governments-for-treaty-amendments. 129 Working documents and reports are available at https://uncitral.un.org/en/working_groups/3/investor-state. See also the report by Members of the Academic Forum: ‘Lack of Consistency and Coherence in the Interpretation of Legal Issues’, https://perma.cc/2HV3-QJLN. 130 The Paris Court of Appeal stressed this point specifically with regard to the enforcement of the trilogy awards. Paris Court of Appeal, RG n°15/11666 (Joined with n°15/11667), 27 June 2017. 131 The self-contained ICSID framework should be distinguished in this respect. 132 Paris Court of Appeal, above (n 130). 133 D Thomson, ‘Yukos shareholders give up on enforcement in France’ (Global Arbitration Review, 10 October 2017), https://globalarbitrationreview.com/article/1148728/yukos-shareholders-give-up-on-enforcement-in-france. 134 Paris Court of Appeal, RG n°18/14721, 24 September 2019. The questions relate to the definition of an investment in the meaning of Arts 1(6) and 26(1) ECT. 135 A Ross, ‘Will Yukos now go before the ECJ?’ (Global Arbitration Review, 15 May 2020), https://globalarbitrationreview.com/article/1226901/will-yukos-now-go-before-the-ecj. 136 The majority shareholders have announced the successful attachment in the Netherlands of a number of alcohol trademarks – including the well-known Stolichnaya and Moskovskaya vodka brands – which were found to belong to Russian state-entity Sojuzplodoimport. See the Reuters report dated 18 May 2020, available at https://perma.cc/ CB5H-HLKM.

The Yukos Saga: A Story of Hidden Matryoshka Dolls 151 often opaque rules on immunity. Several countries, including France, have amended their legislation in this regard in recent years.137 The Yukos saga has in many cases given the impetus for such drastic changes. Interestingly, follow-on threats of BIT claims by Russian agencies whose assets have been seized have already been reported. In 2016, a Russian space agency filed a notice of dispute with France complaining of a freezing order decided by a French court to enforce the award trilogies rendered in The Hague.138 In 2017, the Paris Court of Appeal concluded that said agency was financially independent of the Russian Federation – the freezing order was lifted.139 It is impossible to predict the outcome of ongoing ISDS reform at UNCITRAL. Considering the outcome of the different proceedings in the Yukos saga, one may find intriguing that the Russian Federation’s delegation is in fact one of the few delegations openly defending the current ISDS system at UNCITRAL. In fact, this position is very much understandable as the Russian Federation vigorously opposes the proposed creation of a Multilateral Investment Court. At the end of the day, resisting enforcement of the decisions rendered by such a court might prove to be even more difficult – at least politically – than resisting to ISDS awards. In this regard, it is important to recall the Russian Federation’s decision to entrust its own Constitutional Court with the power to review the decisions rendered by the ECtHR.140 In 2017, the Russian Constitutional Court ruled that the Russian Federation was not bound to enforce the ECtHR decision on the award of pecuniary compensation in the Yukos case.141 In the end, the Yukos saga is certainly not a saga about the Russian Federation’s trust in the rule of (international) law and adjudication.142 Rather than providing definitive answers, the Yukos saga has allowed crucial questions to emerge. These have become key preoccupations in the field. Among these, one may cite shareholders’ standing before tribunals, the consequence of a possible wrongdoing of the investor, the usual difficulties in assessing damages and awarding fair compensation. The saga’s main teachings can also be found on the enforcement plane. The saga confirms that international investment law is crafted today by a large number of actors which certainly do not include only investment arbitrators. Arbitral tribunals were only one part of the story. The dialogue between all the saga’s actors ultimately appears to be rather random, truly tentative and frequently

137 The ‘Sapin 2 law’ (2017) amended the French rules governing enforcement, provisional measures, and state immunity. In particular, a prior authorisation of French courts is now required for all provisional or enforcement measures against property of a foreign state located in France. This change has been occasionally described as the ‘Putin Amendment’ or ‘the Yukos law.’ See for instance F Nodé-Langlois, ‘L’“amendement Poutine” adopté dans la loi Sapin 2’ (10 June 2016) Le Figaro, https://perma.cc/LYZ4-JV8K; T Jones, ‘‘Yukos law’ grabs attention at GAR Live Paris’ (Global Arbitration Review, 25 November 2016), https://globalarbitrationreview.com/ article/1076780/-yukos-law”-grabs-attention-at-gar-live-paris. 138 The Notice in French is available at https://perma.cc/ML7S-LVFL. 139 Paris Court of Appeal, RG n°16/01314, 27 June 2017. 140 In December 2015, the Russian Constitutional Court was granted the powers to rule on enforcement of the decisions of international human rights bodies. See ‘On the Amendments to the Federal Constitutional Law on the Constitutional Court of the Russian Federation’. Interim Opinion no 832/2015 delivered by the European Commission for Democracy through Law (Venice Commission), 15 March 2016. 141 I Marchuk, ‘Flexing Muscles (Yet Again): The Russian Constitutional Court’s Defiance of the Authority of the ECtHR in the Yukos Case’ (EJIL:Talk!, 13 February 2017), https://perma.cc/ZL6U-9W8R. The Court held that damages had been the result of Yukos’ illegal activities: enforcement would thus contravene the constitutional principles of equality and fairness in the area of taxation. 142 The Russian Federation’s ratification of the ECT remains unlikely. See on this, Boute A, ‘The Protection of Russian Investments in the EU Energy Market: A Case in Support of Russia’s Ratification of the Energy Charter Treaty’ (2014) 29(3) ICSID Review 525–47.

152 Alexandre Senegacnik one-sided. But even in the absence of an explicit dialogue, it is clear that all adjudicators were aware of their respective counterparts’ existence. In this vein, a counsel for claimants in one of the arbitrations perhaps best pointed out that he was involved in a case with practical implications going well beyond the disputing parties.143 One may only speculate as to the number of remaining hidden dolls in the saga.

143 Renta 4 SVSA and others v The Russian Federation, Separate Opinion of Charles N Brower at fn 1, reporting the counsel’s statement: ‘This is about more than the few million dollars that are at stake for the Claimants in this case’.

10 The Renewable Energy Arbitrations under the Energy Charter Treaty DANAE AZARIA*

I. INTRODUCTION

T

HE ENERGY SECTOR depends on large, upfront investments, which can only be recovered over a long period of time. This is not only the case for oil and gas, but also of the renewable energy (RE) sector.1 In light of the need to address climate change and energy security concerns, numerous states have been incentivising such long-term investment in the RE sector. This is especially the case of European Union (EU) Member States given the adoption of EU Directive 2001/77/EC on the promotion of electricity produced from renewable energy sources in the internal electricity market (Renewables Directive),2 which aimed at ensuring a 12 per cent input of electricity from RE to the EU’s gross inland energy consumption by 2010.3 Under the Renewables Directive, EU Member States were obliged to attain individual targets for the consumption of electricity from RE.4 Given the significant capital required for investments in this sector, many states, especially those in the EU, have enacted legal frameworks, such as feed-in tariffs (FITs), to incentivise this specific long-term investment.5 Such investments need the continuity of incentive schemes and protection from government policy changes. As the favourable subsidies led to enormous investment in RE and an electricity tariffs deficit, coupled with the financial crisis of 2009, many EU states modified or withdrew their initial incentives. Their measures have given rise to an abundance of investor-state arbitrations under the Energy Charter Treaty (ECT). Up until 30 January 2020, investors in the RE sector have brought 40 arbitration proceedings against Spain,6 9 against Italy7 and 7 against the Czech Republic8 under the ECT.

* Dr Danae Azaria is an Associate Professor at the Faculty of Laws of the University College London (UCL). The auhor is grateful to Anna Morogai and Subana Kerisnan for their research assistance. 1 A recognition of the special nature of foreign investment in the energy sector: RREEF, para 240. 2 Directive 2001/77/EC of the European Parliament and of the Council of 27 September 2001 on the promotion of electricity produced from renewable energy sources in the internal electricity market [2007] OJ L283/33. 3 ibid. 4 Renewables Directive, above (n 2) Preamble (5)–(7). 5 This incentive has also been driven by the need to address climate change and energy security concerns thus leading to the option of diversification from oil and gas (especially from third states). 6 ‘Investment Dispute Settlement Navigator – Spain’: https://perma.cc/P2JB-KQRU. 7 ‘Investment Dispute Settlement Navigator – Italy’: https://perma.cc/5C4E-H7KZ. 8 ‘Investment Dispute Settlement Navigator – Czechia’: https://perma.cc/59GV-B5Q7.

154 Danae Azaria The RE ECT arbitrations are landmark in numerous fields: in international investment law and international energy law, as well as EU law, and public international law. They deal (in the form of jurisdictional objection) with the intricacies of whether the ECT applies to intra-EU investments, given that the latter are subject to the specific regulatory framework of the EU. Some tribunals have addressed this type of (or a variation of this) objection even in the aftermath of the landmark judgment of the Court of Justice (CJ) – Slovak Republic v Achmea (2018).9 In this case the CJ was faced with a preliminary request by a German domestic court to determine whether Articles 344 and 267 of the Treaty for the Functioning of the European Union (TFEU) preclude the application of a provision in a bilateral investment treaty (BIT) between EU Member States under which an investor of a Contracting State may bring arbitral proceedings against the latter state where the BIT was concluded before one of the Contracting States acceded to the EU but the arbitral proceedings are brought after that date. Arbitral tribunals all reject this objection in relation to the ECT, despite the position of the CJ that the TFEU takes precedence over a BIT provision on an arbitration clause. But, crucially these arbitral decisions are landmark in international investment law and international energy law because they clarify the content and function of fair and equitable treatment (FET), including legitimate expectations as an aspect of FET, and contextualise and exemplify the challenges a state may face when endeavouring to incentivise the production of electricity from RE sources, especially in order to address climate change concerns, thus experimenting in a new market sector. Given the sheer number of these cases, not all of them can be reviewed here, and this analysis focuses on four of them but takes others into account. Section 2 examines the regulatory measures that were taken in Spain as an example of the regulatory frameworks some EU Member States have used to incentivise RE, and the type of regulatory changes they have made subsequent to the making of RE investments, which led to arbitral proceedings. Section 3 discusses four landmark arbitrations under the ECT concerning RE against Spain focusing on their findings concerning FET under the ECT. Section 4 discusses how these four cases are landmark for the ECT as a treaty. Section 5 discusses their significance of international investment law, and Section 6 their significance for primary rules of public international law beyond investment protection. Section 7 offers some conclusions.

II. STATE REGULATORY MEASURES IN NEW ENERGY FIELDS: SPAIN’S RENEWABLE ENERGY EXPERIMENTATION

This section focuses on the most contentious regulatory developments in Spain’s domestic legal order. Numerous arbitrations against Spain stemmed from these measures and in some (but not all) cases the same measures were complained of by the investor. Spain liberalised its power generation industry by enacting Law 54/1997, which created a special regime of remuneration10 (Special Regime) entitling producers of electricity from renewable sources to the payment of premiums11 via a FIT. Spain adopted RD 2818/1998 to regulate the Special Regime and the Administrative Registry for Production Facilities under the Special Regime (RAIPRE) was established.12 Producers were guaranteed a premium in excess of the market

9 Judgment of 6 March 2018, Slowakische Republik (Slovak Republic) v Achmea BV, C-284/16, ECLI:EU:C:2018:158. 10 Title

IV Chapter II Law 54/1997. Art 30(4). 12 RD 2818/1998, Art 9. 11 ibid

The Renewable Energy Arbitrations under the Energy Charter Treaty 155 price for RAIPRE-registered facilities.13 In Spain’s case, the EU Renewables Directive required that Spain’s individual target for electricity from RE was 29.4 per cent by 2010.14 Since the existing Spanish measures did not result in an increase in RE investment, RD 2818/1998 was abolished and RD 436/2004 was adopted, pursuant to which tariff rates and premiums were fixed for the lifetime of each facility, subject to a reduction after 25 years.15 The tariffs and premiums were subject to an initial review after two years and every four years thereafter.16 Further, producers could sell their electricity based on the FIT or on the market at a premium.17 Spain further revised the regime by RD 661/2007 (May 2007), in order to better address the requirements of the EU Directive.18 Producers of electricity from RE that qualified under the revised Special Regime were granted priority of access in the Spanish electricity transportation and distribution grid.19 Additionally, the investors were offered the possibility to opt between two incentive schemes: (a) a FIT scheme, entailing the sale of electricity through Spain’s electricity grid in exchange for a regulated fixed tariff per each kw/h; or (b) a feed in premium (FIP) scheme pursuant to which for the sale of electricity on the electricity market Spain paid a premium in addition to the market price.20 The FIT and FIP schemes were applicable for the lifespan of the RE installation21 and were subject to revisions based on a stable index.22 Following the adoption of RD 661/2007 and until 2009, Spain ran marketing campaigns to attract RE investment (in a brochure entitled ‘The sun can be all yours’ issued by the Spanish authorities in order to highlight the incentives provided to foreign investors by RD 661/2007). Within only a few months, by August 2007, Spain had surpassed 85 per cent of its PV target, and the incentives gradually led to an electricity tariff deficit – a shortfall of revenues in the electricity system, when the tariffs for the regulated components of the retail electricity price are set below the corresponding costs borne by the energy companies. To address this, Spain took numerous regulatory measures. RD 1578/2008 was enacted to administer PV investors that registered after the RD 661/2007 regime closed. In 2009, it passed RDL 6/200923 in order to gradually reduce the tariffs deficit throughout 2009–2012 and ultimately to eradicate it by 2013.24 It introduced a Pre-Assignment Registry25 with which PV plants had to register to receive RD 661/2007 tariffs. RD 1614/2010 was also enacted and provided that the ‘revision of tariffs, premiums […] referred to in Article 44(3) of [RD 661/2007] shall not affect those facilities definitively registered in [the RAIPRE as] of 7 May 2009, or those that shall have been registered in the [Pre-Assignment Registry]’.26 13 ibid

Art 23.

14 Renewables

Directive, above (n 2) Art 3 and Annex. Decree 436/2004, Dated 12 March Establishing the Methodology for the Updating and Systematisation of the Legal and Economic Regime for Electrical and Power Production in the Special Regime, BOE No 75, 27 March 2004, Art 36(2). 16 ibid Art 40(1). 17 ibid Art 22. 18 Royal Decree 661/2007, of 25 May, by which the activity of electricity production under special regime is regulated, BOE No 126 of 26 May 2007, 22846–22886, Preamble paras (5) and (7). 19 RD 661/2001, above (n 18) Art 17(e), Annex XI. 20 ibid Art 24(1). 21 ibid Art 36. 22 ibid Art 44(1). 23 Royal Decree-Law 6/2009 of 30 April, laying down specific measures taken in the energy market and approved the special rate, BOE No 111, of 7 May 2009, 39404–39419. 24 ibid Art 3(1)–(3). 25 RDL 6/2009, above (n 23) Art 4. 26 Royal Decree 1614/2010, of 7 December, which regulates and amends certain issues related to the activity of generation of electricity from solar thermal and wind technologies, BOE No 298, 8 December 2010, 101853–101859, Art 4. 15 Royal

156 Danae Azaria As the tariff deficit did not decrease, Spain enacted more drastic measures in 2010 by three consecutive acts: RD 1565/2010,27 RD 1614/201028 and RDL 14/201029 (the 2010 RE Measures). Between 2012 and 2014 Spain enacted several measures that withdrew the entire Special Regime: RDL 1/2012,30 Law 15/201231 RDL 2/1332 RDL 9/201333 and Law 24/201334 (2012–2013 RE Measures). RDL 9/2013 repealed RD 661/2007 thus abolishing the Special Regime, and Law 24/2013 removed all remaining aspects of the Special Regime. A new regime was established by virtue of RD 413/2014 and Order IET/1045/2014 from the Ministry of Industry, Energy and Tourism. The idea behind it was a limited but reasonable return. The FIT and premium options were replaced by a Special Payment,35 which was activated upon a plant reaching a pre-set production threshold. Remuneration was capped at the amount that would be received by a ‘standard installation’ with an operational life of 25 years. Finally, tariff payments received prior to the entry into force of the new regime were counted towards the total remuneration that an installation would receive over its pre-determined operational life.

III. LANDMARK ARBITRAL AWARDS CONCERNING THE PROTECTION OF RE INVESTMENT UNDER FET

The following section discuss four cases under the ECT in the field of RE focusing on the claims concerning the breach of FET under ECT Article 10(1), and explains how their reasoning makes them landmark: Charanne v Spain (section III.A); Eiser v Spain (section III.B); Masdar v Spain (section III.C); RREEF v Spain (section III.D).

A. Charanne v Spain (2016) In 2009, Charanne BV, a Dutch company, and Construction Investments S.à.r.l., a Luxembourgish company (collectively, Charanne) became shareholders of Grupo T-Solar Global S.A, a Spanish company involved in the production and sale of electricity from PV plants (solar energy).36 Because Charanne’s investments occurred over an extended period of time, some of its plants fell within RD 661/2007 and others were governed by RD 1578/2008. Charanne commenced arbitration proceedings against Spain under ECT Article 26, based on

27 Royal Decree 1565/2010, of 19 November, which regulates and amends certain aspects related to the activity of production of electricity under the special regime, BOE No 283, of 23 November 2010, 97428–97446. 28 RD 1614/2010, above (n 26). 29 Royal Decree-Law 14/2010, of 23 December, setting emergency measures for the correction of the tariff deficit in the electricity market, BOE No 312, 24 December 2010, 106386–06394. 30 Royal Decree-Law 1/2012, of 27 January, which proceeds to the suspension of the procedures for pre-allocation of remuneration and to the elimination of economic incentives for new installations producing electricity from cogeneration, renewable energy sources and waste, BOE No 24, 28 January 2012, 8068–8072. 31 Law 15/2012, of 27 December, on tax measures for energy sustainability, BOE No 312, 28 December 2012, 8808–88096. 32 Royal Decree-Law 2/2013, of 1 February, on urgent measures in the electricity system and the financial sector, BOE No 29, 2 February 2013, 9072–9077. 33 Royal Decree-Law 9/2013, of 12 July, by which emergency measures are adopted to ensure the financial stability of the electrical system, BOE No 167, 13 July 2013, 52106–52147. 34 Law 24/2013, of 26 December, regulating the Electricity Sector, BOE No 310, 27 December 2013,105198–105294. 35 RD 661/2007, above (n 18) Amended Art 30(4). 36 Charanne and Construction Investments v Spain, Award (21 January 2016) SCC Case No V 062/2012 (Charanne v Spain) paras 143–144.

The Renewable Energy Arbitrations under the Energy Charter Treaty 157 the Stockholm Chamber of Commerce (SCC) Arbitration Rules. The Tribunal found in favour of the host ECT Contracting Party. Charanne v Spain was the first in a series of cases against Spain (and other ECT Contracting Parties) and has influenced the reasoning (in favour or against) of numerous subsequent tribunals dealing with RE claims under the ECT. Charanne argued that it had invested in Spain relying on the special regime of RD 661/2007 and RD 1578/2008 and, in particular, the right to receive concrete and revised regulated tariffs applicable to the entire net production of electricity facility in operation, which allowed the producer to calculate compensation with a high degree of certainty. These rules were directed to a limited group of investors and constituted ‘specific commitments’ by Spain.37 Further, it argued that in addition to legislation the Government of Spain promoted investment in this sector through various publicity […] which announced that returns on investment in the photovoltaic sector could reach up to 15%, and […] that there were two types of regulated tariffs, one for the first 25 years and another for a period thereafter.38

Charanne’s complaint focused only on the 2010 amendments by Spain (and excluded the 2013 amendments). Spain contended that its measures were reasonable and predictable, and that FET does not mean ‘that a legal system should be frozen, as the obligation of [FET] is not equivalent to a stabilisation clause and States can continue to legislate to respond to changing circumstances’, but that the state is prohibited from ‘acting in inequitable and unreasonable manner when legislating’.39 It argued that legitimate expectations only arise where a state makes a specific commitment, but that in the present case there was no stabilisation clause in the bilateral relationship between Charanne and Spain,40 and its advertising could not give rise to legitimate expectations because it lacked the requisite specificity.41 In relation to Charanne’s argument that Spain violated Article 10(1) by subjecting its investment to ‘regulatory instability and lack of clarity’,42 the Tribunal pronounced that it could not find that there was a breach of FET. This was because it would be required to examine all measures in their entirety, but in the case before it, it could not analyse the 2013 measures.43 In relation to Charanne’s argument on legitimate expectations, the Tribunal pronounced that the legitimate expectations of the investor is a ‘relevant factor’ for analysing whether Spain’s 2010 measures violated ‘other obligations’ in ECT Article 10(1),44 and that legitimate expectations are based on the principle of good faith under customary international law (CIL).45 The Tribunal then based its analysis on the twofold approach taken by UNCTAD’s Study on Fair and Equitable Treatment (2012), pursuant to which legitimate expectations of investors may arise: (a) by specific commitments personally made to the investor; or (b) ‘rules that are not specifically addressed to a particular investor but which are put in place with a specific aim to

37 ibid

para 297. para 299. 39 ibid para 355 (citing EDF (Services) Limited v Romania, Award (8 October 2009) ICSID Case No ARB/05/13; El Paso v Argentina, Award (31 October 2011) ICSID Case No ARB/03/15; Saluka v Czech Republic, Partial Award (17 March 2006) UNCITRAL; Parkerings Compagniet AS v Republic of Lithuania, Award (11 September 2007) ICSID Case No ARB/05/8; Electrabel v Hungary, Award (25 November 2015) ICSID Case No ARB/07/19; Continental Casualty Company v Argentina, Award (5 September 2008) ICSID Case No ARB/03/9. 40 Charanne v Spain, above (n 36) paras 357–360. 41 ibid paras 361–363. 42 ibid paras 479–480. 43 ibid para 484. 44 ibid para 486. 45 ibid. 38 ibid

158 Danae Azaria induce foreign investment and on which the foreign investor relied in making his investment’ may also give rise to legitimate expectations.46 First, the Tribunal rejected the Claimant’s argument that RD 661/2007 and RD 1578/2008 ‘may constitute or be equivalent to a specific commitment’ directed to a specific limited group of investors, including Charanne.47 For the Tribunal, the fact that they were directed to a limited group of investors does not make them […] commitments specifically directed at each investor. The rules at issue do not lose the general nature that characterizes any law or regulation by their specific scope. To convert a regulatory standard into a specific commitment of the state, by the limited character of the persons who may be affected, would constitute an excessive limitation on power of states to regulate the economy in accordance with the public interest.48

Second, the Tribunal considered whether the legal framework at the date of investment (RD 661/2007 and RD 1578/2008) could in itself create legitimate expectations.49 The Tribunal considered that legitimate expectations are not subjective beliefs of investors, but objectively ‘reasonable in the particular case with relevance to representations possibly made by the host State to induce the investment’.50 The Tribunal considered that Spain’s campaign documents could not generate Charanne’s legitimate expectations that the RE economic incentives would remain unchanged,51 because the documents lacked specificity and did not contain language indicating that the tariffs were frozen.52 Further, basing its reasoning on that of Electrabel v Hungary (that ‘subsequent changes should be made fairly, consistently and predictably, taking into account the circumstances of the investment’),53 it pronounced that ‘in the absence of a specific commitment, an investor cannot have a legitimate expectation that existing rules will not be modified’.54 It also recognised (basing itself on the reasoning of other tribunals)55 that the investors would have to have exercised a sufficient degree of due diligence regarding Spain’s legal framework prior to making the investment,56 and it should be established that following such due diligence the investors could not ‘reasonably foresee’ that the regulation could be amended.57 The Tribunal found that Charanne could have reasonably foreseen that the Spanish RE legal regime was not immutable (and thus rejected the argument of legitimate expectations),58 given the 2005 and 2006 decisions of the Spanish Supreme Court which had pronounced that investors had no ‘frozen right’ to receive economic incentives indeterminably.59 The Tribunal acknowledged that these Supreme Court decisions related to different rules than those invoked by the Claimant. However, it considered these decisions relevant, because they ‘establish the

46 ibid

para 489 (emphasis added). para 493. (emphasis added). 49 ibid para 494. 50 ibid para 495. 51 ibid paras 496–504. 52 ibid para 497. 53 It also took into account CMS Gas Transmission Co v Argentina, Award (12 May 2005) ICSID Case No ARB/01/8, and that of El Paso v Argentina, above (n 39). 54 Charanne v Spain, above (n 36) para 499–503. 55 ibid para 505. Frontier Petroleum Services Ltd v Czech Republic, Award (12 November 2010) UNCITRAL, para 287. 56 Charanne v Spain, above (n 36) para 505. 57 ibid. 58 ibid para 511. 59 ibid para 506. 47 ibid 48 ibid

The Renewable Energy Arbitrations under the Energy Charter Treaty 159 principle that national law allowed to provide, within the framework of the [Law 54/1997 of 27 November on the Electricity Sector], changes to an economic system to encourage the generation of renewable energy as it was established with RD 661/2007 and RD 1578/2008’. Rather, it found that the Claimant could have undertaken an analysis of the legal framework of its investment in Spanish law and could have understood that there was a possibility that the regulations adopted in 2007 and 2008 could be amended60 and ‘that is the level of care that would be expected of a foreign investor in a highly regulated as the energy sector, where a preliminary and comprehensive legal framework applicable to the sector analysis is essential to proceed with the investment’.61 Additionally, the Tribunal did not accept that ‘the registration to the RAIPRE gave generators an acquired right to the perception of the tariff ’.62 This was because ‘the RAIPRE was simply an administrative requirement to be able to sell energy and did not imply that the registered facilities had an acquired right to a determined compensation’.63 It thus concluded that these were not specific commitments by Spain to the investor.64 Finally, the Tribunal considered whether by amending the regulatory framework (RD 661/2007 and RD 1578/2008) the 2010 rules by themselves violated FET,65 because the legitimate expectations of the investor were frustrated by the host state unreasonably, disproportionately and against the public interest.66 The Tribunal considered that ‘an investor has a legitimate expectation that, when modifying the existing regulation based on which the investment was made, the State will not act unreasonably, disproportionately or contrary to the public interest’.67 In this respect, the Tribunal found that a legislative change is proportionate if it is ‘not capricious or unnecessary and [does] not amount to suddenly and unpredictably eliminate the essential characteristics of the existing regulatory framework’.68 Because the 2010 amendments did not eliminate the ‘essential characteristics’ of the Special Regime (‘in particular the existence of a guaranteed tariff throughout the life of the facility’),69 they were proportionate, they were based on rational economic considerations they were rational and not arbitrary;70 and were not contrary to public interest, as Spain implemented the amendments with the sole purpose of limiting its electricity tariffs deficit.71 For this reason, the Tribunal found that the 2010 amendment did not violate the ECT (FET through the spectrum of legitimate expectations).72 Arbitrator Guido Santiago Tawil dissented arguing that the system implemented by RD 661/07 and 1578/08 was not aimed at an indeterminate ‘generality’ or an imprecise or indefinite collective, but rather at a limited number of potential recipients, who had sufficient capital for investing in the industry in question and that [Spain] considered it useful for stimulation and to do it, avoiding having to use its own resources.73

60 ibid

para 507.

61 ibid. 62 ibid 63 ibid

para 509. para 510.

64 ibid. 65 ibid

para 512. paras 514–515. 67 ibid para 514. 68 ibid para 517. 69 ibid para 539. 70 ibid paras 533–534. 71 ibid para 535. 72 ibid para 539. 73 ibid para 8. 66 ibid

160 Danae Azaria [W]hen an investor complies with all the requirements established by the legislation in force to be granted a specific and particular benefit, its subsequent ignorance by the host State of the investment violates a legitimate expectation. [S]pain was empowered to modify or remove the promotional regime […]. Nevertheless, if the modification of the benefit granted to those who have invested according to this special regime […] caused damage without adequate compensation it would violate the legitimate expectations created and, consequently, [FET] protected in [ECT Article 10].74

The Tribunal’s approach in Charanne is noteworthy for the following reasons. First, it is framed around the concept of legitimate expectations, and takes a very wide approach to legitimate expectations under FET: it does not distinguish between legislative and administrative actions, and considers that the investor under FET has the legitimate expectation that, when modifying the existing regulation based on which the investment was made, the state will not act unreasonably, disproportionately or contrary to the public interest. Second, it introduces a very high due diligence threshold for the investor to have acquired legitimate expectations from a regulatory regime existing at the time that the investment is made, including that the investor should have knowledge of the domestic courts case law, even in relation to cases that do not concern the same regulatory measures on which the investor would rely to make the investment. It then explains this high threshold by reference to the ‘highly regulated as the energy sector, where a preliminary and comprehensive legal framework applicable to the sector analysis is essential to proceed with the investment’.75 Third, it considers that regulations that do not eliminate the essential characteristics of the investment, are not unreasonable, disproportionate and are taken for reasons of public interest do not violate legitimate expectations in the FET standard of the ECT. Fourth, it does not engage in the interpretation of the ECT in order to address all the above issues, but seems to rely more on the reasoning of other investment arbitrations, implicitly suggesting that FET and legitimate expectations have a meaning developed by international investment tribunals.76

B. Eiser v Spain (2017) In December 2013, Eiser Infrastructure Limited, a UK company, and Solar Energy Luxembourg S.à.r.I., a Luxembourg company (collectively, ‘Eiser’,) invested in a Spanish CSP facility (solar energy) in 2007, which began operating in 2012,77 and brought ICSID proceedings against Spain. The Tribunal found in favour of the investor, which was awarded damages of €128 million.78 Eiser v Spain concerned the 2012–2014 RE Measures (in other words, not the same measures that Charanne concerned). The Tribunal limited its analysis to the breach of FET, and did not consider Eiser’s claims about expropriation and other ECT violations.79 Eiser argued that in order to invest it relied on RD 661/2007, which contained a stabilisation clause (Article 44(3)),80 and that the drastic changes of this regime by the Spanish Government

74 ibid

para 12.

75 ibid. 76 See the wider discussion on FET in SW Schill, ‘Landmark Cases on Fair and Equitable Treatment: Empowering and Controlling Arbitrators as Law-Makers’, ch 22 in this volume. 77 Eiser Infrastructure Limited and Energía Solar Luxembourg Sà rl v the Kingdom of Spain, Award (4 May 2009) ICSID Case No ARB/13/36, paras 1–2, 120–121. 78 ibid para 486. 79 ibid para 352. 80 ibid para 357.

The Renewable Energy Arbitrations under the Energy Charter Treaty 161 violated Eiser’s legitimate expectations.81 Spain rejected that Eiser was denied FET and that the ECT had been violated,82 since Eiser’s expectations were not legitimate, given that Eiser could not reasonably anticipate that RD 661/2007 would be frozen, in the absence of a stabilisation clause and of any specific commitments undertaken by Spain.83 Further, Eiser – according to Spain – failed to exercise sufficient due diligence.84 The Tribunal found that the FET obligation under the ECT did not bar Spain from making appropriate changes to the regulatory regime of RD 661/2007, and so RD 661/2007 did not give immutable economic rights to investors.85 However, Spain’s ECT FET obligation protects investors from ‘a fundamental change to the regulatory regime in a manner that does not take account of the circumstances of existing investments made in reliance on the prior regime’,86 and ‘against the total and unreasonable change that they experienced here’.87 The Tribunal expressly distinguished the legal and factual situation in the case before it from that in Charanne v Spain. It noted that the challenged ‘measures in Charanne had far less dramatic economic effects’ as compared to the measures challenged in Eiser.88 The Tribunal agreed with Charanne v Spain in as much as ‘an investor has a legitimate expectation that, when modifying the existing regulation based on which the investment was made, the State will not act unreasonably, disproportionately or contrary to the public interest’.89 However, the Tribunal implied that this is not a matter of legitimate expectations, but rather about the violation of the FET standard as such.90 Further, while the Tribunal agreed with Charanne v Spain that combating the tariff deficit was ‘a legitimate public policy problem’, and did not question the appropriateness of Spanish authorities adopting reasonable measures to address this situation, it considered that in doing so Spain had to act in a way that would respect its ECT obligation to provide FET.91 Contrary to Charanne v Spain, the Tribunal in Eiser v Spain interpreted the ECT in order to determine the content of FET under that treaty explicitly by relying on Article 31 of the Vienna Convention on the Law of Treaties (VCLT).92 It found that in its context (especially the first sentence of ECT Article 10(1)) and in light of the object and purpose of the ECT, the FET obligation in ECT Article 10(1) second sentence necessarily embraces an obligation to provide fundamental stability in the essential characteristics of the legal regime relied upon by investors in making long-term investments. This does not mean that regulatory regimes cannot evolve. Surely they can. […] However, the Article 10(1) obligation to accord [FET] means that regulatory regimes cannot be radically altered as applied to existing investments in ways that deprive investors who invested in reliance on those regimes of their investment’s value.93

81 ibid

para 358. para 350. 83 ibid para 359. 84 ibid para 360. 85 Eiser v Spain, above (n 77) para 363. 86 ibid para 363 (emphasis added). 87 ibid (emphasis added). 88 ibid paras 368–370. 89 Charanne v Spain, above (n 36) para 514; Eiser v Spain, above (n 77) para 370. 90 Eiser v Spain, above (n 77) para 370 (‘whether viewed as basis for reasonable expectations, or as a statement of a State’s obligations under the ECT, the principle remains the same’). 91 ibid para 371. 92 ibid paras 375–384. 93 ibid para 382 (emphasis added). See also ibid para 387. 82 ibid

162 Danae Azaria On the basis of this reasoning, the Tribunal found that Spain’s measures were contrary to ECT’s FET standard, because by its 2012–2014 measures, Spain abolished the initial legal regime under which Eiser’s investment was made and replaced it with a ‘new and untested regulatory approach, all intended to significantly reduce subsidies to existing plants’,94 which retroactively applied a ‘one size fits all’ approach to installations built under the very different legal framework of RD 661/2007,95 and which reduced Eiser’s incomes by 66 per cent compared to those expected by Eiser under the initial framework.96 The Tribunal concluded that the abolition of the existing regime and its replacement with a completely different regime deprived Eiser of ‘essentially all of the value of their investment’, and that Spain violated its FET obligation under ECT Article 10(1).97 The analysis in Eiser v Spain may be considered landmark (especially in the context of earlier renewables arbitrations under the ECT) for the following reasons. First, in Eiser v Spain, the doctrine of legitimate expectations does not have a prominent place. The Tribunal rather focuses more generally on the scope of regulatory stability under FET in the ECT.98 While earlier investment arbitrations (outside the ECT) have reasoned that legitimate expectations is the ‘dominant element’ of FET,99 Eiser v Spain overtakes such an assumption. Second, methodologically its reasoning is based on the interpretation of the ECT (contrary to Charanne v Spain for instance which hardly touches on the ECT’s interpretation). Third, some of the 2013–2014 measures complained of were executive power measures (eg Royal Decree 413/2014 and Ministry implementing order IET/1045/2014), which would normally be subject to the doctrine of legitimate expectations, according to some scholars, including one of the arbitrators in this case (Professor McLachlan).100 The fact that the Tribunal was unconcerned with the doctrine of legitimate expectations in this case might imply that it did not consider the measures complained of as falling within the scope of administrative action to which the legitimate expectations doctrine applies: it considered them legislative – rather than administrative – measures,101 and did not see merit in addressing legislative measures through this doctrine. Fourth, in Eiser v Spain, the Tribunal placed emphasis on the effects of Spain’s measures on the investment (besides the degree of change of the regime that initially applied when the investment was made). While the fact that the investment is deprived in total or in a significant part of its value constitutes indirect expropriation is accepted,102 Eiser v Spain introduced this as condition for identifying a breach of FET standard.

C. Masdar v Spain (2018) Masdar Solar & Wind Cooperatief U.A. (‘Masdar’), a Danish company, invested in three CSP plants in Spain (solar energy): for the first CSP plant the investment was made in 2008, while 94 ibid

para 391. para 400. para 389. 97 ibid para 418. 98 ibid para 387. 99 Saluka Investments BV v The Czech Republic, Partial Award (17 March 2006) UNCITRAL, para 302. 100 C McLachlan, L Shore and M Weininger, International Investment Arbitration, 2nd edn (Oxford, Oxford University Press, 2017) 309–314. 101 ibid 309–314. 102 M Scherer, International Arbitration in the Energy Sector (Oxford, Oxford University Press, 2018) 10.29, 10.64; ‘Expropriation UNCTAD Series on Issues in International Investment Agreements II’ (2012) https://perma.cc/ HMY3-PB3A. 95 ibid 96 ibid

The Renewable Energy Arbitrations under the Energy Charter Treaty 163 for the remaining two in July 2009.103 Masdar initiated ICSID arbitration against Spain claiming that the latter’s measures between 2012 and 2014 violated FET under ECT Article 10(1).104 The tribunal decided in favour of Masdar and awarded full reparation.105 Masdar argued that by repealing the Special Regime by virtue of RD 661/2007, Spain removed the stability that was promised on the basis of which Masdar made its investments, and violated FET under ECT Article 10(1). Spain relied on Charanne to argue that general legislation, or press releases and others, cannot create legitimate expectations for investors. The Tribunal pronounced that the purpose of [FET] is to ensure that an investor may be confident that (i) the legal framework in which the investment has been made will not be subject to unreasonable or unjustified modification; and (ii) the legal framework will not be subject to modification in a manner contrary to specific commitments made to the investor.106

A state is free to amend its legislation absent explicit undertaking directly extended to investors.107 It then proceeded to determine which kind of specific commitments can give rise to legitimate expectations, and acknowledged the existence of two schools of thought: (a) the one that considers that ‘specific commitments’ can arise from general statements in general laws and regulations; and (b) the one pursuant to which ‘specific commitments’ have to be specific.108 Under the first school of thought, which finds support (and was expressly mentioned by the Tribunal) in the UNCTAD study on legitimate expectations,109 ‘specific commitments’ can arise from any representations comprised in general legislation in force at the date of the investment,110 provided that such legislation was enacted with the purpose of attracting investment.111 In this case, ‘the investor must demonstrate that it has exercised appropriate due diligence and that it has familiarised itself with the existing laws’.112 In the case before it, the Tribunal considered that Masdar carried out a thorough and sufficient due diligence exercise,113 and ‘that it believed that it had a legitimate expectation that the laws would not be modified, as they included stabilisation clauses’.114 In this respect, the Tribunal introduced a subjective criterion for assessing the investor’s due diligence. Further, the Tribunal found that both RD 661/2007 and RD 1614/2010, on which Masdar had relied,115 contained a ‘stabilisation clause’ pursuant to which the revision of the FIT scheme would not affect the installations registered with the Special Regime Registry. This ‘stabilisation clause’ prohibited Spain from passing any legislation modifying the legal regime relied upon by the investors.116

103 Masdar Solar & Wind Cooperatief UA v Kingdom of Spain, Award (16 May 2018) ICSID Case No ARB/14/1, para 343. 104 The tribunal was composed of Beechey, Born and Stern. 105 Considering that granting restitution would materially affect Spain’s legislative authority, the tribunal decided to grant reparation through monetary compensation: €64.5 million plus pre- and post-award compound interest. 106 Masdar v Spain, above (n 103) para 484. 107 ibid para 488. 108 ibid para 490. 109 ibid paras 491–493. 110 ibid para 491. 111 ibid paras 491–493. 112 ibid para 494. 113 ibid para 498. 114 ibid para 499 (emphasis added). 115 ibid para 502 quoting RD 1614/2010, above (n 26) Art 4. 116 Masdar v Spain, above (n 103) para 503.

164 Danae Azaria Then, the Tribunal proceeded to assess how the second school of thought would find application in the present case. According to the Tribunal, the second school of thought rejects the idea that specific commitments can arise from general regulations: a limitation on legislative power can only derive from constitutional requirements or jus cogens rules within the internal order.117 Further, specific commitment cannot arise from political announcements, such as press releases and others.118 Based on El Paso v Argentina, the Tribunal considered that a higher degree of specificity is necessary as to the addressee or regarding the object and purpose of specific commitments.119 It noted that Charanne v Spain considered that a law addressed to a limited group of persons was not a ‘specific commitment’ towards each and every one of those persons,120 and it rejected any reliance on reports, press releases and brochures aimed at attracting investors as a ‘possible legal basis for legitimate expectations’.121 It also noted that Charanne v Spain considered that the investor’s registration with RAIPRE was simply an administrative formality,122 but rejected the assessment of Charanne v Spain; it found that registration with RAIPRE is ‘a very specific unilateral offer from the State, which an investor would be deemed to have accepted once it had fulfilled the substantial condition of construction of the plant and the formal condition of registration within the prescribed “window”’.123 The Spanish authorities issued three separate resolutions confirming in explicit terms that each of Masdar’s plants had duly registered and were guaranteed to benefit from the Special Regime.124 For the Tribunal, these documents were ‘a specific commitment’125 that created legitimate expectations that the benefits granted under the Special Regime would remain unchanged. By withdrawing the Special Regime, Spain breached its FET obligations under ECT Article 10(1).126 The Tribunal focused its analysis on legitimate expectations as an aspect of FET in ECT Article 10(1). It held that only specific commitments give rise to legitimate expectations. These specific commitments can either be made through legislation containing a ‘stabilisation clause’ or through specific governmental representations. In its analysis of both options it seems to depart from established views; it introduces a subjective understanding of the investor’s ‘due diligence’ in case of commitments through legislation, and as a separate matter widens the understanding of ‘specific commitments’. It did not choose between any of these schools of thought, but found that on both of these bases Spain’s measures established legitimate expectations which were defeated by Spain’s regulatory changes.

D. RREEF v Spain RREEF Infrastructure (G.P.) Limited and RREEF Pan-European Infrastructure Two Lux S.à.r.L (RREEF) invested in wind farms and CSP plants. Contrary to earlier cases,

117 ibid

para 504. para 507. 119 ibid para 505. 120 ibid para 508 quoting Charanne v Spain, above (n 36) 493. 121 Masdar v Spain, above (n 103) para 508 quoting Charanne v Spain, above (n 36) para 497. 122 Masdar v Spain, above (n 103) para 508 quoting Charanne v Spain, above (n 36) paras 509–510. 123 Masdar v Spain, above (n 103) para 512. 124 ibid paras 516–517. 125 ibid para 520. 126 ibid para 521–522. 118 ibid

The Renewable Energy Arbitrations under the Energy Charter Treaty 165 RREEF v Spain concerned not only solar energy, but also wind energy. However, the analysis here focuses on CSP plants, where RREEF succeeded in its claim on two grounds. The Tribunal’s reasoning was based on the premise that ‘States enjoy a margin of appreciation in public international law and the exercise of such a power of appreciation must be more particularly recognized when States apply the ECT’,127 but that ‘such a margin of appreciation is not without limits’:128 it can only be exercised in so far as the state party does not violate the ECT.129 The Tribunal found that ‘while it is not expressly mentioned in Article 10(1), […] respect for the legitimate expectations of the investor is implied by this provision and is part of the FET standard’.130 It further explained that not all expectations are ‘legitimate’, and only legitimate expectations are protected under FET. Whilst an ‘expectation’ is subjective, whether or not it is ‘legitimate’ must be objectively assessed. [I]t is necessary, therefore, to assess, first, what are the expectations of an investor and, second, whether those expectations are legitimate. The frustration of a legitimate expectation establishes a wrongful act by the State.131

The Tribunal recognised that because States are in charge of the general interest and, as such, enjoy a margin of appreciation in the field of economic regulations, […] the threshold of proof as to the legitimacy of any expectation is high and only measures taken in clear violation of the FET will be declared unlawful and entail the responsibility of the State.132

It concurred with Eiser, that ‘the obligation to create a stable environment’ in ECT Article 10(1) first sentence, which is the context of FET in the second sentences, ‘excludes any unpredictable radical transformation in the conditions of the investments’.133 It also relied on Blusun v Italy, which had acknowledged that the FET obligation in the ECT sets a high threshold (when read in the context of the first sentence of ECT Article 10(1)), taking also into account earlier arbitral awards outside the ECT, which established the threshold of total alteration of the entire legal setup for foreign investments.134 In light of this reasoning, the Tribunal considered that RD 661/2007 on which RREEF relied to make its investment contained a ‘stability clause’ providing for the immutability of the conditions of the investment (Article 44(3)),135 but that its provisions and the provisions RD 1614/2010 show that adjustments were envisaged.136 As a separate matter, none of the representations invoked by the Claimants could be considered as firm pledges not to change the conditions of the investments in such a way as to neutralise the clear possibility of modification resulting from Articles 4 and 5 of RD 1614/2010.’137 The Tribunal thus considered that these were no specific commitments of immutability.

127 RREEF Infrastructure (GP) Limited and RREEF Pan-European Infrastructure Two Lux Sàrl v Spain, Decision on Responsibility and on the Principles of Quantum (30 November 2018) ICSID Case No ARB/13/30, para 242. 128 ibid para 243. 129 ibid. 130 ibid para 260. 131 ibid para 261 (emphasis added). 132 ibid para 262. 133 ibid paras 314–317. 134 ibid para 317. 135 ibid para 318. 136 ibid para 319. 137 ibid paras 321 and 390.

166 Danae Azaria Instead, it considered that the Claimants can avail themselves of an acquired right to a general regime guaranteeing the essential advantages they could reasonably expect when they made their investments.138 And that [t]he importance of the extent of the alterations suffered by the Claimants to the conditions of their investments must therefore be assessed taking into account the global balance of costs and benefits which they could reasonably expect compared with that which can be expected on the basis of the ulterior modifications.139

Relying on the Tribunal’s reasoning in Eiser v Spain, in relation to this assessment the question is whether or not the Respondent exercised its legislative power unfairly, unreasonably or inequitably, which in turn ‘depends (i) on the scope and content of the legitimate expectations of the Claimants […] and (ii) on whether or not the changes can be held as being reasonable and proportionate’.140 In this respect, the Tribunal focused exclusively on ‘whether the challenged modifications introduced after 2012 constitute ‘a drastic and radical change’ […] affecting unexpectedly the conditions of the investments’.141 Looking at the domestic regulatory regime at the time when the investment was made, the Tribunal considered that ‘the guarantee of “reasonable return” […] was the main [and only] specific commitment [to] the investors in the Special Regime,’ giving rise to a legitimate expectation.142 However, it found that such expectation did not include a guarantee to have the legal regime in place unchanged until the end of the operation of the plants, but it did include to have any modifications reasonable and equitable [even if it entails a lesser return for the Claimants, unless the new regime deprives the Claimants of a reasonable return according to the cost of money in the capital market].143

Whether such a legitimate expectation was violated can only be assessed by way of a global view of the situation that resulted from the modifications introduced by the Respondent after the date of the investment.144 More specifically, the Tribunal explained that the state’s measures have to be reasonable and proportionate. It set out the cumulative criteria by which to assess reasonableness (legitimacy of the measures’ purpose; necessity in the sense of a social need; and suitability in the sense of making it possible to achieve the legitimate purpose)145 and proportionality (ie that the regulation must be closely adjusted to the attainment of its legitimate objective, interfering as little as possible with the effective exercise of the affected rights).146 However, the Tribunal found that in the case before it ‘the determination of a violation of the principles of proportionality and reasonableness is inseparable from an assessment of the damages – if any – endured by the Claimants as a consequence of the measures taken by the Respondent’.147 The Tribunal considered that RREEFF ‘can only get compensation to the extent that such decrease is below the threshold of a reasonable return’.148 It concluded

138 ibid

para 322. (emphasis added). 140 ibid paras 323–324. 141 ibid para 379. 142 ibid paras 384, 386. 143 ibid para 517 (emphasis added). 144 ibid paras 399 and 467. 145 ibid para 464. 146 ibid para 465. 147 ibid para 475. 148 ibid para 523. 139 ibid

The Renewable Energy Arbitrations under the Energy Charter Treaty 167 that ‘with respect to [the] CSP Plants, the Respondent is in breach of its obligation to insure a reasonable return to the Claimants’ investment […]’.149 Arbitrator Volterra dissented150 focusing on the fact that Spain initially endeavoured to attract investments and subsequently, owing to the financial crisis it faced, it changed the regulatory regime affecting the investors. He implied that RREEF had a legitimate expectation of regulatory stability, in contrast with the view of the majority. The reasoning in RREEF v Spain sets a high threshold for establishing legitimate expectations, but did not address the investor’s due diligence. As a separate matter, it considered that general regulatory measures could give rise to specific commitments and by implication legitimate expectations that such commitments cannot be altered totally, unreasonably and disproportionately. It considered that RREEF was only entitled to legitimately expect that it would receive a reasonable return on its investment, not that the regulatory framework would remain unchanged.

IV. THE SIGNIFICANCE OF THE ‘RENEWABLES INVESTMENT ARBITRATIONS’ FOR THE ENERGY CHARTER TREATY

ECT investment arbitrations in the RE sector are significant for the ECT for numerous reasons. First, they may have clarified (or muddled) the content of FET under the ECT (section IV.A below). Second, they may have used different methodology in order to arrive at their determinations. It is important to understand for instance whether they relied on the CIL rules on treaty interpretation when interpreting the content of ECT Article 10(1) (section IV.B below). Third, they raise a wider question about the role of ECT arbitration tribunals for the interpretation of ECT and whether they are concerned with legal consistency among their findings. This is significant in order to encourage states and investors not only to continue to believe in ECT arbitral proceedings but also to actually continue to use them (section IV.C below).

A. The Content of FET under the ECT The central issue in the cases against Spain was the extent to which the host ECT Contracting Party may exercise its right to regulate without violating FET under ECT Article 10(1). All the decisions accepted that an ECT Contracting Party has a right to regulate, including by modifying its domestic legislation, with a view to addressing a tariff deficit and overcoming financial difficulties. In Charanne v Spain,151 Eiser v Spain,152 and Masdar v Spain153 (as well as in numerous other RE cases under the ECT, such as RREEF v Spain,154 Antin v Spain,155

149 ibid

para 600.

150 RREEF Infrastructure (GP) Limited and RREEF Pan-European Infrastructure Two Lux Sàrl v Spain, above (n 127)

Partially Dissenting Opinion of Professor Robert Volterra (30 November 2018). 151 Charanne v Spain, above (n 36) paras 500 and 536. 152 Eiser v Spain, above (n 77) para 371. 153 Masdar v Spain, above (n 103) para 484. 154 RREEF Infrastructure (GP) Limited and RREEF Pan-European Infrastructure Two Lux S.à r.l. v Kingdom of Spain, ICSID Case No ARB/13/30, Award (11 December 2019) para 242–243. 155 Antin Infrastructure Services Luxembourg S.à.r.l. and Antin Energia Termosolar BV v Kingdom of Spain, ICSID Case No ARB/13/31, Award, 15 June 2018, para 555.

168 Danae Azaria Novenergia v Spain,156 and Blusum v Italy157), the tribunals explained that the host state retains some regulatory autonomy, while being an ECT Contracting Party. They all assessed whether or not the host ECT Contracting Party’s regulatory measures exceeded its regulatory autonomy and were thus consistent or inconsistent with the FET requirement under the ECT. However, their reasoning shows some differences, including the issue about the content of FET under ECT Article 10(1), which may need further clarification in order to ensure predictability for investors. These are discussed in section III above, and some of them are highlighted below. In Charanne v Spain, the Tribunal considered whether legitimate expectations had been established for the investor on the basis of the presence of the specific commitments. In that context, investors must comply with their due diligence obligations in order to be able to claim the protection of their legitimate expectations, and the Tribunal adopted a very high threshold for assessing due diligence, which was justified by reference to the highly regulated nature of the particular market in which the investment was made, and which required the investor to be familiar with domestic case law on legislation beyond that specifically relied on by the investor. Yet, in Masdar v Spain, in relation to this approach, the Tribunal seems to introduce some level of subjectivity when assessing due diligence (on the basis of the belief of the investor). Additionally, in Charanne v Spain, the Tribunal, when considering whether by amending the regulatory framework the new regulatory regime violated FET,158 assessed whether the legitimate expectations of the investor were frustrated by the host state acting unreasonably, disproportionately and against the public interest.159 In contrast, in Eiser, the Tribunal was unconcerned with determining whether legitimate expectations had been established. It adopted a broad interpretation of the stability requirement in FET under ECT Article 10(1): the ‘obligation to accord [FET] necessarily embraces the obligation to provide fundamental stability in the essential characteristics of the legal regime relied upon by investors in making long-term investments’. When assessing whether legal stability, as required under FET in ECT Article 10(1), had been violated, it focused on the impact of the regulatory change on the investors.

B. Methodology for Determining the Meaning and Content of FET under ECT Article 10(1) The methodology of the tribunals seems to differ. Eiser v Spain focuses on the interpretation of the ECT and expressly relies on the CIL rules on treaty interpretation.160 However, Charanne v Spain does not rely on treaty interpretation rules when determining the content of FET in ECT Article 10(1) (at least not expressly), but rather on the 2012 UNCTAD Study, and international case law outside the ECT. Additionally, in Masdar v Spain the Claimant had argued at length that the FET standard in the ECT is special compared to bilateral investment treaties (BITs) and other multilateral treaties that protect investors, because it is a treaty that specifically protects investments in the energy sector requiring more stability, which is reflected in

156 Novenergia II – Energy & Environment (SCA) (Grand Duchy of Luxembourg), SICAR v The Kingdom of Spain, SCC Case No 2015/063, Award (15 Febuary 2018) paras 657–658. 157 Blusun S.A., Jean-Pierre Lecorcier and Michael Stein v Italian Republic, ICSID Case No ARB/14/3, Award (27 December 2016). 158 ibid para 512. 159 ibid paras 514–515. 160 See analysis in section III.B above.

The Renewable Energy Arbitrations under the Energy Charter Treaty 169 numerous ECT provisions, including its Article 2 setting out the ECT’s purpose.161 However, the Tribunal in Masdar v Spain did not rely on treaty interpretation, apart from making a passing reference to the general rule on treaty interpretation.162

C. The Role of Investment Arbitration in the Interpretation of the ECT The ECT differs from BITs in that it is a multilateral treaty, which establishes the jurisdiction of ad hoc arbitral tribunals for the resolution of disputes between investors and ECT Contracting Parties. Although in investment treaty arbitration there is no binding jurisprudential precedent or stare decisis,163 against such a background, the need for consistency in the interpretation of the ECT is greater than in relation to the interpretation of similar provisions in different BITs; the interpretation by an arbitral tribunal in relation to a case against one ECT Contracting Party is influential for the interpretation of the same treaty vis-à-vis other ECT Contracting Parties.164 In fact, the decisions of arbitral tribunals which have been given jurisdiction under the ECT upon the consent of the ECT Contracting Parties may be relied on as a supplementary means of interpretation reflected in the rule of Article 32 of the VCLT. Further, legal inconsistency among decisions that interpret and apply the same treaty provisions raises a ‘legitimacy crisis’ for investment treaty arbitration,165 because legal inconsistency gives the impression of bias and may discourage addressees from accepting the results of arbitral awards. The wider perception of failings of the current investment treaty arbitration regime is shown by the termination of BITs by numerous states and the denunciation (or suggestions that denunciation may follow) of the ICSID Convention, and the proposals for an international investment court,166 for an appeal stage in investment arbitration,167 or for preliminary rulings in investment arbitration.168 In the RE cases under the ECT, tribunals frequently relied on previous decisions to support their own reasoning or distinguished their reasoning from previous awards: (a) awards on disputes concerning the ECT; and (b) awards on disputes concerning other investment treaties.

161 Masdar v Spain, above (n 103) paras 391–398. A similar approach was taken in Novenergia II – Energy & Environment (SCA) (Grand Duchy of Luxembourg), SICAR v The Kingdom of Spain, Award (15 February 2018) SCC Case No 2015/063, para 654. 162 ibid para 483. 163 SGS Société Générale de Surveillance SA v Republic of the Philippines, Decision of the Tribunal on Objections to Jurisdiction (29 January 2004) ICSID Case No ARB/02/6, para 97; Gas Natural SDG, SA v The Argentine Republic, Decision on Jurisdiction (17 June 2005) ICSID Case No ARB/03/10, para 52; Garanti Koza LLP v Turkmenistan, Decision on Jurisdiction (3 July 2013) ICSID Case No ARB/11/20, para 149; A Rigo Sureda, ‘Precedent in Investment Treaty Arbitration’ in C Binder and others (eds), International Investment Law for the 21st Century: Essays in Honour of Christoph Schreuer (Oxford, Oxford University Press, 2009) 830–842. 164 For criticisms against investment arbitration generally, see also: G Van Harten, Investment Treaty Arbitration and Public Law (Oxford, Oxford University Press, 2007) 152–184. 165 S Franck, ‘The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law through Inconsistent Decisions’ (2005) 73 Fordham Law Review 1521. 166 European Commission (EC), Recommendation for a Council Decision, authorising the opening of negotiations for a Convention establishing a multilateral court for the settlement of investment disputes (13 September 2017, COM (2017) 493 final. 167 G Bottini, ‘Present and Future of ICSID Annulment: The Path to an Appellate Body?’ (2016) 31 ICSID Review – Foreign Investment Law Journal 712, 725–727; M Feldman, ‘Investment Arbitration Appellate Mechanism Options: Consistency, Accuracy, and Balance of Power’ (2017) 32 ICSID Review – Foreign Investment Law Journal, 528–544. 168 CH Schreuer, ‘Preliminary Rulings in Investment Arbitration’ (2008) 4(6) Transnational Dispute Management 3, www.transnational-dispute-management.com/article.asp?key=1188.

170 Danae Azaria For instance, in Charanne v Spain, the Tribunal based its reasoning on that of Electrabel v Hungary – that ‘subsequent changes should be made fairly, consistently and predictably, taking into account the circumstances of the investment’ – and pronounced that ‘in the absence of a specific commitment, an investor cannot have a legitimate expectation that existing rules will not be modified’.169 In Eiser v Spain, the Tribunal took a different approach in its reasoning to that of the Tribunal in Charanne v Spain, and it expressly distinguished the legal and factual situation before it from that in Charanne v Spain. While the Tribunal agreed with Charanne v Spain that combating the tariff deficit was ‘a legitimate public policy problem’, it noted that the challenged ‘measures in Charanne v Spain had far less dramatic economic effects’ as compared to the measures challenged in Eiser v Spain.170 In Masdar v Spain, the Tribunal considered that it was not bound by Charanne’s reasoning, because of the different legal framework the Tribunal in Masdar had to consider,171 and contrary to Charanne v Spain, the Tribunal considered the registration with the Special Registry requirement ‘a very specific unilateral offer from the State’.172 In RREEF, the Tribunal concurred with Eiser v Spain, that ‘the obligation to create a stable environment’ in ECT Article 10(1) first sentence ‘excludes any unpredictable radical transformation in the conditions of the investments,’173 and with Blusun v Italy.174 Further, while in Masdar v Spain the Tribunal found that the two regulations on which the investor had relied to make the investment (RD 661/2007 and RD 1614/2010) contained a ‘stabilisation clause’ that prohibited Spain from passing any legislation modifying the legal regime relied upon by the investors,175 in RREEF v Spain the Tribunal reached the opposite conclusion vis-à-vis the domestic regulations on the basis of their provisions. It considered that although the regulations contained a stabilisation clause they also foresaw that some adjustments may be made, and thus were not specific commitments. This aspect of RREEF v Spain is important since it has expressly been addressed (and contested) by subsequent arbitral tribunals in other RE arbitrations against Spain under the ECT176 – albeit not on the basis of a clear method of interpreting domestic regulations, which is the crucial matter here, but rather by focusing on the overall reasoning of RREEF v Spain. In this respect, this line of reasoning by tribunals subsequent to RREEF v Spain has given rise to criticism and dissent.177 These arbitrations highlight the need for consistency in the interpretation of the ECT, which is a multilateral treaty, interpreted by ad hoc arbitral tribunals instead of one tribunal that would set consistent jurisprudence over time. The same is needed in relation to the assessment of facts, including domestic regulations, which are identical for arbitral tribunals, in order to establish whether legitimate expectations under the ECT have been established. The reliance on a consistent methodology in relation to both these issues is necessary. This does not mean that all pronouncements by arbitrations will be identical; in fact, each tribunal should engage with and decide on the basis of the arguments and facts before it. However, the express

169 Charanne

v Spain, above (n 36) para 499–503. v Spain, above (n 77) paras 368–370. 171 Masdar v Spain, above (n 103) para 511. 172 Masdar v Spain, above (n 103) para 512. 173 ibid paras 314–317. 174 ibid para 317. 175 Masdar v Spain, above (n 103) para 503. 176 Watkins Holdings Sàrl and others v Kingdom of Spain, Award (21 January 2020) ICSID Case No ARB/15/44, paras 500–502. 177 Dissent on Liability and Quantum by Professor Ruiz Fabri, Watkins Holdings Sàrl and others v Kingdom of Spain, Award (21 January 2020) ICSID Case No ARB/15/44, paras 11–12. 170 Eiser

The Renewable Energy Arbitrations under the Energy Charter Treaty 171 ‘awareness’ and ‘deference’ to other decisions under the ECT (and other treaties) has the potential to bring about greater consistency in the interpretation of the FET standard within the ECT (and beyond). In turn, such consistency will likely assist investors in understanding their rights and ECT Contracting Parties in complying with their obligations. Although inconsistency has not been completely avoided, tribunals have been transparent about their differences from other tribunals showing mostly – albeit not only – that such differences are due to the differences of fact.178 It is quite clear that it is not argued here that arbitral tribunals should engage with the reasoning of other arbitral tribunals over and beyond the arguments made before them by the parties to the dispute.179 But, recognising that arbitral pronouncements do not take place in a vacuum and may have wider implications for the interpretation and clarity of the ECT is a positive development.

V. THE SIGNIFICANCE OF THE ‘RENEWABLES INVESTMENT ARBITRATIONS’ UNDER THE ECT FOR INTERNATIONAL INVESTMENT LAW

The RE arbitrations under the ECT concretise a shift away from earlier cases, outside the ECT – for instance under NAFTA180 or BITs – where investment arbitration tribunals took very strict – if not unworkable – approaches to the host state’s right to regulate. For instance, Tecmed v Mexico (2003) concerned a claim brought against Mexico under the Mexico-Spain BIT, which included the FET standard. The Tribunal found that: The foreign investor expects the host State to act in a consistent manner, free from ambiguity and totally transparently in its relations with the foreign investor, so that it may know beforehand any and all rules and regulations that will govern its investments, as well as the goals of the relevant policies and administrative practices or directives, to be able to plan its investment and comply with such regulations. Any and all State actions conforming to such criteria should relate not only to the guidelines, directives or requirements issued, or the resolutions approved thereunder, but also to the goals underlying such regulations. The foreign investor also expects the host State to act consistently, i.e. without arbitrarily revoking any preexisting decisions or permits issued by the State that were relied upon by the investor to assume its commitments as well as to plan and launch its commercial and business activities.181

Although this formulation of legitimate expectations has been adopted by subsequent tribunals in relation to claims under a variety of BITs,182 authors183 and subsequent investment arbitration tribunals and ad hoc annulment committees184 have criticised this approach for

178 Watkins Holdings Sà rl and others v Kingdom of Spain, Award (21 January 2020) ICSID Case No ARB/15/44, paras 499–504. 179 See also Dissent on Liability and Quantum by Professor Ruiz Fabri, Watkins Holdings Sà rl and others v Kingdom of Spain, Award (21 January 2020) ICSID Case No ARB/15/44, para 4. 180 Metalclad Corporation v The United Mexican States, Award (30 August 2000) ICSID Case No ARB(AF)/97/1. 181 Tecnicas Medioambientales Tecmed SA v Mexico, Award (29 May 2003) ICSID Case No ARB(AF)/00/2, para 154. 182 See Occidental Exploration and Production Co v Ecuador, LCIA, Award [2004] UNCITRAL Case No UN3467, para 185 (Orrego Vicuña, Brower, Sweeney); CMS Gas Transmission Co v Argentina, above (n 53) para 279 (Orrego Vicuña, Lalonde, Rezek). 183 Z Douglas, ‘Nothing If Not Critical for Investment Treaty Arbitration: Occidental, Eureko and Methanex’ (2006) 22 Arbitration International, 27, 28. 184 See also different criticism by MTD Equity Sdn Bhd and MTD Chile SA v Republic of Chile, Decision on Annulment (21 March 2007) ICSID Case No ARB/01/7, paras 67–68; Biwater Gauff (Tanzania) Ltd v United Republic of Tanzania, Award (24 July 2008) ICSID Case No ARB/05/22, para 600; El Paso Energy International Company v The Argentine Republic, Award (31 October 2011) ICSID Case No ARB/03/15, para 343.

172 Danae Azaria imposing unbearable and impractical burdens on the host state, and have departed from it. In Charanne v Spain, Eiser v Spain, and Masdar v Spain (as well as in numerous other RE cases under the ECT, such as RREEF v Spain, Antin v Spain, Novenergia v Spain, and Blusum v Italy),185 the tribunals explained that the host state retains some regulatory autonomy as an ECT Contracting Party, and did not establish the high threshold suggested in Tecmed v Mexico. It could be argued that this departure concerns the interpretation of the ECT. This could be argued to some extent in relation to Eiser v Spain where the Tribunal focused on the interpretation of the ECT. However, given that some decisions examined here – Charanne v Spain and Masdar v Spain – did not place emphasis on the interpretation of the ECT, the departure of the RE cases from earlier cases outside the ECT could also be explained as confirming a more balanced (and narrower) determination of the content of the FET standard generally by arbitration tribunals (irrespective of the treaty being interpreted) in the aftermath of Tecmed v Mexico. More generally regarding the content of FET, however, it cannot be argued that the reasoning of the tribunals discussed here (but also in other RE ECT cases) introduce one consistent methodology for making the application of FET more predictable for ECT Contracting Parties and for investors. It may be argued that this is simply the inherent character of FET under any treaty, and not only the ECT: ‘it is impossible to tie it to a definition […] only such a flexible concept can be adapted to such diversity of factual situations arising in the context of the international law of foreign investment’.186 However, it is also significant that in the context of RE arbitrations under the ECT discussed here only one decision bases its reasoning on the CIL rules on treaty interpretation (along with other arbitral decisions), and that decision does not specifically deal with the legitimate expectations aspect of the FET standard in ECT Article 10(1) but frames its analysis within regulatory stability under the FET standard in ECT Article 10(1). Further, all RE arbitration decisions under the ECT cross-refer to each other and other arbitral decisions under the ECT and outside it. This ‘dialogue’ or ‘awareness’ may offer some layer of legal consistency as to the interpretation of the FET standard more generally in different treaties. However, by not demonstrably adopting a methodology based on treaty interpretation, tribunals in RE cases under the ECT do not overcome the criticism raised against FET that it is a tool that allows arbitrators to make decisions without being bound by the ECT’s normative background.

VI. THE SIGNIFICANCE OF THE ‘RENEWABLES INVESTMENT ARBITRATIONS’ UNDER THE ECT FOR PUBLIC INTERNATIONAL LAW

In 2018, the International Court of Justice (ICJ) pronounced in its Judgment on Bolivia v Chile that references to legitimate expectations may be found in arbitral awards concerning disputes between a foreign investor and the host State that apply treaty clauses providing for fair and equitable treatment.

185 See

also section IV.A above. Tudor, The Fair and Equitable Treatment Standard in the International Law of Foreign Investment (Oxford, Oxford University Press, 2008) 237. 186 I

The Renewable Energy Arbitrations under the Energy Charter Treaty 173 It does not follow from such references that there exists in general international law a principle that would give rise to an obligation on the basis of what could be considered a legitimate expectation.187

In this way, the ICJ rejected Bolivia’s claim that Chile’s representations through its multiple declarations and statements over the years gave rise to ‘the expectation of restoring’ Bolivia’s sovereign access to the sea, and Chile’s refusal to engage in further negotiations with Bolivia ‘frustrates Bolivia’s legitimate expectations’. Bolivia had argued in this respect that the principle of legitimate expectations had been widely applied in investment arbitration.188 On the other hand, Chile had argued that there was no evidence that such a rule of international law existed.189 In its written pleadings, Chile cited Blusun v Italy (2016)190 (one of the ICJ judges was an arbitrator in Blusun v Italy)191 along with numerous other arbitral decisions outside the ECT to support the argument that legitimate expectations are not ‘norms in their own right’.192 In the passage of Blusun v Italy cited by Chile, the Tribunal in Blusun v Italy was implicitly supporting the reasoning of the Tribunal in Charanne v Spain by addressing a criticism by Counsel for Blusun targeted against Charanne v Spain.193 However, none of the RE cases under the ECT (prior or after the ICJ’s pronouncement in Bolivia v Chile) provide any support for the argument that a principle of legitimate expectations exists in general international law and applies to inter-state relationships.

VII. CONCLUSIONS

RE arbitrations under the ECT are landmark not so much because they further clarify the content and function of FET, including legitimate expectations, but rather because they confirm a shift in arbitral decisions away from the unworkable thresholds of earlier decisions that considered that FET allowed for regulatory change under very exceptional circumstances. The host state retains regulatory powers but legitimate expectations – in case of specific commitments – have to be respected or FET in general may be violated if there is a total and unreasonable change of the regulatory regime at the time of making the investment. Crucially, these arbitrations show the need for legal consistency in the interpretation of the ECT, and that investment tribunals show awareness of other ECT arbitrations and deference to the need for consistency – even when they take a different view, they explain this (usually by reference to the difference of facts before them). Further, although the cases on which this contribution has focused have not been relied upon by in inter-state disputes, one of the RE ECT arbitral decisions has been cited by one of the parties to the dispute in Bolivia v Chile (2018) before the ICJ, which shows the potential that these cases and awards have to contribute to the development of international law beyond investment law.

187 Obligation to Negotiate Access to the Pacific Ocean (Bolivia v Chile), Judgment (1 October 2018) ICJ Rep 507, para 162 (emphasis added). 188 ibid para 160. 189 ibid para 161. 190 Blusun SA, Jean-Pierre Lecorcier and Michael Stein v Italian Republic, Award (27 December 2016) ICSID Case No ARB/14/3, para 371. 191 Judge James Crawford was president, and the other two members were Alexandrov and Dupuy. 192 Rejoinder of Chile, 15 September 2017, Vol 1, available at https://perma.cc/C6RR-BYYE. 193 See Blusun v Italy, above n 190 at paras 370–371.

174 Danae Azaria However, despite dealing with legitimate expectations, the RE ETC cases examined in this chapter do not provide evidence that a principle of legitimate expectations exists in interstate relationships. Finally, placing these arbitrations against the wider background of energy security concerns and climate change objectives, these cases exemplify the challenges a state may face in its effort to encourage investment in a new market – power generation from RE.

11 Bifurcation and Trifurcation Decisions IVAN ČAVDAREVIĆ*

I. INTRODUCTION

B

IFURCATION IS A term used in international arbitration to designate situations where arbitral proceedings are split into two or more separate phases. Normally, bifurcation indicates that the proceedings are divided into two separate phases, while trifurcation is used for situations where the proceedings are divided into three phases; however, bifurcation is also commonly used in a more general sense, ie to indicate any split of the arbitral proceedings, and this is the way in which this term will be mostly used in this chapter. Although the exact term ‘bifurcation’ is rarely present in arbitral rules, this technique of dividing the proceedings into separate phases is widely accepted and considered to be a part of a tribunals’ power to conduct the proceedings in the way they deem to be appropriate. The logic behind bifurcation primarily lies in procedural efficiency and fairness, as dividing the proceedings allows the tribunal to deal separately with different groups of issues. This chapter will revisit the notion and specificities of bifurcation in investment arbitration by analysing various investment arbitral decisions on bifurcation. The central idea that will be put forward is that deciding on whether (and in which way) proceedings should be bifurcated, always entails a kind of a balancing exercise by the arbitral tribunal. Arbitral tribunals normally have a significant degree of discretion in conducting the proceedings and when deciding on bifurcation they essentially weigh the potential benefits and risks of different hypothetical scenarios. In light of the selected cases, it will be emphasised that although there are no universal rules on bifurcation, there are certain aspects that tribunals should always bear in mind. In particular, every decision on bifurcation should demonstrate that a tribunal decided by taking into account different interests and by carrying out a balancing exercise between the benefits and risks of different scenarios.

A. General Background As noted by several authors, although the practice of dividing the proceedings into separate phases is not exclusive to arbitration, it is relatively common in this field; it traditionally existed in international commercial arbitration but, in recent years, it has also become

* Ivan Čavdarević is a Research Fellow at the Max Planck Institute Luxembourg for Procedural Law.

178 Ivan Čavdarević increasingly common in investment treaty arbitration.1 However, it should be immediately stressed that arbitral rules rarely regulate the institute of bifurcation explicitly. In other words, usually there are no specific rules or clear criteria on when and under which circumstances a tribunal should (or should not) divide the proceedings. For instance, the Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (ICSID Convention) only indirectly refers to the possibility of dividing the proceedings in provisions that deal with the tribunal’s jurisdiction, and with the tribunal’s general power over the questions of procedure that are not covered under the Convention or ICSID Arbitration Rules.2 A similar situation is that of the United Nations Commission on International Trade Law (UNCITRAL) Arbitration Rules (both the 1976 and 2010 versions). None of these instruments provide any clear legal standards relevant for deciding on bifurcation.3 On the other hand, arbitration is generally characterised by considerable flexibility and most arbitral rules empower tribunals to conduct the proceedings in the way they find the most appropriate, including the power to bifurcate the proceedings. In light of the lack of explicit rules and legal standards on bifurcation, tribunals typically associate this procedural technique with more general principles of arbitration. Two central principles tribunals take into account when deciding on bifurcation are procedural efficiency and fairness.4 Procedural efficiency lies at the core of the decision to bifurcate.5 As tribunals should in general aim to conduct arbitral proceedings efficiently, bifurcation is justified when splitting the proceedings could reduce the overall time and costs of the proceedings. In practice, this essentially means the following: if addressing certain issues separately could eliminate the need for subsequent phase(s) of the proceedings – or, at least, significantly reduce or clarify the scope of issues that are to be analysed – it may be in the interest of procedural efficiency to bifurcate the proceedings.6 Similar considerations apply to the principle of fairness. If bifurcation of proceedings could eliminate the need for subsequent phases or significantly reduce the scope of issues, it would be fairer towards the parties if they are not required to plead the entire case at once. Additionally, the principle of fairness is also relevant in the context of a party’s ability to present its case properly. For example, it is possible that additional evidence, which appeared

1 A Carlevaris, ‘Preliminary Matters: Objections, Bi-furcation, Request for Provisional Measures’ in C Giorgetti (ed), Litigating International Investment Disputes: A Practitioner’s Guide (Leiden, Brill, 2014) 182; L Greenwood, ‘Does Bifurcation Really Promote Efficiency?’ (2011) 28 Journal of International Arbitration 105; and J Commission and R Moloo, Procedural Issues in Investment Arbitration (Oxford, Oxford University Press, 2018) 71. 2 ICSID Convention, Arts 41(2) and 44. 3 1976 UNCITRAL Arbitration Rules, Arts 15(1) and 21; 2010 UNCITRAL Arbitration Rules, Arts 17(1) and 23. 4 Explicit references to these principles have been made by a number of tribunals; see, eg, Accession Mezzanine Capital L.P. and Danubius Kereskedőház Vagyonkezelő v Hungary, Decision on Respondent’s Notice of Jurisdictional Objections and Request for Bifurcation (8 August 2013), ICSID Case No ARB/12/3, para 38; William Ralph Clayton, William Richard Clayton, Douglas Clayton, Daniel Clayton and the Investors of Delaware Inc v Canada, Procedural Order No 20 (5 January 2016), PCA Case No 2009-04, para 37; Cairn Energy PLC and Cairn UK Holdings Limited v India, Procedural Order No 4 (19 April 2017), PCA Case No 2016-7, para 81. 5 For the opposite view regarding the relevance of procedural efficiency and economy in bifurcation, see MV Benedettelli, ‘To Bifurcate or Not To Bifurcate? That is the (Ambiguous) Question’ (2013) 29 Arbitration International. 6 In this context, it should be mentioned that there are views that bifurcation does not in fact contribute to procedural efficiency and economy, as the empirical data arguably does not confirm the view that bifurcation limits the duration of a dispute (see Greenwood, above (n 1) 107; and A Raviv, ‘Achieving a Faster ICSID’ in JE Kalicki and A Joubin-Bret (eds), Reshaping the Investor-State Dispute Settlement System: Journeys for the 21st Century (Leiden, Brill, 2015) 687–9). While this discussion is beyond the scope of this chapter, it should be stressed that these conclusions are usually reached by simply comparing the average duration of bifurcated proceedings with the average duration of non-bifurcated ones, which fails to take into account the specific levels of complexity of different cases.

Bifurcation and Trifurcation Decisions

179

during the second phase, could have had a material impact on findings made in the previous phase.7 Therefore investment tribunals tend to conclude that it would not be in the interest of fairness to allow bifurcation when addressing the relevant issues separately may prejudice a party’s ability to present its case. At this point it should also be stressed that bifurcation in investment arbitration is often closely connected to the issue of preliminary questions. Although arbitral rules typically do not prescribe that a preliminary question has to be resolved separately, the majority of the rules empower tribunals to divide the proceedings and deal with it distinctly. For instance, Article 41(2) of the ICSID Convention leaves it up to the tribunal to decide whether it will deal with an objection as a preliminary question or join it to the merits, and this ‘choice between a preliminary decision and a joinder to the merits is a matter of procedural economy’.8 In this respect, bifurcations arising in connection to preliminary jurisdictional objections are very common in investment arbitration. When it comes to types of bifurcation, ie ways in which the proceedings can be divided, it is generally accepted that tribunals can bifurcate the proceedings in any way they consider to be appropriate.9 However, in the context of investment arbitration, there are usually two types of bifurcation. By far the most common is a bifurcation between jurisdiction and merits.10 This seems to be directly related to the high number of jurisdictional challenges in investor-state disputes. The main potential benefit of bifurcating the proceedings between jurisdiction and merits is related to procedural efficiency; if a tribunal finds that it does not have jurisdiction, the case will not proceed to the merits, which normally means there will be significant savings in cost and time. In the alternative, even if a jurisdictional objection is only partially accepted, that could still result in narrowing the scope of issues to be addressed in the merits phase. The second usual type of bifurcation is between the liability and quantum (damages) phases. Although splitting the merits phase into liability and damages does not occur very frequently in investment arbitration, it is undisputed that tribunals are not precluded from doing this.11 Tribunals usually opt for this type of bifurcation when the merits of the case are very complex and determining the damages may be expensive and time-consuming.12 The logic behind this type of bifurcation is also procedural efficiency. As one author notes, there are three basic scenarios in which this type of bifurcation can save time and costs: first, it may be determined that there is no respondent’s liability, in which case the quantum phase becomes unnecessary; second, a separate decision on liability may stimulate the parties to reach a settlement; and third, throughout the liability phase the issues of quantum may be clarified or narrowed.13 7 AJ van den Berg, ‘Organizing An International Arbitration: Practice Pointers’ in LW Newman and RD Hill (eds), The Leading Arbitrators’ Guide To International Arbitration, 3rd edn (New York, Juris Publishing, 2014) 435. 8 CH Schreuer and others, The ICSID Convention: A Commentary, 2nd edn (Cambridge, Cambridge University Press, 2009), 535, para 69; ibid 537, para 76. 9 See, eg, William Ralph Clayton, William Richard Clayton, Douglas Clayton, Daniel Clayton and the Investors of Delaware Inc v Canada, above (n 4) paras 33–5, where the Tribunal elaborated on this issue in the context of the possibility to bifurcate the quantum phase into more separate stages. 10 An empirical study conducted on 174 ICSID cases showed that, out of 45 cases that were bifurcated, 43 were split between jurisdiction and merits; see Greenwood, above (n 1) 107. 11 Suez, Sociedad General de Aguas de Barcelona S.A., and Vivendi Universal S.A v Argentina, Decision on Liability (30 June 2010), ICSID Case No ARB/03/19, para 273. 12 N Blackaby and others, Redfern and Hunter on International Arbitration, 6th edn (Oxford, Oxford University Press, 2015), 6.54; TJ Tallerico and JA Behrendt, ‘Use of Bifurcation and Direct Testimony Witness Statements in International Commercial Arbitration Proceedings’ (2003) 20 Journal of International Arbitration 297. 13 H Heilbron, ‘Accessing Damages in International Arbitration: Practical Considerations’ in LW Newman and RD Hill (eds), The Leading Arbitrators’ Guide To International Arbitration, 3rd edn (New York, Juris Publishing, 2014) 861.

180 Ivan Čavdarević In addition to these two main types of bifurcation, so-called ‘trifurcation’, which is resorted to only occasionally in investment arbitration, is another way of splitting the arbitral proceedings. As mentioned above, trifurcation represents dividing the proceeding into three separate phases, most commonly jurisdictional, liability, and quantum phases. In theory, arbitral tribunals opt for trifurcation when there are reasons both for separating jurisdiction from the merits and for separating quantum from liability. When it does take place, it is usually with the agreement of the parties.14 However, irrespective of the type of bifurcation, the question of dividing the proceedings is always a practical one. When a tribunal is faced with the question of whether or not to divide the proceedings, the central problem is that in carrying out the balancing exercise mentioned above the tribunal – at the moment of deciding on bifurcation – does not (and, normally, cannot) know what the outcome of the bifurcated phase will be. Therefore it essentially has to make its decision on assuming which scenario would be the most efficient based on the limited information available at that time.15 In the absence of explicit rules and legal standards regarding such decision, the role of a tribunal and the particular approach it adopts becomes even more important. As will be demonstrated below, when deciding on bifurcation most tribunals apply the same or very similar criteria. However, the most convincing decisions on bifurcation are those where tribunals have clearly addressed the facts of the case and issues at stake and conducted the appropriate balancing exercise. Logically, this balancing exercise entails a comparison of the potential benefits and risks of bifurcation, but, depending on the facts of each case, the particular focus can be on various issues. These are precisely the main aspects of bifurcation that this chapter will try to demonstrate through the selected cases.

II. LANDMARK DECISIONS ON BIFURCATION

The cases selected for this analysis are characterised by different levels of legal and factual complexity, so the respective decisions and tribunals’ analyses differ, but they all share some common features. Primarily, tribunals in all of these cases have genuinely engaged with the question of bifurcation, and the relevant legal and factual background instead of simply summarising the parties’ arguments and automatically applying the usual standards for bifurcation. Consequently these decisions have contributed to developing certain aspects related to the institute of bifurcation. Moreover, all of these cases illustrate various types of balancing exercises which tribunals should normally conduct when deciding whether bifurcation would increase procedural efficiency and fairness. Therefore, although some of these cases are more well-known than others (and, thus, more frequently cited in other decisions), they all serve as good examples for understanding both the main features of bifurcation and the role that tribunals have when deciding on this issue.

14 See,

eg, Fireman ’s Fund v Mexico, Award (17 July 2006), ICSID Case No. ARB(AF)/02/01, para 21. compellingly put by the tribunal in Apotex v US, a decision on bifurcation should be made by ‘weighing for both sides the benefits of procedural fairness and efficiency against the risks of delay, wasted expense and prejudice’; Apotex Hodlings Inc and Apotex Inc v US, Procedural Order Deciding Bifurcation and Non-Bifurcation (25 January 2013), ICSID Case No ARB(AF)/12/1, para 10. 15 As

Bifurcation and Trifurcation Decisions

181

A. Glamis Gold Limited v United States Glamis Gold Limited v United States16 is one of the most famous investment arbitral cases dealing with the issue of bifurcation. It has been traditionally considered as the original landmark case on the issue as it was the first one where the Tribunal tried to clearly define the criteria relevant for deciding on a request for bifurcation. As such, Glamis Gold still represents one of the most cited cases (both by parties and tribunals) on the issue of bifurcation. The bifurcation in Glamis Gold v US was requested by the Respondent, which made two preliminary objections to the Tribunal’s jurisdiction and suggested the proceedings be split into jurisdiction and merits. In a nutshell, the Respondent alleged that one part of the Claimant’s claims was time-barred, while the other was not ripe as the Claimant could not assert it had incurred a loss as a result of the Respondent’s acts.17 Since Glamis Gold was conducted under the 1976 UNCITRAL Arbitration Rules, the Tribunal in its analysis firstly stressed the general presumption in favour of deciding on its jurisdiction as a preliminary question, in accordance with Article 21(4) of these rules.18 The Tribunal referred to the drafting history of this provision and stated that the ‘primary motive for the creation of a presumption in favor of the preliminary consideration of a jurisdictional objection was to ensure efficiency in the proceedings’ (emphasis added).19 Having stressed the efficiency as the main motive behind the pro-bifurcation approach, the Tribunal then pointed out that it may decline to bifurcate the proceedings if ‘doing so is unlikely to bring about increased efficiency’(emphasis added).20 After making these preliminary observations, the Tribunal then moved to what can be considered the core of this decision. The Tribunal listed three criteria which, inter alia, may indicate if it is likely or not that bifurcation would increase the efficiency of the proceedings: (1) substantiality of the objection; (2) material reduction of the proceedings; and (3) intertwinement of the issues. The purpose of the first criterion is to differentiate substantial objections that deserve preliminary (separate) consideration, from frivolous objections that, due to the obvious lack of grounds, are ‘very unlikely to reduce the costs of, or time required for, the proceedings’.21 The Tribunal’s reasoning in this respect is clear: if a jurisdictional objection seems frivolous, it is very likely that it will not be accepted. Consequently, granting a separate phase exclusively for analysing such a frivolous objection can hardly lead to the increased efficiency of the proceedings. However, the central task in this respect is determining when an objection should be considered substantial. The Tribunal in Glamis Gold v US did not provide any particular

16 Glamis Gold Limited v United States, Procedural Order No 2 (Revised) (31 May 2005), Ad Hoc Tribunal (UNCITRAL), IIC 118 (2005); Members of the Tribunal: Michael K Young (President), David D Caron & Donald L Morgan. 17 Glamis Gold, above (n 16) para 13. 18 Art 21(4) of the 1976 UNCITRAL Arbitration Rules states that ‘[i]n general, the arbitral tribunal should rule on a plea concerning its jurisdiction as a preliminary question’ (emphasis added). This presumption in favour of bifurcation under the 1976 UNCITRAL Arbitration Rules was later confirmed in a number of other cases; see, eg, Mesa Power Group v Canada, Procedural Order No 2 (18 January 2013), PCA Case No. 2012-17, para 16; President Allende Foundation,Victor Pey Casado and Coral Pey Grebe v Chile, Decision on Respondent’s Request for Bifurcation (27 June 2018), PCA Case No 2017-30, para 100. 19 Glamis Gold, above (n 16) para 11. 20 ibid para 12. 21 ibid.

182 Ivan Čavdarević guidance in this respect, but the standard that was developed through later case law indicates that a substantial objection is one that does not appear frivolous or for which it cannot be prima facie excluded that it might be successful.22 These standpoints seem to correspond with the frivolousness standard applied in Glamis Gold. The second criterion deals with the possibility (or likelihood) that the proceedings at the next phase would be materially reduced if the objection is accepted. The obvious example for this is the following: if a tribunal concludes that it does not have jurisdiction, this should result in a material reduction since the proceedings will not go into merits. However, not all situations are as straightforward; for instance, it may be the case that issues related to a tribunal’s jurisdiction represent the major part of the dispute, while issues on the merits are not very complex. In such a scenario, it is questionable whether bifurcating the proceedings would materially reduce the overall duration and costs of the proceedings since the merits phase would not be lengthy or complex in any case. It seems that this is precisely what the Tribunal in Glamis Gold had in mind when it provided further guidance on this criterion: ‘[T]he tribunal should consider whether the costs and time required of a preliminary proceedings, even if the objecting party is successful, will be justified in terms of the reduction in costs at the subsequent phase of proceedings.’23 This is an important point because it entails a balancing exercise. In essence, the Tribunal was comparing the potential benefits with potential costs of bifurcation. Only if the benefits in the ‘best’ case scenario (ie potential reduction in costs if the jurisdictional objection is upheld) outweigh the costs and time required of the preliminary proceedings, would it be justified to bifurcate the proceedings. Finally, the third criterion concerns the nature and similarity of issues that are to be addressed in different phases of the proceedings. If issues related to the Tribunal’s jurisdiction are so intertwined with the merits of the case, it is ‘very unlikely that there will be any savings in time or cost’.24 The Tribunal did not elaborate further on this, but it is clear why bifurcation would not be efficient if there is a significant overlap between the issues on jurisdiction and on merits. Namely, even if a jurisdictional objection is upheld, the fact that there was an overlap between issues means that the preliminary phase was time-consuming as the tribunal had to deal with aspects related to the merits of the case. Hence, bifurcation in such circumstances would most probably not result in having two separate and less complex phases, but only in extending the overall duration of the proceedings.25 After laying down these general principles, the Tribunal applied them to the factual background of the case. With respect to the first preliminary objection – that part of the Claimant’s claims was time barred – the Tribunal determined that, even if this objection was accepted, the exclusion of the respective events would not ‘in the circumstances of this proceeding exclude the claim in its entirety’.26 Therefore, the Tribunal in this respect mostly relied on the second criterion described above, ie the question of the possibility of material reduction of the proceedings at the next phase. Since the Tribunal considered that the proceedings would in any

22 See, eg, Mesa Power Group v Canada, Procedural Order No 2 (18 January 2013), PCA Case No 2012-17, para 18; Philip Morris Asia Limited v Australia, Procedural Order No 8 (14 April 2014), PCA Case No 2012-12, paras 111, 119 and 125. 23 Glamis Gold, above (n 16) para 12. 24 ibid. 25 Related to this, see, eg, Gavrilović and Gavrilović d.o.o. v Croatia, Decision on Bifurcation (21 January 2015), ICSID Case No Abr/12/39, paras 79 and 93; Global Telecom Holding v Canada, Procedural Order No 2 (14 December 2017), ICSID Case No ARB/16/16, para 109. 26 Glamis Gold above (n 16) para 21.

Bifurcation and Trifurcation Decisions

183

case proceed to the merits, it concluded that ‘the cost and time of [a separate jurisdictional phase] would not be justified in terms of the reduction in costs at the subsequent phase of these proceedings’.27 Regarding the second preliminary objection – that part of the Claimant’s claims was not ripe because the Claimant allegedly had not suffered a loss as a result of the Respondent’s actions – the Tribunal firstly noted that such an objection was not ‘a plea as to jurisdiction for the purposes of Article 21(4) [of the 1976 UNCITRAL Rules]’.28 The Tribunal therefore concluded that, if the proceedings were bifurcated, already in the jurisdictional phase it would be confronted ‘with the issue of whether [the Respondent’s] laws and policies resulted in an expropriation’.29 This is clearly related to the third criterion above, ie intertwinement of the relevant issues with the merits, and the Tribunal decided that proposed bifurcation would be impractical. For these reasons, the overall conclusion reached by the Tribunal was that bifurcation of the proceedings ‘would not ultimately avoid expense for the Parties, contribute to Tribunal efficiency, or be practical’.30 As it can be seen from this summary, in the end the Tribunal in Glamis Gold did not have major difficulties in reaching its final decision on bifurcation. The circumstances of the case and the available facts made the Tribunal’s conclusion on the usefulness of bifurcation relatively straightforward. However, the real value of the decision on bifurcation in Glamis Gold lies in the Tribunal’s analysis, and in explicitly defining the three relevant criteria that may help in determining whether bifurcation is justified or not.

B. Philip Morris v Australia The second decision on bifurcation selected for this analysis was rendered in Philip Morris v Australia.31 This was also an UNCITRAL case but, unlike Glamis Gold v US, it was conducted under the 2010 UNCITRAL Arbitration Rules, and the Tribunal in Philip Morris explicitly addressed the question of differences between the two versions of the Rules. This decision on bifurcation contains very detailed reasoning, and the Tribunal has addressed the arguments presented by both parties at length. More importantly, this decision refers to Glamis Gold v US and builds upon the criteria the tribunal used in that case. Bifurcation of the proceedings in Philip Morris v Australia into jurisdiction and merits was suggested by the Respondent, who raised three preliminary objections on the Tribunal’s jurisdiction. The Respondent claimed that the Tribunal did not have jurisdiction, first, because the Claimant’s investment was allegedly not admitted by the Respondent in accordance with the applicable BIT; second, because at the time when the investment was made ‘there was already a dispute between the Philip Morris group and [the Respondent]’, due to which this dispute fell outside the scope of the BIT; and, third, because the nature of the Claimant’s activities in Australia in any case had not ‘constitute[d] investments for the purpose of the [applicable BIT]’.32 27 ibid. 28 ibid

para 23. para 25. 30 ibid paras 26 and 16. 31 Phillip Morris Asia Limited v Australia, Procedural Order No 8 Regarding Bifurcation of the Procedure (14 April 2014), Ad Hoc Tribunal (UNCITRAL), PCA Case No 2012-12; Members of the Tribunal: Karl-Heinz Böckstiegel (President), Gabrielle Kaufmann-Kohler & Donald M. McRae. 32 ibid paras 35–8, 40–1 and 44–5. 29 ibid

184 Ivan Čavdarević The Tribunal started its analysis by making several general considerations regarding bifurcation. It first noted that, unlike the previous version of UNCITRAL Rules, the applicable 2010 UNCITRAL Rules ‘can only be interpreted as giving the Tribunal a wider discretion and not providing a presumption in favour of bifurcation’.33 This distinction between the 1976 and 2010 UNCITRAL Rules regarding the (non) existence of presumption in favour of bifurcation was re-affirmed by a number of other investment tribunals.34 Second, the Tribunal stated that it would review the Respondent’s jurisdictional objections and decide on bifurcation in accordance with the following three main criteria: (1) whether the objection is prima facie serious and substantial; (2) whether the objection can be examined without prejudging or entering the merits; and (3) whether upholding the objection would dispose of all or an essential part of the claims raised.35 The listed criteria show that, although the Tribunal in Philip Morris v Australia did not use identical language, the three relevant parameters are essentially the same as those used in Glamis Gold v US. Finally, the Tribunal’s third general remark was related to the question of possible intertwinement between the issues on jurisdiction and merits. The Tribunal noted that it could not exclude the possibility that the Parties submit arguments and evidence belonging to the merits already in the jurisdictional phase, but it held that problems arising from such a scenario could be remedied. Namely, the Tribunal could decide not to consider in the first phase of the proceedings those submissions that exclusively deal with the merits. In the alternative, the Tribunal could also – after bifurcating the proceedings – decide to ‘join the objection to the merits’ in such way it would ensure that ‘any decision on the preliminary objections neither prejudges the merits nor is taken in the absence of sufficient information’.36 These points show that the Tribunal in Philip Morris v Australia adopted a rather pragmatic approach, explicitly leaving the possibility to de-bifurcate the proceedings if it turned out that issues on jurisdiction and merits were significantly intertwined. After making these general considerations, the Tribunal moved to the central question of this decision – whether bifurcation would be justified in light of the jurisdictional objections raised by the Respondent. The Tribunal conducted its analysis by addressing the three objections separately and applying the above-mentioned criteria to each of them. However, for the sake of clarity, the following paragraphs will present the Tribunal’s findings in a slightly different order, so that the focus is placed on each of the criteria used by the Tribunal. The Tribunal’s position regarding the first criterion – prima facie seriousness and substantiality of an objection – was straightforward. The Tribunal argued that an objection should be considered substantial whenever it cannot be prima facie excluded that it might be successful.37 This means that the Tribunal adopted a relatively low threshold; in principle, unless it is apparent that the objection cannot be successful (which is rarely the case in practice), the objection should be considered to be prima facie serious and substantial. This position seems to generally correspond to the frivolousness requirement used by the Tribunal in Glamis Gold, although it may be argued that UNCITRAL tribunals take slightly different

33 ibid para 101. Unlike the 1976 UCITRAL Rules, the 2010 version of UNCITRAL Rules does not contain a provision envisaging the presumption in favour of bifurcation (see above n 18). 34 See, eg, Cairn Energy PLC and Cairn UK Holdings Limited v India, above (n 4) para 72; President Allende Foundation, Victor Pey Casado and Coral Pey Grebe, above (n 18) para 100. 35 Philip Morris, above (n 31) para 109. 36 ibid para 108. 37 ibid paras 111, 119 and 125.

Bifurcation and Trifurcation Decisions

185

approaches to this issue depending on which version of the UNCITRAL Arbitral Rules is applicable.38 The second criterion – whether jurisdictional issues can be examined without prejudging or entering the merits – was addressed in great detail by the Tribunal in Philip Morris v Australia. In relation to the first objection, the Tribunal stated that, although it ‘was not persuaded’ that this would happen, it could not exclude the possibility that ‘in case of bifurcation, the Parties in the first phase would submit arguments and evidence which the Tribunal considers to belong to the merits’.39 At this point, the Tribunal referred back to the general observations it had previously made and underlined that, even if it turned out that parties in the jurisdictional phase submit arguments and evidence belonging to the merits, the Tribunal could still refrain from considering such submissions in the jurisdictional phase, or even decide to re-join the two phases of the proceedings.40 This analysis represents a balancing exercise through which the Tribunal was essentially comparing the benefits of a scenario where the proceedings were bifurcated and the objection is upheld, with the risks and costs of a scenario where the proceedings were bifurcated but it turns out that issues on jurisdiction and merits significantly overlap. The conclusion reached by the Tribunal was that the potential benefits of the former scenario – ie material reduction of the proceeding – outweighed potential risks of the latter – ie the costs of rejoining the proceedings. The Tribunal’s reasoning on this criterion was further expanded in relation to the second jurisdictional objection. The Tribunal pointed out that it was not persuaded that analysing the relevant facts in the jurisdictional phase would ‘inevitably require extensive argumentation from the Parties on numerous facts that overlap with the merits of the case’ and that, therefore, bifurcation was appropriate.41 This formulation indicates that the Tribunal took the position that not every overlap between the issues on jurisdiction and merits represents an obstacle for bifurcation; on the contrary, such an overlap has to be of a high intensity, ie it has to exist on numerous facts, so that extensive argumentation from the parties is inevitably required. Furthermore, the Tribunal also addressed the issue of potential prejudice to the parties’ ability to present their case on the merits and concluded that resolving ‘limited factual issues during a preliminary phase of a bifurcated proceedings’ would not have such an effect on the merits phase.42 In light of these conclusions, it may be argued that the Tribunal’s overall position towards these issues and the criterion of intertwinement between issues was pro-bifurcation. Finally, the third criterion used – whether upholding the objection would dispose of all or an essential part of the claims – was the least complex, and the Tribunal did not have to address it in much detail. With respect to the first two objections, the Tribunal easily concluded that upholding any of these objections would ‘dispose of the entire proceedings’. On the other hand, in relation to the third objection, the Tribunal concluded that if the objection was upheld, it would not dispose of the case.43 It is evident the Tribunal’s analysis of this criterion was less detailed than that conducted by the Tribunal in Glamis Gold v US, but this is understandable considering the different factual backgrounds of these two cases. 38 BS Vasani and SZ Vasani, ‘Bifurcation of Investment Disputes’ in Katia Yannaca-Small (ed), Arbitration Under International Investment Agreements: A Guide to the Key Issues, 2nd edn (Oxford, Oxford University Press, 2018) 304. 39 ibid paras 112 and 114. 40 ibid para 116. 41 ibid para 120. 42 ibid. 43 ibid paras 115, 122, and 125–6.

186 Ivan Čavdarević In line with these conclusions, the Tribunal decided to bifurcate the proceedings into jurisdiction and merits concerning the Respondent’s first two objections. In other words the Tribunal concluded that it would deal with the ‘non-admission of investment objection’ and ‘temporal objection’ in the first phase of the proceedings, while the third Respondent’s objection (‘no investments’) would be joined to the merits and addressed in the second phase of the proceedings (should the procedure continue to the phase).44 When it comes to the wider implications of the decision on bifurcation in Philip Morris v Australia, this case is also commonly referred to in other investment arbitral cases.45 As explained above, the Tribunal in this case reaffirmed the three main criteria relevant when deciding on bifurcation, but it also tried to further clarify them. This particularly refers to the question of the intertwinement of jurisdictional and merits issues and the possibility of examining jurisdictional issues without prejudging or entering the merits. The Tribunal has clearly taken the position that a certain degree of overlap between these issues should not automatically be seen as an obstacle for bifurcation, and that the overlap of limited factual issues does not necessarily prejudice a party’s ability to present its arguments in the following phase of the proceedings.46 Finally, the Tribunal explicitly stated that it was entitled to re-join (de-bifurcate) the proceedings after bifurcation if it subsequently turned out that would be the optimal approach.

C. Emmis v Hungary The first ICSID case that will be presented in this analysis is Emmis International, Emmis Radio Operating and MEM Magyar Electronic Media v Hungary.47 The Tribunal’s decision on bifurcation in this case is much shorter than the two decisions presented above, but it effectively demonstrates how complex issues related to bifurcation could be simplified and addressed in a very direct and concise manner. The general background of Emmis v Hungary is similar to the previous two cases; the Respondent raised a jurisdictional objection and suggested bifurcation of the proceedings so that the objection could be addressed as a preliminary issue. The Respondent challenged the Tribunal’s jurisdiction, alleging that the Claimant’s claim did not ‘arise directly out of an investment as required by Article 25 of the ICSID Convention’. The Claimants responded that these allegations were frivolous and opposed to bifurcation since determining the legal issues related to the jurisdictional objection were inextricably intertwined with the merits.48

44 ibid

para 131. eg, Gavrilović, above (n 25); Global Telecom Holding, above (n 25) para 100. 46 Related to the question of overlap, it may be interesting to highlight a case where the Tribunal dealt with the relevance of a previous arbitral award (eg conclusions made there) in the context of a bifurcation request raised in a new treaty-based arbitration. The Tribunal ultimately concluded that analysing the previous arbitral award (which is the kind of exercise that would have to be performed in the first phase of the proceeding, if they were bifurcated) would necessarily entail a substantial overlap with the merits of the present case, and that, therefore, bifurcation would not be desirable; see Lao Holdings N.V. & Sanum Investments Ltd v Lao People’s Democratic Republic, Procedural Order No 2 (23 October 2017), ICSID Case No ARB(AF)/16/2 and ADHOC/17/1, paras 34–51. 47 Emmis International Holding B.V., Emmis Radio Operating B.V. & MEM Magyar Electronic Media Kereskedelmi ES Szolgaltato KFT v Hungary, Decision on Respondent’s Application for Bifurcation (13 June 2013), ICSID Case No Arb/12/2; Members of the Tribunal: Prof. Campbell McLachlan (President), Marc Lalonde & J Christopher Thomas. 48 ibid paras 33 and 39–40. 45 See,

Bifurcation and Trifurcation Decisions

187

The Tribunal first summarised the ‘common ground’ between the parties when it comes to bifurcation. It noted that the parties agreed that ‘the overarching question [related to bifurcation] is one of procedure efficiency’, as well as the fact that they had the same position regarding the factors that might be relevant in the context of bifurcation. These are the same three criteria as in the above two cases: substantiality of the request (objection); possibility for material reduction of the proceedings; and intertwinement of the issues on jurisdiction and merits. As a matter of fact, both the Claimants and the Respondent cited the decision on bifurcation in Glamis Gold v US when arguing that these three criteria should be used, and this point was explicitly noted in the decision on bifurcation.49 The Tribunal then began its analysis by addressing and clarifying the core issue of the dispute. The only claim raised on the merits by the Claimants was related to expropriation, so the Tribunal underlined that the central aspect of this case revolved around the nature of the Claimants’ rights that were allegedly expropriated. On these grounds, the Tribunal concluded that determining ‘which investments of Claimants are capable of giving rise to their expropriation claim’ would be useful irrespective of whether the proceedings were in the end bifurcated, and whether the objection was upheld or rejected.50 In relation to this, the Tribunal clearly separated two distinct issues. The first is the question of the nature of rights and investments held by the Claimants, and this was primarily the question of law. The second issue was whether such rights – if determined to be investments – were expropriated, and this was primarily a question of facts, since addressing this issue ‘would involve a close consideration of the factual evidence […] as to Respondent’s conduct’.51 In making this overview of the aspects involved, the Tribunal laid down the foundations for its further analysis and the balancing exercise it conducted. The Tribunal compared two possible outcomes in case the proceedings were bifurcated. The first is that the jurisdictional objection is rejected, and this scenario would ‘inevitably lead to a significant delay in any merits hearings’. The second possible outcome is that jurisdictional objection is upheld, and such an outcome of the first phase of the proceedings would ‘dispose of the entire case’.52 In light of these two scenarios, by following a very pragmatic line of reasoning the Tribunal argued that the potential benefits of the latter outweighed the potential harms of the former. The Tribunal stated that since the core of the jurisdictional question was the nature of the Claimant’s rights and investments, it would be useful to resolve this legal question as early as possible. The Tribunal held that in order to resolve this question, it had to consider ‘the whole bundle of the Claimant’s rights’, and that, therefore, deferring those questions to the merits phase ‘might lead to confusion and lack of clarity on a fundamental question’ (ie the existence of an investment).53 It can be said that the Tribunal did not see the complexity of questions related to the jurisdictional objection – and a potential risk of a delay – as an obstacle to bifurcating the proceedings; on the contrary, it was precisely the complexity and the relevance of these issues that led the Tribunal to conclude that it would be more efficient to address them separately. 49 ibid

para 37. para 43. para 46. 52 ibid paras 48–9. This kind of comparison is somewhat similar to the one conducted in Philip Morris v Australia, where the Tribunal implicitly compared the benefits of a scenario where the proceedings were bifurcated and the objection is upheld, with the risks and costs of a scenario where the proceedings were bifurcated but it turns out that issues on jurisdiction and merits significantly overlap (above n 40). 53 ibid paras 49–50. 50 ibid 51 ibid

188 Ivan Čavdarević In relation to this, the Tribunal addressed another important point: the risk of overlap or duplication of the issues in separate phases of the proceedings. The Tribunal acknowledged that the ‘question of the nature of the investment can go both to jurisdiction and the merits’, so it reviewed the practical implications of addressing this question in the jurisdictional phase.54 It stressed that a ‘decision as to the nature and incidents of Claimants’ investment’ – made in the jurisdictional phase – ‘would also be final for the purpose of the merits phase of the case’.55 However, given that ‘this [factual] issue can and should be separated from the questions of whether Respondent’s acts constituted expropriation’, the Tribunal concluded the Claimants would ‘not be prejudiced by bifurcation, other than in the increased costs’ if the objection is rejected and the case proceeds into merits.56 This again reflects a particular balancing exercise through which the Tribunal concluded that the benefits of addressing a set of legal issues in the jurisdictional phase are greater than any potential harm that would exist if the jurisdictional objection is eventually rejected. On these grounds, the Tribunal decided to bifurcate the proceedings into jurisdiction and merits. The main significance of the decision on bifurcation in Emis v Hungary lies in its very pragmatic approach. The Tribunal’s analysis shows the importance of addressing and defining the main legal and factual questions very early in the proceedings. By conducting this exercise, a tribunal can make its task of deciding on bifurcation significantly easier; in particular, the Tribunal was able to easily assess the potential benefits and risks of different scenarios precisely because it had a clear picture what the main questions were. In addition to this, a point that should be noted from this decision is the Tribunal’s explicit position on the relevance of findings made in previous phases, for later phases of the proceedings.

D. Gavazzi v Romania The final case in this analysis is another ICSID Case, Marvo Gavazzi and Stefano Gavazzi v Romania.57 The Tribunal’s decision on bifurcation in this case is relatively short, but it serves as a valuable example of bifurcation into liability and quantum. Furthermore, this case illustrates the pro-active approach the Tribunal took concerning bifurcation, and conduct of the proceedings in general. The bifurcation that was initially proposed in Gavazzi v Romania was between jurisdiction and merits. The Respondent filed preliminary objections challenging the Tribunal’s jurisdiction, and, accordingly, suggested that the objections were addressed before the merits. On the other hand, the Claimants also raised objections to the admissibility and jurisdiction of the Respondent’s counterclaim but, unlike the Respondent, they opposed the proposed bifurcation and suggested that all jurisdictional and admissibility objections should be resolved together with the merits.

54 ibid

para 51. is interesting to note that on this point the Tribunal referred to the practice of the International Court of Justice and the Separate Opinion of Judge Higgins in the Oil Platforms Case (Oil Platforms, Iran v United States, Judgment (6 November 2003), ICJ Rep 161), where it was confirmed that decision made in previous phases must be definite and binding for the Tribunal; Emmis, above (n 47) paras 51–2 and 56. 56 ibid paras 55–6. 57 Marvo Gavazzi and Stefano Gavazzi v Romania, Procedural Order No 2, 13 September 2013, ICSID Case No Arb/12/25; Members of the Tribunal: Hans van Houtte (President), V V Veeder & Mauro Rubino-Sammartano. 55 It

Bifurcation and Trifurcation Decisions

189

Since the parties had opposing views as to whether the proceedings should be bifurcated into jurisdiction and merits, early in the proceedings the Tribunal carried out several very useful tasks. First, in Procedural Order No 1 the Tribunal prepared a detailed provisional schedule and timetable for two potential scenarios – for bifurcated proceedings and for joint proceedings.58 Regarding the potential duration of the different scenarios, the Tribunal eventually accepted the Respondent’s remark that bifurcated proceedings would last about two months longer than joint proceedings.59 As will be shown, the fact that Tribunal had at least a general idea about the expected duration of the different scenarios was very useful for deciding which type of bifurcation would be optimal in this case. The other important initiative that the Tribunal in Gavazzi v Romania took before deciding whether or not to bifurcate the proceedings, was to offer the parties an alternative solution. The Tribunal suggested that instead of bifurcating the proceedings between jurisdiction and merits, the proceedings could be split between liability and damages, meaning that the ‘parties preliminary objections would be joined to issues of liability, [which would be] bifurcated from issues of quantum’. The Respondent rejected this proposal and persisted with its request for bifurcation into merits and jurisdiction, while the Claimants found this alternative solution acceptable.60 Against this background, the Tribunal then reviewed whether and which type of bifurcation should be conducted. Interestingly, the Tribunal did not explicitly list the criteria relevant for deciding on bifurcation, but the Tribunal’s reasoning shows that it essentially followed the three usual criteria. The question of overlap between the issues of jurisdiction and merits was particularly in focus in this case as the Tribunal considered that the preliminary objections were closely intertwined with issues of liability. Precisely for this reason, the Tribunal concluded that resolving preliminary – jurisdictional – issues would be ‘greatly assisted by a joint proceeding on preliminary objections and liability’.61 On these grounds, the Tribunal carried out a balancing exercise comparing the potential benefits and harms of bifurcation and non-bifurcation between jurisdiction and merits. It concluded ‘that procedural fairness and efficiency would be better served without bifurcation as between jurisdiction/admissibility and liability’. The Tribunal stressed that this was the case not only because it would save time, but above all due to ‘the benefits of hearing the preliminary objections with the issues of liability’.62 Having decided not to bifurcate the proceedings between jurisdiction/admissibility and merits, the Tribunal then addressed the possibility of bifurcating the proceedings between liability and quantum, and decided that this bifurcation would be appropriate. The Tribunal did not elaborate on this in detail, but its logic in this respect is evident: since bifurcation into jurisdiction and merits was rejected and the parties’ preliminarily objections were joined to issues of liability, it made sense to separate the issue of quantum from this already complex phase (which would include both jurisdictional objections and the entire case on the merits). This is in line with the general assumption that bifurcation between liability and quantum may be desirable when the merits of a case are particularly complex. In sum, the decision on bifurcation in Gavazzi v Romania illustrates what kind of initiatives a tribunal may take when deciding on the appropriate conduct of proceedings and potential

58 Gavazzi

v Romania, Procedural Order No 1 (11 March 2013) para 13. v Romania, Procedural Order No 2 para 10. 60 ibid paras 5–7. 61 ibid para 11. 62 ibid para 12. 59 Gavazzi

190 Ivan Čavdarević bifurcation. The Tribunal in this case was not limited to the parties’ immediate requests, but took a pro-active role, suggesting an alternative solution (which is an approach itself in line with the principle of procedural efficiency).63 Also, an interesting aspect of this decision is that the Tribunal used extensive argumentation and analysis done primarily for the purposes of one type of bifurcation (jurisdiction/merits), to clarify its position and reach the final decision concerning another type of bifurcation (liability/damages).

III. CONCLUSION

Overall, the technique of splitting the proceedings into several separate phases is an exceptionally relevant procedural instrument in international arbitration, and a decision on bifurcation is arguably one of the most important procedural decisions that a tribunal can make. This is even more so the case in a particular system, such as the investment arbitration, where the same categories of actors always have the same roles in a dispute, and where tribunals are usually faced with similar kinds of questions. Since the matter of splitting the proceedings is rarely regulated in detail in international adjudication, the rich practice of investment arbitral tribunals on the matter and the standards that have been developed through the case law in this field can definitely be considered to have played a significant role in the development of the institute of bifurcation in general international law. As presented above, in the absence of explicit rules or standards for dividing the proceedings under the majority of arbitral rules, the overreaching principles of bifurcation lie in procedural efficiency and fairness. Of course, this is not to say that tribunals should not consider other interests when deciding on bifurcation, or that bifurcation is always desirable and will always bring about increased efficiency. On the contrary, decisions on bifurcation should not be about the automatic application of the criteria established through the case law, but rather about providing a convincing analysis of the legal and factual background of the case and of the interests that the tribunal considered when deciding on bifurcation. The practice of investment tribunals has crystalised the three main criteria relevant for deciding on bifurcation – (1) (prima facie) seriousness and substantiality of the objection; (2) possibility of material reduction of the proceedings; and (3) intertwinement of issues on jurisdiction and merits. Furthermore, there is a high degree of consistency among investment tribunals when it comes to reliance on these standards. However, what differentiates approaches of various tribunals is precisely how the criteria were applied, and the aim of this chapter was to present the major decisions that illustrate tribunals’ substantial engagements in determining whether or not bifurcation would be desirable.

63 A similar kind of initiative can be seen in RREEF Infrastructure (G.P.) Limited & RREEF Pan-European Infrastructure Two Lux S.à r.l. v Kingdom of Spain, where the Tribunal decided to split the merits into liability and quantum phases and then, in the quantum phase, it invited the parties, with the assistance of their experts, to reach an agreement on specific financial aspects, ie the actual return on the investments. The Tribunal’s logic was that this approach would be efficient as the parties’ experts could come up with a common or, at least, narrowed proposition, and only if they failed to reach such a proposition, would the Tribunal tackle this issue itself (Decision on Responsibility and the Principles of Quantum, 30 November 2018, ICSID Case No ARB/13/30, paras 596–7). This kind of a proactive approach was then followed by tribunals in other recent cases against Spain: see BayWa Renewable Energy GmbH & BayWa Asset Holding GmbH v Kingdom of Spain, Decision on Jurisdiction, Liability and Directions on Quantum (2 December 2019), ICSID Case No ARB/15/16, paras 616–7; and RWE Innogy GmbH & RWE Innogy Aersa S.A.U. v Kingdom of Spain, Decision on Jurisdiction, Liability, and Certain Issues of Quantum (31 December 2019), ICSID Case No ARB/14/34, paras 745–6.

Bifurcation and Trifurcation Decisions

191

The selected cases are of different types, complexities, and even outcomes, which clearly shows that there is no standard type of bifurcation, or an approach that suits all cases. However, a common feature of all the selected cases is that each tribunal explained the reasons for choosing a particular approach and the relevant standards. Moreover, each of these decisions shows a balancing exercise through which the tribunal weighed the different interests, benefits and risks of various scenarios. These are, in fact, the central aspects of a decision on bifurcation because, while it may retrospectively turn out that a different scenario would have been faster or less expensive, the tribunal’s decision on whether or not to bifurcate the proceedings should be considered appropriate as long as the tribunal took all the available steps to determine which option is most likely to be in the interest of procedural efficiency and fairness. In this context, the selected cases serve as good examples for understanding the institute of bifurcation and the main aspects that tribunals should bear in mind when deciding on this issue.

12 Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga ERIC DE BRABANDERE*

I. INTRODUCTION

P

ARTICIPATION AS AMICUS curiae implies that the intervening party is not actually a party to the dispute but is nevertheless allowed to submit a written statement during the proceedings and, less commonly, is permitted to be heard by the court or tribunal. For more than a decade, the involvement of third parties as amici curiae has been accepted in international investment arbitration. The capacity for third parties to participate as amicus curiae in international investment arbitration has been explicitly accepted under the International Centre for Settlement of Investment Disputes (ICSID) Arbitration Rules since 2006, and more recently in the 2014 United Nations Commission on International Trade Law (UNCITRAL) Rules on Transparency in Treaty-based Investor-State Arbitration and several other arbitration rules. By the end of 2019, there were 94 decisions on amicus briefs or requests to intervene as such, of which several were brought to the same tribunal. Some 67 ICSID Arbitrations and another eight non-ICSID were confronted with requests.1 In the vast majority of these requests, no information was available as to the identity of the submitter nor the outcome of the request. For those of which public information is available, 14 were submitted by the European Commission, and 25 by Non-Governmental Organisations (‘NGO’s), individuals, business corporations, and other international organisations, several of which have been submitted in the same cases. What now thus constitutes an established feature in investor-state arbitration, especially in treaty-based2 arbitration, however, has been and continues to be the subject of an evolution * Eric De Brabandere is Professor of International Dispute Settlement Law and Director, Grotius Centre for International Legal Studies, Leiden University, The Netherlands; Attorney-at-Law at the Brussels Bar (DMDB Law). I would like to thank Mr Neil Nucup, LLM Adv (Leiden) and Mr Rafael Ruschel, LLM (Leiden) for research and editorial assistance. 1 These statistics are based on the publicly available information on the website of the ICSID (https://icsid.worldbank.org/cases/case-database), the UNCTAD Investment Policy Hub – Investment Dispute Settlement Navigator (https://investmentpolicy.unctad.org/investment-dispute-settlement) and ITALAW (www.italaw.com). The statistics reflect the state of affairs on 31 December 2019, the last date on which the websites quoted above were visited. 2 The UNCITRAL Rules on Transparency in Treaty-based Investor-State Arbitration, as the name explicitly indicates, apply only to treaty-based investor-state arbitration, while the ICSID Arbitration rules apply to all ICSID arbitrations irrespective of the foundation of the arbitration agreement. I will however, for the sake of simplicity, refer generally to investor-state arbitration without distinguishing them in function of the legal instrument containing the agreement to arbitrate.

194 Eric De Brabandere to which various decisions have significantly contributed. These contributions have been made essentially in three distinct areas of amicus interventions: 1) the acceptance of the capacity for non-disputing parties to participate as amicus and the reasoning behind such acceptance; 2) the conditions under which non-disputing parties can participate as amicus; and 3) the identity and legal nature of the amicus as such. As far as the latter is concerned, an important evolution has been the ‘EU-saga’, which denotes the long list of cases in which the European Commission (EC) of the European Union (EU) has intervened or sought to intervene as amicus curiae in investment arbitrations between an EU Member State and an investor from another EU Member State (‘intra-EU’ investment disputes). In this chapter, I have identified what I consider to be ‘major decisions’, in line with the general ‘definition’ of what constitutes a ‘major decision’ by the editors of the present volume, in those three distinct areas. The discussion of the decisions, while based on a breakdown in three distinct areas of relevance, will take us from the North American Free Trade Agreement (NAFTA), the treaty under which the first case where amicus briefs were accepted was brought, to the ‘EU-saga’, which raises a series of questions concerning the identity and legal nature of the amici. Before starting the discussion, some clarifications are necessary. I will use the term ‘participation’ of amici, but it should be borne in mind that when non-disputing parties intervene as amicus in a given dispute, this participation in principle is limited to the submission of a brief in that dispute; participation does not equate to intervention in the procedure either as a party or as a non-party, as would be the case, for instance, in proceedings before the International Court of Justice (ICJ) under Articles 62 and 63 of the ICJ Statute. I will also use the term ‘third parties’ to denote the amicus, by which I mean a non-disputing party which equally is not a state party to an applicable investment treaty which provides the basis for the jurisdiction of an arbitral tribunal. This is to distinguish these third parties from other contracting parties under an investment treaty, who may have the right to intervene in arbitral proceedings.3

II. ACCEPTANCE OF AND RATIONALE BEHIND AMICUS CURIAE INTERVENTION

Before the revision of the ICSID Arbitration Rules in 2006 or other revised/adapted arbitration rules such as the 2014 UNCITRAL Rules on Transparency in Treaty-based Investor-State Arbitration,4 voluntary submissions by third parties were not regulated. Nor were procedures for such submissions initially included in applicable investment treaties, which has for several years been a more established practice.5 When investment tribunals were first confronted with voluntary submissions by third parties, tribunals which had to assess the legality and acceptability of such submissions were faced with the absence of any specific regulation or rule in that respect, both in international law generally and in the applicable rules of procedure. They have thus turned to the practice

3 See, eg, Art 1128 of the North American Free Trade Agreement [1992] which gives the right to parties to the NAFTA ‘on written notice to the disputing parties, [to] make submissions to a Tribunal on a question of interpretation of this Agreement.’ For a discussion, see E De Brabandere, ‘Amicus Curiae: Investment Arbitration’ in Hélène Ruiz Fabri (ed), Max Planck Encyclopedia of International Procedural Law (Oxford, Oxford University Press, 2019) s A. 4 UNCITRAL, Rules on transparency in treaty-based investor-State arbitration (2014), https://perma.cc/W9NA-4DWK. 5 See, eg, Art 28(3) US Model BIT (2004), https://perma.cc/E8UU-8NHD.

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

195

developed in other fields of international law, and the discretionary authority they have under arbitration rules to fill procedural gaps. The first major decision in the field of investment arbitration was the Methanex v United States decision in 2001,6 in which a NAFTA Chapter 11 Arbitral Tribunal operating under the UNCITRAL Arbitration Rules concluded that it had the power to accept amicus curiae briefs. Yet, outside of the investment law and arbitration sphere, the current – and now well-established – practice of allowing submissions by amici curiae, in investment treaty arbitration7 was influenced by a similar evolution at the World Trade Organization (WTO).8 In the Shrimp-Turtles case,9 for the first time the WTO Appellate Body confirmed that there is no rule in the WTO Dispute Settlement Understanding (DSU) that prohibits panels from accepting information voluntarily submitted to it, thereby reversing a decision by a Special Panel, which had argued otherwise.10 Contrary to several decisions in international investment law, which will now be addressed, the Appellate Body did not initially clearly state on what grounds such submissions were considered pertinent or appropriate.11 Yet the practice set in motion by the WTO Appellate Body, has influenced similar developments in international investment arbitration.

A. Methanex v United States (2001) The Methanex case12 concerned a claim by a Canadian corporation against the US under NAFTA Chapter 11 following a decision by the State of California to phase out the use of methyl tert-butyl ether (MTBE) as an additive in gasoline.13 Following the commencement of the arbitral proceedings, the Institute for Sustainable Development, an NGO, requested permission to submit an amicus curiae brief.14 The UNCITRAL Arbitration Rules do not explicitly authorise or prohibit an arbitral tribunal to accept an amicus curiae brief, but Article 17(1) UNCITRAL Arbitration Rules (equivalent to Article 15 of the 1976 UNCITRAL Arbitration Rules) affords the Tribunal wide discretion in terms of procedural rules and principles, limited only by contrary party agreement and the principle of equality. The Iran-US Claims Tribunal (IUSCT), which functions under an amended version of the UNCITRAL Arbitration Rules, had earlier adopted an interpretative note15 to Article 15 of the Rules authorising the submission of amicus curiae briefs by parties other than Iran or the US only ‘under special circumstances.’16 6 Methanex Corporation v United States, Decision on Petitions from Third Persons to Intervene as Amici Curiae (15 January 2001) 165 IIC, para 32. 7 See H Ascencio, ‘L’amicus curiae devant les juridictions internationales’ (2001) Revue générale de droit international public 897. 8 See B Stern, ‘L’intervention des tiers dans le contentieux de l’OMC’ (2003) Revue générale de droit international public 257. 9 United States – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report (adopted 12 October 1998) WT/DS58/AB/R. 10 United States – Import Prohibition of Certain Shrimp and Shrimp Products, Panel Report (adopted 15 May 1996) WT/DS58/R (Shrimp/Turtle – Report of the Panel). 11 US – Shrimp/Turtle, above (n 9) paras 102–10. 12 On the other aspects of the case, see R Howse and G Ünuvar, ‘Sowing the Seeds of an ISDS Legitimacy Crisis? The Notorious First Wave of NAFTA Chapter 11 Awards’, ch 7 in this volume. 13 Methanex Corporation v United States, Final Award on Jurisdiction and Merits (19 August 2005) 167 IIC, Part 1, Preface, para 1. 14 Methanex v United States, above (n 6) para 1. 15 Note 5 to Art 15(1) of the Tribunal Rules of Procedure, 3 May 1983, https://perma.cc/U6U7-LA3G. 16 See also Iran v United States (1988) 12 Iran-USCTR 40.

196 Eric De Brabandere Disagreeing with the Claimant, who challenged the possibility of a non-party to present an amicus brief on a number of grounds,17 the Tribunal considered that neither the UNCITRAL Arbitration Rules nor the applicable treaty – Chapter 11 of the NAFTA Agreement – contained any explicit provision in respect of amicus curiae briefs.18 The Tribunal also noted that Article 15 of the 1976 UNCITRAL Arbitration Rules gives the Tribunal much discretion in terms of procedural rules.19 The Tribunal in Methanex v United States referred to both the case law of the IUSCT and the WTO Appellate Body Report in the Shrimp/Turtle dispute. The Methanex Tribunal had also pointed out that accepting amicus curiae briefs from a party other than a disputing party is not the equivalent of adding that entity as a party to the arbitration.20 Methanex is a major decision not only for having accepted the submission of an amicus brief but also for having set out a clear rationale in support of it.21 Because of the extensive reliance by arbitral tribunals on these precise considerations of the Methanex Tribunal, and its general importance in relation to the underlying reasons behind the acceptance of amicus curiae submissions, the relevant excerpt from the award deserves to be quoted in full: There is an undoubtedly public interest in this arbitration … The substantive issues extend far beyond those raised by the usual transnational arbitration between commercial parties. This is not merely because one of the disputing Parties is a State: there are of course disputes involving States which are of no greater general public importance than a dispute between private persons. The public interest in this arbitration miles from its subject- matter-, as powerfully suggested in the Petitions. There is also a broader argument, as suggested by the Respondent and Canada: the Chapter 11 arbitral process could benefit from being perceived as more open or transparency or conversely be harmed if seen as unduly secretive. In this regard, the Tribunal’s willingness to receive amicus submissions might support the process in general and this arbitration in particular; whereas a blanket refusal could do positive harm.22

The Tribunal in this case also clearly distinguished between the general capacity of a Tribunal to accept amicus curiae briefs by NGOs, which is founded on the legal arguments mentioned above, and the appropriateness of the effective acceptance of such briefs in a particular case, which I will address in the next section.23 Since Methanex, the NAFTA Free Trade Commission has issued a statement confirming that no provision in NAFTA limits the discretionary authority of arbitral tribunals from accepting submissions of non-disputing parties24 and which further contains recommendations in terms of procedure.25

17 Methanex

v United States, above (n 6) paras 12–23. para 24. See generally P Dumberry, ‘The Admissibility of Amicus Curiae Briefs by NGOs in Investor-States Arbitration: The Precedent Set by the Methanex Case in the Context of NAFTA Chapter 11 Proceedings’ (2001) 1 Non-State Actors and International Law 3, 201. 19 Methanex v United States, above (n 6) paras 29–32. 20 ibid para 30. 21 ibid. 22 ibid para 49. 23 See discussion below in section III. 24 Statement of the Free Trade Commission on Non-Disputing Party Participation, para A.1, https://perma. cc/59H3-ZBAH. 25 ibid para B.6. 18 ibid

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

197

B. Aguas del Tunari v Bolivia (2005) As far as ICSID arbitration is concerned, it is interesting to note that arbitral tribunals were initially relatively reluctant to take part in the evolution set in motion by the Methanex tribunal and the NAFTA Free Trade Commission, even for cases initiated after the decision of the Tribunal in Methanex. The Tribunal in Aguas del Tunari v Bolivia,26 in a decision of 2005,27 refused such submissions based on a rather restrictive interpretation of the consensual nature of investment arbitration. At the same time, it should be noted that the request made by the petitioner to the Tribunal was rather extensive and far-reaching. The petitioner, Earthjustice, an NGO acting on behalf of a ‘coalition’ of other local and international NGOs, requested to intervene and be authorised to participate as a party in the ongoing dispute.28 Alternatively, if such intervention as a party was not authorised by the Tribunal, the petitioner requested the Tribunal to authorise Earthjustice to participate as amicus curiae, which they interpreted quite broadly as including not only the right to make written submissions on both jurisdiction and merits, but also to attend the hearings, to make oral presentations during the hearings, and to have access to all submissions made to the Tribunal.29 Finally, the petitioner also requested public disclosure of the submissions in the case, the general public’s access to the hearing, and that the Tribunal make an on-site visit.30 This request went far beyond a simple request to make a written submission to the Tribunal as an amicus. The Tribunal’s reply to the request contains a clear reliance on the consensual nature of the proceedings and the treaty-based character of the Tribunal’s jurisdiction. Relying on the absence of consent of the parties to the dispute, and the lack of any provision in either the ICSID Convention or the applicable Bilateral Investment Treaty (BIT) between the Netherlands and Bolivia, the Tribunal noted that it did ‘not have the power’ to join a party to the proceedings, open the hearings to third parties or the general public, or to make the documents of the proceedings public.31 While this may be perfectly understandable in relation to all requests for intervention, the refusal to authorise even the submission of a written brief by the petitioner is more difficult to reconcile with the discretion of the Tribunal under the ICSID Convention and Arbitration Rules as interpreted in subsequent cases, but of course this needs to be seen in the context of that time, as well as in light of the lack of consent by the parties to the dispute to renounce the privacy and confidentiality of the proceedings.32 However, interestingly, and this may in fact explain why the submission of a written brief by the petitioner was rejected, the Tribunal noted that it was of the opinion that there was ‘not at present a need to call witnesses or seek supplementary non-party submissions at the

26 Aguas del Tunari v Bolivia, Decision on Respondent’s Objections to Jurisdiction (21 October 2005) ICSID Case No ARB/02/3, IIC 8. 27 The Tribunal’s decision was embodied in a letter sent by the President of the Tribunal to the petitioner on 29 January 2003, appended as ‘Appendix III’ to the Tribunal’s Decision on Respondent’s Objections to Jurisdiction (21 October 2005). See also paras 15–18 of the decision, which contain a summary of the petition and letter of the President of the Tribunal. 28 Aguas del Tunari v Bolivia, above (n 26) para 16. 29 ibid. 30 ibid. 31 ibid Appendix III, 1–2. 32 ibid 2.

198 Eric De Brabandere jurisdictional phase of its work’.33 Based on this, it is more than likely that the Tribunal considered the added value of an amicus submission on its jurisdiction or on the admissibility of the claim was rather limited. The Tribunal did, however, hint at the possibility for the parties to the dispute to agree to the ‘intervention’ of a third party in the future. In this respect, the Tribunal expressly mentioned that other treaties contain explicit provisions authorising amicus submissions, which removed any formal or principled bar to the authorisation of such submissions.34

C. Suez v Argentina (2005) Following the initial reluctance of the Tribunal in Aguas del Tunari v Bolivia to transpose the developments in the UNCITRAL/NAFTA context to ICSID Arbitration, the Tribunal in Suez/Vivendi v Argentina, for the first time, accepted the authority to receive amicus curiae briefs,35 despite the clear opposition of the Claimant to this request.36 In Suez, a case which is in several respects similar to the Aguas del Tunari case, five NGOs submitted a ‘Petition for Transparency and Participation as Amicus Curiae’.37 Here also, the petition requested not only the participation as an amicus in the proceedings but also several other concessions based on the need for ‘transparency’, including access to the hearings and to all of the documents in the case. The requests that concerned participation beyond the submission of an amicus brief were rejected by the Tribunal based on clear provisions in the ICSID Convention and Arbitration Rules requiring the explicit consent of both parties to the dispute to authorise non-party access to the hearings.38 The Tribunal, however, accepted the request to submit a written brief. The authority of the Tribunal to receive and consider amicus curiae briefs was founded on Article 44 of the ICSID Convention which grants the arbitral tribunal the power to decide on procedural questions that are not regulated by the rules of the ICSID Convention.39 The Tribunal considered that the intervention of a third party as an amicus constituted a ‘procedural’ question falling within the ambit of Article 44 ICSID Convention, re-emphasised that participation as amicus curiae is not the equivalent of participation as a party to the arbitration, and reiterated that the function of amicus curiae is to provide assistance to a court or tribunal by offering expertise and arguments that the parties might not provide.40 The Tribunal also found support in and explicitly referred to the Methanex Tribunal’s interpretation of Article 15 of the UNCITRAL Rules, which the Tribunal found to be ‘substantially similar to Article 44 of the ICSID Convention’.41 33 ibid. 34 ibid. 35 Suez, Sociedad General de Aguas de Barcelona S.A. and Vivendi Universal SA v Argentina, Order in response to a petition for transparency and participation as amicus curiae (19 May 2005) ICSID Case No ARB/03/19, IIC 229. For a similar decision, which concerned a related case and was issued by a tribunal with the same composition as the case just cited, see Suez, Sociedad General de Aguas de Barcelona SA, and InterAguas Servicios Integrales del Agua SA v Argentina, Order in Response to a Petition for Participation as Amicus Curiae (17 March 2006) ICSID Case No ARB/03/17, IIC 234. 36 Vivendi Universal v Argentina, above (n 35) para 3. 37 ibid para 1. 38 ibid paras 5–7. 39 ibid para 16. 40 ibid para 13. 41 ibid para 14.

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

199

Aside from formalising the acceptance of amicus curiae briefs in ICSID Arbitrations, the Suez/Vivendi Tribunal also extensively discussed the conditions for accepting amicus curiae in that specific case. The Tribunal explains that three criteria need to be met in order for an amicus brief to be admitted: a) b) c)

the appropriateness of the subject matter of the case; the suitability of a given non-party to act as amicus curiae in that case, and the procedure by which the amicus submission is made and considered.42

In relation to the first criterion, the Tribunal noted that the case not only involved matters of public interest as does any investment treaty arbitration but also that a decision rendered in a case which concerns ‘the water distribution and sewage systems of a large metropolitan area […] has the potential to affect the operation of those systems and thereby the public they serve’.43 The Tribunal further observed, similar to the Methanex Tribunal, that the acceptance of amicus submissions ‘would have the additional desirable consequence of increasing the transparency of investor-state arbitration’, and that it would also strengthen the ‘acceptance of the legitimacy of international arbitral processes, particularly when they involve states and matters of public interest’.44 On the second criterion, that is the suitability of a given non-party to act as amicus curiae, the Tribunal considered that it could only accept submissions from ‘persons who establish to the Tribunal’s satisfaction that they have the expertise, experience, and independence to be of assistance in this case’.45 The Tribunal thus required potential amici to disclose information relating to, inter alia, the establishment of the amicus, its relationship with the parties, membership in case of an organisation, and the nature of its interest in the specific case.46 The ‘independence’ criterion, however, was not generally accepted,47 nor was it explicitly taken over in the 2006 modification of the ICSID Arbitration Rules. It has nonetheless been considered in subsequent decisions to be ‘implicit’ in the requirement to bring particular knowledge or to provide insight different from that of the parties.48 In the end, the Suez/Vivendi Tribunal accepted that the NGOs submit a single joint amicus curiae brief. The ICSID Rules on Arbitration were amended in 2006 and now require the Tribunal, in determining whether to allow the submission of a brief, to consider, inter alia, the extent to which the amicus ‘has a significant interest in the proceeding’.49

III. CONDITIONS FOR AND LIMITS OF AMICUS CURIAE INTERVENTIONS

Since the formalisation of the participation of third parties as amicus in investment arbitration in both the ICSID Arbitration and the UNICTRAL Rules on Transparency, the various 42 ibid

para 17. para 19. 44 ibid para 22. 45 ibid para 24. 46 ibid para 25. 47 See, eg, Methanex v United States, in which the Tribunal held that ‘Amici are not experts; such third persons are advocates (in the non-pejorative sense) and not ‘independent’ in that they advance a particular case to a tribunal.’ Methanex v United States, above (n 6) para 38. See also the discussion below in section IV.C in relation to the intervention of the European Commission. 48 See notably von Pezold and Others v Zimbabwe, discussed below in section III.A. 49 ICSID Rules of Procedure for Arbitration Proceedings (Arbitration Rules) at ch 4, Rule 37(2)(c). 43 ibid

200 Eric De Brabandere conditions which both enable and limit such interventions have been tested in several important cases. It has become clear that amicus curiae participation is subjected to certain conditions which are effectively applied by arbitral tribunals. The application of the formal conditions of amicus participation has resulted in several tribunals setting the limits of such interventions following a rather formal adherence to the conditions set by the applicable arbitration rules. Yet when it comes down to issues other than the mere submission of a brief, such as access to documents, Tribunals retain much discretion to regulate amicus participation beyond the applicable rules and regulations.50

A. von Pezold v Zimbabwe The von Pezold and Others v Zimbabwe case,51 which was joined to the similar Border Timbers Ltd v Zimbabwe case,52 is important not only because it was one of the first cases in which a petition to request leave to file an amicus curiae submission was denied on substantive grounds after the entry into force of the 2006 modifications to the ICSID Arbitration Rules, but also because the decision is relatively extensive and discusses all requirements which potential amici have to meet in order to be authorised to make a written submission. This case related to the challenge against the Zimbabwean Government’s actions in relation to its land reform programme.53 The European Center for Constitutional and Human Rights (ECCHR) and four indigenous communities of Zimbabwe filed a Petition for leave to make submissions as amicus curiae in the proceedings, based on the revised Rule 37(2) of the ICSID Arbitration Rules. Despite the previous agreement between the parties to the dispute that they would not allow any amicus submissions, the Tribunal considered that such an agreement had been made ‘in a general context’, and that in light of the submissions received there were ‘new circumstances’ which ‘justified the application of Rule 37(2) and a proper consideration of a potential [non-disputing party] application’.54 The petitioners were thus allowed to make a request to intervene as amicus. The Petitioners requested not only permission to make a written submission as joint amici curiae, but also access to the key arbitration documents and permission to attend the oral hearings ‘to reply to any specific questions of the Tribunals on the written submissions’.55 The indigenous communities notably argued that they had a significant interest in the cases because ‘the outcome of the present arbitral proceedings will determine not only the future rights and obligations of the disputing parties with regard to these lands, but may also potentially impact on the indigenous communities’ collective and individual rights.’56

50 See, eg, Infinito Gold Ltd v Costa Rica, Procedural Order No 2 (1 June 2016) ICSID Case No ARB/14/5, paras 41–42. 51 von Pezold and Others v Zimbabwe, ICSID Case No ARB/10/15. 52 Border Timbers Ltd v Zimbabwe, ICSID Case No ARB/10/25. 53 von Pezold and Others v Zimbabwe, Award (28 July 2015) ICSID Case No ARB/10/15, IIC 1333, paras 92–163. 54 von Pezold and Others v Zimbabwe, Procedural Order No 2 (26 June 2012) ICSID Case No ARB/10/15 and Case No ARB/10/25, para 6. 55 ibid para 14. 56 ibid para 21.

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

201

The Tribunal, in considering the application to participate as amicus, noted that it was also required that the potential submission be made by an entity that is ‘independent’ from the parties to the dispute. According to the Tribunal, this criterion was ‘implicit’ in Rule 37(2)(a), which requires that the amicus brings a ‘perspective, particular knowledge or insight that is different from that of the parties’.57 In respect of that specific requirement, the Tribunal considered that the amici lacked the required ‘independence’. Based on the petitioner’s request, the Tribunal noted that there were ‘legitimate doubts as to the independence or neutrality of the Petitioners’, which ‘is a sufficient ground to deny the [non-disputing party] Application’.58 It is not entirely clear which aspect precisely constituted the main reason for the Tribunal to doubt the independence of the Petitioners. While it had refuted several arguments made by the Claimant, the Tribunal nonetheless argued that the indigenous communities ‘appear to be in conflict with the Claimants’ primary position in these proceedings’.59 It also noted that the petitioners received support from the Nyahode Union Learning Centre, an NGO based in South-Eastern Zimbabwe, whose director, Mr Sacco, had supported the government’s actions in respect of land reform and resettlement.60 The Tribunal also further explained that the Petitioners did not meet other requirements set by Rules 37(2) of the ICSID Arbitration Rules. In particular, it noted that the request in effect sought to ‘make a submission on legal and factual issues that are unrelated to the matters before the Arbitral Tribunals’.61 The main arguments of the amici related to issues of human rights and rights of indigenous peoples,62 which the Tribunal considered to be ‘unrelated to the matters before the Arbitral Tribunals’,63 notably in light of the fact that neither party to the proceedings had invoked such arguments. For the same reason, the Tribunal considered that the requested submission would not ‘address a matter within the scope of the dispute’.64 It also argued that the amici had not demonstrated they had ‘a significant interest in the proceeding’.65 As far as the ECCHR was concerned, the Tribunal noted that its expertise was related to corporate responsibility for human rights abuses, which was not an issue in the proceedings. In respect of the indigenous communities, the Tribunal had more difficulty in arguing that they had no ‘significant interest’ since they indeed had interests in the land over which the Claimants asserted ownership. Yet, the Tribunal seems to have outweighed the potential interest in the proceedings of the indigenous communities with their alleged lack of independence, but not without noting that ‘there is a latent tension in the Rule 37(2) criteria which require that a [non-disputing party] be independent yet also possess a significant interest in the proceedings’.66 Finally, the Tribunal ruled that, irrespective of whether or not there was significant interest, allowing a submission would ‘unfairly prejudice’ the Claimant in light of the ‘circumstances surrounding these Petitioners’.67

57 ibid

para 49. para 56. 59 ibid para 51. 60 ibid para 55. 61 ibid para 57. 62 ibid. 63 ibid. 64 ibid para 60. 65 ibid para 61. 66 ibid para 62. 67 ibid. 58 ibid

202 Eric De Brabandere B. Biwater Gauff v Tanzania The amended ICSID Arbitration Rules in 2006 have had a significant impact on NGO participation in investment arbitration. In Biwater Gauff v Tanzania the parties accepted the application of this new rule to the dispute although the dispute had been initiated before the new ICSID rules entered into force.68 Biwater Gauff v Tanzania thus is the first case which in effect applied Rule 37(2) of the ICSID Arbitration Rules. Biwater Gauff Tanzania, a joint venture between UK and German investors, had won a tender related to the repair and expansion of the Dar es Salaam water and sewerage infrastructure.69 At a certain point, the Claimant considered that a series of events constituted an unlawful expropriation of its investment. Petitions for leave to submit an amicus curiae brief were submitted by five NGOs acting jointly.70 The petitioners argued that ‘this arbitration process goes far beyond merely resolving commercial or private conflicts, but rather has a substantial influence on the population’s ability to enjoy basic human rights’.71 The Tribunal, before engaging with the discussion on whether leave to submit an amicus curiae brief should be granted, made some general remarks about the scope of intervention as amicus curiae. It noted that the ICSID Rules, in fact, did not expressis verbis mention granting third parties ‘status as amicus curiae’. The Tribunal, however, noted that subject to its discretion, Rule 32(2) authorised third parties to submit a written brief, as well as to attend the hearings.72 The Tribunal also extensively relied on the orders in both the Methanex and the Suez/ Vivendi cases with respect to the public interest character of these disputes.73 The Tribunal equally noted, again quoting the Methanex decision, that even if it were admitted that there was no special or wider interest at stake, the arbitral process could generally benefit from increased transparency.74 The Tribunal accepted the submission of a single brief by the five NGOs but denied their request for access to the documents filed by the parties and their request to attend or participate in the hearing.75 In its final award, the Arbitral Tribunal noted that the interests, expertise, and perspectives given by the NGOs ‘have been demonstrated to materially differ from those of the two contending parties, and as such have provided a useful contribution to these proceedings’.76

IV. THE IDENTITY AND LEGAL NATURE OF THE AMICI

In practice, amicus curiae submissions in international investment arbitration originally have most often been made by NGOs. In fact, the reasoning behind the acceptance of amicus submissions may be seen as having been tailored to submissions by NGOs, as was discussed in 68 Biwater

Gauff v Tanzania, Procedural Order no 5 (2 February 2007) ICSID Case No ARB/05/22, para 16. the case, see Biwater Gauff v Tanzania, Award (24 July 2008) ICSID Case No ARB/05/22, paras 3ff. 70 ibid para 1. The five NGOs were ‘Lawyers’ Environmental Action Team’, ‘Legal and Human Rights Centre’, ‘Tanzania Gender Networking Programme’, ‘Center for International Environmental Law’, and ‘International Institute for Sustainable Development’. 71 ibid paras 12ff. 72 ibid para 46. 73 ibid paras 51–52. 74 ibid para 54. 75 ibid paras 62–72. 76 ibid para 359. 69 For the facts of

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

203

section II. The Tribunal in Suez/Vivendi, for instance, noted that ‘through the participation of appropriate representatives of civil society in appropriate cases, the public will gain increased understanding of ICSID processes’,77 thereby explicitly linking amicus curiae submissions to ‘civil society’. At the same time, the applicable arbitration rules and investment treaties which contain provisions on amicus curiae submissions are drafted in such a way as to be neutral towards the legal nature of the amicus.78 As a consequence, individuals and international organisations have also been accepted as amici. Notably, over the past decade, the EC has developed a consistent practice of intervening as amicus in intra-EU investment disputes.

A. Glamis Gold v United States of America (2005) The 2005 decision of the Tribunal in Glamis Gold v United States of America on a request for intervention by the ‘Quechan Indian Nation of Fort Yuma Arizona and California USA’79 is the first case in which a submission made by an entity other than an NGO was accepted. Glamis is especially relevant since only non-NGOs were allowed to make amicus submissions. Moreover, it is remarkable to note that this decision was taken very soon after the first formal acceptance of amicus curiae submissions in general, in Methanex. The Glamis Gold case concerned the alleged expropriation of gold mining rights acquired by Glamis, a Canadian company, on federal land in south-eastern California. Glamis claimed that a series of legislative and regulatory measures adopted both by the State of California and by the US were tantamount to expropriation.80 The Tribunal in Glamis decided to follow the procedure as set out in the aforementioned ‘Statement of the Free Trade Commission on non-disputing party participation’,81 and, therefore, no longer had to decide whether, if at all, it had the right to authorise and accept amicus curiae submissions. The Tribunal in Glamis explicitly grounded its authority to receive and consider briefs on the Statement without assessing the conformity of such submission with the UNCITRAL Arbitration Rules.82 Without much explanation, the Tribunal was of the opinion that the request to intervene as amicus met all conditions set in the NAFTA FTC Statement, and this authorised the submission of a brief by the Quechan Indian Nation, thus enabling indigenous peoples to participate in proceedings.83

B. Philip Morris v Uruguay The Philip Morris v Uruguay case constitutes an important decision in which third parties requested and obtained permission to intervene as amicus curiae. Not only were several 77 Vivendi

Universal v Argentina, above (n 35) para 22. K Fach Gómez, ‘Rethinking the Role of Amicus Curiae in International Investment Arbitration: How to Draw the Line Favorably for the Public Interest’ (2012) 35 Fordham International Law Journal 510, 556. 79 Glamis Gold Ltd v United States, Decision on Application and Submission by Quechan Indian Nation (16 September 2005) IIC 122. 80 See generally Glamis Gold Ltd v United States, Award (14 May 2009) IIC 380. 81 Glamis Gold v United States, above (n 79) para 10. For the ‘Statement of the Free Trade Commission on non-disputing party participation’, see above n 24. 82 Glamis Gold v United States, above (n 79) para 8. 83 See generally M Davis, ‘New Developments in International Advocacy: Amicus Curiae and the World Trade Organisation’ (2003) 5 Indigenous Law Bulletin, https://perma.cc/Q7YN-57BM. 78 See

204 Eric De Brabandere requests made by a variety of entities, including international organisations, but the submissions were also extensively discussed by the Tribunal in its decision. The case revolved around the allegation made by Claimant that the so-called ‘plain packaging’ regulations of tobacco products constituted a breach of various provisions of the applicable investment treaty. The Tribunal had received petitions to intervene as amicus from several international organisations, including the World Health Organization (WHO) and the WHO Framework Convention on Tobacco Control (WTO FCTC),84 and the Pan American Health Organization (PAHO).85 The request to submit an amicus brief was rather specific in that it explicitly provided that, because of the WHO and WHO FCTCT’s work, it was able to assist the Tribunal in the determination of factual and legal issues by providing evidence on the relation between ‘large graphic health warnings, bans on misleading branding and the protection of public health, providing facts concerning “tobacco control globally and the regulatory environment in which the Claimant operates”, thereby assisting the Tribunal in determining Claimant’s legitimate expectations’.86 Moreover, an interesting question that arose in this particular context (but also in cases involving amicus submission by the EC, as will be discussed in section V of this chapter) is whether the membership of one of the parties to the dispute, in this case Uruguay, to an international organisation, which is, in turn, seeking leave to submit a brief (WHO and FCTC Secretariat), constitutes an obstacle to the participation of that international organisation in the proceedings given its possible lack of independence.87 The Tribunal did not explicitly address this issue – it simply noted that the request to make a written submission satisfied the criteria of Rule 37(2) ICSID Arbitration Rules. This Tribunal, therefore, seems to have considered that Uruguay’s membership did not pose any specific obstacle. The Tribunal allowed the submission of a brief, and in so doing not only considered the criteria of Rule 37(2) ICSID Arbitration Rules to have been met, but also quoted the Methanex decision in relation to the presence of ‘public interest’ in relation to the subject matter of the dispute.88 It also noted that an amicus submission would ‘support the transparency of the proceeding and its acceptability by users at large’,89 language that is also reminiscent of the Methanex decision mentioned above.90 Finally, the Award in this particular case makes extensive reference to the submissions made by the WHO, the WHO FCTC Secretariat, and the PAHO.91 This stands in sharp contrast to many other decisions in which the submissions of the amici were either not discussed at all or merely briefly taken note of.92 84 Philip Morris Brand SARL and Others v Uruguay, Procedural Order no 3 (17 February 2015) ICSID Case No ARB/10/7. 85 Philip Morris Brand SARL and Others v Uruguay, Procedural Order no 4 (24 March 2015) ICSID Case No ARB/10/7. 86 Philip

Morris v Uruguay, above (n 84) para 7. para 11. A similar argument was made in respect of the request made by the PAHO. Philip Morris v Uruguay, above (n 85) para 12. 88 Philip Morris v Uruguay, above (n 84) para 26. 89 ibid para 28. Similar considerations were also made by the Tribunal in relation to the request made by the PAHO. Philip Morris v Uruguay, above (n 85) paras 28ff. 90 Methanex v United States, above (n 6) para 49. 91 See notably Philip Morris Brand SARL and Others v Uruguay, Award (8 July 2016) ICSID Case No ARB/10/7, paras 391ff. 92 For other cases in which the submissions have been discussed in more detail, see Biwater Gauff v Tanzania, above (n 69) paras 356–91; Micula and Others v Romania, Decision on Annulment (26 February 2016) ICSID Case No ARB/05/20, paras 308–339; and Electrabel v Hungary, Award (25 November 2015) ICSID Case No ARB/07/19, paras 4.89–4.193. See for a discussion of the ‘effects’ of amicus curiae submissions: De Brabandere, above (n 3), 87 ibid

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

205

V. THE EU-SAGA

The ‘EU saga’ has been at the forefront of recent discussions on the role of amicus curiae briefs in investment treaty arbitration, not only because of the specific identity of the amicus in question, but also because of the systematic nature of the EC’s efforts to file requests for leave to make a written submission in (almost) all intra-EU investment disputes. For those cases in which public information is available, almost one-third of the overall amicus submissions have been made by the EC.93 The specificity of the EC intervening in investment arbitrations in intra-EU investment disputes resides in the fact that the Commission is the executive organ of the EU, an international organisation to which both the Respondent, as well as the state of nationality of the investor are members. There is, therefore, as was the case in relation to other international organisations that have been authorised to intervene as amicus in international investment arbitrations,94 a specific link between one of the parties to the dispute and the potential amicus. Furthermore, as will be also shown, the ‘EU-saga’ presents the specificity of an amicus making a submission which in (almost) all cases, in fact, is in full and exclusive support of the Respondent state, in particular in relation to issues that concern the jurisdiction of the Tribunal.95 Finally, contrary to earlier requests to intervene as amicus in investment arbitrations which related more often than not to the public nature and public interest present in the dispute between the investor and host state, such as human rights, rights of indigenous peoples or the protection of the environment, the ‘interest’ protected in intra-EU cases are without doubt of a different nature as in other cases where ‘normally’ one has seen representatives of civil society or professional associates as amicus and is ‘a “direct legal interest” in the outcome of the dispute, as opposed to a broad public interest mandate’.96 The EC’s use of amicus curiae submissions is probably not what was foreseen when investment arbitration ‘opened up’ to third party intervention as amicus,97 although the arbitration rules are in fact neutral towards the identity and legal nature of the amicus. Although the first case in which the EC’s position in respect of the intra-EU nature of the investment dispute was made public is Eastern Sugar BV v Czech Republic,98 it seems that the Commission’s position was based only on several letters and notes sent by the Commission to the Czech Republic and internal documents and notes of the Commission from January and November 2006.99 While it is difficult to pinpoint the exact first case where the EC ‘discovered’ that it was possible to request to intervene as an amicus curiae, the first case in which the EC was allowed to intervene was Electrabel v Hungary in 2008. While the decision stricto sensu authorising the EC to submit an amicus curiae brief in AES v Hungary was taken a couple

and Wei-Chung Lin, ‘Safeguarding the Environment? The Effectiveness of Amicus Curiae Submissions in Investor-State Arbitration’ (2017) 19 International Community Law Review 270–301. 93 cf above Introduction: 14 have been submitted by the European Commission, and 25 by NGOs, individuals, business corporations, and other international organisations. 94 See notably Philip Morris v Uruguay, discussed above in section III.B. 95 See for a discussion of the specificities of the European Commission’s intervention as amicus curiae: F Dias Simões, ‘A Guardian and a Friend? The European Commission’s Participation in Investment Arbitration’ (2017) 25(2) Michigan State International Law Review 233–302. 96 ibid 265. Dias Simões also argues that ‘The EC is not acting as an advocate for social concerns, but rather as the guardian of European Union laws. Its main goal is not to promote the transparency or openness of the proceedings to civil society, but instead to assist the tribunal in reaching to a correct decision.’ ibid 279. 97 ibid 250. 98 Eastern Sugar BV v Czech Republic, Partial Award (27 March 2007) SCC Case No 088/2004, IIC 310 (2007). 99 See ibid paras 119ff.

206 Eric De Brabandere of months before the decision in Eletrabel v Hungary,100 the Electrabel v Hungary case was technically the first case in which the EC made such a request. Moreover, because of the extensive discussion of the request and the way in which the request was handled and subsequently taken into consideration, I have decided to focus mainly on that case.

A. Electrabel v Hungary (2008) Electrabel v Hungary101 concerned an investment made by Electrabel in the electricity sector in Hungary.102 Following a series of events that culminated in the termination of a ‘Power Purchase Agreement’, the Claimant contended that various acts and omissions by the Respondent had resulted in the breach of several investment protection provisions contained in the Energy Charter Treaty (ECT). The EC filed a request to seek leave to make a written submission as a non-disputing party on 13 August 2008.103 The EC considered that it could: assist the Tribunal in the determination of a number of legal issues arising under the Energy Charter Treaty in these proceedings … [and] would seek to address the question whether an ICSID Tribunal has or should exercise jurisdiction over disputes between an investor from an EU Member State against an EU Member State under the Energy Charter Treaty, including on matters that substantially fall under Community competence [, and that it would also] address the question of the applicable law in this dispute and how possible conflicts between several bodies of applicable law should be resolved.104

In particular, and aside from the issue of jurisdiction, questions as to whether the acts of Hungary were in conformity with EU rules and regulations in respect of state aid were issues of concern to and raised by the EC.105 It is interesting to note that the EC sought leave to intervene even before the Respondent had filed any objections to jurisdiction, which was precisely one of the major issues it sought to address in that submission.106 One should also bear in mind that the EC explicitly mentioned that it ‘wished to record that, whilst the Tribunal had invited it to express views as an expert commentator on European Community law, the EC was presenting its Submission as the external representative of the European Communities as a Contracting Party to the Energy Charter Treaty’.107 As a consequence, in this case more than in non-ECT cases, the EC is in fact ‘double-hatting’ in the sense that it is not only acting as a third party but also as a non-disputing party to the treaty, a more specific procedure which is sometimes available to states parties to the investment treaty forming the basis of the dispute, but who are not a party to the dispute as such.108

100 In AES v Hungary, the request to file a written submission as a non-disputing party was made on 3 September 2008 and granted on 26 November 2008 through Procedural Order No 3. See AES v Hungary, Award (23 September 2010) ICSID Case No ARB/07/22, IIC 455, para 3.22. 101 Electrabel SA v Hungary, Decision on Jurisdiction, Applicable Law and Liability (30 November 2012) ICSID Case No ARB/07/19, IIC 567. 102 ibid paras 2.4ff. 103 See Electrabel SA v Hungary, Procedural Order No 4 (28 April 2009) ICSID Case No ARB/07/19, paras 1ff. 104 ibid para 2. 105 For a discussion, see Dias Simões, above (n 95) 267ff. 106 On precisely this point, the EC’s request was refused in future cases since ‘the Commission’s First Application was premature considering that the Respondent had not yet submitted its jurisdictional objections to the Tribunal and therefore dismissed the First Application without prejudice to the Commission’s making a new request in due course’ cf Antin Infrastructure Services Luxembourg SARL and Antin Energia Termosolar BV v Spain, Award (15 June 2018) ICSID Case No ARB/13/31, IIC 1439, para 61. 107 Electrabel v Hungary, above (n 101) para 5.5. See also AES v Hungary, above (n 100) para 8.2. 108 See above section I.

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

207

The Tribunal decided, as a matter of principle that it would allow the EC to file a written submission109 but the precise contours of the authorisation to make the submission were not settled until the adoption of Procedural Order No 4 on 28 April 2009. In the meantime, and although the EC had submitted the request to intervene as amicus first in this case, the Tribunal in AES v Hungary had rendered its decision on a similar request from the EC in that case. As a consequence, the Claimant argued that ‘the scope of the European Commission’s submission, in this case, be the same as in the case of AES v Hungary’, or in the alternative that the EC would ‘[a]ddress issues of substantive EC law relating to state aid and stranded costs only (and not issues of international law nor issues relating to the ECT)’, but would ‘[n]ot address issues of the Tribunal’s jurisdiction’.110 Without much discussion of the requirements in Rule 37(2) ICSID Arbitration Rules, nor of the question of whether and why the request for leave to file an amicus brief would conform to these requirements, the Tribunal authorised the EC to file a written submission. In so doing, the Tribunal did note that the EC ‘is an expert commentator on European Community law and could accordingly assist the Tribunal by addressing several legal issues’111 and directed the EC to focus in its brief on specific issues relating to European Community Law.112 Remarkably, however, the Tribunal also ‘requested’ the parties ‘to assist the European Commission in preparing its written submission’ and provide the EC with access to information about the case, notably the (redacted) memorials of the parties.113 In the Decision on Jurisdiction, Applicable Law and Liability, considerable time was devoted to the amicus submission,114 and the Tribunal not only ‘thanked’ the EC but also noted that: it is unfortunate that the European Commission could not play a more active role as a non-disputing party in this arbitration, given that (as was rightly emphasised in the European Commission’s Submission), the European Union is a Contracting Party to the ECT in which it played from the outset a leading role; and, moreover, that the European Commission’s perspective on this case is not the same as the Respondent’s and still less that of the Claimant. In short, the European Commission has much more than ‘a significant interest’ in these arbitration proceedings.115

Therefore in this particular instance the arguments submitted by the amicus were substantively taken into account, which is not the case with amicus submissions in general.116 Finally, the Tribunal also engaged with jurisdictional objections raised by the EC (but not by the Respondent), which addressed the principle of the competence de la competence.117

109 Electrabel

SA v Hungary, Procedural Order no 1, ICSID Case No ARB/07/19, 19 November 2008. SA v Hungary, Procedural Order No. 4, ICSID Case No ARB/07/19, 28 April 2009, para 16. 111 ibid para 24. 112 ibid. 113 ibid paras 27–29. The majority of the other requests made by the European Commission to intervene as amicus were denied. See, eg, AES Summit Generation Ltd and Another v Hungary, Award (23 September 2010) ICSID Case No ARB/07/22, IIC 455, para 3.22, and more recently Antin v Spain, above (n 106) para 64. In Achmea v Slovakia, the Tribunal only transmitted the Notice of Arbitration and Statement of Claim to the Commission (Achmea BV v Slovakia, Award on Jurisdiction, Arbitrability and Suspension (26 October 2010) PCA Case No 2008-13, para 33). For a discussion of the cases in which Tribunals have been far more reluctant to make documentation and submissions from the parties available to the European Commission, see Dias Simões, above (n 95) 289ff. 114 See Electrabel v Hungary, above (n 101) paras 4.89–4.110. 115 ibid para 4.91. 116 This is however not the only case in which the EC’s arguments were extensively reproduced: see, eg, Achmea v Slovakia, above (n 113) 44–57. For a discussion of cases in which the amicus submissions were subjected to much less discussion, and most notably outside of the EU context, see E De Brabandere, ‘NGOs and the ‘Public Interest’: The Legality and Rationale of Amicus Curiae Interventions in International Economic and Investment Disputes’ (2011) 12 Chicago Journal of International Law 85–113. 117 Electrabel v Hungary, above (n 101) para 4.115. 110 Electrabel

208 Eric De Brabandere In its decision on quantum and costs, the Tribunal acknowledged, amongst other factors, that the Claimant had ‘to expend much time and cost’ to deal with the European Commission submissions,118 and that ‘far from exercising the traditional role of an “amicus curiae”, the Commission became a second respondent more hostile to Electrabel than Hungary itself.’119 The Tribunal, however, did not make any specific arrangements as to costs in light of this finding.

B. Beyond Electrabel v Hungary Beyond the decisions in Electrabel v Hungary, several other decisions in the ‘EU-saga’ are worth a brief mention, not least for having either mitigated or departed from the decisions in Electrabel. First of all, in the now well-known Achmea v Slovakia case, the Tribunal itself had requested that not only the EC, but also the Netherlands as the home state of the investor file submissions as amici curiae, based on a suggestion to this effect by Slovakia.120 Second, the decisions in Electrabel v Hungary can be contrasted with other, more recent decisions in which Tribunals have been more cautious not only with the attention given to the submission but also to the time and costs involved for the parties to the dispute. In Micula v Romania, the Tribunal considered that: In granting leave to the European Community to participate as a non-disputing party, the Arbitral Tribunal is mindful of the need to preserve due process and the good order of the proceeding. In particular, the European Community shall act as amicus curiae and not as amicus actoris vel rei. In other words, the non-disputing party shall remain a friend of the court and not a friend of either Party.121

More recently, in Belenergia v Italy, the Tribunal considered that: The Tribunal has given due consideration to the amicus curiae Brief (‘Amicus Brief’) of the European Commission, which has proven useful. The Tribunal therefore thanks the European Commission for its Amicus Brief. The Tribunal highlights, however, that the European Commission is not a Party to this arbitration. The Tribunal will therefore respond only to the arguments made by the Parties, taking into consideration the observations of the European Commission.122

In Antin v Spain, the Tribunal even went as far as accepting a brief only ‘upon the submission of a written undertaking that it would comply with any decision on costs ordered by the Tribunal’.123 The Tribunal clearly wanted to avoid creating an additional burden on the parties in line with Rule 38(2) ICSID Arbitration Rules which provides that ‘The Tribunal

118 Electrabel

v Hungary, above (n 92) para 234.

119 ibid. 120 See

Achmea v Slovakia, above (n 113) paras 31–33. ‘Invitations’ to submit amicus briefs, however, are not exceptional as such. See for other examples: Press Release, ICSID, Procedural Order Regarding Amici Curiae (2 February 2011), https://perma.cc/98QX-PRDK (in the case Pac Rim v El Salvador, ICSID Case No ARB/09/12). An invitation also was extended by the tribunal in Eli Lilly and Co v Canada: Press Release, ICSID, Eli Lilly and Company v Government of Canada (ICSID Case No UNCT/14/2) – Amici Curiae, 5 November 2015, https://perma.cc/89AK-Z72G. 121 Micula and Others v Romania, Award and separate opinion (11 December 2013) ICSID Case No ARB/05/20, IIC 621, para 27. 122 Belenergia SA v Italy (28 August 2019) ICSID Case No ARB/15/40, para 287. See also Charanne BV and Construction Investments SARL v Spain, Award (21 January 2016) SCC ARB No 062/2012, IIC 758, para 425. 123 Antin v Spain, above (n 106) para 64.

Amicus Curiae Intervention: From NAFTA to the Intra-EU Saga

209

shall ensure that the non-disputing party submission does not disrupt the proceeding or unduly burden or unfairly prejudice either party ….’. The EC, after having requested the Tribunal to alter its decision on the question of costs, which was rejected, finally notified the Tribunal that it was not in a position to file a written submission.124 Finally, in several other more recent cases, tribunals have refused the EC’s requests for submissions principally for reasons of inadmissibility, without providing much information on specific reasons for such decisions.125

VI. CONCLUSION

In general, the increased acceptance in international dispute settlement of NGO participation as amici curiae has been hailed as ‘permitt[ing] the emergence in international law of the idea of civil society as an important participant in the resolution of investment disputes’.126 There is little doubt that the principled acceptance of the intervention of third parties as amicus curiae is a positive development from the perspective of the legitimacy and transparency of the proceedings, in particular in those cases that are of high public interest. However, what is now described as a ‘routine and standardized feature of investor-state arbitration’,127 is in fact the product of a long list of ‘major decisions’ which have not only first accepted and justified the intervention of third parties as a matter of principle, but also shaped the precise contours of the admissibility of amicus curiae submissions in light of the various provisions in arbitration rules and regulations which have been adopted since 2006.

124 Also in other cases not involving the EC as amicus curiae, tribunals have been more attentive to mitigating the implications of amicus curiae briefs for the parties to the dispute. See, eg, the following statement in Philip Morris v Uruguay: ‘The Tribunal reserves the right to make at the appropriate time an order for costs to be paid or reimbursed by the Petitioners should either Party request the reimbursement of properly documented costs it has incurred by reason of the Submission.’ Philip Morris v Uruguay, above (n 84) para 31. 125 See, eg, RREEF Infrastructure (GP) Limited and RREEF Pan-European Infrastructure Two Lux SARL v Spain, Decision on Responsibility and on the Principles of Quantum (30 November 2018) ICSID Case No ARB/13/30, IIC 1537, paras 25 and 30 (a request was submitted twice, and denied twice). 126 F Francioni, ‘Access to Justice, Denial of Justice and International Investment Law’ (2009) European Journal of International Law 729, 742. 127 CH Brower, ‘Introductory Note to International Centre for Settlement of Investment Disputes (ICSID): Biwater Gauff (Tanzania) Ltd v. United Republic of Tanzania, ICSID case no ARB/05/22, Procedural Order No 5’ (2007) 46 International Legal Materials 572.

13 Challenges Decisions in International Investment Arbitration CHIARA GIORGETTI*

I. INTRODUCTION: THE IMPORTANCE OF CHALLENGES PROCEDURES

C

HALLENGES PROCEDURES PROVIDE an essential control mechanism to parties involved in an international investment arbitration. Through challenges proceedings, parties are given the opportunity to ensure that no conflicting interests or previously unknow facts affect the integrity of the Tribunal and thus ensure that the adjudicators chosen in their case continue to possess the qualities necessary to sit and decide on their dispute. More systemically, moreover, challenges provide a tool to ensure the neutrality of a system which is, differently from domestic courts, largely based on parties’ preferences. Neutral and effective challenge procedures are hence fundamental for the actual and perceived legitimacy of Investor-State Dispute Settlement (ISDS). It also follows that if the challenge is denied, then parties should be reassured of the arbitrators’ independence and impartiality; and, as a result, the arbitrators’ final decision is strengthened. Parties and their counsel do not take decisions to challenge an arbitrator easily. Challenges can be disruptive for the entire arbitration process: they may create delays and add time and cost to the proceedings. They can also build mistrust and increase animosity between the parties. Moreover, they also put the party in the difficult position of having to argue, in front of the remaining arbitrators, the suitability of an arbitrator to sit in the proceedings in which they appear. This may create difficult dynamics with and within the Tribunal. As challenges are only rarely accepted, moreover, the party and its counsel may have to continue appearing before an arbitrator they have claimed was not fit to hear the case, and at the very least, counsel will continue to appear in front of the remaining arbitrators. Unconscious – and unwanted – biases and resentment may linger. Except in the rare cases in which challenges are specious and purely dilatory, therefore, a party will start challenging procedures only if it has real concerns and evidence to fear that an arbitrator will not be fair and impartial. It is then important to know what is accepted as normal in terms of an arbitrator’s behaviour, and what is considered off-limit and a possible subject of a successful challenge.

* Chiara Giorgetti is Professor of Law at the Richmond Law School. She is grateful to Gonzalo Flores for his insightful comments on an earlier draft of this chapter and to Cemre Cise Kadioglu for her very helpful research. Her deepest appreciation also goes to Hélène Ruiz Fabri and Edoardo Stoppioni for inviting her to participate in this interesting project and for their excellent comments on previous drafts.

212 Chiara Giorgetti This chapter provides a short overview of challenge procedures generally used in international investment law first. It then focuses on the key decisions that have helped shape the many facets of the meaning of impartiality and independence as key duties of all arbitrators. More specifically, the chapter highlights and assesses decisions related to threshold issues first – and namely the applicable standard of review and who decides the challenge – it then reviews the most important decisions addressing substantive issues, specifically: repeat appointments, double hatting, issue conflict and comportment of arbitrators during proceedings.

II. CHALLENGES PROCEDURES: THE APPLICABLE STANDARD OF REVIEW

Parties in an investment arbitration are generally free to appoint an arbitrator of their choosing to decide their case. In a three-panel arbitration, each party selects one arbitrator and the presiding arbitrator is selected by agreement of the parties or by an appointing authority. A sole arbitrator is selected by agreement or by a neutral appointing authority.1 Even though the choice of arbitrators is momentous, arbitration rules do not generally provide great details on the qualities arbitrators must possess in order to be selected. Parties retain a lot of freedom in making their selection. The requirement that the arbitrator be and remain independent and impartial is present in all rules. So as to ensure that arbitrators conform with the requirement of independence and impartiality, arbitration rules require arbitrators to disclose their past relevant practice and relations. At the same time, rules provide challenge procedures that allow disputing parties to raise specific concerns about the independence and impartiality of arbitrators that may have arisen during the disclosure process or otherwise.2 Thus, if a disputing party believes that an arbitrator lacks or ceases to have the qualities required to serve as an arbitrator, it can challenge that arbitrator and ask the body designated under the applicable arbitration rules to make a determination as to whether the challenge is appropriate and if the arbitrator should consequently be removed.3 Challenges procedures are for the best part similar in the most commonly used arbitration rules, requiring a party to raise any challenge with the arbitral institution or the opposing party within a specific timeframe. Typically, rules provide for relatively tight deadlines for disputing parties to challenge an arbitrator.4 Under the International Convention on the Settlement of Investment Disputes (ICSID) Convention, for example, parties must file any request ‘promptly’.5 A party that intends to challenge an arbitrator under UNCITRAL Rules must send a notice of the challenge within 15 days of appointment notification or after learning of the circumstances giving rise to the challenge.6 The same time limit of 15 days exists in the rules of the Stockholm

1 See generally C Giorgetti, The Selection and Removal of Arbitrators in Investor-State Dispute Settlement (Leiden, Brill, 2019). 2 See generally ibid; L Malintoppi and A Yap, ‘Challenges of Arbitrators in Investment Arbitration: Still Work in Progress?’ in K Yannaca-Small (ed) Arbitration Under International Investment Agreements; A Guide to the Key Issues, 2nd edition (Oxford, Oxford University Press, 2018); J Crawford, ‘Challenges to Arbitrators in ICSID Arbitration’ in D Caron and others (eds), Practicing Virtue: Inside International Arbitration (Oxford, Oxford University Press, 2015). 3 For decision on challenges decided under the ICSID Convention, see https://icsid.worldbank.org/en/Pages/ Process/Decisions-on-Disqualification.aspx. Another excellent source of information on investment arbitrations is the PluriCourts Investment Arbitration Database (PITAD), available at https://pitad.org/index#welcome. 4 See Giorgetti, above (n 1). 5 ICSID Rules, Rule 9 (detailing the procedure to be taken). International Centre for Settlement of Investment Disputes, ICSID Convention, Regulations and Rules, Rules of Procedure for Arbitration Proceedings (Arbitration Rules), https://icsid.worldbank.org [hereinafter ICSID Convention and ICSID Rules]. 6 UNCITRAL Rules (2010), Art 13, https://uncitral.un.org.

Challenges Decisions in International Investment Arbitration

213

Chamber of Commerce (SCC),7 while the London Court of International Arbitration (LCIA)8 allows for 14 days, and the rules of the International Chamber of Commerce (ICC) provide for a 30-day limit from the appointment or the discovery of the relevant facts or circumstances.9 More consequential rules discrepancies exist, however, in relation to the standard to be adopted to review a challenge. Under UNCITRAL, LCIA and SCC Rules, arbitrators may be challenged ‘if circumstances exist’ that give rise to ‘justifiable doubts’ as to the impartiality or independence of an arbitrator.10 The ICC asks for ‘alleged lack of impartiality and independence’.11 Gallo v Canada held that the apprehension of bias should be to an ‘objective observer, reasonable’.12 Differently, the ICSID Convention provides that a party may propose the disqualification of an arbitrator on account of any fact indicating ‘a manifest lack of the qualities’ required to be nominated, which under Article 14 include ‘high moral character and recognized competence in the fields of law, commerce, industry or finance’ and reliability ‘to exercise independent judgment’. This ‘manifest’ standard had, until recently, been applied strictly. In Suez v Argentina, for example, the Tribunal rejected the application to disqualify an arbitrator and recalled that ICSID decisions recognised that the term ‘manifest’ in Article 57 of the Convention should ‘be proven by an objective evidence’.13 More recently, however, ICSID tribunals have adopted a more nuanced interpretation of ‘manifest’ and have moved towards a standard that more closely approaches the ‘justifiable doubts’ standard by adopting the view point of a reasonable third party. The seminal case is Blue Bank v Venezuela, which was decided by the Chair of the ICSID Administrative Council (who is also the President of the World Bank), on 1 November 2013.14 In that case, Venezuela had challenged the majority of the Tribunal. The Chair noted several undisputed facts and applied ‘an objective standard based on a reasonable evaluation of the evidence by a third party’ and interpreted the word ‘manifest’ in the ICSID Convention ‘as meaning “evident” and “obvious”’ in relation ‘to the ease with which the alleged lack of qualities can be perceived’.15 The Chair – focusing on the claimant-appointed arbitrator as the other challenged arbitrator had resigned – held that a reasonable third party ‘would find an evident or obvious appearance of lack of impartiality on a reasonable evaluation of the facts of the case’ and upheld the challenge.16 7 SCC Rules, Art 16, Arbitration Rules of the Arbitration Institute of the Stockholm Chamber of Commerce (1 January 2017), https://sccinstitute.com [hereinafter SCC Arbitration Rules]. 8 LCIA Arbitration Rules, Art 10(3). London Court of International Arbitration (1 October 2014), www.lcia.org [hereinafter LCIA Arbitration Rules 2014]. 9 ICC Rules, Art 14, International Chamber of Commerce Rules of Arbitration (1 March 2017), https://iccwbo.org [hereinafter ICC Arbitration Rules 2017]. 10 UNCITRAL Rules (2010), Art 12(1). 11 LCIA Arbitration Rules, Art 10(2) (‘there are justifiable doubts as to that arbitrator’s impartiality or independence’); SCC Arbitration Rules, Art 19 (‘party may challenge any arbitrator if circumstances exist that give rise to justifiable doubts as to the arbitrator’s impartiality or independence or if the arbitrator does not possess the qualifications agreed by the parties’). ICC Arbitration Rules, Art 14 (‘A challenge of an arbitrator, whether for an alleged lack of impartiality or independence, or otherwise, […]’). 12 Vito G Gallo v The Government of Canada, UNCITRAL, PCA Case No 55798, Decision on the Challenge to Mr JC Thomas, QC (14 October 2009). 13 Suez et al v Argentina, Decision on the Proposal for the Disqualification of a Member of the Arbitral Tribunal (27 February 2012) ICSID Case No ARB/03/17, para 40, https://icsid.worldbank.org. 14 Blue Bank Int’l & Trust (Barbados) Ltd v Bolivarian Republic of Venezuela, Decision on the Parties’ Proposals to Disqualify a Majority of the Tribunal (12 November 2013) ICSID Case No ARB/12/20, paras 66–68, https://icsid. worldbank.org. But for an early use of a similar standard (but not followed by subsequent tribunals) also see Compañía de Aguas del Aconquija et al. v Argentina, Decision on the Challenges to the President of the Committee (3 October 2001) ICSID Case no ARB/97/3, para 28. 15 ibid para 61. 16 ibid para 69.

214 Chiara Giorgetti While Blue Bank is the first ICSID decision to disqualify an arbitrator on the basis of the appearance of lack of impartiality, subsequent decisions have followed suit.17 Shortly thereafter, in Burlington v Ecuador, a decision dated 13 December 2013, the Chair of the Administrative Council upheld a challenge relying on the same standard.18 In that case, also discussed below in the context of arbitrators’ behaviour, Ecuador challenged the claimantappointed arbitrator on grounds that the claimant’s counsel had appointed him too frequently and he had failed to disclose the circumstances of his involvement in cases involving the same counsel. Ecuador also asserted that the arbitrator showed ‘a blatant lack of impartiality towards Ecuador’ during the conduct of proceedings.19 The Chair dismissed much of the proposal concerning the first two grounds, concluding that Ecuador was or should have been aware of the arbitrator’s involvement with the law firm before his accepted appointment. On the last ground, however, the Chair noted that the arbitrator had made some allegations about the ethics of counsel for Ecuador that did ‘not serve any purpose in addressing the proposal for disqualification’.20 He concluded that a third party undertaking a reasonable evaluation of those remarks would conclude that they ‘manifestly evidence[d] an appearance of lack of impartiality’21 with respect to Ecuador and its counsel. In upholding the challenge, the Chair explained that the ICSID Arbitration Rules ‘do not require proof of actual dependence or bias’ rather, it is ‘sufficient to establish the appearance of dependence or bias’.22 With this clear explanation, the Chair endorsed again the revised interpretation of ‘manifest’ as requiring not a demonstration of actual bias, but an appearance of bias, as seen by a reasonable third party.

III. CHALLENGES PROCEDURES: WHO DECIDES?

A neutral authority, either the appointing authority or the institution administering the proceedings, generally decides arbitrators’ challenges. Under the UNCITRAL Arbitration Rules, if the parties do not agree on the challenge, the decision is taken by the appointing authority. In ICC proceedings, it is for the the International Court of Arbitration – which exercises judicial supervision of ICC arbitral proceedings – to decide challenges; and in LCIA proceedings, the LCIA Court – which, similarly to the ICC Court of Arbitration, supervises LCIA proceedings – decides. ICSID stands out in this respect: when only one arbitrator of a panel of three is challenged, the unchallenged co-arbitrators decide the challenge. If they cannot agree on the challenge, or if the challenge involves more than one arbitrator, or the sole arbitrator, the Chair of the Administrative Council of ICSID decides. This system has been criticised because it puts the 17 Caratube International Oil Company v Kazakhstan, Decision on the Proposal for Disqualification of Mr Bruno Boesch (20 March 2014) ICSID Case No ARB/13/13, paras 91 and 110; Burlington Resources, Inc. v Republic of Ecuador, Decision on the Proposal or Disqualification of Professor Francisco Orrego Vicuña (13 December 2013) ICSID Case No ARB/08/5, paras 75, 78–80. Similarly, see Sociedad General de Aguas de Barcelona S.A. and Vivendi Universal S.A v Argentina, Decision on the Proposal for the Disqualification of a Member of the Arbitral tribunal (22 October 2007) ICSID Case No ARB/03/19, para 28. More recent decisions include Italba Corporation v Uruguay, ICSID Case No ARB/16/9; Raiffeisen Bank International AG and Raiffeisen Austria v Croatia, ICSID Case No ARB/17/34; KW Invest GmbH and TLS Invest GmbH v Spain, Elitech B.V. and Razvoj Golf D.O.O. v Croatia, ICSID Case No ARB/17/32; Mathias Kruck and others v Spain, ICSID Case No ARB/15/23. 18 Burlington Resources, Inc. v Republic of Ecuador, Decision on the Proposal or Disqualification of Professor Francisco Orrego Vicuña (13 December 2013) ICSID Case No ARB/08/5, paras 75, 78–80. 19 ibid para 20. 20 ibid para 79. 21 ibid para 80. 22 ibid para 66.

Challenges Decisions in International Investment Arbitration

215

remaining arbitrators in the awkward position of having to decide on the ability of another arbitrator to sit. Indeed, in most cases involving the challenge of one arbitrator under ICSID rules, the two co-arbitrators either denied the challenge or asked the Administrative Council Chair to decide. It is noteworthy that the ICSID’s system of resolving challenges is similar to what is found in other international permanent courts that have jurisdiction over states, such as the International Court of Justice where the entire Court decides on removal proposals.23 The choice of the ICSID’s framers is a testament to this precedent and can therefore be understood as similar in that respect, as ISDS always includes a sovereign in the dispute. Differently, the procedural genesis in the context of LCIA, SCC, ICC and UNCITRAL is commercial arbitration, where the norm is to ask a third party to decide on challenges. In practice, ICSID arbitrators have been generally reluctant to uphold the challenges of another member of the Tribunal. Caratube v Khazakstan was the first case in which the two unchallenged arbitrators upheld a challenge of the third arbitrator. In the case, discussed in more detail below in the context of repeat appointments and issue conflict,24 the co-arbitrators agreed with claimants’ argument and concluded that ‘a third party would find that there is an evident or obvious appearance of lack of impartiality or independence based on a reasonable evaluation of the facts in the present case’.25 It is significant that the proposed amendments to the ICSID Rules, which are under discussion as of December 2021, would allow the remaining arbitrators – by agreement – to refer challenges decisions to the Chair of the Administrative Council for any reason.26 Differently from the present rule, which requires arbitrators to send the challenge to the Chair of the Administrative Council only if they are equally divided, the proposed amended ICSID Arbitration Rule 30 would allow the two deciding arbitrators hearing a challenge of their co-arbitrator to send the challenge to the Chair if they are unable to decide upon the challenge for any reason, without needing to provide an explanation. Should they not be able to decide they would be deemed ‘equally divided’. This would be a welcomed change. Given the constraints in amending the ICSID Convention itself (which requires unanimity of all contracting states) amending the rules as suggested would also be a realistic goal that would allow challenges to be decided by a third party – and not the co-arbitrators – as a matter of routine while complying with the requirement of the Convention. As such, it would be an appropriate solution that also addresses the systemic criticism of the present process. As a general matter, challenges procedures are – overall and as much as specific arbitration rules permit – converging in terms of procedural requirements and standards applied.

IV. SUBSTANTIVE ISSUES: UNDERSTANDING INDEPENDENCE AND IMPARTIALITY

The duties of independence and impartiality are the core components of an arbitrator’s ethical conduct.27 A lack of independence and impartiality would undermine the entire

23 See generally C Giorgetti (ed), Challenges and Recusals of Judges and Arbitrators in International Courts and Tribunals (Leiden, Brill, 2015). 24 Sections IV.A and IV.C below. 25 Caratube International Oil Company LLP & Mr. Devincci Salah Hourani v Republic of Kazakhstan, Decision on the Proposal for Disqualification of Mr Bruno Boesch (20 March 2014) ICSID Case No ARB/13/13, para 62, http:// italaw.com. Ruby Roz Agricol v Kazakhstan, Award on Jurisdiction (UNCITRAL 2013). 26 The draft amendment and the commentaries are available at https://icsid.worldbank.org. 27 See generally C Giorgetti and MA Wahab, ‘A Code of Conduct for Arbitrators and Judges’ (2019) 8 Academic Forum on ISDS Concept Paper, www.jus.uio.no/pluricourts.

216 Chiara Giorgetti arbitration process.28 Independence is characterised by the absence of any external control or pressure, in particular in terms of relations between the arbitrator and a party which may influence the arbitrator’s decision. Lack of independence arises if an arbitrator has shown some form of dependency from one of the parties. Such dependent relations may arise because of economic, personal or social relationships. Impartiality, on the other hand, means the absence of bias or predisposition towards one party and requires that the arbitrator hears the parties without any favour and therefore bases his or her decision only on factors related to the merits of the case.29 Lack of impartiality translates into a partiality of an arbitrator towards a certain party or specific issue. Several challenges decisions have delved into these issues, and have specifically looked at the duties of independence and impartiality through the prism of multiple and repeat appointments and specific connections among arbitration participants, double hatting, issue conflict and behaviour of arbitrators during the arbitration process.30

A. Multiple and Repeat Appointments Critics of ISDS have identified repeat appointments of arbitrators as an element of concern that possibly needs more detailed regulation.31 The fear is that repeat appointments of an arbitrator by the same party or counsel in different cases may create certain allegiances and expectations that an arbitrator decides in favour of the party or counsel that appointed him or her on previous occasions. Thus, critics are concerned that an arbitrator may not be independent and impartial if appointed repeatedly by the same actors or same side in the dispute.32 Repeat appointments have been a frequent subject of disqualification requests. In the practice of investment tribunals, what seems to matter is not the number of cases in which an arbitrator has sat, but rather how close the cases are in terms of both facts and legal issues. In Caratube v Kazakhstan, for example, the issue at stake was the similarities of two cases in which the challenged arbitrator sat and to which he was appointed by the same state, Kazakhstan.33 In the case, Caratube Oil Company and Mr Devincci Salah Hourani, a US national, filed a claim against Kazakhstan at ICSID. Claimants submitted a request for

28 See

generally, UN Commission on International Trade law, Working Group III. Performance Plastics Europe v Bolivarian Republic of Venezuela, Decision on Claimant’s Proposal to Disqualify Mr Gabriel Bottini from the Tribunal under Art 57 of the ICSID Convention (27 February 2013) ICSID Case No ARB/12/13, para 56. Also see cases cited in footnote: Suez, Sociedad General de Aguas de Barcelona S.A. v The Argentine Republic, Decision on Second Proposal for Disqualification, ICSID Cases Nos ARB/03/17 and ARB/03/19, para 29; GEMA International, NCT Necotrans, GETMA International Investissements & NCT Infrastructure & Logistique c La République de Guinée, Décision sur la demande en récusation de Monsieur Bernardo M. Cremades (Arbitre), ICSID Case No ARB/11/29, para 59; Conoco Philipps Company et al. v The Bolivarian Republic of Venezuela, Decision on the Proposal to Disqualify L Yves Fortier, QC (Arbitrator), ICSID Case No ARB/07/30, para 54. 30 See generally, MN Cleis, The Independence and Impartiality of ICSID Arbitrators – Current Case Law, Alternative Approaches, and Improvement Suggestions (Leiden, Brill Nijhoff, 2017). 31 See generally, LA Sobota, ‘Repeat Arbitrator Appointments in International Investment Disputes’ in Giorgetti, above (n 23) 293; D Kapeliuk, ‘The Repeat Appointment Factor: Exploring Decision Patterns of Elite Investment Arbitrators’ (2010) 96 Cornell Law Review 47. 32 See the data collected in S Puig and A Strezhnev, ‘Affiliation Bias in Arbitration: An Experimental Approach (2017) 46 Journal of Legal Studies 371. 33 Caratube International Oil Company LLP & Mr. Devincci Salah Hourani v Republic of Kazakhstan, Decision on the Proposal for Disqualification of Mr Bruno Boesch (20 March 2014) ICSID Case No ARB/13/13, para 62, http:// italaw.com/sites/default. Ruby Roz Agricol v Kazakhstan, Award on Jurisdiction (UNCITRAL 2013). 29 Saint-Gobain

Challenges Decisions in International Investment Arbitration

217

disqualification of the respondent appointed arbitrator alleging he manifestly could not be ‘independent and impartial’ because he had been appointed as arbitrator by Kazakhstan in another related case, Ruby Roz Agricol v Kazakhstan and also argued that several individuals who submitted witness statements in Ruby Roz were also likely to submit witness statements in Caratube.34 The claimants also asserted that the same firm had appointed the same arbitrator in numerous cases, in addition to Caratube and Ruby Roz, which they feared also created a potential conflict. The two unchallenged arbitrators found that multiple appointments by the same firm, without more, did not constitute an objective circumstance that would demonstrate the arbitrator’s inability to exercise independent and impartial judgment in the case at hand.35 In the specific case, however, they agreed with the claimants that Ruby Roz arose out of the same factual context and concluded that the case exhibited an imbalance because of the respondent-appointed arbitrator’s involvement with Ruby Roz. The two unchallenged arbitrators thus upheld the challenge, and explained that they did not question the arbitrator’s ‘moral character, his actual impartiality, or his honesty’ but concluded that ‘a third party would find that there is an evident or obvious appearance of lack of impartiality or independence based on a reasonable evaluation of the facts in the present case.’36 The reluctance to uphold a challenge based only on the number of cases in which an arbitrator has taken part has been a constant finding of deciding authorities. This regardless of whether the appointment was made by the same counsel or by the same party, or the timeframe of such appointments. The conclusion is unchanged when arbitrators have consistently been appointed by the claimants’ side or by the side of respondents, an issue that has also raised concerns.37 A recent case is particularly interesting and on point. In Raiffeisen v Croatia, the respondent challenged the claimant-appointed arbitrator asserting that he had been nominated a total of 38 times, of which 35 was by a claimant.38 Of these cases, four were against Croatia, which meant – at the time – half of all the existing cases against Croatia. The respondent also asserted that the arbitrator had a special relation with the claimant’s counsel, which had led to instances of cross-appointment. The decision on the challenge was taken by the Chair of the Administrative Council, as one of the unchallenged arbitrators explained he could not decide on the case.39 Citing Caratube, the Chair distinguished between cases that had some legal issues overlap and those that had factual overlap,

34 Ruby Roz Agricol v Kazakhstan, a contract case filed in October 2010 under UNCITRAL Rules (as updated in 2010) and decided in favour of the state in August 2013, www.italaw.com/cases/1543. 35 This follows other cases decided along the same line, including Tidewater Investment SRL and Tidewater Caribe, C.A. v Bolivarian Republic of Venezuela, ICSD Case No ARB/10/5, Decision on Claimant’s Proposal to Disqualify Professor Brigitte Stern (23 December 2010); Opic Karimun Corporation v Bolivarian Republic of Venezuela, Decision on the Proposal to Disqualify Professor Philippe Sands, Arbitrator (5 May 2011) ICSID Case No ARB/10/14, para 45; and Universal Compression International Holdings, S.L.U. v. Bolivarian Republic of Venezuela, Decision on the Proposal to Disqualify Prof. Brigitte Stern and Prof. Guido Santiago Tawil (20 May 2011) ICSID Case No ARB/10/09. See also Burlington Resources, Inc. v Republic of Ecuador, Decision on the Proposal or Disqualification of Professor Francisco Orrego Vicuña (13 December 2013) ICSID Case No ARB/08/5, paras 75, 78–80 (challenge on multiple appointment considered untimely), www.italaw.com/cases/181. 36 Caratube, above (n 33) paras 91 and 110. 37 See Sobota and Kapeliuk, above (n 31). 38 Raiffeisen Bank International Ag and Raiffeisenbank Austria D.D. v Republic of Croatia, Decision On The Proposal To Disqualify Stanimir Alexandrov (17 May 2018) ICSID Case No ARB/17/34, www.italaw.com/cases/6634. On the same point, see also Blue Bank International & Trust (Barbados) Ltd. v Bolivarian Republic of Venezuela, above (n 14), in which one of the arbitrators was challenged on the claim that he had too often been appointed by respondents and thus had developed a bias on respondents’ favour. 39 ibid para 10.

218 Chiara Giorgetti and held that only the latter kind were problematic. It concluded that it was ‘insufficient that a similar, or even the same, legal issue arises in two cases’ and specifically that: Here, there is no factual overlap of the kind described in Caratube: this arbitration and [the other case] concern different claimants; different measures; and entirely different sets of facts underlying the respective claims. That a similar legal issue may or may not arise in both cases does not establish a manifest lack of impartiality or independence.40

Deciding authorities have repeatedly asserted that claims of impartiality due only to multiple appointments and without more would be speculative. In Tidewater v Venezuela a case cited with approval in Raiffeisen, the unchallenged arbitrators stated that ‘multiple appointments as arbitrator by the same party in unrelated cases are neutral, since in each case the arbitrator exercises the same independent function’.41 And in Raiffeisen, the Chair found that the same principle applies in cases where there are ‘multiple appointments by different claimants, even if they are in cases against the same respondent’.42 Similarly, in Vivendi I, the unchallenged arbitrators stated that a finding of a lack of independence and impartiality ‘must exclude reliance on speculative assumptions or arguments’ so that the circumstances that are actually established ‘must negate or place in clear doubt the appearance of impartiality’.43 Based on these and similar cases, the Chair in Raiffeisen found that the respondent had submitted no evidence of financial dependency or unconscious bias or submitted evidence of financial dependence, and concluded that: Respondent’s allegations of unconscious bias and financial dependence are the kind of speculative assumptions or arguments that would not lead a third party undertaking a reasonable evaluation of [the arbitrator’s] appointments by claimants to conclude that the alleged lack of impartiality or independence is manifest.44

As a result, the Chair rejected the Respondent’s proposal based on this ground. Other cases have upheld similar conclusions, so that one can conclude that, at the present time, a challenge based only on the number of previous appointments would, without more, unlikely be successful.45 Mere allegations or suspicion of bias, based only on the number of cases, whether such cases come only from claimants or only from respondents, or from the same party or counsel would not – without something more specific – result in 40 ibid

para 56. Investment SRL and Tidewater Caribe, C.A. v Bolivarian Republic of Venezuela, Decision on Claimants’ Proposal to Disqualify Professor Brigitte Stern, Arbitrator (23 December 2010) ICSID Case No ARB/10/5, para 60. 42 Raiffeisen, above (n 38) para 88. 43 Compañía de Aguas del Aconquija S.A. and Vivendi Universal S.A. v Argentine Republic, Decision on the Challenge to the President of the Committee (3 October 2001) ICSID Case No ARB/97/3, para 25, www.italaw.com/ sites/default/files/case-documents/italaw9732.pdf. 44 Raiffeisen, above (n 38) 89. 45 See, eg, the challenge brought against Albert Jan van den Berg in BG Group in disputes against Argentina. The Republic of Argentina v BG Group plc, Case No 08–0485 (rbw) (d.d.c.), Petition to Vacate or Modify Arbitration Award file on 21 March 2008, paras 73, 75 and 76, http://ita.law.uvic.ca/documents/BGvArgentina.pdf. See also the unsuccessful cases brought against the repeat appointments of Ms Stern by the same state (Venezuela) and the same firm (Curtis, Mallet-Prevost, Colt & Mosle LLP), Tidewater Investment SRL and Tidewater Caribe, C.A. v Bolivarian Republic of Venezuela, Decision on the Proposal for the Disqualification of a Member of the Arbitral Tribunal (23 December 2010) ICSID Case No ARB/10/5, para 60 (‘the starting point is that multiple appointments as arbitrator by the same party in unrelated cases are natural, since in each case the arbitration exercise the same independent arbitral function’) and Universal Compression International Holdings, S.L.U. v Bolivarian Republic of Venezuela, Decision on the Proposal for the Disqualification of two Members of the Arbitral Tribunal (20 May 2011) ICSID Case No ARB/10/9. See generally Cleis, above (n 30) 64–72. 41 Tidewater

Challenges Decisions in International Investment Arbitration

219

a successful challenge. Disqualification proposals have only been upheld where more than repeated appointment has been shown. Given the concerns expressed during the UNCITRAL ISDS reform process by many stakeholders in relation to multiple appointments, however, it would be important to review the issue and offer concrete guidance to practitioners on appropriate comportment.

B. Familiarity with Another Participant in the Proceedings Several challenge cases have focused on the specific (rather than general as seen above) relationship between actors in international arbitration. More precisely, critics have raised concerns that arbitrators may lack independence and impartiality because of their specific familiarity with counsel or a party participating in the same proceedings, and because of the specific working relationship that the arbitrator may have with counsel or party. Critics are concerned that an arbitrator may not be independent and impartial if appointed by the same actor because knowing and having worked with them in other cases may create incentives to rule in their favour or bias favouring their arguments or side. The issue has been addressed in a number of cases. In SGS v Pakistan, claimant challenged Pakistan’s appointed arbitrator on the basis that the arbitrator’s firm was acting as counsel on behalf of Mexico in an ongoing investment arbitration presided over by Pakistan’s counsel in SGS. The unchallenged tribunal members rejected the challenge and observed that: It is commonplace knowledge that in the universe of international commercial arbitration, the community of active arbitrators and the community of active litigators are both small and that, not infrequently, the two communities may overlap, sequentially if not simultaneously. It is widely accepted that such an overlap is not, by itself, sufficient ground for disqualifying an arbitrator. Something more must be shown if a challenge is to succeed.46

They therefore concluded that any inference of bias was ‘so far as [they] can see, bereft of any basis in the facts of this proceeding; what we have here is simply a supposition, a speculation merely’.47 Similarly to the finding in relation to repeat and multiple appointments, therefore, specific allegations are needed for a challenge to be upheld, mere inferences or appearances are not sufficient. Analogously, reversed and crossed roles by the same arbitrator and counsel in a variety of cases, without specific allegations of lack of independence, is not sufficient for a removal of the arbitrator. Raiffeisen, seen in the previous section, provides an important example for this situation also. The case looks specifically at the issue of cross-appointments, ie the fear of the existence of a practice of reciprocal appointments as counsel and as arbitrator by the same actors in multiple cases. The concern is that reciprocal appointment would create some form of allegiance or expectation that an arbitrator appointed by a certain counsel would be favourably disposed towards the case, with the expectation that the counsel would do the same when the roles are switched. In Raiffeisen, the challenged arbitrator had been appointed by the claimant counsel as arbitrator previously, and the claimant counsel sat as arbitrator appointed by the challenged arbitrator’s clients. Citing SGS with approval, the Chair in Raiffeisen was persuaded that ‘without “something more”, this tally of “cross-appointments” would not by 46 SGS v Pakistan, SGS Société Générale de Surveillance S.A. v Islamic Republic of Pakistan, Decision on Claimant’s Proposal to Disqualify Arbitrator (19 December 2002) ICSID Case No ARB/01/13, paras 25–26. 47 SGS Société Générale de Surveillance S.A. v Islamic Republic of Pakistan, Decision on Claimant’s Proposal to Disqualify Arbitrator (19 December 2002) ICSID Case No ARB/01/13, para 26.

220 Chiara Giorgetti itself demonstrate to a third party undertaking a reasonable evaluation of the evidence’ that the challenged arbitrator would lack the required qualities to sit. The respondent had not provided any other evidence and so the challenge proposal was rejected.48 What kind of relation would then be necessary to result in the removal of an arbitrator? The issue has been explored in several cases. A rare successful challenge based on existing relations between actors in the same case occurred in Blue Bank v Venezuela, a case already explored above.49 In the case, Venezuela challenged the claimant-appointed arbitrator alleging that the international law firm in which he was a partner also represented a client against Venezuela in another investment arbitration, though the second case was dealt with by a different office of the same firm. The Chair deciding the challenge found that the sharing of a corporate name, the existence of the law firm’s international arbitration steering committee to which the arbitrator was a member, the remuneration structure and the fact that similar issues were likely to be discussed in the two cases made him reach the conclusion that it had ‘been demonstrated that a third party would find an evidence or obvious appearance of lack of impartiality on a reasonable evaluation of the facts of this case’.50 Interestingly, a problematic relationship need not be confined to the area of investment arbitration. The relationship between the arbitrator and one of the parties also became the subject of a challenge in Big Sky Energy Corporation v Kazakhstan,51 also an ICSID case. The claimant, a US investor, filed a proposal for disqualification of the respondentappointed arbitrator arguing that the arbitrator’s past professional ties to members of the Kazakh judiciary put in question his impartiality and independence in the eyes of an independent third-party. These ties rose out of his previous employment by the German Office of International Cooperation, through which he had participated in a project for legal and judicial reform in various newly independent Central Asian states, including Kazakhstan. Actions of the courts were specifically at issue in the arbitration. The proposal for disqualification was upheld by the two remaining co-arbitrators, who decided the challenge under the applicable ICSID Rules. In sum, a relationship between one arbitrator and another actor in the arbitration, being that a party or counsel, would result in the removal of the arbitrator only if there are clear and detailed concerns and not mere allegations. A general argument based on the apprehension of bias, without more, would not result in the acceptance of a challenge. The threshold for disqualification remains very high in relation to specific circumstances that link the arbitrator to the party and requires a demonstrated relation on a relevant issue. Clarifying further the standard and the circumstances that would lead to disqualification is a priority to address stakeholders’ concerns.

C. Double Hatting Another frequent reason for challenges is double hatting, ie situations in which an arbitrator is or has recently also acted as counsel, expert or in other related position in some 48 Raiffeisen,

above (n 38) paras 94–5. Bank, above (n 14), also analysed above in section II on the standard of review. para 69. 51 Big Sky Energy Corporation v Republic of Kazakhstan, Decision on the Disqualification Proposal of Mr Rolf Knieper (3 May 2018) ICSID Case No ARB/17/22; procedural details are available at https://icsid.worldbank.org/en/ Pages/cases/casedetail.aspx?CaseNo=ARB/17/22. The decision is not public, and it has been reported in professional news, including in L Peterson, ‘In a rare development, a part of ICSID arbitrators decide that a colleague must be disqualified’ (IA Reporter, 4 May 2018), www.iareporter.com/articles/30250/. 49 Blue 50 ibid

Challenges Decisions in International Investment Arbitration

221

related matter.52 This situation is peculiar to international investment arbitration – which is characterised by party-appointment of part-time adjudicators – as in most other international fora judges are permanent and are prohibited from sitting in other cases. At the moment, no arbitration rules prohibit double hatting as such, though several proposals are under discussion at the UNCITRAL ISDS reform process and the prohibition has been inserted in the code of conducts included in recent investment treaties. Regulating double hatting is also complicated by the lack of a common understanding of what double hatting is: what kind of cases generate double hatting (only investment arbitration or all public international law cases involving states)? Under which fora (only ICSID or other arbitral institutions also)? Or what kinds of roles (only counsel/arbitrator or also expert, secretary and other roles)?53 Case law suggests some interesting bright lines. For example, in ICS v Argentina, an UNCITRAL case, the appointing authority deciding the challenge found it improper for someone to act at the same time as counsel and arbitrator in two unrelated cases involving the same respondent. A claimant-appointed arbitrator was challenged because he and his firm were representing claimants in another case against Argentina (Vivendi), even if the cases were not related. The appointing authority noted that this was not merely a case in which the arbitrator’s law firm was acting adversely to one of the parties, but ‘rather a case where the arbitrator has personally and recently acted adversely to one of the parties to the dispute’. The appointing authority hence found that doubts about the arbitrator’s impartiality of independence were justifiable in light of the fact that both matters related to investment protection actions, which raised broadly similar concerns against the same state.54 Similarly, in Grand River Enterprises, the US challenged a claimant-appointed arbitrator who simultaneously represented parties adverse to the US in the Inter-American Commission of Human Rights and the Committee on the Elimination of Racial Discrimination (CERD). The ICSID Secretary General deciding the challenge under NAFTA rules found ‘basic similarity’ in the issues to be decided as they all pertained to the US compliance with international commitments and requested the arbitrator to make a choice between continuing as an arbitrator or at CERD. He chose arbitration and the challenge was dismissed.55 The same approach was applied in Vito Gallo v Canada, a NAFTA case applying UNCITRAL rules. The claimant challenged the respondent-appointed arbitrator on the ground that he had been retained as a possible adviser by Mexico, a potential third-party, on issues of international trade and investment. The appointing authority held that from the point of view of a ‘reasonable and informed third party’ there would be justifiable doubts as to his impartiality and independence if he continued to serve as Mexico’s adviser. The arbitrator was given the choice to either withdraw from the advisory position of resign as an arbitrator. He resigned as an arbitrator a few days later.56

52 M Langford, D Behn and RH Lie, ‘The Revolving Door in International Investment Arbitration’ (2017) 20 Journal of International Economic Law 301. 53 See generally UNCITRAL, Report of Working Group III (Investor-State Dispute Settlement Reform) on the work of its thirty-sixth session (Vienna, 29 October–2 November 2018), A/CN.9/964, 6 November 2018, https://undocs.org/ en/A/CN.9/964. 54 ICS Inspection and Control Services Ltd v Argentina, UNCITRAL, PCA Case No 2010-9, Decision on challenge to Mr Stanimir A Alexandrov (17 December 2009) 5, www.italaw.com/sites/default/files/case-documents/ita0415.pdf. 55 Grand River Enterprises Six Nations Ltd et al v United States of America, Decision on the Challenge to Arbitrator James Anaya (28 November 2007) 1, www.italaw.com/sites/default/files/case-documents/ita0382_0.pdf. 56 Vito G. Gallo v Government of Canada, ICSID Deputy Secretary-General, Decision on the Challenge to Mr J. Christopher Thomas, QC (14 October 2009), https://perma.cc/76QN-DEWU.

222 Chiara Giorgetti While setting clear boundaries on double hatting, authorities deciding challenges have also emphasised that there must be a link that is relevant to the arbitration case between the two positions simultaneously held by the challenged arbitrator. So, for example, a challenge filed by a respondent based on an arbitrator sitting as a director of a bank that held a minority interest of two of the claimants was not considered sufficient to create a conflict of interest and an appearance of bias in favour of claimant. The link between the two positions was too remote.57 Double hatting has been criticised in international arbitration and it is important that tribunals have painted some defined red lines for concurrent representations. In this instance, the ISDS reform process, and specifically the work of ICSID and UNCITRAL on a code of conduct for arbitrators that can define and regulate inadmissible double hatting, would be very helpful in addressing the important concerns expressed by many stakeholders.58

D. Issue Conflict Past decisions, academic writings or other public statements authored by an arbitrator may also give rise to concerns that the arbitrator does not come to the case with an open mind, and that the arbitrator has prejudged an issue that is important for the proceedings. In such cases, the alleged lack of impartiality and independence is due to an ‘issue conflict’. This problem has been raised in the literature, and has given rise to some challenges, though very rarely has it resulted in the removal of the arbitrator. Several cases dealt with issue conflicts in a variety of ways.59 For example, in ICSID case Urbaser v Argentina,60 the claimants challenged the appointment of the respondent-appointed arbitrator because of the views he had expressed in his publications on two questions that the claimants considered important to the arbitration. The claimants alleged that arbitrator had prejudged essential elements of the dispute that was the object of the arbitration. However, the two unchallenged arbitrators differed and held that the scholarly opinions expressed by the challenged arbitrator did not meet the threshold of ‘presenting an appearance’ that he would not be prepared to hear and consider the position of the parties ‘with full independence and impartiality’.61 CC/Devas v India, an UNCITRAL case brought under the India–Mauritius BIT is seminal.62 Because of the applicable rules, the challenge was decided by the appointing authority, the 57 Suez, Sociedad General de Aguas de Barcelona S.A. and Interagua Servicios Integrales de Agua S.A. v Argentine Republic, Decision on a Second Proposal for the Disqualification of a Member of the Arbitral Tribunal (12 May 2008) ICSID Case No ARB/0 (Suez II). See also EDF International S.A., SAUR International S.A., Léon Participaciones Argentinas S.A. v Argentine Republic, Challenge Decision Regarding Professor Gabrielle Kaufmann-Kohler (25 June 2008) ICSID Case No ARB/03/23, which was also decided on similar grounds. 58 This is the choice made by several states in new investment treaties, such as the one negotiated between Canada and the EU. 59 On issue conflict, see J Crook and L Boisson de Chazournes, ASIL-ICCA Task Force Report on Issue Conflicts in Investor-State Arbitration (The Hague, International Council for Commercial Arbitration, 2016), https://perma. cc/A7UJ-AJGA. See also IC Popova and JL Polebaum, ‘Emerging Expectations For Arbitrators: ‘Issue Conflict’ In Investor-State Arbitration And Beyond’ (2018) 41 Fordham International Law Journal 937. 60 Urbaser S.A. and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v The Argentine Republic, Decision on Claimants’ Proposal to Disqualify Professor Campbell McLachlan, Arbitrator (12 August 2010) ICSID Case No ARB/07/26, para 20, https://perma.cc/BEG8-ZUZH. See also the discussion on Blue Bank, above (n 14). 61 Urbaser (n 60) para 58. 62 CC/Devas (Mauritius) Ltd., Devas Employees Mauritius Private Limited and Telecom Devas Mauritius Limited v India, PCA Case No 2013-09, Decision on the Respondent’s Challenge to the Hon. Marc Lalonde and Prof. Francisco Orrego Vicuña (30 September 2013), www.italaw.com/sites/default/files/case-documents/italaw3161.

Challenges Decisions in International Investment Arbitration

223

President of the International Court of Justice, Judge Tomka. The respondent challenged the arbitrator appointed by the claimant as well as the Presiding Arbitrator. The respondent argued that the two challenged arbitrators ‘strongly held and articulated positions’ in the case on the issue of essential security interest, an issue likely to be at issue in the case. These views were articulated in decisions in several cases: two cases decided together, and one additional case as well as well as in a chapter in a book published after the decisions for the claimant-appointed arbitrator. President Tomka noted first that it was not surprising that the two arbitrators had decided similarly in the previous cases, because the underlying facts were similar. However, he was more concerned about the chapter published after the decisions. In his view, it showed that the arbitrator’s views had remained unchanged. He was sympathetic to the respondent’s concerns and upheld the challenge concluding that the arbitrator was ‘certainly entitled to his views, including to his academic freedom. Equally, however, the respondent is entitled to have its arguments heard and ruled upon by arbitrators with an open mind. Here, the right of the latter has to prevail.’63 Following CC/Devas, an ICSID case also upheld challenges on issue conflict grounds. In Caratube v Kazakhstan, the unchallenged arbitrators were asked among other things to decide specific questions related to issue conflicts.64 Claimants alleged that the two cases in which the arbitrator sat had many similarities, as both claims relied on essentially the same factual allegations with respect to acts and omissions and pattern of conduct. The claimants also argued that the same witnesses might appear in both cases. The two unchallenged arbitrators agreed with the claimants that both cases arose out of the same factual context. They held that because the arbitrator participated in the other case he might decide ‘based on such external knowledge’65 and concluded that the challenged arbitrator could not be expected ‘to maintain a “Chinese wall” in his own mind’ and hence upheld the challenge.66 The standard adopted in Caratube was cited with approval in the recent Raiffeisen, a case also analysed above, that also explored issue conflict.67 In Raiffeisen, the Chair of the Administrative Council declined the challenge based on fact that the challenged arbitrator was sitting in two cases against Croatia that may touch on similar legal issues and concluded that when a proposal for disqualification rests on consideration of the same legal issue in multiple arbitrations ‘the mere exposure of an arbitrator to the same legal issue in multiple arbitrations is insufficient to disqualify that arbitrator’. Indeed, there must be an additional ‘significant – overlap of facts that are specific to the merits and the parties involved, as was the case in Caratube’.68 The Chair found no similarity between the two cases ‘let alone a significant overlap in facts and parties’ and concluded that a third party ‘undertaking a reasonable evaluation of the facts alleged’ would not find a lack of independence and impartiality.69 Since 2014 no claim of issue conflict has led to a successful challenge.70 Given the lack of a common understanding of ‘issue conflict’ and of its limit, available case law is designing 63 ibid

para 64. See also sections II (applicable standard of review) and IV.A (repeat appointment). 65 Caratube, (n 33) para 89. 66 ibid paras 78, 84, 86, 90. 67 See above (n 38) at section IV.A (repeat appointments) and IV.B (relations between counsel and arbitrator). 68 Raiffeisen, above (n 38) para 91. 69 ibid para 92. 70 See, eg, Valeri Belokon v Kyrgyz Republic, UNCITRAL, Decision on Challenges to Arbitrators Professor Kaj Hober and Professor Jan Paulsson (6 October 2014). Respondent challenged claimant-appointed Jan Paulsson for his writings on denial of justice issues. The deciding authority dismissed the claim and held that he was ‘satisfied that Professor Paulsson’s writings do not reveal fixed views on any specific legal issue likely to be relevant in the present 64 ibid.

224 Chiara Giorgetti important contour of the issue. Raiffeisen shows a more stringent approach to challenges and serves as a counterpoint to CC/Devas and Caratube. Yet, issue conflict represents an important concern of many ISDS stakeholders, which also needs to be balanced with the requirement that arbitrators be competent. It is thus important that it is not based on mere suspicion, but instead based on a significant or specific fact.

E. Behaviour of Arbitrators A final issue for which challenges procedures have been filed relate to the behaviour of arbitrators. This is important as it highlights the ‘practical’ aspect of independence and impartiality so that an arbitrator is expected to behave not only in a respectful and professional way, but also demonstrate independence and impartiality in practice throughout the proceedings. There are not many cases directly on point, but one is important. Burlington v Republic of Ecuador, decided in 2013, tackled issues concerning repeat appointments of the same arbitrator by counsel, duties of disclosure, as well as arbitrator conduct during proceedings.71 The case, applying ICSID rules under the Ecuador–US BIT, was decided by the Chair of ICSID’s Administrative Council after the two remaining co-arbitrators failed to reach a decision on the challenge. Beside issue related to multiple appointments which the Chair found untimely raised, Ecuador also asserted that the arbitrator showed ‘a blatant lack of impartiality towards Ecuador’ during the conduct of proceedings.72 On this issue, the Chair noted that the arbitrator had made some allegations about the ethics of counsel for Ecuador that did ‘not serve any purpose in addressing the proposal for disqualification’.73 He concluded that a third party undertaking a reasonable evaluation of those remarks would conclude that they ‘manifestly evidence[d] an appearance of lack of impartiality’74 with respect to Ecuador and its counsel. In upholding the challenge, the Chair explained that the ICSID Arbitration Rules ‘do not require proof of actual dependence or bias’ rather, it is ‘sufficient to establish the appearance of dependence or bias’.75

V. CONCLUSIONS

Challenges of arbitrators are important safeguards of the investment arbitration system. This is especially significant given the unique role played by private parties in ISDS. The cases analysed in this chapter show that there is a certain convergence among different rules both in procedural and substantive matters. Procedurally, an appearance of bias, as seen by a reasonable third party, is a threshold in which the evidence is evaluated by the authority deciding the challenge. There is no requirement of a demonstration of actual bias, which would be in any event exceedingly difficult to demonstrate.

case on which the Parties have a reasonable expectation of an open mind. I would add that I have found no indication that Professor Paulsson has consistently interpreted the concept of ‘denial of justice’ in a manner favouring investors.’ para 98. 71 Burlington Resources, Inc. v Republic of Ecuador, Decision on the Proposal or Disqualification of Professor Francisco Orrego Vicuna (13 December 2013) ICSID Case No ARB/08/5, paras 75, 78–80. 72 ibid para 20. 73 ibid para 79. 74 ibid para 80. 75 ibid para 66.

Challenges Decisions in International Investment Arbitration

225

Substantively, the contour of a systemic understanding of challenges requirements are also becoming clearer. Bright line cases, such as the prohibition of contemporaneous representation of a party and serving as arbitrator involving the same party, have been drawn. However, issues such as repeat appointments (regardless of whether the appointment was made by the same counsel or by the same party, the timeframe of such appointments or whether it was made by several similarly situated parties if there are allegations of crossappointment) or issue conflicts need ‘something more’ or something ‘significant’ not just allegations of possible bias, refining further the confines of the ‘something more’ will be key to strengthen the legitimacy of ISDS. Removing an arbitrator as a result of a removal procedure is rare. While this ensures that arbitrators are not removed based on simple allegations, the issue remains as to whether this approach satisfies the call to refrain from any appearance of bias when selecting arbitrators. It remains true that each case is unique and that an arbitrator should be able to properly distinguish each one. Yet, issues of repeat appointments, familiarity and double hatting touch an important chord of investment arbitration, which are unique to a system based on partyappointments. The process of ISDS reform presently under way at UNCITRAL is examining this (and other) issues and how to possibly address it. Repeat appointments rise concern on grounds of independence, as well as on the concern of availability of the arbitrator to have the time necessary to fully decide a case. Double hatting and familiarity raise similar concerns related to independence. However, critics point out that the prohibition of double hatting may limit the entry of a new and more diverse pool of arbitrators, though this claim remains to be fully demonstrated. For these reasons, a clarification on whether a limit on the number of cases in which one arbitrator may sit, for example by number at any given time or within a specific timeframe, as well as specific guidelines on incompatible roles – possibly incorporated in a Code of Conduct for arbitrators – would be very helpful to address and alleviate concerns of the many ISDS stakeholders.76

76 C Giorgetti and J Dunoff, ‘Ex Pluribus Unum? On the Form and Shape of a Common Code of Ethics in International Litigation’ (2019) 113 AJIL Unbound 312.

14 Major Decisions on the Definition of Investment SHOTARO HAMAMOTO*

T

HE HOST STATE accepts obligations to accord protection to ‘investments’ and investors may bring disputes concerning ‘investments’ to arbitration pursuant to the applicable investment treaty. The definition of investment thus plays a crucial role in determining the scope of the substantive obligations as well as that of the disputes that can be settled in accordance with the dispute settlement procedure set forth in the applicable investment treaty. Each investment treaty thus contains a clause formulating the definition of investment. Although the definition may vary from one treaty to another, there are a certain number of common problems. However, discussions on the definition of investment were first elicited in the context of the application of the International Centre for Settlement of Investment Disputes (ICSID) Convention. Where the claimant engages arbitration pursuant to an investment treaty that envisages arbitration under the ICSID Convention, the jurisdiction of the Tribunal is based on two instruments, ie the applicable investment treaty and the ICSID Convention. Therefore, for the Tribunal to have jurisdiction, the dispute needs to arise out of an ‘investment’ in the sense of the applicable investment treaty as well as the ICSID Convention (Article 25(1)). As the ICSID Convention contains no definition of investment, the interpretation of the term ‘investment’ in the said article has been a serious problem.

I. ‘INVESTMENT’ UNDER THE ICSID CONVENTION

A. The Requirements of an Investment Fedax is the first case in which the jurisdiction of the Tribunal was objected to on the ground that the underlying transaction did not meet the requirements of an investment under Article 25(1) of the ICSID Convention. The decision of the Fedax Tribunal well summarises the problem.1 In this case, six promissory notes were issued by Venezuela in order to acknowledge its debt for the provision

* Shotaro Hamamoto is Professor at Kyoto University. 1 Fedax v Venezuela, Decision of the Tribunal on Objections to Jurisdiction (11 July 1997) ICSID Case No ARB/96/3, (Francisco Orrego Vicuña (President), Meir Heth, Roberts B Owen) (footnotes partly omitted).

228 Shotaro Hamamoto of services under a contract signed in 1988 with Industrias Metalúrgicas Van Dam C.A., a  Venezuelan company. They were later acquired by Fedax, a company incorporated under the law of Curaçao, Netherlands Antilles, by way of endorsement. When Fedax presented the notes for payment at the maturity date, Venezuela refused to honour them. Fedax then instituted an ICSID arbitration pursuant to the Netherlands-Venezuela Bilateral Investment Treaty (BIT) but Venezuela objected to the jurisdiction of the Tribunal. The Tribunal held that: [¶21] … It is well established that numerous attempts to define investments were made during the negotiations of the Convention, but none were generally acceptable. Because of this difficulty, it was finally decided to leave any definition of the ‘investment’ to the consent of the parties …. [¶22] … Within this broad framework for the definition of investment under the ICSID Convention, the Tribunal also notes that a number of transactions have been identified as qualifying as investments in given circumstances. It has also been noted by commentators of the Convention, and during the history of its negotiation, that jurisdiction over loans, suppliers’ credits, outstanding payments, ownership of shares and construction contracts, among other aspects, was left to the discretion of the parties. [¶29] The Tribunal considers that the broad scope of Article 25 (1) of the Convention and the ensuing ICSID practice and decisions are sufficient, without more, to require a finding that the Centre’s jurisdiction and its own competence are well-founded. In addition, as explained above, loans qualify as an investment within ICSID’s jurisdiction, as does, in given circumstances, the purchase of bonds. Since promissory notes are evidence of a loan and a rather typical financial and credit instrument, there is nothing to prevent their purchase from qualifying as an investment under the Convention in the circumstances of a particular case such as this. This conclusion, however, has to be examined next in the context of the specific consent of the parties and other provisions which are controlling in the matter. [¶39] The claimant has rightly argued that promissory notes of this kind have a legal standing of their own, separate and independent from the underlying transaction. It is not disputed in this case that the Government of Venezuela foresaw the possibility that the promissory notes would be transferred and endorsed to subsequent holders, since they explicitly allow for such a possibility …. [¶40] … To the extent that this credit is provided by a foreign holder of the notes, it constitutes a foreign investment which in this case is encompassed by the terms of the Convention …. [¶42] … The promissory notes were issued by the Republic of Venezuela under the terms of the Law on Public Credit …, which specifically governs public credit operations aimed at raising funds and resources ‘to undertake productive works, attend to the needs of national interest and cover transitory needs of the treasury.’ It is quite apparent that the transactions involved in this case are not ordinary commercial transactions and indeed involve a fundamental public interest …. [¶43] The status of the promissory notes under the Law of Public Credit is also important as evidence that the type of investment involved is not merely a short-term, occasional financial arrangement, such as could happen with investments that come in for quick gains and leave immediately thereafter – i.e. ‘volatile capital.’ The basic features of an investment have been described as involving a certain duration, a certain regularity of profit and return, assumption of risk, a substantial commitment and a significance for the host State’s development.63 … 63

Christoph Schreuer, Commentary on the ICSID Convention (1996) ICSID Review – Foreign Investment Law Journal 11, 316, 372.

The Fedax Tribunal thus makes two important points. First, the travaux préparatoires of the ICSID Convention reveal that it was decided to leave any definition of the ‘investment’ to the consent of the parties. Second, the parties, however, do not have absolute discretion,

Major Decisions on the Definition of Investment 229 since investment needs to be distinguished from ‘ordinary commercial transactions’ or ‘volatile capital’ (see section I.B below). At the end of its analysis, the Tribunal adds that there are a certain number of ‘basic features of an investment’, referring to Christoph Schreuer’s commentary. Although these ‘basic features’ are mentioned only in passing in the Fedax decision, they played a decisive role in Salini v Morocco. In the latter case, Salini and Italstrade, two Italian companies, concluded a contract on the construction of highways in Morocco with the Société Nationale des Autoroutes du Maroc, a company incorporated under the Moroccan law for the purpose of constructing and maintaining highways pursuant to concessions granted by the Government of Morocco. The dispute arose from the different understanding of construction costs, following late completion of the highways. When the investors instituted arbitration pursuant to the Italy-Morocco BIT and the ICSID Convention, the Respondent objected to the Tribunal’s jurisdiction on the ground, among others, that there was no investment protected under the ICSID Convention. The Tribunal held that: [¶52] … [I]l serait inexact de considérer que l’exigence d’un différend « en relation directe avec un investissement » s’est diluée dans le consentement des parties contractantes …. À l’exception d’une décision du Secrétaire général du CIRDI refusant d’enregistrer une requête d’arbitrage concernant un différend né d’une simple vente …, les sentences disponibles ne se sont que très rarement penchées sur la notion d’investissement …. La doctrine considère généralement que l’investissement suppose des apports, une certaine durée d’exécution du marché et une participation aux risques de l’opération (cf. note de E. Gaillard [Journal du droit international, 1999], p. 292) …. La lecture du préambule de la Convention permet d’y ajouter le critère de la contribution au développement économique de l’État d’accueil de l’investissement. [¶53] Les apports effectués par les sociétés italiennes sont détaillés et évalués dans leurs écritures …. [¶54] Alors que la durée totale d’exécution du marché … avait été fixée à 32 mois, celle-ci a été prolongée à 36 mois. L’opération satisfait ainsi à la durée minimale observée par la doctrine, laquelle est de 2 à 5 ans …. [¶55] Quant aux risques encourus par les sociétés italiennes, ceux-ci découlent de la nature du contrat en cause …. [¶57] S’agissant enfin de la contribution du marché au développement économique de l’État marocain, celle-ci ne peut sérieusement être discutée. La construction des infrastructures relève, dans la plupart des pays, des tâches de l’État ou d’autres collectivités publiques. Il ne peut être sérieusement contesté que l’autoroute en cause servira l’intérêt public ….2

Thus, the Salini Tribunal, relying on ‘the doctrine’, established the general criteria to identify an ‘investment’ in the sense of Article 25(1) of the ICSID Convention and then verified if each of these criteria was satisfied in that case. This approach or the ‘Salini test’, which makes the identification of an ‘investment’ less subjective in the sense of Article 25(1) of the ICSID Convention, has been followed by a number of tribunals.3

2 Salini Costruttori v Maroc, Décision sur la compétence (23 juillet 2001) ICSID Case No ARB/00/4, (Robert Briner [President], Bernardo Cremades, Ibrahim Fadallah) [footnotes omitted]. 3 Bayindir v Pakistan, Decision on Jurisdiction (14 November 2005) ICSID Case No ARB/03/29 (Gabrielle Kaufmann-Kohler [President], Franklin Berman, Karl-Heinz Böckstiegel), para 130; Jan de Nul v Egypt, Decision on Jurisdiction (16 June 2006) ICSID Case No ARB/04/13, para 91 (Gabrielle Kaufmann-Kohler [President], Pierre Mayer, Brigitte Stern); Mitchell v Democratic Republic of the Congo, Decision on the Application for Annulment of

230 Shotaro Hamamoto However, various criticisms have been leveled at the Salini decision. Although quite a few tribunals followed the Salini approach itself, many of them refused to consider the element of ‘the contribution to the economic development of the host State’ as one of the criteria on the ground that the economic development of the host state is not a condition but rather a possible consequence of the investment.4 More fundamentally, many tribunals refused to follow the Salini approach, which they considered was not based on the ICSID Convention. Biwater is one of the early cases in which the Tribunal criticised the Salini approach. Biwater Gauff, a company incorporated under the laws of England and Wales, incorporated City Water under the laws of Tanzania. Forty-nine per cent of the shares in City Water were transferred to a Tanzanian company with Biwater Gauff retaining 51 per cent of the shares. City Water concluded contracts with a Tanzanian public corporation representing the state on water and sewerage service in Dar es Salaam, the capital city of Tanzania. Following many difficulties in the execution of the contracts, the local public corporation terminated the contracts and City Water’s senior management members were deported. In the arbitration engaged pursuant to the Tanzania-UK BIT, the host state argued that Biwater Gauff ’s 51 per cent ownership of City Water’s shares did not qualify as investments under Article 25 of the ICSID Convention on the ground that the water and sewerage service project in question was a ‘loss leader’ or an undertaking that was not economically rational standing alone but that might have led to other profitable opportunities later. For Tanzania, the project was an inevitably unprofitable venture and the criteria of ‘risk’ and ‘substantial commitment’ were not satisfied. The Tribunal held that: [¶310] In advancing submissions on Article 25 of the ICSID Convention, parties not infrequently begin with the proposition that the term ‘investment’ is not defined in the ICSID Convention, and

the Award (1 November 2006) ICSID Case No ARB/99/7 (Antonias Dimolitsa [President], Robert SM Dossou, Andrea Giardina), para 27; Saipem v Bangladesh, Decision on Jurisdiction and Recommendation on Provisional Measures (21 March 2007) ICSID Case No ARB/05/07 (Gabrielle Kaufmann-Kohler [President], Christoph H Schreuer, Philip Otton) para 99; Malaysian Historical Salvors v Malaysia, Award on Jurisdiction (17 May 2007) ICSID Case No ARB/05/10 (Michael Hwang [sole arbitrator]) paras 73–75; Kardassopoulos v Georgia, Decision on Jurisdiction (6 July 2007) ICSID Case No ARB/05/18 (L Yves Fortier [President], Francisco Orrego Vicuña, Arthur Watts), para 116; Noble Energy v Ecuador, Decision on Jurisdiction (5 March 2008) ICSID Case No ARB/05/12 (Gabrielle Kaufmann-Kohler [President], Bernardo M Cremades, Henri Alvarez) para 128. 4 Pey Casado v Chili, Award (8 May 2008) ICSID Case No ARB/98/2 (Pierre Lalive [President], Mohammed Chemloul, Emmanuel Gaillard) para 232; Toto Costruzioni Generali v Lebanon, Decision on Jurisdiction (11 September 2009) ICSID Case No ARB/07/12, (Hans van Houtte [President], Alberto Feliciani, Fadi Moghaizel) para 84, n 25; Fakes v Turkey, Award (14 July 2010) ICSID Case No ARB/07/20, (Hans van Houtte [President], Laurent Lévy, Emmanuel Gaillard), para 111; RSM v République Centrafricaine, Décision sur la compétence et la responsabilité (7 December 2010) ICSID Case No ARB/07/2 (Azzedine Kettani [Président] Philippe Merle, Brigitte Stern) para 56; SGS v Paraguay, Decision on Jurisdiction (12 February 2010) ICSID Case No ARB/07/29 (Stanimir A. Alexandrov [President], Donald Francis Donovan, Pablo García Mexía) para 107; Quiborax v Bolivia, Decision on Jurisdiction (27 September 2012) ICSID Case No ARB/06/2 (Gabrielle Kaufmann-Kohler [President], Marc Lalonde, Brigitte Stern) para 220. See also LESI-Dipenta v Algérie, Sentence (10 January 2005) Aff CIRDI No ARB/03/8 (Pierre Tercier [Président], André Faurès, Emmanuel Gaillard), Partie II, para 13; Elsamex v Honduras, laudo, (16 November 2012) ICSID Case No ARB/09/4, (Enrique Gómez Pinzón [sole arbitrator]) para 264; Gavazzi v Romania, Decision on Jurisdiction Admissibility and Liability (21 April 2015) ICSID Case No ARB/12/25 (Hans van Houtte [President], VV Veeder, Mauro Rubino-Sammertano) para 114; İçkale İnşaat Limited Şirketi v Turkmenistan, Award (8 March 2016) ICSID Case No ARB/10/24 (Veijo Heiskanen [President], Carolyn B Lamm, Philippe Sands) para 291; MNSS v Montenegro, Award (4 May 2016) ICSID Case No ARB(AF)/12/8, (Andrés Rigo Sureda [President], Emmanuel Gaillard, Brigitte Stern) paras 189–190; Capital Financial Holdings Luxembourg v Cameroun, Sentence (22 June 2017) ICSID Case No ARB/15/18, (Pierre Tercier [President], Alexis Mourre, Alain Pellet) para 422; Casinos Austria International v Argentina, Decision on Jurisdiction (29 June 2018) ICSID Case No ARB/14/32 (Hans van Houtte [President], Stephan W Schill, Santiago Torres Bernardez) para 190.

Major Decisions on the Definition of Investment 231 then proceed to apply each of the five criteria, or benchmarks, that were originally suggested by the arbitral tribunal in Fedax v. Venezuela, and re-stated (notably) in Salini v. Morocco, namely (i) duration; (ii) regularity of profit and return; (iii) assumption of risk; (iv) substantial commitment; and (v) significance for the host State’s development. [¶312] In the Tribunal’s view, there is no basis for a rote, or overly strict, application of the five Salini criteria in every case. These criteria are not fixed or mandatory as a matter of law. They do not appear in the ICSID Convention …. [¶313] Given that the Convention was not drafted with a strict, objective, definition of ‘investment’, it is doubtful that arbitral tribunals sitting in individual cases should impose one such definition which would be applicable in all cases and for all purposes …. [¶314] Further, the Salini Test itself is problematic if, as some tribunals have found, the “typical characteristics” of an investment as identified in that decision are elevated into a fixed and inflexible test, and if transactions are to be presumed excluded from the ICSID Convention unless each of the five criteria are satisfied. This risks the arbitrary exclusion of certain types of transaction from the scope of the Convention …. [¶316] The Arbitral Tribunal therefore considers that a more flexible and pragmatic approach to the meaning of ‘investment’ is appropriate, which takes into account the features identified in Salini, but along with all the circumstances of the case, including the nature of the instrument containing the relevant consent to ICSID. [¶320] [Biwater Gauff] intended the project to be profitable, albeit with a relatively low rate of return. It invested substantial amounts of equity into City Water; it posted substantial Performance Bonds; and it advanced to the company a shareholder loan of USD 1 million, together with additional support and development funds. It also considered the Project to be in its economic interest and supplied the company with key personnel and also with technology and know-how. The Arbitral Tribunal considers that, viewed overall, and against the backdrop of the scope of consent to ICSID jurisdiction as expressed in the BIT, this was clearly an ‘investment’ within the meaning of Article 25 of the ICSID Convention. Further, even if such are required for the purposes of Article 25 of the Convention, the conditions of ‘risk’ and ‘commitment’ which are disputed by the Republic were present in BGT’s investment.5

Although the Salini decision still continues to be strictly relied on by some arbitral tribunals,6 many more recent tribunals and ad hoc committees take ‘a more flexible and pragmatic approach’ adopted by the Biwater Tribunal, which allows tribunals to carry out a holistic assessment taking into account all relevant elements.7

5 Biwater Gauff v Tanzania, Award (24 July 2008) ICSID Case No ARB/05/22, (Bernard Hanotiau [President], Gary Born, Toby Landau) [footnotes omitted]. 6 Meerapfel Söhne v République centrafricaine, Sentence arbitrale (12 May 2011) ICSID Case No ARB/07/10, (Azzedine Kettani [President], François T’Kint, Marie-Madeleine Mborantsuo), para 184; Flughafen Zürich y Venezuela, laudo, (18 November 2014), ICSID Case No ARB/10/19, (Juan Fernández-Armesto [President], Henri Alvarez, Raúl E Vinuesa) paras 245–247; Houben v Burundi, Sentence (12 January 2016) ICSID Case No ARB/13/7 (Gilbert Guillaume [President], Yas Banifatemi, Brigitte Stern) paras 112–115; Karkey Karadeniz Elektrik Uretim v Pakistan, Award (22 August 2017) ICSID Case No ARB/13/1, (Yves Derains [President], David A O Edward, Horacio A Grigera Naón), paras 633–636; Unión Fenosa Gas v Egypt, Award (31 August 2018) ICSID Case No ARB/14/4, (V V Veeder [President], J William Rowley, Mark Clodfelter) para 6.66. 7 Malaysian Historical Salvors v Malaysia, Decision on the Application for Annulment (16 April 2009) ICSID Case No ARB/05/10 (Stephen M Schwebel [President], Mohamed Shahabuddeen, Peter Tomka), paras 78–79; Inmaris v Ukraine, Decision on Jurisdiction (8 March 2010) ICSID Case No ARB/08/8 (Stanimir A Alexandrov [President], Bernardo Cremades, Noah Rubins) paras 129–131; Alpha Projektholding v Ukraine, Award (8 November 2010) ICSID Case No ARB/07/16, (Davis R Robinson [Chairman], Stanimir A Alexandrof, Yoram Turbowicz) paras 311–314; Deutsche Bank v Sri Lanka, Award (31 October 2012) ICSID Case No ARB/09/2 (Bernard Hanotiau [President], Makhdoom Ali Khan, David A R Williams) para 294; Philip Morris v Uruguay, Decision on Jurisdiction (2 July 2013) ICSID Case No ARB/10/7 (Piero Bernardini [President], Gary Born, James Crawford) paras 204–206; Société civile

232 Shotaro Hamamoto B. The Exclusion of Commercial Transaction This ‘more flexible and pragmatic approach’ is often called a ‘subjective’ approach, as tribunals adopting such approach tend to place much emphasis upon the agreement between the parties to the International Investment Agreement (IIA). If they agree that disputes arising out of an ‘investment’ as defined in their IIA may be brought to arbitration under the ICSID Convention, why should the Tribunal decline its jurisdiction? ‘In reliance on the consensual nature of the [ICSID] Convention, [the drafters] preferred giving the parties the greatest latitude to define these terms themselves, provided that the criteria agreed upon by the parties are reasonable and not totally inconsistent with the purposes of the Convention.’8 However, characterising such an approach as subjective is somewhat misleading as it does not grant the parties to the IIA total discretion. It is in fact commonplace among such tribunals to exclude ‘ordinary commercial transactions’ as the abovementioned Fedax Tribunal did. In the Malaysia Historical Salvors case, a Malaysian marine salvage company owned by a British adventurer contracted with the Government of Malaysia to locate and salvage the cargo of the British vessel Diana, which sank off the coast of Malacca in 1817. The company recovered a number of pieces of cargo, which were sold at auction. The dispute then arose out of the distribution of the amount realised at the auction. The Tribunal followed the Salini approach and declined jurisdiction on the ground that the contract did not make any significant contributions to the economic development of Malaysia.9 The award on jurisdiction was, however, annulled by the ad hoc committee, which held: [¶69] … Little more about the nature of outer limits is indicated in the travaux than is contained in Article 25(1), namely that, ‘[t]he jurisdiction of the Centre shall extend to any legal dispute arising directly out of an investment ….’ It appears to have been assumed by the Convention’s drafters that use of the term ‘investment’ excluded a simple sale and like transient commercial transactions from the jurisdiction of the Centre. [¶71] The preparatory work of the Convention as well as the Report of the Executive Directors thus shows that: (a) deliberately no definition of ‘investment’ as that term is found in Article 25(1) was adopted; (b) a floor limit to the value of an investment was rejected; (c) a requirement of indefinite duration of an investment or of a duration of no less than five years was rejected; (d) the critical criterion adopted was the consent of the parties. By the terms of their consent, they could define jurisdiction under the Convention. [¶80] … It is [the Committee’s] considered conclusion that the Tribunal exceeded its powers by failing to exercise the jurisdiction with which it was endowed by the terms of the [Malaysia-UK] Agreement and the [ICSID] Convention, and that it ‘manifestly’ did so, for these reasons: …

immobilière de Gaëta v Guinée, Sentence (21 December 2015) ICSID Case No ARB/12/36 (Pierre Tercier [President], Horacio A Grigera Naón, Laurent Lévy) paras 206–215; Garanti Koza v Turkmenistan, Award (19 December 2016) ICSID Case No ARB/11/20, (John M. Townsend [President], George Constantine Lambrou, Laurence Boisson de Chazournes) paras 238–242; Gavrilovic v Croatia, Award (26 July 2018) ICSID Case No ARB/12/39 (Michael C Pryles [President], Stanimir A Alexandrov, J Christopher Thomas) paras 191–193; Standard Chartered Bank (Hong Kong) Limited v Tanzania, Award of the Tribunal (11 October 2019) ICSID Case No ARB/15/41 (Lawrence Boo [President], David Unterhalter, Kamal Hossain) para 200. 8 Autopista Concesionada de Venezuela v Venezuela, Decision on jurisdiction (27 September 2001) ICSID Case No ARB/00/5 (Gabrielle Kaufmann-Kohler [President], Karl-Heinz Böckstiegel, Bernardo M. Cremades) para 97. 9 Malaysian Historical Salvors, above (n 3) para 143.

Major Decisions on the Definition of Investment 233 (b) its analysis of these criteria [of investment under Article 25 of the ICSID Convention] elevated them to jurisdictional conditions, and exigently interpreted the alleged condition of a contribution to the economic development of the host State so as to exclude small contributions, and contributions of a cultural and historical nature ….10

C. The Good Faith Requirement Another problem relating to the definition of ‘investment’ under Article 25 of the ICSID Convention is whether or not it includes a good faith element. The problem was first raised in the Phoenix case. Two Czech companies, Benet Praha and Benet Group, were involved in trading of ferroalloys. In 2001, the Czech authorities commenced a criminal investigation against Mr X, who was at that time Benet Praha’s Executive Officer. The investigation was related to an allegation of a series of tax and custom duty evasions and income tax fraud. In 2002, Phoenix, an Israeli company established and entirely owned by Mr X after he had fled to Israel, acquired all interests in the two Czech companies, at the time when they were involved in ongoing legal disputes with a private party and the Czech fiscal authorities. Benet Praha was entirely owned by Mr X’s wife, and Benet Group by Mr X’s wife and daughter. In 2004, Phoneix submitted a request for arbitration to the ICSID pursuant to the Czech Republic-Israel BIT. During the arbitral proceedings, in 2008, Phoenix sold Benet Praha to another Czech company entirely owned by Mr X’s wife. The Tribunal held that: [¶136] … Phoenix bought an ‘investment’ that was already burdened with the civil litigation as well as the problems with the tax and customs authorities …. [¶140] … There are strong indicia that no economic activity in the market place was either performed or even intended by Phoenix. No business plan, no program of re-financing, no economic objectives were ever presented, no real valuation of the economic transactions were ever attempted …. It is not contested by the Claimant that, at the time of the alleged investment, when Phoenix bought the two Czech companies, they had no activity …. But a more important feature is that no activity was either launched or tried after the alleged investment was made …. [¶141] So, why did the Claimant invest in the Czech Republic, if not to develop economic activity, as in any investment? … [¶142] The evidence indeed shows that the Claimant made an ‘investment’ not for the purpose of engaging in economic activity, but for the sole purpose of bringing international litigation against the Czech Republic …. The unique goal of the ‘investment’ was to transform a pre-existing domestic dispute into an international dispute subject to ICSID arbitration under a bilateral investment treaty. This kind of transaction is not a bona fide transaction and cannot be a protected investment under the ICSID system. [¶143] … All the elements analyzed lead to the same conclusion of an abuse of rights. The abuse here could be called a ‘détournement de procédure’, consisting in the Claimant’s creation of a legal fiction in order to gain access to an international arbitration procedure to which it was not entitled ….

10 Malaysian Historical Salvors, annulment, above (n 7). See also Alpha Projektholdin, above (n 7) paras 313–314; Inmaris, above (n 7) paras 106, 131; Global Trading Resource v Ukraine, Award (1 December 2010) ICSID Case No ARB/09/11 (Franklin Berman [President], Emmanuel Gaillard, Christopher Thomas) para 55; SGS v Paraguay, above (n 4) para 93.

234 Shotaro Hamamoto [¶144] The conclusion of the Tribunal is therefore that the Claimant’s initiation and pursuit of this arbitration is an abuse of the system of international ICSID investment arbitration. If it were accepted that the Tribunal has jurisdiction to decide Phoenix’s claim, then any pre-existing national dispute could be brought to an ICSID tribunal by a transfer of the national economic interests to a foreign company in an attempt to seek protections under a BIT. Such transfer from the domestic arena to the international scene would ipso facto constitute a ‘protected investment’ – and the jurisdiction of BIT and ICSID tribunals would be virtually unlimited. It is the duty of the Tribunal not to protect such an abusive manipulation of the system of international investment protection under the ICSID Convention and the BITs. It is indeed the Tribunal’s view that to accept jurisdiction in this case would go against the basic objectives underlying the ICSID Convention as well as those of bilateral investment treaties. The Tribunal has to ensure that the ICSID mechanism does not protect investments that it was not designed for to protect, because they are in essence domestic investments disguised as international investments for the sole purpose of access to this mechanism. [¶145] It follows from these findings that the Tribunal lacks jurisdiction over the Claimant’s request, as the Tribunal concludes that the Claimant’s purported investment does not qualify as a protected investment under the Washington Convention and the Israeli/Czech BIT.11

The Phoenix award is important as it shows that the Tribunal may refuse to exercise its jurisdiction over investments made in bad faith or in violation of the domestic law of the host state. However, while the conclusion arrived at by the Phoenix Tribunal is widely shared, its reasoning is not. Recent tribunals agree that ‘a breach of the general prohibition of abuse of right, which is a manifestation of the principle of good faith, may give rise to an objection to jurisdiction or to a defense on the merits’.12 However, ‘[t]his does not mean that these elements [good faith and legality] are part of the objective definition of the term ‘investment’ contained in Article 25(1) of the ICSID Convention’13 because ‘the principles of good faith and legality cannot be incorporated into the definition of Article 25(1) of the ICSID Convention without doing violence to the language of the ICSID Convention: an investment might be “legal” or “illegal,” made in “good faith” or not, it nonetheless remains an investment’.14 In other words, ‘[i]t may not be a protected investment, i.e. deserve protection in the sense that access to treaty arbitration and/or substantive treaty guarantees may not be granted, but that is a different matter’.15

II. ‘INVESTMENT’ UNDER INTERNATIONAL INVESTMENT AGREEMENTS

A. Diversity of Treaty Definitions Each IIA has its own definition of investment that it protects. Many IIAs define ‘investment’ as ‘every kind of asset’ and add a non-exhaustive list of examples. For example, the 11 Phoenix v Czech Republic, Award (15 April 2009) ICSID Case No ARB/06/5, (Brigitte Stern [President], Andreas Bucher, Juan Fernández-Armesto) (footnotes omitted) (emphasis in the original text). 12 Metal-Tech v Uzbekistan, Award (4 October 2013) ICSID Case No ARB/10/3 (Gabrielle Kaufmann-Kohler [President], John M. Townsend, Claus von Wobeser) para 127. 13 ibid. 14 Saba Fakes v Turkey, Award (14 July 2010) ICSID Case No ARB/07/20, para 112 (Emmanuel Gaillard [President], Hans van Houtte, Laurent Lévy). 15 Quiborax, above (n 4) para 226. It goes without saying that ‘for an investment such as a licence, which is the creature of the laws of the Host State, to qualify for protection, it must be made in accordance with the laws of the Host State’. Cortec v Kenya, Award (22 October 2018) ICSID Case No ARB/15/29 (Ian Binnie [President], Kanaga Dharmananda, Brigitte Stern) para 319.

Major Decisions on the Definition of Investment 235 Netherlands-Venezuela BIT, pursuant to which Fedax v Venezuela was brought to arbitration, provides in its Article 1(a): the term ‘investments’ shall comprise every kind of asset and more particularly though not exclusively: i) movable and immovable property, as well as any other rights in rem in respect of every kind of asset; ii) rights derived from shares, bonds, and other kinds of interests in companies and joint ventures; iii) title to money, to other assets or to any performance having an economic value; iv) rights in the field of intellectual property, technical processes, goodwill and know-how; v) rights granted under public law, including rights to prospect, explore, extract, and win natural resources.

IIAs falling under this category adopt, however, a wide variety of ways to define ‘investment’ by adding some additional requirements. Thus, Article 1(1) of the Argentina-Italy BIT provides: Investment shall mean … any conferment or asset invested or reinvested by an individual or corporation of one Contracting Party in the territory of the other Contracting Party, in compliance with the laws and regulations of the latter party.16

Some of the recent IIAs contain a circular requirement. For example, Article 9.1 of the CPTPP provides: investment means every asset that an investor owns or controls, directly or indirectly, that has the characteristics of an investment, including such characteristics as the commitment of capital or other resources, the expectation of gain or profit, or the assumption of risk.17

There are also IIAs that define ‘investment’ by a closed list. Article 1139 of NAFTA provides: investment means: (a) (b) (c) … (d) … (e)

an enterprise; an equity security of an enterprise; a debt security of an enterprise a loan to an enterprise

an interest in an enterprise that entitles the owner to share in the assets of that enterprise on dissolution, other than a debt security or a loan excluded from subparagraph (c) or (d); (f) real estate or other property, tangible or intangible, acquired in the expectation or used for the purpose of economic benefit or other business purposes; and (g) interests arising from the commitment of capital or other resources in the territory of a Party to economic activity in such territory …, but investment does not mean, i. claims to money that arise solely from 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, or

16 Emphasis 17 Emphasis

added. added.

236 Shotaro Hamamoto ii. the extension of credit in connection with a commercial transaction, such as trade financing, other than a loan covered by subparagraph (d); or iii. any other claims to money, that do not involve the kinds of interests set out in subparagraphs (a) through (h)

It evidently follows that something which qualifies as an investment under one IIA does not necessarily do so under another IIA. The Poštová Banka Tribunal makes that clear. Poštová Banka purchased a series of Greek Government Bonds on the secondary market in 2010. The distribution of bonds occurred electronically through universal depositories. Poštová Banka’s interests in the Bonds were in the form of a book-entry in an account, which did not entitle Poštová Banka to rights in any specific instrument but only to rights to a pool of fungible interests. Following the economic crisis, Greece passed the Greek Bondholder Act of 2012, which led to the restructuring of its sovereign debt, exonerating Greece from reimbursing some of the sovereign bonds detained by the private sector. When Poštová Banka brought Greece to arbitration, the latter argued that the rights held by Poštová Banka did not qualify as ‘investment’ under the Greece-Slovakia BIT. The Tribunal held that: [¶300] The tribunal in Abaclat v. Argentina … paid due regard to the list of examples contained in Article 1(1) of the Argentina-Italy BIT … [¶301] The conclusion of the Abaclat tribunal was that the terms ‘obligations’ and ‘public securities’ were wide enough to encompass the bonds that were the subject of the dispute in that arbitration. [¶302] The Ambiente Ufficio tribunal also considered in detail the list of examples in Article 1(1) of the Argentina-Italy BIT as a key element to its conclusion that bonds are covered either because the word ‘obligaciones’ or ‘obligatzioni’ should be translated as bonds, or because what the tribunal calls a ‘catch-all clause’ covers rights derived from law or contract. [¶304] The language in the Slovakia-Greece BIT, as will be analysed below, is significantly different from the one that led the Abaclat and Ambiente Ufficio tribunals to conclude that government bonds were investments under the Argentina-Italy BIT. [¶306] … [T]he list of examples of Article 1(1) of the Slovakia-Greece BIT, and particularly the sections invoked by Claimants in support of their interpretation of the relevant BIT, are substantially different from the ones invoked by the Abaclat and Ambiente Ufficio tribunals under the Argentina-Italy BIT. Article 1(c) of the Argentina-Italy BIT, invoked by the Abaclat and Ambiente Ufficio tribunals as the basis for their conclusion, includes amongst the illustrative list of what may constitute an investment ‘obligations, private or public titles or any other right to performances or services having economic value, including capitalized revenues.’ (Emphasis added) In contrast, Article 1.1(c) of the Slovakia-Greece BIT refers to ‘loans, claims to money or to any performance under contract having a financial value.’ There is no reference in the Slovakia-Greece BIT to a general concept such as ‘obligations,’ much less to ‘public titles.’18

It follows that it is difficult or even meaningless to establish the overarching definition of ‘investment’ applicable to all IIAs. Each IIA carries its own definition of ‘investment’, which needs to be determined in accordance with the rules of treaty interpretation. That said, there are a certain number of problems common to many IIAs that contain similar requirements.

18 Poštová Banka v Greece, Award (9 April 2015) ICSID Case No ARB/13/8 (Eduardo Zuleta [President], Brigitte Stern, John M Townsend)(footnotes omitted).

Major Decisions on the Definition of Investment 237 B. Requirements Common to Many IIAs As mentioned above, some IIAs provide that they cover investments ‘made in the territory’ of the host state. In Abaclat, the meaning of the term ‘made in the territory’ was hotly debated. In 2001, Argentina defaulted by publicly announcing the deferral of over US$100 billion of external bond debt owed to both Argentine and non-Argentine creditors. In 2005, Argentina launched the Exchange Offer 2005, pursuant to which bondholders could exchange different series of bonds, on which Argentina had suspended payment in 2001, for new debt that Argentina would issue. In 2006, Italian claimants who did not accept the exchange offer filed the request for arbitration with the ICSID. The bonds at stake were divided into smaller negotiable economic values, ie, securities. The security entitlements are the result of the distribution process of the bonds through their division into a multitude of smaller securities representing each a part of the value of the relevant bond. Strictly speaking, the Claimants were not bondholders but had such security entitlements. As mentioned above, Article 1(1) of the Argentina-Italy BIT provides that investments shall mean any conferment or asset invested or reinvested by an individual or corporation of one Contracting Party in the territory of the other Contracting Party. Argentina argued that the Claimants’ security entitlements were not made in Argentina because: (i) they did not cause any transfer of money into the territory of Argentina; (ii) they were located outside of Argentina and were beyond the latter’s scope of territorial jurisdiction based on the foreign law and forum selection clauses contained in the relevant bond documents; and (iii) the indirect holding systems of these entitlements implicated a cut-off point beyond which claims were not permissible because they had only a remote connection with the investment. The Tribunal held that: [¶374] [With regard to the place where the investment is considered made,] [t]he Tribunal finds that the determination of the place of the investment firstly depends on the nature of such investment …. With regard to investments of a purely financial nature, the relevant criteria should be where and/or for the benefit of whom the funds are ultimately used, and not the place where the funds were paid out or transferred. Thus, the relevant question is where the invested funds ultimately made available to the Host State and did they support the latter’s economic development? This is also the view taken by other arbitral tribunals.147 147 See e.g. Fedax N.V. v. Republic of Venezuela (ICSID Case No. ARB/96/3), Decision of the Tribunal on Objections to Jurisdiction of 11 July 1997, § 41. See also Société Générale de Surveillance S.A. v. Islamic Republic of Pakistan (ICSID Case No. ARB/01/13), Decision of the Tribunal on Objections to Jurisdiction of 6 August 2003, §§ 136-140, where emphasis was led on the fact that the aim of SGS‘s activity was to ‘raise the financial revenue of the State’ (§ 139); SGS v. Republic Philippines [Société Générale de Surveillance v. Republic of the Philippines, (ICSID Case ARB/02/6), Decision of 29 January 2004], § 111.

[¶375] A further question is whether it is necessary that investment of purely financial nature be further linked to a specific economic enterprise or operation taking place in the territory of the Host State. Based on the above consideration (see §355) that in Article 1 BIT Argentina and Italy designated financial instruments as an express kind of investment covered by the BIT and thereby intending to provide such investment with BIT protection, the Tribunal considers that it would be contrary to the BIT’s wording and aim to attach a further condition to the protection of financial investment instruments. [¶376] Respondent makes an additional argument out of the fact that the payment of the purchase price occurred after the payment of the lump sum price by the underwriters, and that only the latter payment can be considered to have been made available to Argentina. The Tribunal is of

238 Shotaro Hamamoto the opinion that such argument ignores the reality of the bond issuance process. Indeed, although the payment of the lump sum price for the bonds and the payment of the purchase price by the individual holders of security entitlements happened at different points in time, the latter constitutes the basis for the former. As mentioned above (see §359), the bonds and the security entitlements are part of one and the same economic operation and they make only sense together …. [¶377] Thus, the funds generated by the purchase of the relevant security entitlements are – for the purpose of establishing where they were made – no different than the lumps sum payment paid by the underwriters for the bonds. [¶378] There is no doubt that the funds generated through the bonds issuance process were ultimately made available to Argentina, and served to finance Argentina’s economic development. Whether the funds were actually used to repay pre-existing debts of Argentina or whether they were used in government spending is irrelevant. In both cases, it was used by Argentina to manage its finances, and as such must be considered to have contributed to Argentina’s economic development and thus to have been made in Argentina. [¶379] (ii) With regard to the question whether the presence of forum selection clauses in contractual documents relating to the investment may influence the determination of the place of investment, … [t]he Tribunal considers Respondent’s argument inapposite for two main reasons: – … [F]orum selection clauses are clauses of a procedural nature aiming to determine the place of settlement of a dispute relating to a contractual performance. They have nothing to do with the place where a party is supposed to perform its obligations; – … [R]ights and obligations deriving from the BIT have an independent basis from rights and obligations deriving from the contract, and may as such – in principle – not be affected by contractual provisions dealing only with contractual rights and obligations. [¶380] Consequently, the Tribunal finds that the relevant bonds and Claimants’ security entitlements therein are both to be considered ‘made in the territory of Argentina’.19

This decision was made by majority and Arbitrator Georges Abi-Saab appended a dissenting opinion. [¶77] … The alleged investment, the security entitlements, are not located in Argentina by applying either the legal or the material criteria for determining such location. [¶78] 1) The legal criteria: the financial securities instruments that constitute the alleged investment, i.e. the security entitlements in Argentinean bonds, have been sold in international financial markets, outside Argentina, with choice of law and forum selection clauses subjecting them to laws and fora foreign to Argentina. In fact, they were intentionally situated outside Argentina and out of reach of its laws and tribunals. There is no way then to say (and no legal basis for saying) that they are legally located in Argentina. [¶80] The majority award appears … to suggest that the determinative factor is the place of performance. If this criterion is applied, there can be no doubt that the place of performance under the securities instruments at issue is invariably outside Argentina, given the use of fiscal agents, paying agents, depositories and places of payment all situated outside Argentina. Factors other than the place of performance also point to the location of the securities in question outside Argentina. [¶81] Among these other factors, I consider a clause selecting courts external to the host State, as a prominent one in determining where the security instruments, and the underlying right to repayment of the debt, are located.

19 Abaclat and Others v Argentina, Decision on Jurisdiction and Admissibility (4 August 2011) ICSID Case No ARB/07/5 (Pierre Tercier [President], Georges Abi-Saab, Albert Jan van den Berg) (footnotes partly omitted).

Major Decisions on the Definition of Investment 239 [¶82] To answer the question whether the securities in question are located in Argentinean territory, this Tribunal needs to determine the situs of the debt – i.e. the alleged investment …. The foreign governing law and foreign forum, while not determinative by themselves, are important factors in determining situs. Other factors include the currency of payment, the place of payment and the residence of the intermediaries. On any of these criteria, the transactions at issue here were deliberately structured so as to have their situs outside Argentina. [¶84] [As for the majority’s argument that forum selection clauses are irrelevant,] this facile escape route (échappatoire) is to no avail in the instant case. A treaty claim is necessarily based on a right that has been allegedly violated; here, the debt that was not repaid. If this right is created by contract, it is the contract that governs its legal existence and the modalities of this existence, including the location of this right (and its reciprocal obligation). And the right in the present case has been purposefully located outside Argentina.

The approach taken by the majority in the Abaclat decision is generally adopted by other tribunals dealing with financial activities.20 Another question frequently raised is that of compliance with the domestic law of the host state, as in Fraport v the Philippines. In 1999, Fraport, a company in the international airport business, acquired shares in PIATCO, a Philippine company holding the concession rights for the construction and operation of a new terminal in a Philippine airport. Initially, Fraport acquired 25 per cent of PIATCO. However, as the other shareholders were unwilling or unable to make further contributions or even basic contributions when PIATCO needed further funds, Fraport increased its direct interest to 30 per cent and acquired 31.44 per cent of indirect ownership interests through minority shareholdings in a cascade of Philippine companies. This complex ownership structure was adopted taking into account the Philippine constitutional requirement that foreign nationals could own no more than 40 per cent of the capital of a public utility and the prohibitions under the Anti-Dummy Law (ADL) that prevented foreign nationals from intervening in the management, operation, administration or control of a company that was a public utility. At the same time, Fraport concluded secret shareholders agreements, pursuant to which the shareholders were required to act upon the recommendations made by Fraport on certain issues such as the operation, maintenance and management of the terminal. On all the other issues, if no unanimous position was reached among the shareholders, the issue was to be referred to arbitration. Fraport also extended loans and loan and payment guarantees to PIATCO, the cascade companies and PIATCO’s lenders and contractors. In 2001, public opposition against the project arose, provoked by Philippine businesses that criticised the involvement of a foreign company. The Philippine Government came to the conclusion that the concession agreement was null and void because PIATCO had not possessed the financial prequalification requirements. In 2003, when the terminal was almost fully built, the Philippine Supreme Court held that the concession agreements regarding the terminal project were null and void due to serious violations of Philippine law and public

20 Fedax (Case 1), paras 41–43; Deutsche Bank, above (n 7) paras 288–292. In British Carribean, the parties to the dispute were in agreement that the location of a financial instrument was to be assessed on the basis of the location of the benefit of that investment. British Caribbean Bank v Belize, Award (19 December 2014) PCA Case No 2010-18, (Albert Jan van den Berg [President], John Beechey, Rodrigo Oreamuno) paras 206–207. In Alemanni, the same question as in Abaclat was raised on the basis of similar facts. The Alemanni tribunal, however, decided not to deal with it in its decision on jurisdiction and admissibility and postponed it to the merits. Alemanni and others v Argentina, Decision on Jurisdiction and Admissibility (17 November 2014) ICSID Case No ARB/07/8 (Franklin Berman [President], Karl-Heinz Böckstiegel, J Christopher Thomas) para 297. The case was later discontinued and no decision on the merits was issued.

240 Shotaro Hamamoto policy. Later in the same year, Fraport filed its request for arbitration with the ICSID. In 2004, the Philippine Government instituted expropriation proceedings before a domestic court. Although the Government assured that prompt and just compensation would be paid in accordance with Philippine law, the expropriation proceedings were pending when the ICSID award was rendered. The Tribunal held that: [¶335] Article 1(1) of the [Germany-Philippines] BIT provides that for the purpose of this Agreement ‘the term ‘investment’ shall mean any kind of asset accepted in accordance with the respective laws and regulations of either Contracting State […]’. The qualification ‘accepted in accordance with the respective laws and regulations of either Contracting State’ applies to every form of investment covered by the BIT. Article 2(1) provides: ‘Each Contracting State shall promote as far as possible investments in its territory by investors of the other Contracting State and admit such investments in accordance with its Constitution, laws and regulations as referred to Article 1, paragraph 1. […]’ [¶346] There is, however, the question of estoppel. Principles of fairness should require a tribunal to hold a government estopped from raising violations of its own law as a jurisdictional defense when it knowingly overlooked them and endorsed an investment which was not in compliance with its law. [¶347] But a covert arrangement, which by its nature is unknown to the government officials who may have given approbation to the project, cannot be any basis for estoppel. There is no indication in the record that the Republic of the Philippines knew, should have known or could have known of the covert arrangements which were not in accordance with Philippine law when Fraport first made its investment in 1999. [¶355] … In the context of the internal Fraport documents, the secret shareholder agreements show that Fraport from the outset understood, with precision, the Philippine legal prohibition but believed that if it complied with it, the prospective investment could not be profitable. So it elected to proceed with the investment by secretly violating Philippine law through the secret shareholder agreements. These agreements evidence that Fraport planned and knew that its investment was not ‘in accordance’ with Philippine law. [¶396] When the question is whether the investment is in accordance with the law of the host state, considerable arguments may be made in favour of construing jurisdiction ratione materiae in a more liberal way which is generous to the investor. In some circumstances, the law in question of the host state may not be entirely clear and mistakes may be made in good faith. [¶397] In this case, the comportment of the foreign investor, as is clear from its own records, was egregious and cannot benefit from presumptions which might ordinarily operate in favour of the investor. [¶398] The record indicates that (i) the BIT explicitly and reiteratedly required that an investment, in order to qualify for BIT protection, had to be in accordance with the host state’s law and (ii) local counsel explicitly warned that a particular structural arrangement would violate a serious provision of Philippine law. Moreover, the violation qua violation was explicitly discussed at the level of the Board of Directors …. Thus the violation could not be deemed to be inadvertent and irrelevant to the investment. It was central to the success of the project. The awareness that the arrangements were not in accordance with Philippine law was manifested by the decision to make the arrangements secretly and to try to make them effective under foreign law. All of these facts derive from internal Fraport documents whose credibility can hardly be impeached by Fraport. [¶401] Fraport knowingly and intentionally circumvented the ADL by means of secret shareholder agreements. As a consequence, it cannot claim to have made an investment ‘in accordance with law’. Nor can it claim that high officials of the Respondent subsequently waived the legal requirements

Major Decisions on the Definition of Investment 241 and validated Fraport’s investment, for the Respondent’s officials could not have known of the violation. Because there is no ‘investment in accordance with law’, the Tribunal lacks jurisdiction ratione materiae.21

It should be noted that what was in question in the Fraport case was the legality at the initiation of the investment. Since the Germany-Philippine BIT defines the investment as the one ‘made’ or ‘invested’ in accordance with the domestic law of the host state, any investment made or invested not in accordance with the domestic law does not qualify as ‘investment’ under the BIT. In cases where the applicable IIAs contain such a definition, the legality in the performance of the investment does not bear upon the scope of application of the IIA or the Tribunal’s jurisdiction, though it may well be relevant in the context of the substantive merits of a claim brought under the IIA.22 For the relevance of the conformity with the domestic law of the host state to the definition of investment under Article 25 of the ICSID Convention, see the Phoenix v Czech Republic case and the accompanying text above.

21 Fraport v the Philippines, Award (16 August 2007) ICSID Case No ARB/03/25 (L Yves Fortier [President], Bernardo M Cremades, W Michael Reisman). This award was annulled by the ad hoc committee, which found a serious departure from a fundamental rule of procedure (Art 52(1)(d), ICSID Convention) due to the tribunal’s failure to accord the Claimant an opportunity to make submissions on relevant issues. Fraport v the Philippines, Decision on the application for annulment of Fraport AG Frankfurt Airport Services Worldwide (23 December 2010) ICSID Case No ARB/03/25 (Peter Tomka [President], Dominique Hascher, Campbell McLachlan) paras 218–247. In the resubmitted case, the tribunal also found that it lacked jurisdiction because the investment in question was not made in accordance with Philippine law. Fraport v the Philippines, Award (10 December 2014) ICSID Case No ARB/11/12 (Piero Bernardini [President], Stanimir A Alexandrov, Albert Jan van den Berg) paras 442–468. 22 Hamester v Ghana, Award (18 June 2010) ICSID Case No ARB/07/24 (Brigitte Stern [President], Bernardo Cremades, Toby Landau) para 127; Teinver v Argentina, Decision on Jurisdiction (21 December 2012) ICSID Case No ARB/09/1, (Thomas Buergenthal [President], Henri C Alvarez, Kamal Hossain) para 257; Vannessa Ventures v Venezuela, Award (16 January 2013) ICSID Case No ARB(AF)/04/6 (Vaughan Lowe [President], Charles N Brower, Brigitte Stern) para 167; Urbaser v Argentina, Decision on Jurisdiction (19 December 2012) ICSID Case No ARB/07/26 (Andreas Bucher [President], Pedro J Martinez-Fraga, Campbell McLachlan) para 260; Vladislav Kim v Uzbekistan, Decision on Jurisdiction (8 March 2017) ICSID Case No ARB/13/6 (David D Caron [President], L Yves Fortier, Toby Landau) paras 374–377; Aven v Costa Rica, Final Award (18 September 2018) ICSID Case No UNCT/15/3 (Eduardo Siqueiros [President], C Mark Baker, Pedro Nikken) para 342.

15 The Concept of Protected Investor ALEXANDRA VAN DER MEULEN*

I. THE SIGNIFICANCE OF THE CONCEPT OF ‘INVESTOR’

T

RADITIONALLY, INTERNATIONAL LAW on nationality has been shaped by cases involving diplomatic protection.1 In that context, defining nationality was essential insofar as it was the bond of nationality that conditioned – and provided the basis for – the right of a state to bring an international claim. The proliferation of investment treaties has reshaped the remedies available to injured aliens by providing them with direct access to relief against states, to the exclusion of any need for a state to espouse their claim. The focus of the investment treaty regime has shifted towards the investor, and on whether consent exists under a particular treaty allowing it to bring such a claim (and, in the International Centre for Settlement of Investment Disputes (ICSID) context, whether the nationality conditions are met). International investment treaties will often expressly refer to the requirement of nationality as a threshold condition to allow investors to invoke their provisions and bring proceedings directly against the other contracting state. The question of nationality, therefore, arises when investors intend to rely on the offer to arbitrate that a state has made in its national investment law or in bilateral or multilateral investment treaties to which the state is a party; it serves to determine both the substantive standards that apply, and the jurisdiction ratione personae of an international Tribunal.2 Investment treaties typically define the terms ‘investors’ or ‘nationals’ in broad enough terms to encompass both natural and legal persons. The legal issues arising with respect to each, however, are distinct and therefore addressed separately in this chapter.

II. NATURAL PERSONS AS INVESTORS

A. Investment Tribunals’ Competence to Determine the Opposability of Nationality A claimant must have the nationality of one of the contracting states to the investment agreement being invoked, in accordance with the test for nationality prescribed in that treaty. * Avocat à la Cour in Paris and member of the Bar of the State of New York, counsel at Freshfields Bruckhaus Deringer LLP in Paris. The views expressed in this chapter are entirely my own. 1 Panevezys-Saldutiskis Railway Case, PCIJ Ser A/B, No 76 (1939); Nottebohm Case (second phase) ICJ Rep 1955, 4; Barcelona Traction, Light and Power Company, Limited (Preliminary Objections, Judgment) ICJ Rep 1964, 6. 2 R Dolzer and C Schreuer, Principles of International Investment Law, 2nd edn (Oxford, Oxford University Press, 2012) 44–45.

244 Alexandra van der Meulen In the ICSID context tribunals must also ensure that an investor has standing to bring a claim by ascertaining that the nationality requirements under Article 25 of the Convention have been met.3 Importantly, however, investment protection treaties and ICSID do not themselves attribute nationality. Nationality is a juridical fact which falls to be determined by application of the law of the country the nationality of which is claimed,4 as a result of a renvoi to domestic law in investment treaties.5 A question that arises for arbitral tribunals is whether they can deprive a grant of nationality of legal effects on the international plane and decline their jurisdiction ratione personae? Matters of nationality in principle fall within the sovereign competence of states. It is for each state to determine under its laws who its nationals are (by attributing nationality on the basis of one or more of the following criteria: birth (jus soli), descent (jus sanguinis), naturalisation following a period of residence, marriage, etc).6 Nevertheless, whether a given nationality has legal effects on the international plane is a question of international law:7 arbitral tribunals determine, as part of their power of Kompetenz-Kompetenz, whether a nationality is opposable to another state for the purposes of bringing a claim under a bilateral investment treaty (BIT). The landmark decision in this regard is Soufraki v UAE.8 The claimant, Mr Soufraki, was born of Italian nationals.9 He also acquired Canadian nationality in 1991.10 In 2000, he entered into a concession agreement with the Dubai Department of Ports and Customs, in his personal capacity and expressly as a Canadian national.11 After a dispute arose out of an alleged cancellation of the concession, Mr Soufraki commenced ICSID arbitration and invoked the protection granted by the 1997 Italy-UAE BIT. The respondent objected to the jurisdiction of the Tribunal on the ground that Mr Soufraki had lost his Italian nationality in 1991 when he acquired Canadian nationality.12 To prove his Italian citizenship, Mr Soufraki submitted his Italian passport and identity cards, five certificates of nationality issued by the competent Italian authorities and a letter from the Italian Ministry of Foreign Affairs certifying that he was eligible for protection under the Italy-UAE BIT.

3 The ICSID Convention (Convention on the settlement of investment disputes between states and nationals of other states (adopted 18 March 1965, entered into force 14 October 1966) 575 UNTS 159 (ICSID Convention)) establishes both positive and negative eligibility requirements: an investor must be a national of a Contracting State to the ICSID Convention, but it must not be a national of the host state (Art 25(2)(a)). 4 C Schreuer, ‘Jurisdiction and Applicable Law in Investment Treaty Arbitration’ [2014] McGill Journal of Dispute Resolution 1, 6. 5 eg Italy-UAE BIT (adopted 22 January 1995, entered into force 29 April 1997) Art 1(3); Energy Charter Treaty (adopted 17 December 1994, entered into force 16 April 1998) OJ L380/24 (ECT) Art 1(7)(a)(i); and Comprehensive Economic and Trade Agreement (CETA) between Canada, of the one part, and the European Union and its Member States, of the other part (adopted 30 October 2016) OJ L11/23 (CETA) art 8.1. 6 Nationality Decrees in Tunis and Morocco Case (1923) PCIJ, Ser B, No 4, 24; Nottebohm (Liechtenstein v Guatemala), ICJ Rep 1955, 20. 7 Nationality Decrees in Tunis and Morocco Case, above (n 6); Salem case (United States, Egypt) (1932) 2 RIAA 1184; J Crawford, Brownlie’s Principles of Public International Law, 9th edn (Oxford, Oxford University Press, 2019) 518; J Dugard, ‘Diplomatic Protection’ in Max Planck Encyclopaedia of Public International Law (Oxford, Oxford University Press, 2009) 5–6. 8 Hussein Nuaman Soufraki v The United Arab Emirates, Award (7 July 2004) ICSID Case No ARB/02/7. 9 ibid para 49. 10 ibid para 48. 11 ibid paras 1–3. 12 ibid paras 26, 31.

The Concept of Protected Investor 245 At the outset, the Tribunal held that there was no doubt as a matter of Italian law that Mr Soufraki had lost his Italian nationality in 1991 as a result of his acquisition of Canadian citizenship. The salient question was thus whether Mr Soufraki had reacquired Italian nationality after 1991 by taking up residence in Italy for one year, as Italian legislation permitted him to do.13 In essence, the Tribunal had to determine whether it could look behind the certificates of nationality produced by Mr Soufraki to determine whether he was in fact an Italian national. In an oft-cited passage, the Tribunal held that although ‘great weight’ must be given to the pronouncements of domestic authorities, the Tribunal could decide for itself whether the claimant possessed the requisite nationality: It is accepted in international law that nationality is within the domestic jurisdiction of the State, which settles, by its own legislation, the rules relating to the acquisition (and loss) of its nationality … But it is no less accepted that when, in international arbitral or judicial proceedings, the nationality of a person is challenged, the international tribunal is competent to pass upon that challenge. It will accord great weight to the nationality law of the State in question and to the interpretation and application of that law by its authorities. But it will in the end decide for itself whether, on the facts and law before it, the person whose nationality is at issue was not a national of the State in question.14

The Tribunal therefore accepted the claimant’s certificates of nationality as prima facie evidence but noted that they ‘do … not preclude a decision at variance with [their] contents’.15 It sided with the UAE that the certificates were not determinative of Mr Soufraki’s nationality, as the Italian authorities may not have been aware of the loss of Italian nationality by Mr Soufraki as a result of his acquiring Canadian nationality, or of whether Mr Soufraki had subsequently regained Italian nationality. The Tribunal therefore launched its own investigation to determine whether Mr Soufraki had regained Italian nationality but ultimately concluded that Mr Soufraki had not discharged his burden of demonstrating that he was an Italian national, and declined to exercise jurisdiction on that basis.16 The Tribunal’s findings were upheld by an ad hoc committee: it was ‘clearly within the power, and the duty, of the Tribunal to decide on its jurisdiction and, as a consequence, to verify that Mr Soufraki possessed Italian nationality’.17 The committee further stressed that although international tribunals must apply domestic nationality law, they ‘are not bound by national certificates of nationality or passports or other documentation’18 and ‘are empowered to make [their] own investigation into the nationality of parties regardless of the presence of official government nationality documents’.19 The committee noted that the Tribunal had not improperly withdrawn Mr Soufraki’s nationality for domestic purposes – a power which lies squarely within a sovereign’s eminent domain – it had merely decided that that nationality was not opposable for international arbitration purposes.20

13 ibid

para 53. para 55 (emphasis added). 15 ibid para 63. 16 ibid paras 69–81. 17 Hussein Nuaman Soufraki v The United Arab Emirates, Decision on Annulment Application, 5 June 2007, paras 52, 57. 18 ibid para 64. 19 ibid para 76. 20 ibid paras 55, 60. 14 ibid

246 Alexandra van der Meulen The methodology and salient findings of the Soufraki case have since found ample support in investment cases.21 In particular, it has been widely endorsed that: (i) the nationality of natural persons is governed by the law of the state whose nationality is invoked;22 (ii) investment tribunals are empowered to decide whether a party has the requisite nationality to ascertain their own jurisdiction;23 and (iii) passports and certificates of nationality are accepted as prima facie evidence only and are not binding upon tribunals, particularly in instances where the respondent state shows there to have been fraud or error.24

B. Dual Nationals: The Limited Relevance of the Genuine and Effective Nationality Requirement in the Investment Context The classical statement regarding effective nationality under customary international law is that of the International Court of Justice in the Nottebohm case, in which the Court held that although it was the state’s sovereign prerogative to grant nationality, that grant need not be recognised by other states in circumstances where that nationality was not ‘real and effective’. The Court articulated a test requiring nationality to be based on an individual’s ‘genuine connection’ with the granting state to benefit from the latter’s protection.25 In today’s circumstances, it may be onerous – if not outright impossible in some circumstances – to demonstrate effective nationality following the requirements of Nottebohm, ie the person’s attachment to the state through tradition, interests, family ties and participation in public life.26 Although the Nottebohm principles are still relevant in cases of dual nationality in circumstances where a claim is brought against another state of nationality27 – which appears to have been the underlying rationale for the ruling in the Nottebohm case – they do not prevent a state from espousing the claim of a dual national, if that national does not hold the nationality of the respondent state.28

21 C Schreuer, ‘Criteria to determine Investor Nationality (Natural Persons)’ in M Kinnear and others (eds), Building International Investment Law: The First 50 Years of ICSID (Alphen aan den Rijn, Wolters Kluwer, 2015). 22 eg Waguih Elie George Siag and Clorinda Vecchi v The Arab Republic of Egypt, Decision on Jurisdiction (11 April 2007) ICSID Case No ARB/05/15, para 143; Nations Energy, Inc and others v Republic of Panama, Award (24 November 2010) ICSID Case No ARB/06/19, para 378. 23 Siag v Egypt, above (n 22) para 150 (quoting Soufraki v UAE, Award, para 55); Victor Pey Casado and President Allende Foundation v Republic of Chile, Award (8 May 2008) ICSID Case No ARB/98/2, para 319; Tza Yap Shum v Republic of Peru, Decision on Jurisdiction and Competence (19 June 2009) ICSID Case No ARB/07/6, para 63; Société Civile Immobilière de Gaëta v Republic of Guinea, Award (21 December 2015) ICSID Case No ARB/12/36, para 135. 24 eg Ambiente Ufficio SpA et al v Argentine Republic, Decision on Jurisdiction and Admissibility (8 February 2013) ICSID Case No ARB/08/9, para 318; CEAC Holdings Limited v Montenegro, Award (26 July 2016) ICSID Case No ARB/14/8, para 155; Vladislav Kim and others v Republic of Uzbekistan, Decision on Jurisdiction (8 March 2017) ICSID Case No ARB/13/6, para 209; Caratube International Oil Company LLP v Kazakhstan, Award (5 June 2012) ICSID Case No ARB/08/12, para 377 et seq. 25 Nottebohm, above (n 1) 22–23. 26 United States ex rel Albert Flegenheimer v Italian Republic, Case No 20, Decision No 182, XIV RIAA 377 (1958); ILC, Draft Articles on Diplomatic Protection with Commentaries (2006), Commentary to Article 4, UN Doc A/61/10, 29–30, para 5. 27 Mergé Case (US v Italy), XIV RIAA 247 (1955); Case No 32-A/18-FT, 5 Iran-US Claims Tribunal, Rep 251 (1984-1), 259–260. This rule has been codified in ILC, Draft Articles on Diplomatic Protection, Yearbook of the International Law Commission, 2006, vol II, Part Two (ILC 2006) Art 7. 28 Salem Case, above (n 7) 1188; ILC, Draft Articles on Diplomatic Protection with Commentaries, above (n 26) Art 6.

The Concept of Protected Investor 247 C. ICSID Arbitration In the ICSID context, the Washington Convention does not bar claims by dual nationals so long as they do not hold the nationality of the respondent state.29 Tribunals have seen this as ‘the only jurisdictional bar relating to a natural person’s nationality’ in the Washington Convention.30 In light of this specific and express exclusion, when one of the investor’s nationalities is that of the host state, ICSID tribunals have held the investor not to have standing, irrespective of which of his nationalities is more effective. This was the approach taken by the arbitral tribunal in Champion Trading v Egypt,31 one of the first cases to review the requirements under Article 25(2)(a) of the ICSID Convention. The claimants were three individuals (dual US-Egyptian nationals) and two US companies. When the claimants commenced ICSID proceedings under the Egypt-US BIT, Egypt objected that the three brothers were, as dual US-Egyptian nationals, precluded from bringing claims against it. The claimants for their part invoked the Nottebohm case and argued that their US nationality should be regarded as their effective nationality, such that they were entitled to bring proceedings against Egypt.32 The Tribunal sought to determine whether the claimants could be said to be Egyptian nationals, and it concluded by application of jus sanguinis that they had automatically acquired Egyptian nationality at birth.33 The next question was whether the three brothers could still bring a claim against Egypt notwithstanding their dual nationality. The Tribunal unequivocally excluded the application of the rule of effective nationality in the context of the ICSID Convention stating, first, that the Nottebohm case had not been decided in the context of dual nationality (but of whether Liechtenstein could exercise diplomatic protection on behalf of Nottebohm),34 and, second, that Article 25(2)(a) ‘contain[ed] a clear and specific rule’ excluding dual nationals from ICSID protection where a claim is against a state of nationality.35 Of note however is the fact that the Tribunal suggested there might be instances where the test of effective nationality might be applied (such that dual nationals might not be barred from ICSID arbitration); for instance, where a state applies the jus sanguinis over many generations, such that a third or fourth foreign born generation ‘has no ties whatsoever with the country of its forefathers’.36 It is unclear what the outcome might have been had the claimants made their investment using their US nationality only. Indeed, the fact that the investment was made using ‘exclusively’ the Egyptian nationality appears to have been of particular relevance for the Tribunal.37

29 On the other hand, there is nothing to exclude – and it seems no requirement of effective nationality to restrict – claims of dual nationals if their second nationality is not that of the host state. 30 Saba Fakes v Republic of Turkey, Award (14 July 2010) ICSID Case No ARB/07/20, para 61. 31 Champion Trading Company, Ameritrade International, Inc v Arab Republic of Egypt, Decision on Jurisdiction (21 October 2003) ICSID Case No ARB/02/9. 32 ibid 22–25. 33 ibid 36, 63–64. 34 ibid 50, 57. 35 ibid 57; See also Siag v Egypt, above (n 22) para 198; and Ioan Micula v Romania, Decision on Jurisdiction and Admissibility (24 September 2008) ICSID Case No ARB/05/20, para 100. 36 Champion Trading, above (n 31) 59. 37 ibid 63.

248 Alexandra van der Meulen D. Non-ICSID Arbitration In non-ICSID arbitration, the situation of dual nationals – and the need for an effective nationality test – will depend on the wording of the applicable treaty. Some investment treaties expressly exclude dual nationals from the scope of protection.38 Others adopt a ‘dominant and effective nationality’ criterion in situations of dual nationality.39 Where investment treaties leave the question of dual nationals unaddressed, investment tribunals have not automatically assumed that claims by dual nationals are excluded. This was the approach taken in Serafin García Armas and Gruber v Venezuela,40 in which a tribunal accepted jurisdiction over a claim by dual nationals against a state of nationality. Two dual Venezuelan-Spanish nationals41 brought a claim pursuant to the Spain-Venezuela BIT against Venezuela under the United Nations Commission on International Trade Law (UNCITRAL) Rules, claiming compensation for the alleged expropriation of their food retailer and distribution companies.42 Venezuela objected to the jurisdiction of the Tribunal on the ground that: (i) the treaty did not expressly allow nationals to sue their own state; and (ii) the fact that the BIT referred to ICSID arbitration as a primary method of dispute resolution confirmed the contracting states’ intent to exclude the possibility of being sued by a national.43 In any event, it contended that as a matter of international law, the claimants were precluded from bringing claims against Venezuela since their dominant nationality was Venezuelan.44 The Tribunal rejected Venezuela’s objections. Three points are worth highlighting. First, the Tribunal held that the BIT constituted a lex specialis between the parties; it therefore refused to apply the customary international law rule of effective nationality.45 Whether dual nationals had standing was to be determined on the basis of the BIT at issue, not customary international law.46 The Tribunal found that the treaty did not contain any express limitation on dual nationals, and placed particular emphasis on the fact that a restriction of this kind could have been included, as indeed had been done in other treaties concluded by Venezuela and Spain.47 Second, the Tribunal did not consider it determinative that the BIT provided for ICSID arbitration as a primary method of dispute resolution. Since Venezuela had agreed to arbitrate under the UNCITRAL Rules,48 the exclusion of dual nationals under the ICSID Convention should not be read into a different regime of UNCITRAL arbitration.49 This holding may of course encourage investors to bring a claim that would otherwise be inadmissible under the ICSID convention in another forum; it may also be of particular concern to states that have recently denounced the ICSID Convention, thereby allowing claims of this kind to be brought against them under other rules (assuming the relevant investment treaty provides for this).

38 eg

Canada-Venezuela BIT (adopted 1 July 1996, entered into force 28 January 1998) Art 1(g). US Model BIT (2004), Art 1; CETA (entered into force partially on 21 September 2017) Art 8.1. 40 Serafín García Armas and Karina García Gruber v Bolivarian Republic of Venezuela, Decision on Jurisdiction (15 December 2014) PCA Case No 2013-3. 41 ibid paras 1, 55. 42 ibid para 3. 43 ibid para 87(a). 44 ibid paras 107–117. 45 ibid paras 167–175. 46 ibid paras 176–206. 47 ibid paras 176–181. 48 ibid paras 191–196. 49 ibid paras 192–193. 39 eg

The Concept of Protected Investor 249 Finally, the Tribunal held that the BIT’s definition of ‘investor’ did not exclude dual nationals; it simply required an investor to be a national of at least one of the states parties.50 Even if the claimants’ Spanish nationality was merely formalistic, it was still sufficient. If no additional requirements on nationality had been provided in the BIT, none could be implied.51 Venezuela sought to have the jurisdictional award against it set aside by the Paris Cour d’appel. It contended that it was a principle of international public order that a state could not be sued by its own nationals, and that in the case of dual nationals, the test of effective nationality applies. On 25 April 2017, the court partially annulled the award because it disagreed with the Tribunal’s conclusion regarding the time at which the nationality needed to be held.52 Crucially for present purposes however, the Court upheld the Tribunal’s finding that the Spain-Venezuela BIT protects dual nationals.53 It noted that the BIT’s definition of ‘investors’ did not exclude dual nationals and that the BIT’s object and purpose – to attract and protect investment, without distinction as to the sources of the investments – would only partially be satisfied if dual-national claims were excluded.54 Further, since Venezuela agreed to UNCITRAL arbitration, the reference to ICSID arbitration could not justify the exclusion of benefits for dual nationals.55 Thus, the Court agreed that the relevant question was whether the claimants were Spanish nationals as a matter of domestic law, regardless of whether they were also Venezuelan.56 The reasoning of García Armas is likely to be invoked in claims brought by dual nationals against states of nationality. It has already been followed in Rawat v Mauritius, in which it was held that the silence of the BIT as to dual nationals ‘would seem to point to the inclusion, rather than the exclusion, of dual nationals within the scope of the France-Mauritius BIT’.57 What is less certain is whether tribunals will follow the formalistic approach to nationality adopted in Armas, or whether they will resort to the effective nationality rule to ensure that their jurisdiction extends only to investors that hold the dominant nationality of the home state (to ensure that BITs only protect investments made by foreign – not domestic – investors). In Rawat, the Tribunal held that the claimant’s dominant nationality was that of Mauritius; it did not however draw any consequences from this given that Article 25 of the ICSID Convention58 – which the BIT expressly referred to – in any event excluded his claim.59

50 ibid

para 199. paras 203, 206. 52 République Bolivarienne du Venezuela c Monsieur Serafín García Armas et Madame Karina García Gruber, CA Paris, Civ 1 (25 April 2017) n 15/01040, 7–8. On 13 February 2019, the Cour de cassation overturned the decision of the Cour d’appel partially to annul the tribunal, holding that in order to uphold jurisdiction, all of the requirements under the BIT – including the temporal requirement – needed to be met. Thus, in finding that the tribunal had failed to consider whether the temporal condition was satisfied, the Cour d’appel should have annulled the award in full. Venezuela’s request for annulment was recently heard by a new bench of the Cour d’appel, which followed the Cour de cassation’s direction and set aside the award in its entirety (République Bolivarienne du Venezuela c Monsieur Serafín García Armas et Madame Karina García Gruber, CA Paris, 3 June 2020, n° 19/03588). 53 ibid 4–7. 54 ibid 5–6. 55 ibid 6. 56 ibid 7. 57 Dawood Rawat v The Republic of Mauritius, Award on Jurisdiction (6 April 2018) PCA Case 2016-20, paras 170, 172. 58 ICSID Convention, above (n 3). 59 Rawat, above (n 57) paras 175–179. 51 ibid

250 Alexandra van der Meulen III. LEGAL PERSONS AS INVESTORS

A. Ascertaining the Nationality of Companies Determining whether a legal person satisfies the nationality requirements of an investment protection treaty – and the ICSID Convention where applicable – is sometimes a complex endeavour, given the intricate and multi-layered corporate structures by which investments may be held. Different criteria exist to ascertain the nationality of legal entities for the purposes of establishing the jurisdiction of an arbitral tribunal, and different treaties contain different permutations of each of these criteria.60 The place of incorporation61 and the effective seat of business62 are the most commonly used criteria and the two are sometimes used in combination.63 Certain treaties go beyond formal criteria, and require a bond of economic substance between the corporate investor and the state whose nationality it claims (by requiring both incorporation and management to be located in the home state).64 Treaties may further specify that the requisite economic substance can be evidenced by a controlling interest in a company65 or a predominant or substantial interest.66 The remainder of this chapter examines how arbitral tribunals apply these criteria in practice.

B. Piercing the Corporate Veil As already noted, investment treaties typically adopt the place of incorporation as a criterion to determine the nationality of corporations – and insofar as ICSID arbitration is concerned, there is nothing in the Convention a priori militating against jurisdiction where the claimant is a mere shell company. A question that arises is what approach should be taken in circumstances where the place-of-incorporation test is met but where the economic interests behind the claimant are clearly not located in the same country? Should the Tribunal uphold the parties’ agreement as recorded in the place of incorporation criterion in the applicable BIT, or should the formalistic test be departed from, by piercing the corporate veil – that is, disregard the company’s formal corporate structure and instead consider the economic reality of ownership? The question has come to the fore in particular in cases in which shell corporations were controlled by nationals of the host state. In the landmark case of Tokios Tokelės v Ukraine, in which the applicable BIT provided for a state-of-incorporation criterion, the Tribunal refused to modify the express words of the BIT; it gave effect to the incorporation test, irrespective of the links – or lack thereof – with the state of incorporation.67 60 In the Barcelona Traction case, for example, the dictum of the Court made clear that the right to diplomatic protection was recognised only in respect of the state of incorporation (Case Concerning Barcelona Traction, Light & Power Co, Ltd (Second Phase) [Belgium v Spain] (1970) ICJ Rep 3, para 70). 61 eg ECT, Art 1(7); CAFTA (signed on 5 August 2004, entered into force 1 March 2006) Art 10.28. 62 eg German Model BIT (1991) Art 1(4)(a). 63 eg Belgium-Luxembourg Economic Union-Czech Republic BIT (adopted 24 April 1989, entered into force 13 February 1992) Art 1(1); Argentina-France BIT (adopted 3 July 1991, entered into force 3 March 1993) Art 1(2)(b). 64 eg CETA (entered into force partially on 21 September 2017) Art 8.1. 65 eg Final Act of the European Energy Charter Conference, Understanding 3 (1994), https://perma.cc/3HSM-9URE; Moldova-US BIT (adopted 21 April 1993, entered into force 26 November 1994) Art I(1)(a); Netherlands-Venezuela BIT (adopted 22 October 1991, entered into force 1 November 1993, terminated 1 November 2008) Art 1(b)(iii). 66 eg Lithuania-Sweden BIT (adopted 17 March 1992, entered into force 1 September 1992) Art 1(3)(b); Egypt-US BIT (adopted 11 March 1986, entered into force 27 June 1992) Art I(1)(b)(ii). 67 Tokios Tokelės v Ukraine, Decision on Jurisdiction (29 April 2004) ICSID Case No ARB/02/18.

The Concept of Protected Investor 251 The claimant, Tokios Tokelės, was a company incorporated in Lithuania. In 1994, it set up Taki spravy, a wholly-owned subsidiary in Ukraine. Following certain measures taken by Ukraine against Taki spravy, Tokios Tokelės initiated ICSID arbitration under the 1994 Ukraine-Lithuania BIT. Ukraine argued that the Tribunal should disregard the claimant’s state of incorporation (Lithuania) and find that it did not carry out substantial activities in Lithuania and was 99 per cent-owned by Ukrainian nationals. In essence, it asked the Tribunal to pierce the corporate veil and find that the claimant’s shareholders and managers were in fact Ukrainian nationals acting against their own government, and decline jurisdiction on this basis. The Tribunal refused to do so. It noted that the ICSID Convention is silent on the criteria to determine the nationality of juridical entities and therefore relied on Article 1(2)(b) of the Ukraine-Lithuania BIT, which defined ‘investor’ as ‘any entity established in the territory of the Republic of Lithuania in conformity with its laws and regulations’.68 The ordinary meaning of that provision meant that the states parties had given their consent to jurisdiction over entities incorporated in Lithuania, irrespective of who actually controlled them; that intention, the Tribunal said, had to be upheld.69 The Tribunal went on to verify the consistency of Article 1(2) of the BIT with the ICSID Convention and concluded that the control test in Article 25 could not restrict the scope of consent recorded by states in their BITs without defeating the object and purpose of the Convention, which was to expand ICSID jurisdiction.70 The Tribunal also held irrelevant that capital was being sourced from the host state itself rather than from outside of its territory:71 implying an ‘origin-of-capital’ requirement into the BIT would also be inconsistent with its object and purpose to provide broad protection for investors.72 Therefore the Tribunal upheld jurisdiction rationae personae over the claimant, as a Lithuanian investor. In a dissenting opinion, Professor Prosper Weil, the chairman of the Tribunal, disagreed with the majority’s decision to disregard: (i) the Ukrainian origin of the capital invested by the claimant in its wholly-owned subsidiary; and (ii) the Ukrainian nationality of the claimant’s shareholders.73 In his view, Article 25 of the Convention imposed objective jurisdictional requirements that override the consent of contracting states.74 He further observed that when ruling on its own jurisdiction, an arbitral tribunal had to give effect to the object and purpose of the Convention, which was to resolve disputes between a State and foreign investors. Instead, the result of the majority’s decision was to accept jurisdiction over a dispute ‘between the Ukrainian State and a Ukrainian investor’, allowing Ukrainian nationals to evade the jurisdiction of their domestic law and the application of their national law.75 Given how central ratione personae issues relating to investors’ nationality are to ICSID arbitration, the majority’s decision and Professor Weil’s dissenting opinion have been extensively discussed and referred to in subsequent cases. Most tribunals have followed the

68 ibid

para 28. paras 28, 40. paras 46, 50. 71 ibid para 74. 72 ibid para 77. 73 ibid Dissenting opinion of Prosper Weil (29 April 2004) para 11. 74 ibid para 13. 75 ibid paras 21, 30; see also paras 5, 16, 19, 24. 69 ibid 70 ibid

252 Alexandra van der Meulen majority ruling, upholding a bare place-of-incorporation test – and declining to look behind the corporate veil – by reference to one or more of the following considerations: a.

b.

c.

d.

The ICSID Convention does not limit states to any particular test for determining investors’ nationality; it sets out only an ‘outer limit’ within which states are free to set the parameters of their consent to resort to ICSID arbitration.76 The precise scope of that consent is in turn to be found in the relevant BITs and national legislation on foreign investment.77 Article 25 should not be read in a way that would override parties’ consent by importing additional requirements into the BIT. When states intend to exclude entities from the ambit of their treaties, they can and do specify this (eg by providing for control-based methods for determining nationality). The fact that they do not is evidence of their intention to offer broad protection.78 In the absence of more substantial requirements than incorporation,79 it is to be expected – and permissible – that investors will structure their investments in order to avail themselves of treaty protection; the aim of such instruments is precisely to attract investments. Investors may, as part of their investment structuring, rely on the place of incorporation to foresee whether they will fall within the ambit of a particular treaty. While the place of a company’s incorporation is stable, its shareholders may change over time. Piercing the corporate veil may defeat investors’ expectation that they would be able to rely on an investment treaty and may therefore defeat the object and purpose of such a treaty.

Other tribunals on the other hand have expressly adopted Professor Weil’s approach, in favour of a less formalistic approach, based on the following considerations: a.

b.

Article 25(2)(b) of the ICSID Convention prevents nationals of the host state from acceding to ICSID arbitration. Pursuant to Articles 31 and 32 of the Vienna Convention on the Law of Treaties, and insofar as they expressly provide for ICSID arbitration, BITs should be read in light of the Convention and its object and purpose. The states parties’ intention was that only ‘international’ disputes stricto sensu would be submitted to arbitration and avoid what are in essence disputes between a state and its nationals.80 Applying a strict test of incorporation is tantamount to opening the door to treaty shopping insofar as it allows investors to take advantage of an investment treaty in a country in which they have no real economic activity.

The latter approach was recently taken by a Tribunal in Venoklim v Venezuela, which involved a similar corporate structure to that considered in Tokios Tokeles.81 The dispute arose out of the alleged expropriation of Venoklim’s property through a decree adopted by Venezuela. Venoklim, which was incorporated in the Netherlands but owned and 76 A Broches, ‘The Convention on the Settlement of Investment Disputes between States and Nationals of Other States’ (1972) 136 Collected Courses of the Hague Academy of International Law 331, 359–61, 361. 77 ADC Affiliate Limited and ADC & ADMC Management Limited v The Republic of Hungary, Award (2 October 2006) ICSID Case No ARB/03/16, para 357; KT Asia Investment Group BV v Republic of Kazakhstan, Award (17 October 2013) ICSID Case No ARB/09/8, paras 113–116; The Rompetrol Group v Romania, Decision on Respondent’s Preliminary Objections on Jurisdiction (18 April 2008) ICSID Case No ARB/06/3, para 109; Saluka Investments BV v The Czech Republic, UNCITRAL, Partial Award (17 March 2006) para 241. 78 KT Asia v Kazakhstan, above (n 77) paras 122–123. 79 Tokios Tokelės v Ukraine, above (n 67) para 33. 80 TSA Spectrum de Argentina SA v Argentine Republic, Award (19 December 2008) ICSID Case No ARB/05/5, para 160, in which the tribunal pierced several corporate layers to reach the real source of control of the locally incorporated vehicle so as to determine ‘whether or not the domestic company was objectively under foreign control’. 81 Venoklim v Venezuela, Award of the Tribunal (3 April 2015) ICSID Case No ARB/12/22.

The Concept of Protected Investor 253 controlled by Venezuelan nationals, initiated arbitration proceedings against Venezuela. Venezuela objected to the jurisdiction of the Tribunal on the basis that Venoklim was not a foreign investor, but instead a shell company controlled by Venezuelan nationals.82 Relying on Tokios Tokelés and on the terms of the applicable BIT, Venoklim for its part argued that the place of incorporation was dispositive of the nationality of juridical persons.83 The Tribunal upheld Venezuela’s objection. Referring to Prosper Weil’s dissent, the majority expressed the view that upholding its jurisdiction on the ground that Venoklim was incorporated in the Netherlands would allow ‘formalism to prevail over reality’ and would be contrary to the object and purpose of the Convention, which was to facilitate the settlement of disputes between states and foreign nationals.84 This approach, although in keeping with the concern of states to manage their exposure to investment claims under the ICSID Convention emanating from their own nationals, leads to some uncertainty for investors as to whether or not an investment structure will fall within the protection of a particular investment treaty, including in circumstances where the formal (and unambiguous) criteria for determining nationality have been met.

IV. MINORITY SHAREHOLDERS AS INVESTORS

Under customary international law, the traditional position was that the corporation, not its shareholders, would be protected for company-related claims. The principle was set forth in Barcelona Traction in which the International Court of Justice held that Belgium, the state of nationality of the majority of shareholders of Barcelona Traction, a company incorporated in Canada, could not pursue claims on their behalf against Spain for injury caused to the company.85 Importantly however, the Court did not rule out that treaties in contexts other than diplomatic protection may provide otherwise. Investment treaties provide for such derogations, by including shares as protected qualifying investments and giving standing to shareholders to claim qua shareholders for loss of shareholder value (rather than for loss on behalf of the company); even if the company itself is not endowed with investor status (eg because it is incorporated in the host state).86 In the CMS v Argentina case,87 the Tribunal had to decide whether this protection extended to minority shareholders. The claimant, CMS, owned 29.42 per cent of shares in TGN, a company incorporated in Argentina which had been granted a licence for the transport of gas. Following an economic and financial crisis, Argentina suspended a tariff adjustment formula for gas transportation that according to CMS ultimately had an adverse effect on TGN. CMS initiated arbitration against Argentina under the Argentina-US BIT. Argentina objected to the Tribunal’s jurisdiction on the basis that insofar as TGN was the licensee and CMS was only a minority shareholder, only TGN could bring a claim.88 82 ibid

paras 131–134. para 136. 84 ibid para 154. 85 Case Concerning Barcelona Traction, Light & Power Co, Ltd (Second Phase) [Belgium v Spain] (1970) ICJ Rep 3. 86 eg Siemens AG v The Argentine Republic, Decision on Jurisdiction (3 August 2004) ICSID Case No ARB/02/8; Azurix Corp v The Argentine Republic, Decision on Jurisdiction (8 December 2003) ICSID Case No ARB/01/12. See also Ahmadou Sadio Diallo (Guinea v Democratic Republic of Congo), Preliminary Objections, Judgment, ICJ Rep (2007) para 90, in which the Court clarified that this development was ‘not sufficient to show that there has been a change in the customary rules of diplomatic protection’. 87 CMS v Argentina, Decision of the Tribunal on Objections to Jurisdiction (17 July 2003) ICSID Case No ARB/01/8. 88 ibid paras 36–37. 83 ibid

254 Alexandra van der Meulen Argentina’s contention was that although an investment in shares was a protected investment under the BIT, this would only allow claims for measures affecting the shares qua shares, not claims connected to damage suffered by the company. CMS for its part argued that it had its own right of action, arising directly from the BIT, independently from TGN’s rights under the licence agreement.89 CMS also emphasised that it was not claiming for rights pertaining to TGN but for rights associated with its own investment in TGN.90 The Tribunal analysed the rights of shareholders under general international law, and held that there was ‘no bar in current international law to the concept of allowing claims by shareholders independently from those of the corporation concerned, not even if those shareholders are minority … shareholders’.91 The Tribunal went on to consider the rights of shareholders under the ICSID Convention. It observed that investments were not required to be made by majority or controlling shareholders92 and referred to the growing trend in investment treaty arbitration to allow direct claims by shareholders independently from those of the corporations.93 Finally, the Tribunal noted that the applicable BIT did not contain a requirement as to the number of shares to be owned by the investor.94 It also emphasised that the question of whether a protected investor had rights under a concession or licence agreement is not relevant to determine its jurisdiction since shareholders have a direct right of action.95 It therefore concluded that CMS could bring a claim against Argentina under the BIT and the ICSID Convention, independently from the company.96 Argentina applied for annulment of the award. In its decision the committee noted that treaties that provide for a direct right of action by minority shareholders are permissible under general international law.97 The BIT concluded between the parties and the ICSID Convention therefore had to be applied as lex specialis.98 The committee recalled in particular that the ICSID Convention does not define the term ‘investment’, leaving this task to the parties’ discretion in the instrument on which their consent is based. Relying on the definition in the BIT, the committee held that there was nothing to indicate that an investment could only be made by majority or controlling shareholders.99 It went on to hold that: whether the locally incorporated company may itself claim for the violation of its rights under contracts, licenses or other instruments … does not affect the right of action of foreign shareholders under the BIT in order to protect their own interests in a qualifying investment, as recognized again in many ICSID awards.100

The implication of this holding – and of the fact that indirect shareholders are also entitled to protection101 – is of course that different shareholders (including of interposed companies) 89 ibid

para 40.

90 ibid. 91 ibid

para 48. para 51. 93 ibid paras 52–56. 94 ibid para 63. 95 ibid para 64. 96 ibid paras 48, 65. 97 CMS v Argentina, Decision of the ad hoc Committee on the Application for Annulment of the Argentine Republic (25 September 2007) ICSID Case No ARB/01/8. 98 ibid para 69. 99 ibid. 100 ibid para 74. 101 Azurix Corp v Argentine Republic, Decision on Jurisdiction (8 December 2003) ICSID Case No ARB/01/12 paras 65–66; Siemens AG v Argentine Republic, Decision on Jurisdiction (3 August 2004) ICSID Case No ARB/02/8, paras 141–143. 92 ibid

The Concept of Protected Investor 255 may pursue claims in parallel with respect to the same underlying investment and adverse host state action.102 The affected company itself may pursue one set of remedies while a group of its shareholders may pursue different ones.103 The question that arises is how to balance the injustice that would arise from preventing parallel claims to be brought, on the one hand, against exposing a state to the threat of new proceedings by different parties even after an award against it has been rendered, on the other hand (which in turn raises the threat of contradictory awards – a threat that has already materialised104 – as well as that of abuse by shareholders and companies seeking to exert pressure on host states). Tribunals have sought – not always successfully – to rely on the doctrines of lis pendens and res judicata to mitigate these risks, but it has been suggested that preventing parallel proceedings may be more effectively – although not completely – addressed in investment protection treaties (by adopting more stringent conditions to qualify as a protected investor and by way of denial of benefits clauses).

V. RESTRUCTURING THE CORPORATE STRUCTURE

It is not uncommon for investors to engage in nationality planning and corporate structure, by incorporating intermediate holding companies in jurisdictions perceived to provide a beneficial fiscal and legal environment, and that provide for incorporation as a requirement to fall within the ambit of a favourable investment treaty. Arbitral tribunals have sought to provide guidelines as to what is permissible treaty planning and what is an abus de droit. Some tribunals take the view that strategic corporate restructuring at the pre-investment stage to obtain benefits under a BIT is a right that is consistent with investment treaties, the purpose of which is to afford legal protections. Such tribunals therefore refuse to look behind the corporate veil to determine the true nationality of the company, and thus allow mere shell or mailbox companies to claim the benefit of an investment treaty which provides for protection on the basis of incorporation.105 This approach appears to accept the notion that the primary objective of investment instruments is to promote inflows of capital, irrespective of its origin or of the degree of relations between an investor and the state of incorporation. That is not to say however that corporate restructuring is without limits. Indeed, some tribunals have held that although restructuring at the pre-investment state is not per se illegal, the corporate veil would be pierced – and jurisdiction declined – ‘to prevent the misuse of the privileges of legal personality’106 or if there was a ‘showing of abuse of the corporate form’.107

102 Ampal-American Israel Corp & others v Arab Republic of Egypt, Decision on Jurisdiction (1 February 2016) ICSID Case No ARB/12/11, para 329; See generally, D Müller, La Protection de l’Actionnaire en Droit International, L’Héritage de Barcelona Traction (2015) 59 Revue Générale de Droit International Public, paras 859 et seq. 103 CME Czech Republic BV v Czech Republic, UNCITRAL, Partial Award (13 September 2001). 104 Thus, Ronald Lauder’s investment in the Czech Republic led to two conflicting awards in arbitrations brought under two different BITs against the Czech Republic: CME Czech Republic BV v Czech Republic, Final Award (UNCITRAL, 2003); Lauder v Czech Republic, Final Award (UNCITRAL, 2001). 105 Tokios Tokelės v Ukraine, above (n 67) para 40. 106 ibid paras 53–54 (quoting Barcelona Traction, above (n 60) para 56; emphasis omitted). 107 KT Asia v Kazakhstan, above (n 77) paras 134–138; see also ADC v Hungary, above (n 77) para 358; Saluka v Czech Republic, above (n 77) paras 230–238; Autopista Concesionada de Venezuela, CA v Bolivarian Republic of Venezuela, Decision on Jurisdiction (27 September 2001) ICSID Case No ARB/00/5, para 67.

256 Alexandra van der Meulen Quid in situations where the corporate restructuring takes place with respect to an existing investment that has become exposed to political risk, in order to improve existing legal protections and/or to gain access to investment arbitration against the host state? In these circumstances, investment tribunals have focused on the timing of such restructuring: so long as it occurs prior to the crystallisation of a dispute, it is legitimate corporate planning as opposed to an abuse of right.108 Mobil v Venezuela concerned the legitimacy of corporate restructuring for the purpose of gaining access to ICSID arbitration. Mobil had invested in the exploration and production of petroleum in Venezuela. After Venezuela took a series of measures to regulate the industry, but before Mobil’s investments were nationalised, Mobil interposed a Dutch entity into the existing chain of ownership in order to fall within the ambit of the Netherlands-Venezuela BIT, which allowed Bahamian and US subsidiaries controlled by the Dutch entity to bring ICSID proceedings against Venezuela. Venezuela objected to the jurisdiction of the Tribunal, on the ground that the new Dutch entity was a ‘corporation of convenience’, which had been incorporated in the midst of what Mobil perceived to be a deteriorating environment for investors and as part of its strategy to shop for a more favourable treaty in anticipation of a dispute with Venezuela.109 The Tribunal referred to the general principle of good faith and observed that ‘the main, if not the sole purpose of the restructuring was to protect Mobil investments from adverse Venezuelan measures in getting access to ICSID arbitration through the Netherlands-Venezuela BIT’.110 That did not however, in and of itself, amount to abuse of right.111 In analysing whether in the circumstances Mobil’s restructuring could be said to be an abuse of right, the Tribunal had regard to the following circumstances: a.

b.

First, it was significant that investment in Venezuela did not altogether cease after the restructuring.112 It could not therefore be said that the investment had not been bona fide.113 Second, the Tribunal drew a distinction between those claims that arose prior to Mobil’s corporate restructuring and those that arose after it and concluded that the restructuring process was a ‘perfectly legitimate goal as far as it concerned future disputes’. With respect to pre-existing disputes, however, it noted that: ‘to restructure investments only in order to gain jurisdiction under a BIT for such disputes would constitute … “an abusive manipulation of the system of international investment protection under the ICSID Convention and the BITs”’.114

The Tribunal therefore declined jurisdiction over those claims that were in existence at the time of the restructuring.115

108 Venezuela Holdings, BV, et al (case formerly known as Mobil Corporation, Venezuela Holdings, BV, et al) v Bolivarian Republic of Venezuela, Decision on jurisdiction (10 June 2010) ICSID Case No ARB/07/27, paras 203–206; Phoenix Action, Ltd v The Czech Republic, Award (15 April 2009) ICSID Case No ARB/06/5, paras 113–144. 109 Mobil v Venezuela, above (n 108) para 27. 110 ibid para 190. 111 ibid para 191. 112 ibid paras 193–198. 113 In contrast, in Phoenix v Czech Republic, above (n 108) para 142, no economic activity had taken place after the investment, leading the tribunal to conclude that the investment had not been made for the purpose of engaging in economic activity, but only for bringing international claims against the host state. 114 Mobil v Venezuela, above (n 108) para 205. 115 ibid para 206.

The Concept of Protected Investor 257 Mobil v Venezuela posited a bright line test: was the restructuring carried out before or after a dispute arose?116 That is an objective question – concerning the timing of the investments as compared to the timing of the dispute – which has the advantage of providing certainty and predictability to investors and host states alike. That is not the end of the matter however as some tribunals have since held that even restructuring occurring before a dispute arises may constitute an abuse of right, casting doubt on the ability to invoke an investment treaty in circumstances where a specific future dispute might reasonably have been foreseen by an investor.117 Determining whether a dispute is foreseeable as a result of measures being taken by a host State – such that protection of the BIT will be refused – is a fact-specific inquiry calling for a partly objective, partly subjective assessment of all the relevant circumstances. As the Pacific Rim Tribunal acknowledged, the dividing line ‘will rarely be a thin red line, but will include a significant grey area’.118 In the more recent decision of Philip Morris v Australia, the Tribunal held that the mere announcement by the Australian Government that it intended to introduce plain packaging measures for tobacco products was in and of itself enough to conclude that the dispute that eventually materialised was foreseeable to the claimant at the time of its restructuring, such that the claim was deemed inadmissible.119 This is likely to give rise to a lack of predictability as to when a restructuring would be held to be permissible, at odds with the object and purpose of investment treaties to facilitate investment flows. It is also likely to influence the manner in which investment restructuring takes place, with investors bearing the burden of demonstrating why a restructuring was carried out (by reference to tax and other benefits) and why it was done at a particular point in time, as well as documenting any further investments carried out after the restructuring, in order to demonstrate that it was not for the sole purpose of commencing litigation against the host state.

116 ibid

paras 193–198. Rim Cayman LLC v Republic of El Salvador, Decision on the Respondent’s Jurisdictional Objections (1 June 2012) ICSID Case No ARB/09/12, paras 2.16–2.17, 2.99 (referring to where a ‘party can see an actual dispute or can foresee a specific future dispute as a very high probability’); Philip Morris v Australia, Award on Jurisdiction and Admissibility (17 December 2015) PCA Case No 2012-12, para 554 (referring to a lower threshold, requiring only that a dispute have ‘a reasonable prospect’ of its materialising). 118 Pac Rim v El Salvador, above (n 117) para 2.99. 119 Philip Morris v Australia, above (n 117) para 588. 117 Pac

16 Landmark Investment Cases on State Consent CHRISTOPH SCHREUER*

I. INTRODUCTION

C

ONSENT TO ARBITRATION is an indispensable condition for the jurisdiction of an arbitral tribunal. Consent must be obtained from both or all parties. The parties enjoy a large measure of freedom in how to express their consent. Traditionally, this would take place by way of a direct agreement between the host state and the investor. Consent may also result from a general unilateral offer by the host state, expressed in its legislation or in a treaty, which is subsequently accepted by the investor. Nowadays the vast majority of cases are based on consent given in this indirect way.1 This phenomenon has been called arbitration without privity.2 Here too, the result is an agreement, although it is achieved indirectly and often without direct contact between the parties prior to the institution of proceedings. Consent will be valid according to its own terms, that is, to the extent that disputes are covered by its scope. The shift from contract-based to treaty-based arbitration has had several consequences. The number of potential claimants has increased dramatically. Investors benefitting from arbitration clauses are not individually determined as in contracts, but generically described in the treaties. At the same time, the possibility for states to act as claimants has been reduced. Since investors must accept the offer of arbitration in the treaties, the treaty-based system of investment arbitration in effect provides for investor claims against states but not vice versa. The investor is able to calibrate its acceptance of consent towards a particular dispute, thereby reducing the likelihood of a counterclaim. The shift from contract-based to treaty-based arbitration has also had consequences for the law applied by investment tribunals to the merits. Domestic law and international law used to have roughly the same weight. With the rise of consent based on treaties, tribunals have given increasing weight to treaty law and to international law in general, at the cost of domestic law. The law governing consent to investment arbitration is reflected in many decisions of investment tribunals and cannot usefully be portrayed through one or two cases. This overview seeks to illustrate its most important developments with the help of ten selected cases.

* Christoph Schreuer is a graduate of the Universities of Vienna, Cambridge and Yale, and is currently Of Counsel at Zeiler Floyd Zadkovich in Vienna 1 AM Steingruber, Consent in International Arbitration (Oxford, Oxford University Press, 2012) 60–68. 2 J Paulsson, ‘Arbitration Without Privity’ (1995) 10 ICSID Review/FILJ 232; MD Nolan and FG Sourgens, ‘Limits of Consent – Arbitration without Privity and Beyond’ in MÁ Fernández-Ballesteros and D Arias (eds), Liber Amicorum Bernardo Cremades (Lima, La Ley, 2010) 873.

260 Christoph Schreuer II. AMCO V INDONESIA: CONSENT THROUGH A DIRECT AGREEMENT BETWEEN THE PARTIES

The traditional form of expressing consent to arbitration is through an arbitration clause in a contract. This may be achieved through a compromissory clause in an investment contract between the host state and the investor submitting future disputes arising from the investment operation to arbitration. It is also possible to submit a dispute that has already arisen between the parties through consent expressed in a compromis although this rarely happens. The agreement on consent between the parties need not be recorded in a single instrument. An investment application made by the investor may provide for arbitration. If the application is approved by the competent authority of the host state, there is consent to arbitration by both parties. This is what happened in Amco v Indonesia.3 The investor had submitted an application to the Indonesian Foreign Investment Board to establish a locally incorporated company for the purpose of carrying out the investment operation. The application provided that any disagreements would be submitted to the International Centre for Settlement of Investment Disputes (ICSID) arbitration.4 The application was approved. The Tribunal accepted the validity of the consent clause and found that it was sufficient that its interpretation in good faith showed that the parties had agreed to ICSID arbitration. The Tribunal said: while a consent in writing to ICSID arbitration is indispensable, since it is required by Article 25(1) of the [ICSID] Convention, such consent in writing is not to be expressed in a solemn, ritual and unique formulation. … To conclude, there is a written consent to ICSID arbitration in the investment agreement (Article IX of the Application and its acceptance by the Government).5

III. SPP V EGYPT: CONSENT THROUGH HOST STATE LEGISLATION

A host state can express its consent to the Centre’s jurisdiction through a provision in its national legislation or through some other form of unilateral declaration. References to dispute settlement by arbitration in national investment legislation show a considerable measure of diversity.6 Not all references amount to consent to arbitration. Therefore, in each case the respective provisions in national laws must be studied carefully. In SPP v Egypt,7 the Request for Arbitration relied on Article 8 of Egypt’s Law No. 43 of 1974 which provided in relevant part: Investment disputes in respect of the implementation of the provisions of this Law shall be settled in a manner to be agreed upon with the investor, or within the framework of the agreements in force

3 Amco Asia Corporation and others v Republic of Indonesia, Decision on Jurisdiction (25 September 1983) ICSID Case No ARB/81/1. 4 ibid para 10. 5 ibid paras 23, 25. See also Churchill Mining PLC and Planet Mining Pty Ltd v Republic of Indonesia, Decision on Jurisdiction (24 February 2014) ICSID Case No ARB/12/14 and 12/40, paras 11–18, 199–217. 6 See also AR Parra, ‘Provisions on the Settlement of Investment Disputes in Modern Investment Laws, Bilateral Investment Treaties and Multilateral Instruments on Investment’ (1997) 12 ICSID Review/FILJ 287, 290, 314ff; Steingruber, above (n 1) 199–201; C Schreuer, ‘Investment Arbitration based on National Legislation’ in G Hafner and others (eds), Völkerrecht und die Dynamik der Menschenrechte – Liber Amicorum Wolfram Karl (Vienna, facultas. wuv Universitäts, 2012) 527. 7 Southern Pacific Properties (Middle East) v Arab Republic of Egypt, Decision on Preliminary Objection to Jurisdiction (27 November 1985) ICSID Case No ARB/84/3.

Landmark Investment Cases on State Consent 261 between the Arab Republic of Egypt and the investor’s home country, or within the framework of the Convention for the Settlement of Investment Disputes between the State and the nationals of other countries to which Egypt has adhered by virtue of Law No. 90 of 1971, where such Convention applies.8

About one year before the institution of arbitration, the Claimants sent a letter to Egypt’s Minister of Tourism, which said in relevant part: … we hereby notify you that we accept and reserve the opportunity of availing ourselves of the uncontestable jurisdiction of the International Centre for the Settlement of Investment Disputes, under the auspices of the World Bank, which is open to us as a result of Law No. 43 of 1974, Article 8 of which provides that investment disputes may be settled by ICSID arbitration.9

The Tribunal embarked upon a detailed grammatical analysis of Article 8 of Law No 43, including its Arabic original. This led it to conclude that the text mandated the submission of disputes to the various methods prescribed therein to the extent that such methods were applicable.10 The Tribunal rejected the idea that Article 8 had the consequence only of informing potential investors of Egypt’s willingness, in principle, to negotiate a consent agreement.11 Not every reference to arbitration in domestic legislation amounts to a binding offer of consent. Other legislative provisions referring to the settlement of disputes by arbitration indicate that further action on the part of the host state is necessary to establish consent. Tribunals have rejected attempts to base jurisdiction on legislative provisions of this kind.12

IV. CHURCHILL MINING V INDONESIA AND PLANET MINING V INDONESIA: CONSENT THROUGH TREATIES

The same technique of offering consent to foreign investors is employed by way of treaties, to which the host state is a party. The treaty constitutes the host state’s offer of consent. This offer may then be taken up by a national of the other state party to the treaty. Consent through bilateral investment treaties (BITs) has become accepted practice. Clauses offering investorstate arbitration can be found in many hundreds of BITs. In recent years, the majority of ICSID cases have been brought on the basis of consent offered by BITs or similar treaties.13 8 ibid

para 70. para 40. 10 ibid paras 74–88; See also SPP v Egypt, Dissenting Opinion (14 April 1988) paras 22–26. 11 SPP v Egypt, Decision on Jurisdiction (1985) paras 89–101; see also SPP v Egypt, Dissenting Opinion (1988) para 21. 12 Metal-Tech Ltd v Uzbekistan, Award 4 October 2013 paras 381–388; Biwater Gauff (Tanzania) Ltd v United Republic of Tanzania, Award (24 July 2008) ICSID Case No ARB/05/22, para 329; PNG Sustainable Development Program Ltd v Independent State of Papua New Guinea, Award (5 May 2015) ICSID Case No ARB/13/33, para 286; MNSS BV amd Recupero Credito Acciaio NV v Montenegro, Award (4 May 2016) ICSID No ARB(AF)/12/8, para 171; Venezuela Holdings, BV et al v Bolivarian Republic of Venezuela, Decision on Jurisdiction (10 June 2010) ICSID Case No ARB/07/27, paras 120–140; CEMEX Caracas Investments BV and CEMEX Caracas II Investments BV v Bolivarian Republic of Venezuela, Decision on Jurisdiction (30 December 2010) ICSID Case No ARB/08/15, paras 137, 138; Brandes Investment Partners, LP v Bolivarian Republic of Venezuela, Award (2 August 2011) ICSID Case No ARB/08/3, paras 79–118; Tidewater Inc, Tidewater Investement SRL, Tidewater Caribe, CA et al v Bolivarian Republic of Venezuela, Decision on Jurisdiction (8 February 2013) ICSID Case No ARB/10/5, paras 103–141; OPIC Karimun Corporation v Bolvarian Republic of Venezuela, Award (28 May 2013), paras 100–108, 165–179; ConocoPhillips Petrozuata BV, ConocoPhillips Hamaca BV and ConocoPhillips Gulf of Paria BV v Bolvarian Republic of Venezuela, Decision on Jurisdiction and the Merits (3 September 2013) para 259. 13 The first case brought on the basis of a BIT was Asian Agricultural Products Ltd v Republic of Sri Lanka, Award (27 June 1990) ICSID Case No ARB/87/3, para 2. See also CF Amerasinghe, ‘The Prawn Farm (AAPL) Arbitration’ (1992) 4 Sri Lanka Journal of International Law 155; P Rambaud, Des obligations de l’Etat vis-à-vis de l’investisseur 9 ibid

262 Christoph Schreuer Not every reference to arbitration in a BIT constitutes an offer of consent by the host state. Many clauses do amount to an unequivocal commitment.14 This would be the case where the BIT uses language like ‘hereby consents’ or ‘shall be submitted’. But other BITs merely contain undertakings to give consent in the future or hold out a general prospect of sympathetic consideration.15 Still others simply state that consent may be given by way of agreements with the investor. The fine line between the actual existence of consent and the mere prospect of future consent is illustrated by Churchill Mining v Indonesia and Planet Mining v Indonesia. Notably, in these two cases the same Tribunal deciding on the same day reached different results with respect to similar but not identical clauses in two different BITs. In Churchill Mining v Indonesia,16 Article 7(1) of the UK-Indonesia BIT contained the following clause: (1) The Contracting Party in the territory of which a national or company of the other Contracting Party makes or intends to make an investment shall assent to any request on the part of such national or company to submit, for conciliation or arbitration, to [ICSID].

The Tribunal found that this provision contained a standing offer to arbitrate any dispute that may arise in connection with an investment.17 The Tribunal noted that this clause provided only for ICSID arbitration. It also noted the absence of any indication as to how Indonesia was to give its assent. Together with the absence of a discretion to assent, this favoured the claimant’s position.18 A detailed examination of the BIT’s travaux préparatoires led the Tribunal to the conclusion that ‘the treaty drafters considered the “shall assent” language as functionally equivalent to “hereby consents” or similar wording’.19 In Planet Mining v Indonesia,20 the same Tribunal reached a different result on a somewhat differently worded ICSID clause. Article XI of the Australia-Indonesia BIT provided, that the investor may submit a dispute to ICSID adding that: Where that action is taken by an investor of one Party, the other Party shall consent in writing to the submission of the dispute to the Centre within forty-five days of receiving such a request from the investor;

The Tribunal found that this provision contained no standing offer to arbitrate. Rather, a further act on the part of Indonesia would be required to perfect consent.21 A decisive element in this decision was the time sequence foreseen for Indonesia’s consent after the dispute’s submission to ICSID.22 The Tribunal found confirmation for this result in the treaty practice of the two states, which indicated that Australia ‘deliberately entertains the distinction between advance consent and promise to consent’.23

étranger (Sentence AAPL c Sri Lanka) (1992) 38 Annuaire Français DIntl 501; NG Ziadé, Some Recent Decisions in ICSID Cases (1991) 6 ICSID Rev/FILJ 514. 14 See, eg, AAPL v Sri Lanka, above (n 13); American Manufacturing & Trading Inc v Republic of Zaire, Award (21 February 1997) ICSID Case No ARB/93/1, para 5.19. 15 Millicom International Operations BV and Sentel GSM SA v Republic of Senegal, Decision on Jurisdiction (16 July 2010) ICSID Case No ARB/08/20, paras 55–66. 16 Churchill Mining v Indonesia, above (n 5). 17 ibid para 231. 18 ibid paras 172–175. 19 ibid para 230. 20 Planet Mining v Indonesia, above (n 5). 21 ibid para 198. 22 ibid paras 162–165. 23 ibid para 195.

Landmark Investment Cases on State Consent 263 The Planet Mining Tribunal also noted that the giving of consent by the respondent after the institution of proceedings posed a problem since the ICSID Convention requires proof of consent for the registration of a request for arbitration.24

V. GENERATION UKRAINE V UKRAINE: ACCEPTANCE OF OFFER OF CONSENT BY THE INVESTOR, CONDITIONAL CONSENT

The agreement to arbitrate requires consent by both parties to the dispute. An investor may accept an offer of consent contained in a treaty or in legislation simply by instituting arbitration proceedings. Tribunals have accepted this form of expressing consent in numerous cases. Most investment arbitration cases in recent years are based on consent established in this way. Some tribunals have simply applied this principle without discussing its rationale.25 Other tribunals have explained the combination of the offer given by the host state through a BIT and the acceptance by the investor through the request for arbitration.26 In Generation Ukraine v Ukraine,27 the Ukraine-US BIT contained an offer of arbitration. The claimant had instituted arbitration proceedings on the basis of that offer. The Tribunal said: it is firmly established that an investor can accept a State’s offer of ICSID arbitration contained in a bilateral investment treaty by instituting ICSID proceedings. There is nothing in the BIT to suggest that the investor must communicate its consent in a different form directly to the State; … It follows that the Claimant validly consented to ICSID arbitration by filing its Notice of Arbitration at the ICSID Centre.28

The institution of proceedings is not, however, the only way for the investor to accept the offer of consent in a treaty. In a number of cases investors have expressed their consent in a separate document some time before submitting their request for arbitration.29 Generation Ukraine v Ukraine also serves to illustrate another point. Some expressions of consent are contingent upon the fulfilment of a future condition. In some cases, the conditions ratione personae for the Centre’s jurisdiction have not yet been met when the document containing the consent clause is signed. For instance, in the case of ICSID arbitration, the host state or the state of the investor’s nationality may not yet have ratified the ICSID Convention. In such a case, the date of consent will be the date on which all the conditions have been met.30 In Generation Ukraine v Ukraine, the entry into force of the BIT between Ukraine and the US on 16 November 1996 antedated the entry into force of the ICSID Convention for Ukraine 24 ibid

para 166. eg, AAPL v Sri Lanka, above (n 13) paras 2–4; Fedax v Venezuela, Decision on Jurisdiction (11 July 1997) ICSID Case No ARB/96/3, para 30; Emilio Augustín Maffezini v Kingdom of Spain, Decision on Jurisdiction (25 January 2000) ICSID Case No ARB/97/7, para 19. 26 See, eg, Wena Hotels v Egypt, Decision on Jurisdiction (29 June 1999) 6 ICSID Reports 87; Salini v Morocco, Decision on Jurisdiction (23 July 2001) ICSID Case No ARB/00/4, para 27; Tokios Tokelės v Ukraine, Decision on Jurisdiction (29 April 2004) ICSID Case No ARB/02/18, paras 94–100; Saipem SpA v The People’s Republic of Bangladesh, Decision on Jurisdiction (21 March 2007) ICSID Case No ARB/05/07, para 74. 27 Generation Ukraine, Inc v Ukraine, Award (16 September 2003) ICSID Case No ARB/00/9. 28 ibid paras 12.2, 12.3. 29 See SPP v Egypt, above (n 7) para 40. See also Azurix Corp v Republic of Argentina, Decision on Jurisdiction (8 December 2003) ICSID Case No ARB/01/12, para 56; ADC Affiliate Limited and ADC & ADMC Management Limited v Republic Hungary, Award (2 October 2006) ICSID Case No ARB/03/16, para 363; Wintershall Akitengesellschaft v Argentine Republic, Award (8 December 2008) ICSID Case No ARB/04/14, para 10. 30 Holiday Inns SA and others v Morocco, Decision on Jurisdiction (12 May 1974) 1 ICSID Reports 667/8. See also Cable Television of Nevis, Ltd and Cable Television of Nevis Holdings, Ltd v Federation of St Kitts and Nevis, Award (13 January 1997) ICSID Case No ARB/95/2, paras 2.18, 4.09, 5.24. 25 See,

264 Christoph Schreuer on 7 July 2000. The Respondent argued that its consent, expressed in the BIT, was only ‘preliminary’ and subject to ‘final’ consent once the ICSID Convention came into force for Ukraine.31 The Tribunal rejected this argument. It noted that there was nothing in the BIT that suggested that its consent was only preliminary. The Tribunal said: [12.6] Ukraine’s consent to ICSID arbitration in Article VI(3) of the BIT was naturally conditional upon a future event, viz. Ukraine’s ratification of the ICSID Convention. This no doubt explains the proviso to the consent in Article 3(a)(i) which states: ‘provided that the Party is a party to [the ICSID] Convention’. But Ukraine’s free standing consent to ICSID arbitration was perfected as soon as the ICSID Convention entered into force for Ukraine on 7 July 2000. Ukraine did not make any reservation to the BIT whereby it could reassess the status of its consent once the condition precedent for its full validity had been fulfilled.32

The Tribunal concluded that Ukraine’s consent was valid since the condition precedent to Ukraine’s offer to arbitrate had been fulfilled by the date of the institution of proceedings.33

VI. PING AN V BELGIUM: APPLICATION OF CONSENT TO PAST EVENTS AND PAST DISPUTES

Bilateral investment treaties frequently provide that they shall also apply to investments made before their entry into force.34 Some BITs state, however, that they shall not apply to disputes that have arisen before that date.35 The time of the dispute is not identical with the time of the events leading to the dispute. Typically, the incriminated acts will have occurred some time before the dispute. Therefore, the exclusion of disputes occurring before a certain date cannot be read as excluding jurisdiction over events occurring before that date.36 In Ping An v Belgium37 there were two successive BITs of 1986 and 2009 respectively between BLEU38 and China. The 1986 BIT only provided for jurisdiction for the amount of compensation in case of expropriation, while jurisdiction under the 2009 BIT covered a much wider range of disputes. The dispute at hand arose before the 2009 BIT came into force. The

31 Generation

Ukraine v Ukraine, above (n 27) para 12.1. para 12.6. 33 ibid para 12.8. In the same sense: Toto Costruzioni Generali SpA v Republic of Lebanon, Decision on Jurisdiction (11 September 2009) ICSID Case No ARB/07/12, paras 91–94. 34 Alex Genin, Eastern Credit Limited, Inc and AS Baltoil v Republic of Estonia, Award (25 June 2001) ICSID Case No ARB/99/2, para 326; SGS Société Générale de Surveillance SA v Islamic Republic of Pakistan, Decision on Jurisdiction (6 August 2003) ICSID Case No ARB/01/13, para 153; Pan American v Argentina, Decision on Preliminary Objections (27 July 2006) ISCID Case No ARB/ para 211; Ioannis Kardassopoulos v Republic of Georgia, Decision on Jurisdiction (6 July 2007) ISCID Case No ARB/05/18, paras 196–201; OKO Pankki Oyi, VTB Bank (Deutschland) AG and Sampo Bank Plc v Republic of Estonia, Award (19 November 2007) ICSID Case No ARB/04/6, paras 184–186; Bayindir Insaat Turizm Ticaret Ve Sanayi AS v Islamic Republic of Pakistan, Award (27 August 2009) ICSID Case No ARB/03/29, para 131; H&H Enterprises Investments, Inc v Arab Republic of Egypt, Decision on Jurisdiction (5 June 2012) ICSID Case No ARB/09/15, paras 52–56. 35 Some BITs provide that they will not apply to disputes that have arisen before a certain date in the past. See EuroGas Inc and Belmont Resources Inc v Slovak Republic, Award (18 August 2017) ICSID Case No ARB/14/14, paras 427–461. 36 Renée Rose Levy de Levi v Republic of Peru, Award (9 January 2015) ICSID Case No ARB/10/17, para 167; Philip Morris Asia Limited v The Commonwealth of Australia, Award on Jurisdiction and Admissibility (17 December 2015) PCA Case No 2012-12, para 532. 37 Ping An Life Insurance Company, Limited and Ping An Life Insurance (Group) Company, Limited v The Government of Belgium, Award (30 April 2015) ICSID Case No ARB/12/29. 38 Belgium Luxembourg Economic Union. 32 ibid

Landmark Investment Cases on State Consent 265 Tribunal distinguished between the temporal application of the jurisdictional provisions and the applicability of substantive provisions.39 The Tribunal said: In the present case the Tribunal will not employ any presumption against retroactivity, because in its view the application of a new dispute settlement mechanism to acts which may have been unlawful when they were committed is not in itself the retroactive application of law.40

The dispute settlement clause in the 2009 BIT started with the words ‘[w]hen a legal dispute arises’. The Tribunal reached the conclusion that this clause was not to be read as covering past disputes. It held that ‘[w]hen a legal dispute arises’ must be read as referring only to disputes that arise after the BIT’s entry into force.41 It followed that the claim was dismissed for lack of jurisdiction. Consent expressed in a treaty may well cover events that took place before the treaty’s entry into force.42 Jurisdiction over events that occurred prior to the BIT’s entry into force is not contrary to the principle of non-retroactivity as enshrined in Article 28 of the Vienna Convention on the Law of Treaties (VCLT).43 It follows that the question whether acts and events that occurred prior to an expression of consent to arbitration are covered by the latter should be distinguished from the issue of the applicable substantive law.44 A jurisdictional provision in a treaty may extend to events that took place prior to the treaty’s entry into force. On the other hand, the substantive law in force at the time of the relevant events will have to be applied to the merits of the case.45 A treaty may exclude jurisdiction in respect of acts that occurred before its entry into force.46 If consent to arbitration contained in a treaty is limited to violations of that treaty, the date of the treaty’s entry into force is also necessarily the date from which acts and events are covered by consent to jurisdiction.47 For instance, under the NAFTA48 and under the ECT49 the scope of the consent to arbitration is limited to claims arising from alleged breaches of the respective treaties. In that case, the entry into force of the substantive law also determines the Tribunal’s jurisdiction ratione temporis since the Tribunal may only hear claims for violation of that law.

39 Ping

An v Belgium, above (n 37) para 186. para 218. 41 ibid para 224. See also Salini Costruttori SpA and Italstrade SpA v The Hashemite Kingdom of Jordan, Decision on Jurisdiction (9 November 2004) ICSID Case No ARB/02/13, para 170; Impregilo SpA v The Islamic Republic of Pakistan, Decision on Jurisdiction (22 April 2005) ICSID Case No ARB/03/3, paras 299–300; ABCI Investments NV v Republic of Tunisia, Decision on Jurisdiction (18 February 2011) ICSID No ARB/04/12, paras 169–170. 42 N Gallus, The Temporal Scope of Investment Protection Treaties (London, British Institute of International and Comparative Law, 2009) 138–141. 43 See N Gallus, ‘Article 28 of the Vienna Convention on the Law of Treaties and Investment Treaty Decisions’ (2016) 31 ICSID Review 290. 44 Impregilo v Pakistan, above (n 41) para 309. See also SGS Société Générale de Surveillance SA v The Republic of the Philippines, Decision on Jurisdiction (29 January 2004) ICSID Case No ARB/02/6, para 167. See also C Schreuer, ‘Jurisdiction and Applicable Law in Investment Treaty Arbitration’ (2014) 1 McGill Journal of Dispute Resolution 1–25; Steingruber, above (n 1) 287–288. 45 ‘Island of Palmas case’ in Reports of International Arbitral Awards, vol II (New York, United Nations Publications, 1949), 829, 845; Tradex Hellas SA v Republic of Albania, Decision on Jurisdiction (24 December 1996) 5 ICSID Reports 65; SGS v Philippines, above (n 44) para 166; Salini v Jordan, above (n 41) paras 176, 177; Impregilo v Pakistan, above (n 41) para 311; Ioan Micula, Viorel Micula, SC European Food SA, SC Starmill SRL and SC Multipack SRL v Republic of Romania, Decision on Jurisdiction (24 September 2008) ICSID Case No ARB/05/20, paras 153–157; Jan de Nul NV and Dredging International NV v Arab Republic of Egypt, Award (6 November 2008) paras 132–135. 46 See Railroad Development Corporation v Republic of Guatemala, Second Decision on Jurisdiction (18 May 2010) ICSID Case No ARB/07/23, paras 114–116. 47 See Generation Ukraine v Ukraine, above (n 27) paras 11.1–11.4, 17.1, 17.5. 48 North American Free Trade Agreement (entered into force 1 January 1994) (NAFTA) Art 1116. 49 Energy Charter Treaty (signed 17 December 1994, entered into force 16 April 1998) (ECT) Art 26(1). 40 ibid

266 Christoph Schreuer VII. SGS V PARAGUAY: THE SCOPE OF CONSENT

The scope of consent to arbitration offered in treaties varies.50 Some clauses providing consent are wide and unlimited. Many BITs in their consent clauses contain phrases such as ‘all disputes concerning investments’ or ‘any legal dispute concerning an investment’. These provisions do not restrict a tribunal’s jurisdiction to claims arising from the BIT’s substantive standards. By their own terms, these consent clauses encompass disputes that go beyond the interpretation and application of the BIT itself and would cover claims that arise from a contract in connexion with the investment,51 from customary international law and from other treaties. In SGS v Paraguay,52 a contract entered into between SGS and the Ministry of Finance of Paraguay contained a forum selection clause in favour of the Courts of the City of Asunción under the Law of Paraguay. The clause on the settlement of investor-state disputes in the BIT between Paraguay and Switzerland referred to ‘disputes with respect to investments’. The Tribunal rejected the respondent’s argument that the contractual forum selection ousted consent to arbitration under the BIT. The Tribunal noted that the claimant had not put forward contract claims but was arguing breaches of the BIT. In addition, the Tribunal remarked that the BIT’s dispute settlement clause was broad enough to cover contract claims: Article 9 [of the BIT] provides for the resolution of ‘disputes with respect to investments between a Contracting Party and an investor of the other Contracting Party,’ … There is no qualification or limitation in this language on the types of ‘disputes with respect to investments’ that a Swiss investor may bring against the Republic of Paraguay. The ordinary meaning of Article 9 would appear to give this Tribunal jurisdiction to hear claims for violation of Claimant’s rights under the Contract – surely a dispute ‘with respect to’ Claimant’s investment – should Claimant have chosen to bring them before us.53

The Tribunal added: As previously noted, the BIT’s dispute resolution provisions (Article 9) are not on their terms limited to claims for breach of the BIT itself. Article 9(1) arguably extends the Treaty dispute settlement process to all manner of ‘disputes related to investments’ – a category broad enough to encompass contract disputes.54

A number of other tribunals have endorsed this position.55 It has not, however, found unanimous support. For instance, in SGS v Pakistan56 the Tribunal found that a BIT’s consent 50 For an overview see C McLachlan, L Shore and M Weiniger, International Investment Arbitration (Oxford, Oxford University Press, 2017) 48–52. 51 For discussion of this issue see S Alexandrov, ‘Breaches of Contract and Breaches of Treaty’ (2004) 5 Journal of World Investment & Trade 555, 572; J Griebel, ‘Jurisdiction over ‘Contract Claims’ in Treaty-Based Investment Arbitration on the Basis of Wide Dispute Settlement Clauses in Investment Agreements’ (2007) 5 Transnational Dispute Management; J van Haersolte-van Hof and AK Hoffmann, ‘The Relationship between International Tribunals and Domestic Courts’ in P Muchlinski, F Ortino and C Schreuer (eds), The Oxford Handbook of International Investment Law (Oxford, Oxford University Press, 2008) 962; Steingruber, above (n 1) 299–302. 52 SGS Société Générale de Surveillance SA v Republic of Paraguay, Decision on Jurisdiction (12 February 2010) ICSID Case No ARB/07/29. 53 ibid para 129. 54 ibid para 183. 55 Salini v Morocco, above (n 26) para 61; SGS v Philippines, above (n 44) paras 131–135, 169(3); Tokios Tokelės v Ukraine, above (n 26) para 52, fn 42; Impregilo v Pakistan, above (n 41) paras 57, 82, 102, 188; Siemens AG v Argentina, Award (6 February 2007) ICSID Case No ARB/02/8, para 205; Parkerings-Compagniet AS v Republic of Lithuania, Award (11 September 2007) ICSID Case No ARB/05/8, paras 261–266; Alpha Projektholding GmbH v Ukraine, Award (8 November 2010) ICSID Case No ARB/07/16, para 243; Philip Morris Brands Sàrl, Philip Morris Products SA and Abal Hermanos SA v Oriental Republic of Uruguay, Decision on Jurisdiction (2 July 2013) ICSID Case No ARB/10/7, paras 107, 109; Metal-Tech v Uzbekistan, above (n 12) para 378. 56 SGS v Pakistan, above (n 34).

Landmark Investment Cases on State Consent 267 clause referring to ‘disputes with respect to investments’ did not endow it with jurisdiction over contract claims which did not also constitute breaches of the substantive standards of the BIT.57 Although a number of tribunals have distanced themselves from that decision,58 however, some tribunals, without analysing the matter, seem to be under the impression that once they operate under a treaty-based consent clause, they are restricted to treaty claims regardless of the wording of a BIT’s jurisdiction clause.59 Other BIT clauses offering consent to arbitration circumscribe the scope of consent to arbitration in narrower terms. A provision that is typical of US BITs covers breaches of the treaty’s substantive standards, of an investment authorisation, or of an investment agreement.60 Other treaties restrict consent to disputes involving their substantive provisions. Tribunals operating under these restrictive clauses have found that they could only apply the substantive standards contained in these treaties.61 The scope for the jurisdiction of tribunals is even narrower where consent is limited to one or some of the rights granted under the Treaty. Some BITs restrict jurisdiction to expropriation or to the amount of compensation due after an expropriation. Tribunals have held that clauses of this kind restricted their jurisdiction to claims based on expropriation to the exclusion of other standards of protection.62 However, clauses referring to compensation for expropriation extended to the question of the existence of an expropriation.63

VIII. AMBIENTE UFFICIO V ARGENTINA: CONSENT SUBJECT TO PROCEDURAL CONDITIONS

Offers of consent to arbitration are often subject to procedural conditions that a claimant must comply with before proceeding to arbitration. A common condition is that the parties first seek an amicable settlement through consultations or negotiations. Some national investment laws and numerous BITs contain the condition that a negotiated settlement must be attempted before resort can be had to arbitration. In many treaties this condition is subject to a certain period of time.

57 ibid para 161. For a case with a similar result see Consortium Groupement LESI-DIPENTA v République algérienne démocratique et populaire, Award (10 January 2005) ICSID Case No ARB/03/08, para 25. 58 SGS v Philippines, above (n 44) paras 133–135; Tokios Tokelės v Ukraine, above (n 26) para 52, fn 42; See also the discussions in Salini v Jordan, above (n 41) paras 97–101. 59 Urbaser SA v Republic of Argentina, Decision on Jurisdiction (19 December 2012) ICSID Case No ARB/07/26, para 254; Vladislav Kim and others v Republic of Uzbekistan, Decision on Jurisdiction (8 March 2017) ICSID Case No ARB/13/6, paras 162–164; UAB E energija (Lithuania) v Republic of Lithuania, Award (22 December 2017) ICSID Case No ARB/12/33, paras 498, 846, 853. 60 See KS Gudgeon, ‘Arbitration Provisions of U.S. Bilateral Investment Treaties’ in SJ Rubin and RW Nelson (eds), International Investment Disputes: Avoidance and Settlement (Eagan, West Group, 1985) 41, 45–47; KJ Vandevelde, ‘U.S. Bilateral Investment Treaties – The Second Wave’ (1993) 14 Michigan Journal of International Law 621, 655. 61 Suez, Sociedad General de Aguas de Barcelona, SA and Vivendi Universal, SA v Argentine Republic, Decision on Liability (30 June 2010) ICSID Case No ARB/03/19, para 63; Spyridon Roussalis v Romania, Award (7 December 2011) ICSID Case No ARB/06/1, paras 43, 679–683; Iberdrola Energía v Republic of Guatemala, Award (17 August 2012) ICSID Case No ARB/09/5, paras 296–310. 62 Telenor Mobile Communications AS v Republic of Hungary, Award (13 September 2006) ICSID Case No ARB/04/15, paras 18(2), 25, 57, 81–83; Saipem SpA v The People’s Republic of Bangladesh, Decision on Jurisdiction (21 March 2007) ICSID Case No ARB/05/07, paras 70, 129–133. 63 Señor Tza Yap Shum v Republic of Peru, Decision on Jurisdiction (19 June 2009) ICSID Case No ARB/07/6, paras 129–188; Beijing Urban Construction Co Ltd v Republic of Yemen, Decision on Jurisdiction (31 May 2017) ICSID Case No ARB/14/30, paras 50, 59–69, 74–108.

268 Christoph Schreuer If these ‘waiting periods’ are jurisdictional requirements, they must be complied with by the time of the institution of proceedings. As a consequence, if the request for arbitration is submitted before the expiry of the period foreseen for a settlement, a tribunal would have to decline jurisdiction.64 In Ambiente Ufficio v Argentina,65 Article 8(1) of the Italy-Argentina BIT provided that disputes ‘shall be, insofar as possible, resolved through amicable consultations’. The Tribunal found that this crated an obligation to enter into consultations.66 It did, however, find that there had to be a chance for meaningful consultations. The Tribunal said: there is considerable authority for the proposition that mandatory waiting periods for consultations (let alone a simple duty to consult, as in the present case) do not pose an obstacle for a claim to proceed to the merits phase if there is no realistic chance for meaningful consultations because they have become futile or deadlocked.67

In the particular case, no consultations had taken place. The Tribunal noted that Argentina had passed legislation which forbade any settlement.68 It followed that the claimant had not violated the requirement to engage in consultations.69 Other tribunals too have decided that there was no need to comply with waiting periods prior to the institution of arbitration proceedings since they would have been futile and would hence not have served any useful purpose.70 The reaction of tribunals in cases where the claimants had not complied with the requirement to first attempt an amicable settlement has not, however, been uniform.71 In a number of cases, tribunals found that non-compliance with waiting periods did not affect their jurisdiction.72 Other tribunals have treated waiting periods as requirements for jurisdiction or admissibility that had to be strictly observed.73 64 The International Court of Justice has addressed the issue of non-compliance with a requirement to engage in negotiations in a treaty providing for the Court’s jurisdiction: Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States of America), Judgment (Jurisdiction and Admissibility) [1986] ICJ Rep 14, 427–429. See also PM Dupuy, ‘Preconditions to Arbitration and Consent of States to ICSID Jurisdiction’ in M Kinnear and others (eds), Building International Investment Law (Alphen aan den Rijn, Wolters Kluwer, 2016) 219, 227–228. 65 Ambiente Ufficio SpA and others v Argentina, Decision on Jurisdiction and Admissibility (8 February 2013) ICSID Case No ARB/08/9. 66 ibid paras 577–580. 67 ibid para 582. 68 ibid para 585. 69 ibid para 588. 70 SGS v Pakistan, above (n 34) paras 80, 183–184; LESI–DIPENTA v Algeria, above (n 57) para 32(iv); Teinver SA, Transportes de Cercanías Sa and Autobuses Urbanos del Sur SA v Argentine Republic, Decision on Jurisdiction (21 December 2012) ICSID Case No ARB/09/1, paras 126–129; Philip Morris v Uruguay, above (n 55) paras 227–229; Giovanni Alemanni and others v Argentine Republic, Decision on Jurisdiction and Admissibility (17 November 2014) ICSID Case No ARB/07/8, paras 301–317. 71 C Schreuer, ‘Travelling the BIT Route: Of Waiting Periods, Umbrella Clauses and Forks in the Road’ (2004) 5 Journal of World Investment & Trade 231, 232; G Born and M Šćekić, ‘Pre-Arbitration Procedural Requirements “A Dismal Swamp’’’ in DD Caron and others (eds), Practising Virtue Inside International Arbitration (Oxford, Oxford University Press, 2015) 227. 72 Ethyl Corporation v The Government of Canada, Decision on Jurisdiction (24 June 1998) UNCITRAL, paras 76–88; Ronald S Lauder v The Czech Republic, Final Award (3 September 2001) UNCITRAL, paras 181–191; Wena Hotels v Egypt, above (n 26); Metalclad Corporation v United Mexican States, Award (30 August 2000) ICSID Case No ARB(AF)/97/1, paras 64–67; Bayindir v Pakistan, above (n 34) paras 88–103; Biwater Gauff v Tanzania, above (n 12) paras 343–347; Abaclat and others v Argentine Republic, Decision on Jurisdiction and Admissibility (4 August 2011) ICSID Case No ARB/07/5, para 564. 73 Antoine Goetz et consorts v République du Burundi, Award (10 February 1999) ICSID Case No ARB/95/3, paras 90–93; Enron Corporation v Argentine Republique, Decision on Jurisdiction (14 January 2004) ICSID Case No ARB/01/3, para 88; Burlington Resources Inc v Republic of Ecuador, Decision on Jurisdiction (2 June 2010) ICSID Case No ARB/08/5, paras 312–318, 332–340; Murphy Exploration & Production Company International v Republic

Landmark Investment Cases on State Consent 269 The decisive criterion should be whether there was a promising opportunity for a settlement. There is little point in declining jurisdiction and sending the parties to the negotiating table if negotiations are obviously futile. Negotiations remain possible while the arbitration proceedings are pending. Even if the institution of arbitration was premature, the waiting period will often have expired by the time a decision on jurisdiction is rendered. Under these circumstances, compelling the claimant to start the proceedings anew would be uneconomical. A better way to deal with non-compliance with a waiting period is a suspension of proceedings to allow additional time for negotiations if these appear promising.74 Some consent clauses in BITs provide for a mandatory attempt at settling the dispute in the host state’s domestic courts.75 The period foreseen for this purpose is often 18 months. The investor may proceed to international arbitration if the domestic proceedings do not result in a decision during that period or if the dispute persists after the domestic decision. A requirement of this kind attached to consent to arbitration creates a considerable burden to the party seeking arbitration with little chance of advancing the settlement of the dispute. A final decision by the domestic courts in a complex investment dispute is unlikely within eighteen months. Even if a decision is rendered, the dispute is likely to persist if the investor is dissatisfied with the outcome. Therefore, arbitration remains an option after the expiry of the period of 18 months. It follows that the most likely effect of a clause of this kind is delay and additional cost. One tribunal has called a provision of this kind ‘nonsensical from a practical point of view’.76 Some tribunals have found the requirement to submit the dispute to domestic courts for a certain period to be procedural. Therefore, it could be handled with some flexibility.77 In a number of cases, tribunals decided that resort to domestic courts prior to the institution of arbitration proceedings would have been futile and would have served no useful purpose. Therefore, there was no need to comply with directions contained in treaties to first go to domestic courts.78 In Ambiente Ufficio v Argentina,79 Article 8 of the Argentina-Italy BIT provided that, if amicable consultations were unsuccessful, the dispute is to be submitted ‘to a competent administrative or judicial jurisdiction’ of the host state for 18 months. Only if the dispute

of Ecuador, Award on Jurisdiction (15 December 2010) PCA Case No 2012-16, paras 90–157; Tulip Real Estate and Development Netherland v Republic of Turkey, Decision on Jurisdiction (5 March 2013) ICSID Caso No ARB/11/28, paras 55–72; Supervisión y Control SA v Republic of Costa Rica, Award (18 January 2017) ICSID Case No ARB/12/4, paras 336–348, 351. 74 See Western NIS Entreprise Fund v Ukraine, Order (16 March 2006) ICSID Case No ARB/04/2; Daimler Financial Services AG v Argentine Republic, Award (22 August 2012) ICSID Case No ARB/05/1, para 188. 75 C Schreuer, ‘Calvo’s Grandchildren: The Return of Local Remedies in Investment Arbitration’ (2005) 4 Law and Practice of International Courts and Tribunals 1, 3–5; K von Papp, ‘Biting the Bullet or Redefining “Consent” in Investor-State Arbitration? Pre-Arbitration Requirements After BG Group v Argentina’ (2015) 16 Journal of World Investment & Trade 695; Born and Šćekić, above (n 71); Dupuy, above (n 64) 219; McLachlan, Shore and Weiniger, above (n 50) 55–58. 76 Plama Consortium Limited v Republic of Bulgaria, Decision on Jurisdiction (8 February 2005) ICSID Case No ARB/03/24, para 224. 77 Telefónica SA v Argentine Republic, Decision on Jurisdiction (25 May 2006) ICSID Case No ARB/03/20, paras 91–93; TSA Spectrum de Argentina SA v Argentine Republic, Award (19 December 2008) ICSID Case No 05/5, paras 98–113; Teinver v Argentina, above (n 70) para 135; Philip Morris v Uruguay, above (n 55) paras 98–150; İçkale Inşaat Limited Siketi v Turkmenistan, Award (8 March 2016) ICSID Case No ARB/10/24, paras 195–263. 78 Abaclat v Argentina, above (n 72) paras 585–591; Urbaser SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v Argentina, Decision on Jurisdiction (19 December 2012) ICSID Case No ARB/07/26, paras 106–202; Alemanni v Argentina, above (n 70) paras 302–317. 79 Ambiente Ufficio v Argentina, above (n 65).

270 Christoph Schreuer still exists after that period may the dispute be submitted to international arbitration. In the particular case, the claimants had not brought their case before the domestic courts. The Tribunal found that the requirement to have recourse to domestic courts was a binding precondition for access to international arbitration.80 At the same time, it found that it had to apply the ‘futility exception’ to this obligation: the futility exception to the exhaustion of local remedies rule in the field of diplomatic protection is, in the light of Art. 31(3)(c) of the VCLT, also applicable to clauses requiring recourse to domestic courts in international investment law.81

The Tribunal noted that Argentinean legislation had ruled out redress before Argentine courts and the Supreme Court had confirmed the constitutionality of the legislation. Under these circumstances, the claimants had not violated the BIT’s requirement of having recourse to respondent’s domestic courts.82 The Tribunal said: Given the jurisprudence of the Supreme Court of Argentina and in the light of the circumstances prevailing in the present case, the Tribunal concludes that having recourse to the Argentine domestic courts and eventually to the Supreme Court would not have offered Claimants a reasonable possibility to obtain effective redress from the local courts and would have accordingly been futile. Hence, Claimants did not violate the duty to have recourse to Argentine courts under Art. 8(2) and (3) of the Argentina-Italy BIT when they submitted the Request for Arbitration on 23 June 2008.83

IX. MILLICOM V SENEGAL: TERMINATION OF CONTRACT CONTAINING CONSENT

The termination of an investment contract between the host state and the investor does not necessarily terminate the validity of a clause providing for arbitration contained in that contract. To hold otherwise would largely deprive the arbitration clause of its intended effect. The tribunal must have the power to decide on disputes concerning the agreement’s termination or invalidity even if the tribunal’s very existence depends on the agreement’s validity. International arbitral practice has developed the doctrine of the severability or separability of arbitration agreements. Under this doctrine, the agreement providing for arbitration assumes a separate existence, which is autonomous and legally independent of the agreement containing it.84 The most important argument in favour of this doctrine is the assumption that the parties, in providing for the arbitration of disputes relating to the agreement, intended all disputes, including disputes about the agreement’s validity, to be resolved through arbitration.85 Otherwise, a party could at any time defeat its obligation to arbitrate simply by declaring the agreement void or terminated. Therefore, the ‘intention of the parties and the requirements of effective arbitration combine to give rise to the concept of severability’.86 This principle of

80 ibid

paras 589–593. paras 603, 607. para 628. 83 ibid para 620. 84 For an extensive analysis see SM Schwebel, International Arbitration: Three Salient Problems (Cambridge, Cambridge University Press, 1987) 1–60. 85 ibid 3. 86 ibid 4. 81 ibid 82 ibid

Landmark Investment Cases on State Consent 271 severability of the arbitration agreement is supported by the weight of international arbitral codifications87 and cases as well as by national arbitral practice.88 In Millicom v Senegal,89 the Republic of Senegal had granted the claimant a telephone concession. Article 11 of the Concession Agreement provided for international arbitration in case of a dispute. Just over two years later, Senegal formally terminated the Concession agreement citing violations of the agreement on the part of the investor.90 When Millicom initiated ICSID proceedings under the Concession Agreement, the respondent contested the tribunal’s jurisdiction arguing that the Agreement had ended as of the date of its termination. The Tribunal rejected this argument and said: it is undisputed that an arbitration clause is autonomous and that the end of a contract in which it is incorporated does not eliminate its effects; on the contrary, the arbitration which is agreed to therein also covers the principle and effects of the end of the contract. To admit the contrary would amount to depriving it of an important part of its practical effect since it would suffice for a party to abandon a contract in order to escape from it. This reasoning must apply also to the Concession.91

Despite the clear authority in favour of the severability of a contractual arbitration clause, it is advisable to include an explicit provision in a consent clause to the effect that it applies to the contract’s validity, construction, application and termination.92

X. RUMELI V KAZAKHSTAN: TERMINATION OF LEGISLATION PROVIDING FOR CONSENT

A host state is free to change its investment legislation including a provision offering consent to arbitration to foreign investors. An offer of consent contained in national legislation that has not been taken up by the investor will lapse when the legislation is repealed. The situation is different if the investor has accepted the offer in writing while the legislation was still in force. The consent agreed to by the parties then becomes insulated from the validity of the legislation containing the offer. It assumes a contractual existence independent of the legislative instrument that helped to bring it about. In Rumeli v Kazakhstan,93 the Respondent’s Foreign Investment Law (FIL) of 1994 contained an offer of consent to ICSID arbitration. In addition, Article 6(1) of the FIL granted foreign investors protection against adverse changes in legislation for a period of ten years from the date of the investment.94 The claimants made the relevant investments from 1998 to 2002. The FIL was repealed in 2003. The claimants request for arbitration, containing their acceptance of the offer of arbitration was registered in 2005.95

87 See ICSID Additional Facility Rules, r 45(1); 2012 ICC Rules of Arbitration, Art 6(9); 2014 UNCITRAL Arbitration Rules, Art 23(1); 1985 UNCITRAL Model Law on International Commercial Arbitration, Art 16(1); Institut de Droit International, ‘Resolution on Arbitration between States, State Enterprises or State Entities, and Foreign Enterprises, Art 3(a)’ (1989) 63 Annuaire II 324, 326. 88 Schwebel, above (n 84) 24–59. 89 Millicom International Operation BV and Sentel GSM SA v Republic of Senegal, Decision on Jurisdiction (16 July 2010) ICSID Case No ARB/08/20. 90 ibid paras 8, 11, 12, 88. 91 ibid para 91. See also Duke Energy International Peru Investments No 1 Ltd v Republic of Peru, Decision of the ad hoc Committee (1 March 2011) ICSID Case No ARB/03/28, paras 125–144. 92 GR Delaume, ‘How to Draft an ICSID Arbitration Clause’ (1992) 7 ICSID Review/FILJ 168, 174. 93 Rumeli Telekom AS and Telsim Mobil Telekomunikasyon Hizmetleri v Kazakhstan, Award (29 July 2008) ICSID Case No ARB/05/16. 94 ibid para 333. 95 ibid paras 220–222, 334.

272 Christoph Schreuer The respondent argued that the consent to arbitrate under the FIL was no longer effective since the FIL had been repealed prior to the investor’s acceptance of the offer of arbitration through the request for arbitration.96 The Tribunal found that it had jurisdiction on the basis of the FIL pursuant to its Article 6(1). Therefore, the offer of consent in the FIL remained applicable to the dispute.97 In addition, the Tribunal applied the doctrines of acquired rights and estoppel to the situation. The Tribunal said: Respondent has expressed its consent to ICSID arbitration on December 28, 1994, the date of the entry into force of the FIL, and it remains applicable to the dispute pursuant to Article 6(1). On the other hand, Claimants have consented to ICSID jurisdiction by filing their Request for Arbitration. The Arbitral Tribunal has therefore jurisdiction under the FIL. Besides Article 6(1), it is also well established in international law that a State may not take away accrued rights of a foreign investor by domestic legislation abrogating the law granting these rights. This is an application of the principles of good faith, estoppel and venire factum proprium.98

XI. FÁBRICA DE VIDRIOS V VENEZUELA: CONSENT AND THE DENUNCIATION OF THE ICSID CONVENTION

Under Article 71 of the ICSID Convention, a Contracting State may denounce the Convention with six months’ notice.99 Since participation in the Convention of the host state and of the investor’s state of nationality are conditions for the validity of consent to ICSID arbitration, the termination of either state’s participation in the Convention would vitiate consent. Article 72 of the ICSID Convention blocks this indirect way of withdrawing consent. It provides that the Convention’s denunciation by the host state or the investor’s home state shall not affect the rights and obligations arising out of consent given before receipt of the notice of denunciation.100 Tribunals have given different interpretations to these provisions on denunciation with respect to the timing of expressions of consent. Some tribunals did not see a need for perfected consent at the time of the notice of denunciation and held that a request for arbitration accepting an offer of consent could still be filed within the six-month period until the denunciation takes effect in accordance with Article 71.101 Venezuela submitted its notice of denunciation of the ICSID Convention on 24 January 2012. In accordance with Article 71 of the Convention the denunciation took effect six months

96 ibid

para 308. paras 332–336. 98 ibid paras 334, 335, cf also Caratube International Oil Company LLP v Republic of Kazakhstan, Award (27 September 2017) ICSID Case No ARB/08/12, paras 689–696. 99 ICSID Convention on the Settlement of Investments Disputes Between States and Nationals of other States (entered into force 14 October 1966) Art 71: Any Contracting State may denounce this Convention by written notice to the depositary of this Convention. The denunciation shall take effect six months after receipt of such notice. 100 ICSID Convention, Art 72: Notice by a Contracting State pursuant to Articles 70 or 71 shall not affect the rights or obligations under this Convention of that State or of any of its constituent subdivisions or agencies or of any national of that State arising out of consent to the jurisdiction of the Centre given by one of them before such notice was received by the depositary. 101 Inversión y Gestión de Bienes v Kingdom of Spain, Decision on Jurisdiction (21 June 2013) ICSID Case No ARB/12/17, paras 58–68; Venoklim Holding BV v Bolivarian Republic of Venezuela, Award (3 April 2015) ICSID Case No ARB/12/22, paras 61–68; Blue Bank International & Trust (Barbados) Ltd v Bolivarian Republic of Venezuela, Award (26 April 2017) ICSID Case No ARB/12/20, paras 108–120. 97 ibid

Landmark Investment Cases on State Consent 273 after receipt of the notice, ie on 25 July 2012. In Fábrica de Vidrios v Venezuela,102 the claimants sought to accept the offer of consent to ICSID’s jurisdiction contained in the NetherlandsVenezuela BIT on 20 July 2012, just under six months after Venezuela’s notice of denunciation. The Tribunal found that the investor’s acceptance of the offer of consent after the date of the notice of denunciation was too late. The consent that is preserved by Article 72 is not the offer of consent in the BIT but mutual consent by the host state and the investor. The Tribunal reached this result with the help of the last sentence of Article 25(1) of the ICSID Convention: ‘When the parties have given their consent, no party may withdraw its consent unilaterally’. The Tribunal said: The phrase ‘consent in writing to submit to the Centre’ is equivalent in meaning to an arbitration agreement and thus perfected consent. … Whilst the phrase ‘no party may withdraw its consent unilaterally’ relates to the possible conduct of one party only, the preceding phrase ‘[w]hen the parties have given their consent’ makes it clear that ‘consent’ in the final phrase is not directed to the idea of unilateral consent that arises where a Contracting State has given its consent to ICSID arbitration in an investment treaty or domestic legislation. In other words, it is not being used to describe the legal situation created by a unilateral engagement of a Contracting State to submit to ICSID arbitration (but before that engagement is relied upon by a national of another Contracting State). The last sentence of Article 25(1) simply means that where there is perfected consent, it cannot be undone by the conduct of one of the parties.103

102 Fábrica de Vidrios Los Andeles, CA and Owens-Illinois de Venezuela, CA v Bolivarian Republic of Venezuela, Award (13 November 2017) ICSID Case No ARB/12/21. 103 ibid para 277.

17 Counterclaims ANDREA K. BJORKLUND*

I. COUNTERCLAIMS IN INVESTMENT LAW – INTRODUCTION

C

OUNTERCLAIMS ARE RARE in investment arbitration, though they are becoming more common. Most investment arbitration is now based on investment treaties, rather than on arbitration clauses in concession contracts or on the investment laws of the host state to an investment. Arbitral tribunals can only hear claims if the parties to the arbitration have consented to give the arbitral tribunal the authority to hear the case. The same holds true for counterclaims; the consent to arbitration must encompass consent to both claim and counterclaim. In the case of contract-based investment arbitration, counterclaims are relatively uncontroversial. An arbitration can be commenced by either the foreign investor or the host state, and either can submit a counterclaim. Arbitrations brought under investment treaties raise more complex issues. They are procedurally asymmetrical because only one party can (usually) commence the arbitration. As will be discussed more fully below, whether the investor’s commencement of arbitration can be viewed as automatically including an agreement to submit to counterclaims when reference to them is found in the applicable arbitration rules is not settled. Increasing use of counterclaims is likely inevitable in investment arbitration, but this deceptively simple solution to problems, both real and perceived, regarding the possibility of holding investors to account for their actions in the confines of an investment arbitration raises many questions yet to be answered. Perhaps the most important is that of applicable law. Which laws are relevant in the adjudication of a counterclaim? What effect does governing law have on the desirable characteristics of an adjudicator? For example, an investment tribunal hearing a counterclaim might need expertise in domestic contract law and administrative laws, and/or expertise in international contract law and international human rights law. Procedural questions can also arise. If the basis for the counterclaim is a contract between the investor and the host state or a state entity, parallel proceedings between the same or related entities might be ongoing. Does one tribunal have priority over the others? If so, which one? Is there a case in a municipal court that raises related or duplicative claims? The possibility of counterclaims in investment treaty arbitration burst onto the scene with the decision in Saluka v Czech Republic.1 While the Tribunal in that case disallowed the * Andrea K Bjorklund is a Full Professor and the L Yves Fortier Chair in International Arbitration and International Commercial Law at McGill University. 1 Saluka Investments BV v Czech Republic, Decision on Jurisdiction over the Czech Republic’s Counterclaim (7 May 2004) UNCITRAL (Arbitrators: Sir A Watts, LY Fortier and P Behrens).

276 Andrea K. Bjorklund counterclaim, the case nicely highlighted several of the potential problems associated with the raising of counterclaims in the Bilateral Investment Treaty (BIT) context. A few years later came the decision in Roussalis v Romania, along with the influential dissent in the case, written by Professor Michael Reisman.2 These two cases in particularly highlighted the issue of whether parties had consented to the submission of counterclaims to the Tribunal, thereby conferring on it jurisdiction over them, and where that consent could be found. A related issue, sometimes referred to as one of admissibility, is whether the alleged counterclaim has a sufficiently close connection to the main claim to be considered by the Tribunal. If a tribunal is going to permit a state to assert a counterclaim, the question then arises as to what the basis for that claim – municipal law or international law (or both) – is likely to be. Urbaser v Argentina3 explored the extent of an investor’s obligations under international law, and in particular under international human rights law. These cases, along with selected others, illustrate the myriad procedural and policy considerations raised by the introduction of counterclaims to the investment treaty arbitration universe, including the possibility that filing counterclaims can enhance the legitimacy of investor-state dispute settlement and contribute to its rule-of-law bona fides.

II. TRIBUNAL AUTHORITY TO APPLY COUNTERCLAIMS

Arbitral tribunals only have the authority given to them by the parties to the arbitration, along with what authority is given them by the legal superstructure that supports arbitration.4 In the counterclaims context, this means that a tribunal can only hear a counterclaim if the parties to the arbitration have consented that they do so.

A. Jurisdiction – Party Consent to Hearing Counterclaims In the case of contract-based investment arbitration, counterclaims are fairly straightforward, given that any agreement to arbitrate disputes under the contract may be initiated by either party to the contract, with each party capable of submitting a counterclaim in its turn. The basis for both claim and counterclaim is likely to be contract law. The parties understand that their arbitration agreement can be invoked by either, and that the other party can respond in kind. Investment treaty arbitrations are different. Ordinarily only one party – the investor – can commence the arbitration.5 Whether an investor’s acceptance of a state’s offer to arbitrate can be viewed as encompassing an acceptance of the state’s ability to submit a counterclaim is more complicated. Further questions arise regarding applicable law; given that most investment treaties impose no obligations on investors, the legal basis for the counterclaim is most

2 Roussalis v Romania, Award (7 December 2011) ICSID Case No ARB/06/1 (Arbitrators: B Hanotiau, A Giardina and M Reisman); Roussalis v Romania, Declaration of W Michael Reisman (28 November 2011) ICSID Case No ARB/06/01. 3 Urbaser et al v The Argentine Republic, Award (8 December 2016) ICSID Case No ARB/07/26 (Arbitrators: A Bucher, PJ Martínez-Fraga and C McLachlan). 4 See AK Bjorklund and L Vanhonnaeker, ‘The Powers, Duties, and Rights of International Arbitrators’ in S Kröll, AK Bjorklund and F Ferrari (eds), Cambridge Compendium of International Commercial and Investment Arbitration (Cambridge, Cambridge University Press, forthcoming 2022); J Waincymer, Procedure and evidence in international arbitration (Alphen aan den Rijn, Kluwer, 2012) 26–30. 5 J Paulsson, ‘Arbitration Without Privity’ (1995) 10 ICSID Review – Foreign Investment Law Journal 232, 233–34.

Counterclaims

277

likely to be municipal law. While an investor might be required either implicitly or explicitly to abide by host state laws, the treaty rarely provides a clear basis permitting the state to challenge on substantive grounds the investor’s failure to abide by those laws, and rarely is there a provision in the treaty permitting a state to challenge an investor for its failure to abide by those obligations, or to submit any counterclaim at all.6 The scope of international investment agreements (IIAs) varies, but many limit the jurisdiction of tribunals to hearing claims that certain provisions of the treaty were violated and some explicitly permit only challenges to state measures. Submitting a counterclaim based on international law is also fraught with difficulty, as will be explored more thoroughly in the discussion of Urbaser v Argentina case, below. The question of counterclaims emerged as a topic of interest to the investment arbitration community with the 2004 decision in Saluka v Czech Republic.7 This was the first case in which the question of consent to the counterclaim was squarely before a tribunal in a treaty arbitration. It was followed in 2011 by Roussalis v Romania, a decision featuring both an influential majority decision and a strong dissent by Professor Michael Reisman.8 Together these cases highlight the main challenges tribunals are confronted with when asked whether they have the authority to hear a counterclaim or, to put it differently, whether the investor has consented to have a counterclaim filed against it. Saluka was notable because the Tribunal made clear that the scope of the treaty, which reached any disputes relating to investments, permitted counterclaims in principle, so long as they were closely affiliated with the main claim.9 The counterclaim in this case was not permitted to go forward however, because the specific agreement between the Czech Republic and Nomura (Saluka’s subsidiary) anticipated arbitration under different rules in the event of a dispute,10 thereby negating any suggestion that the investor had consented to the treaty Tribunal’s jurisdiction to hear the counterclaim based on the contract. Moreover, the counterclaim was not, in the Tribunal’s view, closely connected to the allegations regarding breaches of the investment protection rules, thereby falling outside the Tribunal’s jurisdiction.11 In Roussalis, on the other hand, the treaty was more limited in its scope; the dispute resolution clause referred to disputes concerning an obligation of the state.12 The majority thus dismissed the counterclaim because of the limited jurisdictional scope of the treaty. The Roussalis dissent would have grounded consent in the applicable arbitration rules (the International Centre for Settlement of Investment Disputes (ICSID) rules, in this case), which specifically refer to counterclaims. This line of logic is that by invoking the jurisdiction of an ICSID Tribunal the investor agrees to counterclaims given the inclusion of counterclaims in ICSID Convention Article 46 and ICSID Arbitration Rule 40. These two cases thus illustrate the first question a tribunal must face, which is whether consent to hear counterclaims must be found in the investment treaty itself, or whether consent can be implied from procedural provisions referring to counterclaims in the applicable

6 A Vohryzek-Griest, ‘State Counterclaims in Investor-State Disputes: A History of 30 Years of Failure’ (2009) 15 International Law: Revista Colombiana de Derecho Internacional 83, 111–14. 7 Saluka, Decision on Jurisdiction, above (n 1). 8 Roussalis, Award, above (n 2); Roussalis, Declaration of W Michael Reisman, above (n 2). 9 Saluka, Decision on Jurisdiction, above (n 1) para 39. The tribunal in Paushok v Government of Mongolia came to the same conclusion. Paushok v Government of Mongolia, Award on Jurisdiction and Liability (28 April 2011) UNCITRAL, paras 684–99. 10 Saluka, Decision on Jurisdiction, above (n 1) paras 56–58. 11 ibid paras 76–80; cf Perenco Ecuador, Ltd. v Republic of Ecuador, Decision on Jurisdiction (30 June 2011) ICSID Case No ARB/08/6, para 71. 12 Roussalis, Award, above (n 2) paras 869–77.

278 Andrea K. Bjorklund arbitral rules.13 The Roussalis majority leads the line of cases that adopt the former view, whereas the dissent exemplifies the latter, which suggests that in appropriate cases a claimant’s consent can be inferred from the applicable rules and their references to counterclaims, so long as the counterclaims have a close association with the allegations of treaty breach and no other barriers come into play. Saluka arguably straddles these two; the treaty language in Saluka was clear, yet the Tribunal’s reasoning relied on both the treaty and on the United Nations Commission on International Trade Law (UNCITRAL) Rules, which were applicable in that case.14 Most treaties do not specifically give the Tribunal authority over counterclaims, though a few newer treaties have done so.15 This means that consent can be found only implicitly in the treaty itself. It is often inferred from the treaty’s language regarding the scope of investment arbitration. Some investment arbitration provisions confer on tribunals the authority to hear any ‘dispute between an investor of one Contracting Party and the other Contracting Party in connection with an investment’, or another similar formulation. These broad clauses can more plausibly encompass counterclaims as compared to narrow consent clauses specifying that an investor can submit to arbitration a claim for a breach of a specific treaty provision. For example, under NAFTA Article 1116 investors of a party are entitled to ‘submit to arbitration under this Section a claim that another Party has breached an obligation under (a) Section A or Article 1503(2) (State Enterprises), or (b) Article 1502(3)(a) (Monopolies and State Enterprises) …’.16 Because neither Section A, nor Article 1503(2), nor Article 1502(3)(a) imposes any obligations on investors, it is difficult to argue that a state can make a claim against the investor. This type of scope clause is why the Tribunal in Roussalis v Romania refused to entertain counterclaims against the foreign investor.17 A counterclaim must be grounded in law, which means that applicable law clauses might also be relevant to decisions about the availability (or lack thereof) of counterclaims. Some treaties contain no applicable law clause, some direct tribunals to apply the treaty itself (and often refer in addition to other relevant rules of international law), still others designate the domestic law of the host state as one of the sources of applicable law, sometimes alongside international law. When the applicable law clause is limited to international law, it is less clear that a state can submit claims against the investor as the investor is less likely to have obligations under the treaty or international law, though this could be changing. Some states have made creative arguments that investors do have obligations grounded in custom and general principles,18 and at least one tribunal has been receptive to these arguments.19 13 ibid. A roughly contemporaneous case, Goetz v Burundi, involved the Belgium-Burundi BIT, which had a slightly broader clause and also an applicable law clause referring to municipal law as well as international law. Convention entre l’Union Economique Belgo-Luxembourgeoise et la République de Burundi concernant l’Encouragement et la Protection Réciproques des Investissements, 13 April 1989, Arts 8(1) and 8(5), https://perma.cc/THB4-Z33W. Antoine Goetz et al v Burundi, Sentence (21 June 2012) ICSID Case No ARB/01/2. 14 Saluka, Decision on Jurisdiction, above (n 1) para 39. The tribunal in Paushok v Government of Mongolia came to the same conclusion. Paushok, Award, above (n 9) paras 684–99. 15 For example, the Common Market for Eastern and Southern Africa (COMESA) Investment Agreement provides in Art 28(9): ‘A Member State against whom a claim is brought by a COMESA investor under this Article may assert as a defence, counterclaim, right of set off or other similar claim, that the COMESA investor bringing the claim has not fulfilled its obligations under this Agreement, including the obligations to comply with all applicable domestic measures or that it has not taken all reasonable steps to mitigate possible damages.’ Investment Agreement for the COMESA Common Investment Area, 22-23 May 2007, Art 28(9), https://perma.cc/7CXG-XFL. 16 North American Free Trade Agreement (adopted 17 December 1992 entry into force 1 January 1994) 32 ILM 605, 642–43 Art 1116. 17 Roussalis, Award, above (n 2) paras 871–76. 18 Y Kryvoi, ‘Counterclaims in Investor-State Arbitration’ (2012) 21 The Minnesota Journal of International Law 216, 234–36. 19 Urbaser, Award, above (n 3) paras 1193–1199.

Counterclaims

279

Applicable law clauses that refer to domestic law more clearly support counterclaims grounded in domestic law. Even without that reference it is possible to argue that the Tribunal’s authority to apply international law includes its ability to apply international conflict-of-laws rules to select the appropriate law, and that in most cases the law applicable to the conduct of the investor would be host state law.20 The result of this argument is that the Tribunal would have the authority to apply the host state’s domestic law even if the investor’s duty to comply with domestic law is not mentioned in the treaty. As a matter of fundamental due process, it is not problematic to apply host state law to an investor’s conduct: the investor must assume that the laws of the host state apply to its business conduct in that state.21 Yet that does not mean the investor has consented to arbitrate disputes arising under those laws. It is even harder to find the source of that consent if the treaty itself refers only to claims under international law. Because Tribunal authority is derived from consent it is not surprising that tribunals are loath to extend their authority absent textual support. Investment treaty arbitrations are conducted under arbitration rules selected by the disputing parties. Ordinarily the investor chooses to commence arbitration under one of a number of options set out in an investment treaty. Investors most commonly seek arbitration pursuant to the ICSID Convention and its accompanying arbitral rules or pursuant to the UNCITRAL Arbitration Rules. It is thus not surprising that one possible source of a tribunal’s authority to hear counterclaims is the arbitration’s procedural rules. Article 46 of the ICSID Convention refers to counterclaims: Except as the parties otherwise agree, the Tribunal shall, if requested by a party, determine any incidental or additional claims or counter-claims arising directly out of the subject-matter of the dispute provided that they are within the scope of the consent of the parties and are otherwise within the jurisdiction of the Centre.

One could thus argue that an investor’s submission of a claim to ICSID Convention arbitration necessarily encompasses this provision and permits the filing of counterclaims by a host state. Under this line of argument, Article 46 would be considered an integral part of the treatybased consent to arbitration, such that any submission of a claim to arbitration under the ICSID Convention would be deemed to permit a counterclaim. This is the argument proffered by Professor Reisman in his Roussalis dissent.22 A similar argument can be made with respect to the UNCITRAL Rules, which also refer to counterclaims. The UNCITRAL Arbitration Rules of 1976 provided that a respondent may ‘make a counter-claim arising out of the same contract or rely on a claim arising out of the same contract for the purpose of a set-off ’.23 The UNCITRAL Arbitration Rules were revised in 2010 to provide that a tribunal can hear counterclaims provided the tribunal has jurisdiction over them.24

20 See, eg, M Sasson, Substantive Law in Investment Treaty Arbitration: The Unsettled Relationship Between International Law and Municipal Law (Alphen aan den Rijn, Kluwer, 2010) 206–208. 21 If an investor has no reason to expect that a particular law applies to it then the investor might reasonably object to being held responsible for a breach of that law. 22 Roussalis, Declaration of W Michael Reisman, above (n 2). 23 UNCITRAL Arbitration Rules (1976) Art 19.3. 24 UNCITRAL Arbitration Rules (2010) Art 21.3 (‘In its statement of defense, or at a later stage in the arbitral proceedings if the arbitral tribunal decides that the delay was justified under the circumstances, the respondent may make a counterclaim or rely on a claim for the purpose of a set-off provided that the arbitral tribunal has jurisdiction over it.’)

280 Andrea K. Bjorklund However appealing this argument might be on the surface, it disregards the language found in both the ICSID Convention and the 2010 UNCITRAL rules that counterclaims are permitted only if they are within the scope of the consent of the parties – a question that has to be answered by reference to the consent instrument (the treaty) itself.25 As Jean Kalicki has noted with respect to the ICSID Convention: ‘if Article 46 itself provided that consent, then its incorporated requirement of consent (“provided they are within the scope of consent”) would be entirely circular and extraneous’.26 The same remark also holds true for the UNCITRAL rules. The reference to the possibility of counterclaims in the ICSID Convention is not surprising if one keeps in mind that most cases brought to the Centre would be based on contracts, and not treaties. The applicable law and the matrix of facts for both the contract claim and the counterclaim would be the same, and the consent to arbitration in the contract would almost always be bilateral. The UNCITRAL Rules and those of other arbitral bodies are used in commercial as well as in investor-state arbitrations; given that commercial arbitrations are nearly always grounded in contract it is not surprising that they refer to counterclaims. Absent some indication in the treaty itself however, the mere fact that arbitration rules refer to counterclaims is an insufficient basis on which to ground an investor’s consent to counterclaims. The main concerns regarding counterclaims relate to consent and the scope of a tribunal’s jurisdiction. The answer thus relates to the proper interpretation of the consent document – the investment treaty or contract itself. Indeed, the revisions to the UNCITRAL Rules strongly suggest that the rules themselves are not a source of consent to the filing of counterclaims, but that the consent must be found elsewhere.

B. Admissibility Requirement of a ‘Close’ Connection The Saluka Tribunal rejected the counterclaim in part because it was not closely connected to the allegations regarding breaches of the investment protection rules, and thus fell outside the Tribunal’s jurisdiction.27 The ‘close connection’ requirement has sometimes been described as a general principle of law.28 It can also be identified, albeit in different language, in Article 46 of the ICSID Convention: ‘the Tribunal shall, if requested by a party, determine any incidental or additional claims or counter-claims arising directly out of the subject-matter of the dispute …’. The UNCITRAL rules contain no direct textual reference, yet the requirement has generally been regarded as implicit in that context as well. It is also frequently referred to as a matter of admissibility, rather than of jurisdiction.29 The existence of a close connection requirement is not in itself controversial. It is also found in the Statute of the International Court of Justice30 (ICJ) and thus in ICJ jurisprudence.31

25 A

Parra, The History of ICSID (Oxford, Oxford University Press, 2012) 24–25; 132–35. Kalicki, ‘Counterclaims by States in Investment Arbitration’ (14 January 2013) Investment Treaty News, https://perma.cc/UH7G-DU3J. 27 Saluka, Award, above (n 1) paras 76–80; cf Perenco, Decision on Jurisdiction, above (n 11) para 71; Paushok, Award, above (n 9) para 694. 28 Goetz, Sentence, above (n 13). 29 See, eg, T Kendra, ‘State Counterclaims in Investment Arbitration – A New Lease of Life?’ (2013) 29 Arbitration International 575, 592–93; 595–97. 30 Statute of the International Court of Justice, 33 UNTS 993, Art 80. 31 See generally S Murphy, ‘Counter-Claims at the International Court of Justice’ in A Zimmerman and CJ Tams (eds), The Statute of the International Court of Justice – A Commentary, 3rd edn (Oxford, Oxford University Press, 2019). 26 J

Counterclaims

281

The need for this requirement might be seen most clearly when one thinks about the panoply of laws to which an investor is subject under the municipal law of the host state; it could not, and should not, be the case that a state could assert any and every claim against an investor if an investor commences an arbitration under an investment treaty. As is often the case in linedrawing exercises, however, it is difficult to know exactly whether the close connection needs to be legal or merely factual, and even within those two categories just how closely linked the claims or the facts underpinning the claims should be. As Arnaud de Nanteuil has pointed out, in many cases the distinction is more theoretical than actual, given that the legal and factual bases for the claims will be intertwined.32 This linkage might explain why contractually based counterclaims in the context of a concession contract between the investor and the host state seem to raise fewer concerns; there is inevitably both a factual and legal nexus. The Saluka Tribunal took a rather narrow view of the type of connection required, suggesting that the claim must arise from the same legal instrument.33 This requirement would preclude counterclaims in most treaty-based arbitrations, given that very few agreements impose obligations on investors. Even treaties with umbrella clauses impose obligations on states, rather than on investors themselves.34 The Organization of the Islamic Conference (OIC) Investment Agreement at issue in Al Warraq v Indonesia is the leading example of an investment agreement that not only imposes obligations on investors but also specifies that the state may commence claims.35 Most treaties do neither. On the Saluka reading, even a treaty’s broad scope clause (all disputes relating to investments) could not serve as the basis for consent to counterclaims. Post-Saluka, it seems to have become more generally accepted that the close connection required be based primarily on the factual basis for the claims, so long as there is ascertainable consent to the authority of the Tribunal over the counterclaim that also encompasses the Tribunal’s ability to apply the relevant applicable law.36 Thus a counterclaim based on municipal law can go forward so long as there is a broad scope clause and a suitable applicable law clause.

III. INTERSECTION WITH APPLICABLE LAW

The preceding discussion presented the role that applicable law clauses might play in ascertaining whether a treaty permits a tribunal to hear counterclaims. More directly however, the question of applicable law is important because it affects the breadth of a tribunal’s power and could (or should) affect the qualifications desired in the members of an arbitral tribunal.

32 A de Nanteuil, ‘Counterclaims in Investment Arbitration: Old Questions, New Answers?’ (2018) 17(2) The Law and Practice of International Courts and Tribunals 375, 387. 33 Saluka, above (n 1) paras 61, 76. 34 AK Hoffmann, ‘Counterclaims by the Respondent State in Investment Arbitrations – The Decision on Jurisdiction Over Respondent’s Counterclaim in Saluka Investments B.V. v Czech Republic’ (2006) 3(5) Transnational Dispute Management. 35 Al Warraq v Indonesia, Final Award (15 December 2014) UNCITRAL (Arbitrators: BM Cremades, M Hwang, FS Nariman). 36 This has not inevitably been the case. In Gavazzi v Romania, the tribunal rejected Romania’s counterclaim on the ground that it was based on Romanian law and did not directly derive from breaches of the investment treaty. Marco Gavazzi and Stefano Gavazzi v Romania, Decision on Jurisdiction, Admissibility and Liability (21 April 2015) ICSID Case No ARB/12/25 (Arbitrators: H van Houtte, VV Veeder QC and M Rubino-Sammartano) para 161. The tribunal also held that ‘[b]oth in form and substance, it is a free-standing counterclaim not operating merely as a defence to the Claimants’ claim.’; ibid para 151.

282 Andrea K. Bjorklund In a relatively traditional case involving a concession contract between an investor and a host state, a counterclaim might well be based on the contract, and on a failure to abide by the terms of that contract. In Occidental v Ecuador for example, there did not appear to be consent to the Tribunal’s consideration of counterclaims, yet there was no discussion of the Tribunal’s jurisdictional authority as the claims were readily dismissed on the merits.37 In addition to contractual breaches, an investor’s activity might fall foul of the municipal public law of a host state. Two cases against Ecuador for example, raised questions of the investors’ violation of Ecuador’s environmental laws. In Perenco v Ecuador the claimant won a US$449 million award against Ecuador, but was ordered to pay US$54 million towards environmental clean-up.38 Another potential basis for a counterclaim is international human rights law. Whether an investor (as a non-state entity) could have such obligations was squarely before the Tribunal in Urbaser v Argentina. Urbaser is one of many cases arising from Argentina’s privatisation of its water and sewerage services. In 1999, an Argentine company in which Urbaser owned shares, Aguas Del Gran Buenos Aires S.A. (‘AGBA’), signed a concession agreement with the Province of Buenos Aires ‘for the provision of the running water supply and sewage public services in region B (Zone 2) of the Province of Buenos Aires’.39 In the aftermath of the Argentine financial crisis, Argentina changed the legal framework that had hitherto governed the concession agreement, including the tariff that AGBA could charge and the exchange rate between the peso and the dollar. The claimants alleged violations of the Argentina-Spain BIT, including a denial of fair and equitable treatment and the unlawful expropriation of its investment. In its defence, Argentina said its actions had been taken to guarantee the residents’ right to water, and that even before the emergency measures were imposed AGBA had breached the concession contract by failing to make the required infrastructure investment, thereby violating the fundamental right of access to water of Buenos Aires citizens.40 The Argentina-Spain BIT has a broad scope; the Tribunal therefore accepted the possibility that Argentina could submit a counterclaim. The Tribunal also refused to dismiss the counterclaim solely on the basis that it was grounded in human rights law: ‘such argument is not sufficient to go so far as excluding on a simple prima facie basis any such claim as if it could not imply a dispute relating to an investment’.41 The Tribunal further suggested that because investors can hold rights under international law, they can be subject to international law obligations.42 Nonetheless the Tribunal held that the claimants did not have general obligations to provide access to water prior to the time they had entered the concession agreement and the agreement could not in and of itself create such obligations in international law.43 All obligations held by Urbaser were framed by the legal and regulatory environment found in the BIT and in host state laws.44

37 Occidental Petroleum Corp. v Republic of Ecuador, Award (5 October 2012) ICSID Case No ARB/06/11 (Arbitrators: LY Fortier, DAR Williams and B Stern) paras 854–69. The liability of the state was reduced due to the investor’s contributory negligence, which is discussed further in this chapter. 38 Perenco Ecuador Ltd. v Ecuador, Award (27 September 2019) ICSID Case No ARB/08/6 (Arbitrators: P Tomka, N Kaplan, JC Thomas) para 1015. 39 Urbaser, Award, above (n 3) para 63. 40 ibid para 1151. 41 ibid para 1154. 42 ibid para 1194. 43 ibid para 1213. 44 ibid para 1209.

Counterclaims

283

The Urbaser Tribunal thus accepted jurisdiction over the counterclaim but dismissed the case on the merits. Urbaser has generated significant commentary, much of it critical of the Tribunal’s failure to impose an obligation on the investors or to more clearly identify what types of obligations investors can have.45 Yet this is not dissimilar to the hesitation shown by other tribunals insofar as they have been asked to exercise jurisdiction over public law claims. For example, in Paushok v Mongolia, the Tribunal refused to exercise jurisdiction over one of the asserted counterclaims which would have been based on Mongolian tax laws, which the Tribunal found outside its authority due to the public law taboo on the extraterritorial enforcement of tax awards, as opposed to on contractual claims, which the Tribunal would have considered.46

IV. EFFECT OF INVESTOR CONDUCT ON LIABILITY AND ON REMEDIES

This chapter has addressed counterclaims construed narrowly – as the assertion of a legal claim that could give rise to a determination of breach that supports the award of damages for that breach. One should remember however, that the factual underpinnings of counterclaims might have an effect on either liability or on remedies, even if the state is not permitted to submit a counterclaim. These might even be regarded as hidden counterclaims. In the jurisdictional context for example, a state might argue that an investor did not lawfully make its investment. Whilst not a counterclaim as such, the assertion, if accepted and if the treaty denies coverage to investments not made in a lawful manner, would result in the dismissal of the claim. It is thus a fairly significant weapon in the arsenal of the state.47 As to the merits, states can highlight the deficiencies in investor behaviour, even absent the ability to submit a counterclaim, as factual bases to justify their own conduct. Also falling short of a fully-fledged counterclaim is the question of a set-off at the damages phase for injury caused by the investor. For example some cases have seen damages reduced for ‘contributory negligence’.48 In Occidental Petroleum v Ecuador, the Tribunal reduced the damages awarded to Occidental on such grounds. In neither of those examples do states seek affirmative relief from the investor, yet they can diminish some of their substantive liability or the amount of damages they owe.49

V. CERTAIN IMPLICATIONS RELATED TO COUNTERCLAIMS

Those Tribunals that have at least in principle approved the filing of certain counterclaims have recognised that bringing related claims together into the same case has several benefits. These

45 See, eg, E de Brabendère, ‘Human Rights and International Investment Law’ (2018) 75 Grotius Centre Working Paper Series; E Guntrip, ‘Urbaser v Argentina: The Origins of a Host State Human Rights Counterclaim in Investment Arbitration’ (10 February 2017) EJIL:Talk!, https://perma.cc/K3MP-FHAK. 46 Paushok, Award, above (n 9) paras 694–95. 47 For a good discussion of the assertion of counterclaims in different contexts, and in particular as an objection to jurisdiction, see JA Rivas, ‘ICSID Treaty Counterclaims: Case Law and Treaty Evolution’ in JE Kalicki and A JoubinBret (eds), Reshaping the investor-state dispute settlement system: Journeys for the 21st century (Leiden, Brill Nijhoff, 2015) 780, 810. 48 Occidental, Award, above (n 37) paras 670, 678–80. 49 For discussion on contributory negligence cases, see J-M Marcoux and AK Bjorklund, ‘Foreign Investors’ Responsibilities and Contributory Fault in Investment Arbitration’ (2020) 69 International & Comparative Law Quarterly 877.

284 Andrea K. Bjorklund benefits are not always without drawbacks, or at least hesitations. They also give rise to a host of questions for which investment law does not currently have an answer. For the sake of ease I have categorised these into two groups; there could well be some overlap between them.

A. Procedural Observations In the simplest sense, procedural efficiency would presumably be enhanced by permitting governments to submit counterclaims. Complex facts developed through the course of arbitrating the treaty-based claims against the state would likely be relevant for the claims against the investor as well, even if they are based on municipal law. A ‘one-stop-shop’ for all claims related to the same cluster of events would encourage efficient decision-making with the arbitrators well informed about all relevant facts. It would also facilitate the assessment of damages and the calculation of countervailing damages. Having one set of counsel and one hearing would also streamline the process and minimise the duplication of expenses.50 Yet the idealised picture of international arbitration as a speedy and efficient way to resolve international disputes has largely been debunked in both commercial and investment arbitration.51 Enlarging a dispute to include counterclaims will conceivably accelerate this tendency. Permitting counterclaims would also likely raise considerations related to the intersection between domestic and international dispute settlement. Ideally, treaty provisions such as ‘fork-in-the-road’ clauses52 would accommodate these concerns by requiring any actions in the local court to cease once an investor-state dispute claim is filed, if the investor-state claim is going to include claims that would replicate local procedures. The ‘no-u-turn’ approach adopted by NAFTA and its progeny might be more beneficial, in order to minimise duplication between international and domestic proceedings. To give an example, if an investor has filed a municipal-law case against the state in domestic courts, and the state has filed a counterclaim there, can the state file effectively the same counterclaim before the international Tribunal? What if the investor files a counter-counterclaim in the international Tribunal? Would the investor be regarded as having already raised the claim in domestic courts, and thus be barred from filing it in the international procedure? Would the state be deemed to have acquiesced (or would it be estopped) by virtue of having filed the claim first? Would the domestic court be willing to give way to the international Tribunal, in order to avoid duplication of processes? At least for the first several years answering questions such as these would likely occupy many courts and tribunals. One possible solution, adopted in the Burlington Resources case, is for the parties to come to an agreement that the Tribunal can hear the counterclaims, but that solution will only work when the parties agree.53 Effectively both the investor and the state could probably agree to stop any local proceedings and shift all matters to international arbitration (though certain courts might have different ideas.). Yet the parties might not be willing to do this for

50 Jean Kalicki has noted that allowing counterclaims ‘may lead to efficiency, to the centralization of inquiry and the avoidance of duplication’. Kalicki, above (n 26) 1. 51 See, eg, JY Gotanda, ‘An Efficient Method for Determining Jurisdiction in International Arbitrations’ (2001) 40 Columbia Journal of Transnational Law 11, 12–13. 52 For more information on fork-in-the-road clauses, see L Henry, ‘Investment Agreement Claims Under the 2004 Model U.S. BIT: A Challenge for State Police Powers?’ (2010) 31 University of Pennsylvania Journal of International Law 935, 985. 53 See, eg, Burlington Resources Inc v Ecuador, Decision on Liability (14 December 2012) ICSID Case No ARB/08/5, paras 93, 174.

Counterclaims

285

various reasons. Strategically an investor might be trying to avoid a forum in which an award against it is more readily enforceable. The state might seek primary remedies available in local courts that are not readily available in international arbitral tribunals. What is best for either party, or for both parties, is likely to be highly fact-specific and potentially to involve strategic calculations about where each party’s advantage lies. While one might think that states would be reluctant to see their own courts displaced by investment arbitration tribunals, this will not always be the case. In some circumstances states might see an advantage, particularly at the stage of enforcement, if they possess an arbitral award, particularly if the bulk of an investor’s assets are held outside the host state. Awards rendered by international arbitral tribunals are subject to enforcement under the applicable arbitration treaties. ICSID Convention awards are enforceable under the ICSID Convention;54 virtually all other investor-state arbitration awards are enforceable under the New York Convention on Recognition and Enforcement of Arbitral Awards.55 Enforceable awards include those rendered against investors. Because arbitral awards are more readily enforceable than court judgments in most cases,56 states might benefit from obtaining arbitral awards against investors in much the same way that investors benefit from obtaining awards against host states. Permitting counterclaims might well discourage frivolous claims; it might also discourage states from raising frivolous objections.57 If an investor has to take account of claims that might be filed against it, its zeal to file might diminish.58 The possibility of facing counterclaims might also have an effect on third-party funding decisions, as funders would have to assess the likelihood of affirmative liability in addition to the likelihood of success on the merits in the case against the opposing party. In short, the whole cost-benefit calculus would shift. The same is true for a respondent as well; if claims can be submitted only one way there is every incentive for a state to file jurisdictional objections to delay the merits. Should the state be able to recover against the claimant, its willingness to move forward, at least when it has viable claims, would grow. Other procedural problems involve issues such as privity of contract. Against whom can the counterclaim be filed? Against the investor submitting the claim under the treaty, or against the investor’s investment, or both? The claimant is usually the investor submitting a claim on behalf of its investment, yet the counterclaim would often be submitted against the investment as the entity doing business in the host state’s jurisdiction and thus subject to its laws; that is often the entity with whom any concession contract is made as well. In such cases can the counterclaim effectively be submitted against the investor, and would the investor be liable to pay damages?59

54 Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (18 March 1965), 17 UST 1270, 1286, 575 UNTS 159, Arts 46, 53. 55 Convention on the Recognition and Enforcement of Foreign Arbitral Awards (10 June 1958) 21 UST 2517, 2519, 330 UNTS 38 Art 1. 56 See RD Bishop, ‘Introduction: The Enforcement of Arbitral Awards Against Sovereigns’ in RD Bishop (ed), Enforcement of Arbitral Awards Against Sovereigns (Huntington, Juris, 2009) 3–4. 57 See Y Kryvoi, ‘Counterclaims in Investor-State Arbitration’ (2011) 8 LSE Law, Society and Economy Working Papers 4, http://ssrn.com/abstract=1891935. 58 MW Friedman and IC Popova, ‘Can State Counterclaims Salvage Investment Arbitration?’ (2014) 8(2) World Arbitration & Mediation Review 143, 151. 59 Sometimes it might be possible for a foreign-controlled investment itself to submit a claim (and Art 25(2)(b) of the ICSID Convention permits jurisdiction under the Convention to extend to disputes between an investment of a host state that is under foreign control). In those cases the privity issue would presumably not exist. See generally Kryvoi, above (n 18) 230–34.

286 Andrea K. Bjorklund One might have assignment and delegation questions on the submission side as well. Can the federal government submit a counterclaim on behalf of all subnational governmental units, or on behalf of all government agencies? Attribution rules govern when a state can be liable for the acts of its constituents, yet they do not govern offensive uses of authority.60 What law would govern that question? International law? Municipal laws governing the relationship between different branches of government and between state and local governments?

B. Legitimacy-related Observations Permitting governments to file counterclaims redresses concerns about the asymmetry of investor-state dispute settlement in which investors have both rights and remedies, whereas states have neither.61 Though one can readily argue that ISDS redresses the asymmetry that exists in its absence – a situation where states have sovereign power to act as they see fit and investors are often without redress due to poorly functioning, non-independent courts or state immunity rules or the like – it remains true that the spectacle of investors seeking multimillion or even billion dollar awards against host states with no reciprocal claim expected from the state raises questions about procedural and substantive fairness.62 Moreover states may want the opportunity to defend themselves and their reputations vigorously – to win, as opposed merely to ‘not lose’. Yet balance can be over-adjusted. The beginning of this chapter noted the fact that some treaties limit the types of claims that investors can make against states to very specific investment treaty breaches. NAFTA Chapter 11 is one such treaty. Professor Douglas makes the interesting point that extending a tribunal’s jurisdiction to encompass counterclaims in a treaty such as NAFTA Chapter 11 would result in an inequality whereby a host state could submit counterclaims ‘based upon a contractual obligation (if there is an investment agreement in place between the investor and the host state), a tort, unjust enrichment, or a public law act, in circumstances where the investor’s primary claims are limited to breaches of Chapter 11 obligations.’63 Counterclaims, especially if they were contractually based, might also shift the orientation of the proceedings towards private-law issues. Many have written about the administrative law, or even the constitutional law, nature of investment tribunals.64 Treaty obligations

60 ‘Responsibility of States for Internationally Wrongful Acts’ (2001) 2(2) Yearbook of the International Law Commission, Arts 4–11; J Crawford, ‘Treaty and Contract in Investment Arbitration’ (2008) 24 Arbitration International 351, 363 (‘The second qualification is that there can only be contractual jurisdiction under a BIT in respect of an investment contract with the state itself, and not with a separate State entity having its own legal personality, and a fortiori with a third party. It is sometimes argued that the question is one of attribution under Chapter 2 of Part I of the ILC’s Articles on State Responsibility, but attribution has nothing to do with it. (…) The problem here concerns jurisdiction, not merits; the formation of a secondary agreement to arbitrate, not the breach of a primary obligation concerning the protection of investments.’). 61 See, eg, AK Bjorklund, ‘The Role of Counterclaims in Rebalancing Investment Law’ (2013) 17 Lewis & Clark Law Review 465; H Bubrowski, ‘Balancing IIA arbitration through the use of counterclaims’ in A de Mestral and C Lévesque (eds), Improving International Investment Agreements (Abingdon, Routledge, 2012) 212. 62 Bubrowski, above (n 61) 214–15. 63 Z Douglas, The International Law of Investment Claims (Cambridge, Cambridge University Press, 2019) 257. In such a case, he concludes that ‘it would be preferable to construe Chapter 11 of NAFTA as excluding the possibility of counterclaims by the host state respondent’. 64 See, eg, B Kingsbury and SW Schill, ‘Investor-State Arbitration as Governance: Fair and Equitable Treatment, Proportionality and the Emergency Global Administrative Law’ (2009) 9(46) NYU School of Law, Public Law & Legal Theory Research Paper, https://ssrn.com/abstract=1466980.

Counterclaims

287

are meant to constrain the activities of states, so the emphasis on state behaviour is not surprising. Contract-based counterclaims would enlarge those concerns to assess the conduct of the investor and of its investment. This shift might serve to highlight and constrain undesirable investor conduct, thus improving corporate governance. Permitting counterclaims is very likely to encourage the submission of counterclaims. In the first instance, this would likely aggrandise the power of ISDS tribunals, which would be hearing more aspects of complex, multilayered disputes.65 For those concerned about the mechanism of ISDS itself, entrusting tribunals with the authority to hear counterclaims would possibly cement their status as alternatives to domestic courts, particularly if one took an expansive view of counterclaims. To the extent they would apply domestic public law, they might be grappling with an unfamiliar law in which they can claim no special expertise66 and be yet one more step removed from the governmental process that enacted those laws and that might be able to act to correct any misapplication of the law. Permitting counterclaims makes it more likely that investors will be called to account for their actions. Though this can happen in the local courts in the place where the dispute is heard, those decisions sometimes lack force, whether moral or practical, given perceptions about bias in the judiciary. A recent example of this phenomenon is Chevron v Ecuador.67 Regardless of the merits of the case, what appear to be well-documented examples of procedural improprieties raise questions about the legitimacy of the US$18 billion domestic court award rendered against Chevron.68 Had an international tribunal rendered the judgment, there may well have been an outcry, yet the crux of the concerns would be different. Yet there are converse rule-of-law aspects as well. One of the criticisms of ISDS has been its siphoning off from domestic courts disputes that might otherwise have been resolved locally.69 Domestic courts then lose both the opportunity to exhibit their effectiveness and their ability to enhance their capacity to hear complex disputes.70 To the extent that ISDS continues to be an attractive venue, or becomes an even more attractive venue, this problem will continue.

VI. CONCLUSION

The ability of a state to submit counterclaims in investment arbitration can have both positive and negative elements. The lack of clear consensus about whether the pros outweigh the cons is one reason for the ambivalence tribunals have brought to the issue. On the positive side, counterclaims could help to ‘re-balance’ investment law by imposing obligations on investors as well as on states and lead to judicial efficiency by bringing all claims related to the investment claim under the purview of a single tribunal’s authority. Yet the law most likely to govern the investor’s conduct is that of the host state, which raises the question of just how much

65 Some do argue for the exercise of broad powers by arbitral tribunals. See, eg, A Asteriti, ‘Environmental Law in Investment Arbitration: Procedural Means of Incorporation’ (2015) 16(2) Journal of World Investment & Trade 248. 66 Kalicki, above (n 26) 2. 67 The facts of the case can be found in many places. See, eg, Chevron Corp. v Donziger, 768 F. Supp. 2d 581 (SDNY 2011). 68 CA Whytock and C Burke Robertson, ‘Forum Non Conveniens and the Enforcement of Foreign Judgments’ (2011) 111 Columbia Law Review 1444, 1447–48. 69 For an early example of this concern, see United Nations Conference on Trade and Development, Dispute Settlement: Investor-State, 8, 12, 33, UN Sales No E.03.II.D.5 (2003). 70 See WS Dodge, ‘National Courts and International Arbitration: Exhaustion of Remedies and Res Judicata under Chapter 11 of NAFTA’ (2000) 23 Hastings International and Comparative Law Review 357, 381–83.

288 Andrea K. Bjorklund authority an arbitral tribunal should have to apply different municipal laws, including public laws such as tax law and environmental law. Moving in a direction that favours counterclaims will take some state initiative, and states ought to carefully consider how they go about it. If a state wants to ensure that an ISDS tribunal has the authority to hear counterclaims, including an explicit provision in its IIA, along the lines of that in COMESA 2007, would obviate the need to make complex arguments about consent and would ensure a tribunal’s jurisdictional reach. Treaties should also address the privity issues between the investor and the investment and between the state and appropriate state entities. Ensuring that the tribunal is directed appropriately towards the applicable law would be helpful, too. Furthermore, explicitly limiting the counterclaims to matters based on a concession contract (assuming one is at issue) or that are otherwise related to the investment dispute itself would help to allocate authority between the investment tribunal and local courts in a way that reflects their respective areas of expertise.

18 The Applicable Law Saga FRANCK LATTY AND MARINA SIM*

I

NTERNATIONAL INVESTMENT DISPUTES are to be resolved on the basis of the applicable law,1 unless the arbitrators are (expressly) authorised to act as amiable compositeur or to decide the case ex aequo et bono.2 Identifying which law applies, or, in case of conflict, prevails, is of primary importance as it will give the dispute its general character. It may have an impact on the Tribunal’s jurisdiction; it may also determine the success of a particular claim or even the outcome of the whole dispute. Moreover, a failure by the arbitrators to apply the proper law constitutes a ground for annulment of the award under the International Centre for Settlement of Investment Disputes (ICSID) Convention (as manifest excess of powers, Article 52(1)(b))3 and may also lead to the setting aside of the award in non-ICSID proceedings. Notwithstanding the importance of the issue, the ambiguity regarding the applicable law in investment disputes persists and it remains a contentious topic. This ambiguity is exacerbated by the ad hoc and non-hierarchical nature of the system of arbitration and the fact that there is no doctrine of stare decisis in international investment arbitration.4 As a result, the practice

* Franck Latty is Professor of International Law at Paris Nanterre University; Marina Sim is a PhD candidate at Paris Nanterre University. 1 For a recent bibliography on the issue of the applicable law in investment agreements see C Titi, ‘Applicable Law in Investment Agreements’, Oxford Bibliographies, 27 March 2019. 2 See the ICSID Convention, Art 42(3); 2010 UNCITRAL Arbitration Rules, Art 35(2); ICSID Additional Facility Rules, Art 54(2); 2021 ICC Arbitration Rules, Art 21(3); 2020 LCIA Arbitration Rules, Art 22.4; 2017 SCC Arbitration Rules, Art 27(3). See also CME Czech Republic BV v The Czech Republic, Final Award (14 March 2003), para 403 (‘The basic mandate of the Treaty obligates the Tribunal to “decide on the basis of law”, which is a self-explanatory confirmation of the basic principle of law to be applied in international arbitration according to which the arbitral tribunal is not allowed to decide ex aequo et bono without authorization by the parties.’); CH Schreuer, ‘Decisions Ex Aequo et Bono Under the ICSID Convention’ (1996) 11(1) ICSID Review – Foreign Investment Law Journal 37. 3 Klöckner Industrie-Anlagen GmbH et al v United Republic of Cameroon and Société Camerounaise des Engrais, Decision on Annulment (3 May 1985) ICSID Case No ARB/81/2, paras 58–62. For a more extensive interpretation, see MTD Equity Sdn Bhd & MTD Chile SA v Chile, Decision on Annulment (21 March 2007) ICSID Case No ARB/01/7, para 47 (‘The Committee accepts that the notion of endeavouring to apply the law is not a merely subjective matter. An award will not escape annulment if the tribunal, while purporting to apply the relevant law actually applies another, quite different law. But in such a case the error must be “manifest”, not arguable, and a misapprehension (still less mere disagreement) as to the content of a particular rule is not enough.’). See also Sempra Energy International v Argentine Republic, Decision on the Argentine Republic’s Application for Annulment of the Award (29 June 2010) ICSID Case No ARB/02/16, para 164 (‘As a general proposition, this Committee would not wish totally to rule out the possibility that a manifest error of law may, in an exceptional situation, be of such egregious nature as to amount to a manifest excess of powers.’). 4 HE Kjos, Applicable Law in Investor-State Arbitration (Oxford, Oxford University Press, 2013) 9. See also SGS Société Générale de Surveillance SA v Republic of the Philippines, Decision on Jurisdiction (29 January 2004) ICSID Case No ARB/02/ 6, para 97.

292 Franck Latty and Marina Sim on the issue of applicable law ‘varies considerably’.5 At the same time, it is highly important to discern the existing trends in determining the applicable law in order to enhance legal certainty and predictability. The law applicable to investment disputes is reflective of the complex nature of investment arbitration. It presents an intricate interplay between various rules of law, most notably public international law and national law norms.6 Traditionally, international law treats municipal law as a fact ‘with reference to which rules of international law have to be applied, rather than as a rule to be applied on the international plane as a rule of law’.7 The particularity of foreign investments is that they are regulated by both international and national rules.8 This is explained by a diverse range of legal relationships that arises in an investment dispute, which necessitates the application of several different applicable laws by a tribunal.9 A complex approach to the applicable law is necessitated by the status of the claiming party and the private rights and interests that constitute the object of the international protection provided by investment treaties.10 It is also a result of the competing interests of the disputing parties: while the investor often invokes international law as a neutral legal system, the host state frequently advocates the application of its own national law in order to retain the highest possible degree of control over the investor or investment in question.11 On the other hand, a respondent state may seek arguments in general international law (principle of sovereignty, circumstances precluding wrongfulness, etc) or in specific branches of international law (eg international environmental law, international health law) for the arbitral tribunal to broaden its scope and not only to verify the strict observance of the investor’s treaty rights.12 In this respect, the different issues of a case may be governed by different laws and a dépeçage of the applicable law is also possible13 and occurs, eg in case of a combination of treaty-based and contract-based claims. The present chapter deals exclusively with the substantive rules applicable to the merits of investment disputes or lex causae (as opposed to the law applicable to procedural issues of investment disputes). It focuses on the applicable law in investment treaty arbitration.14

5 CH

Schreuer, ‘Investment Arbitration: A Voyage of Discovery’ (2005) 71 Arbitration 73, 75. eg, JW Salacuse, The Three Laws of International Investment. National, Contractual, and International Frameworks for Foreign Capital (Oxford, Oxford University Press, 2013). See also CH Schreuer, ‘International and Domestic Law in Investment Disputes: The Case of ICSID’ (1996) 1 Austrian Review of International and European Law 89, 89 (‘Investment relationships typically involve domestic law as well as international law. The host State’s domestic law regulates a multitude of technical questions such as admission, licensing, labour relations, tax, foreign exchange and real estate. International law is relevant for such questions as the international minimum standard for the treatment of aliens, protection of foreign owned property, especially against illegal expropriations, interpretation of treaties, especially bilateral investment treaties, State responsibility and, possibly, human rights.’). 7 R Jennings and A Watts (eds), Oppenheim’s International Law, Vol 1, 9th edn (London, Longman, 1996) 83. See also Certain German Interests in Polish Upper Silesia (Germany v Polish Republic), Merits (1926) PCIJ Rep Series A, No 7, 19. 8 JE Viñuales, ‘The Source of International Investment Law’ in S Besson and J d’Aspremont (eds), The Oxford Handbook on the Sources of International Law (Oxford, Oxford University Press, 2016) 1074. See also RD Bishop, J Crawford and WM Reisman (eds), Foreign Investment Disputes: Cases, Materials and Commentary, 2nd edn (Alphen aan den Rijn, Kluwer Law International, 2014) 437. 9 Z Douglas, The International Law of Investment Claims (Cambridge, Cambridge University Press, 2009) 40. 10 ibid 41. 11 Kjos, above (n 4) 7. 12 See, eg, Philip Morris Brands Sàrl, Philip Morris Products SA and Abal Hermanos SA v Oriental Republic of Uruguay, Award (8 July 2016) ICSID Case No ARB/10/7, paras 216 et seq. 13 See, eg, CH Schreuer and others, The ICSID Convention: A Commentary, 2nd edn (Cambridge, Cambridge University Press, 2009) 563–64. 14 For contract arbitration, see D Müller, ‘Ad hoc Investment Arbitration Based on State Contracts: From Lena Goldfields to the Libyan Oil Arbitrations’, ch 3 in this volume. 6 See,

The Applicable Law Saga 293 In the case of investment treaty arbitration, the primary source or lex specialis would be an international investment treaty itself. Indeed, almost always the dispute is to be decided in accordance with its provisions.15 At the same time, as was explained in the very first investment treaty case, AAPL v Sri Lanka, such a treaty cannot be read in isolation: Furthermore, it should be noted that the Bilateral Investment Treaty is not a self-contained closed legal system limited to provide for substantive material rules of direct applicability, but it has to be envisaged within a wider juridical context in which rules from other sources are integrated through implied incorporation methods, or by direct reference to certain supplementary rules, whether of international law character or of domestic law nature.16 (emphasis added)

The present contribution focuses on sources other than investment treaties. It categorises these applicable external rules and for each category presents the relevant major arbitral decisions while analysing the function of each set of rules and the articulation between them. In what follows, after addressing the issue of the choice of law, we examine in turn international, national, and EU law as applied by arbitral tribunals. We conclude by questioning the very nature of investment treaty arbitration from the point of view of its applicable law.

I. CHOICE OF LAW

The possibility to choose the law that would be applied to a dispute is a corollary of the principle of party autonomy, which is fundamental in international arbitration. Such choice may be both express or implied;17 at the same time, it should be ‘clear and unequivocal’.18 In general terms, it may be contained, whether in the investment contract, in the investment law of the host state, in a bi- or multilateral investment treaty, or in a subsequent agreement between the parties.19 It may also be incorporated by a renvoi to procedural rules, such as the United Nations Commission on International Trade Law (UNCITRAL) Arbitration Rules, the ICSID Convention or the ICSID Additional Facility Rules. Article 42(1) of the ICSID Convention provides in this respect that ‘The Tribunal shall decide a dispute in accordance with such rules of law as may be agreed by the parties.’ Article 35(1) of the UNCITRAL Rules establishes that ‘The arbitral tribunal shall apply the rules of law designated by the parties as applicable to the substance of the dispute.’

15 Y Banifatemi, ‘The Law Applicable in Investment Treaty Arbitration’ in K Yannaca-Small (ed), Arbitration Under International Investment Agreements: A Guide to the Key Issues (Oxford, Oxford University Press, 2010) 197. 16 Asian Agricultural Products Ltd v Republic of Sri Lanka, Award (27 June 1990) ICSID Case No ARB/87/3, para 21. See also MTD Equity Sdn Bhd and MTD Chile SA v Republic of Chile, Decision on Annulment (21 March 2007) ICSID Case No ARB/01/7, para 61 (‘MTD’s claim is one for “an alleged breach of any right conferred or created by this Agreement with respect to an investment by such investor” (BIT, Article 6(1) (ii)), and thus international law as the proper law of the BIT is applicable to that claim and to any defence thereto. The Respondent insists – and the Claimants do not disagree – that the Tribunal had to apply international law as a whole to the claim, and not the provisions of the BIT in isolation.’) (emphasis added). See E Lagrange, ‘SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?’, ch 6 in this volume. 17 Kjos, above (n 4) 71. 18 Compañía del Desarrollo de Santa Elena, SA v Republic of Costa Rica, Award (17 February 2000) ICSID Case No ARB/96/ 1, paras 63–64 (‘Article 42(1) of the ICSID Convention does not require that the parties’ agreement as to the applicable law be in writing or even that it be stated expressly. However, for the Tribunal to find that such an agreement was implied it must first find that the substance of the agreement, irrespective of its form, is clear. […] the Tribunal is unable to conclude that the parties ever reached a clear and unequivocal agreement that their dispute would be decided by the Tribunal solely in accordance with international law.’). 19 Kjos, above (n 4) 71.

294 Franck Latty and Marina Sim While some investment treaties (eg, the former NAFTA,20 the Energy Charter Treaty (ECT)21 and certain BITs) contain a provision on applicable law, the vast majority of the BITs do not provide for any choice of law.22 In such circumstances, the relevant arbitration rules or the ICSID Convention may circumscribe the applicable law. The residual rule in Article 42(1) of the ICSID Convention states that in the absence of an agreement of the parties, ‘the Tribunal shall apply the law of the Contracting State party to the dispute (including its rules on conflict of laws) and such rules of international law as may be applicable’. Under various arbitration rules, in the absence of choice arbitral tribunals enjoy considerable discretion with regards to the applicable law.23 Interestingly, while the very first investment treaty arbitration, AAPL v Sri Lanka,24 implicitly recognised the possibility of dissociated consent to arbitration by the host state and the investor, it did not go as far as to recognise the possibility of dissociated agreement on the choice of law.25 The Tribunal considered that since the arbitration request was ‘exclusively based on a treaty provision and not in implementation of a freely negotiated arbitration agreement directly concluded between the Parties’ in dispute, they ‘have had no opportunity to exercise their right to choose in advance the applicable law determining the rules governing the various aspects of their eventual disputes’. It drew the conclusion that: the prior choice-of-law referred to in the first part of Article 42 of the ICSID Convention could hardly be envisaged in the context of an arbitration case directly instituted in implementation of an international obligation undertaken between two States in favour of their respective nationals investing within the territory of the other Contracting State.26

The Tribunal stated that under those special circumstances, the choice-of-law process would normally materialise after the emergence of the dispute, by observing and construing the conduct of the Parties throughout the arbitration proceedings.27 Based on the Parties’ written and oral pleadings, the Tribunal concluded that both Parties acted in a manner that demonstrated their mutual agreement to consider the provisions of the Sri Lanka/UK BIT as being the primary source of the applicable legal rules and to apply, within the limits required, the relevant international or domestic legal rules.28 A few years later, the arbitral tribunal in the Antoine Goetz v Burundi case,29 initiated pursuant to the Belgium-Burundi BIT, recognised the possibility of dissociated agreement on

20 North

American Free Trade Agreement, Art 1131(1). Charter Treaty, Art 26(6). 22 E Gaillard and Y Banifatemi, ‘The Meaning of “and” in Article 42(1), Second Sentence, of the Washington Convention: The Role of International Law in the ICSID Choice of Law Process’ (2003) 18(2) ICSID Review – Foreign Investment Law Journal 375, 379; Banifatemi, above (n 15) 200. 23 2010 UNCITRAL Arbitration Rules, Art 35(1); 2021 ICC Arbitration Rules, Art 21(1); 2017 SCC Arbitration Rules, Art 27(1); ICSID Additional Facility Rules, Art 54(1). See also Douglas, above (n 9) rule 3, 40 (‘An investment treaty tribunal has the inherent authority to characterise the issues in dispute and determine the laws applicable thereto.’). 24 Asian Agricultural Products Ltd v Republic of Sri Lanka, Final Award (27 June 1990) ICSID Case No ARB/87/3. For more details on the case see E Lagrange, ‘SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?’, ch 6 in this volume. 25 See E Gaillard, La jurisprudence du CIRDI, Vol I (Paris, Pedone, 2004) 541. 26 Asian Agricultural Products Ltd v Republic of Sri Lanka, Final Award (27 June 1990) ICSID Case No ARB/87/3, paras 18–19. 27 ibid para 20. 28 ibid paras 20–24. This was however criticised by the Arbitrator S Asante in his Dissenting Opinion (See Dissenting Opinion, 576–78). 29 Antoine Goetz et al v Republic of Burundi, Award (10 February 1999) ICSID Case No ARB/95/3 (Arbitrators: P Weil, M Bedjaoui, and JD Bredin); see English translation in ‘Goetz, et al. v The Republic of Burundi, Award, ICSID Case No ARB/95/3, 10 February 1999’ in AJ Van den Berg (ed), Yearbook Commercial Arbitration, Vol XXVI (Alphen aan den Rijn, Kluwer Law International, 2001) 24. 21 Energy

The Applicable Law Saga 295 the choice of law. According to the Tribunal, Burundi accepted the applicable law as determined in the provision of the BIT by becoming a party to it, and that the investors did the same by filing their request for arbitration pursuant to the BIT: A bilateral treaty on investment protection has a juridical impact on both jurisdiction and applicable law. Undoubtedly, the applicable law has not been determined here, strictly speaking, by the parties to this arbitration (Burundi and the investors), but rather by the parties to the Bilateral Treaty (Burundi and Belgium). As was the case with the consent of the parties [to the arbitration], the Tribunal deems nevertheless that Burundi accepted the applicable law as determined in the above provision of the Bilateral Treaty by becoming a party to this Treaty, and that claimants did the same by filing their request for arbitration based on the Treaty. If this is not the first time, as it has been pointed out, that the jurisdiction of the Centre ensues directly from a bilateral treaty on investment protection rather than from a separate agreement between the host State and the investor, this is apparently one of the first instances in which an ICSID Tribunal has to apply the law directly determined by such a treaty.30 (emphasis added)

The Tribunal also predicted that the situation would occur with increasing frequency: The Bilateral Treaty on investment protection is not only the basis for the jurisdiction of the Centre and of the Tribunal; it also determines the applicable law. The present case is one of the first ICSID cases where this happens. Considering the growing use of choice of law clauses in investment treaties, as well as their considerable variety, such situation is equally likely to occur with increasing frequency.31 (emphasis added)

The Tribunal also pointed to ‘a remarkable come-back of international law’: It may be interesting to remark on this subject that choice of law clauses in investment protection treaties frequently refer to the provisions of the treaty itself, and, more broadly, to international law principles and rules. This leads to a remarkable come-back of international law, after a decline in practice and jurisprudence, in the legal relations between host States and foreign investors. This internationalization of investment relations, be they contractual or not, surely does not lead to a radical ‘denationalization’ of the legal relations born of foreign investment, to the point that the national law of the host State is totally irrelevant or inapplicable in favour of the exclusive role played by international law. It merely means that simultaneously – one could say in parallel – these relations depend on both the sovereignty of the host State on its national law and its international obligations.32

It concluded that ‘it suffices that it gives effect, according to the first sentence of Art. 42(1) ICSID Convention, to the parties’ choice as expressed in the BIT’.33 Since the applicable law clause in the BIT provided for a variable hierarchy between international and domestic law, depending on which was the most favourable to the investor, the Tribunal analysed the measures with regard to both legal orders. Today the possibility of dissociated agreement on choice of law is widely recognised both in doctrine and jurisprudence.34 At the same time, it is not always straightforward whether the parties reached an agreement on the applicable law. Thus, in a situation similar to the

30 ibid

36. 31. 32 ibid. 33 ibid 36–37. 34 See, eg, Banifatemi, above (n 15) 195; Siemens AG v The Argentine Republic, Award (17 January 2007) ICSID Case No ARB/02/8, para 76 (‘By accepting the offer of Argentina to arbitrate disputes related to investments, Siemens agreed that this should be the law to be applied by the Tribunal. This constitutes an agreement for purposes of the law to be applied under Article 42(1) of the Convention.’). 31 ibid

296 Franck Latty and Marina Sim AAPL case, the LG&E tribunal concluded that there was no choice of law.35 In any event, in investment treaty arbitration, due to the inherently international nature of the legal acts subject to their jurisdiction, tribunals are naturally led to apply international law.

II. INTERNATIONAL LAW

As many tribunals have stated, investment treaties (and even the ICSID Convention) are not self-contained regimes36 and there is no clinical isolation of these instruments from public international law.37 As a matter of fact, international law is applicable law in all treaty-based investment disputes, as many major decisions have recognised. After dealing with general international law, and more specific instruments, we will discuss the articulation of international law with national law as applicable law.

A. General International Law In investment disputes, the primary role of international law is traditionally ‘the role of guarantor’ – it places restraints on the use of public authority to interfere with private rights under municipal law.38 With investment treaty arbitration, as opposed to state contract disputes, the role of international law has evolved. Investment treaty arbitration is always about the international responsibility of states39 and international law is the lex causae of the investment treaty breaches.40 As Y Banifatemi put it: irrespective of whether or not an investment treaty refers to international law as the law applicable to the merits of the dispute, international law will always be the law governing the interpretation and the application of the treaty providing the basis for the arbitration, to the extent that what is at stake, in investment treaty arbitration, is the international responsibility of a State.41

35 LG&E Energy Corp et al v Argentine Republic, Decision on Liability (3 October 2006) ICSID Case No ARB/02/1, para 85 (‘It is to be noted that the Argentine Republic is a signatory party to the Bilateral Investment Treaty, which may be regarded as a tacit submission to its provisions in the event of a dispute related to foreign investments. In turn, LG&E grounds its claim on the provisions of the treaty, thus presumably choosing the treaty and the general international law as the applicable law for this dispute. Nevertheless, these elements do not suffice to say that there is an implicit agreement by the parties as to the applicable law, a decision requiring more decisive actions. Consequently, the dispute shall be settled in accordance with the second part of Article 42(1).’) (emphasis added). 36 See AAPL v Sri Lanka, Award (27 June 1990) ICSID Case No ARB/87/3, para 257. See also National Grid plc v Argentine Republic, Award (3 November 2008) UNCITRAL, para 87. 37 See Phoenix Action v Czech Republic, Award (15 April 2009) ICSID Case No ARB/06/5, para 78: ‘the ICSID Convention’s jurisdictional requirements – as well as those of the BIT – cannot be read and interpreted in isolation from public international law, and its general principles’. 38 I Alvik, ‘The Hybrid Nature of Investment Treaty Arbitration: Straddling the National/International Divide’ in CC Eriksen and M Emberland (eds), The New International Law: An Anthology (Leiden, Nijhoff, 2010) 92; Bishop, Crawford and Reisman, above (n 8) 445. 39 F Latty, ‘Conditions d’engagement de la responsabilité de l’Etat d’accueil de l’investissement’ in C Leben (ed), Droit international des investissements et de l’arbitrage transnational (Paris, Pedone, 2015) 415. 40 Douglas, above (n 9) rule 10, 81–90 (‘The law applicable to the issue of liability for a claim founded upon an investment treaty obligation is the investment treaty as supplemented by general international law.’); MTD Equity Sdn Bhd and MTD Chile SA v Republic of Chile, Decision on Annulment (21 March 2007) ICSID Case No ARB/01/7, para 72; Compañía de Aguas del Aconquija SA and Vivendi Universal SA v Argentine Republic, Decision on Annulment (3 July 2002) ICSID Case No ARB/97/3, paras 95–103. 41 Banifatemi, above (n 15) 210. See also E De Brabandere, Investment Treaty Arbitration as Public International Law (Cambridge, Cambridge University Press, 2014) 27–28.

The Applicable Law Saga 297 In this context, apart from the investment treaties themselves, tribunals may apply various norms of general international law,42 arising out of customary international law43 or general principles of law.44 As it was indicated in the ADC v Hungary case: The consent [to apply the provisions of the BIT] must also be deemed to comprise a choice for general international law, including customary international law, if and to the extent that it comes into play for interpreting and applying the provisions of the Treaty. This is so since the generally accepted presumption in conflict of laws is that parties choose one coherent set of legal rules governing their relationship […], rather than various sets of legal rules, unless the contrary is clearly expressed.45

The LG&E tribunal also explained in respect of Article 42(1) of the ICSID Convention that the reference to the phrase ‘as may be applicable’ should not be understood as if it were in some way conditioning application of international law; rather, it should be understood as referring, within international law, to the competent rules to govern the dispute at issue. According to the Tribunal, this interpretation could find support in the ICSID Convention’s French version, which refers to the rules of international law ‘en la matière’.46 Among the primary norms of international law, arbitral tribunals often apply the international minimum standards of treatment of aliens and the customary law on expropriation.47 These customary norms may be applied by renvoi in the investment treaty,48 or to interpret its terms,49 or as a ‘gap-filler’,50 when the treaty is silent on a point of law. For example, the Methanex v USA, as well as the Saluka v Czech Republic tribunals applied the police powers of states as customary international law. The Saluka tribunal held: [T]he principle that a State does not commit an expropriation and is thus not liable to pay compensation to a dispossessed alien investor when it adopts general regulations that are ‘commonly accepted as within the police powers of States’ forms part of customary law today. There is ample case law in support of this proposition. As the tribunal in Methanex Corp. v USA said recently in its final award, ‘[i]t is a principle of customary international law that, where economic injury results from a bona fide regulation within the police powers of a State, compensation is not required’.51

The Tribunal acknowledged that the expropriation clause in the Netherlands-Czech and Slovak Federal Republic BIT was drafted very broadly and did not contain any exception for

42 See F Latty, ‘Arbitrage transnational et droit international général’ (2008) 54 Annuaire français de droit international 467, 475 et seq (and the annual review since 2008 on international law in investment arbitration, with P Jacob and A de Nanteuil). 43 On the issue of application of customary international law see, eg, J Kurtz, ‘Building Legitimacy Through Interpretation in Investor-State Arbitration: On Consistency, Coherence, and the Identification of Applicable Law’ in Z Douglas, J Pauwelyn and JE Viñuales (eds), The Foundations of International Investment Law: Bringing Theory Into Practice (Oxford, Oxford University Press, 2013) 280–295. 44 eg, El Paso Energy International Company v Argentina, Award (31 October 2011) ICSID case No ARB/03/15, para 624: ‘there […] seems to be a general principle of law recognised by civilised nations that necessity cannot be recognised if a Party to a contract has contributed to it’. See T Gazzini, ‘General Principles of Law in the Field of Foreign Investment’ (2009) 10 Journal of World Investment and Trade 103. 45 ADC Affiliate Ltd and ADC & ADMC Management Ltd v Republic of Hungary, Award (2 October 2006) ICSID Case No ARB/03/16, para 290. 46 LG&E Energy Corp et al v Argentine Republic, Decision on Liability (3 October 2006) ICSID Case No ARB/02/1, para 88. 47 C McLachlan, ‘Investment Treaties and General International Law’ (2008) 57(2) International & Comparative Law Quarterly 361, 378. 48 See, eg, the case law on Art 1105(1) NAFTA. P Dumberry, The Fair and Equitable Treatment Standard. A Guide to NAFTA Case Law on Article 1105 (Alphen aan den Rijn, Kluwer Law International, 2013). 49 eg, Rompetrol Group NV v Roumania, Award (6 May 2013) ICSID Case No ARB06/3, para 197. 50 L Achtouk-Spivak, ‘Le droit applicable aux arbitrages en matière d’investissements’ in Leben, above n 39 at 853. 51 Saluka Investments BV v The Czech Republic, Partial Award (17 March 2006) UNCITRAL, para 262.

298 Franck Latty and Marina Sim the exercise of regulatory power. However, in using the concept of deprivation, the clause imported into the treaty the customary international law notion that a deprivation could be justified if it resulted from the exercise of regulatory actions aimed at the maintenance of public order.52 The Methanex v USA tribunal also recognised an ‘independent duty’ of an international tribunal to apply peremptory norms: ‘[A]s a matter of international constitutional law a tribunal has an independent duty to apply imperative principles of law or jus cogens and not to give effect to parties’ choices of law that are inconsistent with such principles.’53 More fundamentally, because of the nature of investment arbitration, arbitral tribunals routinely resort to secondary rules of international law, primarily the law of treaties (including the rules governing the interpretation of treaties),54 and the law of responsibility for internationally wrongful acts.55 Thus, customary rules on state responsibility are particularly relevant for determining compensation in case of unlawful expropriations, since expropriation clauses address instances of expropriation carried out pursuant to the BITs. In this respect, the Tribunals rely almost systematically on the PCIJ Chorzow Factory case56 and the ILC Articles on State Responsibility that reflect customary international law in this regard. As the Siemens tribunal explains: ‘The law applicable to the determination of compensation for a breach of such Treaty obligations is customary international law. The Treaty itself only provides for compensation for expropriation in accordance with the terms of the Treaty.’57 The key difference between compensation under the Draft Articles on State Responsibility and the Factory at Chorzów case formula on the one hand, and the expropriation provision in the BIT on the other hand, is that under the former, compensation must take into account ‘all financially assessable damage’ or ‘wipe out all the consequences of the illegal act’ as opposed to compensation ‘equivalent to the value of the expropriated investment’ under the Treaty. Under customary international law, the investor is entitled not just to the value of its enterprise as of the date of expropriation, but also to any greater value that enterprise has gained up to the date of the award, plus any consequential damages.58

B. Non-Investment Instruments Sometimes tribunals apply (or at least consider) non-investment instruments from other international law branches, mainly to address human rights or environmental considerations. This is in line with Article 31(3)(c) of the Vienna Convention on the Law of Treaties (VCLT), which requires that ‘any relevant rules of international law applicable in the relations between the parties’ shall be taken into account together with the context (the so-called ‘principle of

52 ibid

para 254. Corporation v United States of America, Final Award of the Tribunal on Jurisdiction and Merits (3 August 2005) UNCITRAL, para 24. See also Phoenix Action, Ltd v the Czech Republic, Award (15 April 2009) ICSID Case No ARB/06/5, para 78. 54 eg, Daimler Financial Services AG v Argentine Republic, Award (22 August 2012) ICSID case No ARB/05/1, para 254 (concerning the ‘holistic approach’ resulting from Art 31et seq of the Vienna Convention on the Law of Treaties). 55 But see Douglas, above (n 9) 96–97, suggesting that Parts Two and Three of the ILC’s Articles on State Responsibility are not applicable to the responsibility in investment treaty disputes. 56 Case Concerning the Chorzow Factory (13 September 1928) PCIJ, Series A, No 17. But see Douglas, above (n 9) 100–101. 57 Siemens AG v The Argentine Republic, Award (17 January 2007) ICSID Case No ARB/02/8, para 349. 58 ibid para 352. 53 Methanex

The Applicable Law Saga 299 systemic interpretation’ of treaties). Here, the question of applicable law and the rules of interpretation of treaties become inextricably linked.59 As E De Brabandere indicates, when deciding on an investment dispute, there is no reason for the Tribunal to exclude ipso facto human rights considerations as a matter of applicable law.60 Thus, they will be considered, but only in the event that the party to an investment dispute that invokes these considerations can effectively demonstrate a conflict of norms.61 The rules of treaty interpretation make it possible to import into investment litigation substantive norms that can rebalance the distribution of rights and obligations between states and protected investors. In Philip Morris v Uruguay, the interpretation of the Switzerland/ Uruguay treaty by this means justified the application of customary rules on the police powers of the state.62 Later, the Tribunal tasked with assessment of the tobacco regulations adopted by Uruguay for the protection of public health, decided to use the World Health Organization Framework Convention on Tobacco Control as ‘a point of reference on the basis of which to determine the reasonableness of the measures’.63 Since Uruguay merely implemented its obligations assumed under the Convention, it was not held liable. The Urbaser tribunal also relied on Article 31(3)(c) of the VCLT to hold that ‘The BIT cannot be interpreted and applied in a vacuum’; and that it ‘has to be construed in harmony with other rules of international law of which it forms part, including those relating to human rights’. This ‘approach’ was, according to the Tribunal, reflected in the applicable law clause of the treaty which provided for the application of international law. For the Tribunal it proved that ‘the BIT is not framed in isolation, but placed in the overall system of international law’.64 In the same vein, in S.D. Myers v Canada, the Tribunal analysed Canada’s obligations under the Basel Convention on the Control of Transboundary Movements of Hazardous Waste and Their Disposal and held it liable, since while intending to comply with its obligations under the Convention, Canada did not choose the alternative that was least inconsistent with the NAFTA.65 In the Maffezini v Spain case, while addressing the investor’s obligation to conduct an Environmental Impact Assessment, the Tribunal concluded that the procedure was ‘basic for the adequate protection of the environment and the application of appropriate preventive measures’ and that it was true ‘not only under Spanish and EEC law, but also increasingly so under international law’. In stating this, it relied, inter alia, on the Convention on Environmental Impact Assessment in a Trans-boundary Context.66 It should also be noted that external instruments, including soft law ones or non-applicable treaties, may be used by tribunals for argumentative (or even ‘legitimising’) purposes. In order to demonstrate the customary nature of the doctrine of state police powers, the arbitral tribunal in Philip Morris v Uruguay67 referred to the Harvard Draft Convention on the International

59 H Ascensio, ‘Article 31 of the Vienna Convention on the Law of Treaties and International Investment Law’ (2016) 31(2) ICSID Review – Foreign Investment Law Journal 366. 60 De Brabandere, above (n 41) 134. 61 ibid 135. 62 Philip Morris Brands Sàrl, Philip Morris Products SA and Abal Hermanos SA v Oriental Republic of Uruguay, Award (8 July 2016) ICSID Case No ARB/10/7, paras 290 et seq. 63 ibid para 401. 64 Urbaser SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v Argentine Republic, Award (8 December 2016) ICSID Case No ARB/07/26, paras 1200–01. 65 S.D. Myers, Inc. v Government of Canada, Partial Award (13 November 2000) UNCITRAL. 66 Emilio Agustín Maffezini v The Kingdom of Spain (13 November 2000) ICSID Case No ARB/97/7, para 67. 67 Philip Morris Brands Sàrl, Philip Morris Products SA and Abal Hermanos SA v Oriental Republic of Uruguay, Award (8 July 2016) ICSID Case No ARB/10/7, paras 290 et seq.

300 Franck Latty and Marina Sim Liability of States for Damage Caused in their Territory to the Person or Property of Foreigners (1960), to the Third Restatement of the Foreign Relations Law (1987), to an OECD working paper on expropriation of 2004, to which should be added in a more recent period ‘a range of investment decisions [that] have contributed to develop the scope, content and conditions of the State’s police powers doctrine, anchoring it in international law’.68 Finally, to complete its demonstration, the Tribunal referred to the recent treaty practice of certain states (model investment treaties of the US and Canada, as well as the EU’s trade agreements with Canada and Singapore), from which it deduced that the relevant provisions ‘whether or not introduced ex abundanti cautela, reflect the position under general international law’.69 In Urbaser v Argentina, in establishing that corporations, as subjects of international law, are bound by international law obligations, the Tribunal relied successively on the Guiding Principles on Business and Human Rights: Implementing the UN ‘Protect, Respect and Remedy’ Framework (Ruggie Principles), the 1948 Universal Declaration of Human Rights, the General Comment of the Committee on Economic, Social and Cultural Rights No 15 of 2002 on the right to water, the resolution on the same issue of the UN General Assembly 64/292 of 28 July 2010, and the ILO Tripartite Declaration of 1977 on Multinational Enterprises and Social Policy.70 Investment arbitration thus contributes to blurring the boundaries between hard and soft law.71

C. International Law/National Law One of the most controversial questions concerns the respective roles of the national legal order on the one hand, and the international legal order on the other,72 and in particular, the interpretation of the second sentence of Article 42(1) of the ICSID Convention.73 As CH Schreuer stated, ‘The applicable law in investment disputes has turned out to be a dangerous area. It takes great nautical skill to keep the proper balance between the Scylla and Charybdis of the two legal systems.’74 It must be said that Article 42(1) leads ICSID tribunals to articulate norms, not legal orders,75 and as a consequence the primacy of international law may occasionally be affected. Before the Wena Hotels decision,76 the predominant view first formulated by the Klöckner ad hoc committee77 was that international law played ‘complementary’ or ‘corrective’ roles 68 ibid

para 295. para 301. SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v Argentine Republic, Award (8 December 2016) ICSID Case No ARB/07/26, paras 1195 et seq. 71 See F Latty, ‘De la tendresse dans le monde des juges: la soft law devant les juridictions internationales’ in P Deumier and JM Sorel (eds), Regards croisés sur la soft law en droit interne, européen et international (Paris, LGDJ/ Lextenso, 2018) 387 et seq. 72 R Rivier, ‘L’articulation entre droit international et droit national devant les tribunaux arbitraux internationaux d’investissement’ in S Robert-Cuendet (ed), Droit des investissements internationaux ? Perspectives croisées (Brussels, Bruylant, 2017) 413 et seq. 73 Gaillard and Banifatemi, above (n 22) 375. 74 C Schreuer, above (n 5) 75. 75 M Forteau, ‘Le juge CIRDI envisagé du point de vue de son office: juge interne, juge international, ou l’un et l’autre à la fois ?’ in Liber Amicorum Jean-Pierre Cot (Brussels, Bruylant, 2009) 112 et seq. See SAUR International SA v Argentine Republic, Decision on Jurisdiction (6 June 2012) ICSID case No ARB/04/4, para 327 (confronted with the different sources of applicable law provided for by the treaty, the tribunal must not follow the principle of hierarchy but the principle of specialty: to each question must be applied the relevant norm according to its own nature). 76 Wena Hotels LTD v Arab Republic of Egypt, Annulment Proceeding (5 February 2002) ICSID Case No ARB/98/4. 77 Klöckner Industrie-Anlagen GmbH et al v United Republic of Cameroon and Société Camerounaise des Engrais, Decision on Annulment (3 May 1985) ICSID Case No ARB/81/2. See also Amco Asia Corporation et al v Republic of 69 ibid

70 Urbaser

The Applicable Law Saga 301 and applied only in the case of lacunae in the law of the host state or its inconsistency with the requirements of international law. In this regard, the ad hoc committee in Wena Hotels v Egypt recognised the arbitrators’ freedom to resort to international law directly. The committee upheld the Tribunal’s reliance on international law. It considered that there does not seem to be a single answer as to which of the approaches to the role of international law is the correct one – the circumstances of each case may justify one or another solution. It also indicated that the use of the word ‘may’ in the second sentence of Article 42(1) indicated that the Convention ‘does not draw a sharp line for the distinction of the respective scope of international and of domestic law and, correspondingly, that this has the effect to confer on to the Tribunal a certain margin and power for interpretation’.78 On the basis of the negotiation history of the ISCID Convention it concluded that both legal orders could play a role and international law could also be applied by itself: [T]he sense and meaning of the negotiations leading to the second sentence of Article 42(1) allowed for both legal orders to have a role. The law of the host State can indeed be applied in conjunction with international law if this is justified. So too international law can be applied by itself if the appropriate rule is found in this other ambit.79 (emphasis added)

The ad hoc committee also noted that ‘when a tribunal applies the law embodied in a treaty to which Egypt is a party it is not applying rules alien to the domestic legal system of this country’, and that this might also be true for other sources of international law.80 This non-hierarchical interpretation of Article 42(1) of the ICSID Convention was subsequently upheld in numerous awards, such as LG&E,81 Enron,82 Azurix,83 Sempra84 etc. The CMS tribunal described it as ‘a more pragmatic and less doctrinaire approach’ allowing for the application of both domestic law and international law if the specific facts of the dispute so justify.85 The Burlington tribunal referred to it as ‘prevailing case law’.86 Thus, it was established that Article 42(1) of the ICSID Convention empowers the tribunal to apply either the municipal law of the host state (or the municipal laws of other states as per the host state’s conflict of laws rules) or international law depending upon the causes of action advanced by the parties within the scope of the tribunal’s jurisdiction ratione materiae as determined by the relevant instrument (investment treaty, investment agreement or investment authorisation).87

Indonesia, Decision on the Application for Annulment of 16 May 1986, ICSID Case No ARB/81/1, 1 ICSID Rep 509 (1993). 78 Wena Hotels LTD v Arab Republic of Egypt, Annulment Proceeding (5 February 2002) ICSID Case No ARB/98/4, para 39. 79 ibid para 40. 80 ibid para 44. 81 LG&E Energy Corp et al v Argentine Republic, Decision on Liability (3 October 2006) ICSID Case No ARB/02/1, paras 95–96. 82 Enron Corporation and Ponderosa Assets, LP v Argentine Republic, Award (22 May 2007) ICSID Case No ARB/01/3, para 207. 83 Azurix Corp v The Argentine Republic, Award (14 July 2006) ICSID Case No ARB/01/12, para 66. 84 Sempra Energy International v Argentine Republic, Award (28 September 2007) ICSID Case No ARB/0/16, para 236. 85 CMS Gas Transmission Company v The Republic of Argentina, Award (12 May 2005) ICSID Case No ARB/01/8, para 116. 86 Burlington Resources Inc. v Republic of Ecuador, Decision on Ecuador’s Counterclaims (7 February 2017) ICSID Case No ARB/08/5, para 74. 87 Douglas, above (n 9) 132–33.

302 Franck Latty and Marina Sim In this respect, the Annulment Decision in Compañía de Aguas del Aconquija S.A. and Vivendi Universal S.A. v Argentine Republic formulated what would become the so-called ‘Vivendi test’88 based on the ‘essential’ or ‘fundamental basis’ of the claim, allowing a tribunal or ad hoc committee to distinguish between the law applicable to violations of the treaty and the proper law of the contract.89 The case was brought pursuant to the Argentina-France BIT, but some of the investors’ claims were closely interrelated with the concession contract concluded between one of the investors and the province of Tucumán, which provided for the exclusive jurisdiction of the administrative courts of Tucumán. This was the reason the Tribunal declined to consider them on the merits, despite the fact that it accepted Tucumàn’s jurisdiction under the BIT.90 In determining the relationship between Argentina’s responsibility under the BIT and the rights and obligations of the parties to the concession contract, the ad hoc committee stressed that Articles 3 and 5 of the BIT (fair and equitable treatment, full protection and security standards, prohibition of expropriation, war clause) did not relate directly to breach of a municipal contract, rather they set an independent standard. The committee referred to Article 3 of the ILC Articles which provides that ‘The characterization of an act of a State as internationally wrongful is governed by international law. Such characterization is not affected by the characterization of the same act as lawful by internal law.’91 Therefore, it held that each claim should be determined by reference to its own applicable law; in the case of the BIT, by international law: In accordance with this general principle (which is undoubtedly declaratory of general international law), whether there has been a breach of the BIT and whether there has been a breach of contract are different questions. Each of these claims will be determined by reference to its own proper or applicable law – in the case of the BIT, by international law; in the case of the Concession Contract, by the proper law of the contract, in other words, the law of Tucumán. For example, in the case of a claim based on a treaty, international law rules of attribution apply, with the result that the state of Argentina is internationally responsible for the acts of its provincial authorities. By contrast, the state of Argentina is not liable for the performance of contracts entered into by Tucumán, which possesses separate legal personality under its own law and is responsible for the performance of its own contracts.92 (emphasis added)

The committee indicated that where ‘the fundamental basis of the claim’ is a treaty laying down an independent standard by which the conduct of the parties is to be judged, the existence of an exclusive jurisdiction clause in a contract between the claimant and the respondent state or one of its subdivisions cannot operate as a bar to the application of the treaty standard; at most, it might be relevant – as municipal law will often be relevant – in assessing whether there has been a breach of the treaty.93 It also concluded that: [I]t is not open to an ICSID tribunal having jurisdiction under a BIT in respect of a claim based upon a substantive provision of that BIT, to dismiss the claim on the ground that it could or should have

88 Kjos,

above (n 4) 110. de Aguas del Aconquija SA and Vivendi Universal SA v Argentine Republic, Decision on Annulment (3 July 2002) ICSID Case No ARB/97/3. 90 Compañiá de Aguas del Aconquija SA and Vivendi Universal SA v Argentine Republic, Award (21 November 2000) ICSID Case No ARB/97/3. 91 Compañía de Aguas del Aconquija SA and Vivendi Universal SA v Argentine Republic, Decision on Annulment, ICSID Case No ARB/97/3, para 95. 92 ibid para 96. 93 ibid para 101. 89 Compañía

The Applicable Law Saga 303 been dealt with by a national court. In such a case, the inquiry which the ICSID tribunal is required to undertake is one governed by the ICSID Convention, by the BIT and by applicable international law. Such an inquiry is neither in principle determined, nor precluded, by any issue of municipal law, including any municipal law agreement of the parties.94 (emphasis added)

Since the Tribunal declined to decide key aspects of the Claimants’ BIT claims on the ground that they involved issues of contractual performance or non-performance,95 the award was annulled with regards to the acts of the Tucumán authorities for manifest excess of powers (Article 52(1)(b) of the ICSID Convention).96 The Tribunal, constituted anew, applied international law when holding that Argentina was liable for violating the fair and equitable treatment standard and the prohibition of expropriation, as set out in the BIT.97 The conclusions of the Vivendi decision were cited with approval by the ad hoc committee in Azurix, which ruled: Each of Azurix’s claims in this case was for an alleged breach of the BIT. The BIT is an international treaty between Argentina and the United States. By definition, a treaty is governed by international law, and not by municipal law. […] In any claim for breach of an investment treaty, the question whether or not there has been a breach of the treaty must therefore be determined, not through the application of the municipal law of any State, but through the application of the terms of the treaty to the facts of the case, in accordance with general principles of international law, including principles of the international law of treaties. Bearing in mind that an investment treaty, whether bilateral or multilateral, is itself a source of international law as between the States parties to that treaty, the applicable law in any claim for a breach of that treaty can thus be said to be the treaty itself specifically, and international law generally.98

The Tribunal in LG &E Energy Corp. et al. v Argentina indicated that the absence of a contract between the parties favours application of international law in the first place: The fact that there is no contract between the Argentine Republic and LG&E favors in the first place, the application of international law, inasmuch as we are dealing with a genuine dispute in matters of investment which is especially subject to the provisions of the Bilateral Treaty complemented by the domestic law.99

The Tribunal clarified that as part of the Argentine legal system, the BIT prevails over domestic law, ‘especially, inasmuch as in most of the Bilateral Treaty’s assumptions there is an express mention of international law, be it when referring to the treatment to be given to investments, or to the compensation in the event of expropriation or any other like measure, etc’100 and that international law overrides domestic law when there is a contradiction, since a state cannot justify non-compliance of its international obligations by asserting the provisions of its domestic law.101

94 ibid

para 102. para 108. 96 ibid para 119. 97 Compañiá de Aguas del Aconquija SA and Vivendi Universal SA v Argentine Republic, Award (20 August 2007) ICSID Case No ARB/97/3, paras 7.3.8–7.3.10. 98 Azurix Corp v Argentine Republic, Decision on Annulment (1 September 2009) ICSID Case No ARB/01/12, para 146. 99 LG&E Energy Corp et al v Argentine Republic, Decision on Liability (3 October 2006) ICSID Case No ARB/02/1, para 98. 100 ibid para 91. 101 ibid para 94. 95 ibid

304 Franck Latty and Marina Sim The Enron tribunal also justified the supremacy of international law by reference to the Argentine Constitution and the VCLT: It must be noted also that the very legal system of the Argentine Republic, like many modern systems, provides for a prominent role of treaties under both Articles 27 and 31 of the Constitution. Treaties are constitutionally recognized among the sources considered ‘the supreme law of the Nation’. It follows that in case of conflict between a treaty rule and an inconsistent rule of domestic law, the former will prevail. This is not just the consequence of the Constitution so providing, but also the solution dictated by Article 27 of the Vienna Convention on the Law of Treaties in that a State ‘may not invoke the provisions of its internal law as justification for its failure to perform a treaty’. Consistent with this role of international law, regulatory instruments have also made specific reference to the protection of investments under the Treaty (Decree 669/00).102

III. NATIONAL LAW

Unlike international law, national law, usually the host state’s, provides answers to the ‘local’ questions, such as the existence or scope of property rights103 or contract law issues (especially, administrative contracts), to which international law does not give responses. It applies by default for assessing whether there has been a breach of contractual obligations, where the arbitration agreement in the relevant investment treaty is broad enough to encompass contractbased claims.104 As Z Douglas put it: ‘If the investment treaty tribunal has jurisdiction over contractual claims, and the investor has a contract with an emanation of the host state, then its contractual rights fall to be determined by the law governing the contract.’105 Apart from that, in certain cases international investment tribunals need to take municipal law into account in order to determine the international law content of a treaty standard (an incidental question requiring a renvoi to national law).106 It is also particularly relevant for assessing breaches of the so-called umbrella clauses107 and can serve as the basis for counterclaims by host states.108

102 Enron Corporation and Ponderosa Assets, LP v Argentine Republic, Award (22 May 2007) ICSID Case No ARB/01/3, para 208. See also CMS Gas Transmission Company v The Republic of Argentina, Award (12 May 2005) ICSID Case No ARB/01/8, para 120; BG Group Plc v The Republic of Argentina, Final Award (24 December 2007) UNCITRAL, para 97 (‘More importantly, the interplay between international law and municipal law under Article 8(4) of the BIT should not overlook that the former may be deemed incorporated into the latter, depending on the status conferred to international treaties and international law in general by a particular constitutional system. This is particularly relevant to the case of Argentina, whose constitutional framework and doctrine have traditionally admitted the direct application of international law whenever feasible and, at least since the constitutional reform undertaken in 1994, expressly providing for the principle that international treaties preempt provincial and federal law. Accordingly, the challenge of discerning the role that international law ought to play in the settlement of this dispute, vis-à-vis domestic law, disappears if one were to take into account that the BIT and underlying principles of international law, as “the supreme law of the land”, are incorporated into Argentine domestic law, superseding conflicting domestic statutes.’) 103 Douglas, above (n 9) rule 4, 52–72; Vestey Group Ltd v Venezuela, Award (15 April 2016) ICSID Case No ARB/06/4, para 257 (‘The requirements for acquiring property rights over immovable assets situated in Venezuela are governed by specific norms of Venezuelan property law. For a private person to have a claim under international law arising from the deprivation of its property, it must hold that property in accordance with applicable rules of domestic law.’) 104 Kjos, above (n 4) 171. 105 Douglas, above (n 9) 40. 106 See, eg, M Sasson, Substantive Law in Investment Treaty Arbitration: The Unsettled Relationship between International Law and Municipal Law (Alphen aan den Rijn, Kluwer Law International, 2010) 171. 107 See B Samson, ‘Umbrella Clauses and Contract Claims’, ch 23 in this volume. 108 See AK Bjorklund, ‘Counterclaims’, ch 17 in this volume.

The Applicable Law Saga 305 The role of national law for the assessment of treaty-based claims was clarified by the MTD ad hoc committee, in which the parties disagreed on the law applicable to the granting of the permits necessary for the claimants to carry out their investment in Chile: As noted above, the lex causae in this case based on a breach of the BIT is international law. However it will often be necessary for BIT tribunals to apply the law of the host State, and this necessity is reinforced for ICSID tribunals by Article 42(1) of the ICSID Convention. Whether the applicable law here derived from the first or second sentence of Article 42(1) does not matter: the Tribunal should have applied Chilean law to those questions which were necessary for its determination and of which Chilean law was the governing law. At the same time, the implications of some issue of Chilean law for a claim under the BIT were for international law. In short, both laws were relevant.109 (emphasis added)

The committee agreed with the Tribunal that in order to establish the facts of the breach of the BIT, it was necessary to consider the contractual obligations undertaken by the respondent and the claimants and their scope under Chilean law.110 Thus, for assessing the fair and equitable treatment claim, the meaning of a Chilean contract was a matter of Chilean law; its implications in terms of an international law claim were a matter for international law.111 In EnCana v Ecuador, to assess the expropriation claim the Tribunal verified the existence of property rights under domestic law even though the relevant BIT clause only referred to ‘applicable rules of international law’: The relevant clause, Article XIII (7) of the BIT, provides only a tribunal exercising jurisdiction under the BIT ‘shall decide the issues in dispute in accordance with this Agreement and applicable rules of international law’. Unlike many BITs there is no express reference to the law of the host State. However for there to have been an expropriation of an investment or return (in a situation involving legal rights or claims as distinct from the seizure of physical assets) the rights affected must exist under the law which creates them, in this case, the law of Ecuador. The effect of the opening words of Article XII(4) is to permit this Tribunal to determine and apply the taxation law of Ecuador to the extent that it is necessary to do so in order to deal with a claim under Article VIII.112 (emphasis added)

In the Azurix case, since all the claims were for alleged violations of the BIT, the Tribunal adopted a rather limiting approach. It characterised the issues arising out of the Concession Agreement as factual elements to be considered, and dealt with the municipal law of Argentina in a part of the award headed ‘The Facts’. It explained as follows: Azurix’s claim has been advanced under the BIT and, as stated by the Annulment Committee in Vivendi II, the Tribunal’s inquiry is governed by the ICSID Convention, by the BIT and by applicable international law. While the Tribunal’s inquiry will be guided by this statement, this does not mean that the law of Argentina should be disregarded. On the contrary, the law of Argentina should be helpful in the carrying out of the Tribunal’s inquiry into the alleged breaches of the Concession Agreement to which Argentina’s law applies, but it is only an element of the inquiry because of the treaty nature of the claims under consideration.113 (emphasis added) 109 MTD Equity Sdn Bhd and MTD Chile SA v Republic of Chile, Decision on Annulment (21 March 2007) ICSID Case No ARB/01/7, para 72. 110 ibid para 73; MTD Equity Sdn Bhd and MTD Chile SA v Republic of Chile, Award (25 May 2004) ICSID Case No ARB/01/7, para 187. 111 MTD Equity Sdn Bhd and MTD Chile SA v Republic of Chile, Decision on Annulment (21 March 2007) ICSID Case No ARB/01/7, para 75. 112 EnCana Corporation v Republic of Ecuador, Award (3 February 2006) LCIA Case No UN 3481, para 184. 113 Azurix Corp v The Argentine Republic, Award (14 July 2006) ICSID Case No ARB/01/12, para 67. See also Energy International Company v The Argentine Republic, Award (31 October 2011) ICSID Case No ARB/03/15, paras 135, 141.

306 Franck Latty and Marina Sim The Tribunal’s approach was endorsed by the ad hoc committee, which rejected Argentina’s application for annulment of the award: In some cases, it may be an express term of the investment treaty that the host State is required to comply with specified provisions of its own municipal law. In such cases, a breach by the host State of municipal law may thus amount to a breach of the treaty. Although municipal law does not as such form part of the law applicable to a claim for breach of a treaty, in such cases it may be necessary to determine whether there has been a breach of municipal law as a step in determining whether there has been a breach of the treaty.114 (emphasis added)

The committee indicated that, even if it had been necessary to apply Argentine municipal law in determining whether there was a breach of obligations under municipal law, the municipal law would not thereby become part of the applicable law under Article 42 of the ICSID Convention for purposes of determining whether there was a breach of the BIT. Rather, any breach of municipal law that might be established ‘would be a fact or element to which the terms of the BIT and international law would be applied in order to determine whether there was a breach of ’ the BIT.115 It clarified that Article 42(1) of the ICSID Convention ‘cannot possibly be understood as having the effect that, in the absence of an express choice of law clause, the municipal law of the Contracting State will be the applicable law in claims for alleged breaches of an investment treaty’.116 A broader role for domestic law as applicable law was highlighted by the Enron tribunal: The Respondent is right in arguing that domestic law is not confined to the determination of factual questions. It has indeed a broader role, as it is evident in this very case from the pleadings and arguments of the parties that have relied heavily on the Gas Law and generally the regulatory framework of the gas industry, just as they have relied on many other rules of the Argentine legal system, including the Constitution, the Civil Code, specialized legislation and the decisions of courts. The License itself is governed by the legal order of the Argentine Republic and it must be interpreted in its light.117

Since in examining the Argentine law as pertinent to various issues disputed by the parties, the Tribunals concluded that there was generally no inconsistency with international law as far as the basic principles governing the matter were concerned; they applied both Argentine law and international law to the extent relevant to the decision of the claims submitted to them, while noting that in case of inconsistency between Argentine law and the treaties in force, international law would prevail.118 At the same time, as the CMS tribunal observed, it is not always possible to separate the various laws under consideration: Indeed there is here a close interaction between the legislation and the regulations governing the gas privatization, the License and the international law, as embodied in the Treaty and customary international law. All of these rules are inseparable and will, to the extent justified, be applied by the Tribunal.119

114 Azurix Corp v The Argentine Republic, Decision on the Application for Annulment of the Argentine Republic (1 September 2009) ICSID Case No ARB/01/12, para 149. 115 ibid para 151. 116 ibid para 147. 117 Enron Corporation and Ponderosa Assets, LP v Argentine Republic, Award (22 May 2007) ICSID Case No ARB/01/3, para 206. 118 ibid paras 208–09. 119 CMS Gas Transmission Company v The Republic of Argentina, Award (12 May 2005) ICSID Case No ARB/01/8, para 117.

The Applicable Law Saga 307 IV. EUROPEAN UNION LAW

A particular difficulty in respect of the applicable law arises in the context of the so-called intra-EU investment arbitrations, since EU law has direct effect and primacy over national law and may apply to a particular dispute.120 This poses problems with regards to the autonomy of the EU (Community) legal order, first formulated in the landmark decision Costa v ENEL121 and reiterated in particular in Kadi I122 and in Opinion 2/13,123 as it creates the risk of EU law being interpreted and applied by arbitral tribunals without control by the Court of Justice of the European Union (CJEU), or even without residual control by the courts of the Member States.124 In this respect, Article 344 of the Treaty on the Functioning of the European Union (TFEU) prevents Member States from submitting a dispute concerning the interpretation or application of the TEU and TFEU to any method of settlement other than those provided for in the treaties. In fact, EU law can operate on three possible levels: (i) as international law; (ii) as a distinct legal order within the EU, separate both from the national laws of EU Member States, and international law; and (iii) as part of national law.125 One of the major problems in this respect concerned the proceedings initiated pursuant to the ECT, a multilateral agreement to which the EU itself is a party. This was the case in Electrabel S.A. v Republic of Hungary, in which the Tribunal constituted pursuant to Article 26 of the ECT considered that it was required to operate in the international legal framework of the ECT and the ICSID Convention, outside the EU: Under Article 26 ECT and Article 42 of the ICSID Convention, the Tribunal is required to apply the ECT and “applicable rules and principles of international law.” In other words, this Tribunal is placed in a public international law context and not a national or regional context. […] This ICSID arbitration is a dispute resolution mechanism governed exclusively by international law. As a result of the Tribunal’s international status under the ECT and the ICSID Convention, several of the Commission’s submissions cannot be taken into account in this arbitration, because they are based on a hierarchy of legal rules seen only from the perspective of an EU legal order applying within the EU, whereas this Tribunal is required to operate in the international legal framework of the ECT and the ICSID Convention, outside the European Union.126

The Tribunal admitted that EU law was a sui generis legal order, presenting different facets and having a multiple nature.127 When analysing it as part of international law, the

120 See, eg, Emilio Agustín Maffezini v Kingdom of Spain, Award (13 November 2000) ICSID Case No ARB/97/7, where the tribunal applies the EEC Directive 85/337 of 27 June 1985. 121 Judgment of 12 July 2015 1964, Costa v Enel, C 6/64, EU:C:1964:66; J Czuczai, ‘The Autonomy of the EU legal Order and the Law-making Activities of International Organisations: Some Examples Regarding the Council’s Most Recent Practice’ (2012) 31(1) Yearbook of European Law 452. 122 Judgment of 3 September 2008, Yassin Abdullah Kadi and Al Barakaat International Foundation v Council and Commission, C-402/05 P and C-415/05 P, EU:C:2008:461. 123 Opinion 2/13 (Accession of the European Union to the ECHR), 18 December 2014, EU:C:2014:2454, para 170 (‘The autonomy enjoyed by EU law in relation to the laws of the Member States and in relation to international law […]’). 124 Judgment of 6 March 2018, Slovak Republic v Achmea BV, C-284/16, EU:C:2018:158, para 60. 125 Electrabel SA v Republic of Hungary, Decision on Jurisdiction, Applicable Law and Liability (30 November 2012) ICSID Case No ARB/07/19, para 4.20. 126 ibid para 4.112. 127 ibid paras 4.117–8.

308 Franck Latty and Marina Sim Tribunal did not find any fundamental difference in nature that could justify treating EU law, unlike other international rules, differently in an international arbitration requiring the application of relevant rules and principles of international law.128 Thus, the Tribunal concluded that EU law formed part of the rules and principles of international law applicable to the parties’ dispute under Article 26(6) of the ECT. Moreover, EU law, as part of Hungary’s national legal order, was also to be taken into account as a fact relevant to the parties’ dispute.129 On 6 March 2018, the European Court of Justice (ECJ) adopted a controversial decision in Achmea v Slovak Republic declaring the arbitration clauses in intra-EU BITs to be incompatible with EU law, as they were precluded by Articles 267 and 344 of the TFEU.130 However, arbitral tribunals found a way around Achmea, thus restricting its practical consequences. In particular, tribunals considered that the Achmea reasoning did not affect proceedings initiated pursuant to the ECT. In Eskosol v Italy, the Tribunal did not consider that ‘applicable rules and principles of international law’ in Article 26(6) of the ECT operate as ‘a direct and broad renvoi to the EU Treaties’. According to it, the Article could not be interpreted as encompassing EU law, which is ‘a regional and not a worldwide system of law’.131 At the same time, the Tribunal stated that its conclusion that EU law was not part of the ECT’s applicable law did not mean that an ECT tribunal could not consider EU law as a matter of fact if potentially relevant to the merits of a dispute, ‘just as an ECT tribunal may consider a State’s domestic law as part of the factual matrix of a case’; considering EU law to that extent ‘would be perfectly consistent with the Tribunal’s interpretation of Article 26(6)’.132 In RREEF v Spain, the Tribunal concluded that the Achmea judgment was inapposite in that case, since the applicable law in Achmea was the Netherlands–Czech and Slovak Federative Republic BIT exclusively concluded between two EU Member States. This was not the case of the ECT which ‘binds both the EU and its Member States on the one hand and non-EU States on the other hand’.133 In this context, in the Moldova v Komstroy ruling, rendered on 2 September 2021, the ECJ extended the Achmea’s reasoning to intra-EU disputes initiated under ECT.134 Furthermore, on 5 May 2020, the majority of EU Member States adopted an agreement for the termination of the intra-EU BITs, which entered in force on 29 August 2020.135 This agreement is intended to mark the end of intra-EU BIT arbitration, and thus the issue of the applicable law within the EU context may also disappear in the future. At the same time, the question remains whether arbitral tribunals will abide by the Achmea and Komstroy decisions.

128 ibid

para 4.126. paras 4.195, 4.127. 130 Judgment of 6 March 2018, Slovak Republic v Achmea BV, C-284/16, EU:C:2018:158, para 60. 131 Eskosol SpA in liquidazione v Italian Republic, Decision on Termination Request and Intra-EU Objection (7 May 2019) ICSID Case No ARB/15/50, para 121. 132 ibid para 123. See also Statement of Dissent of Pr MG Kohen in Theodoros Adamakopoulos et al v Republic of Cyprus (3 February 2020) ICSID Case No ARB/15/49. 133 RREEF Infrastructure (G.P.) Limited and RREEF Pan-European Infrastructure Two Lux S.à r.l. v Kingdom of Spain, Decision on Responsibility and on the Principles of Quantum (30 November 2018) ICSID Case No ARB/13/30, para 211. See D Azaria, ‘The Renewable Energy Arbitrations Under the Energy Charter Treaty’, ch 10 in this volume. 134 Judgment of 2 September 2021, Republic of Moldova v Komstroy LLC, Case C-741/19, EU:C:2021:655. 135 Agreement for the Termination of Bilateral Investment Treaties between the Member States of the European Union, [2020] OJ L169/1–41. 129 ibid

The Applicable Law Saga 309 V. CONCLUSION

The analysis of the main decisions on the substantive law applied to investment disputes allows to question the very nature of international investment arbitration as public international law. One could agree that investment arbitration is rather of a sui generis nature ‘which cannot be adequately rationalized either as a form of public international or private transnational dispute resolution’136 or a ‘hybrid legal process relying on and applying both municipal and international law in one integrated legal process’.137 It comprises characteristics of commercial arbitration and inter-state adjudication, public (administrative or constitutional) and private law. Regarding the private law aspects, it recognises party autonomy and in particular allows the parties to choose the law that would govern their dispute. It also uses some concepts and techniques peculiar to private international law, such as characterisation, choice of law and dépeçage. Unlike in public international law, in investment dispute settlement the municipal and international law apply conjunctly and perform complementary roles. At the same time, in the context of investment treaty arbitration, the role of national law is quite limited. It may govern contractual issues, where contract-based claims are encompassed by an arbitration clause or apply to counterclaims of the host states. Otherwise, if claims are treaty-based, national law is considered because international law makes it relevant and some tribunals even go as far as to limiting its role to the traditional role of a fact. EU law in inter-EU disputes is usually treated by tribunals as part of national law, that is, as a fact, since a different approach would have deprived them from jurisdiction.

136 Z Douglas, ‘The Hybrid Foundations of Investment Treaty Arbitration’ (2003) 74(1) British Yearbook of International Law 151, 152–53. 137 Alvik, above (n 38) 91.

19 International Investment Tribunals and Treaty Interpretation ELISE RUGGERI ABONNAT*

I. INTRODUCTION

I

N THE FIELD of foreign investment arbitration, the role played by the rules on treaty interpretation has become increasingly important. Initially considered to be quasiirrelevant in the context of contract-based arbitration,1 the pertinence of the ‘principles and rules of treaty interpretation generally recognized in international law’2 was first apprehended by ad hoc Committees in charge of annulment proceedings under the International Convention on Settlement of Investment Disputes (ICSID Convention)3 as a gap-filling authority to construe the meaning of the ICSID Convention.4 By opening the path to transnational treaty-based arbitration, the ruling in AAPL v Sri Lanka5 marked a ‘revolutionary step’6 and by the same token, it strengthened the pertinence of the ‘universally accepted rules of treaty interpretation’7 as codified in the Vienna Convention on the Law of Treaties (VCLT). Ever since, a great majority of arbitral awards in this burgeoning field has to deal with the application of International Investment Agreements (IIA). Consequently, international investment tribunals are recurrently called upon to apply the rules of interpretation prescribed by the VCLT. Within the ‘pandora’s box’ of investment cases dealing with treaty interpretation, it seems appropriate to shed some light on landmark cases where arbitrators dedicated substantial developments on the particular role of the VCLT. Beyond the problem of diverging interpretations

* Elise Ruggeri Abonnat, Ph.D, legal advisor (French Ministry for Europe and Foreign Affairs). The author is grateful to Professor L Boisson de Chazournes and Craig Walsh for their valuable comments. 1 Amco Asia Corporation and others v Republic of Indonesia, Award (20 November 1984) ICSID Case No ARB/81/1, para 248. 2 ibid; Ad hoc Committee Decision on the Application for Annulment (16 May 1986) para 18. 3 ibid. 4 ibid: ‘The ad hoc Committee, having been established under the provisions of an international instrument […] believes that the proceedings before it are governed by the relevant Articles of the Convention […]. Problems of interpretation or lacunae which emerge have to be solved or filled in accordance with the principles and rules of treaty interpretation generally recognized in international law.’ 5 Asian Agricultural Products Ltd (AAPL) v Republic of Sri Lanka, Award (27 June 1990) ICSID Case No ARB/87/3. 6 See in particular E Lagrange, ‘6. SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?’, ch 6 in this volume. 7 AAPL v Republic of Sri Lanka, above (n 5) para 38.

312 Elise Ruggeri Abonnat of standards (addressed in this volume8), this chapter will explore whether interpretative approaches have become specific to the investment field9 and whether interpretation ultimately satisfies its function of assisting the resolution of international investment disputes. Through the lens of six decisions, the proposed analysis will present three intertwined developments that interpretation has undergone in the context of treaty-based investment proceedings. First, it will analyse the ways investment tribunals have used the VCLT as a flexible framework to guide the interpretation of IIA provisions (section II). Building on the observation that going beyond the textual approach, the ‘general rule of interpretation’ of the VCLT ‘does not reign supreme’10 and has undergone numerous adaptations, it will then explain why interpretation has become a power under scrutiny (section III), which in turn leads to the ultimate question of whether interpretation can be the subject-matter of an investment dispute per se (section IV).

II. THE VCLT IN TREATY-BASED ARBITRATION: A ‘ONE-SIZE-FITS-ALL’ FRAMEWORK?

Recalling their customary nature on several occasions,11 investment tribunals have recurrently applied the rules of interpretation as codified in the VCLT to construe the terms of many IIAs as well as the previously adopted ICSID Convention.12 Their customary nature means that the interpretative rules are not optional but binding upon international (investment) tribunals. Among the numerous awards that have dealt with treaty interpretation, two cases have opened the path to major reasoning regarding the relationship between the ‘General rule of interpretation’ enshrined in Article 31 of the VCLT and the non-exhaustive supplementary means of interpretation set forth in Article 32 – a recurrent issue of interpretation in investment arbitration.13 Another specific feature of treaty interpretation is that the interpretative framework offered by Articles 31 and 32 is not self-sufficient, and investment arbitration has been a fertile ‘laboratory’14 for diverse interpretative experiments presumably aligned with the object and purpose of the VCLT.15

A. The Relationships between Primary Means of Interpretation (Article 31) and Supplementary Means of Interpretation of the VCLT (Article 32) While having reconciled several approaches to treaty interpretation, Article 31 of the VCLT also ‘expresses the primacy of the text’ in order to find the true meaning of the treaty terms. 8 See in particular in this volume, Major Procedural Issues, chs 14 and 15, and Major Substantial Issues, chs 20, 21, 22, and 23. 9 F Latty, ‘Les techniques interprétatives du CIRDI’ (2011) 115 Revue Générale de droit International Public 459. 10 DB Hollis, The Oxford Guide to Treaties (Oxford and New York, Oxford University Press, 2012) 479. 11 J Romesh Weeramantry, Treaty Interpretation in Investment Arbitration (Oxford, Oxford University Press, 2012) 207; Malaysian Historical Salvors, SDN, BHD v Malaysia, Decision on the Application for Annulment (16 April 2009) ICSID Case No ARB/05/10, para 56 (Hereinafter: MHS v Malaysia). 12 ibid para 56. 13 See Kılıç İnşaat İthalat İhracat Sanayi ve Ticaret Anonim Şirketi v Turkmenistan, Decision on Article VII.2 of the Turkey-Turkmenistan Bilateral Investment Treaty (7 May 2012) ICSID Case No ARB/10/1, para 9.21; Muhammet Çap & Sehil Inşaat Endustri ve Ticaret Ltd Sti v Turkmenistan, Decision on Respondent’s Objection to Jurisdiction under Article VII (2) (13 February 2015) ICSID Case No ARB/12/6, para 247. 14 L Boisson de Chazournes, ‘Plurality in the Fabric of International Courts and Tribunals: The Threads of a Managerial Approach’ (2017) 28 European Journal of International Law 13. 15 A Pellet, ‘Canons of interpretation under the Vienna Convention’ in J Klingler, Y Parkhomenko and C Salonidis (eds), Between the Lines of the Vienna Convention? Canons and Other principles of Interpretation in Public International law (Alphen aan den Rijn, Kluwer Law International, 2019) 21.

International Investment Tribunals and Treaty Interpretation 313 As its name implies, a textual approach gives prevalence to the text, whereas the teleological approach would give prevalence to the treaty’s overall object and purpose using, if necessary in accordance with Article 31(3), subsequent agreements and practice between the same parties, and any relevant rules of international law as other means of interpretation. The approach established by Article 31 of the VCLT commands the interpreter (‘shall’) to use several elements to evidence the meaning of the term (text, object purpose and context of the terms) before seeking recourse to the external elements surrounding the conclusion of a treaty that are listed in Article 32 (‘recourse may be had to supplementary means’), such as preparatory works or the circumstances of the treaty’s conclusion. In other words, Article 32 of the VCLT allows international tribunals to use supplementary means of interpretation only when a term remains ‘ambiguous or obscure’, when the result is ‘manifestly absurd and unreasonable’, or simply to confirm the result reached by application of Article 31. It requires that the interpretative aids of Article 32 shall be applied after Article 3116 thus providing a ‘step-by-step’17 method. However, as shown in the following decisions, in using the rules of interpretation arbitrators have been criticised for ‘postulat[ing] an outcome and force-fit[ting] it into the actual text’18 the practice of which would suggest that the VCLT is a one-size-fits-all framework for interpretation. i. Decision on the Treaty Interpretation Issue Hrvatska Elektroprivreda D.D. v Republic of Slovenia: Diverging Methodologies of Interpretation to Imply a Term in a Treaty The Decision on the treaty interpretation Issue Hrvatska Elektroprivreda d.d. v Republic of Slovenia19 including the appended Individual Opinion of Jan Paulsson is considered to be a major decision, in that it opened the path to major reasoning on the relationship between Articles 31 and 32 VCLT. It also illustrates that the use of the interpretive rules of the VCLT can be the subject matter of arbitral awards and a source of disagreements among the members of a same tribunal. The dispute was initiated on the basis of the Energy Charter Treaty and arose out of the implementation of an agreement between the newly-independent states of Croatia and Slovenia regarding the delivery of electricity to a jointly managed nuclear power plant (‘the Agreement’). The Claimant claimed damages based on the non-delivery of electricity by Slovenia for a certain period of time following the signature of the Agreement. The issue was whether the Agreement contained an undertaking to deliver electricity and whether a state could be held to owe an obligation by virtue of an implied term introduced into a treaty. Applying the VCLT rules of interpretation, the majority considered that an obligation to resume electricity delivery from a certain date could be implied from the terms of the Agreement on the basis that Articles 31 and 32 VCLT ‘do not categorize treaty provisions as

16 P

Reuter, Introduction to the Law of Treaties, 2nd edn (London and New York, Kegan Paul International, 1995)

97. 17 L

Boisson de Chazournes, ‘Rules of Interpretation and Investment Arbitration CEMEX v Venezuela, ICSID Case No ARB/08/15’ in M Kinnear and others (eds), Building international investment law: the first 50 years of ICSID (Alphen aan den Rijn, Kluwer Law International, 2015) 7. 18 Hrvatska Elektroprivreda dd v Republic of Slovenia, Decision on the Treaty Interpretation Issue (12 June 2009) ICSID Case No ARB/05/24 (Judges C Brower, J Paulsson, D Williams) (Hereinafter: HPP v Slovenia). 19 ibid para 1.

314 Elise Ruggeri Abonnat being either “express” or “implied” [and] [n]o greater or lesser force resides in a term by virtue of the relative magnitude of the clarity with which it has been (or has not been) written’.20 Relying on the ‘context’ as an element of interpretation listed in Article 31(2) VCLT, the majority used a Joint Statement signed by both states parties stating that they should use their best efforts to achieve the ratification of the 2001 Agreement ‘as soon as possible during the first quarter of 2002’.21 According to the majority, this Joint Statement, as a contextual statement, confirmed the textual interpretation of the implied term it found in the Treaty in accordance with Article 31(1) VCLT. The majority further used the supplementary means of interpretation and referred to the travaux préparatoires of the Agreement to imply the existence of an obligation to deliver electricity before the date of entry into force of the treaty, under both qualifying elements of Article 32 VCLT, namely the existence of an ‘ambiguity’ or the need to ‘confirm’ the interpretation made under Article 31.22 In his Individual opinion, the state-appointed arbitrator ‘felt impelled to dissent’23 with the majority’s conclusion which was deemed an ‘impermissible rewriting of the treaty’24 and ‘appear[ed] to turn the VCLT on its head’.25 In particular, he reproached the majority for its ‘desire to reach a result that [it] consider[ed] fair and reasonable lead[ing] to imply terms that are not in the Treaty, to ignore terms that are in the Treaty, and to give retroactive effect to a Treaty when neither its express terms nor its object require retroactivity.’26 His criticisms focused on the misapplication of the rules of interpretation under both Articles 31 and 32. With regard to Article 31, the Opinion criticised the majority for having given an extensive role to the ‘context’ although the various interpretation aids listed in Article 31, as noted by an international tribunal, should be viewed as ‘an integral whole, the constituent elements of which cannot be separated’.27 It added that the VCLT limits the context to ‘the terms of the treaty and not the context of the treaty generally’.28 The Opinion further reproached the majority for having ‘rel[ied] to a great extent on various extrinsic expressions of the Contracting States’ intentions’29 and the subsidiary means of interpretation, although the terms of the Agreement did not leave any ambiguity regarding the date from which the parties consented to create inter se obligation. A method which sought to ‘reverse-engineer from the desire outcome’.30 20 ibid

para 176. para 136. 22 ibid para 183: ‘The result arrived at above by applying Article 31 (…) is confirmed on further investigation pursuant to Article 32, which permits reference to the travaux of a treaty in order to confirm the meaning resulting from the application of Article 31. Alternatively, should application of Article 31 be seen as leaving the meaning “ambiguous or obscure,” or, indeed, “manifestly absurd or unreasonable,” resort to Article 32 produces the same result …’. 23 HPP v Slovenia, above (n 18), Individual Opinion of J Paulsson (Decision on the Treaty Interpretation Issue), para 1. 24 ibid. 25 ibid para 41. 26 ibid para 5. 27 Case concerning the Auditing Accounts between the Kingdom of The Netherlands and the French Republic pursuant to the Additional Protocol of 25 September 1991 to the Convention on the Protection of the Rhine against Pollution by Chlorides of 3 December 1976 (Netherlands v France), Award (12 March 2004) PCA 25 RIAA 267, para 62. 28 ibid para 44. 29 HPP v Slovenia, above (n 18), Individual Opinion of J Paulsson (Decision on the Treaty Interpretation Issue), para 32. 30 ibid para 7. 21 ibid

International Investment Tribunals and Treaty Interpretation 315 As a consequence, contrary to the majority, the dissent considered that ‘a term may be implied only when it is clear beyond peradventure, from the overall text of the relevant instrument, its negotiating history or its actual implementation by the parties, that all Contracting States would have had no hesitation to include that term’.31 This vivid disagreement regarding the use of Articles 31 and 32 illustrates how investment tribunals can introduce a dose of subjectivity into the methodology of interpretation resulting in creating obligations outside the terms of a given treaty. It bears the risk of searching the common intent of the parties outside the treaty, thus diluting the textual meaning of a term from which it derives. However, ‘if terms are to be implied that have not been explicitly included in the treaty, it is important to have a degree of certainty that the omission was unintentional’.32 The interactions of Articles 31 and 32 VCLT can also play a decisive role in overcoming an ambiguity in an IIA. Interestingly, the principles of the interpretation expressed in the minority view in Decision on the treaty interpretation Issue HPP v Slovenia were adopted by the majority in the Daimler v Argentina case and led again to a dissent that criticised the majority for (in that dissenting arbitrator’s opinion) mis-reading and mis-applying the VCLT.33 ii. Daimler Financial Services AG v Argentine Republic: Diverging Methodologies of Interpretation to Overcome an Ambiguous Term In Daimler Financial Services AG v Argentine Republic,34 the VCLT rules of interpretation were applied to address ‘one of the most controversial issues … in the world of international investment law’,35 ie the procedural use of the Most Favoured Nation (MFN) clause.36 From the standpoint of interpretation, the Daimler case stands as a leading decision because it represented a significant change of direction compared to previous decisions on the MFN clause through the use of soft law instruments, external to the conclusion of the Argentina-Germany Bilateral Investment Treaty (BIT). Applying the rules of interpretation, the majority found that the MFN clause of the BIT between Germany and Argentina could not be invoked to circumvent the local exhaustion of remedies requirement and was limited to covering substantial treatment. Adopting a ‘prudent’ approach, the majority prefaced its legal analysis with general observations regarding the interpretive principles flowing from the bilateral nature of a BIT, warning against the risk of ‘speculations’, ‘presumptions’ and ‘hypothetical intentions to be found in consideration external to the text’.37 The majority explained that a good faith interpretation of the terms ‘reinforces the duty of tribunals to limit themselves to interpretations falling within the bounds of the framework mutually agreed to by the contracting state parties’.38 Contrary to the reasoning of the

31 ibid

para 64. Weeramantry, above (n 11) 203. 33 The dissenting arbitrator in Daimler Financial Services AG v Argentine Republic, Dissenting Opinion (22 August 2012) ICSID Case No ARB/05/1 (hereinafter: Daimler v Argentina) was also a member of the Tribunal in HPP v Slovenia, above (n 18). 34 Daimler v Argentina, above (n 33), Award (22 August 2012) (P-M Dupuy, C Brower, D Bello Janeiro). 35 ibid para 160. 36 See G Sacerdoti and JN Moran, ‘Non-Discrimination Clauses: Most-Favoured-Nation (MFN) and National Treatment (NT)’, ch 25 in this volume. 37 ibid para 166. 38 ibid para 173. 32 Romesh

316 Elise Ruggeri Abonnat majority in HPP v Slovenia, it treated the notion of consent as the outer limits of the interpretive mandate39 declaring that ‘[n]on-consent is the default rule; consent is the exception’.40 Following the ‘classic rule of interpretation known as the principle of contemporaneity’41 requiring that the meaning of a term be ascertained at the time the BIT was concluded, the majority ‘look[ed] for clues to the meaning generally ascribed to the term [treatment] by the broader international community of States at the time’,42 observing that the scope of the MFN provision was ‘entirely unexplored’ in the context of treaty-based arbitration. Noting the absence of ‘direct evidence’43 flowing from the drafting history of the BIT, the Tribunal referred to the World Bank Guidelines of Foreign Direct Investment adopted in 1992 as a ‘soft law instrument’ with no binding effect, which ‘nevertheless provide[s] an indication of the prevailing view among the community of states during the period contemporaneous to the adoption of the German-Argentine BIT’.44 In light of this instrument, the majority observed that the ‘prevailing view was that treatment was meant to cover discrete principles of conduct’45 and not to touch upon the settlement of a dispute. The majority did not, however, explicitly refer to the World Bank Guidelines as a supplementary means of interpretation under Article 32. It is worth recalling in this respect, that contrary to the interpretative list of Article 31, Article 32 does not provide an exhaustive list, giving tribunals more latitude to pick and choose among international instruments capable of determining the meaning of a term. However, by exclusively relying on a non-binding instrument to construe the term under scrutiny, the Daimler case adopts an innovative position as to the content of Article 32 VCLT. The majority’s approach, consisting as it did of seeking ‘affirmative evidence’ of consent was strongly criticised by judge Charles N Brower in his dissenting opinion as ‘not simply unconvincing but profoundly wrong’.46 The main reproach addressed to the majority’s interpretive reasoning related to the omissions of ‘mandatory steps’ under the VCLT, that ‘led it to ascribe such great significance to materials so marginally related to the Treaty at issue’47 (ie the 1992 World Bank Guidelines). A cross-reading of the majority and dissent’s reasonings in Daimler exposes the challenge of drawing a sharp line between ‘the establishment of consent to be bound by a specific dispute resolution mechanism and the scope of that consent’.48 Whereas the majority seemed to consider that the rules of interpretation cannot be used to broaden the scope of consent beyond the terms included in the treaty, the minority view, following a similar reasoning as in HPP v Slovenia, considers such approach to be restrictive and unjustified under the VCLT.

39 ibid para 166: ‘as a matter of public international law, the uniform applicability of the Vienna Convention’s customary law interpretive principles to all treaty clauses is beyond doubt. (…) In interpreting dispute resolution provisions in BITs (…) the ultimate goal is to determine what the contracting parties actually consented to. Thus, the fact that dispute resolution clauses should be construed neither liberally nor restrictively does not authorize international tribunals to interpret such clauses in a manner which exceeds the consent of the contracting parties as expressed in the text. To go beyond those bounds would be to act ultra vires.’ (emphasis added) 40 ibid para 175. 41 ibid para 220. 42 ibid. 43 ibid. 44 ibid para 224. 45 ibid para 222. 46 Daimler v Argentina, above (n 33) Dissenting Opinion, para 38. 47 ibid para 18. 48 Daimler v Argentina, above (n 33) Award, para 175.

International Investment Tribunals and Treaty Interpretation 317 At this juncture, the following observations can be made regarding the interactions of Articles 31 and 32 in investment arbitration – both decisions show that investment tribunals may take different stances on the question of whether the ambiguity arrived at by way of Article 31 may be overcome by the non-exhaustive list of supplementary means of interpretation to determine the meaning.49 The majority in HPP v Slovenia accepts that the general rule of interpretation of the VCLT finds application even when a term is absent from the treaty. Article 31 therefore can be used to imply an obligation which is not explicitly formulated in the instrument under scrutiny. As a consequence, the use of supplementary means of interpretation is equally allowed to ‘confirm’ or alternatively ‘determine’ the existence of a (non-written) obligation. To the contrary, the majority in Daimler, and minority in HPP v Slovenia consider that the absence of ‘affirmative evidence’ of a term should be taken at its face value and prevents interpreters from using the VCLT to rewrite the agreement. Focusing on the ‘optional’ nature of Article 32, this approach tends to rein in the excessive use of the supplementary means that would result in ‘adding’ or broaden the scope of the state’s consent. In short, one aspect to be borne in mind is that Article 32 should not be treated as an autonomous method of interpretation.50 This approach is aligned with the function of treaty interpretation as described in clear terms by M Yasseen: ‘la méthode d’interprétation doit viser à exercer une fonction déclarative et non créatrice’.51 Thus, the use of the rules of interpretation should prevent the risk of projecting onto the parties a common intention which is absent from the text. It can be argued in this regard that the inclusion of the term ‘for the avoidance of doubt’52 in a treaty provision, is precisely intended to limit the Tribunals’ use of supplementary means of interpretation in the sense that it aims at excluding a potential ‘ambiguity’ from the text.

B. The Interpretative Methods of the VCLT and Beyond (Mondev International Ltd. v United States of America) Investment tribunals have proven that the framework of the VCLT is not rigid and arbitrators have benefited from a certain latitude in construing the terms of wide-ranging IIA provisions, which are in constant evolution. For example, going one step too far, some tribunals have gone outside the terms of the VCLT postulating that, within investor-state dispute settlement (ISDS), the methodology of interpretation be guided by a presumption in favour of the investor.53 Such practice is questionable as it ‘may serve[s] to dissuade host States from admitting foreign investments and

49 See

on this point Romesh Weeramantry, above (n 11) 111. Mbengue, ‘Rules of Interpretation (Article 32 of the Vienna Convention on the Law of Treaties)’ (2016) 31 ICSID Review 388, 399. 51 MK Yasseen, ‘L’interprétation des traités d’après la Convention de Vienne sur le droit des traités’ (1976) 151 Collected Courses of the Hague Academy of International Law 11. 52 See, eg, the MFN provision of the Agreement between the Government of the United Kingdom of Great Britain and Northern Ireland and the Government of Ukraine for the Promotion and Reciprocal Protection of Investments (adopted and entered into force 10 February 1993) 1993 UKTS 24 Art 3. 53 SGS Société Générale de Surveillance SA v Republic of the Philippines, Decision of the Tribunal on Objections to Jurisdiction (29 January 2004) ICSID Case No ARB/02/6, para 116. 50 MM

318 Elise Ruggeri Abonnat so undermine the overall aim of extending and intensifying the parties’ mutual economic relations’.54 However, the rules of the VCLT are not exhaustive. Some interpretative principles are considered to be implied by the VCLT, as is the case for the principle of effet utile.55 The status of the evolutionary principle under the VCLT remains controversial. It was argued that the principle can also be considered as a subdivision of the general rule of interpretation56 but in practice, it remains ‘a judicial taboo’.57 Indeed, it seems difficult to presume ‘a situation of evolutionary interpretation where no mention is made of it’.58 In contrast to Daimler v Argentina, in which pre-eminence was given to the principle of ‘contemporaneity’ in order to construe the meaning of the term ‘treatment’, the reasoning followed by the Tribunal in Mondev International Ltd. v United States of America59 stands out for its innovative position on the meaning to be ascribed to the Fair and Equitable Treatment (FET) provision of the North Atlantic Free Trade Agreement (NAFTA). Relying on the two constituent elements of the evolutionary interpretation (‘[t]ime and change’60), the Tribunal in Mondev construed the meaning of FET in light of its evolution, and concluded that it ‘cannot be limited to the content of customary international law as recognized in arbitral decisions in the 1920s’.61 Applying the evolutionary principle, the Tribunal observed that the FET ‘is intended to provide a real measure of protection of investments … having regard to its general language and to the evolutionary character of international law, it has evolutionary potential’.62 Relying on the travaux préparatoires and the origins of the NAFTA, the Tribunal added that ‘reasonable evolutionary interpretation of Article 1105(1) is consistent both with the travaux, with normal principles of interpretation …’.63 It is unclear whether investment tribunals have made the VCLT evolutionary, or whether it is the flexibility of the VCLT which opens the way to the application of an evolutionary rule of interpretation. Either way, the prospect that interpretation can be used to broaden the scope of IIA provisions, thereby endowing investment tribunals with a rather unfettered discretion, raises a broader question pertaining to the links, if any, between legal interpretation and competence. This raises the question of whether interpretation brings more order in the application of IIA.64

III. INTERPRETATION AS A POWER UNDER SCRUTINY?

The idea that investment tribunals are given leeway to expand or restrict the meaning of IIA terms, sometimes resulting in inconsistent outcomes, has fostered states to regain control over 54 Saluka

Investments BV v The Czech Republic, Partial Award (17 March 2006) PCA Case No 2001–04, para 300. above (n 51) 74. 56 V Boré Eveno, L’interprétation des traités par les juridictions internationales. Etude comparative (Université Paris 1 Panthéon-Sorbonne, 2004) 187; T Gazzini, Interpretation of International Investment Treaties (Oxford, Hart, 2016) 107. 57 MM Mbengue and A Florou, ‘Evolutionary Interpretation in Investment Arbitration: About a Judicial Taboo’ in G Abi-Saab and others (eds), Evolutionary Interpretation and International Law (Oxford, Hart, 2019) 253. 58 G Marceau and C Marquet, ‘Introduction’ in Abi-Saab and others, above (n 57) 5. 59 Mondev International Ltd v United States of America, Award (11 October 2002) ICSID Case No ARB(AF)/99/2, (N Stephan, SM Schwebel, J Crawford). 60 R Kolb, ‘Evolutionary Interpretation in International Law: Some Short and Less than Trail-Blazing Reflections’ in Abi-Saab and others, above (n 57) 16. 61 Mondev International Ltd v United States of America, above (n 59) para 123. 62 ibid para 119. 63 ibid para 123. 64 Mbengue and Florou, above (n 57) 263. 55 Yasseen,

International Investment Tribunals and Treaty Interpretation 319 interpretation. To achieve this goal, states, as disputing or non-disputing parties, have tried to bring restrictions to investment tribunals’ interpretative mandates by issuing interpretative notes (see section A below). In parallel, interpreters have suggested that failure to apply the rules of interpretation properly, may influence the annulment of an arbitral award (see section B below). Together, these factors are indicative that the use (and misuse) of the rules of interpretation is increasingly under scrutiny.

A. Interpretation as a Shared-Power (Pope & Talbot v Government of Canada) One illustration of an attempt to regain control over interpretation is to be found in the Pope & Talbot case.65 It counts as a leading decision in investment arbitration because it is one of the first occasions in which states parties to an investment treaty adopted a common interpretative declaration with respect to the provisions on National Treatment (Articles 1102 (2)) and Minimum Standard of Treatment (Article 1105) of the NAFTA, while an investor-state proceeding was already underway between Canada and an American investor. In its declaration, the Free Trade Commission (FTC) of the NAFTA, composed of representatives of state parties, declared that the concepts of FET and full protection and security ‘do not require treatment in addition to or beyond that which is required by the customary law minimum standard of treatment of aliens’.66 In light of this Declaration, the Pope & Talbot Tribunal had to determine ‘the extent to which conduct by Canada subsequent to the Investor’s Statement of Claim should be subject to the jurisdiction of the Tribunal’.67 Therefore, the Tribunal engaged in a comprehensive written dialogue with the states’ parties to the NAFTA about the effects of the FTC Declaration over the previous ruling of the Tribunal regarding the content of the FET. Those exchanges shed light on the nature and the effect of interpretive Declarations. Canada argued that ‘notes of interpretation are not amendments to the provisions of the treaty … they state not what the provisions of the Treaty are to mean in the future, but what they always have meant … (they) do not operate to abrogate prior ruling of tribunals. Rather they constitute directions to tribunals’.68 The binding effect of the FTC interpretation was also underlined by Mexico. It declared that the FTC Declaration ‘only clarified the proper interpretation of Article 1105 [and] that the Tribunal has no choice but to find that the interpretation stated the law as it existed at the time of the ruling … [and] must hold that its earlier ruling is “inconsistent with” or “contrary to” the interpretation of the Commission’.69 In two separate rulings, the Tribunal considered that the binding nature of the FTC Declaration was limited to the future and not to the past70 and it would be ‘against elementary rules of due process of justice to compel it to revisit its determination’.71 Applying the

65 Pope & Talbot v Government of Canada, Award on the Merits of Phase 2 (10 April 2001) (hereinafter: Pope & Talbot); Pope & Talbot, Award in Respect of Damages (31 May 2002) (L Dervaird, M Belman, J Greenberg). The presence of Meg Kinnear as representative of Canada is noteworthy. 66 ibid Respondent’s Letter to the Tribunal re Interpretation of NAFTA Chapter 11 Provisions adopted by the NAFTA Free Trade Commission (10 August 2001). 67 ibid Award on the Merits of Phase 2 (10 April 2001) para 2. 68 ibid Canada’s submission re implications of the interpretation of NAFTA article 1105 by the NAFTA Commission (10 September 2001) 2. 69 ibid Mexico’s submission in response to the Tribunal’s letter of October 23rd 2001 (6 November 2001) para 7. 70 ibid Award in Respect of Damages (31 May 2002) para 50. 71 ibid.

320 Elise Ruggeri Abonnat evolutionary principle, it rejected Canada’s ‘static conception’ of the customary standard of minimum protection because there had been ‘an evolution of customary concepts’,72 and observed that several treaties had added an element of ‘fairness’ to the ‘international law minimum’.73 The Tribunal thus concluded that ‘investors under NAFTA are entitled to the international law minimum, plus the fairness elements’ (emphasis added).74 Despite the FTC’s note of interpretation, the Tribunal rejected Canada’s contention regarding the current status of customary international law concerning the protection of foreign property, and proposed ‘a formulation more in keeping with the present practice of states’.75 Whereas the Pope & Talbot case illustrates that interpretation is a shared power between tribunals and states (the latter as ‘masterminds’76 of the treaty), it also shows that the principle of good faith is the outer limit of the state’s power to offer its interpretative views after the adoption of a treaty during an ongoing proceeding. There is a high risk that Member States to a Treaty will use their interpretative power to ‘rewrite’ its provisions. A good faith interpretation then, may lead investment tribunals to second guess whether states’ treaty parties’ interpretative declarations seek to offer a binding interpretation of the meaning of an IIA for the future77 following an evolutionary approach, or constitute an authentic interpretation of the Agreement78 at the time it was made under the principle of contemporaneity. In any case, the Pope & Talbot Tribunal confirms that interpretative declarations from states cannot be considered as an exclusive and dispositive method of treaty interpretation, since they are part of the context under Article 31 VCLT, which, again, does not rank means of interpretation in order of importance. Therefore, contrary to amendments, interpretative declaration ‘can only interpret the treaty terms [but] it cannot change their meaning’.79 Thus, it comes as no surprise that states are increasingly including specific notes of interpretation to their IA80 so as to limit arbitral tribunals’ use of interpretation as law-making power.

B. On the Interaction Between Interpretation, Jurisdiction, and Annulment Under the ICSID Convention (Malaysian Historical Salvors, SDN, BHD v Malaysia) In the absence of an appeal mechanism, the review of arbitral awards under Article 52 of the ICSID Convention is a limited remedy which excludes ‘the review of an award on the merits’.81

72 ibid

para 59. para 60. 74 ibid Award on the Merits of Phase 2 (10 April 2001) para 110. 75 ibid Award in Respect of Damages (31 May 2002) para 60. 76 Magyar Farming Company Ltd, Kintyre Kft and Inicia Zrt v Hungary, Award (13 November 2019) ICSID Case No ARB/17/27, para 212. 77 See the Financial Stability, Financial Services and Capital Markets Union, Declaration of the Representatives of the Governments of the Member States on the Legal Consequences of the Judgment of the Court of Justice in Achmea and on Investment Protection in the European Union (15 January 2019). 78 Theodoros Adamakopoulos, Ilektra Adamantidou, Vasileios Adamopoulos and others v Republic of Cyprus, Statement of Dissent of Professor Marcelo G Kohen, ICSID Case No ARB/15/49, para 55. 79 Magyar Farming Company Ltd, Kintyre Kft and Inicia Zrt v Hungary, above (n 76) para 218. 80 See T Hai Yen, The Interpretation of Investment Treaties (Leiden, Brill Nijhoff, 2014) 203–256; See also Joint Interpretative Instrument on the Comprehensive Economic and Trade Agreement (CETA) between Canada and the European Union and its Member States [2016] OJ L11/3. 81 Maritime International Nominees Establishment v Republic of Guinea, Decision for Partial Annulment of the Arbitral Award (22 December 1989) ICSID Case No ARB/84/4, para 4.05 (S Sucharitkul, K Mbaye, A Broaches). 73 ibid

International Investment Tribunals and Treaty Interpretation 321 At first glance, one might assume that failure to apply the rules of interpretation could be used to influence annulment of an award on the final ground of Article 52;82 namely, ‘failure to state reasons on which the award is based’. However, the first ICSID ad hoc Committee on which A Broches sat, held that ‘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. … even if it made an error of fact or of law’,83 and as a result, any ambiguity left in the arbitral award reasoning will be treated ‘in favorem validitatis sententiae’.84 Thus, failure to properly apply the rules of interpretation is not considered a ground for annulment per se. Indeed, if an Annulment Committee were to annul an award on the basis that ‘… its understanding of facts or interpretation of law … is different from that of the tribunal, it would cross the line that separates annulment from appeal’.85 In contrast, the ruling followed by the ICSID annulment Committee in Malaysian Historical Salvors, Sdn, Bhd v Malaysia,86 suggests that refusal to take into account subsidiary means of interpretation can influence the decision to annul an award on the ground of manifest excess of powers under Article 52(1)(b) of the ICSID Convention. In this respect, it operates as a departure from previous decisions rendered by ICSID ad hoc Committees.87 In the award under the Committee’s scrutiny, the sole arbitrator considered that for an investment to be protected under the ICSID Convention, it was necessary that it made a significant contribution to the economic development of the home state.88 It is in light of this element that it dismissed jurisdiction on that ground as the investment ‘did not benefit the Malaysian public interest … in the sense developed by ICSID jurisprudence, namely that the contributions were significant.’89 The ad hoc Committee was asked to determine whether the original tribunal’s failure to take into account the definition of the term ‘investment’ under the applicable BIT and the preparatory work of the ICSID Convention constituted a manifest excess of powers. From the outset, the ad hoc Committee observed that ‘this case concerns the interpretation of treaties’,90 and that although not applicable as such to the ICSID Convention, the customary nature of the VCLT offered ‘firm ground’91 on which it could rely in conducting the annulment proceeding. It considered that the original Tribunal exceeded its power ‘by failing to exercise the jurisdiction’,92 and that it ‘manifestly’ did so for the reason that its conclusion was ‘not

82 The grounds for annulment are the following: ‘that the Tribunal was not properly constituted; that the Tribunal has manifestly exceeded its powers; that there was corruption on the part of a member of the Tribunal, that there has been a serious departure from a fundamental rule of procedure; or that the award has failed to state the reasons on which it is based.’ 83 Maritime International Nominees Establishment v Republic of Guinea, above (n 81) para 5.09. 84 Klöckner Industrie-Anlagen GmbH and others v United Republic of Cameroon and Société Camerounaise des Engrais, Decision of the Ad Hoc Committee (3 May 1985) ICSID Case No ARB/81/2, 165 (P Lalive, AS El-Kosheri, I Seidi-Hohenveldern). 85 Daimler v Argentina, above (n 33) Decision on annulment (7 January 2015) para 186. 86 Malaysian Historical Salvors, SDN, BHD v Malaysia, above (n 11) Decision on the Application for Annulment (16 April 2009) (S Schwebel, P Tomka, M Shahabuddeen). 87 Fraport AG Frankfurt Service Worldwide v Republic of the Philippines, Decision on the Application for Annulment (23 December 2010) ICSID Case No ARB/03/25, para 113; Daimler v Argentina, above (n 33) Decision on annulment, para 186. 88 MHS v Malaysia, above (n 11) Award on Jurisdiction (17 May 2007) (Sole Arbitrator: M. Hwang) (emphasis added). 89 ibid para 131. 90 ibid Decision on the Application for Annulment (16 April 2009) para 56. 91 ibid para 56. 92 ibid para 74.

322 Elise Ruggeri Abonnat consonant with the travaux in key respects’,93 because the drafters of the ICSID Convention rejected ‘a monetary floor in the amount of an investment’ and left the notion undefined.94 Following a teleological approach, the ad hoc Committee further observed that refusal to exercise jurisdiction is a result ‘difficult to reconcile with the intentions of [the Parties]’95 if the consent to arbitrate investment disputes ‘could be rendered nugatory by a restrictive definition of a deliberately undefined term of the ICSID Convention’.96 Relying on the travaux préparatoires of the ICSID Convention, the Committee found that ‘the failure of the Sole Arbitrator even to consider, let alone apply, the definition of “investment” as it is contained in the Agreement to be a gross error that gave rise to a manifest failure to exercise jurisdiction’.97 It is worth noting that in reaching this conclusion, the Committee had itself departed from the interpretative methodology of the VCLT, since it considered that ‘it is mythological to pretend that [recourse to travaux préparatoires occurred] only when [courts and tribunals] first conclude that the term requiring interpretation is ambiguous or obscure’.98 This statement, when viewed within its larger ‘context’, is somewhat surprising, since it was formulated by two International Court of Justice judges.99 The MHS annulment award illustrates how interpretation may influence a decision to annul an award. It is equally interesting to observe that the interpretative reasoning of arbitral awards can be filtered through the lens of a ‘manifest excess of power’. This ground for annulment confirms that ‘the process of interpretation […] cannot be regarded as a mere mechanical one’100 and may be the result of a subjective exercise of ‘power’ rather than an ‘objective duty’. The existence of annulment mechanisms also raises an interesting question as to whether investment tribunals’ (over) reliance on investment case law resulted in bringing a hierarchy to the precedential value of arbitral awards.101 All in all, correct application of the rules of interpretation is a growing concern for states’ parties to IIAs and ad hoc Committees, as well as to national jurisdictions empowered to set-aside arbitral awards.102 The important role played by interpretation in determining the

93 ibid. 94 ibid. 95 ibid

para 57.

96 ibid. 97 ibid

para 74. para 57. v Malaysia, above (n 11) Dissenting Opinion of Judge Mohamed Shahabuddeen, para 14. 100 Article 19 Interpretation of Treaties (1935) 29 American Journal of International Law 937, 946. 101 H Ruiz Fabri and L Gradoni, ‘La hiérarchisation des précédents’ in Le précédent en droit international (Paris, Pedone, 2016) 191; See also AES Corporation v The Argentine Republic, Decision on Jurisdiction (26 April 2005) ICSID Case No ARB/02/17, para 33 (P-M Dupuy, K-H Böckstiegel, D Bello Janeiro): ‘From a more general point of view, one can hardly deny that the institutional dimension of the control mechanisms provided for under the ICSID Convention might well be a factor, in the longer term, for contributing to the development of a common legal opinion or “jurisprudence constante”.’ 102 Metalclad Corporation v The United Mexican States, Decision of the Supreme Court of British Columbia on the challenge by the Petitioner (2 May 2001) ICSID Case No ARB(AF)/97/1, para 65: ‘With respect, I am unable to agree with the reasoning of the Pope & Talbot tribunal. It has interpreted the word “including” in Article 1105 to mean “plus”, which has a virtually opposite meaning. Its interpretation is contrary to Article 31(1) of the Vienna Convention, which requires that terms of treaties be given their ordinary meaning. The evidence that the NAFTA Parties intended to reject the “additive” character of bilateral investment treaties is found in the fact that they chose not to adopt the language used in such treaties and I find it surprising that the tribunal considered that other evidence was required. The NAFTA Parties chose to use different language in Article 1105 and the natural inference is that the NAFTA Parties did not want Article 1105 to be given the same interpretation as the wording of the provision in the Model Bilateral Investment Treaty of 1987.’ 98 ibid

99 MHS

International Investment Tribunals and Treaty Interpretation 323 scope of investment tribunals’ jurisdiction has been recognised by states, as the authors of IIAs, who are increasingly seeking to regain control over interpretation. It thus comes as no surprise that among the numerous proposals to reform the ISDS regime, many are aimed at ensuring consistency of interpretation through the acceptance of submissions from state treaty parties,103 or through the creation of a permanent arbitration court or an advisory centre.104 The likelihood that rules of interpretation will increasingly be included in IIAs or made subject to scrutiny raises the ultimate question of whether interpretation can itself constitute ground for an arbitrable dispute.

IV. INTERPRETATION AS A LEGAL DISPUTE PER SE? (REPUBLIC OF ECUADOR V UNITED STATES OF AMERICA, PCA CASE NO 2012–05, AWARD, 29 SEPTEMBER 2012)

The interstate arbitration Republic of Ecuador v United States of America, raises an interesting question as to whether legal interpretation of an IIA may itself become the subject-matter of an IIA dispute. This case counts as a reference decision in international investment law since it is one of the rare investment cases initiated at the inter-state level, and the only one in which a state treaty party requested an arbitral tribunal to ‘answer an abstract question of interpretation’.105 Following Chevron Corporation and Texaco Petroleum Company v The Republic of Ecuador106 in which the Tribunal found Ecuador to be in violation of Article II(7) of the Ecuador-US BIT, Ecuador sent a Diplomatic Note to the US offering its interpretation of Article II(7), which provides that ‘Each Party shall provide effective means of asserting claims and enforcing rights with respect to investment, investment agreements, and investment authorizations.’107 Ecuador was in disagreement with the interpretation offered by the Chevron arbitral tribunal, that Article II(7) ‘constituted lex specialis and not a mere restatement of the law on denial of justice’.108 In its Diplomatic Note, Ecuador contended inter alia that the term ‘effective means’ included in Article II(7) of the Treaty could not be interpreted as going beyond the standards of customary international law’. Ecuador’s Diplomatic Note asked the US for formal agreement with Ecuador’s interpretation of Article II(7) of the Treaty. However, the US eventually ceased communication, and never communicated its interpretative position.

103 See ICSID, ‘Proposal for Amendments of the ICSID Rules’ (2019) 1(3) Working Paper 158, rule 77 ‘Participation of Non-Disputing Treaty Party’, https://perma.cc/7RLV-AFTM. 104 United Nations Commission on International Trade Law, Report of Working Group III (Investor-State Dispute Settlement reform) on the work of the thirty-eighth session, A/CN.9/1004, Vienna, 14–18 October 2019, para 28. 105 Republic of Ecuador v United States of America, Award (29 September 2012) PCA Case No 2012–05, para 196 and Dissenting Opinion of Professor RE Vinuesa (Tribunal: RE Vinuesa, D McRae, LO Baptista). 106 Chevron Corp and Texaco Petroleum Co v Republic of Ecuador, Partial Award on the Merits (30 March 2010) PCA Case No 2007–2, UNCITRAL Rules 1976, para 262. 107 Treaty between the United States of America and the Republic of Ecuador concerning the Encouragement and Reciprocal Protection of Investments (adopted on 27 August 1993, entered into force 11 May 1997) TIAS 97-511 (Ecuador-United States BIT). 108 Ecuador-United States BIT, Art VII(1) provides: ‘Any dispute between the Parties concerning the interpretation or application of the Treaty which is not resolved through consultations or other diplomatic channels, shall be submitted, upon the request of either Party, to an arbitral tribunal for binding decision in accordance with the applicable rules of international law.’

324 Elise Ruggeri Abonnat As a result, Ecuador resorted to state-to-state dispute resolution, alleging inter alia that as a result of fruitless attempts at consultations regarding the interpretation of the BIT, a dispute existed between the states’ treaty parties thus requiring the arbitral tribunal to give a binding interpretation of the treaty.109 The US objected to the tribunal’s jurisdiction, alleging that the parties failed to consent to arbitrate interpretative disputes which did not relate to an actual breach of performance of their obligations pursuant to the treaty. It further alleged that Ecuador failed to satisfy the ‘two essential elements necessary to establish the existence of a dispute under international law: concreteness and positive opposition’.110 The arbitral Tribunal was asked to determine whether a dispute relating to interpretation of an investment treaty could arise when there was no allegation of a violation of a treaty, and whether jurisdiction could be exercised to provide a binding interpretation of it. From the outset, the Tribunal remarked on the novelty111 of the case, but eventually dismissed jurisdiction for lack of dispute. Regarding the question of whether the US’ refusal to answer Ecuador’s request to produce a Joint Interpretation of the treaty, the majority explained that there was no obligation to negotiate between the parties.112 Accordingly, in order for interpretation to become the subject matter of a dispute would require not only evidence of ‘positive opposition’ of the parties,113 but also an alleged breach of the treaty which could not be characterised from an omission to respond to a Diplomatic Note. In reaching this conclusion, the Tribunal considered that it was ‘plausible’ that the US’ silence was warranted by ‘a principled stance of not wanting to interfere with the decisions of Article VI investor-State tribunals, be they right or wrong’.114 However, a closer reading of the majority decision suggests that, had Ecuador clearly formulated an allegation of a breach of Article V of the BIT, according to which the Parties ‘agree to consult promptly, on the request of either, to resolve any disputes in connection with the Treaty, or to discuss any matter relating to the interpretation or application of the Treaty’, Ecuador’s claim could well have succeeded.115 Interestingly, the parties submitted several expert opinions on the object and purpose of the inter-state dispute resolution of the BIT and the arbitrability of interpretative dispute.116 In his Expert Opinion, Professor Reisman considered that the possibility of a state pursuing an inter-state proceeding to seek interpretation while an investor-state proceeding was previously introduced, would ‘undermine the validity of final awards tout court, by introducing an advisory or appellate procedure in which the investor would have no standing, hence effectively denying the investor’s rights under the BIT’.117 To the contrary, Professor Pellet considered that a ‘positive opposition’ had never been a pre-requisite for the existence of an interpretative dispute’.118

109 Republic

of Ecuador v United States of America, above (n 105) para 4. para 52. 111 ibid para 188. 112 ibid para 227. 113 See also three companion Judgments rendered by the ICJ, Obligations concerning negotiations relating to cessation of the nuclear arms race and to nuclear disarmament (Marshall Islands v India) (Jurisdiction of the Court and admissibility of the Application) [2016] ICJ Rep 255. 114 Republic of Ecuador v United States of America, above (n 105) para 219. 115 ibid para 227. 116 In chronological order by M Reisman, C Tomuschat, MC Caffrey, A Pellet and CF Amerasinghe. 117 Chevron Corp and Texaco Petroleum Co v Republic of Ecuador, above (n 106) Expert Opinion of Prof WM Reisman (19 January 2010) para 32. 118 ibid Expert Opinion of Prof Pellet (23 May 2012) para 7. 110 ibid

International Investment Tribunals and Treaty Interpretation 325 By dismissing jurisdiction on the ground that an abstract question of interpretation cannot be deemed an arbitrable dispute under an IIA, the Republic of Ecuador v United States of America award substantially restrains the scope of arbitral tribunals’ jurisdiction. This decision marks a significant departure from the investor-state proceedings in which, as we have seen, investment tribunals have resisted states’ treaty parties’ interpretation (Pope & Talbot, Mondev), or accepted that misinterpretation alone can significantly influence the annulment of an award, thereby depriving an investor of a legal remedy (MHS v Malaysia). Perhaps regrettably, the approach followed by the Ecuador v United States tribunal seems to suggest that treaty interpretation falls into the category of non-arbitrable disputes.119 Indeed, if interpretative disputes are not considered to be arbitrable, the terms of state-to-state dispute resolution clauses would be rendered meaningless since many IIAs provisions similar to the one under scrutiny confer jurisdiction to arbitral tribunals for ‘any dispute’ ‘relating to the interpretation and application of the treaty’. In fact, contrary to ISDS proceedings, the Contracting States’ standing under an IA’s is not limited to an existing ‘investment’ dispute between the host state and a foreign investor. It should therefore be assumed that the omission of any required connection with an investment dispute in the state-to-state dispute resolution clause of an IA, is purposely intended to cover other state-to-state legal interests ‘and must have a meaning’.120 Therefore, an interpretative dispute can arise ‘irrespective of whether there is a dispute regarding the treaty’s application’.121 This means that investment tribunals’ administration of justice should include the ability to render a binding interpretation of a treaty when the treaty so provides. Given that Ecuador has subsequently withdrawn from the ICSID Convention and terminated its BITs, it is thus debatable whether this outcome advanced the objective of ‘promotion and protection of investment’, and more generally, whether it contributed to facilitating the resolution of international disputes.

V. CONCLUSIONS

There are a few certitudes that may be distilled from this non-exhaustive exploration of interpretation in investment arbitration. In exercising their interpretative mandate, investment tribunals have certainly advanced investors’ rights which in turn has fostered states’ treaty parties to regain control over treaty interpretation. Whether or not it is considered an ‘obligation’ or a ‘power’, this interpretative mandate will increasingly be shared with states as the principal authors of IIAs. The importance of interpretation can be further understood through the function assigned to the treaties in the international legal order. Unlike contracts in international commercial arbitration, treaties serve a double legal function: they qualify as interpretative guidance by application of Article 31(3)(c) VCLT, and are also listed as a primary source of international law.122

119 A category that was forcefully rejected by H Lauterpacht back in 1930 (H Lauterpacht, ‘La théorie des différends non justiciables en droit international’ (1930) 37 Collected Courses of the Hague Academy of International Law 649). 120 Urbaser SA and Consorcio de Aguas Bilbao Biskaia, Bilbao Biskaia Ur Partzuergoa v Argentine Republic, Award (8 December 2016) ICSID Case No ARB/07/26, para 1187. 121 Republic of Ecuador v United States of America, above (n 104) Dissenting Opinion of Professor Raul Emilio Vinuesa, para 25. 122 Statute of the International Court of Justice, 18 April 1946, 33 UNTS 993, Art 38.

326 Elise Ruggeri Abonnat One direct consequence of this dual function ascribed to treaties in international adjudication is that ‘reference to other rules of international law in the course of interpreting a treaty is an everyday, often unconscious part of the interpretation process’,123 which requires a high standard of diligence from investment tribunals tasked with determining the ordinary meaning of a term.

123 C McLachlan, ‘The principle of systemic integration and article 31(3)(c) of the Vienna Convention’ (2005) 54 International and Comparative Law Quarterly 280.

20 Standard of Review JOSHUA PAINE*

I. INTRODUCTION: UNPACKING THE STANDARD OF REVIEW

I

N INTERNATIONAL INVESTMENT law, the term ‘standard of review’ is often used to refer to the intensity of the scrutiny applied by international arbitrators to determinations of fact and/or law made by the respondent host state.1 The standard of review that is applied by investment arbitrators can fall at a wide range of points ‘on a continuum bounded by total deference to justifications provided or analysis performed by a primary decision-maker at one end, and substitutionary (de novo) review of the relevant measure and its justification at the other’.2 A closely related concept is deference, which ‘refers to a limitation in a … tribunal’s level of scrutiny concerning decisions taken or determinations made by a host state institution because the adjudicator respects the reasons for a state’s decision or conduct even if its own assessment was different’.3 One challenge of the standard of review as a concept is how it relates to the content of applicable substantive norms, such as investment treaty obligations. The standard of review should not be conflated with the content of particular substantive obligations, such as fair and equitable treatment (FET).4 The considerations that inform the appropriate standard of

* Joshua Paine is a senior Lecturer in Law at the University of Bristol. This chapter was partly written while I was a Senior Research Fellow at the Max Planck Institute Luxembourg for Procedural Law. This chapter builds on my earlier work, particularly J Paine, ‘Standard of Review: Investment Arbitration’ in H Ruiz Fabri (ed), Max Planck Encyclopedia of International Procedural Law (Oxford, Oxford University Press, 2019). I thank the editors of the volume and participants in the underlying workshop for comments on prior drafts. 1 See, eg, C Henckels, Proportionality and Deference in Investor-State Arbitration: Balancing Investment Protection and Regulatory Autonomy (Cambridge, Cambridge University Press, 2015) 29–30. 2 C Henckels, ‘Balancing Investment Protection and the Public Interest: The Role of the Standard of Review and the Importance of Deference in Investor-State Arbitration’ (2013) 4 Journal of International Dispute Settlement 197, 201. G Van Harten, Sovereign Choices and Sovereign Constraints: Judicial Restraint in Investment Treaty Arbitration (Oxford, Oxford University Press, 2013) 34. 3 SW Schill, ‘Deference in Investment Treaty Arbitration: Re-Conceptualizing the Standard of Review’ (2012) 3 Journal of International Dispute Settlement 577, 582; Henckels, above (n 1) 34–35; JH Fahner, Judicial Deference in International Adjudication: A Comparative Analysis (Oxford, Hart Publishing, 2020) 5–7. E Shirlow, Judging at the Interface: Deference to State Decision-Making Authority in International Adjudication (Cambridge, Cambridge University Press, 2021) 16 (suggesting the standard of review is one conception of the wider concept of deference and defining deference as ‘techniques used by international adjudicators to recognise a domestic actor’s superior “authority” to decide issues relevant to the settlement of the dispute’). 4 Henckels, above (n 1) 31–32; Schill, above (n 3) 582; Fahner above (n 3) 128–131, 147–148 (distinguishing between interpretation, concerning ‘the substantive content of the legal obligation’, and the standard of review, concerning ‘the distribution of interpretative powers between international and domestic actors’).

328 Joshua Paine review, such as the greater expertise or legitimacy of domestic decision-makers, concern the institutional relationship between international arbitrators and domestic authorities, and, as Van Harten notes, may apply irrespective of the particular substantive standard under which a domestic decision is challenged.5 That said, determining the appropriate standard of review in a particular situation necessarily requires taking into account applicable substantive standards.6 For example, many of the awards discussed in this chapter made statements regarding the standard of review when interpreting and applying the international minimum standard, under the North American Free Trade Agreement (NAFTA).7 At one level, these awards illustrate the relatively high threshold for breach set by the international minimum standard. However, as we will see, these awards also make more wide-ranging statements about the appropriate division of decision-making power between international arbitrators and national authorities. In selecting the cases to discuss as landmark cases in relation to the standard of review, this chapter focuses on awards that have been widely cited in subsequent jurisprudence or that provide a notable consolidation or extension of prior reasoning on particular aspects of the issue. In light of the cases discussed, this chapter argues that the standard of review is fundamentally about the allocation of decision-making power between arbitral tribunals and host states. The standard of review concerns the division of decision-making power between the international and national levels, and also between arbitrators as opposed to other kinds of decision-makers (eg specialist administrators or elected politicians).8 While there is no settled approach to the standard of review in investor-state arbitration, the awards reviewed demonstrate that in a wide variety of circumstances tribunals have recognised that the appropriate degree of scrutiny is something short of full de novo review of a host state’s decision. Furthermore, a number of awards have addressed the underlying considerations that should inform the standard of review. In summary, such considerations can be divided between reasons around the greater expertise or institutional capacity of domestic decision-makers as compared to investment arbitrators for making certain types of determinations, and reasons around the greater legitimacy or accountability of domestic decision-makers for making particular decisions.9 The awards discussed in this chapter, which are mostly from the last decade and all since the year 2000, should be understood in the context of persistent and growing debates over the

5 Van

Harten, above (n 2) 3–6, 98–100; Schill, above (n 3) 582. above (n 3) 582, fn 27, 597–598; Fahner, above (n 3) 128–130, 147; Shirlow, above (n 3) 40–41, 247–253 (discussing the relationship between deference and the content of applicable international obligations). F Ortino and NM Tabari, ‘International Dispute Settlement: The Settlement of Investment Disputes Concerning Natural Resources – Applicable Law and Standards of Review’ in E Morgera and K Kulovesi (eds), Research Handbook on International Law and Natural Resources (Cheltenham, Edward Elgar Publishing, 2016) 507–9 (suggesting the term ‘standard of review’ can include both the issue of the intensity of review and the content of the applicable norm); Henckels, above (n 1) 189. 7 North American Free Trade Agreement (signed 17 December 1992, entered into force 1 January 1994, terminated 1 July 2020) 32 International Legal Materials 296, Art 1105(1) [hereinafter NAFTA]. 8 Henckels, above (n 1) 30. See also Shirlow, above (n 3) 5, 273, 275 (discussing deference as a tool used by international adjudicators ‘to mediate … [and readjust over time] conflicting claims to national and international decision-making authority’). 9 See generally Henckels, above (n 1) 34–43; Fahner, above (n 3) 149–216; Shirlow, above (n 3) 19–30; Schill, above (n 3) 599–605; Van Harten, above (n 2) 3–5. 6 Schill,

Standard of Review

329

legitimacy of investor-state arbitration during this period. A common assessment has been that investment arbitrators have failed ‘to allow States a sufficient margin to determine and implement various policy goals’,10 and arbitrators should employ public law-inspired standards of review, based on experience in other international or national contexts.11 The awards analysed in this chapter suggest that this debate has had concrete effects, in terms of increasingly detailed discussions of the appropriate standard of review by tribunals. Discussions of deference and the standard of review are one way for arbitrators to limit their assumption of decision-making power and address concerns about the intrusiveness of investor-state arbitration.12 The cases discussed in this chapter are all from investment treaty arbitration. However, as a matter of principle, the considerations that inform the appropriate standard of review could also apply in contract-based investment arbitrations. As Ortino and Tabari have argued, where the host state conduct at issue involves an exercise of public or governmental functions, deference may be warranted in contract-based arbitrations, for example based on the greater expertise or accountability of domestic authorities.13 A further challenge regarding the standard of review is that it is heavily fact-specific and partly depends on the particular kind of domestic determination at issue.14 This chapter attempts to give a sense of the variety of settings in which the standard of review question arises, however due to space limitations it does not cover every context. For example, cases concerning the standard of review in relation to the invocation of essential security clauses, or in relation to scrutiny of International Centre for Settlement of Investment Disputes (ICSID) awards by Annulment Committees, are not addressed.15 Within the cases that are discussed, a broad distinction will be drawn between cases involving the application of non-legal forms of expertise or discretion (addressed in section II), and cases concerning the standard of review in relation to determinations of domestic law (addressed in section III). Within the cases concerning determinations of domestic law, a distinction will be drawn depending on whether the determination was made in an adjudicatory setting, reflecting that international law has traditionally afforded particular deference to determinations of domestic law made by domestic courts. Section IV provides some brief conclusions regarding what the cases as a whole illustrate.

10 A Roberts, ‘The Next Battleground: Standards of Review in Investment Treaty Arbitration’ in AJ van den Berg (ed), Arbitration: The Next Fifty Years (Alphen aan den Rijn, Kluwer Law International, 2012) 172. 11 See, eg, Schill, above (n 3); Henckels, above (n 1); Compare the perspective of Caroline Foster, who has consistently resisted the standard of review concept, on the basis that it may represent a shift to something less than full adjudication and weaken international adjudication as a mechanism for securing shared international interests: eg, CE Foster, ‘Adjudication, Arbitration and the Turn to Public Law ‘Standards of Review’: Putting the Precautionary Principle in the Crucible’ (2012) 3 Journal of International Dispute Settlement 525, 525–26, 557–8. CE Foster, Global Regulatory Standards in Environmental and Health Disputes: Regulatory Coherence, Due Regard and Due Diligence (Oxford, Oxford University Press, 2021) 337–42. 12 EM Leonhardsen, ‘Treaty Change, Arbitral Practice and the Search for a Balance: Standards of Review and the Margin of Appreciation in International Investment Law’ in W Werner and L Gruszczynski (eds), Deference in International Courts and Tribunals: Standard of Review and Margin of Appreciation (Oxford, Oxford University Press, 2014) 138–9. 13 Ortino and Tabari, above (n 6) 514–515. 14 Fahner, above (n 3) 89; Henckels, above (n 1) 34–35. 15 For an overview of these issues see, eg, J Paine, ‘Standard of Review: Investment Arbitration’ in H Ruiz Fabri (ed), Max Planck Encyclopedia of International Procedural Law (Oxford, Oxford University Press, 2019) paras 24, 28. On essential security provisions see, eg, Henckels, above (n 1) 87–96.

330 Joshua Paine II. DOMESTIC DETERMINATIONS INVOLVING NON-LEGAL EXPERTISE OR DISCRETION

A. The Beginning: SD Myers v Canada The statements made by this Tribunal concerning deference have proven particularly influential, being cited by numerous subsequent awards. The Tribunal’s key statement was that: When interpreting and applying the ‘minimum standard’, a … tribunal does not have an open-ended mandate to second-guess government decision-making. Governments have to make many potentially controversial choices. In doing so, they may appear to have made mistakes, to have misjudged the facts, proceeded on the basis of a misguided economic or sociological theory, placed too much emphasis on some social values over others and adopted solutions that are ultimately ineffective or counterproductive. The ordinary remedy, if there were one, for errors in modern governments is through internal political and legal processes, including elections.16

The Tribunal added that a determination that the international minimum standard had been breached ‘must be made in the light of the high measure of deference that international law generally extends to the right of domestic authorities to regulate matters within their own borders’.17 The context was that in the surrounding paragraphs the Tribunal was addressing the broad, gap-filling nature of the international minimum standard.18 Ultimately, the majority of the Tribunal held that Canada’s breach of the national treatment standard, through the enactment of a protectionist ban on hazardous waste exports, also established a breach of the international minimum standard.19 In short, the contribution of this award was in providing a generalised statement about deference, rather than itself applying a particularly deferential approach.

B. Glamis Gold v United States This award is notable for its wide-ranging statement about the division of decision-making power between international tribunals and domestic authorities. The award has been widely cited as an example of a deferential approach to applying the international minimum standard. The context was that the claimant was challenging a cultural review process and other measures applied to its proposed mining project, which was located in areas adjacent to sites of indigenous cultural significance. The Tribunal accepted the US’ submission that ‘“[i]t is simply not this Tribunal’s task to become archaeologists and ethnographers”’ and answer de novo questions involved in the cultural review process.20 More generally, it was ‘not the role of this Tribunal, or any international tribunal, to supplant its own judgment of underlying factual material and support for that of a qualified domestic agency’.21 Rather, the Tribunal could ‘assess only whether there was reasonable evidence, and thus the government’s reliance on such was not obviously and actionably misplaced’.22 Specifically, the Tribunal asked in 16 SD

Myers, Inc v Canada, Partial Award (13 November 2000) UNCITRAL, para 261. para 263. 18 ibid paras 259–260, 264–265. 19 ibid para 266. Regarding the prior finding that Canada’s measures were protectionist see paras 193–195, 251, 255. One member of the Tribunal dissented from the finding that a breach of the national treatment obligation established a breach of the international minimum standard: para 267. 20 Glamis Gold Ltd v United States of America, UNCITRAL, Award (8 June 2009) para 779. 21 ibid para 779. 22 ibid para 786. For similar approaches see Apotex Holdings Inc v United States of America, Award (25 August 2014) ICSID Case No ARB[AF]/12/1, para 9.37 (the Tribunal, faced with a challenge to a decision of a domestic drug 17 ibid

Standard of Review

331

applying the international minimum standard whether ‘the agency’s review and conclusions exhibit a gross denial of justice, manifest arbitrariness, blatant unfairness, a complete lack of due process, evident discrimination, or a manifest lack of reasons’.23 For example, in applying this standard to the cultural review process, the Tribunal noted that the US had relied on the advice of qualified professionals and established a prima facie case regarding why particular aspects of the process were reasonable in the circumstances.24 The Tribunal also rejected the claimant’s challenge to separate state-level legislation applying to the project, finding that the fact that this legislation did not address all of the harms posed by open-pit mining to indigenous sacred sites reflected the normal role of government in striking a ‘compromise between the interests of competing parties’.25 In short, Glamis is representative of a line of investment law jurisprudence, beginning with SD Myers, which emphasises that the fact that state measures are arguably sub-optimal or mistaken in public policy terms is insufficient to establish breach of an investment treaty.26

C. Philip Morris v Uruguay This high-profile case is significant as it contains a strong general endorsement of deference towards national-level regulatory determinations and demonstrates how deference may be operationalised in relation to specific regulatory determinations that are challenged under the FET standard. Additionally, the Tribunal was divided over the question of whether the margin of appreciation doctrine, developed in the jurisprudence of the European Court of Human Rights (ECtHR), applies in investment arbitration – an issue that has been debated by other tribunals and commentators.27 The Tribunal’s remarks concerning the standard of review were made in scrutinising Uruguay’s tobacco-labelling measures, which were challenged under

regulator, noted ‘the need for international tribunals to recognize the special roles and responsibilities of regulatory bodies charged with protecting public health and other important public interests … [and] to exercise caution in cases involving a state regulator’s exercise of discretion’). See also para 9.39. Chemtura Corporation (formerly Crompton Corporation) v Canada, Award (2 August 2010) UNCITRAL, paras 123, 134, 138, 143, 145–8, 153, 162 (The Tribunal took into account that ‘certain agencies manage highly specialized domains involving scientific and public policy considerations’, and focused its analysis on whether the process employed by the regulator exhibited bad faith or a lack of due process, sufficient to breach the international minimum standard, rather than the scientific correctness of the regulator’s determinations). 23 Glamis above (n 20) para 779. See also paras 616–617 (disagreeing with the suggestion ‘that domestic deference in national court systems is necessarily applicable to international tribunals’ and finding that deference was present within the international minimum standard rather than being additive to it). 24 ibid paras 783, 787. 25 ibid paras 803–805. See also para 817. 26 See also below text at nn 41–45, 66. Joseph Charles Lemire v Ukraine, Decision on Jurisdiction and Liability (14 January 2010) ICSID Case No ARB/06/18, para 283. Sergei Paushok and ors v Mongolia, Award on Jurisdiction and Liability (28 April 2011) UNCITRAL, para 299. 27 See, eg, Y Fukunaga, ‘Margin of Appreciation as an Indicator of Judicial Deference: Is It Applicable to Investment Arbitration?’ (2019) 10 Journal of International Dispute Settlement 69, 82–86 (suggesting that some of the rationales that underlie the margin of appreciation doctrine also apply in investment arbitration); Paine, above (n 15) paras 22–23 (summarising prior commentary and arbitral awards); Fahner, above (n 3) 192–216 (arguing that specific characteristics of regional human rights courts mean some of the rationales for deference in that context do not apply to other forms of international adjudication). G Born, D Morris and S Forrest, ‘“A Margin of Appreciation”: Appreciating Its Irrelevance in International Law’ (2020) 61 Harvard International Law Journal 65, esp. 92–134 (arguing that there is no basis for transposing the ECtHR’s margin of appreciation doctrine to investment treaty arbitration and any deference afforded must follow from the terms of the applicable treaty provisions, interpreted in accordance with the ordinary rules of treaty interpretation).

332 Joshua Paine the FET standard for lacking a reasonable connection with the relevant regulatory objectives and for having been adopted without due consideration.28 The Tribunal made a general remark that: investment tribunals should pay great deference to governmental judgments of national needs in matters such as the protection of public health. In such cases respect is due to the ‘discretionary exercise of sovereign power, not made irrationally and not exercised in bad faith … involving many complex factors.’29

More specifically, with regard to Uruguay’s ‘single presentation requirement’ for tobacco brands, the majority found that it was not necessary to decide whether the measure actually had the effects intended by Uruguay. Instead, it was sufficient that the measure ‘was an attempt to address a real public health concern that the measure taken was not disproportionate to that concern and that it was adopted in good faith’.30 Similarly, with regard to Uruguay’s so-called ‘80/80’ regulation, concerning the size of health warnings on cigarette packets, the majority emphasised that: Substantial deference is due in that regard to national authorities’ decisions as to the measures which should be taken to address an acknowledged and major public health problem. The fair and equitable treatment standard is not a justiciable standard of good government, and the tribunal is not a court of appeal … the balance to be struck between conflicting considerations was very largely a matter for the government.31

Accordingly, the majority limited its analysis to asking whether the 80/80 regulation ‘was entirely lacking in justification or wholly disproportionate’, and found that the regulation ‘was a reasonable measure adopted in good faith’ and was not ‘arbitrary, grossly unfair, unjust, discriminatory or a disproportionate’.32 Beyond this threshold, ‘[h]ow a government requires the acknowledged health risks of products such as tobacco, to be communicated … is a matter of public policy, to be left to the appreciation of the regulatory authority’.33 The Award in Philip Morris v Uruguay is also significant for its finding ‘that the “margin of appreciation” is not limited to the context of the ECHR but “applies equally to claims arising under BITs,” at least in contexts such as public health’.34 However, the majority’s analysis did not draw on the ECtHR’s case law, but focused on the content of the relevant investment treaty standards, the basis for Uruguay’s measures under the Framework Convention on Tobacco Control (FCTC), and, to a lesser degree, prior investment awards that have considered the issue of the standard of review.35 Accordingly, the award does not necessarily represent an attempt to import the ECtHR’s margin of appreciation concept wholesale into investment arbitration, as opposed to the more general idea of some degree of deference for complex and

28 Philip Morris Brands Sàrl, Philip Morris Products SA and Abal Hermanos SA v Uruguay, Award (8 July 2016) ICSID Case No ARB/10/7, para 389. 29 ibid para 399, partly quoting Electrabel SA v Hungary, ICSID Case No ARB/07/19, Decision on Jurisdiction, Applicable Law and Liability (30 November 2012) para 8.35. 30 Philip Morris v Uruguay, above (n 28) paras 409–410. 31 ibid para 418. 32 ibid paras 419–420. 33 ibid para 419. 34 ibid para 399. 35 eg, ibid paras 391–396, 399–401. Earlier awards had also considered the applicability of the margin of appreciation doctrine, developed in the ECHR context, to investment arbitration: see Born, Morris and Forrest, above (n 27) 92–99, 104–106.

Standard of Review

333

discretionary policy determinations that domestic authorities may be better placed to make.36 While Arbitrator Gary Born dissented from the finding that the margin of appreciation applies in the investment context, on the basis that the standard of review must be focused on the terms of the relevant investment treaty, he accepted that ‘a substantial degree of deference for sovereign regulatory judgments’ was due when applying the FET standard.37 Nevertheless, Arbitrator Born emphasised that ‘deference to sovereign measures is the starting point, but not the ending point’, and, under the FET standard, the Tribunal had to scrutinise whether there was a ‘minimum level of rationality and proportionality between the state’s measure and a legitimate governmental objective’.38 On the facts, Arbitrator Born, dissenting, found that Uruguay’s single presentation requirement failed to satisfy this threshold and thus constituted a breach of the FET standard.39

D. Marfin v Cyprus This case saw the Tribunal provide a detailed discussion of why it refused to second-guess various decisions made by the Cypriot authorities in responding to a financial crisis. The award also contains some significant generalised statements about the division of decision-making power between arbitral tribunals and host states. The context was that the claimants challenged various decisions made by the Cypriot authorities in relation to the second largest bank in Cyprus, Laiki, in which the claimants were shareholders. The bank had encountered financial difficulties during the 2011–2012 Eurozone financial crisis and the claimants argued that Cyprus had used this as a cover to nationalise the bank.40 The decisions challenged included removal of Laiki’s senior managers or directors, and the process employed by Cyprus for recapitalising the bank. Early in its analysis of the merits, in rejecting the claimants’ arguments regarding Cyprus’ failure to negotiate better terms for its banking sector at a Eurozone summit, the Tribunal noted that: It is not up to an arbitral tribunal constituted under an investment treaty to sit in judgment over difficult political and policy decisions made by a State, particularly where those decisions involved an assessment and weighing of multiple conflicting interests and were made based on continuously developing threats to the safety and soundness of the financial system. Unless the measure at issue is shown to be arbitrary, capricious and unrelated to a rational policy, or manifestly lacking even-handedness, a tribunal should not intervene.41

The Tribunal repeatedly found that it was not charged with determining whether the decisions challenged were the best possible decisions in the circumstances or were correct.42 Instead, the 36 Compare CE Foster, ‘Respecting Regulatory Measures: Arbitral Method and Reasoning in the Philip Morris v Uruguay Tobacco Plain Packaging Case’ (2017) 26 Review of European, Comparative and International Environmental Law 287, 292, 296–7 (arguing that the concepts of deference and margin of appreciation had minimal impact as the science on the harmfulness of tobacco was clear and the FCTC provided a basis for determining the reasonableness of Uruguay’s measures). 37 Philip Morris v Uruguay, above (n 28), Concurring and Dissenting Opinion of Gary Born, para 191, see generally paras 137–145, 181–191. 38 ibid paras 139, 142–145. 39 ibid paras 172, 176. 40 Marfin Investment Group Holdings SA, Alexandros Bakatselos and others v Republic of Cyprus, Award (26 July 2018) ICSID Case No ARB/13/27, para 832. 41 ibid para 870. 42 ibid paras 875, 898, 937, 1213.

334 Joshua Paine Tribunal limited its analysis to whether the various measures were ‘arbitrary, capricious and unrelated to a rational policy’,43 or ‘abusive, did not afford due process or … a pretense of form designed to conceal improper ends’.44 As the Tribunal noted: ‘What matters is that BITs do not hold States to an obligation to act following international best practices.’45 Ultimately, the Tribunal found that none of the measures complained of constituted an expropriation or a breach of the FET standard. One of the considerations the Tribunal saw as justifying deference was the greater expertise and institutional capacity of domestic authorities. For example, in relation to determinations made by the Central Bank of Cyprus, the Tribunal was ‘mindful of the fact that a central bank acts as a regulator of a highly technical and sophisticated economic sector, that it has intimate knowledge of the underlying data and is best placed to assess whether one course of action is preferable to another’.46 Nevertheless, the Tribunal emphasised that such deference was limited and did not prevent review of the decisions for abusiveness or lack of due process.47 Similarly to this approach, some commentators have suggested that where arbitrators afford deference to domestic decision-makers based on the latter’s specialised expertise, arbitrators might employ a strict approach to ensuring that standards of procedural propriety are observed.48 The Marfin award is also notable for the Tribunal’s finding that it had to assess the banking regulator’s decision to remove Laiki’s Chief Executive Officer based on the information available at the time of the decision.49 This is similar to the approach suggested by some other tribunals and commentators in relation to scrutiny of state determinations made in the context of rapidly developing scenarios, such as financial crises.50 Finally, the Marfin award also contains a notable statement regarding deference on the basis of the greater legitimacy of domestic authorities. Specifically, the Tribunal held that a decision over whether to inject public funds into a bank in financial difficulty was ‘inherently political in nature and cannot be second-guessed by a tribunal constituted under an investment treaty’.51

III. STANDARD OF REVIEW AND DETERMINATIONS OF DOMESTIC LAW

A. Outside of a Judicial Context: Bilcon v Canada This award raises key questions about the approach of arbitral tribunals to making findings of breach of an investment treaty on the basis of apparent breaches of domestic law. The award also engaged explicitly with the considerations underlying determinations of the appropriate standard of review. The context was that the investor challenged the environmental review process that was applied to its project, which ultimately led to the project not being approved.

43 ibid

paras 870, 875, see also para 1213. para 900, similarly paras 937, 1035. For similar reasoning see Invesmart, BV v Czech Republic, Award (26 June 2009) UNCITRAL, paras 430, 454, 460, 501. 45 Marfin, above (n 40) para 875. See also para 1036; Similarly: Invesmart, above (n 44) para 459. 46 Marfin, above (n 40) para 899. 47 ibid. 48 Schill, above (n 3) 603; Compare Henckels, above (n 1) 149–150. 49 Marfin, above (n 40) para 948. 50 Enron Creditors Recovery Corp and Ponderosa Assets, LP v The Argentine Republic, Decision on the Application for Annulment of the Argentine Republic (30 July 2010) ICSID Case No ARB/01/3, para 372; Henckels, above (n 1) 139–140. 51 Marfin, above (n 40) para 1065; Compare Invesmart, above (n 44) paras 484–487 (states enjoy a margin of appreciation regarding decisions on public expenditure). 44 ibid

Standard of Review

335

The majority of the Tribunal found that the environmental review panel had breached the international minimum standard by failing to follow its mandate in the applicable Canadian laws. Specifically, the review panel had applied a standard to the investor’s project, emphasising community values, which lacked a basis in the relevant domestic legislation, and had failed to assess the project’s adverse effects after potential mitigation measures, as required by the panel’s mandate.52 The majority found that what was at issue was ‘not merely a matter of disputed judgments interpreting grey areas of the law, weighing contested points of evidence, or exercising scientific judgment’.53 Rather, the review panel had ‘departed in fundamental ways from the standard of evaluation required by the laws of Canada rather than merely being controversial in matters of detailed application’.54 The majority emphasised that it was not purporting ‘to conduct its own environmental assessment … [nor] deciding what the actual outcome should have been’.55 Instead, the key was that the investors had been denied a ‘fair opportunity to have the specifics of that case considered, assessed and decided in accordance with applicable [Canadian] laws’.56 Also central to the finding of a breach of the international minimum standard was that the review panel had failed to provide the investor reasonable notice of the approach, based on community values, that would be applied to its project.57 The majority emphasised that ‘mere error in legal or factual analysis’ was insufficient to give rise to a breach of the international minimum standard.58 The Bilcon Award was subject to a vigorous dissent by Arbitrator Donald McRae. He found that there had not been a breach of the international minimum standard as the claimants had only established an arguable breach of domestic law by the environmental review panel.59 Arbitrator McRae emphasised that, beyond this, the majority had not demonstrated why the review panel’s actions were arbitrary, or otherwise sufficient to breach the high threshold set by the international minimum standard, noting that the review panel had not ‘deliberately or wilfully disregarded the [Canadian] law to be applied’.60 Arbitrator McRae also disagreed with the majority’s position that the environmental review panel had applied an unprecedented approach emphasising community values; instead, he found that these issues fell within the panel’s mandate and the investor had been given reasonable notice of the matters its environmental impact assessment needed to cover.61 Arbitrator McRae emphasised that under Canadian law, the claimants could have sought to have the report of the review panel overturned in domestic courts but had not pursued this option.62 In his view, the majority, by conflating a breach of Canadian law with a breach of the international minimum standard, had created a new potential to obtain damages at the international level, which would allow future claimants to ‘bypass the domestic remedy provided for such a departure from

52 William Ralph Clayton, William Richard Clayton, Douglas Clayton, Daniel Clayton and Bilcon of Delaware, Inc v Canada, Award on Jurisdiction and Liability (17 March 2015) PCA Case No 2009–04, paras 451–452, 525–528, 583, 591. 53 ibid para 454. 54 ibid para 594, see also paras 600–602. 55 ibid para 602. 56 ibid paras 603 (emphasis in original), 571. 57 ibid paras 534, 543, 594. 58 ibid para 594. The majority’s finding of a breach of the national treatment standard depended on similar considerations, namely that unlike comparable local projects, the investor’s project was not assessed according to the standards established by the applicable Canadian legislation: paras 697, 700, 704–705. 59 Bilcon v Canada, above (n 52), Dissenting Opinion of Professor Donald McRae, paras 34, 36. 60 ibid para 38, generally paras 36–48. 61 ibid paras 25–26, see also paras 14–16, 22–23. 62 ibid para 42.

336 Joshua Paine Canadian law’.63 In summary, Arbitrator McRae saw the majority decision as a ‘significant intrusion into domestic jurisdiction [that] … will create a chill on the operation of environmental review panels’.64 The Bilcon award also contains a notable more generalised statement regarding the considerations relevant to determining the appropriate standard of review, within the Tribunal’s discussion of the international minimum standard. The Tribunal noted that: third-party adjudicators must, in applying the international minimum standard, take into account that domestic authorities may have more familiarity with the factual and domestic legal complexities of a situation. Secondly, domestic authorities may also enjoy distinctive kinds of legitimacy, such as being elected or accountable to elected authorities. Thirdly, … Third-party adjudicators may have their own advantages including independence and detachment from domestic pressures.65

The Tribunal also noted that: ‘Modern regulatory and social welfare states tackle complex problems. … The imprudent exercise of discretion or even outright mistakes do not, as a rule, lead to a breach of the international minimum standard.’66 This aspect of the award is significant because the majority engaged directly with the key rationales for some degree of deference to national authorities, such as the latter’s greater expertise, familiarity with local conditions or legitimacy. The decision to include these generalised passages in the award likely reflects that Bilcon was an obviously controversial case, involving a finding of breach in relation to a national-level determination made on purportedly environmental grounds.

B. Standard of Review and Domestic Judicial Decisions The next group of cases discussed concern the standard of review in relation to determinations of domestic law by domestic courts. Traditionally, international law has afforded particular deference to determinations of domestic law by the domestic judiciary through certain features of the doctrine of denial of justice. Such deference is manifested in the rule that only a final outcome reached at the domestic level can be challenged, which allows the domestic system to correct errors, and ‘the idea that denial of justice focuses upon the procedural aspects of the adjudication rather than the substantive reasons for the decision’.67 While these ideas have influenced investment treaty cases concerning domestic judicial decisions, some awards have recognised an expanded potential for review of domestic adjudicatory processes.

63 ibid

paras 43, 48. para 48. Arbitrator McRae also dissented from the majority’s finding of a breach of the national treatment standard, as, in his view, the claimants had been treated in accordance with Canadian law: para 65. 65 Bilcon v Canada, above (n 52) para 439. 66 ibid para 437. 67 Z Douglas, ‘International Responsibility for Domestic Adjudication: Denial of Justice Deconstructed’ (2014) 63 International & Comparative Law Quarterly 867, 877–8, 881–2; Chevron Corporation and Texaco Petroleum Corporation v The Republic of Ecuador, Second Partial Award on Track II (30 August 2018) PCA Case No 2009–23, paras 8.37, 8.40–8.42 (also noting that domestic legal systems benefit from a presumption of having acted properly). J Paulsson, Denial of Justice in International Law (Cambridge, Cambridge University Press, 2005) 5, 7–8, 73–76, 82–84, 87. Note, however, that historically international law has recognised that conduct of the domestic judiciary can amount to breach of a primary norm other than denial of justice (eg a domestic judgment that breaches a treaty obligation). No such deference will be owed in these scenarios: see, eg, Paulsson, ibid, 69–72, 84–86. This distinction is related to the question, discussed below, of whether domestic judicial conduct can violate investment treaty standards without constituting a denial of justice: see below n 75 and associated text. 64 ibid

Standard of Review

337

i. Mondev v United States The statements made by this Tribunal regarding the approach to scrutinising domestic judicial decisions have proven influential, being widely cited by later tribunals faced with claims of denial of justice, or other complaints about domestic judicial conduct. The context was that the Tribunal was faced with determining whether a decision of the Massachusetts Supreme Judicial Court, a state-level appellate court, dismissing the claimant’s claims under Massachusetts law, and a decision of the Supreme Court of the United States denying certiorari, constituted a violation of NAFTA’s minimum standard of treatment.68 The Tribunal began its discussion of the issue of denial of justice by noting that: It is one thing to deal with unremedied acts of the local constabulary and another to second-guess the reasoned decisions of the highest courts of a State. Under NAFTA, parties have the option to seek local remedies. If they do so and lose on the merits, it is not the function of NAFTA tribunals to act as courts of appeal.69

The Tribunal then articulated the standard to be applied in deciding denial of justice claims as follows: The test is not whether a particular result is surprising, but whether the shock or surprise occasioned to an impartial tribunal leads, on reflection, to justified concerns as to the judicial propriety of the outcome, bearing in mind on the one hand that international tribunals are not courts of appeal, and on the other hand that Chapter 11 of NAFTA (like other treaties for the protection of investments) is intended to provide a real measure of protection. In the end the question is whether, at an international level and having regard to generally accepted standards of the administration of justice, a tribunal can conclude in the light of all the available facts that the impugned decision was clearly improper and discreditable, with the result that the investment has been subjected to unfair and inequitable treatment.70

On the facts, the claimant argued that the way its domestic law claim was decided by the Massachusetts Supreme Judicial Court had substantially departed from prior jurisprudence and retrospectively applied new rules. The Tribunal doubted that the domestic court ‘made new law’ through its decision and added that ‘even if it had done so its decision would have fallen within the limits of common law adjudication. There is nothing here to shock or surprise even a delicate judicial sensibility.’71 Regarding the claimant’s argument that the domestic court, once it had decided the particular point of law, should have considered whether to apply the new rule retrospectively, the Tribunal noted: ‘it is normally a matter for local courts to determine whether and in what circumstances to apply new decisional law retrospectively’.72 The claimant further argued that the Massachusetts Supreme Judicial Court should have remanded certain questions of fact to the jury. The Tribunal rejected this argument, noting: ‘Questions of fact-finding on appeal are quintessentially matters of local procedural practice. Except in extreme cases, the Tribunal does not understand how the application of local procedural rules about such matters as remand, or decisions as to the functions of juries vis-a-vis appellate

68 Mondev International Ltd v United States of America, Award (11 October 2002) ICSID Case No ARB[AF]/99/2, paras 1, 70. 69 ibid para 126. 70 ibid para 127. The Tribunal partly drew upon ICJ, Elettronica Sicula SPA (ELSI) (USA v Italy) Judgment [1989] ICJ Rep 15, para 128. 71 Mondev, above (n 68) para 133. See also para 137. 72 ibid para 137.

338 Joshua Paine courts, could violate’ the international minimum standard.73 Overall, the Mondev Tribunal was clearly concerned that investment treaty tribunals should not become courts of appeal for determinations of domestic law by domestic courts. Yet Mondev also makes clear that domestic judicial decisions are subject to a degree of control under the international minimum standard, namely where domestic decisions are ‘clearly improper and discreditable’ ‘having regard to generally accepted standards of the administration of justice’.74 ii. OAO Tatneft v Ukraine This award consolidates a line of investment treaty jurisprudence that has held that domestic judicial conduct can be challenged on grounds other than denial of justice. The result of such an interpretation is that a lower threshold may apply for challenging domestic judicial conduct, and unlike for denial of justice, there may be no requirement to exhaust local remedies as a substantive element of establishing the claim.75 Accordingly, this line of jurisprudence has major implications for the potential intrusiveness of investment treaty arbitration vis-à-vis domestic judicial systems. Reflecting an awareness of such considerations, the Tatneft Tribunal included in its analysis a more general discussion of the appropriate standard of review in relation to domestic judicial decisions.76 The context was that the claimant challenged decisions of the Ukrainian courts, and conduct of a prosecutor and other state agencies, which initially concerned a dispute over the dismissal and reinstatement of a company chairperson, and eventually led to the annulment of the claimant’s shareholdings. The Tribunal accepted at the outset that a denial of justice could not be established as the Ukrainian courts had generally been available, and there were no specific examples of nationality-based discrimination or corruption.77 The key step in the Tribunal’s reasoning was to hold that judicial acts are subject to the FET standard understood in an expansive manner, which extends beyond the prohibition of denial of justice, or any requirement of ‘egregiousness or bad faith’.78 The Tribunal immediately added that: ‘This does not alter the conclusion that the mere misapplication of domestic law is not enough to give rise to liability absent some kind of adverse intention.’79 The Tribunal accepted that it was ‘not an appellate court’ and as a general proposition ‘international tribunals ought to be deferential to domestic courts’.80 Nevertheless, it saw the deference owed to domestic courts as limited by the greater protection provided by the FET standard, above the international minimum standard.81 Additionally, ‘deference on the part of international tribunals requires

73 ibid

para 136. para 127. 75 See, eg, Chevron Corporation and Texaco Petroleum Company v Ecuador, Partial Award on the Merits (30 March 2010) UNCITRAL, paras 242–244, 321 (concerning an effective means provision); White Industries Australia Ltd v India, Final award (30 November 2011) UNCITRAL, paras 10.4.22–10.4.24, 11.3.1–11.3.3, 11.4.19 (same). ECE Projektmanagement International GmbH v Czech Republic, Award (19 September 2013) UNCITRAL, paras 4.743, 4.747 4.764, 4.782 (scrutiny of domestic judicial proceedings under FET standard besides denial of justice). See further on these issues: B Demirkol, Judicial Acts and Investment Treaty Arbitration (Cambridge, Cambridge University Press, 2018) 24–57, 192–198; V Prislan, Domestic Courts in Investor-State Arbitration: Partners, Suspects, Competitors (Doctoral thesis, Leiden University, 2019) 239–326, https://hdl.handle.net/1887/74364. Compare: Douglas, above (n 67) 894–896; M Sattorova, ‘Denial of Justice Disguised? Investment Arbitration and the Protection of Foreign Investors from Judicial Misconduct’ (2012) 61 International & Comparative Law Quarterly 223. 76 OAO Tatneft v Ukraine, Award on the Merits (29 July 2014) UNCITRAL, paras 350, 474–481. 77 ibid para 351. 78 ibid para 411, see also paras 394–395, 405–408, 413, 481. 79 ibid para 411. 80 ibid paras 474–475. 81 ibid para 475. 74 ibid

Standard of Review

339

the clear perception that domestic courts are independent, competent and above all clear of suspicion of corruption’.82 On the facts, after examining the specific court decisions complained of, the Tribunal was ‘not at ease with an unrestricted application of the standard of deference’.83 For the Tribunal, the explanations given by the Ukrainian courts were mostly ‘not … convincing and appear rather as an endorsement of the Prosecutor’s arguments’.84 The Tribunal was particularly concerned with the process applied to the claimant when viewed as a whole, which it saw as including numerous due process violations, such as ex parte court orders and failure to serve the claimant.85 Overall, the Tatneft award is an example of the idea that deference to domestic judicial processes may be conditional on whether the domestic process, on the facts, satisfies a sufficient threshold of independence.86 The Tribunal’s expansive approach to the FET standard is potentially very intrusive when applied to domestic judicial processes. iii. Eli Lilly v Canada This award is significant because it suggests that domestic judicial conduct can be challenged under standards other than denial of justice even where the FET standard is limited to the international minimum standard, as is common in contemporary investment treaties. However, in contrast to the Tatneft award, the Eli Lilly award indicates a substantial degree of deference towards reviewing domestic judicial processes. The context was that the claimant had two of its pharmaceutical patents invalidated by decisions of the Canadian courts. The claimant argued that the so-called ‘promise utility doctrine’, which had been developed through decisions of Canadian courts, was ‘radically new, arbitrary and discriminatory’, and the retrospective application of this doctrine to its patents constituted a breach of the international minimum standard and an unlawful expropriation.87 Canada as respondent, and the other NAFTA Parties in their non-disputing party submissions, contended that judicial decisions could only violate the international minimum standard or constitute an expropriation if a denial of justice was established, which the claimant did not allege.88 The Tribunal did not definitively determine this question of legal principle, as it made a factual finding that was fatal to the claimant’s case, namely that there had not been a radical change in Canadian law through the development of the promise utility doctrine.89 Nevertheless, the Tribunal was ‘unwilling to shut the door to the possibility that judicial conduct characterised other than as a denial of justice may’ constitute a breach of the international minimum standard.90 Immediately after this statement, the Tribunal emphasised: that a NAFTA Chapter Eleven tribunal is not an appellate tier in respect of the decisions of the national judiciary. It is not the task of a NAFTA Chapter Eleven tribunal to review the findings of national courts and considerable deference is to be accorded to the conduct and decisions of

82 ibid

para 476. para 479. 84 ibid; see also paras 266–268. 85 ibid paras 480, 397, 399, 405–406. 86 See, eg, J Hepburn, Domestic Law in International Investment Arbitration (Oxford, Oxford University Press, 2017) 128–9. 87 Eli Lilly and Co v Canada, Final Award (16 March 2017) ICSID Case No UNCT/14/2, para 5. 88 ibid paras 186, 188, 196, 202, 204, 206, 208. 89 ibid paras 220, 307, 387. To make this finding, the Tribunal engaged in a detailed examination of the relevant Canadian court decisions, the purpose of which was ‘to assess the factual basis of Claimant’s case’ rather than to judge the domestic rulings against an international legal standard: para 327. 90 ibid para 223. 83 ibid

340 Joshua Paine such courts. It will accordingly only be in very exceptional circumstances, in which there is clear evidence of egregious and shocking conduct, that it will be appropriate for a NAFTA Chapter Eleven tribunal to assess such conduct against the obligations of the respondent State under NAFTA Article 1105(1).91

A subsequent part of the award gives some indication of what such deference may look like. Specifically, beyond the issue of whether there had been a dramatic change in Canadian law, the Tribunal also considered the claimant’s argument that the promise utility doctrine, as adopted by Canadian courts, was arbitrary or discriminatory and constituted a breach of the international minimum standard or an expropriation.92 Notably, in rejecting the arbitrariness claim, the Tribunal found that various aspects of the promise utility doctrine were ‘rationally connected to … legitimate policy goals’ and it was not for the Tribunal to pronounce on whether the particular aspects of the doctrine were ‘the only, or the best, means of achieving’ Canada’s objectives.93 The Tribunal would not ‘question the correctness of the policies or the courts’ decisions’.94 The Tribunal also firmly rejected the suggestion that inconsistency among Canadian court decisions in applying the doctrine could amount to arbitrariness, noting ‘[s]ome level of unpredictability is present in the application of all law’.95 In short, despite its suggestion that domestic judicial conduct can be challenged under investment treaties beyond claims of denial of justice, the Eli Lilly award indicates a substantial degree of deference towards domestic judicial decisions.

IV. CONCLUSION: OBSERVATIONS ON THE CASES VIEWED AS A WHOLE

While the cases discussed in this chapter arose from different contexts, at a more general level several overarching themes can be identified. First, deference based on the specialised expertise or institutional capacity of particular kinds of domestic decision-makers, which may make them better placed than arbitrators to make certain determinations, is generally accepted by arbitrators (Glamis concerning a cultural review process, Marfin concerning a central bank’s management of a financial crisis). This reflects the fact that arbitrators are not specialists in the many non-legal forms of expertise on which state measures may be based.96 Second, there are examples of deference based on the greater accountability or legitimacy of national-level authorities, particularly in cases where the relevant determination involved a discretionary value judgment or an assessment of the domestic public interest (Philip Morris and Marfin, and the general remarks in Bilcon and SD Myers).97 Third, and running against the last two factors, there is a concern that domestic processes may lack independence (Tatneft and Bilcon). Before affording a degree of deference, investment arbitrators will need to be satisfied that a domestic process was not a cover for protectionist ends. Independence as a condition for defence is particularly prominent in relation to domestic judicial processes. Fourth, although

91 ibid para 224. The Tribunal also appeared to leave open the point that a domestic court decision could constitute an expropriation without amounting to a denial of justice: para 225. 92 ibid para 389. 93 ibid para 423. Similarly paras 425–426, 428, 430. 94 ibid para 430. 95 ibid para 421; See also para 429. 96 Similarly: Fahner, above (n 3) 151–152 (concluding ‘deference on this ground is relatively uncontroversial’). 97 A related idea is that national-level decision-makers are potentially more familiar with local conditions: see generally Henckels, above (n 1) 37–8; Schill, above (n 3) 600–601.

Standard of Review

341

the respondent state is treated as a unitary whole for the purposes of state responsibility, investment arbitrators are particularly wary of second-guessing determinations of domestic law by domestic courts (Mondev, Eli Lilly, and the general remarks in Tatneft). This is due to ideas about adjudication being a rational and impartial process,98 domestic courts’ familiarity with their own law, and the concern that international tribunals should not become courts of appeal. Nevertheless, as has been illustrated, investment treaty jurisprudence provides an opening for greater challenges to domestic adjudicatory processes by permitting claims beyond denial of justice. Fifth, throughout the awards discussed, arbitrators emphasised that the mere fact that a measure was arguably sub-optimal or mistaken from a public policy perspective could not establish a breach of an investment treaty. Relatedly, the awards regarding determinations of domestic law frequently highlighted that a breach of domestic law, of itself, was not sufficient to establish a breach of treaty. Put differently, there is general acknowledgment that states must be able to balance competing interests and make mistakes without breaching an investment treaty. In explaining the additional element that is required to establish an investment treaty breach, arbitrators often rely on concepts drawn from the international minimum standard, such as arbitrariness, manifest unfairness, abusiveness or bad faith.99 Finally, many of awards that included important, often somewhat generalised, statements about the standard of review were obviously high profile cases, where one could expect some concern given the sensitive domestic policy interests at issue (SD Myers, Glamis, Philip Morris v Uruguay, Bilcon, Eli Lilly). This is not coincidental. The awards addressing the standard of review have been rendered in a context of growing concerns about the intrusiveness of investor-state arbitration upon domestic policy space. Discussions of the appropriate standard of review and deference are one way in which arbitrators attempt to address such concerns.100

98 See

Douglas, above (n 67) 876–7. criticism is that such terms are evaluative and the reasoning is thus impressionistic: see Henckels, above (n 1)

99 One

117–22. 100 See also Shirlow, above (n 3) 31–32 (discussing the point that deference may be used strategically by international adjudicators to ‘build, maintain or defend their legitimacy’).

21 Expropriation Cases DIANE A DESIERTO*

E

XPROPRIATION HAS BEEN described as ‘the single greatest impairment of the security of an investment. It is an act that eliminates all or substantially all of the value of the investment to the investor. The threat of expropriation was a principal motivating factor in the origin of bilateral investment treaties.’1 In the landscape of international investment arbitration decisions, expropriation is very often one of the litigated issues.2 While expropriation decisions are virtually ubiquitous throughout all international investment arbitral awards,3 various criteria apply in order to warrant discussing or including any expropriation decision in this chapter. First, where a decision contains distinctive doctrinal articulation, interpretive elaboration, or completeness of reasoning as to the nature and scope of expropriation breaches, that decision is included due to its instructive value both for investment law scholars and practitioners. Second, where an expropriation decision is cited with some authoritative frequency in future decisions, such a decision arguably paves the way for the genesis of a ‘jurisprudence constante’.4 Third, if an expropriation decision contains a significant point of departure from previous substantive doctrine, it would likewise be of value to examine the arbitral reasoning underlying such a departure. Any space for doctrinal or interpretive indeterminacy when it comes to expropriation decisions ultimately arises – not just from the variable formulations of expropriation clauses or provisions in international investment treaties5 – but also from continuously evolving legal concepts throughout civil law, common law, and hybrid

* Diane A. Desierto is Professor of Law and Global Affairs, Faculty Director of the LLM in International Human Rights Law, at Notre Dame Law School, with a joint appointment as full Professor at the Keough School of Global Affairs at the University of Notre Dame, where she is also Faculty Fellow in five of the University’s institutes (Klau Center for Civil and Human Rights, Kellogg Institute of International Studies, Liu Institute of Asia and Asian Studies, Pulte Institute of Global Development, Nanovic Institute of European Studies). She is concurrently Professor of International Law and Human Rights at the Philippines Judicial Academy of the Supreme Court of the Philippines, and Chair-Rapporteur of the United Nations Expert Drafting Group on the Right to Development. 1 KJ Vandevelde, Bilateral Investment Treaties: History, Policy, and Interpretation (Oxford, Oxford University Press, 2010) 271. 2 See PM Norton, ‘A Law of the Future or A Law of the Past? Modern Tribunals and the International Law of Expropriation’ (1991) 85 American Journal of International Law 3, 474. 3 See among others JM Cox, Expropriation in Investment Treaty Arbitration (Oxford, Oxford University Press, 2019); United Nations Conference on Trade and Development (UNCTAD), Expropriation, UNCTAD Series on Issues in International Investment Agreements II (2012), https://perma.cc/QVL3-8EEQ. 4 A Bjorklund, ‘Investment Treaty Arbitral Decisions as Jurisprudence Constante’ (2010) 7(1) Transnational Dispute Management 1, www.transnational-dispute-management.com/article.asp?key=1518. 5 See A Rajput, Regulatory Freedom and Indirect Expropriation in Investment Arbitration (The Netherlands, Kluwer Law International, 2018) ch 3 (Types of Expropriatory Measures).

344 Diane A Desierto legal systems as to the constitutive elements of property and community expectations as to what comprises justified or unjustified state takings of property.6

I. ELEMENTS OF AN EXPROPRIATION CLAUSE: ADC V HUNGARY AND AN UNLAWFUL EXPROPRIATION

The arbitral awards involving expropriation are legion,7 but none probably discuss the common elements of an expropriation clause, and the ensuing significant differences in the estimation of damages as between lawful expropriation and unlawful expropriation,8 as clearly as the award in ADC Affiliate Limited and ADC & ADMC Management Limited v The Republic of Hungary.9 The ADC case involved an infrastructure project for the renovation of an existing airport terminal and design of a new terminal in Hungary’s principal airport (Budapest-Ferigehy International Airport) for international and domestic flights.10 The Tribunal in this case was tasked to assess whether various measures taken by the Hungarian government violated Article 4 of the 1989 Hungary-Cyprus bilateral investment treaty. The claimants alleged that this was an unlawful expropriation by Hungary of their investment in the airport, precisely because the Hungarian government did not comply with the requirements under Article 4 of the said treaty.11 The claimants in ADC ultimately sought to establish an entitlement to damages as measured under the international law standard of compensation for an unlawful taking … due to the fact that actual restitution of the contractual rights confiscated by [Hungary] is impractical and considering Article 4 of the BIT in the context of the relevant rules of international customary law, the Claimants are entitled to (a) the consequential damages of the taking, plus (b) the greater of: a. the market value of the expropriated investment at the moment of expropriation; and b. the sum of (x) the market value of the expropriated investment at the date of the award, calculated with the benefit of post-taking information and (y) the value of the income that the Claimant would have earned from the expropriated investments between the date of the taking and the date of the award.12

6 See among others JH Herz, ‘Expropriation of Foreign Property’ (1941) 35 American Journal of International Law 2, 243–262; A Lenhoff, ‘Development of the Concept of Eminent Domain’ (1942) 42 Columbia Law Review 4, 596; EE Meidinger, ‘The ‘Public Uses’ of Eminent Domain: History and Policy’ (1980) 11 Environmental Law 1; AP Fachiri, ‘Expropriation and International Law’ (1925) 6 British Yearbook of International Law 159; JW Salacuse, The Law of Investment Treaties (Oxford, Oxford University Press, 2010). 7 See JM Cox, Expropriation in Investment Treaty Arbitration (Oxford, Oxford University Press, 2019); R Dolzer, ‘New Foundations of the Law of Expropriation of Alien Property’ (1981) 75 American Journal of International Law 3, 553. 8 See the later case of Bernardus Henricus Funnekotter and Others v Republic of Zimbabwe, Award, 22 April 2009 ICSID Case No ARB/05/6, paras 108–110, para 108, https://perma.cc/7DGQ-W2MQ (‘In [the Chorzow Factory] case, the Permanent Court [of International Justice] made a distinction between lawful and unlawful expropriation. It held that, in case of lawful expropriation, the damages suffered must be repaired through the “payment of fair compensation” or “the just price of what was expropriated” at the time of the expropriation. By contrast, it decided that, in case of unlawful expropriation, international law provides for restitutio in integrum or, if impossible, its monetary equivalent at the time of the judgment.’). 9 ADC Affiliate Limited and ADC & ADMC Management Limited v The Republic of Hungary, Award, 2 October 2006 ICSID Case No ARB/03/16 (Tribunal President: the Hon. Charles Brower; Tribunal Members: Professor Albert Jan van den Berg, Neil Kaplan CBE QC), https://perma.cc/3AF2-XJY2. 10 ibid paras 86–94. 11 ibid para 11. 12 ibid paras 241–242.

Expropriation Cases

345

This was a much broader standard for compensation, well beyond the literal terms of compensation explicitly provided for (eg ‘market value of the expropriated investments at the moment of the expropriation’) in the expropriation clause at issue, namely Article 4 of the 1989 Hungary-Cyprus bilateral investment treaty:13 Article 4 1.

Neither Contracting Party shall take any measures depriving, directly or indirectly, investors of the other Contracting Party 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.

2. 3. 4.

5.

The amount of compensation must correspond to the market value of the expropriated investments at the moment of the expropriation. The amount of this compensation may be estimated according to the laws and regulations of the country where the expropriation is made. The compensation must be paid without undue delay upon completion of the legal expropriation procedure, but not later than three months upon completion of this procedure and shall be transferred in the currency in which the investment is made. In the event of delays beyond the three-months’ period, the Contracting Party concerned shall be liable to the payment of interest based on prevailing rates. Investors of either Contracting Party who suffer losses of their investments in the territory of the other Contracting Party due to war or other armed conflict or state of emergency in the territory of the other Contracting Party, shall be treated, with respect to the compensations for these losses, as Investor of any third State.14

The ADC Tribunal prefaced its expropriation analysis by linking the state’s right to regulate with the responsibilities that the same state assumes in entering into an investment treaty: The Tribunal cannot accept the Respondent’s position that the actions taken by it against the Claimants were merely an exercise of its rights under international law to regulate its domestic economic and legal affairs. It is the Tribunal’s understanding of the basic international law principles that while a sovereign State possesses the inherent right to regulate its domestic affairs, the exercise of such right is not unlimited and must have its boundaries. As rightly pointed out by the Claimants, the rule of law, which includes treaty obligations, provides such boundaries. Therefore, when a State enters into a bilateral investment treaty like the one in this case, it becomes bound by it and the investment protection obligations it undertook therein must be honoured rather than be ignored by a later argument of the State’s right to regulate … It is one thing to say that an investor shall conduct its business in compliance with the host State’s domestic laws and regulations. It is quite another to imply that the investor must also be ready to accept whatever the host State decides to do to it.15

Turning to the first two elements of expropriation, namely, the public interest behind the expropriatory measures and the observance of due process of law, the ADC Tribunal categorically declared that it ‘can see no public interest being served by the Respondent’s depriving actions of the Claimants’ investments in the Airport Project’16 and that ‘the taking was not under due

13 See sample expropriation clauses in United Nations Conference on Trade and Development (UNCTAD), Expropriation, UNCTAD Series on Issues in International Investment Agreements II (2012), https://perma.cc/ QVL3-8EEQ. 14 ADC Affiliate Limited and ADC & ADMC Management Limited v The Republic of Hungary, above (n 9) para 368. 15 ibid paras 423–424. 16 ibid para 429.

346 Diane A Desierto process of law as required by Article 4 of the BIT’.17 The Tribunal held that Hungary had failed to provide substantive factual and legal basis for the measures to amount to having been taken for the public interest: A treaty requirement for ‘public interest’ requires some genuine interest of the public. If mere reference to ‘public interest’ can magically put such interest into existence and therefore satisfy this requirement, then this requirement would be rendered meaningless since the Tribunal can imagine no situation where this requirement would not have been met.18

Where other tribunals often appeared more deferential to a host state’s assertion of public interest behind an expropriation,19 the ADC award emphasised the evidentiary burden of the host state to demonstrate the public interest behind any contemplated expropriation measure. Hungary had likewise failed to prove any due process of law was observed in the Hungarian Government’s takeover of the Airport Project, with the Tribunal stressing that: (…) ‘due process of law’, in the expropriation context, demands an actual and substantive legal procedure for a foreign investor to raise its claims against the depriving actions already taken or about to be taken against it. Some basic legal mechanisms, such as reasonable advance notice, a fair hearing and an unbiased and impartial adjudicator to assess the actions in dispute, are expected to be readily available and accessible to the investor to make such legal procedure meaningful. In general, the legal procedure must be of a nature to grant an affected investor a reasonable chance within a reasonable time to claim its legitimate rights and have its claims heard. If no legal procedure of such nature exists at all, the argument that ‘the actions are taken under due process of law’ rings hollow. And that is exactly what the Tribunal finds in the present case.20

The ADC Tribunal likewise found no difficulty establishing that Hungary had not met the remaining two elements of the expropriation clause in Article 4 of the BIT, namely, that the expropriatory measures are not discriminatory and that just compensation was paid. The Tribunal restated the standard for discriminatory treatment and found on the facts that Hungary never satisfied this standard: In order for a discrimination to exist, particularly in an expropriation scenario, there must be different treatments to different parties. However and unfortunately, the Respondent misses the point because the comparison of different treatments is made here between that received by the Respondent appointed operator and that received by foreign investors as a whole. The Tribunal therefore rejects the contentions made by the Respondent and concludes that the actions taken by the Respondent against the Claimants are discriminatory.21

Similarly, because Hungary did not pay any compensation whatsoever when it took over the Airport Project, the Tribunal had little trouble concluding that the requirement in Article 4 of the investment treaty for just compensation was not satisfied.22 Because the ADC Tribunal declared the expropriation to be unlawful, it did not confine itself to the assessment of damages as ‘just compensation’ required in Article 4(1)(c) of the 1989

17 ibid

para 434. para 432. 19 See among others Ioannis Kardassopoulos and Ron Fuchs v The Republic of Georgia, Award, 3 March 2010, ICSID Case Nos ARB/05/18 and ARB/07/15, para 391, https://perma.cc/8JA6-EKSD; LIAMCO v Libya, Award of 12 April 1977, para 241 (‘… Motives are indifferent to international law, each State being free “to judge for itself what it considers useful and necessary for the public good … The object pursued by it is of no concern to third parties” …’). 20 Ioannis Kardassopoulos and Ron Fuchs v The Republic of Georgia, above (n 19) para 435. 21 ibid paras 442–443. 22 ibid para 444. 18 ibid

Expropriation Cases

347

Hungary-Cyprus bilateral investment treaty.23 Instead, it stressed that ‘[s]ince the BIT does not contain any lex specialis rules that govern the issue of the standard for assessing damages in the case of an unlawful expropriation, the Tribunal is required to apply the default standard contained in customary international law in the present case’.24 Citing to the Chorzow Factory standard set by the Permanent Court of International Justice, investor-state arbitral decisions as well as subsequent decisions of the International Court of Justice,25 the Tribunal recalled that ‘reparation must, as far as possible, wipe out all the consequences of the illegal act and re-establish the situation which would, in all probability, have existed if that act had not been committed’.26 The ADC Tribunal then stated: The remaining issue is what consequence does application of this customary international law standard have for the present case. It is clear that actual restitution cannot take place and so it is, in the words of the Chorzów Factory decision, “payment of a sum corresponding to the value which a restitution in kind would bear”, which is the matter to be decided.27

The Tribunal then noted that the case before it was unique, in that the value of the investment after the date of expropriation (1 January 2002) has risen very considerably while other arbitrations that apply the Chorzów Factory standard all invariably involve scenarios where there has been a decline in the value of the investment after regulatory interference. It is for this reason that application of the restitution standard by various arbitration tribunals has led to use of the date of the expropriation as the date for the valuation of damages.28

The Tribunal then ruled sui generis that it would apply the Chorzów Factory standard where ‘the date of valuation is the date of the Award and not the date of expropriation, since this is what is necessary to put the Claimants in the same position as if the expropriation had not been committed’.29 As compensable damages for the expropriation, the Tribunal thus awarded ‘ADC Affiliate US$55,426,973 and ADC & ADMC Management US$20,773,027 both sums to carry interest at 6% per annum with monthly rests until payment’,30 along with costs and expenses of the arbitration.31 The unlawfulness of the host state’s expropriation thus results in varying effects as to how compensation will be estimated (whether strictly according to just compensation under the expropriation clause in the investment treaty, or factoring in other compensable damages arising from the State’s breach of international investment treaty obligations),32 depending on how tribunals see the functions of reparation under the general law of international responsibility, and in relation to international investment law.33

23 See

I Marboe, ‘Reparation cases’, ch 26 in this book. Kardassopoulos and Ron Fuchs v The Republic of Georgia, above (n 19) para 483. 25 ibid paras 484–494. 26 ibid para 484. 27 ibid para 495. 28 ibid para 496. 29 ibid para 497. 30 ibid para 521. 31 ibid para 543. 32 See Siag & Vecchi v Egypt, Award (1 June 2009) ICSID Case No ARB/05/15; Yukos Universal Limited (Isle of Man) v The Russian Federation, PCA Case No AA 227, Final Award, 18 July 2014; Eureko v Poland, UNCITRAL Partial Award, 19 August 2005. 33 See D Khachvani, ‘Compensation for Unlawful Expropriation: Targeting Illegality’ (2017) 32 ICSID Review 2, 385–403; SR Ratner, ‘Compensation for Expropriations in a World of Investment Treaties: Beyond the Lawful/ Unlawful Distinction’ (2017) 111 American Journal of International Law 7, 7– 56. 24 Ioannis

348 Diane A Desierto II. DISTINCTIONS BETWEEN DIRECT AND INDIRECT EXPROPRIATION: METALCLAD CORPORATION V UNITED MEXICAN STATES34

While the Metalclad case is reported to be ‘the most cited authority for the proposition that “value” can be expropriated’,35 it is arguably equally significant in how broadly it drew conceptual distinctions between direct expropriation and indirect expropriation. Claimant Metalclad is an American corporation, whose subsidiary ECO owns 100 per cent of shares in ECONSA, a Mexican corporation.36 ECONSA purchased another Mexican corporation, COTERIN, with a view to acquiring, developing, and operating COTERIN’s hazardous waste transfer station and landfill in La Pedrera, Guadalcazar, in the Mexican State of San Luis Potosi, within the United Mexican States (hereafter, ‘Mexico’). COTERIN owns the landfill property and all permits and licences on which its parent company, Claimant Metalclad, based its investment claim against Mexico, alleging expropriatory actions by the sub-federal units of Guadalcazar and San Luis Potosi.37 The government of San Luis Potosi had given COTERIN a State land use permit to construct the landfill, shortly after Mexico’s federal Government ecological agencies granted COTERIN a federal permit to construct a hazardous waste landfill in La Pedrera.38 Metalclad maintained it received further assurances, representations, and statements from Mexican Government officials that all required permits had been issued for the hazardous waste and landfill project.39 However, after Mexican Government agencies conducted an environmental audit of the project and entered into an agreement (Convenio) that allowed landfill operation, the Governor of San Luis Potosi denounced the Convenio and the municipal local government proceeded to deny all operation and construction permits.40 The same Governor later issued an Ecological Decree declaring an area encompassing the landfill site as a natural area for the protection of rare cactus.41 Metalclad maintained that all of these actions by Mexico, through its local governments San Luis Potosi and Guadalcazar, violated the expropriation clause (Article 1110) in the North American Free Trade Agreement (NAFTA) investment chapter.42 The Tribunal noted that this expropriation provision contains the following elements: NAFTA Article 1110 provides that ‘[n]o party shall directly or indirectly … expropriate an investment … or take a measure tantamount to … expropriation … except: (a) for a public purpose; (b) on a nondiscriminatory basis; (c) in accordance with due process of law and Article 1105(1); and (d) on payment of compensation ….” “A measure” is defined in Article 201(1) as including “any law, regulation, procedure, requirement or practice”’.43

34 Metalclad Corporation v United Mexican States, Award (30 August 2000) ICSID Case No ARB(AF)/97/1 (Tribunal President: Sir Elihu Lauterpacht; Tribunal Members: Messrs Benjamin R Civiletti and Jose Luis Siqueiros), https:// perma.cc/6ZS5-EUTX. 35 Z Douglas, ‘Property Rights as the Object of Expropriation: Emmis v Hungary, ICSID Case No ARB/12/2’ in M Kinnear and others (eds), Building International Investment Law: The First 50 Years of ICSID (The Netherlands, Kluwer Law International, 2016) 341. 36 Metalclad Corporation v United Mexican States, above (n 34) para 2. 37 ibid para 2. 38 ibid paras 28 and 31. 39 ibid paras 32–36, 41–44. 40 ibid paras 48–50. 41 ibid para 59. 42 ibid para 72. 43 ibid para 102.

Expropriation Cases

349

The Tribunal elaborated the above elements to mean that expropriation under NAFTA includes not only open, deliberate and acknowledged takings of property, such as outright seizure or formal or obligatory transfer of title in favour 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.44

Applying this test to the factual circumstances, the Tribunal found that Mexico ‘participa[ted] or acquiesc[ed] in the denial to Metalclad of the right to operate the landfill, notwithstanding the fact that the project was fully approved and endorsed by the federal government … Mexico must be held to have taken a measure tantamount to expropriation in violation of NAFTA Article 1110(1)’.45 In reaching this conclusion, the Tribunal attributed the conduct of the local governments of Guadalcazar and San Luis Potosi to Mexico, and noted Metalclad’s rightful reliance on the representations of the Mexican federal Government: ‘These measures, taken together with the representations of the Mexican federal government, on which Metalclad relied, and the absence of a timely, orderly or substantive basis for the denial by the Municipality of the local construction permit, amount to an indirect expropriation.’46 The Tribunal additionally found the Ecological Decree to be also expropriatory, noting that this finding was itself ‘not necessary for [the Tribunal’s] conclusion [that expropriation took place] … [but] the Decree had the effect of barring forever the operation of the landfill’.47 Neither did the Tribunal deem it relevant to decide or consider the motivation or intent of the adoption of the Ecological Decree. Indeed, a finding of expropriation on the basis of the Ecological Decree is not essential to the Tribunal’s finding of a violation of NAFTA Article 1110. However, the Tribunal considers that the implementation of the Ecological Decree would, in and of itself, constitute an act tantamount to expropriation.48

The Metalclad case significantly introduced to the interpretation of expropriation under NAFTA Article 1110 the concept of varying degrees of prohibited state interference with the use of property, that have the effect of depriving property owners wholly or partly of actual or expected uses of property. This concept significantly distinguishes direct expropriation (which is an open and deliberate state taking of property for a public purpose, in a manner that is nondiscriminatory and pursuant to due process, and always subject to payment of compensation49), from indirect expropriation which can arise from state regulations that create the partial or full effect of interference with, or deprivation of, an investor’s property use.50 It was also illustrative of an arbitral tribunal’s deliberate unwillingness to inquire into the possible genuine public or environmental purposes behind regulations that created such interference with property use, even as a possible ground for balancing the estimation of damages. The Tribunal awarded the fair market value of Metalclad’s actual investment in the project,51 less disallowed expenses, for a total amount of US$16,685,000.00, with 6 per cent monthly

44 ibid

para 103. para 104. 46 ibid para 107. 47 ibid para 109. 48 ibid para 111. 49 See European Media Ventures v Czech Republic, UNCITRAL, Partial Award on Liability, 8 July 2009. 50 See Wena Hotels v Arab Republic of Egypt, Award (8 December 2000) ICSID Case No ARB/98/4, paras 99–101 (cf ‘allowing an entity [over which Egypt could exert effective control] to seize and illegally possess the hotels for nearly a year is more than an ephemeral interference in the use of that property or with the enjoyment of its benefits.’). 51 ibid para 122. 45 ibid

350 Diane A Desierto compounded interest to accrue on any part of the unpaid award after 45 days from the time the award is rendered.52 Metalclad’s introduction of state regulatory interference or deprivation of private property use as part of indirect or ‘creeping’ expropriation reinforces similar distinctions raised in cases decided by the Iran-United States Claims Tribunal, which emphasised factors of substantial impairment, deprivation, or the ‘rendered uselessness’ of such rights as a result of state action, even if there was no formally declared state process for expropriation.53 Metalclad would be cited authoritatively in other cases.54 Metalclad also articulated a much broader test of partial or full deprivation of actual, as well as expected, uses of property as the decisive criterion for characterising a state regulatory action as an indirect or creeping expropriation. Significantly, Metalclad was decided just a few months after Compania del Desarrollo de Santa Elena SA v The Republic of Costa Rica,55 a case which involved direct expropriation of property by Costa Rica, but where the Tribunal referred to a narrower test of deprivation of actual benefit and use of property: ‘a property has been expropriated when the effect of the measures taken by the state has been to deprive the owner of title, possession or access to the benefit and economic use of his property’.56 Nevertheless, some arbitral tribunals have required that claimants demonstrate an intensity or severity of effect on property rights, arising from the degree of state interference or the economic deprivation arising from regulatory actions, in order to qualify the latter as forms of indirect or creeping expropriation. Thus, in a case that rejected the existence of expropriation arising from a governmental ban on imports of polychlorinated biphenyl (PCB) wastes, the Tribunal in SD Myers Inc v Government of Canada stressed that ‘the general body of precedent usually does not treat regulatory action as amounting to expropriation’,57 such that: Expropriations tend to involve the deprivation of ownership rights; regulations a lesser interference. The distinction between expropriation and regulation screens out most potential cases of complaints concerning economic intervention by a state and reduces the risk that governments will be subject to claims as they go about their business of managing public affairs. An expropriation usually amounts to a lasting removal of the ability of an owner to make use of its economic rights although it may be that, in some contexts and circumstances, it would be appropriate to view the deprivation as amounting to an expropriation, even if it were partial or temporary.58

The Tribunal in Pope and Talbot Inc v Government of Canada also required similar consideration for the magnitude or severity of the effects from a regulatory measure,59 ultimately

52 ibid

para 131. Starrett Housing v Iran, IUSCT Case No 32-24-1, Interlocutory Award, 19 December 1983 4 IUSCT Reports 122, 154; Tippetts v TAMS-AFFA Consulting Engineers of Iran, IUSCT Case No 7, 29 June 1984, 6 ISUCT Reports 219. 54 See among others CME Czech Republic BV (Netherlands) v The Czech Republic, UNCITRAL, Partial Award, 13 September 2001, para 606; Tecnicas Medioambientales Tecmed SA v The United Mexican States, Award (29 May 2003) ICSID Case No ARB(AF)/00/2, para 113; Alpha Projektholding GMBH v Ukraine, Award (8 November 2010) ICSID Case No ARB/07/16, para 408; Railroad Development Corporation v The Republic of Guatemala, Award (29 June 2012) ICSID Case No ARB/07/23, para 151; Sempra v The Argentine Republic, Award (28 September 2007) ICSID Case No ARB/02/16, para 403; Telenor Mobile v Hungary, Award (13 September 2006) ICSID Case No ARB/04/15, para 66. 55 Compania del Desarrollo de Santa Elena SA v The Republic of Costa Rica, Final Award (17 February 2000) ICSID Case No ARB/96/1. 56 ibid para 77. 57 SD Myers Inc v Government of Canada, Partial Award, NAFTA 13 November 2000, para 281. 58 ibid paras 282–283. 59 Pope & Talbot v Government of Canada, Interim Award, NAFTA 26 June 2000, para 96. 53 See

Expropriation Cases

351

finding that Canada’s Export Control Regime did not cause any expropriation of the investor’s softwood lumber business.60 The Pope & Talbot v Canada Tribunal found that: The degree of interference with the Investment’s operations due to the Export Control Regime does not rise to an expropriation (creeping or otherwise) within the meaning of Article 1110. While it may sometimes be uncertain 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 the owner.61

Similarly, the Tribunal in Marvin Feldman v Mexico rejected the argument that a regulatory action – denial of tax refunds permitted to certain exporters – supposedly was a measure tantamount to expropriation: (1) … not every business problem experienced by an investor is an expropriation under [NAFTA] Article 1110; (2) NAFTA and customary international law do not require a state to permit ‘gray market’ exports of cigarettes; (3) at no relevant time has the … law, as written, afforded Mexican cigarette resellers such as CEMSA a ‘right’ to export cigarettes (due primarily to technical/legal requirements for invoices stating tax amounts separately and to their status as non-taxpayers); and (4) the Claimant’s ‘investment,’ the exporting business known as CEMSA, as far as this Tribunal can determine, remains under the complete control of the Claimant, in business with the apparent right to engage in the exportation of alcoholic beverages, photographic supplies, contact lenses, powdered milk and other Mexican products – any product that it can purchase upon receipt of invoices stating the tax amounts – and to receive rebates of any applicable taxes under the IEPS law. 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.62

Later tribunals would combine Metalclad with intensity or effects approaches as part of multiple tests and arbitral reasoning in evaluating the nature of ‘creeping’ or indirect expropriation.63 For example, the Tribunal in Azurix Corporation v The Argentine Republic found ‘useful guidance’64 from considering as an additional element for determining the nature of indirect or creeping expropriation the factor of proportionality65 (originally articulated in Tecnicas Medioambientales Tecmed SA v The United Mexican States66), finding that the issue is not so much whether the measure concerned is legitimate and serves a public purpose, but whether it is a measure that, being legitimate and serving a public purpose, should give rise to a compensation claim. In the exercise of their public policy function, governments take all sorts of measures that may affect the economic value of investments without such measures giving rise to a need to compensate …. The public purpose criterion as an additional criterion to the effect of the measures under consideration needs to be complemented ….[with] the case law of the European Court of Human Rights… The Court held that “a measure depriving a person of his property [must] pursue, on the facts as well as in principle, a legitimate aim “in the public interest”’

and bear ‘a reasonable relationship of proportionality between the means employed and the aim sought to be realized’. This proportionality will not be found if the person concerned bears ‘an individual and excessive burden’. The Court considered that such ‘a measure must

60 ibid

para 100. para 102. para 111. 63 See Bear Creek Mining Corporation v Republic of Peru, Award (30 November 2017) ICSID Case No ARB/14/21, 105–167; Generation Ukraine v Ukraine, Award (16 September 2003) ICSID Case No ARB/00/9, 81–96. 64 Azurix Corporation v The Argentine Republic, Award (14 July 2006) ICSID Case No ARB/01/12, para 312. 65 ibid paras 310–311. 66 ibid; Tecnicas Medioambientales Tecmed SA v The United Mexican States, above (n 54). 61 ibid 62 ibid

352 Diane A Desierto be both appropriate for achieving its aim and not disproportionate thereto’. The Court found relevant that non-nationals ‘will generally have played no part in the election or designation of its [of the measure] authors nor have been consulted on its adoption’, and observed that ‘there may well be legitimate reason for requiring nationals to bear a greater burden in the public interest than non-nationals’.67 Finally, the case of Yukos Universal Limited (Isle of Man) v The Russian Federation is also a significant post-Metalclad case, with the Tribunal concluding that while Russia did not take any direct expropriation measures, Russia’s tax (and related enforcement) measures against the claimants ultimately had the effect of total expropriation within the purview of Article 13 of the Energy Charter Treaty: The primary objective of the Russian Federation was not to collect taxes but rather to bankrupt Yukos and appropriate its valuable assets … if the true objective were no more than tax collection, Yukos, its officers and employees, and its properties and facilities, would not have been treated, and mistreated, as in fact they were. Among the many incidents in this train of mistreatment that are within the remit of this Tribunal, two stand out: finding Yukos liable for the payment of more than 13 billion dollars in VAT in respect of oil that had been exported by the trading companies and should have been free of VAT and free of fines in respect of VAT; and the auction of YNG at a price that was far less than its value. But for these actions, for which the Russian Federation for reasons set out above and in preceding chapters was responsible, Yukos would have been able to pay the tax claims of the Russian Federation justified or not; it would not have been bankrupted and liquidated (unless the Russian Federation were intent on its liquidation and found still additional grounds for achieving that end, as the second criminal trial of Messrs. Khodorkovsky and Lebedev indeed suggests). Respondent has not explicitly expropriated Yukos or the holdings of its shareholders, but the measures that Respondent has taken in respect of Yukos … in the view of the Tribunal have had an effect ‘equivalent to nationalization or expropriation.’68

The same Tribunal found that two elements of a lawful expropriation under the Energy Charter Treaty had not been met, in any case, since the ‘effective expropriation of Yukos was [not] carried out under due process of law’69 and there was no payment of ‘any compensation whatsoever’.70

III. PROPERTY SUBJECT OF EXPROPRIATION: WASTE MANAGEMENT INC V UNITED MEXICAN STATES AND ENCANA V ECUADOR

The property that could be subject of direct or indirect expropriation contemplate the full range of all property rights ‘existing under national law that have been taken by the State’.71 As such, property taken as a result of a state’s measures of direct or indirect expropriation could be real property, personal or movable property, as well as rights arising from contracts.72 In Waste

67 ibid;

Azurix Corporation v The Argentine Republic, above (n 64) paras 310–311. Emphasis added. Man) v the Russian Federation, PCA Case No AA 227, Final Award, 18 July 2014, paras 1579–1580, https://perma.cc/B9D9-UJPJ. 69 ibid para 1583. 70 ibid para 1584. 71 Tidewater Investment SRL and Tidewater Caribe CA v The Bolivarian Republic of Venezuela, Award (13 March 2015) ICSID Case No ARB/10/5, para 116. 72 AMOCO v Iran, IUSCT Case No 310-56-3, Award, 14 July 1987, para 108 (‘… expropriation … may extend to any right which can be the object of a commercial transaction, ie freely sold and bought, and thus has monetary value.’). See also SPP v Egypt, Award on the Merits (20 May 1992) ICSID Case No ARB/84/3, para 164 (‘… Nor can the Tribunal accept the argument that the term “expropriation” applies only to jus in rem. The Respondent’s cancellation 68 Yukos Universal Limited (Isle of

Expropriation Cases

353

Management Inc v United Mexican States,73 an investor claim against a series of governmental measures alleging impairment and/or destruction of its concession to provide waste disposal services (in the Mexican city of Acapulco in the State of Guerrero, one of the component States of Mexico), the Tribunal noted the specific language of NAFTA Article 1110, distinguishing between direct or indirect (‘creeping’) expropriation on the one hand, and ‘measures tantamount to expropriation’, on the other hand. The Tribunal clarified: (…) an indirect expropriation is still a taking of property. By contrast where a measure tantamount to an expropriation is alleged, there may have been no actual transfer, taking or loss of property by any person or entity, but rather an effect on property which makes formal distinctions of ownership irrelevant … Evidently the phrase ‘take a measure tantamount to nationalization or expropriation of such an investment’ in Article 1110(1) was intended to add to the meaning of the prohibition, over and above the reference to indirect expropriation. Indeed there is some indication that it was intended to have a broad meaning, otherwise it is difficult to see why Article 1110(8) was necessary. As a matter of international law a ‘non-discriminatory measure of general application’ in relation to a debt security or loan which imposed costs on the debtor causing it to default would not be considered expropriatory or even potentially so. It is true that paragraph (8) is stated to be ‘for greater certainty’, but if it was necessary even for certainty’s sake to deal with such a case this suggests that the drafters entertained a broad view of what might be ‘tantamount to an expropriation.’74

The Tribunal’s analysis yielded the conclusion that there was no direct or indirect expropriation of the concession resulting from regulatory actions taken by Mexican authorities,75 and neither was there a measure tantamount to expropriation. The Waste Management tribunal emphasised: [174]. … The mere non-performance of a contractual obligation is not to be equated with a taking of property, nor (unless accompanied by other elements) is it tantamount to expropriation. Any private party can fail to perform its contracts, whereas nationalization and expropriation are inherently governmental acts, as is envisaged by the use of the term ‘measure’ in Article 1110(1). It is true that, having regard to the inclusive definition of ‘measure’, one could envisage conduct tantamount to an expropriation which consisted of acts and omissions not specifically or exclusively governmental. All the same, the normal response by an investor faced with a breach of contract by its governmental counter-party (the breach not taking the form of an exercise of governmental prerogative, such as a legislative decree) is to sue in the appropriate court to remedy the breach. It is only where such access is legally or practically foreclosed that the breach could amount to a definitive denial of the right (i.e., the effective taking of the chose in action) and the protection of Article 1110 be called into play. [175]. The Tribunal concludes that it is one thing to expropriate a right under a contract and another to fail to comply with the contract. Non-compliance by a government with contractual obligations

of the project had the effect of taking certain important rights and interests of the Claimants. What was expropriated was not the land nor the right of usufruct, but the rights that SPP(ME), as a shareholder of ETDC, derived from EGOTH’S right of usufruct, which had been “irrevocably” transferred to ETDC by the State. Clearly, those rights and interests were of a contractual rather than in rem nature. However, there is considerable authority for the proposition that contract rights are entitled to the protection of international law and that the taking of such rights involves an obligation to make compensation therefore.’) 73 Waste Management Inc v United Mexican States, Award (30 April 2004) ICSID ARB(AF)/003 (Tribunal President: Professor James Crawford; Tribunal Members: Benjamin R Civiletti and Mr Eduardo Magallon Gomez), https:// perma.cc/XDM8-N9AT. 74 ibid paras 143–144. 75 ibid para 162.

354 Diane A Desierto is not the same thing as, or equivalent or tantamount to, an expropriation. In the present case the Claimant did not lose its contractual rights, which it was free to pursue before the contractually chosen forum. The law of breach of contract is not secreted in the interstices of Article 1110 of NAFTA. Rather it is necessary to show an effective repudiation of the right, unredressed by any remedies available to the Claimant, which has the effect of preventing its exercise entirely or to a substantial extent. [176]. In the present case, in the Tribunal’s view, this has not been shown. The question here is not one of final refusal to pay (combined with effective obstruction and denial of legal remedies); it is one of neglect and failure at the contractual level in the context of a marginal enterprise. That does not pass the test for an expropriatory taking of contractual rights as it emerges from the decisions analysed above. [177]. In the Tribunal’s view, it is not the function of the international law of expropriation as reflected in Article 1110 to eliminate the normal commercial risks of a foreign investor, or to place on Mexico the burden of compensating for the failure of a business plan which was, in the circumstances, founded on too narrow a client base and dependent for its success on unsustainable assumptions about customer uptake and contractual performance. A failing enterprise is not expropriated just because debts are not paid or other contractual obligations are not fulfilled. The position may be different if the available legal avenues for redress are blocked or are evidently futile in the face of governmental intransigence. But this was not the case here …76

Thus, what was decisive for the Waste Management v Mexico Tribunal in characterising contract rights as falling within the purview of property that could be subject to expropriation was not just the loss of contract rights per se, but the unavailability of legal redress or ordinary recourse in contract law for the deprivation of contract rights. It is the resultant inability – de jure as well as de facto – to exercise contract rights by an investor that would enable a conclusion that property in the form of contract rights has indeed been expropriated. EnCana Corporation v The Republic of Ecuador77 reiterated the above Waste Management test, but the Tribunal in EnCana drew a significant distinction between the concession contract rights asserted in Waste Management, and the investor’s claimed statutory entitlements to Value Added Tax (VAT) refunds in EnCana: [193]. That case concerned breach of contractual rights by local government bodies, not rights under tax legislation. There are differences between alleged governmental nonperformance of contractual obligations and governmental refusal to make payments allegedly required by statute, in particular tax refunds. Evidently only a government can engage in the latter conduct, and the determined exercise of executive authority by a host State may influence the situation decisively, notwithstanding such separation of powers as may exist under the constitutional practice of that State. As the tribunal said in Revere Copper, it is necessary to look at the substance and not only the form. [194]. All this is true. But there is nonetheless a difference between a questionable position taken by the executive in relation to a matter governed by the local law and a definitive determination contrary to law. In terms of the BIT the executive is entitled to take a position in relation to claims put forward by individuals, even if that position may turn out to be wrong in law, provided it does so in good faith and stands ready to defend its position before the courts. Like private parties, governments do not repudiate obligations merely by contesting their existence. An executive agency does not expropriate the value represented by a statutory obligation to make a payment or refund by mere refusal to pay, provided at least that (a) the refusal is not merely wilful, (b) the courts are open to the aggrieved private party, (c) the courts’ decisions are not themselves overridden or repudiated by the State.

76 ibid

paras 174–177. Emphasis added. Corporation v Ecuador, Award, LCIA 3 February 2006 (Tribunal President: Professor James Crawford; Tribunal Members: Horacio Grigera Naon, Christopher Thomas), https://perma.cc/S863-P8B7. 77 EnCana

Expropriation Cases

355

[195]. This principle applies equally to tax authorities as to other executive agencies. In the Tribunal’s view, the policy of a tax authority … is not reviewable under Article VIII of the BIT (having regard in particular to its Article XII) unless that policy in itself amounts to an actual and effective repudiation of legal rights.78

On the facts, the EnCana Corporation v Ecuador Tribunal declared that the Ecuadorian tax authority’s policy on oil refunds ‘never rose to the level of the repudiation of an Ecuadorian legal right’,79 and as such could not amount to an expropriation. While the deprivation, impairment, or loss of real or personal property could be readily observed, with respect to other forms of property such as private contract rights or claimed public or government entitlements, tribunals have sought to identify an additional test of actual and effective repudiation of such rights and entitlements, focusing on the actual unavailability of legal avenues for redress as would hollow out the value of such rights and entitlements. Other tribunals have generally sought to examine whether local remedies are available for the deprivation of contract rights.80

IV. THE STATE’S INHERENT POLICE POWER AND THE RIGHT TO REGULATE: METHANEX V USA AND PHILIP MORRIS V URUGUAY

Methanex v United States of America stands as the landmark case rejecting the existence of expropriation (direct or indirect) or any compensable regulatory taking when a state exercises its inherent police power in good faith and in a nondiscriminatory manner.81 On the facts, Methanex claimed compensation from the USA in the amount of approximately US$ 970 million (together with interest and costs), resulting from losses caused by the State of California’s ban on the sale and use of the gasoline additive known as ‘MTBE’ (methyl tertiary-butyl ether) which was then intended to become legally effective on 31 December 2002. MTBE is a synthetic, volatile, colourless and organic ether, with a turpentine-like taste and odour. Methanex was (and remains) the world’s largest producer of methanol, a feedstock for MTBE. It has never produced or sold MTBE.82

With respect to the alleged breach of the expropriation clause under Article 1110 of NAFTA, the Methanex Tribunal noted, preliminarily that in this case, there is no expropriation decree or a creeping expropriation. Nor was there a ‘taking’ in the sense of any property of Methanex being seized and transferred, in a single or a series of actions, to California or its designees. Insofar as Methanex can make a claim under Article 1110(1), it is not a claim for nationalization or expropriation, simpliciter, but for ‘measures tantamount to expropriation’.83

Referring back to Waste Management v Mexico,84 the Methanex Tribunal laid the crucial standard of the existence of specific governmental commitments or representations, in order

78 ibid

paras 193–195. para 197. 80 See among others Parkerings-Compagniet AS v Republic of Lithuania, Award (11 September 2007) ICSID Case No ARB/05/8, paras 316–317; R Dolzer, ‘Local Remedies in International Treaties: A Stocktaking’, in DD Caron and others (eds), Practicing Virtue: Inside International Arbitration (Oxford, Oxford University Press, 2015). 81 Methanex Corporation v United States of America, Final Award of the Tribunal on Jurisdiction and Merits, NAFTA, 3 August 2005 (Tribunal President: VV Veeder; Tribunal Members: J William F Rowley, W Michael Reisman), https://perma.cc/2MA5-AW2C. 82 ibid para 1. 83 ibid Pt IV, ch D, 3, para 6. 84 ibid Pt IV, ch D, 4, para 8. 79 ibid

356 Diane A Desierto to determine if a state’s nondiscriminatory exercise of police power and the right to regulate could possibly be expropriatory: … as a matter of general international law, a non-discriminatory regulation for a public purpose, which is enacted in accordance with due process and, which affects, inter alios, a foreign investor or investment is not deemed expropriatory and compensable unless specific commitments had been given by the regulating government to the then putative foreign investor contemplating investment that the government would refrain from such regulation.85

Because no such specific commitments were ever given by the US (or any of its organs) to Methanex, the Tribunal ruled that ‘the California ban was made for a public purpose, was non-discriminatory and was accomplished with due process. Hence, Methanex’s central claim under Article 1110(1) of expropriation under one of the three forms of action in that provision fails. From the standpoint of international law, the California ban was a lawful regulation and not an expropriation.’86 On the facts, the Tribunal found that: Methanex entered a political economy in which it was widely known, if not notorious, that governmental environmental and health protection institutions at the federal and state level, operating under the vigilant eyes of the media, interested corporations, nongovernmental organizations and a politically active electorate, continuously monitored the use and impact of chemical compounds and commonly prohibited or restricted the use of some of those compounds for environmental and/or health reasons. Indeed, the very market for MTBE in the United States was the result of precisely this regulatory process … Methanex entered the United States market aware of and actively participating in this process. It did not enter the United States market because of special representations made to it.87

Other cases have since cited to the landmark affirmation of police power and its boundaries under Methanex v USA.88 The tribunal in Saluka Investments BV v The Czech Republic crystallised the Methanex ruling with the crisp statement 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’.89 Among the post-Methanex cases, however, Philip Morris Brands SARL, Philip Morris Products SA, and Abal Hermanos SA v The Oriental Republic of Uruguay90 stands among the most significant due to its expansive elaboration of police power parameters well beyond the discussions in Methanex. In Philip Morris, the Claimants alleged that, through several tobacco-control measures regulating the tobacco industry, the Respondent violated the BIT in its treatment of the trademarks associated with cigarettes brands in which the Claimants had invested. These measures included the Government’s adoption of a single presentation requirement precluding tobacco manufacturers from marketing more than one variant of cigarette per brand

85 ibid

Pt IV, ch D, 4, para 7. Emphasis added. Pt IV, ch D, 7, para 15. 87 ibid Pt IV, ch D, 5, paras 9–10. 88 See among others Saluka Investments BV v The Czech Republic, PCA Case No 2001–04, Partial Award, 17 March 2006, para 262; Glamis Gold Limited v United States of America, Award, NAFTA 8 June 2009, para 620; Apotex Holdings Inc and Apotex v United States of America, Award (25 August 2014) ICSID Case No ARB(AF)/12/1, paras 6.23–6.24; Chemtura Corporation v Government of Canada, Award, NAFTA 2 August 2010, para 266. 89 Methanex Corporation v United States of America, above (n 81), Pt IV, ch D, 4, para 255. 90 Philip Morris Brands SARL, Philip Morris Products SA, and Abal Hermanos SA v The Oriental Republic of Uruguay, Award (8 July 2016) ICSID Case No ARB/10/7 (Tribunal President: Professor Piero Bernardini; Tribunal Members: Mr Gary Born and Judge James Crawford), https://perma.cc/4DQF-8SDT. 86 ibid

Expropriation Cases

357

family (‘Single Presentation Requirement or SPR’), and the increase in the size of graphic health warnings appearing on cigarette packages (the ‘80/80 Regulation’) ….91

Uruguay relied on the Methanex standard, arguing that the Challenged Measures were adopted in compliance with Uruguay’s international obligations, including the BIT, for the single purpose of protecting public health. According to Uruguay, both regulations were applied in a nondiscriminatory manner to all tobacco companies, and they amounted to a reasonable, good faith exercise of Uruguay’s sovereign prerogatives. The SPR was adopted to mitigate the ongoing adverse effects of tobacco promotion, including the Claimants’ false marketing that certain brand variants are safer than others, even after misleading descriptors such as ‘light,’ ‘mild,’ ‘ultra-light’ were banned. The 80/80 Regulation was adopted to increase consumer awareness of the health risks of tobacco consumption and to encourage people, including younger people, to quit or not to take up smoking, while still leaving room on packages for brand names and logos. Thus for the Respondent, this case is ‘about protection of public health, not interference with foreign investment.’92 The Tribunal in Philip Morris first characterised the property alleged to have been subject of expropriatory measures, finding that the challenged measures affect the claimants’ trademarks but such trademarks did not confer absolute rights of use free from any governmental regulation: ‘under Uruguayan law or international conventions to which Uruguay is a party the trademark holder does not enjoy an absolute right of use, free of regulation, but only an exclusive right to exclude third parties from the market so that only the trademark holder has the possibility to use the trademark in commerce, subject to the State’s regulatory power’.93 Examining the challenged measures, the Tribunal then found that: [T]here is not even a prima facie case of indirect expropriation by the 80/80 Regulation. The Marlboro brand and other distinctive elements continued to appear on cigarette packs in Uruguay, recognizable as such. A limitation to 20% of the space available to such purpose could not have a substantial effect on the Claimants’ business since it consisted only in a limitation imposed by the law on the modalities of use of the relevant trademarks. The claim that the 80/80 Regulation breached Article 5 of the BIT consequently fails.94

Separately from the failed claim of indirect expropriation, however, the Tribunal proceeded to conclude that the challenged measures were a valid exercise of Uruguay’s police powers to protect public health: ‘The Tribunal’s analysis might end here, leading to the dismissal of the Claimants’ claim of expropriation for the above reasons. There is however an additional reason in support of the same conclusion that should also be addressed in view of the Parties’ extensive debate in that regard. In the Tribunal’s view, the adoption of the Challenged Measures by Uruguay was a valid exercise of the State’s police powers, with the consequence of defeating the claim for expropriation under Article 5(1) of the BIT …. It is the Claimants’ contention that Article 5(1) of the BIT prohibits any expropriation unless it is carried out in accordance with the conditions established by said Article and that the existence of a public purpose, one of such conditions, does not exempt the State from the obligation to pay compensation. In the Claimants’ view, the State’s exercise of police powers does not constitute a defense against expropriation, or exclude the requirement of compensation. The Claimants add that there is no room under Article 5(1) or otherwise in the BIT for carving out an exemption based on the 91 ibid

para 9. para 13. para 271. 94 ibid para 276. 92 ibid 93 ibid

358 Diane A Desierto police powers of the State. The Tribunal disagrees. As pointed out by the Respondent, Article 5(1) of the BIT must be interpreted in accordance with Article 31(3) of the VCLT requiring that treaty provisions be interpreted in the light of ‘[a]ny relevant rules of international law applicable to the relations between the parties,’ a reference ‘which includes … customary international law.’ This directs the Tribunal to refer to the rules of customary international law as they have evolved. Protecting public health has since long been recognized as an essential manifestation of the State’s police power, as indicated also by Article 2(1) of the BIT which permits contracting States to refuse to admit investments ‘for reasons of public security and order, public health and morality.’95

The Philip Morris Tribunal referred to Methanex and other cases to emphasise the differentiation between police powers and indirect expropriation: The principle that the State’s reasonable bona fide exercise of police powers in such matters as the maintenance of public order, health or morality, excludes compensation even when it causes economic damage to an investor and that the measures taken for that purpose should not be considered as expropriatory did not find immediate recognition in investment treaty decisions. But a consistent trend in favor of differentiating the exercise of police powers from indirect expropriation emerged after 2000. During this latter period, a range of investment decisions have contributed to develop the scope, content and conditions of the State’s police powers doctrine, anchoring it in international law. According to a principle recognized by these decisions, whether a measure may be characterized as expropriatory depends on the nature and purpose of the State’s action… these provisions … reflect the position under general international law.96

Broader arbitral recognition of states’ valid exercises of police powers and the right to regulate in international investment disputes will impact the future analysis of regulatory measures that challenge today’s constructs of regulatory takings under modern international law. Regulatory measures to achieve climate change targets, the realisation and implementation of human rights obligations, as well as other public policies for security, immigration, wartime expropriations, migration, public health emergencies, cybersecurity and terrorism challenges, among others, call for careful criteria to evaluate state exercises of police power deemed non-expropriatory, as opposed to those that hearken back to obsolescing characterisations of regulatory takings as indirect or creeping expropriations.97 While Methanex was meticulous in prescribing the threshold of specific commitments, governmental assurances, or similar representations that government would ‘refrain from regulation’,98 such a high threshold might also prove unrealistic for states confronting the demands of cross-border cooperation and regulation as part of the next frontier of shared global and domestic challenges on climate change, human rights, automation, global commons management, and other public emergencies. Neither will the wide subjectivities of Philip Morris – employing a context-driven assessment of the nature and purpose of state action to elicit the existence of any ‘expropriatory’ character to such an action – completely suffice. The next series of landmark expropriation decisions in international investment law will have to confront the unique global, local, and cross-border paradigms of future public policy-making amid shared human development challenges.

95 ibid

paras 287–291. paras 295 and 301. Emphasis added. 97 See among others V Vadi, Public Health in International Investment Arbitration (London, Routledge, 2013); C Titi, The Right to Regulate in International Investment Law (Baden-Baden, Nomos, 2014); LW Mouyal, International Investment Law and the Right to Regulate (London, Routledge, 2016). 98 Azurix Corporation v The Argentine Republic, above (n 64) paras 310 and 311. 96 ibid

22 Landmark Cases on Fair and Equitable Treatment: Empowering and Controlling Arbitrators as Law-Makers STEPHAN W SCHILL*

I. INTRODUCTION: THE PROBLEM OF NORMATIVE VAGUENESS

F

AIR AND EQUITABLE treatment (FET) constitutes one of the core standards of international legal protection for foreign investors and their investments. FET clauses appear prominently in the majority of the more than 3,300 international investment agreements (IIAs);1 and even where omitted, FET may apply by operation of a most-favoured-nation clause.2 Historically, FET has also played an important role, as its connection to the customary international law minimum standard of treatment of aliens,3 and its inclusion in treaties that qualify as precursors of the present-day investment treaty regime,4 show FET clauses come with certain textual variations, the most important of which is a difference between clauses tying FET to the customary international law minimum standard and FET clauses defining FET autonomously.5 Most tribunals, however, loathe attributing much relevance to variations in the

* Stephan W Schill is Professor of International and Economic Law and Governance at the University of Amsterdam, Faculty of Law. 1 See United Nations Conference on Trade and Development (UNCTAD), World Investment Report 2019 – Special Economic Zones (New York, United Nations, 2019) 99. 2 See, eg, Rumeli Telekom AS and Telsim Mobil Telekomunikasyon Hizmetleri AS v Kazakhstan, Award (29 July 2008) ICSID Case No ARB/05/16, para 575; Hesham T M Al Warraq v Indonesia, Final Award (15 December 2014) UNCITRAL, para 555. 3 See J Paulsson, Denial of Justice in International Law (Cambridge, Cambridge University Press, 2005); M Paparinskis, The International Minimum Standard and Fair and Equitable Treatment (Oxford, Oxford University Press, 2013); T Weiler, The Interpretation of International Investment Law: Equality, Discrimination and Minimum Standards of Treatment in Historical Context (Leiden, Martinus Nijhoff, 2013). 4 Precursors include trade treaties in the inter-war period as well as the League of Nations Covenant, the failed Havana Charter of 1948, post-World War II friendship, commerce and navigation treaties concluded, amongst others, by the US, the 1959 Abs-Shawcross Draft Convention, and the 1967 OECD Draft Convention on the Protection of Foreign Property. See M Pinchis, ‘The Ancestry of ‘Equitable Treatment’ in Trade – Lessons from the League of Nations During the Inter-War Period’ (2014) 15(1-2) Journal of World Investment & Trade 13; S Vasciannie, ‘The Fair and Equitable Treatment Standard in International Investment Law and Practice’ (1999) 70 British Yearbook of International Law 99; C Yannaca-Small, ‘Fair and Equitable Treatment Standard in International Investment Law’ (2004) 3 OECD Working Papers on International Investment 3 ff. 5 For a typology of FET clauses, see M Waibel, ‘Fair and Equitable Treatment as Boilerplate’ (2019) 30 American Review of International Arbitration 85, 89–95.

360 Stephan W Schill formulation of FET clauses.6 Instead, FET is widely considered, as stated in Rosalyn Higgins’ Separate Opinion in the Oil Platforms case, as ‘a legal term of art well known in the field of overseas investment protection’.7 It is a standard for assessing the legality of host government conduct under international law, not a standard that empowers tribunals to render decisions ex aequo et bono.8 Arbitral tribunals have interpreted and applied FET in numerous cases;9 the literature on FET is abundant.10 Yet, despite its great practical importance, the FET standard, for a very long time, has created conceptual problems and controversy about how to resolve the tensions between investor and state interests because the normative content of what FET requires lacks clarity. As Jeswald Salacuse aptly put it ten years ago, the FET standard is ‘maddeningly vague, frustratingly general, and treacherously elastic’.11 Indeed, when FET claims were first arbitrated in the late 1990s and early 2000s, in particular under the North American Free Trade Agreement (NAFTA),12 there was no more than a glimpse of an idea of what FET required. In this situation of normative vagueness, or indeterminacy, it was arbitral tribunals that took it upon themselves to concretise and further develop the normative content of FET. They did so case-by-case and rather independently of the formulation of the FET clause, by building on arbitral precedent in ways that resemble a common law system of (persuasive) precedent.13 In concretising the normative content of FET in that way, arbitral tribunals in effect acted as important law-makers because they, rather than the states parties to IIAs, shaped and concretised the present-day meaning of FET clauses.14 FET in that sense empowered arbitral tribunals and delegated considerable law-making powers to them.

6 See, eg, Staur Eiendom AS, EBO Invest AS & Rox Holding AS v Republic of Latvia, Award (28 February 2020) ICSID Case No ARB/16/38, para 416; Murphy Exploration & Production Company International v Republic of Ecuador, Partial Final Award (6 May 2016) UNCITRAL, PCA Case No 2012-16, paras 205–206. Many arbitral tribunals also conclude that the distinction between autonomous FET clauses and those tied to custom is ‘more theoretical than real’. See, eg, Rumeli v Kazakhstan, above (n 2) para 611. See further references below n 104. 7 Oil Platforms (Islamic Republic of Iran v United States of America) (Preliminary Objections, Judgment, 12 December 1996) Separate Opinion of Judge Higgins, ICJ Rep 1996, 847, 858, para 39. 8 See Yannaca-Small, above (n 4) 40; C Schreuer, ‘Fair and Equitable Treatment in Arbitral Practice’ (2005) 6(3) Journal of World Investment & Trade 357, 365. 9 Statistically, breach of FET is claimed in more than 50% of all investment treaty arbitrations. See UNCTAD, Investment Dispute Settlement Navigator, https://investmentpolicy.unctad.org/investment-dispute-settlement. 10 See the following monographs that touch exclusively, or to a significant extent, on FET: I Tudor, The Fair and Equitable Treatment Standard in the International Law of Foreign Investment (Oxford, Oxford University Press, 2008); S Montt, State Liability in Investment Treaty Arbitration – Global Constitutional and Administrative Law in the BIT Generation (Oxford, Hart, 2009); R Kläger, ‘Fair and Equitable Treatment’ in International Investment Law (Cambridge, Cambridge University Press, 2011); A Diehl, The Core Standard of International Investment Protection (Alphen aan den Rijn, Wolters Kluwer, 2012); Paparinskis, above (n 3); P Dumberry, The Fair and Equitable Treatment Standard: A Guide to NAFTA Case Law on Article 1105 (Alphen aan den Rijn, Kluwer Law International, 2013); Weiler, above (n 3); R Islam, The Fair and Equitable Treatment (FET) Standard in International Investment Arbitration (New York, Springer, 2018); FM Palombino, Fair and Equitable Treatment and the Fabric of General Principles (New York, Springer, 2018). 11 J Salacuse, The Law of Investment Treaties (Oxford, Oxford University Press, 2010) 221. 12 See R Howse and G Ünüvar, ‘Sowing the Seeds of an ISDS Legitimacy Crisis? The Notorious First Wave of NAFTA Chapter 11 Awards’, ch 7 in this volume. See further Dumberry, above (n 10). 13 See SW Schill, The Multilateralization of International Investment Law (Cambridge, Cambridge University Press, 2009) 321–357. 14 In this sense see P Juillard, ‘L’évolution des sources du droit des investissement’ (1994) 250 Recueil des cours 9, 131 (stressing the ‘œuvre prétorienne des tribunaux arbitraux’ because of the ‘contours imprécis’ of several core standards of treatment, including FET).

Landmark Cases on Fair and Equitable Treatment

361

Today, the doctrinal contours of how arbitral tribunals interpret FET are rather straightforward. Tribunals apply FET to the exercise of public authority by any branch of government – executive, legislator, and courts – and have developed recurring elements that are widely accepted as forming part of FET.15 These include the requirement for the host state to provide a certain degree of stability, predictability and consistency, not to engage in arbitrary and discriminatory conduct, not to deny justice, to provide due process and transparency, to protect legitimate expectations of foreign investors, and not to engage in unreasonable and disproportionate conduct.16 Conceptually, these elements can be grasped by explaining FET as an embodiment of the normative demands flowing from the concept of the rule of law17 as it is laid down in many domestic legal systems and in a variety of international regimes, including human rights treaties or systems of regional economic integration, such as the European Union (EU). The jurisprudence on FET has been aptly summarised, for example, by the Tribunal in Murphy v Ecuador: It is clear from the repeated reference to ‘fair and equitable’ treatment in investment treaties and arbitral awards that the FET treaty standard is now generally accepted as reflecting recognisable components, such as: transparency, consistency, stability, predictability, conduct in good faith and the fulfilment of an investor’s legitimate expectations. The precise application of these components, and the stringency of the standard applicable, may vary from case to case depending on the terms of the clause and the specific circumstances of the case. Notwithstanding, the function of the FET clause in investment treaties is broadly the same: it ensures the stability and predictability of the legal and business framework in the State party subject to any qualifications otherwise established by the treaty and under international law.18

The consensus on the normative content of FET notwithstanding, the Tribunals’ power to shape the normative contours of FET has been subject to criticism. Controversy arose not only in respect of the individual doctrinal elements of FET that tribunals developed, including above all the protection of legitimate expectations,19 but also regarding the very 15 See also Institut de Droit International, ‘Legal Aspects of Recourse to Arbitration by an Investor Against the Authorities of the Host State under Inter-State Treaties’ Resolution (13 September 2013) art 13 (concerning the content of the FET standard). 16 For details, see M Jacob and SW Schill, ‘Fair and Equitable Treatment: Content, Practice, Method’ in M Bungenberg and others (eds), International Investment Law: A Handbook (Oxford, Beck/Nomos/Hart, 2015) 700; R Dolzer and C Schreuer, Principles of International Investment Law, 2nd edn (Oxford, Oxford University Press, 2012) 145–160; C McLachlan and others, International Investment Arbitration: Substantive Principles, 2nd edn (Oxford, Oxford University Press, 2017) ch 7. 17 SW Schill, ‘Fair and Equitable Treatment, the Rule of Law, and Comparative’ in SW Schill (ed), International Investment Law and Comparative Public Law (Oxford, Oxford University Press, 2010) 151 (this chapter is based on work first published as SW Schill, ‘Fair and Equitable Treatment under Investment Treaties as an Embodiment of the Rule of Law’ (2006) 6 Institute for International Law and Justice Working Paper (Global Administrative Law Series); similarly KJ Vendevelde, ‘A Unified Theory of Fair and Equitable Treatment’ (2010) 43 New York University Journal International Law and Politics 43, 49–53. 18 Murphy v Ecuador, above (n 6). For a similar synthesis of the sub-elements of FET, see Rumeli v Kazakhstan, above (n 2) paras 609–610; Joseph C Lemire v Ukraine, Decision on Jurisdiction and Liability (14 January 2010) ICSID Case No ARB/06/18, paras 284–285; Electrabel SA v Republic of Hungary, Decision on Jurisdiction, Applicable Law and Liability (30 November 2012) ICSID Case No ARB/07/19, para 7.74; The PV Investors v Spain, Final Award (28 February 2020) UNCITRAL, PCA Case No 2012-14, paras 561–565. 19 See, eg, M Sornarajah, ‘Mutations of Neo-Liberalism in International Investment Law’ (2011) 3 Trade, Law and Development 203, 257–268; TJ Zeyl, ‘Charting the Wrong Course: The Doctrine of Legitimate Expectations in Investment Treaty Law’ (2011) 49 Alberta Law Review 203; M Potestà, ‘Legitimate Expectations in Investment Treaty Law: Understanding the Roots and the Limits of a Controversial Concept’ (2013) 28 ICSID Review – Foreign Investment Law Journal 88; J Ostřanský, ‘An Exercise in Equivocation: A Critique of Legitimate Expectations as a General Principle of Law under the Fair and Equitable Treatment Standard’ in A Gattini and others (eds), General Principles of Law and International Investment Arbitration (Leiden, Brill, 2018) 344.

362 Stephan W Schill legitimacy of how arbitral tribunals have developed and continue to develop FET’s normative content.20 Core concerns relate to two issues in particular: first, whether arbitral jurisprudence is biased in favour of investors and leaves enough room for States to pursue public interests; and second, how to embed, and potentially limit, the way tribunals interpret FET clauses. A prominent reaction of states consists in tying FET expressly to custom.21 But arbitral tribunals are also increasingly looking at other sources of law, both domestic and international, as a means to guide, embed, and legitimise their interpretations of FET. The purpose of this chapter is not to address the doctrinal details of FET and discuss landmark cases that developed the various sub-elements of FET.22 Instead, it focuses on understanding the interaction between arbitral tribunals and contracting states in determining the normative content of FET. Its central thesis is that FET’s normative vagueness empowers arbitral tribunals as central law-makers in investment law and is responsible for attempts to embed FET into a broader normative framework that ensures policy space for states to take measures in the public interest and implements limits on the law-making activities of arbitral tribunals. Against this background, section II zooms in on three landmark cases from the ‘early’ era of investor-state arbitration that translated FET’s textual indeterminacy into an autonomous check on government conduct. Section III discusses three landmark cases that address how the interpretive authority of arbitral tribunals can be embedded and limited by connecting FET to other sources of law, in particular customary international law and comparative public law. Section IV concludes.

II. EMPOWERING ARBITRATORS AS LAW-MAKERS: TECMED – WASTE MANAGEMENT II – SALUKA

When first confronted with claims by investors in the 1990s that certain state behaviour was contrary to FET, arbitral tribunals were faced with a tabula rasa. Both meaning and normative content of FET clauses were untested in international dispute settlement. There was neither any visible precedent on the interpretation of FET clauses, nor did other sources of law, whether domestic or international, provide clear guidance. The rules on treaty interpretation were equally unhelpful in narrowing down the meaning of FET.23 In this situation, it was arbitral tribunals that were at the forefront of developing the normative content of FET. Three landmark cases from the early period of investment treaty arbitration – Tecmed v Mexico, Waste Management II v Mexico, and Saluka v Czech Republic – illustrate how arbitral tribunals turned from institutions for the settlement of individual investment disputes into important lawmakers that had to mitigate competing claims by disputing parties for, on the one hand, investment protection and, on the other hand, the need to regulate foreign investment.

20 See only M Sornarajah, Resistance and Change in the International Law on Foreign Investment (Cambridge, Cambridge University Press, 2015); M Koskenniemi, ‘It’s not the Cases, It’s the System’ (2017) 18(2) Journal of World Investment & Trade 343. 21 See below sections II.A and B. 22 For further details on landmark cases addressing the various sub-elements of FET, especially the concept of legitimate expectations in the context of the Spanish renewable energy cases, see also D Azaria, ‘The Renewable Energy Arbitrations Under the Energy Charter Treaty’, ch 10 in this volume. 23 The terms ‘fair and equitable’ could at best be replaced with similarly vague phrases, such as ‘just,’ ‘even-handed,’ ‘un-biased’ or ‘legitimate;’ teleological considerations proved similarly unhelpful in concretising the meaning of FET. See, eg, MTD Equity Sdn Bhd and MTD Chile SA v Republic of Chile, Award (25 May 2004) ICSID Case No ARB/01/7, para 113; Saluka Investments BV v The Czech Republic, Partial Award (17 March 2006) UNCITRAL, para 297.

Landmark Cases on Fair and Equitable Treatment

363

A. Tecmed v Mexico One of the cases that shows how little constrained arbitral tribunals were during investment treaty arbitration’s early years, and how they made use of their authority in shaping the normative content of FET, is the Award in Tecnicas Medioambientales Tecmed SA v Mexico.24 Rendered in Spanish at the end of May 2003 under the Arbitration (Additional Facility) Rules of the International Centre for Settlement of Investment Disputes (ICSID), it became accessible to a wider audience through an unofficial English translation on ICSID’s website later that same year.25 Its 81 pages of reasoning are fact-heavy and terse on expounding the law; they reflect not only the dearth of material (including precedent and scholarship) on BITs at the time, but also the typical approach of commercial arbitrators to reason their decisions for the disputing parties rather than the wider public, which public international lawyers regularly cater to.26 The case involved the claim that the refusal by Mexican authorities to renew the operating licence for a hazardous waste landfill that was run by a Spanish-owned, but locally incorporated company breached the bilateral investment treaty (BIT) between Mexico and Spain, including its requirement in Article 4(1) that ‘[e]ach Contracting Party will guarantee in its territory fair and equitable treatment, according to International Law, for the investments made by investors of the other Contracting Party’.27 While the Mexican agency in charge had argued that the operator was unreliable, inter alia because it had processed toxic waste without permission and had exceeded the landfill’s capacity, the Tribunal found that the agency’s decision was motivated by purely political reasons, namely to respond to massive protests by the local population against the landfill. Although the investor had agreed already to relocate the landfill, the requested renewal of the operating licence for five more months was refused. The Tribunal found the Mexican authority’s conduct to constitute both a compensable indirect expropriation and a breach of the FET clause, and ordered Mexico to pay approximately US$5.5 million in damages. Although important in respect of the concept of indirect expropriation,28 the Tribunal’s approach to interpreting the FET clause in the Mexico-Spain BIT is of principal interest here. The Tribunal’s approach in setting out the clause’s normative content could hardly be sparser; it involves reasoning by affirmation and praetorian proclamation rather than in-depth engagement with the applicable sources of international law and the details of treaty interpretation. As a starting point, the Tribunal spends no more than a few lines on noting the interpretation of the FET clause in Article 1105(1) of NAFTA as support for the proposition that Article 4 of the Mexico-Spain BIT is an expression of the international law principle of

24 Técnicas Medioambientales Tecmed, SA v The United Mexican States, Award (29 May 2003) ICSID Case No ARB (AF)/00/2. 25 See G Alvarez Avila, ‘Técnicas Medioambientales Tecmed, SA v United Mexican States (Case No ARB(AF)/00/2) – Introductory Note’ (2004) 19 ICSID Review 154, 157. 26 All three arbitrators – Horacio A Grigera Noan (President), José Carlos Fernandez Rozas and Carlos Bernal Verea – come from the world of commercial arbitration and private international law. On the different epistemic communities of commercial arbitrators and public international lawyers and their approaches to investment treaty arbitration and investment law, see SW Schill, ‘W(h)ither Fragmentation? On the Literature and Sociology of International Investment Law’ (2011) 22 European Journal of International Law 875, 887–890; A Roberts, ‘Clash of Paradigms: Actors and Analogies Shaping the Investment Treaty System’ (2013) 107 American Journal of International Law 45. 27 For the Tribunal’s analysis of the FET claim, see Tecmed v Mexico, above (n 24) paras 152–174. 28 See DA Desierto, ‘Expropriation Cases’, ch 21 in this volume.

364 Stephan W Schill good faith (‘bona fides’).29 After replacing one vague concept (‘fair and equitable treatment’) with another (‘good faith’), the Tribunal declares, in a long, but central paragraph, its understanding of FET: [154]. The Arbitral Tribunal considers that [the FET clause in] the Agreement, in light of the good faith principle established by international law, requires the Contracting Parties to provide to international investments treatment that does not affect the basic expectations that were taken into account by the foreign investor to make the investment. The foreign investor expects the host State to act in a consistent manner, free from ambiguity and totally transparently in its relations with the foreign investor, so that it may know beforehand any and all rules and regulations that will govern its investments, as well as the goals of the relevant policies and administrative practices or directives, to be able to plan its investment and comply with such regulations. Any and all State actions conforming to such criteria should relate not only to the guidelines, directives or requirements issued, or the resolutions approved thereunder, but also to the goals underlying such regulations. The foreign investor also expects the host State to act consistently, i.e. without arbitrarily revoking any preexisting decisions or permits issued by the State that were relied upon by the investor to assume its commitments as well as to plan and launch its commercial and business activities. The investor also expects the State to use the legal instruments that govern the actions of the investor or the investment in conformity with the function usually assigned to such instruments, and not to deprive the investor of its investment without the required compensation. In fact, failure by the host State to comply with such pattern of conduct with respect to the foreign investor or its investments affects the investor’s ability to measure the treatment and protection awarded by the host State and to determine whether the actions of the host State conform to the fair and equitable treatment principle. Therefore, compliance by the host State with such pattern of conduct is closely related to the above-mentioned principle, to the actual chances of enforcing such principle, and to excluding the possibility that state action be characterized as arbitrary; i.e. as presenting insufficiencies that would be recognized ‘… by any reasonable and impartial man,’ or, although not in violation of specific regulations, as being contrary to the law because ‘… (it) shocks, or at least surprises, a sense of juridical propriety.’30

The key proposition in Tecmed is the link the Tribunal proclaims between FET and the protection of a foreign investor’s expectations. Legitimate expectations are not just an element of FET but the core of its normative content. For the Tribunal, consistent and transparent decision-making, the absence of arbitrary conduct, compliance with domestic law etc become elements of FET not because this can be derived through treaty interpretation,31 or some other process of law ascertainment from the applicable BIT, but because it is what foreign investors expect. While linking this understanding of FET to a handful of precedents of other international courts and tribunals,32 it should be clear to any unbiased observer that the Tribunal’s

29 Tecmed v Mexico, above (n 24) para 153 (referring to SD Myers, Inc v Government of Canada, Partial Award (13 November 2000) NAFTA/UNCITRAL, para 134 and Mondev International Ltd v United States of America, Award (11 October 2002) NAFTA, ICSID Case No ARB(AF)/99/2, para 116). 30 Tecmed v Mexico, above (n 24) para 154 (quoting first from Neer v México (1926) IV RIAA 60 and then from Elettronica Sicula SpA (ELSI) (United States of America v Italy) (Judgment) (1989) ICJ Rep 65, 128). 31 The Tribunal, however, tries to mask this deliberate choice by claiming that its understanding of FET flowed necessarily from an ‘autonomous interpretation’ of the Mexico-Spain BIT pursuant to the rules on treaty interpretation in the Vienna Convention on the Law of Treaties (VCLT) (adopted 23 May 1969, 1155 UNTS 331). Considering in particular the object and purpose of the BIT to ‘strengthen and increase the security and trust of foreign investors’, the Tribunal argues that under any other interpretation, the FET clause in question ‘would be deprived of any semantic content or practical utility of its own.’ See Tecmed v Mexico, above (n 24) paras 155–156. 32 See above nn 29, 30. In addition, the Tribunal referenced I Brownlie, Principles of International Law, 5th edn (Oxford, Oxford University Press, 1998) 19. See Tecmed v Mexico, above (n 24) para 153.

Landmark Cases on Fair and Equitable Treatment

365

determination of FET was nothing but a deliberative and contingent choice in the face of normative indeterminacy.33 In fact, the Tribunal’s arguments (connection of FET to good faith and to the objective and purpose of the BIT) could have been invoked just as easily to justify a different, more sovereignty-friendly interpretation of FET. Rather than arriving at its position by accepted means of law ascertainment, the Tribunal, in effect, proclaimed the normative content of FET in an act of delegated law-making similar to how domestic courts fill vague notions, such as ‘good faith’ under private law, with concrete content. Indeed, as the International Court of Justice (ICJ) recently confirmed in Obligation to Negotiate Access to the Pacific Ocean (Bolivia v Chile), there was no grounding in general international law for the protection of legitimate expectations.34 Applied to the case at hand, the Tribunal reasoned that the investor’s legitimate expectations were based on an agreement reached with the domestic agency to relocate the landfill.35 Due to this agreement, the investor could reasonably expect to have his operating licence renewed until the relocation of the landfill was accomplished. In addition, the landfill operator also had a right to a fair hearing and should have been given the opportunity to remedy any breach of the operating conditions. Finally, the Tribunal pointed out that the domestic agency was not entitled to use the non-renewal of the operating licence as a coercive measure to force the investor to relocate the landfill more rapidly. The Tribunal thus found that the unexpected non-renewal of the licence violated the investor’s right to be heard, his right to a transparent administrative proceeding, and his expectation to see his operating licence renewed. Tecmed’s interpretation of FET has become widely invoked by investors in subsequent arbitration proceedings and has been cited with approval by arbitral tribunals under a variety of different BITs.36 In addition, tribunals have relied on certain elements first developed in Tecmed, in particular the idea of FET encompassing the protection of legitimate expectations.37 At the same time, the Tecmed award has been criticised for one-sidedly taking foreign investors’ expectations into account in shaping the content of FET,38 and suggesting

33 Some even criticise this as an ‘invention’ of normative content by the Tribunal; see C Campbell, ‘House of Cards: The Relevance of Legitimate Expectations under Fair and Equitable Treatment Provisions in Investment Treaty Law’ (2013) 30 Journal of International Arbitration 361, 379. Similarly, Suez, Sociedad General de Aguas de Barcelona SA and Vivendi Universal SA v Argentine Republic, Award (9 April 2015) ICSID Case No ARB/03/19, and AWG Group v Argentine Republic, Separate Opinion Pedro Nikken (30 July 2010) UNCITRAL, para 25. 34 Obligation to Negotiate Access to the Pacific Ocean (Bolivia v Chile) (Judgment, 1 October 2018) ICJ Rep 2018, 507, 559, para 162. See CJ Tams and E Methymaki, ‘The World Court’s Influence on Contemporary Investment Law’, ch 4 in this volume. 35 For the application of the proclaimed standard to the facts at hand, see Tecmed v Mexico, above (n 24) paras 157–174. 36 A good overview over Tecmed’s impact on subsequent arbitral jurisprudence on FET is given by L Reed and S Consedine, ‘Fair and Equitable Treatment: Legitimate Expectations and Transparency’ in M Kinnear and others (eds), Building International Investment Law – The First 50 Years of ICSID (Alphen aan den Rijn, Wolters Kluwer, 2016) 283, 289–293. For some of the early endorsements of Tecmed’s approach, see, eg, MTD Equity Sdn Bhd & MTD Chile SA v Republic of Chile, Award (25 May 2004) ICSID Case No ARB/01/7, paras 113 ff; Occidental Exploration and Production Company v The Republic of Ecuador, Final Award (1 June 2004) LCIA Case No UN3467, para 185; Eureko BV v Republic of Poland, Partial Award (19 August 2005) UNCITRAL, paras 231–235; Siemens AG v Argentine Republic, Award (6 February 2007) ICSID Case No ARB/02/8, paras 298–299. 37 See, eg, Suez and AWG v Argentina above n 33 Decision on Liability (30 July 2010) paras 222–226; El Paso Energy International Company v Argentine Republic, Award (31 October 2011) ICSID Case No ARB/03/15, para 348; Electrabel v Hungary, above (n 18) para 7.75; Franck Charles Arif v Republic of Moldova, Award (8 April 2013) ICSID Case No ARB/11/23, para 538; Venezuela Holdings BV and others v Bolivarian Republic of Venezuela, Award (9 October 2014) ICSID Case No ARB/07/27, paras 256 and 264; Gold Reserve Inc v Bolivarian Republic of Venezuela, Award (22 September 2014) ICSID Case No ARB(AF)/09/1, paras 570–576; Foresight Luxembourg Solar 1 Sárl et al v Kingdom of Spain, Final Award (14 November 2018) SCC Case No 2015/150, para 352. 38 MTD Equity Sdn Bhd & MTD Chile SA v Republic of Chile, Decision on Annulment (21 March 2007) ICSID Case No ARB/01/7, para 67.

366 Stephan W Schill an interpretation that ‘looks like a programme of good governance that no State in the world is capable of guaranteeing at all times’.39 However, despite this criticism, many of the sub-elements Tecmed proclaimed – including the need for consistent and transparent decision-making, the absence of arbitrariness, the duty to apply domestic law, and the protection of an investor’s detrimental reliance on specific host state representations – can be justified, provided that state interests are also taken into account, as requirements that the rule of law demands of administrative action. What is more, criticism of the Tecmed award in arbitral jurisprudence has mostly had the effect of ‘add[ing] depth, breadth, and predictability to the conceptual framework’40 Tecmed first set out, rather than suggesting burying it altogether. In this way, Tecmed has become the basis of a rather consistent jurisprudence that has forged the normative content and the doctrinal contours of FET.

B. Waste Management II v Mexico While Tecmed shows how in the early years of investment treaty arbitration arbitral tribunals were able to determine the normative content of FET rather autonomously, the 2004 award by a NAFTA tribunal established under the ICSID Additional Facility Rules in Waste Management, Inc v United Mexican States (Waste Management II)41 illustrates the jurisprudential mechanism that enables arbitral tribunals, collectively, to further develop FET once its normative content had been laid down in the very first cases. Waste Management II is a particularly clear and influential example of how tribunals make use of arbitral precedents as the main source for determining the content of FET, thus asserting their collective interpretive authority over that of states. The dispute in Waste Management II – a resubmitted case after the claimant had remedied procedural defects that had led to a defeat on jurisdiction on its original case42 – arose out of difficulties in implementing a concession granted to Acaverde, a wholly owned Mexican subsidiary of the US claimant, for the provision of waste disposal services in the City of Acapulco.43 Under the concession, the City was to ensure Acaverde’s exclusivity by forbidding the collection of waste by unauthorised third parties and requiring residents to conclude service agreements with Acaverde. The City, however, was unable to ensure Acaverde’s exclusivity as residents resisted paying for Acaverde’s services and other waste collectors continued operating. In addition, the City failed to make a landfill site available to Acaverde and did not meet approximately 80 per cent of its payment obligations under the concession agreement. At the same time, there were complaints that Acaverde failed to meet its obligations to keep the concession area free of waste. After the difficulties in implementing the concession agreement resulted in financial difficulties of Acaverde, ultimately leading it to withdraw and sell its business, Waste Management, Acaverde’s owner, claimed that its investment had been subject to arbitrary

39 El Paso v Argentina, above (n 37) para 342 (emphasis in the original). Similarly, Z Douglas, ‘Nothing If Not Critical for Investment Treaty Arbitration: Occidental, Eureko and Methanex’ (2006) 22 Arbitration International 27, 28. 40 Reed and Consedine, above (n 36) 294. 41 Waste Management, Inc v United Mexican States, Award (30 April 2004) ICSID Case No ARB(AF)/00/3 (Waste Management II). 42 See Waste Management, Inc v United Mexican States, Arbitral Award (2 June 2000) ICSID Case No ARB(AF)/98/2. For in depth discussion, see AK Bjorklund, ‘Waiver of Local Remedies and Limitation Periods’ in Kinnear and others, above (n 36) 237–244. 43 For the facts of the case, see Waste Management II v Mexico, above (n 41) paras 40–72.

Landmark Cases on Fair and Equitable Treatment

367

conduct by the Mexican authorities, that it had not received due process, and that it had fallen victim to a denial of justice when trying to bring claims for breach of the concession agreement in Mexican courts.44 All of this, the claimant argued, violated the FET clause in Article 1105(1) of NAFTA, which provides that ‘[e]ach party shall accord to investments of investors of another Party treatment in accordance with international law, including fair and equitable treatment and full protection and security.’ Unlike the Tribunal in Tecmed, the Tribunal in Waste Management addressed the parties’ arguments about how to determine the normative content of Article 1105(1) of NAFTA at great length and in considerable detail.45 It recalled not only the text of the provision, and its interpretation by NAFTA’s Free Trade Commission, which clarified that FET under Article 1105(1) of NAFTA was equivalent to the customary international law minimum standard of treatment,46 but addressed in detail how earlier NAFTA tribunals had interpreted the FET standard. On the basis of this assessment of precedent, the Tribunal extrapolated the following definition: Taken together, the S.D. Myers, Mondev, ADF and Loewen cases suggest that the minimum standard of treatment of fair and equitable treatment is infringed by conduct attributable to the State and harmful to the claimant if the conduct is arbitrary, grossly unfair, unjust or idiosyncratic, is discriminatory and exposes the claimant to sectional or racial prejudice, or involves a lack of due process leading to an outcome which offends judicial propriety – as might be the case with a manifest failure of natural justice in judicial proceedings or a complete lack of transparency and candour in an administrative process.47

As this quote shows, the Tribunal’s principal approach was not to apply the methods of treaty interpretation or determine state practice and opinio juris in order to give concrete meaning to the FET clause in question; instead it principally relied on arbitral precedent. In applying the standard thus developed to the facts of the case, the Tribunal ultimately found no breach of FET because it attributed many of the difficulties faced by Acaverde to the ‘weaknesses of the original business plan’.48 The Tribunal also clarified that what was primarily at stake in the case was a breach of contractual obligations of the concession agreement. Such breaches did not, however, translate in and of themselves into a breach of the international minimum standard under customary international law.49 This required something more, such as wholly arbitrary or grossly unfair conduct.50 The Tribunal found the persistent non-payment of debt not to be wholly arbitrary, as long as it did ‘not amount to an outright and unjustified repudiation of the transaction and provided that some remedy is open to the creditor to address the problem’.51 Therefore in comparison to Tecmed, the Tribunal in Waste Management II announced a more limited normative content of FET and emphasised that attention had to be paid to what disputes should be resolved by domestic courts rather than by an investment treaty tribunal. After all, in the Tribunal’s view, claims for breach of FET should not function as a simple debt collection mechanism for public contracts.52

44 ibid

para 87. ibid paras 89–99. 46 See NAFTA Free Trade Commission, Notes of Interpretation of Certain Chapter 11 Provisions (31 July 2001), https://perma.cc/VH7Z-ERXG. 47 Waste Management II v Mexico, above (n 41) para 98. 48 ibid para 113. 49 ibid para 114. 50 ibid para 115. 51 ibid. 52 ibid para 116. 45 See

368 Stephan W Schill Despite the greater deference Waste Management II accorded to host state conduct, the Award shows that arbitral tribunals, already in the early period of investment treaty arbitration, turned prior arbitral decisions from their status as ‘subsidiary means for the determination of rules of law’ in the sense of Article 38(1)(d) of the ICJ Statute into principal reference points for understanding the content of IIAs.53 This has the important effect that it is no longer so much the rules and principles in IIAs themselves that determine the outcome of investment treaty disputes, but the way in which investment tribunals have applied, shaped, and further developed these rules and principles in past cases. What is more, even though arbitral tribunals have not become tired of emphasising that arbitral precedent is not binding, the authority of arbitral precedent in the determination of treaty rules like FET is reinforced by the view many tribunals express that a continuous line of precedents, a jurisprudence constante, should be followed, unless there are strong reasons to diverge.54 Tribunal decisions, while de jure non-binding beyond the individual case, de facto determine how investment treaties are interpreted and investment disputes decided, including in particular in respect of FET. Waste Management II attests to the existence of this mechanism.

C. Saluka v Czech Republic Although tribunals like the one in Waste Management II had already indicated that the standard of review under FET should be more limited than the sweeping pronouncements made in Tecmed, an important concern relating to early arbitral jurisprudence was the risk that host state interests would not be sufficiently taken into account in shaping the normative content of FET. However, it did not take long after Tecmed for tribunals to start emphasising the importance of host state interests in shaping the content of FET. Of particular importance in this respect is the 2006 Partial Award in Saluka Investments BV v Czech Republic, an UNCITRAL arbitration based on the Netherlands-Czech Republic BIT.55 The case concerned a claim by Saluka Investments, a Dutch company owned and controlled by the Japanese Nomura Group, that the Czech authorities’ conduct in resolving the country’s banking crisis in early 2000 affected its shareholding in IPB, one of four major Czech banks, in breach of its right to FET.56 More specifically, Saluka claimed that, after the stability of the country’s banking system was threatened in the late 1990s due to problems with nonperforming loans, the Czech Government gave state assistance to the three other major Czech banks, in which the state had a major shareholding interest, but not to IPB. IPB was instead

53 Even contemporary quantitative citation studies supported this. See JP Commission, ‘Precedent in Investment Treaty Arbitration: A Citation Analysis of a Developing Jurisprudence’ (2007) 24 Journal of International Arbitration 129, 148 (concluding that ‘citations to supposedly subsidiary sources, such as judicial decisions, including arbitral awards, predominate’); OK Fauchald, ‘The Legal Reasoning of ICSID Tribunals – An Empirical Analysis’ (2008) 19 European Journal of International Law 301, 333–343. For a more recent study, see, eg, W Alschner, ‘Ensuring Correctness or Promoting Consistency? Tracking Policy Priorities in Investment Arbitration Through Large-Scale Citation Analysis’ in D Behn and others (eds), Empirical Perspectives on Investment Arbitration (Cambridge, Cambridge University Press, 2021) 230. 54 See, eg, Saipem SpA v People’s Republic of Bangladesh, Decision on Jurisdiction and Provisional Measures (21 March 2007) ICSID Case No ARB/05/07, para 67; Daimler Financial Services AG v Argentine Republic, Award (22 August 2012) ICSID Case No ARB/05/1, para 52; International Thunderbird Gaming Corp v United Mexican States, Arbitral Award (26 January 2006) UNCITRAL (NAFTA), Separate Opinion of Thomas Wälde, para 16 (‘A deviation from well and firmly established jurisprudence requires an extensively reasoned justification’). 55 Saluka Investments v Czech Republic, above (n 23). 56 For the facts of the case, see ibid paras 26–163.

Landmark Cases on Fair and Equitable Treatment

369

put into forced administration in mid-2000, its banking licence revoked, and an administrator installed who, within a matter of days, transferred IPB’s business to one of the other Czech banks. After rejecting that Saluka was subject to an unlawful expropriation,57 the Tribunal found a breach of the FET provision in the Netherlands-Czech Republic BIT, because the Czech Ministry of Finance and the Czech Central Bank had ‘unreasonably frustrated IPB’s and its shareholders’ good faith efforts to resolve the bank’s crisis,’ had ‘failed to deal with IPB’s as well as Saluka’s/Nomura’s proposals in an unbiased, even-handed, transparent and consistent way,’ and ‘refused to communicate with IPB and Saluka/Nomura in an adequate manner’.58 Furthermore, while finding that giving financial assistance to banks was not contrary to FET, nor that an investor was protected against its competitors receiving such assistance,59 the Tribunal found that provisions on state aid ‘must not be discriminatory or unreasonably harmful for the foreign investor,’60 ultimately concluding that the Czech Republic had violated FET. Most importantly, the Tribunal’s approach diverged considerably from that of Tecmed. Saluka agreed with earlier jurisprudence that the concept of legitimate expectations formed part of FET, and was even ‘the dominant element of that standard’.61 Yet the Saluka tribunal specifically warned of the danger of taking the idea of the investor’s expectation ‘too literally’ since this would ‘impose upon host States’ [sic] obligations which would be inappropriate and unrealistic’.62 It also emphasised that one could not make the content of FET dependent on the investor’s ‘subjective motivations and considerations’.63 Rather, an investor’s ‘expectations, in order for them to be protected, must rise to the level of legitimacy and reasonableness in light of the circumstances’.64 For this to be the case, the Tribunal reasoned, FET required a balance between the investor’s legitimate expectations and the host state’s interests: 305. No investor may reasonably expect that the circumstances prevailing at the time the investment is made remain totally unchanged. In order to determine whether frustration of the foreign investor’s expectations was justified and reasonable, the host State’s legitimate right subsequently to regulate domestic matters in the public interest must be taken into consideration as well … 306. The determination of a breach of Article 3.1 by the Czech Republic therefore requires a weighing of the Claimant’s legitimate and reasonable expectations on the one hand and the Respondent’s legitimate regulatory interests on the other. 307. A foreign investor protected by the Treaty may in any case properly expect that the Czech Republic implements its policies bona fide by conduct that is, as far as it affects the investors’ investment, reasonably justifiable by public policies and that such conduct does not manifestly violate the requirements of consistency, transparency, even-handedness and non-discrimination. In particular, any differential treatment of a foreign investor must not be based on unreasonable distinctions and demands, and must be justified by showing that it bears a reasonable relationship to rational policies not motivated by a preference for other investments over the foreign-owned investment.65

Different from Tecmed, the Tribunal in Saluka emphasised the importance of the host state’s interests and, in terms of the underlying legal analysis involved, took a fundamentally 57 ibid

paras 245–278. para 407. 59 ibid para 445. 60 ibid para 446. 61 ibid para 301. 62 ibid para 304. 63 ibid. 64 ibid (emphasis in the original). 65 ibid paras 305–307. 58 ibid

370 Stephan W Schill different approach. Rather than looking one-sidedly at the position of the foreign investor, the relationship between investor and state interests was key. Only if an investor’s interest was disproportionately interfered with, could FET be violated. This approach has since been taken up repeatedly by disputing parties in investment treaty arbitrations66 and was endorsed by a number of investment tribunals as appropriately reflecting the content of FET.67 In sum, while confirming the importance of arbitral precedent for concretising FET, Saluka is a reminder to assess precedent critically and strive for a balance of investor and host state interests, including in the interpretation of FET clauses. Still, arbitral jurisprudence develops path dependencies that are typical for any precedent-oriented system. Tribunals frame their reasoning – even if they do not, or do not fully, agree with earlier awards – as part of the discursive framework established by earlier decisions. This shows that arbitral tribunals are the principal actors that shape the normative content of FET and the balance between competing interests within the investment treaty regime. Precedent-orientation brings advantages and concerns: it guides future interpretations, and thereby contributes to consistency, but it also constrains reinterpretation and change. How such change can be effectuated notwithstanding the jurisprudential path dependencies, entails reliance on precedent, as set out above, is addressed in the next section.

III. CONTROLLING ARBITRATORS AS LAW-MAKERS: NEER – GLAMIS GOLD – INTERNATIONAL THUNDERBIRD GAMING

Although arbitral tribunals, as Waste Management II and Saluka indicate, started to nuance their interpretations of FET early on in order to find balanced results, states have remained wary of the tribunals’ interpretive authority. As a result, various attempts to link FET to other sources of law have sprung up in order to control the tribunals’ law-making activities. One approach is to tie FET to customary international law.68 This has revived (partly quite dated) cases that preceded the modern investment treaty regime by many decades, most importantly the 1926 Neer case,69 and has given rise to lively controversy, as illustrated in Glamis Gold v United States, about the exact content of custom, the means of determining it, and its relationship to autonomous FET clauses. Another approach, which is pursued by several arbitral tribunals following Thomas Wälde’s Separate Opinion in International Thunderbird

66 See, eg, Blusun SA and others v Italian Republic, Award (27 December 2016) ICSID Case No ARB/14/3, paras 163, 184; Casinos Austria International GmbH and Casinos Austria Aktiengesellschaft v Argentine Republic, Decision on Jurisdiction (28 June 2018) ICSID Case No ARB/14/32, para 242. 67 See, eg, BG Group Plc v The Republic of Argentina, UNCITRAL, Final Award (24 December 2007) para 298; Rumeli above (n 2) para 679; Joseph Charles Lemire v Ukraine, above (n 18) paras 262, 264, 356, 371, 418 (referring to the ‘Saluka test’); El Paso, above (n 37) paras 358, 365; Franck Charles Arif, above (n 37) paras 532, 537; Ioan Micula and others v Romania, Award (11 December 2013) ICSID Case No ARB/05/20, para 516; Philip Morris Brands Sàrl and others v Oriental Republic of Uruguay, Award (8 July 2016) ICSID Case No ARB/10/7, paras 320–324, 399; Flemingo DutyFree Shop Private Limited v Republic of Poland (12 August 2016) UNCITRAL, para 551; Foresight Luxembourg Solar 1 v Spain, above (n 37) para 352. 68 States have done so through both authoritative interpretations, as was the case under NAFTA, or by using language in newer IIAs that clarifies that FET is equivalent to the international minimum standard. On the former, see NAFTA Free Trade Commission, above (n 46); on the latter, see, eg, Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) (signed 8 March 2018, entered into force 30 December 2018) Art 9.6. 69 For a broader overview over the use of historical cases by investment treaty tribunals, see SW Schill, CJ Tams and R Hofmann, ‘International Investment Law and History: An Introduction’ in SW Schill, CJ Tams and R Hofmann (eds), International Investment Law and History (Cheltenham, Edward Elgar, 2018) 3, 7–11.

Landmark Cases on Fair and Equitable Treatment

371

Gaming v Mexico, is to draw connections between FET and comparative public law as a means to concretise FET.

A. The Neer case The Decision by the United States-Mexico General Claims Commission in L.F.H. Neer and Pauline Neer (U.S.A.) v United Mexican States has lain dormant for many decades after it was handed down in 1926.70 Attempts at tying the interpretation of FET to the customary international law minimum standard of treatment, which is pursued by a number of states, including the US,71 have revived this case as a landmark for the interpretation of FET clauses. The case concerned a claim brought by the US against Mexico on behalf of the heirs of Paul Neer who was employed as superintendent of a mine in Mexico. In November 1924, Mr Neer was killed by a group of armed men when riding home on horseback. The culprits, however, were never apprehended. The US claimed before the Commission that the Mexican authorities had failed to investigate diligently, thus breaching (customary) international law.72 The Commission dismissed the claim, but made important statements on the impact and scope of international law if aliens sustained damage abroad. It reasoned: The Commission recognizes the difficulty of devising a general formula for determining the boundary between an international delinquency of this type and an unsatisfactory use of power included in national sovereignty. … Without attempting to announce a precise formula, it is in the opinion of the Commission possible to … hold (first) that the propriety of governmental acts should be put to the test of international standards, and (second) that the treatment of an alien, in order to constitute an international delinquency, should amount to an outrage, to bad faith, to wilful neglect of duty, or to an insufficiency of governmental action so far short of international standards that every reasonable and impartial man would readily recognize its insufficiency. Whether the insufficiency proceeds from deficient execution of an intelligent law or from the fact that the laws of the country do not empower the authorities to measure up to international standards is immaterial.73

Two propositions are at the heart of the Commission’s decisions: first, that compliance with domestic law alone was not sufficient, but that an independent standard existed under (customary) international law that protected aliens against certain government conduct; and second, that the dividing line between a breach of international law (an ‘international delinquency’) and ‘an unsatisfactory use of power included in national sovereignty’ was treatment that ‘amount[ed] to an outrage, to bad faith, to wilful neglect of duty, or to an insufficiency of governmental action so far short of international standards that every reasonable and impartial man would readily recognize its insufficiency’.74

70 LFH Neer and Pauline Neer (USA) v United Mexican States, Decision (15 October 1926) IV RIAA 60–66. For a comprehensive assessment of the Mexican Claims Commissions, see AH Feller, The Mexican Claims Commissions, 1923–1934 (New York, Macmillan, 1935). 71 See Treaty Between the Government of the United States of America and the Government of [Country] Concerning the Encouragement and Reciprocal Protection of Investment (2012 US Model BIT) Art 5, https://perma. cc/GVX3-59JM; see also the references, below n 126. 72 See Neer v Mexico, above n 70 paras 60–61. 73 ibid paras 61–62. 74 The Commission added that ‘[i]t is not for an international tribunal such as this Commission to decide, whether another course of procedure taken by the local authorities at Guanacevi might have been more effective’ (ibid 62).

372 Stephan W Schill While not addressing the concept of FET as such, nor even dealing with a case of foreign property or investment protection, the holding in Neer became widely considered in investment treaty arbitration75 as ‘the landmark case for the international minimum standard’.76 Most notably, although Neer concerned the obligation to protect foreigners against acts committed by private parties – something one would assess under full protection and security, rather than FET, under modern IIAs – investment treaty arbitration turned the standard of review the Commission pronounced (outrage, bad faith, etc) into the generally applicable standard to determine breach of the customary international minimum standard, including in cases where active interference by the host state was at stake.77 Neer thus became, through subsequent use in investment treaty arbitrations, a landmark case on the generally applicable standard of review under the customary international law minimum standard, even though it never stood for that proposition itself. In fact, the United States-Mexico Claims Commissions had formulated the standard of review in cases of active interference of host state authorities with rights of aliens differently from Neer. In Harry Roberts v United Mexican States,78 a case rendered shortly after Neer, which concerned the allegedly arbitrary arrest of an American citizen and therefore an issue that would in an investment treaty context today fall under FET,79 the Commission formulated a more demanding standard: the ‘test is, broadly speaking, whether aliens are treated in accordance with ordinary standards of civilization’.80 Accordingly, as per the Commission itself, active interference by host state authorities, in order to be lawful under international law, had to conform to ordinary standards of civilisation, not to the lower standard of outrage, bad faith, or wilful neglect of duty. Neer’s own limitations notwithstanding, in many investment treaty arbitrations the decision became the starting point for discussions about the relationship between the international law minimum standard of treatment and treaty-based FET. While some tribunals endorsed the standard of review declared in Neer, others resisted being bound by a standard that may appear anachronistic and insufficiently adapted to the needs of today’s global economy. Glamis Gold v United States discussed in the next section illustrates the varying reception of Neer.

75 Neer is particularly prominently invoked by parties and tribunals in NAFTA Chapter 11 arbitrations. See Dumberry, above (n 10) 16–19, 106–124. See also Bilcon of Delaware Inc et al v Canada, Award on Jurisdiction and Liability (17 March 2015) UNCITRAL, PCA Case No 2009-04, paras 427–445; Mesa Power Group LLC v Canada, Award (24 March 2016) UNCITRAL, PCA Case No 2012-17, paras 496–507. For examples beyond NAFTA, see, eg, Azurix Corp v Argentine Republic, Award (14 July 2006) ICSID Case No ARB/01/12, para 365; Siemens AG v Argentine Republic, Award (6 February 2007) ICSID Case No ARB/02/8, para 293; Adel A Hamadi Al Tamimi v Sultanate of Oman, Award (3 November 2015) ICSID Case No ARB/11/33, para 383. See also Alex Genin, Eastern Credit Limited, Inc and AS Baltoil v Republic of Estonia, Award (25 June 2001) ICSID Case No ARB/99/2, para 367 (paraphrasing Neer through a reference to Brownlie, above (n 31) 527–531. 76 Yannaca-Small, above (n 4) 9, fn 37. See further UNCTAD, Fair and Equitable Treatment (New York, United Nations, 1999) 39–40; JC Thomas, ‘Reflections on Art. 1105 of NAFTA: History, State Practice and the Influence of Commentators’ (2002) 17 ICSID Review – Foreign Investment Law Journal 21; Paparinskis, above (n 3) 53. 77 For criticism, see J Paulsson and G Petrochilos, ‘Neer-ly Misled?’ (2007) 22 ICSID Review – Foreign Investment Law Journal 242; SM Schwebel, ‘Is Neer Far from Fair and Equitable?’ (2011) 27 Arbitration International 555. Similar criticism is made in Mondev, above (n 29) para 115. 78 Harry Roberts (USA) v United Mexican States, Decision (2 November 1926) IV RIAA 77. 79 For references to Roberts in investment treaty cases, see, eg, Ol European Group BV v Venezuela, Award (10 March 2015) ICSID Case No ARB/11/25, paras 486–489; Joseph C Lemire v Ukraine, above (n 18) paras 248–249. 80 Roberts v Mexico, above (n 78) 80.

Landmark Cases on Fair and Equitable Treatment

373

B. Glamis Gold v United States At the heart of the Award in Glamis Gold v United States81 is the issue regarding the extent to which FET can be embedded in, and the law-making activity of arbitral tribunals limited by, customary international law. The Tribunal was faced, inter alia, with a claim by a Canadian mining company for breach of the FET clause in Article 1105(1) NAFTA arising out of delays in approving a mining project in California and the regulatory changes that made the project unviable.82 The so-called Imperial Project involved mining for gold and silver in three open pits with only partial backfilling, leaving a pit approximately 1,400 meters long, 800 meters wide, and 240 meters deep. It was located in an area that was home to significant cultural and religious sites of the Native American Quechan Tribe – most notably, the ‘Trail of Dreams,’ which is part of a network of ceremonial paths and sacred sites likened in their importance for the tribe to Mecca or Jerusalem. Although the Imperial Project seemed to fulfil the statutory requirements when approval was sought in December 1994,83 it was rejected in January 2001, after a long and winding administrative process, due to opposition by the Quechan. The process, inter alia, involved the issuance by the government of a legal opinion that reinterpreted the applicable regulatory framework in order to allow denying mining projects that caused irreparable harm to cultural values completely (the M-Opinion). Yet, towards the end of 2001, the Bush administration repealed the M-Opinion and any decision based on it. Notwithstanding this, a reconsideration of the Project had still not taken place in 2009 when the Tribunal handed down its Award. In addition, in April 2003 the State of California had enacted legislation that introduced a mandatory backfilling requirement for mining activities affecting Native American sacred sites in order to stop the Imperial Project. The Tribunal ultimately denied Glamis Gold’s claim under Article 1105(1) NAFTA.84 At the outset, the Tribunal agreed with the disputing parties (and with all NAFTA parties) that the provision had to be interpreted, as per the NAFTA Free Trade Commission’s Note of Interpretation, as referring to the customary international law minimum standard.85 Determining the content of the customary international law minimum standard, however, was controversial: the claimant argued that the Tribunal should follow arbitral precedent on FET; the respondent insisted instead that a general and consistent practice of states plus opinio juris had to be shown, as arbitral tribunals could not create customary international law.86 The Tribunal, essentially siding with the respondent, found that its task under Article 1105(1) NAFTA was not one of treaty interpretation, but of determining the content of customary international law. As compared to earlier jurisprudence, this constituted a considerable shift in the methodology of determining the content of FET. Furthermore, since the parties disagreed as to whether the customary international law 81 Glamis

Gold, Ltd v United States, Award (8 June 2009) UNCITRAL. the facts of the case, see ibid paras 27–185. the applicable Californian legislation, the land in question was, in principle, open for mining, but restrictions were possible ‘to protect the scenic, scientific, and environmental values … against undue impairment’ (ibid para 48). This legislation, and its implementing regulation, was originally understood to require nondiscretionary approval of mining operations by the competent federal agency, the Bureau of Land Management (BLM), subject only to measures for ‘reasonable reclamation’, such as reshaping disturbed land and revegetating. The implementing regulations did not require mandatory backfilling and did not allow denying mining operations completely in order to give precedence to competing land uses. 84 See ibid paras 537–830. 85 ibid para 599. 86 ibid paras 542–542. 82 For

83 Under

374 Stephan W Schill minimum standard had evolved since the 1926 Neer case, the Tribunal determined that it fell on the claimant to establish such an evolution through the showing of state practice and opinio juris.87 Arbitral precedent, by contrast, could not create or prove custom, but only illustrate its content, provided that it actually involved an examination of custom rather than an autonomous interpretation of FET.88 For this reason, the Tribunal expressly discarded the relevance of awards interpreting treaty-based FET clauses, including the award in Tecmed.89 Analysing arbitral precedent the Tribunal considered relevant, such as prior NAFTA awards,90 the Tribunal concluded that the claimant was unable to prove an evolution of the international minimum standard.91 This, the Tribunal noted, ‘effectively freezes the protections provided for in [Article 1105(1) of NAFTA] at the 1926 conception of egregiousness’.92 Although it conceded that ‘it is entirely possible … that, as an international community, we may be shocked by State actions now that did not offend us previously’,93 the Tribunal circumscribed the content of what FET required when tied to customary law as follows: a violation of the customary international law minimum standard of treatment, as codified in Article 1105 of the NAFTA, requires an act that is sufficiently egregious and shocking – a gross denial of justice, manifest arbitrariness, blatant unfairness, a complete lack of due process, evident discrimination, or a manifest lack of reasons – so as to fall below accepted international standards and constitute a breach of Article 1105.94

In applying this standard, the Tribunal determined that none of the measures violated customary international law. The change of the settled approval standard through the M-Opinion did ‘not exhibit blatant unfairness or evident discrimination to this particular investor’,95 nor did it upset legitimate expectations, which would have required ‘a quasi-contractual relationship … whereby the State has purposely and specifically induced the investment’.96 The delay in the administrative approval was also not manifestly arbitrary since complicated issues were at stake.97 As regards California’s measures, the Tribunal found no evidence that the claimant’s investment was specifically targeted; instead, the measures were of general application, both in form and effect.98 The Tribunal also saw no reason to take issue with the process of adopting new legislation, as the claimant, absent specific assurances, could not have had legitimate expectations that the regulatory framework remained unchanged.99 In substance, California’s legislation was also not arbitrary; the measures were rationally related to a legitimate government purpose.100

87 ibid

paras 600–603. paras 605–611. para 610. 90 These included in particular Mondev, above (n 29) paras 116 and 127; SD Myers, Inc v Canada, Partial Award (13 November 2000) UNCITRAL, para 263; International Thunderbird Gaming v Mexico, above (n 54) para 194. See Glamis Gold v US, above (n 81) para 614. 91 ibid paras 612–627. 92 ibid para 604. 93 ibid para 616. 94 ibid para 627. 95 ibid para 765. 96 ibid para 766. 97 ibid paras 773–777. 98 ibid paras 791–797, 819–821. 99 ibid paras 800–802, 809–815. 100 ibid paras 803–806, 816–818. 88 ibid 89 ibid

Landmark Cases on Fair and Equitable Treatment

375

The application of the FET clause in Article 1105(1) of NAFTA in Glamis Gold is significant for a number of reasons. First of all, it shows the Tribunal’s sensitivity in reviewing a state’s decision to restrict economic rights in order to protect non-investment concerns – here the protection of Native American sacred sites against harm caused by open-pit mining operations – finding that, without a contractual or quasi-contractual bond between investor and state, FET does not offer protection against measures that, in pursuing a legitimate government interest, merely make use, enjoyment, and exploitation of property more costly, but not impossible. Second, the award illustrates a trend to apply FET primarily in a procedural fashion, with the Tribunal examining the procedural propriety of the administrative proceedings and of Californian legislation without, however, reviewing the balance struck between competing rights and interests other than for unreasonable excesses. Both counts ensure that FET respects the host state’s regulatory autonomy, while protecting foreign investors against unreasonable and disproportionate measures. It is questionable, however, to what extent the Tribunal’s approach to FET is a necessary consequence of de facto endorsing the standard of review announced in Neer. Instead, there are good arguments to be made that respect for a host state’s regulatory autonomy, absent any specific representations to the contrary, and limiting, in principle, a tribunal’s standard of review to procedural propriety is appropriate not only for FET clauses tied to the customary international law minimum standard, but also for autonomous FET clauses. After all, FET clauses can hardly be read as abrogating a host state’s right to regulate, nor do they put investment treaty tribunals into a position where they can replace a host state’s value judgment about the relationship between economic and non-economic interests with their own. In that sense, despite Glamis Gold’s endorsement of the Neer standard, differences between autonomous FET clauses and those tied to custom may be negligible in substance – except for certain subelements of FET, such as transparency. What makes a significant difference, however, is the methodology adopted in Glamis Gold for determining the content of the customary international law minimum standard. However, the approach of burdening the parties with showing state practice and opinio juris is deeply problematic. In particular, it is difficult to justify that the claimant, a private investor, should bear that burden. Instead, the Tribunal should have applied the principle iura novit curia and determined the content of the international minimum standard itself.101 In doing so, the Tribunal would have also had to grapple with the fact that many other tribunals that apply FET clauses that are linked to custom do not feel tied to the standard adopted by Neer. On the contrary, given that the exact contours of what the international law minimum standard demands of host states is itself vague and hardly concrete,102 many other tribunals refer – faute de mieux – to decisions of arbitral tribunals, without distinguishing in detail whether findings were based on the customary variant of FET or an autonomous standard, considering that there are indeed few, if any, practical differences between the two types of FET clauses.103 Many tribunals also consider that the international law minimum standard itself has evolved since 1926 101 See F Rosenfeld, ‘Iura Novit Curia in International Law’ (2017) 6 European International Arbitration Review 131. See also HE Veenstra-Kjos, The Interplay between National and International Law in Investor-State Arbitration (PhD thesis, University of Amsterdam, 2010) ch 5; FG Sourgens, A Nascent Common Law: The Process of Decision making in International Legal Disputes between States and Foreign Investors (Leiden, Brill/Nijhoff, 2015) ch 4. 102 See UNCTAD, above (n 76) 28–29; see also Railroad Development Corporation (RDC) v Republic of Guatemala, Award (29 June 2012) ICSID Case No ARB/07/23, paras 216–217; SAUR International SA v Argentine Republic, Decision on Jurisdiction and Liability (6 June 2012) ICSID Case No ARB/04/4, para 493. 103 See, eg, Sempra Energy International v Argentine Republic, Award (28 September 2007) ICSID Case No ARB/02/16, para 302; Azurix v Argentina, above (n 75) paras 364–372; Biwater Gauff (Tanzania) Ltd v Tanzania, Award (24 July 2008) ICSID Case No ARB/05/22, paras 586–603.

376 Stephan W Schill and is now more demanding.104 All of this might ultimately suggest that there is no categorical difference between the content of the customary international law minimum standard and an autonomously worded FET treaty standard.105 At the same time, the tribunals’ reactions to attempts to tie FET to customary international law could also be seen as instances of resisting states’ introduction of tighter control mechanisms for the tribunals’ interpretive authority and law-making power.

C. International Thunderbird Gaming v Mexico Stressing the connections between FET and the customary international law minimum standard of treatment is but one way to guide the interpretation of FET clauses and embed it in a normative framework that limits the activity of arbitral tribunals as lawmakers. Interestingly, arbitral tribunals themselves also draw connections between FET and other sources of law, both international and domestic, as a mechanism to concretise the meaning of FET. The conceptual foundations of this approach were laid out most prominently in Thomas Wälde’s Separate Opinion in International Thunderbird Gaming v Mexico.106 The case concerned a dispute under Chapter 11 of NAFTA that arose out of the closedown of the claimant’s gaming facilities in Mexico because the gaming machines used were found to be contrary to Mexican law; cash prizes that players could win were not only dependent on the players’ skills – which would have been lawful gaming under Mexican law – but involved elements of chance – which Mexican law prohibited.107 Yet, the claimant relied on a letter issued by the competent Mexican authority that it claimed contained an individual representation that the gaming machines used were lawful. This, claimant argued, created a legitimate expectation that its operations were lawful and could not be closed down. The Tribunal accepted the Free Trade Commission’s interpretation that Article 1105(1) of NAFTA was equivalent to the customary international law minimum standard.108 It also confirmed that the protection of legitimate expectations was a relevant aspect for the protection offered to foreign investors under Article 1105(1) of NAFTA.109 Ultimately, however, the Tribunal’s majority found that no breach of Article 1105(1) of NAFTA had occurred 104 See, eg, Pope & Talbot Inc v Government of Canada, Award in Respect of Damages (31 May 2002) UNCITRAL, para 59; ADF Group Inc v United States of America, Award (9 January 2003) ICSID Case No ARB(AF)/00/1, para 179; Mondev, above (n 29) paras 116, 125; International Thunderbird Gaming v Mexico, above (n 54) para 194; Cargill Incorporated v United Mexican States, Award (18 September 2009) ICSID Case No ARB(AF)/05/2, paras 281–282; Merrill & Ring Forestry LP v Government of Canada, Award (31 March 2010) UNCITRAL, para 213; Chemtura Corporation v Canada, Award (2 August 2010) UNCITRAL, paras 121–122; SAUR International SA v Argentina, Décision sur la Compétence et sur la Responsabilité (6 June 2012) ICSID Case No ARB/04/4, para 494; Railroad Development Corporation (RDC) v Republic of Guatemala, Award (29 June 2012) ICSID Case No ARB/07/23, para 218; Bilcon v Canada, above (n 75) para 435. Some tribunals also concluded that the inclusion of FET in the vast web of investment treaties made FET itself into a norm of customary international law. See, eg, Pope & Talbot v Talbot, above (n 104) para 62; Mondev v US, above (n 29) para 125; Merril & Ring v Canada, above (n 104) para 210. 105 For further discussion of the relationship between the customary international law minimum standard and FET, see, eg, A Orakhelashvili, ‘The Normative Basis of ‘Fair and Equitable Treatment’ (2008) 46 Archiv des Völkerrechts 74; A Newcombe and L Paradell, Law and Practice of Investment Treaties: Standards of Treatment (Alphen aan den Rijn, Kluwer Law International, 2009) 264–275; H Haeri, ‘A Tale of Two Standards: ‘Fair and Equitable Treatment’ and the Minimum Standard in International Law’ (2011) 27 Journal of International Arbitration 27; Paparinskis, above (n 3). 106 See International Thunderbird Gaming v Mexico, above (n 54). 107 For the facts of the case, see ibid, paras 41–84. 108 ibid paras 192–193. 109 See ibid para 147.

Landmark Cases on Fair and Equitable Treatment

377

because the representation in question did not in fact create an expectation that the claimant’s gaming machines were lawful.110 The Tribunal further explained that FET translated into a number of concrete requirements for administrative proceedings, including the need to grant due process, fairness, not to act arbitrarily, to provide a right to be heard, to base decisions on evidence, to give reasons, and to provide an opportunity for judicial review of administrative decisions.111 The Tribunal found, however, that none of these aspects were violated in the case at hand. The majority’s reasoning is a concise and worthwhile read in its own right on what impact FET can have on the administrative law and practice of host states. The more remarkable part of the Award, however, is the Separate Opinion by Thomas Wälde. It is striking not so much because of his disagreement with the majority concerning the assessment of the facts of the case,112 but because it develops the conceptual foundations of international investment law and investor-state dispute settlement as a genuinely public law discipline and shows what this conceptualisation means for the application of FET as a public law concept. Wälde’s starting point is that the nature of investment treaty disciplines in general, and FET in particular, has to be understood as part of a sui generis normative framework. This framework is different from both traditional public international law and international commercial arbitration, the two bodies of law and practice that come together in investment treaty arbitration. In Wälde’s view, investment law differs from public international law, because it is not part of an inter-state system of dispute resolution; instead ‘[a]t its heart lies the right of a private actor to engage in an arbitral litigation against a (foreign) government over governmental conduct affecting the investor’.113 Similarly, international commercial arbitration provide no suitable analogy, because investment law does not deal with equal parties engaging in arm’s length transactions in a transnational context, but involves a foreign investor’s ‘exposure to political risk, lack of familiarity with and integration into, an alien political, social, cultural, commercial, institutional and legal system, at a disadvantage’.114 Consequently, Wälde reasoned, investment treaty disciplines are better analogised with judicial review of governmental conduct under administrative or constitutional law: [M]ore appropriate for investor-state arbitration are analogies with judicial review relating to governmental conduct – be it international judicial review (as carried out by the WTO dispute panels and Appellate Body, by the European- or Inter-American Human Rights Courts or the European Court of Justice) or national administrative courts judging the disputes of individual citizens’ [sic] over alleged abuse by public bodies of their governmental powers. In all those situations, at issue is the abuse of governmental power towards a private party that did and could legitimately trust in governmental assurances it received; in commercial arbitration on the other hand it is rather a good-faith interpretation of contractual provisions that is at stake. Abuse of governmental powers is not an issue in commercial arbitration, but it is at the core of the good-governance standards embodied in investment protection treaties. The issue is to keep a government from abusing its role as sovereign and regulator after having made commitments.115

From this perspective, international investment law constitutes a public law discipline, because it imposes restraints on a state’s exercise of powers vis-à-vis private investors. Investment 110 ibid

paras 148–167. paras 194–201. 112 Wälde explained that the letter in question, in his view, had created expectations in the claimant. See ibid, Separate Opinion Thomas Wälde, paras 59–95. 113 ibid para 13. 114 ibid para 12. 115 ibid para 13. 111 ibid

378 Stephan W Schill treaty arbitration, in turn, is functionally analogous to administrative or constitutional judicial review. International investment law, in other words, is best understood as forming part of a newly emerging body of public law that escapes rigid classifications into categories of either classical public international law, domestic administrative or constitutional law, or commercial law and arbitration: International investment law is international law as regards its sources, functionally comparable to administrative or constitutional law in limiting government conduct, and linked to international commercial arbitration as regards the form and practice of dispute settlement. After setting out these conceptual foundations, Wälde further reasoned that comparative public (administrative, constitutional and international) law should become part of the standard methodology for interpreting and applying investment protection standards, including FET. In order to concretise FET, and determine whether the protection of legitimate expectations could be seen as forming part of it, Wälde drew on a wide range of sources. This included contract law principles, such as estoppel and venire contra factum proprium,116 but above all public law concepts found in a variety of international and domestic legal systems, including EU law, human rights law,117 international trade law,118 and ‘developed systems of administrative law’.119 In fact, for Wälde, ‘[t]he common principles of the principal administrative law systems are … an important point of references for the interpretation of investment treaties to the extent investment treaty jurisprudence is not yet firmly established’.120 Because the protection of legitimate expectations was widely recognised under these systems of law, Wälde found the same should be the case under FET. The public law approach first laid out in Wälde’s Separate Opinion in International Thunderbird Gaming has had a profound impact on the understanding of the nature of investment treaty disciplines and has fuelled much literature in the field.121 It has also impacted the understanding of FET in arbitral jurisprudence, with an increasing amount of cases drawing on the conceptual approach set out by Thomas Wälde. Several tribunals have since drawn on public law sources outside the investment treaty regime, both domestic and international, in order to concretise FET, including human rights law.122 This development is significant not only for understanding and concretising the doctrinal structure of FET, but also as a means for coming to terms with the vagueness of FET clauses and limiting the law-making activities of arbitral tribunals. Contrary to attempts by states to tie FET to customary international law, and thereby reduce the discretion of arbitral tribunals, the tribunal-driven process of tying FET to comparative public law has not generated any significant opposition in arbitral jurisprudence. This may suggest that tribunal-driven processes of embedding FET in a specific normative framework are more readily palatable to arbitral tribunals than approaches coming from ‘outside’ actors.

116 ibid

para 27.

117 ibid. 118 ibid

para 29. para 28. 120 ibid. 121 See, eg, G Van Harten, Investment Treaty Arbitration and Public Law (Oxford, Oxford University Press, 2007); Montt, above (n 10); Schill, above (n 17). 122 See, eg, Noble Ventures, Inc v Romania, Award (12 October 2005) ICSID Case No ARB/01/11, para 178; Joseph Charles Lemire v Ukraine, above (n 18) para 506; Total SA v Argentine Republic, Decision on Liability (27 December 2010) ICSID Case No ARB/04/1, para 111; Toto Costruzioni Generali SpA v Republic of Lebanon, Award (7 June 2012) ICSID Case No ARB/07/12, para 166; Gold Reserve v Venezuela, above (n 37) para 576; Al Warraq v Indonesia, above (n 2) para 577. 119 ibid

Landmark Cases on Fair and Equitable Treatment

379

Notwithstanding this, comparative public law approaches come with their own risks and pitfalls, and are not necessarily more objective than free-floating interpretations and applications of FET if conducted without the necessary methodological rigor.123 However, if practiced in a sophisticated fashion, comparative public law approaches have the potential to embed the interpretation and application of FET in a normative framework that transcends investment law and is able to guide its application and thereby limit the risk of further developing FET clauses in ways that contravene the interests and intentions of states parties to IIAs.

IV. CONCLUSION

This chapter has argued that FET has become the central norm in the investment treaty regime precisely because of its normative vagueness. This has empowered arbitral tribunals to concretise and further develop the normative content of FET clauses and shape their application to host state conduct. Although tribunals recognised early on that the purpose and content of FET clauses was not to provide one-sided protection to foreign investors, the broad discretion they enjoyed in interpreting and applying FET has resulted in concerns that arbitral decisionmaking is overly restrictive of legitimate state conduct. Consequently, various attempts have been made to tie FET clauses to outside normative frameworks: contracting states have counted principally on tying FET clauses to customary international law, which was received with reservations in arbitral practice; an important strand in arbitral jurisprudence has reacted by concretising FET clauses through a comparative public law analysis of how other domestic and international legal regimes limit government conduct under the concept of the rule of law. The relation to both custom and comparative public law is likely going to be of continuous importance in how the jurisprudence on FET will develop in the future. The landmark cases discussed in this chapter will thus not only be of historic interest, but remain influential. In addition, contracting states have started to react to the problem of vagueness of traditional FET clauses also on the treaty-drafting front. Apart from the radical, and still little practiced, option of excluding FET clauses from IIAs altogether,124 several states, including the contracting parties to the Comprehensive and Progressive Agreement for Trans-Pacific Partnership and the United States-Mexico-Canada Agreement continue tying FET to custom.125 Others, such as Canada and the EU (and its Member States), as illustrated in their 2016 Comprehensive Economic and Trade Agreement, react by drafting FET clauses in a much more detailed manner than before, often codifying standards developed by certain arbitral tribunals, and, in addition, provide for treaty-based mechanisms for the contracting states to influence future interpretations of FET more directly.126 All of this points to the continued importance of debates about the conceptual foundations of FET and of the interaction between arbitral tribunals and contracting states in working jointly in concretising the normative content and doctrinal structures of FET. 123 See SW Schill, ‘General Principles of Law and International Investment Law’ in T Gazzini and E De Brabandere (eds), International Investment Law: The Sources of Rights and Obligations (Leiden, Martinus Nijhoff, 2012) 133, 145–154. 124 See, eg, Thailand–New Zealand Closer Economic Partnership Agreement (signed 19 April 2005, entered into force 1 July 2005) ch 9; India-Singapore Comprehensive Economic Cooperation Agreement (signed 26 May 2005, entered into force 1 August 2005). 125 See CPTPP, above (n 68) Art 9.6; United States-Mexico-Canada Agreement (signed 30 November 2018, entered into force 1 July 2020) Art 14.6. 126 See Comprehensive Economic and Trade Agreement (signed 30 October 2016, in parts provisionally applied since 21 September 2017) Art 8.10.

23 Umbrella Clauses and Contract Claims BENJAMIN SAMSON*

U

MBRELLA CLAUSES HAVE proved to be one of the mysteries of international investment law.1 A typical2 umbrella clause requires that each state party to the investment protection treaty ‘shall observe any other obligation it may have entered into with regard to investments by nationals or companies of the other party’.3 Despite their apparent simplicity, they are the source of numerous complex – and interrelated – issues,4 be it their nature and effect, the law applicable to the claim, the types of obligation covered, the admissibility of shareholder claims, the nature of the claim or the articulation between the dispute resolution clause (DRC) in the treaty and the contract. Each of these questions has been the object of a debate in the case law. It is not easy to find one’s way within the case law on umbrella clauses. Arbitral tribunals often disagree with one another on issues of interpretation and application of these clauses, either entirely or only partially. Decisions and awards may be difficult to assess. Some are supported by just a few reasons only or do not identify the right question at hand. The position of tribunals and their reasoning are sometimes unclear, if not confusing. Nevertheless, certain decisions may serve as beacons to help in grasping the arbitral practice. Some of these major decisions will be discussed in the present chapter. As it cannot aim at exhaustivity, this chapter will focus on the major decisions which address the most complex and cleaving issues,5 ie the effect of umbrella clauses (section II),

* Benjamin Samson holds a doctorate in law from University Paris Nanterre. This chapter reflects developments as of April 2020. 1 See S Lemaire, ‘La mystérieuse ‘Umbrella Clause’ (interrogations sur l’impact de la clause de respect des engagements sur l’arbitrage en matière d’investissements)’ (2009) 3 Revue de l’arbitrage 479–502. 2 One cannot emphasise enough that the text of an umbrella clause varies, sometimes greatly, from one clause to another. As in the case of the ECT, it may also be useful to compare the different linguistic versions of the text of the umbrella clause in a treaty (see RREEF Infrastructure (GP) Limited and RREEF Pan-European Infrastructure Two Lux S.à r.l. v Kingdom of Spain, ICSID Case No ARB/13/30, Decision on Responsibility and on the Principles of Quantum (30 November 2018) para 284. 3 Treaty between the Federal republic of Germany and Pakistan for the promotion and protection of investments (Signed on 25 November 1959, entered into force on 26 March 1963) [Germany-Pakistan BIT], Art 7. See also, eg Agreement between the Government of the People’s republic of China and the Government of Malta on the promotion and protection of investments (Signed on 22 February 2009, entered into force on 1 April 2009) [China-Malta BIT], Art 10(2) or Agreement between the Islamic Republic of Iran and Japan (Signed on 05 February 2016, entered into force on 26 April 2017) [Iran-Japan BIT], Art 6. 4 For a complete overview of these questions, see A Sinclair, ‘Umbrella Clause’ in M Bungenberg and others (eds), International Investment Law (Baden-Baden, CH Beck/Hart/Nomos, 2015) 887–958 and A Reinisch and C Schreuer, International Protection of Investments: The Substantive Standards (Cambridge, Cambridge University Press, 2020) 855–969. Both identify the relevant literature on umbrella clauses. 5 As it concentrates on the relationship between umbrella clauses and contracts, the major decisions on the question whether these clauses apply to non-contractual obligations will not be dealt with.

382 Benjamin Samson the nature of the claim when an umbrella clause is applicable (section III) and the possible claims by shareholders (section IV). For a better understanding, the major decisions on umbrella clauses must be considered in the wider context of the distinction between treaty and contract claims, set out in the 2002 Decision of the first ad hoc Committee in the Vivendi v Argentina (I) case (section I).

I. VIVENDI V ARGENTINA (I) AND THE SUMMA DIVISIO TREATY CLAIMS / CONTRACT CLAIMS

The 3 July 2002 Decision of the ad hoc Committee in the Vivendi (I) case is undoubtedly the landmark decision with respect to the relationship between treaty and contract in contemporary international investment law.6 It was rendered at an early stage of the BIT arbitration boom. The Committee’s findings on this issue were ‘incontestablement destinés à faire jurisprudence’.7 They have been adopted or referred to by virtually every arbitral tribunal constituted in investor-state cases involving contractual issues.8 In this decision, the Committee articulated what has become the summa divisio of international investment law: the distinction between treaty claims and contract claims.9 It discussed both substantial (section A) and jurisdictional (section B) aspects of this distinction.

A. The Substantial Aspect of the Treaty Claims/Contract Claims Distinction In its 2002 Decision, the Vivendi (I) Committee stated that ‘[a] state may breach a treaty without breaching a contract, and vice versa’10 and that, accordingly, ‘whether there has been a breach of the BIT and whether there has been a breach of contract are different questions’.11 In order to distinguish the two types of claim, the Committee borrowed the criterion of the ‘fundamental basis of the claim’ from the decision of the American-Venezuelan Mixed Commission in the Woodruff case.12 Although the Committee did not elaborate on this criterion, it appears rather simple. A claim is a treaty claim if the norm of conduct – the primary rule of obligation in the sense given by Hart13 – or as the Committee put it, the ‘independent standard by which the conduct of the parties is to be judged’,14 is contained in

6 The 2002 decision is part of a 13-year long arbitration process concerning a dispute between a French investor Compagnie Générale des Eaux, then its successor Vivendi, and the Republic of Argentina arising out of a water and sewage services contract signed by Compagnie Générale des Eaux and the Argentinian Province of Tucumán. For a detailed summary of the case, see S Luttrell, ‘Vivendi v Argentina’ in C Miles and E Bjorge (eds), Landmark Cases in Public International Law (Oxford, Hart Publishing, 2017). 7 E Gaillard, La jurisprudence du CIRDI (Vol 2) (Paris, Pedone, 2004) 752. 8 For a recent example, see, eg, Glencore International AG and CI Prodeco SA v Republic of Colombia, Award (27 August 2019) ICSID Case No ARB/16/6 [Glencore v Colombia] para 1032. 9 There is a vast literature on the distinction. Among the recent publications, see I Fadlallah, ‘La distinction Treaty Claims – Contract Claims’ in C Leben (ed), Droit international des investissements et de l’arbitrage international (Paris, Pedone, 2015) 759–772 or S Alexandrov, ‘Breach of Treaty Claims and Breach of Contract Claims. When Can an International Tribunal Exercise Jurisdiction’ in K Yannaca-Small (ed), Arbitration Under International Investment Agreements – A Guide to the Key Issues, 2nd edn (Oxford, Oxford University Press, 2018) 370–394. 10 Vivendi v Argentina (I), Decision on Annulment (3 July 2002) ICSID Case No ARB/97/3, para 95. 11 ibid para 96. 12 See UN, RIAA (vol IX, UN 2009) 223. 13 HLA Hart, The Concept of Law, 3rd edn (Oxford, Oxford University Press, 2012) 81. 14 Vivendi v Argentina (I), above (n 10) para 98.

Umbrella Clauses and Contract Claims

383

a treaty; a claim is contractual if the norm of conduct – for instance the payment of a sum of money – is embodied in a contract.15 The Committee, however, unduly complicated this clear, if not self-evident, distinction. It tied the distinction between treaty and contract claims to the separation between the international and national legal systems. The Committee explained that the distinction stems from the general principle codified in Article 3 of the ILC’s Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA),16 which provides that ‘[t]he characterization of an act of a State as internationally wrongful is governed by international law. Such characterization is not affected by the characterization of the same act as lawful by internal law.’17 This link leads to locating treaties and contracts in two different places, treaties in the international legal order and contracts in national legal orders. This is inaccurate or at least imprecise. Contractual partners may choose international law as the applicable law and international arbitration as the dispute resolution mechanism. A breach of treaty may entail the state’s responsibility under a municipal law triggering a specific regime.18 Treaty claims and contract claims are different not because they are ‘grounded on differing legal orders’19 but merely because they are based on different instruments. It is only once the source of the obligation allegedly breached has been identified that the law applicable to that source and the regime of responsibility is to be determined. The choice of the court or tribunal before which the claim is brought may affect this determination.20

B. The Jurisdictional Aspects of the Treaty Claims/Contract Claims Distinction As regards the jurisdictional aspect of the treaty/contract divide, the Vivendi (I) Committee discussed two questions. The first was the scope of the DRC in the Argentina/France BIT. In the Vivendi (I) case, the DRC in the BIT applied to ‘[a]ny dispute relating to investments’.21 The Committee was not the first jurisdiction to interpret such wording. In the Salini v Morocco case, the Tribunal found that: The terms of Article 8 are very general. The reference to expropriation and nationalisation measures, which are matters coming under the unilateral will of a State, cannot be interpreted to exclude a claim based in contract from the scope of application of this Article.22

15 See Glencore International AG and CI Prodeco SA v Republic of Colombia, Award (27 August 2019) ICSID Case No ARB/16/6, para 1031. 16 Vivendi v Argentina (I), above (n 10) paras 95–96, quoting also the Commentary of this Article. 17 International Law Commission, The Internationally Wrongful Act of a State – General Principles (2001) II International Law Commission Yearbook 2, 36, Art 3. 18 For instance, in French administrative law, the regime of responsibility for the breach of a treaty is distinct from the regime of international responsibility (see Conseil d’Etat, 8 February 2007, Gardedieu v The State, No 279522, and the comments of M Forteau in A Pellet and A Miron (eds), Les grandes décisions de la jurisprudence française de droit international public (Paris, Pedone, 2015) 454–467). 19 SGS Société Générale de Surveillance SA v Islamic Republic of Pakistan, Decision on Jurisdiction (6 August 2003) ICSID Case No ARB/01/13, para 147. 20 See M Forteau, ‘Repenser la logique de traitement des rapports entre ordres juridiques. Changer de regard: tout ne serait-il pas affaire de droit applicable, plutôt que d’ordres juridiques’ in B Bonnet (ed), Traité des rapports entre ordres juridiques (Paris, Librairie générale de droit et de jurisprudence, 2016) 633–649 and the distinction made by the author between the law applicable to and the law applicable by. 21 Agreement between Argentina and France for the promotion and the protection of investments (signed on 3 July 1991, entered into force on 3 March 1993) [Argentina-France BIT] Art 8(1). 22 Salini Costruttori SpA and Italstrade SpA v Kingdom of Morocco, Decision on Jurisdiction (23 July 2001) ICSID Case No ARB/00/4, para 61.

384 Benjamin Samson The Committee adopted the same reasoning. It explained that: Article 8 does not use a narrower formulation, requiring that the investor’s claim allege a breach of the BIT itself. Read literally, the requirements for arbitral jurisdiction in Article 8 do not necessitate that the Claimant allege a breach of the BIT itself: it is sufficient that the dispute relate to an investment made under the BIT.23

The Committee did not – and arguably had not to – discuss the parallel issue of the scope of contractual DRC. These clauses may also be drafted in wide terms and apply to non-contractual claims. As an arbitral Tribunal said: While the basis for arbitral jurisdiction is contractual, the subject matter jurisdiction of arbitral tribunals need not be limited to contractual claims. Other grounds of liability may be included, depending on the scope of the arbitration agreement and limits on arbitrability.24

The second question addressed by the Committee was the articulation of the DRC in the BIT and a DRC in a contract. At first sight, the position of the Committee seems straightforward. It stated that: – In a case where the essential basis of a claim brought before an international tribunal is a breach of contract, the tribunal will give effect to any valid choice of forum clause in the contract; – On the other hand, where ‘the fundamental basis of the claim’ is a treaty laying down an independent standard by which the conduct of the parties is to be judged, the existence of an exclusive jurisdiction clause in a contract between the claimant and the respondent state or one of its subdivisions cannot operate as a bar to the application of the treaty standard. […] In the Committee’s view, it is not open to an ICSID tribunal having jurisdiction under a BIT in respect of a claim based upon a substantive provision of that BIT, to dismiss the claim on the ground that it could or should have been dealt with by a national court.25

The Committee appears to say that, absent express words to the contrary, a contract DRC cannot override a treaty DRC with respect to treaty claims and that, conversely, a treaty DRC cannot override a contract DRC with regard to contract claims. There may be a doubt that the Committee adopted such a clear-cut articulation, without a thorough reasoning.26 Indeed, in footnote 69 of its Decision, the Committee indicated that it ‘does not need to consider whether the effect of Article 8 of the BIT [the DRC] is to override exclusive jurisdiction clauses in contracts underlying investments to which the BIT applies’. Be that as it may, this is how arbitral tribunals have interpreted or used the 2002 Decision, including the Tribunals in the SGS v Philippines and BIVAC v Paraguay cases27 in each of which sat a member of the Vivendi (I) Committee.28 It is against this background that the major decisions on umbrella clauses may be discussed. 23 Vivendi

v Argentina (I), above (n 10) para 55.

24 Niko Resources (Bangladesh) Ltd v Bangladesh Petroleum Exploration & Production Company Limited (‘Bapex’)

and Bangladesh Oil Gas and Mineral Corporation (‘Petrobangla’), Decision on Jurisdiction (19 August 2013) ICSID Cases No ARB/10/11 and No ARB/10/18, para 499; BIVAC v Paraguay, Decision on Jurisdiction (29 May 2009) ICSID Case No ARB/07/9, para 145 and SGS v Paraguay, Decision on Jurisdiction (12 February 2010) ICSID Case No ARB/07/29, para 173. 25 Vivendi v Argentina (I), above (n 10) paras 98 and 101–102. 26 Such articulation between contract and treaty DRCs is far from self-evident. In particular, if one considers that an arbitral tribunal must give effect to an applicable contractual DRC, it is at best unclear why the effect of the clause should be limited to contract claims, when the scope of the DRC includes non-contractual claims. 27 See SGS Société Générale de Surveillance SA v Republic of the Philippines, Decision on Jurisdiction (29 January 2004) Case No ARB/02/6, para 153 and Bureau Veritas, Inspection, Valuation, Assessment and Control BIVAC BV v The Republic of Paraguay, Decision on Jurisdiction (29 May 2009) ICSID Case No ARB/07/9, paras 152–153. 28 Respectively LY Fortier (President) and JR Crawford (Member).

Umbrella Clauses and Contract Claims

385

II. THE EFFECT OF UMBRELLA CLAUSES – SGS V PHILIPPINES, A DEAD-END?

The effect of umbrella clauses is one of the most debated issues in international investment law. In the case law, different arbitral tribunals have offered different – sometimes radically opposed – solutions. The decisions of four arbitral Tribunals have framed the debate over the effect of umbrella clauses. The 2004 Decision on jurisdiction in the SGS v Philippines case is one of them. The four decisions share the same theoretical basic premise: the principle of customary international law (CIL) that ‘a violation of a contract entered into by a State with an investor of another State, is not, by itself, a violation of international law’.29 The following paragraphs summarise their position. In the SGS v Pakistan case, the Tribunal adopted a restrictive position. It rejected the claimant’s argument that the effect of umbrella clauses is to assimilate the breach of contract to a breach of the BIT.30 It identified two possible limited effects of the applicable umbrella clause.31 It first explained in loose terms that the umbrella clause merely ‘signal[ed] an implied affirmative commitment to enact implementing rules and regulations necessary or appropriate to give effect to a contractual or statutory undertaking in favor of investors of another Contracting Party that would otherwise be a dead letter’.32 In even vaguer terms, the Tribunal then stated that: we do not preclude the possibility that under exceptional circumstances, a violation of certain provisions of a State contract with an investor of another State might constitute violation of a treaty provision (like Article 11 of the BIT) enjoining a Contracting Party constantly to guarantee the observance of contracts with investors of another Contracting Party.33

As an example, the Tribunal referred to situations where the host state would frustrate the investor’s right stemming from a compromissory clause.34 Conversely, in SGS v Philippines,35 the Tribunal accepted that an umbrella clause ‘makes it a breach of the BIT for the host State to fail to observe binding commitments, including contractual commitments, which it has assumed with regard to specific investments’.36 Interpreted as such, umbrella clauses could be compared to the clauses de renvoi from international law to domestic law, as identified by Anzilotti.37 The violation of the

29 See SGS v Pakistan, above (n 19) para 167 and SGS v Philippines, above (n 27) para 122. See also Noble Ventures, Inc v Romania, Award (12 October 2005) ICSID Case No ARB/01/11, para 53 (hereinafter Noble Ventures v Romania) and El Paso Energy International Company v Argentine Republic, Decision on Jurisdiction (27 April 2006) ICSID Case No ARB/03/15, para 77. 30 SGS v Pakistan, above (n 19) paras 99 and 165–171. 31 The clause read: ‘Either Contracting Party shall constantly guarantee the observance of the commitments it has entered into with respect to the investments of the investors of the other Contracting Party’ (Agreement between the Islamic republic of Pakistan and the Swiss Confederation on the promotion and reciprocal protection of investments (Signed on 11 July 1995, entered into force on 06 May 1996) (Pakistan-Switzerland BIT) Art 11). 32 SGS v Pakistan above (n 19) para 172. 33 ibid. 34 ibid. 35 The applicable umbrella clause read: ‘Each Contracting Party shall observe any obligation it has assumed with regard to specific investments in its territory by investors of the other Contracting Party’ (Agreement between the republic of Philippines and the Swiss confederation on the promotion and protection of investments (Signed on 31 March 1997, entered into force on 23 April 1999) (Philippines-Switzerland BIT) Art X(2)). 36 SGS v Philippines, above (n 27) para 128. 37 See Y Nouvel, ‘La compétence matérielle: contrat, traité et clauses parapluie’ in C Leben (ed), La procédure arbitrale relative aux investissements internationaux, aspects récents (Paris/Bruxelles, Librairie générale de droit et de jurisprudence/Anthémis, 2010) 23ff and G Cahin, ‘La clause de couverture’ (dite umbrella clause) (2015) 119 Revue Générale de Droit International Public 130ff.

386 Benjamin Samson contract would be a ‘fait interne’,38 ‘une donnée conditionnant la mise en œuvre de la règle internationale’.39 In the Noble Ventures v Romania case,40 the Tribunal has been understood as considering that umbrella clauses do internationalise contracts.41 In its 2005 Award, it explained that ‘[a] n umbrella clause is usually seen as transforming municipal law obligations into obligations directly cognizable in international law’ and ‘proceed[ed] on the basis that, in including Art. II(2)(c) in the BIT, the Parties had as their aim to equate contractual obligations governed by municipal law to international treaty obligations as established in the BIT’.42 In El Paso v Argentina,43 the Tribunal adopted a fourth view. While endorsing the reasoning and arguments of the SGS v Pakistan Tribunal,44 it gave a different effect to umbrella clauses. The Tribunal considered that an umbrella clause extends the protection of the treaty to breaches of ‘additional investment protections contractually agreed by the State as a sovereign – such as a stabilization clause – inserted in an investment agreement’ concluded by the host state,45 such breaches being assimilated to treaty claims.46 As apparent, the 2004 Decision of the SGS v Philippines Tribunal did not instantly become a major decision. At the time it was rendered, the arbitral debate on the effect of umbrella clauses had just started and this tribunal was only the second to discuss this issue in some depth. Over time however, its position was deemed the most convincing by the majority of arbitral tribunals.47 The solution proposed by the SGS v Pakistan was never upheld by another tribunal. Similarly, no tribunal found that umbrella clauses internationalise contracts. The position of the El Paso Tribunal was endorsed by a few tribunals only.48

38 J Cazala, ‘La clause de respect des engagements’ in C Leben (ed), Droit international des investissements et de l’arbitrage international (Paris, Pedone, 2015) 351. 39 Nouvel, above (n 37) 78–79; R Rivier, ‘L’articulation entre droit national et droit international devant les tribunaux arbitraux internationaux d’investissement’ in S Robert-Cuendet (ed), Droit des investissements étrangers: Perspectives croisées (Bruxelles, Bruylant, 2017) 435. 40 The applicable umbrella clause read: ‘Each Party shall observe any obligation it may have entered into with regard to investments’ (Treaty between the Government of Romania and the United States of America concerning the reciprocal encouragement and protection of investment (signed on 28 May 1992, entered into force on 15 January 1994) (Romania-United States BIT) Art II(2)(c)). 41 See, eg, Toto Costruzioni Generali SpA v Republic of Lebanon, Decision on Jurisdiction (11 September 2009), ICSID Case No ARB/07/12, para 199 or J Crawford, ‘Treaty and Contract in Investment Arbitration’ (2008) 24 Arbitration International 351, 368. 42 Noble Ventures v Romania, Award (12 October 2005) ICSID Case No ARB/01/11, paras 53 and 61. The position of the Tribunal is not however as clear as it may seem. It may well be that it adopted the same position as the SGS v Philippines Tribunal since, as the latter, it explained that umbrella clauses assimilate breaches of contract to breaches of treaties (see ibid para 54). 43 The applicable umbrella clause read: ‘Each Party shall observe any obligation it may have entered into with regard to investments’ (Agreement between Argentina and the United-States of America concerning the reciprocal encouragement and protection of investment (Signed on 14 November 1991, entered into force on 20 october 1994) (Argentina-United states BIT) Art II(2)(c)). 44 See El Paso v Argentina, above (n 29) para 85. 45 ibid para 81, see also paras 84–86. 46 ibid para 84. See also the Award of 31 October 2011, paras 531–532. 47 Among the more recent of the 20 or more decisions, see, eg, Ioan Micula, Viorel Micula and others v Romania (I), Final Award (11 December 2013) ICSID Case No ARB/05/20, para 417; Electricité de France (EDF) International S.A. v Republic of Hungary, Swiss Federal Supreme Court Decision No 4A_34/2015 on Set-Aside of Award (6 October 2015) para 3.2.2.; Georg Gavrilovic and Gavrilovic doo v Republic of Croatia, Award (25 July 2018)ICSID Case No ARB/12/39, para 420 or Belenergia SA v Italian Republic, Award (6 August 2019) ICSID Case No ARB/15/40, para 356. 48 See, eg, Sempra Energy International v Argentine Republic, Award (28 September 2007) ICSID Case No ARB/02/16, para 310 or Supervision y Control S.A. v Republic of Costa Rica, Award (18 January 2017) ICSID Case No ARB/12/4, para 282.

Umbrella Clauses and Contract Claims

387

What explains the ‘success’ of the SGS v Philippines 2004 decision on jurisdiction? In comparison to the other three decisions, it may have appeared better reasoned and subsequent tribunals have maybe found the solution adopted more reasonable. The Tribunal started the interpretation of the umbrella clause by analysing its wording in their context and in the light of the treaty’s object and purpose, in accordance with the basic rule of treaty interpretation. It noted the use of ‘the mandatory term “shall”’.49 The Tribunal then explained that the term ‘obligation’ means legal obligations, which existence ‘will be a matter for determination under the applicable law, normally the law of the host State’50 and not international law. Finally, it examined the object of purpose of the BIT as expressed in its preamble.51 The Tribunal then confronted its interpretation to the arguments put forward by the SGS v Pakistan Tribunal. It clarified that: – Umbrella clauses do not apply to all municipal obligations of the host state but only those entered into with respect to investments. There must be a link between the obligation invoked and the investment, which protection is sought by the claimant. Therefore, ‘legal obligation[s] of a general character’ are outside of their scope.52 – An umbrella clause does not internationalize contracts. As the Tribunal stated, ‘[i]t does not convert questions of contract law into questions of treaty law. In particular it does not change the proper law of the [contract] from the law of the Philippines to international law.’53 – It does not follow from its interpretation of the umbrella clause that contractual DRCs are overridden by the DRC in the BIT.54 The former remain binding upon the investor.

By contrast, the reasoning of the three other tribunals displays important defects. In SGS v Pakistan, ‘the Tribunal failed to give any clear meaning to the “umbrella clause”’.55 Despite the clear and mandatory wording of this clause, it would merely ‘signal an implied affirmative commitment’ to provide a legal framework ‘appropriate to give effect to’ the host state’s commitments towards foreign investors and maybe mean that a breach of certain contract provisions ‘might constitute’ – but only ‘under exceptional circumstances’ – a breach of the umbrella clause.56 Neither those contractual provisions nor these exceptional circumstances were identified by the Tribunal. The main flaw of the interpretation of the umbrella clause by the El Paso Tribunal is the lack of textual basis. The Tribunal based its decision on the alleged ‘necessity to distinguish the State as a merchant […] from the State as a sovereign’.57 However, the applicable umbrella clause embodied no such important limitation. For its part, the weakness in the reasoning of the Noble Ventures Tribunal is the lack of explanation of the means by which a transforming or an equating of municipal obligations to international treaty obligations happens and what the consequences are with regard to the applicable law to the claim. Despite the quality of its reasoning, one may wonder whether the solutions adopted by the SGS v Philippines Tribunal did not render umbrella clauses useless. The issue of the applicable law is again key in this respect. The Tribunal explained that the umbrella clause did not substitute the lex contractus for international law.58 It specified that ‘the issue of the 49 SGS

v Philippines, above (n 27) para 115. paras 117 and 126–128. 51 ibid para 116. 52 ibid para 121. 53 ibid para 126. 54 ibid para 123. 55 ibid para 125. 56 SGS v Pakistan, above (n 19) para 172. 57 El Paso v Argentina, above (n 29) para 81. 58 SGS v Philippines, above (n 27) para 126. 50 ibid

388 Benjamin Samson extent or content of ’ the contractual obligations continued to be governed by that lex but remained silent on the law applicable to the performance of such obligations and the regime of responsibility of the host state. Yet, this is a crucial question because, absent specific contract provisions, circumstances precluding wrongfulness and rules on attribution and reparation are set out in the law applicable to the host state’s responsibility. In the absence of an umbrella clause, the lex contractus, usually the law of the host state, undoubtedly governs this question as well. If, when an umbrella clause applies, they remain governed by that law, the umbrella clause would be deprived of substantive effet utile. Indeed, the situation would be no different whether an umbrella clause applies or not, both the existence of the contractual obligations and the responsibility of the host state being governed by the domestic law of this state. It would follow that the only function of an umbrella clause would be to confer jurisdiction to the arbitral tribunal over contractual claims.59 A fifth and more recent line of case law adopts this view. For instance, in Kontinental Conseil Ingénierie v République Gabonaise, the Tribunal stated that ‘[u]ne clause “parapluie” a en principe pour effet d’ouvrir la compétence du tribunal du traité aux réclamations fondées sur des obligations contractuelles nées de contrats conclus entre l’investisseur et l’Etat partie à ce même traité.’60 However, as widely accepted including by the SGS v Philippines Tribunal itself,61 such jurisdiction is already granted by a broad treaty DRC. In the vast majority of BITs containing an umbrella clause, the DRC is broad in scope. Hence in most cases, under the interpretation proposed by the SGS v Philippines Tribunal, umbrella clauses would add nothing, neither as matter of substance nor as one of jurisdiction.

III. UMBRELLA CLAUSES AND THE NATURE OF THE INVESTOR’S CLAIM – BIVAC V PARAGUAY / SGS V PARAGUAY, A CROSSROAD

The particularity of an umbrella clause is that it is a treaty commitment referring to contract commitments. This feature raises the question of the nature of the investor’s claim when it invokes an umbrella clause. Is the fundamental basis of the claim the contract or the treaty containing the umbrella clause? The answer may affect the jurisdiction of a treaty-based tribunal. If the claim is contractual in nature, the jurisdiction of such tribunal may be barred by a contractual DRC. On the contrary, if the basis of the claim is the treaty, a DRC in the relevant contract will not impinge on the jurisdiction of that tribunal. The question of the nature of the claim first arose in SGS v Philippines but the Tribunal avoided providing a straight answer to it. In its 2004 Decision on jurisdiction, it began by examining with satisfaction its ‘[j]urisdiction under the “umbrella clause”’,62 which it distinguished from its ‘[j]urisdiction over contractual claims’.63 Then, the Tribunal discussed the articulation between the DRCs in the treaty and in the contract and decided that the mechanism provided for in the contract should take precedence with respect to contract claims.64 In view of this

59 Interestingly, a member of that tribunal described umbrella clauses as an ‘extra mechanism for the enforcement of claims’ (J Crawford, ‘Treaty and Contract in Investment Arbitration’, above (n 41) 370). 60 See, eg, KCI Kontinental Conseil Ingénierie v Gabonese Republic, Final Award (23 December 2016) PCA Case No 2015–25, para 177. See also Toto v Lebanon, Decision on Jurisdiction (11 September 2009) ICSID Case No ARB/07/12, paras 201–202 and Consutel Group v Algeria, Final Award (3 February 2020) PCA Case No 2017–33, para 375. 61 SGS v Philippines, above (n 27) para 135. 62 ibid paras 113–129. 63 ibid paras 130–135. 64 ibid para 155.

Umbrella Clauses and Contract Claims

389

finding, the Tribunal found it ‘necessary to consider whether SGS has stated a case under the BIT which can be determined independently from the contractual issues’.65 It found that ‘[n]o question of a breach of the BIT independent of a breach of contract claim is raised’66 and that SGS’s claims ‘fairly raise questions of breach of ’67 Articles IV (FET) and X(2) (umbrella clause) of the BIT.68 The Tribunal concluded that it should stay proceedings because the treaty claims were subject to a factual determination to be made by the court designated in the contract.69 In sum, the Tribunal appeared to consider that SGS’s claims were treaty claims as a matter of law but contract claims as a matter of fact. In contrast, the BIVAC v Paraguay and SGS v Paraguay Tribunals tackled the issue of the nature of the claim head on. Their decisions on jurisdiction highlight one of the main crossroads in the arbitral case law. Despite the striking similarities between the two cases, the tribunals reached opposite decisions, which opened two distinct paths in radically opposed directions. Both disputes concerned a contract for pre-shipment inspection and certification services.70 The contracts were drafted in similar terms71 and signed after the applicable BIT entered into force.72 They both embodied a DRC conferring jurisdiction to the tribunals of the City of Asunción.73 Both BITs contained a broad DRC. The Tribunals were presented with similar, if not identical arguments. Both tribunals endorsed the solution of the SGS v Philippines tribunal with respect to their effect.74 However, they disagreed on the nature of the claimant’s claim(s). In BIVAC v Paraguay, the Tribunal found that the claim was contractual and that, therefore, the DRC in the contract should be given effect.75 It concluded that the claim was inadmissible.76 In SGS v Paraguay, the Tribunal considered that the claims were treaty claims77 and, consequently, that the DRC in the contract could not preclude it from exercising its jurisdiction.78 Subsequent arbitral tribunals followed one path or the other.79 Both the BIVAC v Paraguay and SGS v Paraguay Tribunals advanced reasons in support of their decisions. The latter explained that, like the former, it concluded ‘that the umbrella clause before us “establishes an international obligation for the parties to the BIT to observe contractual obligation[s] with respect to investors”’,80 and hence, that [e]ven if the alleged breach of the treaty obligation depends upon a showing that a contract or other qualifying commitment has been breached, the source of the obligation cited by the claimant, and hence the source of the claim, remains the treaty itself.81 65 ibid

para 156. para 159. 67 ibid para 157. 68 ibid paras 162–163. 69 ibid paras 174–176. 70 See BIVAC v Paraguay, above (n 27) para 7 and SGS Soc Société Générale de Surveillance SA v The Republic of Paraguay, Decision on Jurisdiction (12 February 2010) ICSID Case No ARB/07/29, para 26. 71 See ibid para 42. 72 See ibid para 178. 73 See BIVAC v Paraguay, above (n 27) para 144 and SGS v Paraguay, above (n 70) para 34. 74 See ibid para 170. 75 See BIVAC v Paraguay, above (n 27) paras 143–154. 76 See ibid para 159. 77 See SGS v Paraguay, above (n 70) paras 138, 142 and 166–171. 78 ibid paras 138–141 and 185. 79 For tribunals that adopted the same position as the BIVAC v Paraguay Tribunal, see, eg, Toto v Lebanon, above (n 60) paras 201–202; Bosh International, Inc and B&P, LTD Foreign Investments Enterprise v Ukraine, Award (25 October 2012) ICSID Case No ARB/08/11, paras 251–259 or KCI v Gabon, above (n 60) paras 180–183. For tribunals that adopted the same position the SGS v Paraguay Tribunal, see, eg, Nissan Motor Co, Ltd v Republic of India, Decision on Jurisdiction (29 April 2019) PCA Case No 2017-37, paras 276–281 or Belenergia SA v Italian Republic, Award (6 August 2019) ICSID Case No ARB/15/40, para 356. 80 SGS v Paraguay, above (n 70) para 170. 81 ibid para 142. 66 ibid

390 Benjamin Samson If, indeed, the effect of umbrella clauses is to make a breach of a contract a breach of the BIT, or as the BIVAC v Paraguay Tribunal put it, to ‘import the obligations under the Contract into the BIT’,82 the source of the claim seems to be the treaty. Despite these findings, the Tribunal in BIVAC v Paraguay convincingly established the artificiality of this reasoning: There is no freestanding international treaty obligation, under the BIT or anywhere else, which would allow this Tribunal to determine whether the acts of Paraguay gave rise to a violation of Article 3(4) of the BIT [the umbrella clause]. The reality is that the Tribunal would have to interpret and apply the Contract, to determine whether Paraguay was or was not in breach of its obligations thereunder to make certain payments on the nineteen invoices. If the Tribunal was to find that there was a breach of the Contract, on BIVAC’s argument it would have to find also a breach of the BIT. But if the Tribunal were to find no breach of the Contract, there would be no breach of the BIT. Everything turns on the meaning and effect of the Contract. BIVAC has provided no explanation as to where in the BIT, or anywhere else in the rules of international law, the Tribunal might find the ‘independent standard’ that might allow it to determine whether or not Article 3(4) has been breached. […] In the present case, in relation to Article 3(4) we do not see how it could be concluded that ‘the fundamental basis of the claim’ was the BIT rather than the Contract. Any other approach strikes us as being so artificial as to be unreasonable.83

The source of this artificiality may be found in the way the debate was framed. The question whether umbrella clauses turn breaches of contract into breaches of treaty was influenced by the way the rule of CIL is itself set out: a breach of contract is not per se a breach of international law.84 In domestic law, one does not ask whether a breach of contract is a breach of the law. The question is whether the contract has binding force, whether the domestic legal system considers the contract as one of its primary rules of obligation. Similarly, what the rules of CIL actually mean is that ‘[i]nvestor-state contracts do not […] constitute primary rules of international law’.85 Hart’s distinction between primary and secondary rules may be a key to understanding umbrella clauses. As exceptions to CIL, theses clauses may be considered as (international) secondary rules of recognition as defined by Hart,86 ie rules enabling ‘a conclusive identification of the primary rules of obligation’ of a legal system.87 Umbrella clauses may be understood as recognising contractual obligations of the host state as primary rules of international law. It, however, does not mean that issues of domestic contract law are converted into issues of treaty law. The SGS v Philippines Tribunal is correct in this respect. The ‘new’ primary rule of international law is the host state’s contractual obligation ‘lock, stock and barrel’, the content and extent of which are defined by the provisions of contract and the lex contractus.88 The history of the concept of umbrella clause confirms this interpretation.89 The 1962 OECD Draft Convention on the Protection of Foreign Property

82 BIVAC

v Paraguay, above (n 27) para 142. para 149. 84 ibid para 11. 85 I Alvik, ‘Investor-State Contracts’ in M Krajewski and RT Hoffmann (eds), Research Handbook on Foreign Direct Investment (Cheltenham/Northampton, Edward Elgar, 2019) 274. 86 HLA Hart, The Concept of Law, 3rd edn (Oxford, Oxford University Press, 2012). 87 ibid 95. 88 See below para 35. 89 On this history, see AC Sinclair, ‘The Origins of the Umbrella Clause in the International Law of Investment Protection’ (2004) 20 Arbitration International 411–434. 83 ibid

Umbrella Clauses and Contract Claims

391

contained an umbrella clause (Article 2). Its commentary is instructive. It explains that the host state: is free to provide that, after a period, the terms of its undertaking might be altered or that the undertaking might altogether lapse; the undertaking itself might be governed by its own national law. However, any right originating under such an undertaking gives rise to an international right that the Party of the national concerned […] is entitled to protect.90 (Emphasis added)

Such interpretation resolves two problems. First, it eliminates the fiction of the characterisation of the claim as a treaty claim. It is unnecessary to assimilate the breach of contract to a breach of treaty. Once it has produced its ‘recognition effect’, the umbrella clause fades away. The arbitral tribunal is left with an international responsibility claim based on the breach of a contractual obligation. Second, it gives some substantial meaning to umbrella clauses, without which these clauses would be deprived of effet utile in most cases.91 When an umbrella clause enters into force, the trap closes on the host state’s will. It then cannot renege on its contractual commitments, except as provided by international law. By making contractual obligations international obligations, umbrella clauses also trigger the applicability of the law on international state responsibility, including its circumstances precluding wrongfulness and its rules on attribution and reparation. This would leave tribunals with the unsettled issue of the articulation between the DRCs in the treaty and the contract. The BIVAC v Paraguay and SGS v Paraguay Tribunals also disagreed on this issue. Indeed, while the former gave priority to the DRC in the contract,92 the latter considered that, even if the claim was contractual in nature, the investor had a choice between the mechanisms available under the treaty and the contract.93 This question continues to divide arbitral tribunals94 and is likely to do so in future.

IV. UMBRELLA CLAUSES AND SHAREHOLDERS CLAIMS – BURLINGTON V ECUADOR, A LANDMARK

The third key question with respect to umbrella clauses is whether affiliates of the private contractual partner of the host state may invoke the umbrella clause. More precisely, the issue is whether the host state’s contractual obligations to which it refers may be enforced by a person who is not a beneficiary of these obligations, as a party to the contract or otherwise. In early BIT arbitrations, tribunals dealt rather expeditiously with this issue. For instance, in Siemens v Argentina, the Tribunal merely noted that the claimant was not a party to the relevant contract95 and dismissed the claim. In Enron v Argentina, the claimant invoked, through the umbrella clause, obligations contained in a licence granted by Argentina to its local subsidiary.96 The Tribunal entertained and upheld Enron’s claim, for the sole reason that the umbrella clause referred to ‘any obligation it may have entered into with respect to investments’.97 In CMS v Argentina, the Tribunal did not advance any legal justification. 90 OECD

Commentary of Article 2 of the Model tax convention (1962) OECD Publications, No 15637, 14. para 20. 92 ibid para 24. 93 SGS v Paraguay, above (n 70) 176–185. 94 See above the references at (n 79). 95 See Siemens v Argentina, Award (6 February 2007) ICSID Case No ARB/02/8, para 204. See also Azurix Corp v The Argentine Republic (I), Award (14 July 2006) ICSID Case No ARB/01/12, para 384. 96 See Enron v Argentina, Award (22 May 2007) ICSID Case No ARB/01/3, para 270. 97 See ibid paras 274 and 276 – emphasis added. 91 ibid

392 Benjamin Samson In its 2005 Award, it upheld CMS’s claims that Argentina had breached contractual obligations also contained in a licence,98 which were not granted to the claimant but to its Argentinian subsidiary. In 2007, an ad hoc Committee partially annulled this Award, not for manifest excess of power but for failure to state reasons. The Committee found ‘major difficulties’ with the tribunal’s decision, which made the absence of reasoning problematic. In particular, it hinted that: (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.99

In 2012, the Tribunal in the Burlington v Ecuador case delivered its Decision on Liability. This Decision was the first to develop such a detailed reasoning in support of the tribunal’s findings on this issue100 and to assess prior case law.101 The Decision addresses difficult questions, such as the meaning of the words ‘with regard to investments’ in the umbrella clause and the impact of the protection of indirect investments by the BIT. The scope of the Tribunal’s reasoning makes its decision a landmark in the case law, to which subsequent tribunals should compare their views and in the light of which their decisions must be assessed.102 The articulation of the reasoning in this decision was likely fostered by the strong disagreement within the tribunal between President Kaufmann-Kohler and Arbitrator Stern on the one hand, and Arbitrator Orrego Vicuña on the other.103 The majority decided that Burlington was not entitled to ‘enforce against Ecuador its subsidiary’s contract rights’.104 To reach this conclusion, the Tribunal mainly focused on the term ‘obligation’ in the umbrella clause. It established that the beneficiary of the obligation protected by an umbrella clause is to be determined in accordance with the proper law of the contract. It explained that ‘in its ordinary meaning’, the term obligation ‘entails a party bound by it and another one benefiting from it, in other words, entails an obligor and an obligee’105 and that considered as context of the word obligation, the term ‘entered into’ reinforces the ‘idea of privity’.106 The Tribunal then specified that ‘an obligation does not exist in a vacuum’107

98 See CMS Gas Transmission Company v The Argentine Republic, Award (12 May 2005) ICSID Case No ARB/01/8, paras 302–303. 99 CMS v Argentina, above (n 98), Decision on Annulment (25 September 2007) para 95 – italics in the original text. 100 Burlington v Ecuador, Decision of Liability (14 December 2012) ICSID Case No ARB/08/5, paras 212–220. 101 ibid paras 221–234. 102 Compare Supervision v Costa Rica, Award (18 January 2017) ICSID Case No ARB 12/4, paras 279–291 and WNC Factoring Ltd (WNC) v The Czech Republic, Award (22 February 2017) PCA Case No 2014-34, paras 317–341. 103 By comparison, in El Paso v Argentina, the Tribunal, on which Brigitte Stern also sat, reached the same decision but addressed the issue in a more expeditious manner (El Paso v Argentina, above (n 29) paras 533–538). 104 Burlington v Ecuador, above (n 100) paras 220 and 234. 105 ibid para 214. 106 ibid para 216. 107 ibid para 214.

Umbrella Clauses and Contract Claims

393

but only within a legal system. That legal system, in this case, the law of Ecuador, ‘defines the content of the obligation including the scope of and the parties to the undertaking, i.e. the obligor and the obligee’.108 It is thus not enough to assert, as Orrego Vicuña did, that the issue is ‘who are the obligor and obligee in the context of complex business transactions usually characterizing investments’.109 One must identify the law applicable to that question and Orrego Vicuña did not challenge the Tribunal’s findings in this respect. The Tribunal’s conclusion is by no means affected by the protection of indirect investment offered by the BIT. The Tribunal rightly rejected Burlington’s argument, adopted by Orrego Vicuña,110 that ‘because the definition of investment in Article I of the Treaty covers both direct and indirect investment’ Burlington, as indirect investor, became ‘a co-obligee of Ecuador’s obligations under the [contract]’.111 The reason given by the Tribunal is however not entirely convincing: The umbrella clause is only one of the various substantive protections that the Treaty bestows upon investors, with the scope of protection depending on the terms of each specific provision. Other Treaty provisions unquestionably protect both direct and indirect investments, such as for instance the expropriation clause. The object and purpose of the Treaty do not impose that all standards of protection have the same scope.112

The question here is not so much that of the scope of the various protections in the BIT than that of the effect of the protection of indirect investment. As persuasively demonstrated by D Müller, being an indirect investor does not mean that one is subrogated to the rights of the state’s contractual partner or that one becomes a party to all contracts signed by that partner.113 Similarly, being an indirect investor does not confer a right to enforce contractual obligations on behalf of that party: ‘La protection des investissements contrôlés […] indirectement par l’investisseur ne doit pas être confondue avec le mécanisme procédural des réclamations indirectes (derivative claims).’114 This however does not mean that umbrella clauses do not cover contracts concluded with local vehicles of foreign investors. In this respect, the Burlington v Ecuador Tribunal may be wrong in stating that the words ‘with respect to investments’ do not broaden the scope of protection of these clauses.115 Within its ordinary meaning, this expression embraces contracts with local subsidiaries.116 The difficulty is that one tends to conflate two distinct questions: which contracts are protected and who may invoke their violation. If the terms ‘with respect to investments’ extend the reach of umbrella clauses on contracts with local subsidiaries, they do ‘not expand the universe of obligees who may rely on the underlying obligation’, as the Tribunal correctly noted.117 Umbrella clauses do not identify the person or persons entitled to enforce the host state’s contractual obligations. Only the law applicable to the responsibility of the host state does.

108 ibid. 109 Burlington

v Ecuador, above (n 100), Dissenting Opinion of Arbitrator Orrego Vicuña, para 5. paras 7–9. 111 Burlington v Ecuador, above (n 100) para 217. 112 ibid para 218. 113 See D Müller, La protection de l’actionnaire en droit international. L’héritage de la Barcelona Traction (Paris, Pedone, 2015) 412–417. 114 ibid 409. 115 Burlington v Ecuador, above (n 100) para 216. 116 See Energy Charter Secretariat, The Energy Charter Treaty, A Reader’s Guide (2002) 26. 117 Burlington v Ecuador, above (n 100) para 228 – emphasis added. 110 ibid

394 Benjamin Samson The effect one recognises as due to umbrella clauses influences the determination of that law. If one considers that umbrella clauses are merely jurisdictional in nature, that law is the lex contractus. However, if one accepts that umbrella clauses afford a substantial protection, the applicable law is international law because it is the international responsibility of the host state, which the Claimant seeks to implement before the arbitral tribunal. As set out in the ARSIWA, the general principle is that only the injured person may invoke the state’s responsibility for the violation of non-erga omnes obligations.118 The notion of injured person in the ILC’s Articles is defined not by reference to the damage suffered but to the beneficiary of the obligation breached by the responsible state.119 Absent an exception, a claimant may only enforce the right it owns, as beneficiary of a host state’s obligation. In this case, the obligation invoked was enshrined in a contract, of which the claimant, Burlington, was not a beneficiary. Its claim was therefore rejected. It is true that the protection offered by umbrella clauses, invented in the 1950’s, may not fit modern investment schemes and that states may circumvent it by requiring that investments be channeled through a local company.120 This however is not a reason to read procedural rights in these clauses that they do not provide for. It is not for tribunals to adapt the protection offered by in BITs.

V. CONCLUSION

Clauses de renvoi, jurisdictional provisions or secondary rules of recognition? After more than 15 years of case law, the mystery surrounding the nature of umbrella clauses has not yet been resolved. It has however lessened. As the Tribunal noted in BIVAC v Paraguay, ‘with the passage of time the issues have crystallised as there has been greater time for reflection’.121 Major decisions have facilitated the task of subsequent tribunals.122 They have also helped states to clarify the meaning they intend to give to umbrella clauses. For instance, Austria and Kirghizstan have specified in their umbrella clause that it ‘means, inter alia, that the breach of a contract between the investor and the host State will amount to a violation of this treaty’,123 thus adopting the position of the SGS v Philippines Tribunal. Moreover, major decisions have shed light on practical problems umbrella clauses raise, and available adjustment variables enabling states to strengthen or weaken the protection offered by umbrella clauses. States may expand the definition of ‘investor’ in their BIT to include local subsidiaries of foreign investors.124 Any umbrella clause would hence extend their reach on contracts concluded between the host state and these subsidiaries. It would also

118 See

ILC, above (n 17) Art 42, 117. ibid Art 42, 118. See also Reparation for Injuries Suffered in the Service of the United Nations, Advisory Opinion, ICJ Rep 1949, 181–182. 120 Burlington v Ecuador, above (n 100), Dissenting Opinion of Arbitrator Orrego Vicuña, paras 9–10. 121 BIVAC v Paraguay, above (n 27) para 150. 122 See ibid. 123 Agreement between Austria and Kirghizstan concerning the reciprocal encouragement and protection of investment (Signed on 22 April 2016) [Austrie-Kirghizstan BIT] Art 11(1). 124 See, eg, Agreement between Burundi and the Netherlands concerning the reciprocal encouragement and protection of investment (Signed on 24 May 2007, entered into force on 01 August 2009) [Burundi-The Netherlands BIT] Art 1(b)(ii)-(iii). 119 see

Umbrella Clauses and Contract Claims

395

open BIT dispute settlement mechanisms to the latter. States may also – and should – provide for articulation solutions in their investment protection treaties. Some umbrella clauses indicate that ‘disputes arising from such obligations shall be settled only under the terms of the contracts underlying the obligations’.125 States can also include contractually agreed dispute settlement mechanisms among the choices opened to the investor by the DRC.126

125 See, eg, Agreement between Iceland and Mexico concerning the reciprocal encouragement and protection of investments (Signed on 24 June 2005, entered into force on 28 April 2006) [Iceland-Mexico BIT] Art 8. 126 See, eg, Agreement between Argentina and the United-States of America concerning the reciprocal encouragement and protection of investment (Signed on 14 November 1991, entered into force on 20 October 1994) [Argentina-United states BIT] Art VII(2).

24 Denial of Benefits Clause PAUL JEAN LE CANNU AND LUISA F. TORRES*

I. INTRODUCTION

D

ENIAL OF BENEFITS (DoB) clauses are often described as a treaty-based tool designed to counter-balance broad treaty definitions of ‘investor’, by providing a state the ability to deny, in specific circumstances, certain benefits of the investment treaty to juridical persons that might otherwise be protected by the treaty.1 Not all investment treaties include a DoB clause, but these clauses are not uncommon either. Treaties in the US investment treaty program constitute a salient example that has featured prominently in the arbitral decisions on the subject.2 Another one is Article 17 of the Energy Charter Treaty (ECT).3 DoB provisions also feature more commonly in modern model investment treaties.4 The interpretation and application of DoB clauses in the context of Investor-State Dispute Settlement (ISDS) has given rise to a rich body of case law.5 The discussion in arbitral decisions on the subject has primarily revolved around three main aspects: (i) the nature/characterisation of DoB objections; (ii) the substantive requirements for triggering a DoB provision; and (iii) the procedural requirements for the exercise of such provision. As treaty language varies, the answer to some of these questions is informed by the specificities of the treaty at issue. Tribunals have alluded to this point. For example, in Pac Rim, the tribunal observed that it

* The authors are Legal Counsel at the International Centre for Settlement of Investment Disputes (ICSID) Washington DC, USA. The views expressed in this article are those of the authors and should not be attributed to ICSID. Research for this article was generally concluded by December 2020. 1 See, eg, L Mistelis and C Baltag, ‘Denial of Benefits’ Clause in Investment Treaty Arbitration’ (2018) 293 Queen Mary School of Law Legal Studies Research Paper 1, https://ssrn.com/abstract=3300618; M Feldman, ‘Denial of Benefits after Plama v Bulgaria’ in M Kinnear and others (eds), Building International Investment Law: The First 50 Years of ICSID (Alphen aan den Rijn, Kluwer, 2016) 465; L Gastrell and PJ Le Cannu, ‘Procedural Requirements of ‘Denial of Benefits’ Clauses in Investment Treaties: A Review of Arbitral Decisions’ (2015) 30(1) ICSID Review – Foreign Investment Law Journal 78, 82. 2 For a table of publicly available decisions on the subject and the treaties at issue in those cases, see Annex A. For treaties in the US BIT and FTA programmes, see https://tcc.export.gov/Trade_Agreements/Bilateral_Investment_ Treaties/index.asp and https://ustr.gov/trade-agreements/free-trade-agreements. 3 Energy Charter Treaty [ECT], Art 17, https://perma.cc/BY4Y-K7XV. 4 See, eg, 2012 US Model Bilateral Investment Treaty [2012 US Model BIT], Art 17, https://perma.cc/R9Z7-DSEE; 2014 Canada Model Foreign Investment Protection Agreement [2014 Canada Model FIPA], Art 19, https://perma.cc/ JH23-RZZW. 5 See Annex A.

398 Paul Jean Le Cannu and Luisa F. Torres did not find it desirable to draw from past decisions under other treaties ‘given their different wording, context and effects’.6 As shown in this chapter, certain decisions on this topic have been considered as influential in shaping particular aspects of the debate, and figure more prominently in subsequent cases or doctrinal analysis. It is also fair to say, however, that over time the discussion on the different aspects of the subject matter has been enriched by the contributions of many decisions. It is therefore difficult to single out one case as the only ‘landmark’ case in all aspects of the interpretation and application of DoB provisions. This chapter will provide an overview of the decisions that have addressed the questions of: (i) the nature/characterisation of DoB objections (section II); and (ii) the substantive requirements for triggering the provision (section III). An in-depth analysis of the discussion and decisions around the procedural aspects of the exercise of a DoB provision is available in an earlier publication by one of the authors.7

II. NATURE OF THE DENIAL OF BENEFITS OBJECTION

Before any attempt is made to determine which case(s) addressing the nature of DoB objections may qualify as a ‘landmark’ case, one should very briefly recall that the distinction between such concepts as jurisdiction, admissibility and the merits has long been the subject of debate in international litigation. Some have argued that this debate has not achieved the desired level of clarity. In his 2006 study of the law and practice of the International Court of Justice, Prof. Shabtai Rosenne came to the conclusion that there was no ‘certainty or unanimity over the categorization of preliminary objections’ and that ‘in each instance the decisive appreciation will always be based on a combination of specifics and the subjective assessment of the majority’.8 A number of scholars see a similar state of ‘confusion’9 in the field of international investment law,10 and have suggested a variety of solutions and criteria to distinguish the different categories of preliminary objections.11

6 Pac Rim Cayman LLC v Republic of El Salvador, Decision on the Respondent’s Jurisdictional Objections (1 June 2012) ICSID Case No ARB/09/12 [Pac Rim], paras 4.3–4.5. See also: Ampal-American Israel Corp., EGI-FUND (08-10) Investors LLC, EGI-Series Investments LLC, BSS-EMG Investors LLC, and Mr. David Fischer v Arab Republic of Egypt, Decision on Jurisdiction (1 February 2016) ICSID Case No ARB/12/11 [Ampal], para 129; Luxtona Limited v Russian Federation, Interim Award on Respondent’s Objections to the Jurisdiction of the Tribunal (22 March 2017) PCA Case No 2014-09 [Luxtona], para 281. 7 Gastrell and Le Cannu, above (n 1) (discussing two broad lines of cases, similar to those explored in section II below). For other decisions addressing the procedural requirements after 2015, see, eg, Ampal, above (n 6) paras 124–173; Isolux Netherlands, B.V. v Kingdom of Spain, Award (12 July 2016) SCC Case No V2013/153 [Isolux], para 715; Luxtona, above (n 6) paras 256–282; NextEra Energy Global Holdings B.V. and NextEra Energy Spain Holdings B.V. v Kingdom of Spain, Decision on Jurisdiction, Liability and Quantum Principles (12 March 2019) ICSID Case No ARB/14/11 [NextEra], paras 262–271; Gran Colombia Gold Corp. v Republic of Colombia, Decision on the Bifurcated Jurisdictional Issue (23 November 2020) ICSID Case No ARB/18/23 [Gran Colombia], paras 126–132. 8 S Rosenne, The Law and Practice of the International Court of Justice 1920–2005 (Leiden, Brill, 2006) 848. 9 H Wehland, ‘Jurisdiction and Admissibility in Proceedings under the ICSID Convention and the ICSID Additional Facility Rules’ in C Baltag (ed), ICSID Convention after 50 Years: Unsettled Issues (Alphen aan den Rijn, Kluwer, 2016) 227. 10 See, eg, F Fontanelli, Jurisdiction and Admissibility in Investment Arbitration: The Practice and the Theory (Leiden, Brill, 2018) 99 et seq; S Pauker, ‘Admissibility of Claims in Investment Treaty Arbitration’ (2018) 34(1) Arbitration International 1, 6; C Söderlund and E Burova, ‘Is There Such a Thing as Admissibility in Investment Arbitration?’ (2018) 33(2) ICSID Review – Foreign Investment Law Journal 525, 526; M Hwang and S Cheng Lim, ‘The Chimera of Admissibility in International Arbitration’ in N Kaplan and M J Moser (eds), Jurisdiction, Admissibility and Choice of Law in International Arbitration: Liber Amicorum Michael Pryles (Alphen aan den Rijn, Kluwer, 2018) 265; F Santacroce, ‘Navigating the Troubled Waters Between Jurisdiction and Admissibility: An Analysis of Which

Denial of Benefits Clause 399 Amidst this ongoing debate, a review of the cases, whether under the ICSID Convention or other arbitration rules, shows that a number of tribunals have had recourse to the concepts of jurisdiction, admissibility and the merits to characterise the nature and effect of DoB objections. Importantly, differences in treaty language appear to have been a key factor in this categorisation exercise. Indeed, while two broad lines of cases can be identified – one inaugurated by Plama12 under the ECT and the other by Ulysseas13 and Pac Rim14 under BITs and FTAs – the general approach followed by tribunals in both sets of cases focused particularly on one issue: whether the language of the treaty suggests that the dispute resolution provisions are among the ‘benefits’ or ‘advantages’ to which the DoB clause applies. It seems that only one tribunal – in Generation Ukraine15 – took a different approach when interpreting treaty text similar to those discussed in the Ulysseas/Pac Rim line of cases. The analysis offered by the Generation Ukraine tribunal does not appear to have been followed in subsequent cases.

A. Denying the Benefit of Substantive (Treaty) Provisions: Plama Plama is the first case in which a tribunal analysed in detail the nature of the DoB clause in Article 17 of the ECT. The tribunal’s analysis in its Decision on Jurisdiction has been abundantly referred to and discussed by subsequent tribunals16 and scholars.17 Law Should Govern Characterisation of Preliminary Issues in International Arbitration’ (2017) 33(1) Arbitration International 539, 539–540; F Fontanelli and A Tanzi, ‘Jurisdiction and Admissibility in Investment Arbitration: A View from the Bridge at the Practice’ (2017) 16 The Law and Practice of International Courts and Tribunals 3, 6 et seq; Y Shany, Questions of Jurisdiction and Admissibility before International Courts (Cambridge, Cambridge University Press, 2016) 2–3; L Gouiffès and M Ordonez, ‘Jurisdiction and Admissibility: Are We Any Closer to a Line in the Sand?’ (2015) 31(1) Arbitration International 107, 110–113; Y Banifatemi and E Jacomy, ‘Compétence et recevabilité dans le droit de l’arbitrage en matière d’investissements’ in C Leben (ed), Droit International des Investissements et de l’Arbitrage Transnational (Paris, Pedone, 2015) 774; V Heiskanen, ‘Ménage à trois? Jurisdiction, Admissibility and Competence in Investment Treaty Arbitration’ (2014) 29(1) ICSID Review – Foreign Investment Law Journal 231; M Waibel, ‘Investment Arbitration: Jurisdiction and Admissibility’ (2014) 9 University of Cambridge Faculty of Law Legal Studies Research Paper Series 3, 7; Z Douglas, International Law of Investment Claims (Cambridge, Cambridge University Press, 2009) paras 291, 295 and fn 16; D Williams, ‘Jurisdiction and Admissibility’ in P Muchlinski, F Ortino and C Schreuer (eds), The Oxford Handbook of International Investment Law (Oxford, Oxford University Press, 2008) 919; G Zeiler, ‘Jurisdiction, Competence, and the Admissibility of Claims in ICSID Arbitration Proceedings’ in C Binder and others (eds), International Investment Law for the 21st Century: Essays in Honour of Christoph Schreuer (Oxford, Oxford University Press, 2009) 81–82; I Laird, ‘A Distinction without a Difference? An Examination of the Concepts of Admissibility and Jurisdiction in Salini v Jordan and Methanex v USA’ in T Weiler (ed), International Investment Law and Arbitration: Leading Cases from ICSID, NAFTA and Customary International Law (Berlin, Cameron May Publishing, 2005) 216; C Santulli, Droit du Contentieux International (Paris, Montchrestien, 2005) 255. 11 See in particular on the distinction between jurisdiction and admissibility, T Obamuroh, ‘Jurisdiction and admissibility: a case study’ (2020) 36(3) Arbitration International 373, 377 et seq; Fontanelli, above (n 10) 146 et seq; Santacroce, above (n 10) 540 et seq; Hwang and Lim, above (n 10) 266, 287–288; A Reinisch, ‘Jurisdiction and Admissibility in International Investment Law’ (2017) 16 The Law and Practice of International Courts and Tribunals 21, 23; Wehland, above (n 9) 234; Banifatemi and Jacomy, above (n 10) 775 et seq; L Velarde Saffer and J Lim, ‘Judicial Review of Investor Arbitration Awards: Proposals to Navigate the Twilight Zone Between Jurisdiction and Admissibility’ (2014) 8(1) Dispute Resolution International 85, 87, 90–93; Heiskanen, above (n 10) 237–242; Waibel, above (n 10) 2–3, 5–9; Douglas, above (n 10) paras 291–295, 310–311; J Paulsson, ‘Jurisdiction and Admissibility’ in G Aksen and R Briner (eds), Global Reflections on International Law, Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner (Paris, ICC Publishing, 2005) 608, 616; G Fitzmaurice, The Law and Procedure of the International Court of Justice (Cambridge, Cambridge University Press, 1986) 438–439. 12 Plama Consortium Limited v Republic of Bulgaria, Decision on Jurisdiction (8 February 2005) ICSID Case No ARB/03/24 [Plama Decision]. 13 Ulysseas, Inc. v Republic of Ecuador, Interim Award (28 September 2010) UNCITRAL [Ulysseas]. 14 Pac Rim, above (n 6). 15 Generation Ukraine, Inc. v Ukraine, Award (16 September 2003) ICSID Case No ARB/00/9 [Generation Ukraine]. 16 See examples below. 17 See inter alia the commentary referred to below. See also the views and references on Plama provided in Feldman, above (n 1) 463, fn 3.

400 Paul Jean Le Cannu and Luisa F. Torres Article 17 of the ECT provides in relevant part that: ‘[e]ach Contracting Party reserves the right to deny the advantages of [] Part [III of the ECT]’ under certain conditions. Part III contains a number of provisions concerning the promotion and protection of investments.18 In Plama, the respondent contended that a denial by the host state of the advantages contemplated in Article 17(1) of the ECT would deprive the tribunal of all jurisdiction. According to the respondent, ‘the denial of such “advantages” [was] not limited to those conferred by Part III of the ECT but include[d] also all advantages relating to Part III, including the right to invoke international arbitration under Article 26 of Part V of the ECT alleging any breach of Part III’.19 The respondent argued that Article 17 of the ECT conferred on the host state ‘a direct and unconditional right of denial which may be exercised at any time and in any manner’.20 By contrast, the claimant saw the respondent’s reliance on Article 17(1) not as ‘a true objection to jurisdiction but rather [as] a defense on the merits’21 because Part III of the ECT does not include the dispute resolution provisions of the treaty.22 For the Plama tribunal, Article 17(1) of the ECT could not be interpreted as going to the jurisdiction of the tribunal for two main reasons: (i) this interpretation would not only have been incompatible with the ordinary meaning of the terms of Article 17(1); (ii) it would have also been contrary to the ECT’s object and purpose. Invoking Article 31(1) of the Vienna Convention on the Law of Treaties (VCLT), the tribunal first emphasised that the terms of Article 17, including its title, unambiguously refer to a denial of the advantages of ‘this Part’, namely ‘Part III’ of the ECT,23 contrary to ‘most modern investment treaties’ where the DoB provision ‘operate[s] as a denial of all benefits to a covered investor under the treaty […]’.24 The tribunal reasoned that: Article 26 […] is not physically or juridically part of the ECT’s substantive advantages enjoyed by that investor under Part III. As a matter of language, it would have been simple to exclude a class of investors completely from the scope of the ECT as a whole, as do certain other bilateral investment treaties; but that is self-evidently not the approach taken in the ECT. […].25

In the eyes of the tribunal, only a ‘gross manipulation’ of the language of Article 17 would allow one to read it as referring to Part V of the ECT, which contains the dispute resolution provisions of the treaty.26 Even if such ‘manipulation’ was ‘permissible’,27 the tribunal was of the view that the object and purpose of the ECT required Article 26 of the ECT – the dispute resolution provision – to remain ‘unaffected by the operation of Article 17(1)’ because Article 26 ‘is a very important feature of the ECT’,28 ‘perhaps the most important aspect of the ECT’s investment regime […]’,29 as it provides covered investors ‘an almost unprecedented remedy’ for their claims against the host state.30

18 ECT,

above (n 3) Arts 10–17. Decision, above (n 12) para 146. 20 ibid para 144. 21 ibid para 41. 22 ibid paras 41, 70. 23 ibid para 147. 24 ibid para 149 (emphasis added). 25 ibid para 148. On the separability of the provisions of Art 26, see ibid para 130. 26 ibid para 147. 27 ibid. 28 ibid para 148. 29 ibid para 141. 30 ibid paras 140 and 141. 19 Plama

Denial of Benefits Clause 401 In addition, the tribunal considered that, if Article 17 was read as raising jurisdictional issues and Article 26 was therefore unavailable to investors, a whole range of potentially controversial issues that may arise out of the exercise of the right to deny benefits could not be determined, such as ‘citizenship, nationality, ownership, control and the scope and location of business activities […]’.31 In the tribunal’s view, the respondent’s argument that ‘there was no remedy available to a covered investor under the ECT at all’ would make the Contracting State invoking Article 17(1) ‘the judge of its own cause’.32 The Plama tribunal further noted that if the host State’s invocation of Article 17(1) were to prevent an investor from invoking Article 26, Article 27 of the ECT would still leave intact the right of the investor’s home state to invoke the state-to-state dispute resolution provisions against the host state. For the tribunal, it seemed ‘an unnecessarily complicated result to resolve that dispute when, on the ordinary meaning of Article 17(1), the covered investor could invoke Article 26 directly against the host state without the assistance of its home state’.33 The tribunal thus concluded that ‘the Respondent’s case on Article 17(1) cannot support a complaint to the jurisdiction of the Tribunal in this case’;34 and that the application of this provision ‘related to the merits’ of the parties’ dispute.35 In the merits phase of the proceedings, Bulgaria argued that Plama’s claims were ‘inadmissible’ on the ground that Plama had no substantial business activities in Cyprus,36 and had failed to prove that it was a legal entity owned or controlled by nationals of a Contracting Party to the ECT within the meaning of Article 17(1).37 In its Award, while the tribunal addressed the issue in a ‘preliminary discussion’,38 it did not expressly characterise it as one of admissibility.39 Rather, the tribunal confirmed its earlier conclusion that ‘Article 17(1) of the ECT ha[d] no relevance to the Tribunal’s jurisdiction to determine Claimant’s claims against Respondent under Part III of the Treaty […]’.40 As noted above, the Plama Decision, including its analysis on the nature and effect of the ECT’s DoB clause, has been abundantly discussed. A number of authors have questioned the view that Article 17 raises a ‘merits’ issue, as held by the Plama tribunal,41 with some suggesting that Article 17 should be read as raising an issue of jurisdiction.42 Some states have similarly contended that the exercise of the state’s right under Article 17 is ‘a matter of jurisdiction’

31 ibid

para 149.

32 ibid. 33 ibid

para 150. para 151. 35 ibid. 36 Plama Consortium Limited v Republic of Bulgaria, Award (27 August 2008) ICSID Case No ARB/03/24 [Plama Award], para 81. 37 ibid para 87. 38 ibid Section IV, ‘Preliminary Discussion: Claimant’s ‘Ownership’ and ‘Control’ and The Allegations of Misrepresentation’. 39 The terms ‘admissibility’ or ‘admissible’ are not used by the tribunal in the Plama Decision. 40 Plama Award, above (n 36) para 89. 41 Pauker, above (n 10) 55. See also Douglas, above (n 10) paras 874, 878. 42 L Shore, ‘The Jurisdiction Problem in Energy Charter Treaty Claims’ (2007) 10 International Arbitration Law Review 58, 63. See also J Chalker, ‘Making the Energy Charter Treaty Too Investor Friendly: Plama Consortium Limited v the Republic of Bulgaria’ (2006) 3(5) Transnational Dispute Management 1, 7; C Baltag, Jurisdictional Limits of The Energy Charter Treaty and Its Interplay with Related Treaties and Arbitration Rules: The Notion of Investor, Thesis (London, Queen Mary University of London, 2011) 186, https://core.ac.uk/download/ pdf/77038605.pdf. 34 ibid

402 Paul Jean Le Cannu and Luisa F. Torres arguing that ‘the consent to arbitration given in Article 26(1) of the ECT refers to disputes that “concern an alleged breach of an obligation […] under Part III”’.43 In addition, some authors have disputed the Plama tribunal’s view that if Article 17(1) were treated as jurisdictional, it would make the denying Contracting Party ‘the judge of its own cause’44 and create an ‘unnecessary complicated result’45 at odds with the ordinary meaning of this provision.46 Some of them submit that, if it were assumed that Article 17 raises an issue of jurisdiction, it would not amount to a self-judging clause in light of the principle of compétence de la compétence.47 As to the unnecessary complications that the Plama tribunal identified, it has been suggested that difficulties could equally arise for the respondent state ‘from reserving to the merits stage the consideration of whether the ‘Investor’ is a legal entity [as] to which the state can avoid liability’.48 As shown below, however, the jurisdictional analysis has not prospered before tribunals constituted under the ECT. Other commentators have argued that Article 17 raises an issue of admissibility rather than jurisdiction.49 Zachary Douglas takes the example of Generation Ukraine,50 where the tribunal ruled on the basis of a differently-worded provision – Article I(2) of the US-Ukraine BIT – pursuant to which ‘[e]ach party reserves the right to deny to any company the advantages of this Treaty’51 under certain conditions. The Generation Ukraine tribunal was of the view that Article I(2) did not raise jurisdictional issues: […] the burden of proof to establish the factual basis of the ‘third country control’, together with the other conditions [set forth in Article 1(2)], falls upon the State as the party invoking the ‘right to deny’ conferred by Article 1(2). This is not, as the Respondent appears to have assumed, a jurisdictional hurdle for the Claimant to overcome in the presentation of its case; instead it is a potential filter on the admissibility of claims which can be invoked by the respondent State.52

While some scholars suggest that ‘[…] the Tribunal did not fully explain how it reached this conclusion’,53 others argue that the tribunal ‘helpfully’ described the right to deny treaty

43 See,

eg, NextEra, above (n 7) para 213; Isolux, above (n 7) paras 236–237. Decision, above (n 12) para 149. 45 ibid para 150. 46 Shore, above (n 42) 63; Douglas, above (n 10) para 878, fns 1 and 3; D Burriez, ‘La clause de déni des avantages et l’accès à l’arbitrage en matière d’investissement’ in A de Nanteuil (ed), L’accès de l’investisseur à la justice arbitrale (Paris, Pédone, 2015) 75. 47 Douglas, above (n 10) para 878, fns 1 and 3; Burriez, above (n 46) 75. 48 Shore, above (n 42) 63. 49 H Essig, ‘Balancing Investors’ Interests and State Sovereignty: The ICSID-Decision on Jurisdiction Plama Consortium Ltd. v Republic of Bulgaria’ (2007) 4(5) Transnational Dispute Management 1, 8. 50 Douglas, Rule 54, based on Generation Ukraine states: ‘Where a ‘denial of benefits’ provision is successfully invoked by the host state in arbitration proceedings against the claimant, the substantive protection of the investment treaty is denied to the claimant and its claims must be dismissed as inadmissible.’ Douglas, above (n 10) 468 and fn 1. 51 Treaty Between the United States of America and Ukraine Concerning the Encouragement and Reciprocal Protection of Investment, signed 4 March 1994, in force 16 November 1996 [US-Ukraine BIT], Art I(2) (emphasis added). 52 Generation Ukraine, above (n 15) para 15.7 (emphasis added). 53 C McLachlan, L Shore and M Weiniger, International Investment Arbitration, Substantive Principles, 2nd edn (Oxford, Oxford University Press, 2017) para 5.187. For a critical approach to the DoB ruling in Generation Ukraine, including on the burden of proof analysis, see M Feldman, ‘Setting Limits on Corporate Nationality Planning in Investment Treaty Arbitration’ (2012) 27(2) ICSID Review – Foreign Investment Law Journal 281, 298 and fn 111. 44 Plama

Denial of Benefits Clause 403 benefits as an admissibility filter.54 Building on the tribunal’s reasoning, Douglas offers the following analysis: A jurisdictional requirement must be positively established by the claimant. A ‘denial of benefits’ provision must be positively invoked by the respondent. A ‘denial of benefits’ provision is not selfjudging. The burden of proof clearly falls upon the respondent host state and if that burden is discharged before the tribunal, then the claimant’s claims must be dismissed. A ‘denial of benefits’ clause obviously does not supply a defence on the merits of the claims; if the provision is properly invoked then the merits of the claims will never be tested. It is thus a matter of admissibility.55

Douglas further argues that where this analysis is applied to the ECT, it should be concluded that a denial of advantages under that treaty goes to the admissibility of the claims because ‘it is the substantive protection that is denied by Article 17 of the ECT […]’,56 ie Part III of the treaty. However, Douglas argues, the interpretation would be different ‘if there were an express reference to Part V of the ECT in Article 17’.57 In that case, he submits, the invocation of Article 17 would then amount to a jurisdictional objection.58 Stephen Jagusch and Anthony Sinclair suggest a somewhat different reading of the Plama Decision, which they understand as addressing the issue of ‘whether Article 17 could support an objection to the admissibility of Plama’s ECT claims on the merits’.59 The divergence of views expressed by investment arbitration scholars does not appear to have found any echo in post-Plama ECT cases. While offering less detailed analysis than Plama, the tribunals which have subsequently addressed the nature of the DoB clause under Article 17 of the ECT all appear to agree that the clause does not raise jurisdictional issues. However certain differences do remain. In a number of cases the tribunals took a similar approach to that of the Plama tribunal and found that Article 17 of the ECT raised a ‘merits’ issue. For the tribunals in the Yukos cases, ‘[w]hether or not Claimant is entitled to the advantages of Part III is a question not of jurisdiction but of the merits.’60 Like the Plama tribunal, the Yukos tribunals emphasised that Article 17 of the ECT specified in both its title and its text that denial of treaty advantages was limited to ‘this Part’, ie Part III of the ECT, and that the ECT dispute settlement provisions were found in Part V of the ECT.61 The tribunals in Khan,62 and 54 S Jagusch and 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 (New York, Juris Publishing, 2006) 103. See also the analysis of the similarly-worded DoB clause of the 2012 US Model BIT in AM Steingruber, ‘Some remarks on Veijo Heiskanen’s Note-Ménage à trois? Jurisdiction, Admissibility and Competence in Investment Treaty Arbitration’ (2014) 29(3) ICSID Review – Foreign Investment Law Journal 675, 682, 684, 688–689. 55 Douglas, above (n 10) para 874. 56 ibid para 876. 57 ibid (emphasis added). 58 ibid. 59 Jagusch and Sinclair, above (n 54) 98 (emphasis in the original). 60 Hulley Enterprises Limited (Cyprus) v The Russian Federation, Interim Award on Jurisdiction and Admissibility (30 November 2009) UNCITRAL, PCA Case No AA 226 [Hulley], para 440 (emphasis added); Veteran Petroleum Limited (Cyprus) v The Russian Federation, Interim Award on Jurisdiction and Admissibility (30 November 2009) UNCITRAL, PCA Case No AA 228 [Veteran], para 497 (emphasis added); Yukos Universal Limited (Isle of Man) v The Russian Federation, Interim Award on Jurisdiction and Admissibility (30 November 2009) UNCITRAL, PCA Case No AA 227 [Yukos], para 441 (emphasis added). 61 Hulley, above (n 60) paras 441–442; Veteran, above (n 60) paras 497–499; Yukos, above (n 60) paras 441–443. The Yukos tribunals expressly adopted the reasoning from the Plama Decision, above (n 12) para 148. 62 Khan Resources Inc., Khan Resources B.V., and CAUC Holding Company Ltd. v The Government of Mongolia and MonAtom LLC, Decision on Jurisdiction (25 July 2012) UNCITRAL, PCA Case No 2011-09 [Khan], paras 411–412.

404 Paul Jean Le Cannu and Luisa F. Torres Luxtona,63 expressly referred to Plama and/or Yukos (or Khan in the case of Luxtona) and offered a similar textual analysis. Having noted that Article 17 of the ECT does not raise a jurisdictional issue, other tribunals, by contrast, have not expressly stated whether the issue is one of admissibility or the merits. While the Liman tribunal suggested that denial of benefits was not a jurisdictional issue ‘stricto sensu’;64 the Stati tribunal relied on Plama to reject the respondent’s characterisation of Article 17 as jurisdictional,65 but did not expressly specify whether Article 17 of the ECT raised ‘admissibility’ or ‘merits’ issues. Only one ECT tribunal has characterised Article 17 as raising an admissibility issue. In Isolux, the tribunal had ‘no doubt that the denial of benefits under Article 17 of the ECT raises a question of admissibility’.66 The reasoning was similar to that of the Plama tribunal; analysing the DoB provision as jurisdictional in nature ‘would deprive the investor of any necessary forum to decide this issue’.67 The Isolux tribunal found support in the Plama and Yukos decisions, which in its view had confirmed that denial of benefits was ‘an issue of admissibility or of the merits’ and not a jurisdictional one.68 B. Denying the Benefit of Jurisdictional (Treaty) Provisions: Ulysseas and Pac Rim While, as shown above, Plama may have been a reference decision for tribunals seeking to determine the nature of the DoB clause under Article 17 of the ECT, it has not been seen as such by tribunals constituted under BITs and FTAs, whose DoB provisions are worded differently from the ECT’s. Rather, a majority of these tribunals have analysed these provisions as raising a jurisdictional issue.69 The first two cases in which the nature of objections under these clauses was analysed in some detail were Ulysseas – under the US-Ecuador BIT – and Pac Rim – under the CAFTA-DR – which together inaugurated a second and separate line of cases upholding the jurisdictional nature of DoB clauses under similarly-worded instruments. In Ulysseas, the basis of jurisdiction invoked by the claimant company was the US-Ecuador BIT,70 which provided in relevant part that ‘[e]ach Party reserves the right to deny to any

63 Luxtona, above (n 6) para 239. In AMTO, the respondent appeared to argue that the claimant’s case was ‘inadmissible’ on the basis of Art 17(1) of the ECT and that Art 17(1) was outside the scope of the tribunal’s jurisdiction ratione materiae. Limited Liability Company AMTO v Ukraine, SCC Case No 080/2005, Final Award (26 March 2008) [AMTO], paras 26(h), 59, 60. The tribunal dismissed these arguments, but did not specifically state whether a DoB objection under the ECT went to jurisdiction or admissibility. 64 Liman Caspian Oil BV and NCL Dutch Investment BV v Republic of Kazakhstan, Excerpts of the Award (22 June 2010) ICSID Case No ARB/07/14 [Liman], para 215. 65 Anatolie Stati, Gabriel Stati, Ascom Group SA and Terra Raf Trans Traiding Ltd v Kazakhstan, Award (19 December 2013) SCC Case No V116/2010 [Stati], para 745. 66 Isolux, above (n 7) para 711 (authors’ free translation). 67 ibid para 712 (authors’ free translation). See also Plama Decision, above (n 12) para 149. 68 Isolux, above (n 7) para 713 (authors’ free translation). 69 Other tribunals have held that the characterisation of similarly worded provisions as going to jurisdiction or admissibility was of little consequence. See, eg, Pan American Energy LLC, and BP Argentina Exploration Company v The Argentine Republic, ICSID Case No ARB/03/13, and BP America Production Company et al v The Argentine Republic, Decision on Preliminary Objections (27 July 2006) ICSID Case No ARB/04/8 [Pan American], para 54 (footnote omitted). See also R Happ, ‘Denial of Benefit Clauses any Other Mechanisms that Limit the Scope of BITs for Investors’ in ILA German Branch Sub-Committee on Investment Law, The Determination of the Nationality of Investors in Investment Protection Treaties (Halle, Transnational Economic Law Research Centre, 2011) 66–67. 70 In EMELEC, Ecuador raised an objection to jurisdiction on the basis of the DoB provision of the US-Ecuador BIT, but the tribunal dismissed the case on a different jurisdictional basis and did not pass upon the respondent’s DoB argument. It did hold, however, that ‘Ecuador [had] announced the denial of benefits to EMELEC at the proper stage of the proceedings, i.e. upon raising its objections on jurisdiction’. Empresa Eléctrica del Ecuador, Inc. v Republic of Ecuador, Award (2 June 2009) ICSID Case No ARB/05/9 [EMELEC], para 71.

Denial of Benefits Clause 405 company the advantages of this Treaty’ if certain requirements are met.71 The award indicates that Ecuador invoked the ‘denial of benefits’ provision of the treaty in its first written submission to deny the claimant ‘the advantages of the BIT, both substantive and procedural, including the right to have recourse to arbitration before this Tribunal’.72 The Ulysseas tribunal agreed with the respondent that its DoB objection was of a jurisdictional nature. Indeed, the tribunal stated that, in its view: ‘since [the] advantages [of this Treaty] include BIT arbitration, a valid exercise of the right would have the effect of depriving the Tribunal of jurisdiction under the BIT’.73 The reference in Article I(2) of the BIT to ‘the right to deny […] the advantages of this Treaty’ appears to have been understood as including all the advantages of the Treaty, with the tribunal considering the dispute resolution provisions as one of those ‘advantages’. In Pac Rim, the state’s DoB objection was based on Article 10.12.2 of CAFTA,74 which provides that ‘[…] a Party may deny the benefits of this Chapter’ (ie Chapter 10 on investment) subject to certain conditions being met.75 The parties disagreed as to whether an objection based on this provision went to the jurisdiction of the tribunal.76 While the Pac Rim tribunal took a similar approach to that of the Ulysseas tribunal,77 it made it clear that reliance on decisions under different treaties, including the ECT, was not ‘desirable […] given their different wording, context and effect’.78 The Pac Rim tribunal specifically referred to Plama, AMTO,79 Petrobart,80 Generation Ukraine, and EMELEC.81 In the tribunal’s view, a DoB under CAFTA raised an issue of jurisdiction and had to be distinguished from the ‘more limited issue’ arising under Article 17 of the ECT: As expressly worded in CAFTA, it is significant that the ‘benefits’ denied under CAFTA Article 10.12.2 include all the benefits conferred upon the investor under Chapter 10 of CAFTA, including both Section A on ‘Investment’ and Section B on ‘Investor-State Dispute Settlement.’ Section B specifically includes CAFTA Article 10.16(3)(a) providing for ICSID arbitration, as here invoked by the Claimant for its claims under CAFTA to establish the Tribunal’s jurisdiction to decide those CAFTA claims against the Respondent. This jurisdictional issue under CAFTA does not

71 Treaty Between United States of America and the Republic of Ecuador Concerning Encouragement and Reciprocal Protection of Investment, signed 27 August 1997, in force 11 May 1997, date of termination 18 May 2018 [US-Ecuador BIT], Art I(2). 72 Ulysseas, above (n 13) para 164. 73 ibid para 172. The Ulysseas tribunal concluded that the claimant’s claims were ‘within the Tribunal’s jurisdiction and are admissible.’ ibid para 190. This appears to be in response to the parties’ requests for relief: the respondent requested a ruling that the ‘Claimant’s claims in these proceedings are outside the jurisdiction of this Tribunal or, in the alternative, inadmissible;’ and conversely, the claimant requested that the tribunal confirm that ‘its claims [were] within the Tribunal’s jurisdiction and are admissible.’ ibid paras 143, 146. The Interim Award does not contain any discussion with respect to admissibility in connection with the respondent’s DoB objection. On the characterisation of the DoB objection in this case, see the views expressed in Burriez, above (n 46) 89–90. 74 Pac Rim, above (n 6) para 1.17. The other source of consent invoked by the claimant (a law) did not contain any DoB clause. 75 Dominican Republic-Central America-United States Free Trade Agreement, signed 5 August 2004 [CAFTA], Art 10.12.2. 76 See, eg, Claimant’s Rejoinder on Respondent’s Objections to Jurisdiction, 2 March 2011, para 216, https://icsid. worldbank.org/cases/case-database/case-detail?CaseNo=ARB/09/12. 77 There is no reference to the Ulysseas Interim Award in the Pac Rim Decision. When the Pac Rim Decision was issued, it appears that the Ulysseas Interim Award had not yet been published. See L Peterson, ‘Analysis: Choice of BIT arbitration over contractual remedies fails to pay off in Ulysseas v Ecuador case; stabilization clause also hurts case’ (IA Reporter, 18 June 2012), referring to the ‘unpublished interim award’ in that case. 78 Pac Rim, above (n 6) para 4.3. 79 AMTO, above (n 63). 80 Petrobart Limited v The Kyrgyz Republic, Award (29 March 2005) SCC Case No 126/2003 [Petrobart]. 81 Pac Rim, above (n 6) para 4.3.

406 Paul Jean Le Cannu and Luisa F. Torres therefore resemble the more limited issue under Article 17(1) of the Energy Charter Treaty, although in this respect it resembles the position under Article 1113(1) of NAFTA. […].82

As in Ulysseas, the treaty – this time CAFTA – was read in such a way as to include the dispute resolution provisions among the ‘benefits’ of the treaty chapter on investment. The same approach was taken in the commentary to the 2012 SADC Model Bilateral Investment Treaty Template whose drafters expressly note that the denial of benefits provision of the model treaty ‘provides for two types of situations where a State Party may exercise its right to deny an investor the benefits of the treaty, including access to any dispute settlement benefits’.83 Since then several tribunals have adopted a similar stance, both under BITs and FTAs with similarly worded DoB clauses. In Guaracachi,84 the basis of jurisdiction invoked by Guaracachi America, Inc. (GAI), a company incorporated in the US, was the US-Bolivia BIT,85 whose DoB clause is very similar to that of the US-Ecuador BIT and refers to ‘the benefits of this Treaty’.86 There as well, the tribunal interpreted the language of the clause as encompassing the dispute resolution provisions: No one can accept more than what is being offered. In this case, what was offered by both Bolivia and the US, in the BIT concluded between them, was a package of benefits to investors of both countries, including the benefit of being able to submit disputes to arbitration, coupled with an express prior reservation of the right to deny those benefits if and when the Respondent so decides (subjective requirement) and if the investor’s company is or becomes a ‘shell company’ controlled by a company incorporated in a third country (objective requirement). The reservation of the right of denial of benefits contained in Article XII operates on the Contracting Parties’ offer of consent to arbitration as much as every other benefit conferred by the BIT.87

In Bridgestone, this time under the US-Panama TPA, the tribunal was also satisfied that a DoB raised an issue of jurisdiction in light of the wording of the TPA. Similar to CAFTA, the US-Panama TPA provides that ‘a [TPA] Party may deny the benefits of this Chapter’, ie Chapter 10 on investment.88 This chapter itself comprises two sections, one on investment (Section A) where the DoB provision is found, and another on investor-state dispute settlement (Section B). The Bridgestone tribunal held that the DoB objection raised by the state went ‘to competence […] because one of the benefits of Chapter 10 is the right to arbitrate. That right is subject to the condition subsequent that it is not removed by a valid denial of benefits’.89 82 ibid para 4.4 (emphasis added). At the time of writing, the authors are not aware either of any decisions or awards under NAFTA or CAFTA that may have confirmed, or departed from, the Pac Rim tribunal’s analysis. See the references to DoB in Waste Management, Inc. v United Mexican States, Award (30 April 2004) ICSID Case No ARB(AF)/00/3 [Waste Management], para 80; Saint Marys VCNA, LLC v Government of Canada, Consent Award (12 April 2013) UNCITRAL [Saint Marys], paras 4, 5; B-Mex, LLC and Others v United Mexican States, Partial Award (19 July 2019) ICSID Case No ARB(AF)/16/3 [B-Mex], para 202. 83 2012 SADC Model Bilateral Investment Treaty Template with Commentary, 48 (emphasis added). Art 26.1 of the SADC Model BIT Template provides in relevant part that ‘[a] Party may at any time deny the benefits of this Agreement to an investor of another Party […]’ (emphasis added). 84 Guaracachi America, Inc. and Rurelec PLC v The Plurinational State of Bolivia, Award (31 January 2014) UNCITRAL, PCA Case No 2011-17 [Guaracachi]. 85 The UK-Bolivia BIT, which the other claimant relied upon, does not contain any DoB provision. 86 Treaty between The Government of The United States of America and The Government of The Republic of Bolivia Concerning The Encouragement and Reciprocal Protection of Investment, signed 17 April 1998, in force 6 June 2001, date of termination 10 June 2012 [US-Bolivia BIT], Art XII (‘[e]ach party reserves the right to deny to a company of the other Party the benefits of this Treaty […]’) (emphasis added). 87 Guaracachi, above (n 84) para 373 (emphasis added.) See also: ibid paras 366, 377. 88 US-Panama Trade Promotion Agreement, signed 28 June 2007, entered into force 31 October 2012 [US-Panama TPA], Art 10.12.2. 89 Bridgestone Licensing Services, Inc. and Bridgestone Americas, Inc. v Republic of Panama, Decision on Expedited Objections (13 December 2017) ICSID Case No ARB/16/34 [Bridgestone], para 288 (emphasis added).

Denial of Benefits Clause 407 More recently, in Gran Colombia, a tribunal constituted under the investment chapter of the Canada-Colombia FTA and faced with similar provisions reached an analogous conclusion, holding that ‘[a] State’s well-grounded denial of benefits has the effect of denying jurisdiction to a claimant seeking to rely on the substantive and dispute resolution provisions of Chapter Eight of the FTA […]’.90 The above review of arbitral decisions and awards shows a number of different approaches to the nature of DoB provisions in investment treaties. This has led some scholars to observe that ‘[t]ribunals […] continue to disagree on where such clauses fall on the ‘jurisdiction or admissibility’ divide’.91 Still, some of these decisions appear to suggest that there may be some common ground between the approach taken under the Plama/ECT line of cases and that taken under the Ulysseas/Pac Rim/non-ECT line. Some might argue that the analyses of the Plama and Pac Rim tribunals as to the scope of the terms used in the respective DoB provisions of the ECT and the CAFTA are in reality not so dissimilar. While the Plama tribunal contrasted the ECT provision, which refers only to ‘this Part’92 with those of other investment treaties whose DoB provision ‘operate[s] as a denial of all benefits to a covered investor under the treaty […]’,93 the Pac Rim tribunal also noted that ‘CAFTA Article 10.12.2 include all the benefits conferred upon the investor under Chapter 10 of CAFTA’ and compared ‘[t]his jurisdictional issue under CAFTA’ with ‘the more limited issue’94 arising under Article 17(1) of the ECT. As the survey of arbitral decisions seems to suggest, a reference to the benefits of ‘the Treaty’ or the ‘investment chapter of the Treaty’ as opposed to a ‘part’ of a treaty addressing substantive obligations could thus have a decisive impact on a tribunal’s analysis on the nature of the DoB clause.

III. SUBSTANTIVE REQUIREMENTS

The substantive requirements for the exercise of a DoB clause are governed by the underlying treaty, and may differ among treaties. The diverse wording of DoB clauses coupled with the various factual contexts underlying the exercise of the clause has given rise to a rich body of decisions addressing the substantive elements for a denial of benefits. Subject to treaty language specificities, it can generally be said that the conditions most commonly considered by arbitral decisions have included the investor’s: (i) lack of substantial business activities in the home state; and (ii) ownership or control by a national of a third state.95 Some treaties also afford states the right to deny the benefits of the treaty in additional circumstances, such as: (i) ownership or control by a national of the host/denying state (often paired

90 Gran Colombia, above (n 7) para 131. Chapter 8 of the FTA is also divided into two sections (Section A – Investment and Section B – Settlement of Disputes between an Investor and the Host Party), with the DoB clause located in Section A and referring to ‘the benefits of this Chapter’. Free Trade Agreement between Canada and the Republic of Colombia, signed 21 November 2008, entered into force 15 August 2011 [Canada-Colombia FTA], Art 814. 91 McLachlan et al, above (n 53) para 5.187. 92 See Plama Decision, above (n 12) para 147. 93 ibid para 149 (emphasis added). 94 Pac Rim, above (n 6) para 4.4. It is worth noting in the context of the discussion on the modernisation of the ECT an amendment proposed by the EU to Art 17 of the ECT, which would include an express reference to the dispute resolution provisions of the ECT (Arts 26 and 27). See EU Text Proposal for the Modernisation of the Energy Charter Treaty (ECT), May 2020, https://perma.cc/2QB3-3QHV. 95 See, eg, ECT, above (n 3) Art 17(1); CAFTA, above (n 75) Art 10.12.2; US-Panama TPA, above (n 88) Art 10.12.2; Canada-Colombia FTA, above (n 90) Art 814(2); US-Ecuador BIT, above (n 71) Art I(2). For a list of cases that have addressed these DoB provisions, see Annex A.

408 Paul Jean Le Cannu and Luisa F. Torres with lack of substantial business activities in the home state);96 or (ii) ownership or control by nationals of a third state lacking diplomatic relations or lacking normal economic relations with the denying state.97 However, these other situations have seldom been dealt with in available arbitral decisions to date.98 Tribunals have also often been confronted with preliminary questions that bear on the analysis of compliance with the substantive requirements for exercising a DoB provision, namely: (i) whether the conditions established in the treaty are cumulative; and (ii) who bears the burden of proof to establish the existence of those conditions. Section A below provides an overview of arbitral decisions on these two preliminary issues. Sections B and C discuss arbitral decisions on the application of the substantive elements most commonly addressed by tribunals. A. Preliminary Issues: Cumulative or Individual Requirements and Burden of Proof Tribunals have generally dealt with these preliminary matters briefly, which poses a challenge to the identification of a ‘major’ decision in these two areas. Presumably, the initial cases that addressed these issues might have played some role in shaping the discussion thereafter, although tribunals seldom refer to one another when addressing these questions. Whether the conditions for exercising a DoB are to be individually or cumulative met is also a question dependent on the treaty language. This matter was the subject of controversy in one of the early cases on the subject. In Generation Ukraine, the parties debated whether Article I(2) of the US-Ukraine BIT should be interpreted in such a way that ‘the right of denial may extend to companies without substantial activities in the place of incorporation, whether or not they are subject to “third country control”’.99 Relying primarily on the comments in the US State Department Letter of Submittal at the time of ratification of the treaty, the tribunal was satisfied that ‘third country control’ was also a prerequisite for invoking the DoB provision.100 A number of tribunals that have since dealt with this question in the context of a variety of treaties have also found that, given the language of the treaty before them, the conditions at issue in the case were cumulative.101 As to burden of proof, the survey of arbitral decisions suggests a degree of uniformity on the conclusion that the burden of establishing the factual underpinnings for a denial of benefits falls on the party advancing the allegation (the denying state). However, a few tribunals have been receptive to the notion that the investor has a role in the establishment of those facts, with some recognising that the burden might be shifted to the investor in some circumstances. 96 See, eg, Treaty Between United States of America and The Argentine Republic Concerning Reciprocal Encouragement and Protection of Investment, signed 14 November 1991, in force 20 October 1994 [US-Argentina BIT], Art I(2); 2012 US Model BIT, above (n 4) Art 17(2); Australia-Korea Free Trade Agreement, signed 8 April 2014, in force 12 December 2014 [Australia-Korea FTA], Art 11.11(2). 97 See, eg, 2012 US Model BIT, above (n 4) Art 17(1); 2014 Canada Model FIPA, above (n 4) Art 19(a); ECT, above (n 3) Art 17(2); Australia-Korea FTA, above (n 96) Art 11.11(1). 98 Libananco briefly dealt with an allegation based on lack of diplomatic relations. Libananco Holdings Co. Limited v Republic of Turkey, Award (2 September 2011) ICSID Case No ARB/06/8 [Libananco], paras 1, 104(g), 551. 99 Generation Ukraine, above (n 15) para 15.2; US-Ukraine BIT, above (n 51) Art I(2) (‘Each Party reserves the right to deny to any company the advantages of this Treaty if nationals of any third country control such company and, in the case of a company of the other Party, that company has no substantial business activities in the territory of the other Party or is controlled by nationals of a third country with which the denying Party does not maintain normal economic relations.’) 100 Generation Ukraine, above (n 15) paras 15.4 to 15.7. 101 See, eg, AMTO, above (n 63) para 62 (referring to ECT, above (n 3) Art 17(1)); Plama Award, above (n 36) para 80 (referring to ECT, above (n 3) Art 17(1)); Ulysseas, above (n 13) paras 167–168 (referring to US-Ecuador BIT, above (n 71) Art I(2)); Hulley, above (n 60) para 459 (referring to ECT, above (n 3) Art 17(1)); Veteran, above (n 60) para 516 (referring to ECT, above (n 3) Art 17(1)); Yukos, above (n 60) para 460 (referring to ECT, above (n 3) Art 17(1)); Pac Rim, above (n 6) para 4.61 (referring to CAFTA, above (n 75) Art 10.12.2); Gran Colombia, above (n 7) paras 133–134 (referring to Canada-Colombia FTA, above (n 90) Art 814(2)).

Denial of Benefits Clause 409 Two of the first cases to address burden of proof were Generation Ukraine and AMTO. In Generation Ukraine, the tribunal held that the burden of establishing the factual basis for the exercise of the DoB provision fell ‘upon the State as the party invoking the “right to deny”’,102 reasoning that the DoB clause did not constitute a jurisdictional hurdle for the claimant to meet, but rather, ‘a potential filter on the admissibility of claims which can be invoked by the respondent State’.103 In turn, in AMTO, the tribunal relied on the principle that in international arbitration the ‘burden of proof of an allegation […] rests on the party advancing the allegation’;104 and confirmed that the state was responsible for establishing the factual prerequisites for its exercise of the DoB clause.105 In doing so, the tribunal rejected the respondent’s allegation that the right to deny benefits embodied in Article 17(1) of the ECT was not subject to arbitration, and was therefore self-judging.106 While the AMTO tribunal recognised that it might be difficult for a state to show compliance with the elements in Article 17(1) of the ECT and that evidence might be more easily accessible to the investor, it did not find this to be a sufficient reason to shift the burden.107 It did accept, however, that the ‘relative accessibility of evidence’ may support the claimant’s duty to disclose the evidence or the drawing of negative inferences for failure to do so.108 Subsequently, tribunals such as Liman, Ulysseas, Pac Rim and Luxtona have also confirmed the principle that the burden of proof falls on the denying state.109 Others have not addressed the matter expressly, although the manner in which the tribunals phrase their findings on the DoB objection might suggest alignment with this same principle.110 A few decisions have offered some alternative views, however. For instance, Plama found that the burden of proof on the question of the claimant’s owner or controller fell on the claimant;111 a conclusion that appears to have been informed by the holding that the application of the DoB provision in the ECT was not a matter of jurisdiction.112 In Guaracachi the tribunal made no statement of principle on the question of burden, but its reasoning seems to suggest that the burden on the substantial business activities element was placed on the claimant.113

102 Generation

Ukraine, above (n 15) para 15.7. The tribunal ultimately dismissed the DoB objection on grounds that the state was a ‘long way from displacing’ the clear manifestation of control of the investor by a national of the home state (US). ibid para 15.9. 104 AMTO, above (n 63) para 64. 105 ibid. 106 ibid paras 26(h), 59–60. 107 ibid para 65. 108 ibid. 109 Liman, above (n 64) para 164; Ulysseas, above (n 13) para 166; Luxtona, above (n 6) paras 284, 300–301, 304; Pac Rim, above (n 6) paras 4.60, 4.92. 110 See, eg, Masdar Solar & Wind Cooperatief U.A. v Kingdom of Spain, Award (16 May 2018) ICSID Case No ARB/14/1 [Masdar], para 254 (dismissing the DoB objection, on the ground that the ‘Respondent […] failed to demonstrate the Claimant has no substantial business activity in the Netherlands’); 9REN Holding S.À.R.L. v Kingdom of Spain, Award (31 May 2019) ICSID Case No ARB/15/15 [9REN], para 182 (dismissing the DoB objection stating that the respondent had ‘failed to establish that 9REN lacks substantial business activities in Luxembourg’). 111 Plama Award, above (n 36) paras 82, 89. 112 ibid para 89. While the Plama approach to the nature of the DoB objection and the procedural requirements of the DoB provision under the ECT has broadly been followed in subsequent ECT cases, the same cannot be said of its position on the burden of proof. See eg: Luxtona, above (n 6) paras 239, 300. The authors are not aware of any decision in the ECT context following Plama on the matter of burden of proof. 113 See Guaracachi, above (n 84) para 370 (holding that the ‘no substantial business activities’ prong was met because ‘[i]nsufficient evidence ha[d] been provided to prove that [the claimant] carried on substantial business activities in the US at any point in time.’) 103 ibid.

410 Paul Jean Le Cannu and Luisa F. Torres Finally, in Bridgestone, the dispute concerned whether the claimant subject of the putative denial of benefits had ‘no substantial business activities’ in the US.114 In line with the majority of cases, the tribunal held that the overall burden was on the state ‘as the Party asserting an entitlement to deny benefits’.115 It added, however, that given that the element at issue entailed a requirement ‘to prove a negative in relation to matters that fall essentially within the knowledge of the Claimants’, ‘the evidential burden [was] readily shifted’.116 By contrast with Plama, this shifting of the evidentiary burden to claimant was not premised on the nature of the DoB objection.117 Rather, the Bridgestone tribunal referred to the nature of the element to be proven (a negative) and the claimant’s knowledge of the evidence, contrasting in this latter respect with AMTO.118 In the end however, the Bridgestone tribunal observed that the question of burden of proof was ‘immaterial’ given the ‘ample material’ before it to reach a decision,119 and held that the claimant had substantial business activities in the US.120

B. Lack of Substantial Business Activities DoB clauses commonly provide that a host state may exercise the denial against an investor who has no ‘substantial business activities’ in the territory of its home state.121 With varying degrees of depth, this element has been analysed in at least 10 identified decisions.122 In some other cases however, the issue has not been fully analysed or ruled upon for a variety of reasons, including that: (i) the investor conceded its lack of substantial business activities in the home state;123 (ii) the DoB objection was dismissed on grounds of lack of compliance with another substantive element;124 (iii) the DoB objection was dismissed on other grounds, such as untimeliness,125 or the nature of the treaty advantages impacted;126 or (iv) the specific DoB provision at issue did not include a lack of substantial business activities requirement.127

114 Bridgestone, 115 Bridgestone,

above (n 89) para 287. See US-Panama TPA, above (n 88) Art 10.12. above (n 89) para 289.

116 ibid. 117 The Bridgestone tribunal had accepted that the DoB objection addressed a matter that ‘goes to competence.’ ibid para 288. 118 See AMTO, above (n 63) para 65. 119 Bridgestone, above (n 89) para 289. 120 ibid para 302. 121 See above (n 95). 122 Petrobart, above (n 80) 63; Pan American, above (n 69) para 221; AMTO, above (n 63) paras 68–70; Pac Rim, above (n 6) paras 4.63–4.78; Guaracachi, above (n 84) paras 366–370; Bridgestone, above (n 89) paras 290–302; Masdar, above (n 110) paras 252–256; NextEra, above (n 7) paras 253–261; 9REN, above (n 110) para 182; Gran Colombia, above (n 7) paras 136–141. 123 See, eg, Plama Decision, above (n 12) paras 168–169, 240; Plama Award, above (n 36) para 81; Hulley, above (n 60) para 460; Veteran, above (n 60) para 517; Yukos, above (n 60) para 461; Khan, above (n 62) para 415; Luxtona, above (n 6) paras 227, 286. See also: Libananco, above (n 98) paras 104(g), 551 (referring to respondent’s invocation of Art 17(1) of the ECT arguing inter alia that ‘Libananco […] conducts no business in Cyprus,’ but later stating that there was not ‘any serious difference between the Parties as to the nature of Libananco’s business activities in Cyprus.’) 124 See, eg, Generation Ukraine, above (n 15) para 15.9; Ulysseas, above (n 13) paras 168, 189–190. 125 See, eg, NextEra, above (n 7) paras 261, 270; Stati, above (n 65) para 745; Isolux, above (n 7) paras 715–716. 126 See, eg, Stati, above (n 65) para 745; Isolux, above (n 7) paras 712–714. 127 See Ampal, above (n 6) para 88 (referring US-Egypt BIT, Protocol, para 1).

Denial of Benefits Clause 411 By its nature, the analysis of the ‘substantial business activities’ requirement entails a highly fact-specific inquiry, as a survey of the arbitral decisions available on the matter confirms. In at least one of its non-disputing party interventions in cases involving a US treaty, the US observed that ‘its long practice is to omit a precise definition’ of the term ‘substantial business activities’, so that the existence of such activities is ‘evaluated on case-by-case basis’.128 In the context of addressing this first and foremost fact-specific issue however, some tribunals have proposed a few guiding principles that have contributed to shaping the debate on the subject, as discussed below. i. The Initial Cases on the Substantial Business Activities Prong The substantial business activities requirement appears to have been addressed for the first time in Petrobart and Pan American, albeit only in a succinct fashion.129 In Petrobart, an ECT case, the tribunal concluded that the claimant (a company incorporated in Gibraltar) had ‘substantial business activities in the Area of a Contracting Party i.e. the United Kingdom’ because it was managed by another company registered in England and with its principal office in London.130 In Pan American, the tribunal dismissed a DoB objection under the US-Argentina BIT,131 finding that the claimant targeted by the denial, a Delaware company, was controlled by two US companies with ‘substantial business contacts in the United States’.132 However the decision offers no further insight as to what those ‘substantial business contacts’ consisted of. The first two decisions to engage in some depth in the analysis of lack of substantial business activities prong were AMTO and Pac Rim. Both decisions are often discussed by commentators on this subject,133 and it is not uncommon for their holdings to feature in parties’ arguments or in the analysis of subsequent tribunals on this subject.134 128 See,

eg, Bridgestone, above (n 89) para 283. above (n 80) 63; Pan American, above (n 69) para 221. the ECT at issue in Petrobart, refers to a legal entity with ‘no substantial business activities in the Area of the Contracting Party in which it is organised’. The respondent had argued that the ‘alleged presence of Petrobart in […] London would not place its principals in the “Area of the Contracting Party in which [Petrobart] is organized,” […], because Petrobart is, […] incorporated and registered in Gibraltar’ and ‘Gibraltar is not a component country of the United Kingdom but a crown colony that looks to the United Kingdom for matters of defence and international relations only. Gibraltar registers its own legal entities.’ ibid 42. The tribunal appears to have considered that the Area of the UK as a whole was relevant to the substantial business activities analysis, but it did not elaborate on the point. ibid 63. The parties in the case had debated whether the ECT applied to Gibraltar. The discussion centred on the fact that, at the time of signature of the ECT, the UK had included Gibraltar in its declaration concerning provisional application of the ECT, but it did not do so at the time of ratification. The tribunal found that the ECT continued to apply provisionally to Gibraltar. ibid 62–63. 131 Pan American, above (n 69) paras 182, 204, 221. See: US-Argentina BIT, above (n 96) Art I (2). Although the decision initially indicates that the objection concerned lack of substantial activities in the host state (Argentina), it later states that the matter concerned lack of business activities in the investor’s home state (US). cf Pan American, above (n 69) paras 182, 221. 132 Pan American, above (n 69) paras 1, 221. 133 See, eg, Y Banifatemi, ‘Taking Into Account Control Under Denial of Benefits Clauses’ in Y Banifatemi (ed), Jurisdiction in Investment Treaty Arbitration (Paris, International Arbitration Institute, 2018) 235–237; Mistelis and Baltag, above (n 1) 34–35. 134 For references to AMTO, see, eg, Masdar, above (n 110) paras 223 (summarising claimant’s argument), 253 (tribunal’s discussion); NextEra, above (n 7) paras 255, 257–258 (tribunal’s discussion); 9REN, above (n 110) para 181 (summarising claimant’s argument); Guaracachi, above (n 84) para 217 (summarising claimant’s argument); Bridgestone, above (n 89) para 271 (summarising claimants’ arguments). For references to Pac Rim, see, eg, NextEra, above (n 7) paras 254, 256 (tribunal’s discussion); Guaracachi, above (n 84) para 215 and fn 196 (summarizing respondent’s argument); Bridgestone, above (n 89) paras 252–253 and fn 306 (summarising respondent’s argument), para 271 (summarising claimants’ arguments). 129 Petrobart,

130 Petrobart, above (n 80) 4, 63. Article 17(1) of

412 Paul Jean Le Cannu and Luisa F. Torres a. AMTO In AMTO the question arose as to whether the investor had ‘no substantial business activities’ in its home state (Latvia)135 for purposes of application of the DoB clause Article 17(1) of the ECT; and the tribunal ultimately concluded that the requirement was not met.136 Noting that the ECT contained no definition of the term ‘substantial’, the tribunal turned to the purpose of Article 17(1) of the ECT to establish its meaning, concluding that such purpose consisted of excluding from protection investors that had adopted a ‘nationality of convenience’.137 The tribunal went on to hold that: ‘[…] “substantial” in this context means “of substance, and not merely of form.” It does not mean “large,” and the materiality not the magnitude of the business activity is the decisive question.’138 This is perhaps one of the most recognisable and often cited holdings of this decision on this issue.139 On the facts, the claimant had argued that its main business activity consisted of acting as an ‘investment company’,140 and that its business activities in Latvia were substantial. In particular, the claimant relied on: (i) activities in the ‘field of financial investments by participating as a shareholder in companies in Finland, Ukraine, USA and Russia’; (ii) a ‘project on real estate acquisition in Riga’ for which the tribunal only found a reference to a ‘preliminary agreement’; (iii) tax payments for a period of seven years; (iv) employment of two full-time staff; (v) holding of a multi-currency account in a Latvian bank; and (vi) the rental of an office in Latvia for the past seven years.141 The tribunal was persuaded that the claimant’s ‘investment related activities conducted from premises in Latvia, and involving the employment of a small but permanent staff ’ constituted ‘substantial business activity in Latvia’.142 As a result, the state was prevented from denying to the claimant the advantages of Part III of the ECT.143 This finding alone was dispositive of the objection as a whole.144 b. Pac Rim Some years later, the Pac Rim tribunal analysed the non-substantial business activities prong of Article 10.12.2 of CAFTA. Pac Rim Cayman LLC, the claimant, was a legal person organised under the laws of the State of Nevada, US.145 By contrast with AMTO, Pac Rim concluded that the claimant did not have substantial business activities in its home state (US).146 The tribunal acknowledged that the ‘group of companies’ to which the claimant belonged had substantial business activities in the US.147 In this regard, it referred to evidence to the 135 AMTO,

above (n 63) paras 1, 62, 68. para 69. 137 ibid. 138 ibid. 139 See, eg, Masdar, above (n 110) paras 223 (summarising claimant’s argument), 253 (tribunal’s discussion); NextEra, above (n 7) paras 255, 257–258 (tribunal’s discussion); 9REN, above (n 110) para 181 (summarising claimant’s argument); Guaracachi, above (n 84) para 217 (summarising claimant’s argument); Bridgestone, above (n 89) para 271 (summarising claimants’ arguments). 140 AMTO, above (n 63) para 16. 141 ibid para 68. See also: ibid para 16. 142 ibid para 69. 143 ibid. 144 ibid. The tribunal did not rule on the question of compliance with the ‘third State’ control requirement, or the timing elements. ibid. 145 Pac Rim, above (n 6) para 1.1. The claimant was wholly owned by Pacific Rim Mining Corporation, a legal person organised under the laws of Canada. ibid. 146 ibid para 4.78. 147 ibid para 4.63. 136 ibid

Denial of Benefits Clause 413 effect that: (i) the CEO of the claimant’s parent company maintained his offices in Nevada, US, from where he managed the claimant and other companies in the group; and (ii) other group subsidiaries were based in Nevada, US including one that had provided a substantial portion of the capital invested in the host state (Salvador).148 That said, the tribunal explained that the substantial business activities element in Article 10.12.2 of CAFTA, did not refer to ‘the collective activities of a group of companies, but to activities attributable to the “enterprise” itself, […]’.149 Thus, for purposes of Article 10.12.2, the claimant could not ‘aggregate to itself the separate activities of other natural or legal persons […]’.150 While the tribunal recognised that the claimant had ‘certain’ activities in the US, it found that it was only a ‘passive actor’ in the US.151 In so concluding, the tribunal relied on witness evidence to the effect that the claimant was a ‘holding company’ with: (i) no employees; (ii) no office space lease; (iii) no bank account; (iv) no board of directors – only two managers; and (v) no physical existence other than its registration documents.152 The tribunal remarked, however, that its conclusion did not amount to a general finding that a ‘traditional holding company could never’ satisfy the element of ‘substantial business activities’ for purposes of Article 10.12.2 of CAFTA.153 The tribunal’s view was that, in this specific case, the claimant did not have the characteristics of a traditional holding company.154 It explained: Generally, such holding companies are passive, owning all or substantially all of the shares in one or more subsidiary companies which will employ personnel and produce goods or services to third parties. The commercial purpose of a holding company is to own shares in its group of companies, with attendant benefits as to control, taxation and risk-management for the holding company’s group of companies. It will usually have a board of directors, board minutes, a continuous physical presence and a bank account.155

As further support for its conclusion, the tribunal relied on the fact that the claimant’s activities ‘principally’ consisted of holding ‘the shares of its subsidiaries in El Salvador’.156 It observed obiter dicta that the ‘position might arguably be different if it was acting as a traditional holding company owning shares in subsidiaries doing business in the USA’, but did not expand further on this.157

148 ibid 149 ibid

para 4.65. para 4.66.

150 ibid. Plama had made a similar statement in 2005. See, Plama Decision, above (n 12) para 169 (stating that a short-

fall on business activities ‘cannot be made good with business activities undertaken by an associated but different legal entity’). However, the statement was of less impact in Plama, in light of claimant’s concession that the no-substantial business activities prong was met. ibid paras 168–169. Most recently, the tribunal in Gran Colombia has also confirmed the view that the relevant inquiry is ‘whether the Claimant entity – viewed on its own – has any substantial business activities in [the home State], in any business sector’. Gran Colombia, above (n 7) para 140 (emphasis added). Other cases appear to suggest a different view. Thus, in Pan American, the tribunal dismissed a DoB objection arising under Art I(2) of the US-Argentina BIT, holding that the claimant, PAE, was ‘controlled by BP America and BP Argentina, which are both US companies and have both substantial business contacts in the United States’, seemingly suggesting that the activities of the controlling companies were determinative. Pan American, above (n 69) para 221 (emphasis added). 151 Pac Rim, above (n 6) paras 4.67–4.68. 152 ibid paras 4.69–4.70. 153 ibid para 4.72. 154 ibid paras 4.73, 4.75. 155 ibid para 4.72. 156 ibid para 4.74. 157 ibid.

414 Paul Jean Le Cannu and Luisa F. Torres Also relevant to the Pac Rim tribunal’s conclusion was its finding that, despite a change of nationality in December 2007, there had not been ‘any material difference between the Claimant’s activities as a company established in the Cayman Islands and its later activities as a company established in the USA’, which for the tribunal remained ‘equally unsubstantial’.158 The tribunal observed that the claimant could not claim geographical activities in the territory of the US when such activities were not different from those it had as a Cayman Islands company.159 Ultimately, the tribunal concluded that the claimant was ‘not a traditional holding company actively holding shares in subsidiaries but more akin to a shell company with no geographical location for its nominal, passive, limited and insubstantial activities’.160 The distinction Pac Rim drew between a ‘traditional holding company’ and a ‘shell company’ commonly features in parties’ arguments or tribunal’s discussions when addressing the matter of ‘substantial business activities’.161 ii. The Substantial Business Activities Prong in Recent Cases After AMTO and Pac Rim, the body of case law on the lack of substantial business activities prong has continued to grow. In the last seven years, with varying degrees of depth, at least six additional decisions (Guaracachi, Bridgestone, Masdar, NextEra, 9REN and most recently, Gran Colombia) have devoted attention to this subject. Only one of these (Guaracachi) has upheld the DoB objection, concluding that the lack of substantial business activities requirement was met.162 Four other cases (Bridgestone, Masdar, 9REN and Gran Colombia) reached the opposite result.163 And one (NextEra) ultimately left the question of substantial business activities open, although it provided some discussion on the matter.164 Guaracachi, Masdar and 9REN are succinct in their treatment of this requirement. Both Masdar and 9REN addressed the issue under Article 17(1) of the ECT,165 and in both cases the finding on the substantial business activities prong was dispositive of the overall DoB objection.166 By contrast, Guaracachi, addressed the issue under Article XII of the US-Bolivia BIT, and it concluded that there was ‘[i]nsufficient evidence’ of the claimant’s substantial business activities in its home state (the US).167

158 ibid

para 4.73. para 4.74. para 4.75. 161 See, eg, NextEra, above (n 7) paras 254, 256 (tribunal’s discussion); Guaracachi, above (n 84) para 215 and fn 196 (summarising respondent’s argument); Bridgestone, above (n 89) paras 252–253 and fn 306 (summarising respondent’s argument), para 271 (summarising claimants’ arguments). 162 Guaracachi, above (n 84) paras 370, 384. 163 Bridgestone, above (n 89) para 302; Masdar, above (n 110) paras 251–256; 9REN, above (n 110) para 182; Gran Colombia, above (n 7) paras 139–141. 164 NextEra, above (n 7) paras 261, 270. The tribunal considered that the tribunal’s conclusion that the state had not exercised its right to deny benefits in a timely fashion obviated the need to decide on the substantial business activities prong. ibid. 165 See Masdar, above (n 110) paras 203, 251–256; 9REN, above (n 110) paras 174–175, 182. 166 See Masdar, above (n 110) paras 252–256; 9REN, above (n 110) para 182. 167 Guaracachi, above (n 84) paras 1, 366, 370. The tribunal dismissed the claimant’s contention that the claimant ‘conducted substantial commercial activities in the United States, since it maintains offices in said territory, holds shareholders’ meetings in Ohio as well as Board of Directors’ meetings, prepares the minutes of said meetings, etc’. ibid para 217. The tribunal also dismissed a separate, but related allegation that the exercise of the DoB would be ‘unfair’, given that it had been the state who had required the establishment of claimant as a special purpose vehicle, only allowing it to hold shares. ibid paras 214, 369. 159 ibid 160 ibid

Denial of Benefits Clause 415 Bridgestone addressed the substantial business activities prong in the context of Article 10.12.2 of the US-Panama TPA,168 at some additional length. Here, the target of the purported denial of benefits was a company incorporated in Delaware, US, one of the two claimants in this case.169 The tribunal did not attempt a definition of the notion of substantial business activities. Instead, it observed that whether the term ‘substantial’ was understood as ‘activities of substance’ or ‘significant activities’, the claimant carried out those activities in the US, and, after a detailed factual discussion, it dismissed the DoB objection.170 This said, the tribunal did disagree with the respondent’s submission of principle that business activities ‘will not be substantial unless they are money-making’.171 Gran Colombia offers the most recent iteration of the debate around the prong of substantial business activities. Although it remains to be seen whether and to what extent its contribution on this subject will have a lasting impact on future cases, it is perhaps one of the cases that has delved with greater depth into the analysis of the ‘substantial business activities’ prong to date, and for that reason it is worthy of separate note. a. Gran Colombia In Gran Colombia the issue arose in the context of Colombia’s invocation of the DoB clause in Article 814(2) of the Canada-Colombia FTA.172 The claimant was a company incorporated in British Columbia, Canada,173 and as such the inquiry concerned its business activities in Canada. As in Bridgestone, the tribunal’s conclusion on the ‘substantial business activities’ prong was dispositive of the DoB objection.174 Before addressing the facts, the Gran Colombia tribunal made a number of preliminary observations concerning the interpretation of the ‘no substantial business activities’ requirement in the treaty. Most notably, it held: First, that the treaty did not require a comparative analysis of the claimant’s activities in various jurisdictions in search of the ‘dominant’ connection; the inquiry was not whether the claimant had the ‘most’ substantial connections to its home state, but whether the claimant had ‘some substantial business activity’ in its home state.175 Second, that the word ‘substantial’ provided an ‘important materiality threshold’.176 More particularly: A business activity may not be mere[ly] cursory, fleeting or incidental, but must be of sufficient extent and meaning as to constitute a genuine connection by the company to its home State. […] The connection between the company and its home State cannot be merely a sham, with no business reality whatsoever, other than an objective of maintaining its own corporate existence.177

Third, that the term ‘business activities’ required activities of ‘a “business” nature’ in the home state, a requirement not met if the company had ‘no activities in its home jurisdiction 168 Bridgestone,

above (n 89) paras 286–287. See US-Panama TPA, above (n 88) Art 10.12. above (n 89) para 2. para 302. For the tribunal’s discussion of the various elements of fact underlying this conclusion, see, eg, ibid paras 292–302. The substantial activities prong was the dispositive question, as the element of ownership by a person of a non-Party was uncontroversial. ibid para 287. 171 ibid para 300. 172 Gran Colombia, above (n 7) para 47. See Canada-Colombia FTA, above (n 90) Art 814. 173 Gran Colombia, above (n 7) para 2. 174 ibid paras 135, 141. 175 ibid para 136 (emphasis in original). 176 ibid para 137. 177 ibid. 169 Bridgestone, 170 ibid

416 Paul Jean Le Cannu and Luisa F. Torres other than those required to maintain its bare registration’.178 However as the treaty did not limit the ‘nature of th[e] business’, the company’s activities in its home state could be of a different nature from those of that same company in other jurisdictions.179 On the facts, the tribunal concluded that the claimant satisfied the threshold of ‘substantial business activities’ in its home state (Canada).180 It did so relying on the following factors: (i) ‘core corporate functions’ in Canada, including ‘corporate finance, fundraising, accounting, shareholder relations, legal, administration and IT support’; (ii) an office space in Canada, including rental and utility payments and related expenses; (iii) eight full time employees in Canada, including payments in Canada on compensation and benefits; (iv) several (six) bank accounts in Canada, through which the claimant conducted its business; (v) annual purchases of goods and services in Canada, in ‘accounting and advisory services, legal services, and shareholder and investor related activities, as well as miscellaneous services such as IT, liability policies and a listing fee for the Toronto Stock Exchange’; and (vi) financing activities by way of fund raising ‘transactions on the Canadian debt and equity markets, in order to support its operations’.181 For the tribunal, the above mentioned activities were ‘substantial’, as they had the ‘depth and materiality to demonstrate a genuine and meaningful connection (and contribution) to Canada’.182 They qualified as ‘business’ even though ‘the business in question in Canada appears […] to be essentially investment management and financing, to support a different type of business (mining operations)’.183 The tribunal remarked that the treaty simply required that ‘any substantial business activities in Canada, in any business sector’.184

C. Ownership or Control by a National of a Third State The second condition that has been most commonly at issue concerns the question of ownership or control of the investor targeted by the denial of benefits. With different degrees of depth, this requirement has been addressed in at least 15 publicly known awards or decisions to date.185 The decisions most frequently deal with the condition that the claimant be owned or controlled by a national of a ‘third country,’ ‘third State’ or ‘non-Party’.186 There does not seem to have been any arbitral decision dealing with the exercise of a DoB provision targeting an entity owned or controlled by nationals of the host/denying state.187 178 ibid. 179 ibid

para 138. paras 139–140. 181 ibid para 139. 182 ibid para 140. 183 ibid. 184 ibid (emphasis in original). 185 See Generation Ukraine, above (n 15) paras 15.7–15.9; Plama Decision, above (n 12) paras 170–178; Petrobart, above (n 80) 63; Pan American, above (n 69) para 221; AMTO, above (n 63) para 66; Plama Award, above (n 36) paras 79–95; Hulley, above (n 60) paras 459–536; Veteran, above (n 60) paras 516–548; Yukos, above (n 60) paras 460–537; Ulysseas, above (n 13) paras 164–165, 190; Libananco, above (n 98) paras 551, 553–556; Pac Rim, above (n 6) paras 4.79–4.82; Guaracachi, above (n 84) para 370; Luxtona, above (n 6) paras 284, 301–314; NextEra, above (n 7) paras 250–252. 186 See: section i below. 187 In some ECT cases, the respondent has argued (and the tribunals have dismissed) that Art 17(1) of the ECT applies a fortiori in situations of control by host state nationals. See, eg, Yukos, above (n 60) para 542; Hulley, above (n 60) para 541; Veteran, above (n 60) para 553. When addressing ratione personae jurisdictional objections on grounds that the claimant was controlled by nationals of the host state, other tribunals have referred to the lack of a denial of benefits provision in the treaty, or to the lack of the host state control element in the DoB clause in the treaty, 180 ibid

Denial of Benefits Clause 417 As the concepts of ‘ownership’ and ‘control’ employed in DoB clauses are often left undefined in the underlying treaties, it has generally fallen on tribunals to interpret them and apply them to the facts before them. Unsurprisingly, here too, the survey of the arbitral decisions reveals that the analysis of the ownership or control element regularly involves a fact-specific inquiry. For example, in Luxtona the tribunal observed that the assessment of control involves an examination of the ‘actual circumstances in each situation’ and consideration of ‘all relevant factors’.188 The interpretation of the notion of ‘third State’ has also been debated in a few cases.189 In the context of this fact-specific inquiry, it is again difficult to point to a single decision as predominantly influential in shaping the debate. It is also the case, however, that some decisions, in particular earlier ones, have delved somewhat more deeply into the analysis of the third state ownership or control prong, and in that sense have contributed to a better understanding of this requirement for the invocation of a DoB clause. i. Ownership or Control The control element of a DoB clause appears to have been addressed for the first time, albeit briefly, in Generation Ukraine, under the US-Ukraine BIT.190 The claimant was a US company wholly-owned by a natural person of US nationality,191 yet Ukraine argued that the claimant’s control rested in Canada.192 The tribunal dismissed the allegation observing that the facts relied upon by Ukraine were ‘a long way from displacing the clear manifestation of control by a US national […] who owns 100% of the share capital of the Claimant’.193 Thereafter, Plama seems to be the first case to have considered in some detail the element of ‘third state own[ership] or control’, in Article 17(1) of the ECT.194 While the Plama tribunal deferred the conclusion on compliance with this element to the merits phase,195 in its Decision on Jurisdiction it did make certain preliminary observations of principle relevant to the interpretation of the control prong. Thus, the tribunal: (i) addressed the meaning of the words ‘own or control’, holding that ‘ownership includes indirect and beneficial ownership; and control includes control in fact, including an ability to exercise substantial influence over the legal entity’s management, operation and the selection of members of its board of directors or any other managing body’;196 (ii) confirmed that third state ‘ownership’ or ‘control’ were alternatives, such that only one needed to be satisfied; and (iii) observed that ‘third state’ referred to a non-Contracting State to the ECT.197

as a relevant factor to their analysis. See, eg, Tokios Tokèles v Ukraine, Decision on Jurisdiction (29 April 2004) ICSID Case No ARB/02/18 [Tokios Tokèles], paras 21, 35–36; Charanne B.V. and Construction Investments S.à.r.l. v Kingdom of Spain, Award (21 January 2016) SCC Case No V2012/062 [Charanne], paras 411, 415–416. 188 Luxtona, above (n 6) para 305. 189 See: section ii below. 190 Generation Ukraine, above (n 15) para 1.2. US-Ukraine BIT, above (n 51) Art I(2). 191 Generation Ukraine, above (n 15) para 1.1. 192 ibid para 15.8. 193 ibid para 15.9. The respondent had argued that control rested in Canada on the ground that the claimant: (i) had once been assisted by a Canadian ambassador; (ii) had been presided by a Canadian national at some point; (iii) had an office and bank account in Canada; and (iv) identified itself in its letterhead as a ‘United States, Canada, Ukraine venture.’ ibid para 15.8. 194 Plama Decision, above (n 12) paras 170–178. See ECT, above (n 3) Art 17(1). 195 Plama Decision, above (n 12) paras 177–178. 196 This interpretation appeared to have been ‘common ground’ between the Parties. Plama Decision, above (n 12) para 170. See also Plama Award, above (n 36) para 80. 197 Plama Decision, above (n 12) para 170.

418 Paul Jean Le Cannu and Luisa F. Torres The tribunal went on to analyse, at the merits stage, whether this limb of Article 17(1) was met on the facts of the case. While observing that Article 17(1) requires that ‘citizens or nationals of a third state own or control’ the claimant,198 the tribunal framed the question before it as ‘whether Claimant is a legal entity owned or controlled by citizens or nationals of a State Party to the ECT’.199 As previously noted, the Plama tribunal held that the burden of proof on the question so formulated fell on the claimant;200 and ultimately concluded that it had been established that the claimant was indirectly owned and controlled by a French national.201 France being an ECT Contracting Party, the tribunal ruled that the right to deny benefits under Article 17(1) could not be exercised against the claimant.202 After the inaugural analysis in Plama, a number of additional decisions (Petrobart, Hulley, Veteran, Yukos, AMTO, NextEra and Luxtona) have also dealt with the element of ‘third state own[ership] or control’ in Article 17(1) of the ECT. Petrobart offers little insight into the analysis of the question of ownership and control.203 Other tribunals have offered discussions that, for the most part, have revolved around the factual specificities of each case.204 For example, in Luxtona, while the tribunal had found that the purported denial of benefits did not have retrospective effect,205 it still analysed the control prong, ultimately finding that the state had failed to establish that nationals of a state not party to the ECT controlled the claimant.206 The claimant, a company incorporated in Cyprus, was indirectly owned by a Dutch entity, and the issue was whether that Dutch entity was controlled by US nationals.207 In particular, whether the US nationality of four board members of the Dutch entity supported the conclusion that said entity, and ultimately, Luxtona, were under US control.208 Relying on specificities of Dutch law as to the governance of the type of entity in question, the tribunal concluded that the nationality of the board members had no bearing on the issue.209 The tribunal added that the purpose Article 17(1) required it to analyse the question of control making a ‘realistic assessment of the economic realities’,210 which were that the US board members were barred from deriving economic benefits from their role – other than their compensation as directors – and that an award would benefit the Dutch entity, not the individual board members.211

198 Plama

Award, above (n 36) para 79. See also Plama Decision, above (n 12) para 170. Award, above (n 36) paras 77, 82; Plama Decision, above (n 12) para 171. 200 Plama Award, above (n 36) paras 82, 89. 201 ibid paras 91–95. The tribunal did add, however, that ‘without losing sight of the fact that Claimant bears the burden of proof on this issue’, the tribunal had found that the state’s attempt to ‘cast doubt’ on this French national’s ownership and control was not convincing, and the respondent had provided no cogent evidence as to who owned or controlled the claimant other than this French national. ibid para 94. 202 Plama Award, above (n 36) para 95. 203 Petrobart, above (n 80) 4, 63 (holding that claimant, a Gibraltar company, had successfully contradicted the allegation that it was ‘a company owned or controlled by citizens or nationals of a state other than the United Kingdom’.) 204 See, eg, AMTO, above (n 63) paras 1, 66–67, 70; Hulley, above (n 60) paras 460–536; Veteran, above (n 60) paras 517–548; Yukos, above (n 60) paras 461–537; NextEra, above (n 7) paras 2, 250–252. 205 Luxtona, above (n 6) para 282. 206 ibid paras 284, 304, 314. 207 ibid paras 27, 286. There was no contention that third party nationals owned the claimant. ibid para 301. 208 ibid para 307. 209 ibid para 311. 210 ibid para 312. 211 ibid. 199 Plama

Denial of Benefits Clause 419 In addition to their factual analysis, some tribunals have contributed certain general remarks on interpretation of the third country control element of Article 17(1). For example, in NextEra the tribunal dismissed the contention that only direct ownership or control were relevant for purposes of Article 17(1), and held that the relevant inquiry was ‘control in fact’.212 In turn, the Luxtona tribunal rejected the submission that Understanding 3 of Article 1(6) of the ECT (addressing the reference to control in the context of the definition of ‘Investment’) was determinative of the meaning of ‘control’ in the context of Article 17(1), yet it accepted that some of its elements provided useful insight, in particular, that assessment of control involves an examination of the ‘actual circumstances in each situation’ and consideration of ‘all relevant factors’.213 It also disagreed that, for purposes of assessing ownership or control, Article 17(1) of the ECT compelled the finding of the natural persons at the top of the structure, by contrast with the decision in Ulysseas (under a different treaty) discussed further below.214 Aside from the ECT context, the ‘third country’ ownership or control element has also been at issue in a number of decisions under a variety of US treaties. Thus, after the first decision in Generation Ukraine, the Pan American, Ulysseas, Pac Rim and Guaracachi tribunals also addressed the issue. Only in the last two was it found that the third country ownership or control element was met. Pan American and Guaracachi addressed the matter succinctly.215 The issue in Pac Rim was also discrete but is worth discussing as it concerned the element of ‘non-Party’ ownership in the DoB provision in Article 10.12.2 of CAFTA.216 It was undisputed that claimant, a US company, was wholly owned by a Canadian company, a non-CAFTA Party for purposes of Article 10.12.2.217 The argument was made however, that because a majority of the shareholders of the Canadian company resided or at least had postal addresses in the US, the claimant was owned (albeit indirectly) by natural and juridical persons of a CAFTA Party.218 The tribunal dismissed the contention, observing that the postal addresses were insufficient to determine the US nationality of the shareholders.219 The tribunal also noted, however, that it was not reaching a conclusion as to whether the nationality of the claimant’s ownership could be established on the basis of indirect ownership.220 The tribunal eventually concluded that the claimant was owned by a national of a non-CAFTA Party (Canada), thereby meeting the requirement of Article 10.12.2 of CAFTA, and did not reach the question of control.221

212 NextEra, above (n 7) para 250 (stating ‘[i]t is control in fact that counts, just as control in fact is what is meant in respect of Art 1(6) (Understanding 3).’) 213 Luxtona, above (n 6) para 305. 214 ibid para 313. 215 Pan American, above (n 69) paras 1, 221 (holding that the third country control element in Art I(2) of the US-Argentina BIT was not met because the claimant, a US company, was controlled by two US companies). See US-Argentina BIT, above (n 96) Art I(2); Guaracachi, above (n 84) para 370 (concluding that the third country control element of Art XII of the US-Bolivia BIT was met as the claimant, a US company, was ‘owned and controlled’ by non-US companies). See US-Bolivia BIT, above (n 86) Art XII. 216 Pac Rim, above (n 6) paras 4.79–4.82. See CAFTA, above (n 75) Art 10.12.2. 217 Pac Rim, above (n 6) paras 1.1, 4.79. 218 ibid para 4.80. 219 ibid para 4.81. 220 ibid. 221 ibid para 4.82. The tribunal had held that, under CAFTA, ownership or control by a non-CAFTA party were alternative requirements. ibid para 4.61.

420 Paul Jean Le Cannu and Luisa F. Torres Ulysseas also offered a detailed analysis of the control element in the context of the DoB provision of Article I(2) of the US-Ecuador BIT. After accepting that control by ‘nationals’ of a ‘third country’ was one of two cumulative conditions for the exercise of the provision before it,222 the tribunal went on to decide whether the claimant, a company incorporated in the US,223 was controlled by a Brazilian national (as the respondent argued), or by a US national (as the claimant contended).224 In so doing, the tribunal made various observations concerning the interpretation of Article I(2) of the BIT, including that: (i) given that the definition of the term ‘national’ of a Party in the BIT referred to a ‘natural person’, only a natural person could be at the top of the chain of control of a company for purposes of Article I(2), and it was necessary to determine the natural person controlling the claimant;225 and that (ii) the relevant date for determining compliance with this condition was the date on which the claimant had claimed the advantages that the state intended to deny, namely the date of the notice of arbitration.226 On the facts, the Ulysseas tribunal ultimately found that the claimant had ‘conclusively proven’ that its control ultimately lay with a natural person national of the US.227 The tribunal considered the issue whether ‘control’ as the legal capacity to direct the actions of a company could be exercised through contractual rights, not only through ownership.228 In the respondent’s submission, the claimant was subject to Brazilian ‘legal control’, because by virtue of a Joint Venture Agreement (JVA) between one of the claimant’s parent companies and a Bahamian company, the claimant’s conduct with respect to two power barges was beholden to the consent of the Bahamian company, in turn controlled by a Brazilian national.229 The tribunal disagreed, reasoning that: (i) the JVA did not afford the Bahamian company control over the entirety of the claimant’s affairs, and impacted, at most, only one line of business;230 and (ii) the nationality of the alleged Brazilian person had not been proven.231 ii. Third State A few decisions have also addressed the notion of ‘third state’ in the context of the ownership or control prong of a DoB provision. Plama first held that, for purposes of Article 17(1) of the ECT, ‘third state’ referred to a non-Contracting State to the ECT.232 Since then, some other tribunals have addressed issues around the notion of ‘third state’ in Article 17 of the ECT, reaching conclusions consistent with Plama. In Yukos, Hulley and Veteran, the tribunals addressed obiter dicta whether control residing in Russian nationals amounted to control by nationals of a ‘third state’ for purposes of Article 17 of the ECT.233 Russia had argued that, while the term was undefined in the treaty, it was 222 Ulysseas, above (n 13) paras 164, 167 (referring to control by third party nationals and lack of substantial business activities). See US-Ecuador BIT, above (n 71) Art I(2). 223 Ulysseas, above (n 13) para 48. 224 ibid paras 114, 119. 225 ibid paras 169–170. 226 ibid para 174. 227 ibid paras 175–179. 228 ibid para 181. 229 ibid paras 182–185. 230 ibid paras 186–188. 231 ibid para 189. 232 Plama Decision, above (n 12) para 170. 233 Yukos, above (n 60) paras 538–546; Hulley, above (n 60) paras 537–545; Veteran, above (n 60) paras 549–557. A similar issue arose but was left open in AMTO. AMTO, above (n 63) paras 67, 70.

Denial of Benefits Clause 421 ‘used in a manner that [did] not exclude the possibility that a third State may be a Contracting Party or signatory (Respondent being the later)’, and it also applied ‘a fortiori’ to nationals of the host state.234 Observing that the ECT ‘clearly distinguishes between a Contracting Party (and a signatory), on the one hand, and a third State, which is a non-Contracting Party, on the other’, the tribunals held that Russia was not a third state for purposes of Article 17.235 A few years later, Libananco also addressed obiter dicta the question whether the words ‘third state’ in Article 17 were ‘capable of including [ECT] Treaty Parties’.236 The tribunal answered in the negative, observing that the term referred to a state not party to the ECT.237

IV. CONCLUSIONS

As shown in this article, a number of decisions, especially some of the early ones such as the Plama or Pac Rim decisions, have offered analyses on several important aspects of the interpretation and application of DoB provisions, which many tribunals have subsequently discussed and/or followed. However, given the multiplicity of issues arising out of the exercise of DoB provisions, the importance and contribution of other decisions, as we have attempted to demonstrate, cannot be understated. With the increased popularity of DoB clauses, including in so-called mega-regional treaties,238 DoB provisions may well take on an increasingly ‘significant role in providing clearer guidance to companies regarding the level of connection they must have to their respective home States in order to be assured of investment treaty protections’.239 So will tribunals whose interpretation and application of these provisions are also key to building this clear and predictable framework.

ANNEX A: TABLE OF CASES Date 1.

2003

234 Yukos,

Case Name CCL v Republic of Kazakhstan, SCC Case No 122/2001, Award on Jurisdiction

Tribunal Chair unnamed

Treaty US-Kazakhstan BIT240

Carter Söderlund

above (n 60) para 542; Hulley, above (n 60) para 541; Veteran, above (n 60) para 553. above (n 60) paras 544–546; Hulley, above (n 60) paras 543–545; Veteran, above (n 60) paras 555–557. 236 Libananco, above (n 98) paras 551, 555. The tribunal had already concluded that it lacked jurisdiction on the ground that the claimant had failed to establish ownership over the alleged investment at the relevant time. ibid paras 537–538. 237 ibid para 553. 238 See, eg, Comprehensive Economic Trade Agreement [CETA], signed on 30 October 2016 and entered into force provisionally on 21 September 2017, Art 8.16; Comprehensive and Progressive Agreement for Trans-Pacific Partnership [CPTPP], signed on 8 March 2018 and entered into force on 30 December 2018, Art 9.15; Regional Comprehensive Economic Partnership [RCEP], signed on 15 November 2020, Art 10.14. 239 Feldman, above (n 53) 302. 240 Treaty Between United States of America and the Republic of Kazakhstan Concerning Reciprocal Encouragement and Protection of Investment, signed 19 May 1992, in force 12 January 1994 [US-Kazakhstan BIT], Art I(2), https:// perma.cc/D3LX-6BFW. 235 Yukos,

422 Paul Jean Le Cannu and Luisa F. Torres Date 2.

3.

4.

5.

6.

7.

8.

9.

Case Name

Tribunal

16 September Generation Ukraine, Inc. v 2003 Ukraine, ICSID Case No ARB/00/9, Award

Paulsson (P)

8 February 2005

Plama Consortium Limited v Republic of Bulgaria, ICSID Case No ARB/03/24, Decision on Jurisdiction

Salans (P)

Petrobart Limited v Kyrgyz Republic, SCC Case No 126/2003, Award

Danelius (P)

Pan American Energy LLC and BP Argentina Exploration Company v Argentine Republic, ICSID Case No ARB/03/13, Decision on Preliminary Objections

Caflisch (P)

Limited Liability Company AMTO v Ukraine, SCC Case No 080/2005, Award

Cremades (P)

Plama Consortium Limited. Republic of Bulgaria, ICSID Case No ARB/03/24, Award

Salans (P)

Empresa Eléctrica del Ecuador, Inc. (EMELEC) v Republic of Ecuador, ICSID Case No ARB/05/9, Award

Sepúlveda (P)

29 March 2005

27 July 2006

26 March 2008

27 August 2008

2 June 2009

30 November Hulley Enterprises 2009 Limited (Cyprus). Russian Federation, UNCITRAL, PCA Case No AA 226, Interim Award on Jurisdiction and Admissibility

Treaty US-Ukraine BIT241

Salpius Voss ECT242

van den Berg Veeder ECT

Bring Smets US-Argentina BIT243

Stern van den Berg

ECT

Runeland Söderlund ECT

van den Berg Veeder US-Ecuador BIT244

Reisman Rooney Fortier (P)

ECT

Poncet Schwebel

241 Treaty Between United States of America and Ukraine Concerning Reciprocal Encouragement and Protection of Investment, signed 4 March 1994, in force 16 November 1996 [US-Ukraine BIT], Art I(2), https://perma.cc/Z9SH-93WW. 242 Energy Charter Treaty [ECT], Art 17, https://perma.cc/MPQ7-V4M9. 243 Treaty Between United States of America and The Argentine Republic Concerning Reciprocal Encouragement and Protection of Investment, signed 14 November 1991, in force 20 October 1994 [US-Argentina BIT], Art I(2), https://perma.cc/V755-47KD. 244 Treaty Between United States of America and the Republic of Ecuador Concerning Encouragement and Reciprocal Protection of Investment, signed 27 August 1997, in force 11 May 1997, date of termination 18 May 2018 [US-Ecuador BIT], Art I(2), https://perma.cc/KR6B-4F4H.

Denial of Benefits Clause 423 Date 10.

11.

12.

13.

14.

15.

16.

17.

Case Name

Tribunal

30 November Veteran Petroleum 2009 Limited (Cyprus) v Russian Federation, UNCITRAL, PCA Case No AA 228, Interim Award on Jurisdiction and Admissibility

Fortier (P)

30 November Yukos Universal 2009 Limited (Isle of Man) v Russian Federation, UNCITRAL, PCA Case No AA 227, Interim Award on Jurisdiction and Admissibility

Fortier (P)

22 June 2010

Böckstiegel (P)

Liman Caspian Oil B.V. and NCL Dutch Investment B.V. v Republic of Kazakhstan, ICSID Case No ARB/07/14, Excerpts of Award

Schwebel

Schwebel

ECT

Crawford Hobér

2 September 2011

Libananco Holdings Co. Limited v Republic of Turkey, ICSID Case No ARB/06/8, Award

Hwang (P)

Pac Rim Cayman LLC v Republic of El Salvador, ICSID Case No ARB/09/12, Decision on Jurisdiction

Veeder (P)

Khan Resources Inc., Khan Resources B.V., and CAUC Holding Company Ltd. v Government of Mongolia, UNCITRAL, PCA Case No 2011-09, Decision on Jurisdiction

Williams (P)

Anatolie Stati, Gabriel Stati, Ascom Group S.A., and Terra Raf Trans Traiding Ltd. v Republic of Kazakhstan, SCC Case No V2010/116, Award

Böckstiegel (P)

19 December 2013

ECT

Poncet

Bernardini (P)

25 July 2012

ECT

Poncet

28 September Ulysseas, Inc. v Republic 2010 of Ecuador, UNCITRAL, Interim Award

1 June 2012

Treaty

US-Ecuador BIT

Pryles Stern ECT

Álvarez Berman CAFTA245

Stern Tawil ECT

Fortier Hanotiau

ECT

Haigh Lebedev

245 Dominican Republic-Central America-United States Free Trade Agreement, signed 5 August 2004 [CAFTA], Art 10.12.2, https://ustr.gov/trade-agreements/free-trade-agreements/cafta-dr-dominican-republic-central-america-fta/ final-text.

424 Paul Jean Le Cannu and Luisa F. Torres Date 18.

19.

20.

21.

22.

23.

31 January 2014

1 February 2016

12 July 2016

22 March 2017

13 December 2017

16 May 2018

Case Name

Tribunal

Treaty

Guaracachi America, Inc. and Rurelec PLC v Plurinational State of Bolivia, UNCITRAL, PCA Case No 2011-17, Award (with Dissenting Opinion of Co-Arbitrator Manuel Conthe)

Júdice (P)

US-Bolivia BIT246

Conthe

U.K.-Bolivia BIT

Ampal-American Israel Corp., EGI-FUND (08-10) Investors LLC, EGI-Series Investments LLC, BSS-EMG Investors LLC, and Mr. David Fischer v Arab Republic of Egypt, ICSID Case No ARB/12/11, Decision on Jurisdiction

Fortier (P)

Isolux Netherlands, B.V. v Kingdom of Spain, SCC Case No V2013/153, Award and Dissenting Opinion of Professor Guido Santiago Tawil

Derains (P)

Luxtona Limited v Russian Federation, PCA Case No 2014-09, Interim Award on Respondent’s Objections to the Jurisdiction of the Tribunal

Crook (P)

Bridgestone Licensing Services, Inc. and Bridgestone Americas, Inc. v Republic of Panama, ICSID Case No ARB/16/34, Decision on Expedited Objections

Phillips (P)

Masdar Solar & Wind Cooperatief U.A. v. Kingdom of Spain, ICSID Case No. ARB/14/1, Award

Beechey (P)

Vinuesa

US-Egypt BIT247

McLachlan Orrego Vicuña

ECT

Tawil Von Wobeser

ECT

Radicati Oreamuno

US-Panama TPA248

Grigera Naón Thomas

ECT

Born Stern

246 Treaty Between The Government of The United States of America and The Government of The Republic of Bolivia Concerning The Encouragement and Reciprocal Protection of Investment, signed 17 April 1998, in force 6 June 2001, date of termination 10 June 2012 [US-Bolivia BIT], Art XII, https://perma.cc/2XYE-NWB3. 247 Treaty Between The United States of America and The Arab Republic of Egypt Concerning The Reciprocal Encouragement and Protection of Investments, signed 29 September 1982, Protocol 11 March 1986, entered into force 27 June 1992 [US-Egypt BIT], Protocol, para 1, https://perma.cc/5QGG-5U99. 248 US-Panama Trade Promotion Agreement, signed 28 June 2007, entered into force 31 October 2012 [US-Panama TPA], Art 10.12.2, https://ustr.gov/trade-agreements/free-trade-agreements/panama-tpa/final-text.

Denial of Benefits Clause 425 Date 24.

25.

26.

12 March 2019

31 May 2019

Case Name

Tribunal

NextEra Energy Global Holdings B.V. and NextEra Energy Spain Holdings B.V. v Kingdom of Spain, ICSID Case No ARB/14/11, Decision on Jurisdiction, Liability and Quantum Principles

McRae (P)

9REN Holding S.À.R.L. v Kingdom of Spain, ICSID Case No ARB/15/15, Award

Binnie (P)

23 November Gran Colombia Gold Corp. 2020 v Republic of Colombia, ICSID Case No ARB/18/23, Decision on the Bifurcated Jurisdictional Issue

Treaty ECT

Fortier Boisson de Chazournes

ECT

Haigh Veeder Kalicki (P) Hanotiau

Canada-Colombia FTA249

Stern

249 Free Trade Agreement between Canada and the Republic of Colombia, signed 21 November 2008, entered into force 15 August 2011 [Canada-Colombia FTA], Art 814(2), www.international.gc.ca/trade-commerce/tradeagreements-accords-commerciaux/agr-acc/colombia-colombie/fta-ale/index.aspx?lang=eng.

25 Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment GIORGIO SACERDOTI AND NIALL MORAN*

I. THE RATIONALE FOR NON-DISCRIMINATION CLAUSES, FROM TRADE TO INVESTMENT TREATIES

N

ON-DISCRIMINATION CLAUSES IN international investment agreements (IIAs) traditionally ensure equality of treatment between foreign investors and their investments protected under a given IIA, and investors from other countries or local investors of the host state. The clauses guaranteeing this treatment are respectively known as Most-Favoured-Nation (MFN) treatment and National Treatment (NT) clauses.1 Both types of clauses aim to prevent nationality-based discrimination. NT clauses ensure that treaty-protected foreign investments are not discriminated against compared to domestic investors and investments, while the MFN obligation ensures that the treaty’s parties do not discriminate against treaty-protected foreign investors compared to investors from other countries. As a consequence of the extensive number of Bilateral Investment Treaties (BITs) and IIAs in place, their NT clauses ensure that there is no discrimination as to their treatment by host states against foreign investors in comparison to nationals, while the MFN clause ensures non-discrimination, ie by and large equality of treatment, among foreign investments in respect of other IIAs.2 Such clauses in investment treaties are patterned after and derive from the model of bilateral commercial treaties and their treatment of nationals – such as the friendship, commerce and navigation bilateral treaties that were widely concluded in the nineteenth and twentieth centuries. They have been continued in modern BITs and regional trade agreements that include provisions on investment. MFN treatment and national treatment are thus regularly * Giorgio Sacerdoti is Emeritus Professor of International Law at Bocconi University in Milan; Niall Moran is an Assistant Professor in Law at Dublin City University. 1 For a taxonomy of the different standards of protection found in international practice and more specifically spelled out in Bilateral Investment Treaties (BITs) see G Sacerdoti, ‘Bilateral Treaties and Multilateral Instruments on Investment Protection’ (1997) 269 Hague Academy Courses 251. 2 For a general review of issues and investment arbitration case law see A Reinisch, ‘Most Favoured Nation treatment’ in M Bungenberg and others (eds), International Investment Law (Baden-Baden, Beck/Nomos/Hart, 2015). See also UNCTAD, ‘Most-Favoured-Nation treatment’ (2010) UNCTAD Series on Issues in International Investment Agreements II, UNCTAD/DIAE/IA/2010/1.

428 Giorgio Sacerdoti and Niall Moran included in bilateral investment treaties to the point that they can be considered, as in the GATT in relation to trade, ‘cornerstone obligations’.3 The object and purpose of MFN and NT clauses is essentially to provide for non-discrimination between states and can therefore be seen as a reflection of the principle of sovereign equality.4 They aim at preventing state-caused distortions of competitive opportunities in the economic sphere; a concern that lies at the foundation of the notions of comparative advantage, free trade, unimpeded fair competition and economic liberalism.5 Today, MFN clauses are widely found in bilateral and multilateral treaties dealing with the treatment of persons, rights of establishment, trade in goods and services, investments, and the protection of intellectual property rights.6 The subject matter in respect of which MFN treatment is required depends on the text and scope of the relevant treaty.7 The two clauses operate differently however in the trade and investment spheres.8 Concerning trade, the MFN obligation essentially concerns border measures in respect of regulation governing the importation of goods, while the NT obligation is meant to prevent regulatory discrimination as compared to local competing products to the detriment of imported products once they have gained market access in the importing country (behind-the border treatment). As for foreign investment, both clauses tend to cover post-investment matters only. They operate once an investment has been admitted, except in the rare instances in which investment treaties also grant the right to make investments. In such treaties both clauses are aimed at preventing discrimination behind the border either between foreign protected investors and their investments in respect of those of other origin (MFN clauses) or in respect of local, national investors (NT clauses). The most widely used MFN clause in trade matters is that found in Article I of the General Agreement on Tariffs and Trade (GATT). Similar clauses can be found in other World Trade Organization (WTO) agreements such as the General Agreement on Trade in Services (GATS) and the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS). It is worth recalling the text of GATT Article I because its basic elements are also found in MFN clauses of investment agreements, albeit they are shortened, replacing the term ‘product’ with ‘investor’ or ‘investment’. Article I:1 requires WTO Members to treat ‘like products equally, irrespective of their origin’. In other words it prohibits discrimination ‘among “like products” originating in or destined for different countries’. According to WTO case law, Article I also prohibits ‘in fact’ or de facto discrimination, a principle that may be looked at also for the application of MFN clauses in investment treaties. It is immaterial whether the more favourable treatment invoked is found in national legislation or in another treaty.9 3 See ILC, Report of the International Law Commission, Sixty-seventh session (2015) UN Doc A/70/10, Annex, Final report of the Study Group on the Most-Favoured-Nation clause. 4 See T Weiler, The Interpretation of International Investment Law: Equality, Discrimination, and Minimum Standards of Treatment In Historical Context (Leiden, Martinus Nijhoff Publishers, 2013) 415–416. 5 See J Jackson, The World Trading System: Law and Policy of International Economic Relations, 2nd edn (Cambridge, MIT Press, 1997) ch 2. 6 Within the WTO, in the context of trade in goods, MFN treatment is required by Article I GATT. Additional MFN-type obligations can be found in Arts III:7, IV lit. b, V, IX:1 and XX lit. j. In the context of trade in services and intellectual property rights, MFN obligations arise under Article II GATS and Article 4 TRIPS Agreement respectively. 7 In other words, MFN clauses are to be construed according to the ejusdem generis principle. See, for example, E Ustor, ‘Most-Favoured Nation Clause’ in R Bernhardt (ed), Encyclopedia of International Public Law, Vol 8 (Amsterdam, North-Holland Publishing, 1985) 471. 8 This difference has been emphasised by N DiMascio and J Pauwelyn, ‘Nondiscrimination in Trade and Investment Treaties: Worlds Apart or Two Sides of the Same Coin?’ (2008) 102(1) American Journal of International Law 48. 9 This also holds true in investment law although the better treatment invoked is usually granted to nationals of other countries by treaty. In Menzies v Senegal, ICSID Case No ARB/15/21, Award (5 August 2016) (Members of the

Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment 429 MFN clauses in investment treaties also use the concept found in Article III GATT relating to NT, that of ‘treatment no less favourable’ than that granted to ‘like products of national origin in respect of all laws, regulations and requirements affecting their internal sale’. It comes as no surprise therefore that in many BITs a single clause establishes the entitlement to both standards of treatment. For example, the BIT between Austria and Czechoslovakia signed on 15 October 1990 provides: ‘Each Contracting Party shall accord investors of the other Contracting Party treatment no less favourable than that accorded to its own investors or investors of a third State and their investments.’10 In other BITs the MFN treatment is linked to fair and equitable treatment.11 The International Law Commission (ILC) dealt with MFN clauses in 1969–1978, when its work led to Draft Articles on the subject,12 and again in 2011–2015 leading to a Report by an ad hoc Study Group. The 2015 Report confirmed the 1978 definition of such clauses: ‘treatment accorded by the granting State to the beneficiary State, or to persons or things in a determined relationship with that State, not less favourable than treatment extended by the granting State to a third State or to persons or things in the same relationship with that third State’.13

II. MAJOR ARBITRAL DECISIONS ON THE APPLICATION OF MFN CLAUSES TO ESTABLISH MORE FAVOURABLE TREATMENT

This section looks at major cases concerning MFN clauses, including cases where treatment not provided in the ‘basic treaty’ can be extended via a MFN clause (II.A and B); cases where limitations to treatment provided in the basic treaty can be disregarded because of a MFN clause (II.C); and cases where a MFN clause has been invoked in respect of Investor-State Dispute Settlement (ISDS) provisions (II.D).

A. MFN Clauses Invoked to Benefit from Better Specific Treatment i. Occidental v Ecuador This is a significant case, decided on 1 July 2006, where the successful invocation of the MFN clause by the claimant led to extending more favourable substantive treatment to a foreign investor.14 Tribunal: Prof. Bernard Hanotiau, President Dr Hamid G Gharavi and Prof. Pierre Mayer), the Tribunal rejected the claim that the MFN clause of Art II GATS covering services could extend to consent to investment arbitration. 10 See also the Argentina-United Kingdom agreement of 11 December 1990, Art 3(1): ‘Neither Contracting Party shall in its territory subject investments or returns of investors or companies of the other Contracting Party to treatment less favourable than that which it accords to investments or returns of its own investors or companies or to investments or returns of nationals or companies of any third State.’ 11 For example, Article 3 of the China-Peru agreement of 9 June 1994 provides: ‘1. Investments and activities associated with investments of investors of either Contracting Party shall be accorded fair and equitable treatment and shall enjoy protection in the territory of the other Contracting Party. 2. The treatment and protection referred to in Paragraph 1 of this Article shall not be less favourable than that accorded to investments and activities associated with such investments of investors of a third State.’ 12 ILC, Draft Articles on most-favoured-nation clauses with commentaries (1978) II Yearbook of the International Law Commission, Part Two. 13 ILC, Final report, Study Group on the Most-Favoured-Nation clause (2015) UN doc A/CN.4/L.852, para 13. The ILC 2015 Report lists certain key elements that make up MFN provisions: 1) Under such a provision each state agrees to grant a particular level of treatment to the other state or states, and to persons and entities in a defined relationship with that state or those states. 2) The level of treatment provided by a MFN provision is determined by the treatment

430 Giorgio Sacerdoti and Niall Moran The better treatment invoked by the investor concerned the entitlement of exporters to receive a VAT refund. The issue was whether Occidental should be entitled to such a refund because it was granted to other national and foreign exporters based on Article II(1) of the US-Ecuador BIT (the ‘basic treaty’). This provision established the obligation to treat investments and associated activities of investors of either country ‘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 most favorable’. In respect of this matter, the Tribunal summarised the arguments of the parties as follows: The claimant based its entitlement on the fact that a ‘number of companies involved in the export of other goods, particularly flowers, mining and seafood products are entitled to receive VAT refund and continuously enjoy this benefit.’15 (…) The Claimant also asserts that the meaning of ‘in like situations’ does not refer to those industries or companies involved in the same sector of activity, such as oil producers, but to companies that are engaged in exports even if encompassing different sectors.16

On the other hand: The Respondent opposes all such arguments on the basis that ‘in like situations’ 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, it is argued, 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. (…) Ecuador also opposes the arguments concerning the most-favored-nation clause as no example is given of a Spanish or Argentine company in the oil sector.17

The Tribunal sided with the claimant based on the following reasoning on non-discrimination, which includes both MFN and NT treatment: In fact, ‘in like situations’ 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.18

The approach of the Occidental v Ecuador tribunal raises concerns because of the broad notion of ‘like circumstances’ (ejusdem generis), which is used without a satisfactory explanation. Such an approach ultimately deprives the host state of the right to differentiate the tax regime of different economic sectors (in this instance granting VAT refunds), a practice that is widely followed by many countries in the exercise of their legitimate regulatory powers.19

given by the state granting MFN to third states (‘no less favourable’). 3) A MFN commitment applies only to treatment that is in the same category as the treatment granted to the third state (‘ejusdem generis’). 4) The persons or entities entitled to the benefit of MFN treatment are limited to those in the same category as the persons or entities of the third state that are entitled to the treatment being claimed (such as products or investors). It is noted that the second and third of these elements create the greatest difficulty. 14 Occidental Exploration and Production Company (OEPC) v Republic of Ecuador, LCIA Case No UN3467 (Members of the Tribunal: Patrick Barrera Sweeney, Judge Charles N Brower and Prof. Francisco Orrego Vicuña – President), available at jurismundi.com. 15 ibid, Award, para 168. 16 ibid, Award, paras 167–169. 17 ibid, Award, paras 171–172. 18 ibid, Award, paras 173–177. 19 See, eg, G Harrison and R Krelove, ‘VAT Refunds: A Review of Country Experience’ (2005) 5(218) IMF Working Paper sec 3A.

Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment 431 ii. Parkerings v Lithuania20 The claimant had established a local consortium that had obtained a permit to design, build and operate a ‘modern integrated parking system’, in the Old City of Vilnius, a UNESCO heritage site. After the permit had been revoked by the municipality due to environmental and cultural concerns, the claimant complained inter alia that it had received de facto less favourable treatment than another foreign-owned operator of a similar parking system in breach of the MFN clause in the applicable Norway-Lithuania BIT. The Tribunal disagreed, holding that a host state may validly differentiate between investors based on: (1) the social, cultural and environmental impacts of the investors’ investment projects; and (2) the costs and benefits the investors’ projects would provide for the host state.21 To judge those claims, the Tribunal applied the rule that a breach of the MFN obligation arises when a state accords different treatment to another foreign investor in ‘like circumstances’ (para 369). It clarified that investors will only be in ‘like circumstances’ if they are ‘in the same economic or business sector’ (para 371). The Tribunal also added that if the state possesses a legitimate objective for treating the two investors differently, no violation of the MFN provision will be found.22 The Tribunal held that Lithuania did not breach the MFN obligation because although it treated Parkerings and the other investor (Pinus Proprius) differently, there were legitimate reasons for distinguishing between the two investors’ investments. In particular, the Tribunal stated: The fact that [Parkerings’] MSCP project extended significantly more into the Old Town as defined by the UNESCO is decisive …. The [goals of] historical and archaeological preservation and environmental protection could be and in this case were a justification for the refusal of the project. The potential negative impact of the [Parkerings] project in the Old Town was increased by its considerable size and its proximity with the culturally sensitive area of the Cathedral. Consequently, [Parkerings’] MSCP … was not similar with the MSCP constructed by Pinus Proprius.23

According to this award environmental concerns, cultural values, domestic and international obligations, and other assessments of relative costs and benefits may influence ‘likeness’ determinations, a novel approach that has no precedent in GATT or in investment cases.24 It also suggests that characteristics of the actual investment projects are relevant to determining whether the investors are in ‘like circumstances’ for purposes of MFN analysis (para 410).

B. MFN Clauses as a Basis for Claiming General Treatment not Provided for in the ‘Basic Treaty’ In another series of cases, MFN clauses in BITs have been invoked more broadly to extend a standard of treatment, such as fair and equitable treatment (FET), that the basic treaty did not include but was provided for in other BITs entered into by the host state. 20 Parkerings Compagniet AS v Lithuania, ICSID Case No ARB/05/8, Award (11 September 2007) (Members of the Tribunal: Dr Julian Lew QC, The Hon. Marc Lalonde PC, OC, QC and Dr Laurent Lévy – President). 21 ibid paras 392–396, 410, 430. 22 ibid paras 371, 375–376. 23 ibid para 392; see also 393–396. 24 For a review of the GATT/WTO case law by the Appellate Body on the likeness concept in Article I GATT see G Sacerdoti and others, ‘Commentary to Article I GATT, General Most-Favoured-Nation Treatment’ in R Wolfrum, PT Stoll and H Hestermeyer (eds), WTO Trade in Goods (Leiden, Brill, 2011) 53–77.

432 Giorgio Sacerdoti and Niall Moran i. Bayindir v Pakistan25 This dispute arose under the 1995 BIT between Turkey and Pakistan, which does not contain the fair and equitable treatment standard as an obligation, this standard being just mentioned in the preamble as ‘desirable’. Therefore in order to rely on that standard the claimant invoked the MFN clause of the BIT to ‘import’ the fair and equitable treatment obligation found in several BITs between Pakistan and other countries. In its Decision on Jurisdiction the Tribunal found that the MFN clause in Article II(2) of the BIT permitted the Turkish investor prima facie to rely on the fair and equitable treatment provisions provided for in Pakistan’s other BITs.26 In the merits the Tribunal confirmed this conclusion stating that: the ordinary meaning of the words used in Article II(2) together with the limitations provided in Article II(4) show that the parties to the Treaty did not intend to exclude the importation of a more favourable substantive standard of treatment accorded to investors of third countries.27

ii. Magdenli v Kazakstan Following this approach the MFN clause in a similarly-worded BIT between Turkey and Kazakhstan, which also only contained a reference to FET in its preamble, was invoked to import the FET standard from another BIT of the host country as a legal obligation. In an International Chamber of Commerce (ICC) award between Magdenli, a Turkish investor, and Kazakhstan the Tribunal summed up the issue as follows: The Tribunal does not see why the broader protections granted to investors of other countries under the BITs in force, which they have concluded with Kazakhstan, could not be the basis to extend those protections to Turkish investors thanks to the MFN clause of Article II(2) of the BIT. The object and purpose of a MFN clause is exactly to enable the beneficiaries of rights under the treaty containing such clause (the basic treaty) to benefit from the broader protection that the country which has granted thereby those rights has granted to nationals of other countries under other (‘comparator’) treaties, or even unilaterally. In this way, especially in respect of treaties in the area of trade, investments and treatment of aliens, equality of treatment as to rights and competitive opportunities is maintained over time between nationals, natural and legal persons, of different countries. It is true that the operation of the MFN clause is not without limits. It is generally considered that it operates in respect of ‘eiusdem generis’ matters, that is in respect of the same or similar subject matter covered by the treaty containing the MFN clause. This is however what is at issue here, where Magdenli invokes other BITs to obtain for its investment in Kazakhstan the more extensive protection that Kazakhstan grants in its territory to investors of other countries under the respective BITs.28

25 Bayindir Insaat Turizm Ticaret Ve Sanayi AS v Pakistan, ICSID Case No ARB/03/29 (Members of the Tribunal: Prof. Gabrielle Kaufmann-Kohler – President, Sir Franklin Berman and Prof. Karl-Heinz Böckstiegel). 26 ibid, Decision on Jurisdiction, 14 November 2005. 27 ibid, Award 27 August 2009, para 157. 28 See Magdenli v Kazakhstan, ICC award of 8 November 2018, paras 125–127, www.iareporter.com/news-andanalysis, under the Turkey-Kazakhstan BIT of 1992 (Members of the Tribunal: Prof. Giorgio Sacerdoti, – President, Jeffrey M Herzberg and Prof. Sebnem Akipek Okal).

Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment 433 C. MFN Clauses as a Basis for Disregarding Limitations to Treatment Provided for in the ‘Basic Treaty’ In other cases the clause has been invoked to extend the application of a given treatment, such as the full protection and security of fair and equitable treatment, or protection from expropriation, by disregarding limitations set forth in the basic treaty that other BITs did not include. In one of the first ICSID cases, Asian Agricultural Products Ltd. (AAPL) v Sri Lanka,29 the claimant argued that the full protection and security clause of the Switzerland-Sri Lanka BIT should also apply in the case of a civil disturbance, which that BIT excluded, because another of Sr Lanka’s BITs did not provide for such an exemption. While rejecting the claim on the merits, the Tribunal did not rule out that such better treatment could be imported.30 i. Pope & Talbot Inc. v Canada The intention of the parties has been the key factor for some tribunals in deciding whether a provision may be ‘expanded’ based on a MFN clause. This intention is influenced by the context in which the basic treaty was concluded and whether or not it was intended that the treatment in question could be subject to MFN extensions.31 In Pope & Talbot Inc. v Canada the Tribunal admitted such an expansion in principle. It stated in an obiter dictum, since it did not reach the merit of the claim, that the MFN treatment provided for in Article 1103 of NAFTA justified an ‘additive’ interpretation of the Minimum Standard of Treatment of NAFTA Article 1105 with reference to the more favourable fair and equitable treatment standard found in various BITs concluded by Canada.32 ii. CME v Czech Republic In this UNCITRAL Award, the Tribunal relied on the MFN clause in Article 3.2 of the Netherlands-Czech Republic BIT in order to determine the content of ‘just compensation’ by importing the ‘fair market value’ standard found in the US-Czech Republic BIT. The Tribunal reasoned as follows: The determination of compensation under the Treaty between the Netherlands and the Czech Republic on basis of ‘fair market value’ finds further support in the ‘most favored nation’ provision of Article 3(5) of the Treaty. That paragraph specifies that if the obligations under national law of either party in addition to the present Treaty contains rules, whether general or specific, entitling investments by investors of the other party to a treatment more favourable than provided by the present treaty ‘such rules shall to the extent that they are more favourable prevail over the present Agreement’. The bilateral investment treaty between the United States and the Czech Republic provides that compensation shall be equivalent to the fair market value of the expropriated investment …. The Czech Republic is therefore obliged to provide no less than fair market value to Claimant in respect of its investment …33 29 Asian Agricultural Products Ltd (AAPL) v Republic of Sri Lanka, ICSID Case No ARB/87/3 (1990) (Members of the Tribunal: Dr Ahmed Sadek El Kosheri – President, Prof. Berthold Goldman and Dr Samuel KB Asante). See also E Lagrange, ‘SPP v Egypt, AAPL vs Sri Lanka: Some Revolutionary Steps?’, ch 6 in this volume. 30 Asian Agricultural Products Ltd (AAPL) v Republic of Sri Lanka, above (n 29) para 54. 31 ILC, Final Report, above (n 13) section C, paras 174–193. 32 UNCITRAL (NAFTA), Procedural Order No 2, 28 October 1999, in Pope & Talbot Inc v Government of Canada, NAFTA/UNCITRAL, Tribunal Damages Award IIC 195 (2002) para 75 (Members of the Tribunal: The Hon Lord Dervaird – Presiding Arbitrator, The Hon Benjamin J Greenberg, QC and Munay J Belman). 33 CME Czech Republic BV v The Czech Republic, Final Award (14 March 2003) para 500 (Members of the Tribunal: Dr Wolfgang Kühn – Chairman, Judge Stephen M Schwebel and Ian Brownlie, QC).

434 Giorgio Sacerdoti and Niall Moran D. Major Decisions on the Application of MFN Clauses to Investor-State Dispute Settlement Provisions In arbitral practice violation or breaches of MFN treatment clauses in investment treaties have not been raised frequently and their resolution has rarely been controversial despite the decisions in Occidental, Parkering and Bayindir mentioned above. However, an ‘unexpected’34 application of MFN treatment in investment treaties has provoked endless debate and generated different, and some inconsistent, decisions by arbitral tribunals as well as many dissenting opinions. The issue at stake is the application of MFN treatment to ISDS provisions from third treaties considered more favourable to solve issues relating to admissibility and jurisdiction over a claim, such as the elimination of a preliminary requirement to attribute or to extend the scope of jurisdiction. In this respect, MFN clauses in a ‘basic treaty’ (the one invoked, containing the MFN clause) have been invoked to expand the scope of a treaty’s dispute settlement provisions in several ways. These include: (a) invoking a dispute settlement process not available under the basic treaty;35 (b) broadening the jurisdictional scope where the basic treaty restricted the ambit of the dispute settlement clause to a specific category of disputes, such as disputes relating to compensation for expropriation; and (c) overriding the applicability of a provision requiring the submission of a dispute to a domestic court for a ‘waiting period’ of 18 months, prior to submission to international arbitration. It is in this third circumstance that MFN has been most commonly invoked. Two of the major cases concerning the application of a MFN clause to ISDS provisions and their impact are now considered. i. Maffezzini v Spain36 This was the first award that had to decide on the applicability of the MFN clause to the dispute settlement mechanism provided by the basic treaty invoked by the claimant in order to avail itself of the less rigorous conditions of another BIT concerning arbitration. The Argentinian claimant invoked other BITs Spain had entered into in order to avoid the requirement of the applicable Argentina-Spain BIT that a claimant pursue its claim in domestic courts for an 18-month period before accessing international arbitration. The Tribunal accepted the claimant’s contention, considering that ‘all matters’ in the MFN clause also included dispute settlement provisions since these were ‘essential to the protection of the rights envisaged under the pertinent treaties; they are also closely linked to the material aspects of the treatment accorded’.37 The Tribunal noted that the waiting period did not preclude access to international arbitration and thus did not reflect a fundamental question of public policy. It would have been different in respect of an exhaustion of local remedies requirement, since in the Tribunal’s view, ‘[T]he beneficiary of the clause should not be able to override public policy

34 As

defined in the presentation of the UNCTAD, above (n 2) xiv (2010). Menzies v Senegal, above (n 9). 36 Emilio Augusto Maffezzini v Spain, ICSID Case No 97/7, Decision on Jurisdiction (25 January 2000) (Members of the Tribunal: Prof. Francisco Orrego Vicuña – President, Judge Thomas Buergenthal and Maurice Wolf). 37 ibid para 55. Article IV(2) of the BIT provides that ‘in all matters subject to this Agreement, this treatment shall not be less favorable than that extended by each Party to the investments made in its territory by investors of a third country’. 35 See

Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment 435 considerations that the contracting parties might have envisaged as fundamental conditions for their acceptance of the agreement in question …’38 Following the Maffezzini decision, several investment tribunals have faced the same issue in relation to the MFN clause in the Argentina-Spain BIT in cases brought against Argentina. These tribunals relied inter alia on the broad wording of the MFN clause (‘in all matters’) and on the close relation of the dispute settlement clause to the substantive protection granted by the BIT (the ‘basic treaty’), considering the exemption from the 18-month requirement as a more favourable treatment.39 The definition of the requirement as a temporary and merely suspensive admissibility condition rather than a jurisdictional prerequisite has also been used to justify the application of the MFN clause.40 In addition, the Tribunal in Telefónica v Argentina observed that the local litigation requirement pertained more to the substantive treatment by Argentina in its territory than to the international dispute settlement mechanism and was thus undoubtedly covered by the MFN requirement.41 ii. Plama v Bulgaria Other tribunals have expressed the contrary view that the limits of the consent to arbitration and the conditions thereof cannot be circumvented by relying on a MFN clause.42 According to Plama jurisdictional limits agreed in the basic treaty under which a claimant challenges measures of the host state allegedly in breach of the treaty obligations cannot be sidelined as a result of a MFN clause. In Plama the claimant invoked the MFN clause of the Bulgaria-Cyprus BIT that provided international arbitration only in respect of the amount of compensation in order to extend ICSID jurisdiction also to the legality of a taking (expropriation). The Tribunal rejected the argument and dissented from Maffezzini by recourse to a more restrictive interpretation, excluding any presumption of the application of the MFN clause to dispute settlement. An agreement to arbitrate requires in the tribunal’s view a ‘clear and unambiguous intention’.43 The Tribunal considered that: a MFN provision in a basic treaty does not incorporate by reference dispute settlement provisions in whole or in part set forth in another treaty, unless the MFN provision in the basic treaty leaves no doubt that the Contracting Parties intended to incorporate them.44

38 ibid

para 62. Siemens v Argentina, Decision on Jurisdiction (3 August 2004) ICSID Case No ARB/02/08 (Members of the Tribunal: Dr Andrés Rigo Sureda – President, Judge Charles N Brower and Prof. Domingo Bello Janeiro); Hochtief v Argentina, ICSID Case No ARB/07/31, Decision on Jurisdiction (17 June 2005) (Members of the Tribunal: Prof. Vaughan Lowe, QC – President, Judge Charles N Brower and J Christopher Thomas, QC); Gas Natural v Argentina, ICSID Case No ARB/03/10 (Members of the Tribunal: Prof. Andreas F Lowenfeld – President, Henri C Álvarez and Dr Pedro Nikken); National Grid v Argentina, UNCITRAL Decision on Jurisdiction (24 October 2011); Impregilo v Argentina, ICSID Case No ARB/07/17, Award (21 June 2011) (Members of the Tribunal: Judge Hans Danelius – President, Judge Charles N Brower and Prof. Brigitte Stern) (with a strong dissenting opinion by B Stern, thoroughly reviewing the opposite stream of cases). 40 Teinver v Argentina, Decision on Jurisdiction (21 December 2012) ICSID Case No ARB/09/1 (Members of the Tribunal: Judge Thomas Buergenthal – President, Henri C Alvarez, QC and Dr Kamal Hossain). 41 Telefónica v Argentina, Decision on Jurisdiction (25 May 2006) ICSID Case No ARB/03/20 (Members of the Tribunal: Prof. Giorgio Sacerdoti – President, Judge Charles N Brower and Eduardo Siqueiros), para 102–104. 42 See Salini and Italstrade v Jordan, Decision on Jurisdiction (29 November 2004) ICSID Case No ARB/02/13 (Members of the Tribunal: HE Judge Gilbert Guillaume – President, Bernardo Cremades and Sir Ian Sinclair); Wintershall AG v Argentina, Award (8 December 2008) ICSID Case No ARB/04/14 (Members of the Tribunal: Fali S. Nariman – President, Dr Santiago Torres Bernárdez and Prof. Piero Bernardini). 43 Plama Consortium Ltd v Bulgaria, Decision on Jurisdiction (8 February 2005) ICSID Case No ARB/03/24 (Members of the Tribunal: Carl F Salans – President, Prof. Albert Jan van den Berg and VV Veeder, QC), para 199. 44 ibid para 208. 39 See

436 Giorgio Sacerdoti and Niall Moran Clear support for the Plama decision came shortly thereafter from the Tribunal in Berschader v Russia, in respect of a similar request to that of Plama to extend jurisdiction based on the MFN clause beyond the quantum of compensation. After a detailed review of relevant cases, the Tribunal held that: (…) the expression ‘all matters covered by the present Treaty’ does not really mean that the MFN provision extends to all matters covered by the Treaty. Therefore ‘the ordinary meaning’ of that expression is of no assistance in the instant case, and the expression as such does not warrant the conclusion that the parties intended the MFN provision to extend to the dispute resolution clause.45

In the same vein, the Tribunal in Telenor v Hungary rejected the application of the MFN clause as a basis to expand the arbitral jurisdiction to disputes not covered by the dispute settlement provision. The Tribunal reasoned as follows: Where, as in the present case, both parties to a BIT which restricts the reference to arbitration to specified categories have entered into other BITs which refer all disputes to arbitration or where they have concluded other BITs some of which refer all disputes to arbitration while others limit such a reference to specified categories of dispute, then it can fairly be assumed that in the BIT in question the two parties share a common intention to limit the jurisdiction of the arbitral tribunal to the categories so specified. In these circumstances, to invoke the MFN clause to embrace the method of dispute resolution is to subvert the intention of the parties to the basic treaty, who have made it clear that this is not what they wish.46

The above divergent positions have prompted several states in subsequent BITs to specify that dispute settlement is not covered by the MFN clause.47 As to existing BITs, in its conclusions the ILC 2015 Report on the MFN standard distinguishes between a jurisdictional and a contractual approach followed by investment tribunals. In principle the limits to agreed jurisdiction as defined by the parties cannot be extended through recourse to a MFN clause, while other procedural requirements and admissibility conditions may be extended in this way. The Study Group observes ‘that conclusions about the applicability of MFN clauses to dispute settlement provisions should be based on the interpretation and analysis of the provisions in question and not on assumptions about the nature of investment agreements or of the rights that are granted under them’.48

III. MAJOR NATIONAL TREATMENT CASES IN INTERNATIONAL INVESTMENT LAW

There are three main stages in a national treatment claim in investment law: the first involves a determination of whether or not an investor or investment is in like circumstances with the

45 Beschader v Russia, SCC Case No 080/2004, Award (21 April 2006) (Members of the Tribunal: Advokat Bengt Sjövall, Prof. Sergei Lebedev and Prof. Todd Weiler), para 194. 46 ibid para 95. 47 Thus, while the BIT between Colombia and Switzerland of 2006 exclude at Note to Article 4.2 dispute settlement from the clause (https://investmentpolicy.unctad.org/international-investment-agreements/treaty-files/803/download), the UK Model investment treaty (2008) explicitly includes it at Article 3.3 (https://investmentpolicy.unctad.org/ international-investment-agreements/treaty-files/2847/download). 48 ILC, above (n 13) para 173. On the basis of the Report, the Summary Conclusions of the ILC on Most-Favoured-Nation clause at point (e) states: ‘Whether MFN clauses are to encompass dispute settlement provisions is ultimately up to the States that negotiate such clauses. Explicit language can ensure that a MFN provision does or does not apply to dispute settlement provisions. Otherwise the matter will be left to dispute settlement tribunals to interpret MFN clauses on a case-by-case basis’, ILC 3277th meeting (23 July 2015), UN doc A/CN.4/SR.3277.

Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment 437 identified domestic comparator. At the second stage, the Tribunal determines whether the investor or investment has received less favourable treatment (LFT) than the domestic counterpart, which has been defined at the first stage. At this point, the burden of proof shifts to the defendant, and in the third stage the host state may attempt to justify the differences in its treatment of investors. This section gives an overview of major decisions relating to each stage and shows that the case law has been considerably developed, but not necessarily uniformly.

A. Likeness in the Case Law During the likeness stage, a tribunal must determine the appropriate domestic comparator and whether the foreign investor is in ‘like circumstances’ with this comparator. Tribunals have found that the elements that make up likeness depend upon the ‘facts of a given case’,49 the ‘legal context and the specific circumstances of any individual case’,50 and taking into account ‘all the circumstances of each case’.51 The ‘remarkable’ variance52 in the likeness tests employed by tribunals has led to unpredictable outcomes and this can arise even where the wording of the relevant national treatment provisions are quite similar. This section examines several approaches that have been employed by tribunals in major decisions establishing the relevant comparator. The Tribunal in Methanex v U.S.A.53 employed a ‘narrow approach’ to identifying the proper comparator. In this case, methanol and ethanol were competing products that were treated differently. This was not deemed to be discriminatory however as there was domestic investment that was deemed to be identical to Methanex that also received less favourable treatment. The US proposed methodology that looked at comparators that were close to the foreign investment ‘in all relevant respects’.54 The Tribunal followed this methodology and deemed this comparator to be identical to Methanex and that it would be ‘perverse’ to ignore such an identical comparator and to use comparators that were less ‘like’.55 It may seem that there are many situations where the requirement of an identical comparator is overly burdensome given the many possible differences between operators in terms of size. However, the Tribunal in Methanex found that ‘where there is no identical domesticallyowned counterpart to the foreign-owned investment. [...] a tribunal may look farther afield and expand the scope of domestically-owned comparators’.56

49 Pope & Talbot Inc v Government of Canada, NAFTA/UNCITRAL, Tribunal Damages Award IIC 195 (2002) Award on Merits, para 75. 50 Total v Argentina, Award and Dissenting Opinion(2013) ICSID Case No ARB/04/01 para 210 (Members of the Tribunal: Prof. Giorgio Sacerdoti – President, Henri C Alvarez and Luis Herrera Marcano). 51 SD Myers, Inc v Government of Canada, NAFTA/UNCITRAL Tribunal, 1st Partial Award and Separate Opinion, IIC 249 (2000) para 244 (Members of the Tribunal: Prof. J Martin Hunter – President, Edward C Chiasson, QC and Prof. Bryan P Schwartz). 52 A Reinisch, ‘National Treatment’ in Bungenberg and others, above (n 2) 859. 53 Methanex Corp v USA, UNCITRAL, Final Award of the Tribunal on Jurisdiction and Meris (3 August 2005) (Members of the Tribunal: J William F Rowley, Prof. W Michael Reisman and VV Veeder, QC – President). 54 ibid Part IV, ch B, para 16. 55 ibid para 17. 56 ibid para 15.

438 Giorgio Sacerdoti and Niall Moran The Tribunal in Occidental v Ecuador57 took a ‘broad approach’ in identifying the proper comparator. Occidental, like another domestic oil producer, had not received a VAT refund. The purpose of VAT refunds is to ensure unchanged conditions of competition, a goal that is only relevant to producers in the same sector. Occidental claimed that other companies involved in the export of goods were the appropriate comparator. The Tribunal accepted Occidental’s argument 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 As in the previous 2006 case discussed above in respect of MFN treatment, the Tribunal never addressed the question of whether or not the discrimination was nationality-based. The Tribunal stated that it compared Occidental to all domestic exporters as ‘no exporter ought to be put in a disadvantageous position as compared to other exporters’.59 When the relevant comparator is construed so broadly, the result can be ‘seemingly absurd’ with flowers, seafood and oil being considered together.60 The reasoning in the award assumes that the de facto discrimination between Occidental and other exporters was based on nationality, without considering whether the flower and seafood exporters in question were foreign or domestically owned. An approach that is in between the broad and narrow ones above was used in Pope & Talbot Inc v Canada.61 The Tribunal stated that the first stage in national treatment analysis is essentially a comparison of whether the investments are in the same business or economic sector. This prompts the question of whether a competitive relationship between investors is a necessary condition for likeness (in line with the WTO approach to determining ‘like products’). SD Myers v Canada was the first national treatment claim under the NAFTA and it endorsed a competition-based approach to likeness62 having due regard to the ‘overall legal context’ in which the term ‘like circumstances’ appears in the NAFTA and the GATT. This case concerned Canada’s export ban on hazardous waste. SD Myers is an Ohio-based firm that was providing for the remediation of polychlorinated biphenyl (PCB) as part of a business model that relied on an open border between the US and Canada. An interim order had banned the export of PCBs because of a significant danger to ‘the environment and human life’.63 The Tribunal considered likeness (via a competition test), less favourable treatment (via a test for disparate impact) and finally the purpose of the regulatory measure.

57 Occidental Petroleum Corp v The Republic of Ecuador, Final Award (5 October 2012) ICSID Case No ARB/06/11 (Members of the Tribunal: David Williams, QC, Prof. Brigitte Stern and L. Yves Fortier, QC – President) (2012). 58 ibid para 173. 59 ibid para 176. 60 Di Mascio and Pauwelyn, above (n 8). 61 Pope & Talbot Inc v Government of Canada, Tribunal Report, UNCITRAL, Award on the Merits of Phase 2 (10 April 2001). 62 SD Myers, Inc v Government of Canada, NAFTA/UNCITRAL Tribunal, 1st Partial Award and Separate Opinion, (2000) para 251. Further authority for this position includes: 1) Archer Daniels Midland v Mexico, ICSID Case No ARB (AF)/04/5 (Members of the tribunal: Bernardo M Cremades – President, Arthur W Rovine and Eduardo Siqueiros), paras 201–204, which endorsed this competition-based approach; and 2) United Parcel Service of America Inc v Government of Canada, ICSID Case No UNCT/02/1, Award on the Merits (Members of the Tribunal: Dean Ronald A Cass, L Yves Fortier, QC and Judge Kenneth Keith – President), para 101, where it was found that a competitive relationship establishes a prima facie case of like circumstances. 63 SD Myers v Canada, above (n 62) para 123.

Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment 439 The Tribunal found that SD Myers was in like circumstances with Canadian operators based on the competitive relationship between them as SD Myers was ‘in a position to attract customers that might otherwise have gone to the Canadian operators’.64 Other tribunals have however found that a competitive relationship is not decisive in determining the likeness of investors: where there is a legitimate reason for the different treatment of investments, as discussed hereafter in section III.C.

B. Less Favourable Treatment in the Case Law This second step of national treatment claims involves determining whether state conduct has resulted in the investor or investment receiving less favourable treatment (LFT) compared to the relevant comparator as a result of the application of domestic policy, laws or regulations. This section examines how tribunals in several major decisions have determined whether or not there has been LFT. Tribunals must determine whether the investor received LFT compared to their domestic comparator. Thus a critical question is how broadly or narrowly the domestic comparator has been construed at the likeness stage. If the domestic comparator is ‘identical’ investors, it is more difficult to show LFT. Once the relevant comparator in Methanex was found to be an identical domestic comparator, it is unsurprising that the Tribunal found that ‘There is no more or less favourable treatment here. The treatment is uniform, for the ban applies to all MTBE manufacturers.’65 Methanex had not received LFT compared to identical methanolproducing domestic operators. The narrow approach employed in Methanex v U.S.A ‘runs the risk of excluding swathes’ of discriminatory conduct from the scope of national treatment.66 This would be a particular risk where LFT is accorded to an investor and an identical domestic firm, where there are other non-identical domestic competitors. Where there is a broad reading of likeness such as described above in Occidental v Ecuador, this makes it easier to show LFT. Other companies involved in the export of goods were deemed to be the appropriate comparator here and Occidental was deemed to have received less favourable treatment as it was denied a VAT refund, which was paid to exporters of seafood products inter alia. The fact that its domestic competitor was also denied such a refund did not affect the outcome in this case (the competitor was the state-owned Petroecuador and any tax rebate would have been revenue neutral). In Pope & Talbot v Canada, the Tribunal found that LFT could be demonstrated by showing that an investor has not received treatment that is the equivalent to that received by any like domestic investor.67 By this standard, it is sufficient that one domestic company has received preferential treatment to find a breach of the treaty even if the investor has been treated in the same way as 99 per cent of domestic companies. The Tribunal in AES v Hungary found that in order to conclude that a claimant has been treated less favourably, there has to have been actual discrimination that goes beyond a mere difference in treatment. The different fees charged to producers were deemed not to be

64 ibid

para 251. Corp v USA, above (n 52) para 21. 66 J Kurtz, ‘The Use and Abuse of WTO Law in Investor–State Arbitration: Competition and its Discontents’ (2009) 20 European Journal of International Law 769. 67 Pope & Talbot Inc v Government of Canada, NAFTA/UNCITRAL, Award on the Merits of Phase 2 (2001) para 42. 65 Methanex

440 Giorgio Sacerdoti and Niall Moran discriminatory as they reflected the different ‘underlying costs’ of producers.68 These fees were arrived at using the same methodology for domestic and foreign producers and as such were not evidence of actual discrimination. In Mercer v Canada (2018)69 the investor established that it was treated differently to other self-generating pulp mills in the same sector. However, the different treatment was not proven to be ‘discriminatory treatment’. Other pulp mills ‘whilst ostensibly comparators’, were not found by the Tribunal to be in like circumstances in relation to their particular circumstances under BC Hydro’s consistent application of its GBL methodology. The Tribunal found that although investors may be operating within the same industry, for the purpose of a specific regulatory treatment differences in the individual circumstances of the investors may entitle the regulator to treat them differently based on the particular circumstances relevant to both investments. In ADF Group v Buy America, the determination of whether there had been less favourable treatment largely concerned the characterisation of the investment.70 The investment of the investor was deemed to be ‘its steel in the US’ rather than its contractual right to provide fabricated steel.71 ADF’s steel in the US was not found to have been treated differently to steel owned by US investors because all steel was required to be fabricated in the US. The Tribunal found that evidence of discrimination was required to find that there had been less favourable treatment. The Tribunal was however unable to determine that there had been LFT of the foreign investor as a result of the ‘scant evidence’ demonstrating the comparative economics and the relevant competitive situation of fabricators in the US and Canada.72 The Tribunal highlighted the cost of fabrication, fabrication capacity and transportation as factors that would be relevant to a determination of LFT. ADF could not source fabricated steel from its own facilities in Canada, unlike competitors whose facilities were based in the US. An important context however is that the NAFTA has exceptions for government procurement and thus breach would have been justifiable under Article 1108(7), NAFTA.73

68 ibid

para 10.2.7. International Inc v Government of Canada, Award (6 March 2018) ICSID Case No ARB(AF)/12/3 (Members of the Tribunal: VV Veeder, QC – President, Prof. Francisco Orrego Vicuña and Prof. Zachary Douglas, QC). 70 ADF Group Inc. v United States of America, Award (9 Jan 2003) ICSID Case No. ARB (AF)/00/1, para 155. (Members of the Tribunal: Judge Florentino P. Feliciano– President, Professor Armand de Mestral and Ms. Carolyn B Lamm). Newcombe reminds us that if the investment had been characterised as the contractual right to provide fabricated steel, then the Tribunal might have found there had been LFT and that this required use of local producers was a performance requirement. See A Newcombe and L Paradell, ‘Law and Practice of Investment Treaties: Standards of Treatment’ (Zuidpoolsingel, Kluwer Law International, 2009) para 4.24. 71 This case concerned the changing of the layout of a highway junction in Northern Virginia that was prone to accidents. For this project, ADF was sub-contracted for the supply of all steel components for nine bridges. The Tribunal found measures requiring the use of steel fabricated in the US in the construction products did not constitute LFT for a Canadian investor. 72 para 157: ‘… For instance, it appears to the Tribunal that specific evidence concerning the comparative economics of the situation would be relevant, including: whether the cost of fabrication was significantly lower in Canada; whether fabrication capacity was unavailable at that time in the United States and whether transportation costs to Canada were sufficiently low to make up the differential. We note the US did submit evidence of available capacity and Mr Paschini referred to massive increases in costs due to fabrication in the US. This scant evidence is, however, not sufficient to show what the relevant competitive situation of Canadian fabricators and US fabricators was in general, nor was it evidence of the comparative costs of steel fabrication in the US and Canadian facilities, in particular …’. (emphasis added) 73 Para 4.24. NAFTA Article 1108(7): ‘Articles 1102, 1103 and 1107 do not apply to: (a) procurement by a Party or a state enterprise; or (b) subsidies or grants provided by a Party or a state enterprise.’ 69 Mercer

Non-Discrimination Clauses: Most-Favoured-Nation and National Treatment 441 C. Regulatory Purpose and Host State Justification Likeness based on investors being in the same economic sector can be rebutted where there is a reason to distinguish investors for a legitimate regulatory purpose.74 The burden of proof is on the host state to show there was a legitimate regulatory purpose behind the discrimination and to reverse the finding of likeness. The regulatory purpose of the treatment must be taken into account in establishing breach of national treatment provisions in investment agreements. In Feldman v Mexico,75 the company was found to be in like circumstances with cigarette reseller-exporters but not cigarette producer-exporters as there were ‘rational bases’ for treating them differently.76 The tribunal found that firms reselling/exporting cigarettes were in like circumstances while those that merely produced cigarettes were not. It was acknowledged that there are ‘at least some rational bases for treating’ these investors differently including inter alia discouraging smuggling and protecting intellectual property rights.77 In Cargill v Mexico, it was found that like circumstances are not determined in the abstract but rather ‘by reference to the rationale for the measure that was being challenged’.78 The tribunal found that it would have been possible in respect of other, different measures that ‘the mills in GAMI and the lumber producers in Pope & Talbot’ could have been found to be in ‘like circumstances’.79 Thus the question is whether investors are in like circumstances with respect to the particular measure in question.80 Considering the regulatory purpose of a measure in determining likeness has allowed investment tribunals to balance investor interests ‘with an unlimited list of legitimate government concerns a list far broader than the exceptions in GATT Article XX’.81 Tribunals must consider the regulatory purpose of the measure as well as any nonprotectionist justifications for differences in the treatment of investors or investments by the state. These justifications may be considered as part of the national treatment article itself (eg as is the case with Article 1102 NAFTA) or may be based on treaty exceptions.82 When identifying the relevant subject for comparison, tribunals must take into account the regulatory purpose of the treatment. Although the broad test employed in the Occidental award was questioned above, in certain regulatory circumstances investors from different sectors should be treated in the same way. Investors that compete with and are ‘identical’ to their domestic counterparts, such as in Methanex v U.S.A, will not necessarily be in like circumstances where the regulatory purpose of a measure is considered. Competing products

74 See, eg, Windstream Energy LLC v Government of Canada, Award (27 September 2016) PCA Case No 2013-22 (Members of the Tribunal: Dr Veijo Heiskanen – President, R Doak Bishop and Dr Bernardo Cremades). 75 Marvin Feldman v Mexico (2002) ICSID Case No ARB(AF)/99/1 (Members of the Tribunal: Prof. Konstantinos D Kerameus, – President, Jorge Covarrubias Bravo and Prof. David A Gantz). 76 Di Mascio and Pauwelyn, above (n 8) 74. 77 Marvin Roy Feldman Karpa v Mexico, see above (n 75) paras 170–171. 78 Cargill v Mexico, Award (18 September 2009) ICSID Case No ARB(AF)/05/Z (Members of the Tribunal: Dr Michael C Pryles – President, Prof. David D Caron and Prof. Donald M McRae), para 206. 79 ibid. 80 See Mercer International Inc v Government of Canada, Award (6 March 2018) ICSID Case No ARB(AF)/12/3 (Members of the Tribunal: Prof. Francisco Orrego Vicuña, Prof. Zachary Douglas and VV Veeder, QC – President), para 7.21. 81 Di Mascio and Pauwelyn, above (n 8) 83. 82 See the Tribunal decision in Total v Argentina, Award, above (n 50).

442 Giorgio Sacerdoti and Niall Moran have also been deemed not to be ‘like’ for various regulatory reasons in cases such as UPS v Canada83 and Champion Trading Co. v Egypt84 as well as those discussed above. The tribunal in Pope & Talbot states that there must be a ‘reasonable nexus to rational government policies’ that are not motivated by favouring domestic investors.85 It found that any difference in treatment must bear a ‘reasonable relationship to rational policies’. In SD Myers v Canada, the tribunal carefully considered the contextual difference between national treatment under the WTO Agreements and under the NAFTA, with regard to the fact that there are no general exceptions to NAFTA. Schill describes this as a sophisticated approach ‘that marries adverse competitive impact with an assessment of impermissible regulatory purpose’.86

IV. CONCLUSION

The above cases illustrate the major trends, and conflicts, in the jurisprudence of nondiscrimination cases in international investment law.87 As seen in relation to MFN clauses, major cases have centred around the following issues: whether treatment not provided in the basic treaty can be extended via a MFN clause; whether limitations to treatment provided in the basic treaty can be disregarded because of a MFN clause; and whether a MFN clause applies to ISDS provisions. While tribunals have been willing to disregard some limitations of the basic treaty and to extend substantive protections such as the fair and equitable treatment via MFN clauses, they have taken divergent positions on ISDS provisions. The differing outcomes in cases such as Maffezzini and Plama have prompted treaty drafters to clarify their positions in subsequent agreements. There also seems to be an emerging acceptance that MFN clauses can import procedural elements of ISDS, but not in instances where ISDS, or a given model thereof, is not provided for at all in the treaty as this would go against the intention of the parties and belie the consensual basis of jurisdiction in international adjudication. Concerning national treatment cases, the case law has been dominated by three major questions: determining likeness and less favourable treatment, and whether discrimination may be justified by some legitimate regulatory purpose. Tribunals have diverged in how they have answered these fundamental questions. This chapter looked at the different ends of the spectrum in terms of how tribunals have construed likeness; at how determinations of likeness, be they broad or narrow, impact findings of less favourable treatment, as well as factors such as the characteristics of investments and whether their treatment is discriminatory. It also looked at how tribunals have approached various justifications put forth for differences in treatment between investors and what constitutes a legitimate regulatory purpose. Despite divergences in the merits, tribunals have been consistent in this basic approach.

83 United Parcel Service of America Inc. v Canada, Award (22 November 2002) ICSID Case No UNCT/02/1, paras 87–120. The Tribunal concluded in this case that the different characteristics of goods imported by mail and courier required different customs treatment, justifying the differential treatment. 84 Champion Trading Company, Ameritrade International, Inc v Arab Republic of Egypt, (2006) ICSID Case No ARB/02/9 (Members of the Tribunal: Robert Briner – President, L Yves Fortier, QC and Prof. Laurent Aynès), para 154 in particular. The justification for the difference of treatment in this case came down to one of the party’s trading on the free market while others were buying from Collection Centres at fixed prices, which was considered not to be a like situation. 85 See Pope & Talbot, above (n 61). 86 S Schill, The Multilateralisation of International Investment Law (Cambridge, Cambridge University Press, 2009) 273. 87 See N Piracha, Towards Uniformly Accepted Principles for Interpreting MFN Clauses, Kluwer Law, 2021.

26 Reparation Cases: Applicable Principles in International Investment Arbitration IRMGARD MARBOE*

I. INTRODUCTION

T

HE OBLIGATION TO provide reparation is the consequence of a wrongful act. This general principle of law has already been highlighted by the Permanent Court of International Justice (PCIJ) in Factory at Chorzow1 and has also gained the status of a rule of customary international law.2 In investment arbitration, the wrongful act generally consists in a state’s breach of its international obligations usually contained in an international investment protection treaty. In addition, if other obligations are violated, such as contractual obligations, the question of the type and extent of reparation also arises. Thus, all investment arbitration cases, which have reached the stage of a decision on the merits and established the breach of an obligation, are ‘reparation cases’. The purpose of this chapter is to highlight issues connected to the applicable legal rules and principles on reparation and valuation, to take account of some controversial topics, such as the choice of the valuation date and the role of ‘country risk’, to discuss the question of applicable valuation methods, and to provide some insight into the issue of interest. The abundant practice of investment arbitration on the matter of valuation contributes significantly to the reflection on reparation and compensation in international law in general. Notably, that practice has undergone an important evolution: while in the 1980s the decisions on quantum were hardly ever reasoned in detail,3 and forward looking valuation methods were regarded as ‘Nostradamus prophecies’,4 investment tribunals have increasingly clarified and specified the applicable rules and approaches to valuation.

* Irmgard Marboe is Professor of International Law at the Department of European, International and Comparative Law at the Law Faculty of the University of Vienna. 1 Factory at Chorzow, Germany v Poland (Claim for Indemnity) (Merits) (1928) PCIJ Series A, No 17, 47. 2 See S Ripinsky and K Williams, Damages in International Investment Law (London, British Institute of International and Comparative Law, 2008) 27; B Sabahi, Compensation and Restitution in Investor-State Arbitration (Oxford, Oxford University Press, 2011) 47 with further references and discussion; see also I Marboe, ‘Restitution, Damages and Compensation’ in M Bungenberg and others (eds), International Investment Law (Baden-Baden, Nomos, 2015) 1031, 1034 et seq with further references. 3 The notable exception and the beginning of a new trend towards the application of forward looking valuation methods was Amco Asia v Republic of Indonesia (Amco I), Award (20 November 1984) paras 251–282; Amco Asia v Indonesia (Amco II), Award (5 June 1990), paras 169–284. 4 I Seidl-Hohenveldern, ‘L’évaluation des dommages dans les arbitrages transnationaux’ (1987) 33 Annuaire français de droit international 7, 24.

444 Irmgard Marboe II. THE CUSTOMARY LAW PRINCIPLE OF ‘FULL REPARATION’ AND THE BIT STANDARD: ADC V HUNGARY

ADC v Hungary5 can be regarded as a landmark case as it discussed the applicable standard and the principles for the calculation of damages in detail and raised some important aspects in this context. In particular, the Tribunal came to the conclusion that: (1) a distinction between lawful and unlawful expropriation was necessary; and (2) the date of the award should be the valuation date and not the date of the expropriation if the expropriation was unlawful and the value of the expropriated asset has increased. The dispute had arisen over a concession to reconstruct and operate Budapest airport. The Tribunal decided that the respondent state had indirectly expropriated the concession in breach of the applicable Bilateral Investment Treaty (BIT).6 According to the Tribunal, the expropriation was unlawful because it was not in the public interest, was discriminatory, did not comply with due process, and was not accompanied by the payment of compensation.7 At the beginning of its analysis of the issue of damages, the Tribunal noted that the applicable standard for assessing damages had given rise to considerable debate between the Parties, in particular ‘whether the BIT standard is to be applied or the standard of customary international law’.8 The claimants argued that the respondent’s deprivation of their investment was a breach of the BIT and as an internationally wrongful act was subject to the customary international law standard as set out in Chorzów Factory. The respondent, by contrast, contended that the BIT standard was a lex specialis, which superseded the customary international law standard. The Tribunal accepted that a BIT could be considered as a lex specialis, whose provisions would prevail over rules of customary international law,9 but it pointed out that the BIT did not stipulate any rules relating to damages payable in the case of an unlawful expropriation. In an often-quoted formulation the Tribunal held: The BIT only stipulates the standard of compensation that is payable in the case of a lawful expropriation, and these cannot be used to determine the issue of damages payable in the case of an unlawful expropriation since this would be to conflate compensation for a lawful expropriation with damages for an unlawful expropriation.10

The Tribunal went on to distinguish the provisions of the BIT and the rules of customary law. Article 4 of the Hungary-Cyprus BIT referred to ‘just compensation’ in case of an expropriation and provided: 2. The amount of compensation must correspond to the market value of the expropriated investments at the moment of the expropriation. 3. The amount of this compensation may be estimated according to the laws and regulations of the country where the expropriation is made.11

5 ADC Affiliate Ltd and ADC & ADMC Management Ltd v Hungary, Award (2 October 2006) ICSID Case No ARB/03/16. 6 ADC v Hungary, above (n 5) para 476. 7 ibid paras 426–444, 476. 8 ibid para 480. 9 ibid para 481; the tribunal referred, amongst others, to Phillips Petroleum Co Iran v Iran, 21 Iran-US Cl Trib Rep 121. 10 ADC v Hungary, above (n 5) para 481. 11 ibid para 482. The laws and regulations of Hungary include Section 132 of the Hungarian Constitution which provides that expropriation of ownership must be accompanied by ‘full, unconditional and prompt compensation’, as the tribunal referred to the Respondent’s Counter-Memorial, at para 584.

Reparation Cases: Applicable Principles in International Investment Arbitration 445 Previous tribunals had hardly paid attention to the distinction between the standard of compensation for expropriation and the assessment of damages after an unlawful act. This distinction was generally not regarded as relevant as it seemed that there was no difference with respect to the financial consequences.12 A notable exception was the Iran-US Claims Tribunal, which pointed out in Philips Petroleum v Iran: The tribunal believes that the lawful/unlawful taking distinction, which in customary international law flows largely from the Case Concerning the Factory at Chorzów (Claim for Indemnity) (Merits), P.C.I.J. Judgment No. 13, Ser. A., No. 17 (28 September 1928), is relevant only to two possible issues: Whether restitution of the property can be awarded and whether compensation can be awarded for any increase in the value of the property between the date of taking and the date of the judicial or arbitral decision awarding compensation.13

The Tribunal in ADC v Hungary pointed out that, as the BIT did not contain any lex specialis governing the issue of the standard for assessing damages in the case of an unlawful expropriation, it had to apply the default standard contained in customary international law. According to his standard, it is necessary that: (1) ‘full reparation’ is provided, and (2) full reparation is achieved by a comparison of the financial situation of the injured party with and without the breach. The Tribunal referred to the PCIJ’s often-quoted formulation in Chorzów Factory, according to which ‘reparation must, as far as possible, wipe out all the consequences of the illegal act and re-establish the situation which would, in all probability, have existed if that act had not been committed’.14 In order to confirm that this formulation was still valid the Tribunal made reference to a number of arbitral awards by international investment tribunals15 as well as judgments and advisory opinions of the International Court of Justice,16 which had quoted and followed the PCIJ’s dictum. The Tribunal concluded that ‘there can be no doubt about the present vitality of the Chorzów Factory principle, its full current vigor having been repeatedly attested’.17 In addition, the Tribunal noted that the International Law Commission (ILC), in its Articles on the Responsibility of States for Internationally Wrongful Acts, expressly relied on Chorzów Factory,18 when it formulated in Article 31(1): The responsible State is under an obligation to make full reparation for the injury caused by the internationally wrongful act.19 12 See A Sheppard, ‘The Distinction between Lawful and Unlawful Expropriation’ in C Ribeiro (ed), Investment Arbitration and the Energy Charter Treaty (New York, JurisNet, 2006) 169, 198 et seq; more recently this view has been maintained by S Ratner, ‘Compensation for Expropriations in a World of Investment Treaties: Beyond the Lawful/Unlawful Distinction’ (2017) 111 American Journal of International Law 7–56. 13 Phillips Petroleum Company Iran v Iran, Award (29 June 1989) 21 Iran-US CTR (1989) 79, para 110. 14 Factory at Chorzow (Claim for Indemnity) (Merits), above (n 1) 47. 15 The tribunal referred to SD Myers v Canada, UNICTRAL (NAFTA) Award (Merits) (13 November 2000) para 311; Metalclad v Mexico, Award (30 August 2000) ICSID Case No ARB(AF)/97/1, para 122; CMS Gas Transmission Company v Argentina, Award (12 May 2005) ICSID Case No ARB/01/8, para 400; Petrobart Limited v Kyrgyzstan Republic, Arbitration case No 126/2003, Arbitration Institute of the Stockholm Chamber of Commerce (Energy Charter Treaty) (29 March 2005) paras 78 and 78; Amoco International Finance Corporation v Iran, 15 IRAN–US CTR, 189, 246 (paras 191–194); and MTD Equity v Chile (25 May 2004) ICSID Case No ARB/01/7, para 238. 16 Case Concerning the Gabcíkovo-Nagymaros Project (Hungary v Slovakia) (1997) ICJ Reports 7, para 152; LaGrand Case (Germany v US) (2001) ICJ Reports 466, para 125; Case Concerning the Arrest Warrant (Democratic Republic of Congo v. Belggium) (2002) ICJ Reports 3, para 76; Case Concerning Avena and other Mexican Nationals (Mexico v US) (2004) ICJ Reports 12, paras 119–121; Advisory Opinion on the Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory (2004) ICJ Reports 136, para 152. 17 ADC v Hungary, above (n 5) para 493. 18 ibid para 494. 19 Articles on the Responsibility of States for Internationally Wrongful Acts, concluded in 2001, UN GA Res UN-Doc A/ Res/ 56/ 83, Annex.

446 Irmgard Marboe The Tribunal in ADC v Hungary continued by stressing that the ILC, in Article 35, stipulated that restitution in kind was the preferred remedy for an internationally wrongful act and that, according to Article 36, ‘only where restitution cannot be achieved can equivalent compensation be awarded.’20 The Tribunal highlighted that the ILC, in its Commentary on the Articles, stated that ‘[t]he general principle of the consequences of the commission of an internationally wrongful act was stated by the Permanent Court in the Factory at Chorzów case’.21 Whether the above-mentioned rules on reparation as referred to by the Tribunal in ADC v Hungary are applicable in investment arbitration and not in purely inter-state disputes needs, however, some additional reflection. The PCIJ in Factory at Chorzów held to this effect: Rights or interests of an individual the violation of which rights causes damages are always in a different plane to rights belonging to a State, which rights may also be infringed by the same act. The damage suffered by an individual is never therefore identical in kind with that which will be suffered by a State; it can only afford a convenient scale for the calculation of the reparation due.22

Furthermore, Article 33 of the 2001 ILC Articles on state responsibility provides that the obligations of the responsible state owed to other states are ‘without prejudice to any right, arising from the international responsibility of a State, which may accrue directly to any person or entity other than a State’.23 The existence and scope of a claim for damages by an investor against a state thus depends on the contents of the primary rule of international law, usually an international investment protection treaty. However, while it is possible, in principle, that an international investment treaty provides for a different standard of the assessment of compensable damage, this does not seem to be the case.24 On the contrary, recent investment protection agreements confirm that the ‘full reparation’-standard remains unchanged.25 The consequence of the standard of full reparation as applied in the case ADC v Hungary was that the date of the valuation should be the date of the award, and not the date of the expropriation. While the BIT demanded the valuation date to be ‘at the moment of the expropriation’,26 the Tribunal held that full reparation under the customary law standard would require restitution as the primary remedy. It explained in this respect: It is clear that actual restitution cannot take place and so it is, in the words of the Chorzów Factory decision, ‘payment of a sum corresponding to the value which a restitution in kind would bear’, which is the matter to be decided.27

20 ADC

v Hungary, above (n 5) para 494. para 494. at Chorzow (Claim for Indemnity) (Merits), above (n 1)t 28. 23 Article 33 of the ILC Articles, above (n 19). 24 In contrast, Art 41 of the European Convention on Human Rights provides that ‘just satisfaction’ can be awarded by the Court, if it finds that there has been a violation of ’the Convention. For a more detaild discussion see I Marboe, Damages in Investor-State Arbitration. Current Issues and Challenges (Leiden, Brill Research Perspectives, 2018) 8–18. 25 See, eg, Art 8.39(3) of the Comprehensive Economic and Trade Agreement between Canada and the EU (CETA), signed on 30 October 2016: ‘Monetary damages shall not be greater than the loss suffered by the investor or, as applicable, the locally established enterprise, reduced by any prior damages or compensation already provided. For the calculation of monetary damages, the Tribunal shall also reduce the damages to take into account any restitution of property or repeal or modification of the measure.’ This formulation confirms the full reparation-standard and adds a few specifications. For a more detailed discussion see Marboe, above (n 23) 8–18. 26 See Art 4 of the agreement between Hungary and Cyprus and for the promotion and protection of investment law (Signed on 24 May 1989, entered into force on 25 May 1990) [Hungary-Cyprus BIT] 27 ADC v Hungary, above (n 5) para 495 (emphasis in original). 21 ibid

22 Factory

Reparation Cases: Applicable Principles in International Investment Arbitration 447 According to the Tribunal the case at hand was ‘almost unique’ since the value of the investment after the date of expropriation (1 January 2002) had risen very considerably.28 The airport concession had been handed over to another operator who managed the airport with considerable commercial success. In other arbitrations, there had been a decline in the value of the investment after regulatory interference, so that the application of the restitution standard by various tribunals had led to use of the date of the expropriation as the date for the valuation of damages.29 In order to confirm its conclusion, the Tribunal notably quoted the jurisprudence of the European Court of Human Rights (ECtHR). In particular, it highlighted the judgment in Papamichalopoulos and Others v Greece,30 where the Court had decided to compensate the expropriated party the higher value the property enjoyed at the moment of the Court’s judgment rather than the considerably lesser value it had had at the earlier date of dispossession.31 This corresponded to the long-established continuous practice of the ECtHR, a practice which was however later adapted.32 The Tribunal also made reference to the jurisprudence of the Iran-US Claims Tribunal focusing on the award in Amoco International Finance Corporation v Iran,33 where the Tribunal extensively cited with approval the PCIJ in Chorzów Factory according to which damages are ‘not necessarily limited to the value of the undertaking at the moment of dispossession’.34 The Tribunal in ADC v Hungary therefore concluded that the claimants should be compensated the market value of the expropriated investments as at the date of the award, which the Tribunal took as 30 September 2006. The choice of the valuation date in this case was the consequence of the primacy of the remedy of restitution under the general rules of state responsibility. While only a few investment tribunals have awarded restitution in the past,35 commentators largely agree that restitution remains an available remedy in investor-state arbitration in principle.36 With respect to the appropriate valuation method under the premise of the full reparationstandard, the Tribunal decided to apply the Discounted Cashflow (DCF) method37 even 28 ibid

para 496. para 496. 30 Papamichalopoulos et al v Greece (1995) (just satisfaction) ECHR Ser A, No 330-B, para 36. 31 ADC v Hungary, above (n 5) para 497. 32 See, eg, the judgment in Guiso-Gallisay v Italy (2009) ECHR 58858/00, para 102 and subsequent judgments; see I Marboe, Calculation of Compensation and Damages in International Investment Arbitration, 2nd edn (Oxford, Oxford University Press, 2017) 140 et seq. with further references. 33 Amoco International Finance Corporation v Iran, 15 IRAN-US CTR, 189, 247, para 196. 34 Factory at Chorzow, above (n 1) 47. 35 Texaco v Libya, Award (19 January 1977) 53 ILR 389, 511 (repondent legally bound to perform the contracts and to give them full effect); ATA Construction v Jordan, Award (18 May 2010) ICSID Case No ARB/08/2, paras 131–132 (restoration of the claimants’ right to arbitration); Enron v Argentina, Award (14 January 2004) ICSID Case No ARB/01/3, paras 79–81 (the tribunal accepted in principle the power to award restitution, which it eventually did not in its award of 22 May 2007); Arif v Moldova, Award (8 April 2013) ICSID Case No ARB/11/23, para 572 (tribunal offered respondent choice of restitution of duty free stores and their exclusivity or compensation). 36 Sabahi, above (n 2) 61–90; E Stopponi, La réparation dans le contentieux international de l’investissement – contribution à l’étude de la restitution in integrum (Paris, Pedone, 2015); C Schreuer, ‘Alternative Remedies in Investment Arbitration’ (2016) 3(1) Journal of Damages in International Arbitration 1, 20. 37 According to the World Bank Guidelines on the Treatment of Foreign Direct Investment, the Discounted Cashflow (DCF) method is a valuation method that looks at ‘the cash receipts realistically expected from the enterprise in each future year of its economic life as reasonably projected minus that year’s expected cash expenditure, after discounting this net cash flow for each year by a factor which reflects the time value of money, expected inflation, and the risk associated with such cash flow under realistic circumstances. Such discount rate may be measured by examining the rate of return available in the same market on alternative investments of comparable risk on the basis of their present value.’ World Bank, ‘Legal Framework for the Treatment of Foreign Investment, vol 2, Report to the Development Committee and Guidelines on the Treatment of Foreign Direct Investment’ (1992) 31 International Legal Materials 1363, 1376. 29 ibid

448 Irmgard Marboe though it was mindful of the respondent’s admonishment that ‘international tribunals have exercised great caution in using the [DCF] method due to its inherently speculative nature’.38 However, the Tribunal found the adoption of the DCF method adopted by the claimants’ expert was fully justified and that the calculations carried out were in line with the respective standard. The approach and method were reasonable and reliable so that they could be endorsed by the Tribunal in calculating the final amount of damages.39 In addition, the amount paid in the subsequent acquisition of the airport operation contract by another investor confirmed this calculation.40 This combination of forward looking methods and a market approach, which looks at prices actually paid for identical or comparable assets, is generally recommended in the valuation of large infrastructure projects. The Tribunal in OI European Group v Venezuela called this a ‘sanity check’ of the DCF method.41 The Tribunal in Rusoro v Venezuela, pointed out that the DCF method could not be applied to all types of circumstances, but works properly only if certain criteria are met, such as: (1) an established historical record of financial performance; (2) reliable projections of its future cash flow, ideally in the form of a detailed business plan, verified by an impartial expert; (3) the price at which the enterprise will be able to sell its products or services can be determined with reasonable certainty; (4) the business plan can be financed with self-generated cash, or at least without uncertainty regarding the availability of financing; (5) it is possible to calculate a meaningful Weighted Average Cost of Capital (WACC), including a reasonable country risk premium, which fairly represents the political risk in the host state; and (6) the enterprise is active in a sector with low or at least predictable regulatory pressure.42 The Tribunal rejected the DCF method in that case and applied a combination of market based and cost-based approaches.43 Despite remaining uncertainties, the DCF method has gained increasing acceptance in international arbitration as it seems to best reflect valuation practices actually applied in investment decisions and thus the damage incurred by a wrongful act, as also discussed in the Spanish solar energy cases.44 In ADC v Hungary another consequence of the choice of the date of the award as the valuation date was that the Tribunal decided to award post-award interest only. A ‘grace period’ of 30 days was given at which interest would start to accrue if the respondent had not paid the award in full.45 The interest rate was set at six per cent, compounded monthly. This rather high interest rate and the short compounding interval should give an incentive for quick compliance with the award.46

38 ADC

v Hungary, above (n 5) para 502. para 514. The tribunal even explicitly applauded the presentations made by the claimants’ expert as ‘an example as to how damages calculations should be presented in international arbitration; they reflect a high degree of professionalism, clarity, integrity and independence by financial expert witnesses’. ibid para 516. The amount awarded was US$76.2 million. 40 On 22 December 2005, BAA acquired Budapest Airport Rt for US$2.23 billion (£1.26 billion) for 75% minus one share and a 75-year assets management contract plus moveable assets. ibid para 516. 41 OI European Group v Venezuela, Award (10 March 2015) ICSID Case No ARB/11/25, paras 852–879. 42 Rusoro v Venezuela, Award (22 August 2016) ICSID Case No ARB(AF)/12/5, para 759. 43 ibid paras 762–770. 44 See, eg, Eisler Infrastructure v Spain, Award (4 May 2017) ICSID Case No ARB/13/36, para 465; Novenergia v Spain, Award (15 February 2018) SCC Case No 2015/063, paras 820–821; Masdar Solar v Spain, ICSID Case No ARB/14/1, Award (16 May 2018) paras 575–587; Antin Infrastructure v Spain, Award (15 June 2018) ICSID Case No ARB/13/31, para 688; RREEF v Spain, Award (11 December 2019) ICSID Case No ARB/13/30, paras 19–20. 45 The tribunal decided that the respondent should pay to the claimants US$7,623,693 in full satisfaction of their costs and expenses of the arbitration. ibid para 543. 46 See Marboe, ‘Calculation’, above (n 23) 393 et seq, with further references. 39 ibid

Reparation Cases: Applicable Principles in International Investment Arbitration 449 The value of the ADC v Hungary award lies in its detailed reasoning with respect to the valuation of damages, which at that time was quite unusual. It highlighted that the standard of compensation defined in BITs only referred to ‘lawful’ expropriations, thus expropriations carried out in accordance with the conditions laid down in the BIT. Unlawful expropriations, by contrast, would entail state responsibility and thus require full reparation. While the award is conclusive in its reasoning, commentators and tribunals are not always convinced of its result. One issue concerns the question of whether non-payment of compensation alone would render the expropriation unlawful. Another issue is the allegedly arbitrary choice of the valuation date, if it is the date of the award. Nevertheless, it is still an unusual situation that an asset increases in value after its expropriation. In most cases the value of the expropriated asset declines after the expropriation. In such cases the standard of compensation contained in the BIT represents the minimum amount, which has to be paid to the former owner. If the value increases, restitution of the asset or an amount equivalent to the higher value represents the necessary consequence.

III. YUKOS V RUSSIA: THE LARGEST AMOUNT EVER AWARDED IN INVESTMENT ARBITRATION

The arbitration against the Russian Federation initiated by some shareholders of the Yukos group,47 is the culmination of disputes which arose from the re-nationalisation of Yukos, the largest oil company in the Russian Federation.48 It is a landmark case as it highlights uncertain or controversial issues in an investment arbitration in which the stakes were very high. The arbitration resulted in an award of US$50 billion plus interest (accruing since 180 days after the award of 18 July 2014) in favour of the claimants. The controversial issues included the valuation date of an expropriation carried out through a series of measures, the method of valuation – not the DCF-method but a method based on share prices, the valuation of contributory negligence, and the calculation of interest. In its decision on the merits the Tribunal came to the conclusion, as had other tribunals engaged in arbitrations by other Yukos shareholders,49 that Yukos had been indirectly and unlawfully expropriated by the respondent.50 With regard to the valuation date, the Tribunal addressed two issues, namely: (a) the date of the expropriation of the claimants’ investment by the respondent; and (b) whether the claimants were entitled to choose between a valuation based on the date of expropriation and a valuation based on the date of the award.51 As the expropriation of Yukos was the result of a number of measures by the respondent over a longer period of time, identifying a specific expropriation date represented a particular challenge. The Tribunal pointed out that both parties had agreed that the date of the

47 In the case at hand, these were Hulley Enterprises Limited (Cyprus), Yukos Universal Ltd (Isle of Man), and Veteran Petroleum Limited (Cyprus). The three claimants filed arbitrations separately but the arbitral proceedings were conducted and decided by the same tribunal. The three awards were published separately but with almost identical wording. See Hulley Enterprises Limited (Cyprus) v Russia, Award (18 July 2014) PCA Case No AA 226; Yukos Universal Ltd (Isle of Man) v Russia, Award (18 July 2014) PCA Case No AA 227; Veteran Petroleum Limited (Cyprus) v Russia, Award (18 July 2014) PCA Case No AA 228 (hereafter: Yukos v Russia). 48 See also RosinvestCo UK Ltd v Russia, Final Award (12 September 2010) SCC Arbitration V (079/2005); Quasar de Valores SICA SA and others v Russia, Award (20 July 2012) SCC Arbitration No 24/2007. 49 ibid. 50 Yukos v Russia, above (n 48) paras 1580–1585. 51 ibid para 1760.

450 Irmgard Marboe expropriation should be ‘the date on which the incriminated actions first lead to a deprivation of the investor’s property that crossed the threshold and became tantamount to expropriation’.52 The claimants had submitted that this was 21 November 2007, the day on which Yukos was struck off the Russian register of legal entities. The respondent did not propose an alternative date but argued that claimants had lost effective control over their investment long before this date.53 The Tribunal agreed with the respondent that ‘the threshold to the expropriation of Claimants’ investment was crossed earlier than in November 2007’ and held that ‘a substantial and irreversible deprivation of Claimants’ assets occurred on 19 December 2004, the date of the YNG auction [which was] Yukos’ main production asset’.54 The determination of 19 December 2004 as the date of the expropriation differs from the award in Quasar de Valores v Russia which, when confronted with the same facts, had accepted the claimants’ submissions that it was 23 November 2007.55 In its reasons, the Tribunal in Yukos v Russia did not mention the different choice made by the earlier Tribunal, but contended that ‘the auction on that date […] marked a substantial and irreversible diminution of Claimants’ investment’, that the claimants had ‘lost the power to govern the financial and operating policies of Yukos so as to obtain the benefits from its activities’ in December 2004, and that Yukos had become ‘incapable of operating as a business’.56 The Tribunal in Rosinvest v Russia, also confronted with the same facts,57 had chosen yet another valuation date. The Tribunal detached the date of the expropriation from the valuation date pointing out that the claimant had made a speculative investment in Yukos at a time when the market had overreacted to transient events and the price of shares was unjustifiably low.58 The Tribunal did not accept claimant’s alleged optimistic expectations regarding the future development of the value of the investment and decided that, for valuation purposes, the date must be applied where the economic risk connected to the investment was taken over by claimant.59 The different approaches taken by the three Tribunals can be analysed in light of Article 15 of the ILC Articles on State Responsibility, which stipulates that when a breach consists of a composite act it is necessary to decide which action or omission is sufficient to constitute the wrongful act based on the precise facts of the case.60 This does not necessarily have to be the last in the series.61 Article 15 leaves a rather wide margin of discretion to arbitral tribunals

52 ibid

para 1761. paras 1736–1738. 54 ibid para 1762. 55 Quasar de Valores v Russia, above (n 3) para 189; see the Tribunal’s conclusion at para 215. 56 Yukos v Russia, above (n 48) para 1762, with references to other parts of the award and exhibits presented by the parties. 57 The claimant had submitted that the expropriation of Yukos’ assets had commenced with the auction of YNG on 19 December 2004 and had been concluded with the Russian Federation’s auction of the last of Yukos’ assets during the final bankruptcy auction on 15 August 2007. Rosinvest v Russia, para 391. 58 ibid paras 666–669. 59 ibid para 672. According to the tribunal this occurred when an internal ‘Participation Agreement’ with claimant’s parent company was terminated. ibid para 673. 60 Article 15(1) of the ILC Articles on State Responsibility provides: ‘The breach of an international obligation by a State through a series of actions or omissions defined in aggregate as wrongful occurs when the action or omission occurs which, taken with the other actions or omissions, is sufficient to constitute the wrongful act.’ See Responsibility of States for Internationally Wrongful Acts, Annex to General Assembly Resolution 56/83 of 12 December 2001 (ILC Articles on State Responsibility). 61 See Commentary to Art 15(1): ‘Paragraph 1 of article 15 defines the time at which a composite act ‘occurs’ as the time at which the last action or omission occurs which, taken with the other actions or omissions, is sufficient to constitute the wrongful act, without it necessarily having to be the last in the series.’ United Nations, ‘State Responsibility’ (2001) 2(2) Yearbook of the International Law Commission 20, 63. 53 ibid

Reparation Cases: Applicable Principles in International Investment Arbitration 451 for the determination of the expropriation date in cases of indirect expropriation. The arguments and results in Yukos v Russia and Quasar de Valores are different, but can in principle be aligned with the main idea contained in Article 15. By contrast, the Tribunal in Rosinvest v Russia took a different approach and was foremost concerned with avoiding rewarding an allegedly speculative investment made by the claimant. As to the right to choose the valuation date, the Tribunal in Yukos v Russia followed the claimants’ submissions and held that ‘in the case of an unlawful expropriation, as in the present case, Claimants are entitled to select either the date of expropriation or the date of the award as the date of valuation’.62 It pointed out that the standard of compensation provided in Article 13(1) ECT was applicable only to lawful expropriations and inferred from this, conversely, that damages for an unlawful taking need not be calculated as of the date of taking. Affirming the approach taken in ADC v Hungary63 it contended that ‘conflating the measure of damages for a lawful taking with the measure of damages for an unlawful taking is, on its face, an unconvincing option’.64 With respect to the valuation method, the Tribunal did not consider the claimants’ DCF model as sufficiently reliable and found that the comparable transactions method could not ground the determination of damages either, primarily because there were no comparable transactions.65 After having considered various valuation approaches it concluded that the ‘comparable companies’ measure as put forward by the claimants and as commented by the respondent’s expert was the best available estimate for the value of Yukos.66 The comparable companies approach was based on data available for a pool of Russian and international oil companies with characteristics similar to Yukos, in terms of production, reserves, profitability, revenue, growth and financing structure. The claimants’ expert established the ratio between the enterprise value of these companies and relevant operating or financial metrics (EBITDA,67 reserves and production) and then applied these ratios to the relevant metrics of Yukos. The net income, EBITDA, reserves and production of Yukos were derived from the ‘pro-forma financial statements’ prepared for the DCF method.68 The Tribunal applied the RTS Oil and Gas Index, an index based on prices of trades executed in securities admitted to trading on the Moscow Stock Exchange,69 to which both parties had referred to as a reliable indicator reflecting the changes in the value of Russian oil and gas companies.70 As regards the value of the lost dividends from shareholding, the Tribunal compared their value up to the expropriation date with their value up to the date of the award and added pre-award interest.71 It came to the conclusion that the damage calculated as of the expropriation date was US$21.988 billion, in contrast to US$66.694 billion as of the date of the award, taking 20 June 2014 as a proxy.72 Since the claimants were entitled to the higher of the two amounts, the Tribunal took the latter sum as the basis for the award.73 62 Yukos

v Russia, above (n 48) para 1763. para 1769. para 1765. 65 ibid para 1785. 66 ibid paras 1784, 1787. 67 Earnings Before Interest, Tax, Depreciation and Amortisation. 68 ibid para 1715. 69 Russia Trading System Oil and Gas Index. 70 ibid para 1788. 71 ibid para 1792. 72 ibid para 1826. The calculations are derived from tables which are contained in the Annexes to the award. 73 ibid paras 1826–1827. 63 ibid 64 ibid

452 Irmgard Marboe It is notable that the Tribunal did not at all mention the price of the Yukos shares at the various stages of expropriation and thereafter, even though both in Rosinvest v Russia and Quasar de Valores v Russia the prices of Yukos’ shares had been the determining factors in the calculation of damages.74 The Tribunal in Yukos v Russia did not apply the DCF method either, even though this method had become increasingly used in international investment arbitration and there was a proven record of profitability and success of the economic activities of Yukos. Finally, the Tribunal in Yukos v Russia addressed the alleged ‘illegal and bad faith conduct’ by claimants starting from the privatisation of Yukos in the mid-1990 until its liquidation.75 Referring to Article 39 of the ILC Articles on State Responsibility according to which ‘account shall be taken of the contribution to the injury by willful or negligent action or omission of the injured’ the Tribunal pointed out that not any contribution would trigger a finding on contributory fault but that the contribution must be ‘material and significant’. The Tribunal decided that, out of the 28 actions or omissions argued by the respondent, two of them, namely Yukos’ ‘abuse of the low-tax regions by some of its trading entities’ and its ‘questionable use of the Cyprus-Russia DTA’, stood out as having contributed in a material way to the prejudice subsequently suffered.76 The Tribunal needed then to determine to what extent this conduct could ‘lessen the responsibility of Respondent’77 and to quantify the contribution. In so doing the Tribunal referred to the ICSID Annulment Committee in MTD v Chile, which emphasised that the role of the two parties contributing to the loss is ‘only with difficulty commensurable and the Tribunal [has] a corresponding margin of estimation’.78 It then decided, ‘in the exercise of its wide discretion’79 that the claimants had contributed to the prejudice which they suffered as a result of the destruction of Yukos to the extent of 25 per cent.80 The rough estimation of contributory fault appears problematic, in particular in view of the very high amount of damages at stake. However previous tribunals have also used overall percentages.81 In the present case, a more precise quantification could probably have been achieved by reference to comparable benchmarks. The ‘abuse of the low-tax regions by some of its trading entities’, for example, could be quantified by comparing the extraordinary tax advantages with more normal tax paying regimes. The effects of the ‘questionable use of the Cyprus-Russia DTA’82 could possibly be levelled out by eliminating or reducing the tax benefits granted therein. 74 In these two cases, the claimants had submitted that the value of their shares in Yukos was best represented through the expectation to receive dividends, which, as such, was priced in the value of the shares. 75 Already in the discussion on jurisdiction and admissibility, the respondent had listed 28 instances of alleged misconduct but the tribunal decided that it would deal with them in the merits phase of the arbitration, most importantly during the assessment of liability and damages. ibid para 1374. 76 ibid para 1634. 77 ibid para 1635. This seems to suggest that the responsibility of the respondent as such was ‘lessened’. However, the behaviour of the claimants did not influence the respondent’s responsibility for the breach of its treaty obligations as such. Only the amount of damages could be ‘lessened’. This corresponds to the systematic order of the ILC Articles that put Art 39 on ‘Contribution to the Injury’ in the context of the other articles on ‘Reparation’. See Arts 34–39 of the ILC Articles on State Responsibility. 78 MTD Equity Sdn Bhd v Chile, Award (25 May 2004) ICSID Case No ARB/01/7, paras 243, 246. 79 Yukos v Russia, above (n 48) para 1637. 80 ibid para 1637. 81 For example, the tribunal in Occidental Petroleum v Ecuador also applied 25%, while in MTD v Chile it was 50%. Occidental Petroleum Corporation v Ecuador, para 687; MTD Equity Sdn Bhd v Chile, Award (25 May 2004) ICSID Case No ARB/01/7, paras 243, 246. By contrast the tribunals in Rosinvest v Russia and in Quasar de Valores v Russia, confronted with the same facts as in the present case, did not refer to the concept of ‘contributory negligence’ at all. 82 Double Taxation Agreement.

Reparation Cases: Applicable Principles in International Investment Arbitration 453 The Tribunal in Yukos v Russia correctly pointed out that the concept of contributory fault must not be confused with the investor’s duty to mitigate its losses, a duty that arises after a breach of an obligation has occurred. Furthermore, the fault of claimants may not necessarily be related to the wrongdoing of the respondent, but nevertheless reduce the latter’s obligation to pay damages. Finally, as the Tribunal considered relevant in the case at hand, the victim may have provoked the state’s wrongful conduct. An alleged provocation, however, does not abrogate the state’s responsibility but may decrease the amount of damages payable by it. In Yukos v Russia, contributory fault resulted in a decrease of the amount of damages from US$66.694 billion to US$50,021 billion. In addition, it needs to be mentioned that the award in Yukos v Russia stands out for its detailed reasoning on interest.83 The Tribunal was not convinced by either of the parties’ submissions on the interest rate. It therefore carried out its own analysis of the pertinent legal and economic factors and extensively discussed the most suitable approach for the selection of an appropriate rate.84 It evaluated the borrowing rate of the investor, the prime rate, inter-bank rates, the borrowing rate of the respondent, and the alternative investment rate. After having considered the various approaches, it decided that the ‘investment alternative approach’ was the most suitable in the present case.85 The Tribunal referred to US Treasury bond rates as an appropriate ‘alternative investment’, which several other tribunals had also chosen.86 It determined that the rate should be equal to the average yield of ten-year US Treasury bonds in the period between 1 January 2005 and 30 May 2014, which was 3.389 per cent.87 It found support of this choice in the fact that the US was still the reference economy for many international investors, that the US dollar was still the most important reference currency for international transactions, and that, most relevant for the present case, all the calculations of claims, including the share values and dividends, were calculated in US dollars. The choice is however not entirely convincing because it assumes that investors would put their money in US treasury bonds which in reality is rather unlikely.88 With regard to the question of compounding, the Tribunal held that pre-award interest on the stream of dividends foregone should not be compounded in spite of the fact that awarding compound interest meanwhile represented a form of jurisprudence constante in investor-state arbitration.89 By contrast, it decided that, after a grace period of 180 days, post-award interest at a rate of the ten-year US Treasury bonds should be compounded annually.90 In view of the compensation of the loss of shares, similar to ADC v Hungary, no pre-award interest was awarded. 83 Yukos

v Russia, above (n 48) paras 1638–1692. so doing the tribunal explicitly referred to two books, Ripinsky and Kevin Williams, above (n 2) and I Marboe, Calculation of Compensation and Damages in Investment Arbitration, 2nd edn (Oxford, Oxford University Press, 2017), whose arguments it discussed in detail. 85 ibid para 1682–1684, referring to Siemens v Argentina (2002) ICSID Case No ARB/02/8, para 396; Compañía de Desarollo de Santa Elena v Costa Rica, Award (17 February 2000) ICSID Case No ARB/96/1, para 104; Alpha Projektholding GmbH (Austria) v Ukraine, Award (8 November 2010) ICSID Case No ARB/07/16, para 514; EDS International SA, SAUR Internationl SA, Leon Participaciones Argentinas SA v Argentina, Award (11 June 2012) ICSID Case No ARB/03/23, paras 1325 et seq; Gemplus SA et al v Mexico, Award (16 June 2010) ICSID Case Nos ARB(AF) 04/3 and 04/4, paras 16–24. 86 Yukos v Russia, above (n 48) para 1684–1685. 87 ibid para 1687. 88 See on other investment alternatives Marboe, above (n 84) 356 et seq, with further references to pertinent arbitral practice. 89 ibid para 1689. 90 ibid paras 1691–1692. 84 In

454 Irmgard Marboe IV. NECESSITY AND THE ASSESSMENT OF DAMAGES: LG&E V ARGENTINA

The arbitration in LG&E v Argentina is one of the few in which the Tribunal affirmed the existence of a ‘state of necessity’ in Argentina during its economic crisis of the early 2000s.91 Based on Article 25 of the ILC Articles on State Responsibility and Article XI of the BIT between the US and Argentina the Tribunal concluded that Argentina was justified in not observing its international obligations with respect to the claimants’ investment for a certain period of time.92 What did this mean for Argentina’s obligation to make reparation? According to Article 27 of the ILC Articles on State Responsibility, the invocation of a circumstance precluding wrongfulness ‘is without prejudice to […] the question of compensation for any material loss caused by the act in question’.93 On the other hand, Article XI of the BIT was a so-called ‘non-precluded measures provision’ (NPM)94 which stated: This Treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfilment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.

In its decision on liability the Tribunal primarily relied on Article IX of the BIT, but in its interpretation and application it used the conditions developed under customary international law for determining a state of necessity.95 Despite the different wording of Article 25 and of the NPM clauses in BITs, the respective conditions were also applied vice versa by other investment tribunals, whose decisions were sometimes annulled in whole or in part because of this imprecision.96 In LG&E v Argentina the Tribunal came to the conclusion that the conditions of Article XI were met and Argentina was justified in its action in contravention to its international obligations during a certain period of time. It would therefore not have been necessary to discuss the effect of Article 27 of the ILC Articles on the issue of reparation. The Tribunal nevertheless referred to the ILC Articles in its reasoning: 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 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 in what circumstances compensation would be payable’. The rule does not specify

91 For furthr discussion of the cases against Argentina in the wake of the economic crisis of 2000 see A Reinisch and J Tropper, ‘The Argentinian Crisis Arbitrations’, ch 8 in this book. 92 LG&E Energy Corp, LG&E Capital Corp and LG&E International Inc v Argentina, Decision on Liability (3 October 2006) ICSID Case No ARB/02/1, para 266. 93 The ILC, in its Commentary to Art 27, refers to the ICJ’s judgment in Gabcíkovo-Nagymaros Project (Slovakia v Hungary) (Judgment) [1997] ICJ Reports 1997, 39, para 48, where the ICJ noted that ‘Hungary expressly acknowledged that, in any event, such a state of necessity would not exempt it from its duty to compensate its partner’, Report of the International Law Commission on the work of its fifty-third session (2008) UN, 86. 94 See Wi Burke-White and A von Staden, ‘Investment Protection in Extraordinary Times: The Interpretation and Application of Non-Precluded Measures provisions in Bilateral Investment Treaties’ (2008) 48 Virginia Journal of International Law 307. 95 LG&E Energy Corp v Argentina, Decision on Liability, above (n 92) paras 226 et seq, in particular para 257. 96 See, eg, Sempra v Argentina, Decision on the Application for Annulment (29 June 2010) ICSID Case No ARB/02/16, paras 159–185. By contrast, the ad hoc committee in CMS v Argentina limited itself to heavily critising this lack of distinction, but ultimately refrained from annulment. See CMS v Argentina, Decision on the Application for Annulment (25 September 2007) ICSID Case No ARB/01/8, paras 134–136. Another award annulled for misconception of the state of emergency was Enron v Argentina, Decision on the Application for Annulment (30 July 2010) ICSID Case No ARB/01/3, paras 347–405. See the discussion in I Marboe, Calculation of compensation and damages in international investment law, 2nd edn (Oxford, Oxford University Press, 2017) 160–162.

Reparation Cases: Applicable Principles in International Investment Arbitration 455 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 provides the answer.97

In its conclusion the Tribunal again used the state of necessity to explain the effects of Article XI: 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. This exception is appropriate only in emergency situations; and once the situation has been overcome, i.e. certain degree of stability has been recovered; the State is no longer exempted from responsibility for any violation of its obligations under the international law and shall reassume them immediately.98

When the Tribunal turned to the consequences of this conclusion with respect to the issue of reparation it noted that neither Article 27 of ILC’s Draft Articles nor Article XI of the Treaty specified whether any compensation was payable to the party affected by losses during the state of necessity.99 In view of its earlier considerations it decided that the damages suffered during the state of necessity ‘should be borne by the investor’.100 However, it pointed out that the state of necessity only lasted from 1 December 2001 until 26 April 2003. During that period Argentina was exempt of responsibility so that the claimants had to bear the consequences of the measures taken by the state. The Tribunal emphasised that Argentina should have restored the tariff regime on 27 April 2003, after the end of the state of necessity. The exemption from liability was thus only limited to the specific period of time which the Tribunal had identified as an emergency situation. Immediately after its end, the original situation should have been restored. It was therefore the remaining period of time, which the Tribunal took as relevant for compensating the investors for their losses. In the final award, the Tribunal concentrated on the assessment of damages, interest, and costs.101 It decided that the appropriate standard for reparation was ‘full’ reparation as set out by the PCIJ in the Factory at Chorzów case and codified in Article 31 of the ILC Articles on State Responsibility.102 While the claimants had requested compensation by the fair market of their loss, the Tribunal explained that the fair market value was inapplicable in the case at hand.103 It noted, amongst others, that fair market value was referred to in the BIT as the measure of compensation in cases of expropriation, and its application did not extend similarly to other treaty standards.104 Furthermore, it suggested that there might be a difference between ‘compensation’ as the consequence of a legal act and ‘damages’ as the consequence of the committing of a wrongful act, a distinction, which had been noticed by various tribunals.105 If fair market value was not the proper measure of compensation for unlawful expropriation, it would a fortiori not be appropriate for breaches of other 97 LG&E Energy Corp v Argentina, Decision on Liability, above (n 92) para 260 (footnote omitted, emphasis in original). 98 ibid, para 261. 99 ibid para 264. 100 ibid. 101 LG&E Energy Corp, LG&E Capital Corp and LG&E International Inc v Argentina, Award (25 July 2007) ICSID Case No ARB/02/1. 102 ibid para 31. 103 ibid paras 33 et seq. 104 ibid para 37, referring to SD Myers v Canada, UNCITRAL, First Partial Award (13 November 2000) para 307. 105 ibid para 38, with references to literature and a number of arbitral awards.

456 Irmgard Marboe treaty standards,106 such as the breach of the obligation to fair and equitable treatment, non-discrimination, and umbrella clause contained in the case at hand. Instead of the fair market value the Tribunal chose the ‘actual loss’ incurred as a result of the wrongful acts as the appropriate measure of compensation.107 The core question was that of ‘causation’.108 As a starting point, the Tribunal identified the abrogation of the specific guarantees provided by Argentina in the gas regulatory framework as the fundamental act giving rise to the breach of the BIT obligations. In particular, it considered that: (i) the abolition of the right to calculate tariffs in dollars before conversion to pesos; (ii) the abandonment of the Producer Price Index (PPI) adjustments; (iii) the suspension of the tariff reviews; and (iv) the forced renegotiation of the licences violated the standard of the fair and equitable treatment and the umbrella clause and resulted in discriminatory treatment against the gas distribution companies.109 With respect to the valuation method, the Tribunal found that neither the stock market share price nor the DCF method would properly account for the accrued loss.110 Instead, it decided to assess the ‘dividends’ lost as the most appropriate measure of damage and summarised in a procedural order: A calculation will be made of the dividends that would or could have been generated without any change in the tariff system. Dividends received by the Claimants will be subtracted from this figure, after which the damages suffered during the State of Necessity will be subtracted from this amount.111

With respect to the valuation date, the Tribunal first considered the entire period between 18 August 2000, the date of the injunction to suspend the PPI adjustments,112 and 28 February 2005, the date of the post-hearing brief.113 It then deducted the losses incurred during the ‘State of Necessity’ (between 1 December 2001 und 26 April 2003).114 In addition, the Tribunal awarded pre-award interest, disagreeing with the respondent’s contention that payment of interest from 2000 would result in double recovery. It stated that interest was due on the amount of dividends that the claimants would have received but for abrogation of the tariff regime minus the dividends actually received. It underlined that ‘[l]ost dividends compensate claimants for Argentina’s breach and interest compensates Claimants for the impossibility to invest the amounts due.’115 After a grace period of 30 days post-award interests were to be paid at a rate of six-month US Treasury bills, compounded, until the date of payment.116 While this award has become known for its alleged application of the state of necessity as a circumstance precluding wrongfulness as codified in Article 25 of the ILC Articles, this is not entirely correct. The Tribunal primarily relied of the NPM-clause contained in the BIT and effectively found that the respondent state was justified in not complying with the BIT during a certain period of time. Therefore, the question of the secondary obligation to reparation and

106 ibid. 107 ibid

paras 41 et seq. para 45. 109 ibid para 46. 110 ibid para 59 111 ibid. 112 ibid para 49. 113 ibid para 94. 114 ibid paras 61, 108. 115 ibid para 102. 116 ibid. 108 ibid

Reparation Cases: Applicable Principles in International Investment Arbitration 457 compensation in accordance with Article 27 of the ILC Articles on State Responsibility did actually not arise. However, the Tribunal itself contributed to this misunderstanding by applying the conditions of the state of necessity to the BIT-clause. Regarding the result, however, the decision cannot be criticised as the ILC Articles leave the question of compensation open and may well lead to the non-payment of damages. When it comes to the calculation of damages, the decision taken by the Tribunal represents a landmark case for its clarity and consistency as well as for its detailed and logical reasoning. The Tribunal did not automatically apply the standard of fair market value or a valuation method only because they were presented or even agreed by the parties. In this sense the award in LG&E v Argentina shows that iura novit curia exists also for the determination of the applicable principles in reparation cases.

V. ASSESSING COUNTRY RISK: TIDEWATER V VENEZUELA

Several investment arbitration cases against Venezuela in the wake of the ‘Bolivarian Revolution’ initiated by Hugo Chavez after the turn of the millennium raised the question to what extent ‘country risk’ should be included for valuation purposes. In times of political change, public announcements directed against foreign investors may represent a threat of expropriation. Can these announcements be used to decrease compensation once the expropriation, directly or indirectly, materialises? One of the landmark cases in this respect is Tidewater v Venezuela, an award rendered under the auspices of ICSID in 2015.117 Typically, international investment protection treaties stipulate that, in case of expropriation, ‘compensation shall amount to the market value of the investment expropriated immediately before the expropriation or before the impending expropriation become public knowledge, whichever is earlier’.118 The reason is that the expropriation itself or an impending expropriation decreases the value of property considerably. Therefore, the common standard in investment protection treaties is to exclude such a negative influence from the valuation in order to ensure that the expropriated investor receives ‘adequate compensation’, as intended by the treaty drafters.119 The effect of negative commercial prospects is particularly relevant for valuation methods that rely on future income of cashflow, as opposed to historical cost approaches.120 One possibility to reflect risk and uncertainty is the ‘discount factor’. Several tribunals in the cases against Venezuela were confronted with submissions by the parties as to how the country risk should be reflected in the discount factor.121 The Tribunal in Tidewater v Venezuela noted that ‘the element of greatest difference between the approaches of the experts is the premium 117 Tidewater Investment Srl and Tidewater Caribe, CA v Venezuela, Award (13 March 2015) ICSID Case No ARB/10/5. 118 This is, for example, the wording of Art 5 of the agreement between Barbados and Venezuela for the promotion and protection of investments (Signed on 15 July 1994, entered into force on 31 October 1995) [Barbados-Venezuela BIT], which was applied in Tidewater v Venezuela. 119 See T Wälde and B Sabahi, ‘Compensation, Damages, and Valuation’ in P Muchlinski, F Ortino and C Schreuer (eds), The Oxford Handbook on International investment Law (Oxford, Oxford University Press, 2008) 1049, 1081–1082. 120 With respect to the different valuation approaches based on market transactions, future income, or historical costs, see I Marboe, Calculation of compensation and damages in international investment law, 2nd edn (Oxford, Oxford University Press, 2017) 184 et seq. 121 See also Mobil Cerro Negro v Venezuela, Award (9 October 2014) ICSID Case No ARB//07/27; Gold Reserve v Venezuela, Award (22 September 2014) ICSID Case No ARB(AF)/09/1; Flughafen Zürich v Venezuela, Award (18 November 2014) ICSID Case No ARB//10/19; OI European Group v Venezuela, Award (10 March 2015) ICSID Case No ARB//11/25.

458 Irmgard Marboe to be applied to the risk of investing in a particular country, here Venezuela’.122 The claimants’ expert had adopted a country risk premium of 1.5 per cent, in contrast to the respondent’s 14.75 per cent. He explained that the country risk premium was rather low because political risk ought to be excluded from it. The existence of the investment protections contained in the Venezuela- Barbados BIT meant that the ‘real risks of the Government acting in a very negative way towards any private investment’123 should be excluded. To include such risks would confer an illegitimate benefit on the state. The Tribunal rejected this view pointing out that the BIT did not prohibit all state taking of private property. It only prescribed that in such a case the state must pay compensation in the amount of the ‘market value’ of the investment.124 According to the Tribunal, it should consider the value that a willing buyer would have placed on the investment. In determining this value, one element that a buyer would consider is the risk associated with investing in a particular country. Such a factor is not specific to the particular State measure that gives rise to the claim […] Rather the country risk premium quantifies the general risks, including political risks, of doing business in the particular country, as they applied on that date and as they might then reasonably have been expected to affect the prospects, and thus the value to be ascribed to the likely cash flow of the business going forward.125

The Tribunal found that, in case of a lawful expropriation, ie complying with the conditions of lawful expropriations under the BIT, the amount of compensation must reflect how the market evaluates the general risk of expropriation before the actual expropriation materialised, which may lead to a decrease of the amount of compensation. In the case before it the Tribunal found that the expropriation was lawful because it was carried out on the basis of a law which also provided for compensation. It noted that there was no refusal of the state to pay compensation, rather the parties were unable to agree on the bases of valuation and the procedure.126 The Tribunal therefore decided that it should determine the amount of compensation according to the standard provided in the BIT. In accordance with this line of argument, cases of unlawful expropriations and other BIT breaches, by contrast, would require full reparation and the financial situations with and without the unlawful act would have to be compared. The Tribunal in Gold Reserve v Venezuela represents a suitable example because, after having found that Venezuela had breached Article 2 of the Canada–Venezuela BIT by terminating the concessions without proper reasoning and by physically seizing the claimant’s assets, actions which the Tribunal considered as ‘particularly egregious’,127 it agreed with the claimant’s expert that ‘it is not appropriate to increase the country risk premium to reflect the market’s perception that a State might have a propensity to expropriate investments in breach of BIT obligations’.128 Similarly, in Flughafen Zürich v Venezuela, the Tribunal decided that the measures adopted by Venezuela constituted a ‘direct expropriation, more concretely, a nationalization of the investment’.129 The measures were not lawful, because they were in violation of Venezuelan

122 Tidewater

v Venezuela, above (n 117) para 182. para 183. para 185. 125 ibid para 186. 126 ibid para 145. 127 Gold Reserve v Venezuela, above (n 121) para 615. 128 ibid para 841. 129 Flughafen Zürich v Venezuela, above (n 121) para. 509. 123 ibid 124 ibid

Reparation Cases: Applicable Principles in International Investment Arbitration 459 law, as confirmed by the national administrative court, and were not accompanied by the payment of compensation.130 It can be concluded that Tidewater v Venezuela refines the approaches discussed earlier with respect to the distinction between compensation for lawful expropriations and full reparation for unlawful expropriations. Even in situations of political turmoil, states are called to observe their international obligations with respect to due process and non-discrimination  – which are generally the other conditions for lawful expropriations – and are rewarded with a lower amount of compensation. As the outcomes in the cases involving Venezuela demonstrate, the lack of agreement over a specific amount of money alone does not necessarily render the expropriation unlawful.

VI. CONCLUSION

The brief overview of a few landmark cases in international investment arbitration shows that the assessment of reparation in investor-state disputes plays an increasingly important role. Tribunals have refined their ways and means of arriving at an outcome in a remarkable way. Compared to earlier practice where tribunals often dedicated very little time and space to their considerations on valuation, the more recent awards demonstrate how important the discussion on quantum has become. This is not only true for the respective parties of the arbitration who seek a solution of their dispute but also to the economic and legal communities at large. The nature of reparation follows the rules of the law of state responsibility, which despite some doubts to this effect have been repeatedly confirmed by arbitral tribunals. While the principle of ‘full reparation’ is paramount, the primacy of restitution provides an important yardstick. Even if in arbitral practice restitution has hardly ever been awarded, its applicability in principle has been confirmed by numerous tribunals. With respect to the assessment of the amount of compensation, the appropriate valuation method and the valuation date have to be chosen so as to arrive as closely as possible at restitution. Numerous tribunals have confirmed that, in their damage assessment, they follow the PCIJ in its seminal judgment in Factory at Chorzów according to which reparation ‘must as far as possible wipe out’ the consequences of the unlawful act. Investment protection treaty clauses on compensation usually address lawful expropriations, thus expropriations, which are undertaken in the public interest, in accordance with due process of law, on a non-discriminatory basis, and against compensation. However, if some of these conditions, most importantly the requirement of due process, are not met, these primary rules determine the minimum amount of compensation, usually the fair market value of the expropriated asset at the expropriation date. While the difference between lawful and unlawful expropriations has not always been evident in the practice of tribunals, some of the cases quoted above have shown in which circumstances the difference does matter and leads to a different financial outcome. The amount of compensation can be reduced by contributory fault on the part of the claimant. Several cases have demonstrated that sometimes acts of the investor are not respectful of national law or contractual obligations. Such acts by the investor do not generally justify a violation of a host state of its international obligations, but they may reduce the state’s

130 ibid

para 511.

460 Irmgard Marboe payment obligations. Tribunals struggle, however, with the quantification of that contribution in practice. Circumstances precluding wrongfulness, such as the state of necessity, can justify the breach of an international obligation of a state but do not necessarily lead to an exemption of compensation. Nevertheless, practice has shown that tribunals are not inclined to order compensation for the time the respective circumstances prevail but want states to resume their obligations as quickly as possible thereafter. The circumstances precluding wrongfulness under custmary international law – as codified by the ILC in its Articles on State Responsibility – should not be confused with so-called clauses addressing NPMs, non-precluded measures, which are allowed in investment protection treaties to safeguard important interests of the host state. With respect to the choice of the valuation method, it has turned out that methods based on expected future income or cashflow have become increasingly accepted in investment arbitration. In addition, for a ‘sanity check’, as one tribunal put it, other methods based on market values or asset values could be used. The criteria for selecting the DCF method, even if some tribunals have formulated proposals to provide clearer guidelines, are not strict and vary from case to case. Finally the importance of the determination of interest should not be underestimated, both of pre-award and post-award interest. So far there is no generally accepted solution for the selection of an interest rate. The predominant view seems to be that the interest rate should reflect the claimant’s alternative investment opportunities. However, tribunals have often not explained clearly which alternative investment would be the most appropriate to choose and why. Another approach would be the borrowing rate of the host state or the choice of certain benchmarks, such as interbank interest rates. Compounding of interest has become increasingly accepted in arbitral practice even though it seems that exceptions are made when the amount awarded is already quite high. Investment arbitration can only fulfil its assigned role if the final – including financial – outcomes of the procedures are based on objectively verifiable and understandable principles. While valuation will never be an exact science and estimations and assumptions are inevitable, well-reasoned awards will continue to improve consistency and predictability of this aspect of international investment arbitration.

27 The Contribution of International Commercial Arbitration to Investment Arbitration GEORGE A BERMANN*

I. INTRODUCTION

I

NVESTMENT ARBITRATION IS structured and has been developed largely on the basis of existing practice in international commercial arbitration. Although the two species of arbitration each exhibit distinctive features – some procedural, such as confidentiality, and others substantive, such as applicable law – the contribution of commercial arbitration to investment arbitration is palpable. This is largely unsurprising. International commercial arbitration had enjoyed centuries of history by the time international investment arbitration came to the fore, and could not help but have a ‘formative impact’ on it.1 There would appear to be no reason why, as international investment arbitration developed into a robust mechanism for the settlement of investment disputes,2 it should shed its affinity with international commercial arbitration. However, history alone does not account for the heavy influence that international commercial arbitration has had, and continues to have, on international investment arbitration. A key factor has also been the high degree of overlap among counsel and arbitrators operating in both fields.3 It is small wonder that, as practitioners expanded their arena of activity to international investment arbitration, they brought with them the methods they had perfected in the conduct of international commercial arbitration. This commonality among actors contributed

* George Bermann is Professor of Law at Columbia Law School and Director of the Center for International Commercial and Investment Arbitration. 1 B Legum, ‘Investment Treaty Arbitration’s Contribution to International Commercial Arbitration’ (2005) 60 Dispute Resolution Journal 70, 74. 2 K Bockstiegel, ‘Commercial and Investment Arbitration: How Different are they Today?’ (2012) 28 Arbitration International 577, 577–578 (‘Investment arbitrations have also existed to some extent for quite some time as we know from older cases. But it became a widely used general field of international dispute settlement only when the first bilateral investment treaties (BITs) were concluded starting in 1959 and when the World Bank initiated the ICID-Convention in 1965.’). 3 GB Born, ‘A New Generation of International Adjudication’ (2012) 61 Duke Law Journal 775, 835 (‘In practice, the procedures used in international commercial arbitration are the model for investment arbitration, including the number and selection of arbitrators, the presentation of evidence, the conduct of hearings, and the awards – in part because of overlaps in the individuals and law firms that serve as arbitrators and counsel in both sets of proceedings.’).

464 George A Bermann powerfully not only to a harmonisation of prevailing practices in the two domains,4 but to what may be termed more generally a common epistemic community.5 The impact of the development of an epistemic community comprising the actors in both international commercial and investment arbitration simply cannot be overstated. While public international law lies at the centre of investment as distinct from commercial arbitration, the actors in that field come predominantly from private international law practice.6 Recent empirical studies show that less than a third of arbitrators sitting in international investment arbitration are specialised in public international law, leaving the lion’s share of activity to commercial law specialists.7 This has resulted in what one scholar characterises as a ‘unique marriage’ of ‘epistemic communities’,8 albeit one in which, from a broadly procedural point of view, the commercial arbitration actors have had the greatest impact.9 The sections that follow document the impact that practice in international commercial has had on the practice of investment arbitration. Section II traces the impact of international commercial arbitration on general principles, identifying the resulting commonalities, but also the divergences due to the special nature of international investment arbitration. Section III then examines a number of particular procedural features that, as will be seen, the two practices mostly have in common. Overall, the contribution of international commercial arbitration to international investment arbitration has been striking.

4 Bockstiegel,

above (n 2) 584–585.

5 According to international relations theory, an epistemic community can be defined as a ‘network of

professionals with recognized expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge within that domain or issue-area’. See PM Haas, ‘Epistemic Communities and International Policy Coordination’ (1992) 46 International Organization 1, 3. See also SI Strong, ‘Clash of Cultures: Epistemic Communities, Negotiation Theory and International Lawmaking at the United Nations Commission on International Trade Law (UNCITRAL)’ (2016) 50 Akron Law Review 495. 6 SW Schill, ‘W(h)ither Fragmentation? On the Literature and Sociology of International Investment Law’ (2011) 22 The European Journal of International Law 875, 888 (‘In a sense, investment treaty arbitration has the effect of commercializing international investment law also from a sociological perspective. In fact, with rising numbers of disputes, the centre of gravity increasingly moved to the commercial bar.’). 7 M Waibel and Y Wu, ‘Are Arbitrators Political?’ (2012) SSRN Electronic Journal, www.researchgate.net/publication/256023521_Are_Arbitrators_Political (‘Arbitrators are typically recruited from the pool of individuals with substantial experience in advising clients on international investment law. There is thus little prospect for recruitment for lawyers working outside the large and well-known European and US law firms. More than sixty percent of arbitrators are in full-time private practice. More than half of the arbitrators wear a second hat as counsel to investor in other ICSID arbitrations. Fewer than one third of the arbitrators are full-time academics and specialists in public international law.’). 8 A Roberts, ‘Clash of Paradigms: Actors and Analogies Shaping the Investment Treaty System’ (2013) 107 American Journal of International Law 45, 54–55 (‘The unique marriage of public international law as the applicable law with dispute resolution rules resembling those in international commercial arbitration means that the field was historically populated by two very different professional communities: one from the side of public international law and interstate dispute resolution, and the other from the other side of private law and commercial arbitration.’). However, the author further observes that: ‘As the regulatory impact of the system has become more evident, other professional communities have joined the field, including those with backgrounds in public law, international human rights law, environmental law and trade law. This is particularly evident in the academic sphere where numerous scholars are articulating approaches to the investment field built upon related areas with which they are familiar, such as Schneiderman (constitutional law), Kingsbury (global administrative law), Kurtz (international economic law) and Simma (public international law and human rights). As the investment regime intersects with new domains, like European Union law, we can expect other professional communities to join the field, bringing their analogies with them.’). 9 K Miles, The Origins of International Investment Law: Empire, Environment and the Safeguarding of Capital (Cambridge, Cambridge University Press, 2013) 345 (‘Arbitrators involved in international commercial and investorstate arbitration exert a powerful influence over the culture underlying decision-making in investment disputes.’). See generally M Sornarajah, ‘The Clash of Globalizations and the international law on foreign investment’ (2003) 10 Canadian Foreign Policy Journal 1.

The Contribution of International Commercial Arbitration to Investment Arbitration 465 II. GENERAL PRINCIPLES OF INTERNATIONAL ARBITRATION

It is an inescapable fact that international commercial arbitration has lent international investment arbitration some of its most basic and general characteristics. This section addresses the matter in terms of the general principles prevailing within the international commercial and investment spheres alike – principles that international investment arbitration may be said to have derived from international commercial arbitration. On the other hand, the fact that international commercial arbitration has supplied international investment arbitration some of its most important general features does not mean that international investment arbitration has not taken some of them in a somewhat different direction. That pattern is explored in the section that follows.

A. Commonalities International commercial arbitration’s influence over international investment arbitration manifests itself in a good number of commonalities in general principles. Among them are those that follow. i. The Principle of Equality of Arms A fundamental tenet of procedural law in commercial arbitration is the principle of equality of arms, a principle shared by any institution, such as national courts, overseeing proceedings between horizontally situated parties.10 Equal treatment of the parties is regarded as ‘inherent in consent to arbitration’.11 To be sure, the relationship between states and investors has a strong vertical dimension. Even while participating in an arbitral proceeding, the state retains its governing authority, which includes authority to take measures antagonistic to an investor’s interests.12 States have a continuing interest in doing what they can to preserve public resources that may be diverted by adverse awards of damages.13 The possibility also cannot

10 See, eg, HLA Hart, The Concept of Law, 3rd edn (Oxford, Clarendon Press, 2012) 160 (‘the structure of the idea of justice … consists of two parts: a uniform or constant feature, summarized in the precept “Treat like cases alike” and a shifting or varying criterion used in determining when, for any given purpose, cases are alike or different.’); M Scherer, D Prasad and D Prokic, ‘The Principle of Equal Treatment in International Arbitration’ (2018) 1 (‘the principle of equal treatment is fundamental to the idea of justice, in international arbitration as in any adjudicative system. Arbitration conventions, rules and national laws unanimously impose a requirement, either express or implied, that the parties be treated equally throughout the arbitral process.’) (citations omitted). 11 Roberts, above (n 8) 55 (‘the “fundamental equality that is inherent in consent to arbitration” means that arbitration law and tribunals should not accord states certain privileges or deference unless clearly required by the governing law.’) (citation omitted). 12 P Lalive, ‘Some Threats to International Investment Arbitration’ (1986) 1 ICSID Review – Foreign Investment Law Journal 26, 37 cited by TW Wälde, ‘‘Equality of Arms’ in Investment Arbitration: Procedural Challenges’ in K Yannaca-Small (ed), Arbitration Under International Investment Agreements: A Guide to Key Issues (Oxford, Oxford University Press, 2010) 163, above (n 5) (‘Lalive noted the “dual role” of the state, as contract party (in contractual arbitration) or for purposes of investment promotion by guarantees of protection it promises submission to an adjudicatory regime it does not control; but, as government, sovereign and involved in the domestic political process, it wishes to avoid, evade, frustrate and, most of all, not lose a case initiated by a foreign investor.’). See also generally R Boivin, ‘International Arbitration with States: An Overview of the Risks’ (2002) 19 Journal of International Arbitration 285; H Fox, ‘States and the Undertaking to Arbitration’ (1988) 37 ICLQ 1. 13 Wälde, above (n 12) 163 (‘A loss of such an international arbitration claim can be, therefore, politically very embarrassing. The government, and the particular politicians and civil servants responsible, will therefore be under immense pressure not to lose – at least not during their tenure.’).

466 George A Bermann be excluded that tribunals may show a modest degree of procedural deference to states, for example, demonstrating somewhat greater liberality in the grant of extensions.14 Even so, the principle of equality of arms characteristic of international commercial arbitration is fundamentally intact in international investment arbitration as well.15 In its annulment decision in the case of Malicorp v Egypt, the ad hoc committee had this to say: The Committee notes that the Parties agree on the substance of the principe du contradictoire and on the fact that it is a rule of procedure that ensures equality of the parties in an adversarial proceeding. The Committee further notes that this principle is closely related to the right to be heard. This right of parties to present their case has been recognized as part of that ‘set of minimal standards’ considered fundamental for a fair hearing. The Committee thus concludes that the principe du contradictoire is a fundamental rule of procedure.16

ii. Arbitrator Appointment and Challenge Many would argue that, in international arbitration, the most important decision that counsel can make is the selection of arbitrators.17 On this aspect of international arbitration, international investment arbitration has largely imported the mechanisms and standards long prevailing in international commercial arbitration.18 The main distinction one sees is a purely factual one. International commercial arbitrators generally range in their work over a wide variety of fields of commerce and industry, though concentration in certain sectors (such as construction, energy, mining and intellectual property) is far from unknown. In investment arbitration, the issues – such as whether an expropriation occurred, or fair and equitable treatment was denied – regularly recur, as do general principles of public international law, and, thanks to the transparency that characterises investment arbitration, the views of arbitrators on those issues can become quite well known.19 There also may be some differences among standards of independence and impartiality. For example, the International Centre for Settlement of Investment Disputes (ICSID) Convention requires only that arbitrators

14 See M Hirsch, ‘Invitation to the Sociology of International Law’ in Z Douglas, J Pauwelyn and JE Viñuales (eds), The Foundations of International Investment Law: Bringing Theory into Practice (Oxford, Oxford University Press, 2014) 147 (‘The present-day investment arbitration community (including investment tribunals) is dominated by the commercial arbitration paradigm.’) (citations omitted). 15 TW Wälde, ‘Procedural Challenges in Investment Arbitration under the Shadow of the Dual Role of the State: Asymmetries and Tribunals’ Duty to Ensure, Pro-actively, the Equality of Arms’ (2010) 26 Arbitration International 3, 39. 16 Malicorp Limited v The Arab Republic of Egypt, Decision on the Application for Annulment of Malicorp Limited (3 July 2013) ICSID Case No ARB/08/18, para 36. 17 AJ Van Der Berg, ‘Qualified Investment Arbitrator? Comment on Arbitrators in Investment Arbitrations’ in P Wautelet, T Kruger and G Coppens (eds), The Practice of Arbitration: Essays in Honour of Hans van Houtte (Oxford, Hart Publishing, 2012); RH Smit, ‘Thoughts on Arbitrator Selection: Why my Father was usually a Good Choice’ (2012) 23 American Review of International Arbitration 575. 18 CF Dugan and others, Investor-State Arbitration (Oxford, Oxford University Press, 2008) 128–132. 19 Bockstiegel, above (n 2) 581–82 (‘In commercial arbitration, in view of the very wide variety of fields of commerce such as trade, manufacturing, construction, service, finance, insurance or transport, it is obviously important for the parties to be able to select arbitrators well acquainted with such fields. In investment arbitration, the situation is somewhat different. The usual issues in such disputes are more limited, since the bilateral as well as multilateral investment protection treaties contain very similar protective provisions dealing with expropriation, fair and equitable treatment, discrimination and sometimes contracts by umbrella clauses. In view of this, the typical expertise required from arbitrators is one of public international law and particularly its application to such protection. … In practice, the result is that many arbitrators of commercial arbitration do not feel comfortable or are not chosen by the parties in investment arbitrations, and vice versa, many experts of international law selected for investment arbitration are not active in commercial arbitration. Generally, the number of arbitrators active in investment arbitration is much smaller than that in commercial arbitration. However, there is quite a group of arbitrators who do both kinds of arbitration.’) (emphasis added).

The Contribution of International Commercial Arbitration to Investment Arbitration 467 exercise ‘independent judgment’,20 while under most other regimes the existence of doubts about an arbitrator’s independence and impartiality may disqualify him or her.21 It is unclear how important in practice these differences in formulation really are. The IBA Guidelines on Conflict of Interest in International Arbitration,22 which are widely followed in international commercial arbitration,23 are also ‘regularly relied upon by parties challenging an arbitrator and seeking annulment of [ICSID] awards’,24 by the ICSID tribunal, and the ICSID Annulment Committee or ICSID Administrative Council.25 In the case of Perenco v Ecuador, for example, the Permanent Court of Arbitration applied the standard of ‘justifiable doubts’ from ‘the point of view of a reasonable third person’ provided by General Standard 2(b) IBA Guidelines in sustaining a challenge against Judge Brower.26 iii. Ethics in International Arbitration Of growing salience in international investment arbitration, as in international commercial arbitration, are concerns surrounding professional ethics. Despite the lack of a binding code of ethics in international arbitration,27 impartiality and independence, as discussed above, 20 Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (International Centre for Settlement of Investment Disputes [ICSID]), 575 UNTS 159 (14 October 1966) [ICSID Convention], Art 14(1) (‘Persons designated to serve on the Panels shall be persons of high moral character and recognized competence in the fields of law, commerce, industry or finance, who may be relied upon to exercise independent judgment. Competence in the field of law shall be of particular importance in the case of persons on the Panel of Arbitrators.’) and Art 57 (‘A party may propose to a Commission or Tribunal the disqualification of any of its members on account of any fact indicating a manifest lack of the qualities required by paragraph (1) of Article 14. A party to arbitration proceedings may, in addition, propose the disqualification of an arbitrator on the ground that he was ineligible for appointment to the Tribunal under Section 2 of Chapter IV.’). Despite the wording of the Article, both independence and impartiality are required under Article 14(1). See Urbaser SA and Consorcio de Aguas Bilbao Bizkaia, Bilbao Biskaia Ur Partzuergoa v The Argentine Republic, Decision on Claimant’s Proposal to Disqualify Professor Campbell McLachlan, Arbitrator (12 August 2010) ICSID Case No ARB/07/26, para 36; Saint-Gobain Performance Plastics Europe v Bolivarian Republic of Venezuela, Decision on Claimant’s Proposal to Disqualify Mr. Gabriel Bottini from the Tribunal under Article 57 of the ICSID Convention (27 February 2013) ICSID Case No ARB/12/13, para 55. 21 Article 12(1) of the UNCITRAL Model Law on International Arbitration 1985 (United Nations Commissions on International Trade [UNCITRAL]), UN Doc A/40/17, Annex I (21 June 1985) [UNCITRAL Model Law] requires disclosure and justifies disqualification of persons where facts are ‘likely to give rise to justifiable doubts’ as to their impartiality or independence. 22 International Bar Association, IBA Guidelines on Conflicts of Interest in International Arbitration (2014) [hereinafter IBA Guidelines], General Standard 2, requiring an arbitrator to decline appointment ‘if facts or circumstances exist, or have arisen since the appointment, which, from the point of view of a reasonable third person having knowledge of the relevant facts and circumstances, would give rise to justifiable doubts as to the arbitrator’s impartiality or independence’. 23 M Scherer, ‘The IBA Guidelines on Conflict of Interest in International Arbitration: The First Five Years 2004–2009 (The IBA Conflicts of Interest Subcommittee, a subcommittee of the IBA Arbitration Committee)’ (2010) 4 Dispute Resolution International 5. See also C Santoro and L Babiy, ‘Corruption of Arbitrators’ in PE Mason and HA Grigera Naón (eds), International Arbitration: 21st Century Perspectives, s 5.03 (New York, Lexis Nexis, 2019). 24 Scherer, above (n 23) 37–46. 25 ibid citing Hrvatska Elektroprivreda v Republic of Slovenia, Tribunal’s Ruling regarding the participation of David Mildon QC in further stages of the proceedings (6 May 2008) ICSID Case No ARB/05/24; EDF International S A, SAUR International S A and León Participaciones Argentinas S A v Argentine Republic, Challenge Decision Regarding Professor Gabrielle Kaufmann-Kohler (25 June 2008) ICSID Case No ARB/03/23; Azurix v Argentine Republic, Annulment Proceeding, Decision on the Application for Annulment of the Argentine Republic (1 September 2009) ICSID Case No ARB/01/12; Vito G Gallo v Government of Canada, UNCITRAL, Decision on the Challenge to Mr J Christopher Thomas, QC (14 October 2009); Participaciones Inversiones Portuarias SARL v Gabonese Republic, Decision on the Proposal to Disqualify an Arbitrator (12 November 2009) ICSID Case No ARB/08/17; The Rompetrol Group N V v Romania, ICSID Case No ARB/06/3, Decision on the Participation of a Counsel (14 January 2010). 26 Perenco Ecuador Limited v the Republic of Ecuador & Empresa Estatal Pertoleos Del Ecuadorpca, Decision on Challenge to Arbitrator (8 December 2009) PCA Case No IR-2009/1, paras 9, 42. 27 A code of ethics in international arbitration has been long sought and unsuccessfully realised. See T Landau and JR Weeramantry, ‘A Pause for Thought’ in AJ van den Berg (ed), International Arbitration: The Coming of

468 George A Bermann remain at the core of international commercial and investment arbitration alike. They are regulated by institutional rules, such as those imposing disclosure requirements, that are common to both. Institutions apply the same rules to all arbitrations conducted under their aegis, whether commercial or investment.28 Even ICSID rules on impartiality and independence, though confined to investment cases, are not fundamentally different29 and awards under the New York and ICSID Conventions are equally subject to annulment on such grounds. The similarity of issues in investment and commercial arbitration is such that scholars regularly discuss ethics in international arbitration without distinguishing between the two types of arbitration.30 This is largely because the soft law instruments that exist on such matters as conflicts of interests and the conduct of proceedings are equally applicable to commercial and investment arbitration.31 These include the 1987 IBA Rules of Ethics for International Arbitration (IBA Rules of Ethics), the 2014 IBA Guidelines on Conflicts of Interest in International Arbitration (IBA Guidelines on Conflict of Interest) and the 2013 Guidelines on party representation in international arbitration (IBA Guidelines on Party Representation).32 Consequently, ISDS tribunals have ruled upon challenges to the conduct of arbitrators and counsel on the basis of the same criteria prescribed by soft law instruments for international commercial arbitration.33 iv. The Seat of Arbitration The seat of arbitration plays an essential role in international commercial arbitration. The arbitration law of the seat (lex arbitri) directly governs the conduct of arbitrations situated there, and the courts of the seat may themselves take action in support of those arbitrations and resolve challenges to the resulting awards.34 To the extent that investment a New Age, ICCA Congress Series, Vol. 17 (Alphen aan den Rijn, Kluwer Law International, 2013) 496, 501 (‘As observed by certain scholars, “[g]iven the local or regional differences in the formulation and application of ethics rules, it is unclear how divergences may simply be ‘papered over’, resolved or erased with the imposition of a single, universal, uniform code.”’). 28 The issue of arbitrators, parties and counsels’ conflicts in investment arbitration has also been subject of a ASIL-ICCA Task Force study in preparation for the ongoing revision to the Arbitration Rules of the International Center for Settlement of Investment Disputes. See IC Popova and JL Polebaum, ‘Emerging Expectations for Arbitrators: ‘Issue Conflict’ in Investor-State Arbitration and Beyond’ (2018) 41 Fordham Law Journal 4, 937, 939. 29 ICSID Convention, Art 14, para 1: ICSID panel members must be ‘persons who may be relied upon to exercise independent judgment’. See also Art 40. 30 See, eg, CA Rogers, Ethics in International Arbitration (Oxford, Oxford University Press, 2014); P Halprin and S Wah, ‘Ethics in International Arbitration’ (2018) 1 Journal of Dispute Resolution 87, 88–89. 31 For the purposes of this chapter, soft laws include all those instruments that provide ‘[n]on-mandatory or discretionary legal techniques that can nevertheless impact decision-makers’,.see M Jacob and SW Schill, ‘Going Soft: Towards a New Age of Soft Law in International Law?’ (2014) 8 World Arbitration and Mediation Review 1, 3. Some scholars have defined soft law instruments more broadly, as to include any norms, treaties and non-binding instruments that are too vague to be applied or that do not give rise to a cause of action in court. See G Kaufmann-Kohler, ‘Soft Law in International Arbitration: Codification and Normativity’ (2010) 1 Journal of International Dispute Settlement 283, 284. 32 See also the guidelines and notes issued by some arbitral institutions, such as the SCC Practice Note – SCC Board Decisions on Challenges to Arbitrators 2013–2015 [hereinafter SCC practice note (2013–2015)]. 33 Halprin and Wah, above (n 30) 94; K Duggal, ‘Evidentiary Principle in Investor-State Arbitration’ (2017) 28 American Review of International Arbitration 1, 21–22. For use of soft law in international commercial arbitration see also S Orsi, ‘Ethics in International Arbitration: New Considerations for Arbitrators and Counsel’ (2013) 1 The Arbitration Brief 3, 92, 95–98, 103–107; E Mereminskaya, ‘Results of the Survey on the Use of Soft Law Instruments in International Arbitration’ (Kluwer Arbitration Blog, 6 June 2014), https://perma.cc/9B7J-RPNU; D Arias, ‘Soft Law Rules in International Arbitration: Positive Effects and Legitimation of the IBA as a Rule-Maker’ (2018) 6 Indian Journal of Arbitration Law 29, 40–41. 34 GA Bermann, International Arbitration and Private International Law (Leiden, Martinus Nijhoff, 2017) 215–16, para 230.

The Contribution of International Commercial Arbitration to Investment Arbitration 469 arbitration proceeds under the aegis of an arbitral institution (other than ICSID, as shown below), the role of the law and courts of the seat is essentially the same.35 Unless a treaty provides otherwise, the parties to international investment arbitration enjoy the same freedom as parties to international investment arbitration to determine the seat of arbitration.36 The seat has much the same importance in non-ICSID investment arbitration as in international commercial arbitration; the choice of seat being potentially outcome-determinative. In the famous Achmea case,37 the pronouncement by the Court of Justice of the European Union (CJEU) that the dispute resolution provisions of intra-EU BITs are invalid38 led the courts of Germany, where the arbitration had been seated, to annul the award without further question. Interestingly, although the investment agreements recently negotiated by the EU with Canada, Singapore and Vietnam include a two-tier court-like mechanism for the resolution of disputes arising under those agreements,39 they do not preclude parties from bringing annulment actions in national courts.40 Investors who opt instead for UNCITRAL arbitration, as they may, are entitled to seek annulment of an award in the national courts of the seat of arbitration.41 The regime established under the ICSID is of course a conspicuous exception, for it establishes what is widely characterised as a self-contained regime that in principle excludes any appeal or review by national courts.42 This includes most notably the annulment of awards,

35 Mesa Power Group, LLC v Canada, UNCITRAL, PCA Case No 2012–17, Procedural Order No 3 (28 March 2013) para 39. See generally J Commission and R Moloo, Procedural Issues in International Investment Arbitration (Oxford, Oxford University Press, 2018) 1, para 1.04; K Hobér and N Eliasson, ‘Review of non-ICSID Awards by National Courts’ in Yannaca-Small, above (n 12) 790, para 28.140. 36 Commission and Moloo, above (n 35) 147, para 9.04; Arbitration Rules United Nations Commission on International Trade Law (UNCITRAL) (as adopted in 2013), UN A/Res/68/109 (18 December 2013) [hereinafter UNCITRAL Rules] Art 18; Arbitration Rules of the Arbitration Institute of the Stockholm Chamber of Commerce 2017 (1 January 2017) [hereinafter SCC Rules] Art 25; Rules of Arbitration of the International Chamber of Commerce (1 March 2017) [hereinafter ICC Rules] Art 18. 37 Achmea BV v The Slovak Republic, Award on Jurisdiction, Arbitrability and Suspension (26 October 2010) UNCITRAL, PCA Case No 2008–13, para 16. 38 Slovak Republic v Achmea EU:C:2018:158 (6 March 2018), para 60; J Hillebrand Pohl, ‘Intra-EU Investment Arbitration after the Achmea Case: Legal Autonomy Bounded by Mutual Trust?’ (2018) 14 European Constitutional Law Review 767, 789. It is speculated that at the outset of the arbitration, Slovakia requested the seat of arbitration be in Germany, because German courts were the likeliest to refer to the CJEU in a preliminary ruling. Had the seat been in a non-EU Member State or in a jurisdiction where courts would avoid referring to the CJEU for a preliminary ruling in such a question, the far-reaching impact of Achmea would have hardly occurred. See CJEU, Annual Report 2010 – Synopsis of the work of the Court of Justice, the General Court and the European Union Civil Service Tribunal (Luxembourg, Publications Office of the European Union 2011) 104 (showing German courts making a total of 1802 references for a preliminary ruling whereas French courts made 816, British courts 505 and Italian courts 1056). 39 Comprehensive Economic and Trade Agreement (CETA) between Canada, of the one part and the European Union and its Member States, of the other part, L:2017:011:TOC (provisionally in force since 21 September 2017) [CETA], Art 8.28(d); European Union-Viet Nam Investment Protection Agreement (signed 30 June 2019) [EU-Viet Nam Investment Protection Agreement] Arts 3.53, 3.55.1; European Union-Singapore Investment Protection Agreement (signed 19 October 2018, entered into force 21 November 2019) [EU-Singapore Investment Protection Agreement] Arts 3.18.4, 3.19.2. 40 CETA, above (n 39) Arts 8.41.5, 8.28.9(b); EU-Viet Nam Investment Protection Agreement, above (n 39) Arts 3.57.7, 3.57.1(b); EU-Singapore Investment Protection Agreement, above (n 39) Arts 3.22.5, 3.22.1. 41 J Dahlquist Cullborg, ‘Place of Arbitration in the Proposed ‘Investment Court’ Scenario: An Overlooked Issue?’ (Kluwer, 23 March 2017) arbitrationblog.kluwerarbitration.com/2017/03/23/joel-booked accessed 10 November 2019; see also A Reinisch, ‘Will the EU’s Proposal Concerning an Investment Court System for CETA and TTIP Lead to Enforceable Awards? The Limits of Modifying the ICSID Convention and the Nature of Investment Arbitration’ (2016) 19 Journal of International Economic Law 761, 783. 42 CH Schreuer and others, The ICSID Convention: A Commentary, 2nd edn (Cambridge, Cambridge University Press, 2009) 1102, para 18; F Nitschke and K Ait-El-Hadj, ‘Determining the Place of Arbitration in ICSID Additional Facility Proceedings’ (2015) 30 ICSID Review – Foreign Investment Law Journal 243, 243–44.

470 George A Bermann which is the exclusive prerogative of ad hoc committees set up by ICSID itself for that specific purpose.43 The ICSID example illustrates well how construction of an arbitral regime dedicated exclusively to investment disputes may entail a change in the usual rules of the game. Intervention of national courts at the annulment stage was viewed as especially dangerous in that investor-state awards, at least those resulting from treaty-based, as opposed to contractbased, arbitration can ordinarily result in damages awards rendered only against states and not against investors. v. Kompetenz-Kompetenz and Separability Among the most signal contributions of international commercial arbitration to international investment arbitration are the principles of Kompetenz-Kompetenz and separability. There is no room for doubt that international investment tribunals enjoy authority to determine their own jurisdiction. Their Kompetenz-Kompetenz principle may even be more complete than in international commercial arbitration because, while commercial disputes may easily start in a national court with the court deciding whether an arbitration agreement invoked as a basis for dismissal is applicable and enforceable, resolution of treaty-based investment claims will invariably begin with the institution of arbitration, operating as acceptance by the investor of the host state’s treaty-based standing offer to arbitrate.44 It therefore makes no difference, from a Kompetenz-Kompetenz point of view, whether an agreement to arbitrate is entered into by contract in commercial arbitration or by treaty in investment arbitration. In international commercial arbitration, the principle of separability preserves a tribunal’s authority to resolve a dispute on the merits even if it invalidates the arbitration agreement on which its authority to rule in principle depends. The principle is recognised virtually universally. The leading US case is Prima Paint v Flood Conklin.45 Technically, the Court relied on Section 4 Federal Arbitration Act, according to which a court must compel arbitration of a dispute covered by an arbitration agreement as long as ‘the making of the agreement for arbitration or the failure to comply [with the agreement] is not in issue’.46 It understood the statute to mean that a court ‘may consider only issues relating to the making and performance of the agreement to arbitrate’. However, it also cited a policy reason, namely, that ‘[an] arbitration procedure, when selected by the parties to a contract, be speedy, and not subject to delay and obstruction in the courts’. Since the claimant had argued that the entire contract, and not the arbitration agreement as such, had been fraudulently induced, the case was sent to arbitration. Separability serves the same purpose in international investment arbitration as it does in international commercial arbitration. Specifically addressing investment arbitration, a leading scholar observes that, without separability ‘an illegality tainting the commercial bargain would often infect that arbitration agreement given the reality that both [agreements] are negotiated in a single package’.47 International investment tribunals make use themselves of

43 See M Burgstaller and CB Rosenberg, ‘Challenging International Arbitral Awards: To ICSID or not to ICSID?’ (2011) 27 Arbitration International 91, 93. 44 J Paulsson, ‘Arbitration Without Privity’ (1995) 10 ICSID Review – Foreign Investment Law Journal 232. 45 Prima Paint Corp v Flood & Conklin Mfg Co, 388 US 395 (1967). 46 Federal Arbitration Act 1926, 9 USC, s 4. 47 Z Douglas, ‘The Plea of Illegality in Investment Treaty Arbitration’ (2014) 29 ICSID Review – Foreign Investment Law Journal 155, 162.

The Contribution of International Commercial Arbitration to Investment Arbitration 471 the separability principle. Faced with a challenge to their jurisdiction based on the alleged invalidity of the main contract, tribunals do not suspend proceedings on the merits in order to determine the matter; they do so only if the challenge is directed specifically to the arbitration clause.48 This practice is routine.49

B. Divergences The fact that international investment arbitration has drawn a great many of its general principles from international commercial arbitration has not prevented it from developing, even on the general level, in a somewhat different direction. Considering aspects of international investment arbitration’s distinctiveness helps put the contribution of international commercial arbitration to international investment arbitration in fuller perspective. i. Methodology Notwithstanding the numerous and important commonalities between international commercial and investment arbitration, international investment arbitration is also marked by certain distinctive features that find their source in its public international law pedigree,50 largely accounted for by the critical presence of public international law-oriented practitioners in the international investment arbitration world. Some of these distinctions are highly nuanced and risk overgeneralisation. 48 See, eg, World Duty Free Co Ltd v Republic of Kenya, Award (4 October 2006) ICSID Case No ARB/00/7, s IV; Metal-Tech Ltd v Republic of Uzbekistan, Award (4 October 2013) ICSID Case No ARB/10/3, paras 278 ff. 49 Plama Consortium v Republic of Bulgaria, Decision on Jurisdiction (8 February 2005) ICSID Case No ARB/03/24, para 130 (‘[T]he Respondent’s charges of misrepresentation are not directed specifically at the parties’ agreement to arbitrate found in Article 26 ECT. The alleged misrepresentation relates to the transaction involving the sale of the shares of Nova Plama by EEH to PCL and the approval thereof given by Bulgaria in the Privatization Agreement and elsewhere. It is not in these documents that the agreement to arbitrate is found. Bulgaria’s agreement to arbitrate is found in the ECT, a multilateral treaty, a completely separate document. The Respondent has not alleged that the Claimant’s purported misrepresentation nullified the ECT or its consent to arbitrate contained in the ECT. Thus not only are the dispute settlement provisions of the ECT, including Article 26, autonomous and separable from Part III of that Treaty but they are independent of the entire Nova Plama transaction; so even if the parties’ agreement regarding the purchase of Nova Plama is arguably invalid because of misrepresentation by the Claimant, the agreement to arbitrate remains effective.’). See also Malicorp Ltd v Arab Republic of Egypt, Award (7 February 2011), ICSID Case No ARB/08/18, para 119 (‘According to [the doctrine of separability], defects undermining the validity of the substantive legal relationship, which is the subject of the dispute on the merits, do not automatically undermine the validity of the arbitration agreement. Thus, an arbitral tribunal is competent to decide on the merits even if the main contract was entered into as a result of misrepresentation or corruption. Only defects that go to the consent to arbitrate itself can deprive the tribunal of jurisdiction. In the present case, there is nothing to indicate that the consent to arbitrate, as distinct from the consent to the substantive guarantees in the [BIT], was obtained by misrepresentation or corruption or even by mistake. The allegations of the Respondent relate to the granting of the Concession. However, it is not the Contract that provides the basis for the right to arbitrate, but the State’s offer to arbitrate contained in the [BIT] and the investor’s acceptance of that offer. The offer to arbitrate thereby covers all disputes that might arise in relation to that investment, including its validity.’). 50 D Kennedy, ‘The Politics of the Invisible College: international Governance and the Politics of Expertise’ [2001] European Human Rights Law Review 463, 478 (‘Unlike the common man, or, for that matter politicians, whose minds are expanded by campaigning about the polity, experts have had their thinking narrowed by professional specialization.’). On the links between arbitrators’ professional experience and their awards in employment and labour arbitration, see also BS Klaas, D Mahony and HN Wheeler, ‘Decision Making About Workplace Disputes: A Policy-Capturing Study of Employment Arbitrators, and Jurors’ (2006) 45 Industrial Relations: A Journal of Economy and Society 68. See, generally Y Dezalay and BG Garth, Dealing in Virtue: International Commercial Arbitration and the Construction of a Transnational Legal Order (Chicago, University of Chicago Press, 1996); Waibel and Wu, above (n 7).

472 George A Bermann Among the traces of investment arbitration’s public international law origins, those relating to mode of reasoning may be most salient.51 First, arbitrators may differ in the way they approach the interpretation of investment treaties.52 A public international lawyer would be naturally inclined to resort to the authority of the Vienna Convention on the Law of Treaties (VCLT), and public international law more generally, in understanding the import of treaty provisions upon which investment arbitration outcomes depend, while an investment arbitrator coming from commercial arbitration circles, would be more likely to invoke national law and analogise the interpretation of treaties to the interpretation of contracts. Second, actors coming from public and private international law worlds may differ in terms of argumentation and the reference points of which that argumentation makes use. Again, at the risk of overgeneralisation, public international lawyers are more likely than private international lawyers to invoke broadly conceived general principles, whereas private international lawyers are more likely to conduct stricter fact-based inquiries.53 In the Corn Products case, for instance, the presence of a professor of public international law as chair may help explain the multiple references to international law jurisprudence in the award.54 In Continental Casualty, where the chairman was a member of the WTO Appellate Body, the award was largely based on international trade law jurisprudence.55 Third, different backgrounds may also be reflected in arbitrators’ and counsel’s conception of investment arbitration’s fundamental nature. International commercial arbitrators may continue to view investment arbitration as, to some degree, private in nature, governed by certain cardinal principles such as party autonomy or confidentiality. This may account for the observation that ‘dispute resolution in investment treaty arbitration is of a discrete nature [and] has no systemic effect’.56 By comparison, lawyers schooled in public international law may more readily consider investment arbitration as a means of serving politico-judicial

51 Roberts, above (n 8) 54 (‘As an empirical matter, it is difficult to prove what impact, if any, arbitrators’ backgrounds will have on their approach to investor-state disputes. Whether and to what extent arbitrators’ backgrounds or personal views play a role in their decision-making may vary between hard and easy cases, and different aspects of arbitrators’ backgrounds may pull them in different directions.’). For criticism that we lack sufficient data from which to draw reliable statistical conclusions, see G Van Harten, ‘Fairness and independence in investment arbitration: A critique of Development and Outcomes of Investment Treaty Arbitration’ (Investment Treaty News, 16 December 2010), https://perma.cc/88EG-3UZ5 (in reference to SD Franck, ‘Development and Outcomes of Investment Treaty Arbitration’ (2009) 50 Harvard International Law Journal 435). 52 See generally T Wälde, ‘Interpreting Investment Treaties: Experiences and Examples’ in C Binder and others (eds), International Investment Law for the 21st century – Essays in Honour of Christoph Schreuer (Oxford, Oxford University Press, 2009) 724. 53 J Paulsson, ‘Avoiding Unintended Consequences’ in K Sauvant and M Chiswick-Patterson (eds), Appeals Mechanism in International Investment Disputes (Oxford, Oxford University Press, 2008) 241, 262–3 (‘Commercial lawyers venturing into finely balanced matters of public international law may also be tempted, perhaps by an excess of self-confidence, to deliver themselves of a broad general exposition with the intent of clearing up a troubling issue … [public international lawyers], like surgeons operating on someone who has the flu, do what they know rather than what they should, avoiding areas with which they are not familiar but which are at the core of the task at hand.’). See also Wälde, above (n 52) 724–27. 54 Corn Products International, Inc v United Mexican States, ICSID Case No ARB (AF)/04/1, Decision on Responsibility (15 January 2008) paras 161–79; A Roberts, ‘Clash of Paradigms: Actors and Analogies shaping the investment treaty system’ (2013) 107 American Journal of International Law 45, 55. 55 Continental Casualty Company v The Argentine Republic, Award (5 September 2008) ICSID Case No ARB/03/9, para 192; Roberts, above (n 8) 55. 56 SW Schill, ‘Crafting the International Economic Order: The Public Function of Investment Treaty Arbitration and Its Significance for the Role of the Arbitrator’ (2010) 23 Leiden Journal of International Law 401, 407 (‘the description of the ethos of commercial arbitrators in the text is deliberately stereotypical in order to squeeze out the essence of how they perceive arbitration, … is quite similar to the perception of dispute settlement under the old GATT system.’).

The Contribution of International Commercial Arbitration to Investment Arbitration 473 purposes relating to the proper administration of international economic justice.57 Centre stage is the so-called ‘right to regulate’ which is often invoked by tribunals in justifying challenged measures. In Saluka v Czech Republic,58 the Tribunal expressed the right to regulate this way: In the opinion of the Tribunal, the principle that a State does not commit an expropriation and is thus not liable to pay compensation to a dispossessed alien investor when it adopts general regulations that are ‘commonly accepted as within the police power of States’ forms part of customary international law today. There is ample case law in support of this proposition.

The Tribunal cited in support the ruling in Methanex Corp. v United States to the following effect: It is a principle of customary international law that, where economic injury results from a bona fide regulation within the police powers of a State, compensation is not required.59

Thus, investment arbitration not only resolves a dispute between two parties regarding their private rights, but also contributes to establishing a global normative framework ‘that overarches the individual bilateral treaty relations and establishes uniform rules for the conduct of host states that consist in adopting a liberal attitude towards market mechanisms and that accept the limited role of the State vis-à-vis the economy’.60 ii. The Applicable Law The influence of international commercial arbitration on international investment arbitration is least apparent in choice of applicable law,61 due notably to the impact of the applicable law on arbitral outcomes.62 International commercial arbitration recognises the freedom of parties to choose the law applicable to the merits of the dispute,63 which they commonly do.64 57 C Brown, ‘The Inherent Powers of International Courts and Tribunals’ (2006) 76 British Yearbook of International Law 195, 231 (‘One of these is the function of ensuring the proper administration of international justice. This is distinct from the function of settling disputes, in that it emphasizes the need for effectiveness and efficiency in judicial decision-making, and it is well established in the jurisprudence of international courts, as well as in the literature.’) (citations omitted). Against, J Crawford, ‘Treaty and Contract in Investment Arbitration’ (2008) 24 Arbitration International 351, 353 (‘According to an influential view of the international arbitral function, it is an illusion to think that there is a right or correct method of resolving such issues as these. Arbitrators are not judges; there is no method of ensuring “correct” decisions, and there will be unresolvable disagreements at the general level between arbitrators depending on their own legal traditions and approaches.’). 58 Saluka Inv BV v The Czech Republic, UNCITRAL, Partial Award (17 March 2006) para 262. 59 MethanexCorp v United States, Final Award (3 August 2005) 44 International Legal Materials 1343, para 410. 60 SW Schill, The Multilateralization of International Investment Law (Cambridge, Cambridge University Press, 2009) 17; A Mills, ‘Antinomies of Public and Private at the Foundations of International Investment Law and Arbitration’ (2011) 14 Journal of International Economic Law 469, 484 (‘This form of reasoning, that a rule should be evaluated based on its capacity for generalization, is most closely associated with Kant’s “categorical imperative”, or its modern reinterpretation in John Rawls, A Theory of Justice.’). 61 P Bernardini, ‘International Commercial Arbitration and Investment Treaty Arbitration Analogies and Differences’ in DD Caron and others (eds), Practising Virtue: Inside International Arbitration (Oxford, Oxford University Press, 2015) 57. 62 Y Banifatemi, ‘The Law Applicable in Investment Treaty Arbitration’ in Yannaca-Small, above (n 12) para 19.02 (‘That is not to say that choice of law is of no consequence. To the contrary, in investment treaty arbitration, just as in international commercial arbitration, it is a fundamental process in that the outcome of the dispute may sometimes greatly depend on the rules determined to be applicable.’). 63 N Blackaby and others, Redfern and Hunter on International Arbitration, 6th edn (Oxford, Oxford University Press, 2015) 187, para 3.97. 64 GB Born, International Commercial Arbitration, 2nd edn (Alphen aan den Rijn, Wolters Kluwer, 2014) 2670–71. (‘International arbitration agreements typically include, or accompany, a choice-of-law provision addressing the substantive law applicable to the parties’ contract, or, more broadly, their entire relationship.’).

474 George A Bermann Only if the parties fail to select a law will the Tribunal select one, directed by most rules to apply the law, or the choice of law principle, it deems most appropriate.65 That gives tribunals a great deal of latitude in determining the applicable law in the absence of choice of law. The principle of party autonomy nominally applies in international investment arbitration as well, as typified not only by the ICSID Convention itself,66 but also by the rules of UNCITRAL and the Stockholm Chamber of Commerce which also govern a very large number of international investment arbitrations.67 However, the ICSID Convention offers greater guidance in determination of the applicable law, by typically directing tribunals, in the absence of party choice, ‘to apply the law of the Contracting State party to the dispute … and such rules of international law as may be applicable’. The direct application of international law is among the factors that most sharply distinguish international investment arbitration from international commercial arbitration. The distinctiveness of investment arbitration in this regard is readily attributed not only to the presence of a state in the dispute, but also to the fact that, at least in treaty-based arbitration, a tribunal ultimately draws its authority from an international agreement between states. Equally distinctive is the fact that the investment treaties on which international investment arbitrations are based do not merely authorise the arbitration of disputes arising under them, but contain in their very terms the most essential principles of substantive law in the resolution of those disputes. Accordingly, international investment tribunals first and foremost find the applicable legal principles in the terms of the treaty itself. For example, the guarantees of fair and equitable treatment or the right to compensation in the event of expropriation of which investors avail themselves are not derived so much from general public international law as from the investment treaty itself, the instrument that established arbitration as an available means of dispute resolution. Under the terms of most investment treaties, general principles of international law come into play not only as independently applicable law, but also in whatever gap-filling is required in application of the treaty’s substantive terms. While investment treaties most often also include the law of the host state in the applicable law, they do not generally state its hierarchical relationship to the substantive rights established by the treaties and general principles of international law or delineate the issues to which national law shall have application.68 In Wena v Egypt, the Tribunal simply stated: The Tribunal finds that, beyond the provisions of the [U.K.-Egypt BIT], there is no special agreement between the parties on the rules of law applicable to the dispute. Rather, the pleadings of both parties

65 UNCITRAL Rules, Art 35(1) (‘Failing such designation by the parties, the arbitral tribunal shall apply the law which it determines to be appropriate.’); SCC Rules, Art 27(1) (‘In the absence of such agreement, the Arbitral Tribunal shall apply the law or rules of law that it considers most appropriate.’); ICC Rules, Art 21(1) (‘In the absence of any such agreement, the arbitral tribunal shall apply the rules of law which it determines to be appropriate.’). 66 Under Art 42 of the ICSID Convention, ‘[t]he 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’. See generally Banifatemi, above (n 62) 486, para 19.03; ICSID Convention, Art 42(1) (‘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.’). 67 UNCITRAL Rules, Art 35(1) (‘The arbitral tribunal shall apply the rules of law designated by the parties as applicable to the substance of the dispute.’); SCC Rules, Art 27(1) (‘The Arbitral Tribunal shall decide the merits of the dispute on the basis of the law(s) or rules of law agreed upon by the parties.’). 68 Banifatemi, above (n 62) 490, para 19.10 (‘Where such choice exists, the situations can broadly be categorized as follows. Almost always, the dispute is to be decided “in accordance with the provisions of the Agreement” itself. Frequently, the BIT is applicable in conjunction with “the principles of international law” or “the applicable rules of international law”. This is also the case for multilateral treaties containing investment protection rules such as the

The Contribution of International Commercial Arbitration to Investment Arbitration 475 indicate that, aside from the provisions of the [BIT], the Tribunal should apply both Egyptian law (i.e., ‘the law of the Contracting State party to the dispute’) and ‘such rules of international law as may be applicable.’69

Theoretically, international commercial tribunals may, in the absence of a choice of law by the parties, also apply non-national law if they deem it appropriate to do so,70 and the Tribunal in the leading Lena Goldfields v USSR case did just that in giving effect to what it found to be an international law remedy of unjust enrichment.71 The Lena tribunal accepted that international law applied to a dispute in conjunction with the Soviet law on domestic matters. However, the application of non-national law in international commercial arbitration is virtually unknown. As far as the law applicable to the arbitral proceeding itself is concerned, both commercial and non-ICSID investment arbitrations operate within the framework of the arbitration law of the seat72 and the arbitral rules that parties may have selected. In this respect, as in so many, commercial and non-ICSID arbitrations are conducted in the same fashion. They are also both conducted on the understanding that the eventual awards will be subject to the 1958 New York Convention.73 In sum, the principles and rules applicable to international commercial arbitration and non-ICSID investment arbitration are essentially no different.74

NAFTA and the Energy Charter Treaty. The choice of applicable law may include, in addition, the law of the host state in its entirety. Some bilateral investment treaties refer to the treaty, the law of the host state, and particular agreements between the parties, but not necessarily to the rules of international law.’). See also CH Schreuer and others, The ICSID Convention: A Commentary, 2nd edn (Cambridge, Cambridge University Press, 2009) 595. 69 Wena Hotels Limited v Arab Republic of Egypt, (8 December 2000) ICSID Case No ARB/98/4, para 79. 70 O Spiermann, ‘Applicable Law’ in PT Muchlinski, F Ortino and C Schreuer (eds), The Oxford Handbook of International Investment Law (Oxford, Oxford University Press, 2008) 93 (‘Arbitral tribunals adopt an approach to applicable law different from most national courts and their choice-of-law processes. Thus, it is commonly held that arbitral tribunals are without a lex fori on the basis of which to choose applicable law. An early example was the award from 1958 in Saudi Arabia v Aramco, in which the tribunal founded this approach on principles of equality between the parties as well as state immunity. The view was upheld in subsequent awards (albeit on different grounds), and it is now commonplace in international commercial arbitration at large.’). 71 VV Veeder, ‘The Lena Goldfields Arbitration: The Historical Roots of Three Ideas’ (1998) 47 ICLQ 747, 772 (‘As Lena Goldfields’ counsel in the Lena arbitration, Idelson’s internationalization of a transnational contract was a gigantic first step for international commercial arbitration, almost equivalent to the caveman’s discovery of fire. His innovation eventually spawned a mass of followers in France, Switzerland and elsewhere, not limited to State contracts or concessions. […] Under section 46(l)(b) of the English Arbitration Act 1996, based upon Article 28 of the UNCITRAL Model Law, an English arbitration tribunal may now decide a dispute (if the parties so agree) in accordance with considerations other than a national law, as are either agreed by the parties or determined by the tribunal; and the new 1998 LCIA Rules allow an arbitration tribunal to decide the parties’ dispute applying the law(s) or rules of law which it considers appropriate, both where the parties have so chosen but also in the absence of any chosen national law (as do the ICC rules, old and new). Yet the historical roots of this original idea lie not in Paris, Geneva, Vienna or Rome but in the mind of an Anglophile Russian exile in London over 65 years ago.’). See also A Ernst, ‘Lena Goldfields Arbitration’ (August 2014) Max Planck Encyclopedia of Public International Law, para 14. 72 P Bernardini, ‘International Commercial Arbitration and Investment Treaty Arbitration Analogies and Differences’ in DD Caron and others (eds), Practising Virtue: Inside International Arbitration (Oxford, Oxford University Press, 2015) 55 (‘Under all these systems of arbitration, the proceeding is administered in accordance with the will of the parties, or failing this, by the arbitral tribunal’s determination and develops, as usual, through a written and an oral phase, with full respect at all times to the parties’ equality and their right to be heard.’). 73 Blackaby and others, above (n 63) 61, paras 1.211–12; R Dolzer and C Schreuer, Principles of International Investment Law, 2nd edn (Oxford, Oxford University Press, 2012) 310. 74 See ibid 241.

476 George A Bermann iii. Structural Reform A truly salient feature of international investment arbitration for which international commercial arbitration is far from responsible is the impetus for major structural reform. The general framework within which international commercial arbitration takes place, while subject to modification, has not come in for radical re-examination. The same cannot be said of international investment arbitration. The movement observable in that field is plainly fuelled by concern over its impact on states’ authority to adopt measures that they deem, or at least assert, to be in the public interest. The best evidence is the adoption by a new generation of international arbitration agreements, typified by the EU-Canada Comprehensive Economic and Trade Agreement (CETA),75 of features traditionally unknown to international commercial arbitration, but also missing from international investment arbitration until relatively recent times. The most salient of these features is the replacement of ad hoc investment tribunals by a standing body of government-named decisionmakers. The pressures in this direction are most evident in the workings of UNCITRAL’s Working Group III, assembled precisely to identify what are widely viewed as the predominant ills of international investment arbitration and to develop appropriate responses. By way of concerns, the Working Group has highlighted three categories: (1) concerns pertaining to consistency, coherence, predictability and correctness of arbitral awards; (2) concerns pertaining to arbitrators and decision-makers; and (3) concerns pertaining to cost and duration of proceedings. To these, the Working Group added, in the course of deliberations, concerns over third-party funding. Having defined these concerns, Working Group III is directing its attention to a range of reforms, each in its own way meant to better secure the public interest in the resolution of international investment disputes. The reform proposals are arrayed on a spectrum running from incremental to radical change. Illustrative of the specific concerns falling within these categories are a requirement of exhaustion of local remedies and counterclaims by states. At the time of writing, development of a consensus among the state delegations is a distance away. Whatever the outcome, international investment arbitration will have undergone a comprehensive re-examination that dwarfs in magnitude the discrete modifications – for example, greater transparency or the introduction of expedited proceedings – that have been discussed and introduced into international commercial arbitration over recent years. It cannot in the least be said that this development represents a contribution of international commercial to international investment arbitration. It reflects a series of preoccupations that are, for the most part, though not entirely, unique to international investment arbitration.

III. PROCEDURAL FEATURES OF INTERNATIONAL ARBITRATION

The previous section sought to depict the extent to which international commercial arbitration may have contributed to the establishment or development in international investment arbitration of certain very general principles upon which investment arbitration, no less than commercial arbitration, currently operates. Both commonalities and divergences are readily observed. By contrast, this section looks at much more particularised features of international commercial arbitration, mostly procedural, in an effort to determine their impact on international investment arbitration. 75 CETA.

The Contribution of International Commercial Arbitration to Investment Arbitration 477 That international commercial arbitration should have a paramount influence on international investment arbitration in matters of arbitral procedure should come as no surprise. One need only consider the fact that a large percentage of international investment cases – the non-ICSID cases – are conducted on the basis of the very same national arbitration laws and the very same institutional and other arbitration rules as international commercial cases, and influenced by essentially the same soft law instruments, such as the 2010 IBA Rules on the Taking of Evidence.76 An obvious exception in arbitral procedure is the allowance of amicus curiae participation in international investment arbitration, a vehicle by which non-parties – often non-governmental organisations – may bring certain public interest considerations to a tribunal’s attention.77 Investment tribunals have largely developed on their own the principles governing amicus curiae participation. By contrast, involvement of amici has no real place in international commercial arbitration, if only due to the latter’s largely private contractual origin. The real difference in arbitral procedure lies, as is so often the case, in ICSID investment arbitration. Obviously these arbitrations are conducted under the ICSID Rules though, as noted by commentators,78 the ICSID Rules are not strikingly different from the procedural rules – from commencement of the proceedings, to selection of arbitrators, to the proceedings themselves, to the final award – under which most international commercial arbitrations are conducted.79 In fact, the 2006 amendments to the ICSID Rules, as well as the rule revision taking place as this chapter is written, introduced and will introduce numerous procedural mechanisms that originated in the rules of the major international commercial arbitration institutions, an example being the early dismissal mechanism introduced in 2006 for claims

76 International Bar Association, IBA Rules on the Taking of Evidence in International Arbitration (29 May 2010) [IBA Rules of Evidence]. In recognition of the applicability of the Guidelines to investment arbitration proceedings, the IBA Committee deleted the word ‘Commercial’. See Commentary on the Revised Text of the 2010 IBA Rules on the Taking of Evidence in International Arbitration, www.ibanet.org/Publications/publications_IBA_guides_and_free_ materials.aspx (‘The word “commercial” was deleted from the title of the Rules to acknowledge the fact that the IBA Rules of Evidence may be and are used both in commercial and investment arbitration.’). See also Blackaby and others, above (n 63) 381, para 6.95. 77 See A Saravanan and S Subramanian, ‘The Participation of Amicus Curiae in Investment Treaty Arbitration’ (2016) 5 Journal of Civil & Legal Sciences 201. 78 F Gelinas, ‘Investment Tribunals and the Commercial Arbitration Model: Mixed Procedures and Creeping Institutionalization’ in Markus W Gehring and Marie-Claire Cordonier Segger (eds), Sustainable Development in World Trade Law (Alphen aan den Rijn, Wolters Kluwer, 2005) 577 (‘Investors have been granted the right to bring claims to an arbitral tribunal constituted under one of various sets of international arbitration rules that are strongly inspired by commercial arbitration rules, such as the International Centre for the Settlement of Investment Disputes (ICSID) Rules and the ICSID Additional Facility Rules.’); Z Douglas, ‘The Plea of Illegality in Investment Treaty Arbitration’ (2014) 29 ICSID Review – Foreign Investment Law Journal 155, 157 (‘The two most common sets of arbitration rules used for investment treaty arbitration – the ICSID Arbitration Rules and the UNICITRAL Arbitration Rules were originally conceived for commercial arbitration and, at least in respect of UNCITRAL Rules, the primary field for their application remains commercial arbitration.’). 79 GB Born, ‘A New Generation of International Adjudication’ (2012) 61 Duke Law Journal 775, 834–835 (‘Virtually all forms of investment arbitration are conducted pursuant to procedures that parallel international commercial arbitration procedures. The arbitration provisions of the ICSID Convention and the associated ICSID Arbitration Rules are modelled closely on international commercial-arbitration rules. Most BITs provide for arbitration either pursuant to the ICSID Rules, the UNCITRAL Rules, or the rules of a commercial arbitral institution. In practice, the procedures used in international commercial arbitration are the model for investment arbitration, including the number and selection of arbitrators, the presentation of evidence, the conduct of hearings, and the awards – in part because of overlaps in the individuals and law firms that serve as arbitrators and counsel in both sets of proceedings.’); S Wilske, M Raible and L Markert, ‘International Investment Treaty Arbitration and International Commercial Arbitration – Conceptual Difference or only a ‘Status Thing’?’ (2008) 1 Contemporary Asia Arbitration Journal 213, 218 (‘[T]he ICSID Rules are not fundamentally different from other procedural frameworks for arbitration.’) (citation omitted).

478 George A Bermann manifestly lacking in merit.80 Recent innovations, such as expedited proceedings,81 inspired by recent changes in the rules of other institutions,82 are currently under consideration before the ICSID Reform Committee. As noted above, ICSID tribunals, having no seat, are not subject to the arbitration law of any particular jurisdiction, including the jurisdiction where hearings may have been held. Similarly, the enforcement regime applicable to ICSID awards is not the New York Convention, but the ICSID Convention itself. The procedural influence of international commercial arbitration over ICSID arbitration is somewhat less than the influence of international commercial arbitration over international investment arbitration conducted under other auspices. The list of procedural features reflecting an identity between international commercial and investment arbitration is extremely long. It includes such features as case management,83 document production and discovery,84 cost allocation,85 third-party funding, and treatment of conditions precedent to arbitration. On the latter issue, the US Supreme Court decision in BG Group PLC v Argentina,86 is telling. The Court, albeit over a vigorous dissent, explicitly borrowed from its international commercial arbitration case law87 to hold that the question whether an investor was justified in instituting arbitration without litigating for a period of time in the host state’s courts is primarily for the Tribunal and not the court.88 80 Rules of Procedure for Arbitration Proceedings (10 April 2006) [ICSID Arbitration Rules] r 41(5), http://icsidfiles. worldbank.org/icsid/icsid/staticfiles/basicdoc/partf.htm. 81 ‘Proposals for Amendment of the ICSID Rules’ (2019) 3(1) ICSID Working Paper, Arbitration Rules, r 74 ‘Consent of Parties to Expedited Arbitration’, https://perma.cc/G5LS-UC8P (‘The parties to an arbitration conducted under the Convention may consent at any time to expedite the arbitration in accordance with this Chapter (“expedited arbitration”) by jointly notifying the Secretary-General in writing of their consent.’). 82 See ‘Using fast track arbitration for resolving commercial disputes’ (2018) 6 Norton Rose Fullbright 25, https:// perma.cc/ATM4-FKDL (‘The first expedited arbitration procedure was introduced in 1992 by the Geneva Chamber of Commerce in its Arbitration Rules (which are now a part of the Swiss Arbitration Rules). Since then, many other international arbitration institutions have followed suit including the American Arbitration Association (AAA), Stockholm Chamber of Commerce (SCC), Hong Kong International Arbitration Centre (HKIAC), Singapore International Arbitration Centre (SIAC)) and most recently, the International Chamber of Commerce (ICC), which introduced an expedited procedure in January 2017.’); ICDR 2014 International Dispute Resolution Procedures (2014), Art 1(4), Art E-1 to E-10; SCC 2017 Expedited Arbitration Rules (2017); ICC 2017 Rules of Arbitration (2017), Art 30 and Appendix VI; SIAC 2016 Arbitration Rules (2016), r 5. With respect to security for costs, the Hong Kong International Arbitration Centre (HKIAC) provided an express provision on security for costs as early as 2013. 83 I Bantekas, An Introduction to International Arbitration (Cambridge, Cambridge University Press, 2015) 181. 84 See, eg, FG Sourgens, K Duggal and IA Laird, Evidence in International Investment Arbitration (Oxford, Oxford University Press, 2018) paras 9.37–9.45. 85 See, eg, for commercial arbitration ICC Rules, Art 38.1; LCIA Rules, Art 28.3; SCC Rules, Art 50; for investment arbitration ICSID Arbitration Rules, r 28.2; UNCITRAL Rules, Art 42.1. It has been suggested, however, that commercial and investment arbitration differ in principle on the default rule concerning which party shall presumptively bear the costs of arbitration. Most commercial arbitration rules embrace the ‘costs follow the event’ principle (English rule), imposing a rebuttable presumption that the party who lose must pay.85 In contrast, the general principles applied in investor-state disputes – although the ICSID Arbitration Rules do not take any default position in this regard – is the ‘pay your own way’ rule, which imposes a rebuttable presumption that each party shall bear its own cost (American rule). See M Hodgson and A Campbell, ‘The allocation of Costs in Investment Treaty Arbitration’ in CL Beharry (ed), Contemporary and Emerging Issues on the Law of Damages and Valuation in International Investment Arbitration (Brill, Martinus Nijhoff, 2018) ch 15, 402. Yet, even this putative difference appears to be lessening. See, eg, ADC Affiliate Limited and ADC & ADMC Management Limited v The Republic of Hungary, Award (2 October 2006) ICSID Case No ARB/03/16, para 533. See also PSEG Global Inc & Konya Ilgin Elektrik Üretim ve Ticaret Limited Sirketi v Republic of Turkey, Award (19 January 2007) ICSID Case No ARB/02/5, para 352; International Thunderbird Gaming v Mexico UNCITRAL, Award (26 January 2006) paras 213–14, 218–20; Telenor Mobile Communications AS v The Republic of Hungary, ICSID Case No ARB/04/15, Award (13 September 2006) para 107; EDF (Services) Limited v Romania, Award (8 October 2009) ICSID Case No ARB/05/13, paras 325, 327. 86 BG Group PLC v Republic of Argentina, 572 US 25 (2014). 87 Howsam v Dean Witter Reynolds, Inc, 537 US 79 (2002). 88 According to the Court in BG Group, above (n 87) 41, ‘[t]he upshot is that our ordinary presumption applies and it is not overcome. The interpretation and application of the local litigation provision is primarily for the arbitrators. Reviewing courts cannot review their decision de novo. Rather, they must do so with considerable deference’.

The Contribution of International Commercial Arbitration to Investment Arbitration 479 International commercial arbitration makes its influence over international commercial and investment arbitration alike especially felt through the numerous soft law instruments that were mostly developed with international commercial arbitration in mind. These ‘[n]on-mandatory or discretionary legal techniques that can nevertheless impact decisionmakers’89 have largely seamlessly extended their scope of application to international investment arbitration.90 The ready applicability of soft law instruments, adopted chiefly with international commercial arbitration in mind, to international investment arbitration is testament to the fact that the concerns these instruments are meant to address apply equally to international commercial and investment arbitration.91 To be sure, common rules may perform somewhat differently in commercial and investment contexts. For example, the heightened political sensitivity of documents in the latter context may result in heightened objections to discovery requests.92 Similarly, US courts may be somewhat more amenable to making documents available for use in international investment as compared to international commercial arbitration, perhaps due to the fact that the former regularly entails the application of international law and to the prominence of international treaties as compared to private contracts as a source of the obligation to arbitrate.93 Third-party funding (TPF) is a particularly interesting and rather exceptional matter in this regard. Far from a contribution of international commercial to international investment arbitration, third-party funding represents the converse: TPF initially figured predominantly in international investment arbitration, only to demonstrably make its way into international commercial arbitration.94 At the same time, only now is ICSID poised to address third-party funding, replicating the prevailing rule in international commercial arbitration that requires the disclosure of the identity of the funder and permits tribunals to consider TPF in ordering

89 Jacob and Schill, above (n 31) 3. Some scholars have defined soft law instruments in broader terms, as any norms, including treaties and non-binding instruments, that are too vague to be applied or that do not give rise to a cause of action in court. Kaufmann-Kohler, above (n 31) 284. 90 See Mereminskaya, above (n 33); Aras, above (n 33) 40–41. Among the soft law instruments that now operate in international investment arbitration no less than in international commercial arbitration figure the 2010 IBA Rules on the Taking of Evidence in International Arbitration (‘IBA Evidence Rules’), the 2014 IBA Guidelines on Conflicts of Interest in International Arbitration (‘IBA Guidelines on Conflict of Interest’), the 2013 Guidelines on party representation in international arbitration (‘IBA Guidelines on Party Representation’) and guidelines and practice notes issued by arbitral institutions like the ICC. 91 See, eg, Sourgens and Duggal and Laird, above (n 84) paras 9.37–9.45. 92 IBA Rules of Evidence, Arts 9.2(b), (e), (f). When dealing with issues of state secrecy, investment arbitration tribunals may have to delve into both international and national law. In Corfu Channel, for instance, the International Court of Justice refused to draw an adverse inference based on the UK’s refusal to disclose documents which it claimed were protected by the naval secrets privilege. Corfu Channel Case (UK v Albania) (Merits) [1949] ICJ Rep 4, 32. 93 See, eg, In re Application of Chevron Corporation, 709 F Supp 2d 283 (SDNY 2010). See generally SI Strong, Discovery Under 28 U.S.C. §1782: Distinguishing International Commercial Arbitration and International Investment Arbitration (2013) 1 Stanford Journal of Complex Litigation 295, 371 (‘[E]merging case law and commentary has suggested that a distinction could be developing between the two types of arbitration, with Section 1782 requests appearing to be more likely to be granted in cases involving investment arbitration. … Although Section 1782 requests in the context of international commercial and investment arbitration have increased significantly over the last few years, the case law and commentary are nowhere near developed enough to foreclose new avenues of analysis’) (citations omitted). 94 See Legum, above (n 1) 72; N Pitkowitz (ed), Handbook on Third-Party Funding in International Arbitration (New York, JurisNet, 2018) 7; J Von Goeler, Third-Party Funding in Arbitration and Its Impact on Procedure (Alphen aan den Rijn, Wolters Kluwer, 2016) 75; J Honlet, ‘Recent decisions on third-party funding in investment arbitration’ (2015) 30 ICSID Review – Foreign Investment Law Journal 699.

480 George A Bermann security for costs.95 International investment arbitration’s influence on international commercial arbitration can be seen elsewhere as well, for example, in the movement toward greater transparency. It is a reality that, even taken together, procedural governance of arbitral proceedings by national arbitration laws and rules, as well as soft law, is anything but comprehensive. Procedural gap-filling is a constant in international arbitration and is performed by some combination of party autonomy over procedural matters and the exercise of a wide-ranging discretion on the part of tribunals themselves. In this respect, international commercial and investment arbitration are fully alike.96 Not to be overlooked in this context is once again the already-mentioned prominence of international commercial arbitrators on international investment tribunals, as a result of which practices common to commercial arbitration readily make their way into international investment arbitration.97

IV. CONCLUSION

The contribution of international commercial arbitration to international investment arbitration is readily documented, the most plausible reason being the heavy participation of international commercial arbitrators in the newer international investment arbitration. This is most clearly evidenced in some, though not all, the general principles on which international investment arbitration has come to rest. Where international investment arbitration most reflects the contribution of international commercial arbitration is in the procedural realm. This is not to suggest that the relationship between international commercial and investment arbitration is a one-way street.98 Nevertheless, influence has until now been heavily in one direction. Whether that continues to be the case remains to be seen.

While parties to commercial arbitrations have been relying on TPF since the 1990s, the use of TPF in international arbitration has increased starting from early 2000, in concomitance with the increased use of ISDS mechanisms by investors. 95 Proposals for Amendment of the ICSID Rules, above (n 81) 52. ICSID has rejected proposals to prohibit TPF, noting that receiving funding from a third party does not mean that a claim is necessarily frivolous and that whether to prohibit TPF remains a state’s prerogative. Proposals for Amendment of the ICSID Rules, Working Paper #1, vol 3 (August 2018) 131, paras 242–243. 96 Böckstiegel, above (n 2) 584; see Sourgens and Duggal and Laird, above (n 84) para 9.27. A number of rules expressly empower arbitrators to manage parties’ expectations in connection with conduct of the proceedings, requesting them to be responsive to, and efficient in dealing with, parties’ repeated procedural requests. See Born, above (n 64) 2337–2342. 97 See Böckstiegel, above (n 2) 584–585; Roberts, above (n 8) 55. 98 Legum, above (n 1) 74.

28 The European Court of Human Rights’ Case Law on International Investment Issues URSULA KRIEBAUM*

I. INTRODUCTION

T

HE FIRST TIME that the European Court of Human Rights (ECtHR) decided a case on interference with the property rights of a foreigner was in 1986.1 Yet, it was not until the new millennium that the Court decided cases containing facts typical for investment arbitration. Similarities between different standards of protection in the European Convention of Human Rights (ECHR) and Investment Protection Treaties create potential overlaps. Article 6 ECHR which guarantees a right to a fair trial in civil proceedings2 overlaps with investment law standards of fair and equitable treatment. Article 1 of the First Additional Protocol to the ECHR3 which contains guarantees for the peaceful enjoyment of property overlaps with investment law standards of expropriation. Thus, where the ECHR applies, and the facts of the case allow, an investor may also potentially frame their case as a claim under the ECHR. * Ursula Kriebaum is Professor of International Law at the Department of European, International and Comparative Law at the Law Faculty of the University of Vienna. 1 Agosi v United Kingdom (1986) Series A, no 108. 2 Article 6, Right to a fair trial: 1. In the determination of his civil rights and obligations or of any criminal charge against him, everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law. Judgment shall be pronounced publicly but the press and public may be excluded from all or part of the trial in the interests of morals, public order or national security in a democratic society, where the interests of juveniles or the protection of the private life of the parties so require, or to the extent strictly necessary in the opinion of the court in special circumstances where publicity would prejudice the interests of justice. … 3. Everyone charged with a criminal offence has the following minimum rights: … (b) to have adequate time and facilities for the preparation of his defence; … For an overview see, eg, C Grabenwarter, European Convention on Human Rights: ECHR (Oxford, Hart, 2014) 98 f; WA Schabas, The European Convention on Human Rights: A Commentary (Oxford, Oxford University Press, 2015) 264 f. 3 Article 1, Protection of property: Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law. The preceding provisions shall not, however, in any way impair the right of a State to enforce such laws as it deems necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties. For an overview see, eg, Grabenwarter, above (n 2) 365 f; Schabas, above (n 2) 958 f.

482 Ursula Kriebaum This chapter will shed light on some elements in ECtHR case law that lie at the intersection of the two systems of property protection provided by international law, using leading cases to exemplify these elements.4

II. LE BRIDGE V MOLDOVA – COMPETING JURISDICTION AND PARALLEL PROCEEDINGS

There are a few cases where investors have opted for ‘parallel proceedings’ before investment arbitral tribunals and the ECtHR. National legal orders commonly use the principles of res judicata, lis pendens and ne bis in idem to prevent the same dispute from being litigated in different courts or tribunals.5 Article 35 ECHR deals with re-litigation and parallel proceedings. Its relevant part, Article 35 § 2(b), reads as follows: 2. The Court shall not deal with any application submitted under Article 34 that … (b) is substantially the same as a matter that has already been examined by the Court or has already been submitted to another procedure of international investigation or settlement and contains no relevant new information.

In a situation where several international bodies would be dealing simultaneously with substantially similar applications the ECtHR has to decide two questions: 1) whether the case is substantially the same as a matter that has already been submitted to a parallel set of proceedings; 2) if yes, whether the simultaneous proceedings may be seen as ‘another procedure of international investigation or settlement’. Le Bridge v Moldova, a case concerning airport duty free shops in Moldova, illustrates the problem.6 In 2009, Le Bridge, a Moldovan company, lodged an application before the ECtHR against Moldova. It alleged a violation of the right to a fair trial under Article 6(1) of the Convention in relation to the award of a tender to operate duty-free shops. Subsequently, in July 2012 the sole shareholder and CEO of Le Bridge, Mr Arif, a French national, started an ICSID arbitration against Moldova relying on the Moldova-France BIT.7 In April 2013, the ICSID Tribunal issued its award and rejected the majority of the investor’s claims but found that Moldova had violated in one instance the obligations to provide fair and equitable treatment. The ICSID Tribunal awarded only US$2.8 million out of over US$30 million claimed.8 Both the ECtHR (in Le Bridge v Moldova) and an investment Tribunal (in Arif v Moldova) had to deal with the issue of parallel proceedings. Moldova filed a request in the arbitration to postpone the filing of its counter-memorial in deference to the ECtHR

4 For a comparison of the property protection under international investment law and human rights law see: U Kriebaum, Eigentumsschutz im Völkerrecht, Eine vergleichende Untersuchung zum internationalen Investitionsrecht sowie zum Menschenrechtsschutz (Berlin, Duncker & Humblot, 2008); U Kriebaum, ‘Is the European Court of Human Rights an Alternative to Investor-State Arbitration?’ in PM Dupuy, EU Petersmann and F Francioni (eds), Human Rights in International Investment Law and Arbitration (Oxford, Oxford University Press, 2009). 5 On the issue of competing jurisdiction between international courts and tribunals, see, eg, Y Shany, The Competing Jurisdiction of International Courts and Tribunals (Oxford, Oxford University Press, 2004); N Lavranos, ‘Regulating Competing Jurisdictions Among International Courts and Tribunals’ (2008) 68 Zeitschrift fuer Auslaendisches Oeffentliches Recht und Völkerrecht 575; C McLachlan, Lis Pendens in International Litigation (Leiden, Martinus Nijhoff, 2009); J Magnaye and A Reinisch, ‘Revisiting Res Judicata and Lis Pendens in Investor-State Arbitration’ (2015) 15 The Law and Practice of International Courts and Tribunals 264; C Titi, ‘Res Iudicata and Interlocutory Decisions under the ICSID Convention: Antinomies over the Power of Tribunals to Review’ (2018) 33 ICSID Review 358. 6 Le Bridge v Moldova, App No 48027/10 (ECtHR, 19 April 2018). 7 Arif v Moldova, Award, 8 April 2013. 8 ibid paras 306, 633.

The European Court of Human Rights’ Case Law on International Investment Issues 483 proceedings. The Tribunal noted that the respondent had raised before the ECtHR an objection to admissibility in favour of the ICSID proceedings. Moreover, the Tribunal indicated that it was not persuaded that the proceedings before the ECtHR were substantially similar to the ICSID proceedings, given that they relate to different claimants, different scope of claims and different relief.9 Mentioning the respondent’s own pleading before the ECtHR, the Tribunal denied the request to postpone the filings.10 Le Bridge continued its case before the ECtHR. It accepted that the parallel ICSID claim did constitute ‘another procedure of international investigation or settlement’ in the sense of Article 35 § 2(b) ECHR.11 However, Le Bridge claimed that the claimants were different.12 The Court had to assess whether the two matters were ‘substantially the same’ as prohibited by Article 35 § 2(b).13 It held that the essence of the argument in both proceedings was that the domestic civil proceedings were unfair. Then it turned to the complainant and stressed that the applicant company only had one shareholder. That sole shareholder, Mr Arif, was directly involved in the proceedings before the ECtHR and was the claimant in the arbitration. Therefore, the Court decided to pierce the ‘corporate veil’ and disregarded the company’s legal personality.14 The Court had previously held in Agrotexim v Greece,15 a case that concerned the legal standing of a shareholder, that piercing the corporate veil would only be justified in exceptional circumstances. In Le Bridge, it decided that these exceptional circumstances were fulfilled because the sole owner and shareholder brought the arbitral complaint. It held that, therefore, the complaints before the Court and the case before the ICSID Arbitral Tribunal were in substance submitted by the same complainant.16 Furthermore, the complaint was substantially the same as a matter that had already been submitted to another procedure.17 The Court did not discuss the difference of the legal basis – a BIT and the ECHR – and rejected the application as inadmissible in 2018.

III. SOVTRANSAVTO HOLDING V UKRAINE – SHARES

The legal standing of shareholders is an important issue under investment law as well as under general international law18 and human rights law. Investments often take place through the acquisition of shares in a company that has a nationality different from that of the investor. The protection of shareholders’ rights raises a number of issues. The most basic issue is whether shareholding qualifies as a protected right at all. The second issue concerns the position of shareholders in relation to the company. In particular, does a shareholder have the right

9 ibid

para 14.

10 ibid. 11 Le

Bridge v Moldova, above (n 6) paras 23, 26. para 23. 13 ibid para 26. 14 ibid para 30. 15 Agrotexim and others v Greece (1995) Series A no 330-A, para 66. 16 Le Bridge v Moldova, above (n 6) paras 31–33. 17 ibid para 33. 18 Barcelona Traction, Light and Power Co, Ltd (Belgium v Spain) (Judgment) [1970] ICJ Reports 3, 4. The ICJ recognised that this might be different under treaties: paras 89/90. The ICJ also recognised that the exclusion of shareholders’ rights might not apply if the company was registered in the state inflicting the damage: para 92. See however, Diallo (Guinea v DRC) (Preliminary Objections, Judgment) [2007] ICJ Reports 582, 614–616, paras 87–94. 12 ibid

484 Ursula Kriebaum to pursue a claim independently of the company? Finally, are shareholders entitled to pursue a claim only in respect of their ownership of the shares or also for diminution of the value of the company? Under the ECHR, it is well established that ownership of shares is protected, in principle, by Article 1, Protocol No 1.19 A prominent example is Sovtransavto Holding v Ukraine. The applicant was a Russian company which held 49 per cent of the shares in a Ukrainian company. The applicant alleged violations of Article 1 of Protocol No 1, contending that it had lost control over the management and assets of the Ukrainian company. It complained that the compensation it had received after the liquidation of the company was not commensurate with its original share in the Ukrainian company’s capital. The Court pointed out that a company share certifies that the holder possesses a portion of in the company together with the corresponding rights.20 This was not just an indirect claim on the company’s assets but also covered other rights like voting rights and the right to influence the company.21 The shares undoubtedly had an economic value and constituted ‘possessions’ within the meaning of Article 1, Protocol No 1.22 When addressing the interference, the Court stressed that the Convention also contains positive obligations for the protection of property. States have to protect property rights by providing effective judicial remedies for instances of interference by private parties. These judicial procedures have to fulfil fair trial guarantees. It stated: As regards the right guaranteed by Article 1 of Protocol No. 1, those positive obligations may entail certain measures necessary to protect the right of property (…), even in cases involving litigation between individuals or companies. This means, in particular, that the States are under an obligation to afford judicial procedures that offer the necessary procedural guarantees and therefore enable the domestic courts and tribunals to adjudicate effectively and fairly any disputes between private persons.23

Since the Court found serious procedural shortcomings in the domestic procedures and considered them to be unfair, it held that Ukraine had violated its positive obligation under Article 1 of Protocol No 1.24 The ECtHR did not discuss whether the Russian company, a shareholder in the Ukrainian company, had standing independently from the Ukrainian company. This might have two reasons. First, the preliminary exceptions of the Ukrainian Government did not contain an allegation that the applicant would not have standing because it was a shareholder and that only the Ukrainian company would have victim status and therefore standing. Second, the facts of the case clearly indicated that the interference concerned rights linked to the shares and not only financial losses of the applicant due to interference with the Ukrainian company’s property rights. At the same time, the ECtHR has adopted a restrictive attitude towards shareholders who have acted independently of the company in pursuit of claims arising from acts that adversely affected the company. In some cases, claims by majority shareholders have been admitted on the ground that the claimants had carried out their own business through the medium of the

19 See, eg, Bramelid & Malmström v Sweden, App No 8588/79, 8589/79, (ECtHR, Decision of 12 October 1982) DR 29 (1982), 64, 81; Lithgow et al v United Kingdom, App No 9006/80; 9262/81; 9263/81; 9265/81; 9266/81; 9313/81; 9505/81 (ECtHR, 8 July 1986). 20 Sovtransavto Holding v Ukraine, App No 48553/99 (ECtHR, 27 September 2001), para 13. 21 ibid 13. 22 ibid 14; Sovtransavto Holding v Ukraine, App No 48553/99 (ECtHR, 25 July 2002), paras 91, 92. 23 ibid para 96. 24 ibid paras 97, 98.

The European Court of Human Rights’ Case Law on International Investment Issues 485 companies and were hence directly affected. The fact that it was not their shareholding as such that was affected, but rights of the company which in turn led to a loss in the value of the shares, did not affect the standing of these shareholders.25 By contrast, claims by minority shareholders have been declared inadmissible even though the value of their shares was affected.26 What was decisive as to the independent standing of shareholders in these cases was not mere majority shareholding in itself but the fact that the company was a vehicle through which shareholders conduct their business. In international investment law the situation is different. Most treaties include shareholding or participation in a company in their definitions of ‘investment’.27 Therefore, shareholding is an investment and shareholders are protected separately from the corporation. The nationality of the shareholder of the corporation (the investor) is decisive for protection by an investment protection treaty. The shareholder may pursue claims for adverse action by the host state against the local company that affects its value and profitability. Arbitral practice on this point is extensive.28 Investment protection treaties also protect minority shareholders.29 The protection is not restricted to ownership of the shares but extends to adverse action by the host state in violation of treaty guarantees affecting the local company’s economic position and hence the value of shares.30 25 X v Austria, App No 1706/62 (Commission Decision 4 October 1966) 130; Kaplan v United Kingdom, App No 7598/76 (Commission Decision, 17 July 1980) para 23. See also Pine Valley Developments Ltd and Others v Ireland, App No 12742/87 (ECtHR, 29 November 1991) para 42. 26 Yarrow PLC et al v United Kingdom, App No 9266/81 (Commission Decision, 28 January 1983) para 185. 27 See, eg, Article 1 of the Treaty between the United States of America and the Argentine Republic Concerning the Reciprocal Encouragement and Protection of Investment, entered into force 20 October 1994, https://perma. cc/6N8M-QGJB. 28 See, eg, Antoine Goetz and others v Republic of Burundi, Award (10 February 1999) 6 ICSID Reports 5; Emilio Augustín Maffezini v The Kingdom of Spain, Decision on Jurisdiction (25 January 2000) 5 ICSID Reports 396; Compañía de Aguas del Aconquija, PA & Compagnie Générale des Eaux v Argentine Republic (the Vivendi case), Decision on Annulment (3 July 2002) 6 ICSID Reports 340; Azurix Corp v Argentine Republic, Decision on Jurisdiction (8 December 2003) 43 ILM 259; LG&E Energy Corp v Argentine Republic, Decision on Jurisdiction (30 April 2004); AMT v Zaire, Award (21 February 1997) 5 ICSID Reports 11; Alex Genin, Eastern Credit Limited, Inc and AP Baltoil v The Republic of Estonia, Award (25 June 2001) 6 ICSID Reports 241; CME Czech Republic B V (The Netherlands) v The Czech Republic, Partial Award (13 September 2001); Camuzzi v Argentina, Decision on Jurisdiction (11 May 2005) paras 12, 78–82, 140–142; Gas Natural v Argentina, Decision on Jurisdiction (17 June 2005) paras 32–35, 50–51; AES Corporation v Argentina, Decision on Jurisdiction (26 April 2005) paras 85–89; Compañía de Aguas del Aconquija, SA & Vivendi Universal SA v Argentina (Vivendi II), Decision on Jurisdiction (14 November 2005) paras 88–94; Continental Casualty v Argentina, Decision on Jurisdiction (22 February 2006) paras 51–54, 76–89. Decisions of investment tribunals that have not yet appeared in print are available at www.italaw.com; Suez, Sociedad General de Aguas de Barcelona SA, and InterAguas Servicios Integrales del Agua SA v Argentina, Decision on Jurisdiction (16 May 2006) paras 46–51; National Grid v Argentina, Decision on Jurisdiction (20 June 2006) paras 147–158, 165, Award (3 November 2008) para 126; Pan American v Argentina, Decision on Preliminary Objections (27 July 2006) paras 209–222; Parkerings v Lithuania, Award (11 September 2007) paras 250–254; Impregilo v Argentina, Award (21 June 2011) paras 110–140, 238–246, 271; Busta v Czech Republic, Award (10 March 2017) para 191. 29 See, eg, AAPL v Sri Lanka, Award (27 June 1990) 4 ICSID Reports 246; LANCO v Argentina, Decision on Jurisdiction (8 December 1998) 5 ICSID Reports 367; Vivendi v Argentine Republic, Decision on Annulment (3 July 2002) 6 ICSID Reports 340; CMS Gas Transmission Company v Republic of Argentina, Decision on Jurisdiction (17 July 2003) 42 ILM 788; Champion Trading Co and Ameritrade International Inc v Arab Republic of Egypt, Decision on Jurisdiction (21 October 2003); GAMI Investments, Inc v Mexico, Award (15 November 2004); LG&E Energy Corp v Argentine Republic, Decision on Jurisdiction (30 April 2004) paras 50–63; Sempra Energy v Argentina, Decision on Jurisdiction (11 May 2005) paras 92–94; El Paso Energy v Argentina, Decision on Jurisdiction (27 April 2006) para 138; Phoenix v Czech Republic, Award (15 April 2009) paras 121–123; Hochtief v Argentina, Decision on Jurisdiction (24 October 2011) paras 112–119. 30 See, eg, CMS v Argentina, Decision on Jurisdiction (17 July 2003) paras 59, 66–69, Decision on Annulment, 25 September 2007, paras 58–76; Azurix v Argentina, Decision on Jurisdiction (8 December 2003) paras 69, 73, Decision on Annulment (1 September 2009) paras 57–62, 76–80, 86–130; Enron v Argentina, Decision on Jurisdiction (14 January 2004) paras 35, 43–49, 58–60 and Decision on Jurisdiction (Ancillary Claim) (2 August 2004) paras 17, 34–35; Siemens v Argentina, Decision on Jurisdiction (3 August 2004) paras 125, 136–150; GAMI v Mexico, Award

486 Ursula Kriebaum IV. AGROTEXIM HELLAS S.A. ET AL V GREECE – SHAREHOLDERS

Agrotexim Hellas SA et al v Greece illustrates that majority shareholding in itself does not automatically lead to the victim requirement in Article 34 of the Convention being satisfied. It is the leading case on how the ECtHR handles the problem of the so-called ‘corporate veil’ and shareholders acting independently of the company before the Court.31 First, the European Commission of Human Rights, which existed until 1998, decided in 1992 on the admissibility of the case. Thereafter, the Court, although not bound by positive admissibility decision of the Commission, passed its judgment in 1995. The applicants were six Greek limited liability companies. Together, they were majority shareholders (51.35 per cent) in a brewery. The state-controlled National Bank of Greece was the main creditor of the brewery. To overcome financial difficulties, the brewery negotiated development projects for two factory plants in central Athens. The City of Athens wanted to turn the areas of the factory plants into public parks and announced that it was going to expropriate the plots but never formally expropriated them. This led to the breakdown of negotiations with potential investors in these plants and caused significant financial losses for the brewery’s shareholders. Because of the financial difficulties of the brewery, the General Meeting of Shareholders decided to wind up of the company. In August 1983, it appointed one liquidator according to the facts as presented in the Commissions’ Report32 and two liquidators according to the facts as presented to the Court.33 In November 1983, the Greek authorities commenced liquidation of the company under a special procedure.34 As reported in the facts in the Commission’s Report, this required the appointment of two liquidators, one representing the interests of the brewery’s management and the other representing the interests of the National Bank.35 In 1984, these liquidators replaced the liquidator(s) appointed by the General Meeting of Shareholders and had to act in common.36 The liquidators did not pursue remedies against disadvantageous domestic judgments concerning the two factory plots.37 The City undertook various measures to highlight that it intended to turn the two plots into public parks. Furthermore, in 1989 services of the municipality occupied one of the plots and began demolition works. In 1991, the Court of Appeal replaced these two by a single liquidator representing the interest of the National Bank of Greece.38 The shareholders complained that the various measures taken by the municipality of Athens constituted a violation of their right to peaceful enjoyment of their possessions contrary to Article 1 of Protocol No 1.

(15 November 2004) paras 26–33; Camuzzi v Argentina, Decision on Jurisdiction (11 May 2005) paras 45–67; Sempra Energy v Argentina, Decision on Jurisdiction (11 May 2005) paras 73–79; Continental Casualty v Argentina, Decision on Jurisdiction (22 February 2006) para 79; Bogdanov v Moldova, Award (22 September 2005) para 5.1; Total v Argentina, Decision on Jurisdiction (25 August 2006) para 74; RosInvest v Russia, Award (12 September 2010) paras 605–609, 625; Pezold v Zimbabwe, Award (28 July 2015) para 326. 31 For a critical discussion of the case see: M Emberland, ‘The Corporate Veil in the Case Law of the European Court of Human Rights’ (2003) 63 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 945; SCC Tishler, ‘A New Approach to Shareholder Standing before the European Court of Human Rights’ (2014) 25 Duke Journal of Comparative & International Law 259. 32 Agrotexim Hellas PA and Others v Greece, App No 14807/89 (Commission Decision, 12 February 1992) para 151. 33 Agrotexim and Others v Greece (1), App No 14807/89 (ECtHR, 24 October 1995) para 15. 34 ibid para 17. 35 Agrotexim Hellas PA and Others v Greece, above (n 32) paras 151, 152. 36 Agrotexim and Others v Greece, above (n 33) para 20. 37 Agrotexim Hellas P.A. and Others v Greece, above (n 32) para 152. 38 ibid para 156.

The European Court of Human Rights’ Case Law on International Investment Issues 487 None of the applicants alone held a majority of the shares but jointly they owned 51.35 per cent. The Commission decided that majority ownership could not be the only criterion to determine whether a shareholder may claim victim status. It identified two exceptions to the rule of ‘no piercing of a corporate veil’ in the case of corporate legal personalities. One exception is where the shareholders are using the company simply as a vehicle to carry out their own business. The other exception is where the company was not capable itself of lodging an application. The Commission said: the question whether a shareholder may claim to be victim of measures affecting a company cannot be determined on the sole criterion of whether the shareholder holds the majority of the company shares. This element is an objective and important indication but other elements may also be relevant  … it [the Commission] has previously taken into account the fact that an applicant shareholder was carrying out its own business through the medium of the company and that he had a personal interest in the subject matter of the complaint (…). It has also considered whether it was open to the company itself, being the direct victim, to lodge an application with the Commission.39

The Commission took note of the fact that the company was in liquidation. It stressed that after the imposition of the special liquidation procedure in 1983, the company was essentially under state control. This was so because of the manner in which the liquidators had been appointed and had to act.40 Therefore, the Commission concluded that the brewery could not reasonably be expected to lodge an application with it against the Greek state.41 Hence, the Commission found that one of its two exceptions for a piercing of the corporate veil was satisfied and considered the shareholders to be victims and entitled to standing before it. Therefore, it rejected the respective objection of the government to its jurisdiction.42 The Court mentioned this exception and the assessment of the Commission: The fact that Fix Brewery was subject to a special liquidation procedure meant that it was essentially and effectively under the control of the State so that it was not reasonably an option for the company to lodge a complaint against Greece.43

However, in the following paragraph the Court mentioned that the Commissions seems to accept that where a violation of a company’s rights protected by Article 1 of Protocol No. 1 (P1-1) results in a fall in the value of its shares, there is automatically an infringement of the shareholders’ rights under that Article (P1-1).44

The Court disagreed with this suggestion and pointed out that this approach would create difficulties in establishing who is entitled to apply to the Strasbourg institutions: the shareholders, the creditors or only the company. Furthermore, it pointed out that difficulties concerning the exhaustion of local remedies requirement under Article 35(1) ECHR would also arise since most domestic legal systems would not grant shareholders standing in respect of acts or omissions that are prejudicial to the company.45

39 ibid

para 156.

40 ibid. 41 ibid. 42 ibid. 43 Agrotexim

and Others v Greece, above (n 33) para 63. para 64. 45 ibid para 65. 44 ibid

488 Ursula Kriebaum The Court in accordance with the criteria established in the case law of the Commission, stated that it would allow the piercing of the corporate veil only in exceptional circumstances and mentioned the impossibility of an application by the company itself to the Court. It said: … the piercing of the ‘corporate veil’ or the disregarding of a company’s legal personality will be justified only in exceptional circumstances, in particular where it is clearly established that it is impossible for the company to apply to the Convention institutions through the organs set up under its articles of incorporation or – in the event of liquidation – through its liquidators. … This principle has also been confirmed with regard to the diplomatic protection of companies by the International Court of Justice (Barcelona Traction, Light and Power Company Limited, judgment of 5 February 1970, Reports of judgments, advisory opinions and orders 1970, pp. 39 and 41, paras. 56–58 and 66).46

In contrast to the assessment of the Commission, the Court concluded that the liquidators had had the possibility to complain and that, therefore, the conditions for the exception had not been met.47 It follows from the above practice that under the European system an independent right of shareholders to claim a violation of Article 1 of Protocol No 1 is subsidiary to the right of the company itself and will be recognised only in exceptional cases. This will be the case, in particular, where the company itself does not have the possibility to pursue the claim. Unlike the Commission, the Court considered this condition not to be fulfilled in the case at issue.48 It found that it was not clearly established that the brewery could not have applied to the Commission through its liquidators. Therefore, it denied the shareholders the right to apply to the Convention institutions.49 It did not apply different criteria to come to this conclusion but merely characterised the facts of the case differently.50 As observed in the chapter on Sovtransavto Holding v Ukraine, in investment law, unlike human rights law, the claimant’s nationality is essential. In particular, only foreign investments are protected. To honour this difference, contemporary investment treaties give independent standing to foreign shareholders including minority shareholders. They may pursue claims for adverse action by the host state against the local company that affects its value and profitability. Developments in international investment law in this respect have been far more generous towards shareholders than in human rights law.51

V. BIMER S.A. V MOLDOVA – LICENCES OF A FOREIGN INVESTMENT

Bimer SA v Moldova is a case of a foreign investment that turned to the ECtHR for protection.52 Bimer SA was a company incorporated in the Republic of Moldova. From the moment of incorporation Moldovan, American and Bahamian investors owned its shares. The company 46 ibid

para 66. paras 68, 71. 48 ibid para 67. 49 ibid para 71. 50 For conflicting views on this case see: Emberland, above (n 31); Tishler, above (n 31). 51 For detailed treatment see: PA Alexandrov, ‘The ‘Baby Boom’ of Treaty-Based Arbitrations and the Jurisdiction of ICSID Tribunals: Shareholders as ‘Investors’ and Jurisdiction Ratione Temporis’ (2005) 4 The Law and Practice of International Courts and Tribunals 19; C Schreuer, ‘Shareholder Protection in International Investment Law’ (2005) 2 Transnational Dispute Management 3; Emberland, above (n 31); C Schreuer and U Kriebaum, ‘The Concept of Property in Human Rights Law and International Investment Law’ in S Breitenmoser and others (eds), Liber Amicorum Luzius Wildhaber, Human Rights Democracy and the Rule of Law (Baden-Baden, Nomos, 2007); Kriebaum, Eigentumsschutz im Völkerrecht, above (n 4) 75–110; D Gaukrodger, ‘Investment Treaties and Shareholder Claims: Analysis of Treaty Practice’ (2014) OECD Working Papers on International Investment, http://dx.doi.org/10.1787/5jxvk6shpvs4-en; Tishler, above (n 31). 52 Bimer SA v Moldova, App No 15084/03 (ECtHR, 10 July 2007). 47 ibid

The European Court of Human Rights’ Case Law on International Investment Issues 489 qualified as a company owned by foreign investors and thus benefited from special incentives and guarantees under the Moldovan Law on Foreign Investments.53 In 1998 the company obtained two licences to operate a duty-free shop and a bar at the border between Moldova and Romania. In 2002, the Moldovan Parliament amended the Customs Code whereby dutyfree sales outlets were thenceforward restricted to international airports. Following this legal change, the customs department prevented the applicant through an order from continuing to operate its duty-free business. The order terminated the licence to carry on the duty-free business at the customs office. In line with its case law, the ECtHR confirmed that a licence constituted ‘possessions’ within the meaning of Article 1 of Protocol No 154 and held the interference to be a ‘measure of control of use of property’.55 Since the Court did not consider the measure to be an expropriation, the issue whether the applicant would be considered a foreigner under the Convention and thus able to profit from the reference to the ‘general principles of international law’ in Article 1 of Protocol No 1 did not arise.56 The Court observed that for a measure of control of use to be justified, it must be lawful, in the general interest, and proportionate to the aim pursued.57 The main point in dispute between the parties was whether the applicant’s business qualified as a foreign investment in the sense of the Law on Foreign Investment and therefore enjoyed a right to keep incentives concerning customs and tax for a period of 10 years.58 In general, the ECtHR defers to the decision of local courts in the interpretation of domestic law. The Court stresses in this context that its power to review compliance with domestic law is limited. It is in the first place for the national authorities to interpret and apply that law.59 It accepts national court decisions unless they applied the legal provisions in question manifestly erroneously or so as to reach arbitrary conclusions.60 To answer this question the ECtHR deferred to the decision of the Court of Appeal that had found the Law on Foreign Investment applicable to the applicant’s business. The Court of Appeal had considered the amendment of the Customs Code not applicable to the claimant’s business. The Supreme Court set aside the Court of Appeal’s judgment but on different grounds. The Supreme Court found that no interference had occurred but did not put the application of the Law on Foreign Investment to the applicant’s business into doubt. Therefore, by deferring to the Appeal Court’s assessment as to the applicability of the Law on Foreign Investment the ECtHR considered the order of the customs department as unlawful interference with the applicant’s property.61 It therefore found that the interference was a violation of Article 1 of Protocol No 1.62 53 ibid

para 7. para 29. See also, eg, Centro Europa / SRL and Di Stefano v Italy [GC], App No 38433/09 (ECtHR, 7 June 2012), paras 177–178. 55 Bimer SA v Moldova, above (n 52) para 51. 56 On this issue, see: U Kriebaum, ‘Nationality and the Protection of Property under the European Convention on Human Rights’ in I Buffard and others (eds), International Law between Universalism and Fragmentation (Brill, Nijhoff, 2008). 57 Bimer SA v Moldova, above (n 52) para 52. 58 ibid para 53. 59 Tre Traktörer Aktiebolag v Sweden, App No 10873/84 (ECtHR, 7 July 1989), para 58; Allan Jacobsson v Sweden, App No 8/1997/792/993 (ECtHR, 25 October 1989), para 57; Håkansson and Sturesson v Sweden, App No 12585/86 (ECtHR, 21 February 1990), para 47; Beyeler v Italy [GC], App No 33202/96, (ECtHR, 5 January 2000), para 108; Allard v Sweden, App No 35179/97 (ECtHR, 24 June 2003), para 53. 60 See, eg, Beyeler v Italy [GC], App No 33202/96 (ECtHR, 5 January 2000), para 108. 61 Bimer SA v Moldova, above (n 52) paras 55–58. 62 ibid paras 59, 60. 54 ibid

490 Ursula Kriebaum VI. VÉKONY V HUNGARY – TOBACCO RETAIL LICENCE

Vékony v Hungary concerned restrictions on tobacco sales to protect the health of young people, an issue that has led to considerable public debate in the sphere of international investment arbitration. The case arose in the context of a family business that had a licence to sell tobacco products and did so subject to excise tax. In 2012, the Hungarian Parliament enacted a law whereby tobacco retail was to become a state monopoly and tobacco retailers had to be licensed through a concession tender. As a consequence, the applicant lost his tobacco retail licence. He was unable to obtain a new licence under the new rules. The ECtHR considered the licence to be a possession since it guaranteed an important share of the applicant’s turnover.63 Consistent with its case law, it considered the withdrawal of the licence as amounting to an interference with the right to peaceful enjoyment of possession in the sense of Article 1 of Protocol No 1.64 The harmful effects on this branch of the applicant’s business did not change anything in the qualification of the law as an interference with the applicant’s property rights. The statutory removal of the licence amounted to an interference despite the harmful consequences of smoking as facilitated by tobacco retail.65 In line with its case law, the ECtHR held the cancellation and non-renewal of the tobacco licence to be a measure of control of the use of property.66 It did not question the position of the government that the measure was to serve the purpose of combatting underage smoking. Furthermore, it considered this to be in the general interest.67 This is in line with the Court’s case law. The ECtHR considers the national authorities, in principle, to be better placed than international judges to appreciate what is ‘in the public interest’. Therefore, it is for the national authorities to make the initial assessments both of the existence of a problem of public concern and of the remedial action to be taken. The Court allows for a considerable margin of appreciation in this field and will only intervene if a national authority’s judgment is manifestly without reasonable foundation.68 As always in such cases, the Court assessed whether the measure struck a ‘fair balance’ between the demands of the general interest of the community and the requirements of the protection of the individual’s fundamental rights. It stated that a proper balance would not be found if the person concerned had to bear an individual and excessive burden.69 To assess whether the measure was proportional the Court considered the economic impact of the measure on the applicant. Since the loss of the licence reduced the turnover of the business by one-third and led eventually to the winding-up of the shop, it considered the economic impact to be severe.70 63 Vékony 64 ibid

v Hungary, App No 65681/13 (ECtHR, 13 January 2015), para 29. para 29.

65 ibid. 66 ibid para 30. It referred to Tre Traktörer Aktiebolag v Sweden, above (n 59) para 55; Megadat.com SRL v Moldova, App No 21151/04 (ECtHR, 8 July 2008), para 65; Malik v United Kingdom, App No 23780/08 (ECtHR, 13 March 2012), paras 88–89. 67 Vékony v Hungary, above (n 63) para 31. 68 See, eg, James and Others v United Kingdom, App No 8793/79 (ECtHR, 21 February 1986), para 46; Papamichalopoulos v Greece, App No 14556/89 (ECtHR, 23 June 1993), para 37; Former King of Greece v Greece, App No 25701/94 (ECtHR, 23 November 2000), para 87; Malama v Greece, App No 43199/98 (ECtHR, 1 March 2001), para 46; Pincová and Pinc v Czech Republic, App No 36548/97 (ECtHR, 5 November 2002), para 48; Zvolský and Zvolska v Czech Republic, App No 46129/99 (ECtHR, 12 November 2002), para 67; Olczak v Poland, App No 30417/96, (ECtHR, 7 November 2002), para 77; Jahn and Others v Germany [GC], App No 46720/99, 72203/01, 72552/01, (ECtHR, 30 June 2005), para 91. 69 Vékony v Hungary, above (n 63) para 32. 70 ibid para 33.

The European Court of Human Rights’ Case Law on International Investment Issues 491 It then considered the transitional periods between the enactment of the impugned law and the deadline for terminating the tobacco retail. Only 10 months passed between the enactment of the law and the deadline for the termination of the retail. From the moment the applicant learned that he had not been granted a new licence to the moment he had to stop selling tobacco, only three months elapsed.71 The Court considered these periods as too short for a business that had for neatly twenty years benefitted from such a licence. Furthermore, the Court had already previously found that licencing policies that are arbitrary, discriminatory or disproportionately harsh violate Article 1 of Protocol No 1.72 In addition, the Court requires from the state parties that an interference with the peaceful enjoyment of possessions has to be accompanied by procedural guarantees even though the text of Article 1 of Protocol No 1 does not explicitly contain such a requirement. Among these guarantees is the right of the applicant to have a reasonable opportunity to present his case to the authorities and to effectively challenge the measures. Another factor is the transparency of the procedure leading to the withholding of the licence. Furthermore, the Court took into consideration that the applicant had not been in breach of the law. In view of all these factors, the Court found that these measures imposed an excessive individual burden on the applicant.73 Therefore, like under most BITs, licences are a protected property under the European Convention on Human Rights. Interference with licences will most likely be considered as measures to control the use of property. Even where states revoke licences to protect public health and the Court does not put that into question, the measures must still not be arbitrary, must be applied in a transparent manner, and provide for a possibility of review.

VII. ANHEUSER-BUSCH INC V PORTUGAL – TRADEMARK

Intellectual property rights regularly feature in BITs.74 They have led to high-profile cases before investment tribunals. In the Philipp Morris cases75 investment tribunals had to deal with trademarks, in Eli Lilly v Canada76 with patent rights.77 Anheuser-Busch Inc v Portugal78 is the leading case on intellectual property rights under the European Convention on Human Rights, where the Grand Chamber of the ECtHR had to 71 ibid

para 34. referring to Megadat.com SRL v Moldova, above (n 66) para 79. 73 Vékony v Hungary, above (n 63) para 37. 74 On intellectual property rights and investment protection, see, eg, S Klopschinski, Der Schutz geistigen Eigentums durch völkerrechtliche Investitionsschutzverträge (Köln, Carl Heymanns Verlag, 2011); B Mercurio, ‘Awakening the Sleeping Giant: Intellectual Property Rights in International Investment Agreements’ (2005) 15 Journal of International Economic Law 871; HG Ruse-Khan, ‘Investment Law and Intellectual Property Rights’ in Bungenberg and others (eds), International Investment Law (Oxford, Hart, 2015); C Correa and JE Vinuales, ‘Intellectual Property Rights as Protected Investments: How Open are the Gates’ (2016) 19 Journal of International Economic Law 91; K Lidell and M Waibel, ‘Fair and Equitable Treatment and Judicial Patent Decisions’ (2016) 19 Journal of International Economic Law 145; HG Ruse-Khan, ‘Challenging Compliance with International Intellectual Property Protection Norms in Investor-State Dispute Settlement’ (2016) 19 Journal of International Economic Law 241. 75 Philipp Morris Asia Ltd v Australia, Award on Jurisdiction and Admissibility (17 December 2015); Philipp Morris v Uruguay, Award (8 July 2016). 76 Eli Lilly v Canada, Final Award (16 March 2017). 77 On investment cases and intellectual property law, see, eg, D Khayar and W Ahern, ‘Reliance on Investment Treaty Standards to Claim for Failures to Recognize or Protect Intellectual Property Rights’ (2016) 3(2) BCDR International Arbitration Review 399. 78 Anheuser-Busch Inc v Portugal [GC], App No 73049/01 (ECtHR, 11 January 2007). 72 ibid

492 Ursula Kriebaum examine whether Article 1 of Protocol No 1 applies to trademarks and an application for the registration of a trademark.79 In 1981, Anheuser-Busch, a US company had applied to register the trademark ‘Budweiser’ in Portugal. That application did not immediately lead to a registration since there existed a conflicting prior appellation of origin for beer containing the word ‘Budweiser’ from 1968. In 1986, Czechoslovakia and Portugal concluded a bilateral agreement on the protection of geographical indications, which entered into force in 1987. It provided inter alia for the protection of the terms ‘Českébudějovické pivo’ and ‘Českébudějovický Budvarà’ for beer from the Czech Republic.80 In 1989, Anheuser-Busch brought an action to cancel these protected appellations of origin before a Lisbon court. In 1995, the Court of first instance granted this application and cancelled the protected appellations. As a consequence, Anheuser-Busch’s application for the registration of a trademark was registered. However, the Court of Appeal reversed the decision and refused the registration because of the bilateral treaty. The Supreme Court confirmed the judgment of the Court of Appeal that also covered translations of ‘Českébudějovický Budvar’ and therefore Budweis.81 It did so despite the fact that Anheuser-Busch’s application predated the bilateral treaty by six years. Anheuser-Busch filed a complaint against this judgment with the ECtHR. The Court had to decide two main questions: Are trademarks protected and if so, does this also extend to applications to register a trademark? The case was decided, first, by a Chamber of the ECtHR82 and then was referred to the Grand Chamber which delivered the final judgment in this case.83 Concerning the concept of property protected by the Convention, the Chamber of the ECtHR and the Grand Chamber referred to the autonomous interpretation of the term ‘possessions’ by the Court which is independent of domestic law. In the famous Beyeler v Italy case it had said: [P]ossessions in the first part of Article 1 has an autonomous meaning which is not limited to ownership of physical goods and is independent from the formal classification in domestic law.84

Both the Chamber and the Grand Chamber observed that intellectual property rights enjoy the protection of Article 1 of Protocol No 1 and that a registered trademark constitutes a ‘possession’.85 The Chamber discussed whether the application for the registration of the trademark created a ‘legitimate expectation’ that attracted the protection of Article 1 of Protocol No 1 and denied this. Despite the fact that the Chamber recognised that the legal position possessed ‘a definite financial value’, could in certain circumstances entitle the company to compensation, and conferred on the applicant company a right of priority over

79 On this case see, eg, Kriebaum, Eigentumsschutz im Völkerrecht, above (n 4) 72–73; R Helfer, ‘The New Innovation Frontier – Intellectual Property and the European Court of Human Rights’ (2008) 49 Harvard International Law Journal 1; B Goebel, ‘Trademarks as Fundamental Rights – Europe’ (2009) 99 Trademark Reporter 931; JW Reiss, ‘Commercializing Human Rights: Trademarks in Europe After Anheuser-Busch v Portugal’ (2011) 14 The Journal of World Intellectual Property 176. 80 Anheuser-Busch Inc v Portugal, App No 73049/01 (ECtHR, 11 October 2005), para 18. 81 Anheuser-Busch Inc v Portugal [GC], above (n 78) para 24. 82 Anheuser-Busch Inc v Portugal, above (n 80). 83 Anheuser-Busch Inc v Portugal [GC], above (n 78). 84 Beyeler v Italy [GC], App No 33202/96, (ECtHR, 5 January 2000), para 100. See also Matos e Silva Lda v Portugal, App No 15777/89 (ECtHR, 27 August 1996), para 75; Former King of Greece v Greece, above (n 68) para 60; Tsirikakis v Greece, App No 46355/99 (ECtHR, 17 January 2002), para 53; Forrer-Niedenthal v Germany, App No 47316/99 (ECtHR, 20 February 2003), para 32; Broniowski v Poland [GC], App No 31443/96 (ECtHR, 22 June 2004), para 129; Öneryildiz v Turkey [GC], App No 48939/99 (ECtHR, 30 November 2004), para 124. 85 Anheuser-Busch Inc v Portugal, above (n 80) paras 43, 52.

The European Court of Human Rights’ Case Law on International Investment Issues 493 subsequent applications, it denied the protection of Article 1 of Protocol No 1.86 It justified its finding with the fact that Article 1 of Protocol No 1 only protects existing possessions. The right to use the mark was already disputed at the time of the application for registration. At the time when Anheuser challenged the registration of the appellation of origin, the bilateral treaty had already been in force for more than two years. The Chamber observed in this context that the applicant company ‘could not be sure of being the owner of the trademark in question until after final registration and then only on condition that no objection was raised by a third party’.87 Therefore, the Chamber concluded that Article 1 of Protocol No 1 was not applicable to the case.88 Two dissenting judges disagreed and took the view that the applicant had legitimate expectations that the trademark would be registered. They saw the refusal to register it because of the bilateral treaty Portugal had entered into after the application for registration of the trademark as interference that caused an individual and excessive burden on the applicant and therefore violated Article 1 of Protocol No 1.89 They based their finding on the Portuguese Code of Industrial Design that conferred a right to have their application examined in accordance with the rules in force at the time of an application.90 The Grand Chamber decided that not only a registered trademark but also the application for the registration of a trademark is a substantive interest protected by Article 1 of Protocol No 1.91 It took note of the ‘bundle of financial rights and interests’ that comes into existence with the application for registration of the trademark and the number of potential legal transactions with a substantial financial value that follow the application for registration. It did not consider it to be necessary to enter into a discussion on the existence of legitimate expectation. It agrees with the Chamber that such applications (an application for the registration of a trademark) may give rise to a variety of legal transactions, such as a sale or licence agreement for consideration, and possess – or are capable of possessing – a substantial financial value. The applicant company therefore owned a set of proprietary rights – linked to its application for the registration of a trademark – that were recognised under Portuguese law, even though they could be revoked under certain conditions. This suffices to make Article 1 of Protocol No 1 applicable in the instant case and to make it unnecessary for the Court to examine whether the applicant company could claim to have had a ‘legitimate expectation’.92 After having established that a protected right exists, the Grand Chamber examined whether Portugal had interfered with that right. It decided that the critical question was whether the Supreme Court judgment denying the registration of the trademark was a violation of the Convention. The Supreme Court had denied the registration because of a bilateral agreement on the protection of geographical indications that had been concluded between the two states after the application for the registration of the trademark by Anheuser-Busch but before Anheuser-Busch challenged the registered appellations of origin. The Court found that the dispute was not between the applicant and the state, but between two private parties over the interpretation of Portuguese law and the interpretation of the bilateral treaty. It mentioned in this context that the interpretation of the treaty was disputed. By referring to the treaty 86 ibid

paras 47–48. para 50. para 53. 89 Anheuser-Busch Inc v Portugal, App No 73049/01, (ECtHR, 11 October 2005), Dissenting Opinion of Judges Costa and Cabral Barreto, paras 5–8. 90 ibid para 3. 91 Anheuser-Busch Inc v Portugal [GC], above (n 78) paras 74–76. 92 ibid paras 76, 77. 87 ibid 88 ibid

494 Ursula Kriebaum interpretation as being in dispute, the Court reconciled its case law on retrospective legislation that deprives a property owner of rights with the case law on private to private litigations.93 The ECtHR did not ask whether the conclusion of the bilateral treaty or its application by the Supreme Court constituted an expropriation. Rather, it examined whether the decision by the Supreme Court that gave priority to an appellation of origin protected by the bilateral treaty over the prior application for a trademark was arbitrary or manifestly unreasonable. To decide this, the ECtHR considered whether the proceedings fulfilled procedural guarantees such as a possibility to present one’s case and the right to be heard. The Court decided that this was the case and denied the existence of an interference with the applicant’s rights.94 Two dissenting judges criticised this approach and considered the conclusion of the bilateral treaty and its application by the Supreme Court to be expropriations of foreign property. They considered that a retroactive application of the bilateral treaty caused a dispute opposing the applicant and the state. Since they considered the act to be an expropriation of foreign property, the reference to ‘general principles of international law’ would be triggered.95 Therefore, the principles of non-discrimination, and the requirement to pay prompt, adequate and effective compensation would have to be respected. They therefore found that a violation of Article 1 of Protocol No 1 had occurred.96 The two judgments and the dissenting opinion illustrate the various ways in which the case can be approached. What is common ground is that registered trademarks are a protected property. Otherwise, the Chamber adopted a much narrower approach than the Grand Chamber to intellectual property rights. The Chamber judgment provided for a more limited Court control of state measures in the domain of intellectual property law. The Grand Chamber enlarged the scope of control from registered trademarks to applications for trademarks. However, it distinguished between disputes opposing the state and the individual and litigation between private parties concerning intellectual property rights. In the second situation, it will only examine whether the procedure is fair and will not second-guess whether domestic courts assessed the facts and the law in a correct manner. The Court observes that, even in cases involving litigation between individuals and companies, the obligations of the state under Article 1 of Protocol No 1 entail the taking of measures necessary to protect the right of property. In particular, the state is under an obligation to afford the parties to the dispute judicial procedures which offer the necessary procedural guarantees and therefore enable the domestic courts and tribunals to adjudicate effectively and fairly in the light of the applicable law. However, the Court reiterates that its jurisdiction to verify that domestic law has been correctly interpreted and applied is limited and that it is not its function to take the place of the national courts, its role being rather to ensure that the decisions of those courts are not flawed by arbitrariness or otherwise manifestly unreasonable.97 This finding implies that the state parties to the Convention are under a positive obligation to provide for effective judicial procedures to enforce intellectual property rights.98

93 LR Helfer, ‘The New Innovation Frontier – Intellectual Property and the European Court of Human Rights’ (2008) 49 Harvard International Law Journal 1, 31. 94 Anheuser-Busch Inc v Portugal [GC], above (n 78) paras 79–87. 95 See on this issue: Kriebaum, above (n 56). 96 Anheuser-Busch Inc v Portugal [GC], App No 73049/01, (ECtHR, 11 January 2007), Joint Dissenting Opinion of Judges Caflisch and Cabral Barreto, paras 7–10. 97 Anheuser-Busch Inc v Portugal [GC], above (n 78) para 83. 98 See also Sovtransavto Holding v Ukraine, App No 48553/99, (ECtHR, 25 July 2002), para 96, mentioned above in the section discussing this case. There, the Court also stressed the procedural aspects of the right to the protection of property.

The European Court of Human Rights’ Case Law on International Investment Issues 495 VIII. YUKOS V RUSSIA – TAX COLLECTION

Another investment law case of interest decided by the ECtHR is Yukos v Russia.99 It provides interesting insights into the possibility of parallel litigation in different international fora on the one hand and different approaches by these international fora to legal questions that appear to be similar at first view on the other. The case concerned tax reassessments that led to tax claims by the Russian state against Yukos and the subsequent enforcement measures that led to the liquidation of the company. Russia accused Yukos of having abused a ‘tax optimization scheme’.100 Shareholders of Yukos brought cases before investment arbitral tribunals,101 which then had to deal with the ‘Yukos saga’ and a number of cases were brought before the ECtHR. The OAO Neftyanaya Komapaniya Yukos (Yukos) itself introduced one of the cases before the ECtHR. The company argued that Russia had breached Article 6 of the Convention due to numerous fair trial violations as well as Article 1 of Protocol No 1 taken alone and in conjunction with Articles 1, 7, 13, 14 and 18 of the Convention. It based the complaint concerning a violation of the right to the protection of its property upon ‘the allegedly unlawful, arbitrary and disproportionate imposition and enforcement of the 2000–2003 Tax Assessments’ as well as on the allegedly ‘unlawful, arbitrary and disproportionate’ sale of OAO Yuganskneftegaz, Yukos’s most valuable subsidiary.102 Out of the many aspects of interest in this case, the following analysis will focus on parallel arbitration and the different approaches of investment tribunals and the ECtHR concerning the tax collection and enforcement issue. Yukos lodged its ECtHR application against the Russian Federation in April 2004. Shortly thereafter, in February 2005, the applicant company’s former majority shareholders, Hulley Enterprises Ltd, Yukos Universal Ltd and Veteran Petroleum Ltd, brought arbitration proceedings against the Russian Federation for alleged breaches of the Energy Charter Treaty (ECT) before an investment tribunal.103 They had jointly owned over 60 per cent of the shares in the applicant company. The ECtHR declared Yukos’s application admissible in January 2009 and decided to hold a hearing on the merits of the case. Subsequent to this decision on admissibility, but prior to the hearing, the Court was informed that arbitration proceedings were pending. It invited the parties to address this question at the hearing.

99 Case of OAO Neftyanaya Kompaniya Yukos v Russia, App No 14902/04 (ECtHR, 20 September 2011). For a comparison of the approaches of the ECtHR and the investment arbitration case, see, eg, E De Brabandere, ‘Yukos Universal Limited (Isle of Man) v The Russian Federation’ (2015) 30 ICSID Review 345; HG Dederer, ‘The Yukos Case. A Comparative Case Note on the ECtHR’s Decisions and the PCA Tribunal’s Awards’ (2015) Journal of Siberian Federal University. Humanities & Social Sciences 2062; C Brown, ‘Investment Treaty Tribunals and Human Rights Courts’ (2016) 15 The Law and Practice of International Courts and Tribunals 287. 100 Yukos Universal Limited (Isle of Man) v The Russian Federation, PCA Case No AA 227, Final Award, 18 July 2014, para 90. 101 ibid; Hulley Enterprises Ltd v The Russian Federation, PCA Case No AA 226, Final Award, 18 July 2014 and Veteran Petroleum Ltd v The Russian Federation, PCA Case No AA 228, Final Award 18 July 2014 are virtually identical. The Tribunals were composed of the same three arbitrators and all three were based on the ECT. The references are to the Yukos Universal award. Quasar de Valores SICAV SA et al v The Russian Federation (Spain/Russia BIT), Award (20 July 2012). On this case brought by minority shareholders see I Marboe, ‘Quasar de Valores SICAV SA and others v The Russian Federation’ (2013) 28 ICSID Review 247–253; RosinvestCo UK Ltd v The Russian Federation (UK/Russia BIT), Final Award (12 September 2010). 102 Case of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) para 552. 103 The only cases discussed by the ECtHR were the three investment arbitrations brought by Yukos’s majority shareholders: Yukos Universal Limited (Isle of Man) v The Russian Federation, above (n 100); Hulley Enterprises Ltd v The Russian Federation, above (n 101); Veteran Petroleum Ltd v The Russian Federation, above (n 101).

496 Ursula Kriebaum The ECtHR’s provision dealing with re-litigation and parallel proceedings is Article 35.104 The Court explained that Article 35 § 2(b) of the Convention is intended to avoid a situation where several international bodies would simultaneously be dealing with applications that are substantially the same.105 It had to decide whether the case before it was substantially the same as a matter that had already been submitted to a parallel set of proceedings. To answer the question whether this was the case it relied on the triple identity test.106 The Court said that for assessing the similarity of the cases it would rely on a comparison of the parties in the respective proceedings, the relevant legal provisions relied on by them and the scope of their claims as well as the types of the redress sought.107 The ECtHR observed that the applicants in the case under consideration were different from the claimants in the arbitration. It was the Yukos company itself that had introduced a complaint before the Court. In the arbitrations before the PCA, the applicant company’s former majority shareholders relied on the ECT: Hulley Enterprises Ltd, Yukos Universal Ltd and Veteran Petroleum Ltd. In addition, some minority shareholders were pursuing separate BIT arbitrations.108 The shareholders in the arbitrations had never taken part in the Strasbourg proceedings and Yukos and the shareholders did not coordinate their actions. Therefore, the ECtHR decided that the parties in the arbitration proceedings and in the case before it were different. Hence the two matters were not ‘substantially the same’ within the meaning of Article 35 § 2(b) of the Convention. It therefore did not have to answer the question whether the arbitral proceedings were ‘another procedure of international investigation of settlement’ and declared the case admissible.109 Concerning the merits of the cases, it is striking that all of the investment tribunals decided that the sequence of the Russian measures against Yukos amounted to an indirect expropriation.110 By contrast, the ECtHR did not consider these measures to amount to an indirect expropriation or a de facto expropriation as the Court would call such measures.111 It was common ground between the ECtHR and the investment tribunals that the following facts had occurred: The Russian courts found Yukos guilty of repeated tax fraud, in particular for using an illegal tax evasion scheme involving the creation of sham companies. In the enforcement proceedings Yukos’s assets in Russia were attached, its domestic bank accounts partly frozen and the shares of its Russian subsidiaries seized. Concerning the second year of the tax assessment, the tax ministry had doubled

104 See

above ch 1. of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) para 520. 106 Identity of parties / object of relief / legal grounds. 107 Case of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) para 521. See Vesa Peltonen v Finland (dec), App No 19583/92 (ECtHR, 20 February 1995); Cereceda Martin and Others v Spain, App No 16358/90 (ECtHR, 12 October 1992); Decision on the competence of the Court to give an advisory opinion [GC], para 31. 108 Quasar de Valores SICAV SA et al v The Russian Federation, above (n 101); RosinvestCo UK Ltd v The Russian Federation (UK/Russia BIT), Final Award (12 September 2010). 109 Case of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) paras 523–526. For another example of parallel proceedings with a different outcome see the Le Bridge v Moldova, App No 48027/10 (ECtHR, 19 April 2018), case discussed above at (n 6). 110 Yukos Universal Limited (Isle of Man) v The Russian Federation, above (n 100) paras 1580–1585; Quasar de Valores v Russia, Award, above (n 101) para 177; RosinvestCo UK Ltd v The Russian Federation (UK/Russia BIT), Final Award (12 September 2010) para 633. 111 For de facto expropriations, see, eg, Papamichalopoulos v Greece, above (n 68); Vasilescu v Romania, App No 60868/00 (ECtHR, 22 May 1998); Brumarescu v Romania, [GC], App No 28342/95 (ECtHR, 28 October 1999); Zwierzynski v Poland, App No 34049/96 (ECtHR, 19 June 2001); Karagiannis and others v Greece, App No 51354/99 (ECtHR, 16 January 2003); Popov v Moldavia (no 2), App No 19960/04 (ECtHR, 6 December 2005). 105 Case

The European Court of Human Rights’ Case Law on International Investment Issues 497 the penalty rate from 40% to 80% since Yukos had used the same tax evasion methods in the following years. Yukos was also required to pay bailiffs an enforcement fee, calculated at 7% of the total debt. The authorities refused to suspend or reschedule the payments. The company was required to pay all those amounts within very short deadlines and it made numerous unsuccessful requests to defer payment. The Russian authorities announced the sale of OAO Yuganskneftegaz to collect the tax debt. Its shares were auctioned at considerably below market value and Yukos was declared insolvent and liquidated. 93% of the former assets of Yukos ended up in the hands of the Russian State as a result of the tax debt enforcement measures.112

The investment tribunals found that these acts had been politically motivated attacks by the Russian authorities that eventually led to the destruction of Yukos.113 The ECtHR did not share this finding. It is unclear what led to this different assessment of the facts. One factor that could have contributed is the different manner in which investment tribunals and the ECtHR gather evidence. In most cases the proceedings before the European Court are entirely written and no hearing is conducted.114 In the Yukos case, a public hearing did take place115 but no experts or witnesses were heard.116 By contrast, all the investment tribunals heard witnesses extensively117 and some of the witnesses were also cross-examined.118 This different form of establishing the facts may have led to a different assessment of the intention behind the Russian measures.119 The different assessment of the facts was probably one factor contributing to the different characterisation of the interferences by the dispute settlement bodies. The ECtHR found a violation of the right to a fair trial (Article 6 ECHR) because Yukos had insufficient time to prepare its case before the lower courts. Furthermore, the Court found a violation of the right to the protection of property (Article 1 of Protocol No 1) in two regards. The first concerns the 2000–2001 tax assessment proceedings regarding the imposition and calculation of penalties. The second concerns the disproportionate enforcement proceedings with regard to the debt collection and the penalties. The investment tribunals did not decide whether Russia had violated the fair and equitable treatment (FET) standard which is the closest equivalent to the right to a fair trial protected by Article 6 of the Convention. This was due to the jurisdictional clauses in the UK-Russia and in the Spain-Russia BIT which do not provide for jurisdiction with regard to the fair and equitable treatment standard.120 The ECT contains a provision guaranteeing FET, but the question arises, whether the Russian measures were covered by the tax carveout of the ECT in Article 21.121 This carveout

112 See, eg, Quasar de Valores v Russia, above (n 101) para 169; Case of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) para 555. 113 Yukos Universal Limited (Isle of Man) v The Russian Federation, above (n 100) paras 1180, 1253. 114 Schabas, above (n 2) 806. See also: Practice direction issued by the President of the Court in accordance with Rule 32 of the Rules of Court on 1 November 2003 and amended on 22 September 2008 and 29 September 2014, https://perma.cc/WMQ9-PDHM. 115 Case of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) para 5. 116 In the exceptional case where experts or witnesses were heard the ECtHR noted this in the judgments. See, eg, Husayn (Abu Zubaydah) v Poland, App No 7511/13 (ECtHR, 24 July 2014), para 8. 117 See, eg, Yukos Universal Limited (Isle of Man) v The Russian Federation, above (n 100) paras 117–246. 118 ibid para 120. 119 Dederer, above (n 99). 120 Article 8 Russia-United Kingdom BIT; Article 10 Russia-Spain BIT. 121 Article 21 ECT-Taxation: ‘(1) Except as otherwise provided in this Article, nothing in this Treaty shall create rights or impose obligations with respect to Taxation Measures of the Contracting Parties. In the event of any inconsistency between this Article and any other provision of the Treaty, this Article shall prevail to the extent of the inconsistency. (…)’

498 Ursula Kriebaum excludes FET from the scope of protection of the ECT. It has a claw back, but this covers only expropriations and not violations of the FET standard. In their decisions on jurisdiction, the Tribunals decided that this would be a matter to be decided at a later stage.122 In their final awards, the Tribunals found that in view of the fact that they had found a breach of Article 13 ECT (expropriation) they did not need to consider whether the measures were also in breach of Article 10 ECT that provides for FET. Therefore, none of the Investment Tribunals dealing with the Yukos affair made a substantive statement on the fair trial issue. While all of the Investment Tribunals found that an expropriation had occurred,123 the ECtHR reached a different conclusion. The property protection clause in Article 1 of Protocol No 1 ECHR provides for a differentiated concept of interference comprising three rules.124 The first rule contains the principle of peaceful enjoyment of property; it is set out in the first sentence of the first paragraph (and is called: ‘other interference’ in the case law of the ECHR125). The second rule covers expropriation and subjects it to certain conditions; it appears in the second sentence of the first paragraph. The third rule recognises that the states are entitled, amongst other things, to control the use of property in accordance with the general interest, by enforcing such laws as they deem necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties. The Court, without much discussion, found that in view of the circumstances of the case the complaints had to be examined under the third rule. It allows for the regulation of use and explicitly provides for the possibility to enforce tax laws and penalties.126 The Court discussed the complaint regarding an intentional destruction of the company under the angle of Article 18 of the Convention in combination with Article 1 of Protocol No 1. It decided that the applicants were not able to prove that Russia had pursued the tax collection proceedings with a view to destroying the company and taking control of its assets.127 Therefore, the ECtHR denied a violation of Article 18 ECHR128 in combination with Article 1 of Protocol No 1. By contrast, the investment tribunals found that the government had the intention to destroy Yukos, which led to their finding of an expropriation. Having opted for the third rule, the Court pointed out that it would use three criteria to assess whether Russia’s regulation of use to collect taxes violated the Convention: lawfulness of the measures, whether the measures pursued a legitimate aim and whether they were proportionate to the aim pursued.129 First, the Court found that some of the measures were not lawful. It identified two reasons for unlawfulness: the retroactive change in the rules on the applicable statutory time limit for 122 Yukos

Universal Limited (Isle of Man) v The Russian Federation, above (n 100) paras 585, 586. paras 1580–1585; Quasar de Valores v Russia, above (n 101) para 177; RosinvestCo UK Ltd v The Russian Federation (UK/Russia BIT), Final Award (12 September 2010) para 633. 124 For the text of Article 1 of Protocol No 1, see n 3. 125 See, eg, A van Rijn, ‘Right to the Peaceful Enjoyment of One’s Possessions (Article 1 of Protocol No 1)’ in P van Dijk and others, Theory and Practice of the European Convention on Human Rights (Cambridge, Intersentia, 2006). See, eg, Sporrong and Lönnroth v Sweden, App No 7151/75 and 7152/75 (ECtHR, 23 September 1982); Beyeler v Italy, App No 33202/96 (ECtHR, 5 January 2000), para 106; Sovtransavto Holding v Ukraine, App No 48553/99 (ECtHR, 25 July 2002), para 93; SA Dangeville v France, App No 36677/97 (ECtHR, 16 April 2002), para 51; Broniowski v Poland [GC], App No 31443/96 (ECtHR, 22 June 2004), para 136; Piven v Ukraine, App No 56849/00 (ECtHR, 29 June 2004), para 49. 126 Case of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) para 557. 127 ibid paras 665, 666. 128 Article 18 ECHR: ‘The restrictions permitted under this Convention to the said rights and freedoms shall not be applied for any purpose other than those for which they have been prescribed.’ 129 Case of OAO Neftyanaya Kompaniya Yukos v Russia, above (n 99) para 558. 123 ibid

The European Court of Human Rights’ Case Law on International Investment Issues 499 the collection of fines for tax evasion by a judgment of the Supreme Court and the doubling of the penalties for the tax year 2001 because of a violation of the tax laws in the year 2000. Therefore, the Court considered that the imposition and calculation of the penalties concerning the 2000–2001 tax assessments had been a violation of Article 1 of Protocol No 1. It found the rest of the tax assessments and the fines and interest payments not to be in violation of Article 1 of Protocol No 1.130 Second, the ECtHR found that the measures pursued a legitimate aim of securing the payment of taxes and constituted a proportionate measure.131 Third, the Court decided that the enforcement proceedings had violated Article 1 of Protocol No 1. This was because of the pace of the enforcement proceedings. Furthermore, the Court found the obligation to pay the full enforcement fee of 7 per cent of the tax debt to be disproportionate. Furthermore, the authorities had failed to take the impact of their actions on the company into account and did not strike a fair balance between the legitimate aims sought and the measures employed. This was in violation of Article 1 of Protocol No 1.132 As a conclusion, the ECtHR found that no expropriation had occurred and that Russia was only responsible for a limited number of violations of the rights to a fair trial and of the right to the protection of property. This made a significant difference with respect to the amount of damages granted by the ECtHR. It was substantially lower than the one granted by the investment arbitral tribunals.133 Furthermore, the Court did not grant any damages as a consequence of the fair trial violations because the applicants were not able to prove the causal link between the violations and the pecuniary damage allegedly sustained by the company.134

IX. JAMES V UNITED KINGDOM – AMOUNT OF COMPENSATION (FOREIGNERS/NATIONALS)

The provision on the protection of property in the ECHR does not contain any explicit obligation of compensation for an expropriation or an illegal interference with property rights. The Article only contains an implicit reference when it refers to ‘the conditions provided for by the general principles of international law’: No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law. ….135

The cases presented below shall illustrate the approach of the Court to the amount of compensation required in case of a violation of Article 1 of Protocol No 1.136

130 ibid

para 607. of OAO Neftyanaya Kompaniya Yukos v Russia [GC], App No 14902/04 (ECtHR, 31 July 2014), para 606. 132 ibid paras 657, 658. 133 For an analysis of the Yukos awards see A Senegacnik, ‘The Yukos Saga: A Story of Hidden Matryoshka Dolls’, ch 9 in this volume. 134 Case of OAO Neftyanaya Kompaniya Yukos v Russia [GC], above (n 131) para 19. 135 Emphasis added. 136 See on this issue eg: H Ruiz Fabri, ‘The Approach Taken by the European Court of Human Rights to the Assessment of Compensation for ‘Regulatory Expropriations’ of the Property of Foreign Investors’ (2003) 11 NYU Environmental Law Journal 148; C Kissling and D Kelliher, ‘Compensation for Pecuniary and Non-Pecuniary Loss’ in A Fenyves and others (eds), Tort Law in the Jurisprudence of the European Court of Human Rights 131 Case

500 Ursula Kriebaum One of the early judgments, James v United Kingdom, concerned UK owners of property that were obliged to sell their property (a freehold) by giving the leaseholders a right of purchase. The Court clarified a number of important questions concerning the protection of property under the Convention in this judgment. It held that the nationals of the interfering state could not profit from the reference to the general principles of international law.137 The Court treated the reference to the general principles of international law as including not only their substantive protection but also the requirement of foreign nationality for their application. In other words, according to the Court, foreigners are protected by these general principles of international law while nationals are not.138 The Court did not take any position concerning the content of the clause. It simply repeated the assertion of the applicants:139 The applicants argued in the alternative that the reference in the second sentence of Article 1 (P1-1) to ‘the general principles of international law’ meant that the international law requirement of, so they asserted, prompt, adequate and effective compensation for the expropriation of property of foreigners also applied to nationals.140

Nevertheless, according to the Court’s interpretation of Article 1 of Protocol No 1 the payment of compensation is a necessary condition for the lawful taking of property of anyone within the jurisdiction of a Contracting State. The Article itself implies this. Therefore, the terms of compensation are an important factor in assessing whether an interference imposed a disproportionate burden on the applicant.141 The Court has developed this requirement out of the proportionality principle, which it found to be inherent in the Convention as a whole and which is also reflected in the structure of Article 1 of Protocol No 1. In addition, the Court relied on the national legal systems of the Contracting States. It observed in that context: under the legal systems of the Contracting States, the taking of property in the public interest without payment of compensation is treated as justifiable only in exceptional circumstances …. As far as Article 1 (P1-1) is concerned, the protection of the right of property it affords would be largely illusory and ineffective in the absence of any equivalent principle. Clearly, compensation terms are material to the assessment whether the contested legislation respects a fair balance between the various interests at stake and, notably, whether it does not impose a disproportionate burden on the applicants (…).142

(Berlin, De Gruyter, 2011); V Fikfak, ‘Changing State Behaviour: Damages before the European Court of Human Rights’ (2018) 29 European Journal of International Law 1091. 137 James and Others v United Kingdom, above (n 68) paras 59–66; Lithgow v United Kingdom, App No 9006/80; 9262/81; 9263/81; 9265/81; 9266/81; 9313/81; 9505/81 (ECtHR, 8 July 1986), paras 111–119. 138 See, eg, U Kriebaum, above (n 56) 650–653. 139 James and Others v United Kingdom, above (n 68) para 58; Lithgow v United Kingdom, above (n 137) para 111. 140 James and Others v United Kingdom, above (n 68) para 58. 141 See, eg, ibid para 54; Lithgow v United Kingdom, above (n 137) para 120. See also: The Holy Monasteries v Greece, App No 13092/87 and 13984/88 (ECtHR, 9 December 1994), para 71; Pressos Compania Naviera SA and Others v Belgium, App No 17498/91 (ECtHR, 20 November 1995), para 38; The Former King of Greece and Others v Greece [GC], App No 25701/94 (ECtHR, 23 November 2000), para 89; Platakou v Greece, App No 38460/97 (ECtHR, 11 January 2001), para 55; Malama v Greece, above (n 68) para 48; Zvolský and Zvolska v Czech Republic, above (n 68) para 70; Jahn and Others v Germany [GC], above (n 68) para 94; Straĭn and Others v Romania, App No 57001/00, Judgment of 21 July 2005, para 52; Draon v France [GC], App No 1513/03, Judgment of 6 October 2005, para 79. 142 James and Others v United Kingdom, above (n 68) para 54.

The European Court of Human Rights’ Case Law on International Investment Issues 501 As far as the amount of required compensation is concerned, the Court stated, as it has regularly done ever since, that in principle the compensation must be reasonably related to the value of the property taken:143 the taking of property without payment of an amount reasonably related to its value would normally constitute a disproportionate interference which could not be considered justifiable under Article 1 (P1-1). Article 1 (P1-1) does not, however, guarantee a right to full compensation in all circumstances.144

However, there will be exceptions to this rule if there is a special interest of the state to expropriate and if at the same time less compensation than the fair market value causes no excessive burden for the individual.145 Furthermore, the Court set out in James v United Kingdom that in the context of largescale expropriations for a social reform, nationals may have to accept a heavier burden than non-nationals: Especially as regards a taking of property effected in the context of a social reform, there may well be good grounds for drawing a distinction between nationals and non-nationals as far as compensation is concerned. To begin with, non-nationals are more vulnerable to domestic legislation: unlike nationals, they will generally have played no part in the election or designation of its authors nor have been consulted on its adoption. Secondly, although a taking of property must always be effected in the public interest, different considerations may apply to nationals and non-nationals and there may well be legitimate reason for requiring nationals to bear a greater burden in the public interest than non-nationals.146

In its later case law, the Court clarified that a total lack of compensation is only justifiable in exceptional circumstances.147

143 See, eg, ibid; Lithgow v United Kingdom, above (n 137) para 121. The Holy Monasteries v Greece, above (n 141) para 71; Pressos Compania Naviera SA and Others v Belgium, above (n 141) para 38; The Former King of Greece and Others v Greece, above (n 141) para 89; Lallement v France, App No 46044/99 (ECtHR, 11 April 2002), para 18; Motais de Narbonne v France, App No 48161/99 (ECtHR, 2 July 2002), para 19; Pincová and Pinc v Czech Republic, App No 36548/97 (ECtHR, 5 November 2002), para 53; Broniowski v Polen [GC], App No 31443/96 (ECtHR, 22 June 2004), para 176; Jahn and Others v Germany [GC], above (n 68) para 94; Straĭn and Others v Romania, above (n 141) para 52; Draon v France [GC], above (n 141) para 79. 144 James and Others v United Kingdom, above (n 68) para 54. 145 For cases where the Court found that there is such an exceptional situation, see, eg, James and Others v United Kingdom, above (n 68) para 54; Lithgow v United Kingdom, above (n 137) para 121; The Former King of Greece and Others v Greece, above (n 141) para 78; Senkspiel v Germany, App No 77207/01 (ECtHR, 12 January 2006). For cases where the Court found the compensation to be disproportionally low, see, eg, Platakou v Greece, above (n 141) paras 56, 57; Pincová and Pinc v Czech Republic, App No 36548/97 (ECtHR, 5 November 2002), para 61; Scordino v Italy (no 1) [GC], App No 36813/97 (ECtHR, 29 March 2006), paras 103, 104. 146 James and Others v United Kingdom, above (n 68) para 65. 147 The Holy Monasteries v Greece, above (n 141) para 71. See also eg: Pressos Compania Naviera SA and Others v Belgium, above (n 141) para 38; The Former King of Greece and Others v Greece, above (n 141) para 89; Malama v Greece, above (n 68) para 48; Zvolský and Zvolska v Czech Republic, above (n 68) para 70; Broniowski v Poland [GC], App No 31443/96 (ECtHR, 22 June 2004), para 176; IRS v Turkey, App No 26338/95 (ECtHR, 20 July 2004), para 49; Jahn and Others v Germany [GC], above (n 68) para 94; Straĭn and Others v Romania, above (n 141) para 52; Draon v France [GC], App No 1513/03 (ECtHR, 6 October 2005), para 79. In the case of Jahn and Others v Germany, the Court surprisingly accepted the lack of any compensation because of the exceptional circumstances prevailing in the unique context of German reunification. It stressed the fact that the expropriation had been taken for reasons of social justice and the lack of legitimate expectations of the applicants with regard to the continuance of their ownership position concerning the expropriated land.

502 Ursula Kriebaum X. PAPAMICHALOPOULOS V GREECE – AMOUNT OF JUST SATISFACTION (LAWFUL/UNLAWFUL EXPROPRIATION)

Like general international law and many international investment tribunals, the ECtHR has always distinguished between lawful and unlawful expropriations.148 It did so for example in Papamichalopoulos v Greece. The dispute arose out of a transfer of land to a Navy fund in Greece. The Navy constructed a holiday resort on the applicant’s land.149 The applicant had the title to the land recognised by a Greek court but could not enforce the judgment.150 Several attempts to obtain compensation from the Greek state failed. The ECtHR found that the applicants were ‘unable either to make use of their property or to sell, bequeath, mortgage or make a gift of it’.151 Therefore, the Court found that a de facto expropriation had occurred and that there was and continued to be a breach of Article 1 of Protocol No 1.152 In its judgment on just satisfaction, the ECtHR pointed out that states are free to choose the means to comply with a judgment of the Court. It stated that it is for the state to grant a restitution in integrum if the nature of the breach of the Convention allows for this. If this is not or only partially possible, the Court may afford the injured victim such satisfaction as it decides to be appropriate.153 The ECtHR explicitly referred to general international law, especially the Chorzów case of the Permanent Court of International Justice and its famous quote on reparation for illegal acts.154 The European Court distinguished between lawful and unlawful expropriations. It held that in the case of an unlawful expropriation the person concerned has to be put in a situation equivalent to the one in which they would have been if there had not been a breach of Article 1 of the Additional Protocol.155 The unlawfulness of such a dispossession inevitably affects the criteria to be used for determining the reparation owed by the respondent state, since the pecuniary consequences of a lawful expropriation cannot be assimilated to those of an unlawful dispossession.156 A comparison between the approach of the Court with regard to compensation for expropriations from nationals with the general international law rules applicable to expropriations from foreigners shows that a difference arises only in cases of lawful expropriations. Even then, the difference will only arise in those exceptional situations where two conditions are met: 1) there is a special interest of the state to expropriate and, at the same time; 2) less than full compensation does not cause an excessive burden for the expropriated individual. In general, international law and international investment law, no such exception exists – full

148 See M Pellonpää, ‘Does the European Convention on Human Rights Require “Prompt, Adequate and Effective” Compensation for Deprivation of Possessions?’ in M Tumpamäke (ed), Liber Amicorum Bengt Broms (Helsinki, Finnish Branch of the International Law Association, 1999) 383 et seq. 149 Papamichalopoulos and Others v Greece, App No 14556/89 (ECtHR, 24 June 1993), para 43. 150 ibid. 151 ibid para 45. 152 ibid para 46. 153 Papamichalopoulos and Others v Greece, App No 14556/89 (ECtHR, 31 October 1995), para 34. 154 PCIJ, Case Concerning the Factory at Chorzów (Germany v Poland), Claim for Indemnity – The Merits, Judgment of 13 September 1928, Collection of Judgments, Series A no 17, 47. 155 Papamichalopoulos and Others v Greece, above (n 153) paras 36–40; see also eg: Hentrich v France, App No 13616/88 (ECtHR, 22 September 1994), para 71; Hentrich v France, App No 13616/88 (ECtHR, 3 July 1995), para 11; Bimer SA v Moldova, App No 15084/03, Judgment of 10 July 2007, paras 59, 69–72. 156 Papamichalopoulos and Others v Greece, above (n 153) para 36.

The European Court of Human Rights’ Case Law on International Investment Issues 503 compensation is always required in the case of an expropriation. However, general international law and international investment law do not protect nationals against expropriations without compensation.

XI. GUISO GALLISAY V ITALY – AMOUNT OF JUST SATISFACTION (LAWFUL/UNLAWFUL EXPROPRIATION)

In Guiso Gallisay v Italy the Chamber, as well as the Grand Chamber, departed from this wellestablished case law on the application of Article 41 of the Convention (just satisfaction) to cases of constructive expropriation.157 It did so with an express reference to its previous case law and after careful reflection.158 Under Italian law, the authorities may occupy a plot of land and build on it prior to a formal expropriation. The formal expropriation has to take place within the authorised period of occupation that may not exceed five years.159 The authorised occupation creates an entitlement to compensation.160 In its judgment on the merits the Court found that the authorities had profited from an unlawful occupation of land. In other words, the authorities were able to take possession of the land in breach of the rules governing expropriation in due form, and, inter alia, without compensation being made available at the same time to the applicants.161 In analysing the legal situation, the Grand Chamber acknowledged that it was a case of unlawful expropriation.162 However, it departed openly from its well-established case law. The Court only granted the value of the land as of the date on which the applicants lost their rights of ownership over the property concerned instead of the date of the ECtHR’s judgment as in all previous cases. It also did not grant compensation for loss of enjoyment in the amount of the value of the building as in its previous case law.163 Rather, it made an award for loss of opportunities because of the unavailability of the land during the period of the lawful occupation by the authorities until the loss of ownership. This amount was reduced by compensation already paid at the national level and it ruled on an equitable basis.164 The Court justified its departure from general international law and its previous case law with the statement that the previous method could lead to unequal treatment between applicants, depending on the nature of the public works carried out by the public authorities, which was not necessarily linked to the potential of the land in its original state.165 Judge Tulkens dissented at the Chamber level and Judge Spielmann at the Grand Chamber level. Both were of the opinion that in the case of an unlawful de facto expropriation the Court should not depart from its previous case law that was moreover in line with general international law. That case law had reflected the established practice of distinguishing

157 Guiso Gallisay v Italy [GC], App No 58858/00 (ECtHR, 21 October 2008), para 27; Guiso Gallisay v Italy [GC], App No 58858/00 (ECtHR, 22 December 2009), para 103. 158 ibid Guiso Gallisay v Italy (2008), para 27: ‘La Cour, après mûre réflexion, estime que l’extension de la jurisprudence Papamichalopoulos et autres c Grèce aux cas d’expropriation indirecte, dont la jurisprudence précitée est la conséquence, ne se justifie pas’. Guiso Gallisay v Italy [GC] (2009) paras 98–107. 159 ibid Guiso Gallisay v Italy [GC] (2009) para 16. 160 ibid 17. 161 Guiso Gallisay v Italy [GC], App No 58858/00 (ECtHR, 8 December 2005), para 94. 162 Guiso Gallisay v Italy [GC] (2009), above (n 157) paras 94, 95. 163 ibid para 103. 164 ibid para 107. 165 ibid para 103.

504 Ursula Kriebaum between lawful and unlawful expropriations. Tulkens pointed out that the rule of law has to be respected and that states acting in the public interest have to act in accordance with domestic law.166 Judge Spielmann agreed with Judge Tulkens and laid out the principles that regulate damages for illegal acts in general international law. He pointed to the fact that the new approach is a departure from principles of international law167 and abolishes the distinction between an unlawful taking and a lawful expropriation.168 He considered that the judgment reduced the level of compensation ‘in an arbitrary fashion’.169 This judgment therefore has led to an unhappy and unnecessary fragmentation of international law.

XII. CONCLUSION

Although not originally designed as a body for the settlement of investment disputes, the ECtHR today must be counted as such. From the investor’s perspective, the property protection offered under the ECHR, however, has a number of disadvantages compared to international investment protection, only some of which have become apparent in the cases analysed here. One obvious example is shareholder protection. With regard to licences and trademarks, the analysed cases did not unveil such obvious disadvantages. In the area of tax collection, the differences in the manner of operation of the two systems and in the construction of the property protection norms become obvious. The manner in which investment tribunals and the ECtHR gather evidence differs largely. The property protection clause in the ECHR differs significantly from those in investment protection treaties. It explicitly provides for the right of the state to collect taxes and enforce penalties. The wording suggests a large margin of appreciation of the states in these matters. These factors could be an explanation for the different outcomes of the cases in the ECtHR and the investment arbitral tribunals. Furthermore, the possibility to be awarded less than full compensation for a violation of the Convention is a significant disadvantage for investors compared to the investment law property protection system. Another significant difference, not discussed in the cases selected here, is the requirement of the ECHR to exhaust local remedies before being able to bring a claim. The same is true for the biggest advantage of the ECHR, the absence of the nationality requirement found in international investment law. Therefore, an investor who has no BIT or other investment protection treaty providing for the jurisdiction of an arbitral tribunal at its disposal and has invested in one of the states parties to the ECHR and Protocol No 1 can rely on the property protection offered by the Convention. In many cases, the two systems will not be alternatives since they typically do not exclude each other. This is especially so if the national company turns to the ECtHR and the foreign investor complains before an investment tribunal. But as is illustrated by the Le Bridge case, the situation may be different if the foreign investor is de facto identical with the local company.

166 Guiso

Gallisay v Italy [GC], App No 58858/00 (ECtHR, 21 October 2008), Dissenting Opinion, para 7. Gallisay v Italy [GC], App No 58858/00 (ECtHR, 22 December 2009), Dissenting Opinion, para 11. para 13. 169 ibid para 16. 167 Guiso 168 ibid

The European Court of Human Rights’ Case Law on International Investment Issues 505 In such a situation the ECtHR would disregard the ‘corporate veil’. It would do so probably with regard to shareholder protection and allow a claim by the shareholder independently of the company; but probably also with regard to the question whether the investment claim is substantially the same as the one before the Court and deny standing. In typical investment cases investment arbitration offers clear advantages. But the human rights system can serve as a useful complement.170

170 For a comparison of the two systems, see, eg, Schreuer and Kriebaum, above (n 51); Kriebaum, Eigentumsschutz im Völkerrecht, above (n 4); Kriebaum, ‘Is the European Court of Human Rights an Alternative to Investor-State Arbitration?’, above (n 4).

29 The CJEU Saga GILLIAN CAHILL AND DANIEL SARMIENTO*

I. INTRODUCTION

O

VER THE PAST few years, there have been rapid significant developments in the legal space between EU law and International Arbitration, and in particular in the area of International Investment Arbitration. The reasons for this are multiple. First, historical. The accession of the 12, newest Member States to the EU brought with it a proliferation of what are now commonly known as intra-EU Bilateral Investment Treaties (intra-EU BITs). These BITs all provided for investment arbitration. And with this proliferation came disputes arising out of these intra-EU BITs. The fact that this wave of new investment disputes came as a surprise to many but not all actors in the area is itself surprising when examined from a historical perspective. In fact, during the 1990’s, the EU Commission (the Commission) had actively encouraged the then Member States of the European Economic Community to invest in many Eastern European Countries. These countries similarly had obligations to encourage more foreign direct investment as part of their path to accession.1 Many such foreign direct investments did indeed take place and investment protections were offered to these investors on the basis of fairly standard Bilateral Investment Treaties. However, there was no condition of accession which required the newest Member States to end or modify these Treaties once they formally acceded to the Union. And thus, their extra-EU BITs were transformed into intra-EU BITs. It is this transformation that has raised multiple issues pertaining to the cross-over of EU law and investment law and in particular whether it was possible for two systems of investment protection to operate within the unique sphere of EU law. A particular sub-set of these issues related to whether or not the dispute resolution provisions provided for in these BITs, namely investment arbitration (hereafter referred to as ISDS), were or could be compatible with EU law. The EU law concern was these provisions could result in an investment arbitral tribunal being required to decide issues of EU law that might risk falling otherwise within the jurisdiction of the Court of Justice of the European Union (CJEU). Accordingly, even a potential risk to EU law’s nebulous concept of autonomy was going to be a concern for the CJEU. These new legal issues and disputes helped carve out a * Gillian Cahill is a practising Barrister at law; Daniel Sarmiento is a Professor of Administrative and European Law at the Complutense University of Madrid. 1 See, eg, Art 74 of the Europe Agreement between Romania and the EEC which provided that the aims of investment promotion and protection should be, inter alia, ‘… the conclusion by the Member States and Romania of Agreements for the promotion and protection of investment (…)’.

508 Gillian Cahill and Daniel Sarmiento series of cases from the CJEU on these cross-over questions and put into focus the concept of autonomy of EU law both internally within the EU (in the Achmea decision, see below) and externally (in the Canada Europe Trade Agreement CETA Opinion). Second, political. A series of geopolitical shifts also influenced the cases that eventually found their way to the CJEU in this area of investment law. In 2013, the initial negotiations on a proposed transatlantic trade partnership (TTIP) between the EU and the US included investment arbitration provisions in the trade deal as the mechanism for resolving disputes.2 This in itself was fairly uncontroversial as investment arbitration had been used in trade agreements for many years, having been touted as being the most neutral and appropriate manner in which international disputes involving states and investors could be resolved. But in and around 2013, the tides shifted significantly in Europe as regards the conclusion of TTIP and the inclusion of the ISDS provisions became the focus of public protest and negative media attention.3 ISDS provisions came under fire on numerous fronts. The European public complained of a lack of transparency in the process. In particular a perceived secrecy was attributed to the fact that ISDS arbitrations were to be conducted in private by ‘private’ arbitrators who could therefore be less easily held to account. The impact on a state’s right to regulate was also raised as a concern. As were arbitrator ethics and the potential for frivolous claims. Finally, many EU citizens were concerned by the lack of an appeal against a decision from an ISDS arbitral tribunal. In response, the Commission launched a public consultation on TTIP in 2014 which demonstrated EU citizens having an overwhelmingly negative consideration of ISDS provisions.4 The political sands then shifted further as for this and many other reasons beyond the scope of this chapter, the TTIP negotiations fell apart in 2016. However, armed with the public response to ISDS the Commission adapted its approach to the inclusion of these dispute resolution provisions, proposing the adoption of an investment court in the series of new trade deals actively being negotiated with countries such as Canada. These new dispute resolution provisions, ultimately ended up before the CJEU for an EU legality check prior to the conclusion of the Canada Europe Trade Agreement (CETA) in Opinion 1/17.5 But tricky legal issues remained as regards trade agreements that had been finalised prior to CETA and that contained ISDS provisions such as in the Singapore Free Trade Agreement (Singapore FTA). The Commission itself therefore approached the CJEU seeking approval on the allocation of competences and responsibilities between the Union and Member States as proposed in the Singapore FTA in Opinion 2/15. In the meantime, the CJEU was also seised of a number of preliminary references pertaining to whether ISDS provisions in intra-EU BITs were compatible with EU law (the Achmea decision). The General Court was concerned with a case that raised front and central how EU law was to respond where an arbitral award itself raised a direct clash with state aid obligations (the Micula decision). This chapter looks at how the CJEU’s has responded to this saga of investment law issues raised in EU law terrain over the last 10 to 15 years. 2 For

a history of the EU negotiating texts on TTIP, see https://trade.ec.europa.eu/doclib/press/index.cfm?id=1230. example, see press reports that over 250,000 people reportedly protested at an anti-TTIP rally in Berlin in October 2015 (https://perma.cc/6D7L-DV4U). 4 The Commission sets out the issues it identified and its request for public comment on same in: Commission Staff Working Document, Online public consultation on investment protection and investor-to-state dispute settlement (ISDS) in the Transatlantic Trade and Investment Partnership Agreement (TTIP), 13 January 2015, https://ec.europa. eu/transparency/regdoc/rep/10102/2015/EN/10102-2015-3-EN-F1-1.Pdf. 5 Opinion 1/17, ECLI: EU:C.2019:341. 3 For

The CJEU Saga 509 II. THE SINGAPORE OPINION

On 16 May 2017, the CJEU issued Opinion 2/15 relating to the Singapore FTA between the European Union (EU) and Singapore.6 The decision marked the first of four CJEU encounters with the points of resistance between EU law and Investment Arbitration examined in this chapter. The scope of Opinion 2/15 was however circumscribed to determining whether the EU had exclusive competence to ratify the Singapore FTA. The thornier question of whether the dispute resolution provisions that provide for investment arbitration contained in the Singapore FTA conflicted with EU law in an internal or external EU context was therefore left over to the Achmea decision and the CETA opinion.

A. Background to the Request for the Opinion In 2010, the EU began negotiations for a bilateral trade agreement with Singapore, envisaged as the first step in closer economic integration with the Association of South East Asian Nations (ASEAN). Negotiations ended in October 2014, but their conclusion prompted differences of opinion within the EU’s Trade Policy Committee as to the EU’s competence to ratify the Agreement without the consent of the Member States. The Commission and Parliament held that the EU had capacity to sign and conclude the Agreement on the basis of Article 3(2) of the Treaty on the Functioning of the EU (TFEU) which grants it, ‘exclusive competence for the conclusion of an international agreement when its conclusion is provided for in a legislative act of the Union’. In short, where the TFEU grants the EU exclusive competence over specific areas, it may conclude international agreements pertaining to those areas without the ratification of Member States. The Commission determined that the ‘legislative act’ providing for conclusion of the Singapore FTA was Article 3(1)(e) TFEU, an umbrella provision under which the EU gains exclusive competence over Common Commercial Policy (CCP), whose scope is defined in Article 207(1) TFEU.7 Significantly, the purview of Article 207(1) TFEU was extended in 2009 by the Treaty of Lisbon to include ‘foreign direct investment’ (FDI). This new europeanisation of investment competence forms part of the key political backdrop to the diverging views between the Council and Commission that led to the request for the Singapore FTA Opinion. Contrary to the Commission’s view, the Council and 25 Member States submitted observations stating that the investment provisions in Chapter 9 of the Agreement relating to non-direct foreign investment fell within the exclusive competence of the Member States.

B. Request for an Opinion Arising from the difference in the various negotiating positions, in 2015, the European Commission made a request for the CJEU’s Opinion8 on whether the EU had the requisite 6 Opinion

2/15, ECLI:EU:C:2017:376. 207(1) TFEU: ‘The common commercial policy shall be based on uniform principles, particularly with regard to changes in tariff rates, the conclusion of tariff and trade agreements relating to trade in goods and services, and the commercial aspects of intellectual property, foreign direct investment, the achievement of uniformity in measures of liberalisation, export policy and measures to protect trade such as those to be taken in the event of dumping or subsidies. The common commercial policy shall be conducted in the context of the principles and objectives of the Union’s external action.’ 8 Requested pursuant to the mechanism set out in Art 218(11) TFEU. 7 Art

510 Gillian Cahill and Daniel Sarmiento exclusive competence to conclude the Singapore FTA, specifically querying whether any provisions of the Agreement fell within the EU’s shared competence or within the exclusive competence of the Member States.

C. The CJEU Opinion In its Opinion, the CJEU divided the Singapore FTA into two parts. First, the parts of the Agreement which fell under the EU’s exclusive competence pursuant to Article 3 TFEU. Under Article 3(1)(e) TFEU, the EU has exclusive competence in the area of the common commercial policy (CCP), itself defined in Article 207(1) TFEU, such that only the EU can act. Second, the elements that fell under the shared competence regime and accordingly required unanimous Member State agreement pursuant to Article 4 TFEU. Two main points can be drawn from the Opinion in relation to each of the Court’s divided parts. On the one hand, the Court used a broad interpretation of the scope of the CCP, to determine that the vast majority of the Singapore FTA’s provisions pertain to the exclusive competence of the EU. The Court’s decision was based on applying a wide reading of the ‘direct and immediate effect on trade’ criterion developed in previous CJEU jurisprudence.9 The Court made its ruling, relying on the notion that provisions purely instrumental to the achievement of the objective of trade liberalisation, and therefore subsumed within Article 207(1), fall within the CCP.10 This practical approach enabled a clean separation of the trade and investment elements of the Singapore FTA, theoretically facilitating fast track ratification of the parts of the Singapore FTA that relate to trade. The CJEU’s brief affirmation of the EU’s competence to unilaterally terminate existing treaties between Singapore and the Member States (individually), as envisaged in Article 9.10 of the Singapore FTA, on the basis of the Lisbon Treaty’s expansion of Article 207 TFEU appears to reflect the post-Lisbon Treaty shift in investment competence to the EU.11 On the other hand, two important areas of the Singapore FTA contained provisions that were deemed to be of shared competence and therefore required Member State ratification. As a result, in the Court’s view, the Agreement, in the form presented to the CJEU, could not be concluded by the EU alone and required the unanimous backing of the Member States. These two areas were foreign non-direct investment and investor state dispute resolution.

9 According to the CJEU, an EU act falls within the CCP ‘if it relates specifically to such trade in that it is essentially intended to promote, facilitate or govern such trade and has direct and immediate effects on it’. See, Opinion 2/15, ECLI:EU:C:2016:656, para 36 and related case law, Daiichi Sankyo Co. Ltd and Sanofi-Aventis Deutschland GmbH v DEMO Anonimos Viomikhaniki kai Emporiki Etairia Farmakon, Case C-414/11, ECLI:EU:C:2013:520, para 51; Commission v Parliament and Council, C-411/06, ECLI: EU:C:2009:518, para 71; Regione autonoma Friuli-Venezia Giulia and Agenzia regionale per lo sviluppo rurale (ERSA) v Ministero delle Politiche Agricole e Forestali, C-347/03, ECLI:EU:C:2005:285, para 75. In making this finding in Opinion 2/15, the CJEU departed from Advocate General Sharpston’s Opinion in relation to Transport Services, moral IP rights and trade and sustainable development provisions, all of which areas she had determined not to be within the EU’s exclusive competence. See to that effect, Opinion 2/15, ECLI:EU:C:2016:656, para 305 and Opinion of Advocate General Sharpston, 2/15, ECLI:EU:C:2016:992, para 570. 10 Opinion 2/15, ECLI:EU:C:2016:656, paras 134–135. 11 ibid paras 247–249.

The CJEU Saga 511 D. Foreign Non-Direct Investment The CJEU held that foreign non-direct investment12 does not fall within the ambit of Article 207(1) TFEU’s exclusive competencies, given that that Article explicitly refers to foreign direct investment and therefore implicitly excludes non-direct investment from its scope. As such, it is shared competence and agreements covering this area that are subject to Member State ratification. The CJEU rejected the Commission’s creative submission that foreign non-direct investment could be nonetheless caught by the ERTA doctrine/Article 3(2) TFEU, which grants the EU competence to conclude an international agreement where it might affect common rules or alter their effect.13 Rather, the Court determined that the case law on which the Commission purported to rely was not applicable to a situation where the EU rule referred to is a provision of the FEU Treaty and not a rule adopted on the basis of the FEU Treaty.14 In other words, the ERTA doctrine and the common EU rules find their limits in secondary legislation but cannot be contained in or consist of primary sources of EU law. Accordingly, the argument that foreign non-direct investment might affect the free movement of capital provisions and thereby Article 63 TFEU was firmly rejected. Title IV of the Singapore FTA which contained the provisions on foreign non-direct investment were deemed to be a shared competence between the EU and the Member States. i. Investor-State Dispute Settlement One of the most contentious issues raised in the request for an Opinion on the Singapore FTA was whether the competence to ratify the ISDS mechanism, contained in the Agreement, was shared between the EU and the Member States or exclusive to the EU. The CJEU emphasised that pursuant to the ISDS provisions of the Singapore FTA, a wouldbe claimant was required to ‘withdraw[..] any pending claim submitted to a domestic court or tribunal concerning the same treatment as alleged to breach the agreement’s investment protection provisions’.15 Accordingly, a regime that removes disputes from the jurisdiction of domestic courts may not be regarded as ancillary or an accessory to such substantive obligations. Consequently, it ‘cannot be established without the Member States’ consent’ and the competence to enter into an ISDS regime was deemed to be a shared one.16 The Court’s reliance on the notion that the ISDS regime removes disputes from the jurisdiction of domestic courts would be revisited by the CJEU in later case law, becoming a key component of the reasoning in the Achmea case. Although the CJEU did not make any express finding on the validity of the ISDS regime, the very mention of it in terms such as removal of disputes from the jurisdiction of domestic courts should have been a warning sign that the Court had concerns about the use of the state prerogative to avoid domestic courts, and by implication, the control of the CJEU.

12 As

to the CJEU’s definition of foreign non-direct investment, see Opinion 1/15, ECLI:EU:C:2016:656, para 227. ERTA doctrine was set out by the CJEU in Commission v Council, C-22/70; ECLI:EU:C:1971:32, and concerned the implied powers of the EU to enter into international agreements. The Court explained that ‘[.]to the extent to which [Union] rules are promulgated for the attainment of the objectives of the Treaty, the Member States cannot, outside the framework of the [Union] institutions, assume obligations which might affect those rules or alter their scope’. 14 Opinion 2/15, ECLI: EU:C:2016:656, para 230. 15 Opinion 2/15, EU:C:2016:656, paras 289–291. 16 ibid para 292. 13 The

512 Gillian Cahill and Daniel Sarmiento Given the circumscribed nature of the issue of the division of competence that formed the request for the Court’s Opinion, it is necessarily silent on a number of important investment law issues. These were later filled in by the CJEU in the Achmea decision.

E. Conclusion Opinion 2/15 walks a careful line between the competing forces of EU policy and Member State control over the same. It departed from the AG’s opinion reserving competence to the EU to allow the Singapore FTA to be divided procedurally, giving effect to the policy impetus behind the Lisbon Treaty’s ‘europeanisation’ of Article 207 TFEU and external investment policy, whilst allowing Member States to retain a shared control over the contentious issue of ISDS. As a consequence of the CJEU’s Opinion, the EU-Singapore FTA as well as the EU-Vietnam Agreement were split into two standalone agreements, with the intention of accelerating the ratification process. Conversely, the reservation of ISDS provisions as a shared Member State competence arguably reflects the competing pressures from Member States to retain a say in the wake of concerns regarding the investment arbitral process as a means of resolving disputes arising from EU trade agreements. As would be seen, what was left over from Opinion 2/15 was later filled in the Achmea decision and the CETA Opinion.

III. THE ACHMEA DECISION

On 6 March 2018, the Court gave its preliminary ruling on the Slowakische Republik v Achmea BV case.17 The CJEU ruled that the investor-state arbitration clause contained in an intra-EU BIT was not compatible with EU law. The Court’s decision caused instant havoc in the investment arbitration world because depending on how widely the Court’s ruling is understood, Achmea has now sounded the death-knell for intra-EU investment arbitration.18 Despite repeated suggestions to the contrary, the CJEU’s findings in Achmea, however, were predicated neither on an innate misunderstanding of investment arbitration nor a hostility towards it. Rather, the Court conducted a formal analysis as it had done in other areas where similar, although not identical, issues concerning the autonomy of EU law as raised through the prism of other parallel systems of dispute resolution had arisen.19 In so doing, the Achmea decision may be understood as a constitutional re-statement of the primacy of the Court of Justice’s jurisdiction, enshrined in Article 344 TFEU and given particular expression by the preliminary reference procedure laid down in Article 267 TFEU.

17 Achmea

BV, C-284/16, EU:C:2018:158. has been repeatedly argued in the investment arbitration community that, as Achmea did not concern an ICSID investment arbitration, then intra-EU arbitrations run under the auspices of ICSID do not fall within the scope of the CJEU’s ruling in Achmea and may therefore still occur within the EU. Arbitral Tribunals have also been receptive to this argument. See, eg, UP and CD Holding Internationale v Hungary, Award (9 October 2018) ICSID Case No Arb 13/35, paras 259 to 267. In this regard, see H Ruiz Fabri and E Gaillard (eds), EU Law and International Investment Arbitration (New York, JurisNet, 2018). 19 See, eg, Opinion 1/91: ECLI:EU:C:1991:490 (Agreement relating to the creation of the EEC), Commission v Ireland, C-459/03; ECLI:EU:C:2006:345, Opinion 1/09, ECLI:EU:C:2011:123 (Patents Court); Opinion 2/13, ECLI:EU:C:2014:2454 (Accession of the EU to the European Convention on Human Rights (ECHR)). 18 It

The CJEU Saga 513 A. Background to the Court’s Judgment The Achmea decision arose out of an investment arbitration brought on the basis of the 1991 Netherlands-Czechoslovakia BIT. This BIT, like many others of its kind, was concluded before Slovakia’s accession to the EU in 1995. It was therefore negotiated and ratified as a third party or extra-EU BIT. Thereafter, in 2004, the Dutch health insurance company Achmea entered the Slovakian market, following economic liberalisation reforms. In 2006, the Slovak Republic partly reversed this liberalisation, altering the environment in which Achmea had made their initial investment. Achmea initiated investment arbitration proceedings challenging this reverse course and was ultimately awarded €22 million plus interest. Setting aside proceedings were then initiated by Slovakia in Frankfurt (the designated seat of the arbitration) on the basis that Slovakia’s accession to the EU had removed the jurisdiction of the Tribunal as otherwise provided for by the BIT. The Oberlandesgericht Frankfurt am Main (Higher Regional Court, Frankfurt am Main, Germany) rejected this challenge, a ruling which Slovakia appealed before the Bundesgerichtshof (Federal Court of Justice), who then subsequently referred the matter for a preliminary ruling from the CJEU. The Bundesgerichtshof referred three preliminary questions to the Court, asking the CJEU to determine whether the ISDS mechanism provided for in Article 8 of the BIT, concluded before one of the Contracting States acceded to the EU but where arbitral proceedings were not to be brought until after that date, was precluded by any of the following tenets of EU law: i. The exclusive jurisdiction of the CJEU (Article 344 TFEU), ii. The preliminary reference procedure (Article 267 TFEU), or iii. The principle of non-discrimination (Article 18 TFEU)?

B. The Court’s Decision Prior to entering into its substantive response, the CJEU spent a considerable part of its judgment outlining the context of this case. The key element for the Court, echoing what has been alluded to earlier in Opinion 1/15, was whether or not any other system for resolution of disputes that is to operate within the EU threatens, even potentially, the autonomy of the EU legal order. For the Court, the structural and symbiotic functioning of the EU’s legal order is dependent on the autonomy of EU law from that of Member States and international law.20 The autonomy of EU law is based in Article 344 TFEU, through which Member States undertake not to submit a dispute concerning the interpretation or application of the Treaties to any method of settlement other than those provided for therein. The Court observed that this autonomy allows for the alignment and application of Union law across the many and varied judicial systems that compose the EU.21 But as repeatedly stated by the Court in its constitutional jurisprudence, it is the preliminary ruling procedure laid out in Article 267 TFEU, that is the fundamental watch-guard of that EU law unity which Article 344 seeks to ensure. The preliminary ruling mechanism, is nothing less than ‘the keystone’ of the EU judicial system 20 Opinion

2/13 (Accession of the EU to the ECHR), EU:C:2014:2454, paras 174 to 176 and the case law cited. relation to case law which shaped the principal of the autonomy of EU law, see further Case 26/62 Van Gend en Loos ECLI:EU:C:1963:1 and Case 6/64 Costa v Enel ECLI:EU:C:1964:66. 21 In

514 Gillian Cahill and Daniel Sarmiento which has ‘the object of securing uniform interpretation of EU law, thereby serving to ensure its consistency, its full effect and its autonomy as well as, ultimately, the particular nature of [EU] law’.22 The Court has, in previous jurisprudence, made very similar pronouncements regarding the autonomy of EU law and the importance of the preliminary reference procedure in preserving this autonomy.23 From an EU law perspective this makes perfect sense. The CJEU’s responsibility is to try and ensure that cracks in the EU legal system that might arise via a disharmonious application of EU law, whereby each Member State picks and chooses how to apply EU law, are to be avoided at all cost. It can only fulfil this responsibility if the preliminary reference procedure functions correctly. Accordingly, the fact that Achmea is an investment arbitration case is – from the CJEU’s perspective – mere context of a legal issue that it has looked at before.24 What matters most is what effect would ISDS have on the keystone preliminary reference procedure itself.25 In this regard, the CJEU examined first, whether the disputes addressed by the Tribunal relate to the interpretation and application of EU law, and second, whether the Arbitral Tribunal constitutes a court or tribunal of a Member State. i. Is an Investment Arbitral Tribunal Required to Apply EU Law? Many BITs contain applicable law clauses where the parties pre-select which law they wish to be applicable to the resolution of their disputes. In many of the previous arbitral awards where Tribunals had been asked to rule on a state’s ‘EU law Defence’, ie where a Respondent state raises the ‘EU law made me do it’, Tribunals avoided the issue by determining that EU law was not part of the applicable law framework that the parties had chosen to apply to their dispute. However, for the CJEU, Article 8(6) of the BIT at stake in Achmea required the Tribunal to take account in particular of the law in force of the contracting party concerned and other relevant agreements between the contracting parties.26 As EU law is an autonomous legal order, first, it ‘must be regarded both as forming part of the law in force in every Member State’ and, second ‘as deriving from an international agreement between the Member States’.27 EU law is therefore doubly applicable. For that reason, a Tribunal may be called on to interpret or indeed to apply EU law and in particular the provisions concerning the fundamental freedoms, including freedom of establishment and free movement of capital. 22 Achmea BV, C-284/16, ECLI:EU:C:2018:158, para 37; Opinion 2/13 (Accession of the EU to the ECHR) of 18 December 2014, EU:C:2014:2454, para 176 and the case law cited. 23 See, eg, in Opinion 1/09, ECLI:EU:C:2011:123 (Patents Court), para 85 where the CJEU could not allow creation of a Patent court that would bypass domestic courts and essentially have exclusive jurisdiction over a part of EU law, even if the Patents’ Court was to engage with the CJEU itself. The CJEU held that the rights of domestic courts to refer issues of EU law to the CJEU pursuant to the preliminary reference procedure would be infringed by the proposed system. In so doing, the Court emphasised that ‘[…] the tasks attributed to the national courts and to the Court of Justice respectively are indispensable to the preservation of the very nature of the law established by the Treaties’. See also, Opinion 2/13 (Accession of the EU to the ECHR), EU:C:2014:2454, paras 176 and 196 to 199. 24 ibid. 25 See Opinion 1/09 (Patents Court), EU:C:2011:123, paras 66 and 85. 26 Art 8(6) of the Netherlands-Czechoslovakia BIT provides: ‘The arbitral tribunal shall decide on the basis of the law, taking into account in particular though not exclusively:

– – – –

the law in force of the Contracting Party concerned; the provisions of this Agreement, and other relevant agreements between the Contracting Parties; the provisions of special agreements relating to the investment; the general principles of international law.’

27 Achmea

BV, C-284/16, ECLI:EU:C:2018:158, para 41.

The CJEU Saga 515 In making this finding, the CJEU implicitly determined that the parties’ choice of applicable law cannot exclude the application of EU law in an intra-EU BIT, where that choice includes the law in force of an EU Member State. ii. Is an Investment Arbitral Tribunal a Court or Tribunal of a Member State Pursuant to Article 267 TFEU? The CJEU responded with uncharacteristic frankness that an investment arbitral tribunal is not a court or tribunal of a Member State pursuant to Article 267 TFEU. In definitively closing the door on this issue, the Court rejected what had been proposed as a potential way to reconcile intra-EU BITs in their current form with the autonomy of the EU legal order. It had been suggested by Advocate General Wathelet that the CJEU had implicitly relaxed the criteria it normally used to determine what is a court of a Member State pursuant to Article 267 TFEU in a series of recent cases to include arbitration between the individual and the state.28 Consequent on this relaxation, an investment arbitral tribunal could fulfil the Court’s criteria29 and ask the CJEU for a preliminary ruling on an issue of EU law that might arise during the course of the investment arbitration. This suggested approach has a number of, at the very least superficially, attractive features. First, it would have allowed a way to bring investment arbitral tribunals that were constituted under an intra-EU BIT into the EU legal order. Having done so, the CJEU could then retain its role as ultimate arbiter of EU law under Article 344 TFEU and could therefore (at least potentially) ensure that the uniform interpretation of EU law was preserved via the preliminary reference mechanism. There would therefore be no risk to the autonomy of the EU legal order, which lest it not be forgotten, is the CJEU’s ultimate concern. This approach would be consistent with what the Arbitral Tribunal in Achmea had observed having found that, ‘Courts and tribunals throughout the EU interpret and apply EU law daily. What the ECJ has is a monopoly on the final and authoritative interpretation of EU law’.30 However, this suggestion also raises serious issues that may have played a role in the Court’s definitive rejection of this approach, although it did not comment on them in the Achmea ruling. For example, in order to determine that an investment arbitral tribunal fulfilled the criteria of Article 267 TFEU, it would have to be determined that either the Arbitral Tribunal, or as was suggested by Advocate General Wathelet,31 the institution behind the Arbitral Tribunal fulfilled the CJEU’s criterion of permanency. However, permanency and arbitral tribunals are, in general, somewhat of a contradiction in terms. Tribunals are not permanent and derive their mandate solely for the time of the arbitration. Once the process is over they are functus officio. Whilst the institutions behind investment arbitral tribunals (for example ICSID or the Permanent Court of Arbitration) are permanent, it seems difficult and unsatisfactory

28 See,

Opinion of Advocate General Wathelet, C-284/16, ECLI:EU:C:2017:699, para 88. to settled case law of the CJEU, in order for a judicial body to be a ‘court or tribunal’ for the purposes of Art 267 TFEU, a number of factors must be taken into account: ‘(i) whether the body is established by law, (ii) whether it is permanent, (iii) whether its jurisdiction is compulsory, (iv) whether its procedure is inter partes and (v) whether it applies rules of law and whether it is independent’. See to that effect, Belov (C-394/11, ECLI: EU:C:2013:48, para 38 and the case law cited; Ascendi Beiras Litoral e Alta, Auto Estradas das Beiras Litoral e Alta (C-377/13, ECLI:EU:C:2014:1754, para 23) and Consorci Sanitari del Maresme (C-203/14, ECLI:EU:C:2015:664, para 17). 30 Eureko B.V v Slovak Republic, Award on Jurisdiction, Arbitrability and Suspension (26 October 2010) PCA Case No 2008–13, para 282. 31 See to that effect, Opinion of Advocate General Wathelet, C-284/16, ECLI:EU:C:2017:699, paras 107 and 108. 29 Pursuant

516 Gillian Cahill and Daniel Sarmiento to ascribe those bodies, who are not exercising the judicial power of the arbitral tribunal, the qualification of permanent. This is particularly so when it is the Tribunal who would be tasked with the interpretation of EU law and with making an eventual preliminary ruling to the CJEU.32 Moreover, there is a significant practical issue in how to follow up on any eventual ruling that the CJEU would make. In the current system, the Treaties are designed to ensure that a failure to comply with EU law and the rulings that the CJEU makes. Thus, the Commission can bring infringement proceedings under Article 258 TFEU. The Court has also, on occasion, had recourse to the idea that a failure to make a preliminary reference by national court judges, leading to Member State sanctions, can also ground a claim for legality on those cases.33 If part of the Commission’s role is therefore to overview Member State compliance with EU law and of national court rulings made to the CJEU, how would an arbitral tribunal which is not attached to any particular Member State fit into this control and compliance regime? Would the Commission then take on that role also in relation to arbitral tribunals? How could a follow up for failure to comply with a preliminary ruling by a Tribunal ultimately be enforced? Would that failure be ascribed to the Member State of the seat of the arbitration? Or of the nationality of one of the contracting parties of an intra-EU BIT? If so, what criteria would be used to select which party? Neither of these options seem satisfactory when one of the purposes of choosing the seat is neutrality and neither contracting party may be desirous to take on the ‘policing responsibility’ of enforcing correct application of preliminary reference. This latter issue raises two further sub-set of issues. The first relates to the arbitral process and whether parties would be amenable to having their arbitrations suspended for a period of at least 16 months whilst a preliminary ruling on EU law takes place. The second relates to the Tribunal. It appears highly improbable that any Tribunal, vested with power to determine the dispute, would want to cede some of that jurisdiction to the CJEU whose ruling, if the Tribunal is assimilated into the EU regime, it would be bound to apply. None of these issues are easily resolved but many are encapsulated in the determination that the CJEU made in rejecting that arbitral tribunals could fall under Article 267 TFEU. The Court focused on the difficulty that arises in trying to integrate an investment arbitral tribunal – that effectively operates on an entirely different legal plane – into the jurisdiction of a Member State for the purposes of Article 267 TFEU. In this regard, the CJEU was clear that the consequence of a tribunal set up by Member States being situated within the EU judicial system is that ‘its decisions are subject to mechanisms capable of ensuring the full effectiveness of the rules of the EU’.34 And subject to all the mechanisms available in the Treaties for ensuring that full effectiveness of the rules of the EU are complied with. The Court found, however, that the Achmea arbitral tribunal is not part of the judicial system of either the Netherlands or Slovakia (the contracting parties to the BIT), on the basis that the exceptional nature of the Tribunal’s jurisdiction is one of the principal reasons for the

32 Despite the Advocate General’s persuasive attempts to argue the contrary, the CJEU’s later jurisprudence does not appear to have relaxed the application of the criteria for establishing a court or tribunal pursuant to Art 267 TFEU. Rather, it is suggested that the Merck decision, on which this line of argument was based, should be viewed in its context; an order of three judges of the Court which formation necessarily presupposes the Court’s decision is limited to applying its settled jurisprudence. On the substance of that case, it is noted that the Portuguese tax arbitration tribunal at issue therein was established by law and tangently attached to the Portuguese judicial system. Accordingly, the issues raised above did not arise. 33 C-416/17, Commission v France, ECLI:EU:C:2018:811 and C-224/01, Kobler, ECLI:EU:C:2003:513. 34 See, to that effect, Opinion 1/09 (Patents Court), ECLI:EU:C:2011:123, para 82 and the case law cited.

The CJEU Saga 517 existence of the BIT’s arbitration clause.35 If the Arbitral Tribunal is not part of the judicial system then its decisions cannot be subject to those same mechanisms and as a consequence the full effectiveness of the rules of the EU is at risk. Moreover, the Court rejected that an arbitral tribunal was akin to a court ‘common to the Member States’ as it had already held that this requires a tribunal collectively set up between Member States and which is vested with the power to refer questions. The distinction here seems based in large part on the fact that Member State courts had supervisory powers over this tribunal thus re-attaching it so to speak to a part of the judicial system of a national court. But no such re-attachment was possible for the Achmea Tribunal which has no power to make a reference to the CJEU for a preliminary ruling. iii. Distinction between Commercial and Investment Arbitration In addition to the findings above, one of the Court’s findings that appeared to cause the most consternation in arbitral circles was the Court’s distinction between commercial and investment arbitration. The Court held that whilst issues of efficiency meant that it was sufficient that commercial arbitral awards were subject to limited judicial intervention within the EU legal system,36 the same considerations did not apply in an intra-EU investment arbitration context. The Court based this finding on the fact that by signing the BIT, two Member States effectively used state power to remove a section of disputes from the confines of the mechanisms set out in the EU Treaties. Such a subversion of the rules of EU Treaties was simply not permitted by Member States. This consideration was notably absent from commercial arbitration cases as they only involve private parties, whereas once a state prerogative has been engaged at Member State level, it sets off a series of corresponding EU law obligations. In this regard, Member States are obliged by the principle of sincere cooperation set out in Article 4(3) TFEU, and which ensures the ‘grounding principles of mutual trust and sincere cooperation’37 among national courts, whose submission to the CJEU ensures its primacy, and in so doing, guarantees the integrity of the Union’s legal order. This principle of mutual trust is therefore the element that binds the Member States within the Union ecosystem, and this is the reason why the Court considers it to be the ‘fundamental premise’ of EU law.38 That two Member States bilaterally agree to circumvent it, is thus on principle a threat to the very foundations of the EU order. The intra-EU BIT is, in symbolic terms, the enemy at the gates of autonomy. The Court also firmly rejected the possibility that the autonomy of EU law may be ensured if the arbitral seat chosen is that of a Member State whose laws allow for judicial review of the final arbitral award.39 The choice of arbitral seat is, entirely within the parties’ discretion, which in turn means that the level of EU law review could vary from EU Member State of seat to EU Member State of seat. Leaving the application of EU law to such a random decision would simply not be permissible from the Court’s perspective.

35 Achmea

BV, C-284/16, EU:C:2018:158, para 45. the possibility of a preliminary reference being made by a national court via an annulment or enforcement application. 37 Achmea BV, C-284/16, EU:C:2018:158, para 34. 38 ibid. 39 Achmea BV, C-284/16, EU:C:2018:158, paras 53 to 55. See further to that effect, Eco Swiss, C-126/97, ECLI: EU:C:1999:269, paras 35, 36 and 40, and Mostaza Claro, C-168/05, ECLI:EU:C:2006:675, paras 34 to 39. 36 ie

518 Gillian Cahill and Daniel Sarmiento iv. International Agreements Where the EU is Actor? A further issue left over in Achmea for a future preliminary reference is what to do with international agreements which could contain similar ISDS provisions as in the Achmea case but to which the EU is a party. Where the EU is the only signatory, on behalf of the Member States, the CJEU has already held that it is not, in principle, incompatible with EU law for the EU to enter into an international agreement providing for the establishment of a court responsible for the interpretation of its provisions and whose decisions are binding on the institutions, including the Court of Justice.40 Thus, the Court reiterated its long stated principle that the competence of the EU in the field of international relations and its capacity to conclude international agreements necessarily entail the power to submit to the decisions of a court which is created or designated by such agreements as regards the interpretation and application of their provisions, provided that the autonomy of the EU and its legal order is respected.41 In the Achmea judgment, the crucial issue for the Court was not that ISDS provisions of an intra-EU BIT entail the submission to the jurisdiction of courts created by an international agreement but rather that such agreements ensure that the autonomy of the EU legal order is respected. In Achmea, an investment arbitral tribunal constituted pursuant to ISDS provisions contained in that agreement could not ensure that the autonomy of the EU legal order was so respected. But perhaps another international agreement, even one containing similar provisions, could do so. In so finding, the CJEU left the door wide open as to the compatibility with EU law of the provisions of an international agreement such as the Energy Charter Treaty (ECT) which also contains ISDS provisions and to which both the Member States and the EU, as well as third countries are parties. Whilst the Commission has long held the view that the ECT cannot operate between Member States, the Court has not yet had the opportunity to rule on this point. The operation of the ISDS provisions contained in the ECT is, however, not expressly excluded by the terms of the ECT itself. Accordingly, when an eventual preliminary reference on this issue arrives at the Court, it will likely conduct the same analysis as it did in Achmea to determine whether it is possible, in an international agreement such as the ECT, for the autonomy of the EU legal order to be preserved and for ISDS provisions to operate intra-EU. In light of the frank assessment made in Achmea and the low threshold for engaging even a potential risk to the autonomy of the EU legal order, it seems likely that the Court’s analysis will be negative. The question left over for the CETA Opinion, was whether there could be a role for extra EU investment arbitration at all within the EU in the post-Achmea world.

C. Conclusion The Achmea judgment has had a seismic impact on the investment arbitration communities, essentially sounding a death knell for investment arbitrations brought consequent on an intra-EU BIT. The reaction from the investment arbitration community was immediate. Respondents in investment arbitration cases raised Achmea as a jurisdictional objection to

40 Achmea BV, C-284/16, EU:C:2018:158, para 57; Opinion 1/91 (EEA Agreement – I), ECLI;EU:C:1991:490, paras 40 and 70; Opinion 1/09 (Patents Court), ECLI:EU:C:2011:123, paras 74 and 76; and Opinion 2/13 (Accession of the EU to the ECHR), EU:C:2014:2454, paras 182 and 183. 41 ibid.

The CJEU Saga 519 the Tribunal seised of their case.42 Tribunals already seised of cases arising out of intra-EU BITs subsequently tried to distinguish Achmea or minimise the ratio of the Court’s decision.43 Not one Arbitral Tribunal conceded that Achmea had any impact on the case they were dealing with or accepted a challenge to jurisdiction based on the CJEU’s ruling. Subsequently, the Commission requested Member States to immediately end their intra-EU BITs which many of them agreed to do.44 Subsequently Member States signed a plurilateral treaty to end existent BITs, pending disputes continue to come before tribunals, under a somewhat unclear ‘structured dialogue’ system provided for in the Termination Agreement.45 Awards that have been rendered arising out of intra-EU BITs have not been complied with and attempted enforcements have had varying degrees of success.46 Considered together, the Court’s finding in Opinion 2/15 that the EU does not have exclusive competence to enter into ISDS provisions in its trade agreements and its finding in Achmea, that these provisions in intra-EU BITs agreed by Member States, are reconcilable. Where the EU will be the party of a trade agreement, then the Member States must agree to the manner in which the EU will resolve those disputes even if, on an external plane, that does not involve the jurisdiction of the CJEU as ultimate arbiter. However, where the agreement operates amongst Member States alone without the ‘Union actor’ element then the full autonomy of EU law remains the key principle that cannot be avoided or minimised via an ISDS provision. Post Achmea, the question remained as to whether or not the Court’s pronouncements on investment arbitration as a mechanism for disputes left any room for it still to be considered a viable option in future EU trade agreements. As will be seen, the CETA opinion showed us that the door on investment arbitration, as least in its traditional form and as originally envisaged in the TTIP agreement, has now been definitively shut. However, the CETA opinion showed a significant evolution on the part of the CJEU over the external facets of the EU’s investment policy and its validation of what, in light of the response given by the Court in Achmea, could very well have been decided in a very different way.

IV. THE CETA OPINION

On 30 April 2019, the Court gave its Opinion 1/17 on the compatibility of EU law with the CETA, following a September 2017 request made by Belgium in particular on the Investment Court System (ICS) provided for in CETA.47 After the failure of the TTIP negotiations, the Commission had determined to change approach, starting with CETA. It had strongly argued that a permanent body such as a multilateral investment court was the best way of resolving the issues surrounding ISDS, whilst

42 See, eg, Gabriel Resources Ltd. and Gabriel Resources (Jersey) v Romania, ICSID Case No. ARB/15/31, where Romania filed an additional objection to jurisdiction based on Achmea. 43 See, eg, Award in Masdar Solar & Wind Cooperatief U.A.v The Kingdom of Spain where the tribunal rejected Spain’s EU law objection to jurisdiction, and attempted to restrict the scope of the Achmea judgment, available at https://perma.cc/R8YW-HEAV. 44 On 15 January 2019, 22 Member States signed a political declaration agreeing to end their intra-EU BITs. 45 On 24 October 2019, a draft plurilateral treaty ending intra-EU BITs was published. See further statement on the draft agreement, accessible at https://ec.europa.eu/info/publications/191024-bilateral-investment-treaties_en. 46 See section IV below concerning attempted enforcement of the award rendered in the Micula case. 47 Opinion 1/17, ECLI:EU:C:2019:341.

520 Gillian Cahill and Daniel Sarmiento continuing to promote international trade and investment.48 The Commission’s aim was that this mechanism would eventually replace the bilateral investment tribunals in future EU FTAs.

A. Request for an Opinion Opinion 2/15, which had determined that ratification of Member States was now required for ISDS provisions contained in EU trade Agreements, triggered the request for the CETA Opinion. The Walloon Region of Belgium’s agreement to the signature of CETA by the Belgian federal Government was precisely conditioned to Belgium referring the validity of the CETA’s Investment Court System (ICS) to the CJEU. Belgium then sought an Opinion from the CJEU on three main issues: i. Whether the CETA ISDS mechanism is compatible with the autonomy of the EU legal order; ii. Whether the envisaged ISDS mechanism is compatible with the general principle of equal treatment and with the requirement of effectiveness, and iii. Whether the Tribunal envisaged by the investment Court provides access to an independent and impartial judiciary?

B. Investment Court System In Section F of CETA, the ICS seeks to transform ad hoc investment arbitration via the creation of a permanent court with an appellate tribunal. Article 8.27 of the CETA, thus provides for the establishment of a permanent tribunal of 15 members (three members appointed by Canada, three by the EU and three from neutral countries). Article 8.27.7 sets out that a division of three Members will ‘[hear] the case on a rotation basis, ensuring that the composition of the divisions is random and unpredictable’. Article 8.28.5 states that the division of the CETA Appellate Tribunal constituted to hear the appeal consists of three randomly appointed Members. Pursuant to Article 8.28.9, an appeal of a CETA Tribunal’s decision to the appellate body is permissible. The CJEU also retains the final word on who should be the Respondent (ie the Union or a Member State) in casu, according to Article 8.21. The Court’s role as the sole interpreter of the division of competencies as set out in the treaties is therefore preserved and the Investment Court must respect the CJEU’s finding in this regard.

C. No Jurisdiction to Interpret EU law Following the Achmea decision, one of the biggest issues for the CJEU in CETA was whether the Investment Court would have jurisdiction to interpret EU law. Similar to Arbitral Tribunals constituted under intra-EU BITs, the Investment Court could also potentially, be required to interpret EU law, therefore raising the same threats to the autonomy of EU law as had occurred in Achmea. 48 See further speech given by Colin Brown, Deputy Head of Unit, Dispute Settlement and Legal Aspects of Trade Policy, Directorate General for Trade, at the 3rd Vienna Investment Arbitration Debate (22 June 2018) entitled ‘The European Union’s Approach to Investment Dispute Settlement’, available at https://perma.cc/HTP8-F85Z.

The CJEU Saga 521 In light of what had been decided in Achmea, the CJEU’s response to this issue, was rather surprising. The Court held that, first, there was no need for any preliminary ruling mechanism as the Investment Court will never interpret EU law but only take it into account as a matter of fact.49 Second, even if the Investment Court did need to take into consideration a domestic measure (which could very well be executing EU law or be EU law), Article 8.31.2 of CETA obliges the Investment Court to follow the CJEU’s prevailing interpretation (as a matter of fact) hence respecting the EU acquis.50 Third, given the nature of arbitration and the need to have speedy decisions, it is absolutely understandable that there is no interim moment of consultation as that would delay the process. The CJEU also justified the lack of domestic review (for example by Member State courts), which had been a critical component of Achmea, because it was precisely to avoid such courts and review that the ISDS process was created.51 Thus, having been extremely critical of the use of a mechanism to avoid the jurisdiction of domestic courts in Achmea and in an intra-EU context, the CJEU appeared to have no difficulty validating such avoidance in external, international context. This apparent double standard may find its conciliation in the fact that on an external plane there is no mutual trust at issue and no Member State obligations under the Treaty triggered. Accordingly, the CJEU was satisfied that the investment court had no jurisdiction to interpret EU law.

D. Compatibility of the CETA ISDS Mechanism with the Autonomy of the EU Legal Order As regards the ICS system, echoes of the Achmea decision are immediately evident, with the Court outlining the fundamental importance of the autonomy of the EU legal order at the outset of its Opinion. Having done so, the Court revisits the issue it raised but left unaddressed in Achmea, stating at paragraph 106 of its Opinion, that an international agreement providing for the creation of a court whose decisions are binding on the EU, is, in principle, compatible with EU law. For the Court, such a possibility is ‘necessary to maintain the powers of the Union in international relations’.52 Further, CETA tribunals’ jurisdiction would be confined to interpretation of the CETA agreement itself, analysed in light of rules and principles of international law applicable to the parties.53 The CETA tribunals do not therefore suffer from the same clash with the autonomy of EU law as had befallen the Patent Court considered in Opinion 1/09, and the Accession Agreement on the European Convention on Human Rights.54 Moreover, the CJEU was quite clear that ‘any meaning given to domestic law by the Tribunal shall not be binding upon the courts or the authorities of that Party’.55 Thus, the Investment Court’s decisions have no direct effect and Member State courts would still be at liberty to raise preliminary issues concerning related questions of EU law via domestic law, even if a CETA tribunal had opined on that issue.

49 Opinion

1/17, ECLI:EU:C:2019:341, para 131. 132. 51 ibid paras 134 and 135. 52 ibid para 117, which references the following jurisprudence: Opinion 2/13 (Accession of the Union to the ECHR), ECLI:EU:C:2014:2454, para 182; see also Opinion 1/91 (EEA Agreement – I), ECLI:EU:C:1991:490, paras 40 and 70, and Opinion 1/09 (Patent Court), ECLI:EU:C:2011:123, para 74). 53 Opinion 1/17, ECLI:EU:C:2019:341, para 118. 54 See fn 56. 55 CETA, Art 8.31.2. 50 ibid

522 Gillian Cahill and Daniel Sarmiento Unlike the intra-EU nature of the Achmea tribunals, CETA also avoided any issues pertaining to the principle of mutual trust that ensures the unity of EU law because this principle is not applicable between the EU and a non-Member State.56 The Court therefore considered it appropriate that the Investment Court would not have the capacity to make a reference for a preliminary ruling to the CJEU and the central stumbling block of Achmea was removed. Although raised, the CJEU did not fully engage with what might occur if the Investment Court failed to take account of EU law (as a fact) correctly or erroneously? Instead the CJEU fell back on the fact that Article 8.28.2(b) of CETA provides that the Appellate Tribunal may also identify ‘manifest errors in the appreciation of the facts, including the appreciation of relevant domestic law’, and that it was in no way the intention of the Parties to confer on the Appellate Tribunal jurisdiction to interpret domestic law.57 From a practical perspective, the CJEU’s approach raises significant issues and conflicting responsibilities for the Member States. What if the Investment Court and/or appellate body wrongly take account of EU law and a Member State or the EU is required to pay out significant monetary sums consequent on an award made by the Investment Court? Or can a Member State refuse to comply with an award, raising a preliminary issue on the issue? No easy answers are found to this conundrum in the CETA Opinion.

E. Possible Threat of the CETA ISDS Mechanism to EU Institutions’ Right to Regulate A further issue particular to CETA had been the perceived threat that CETA might pose to the EU institutions’ right to regulate. This criticism, as mentioned above, is often levied at investment arbitration more generally. By giving investors broad right to sue over alleged failures of investment protection, it has been suggested that a regulatory chill can occur whereby states, fearful of these claims, simply avoid the issue by not enacting potentially important pieces of legislation that might trigger investment claims.58 This regulatory chill is of particular concern where a state is considering enacting legislation that impacts upon the environment and public health rights of the general public. Those rights must be very carefully considered as against those of any individual investor if regulatory chill is to be avoided. In its Opinion, the Court dismissed the possibility that the CETA tribunal may indirectly prevent EU institutions from operating in accordance with the EU’s constitutional framework. This threat hypothetically arises where a tribunal considers an individual investor has not received ‘fair and equitable treatment’ under Article 8.9 of Section D (CETA), or that the investor’s treatment by a Member State has been marked by any of the defects listed in Section C or D of Chapter 8 of CETA. For the Court, this situation results in a substantive as opposed

56 Opinion

1/17, ECLI:EU:C:2019:341, para 129. para 133. eg, SA Spears, ‘The Quest for Policy Space in a New Generation of International Investment Agreements’ (2010) 13 Journal of International Economic Law 1040. ‘Concerns about regulatory chill arise not just from the actual outcome of investment cases – host states have prevailed more often than they have lost in investor-state arbitrations – but also from the inconsistent legal conclusions and reasoning found in arbitral awards. Arbitrators are producing inconsistent international investment law because they are interpreting and applying a patchwork quilt of different IIAs and very broadly worded substantive guarantees, and are not constrained by a formal doctrine of precedent. The uncertainty this creates – coupled with a suspicion that arbitrators drawn from the ranks of economic specialties within international law will interpret substantive guarantees expansively and will find in favor of investors when assessing politically sensitive actions taken by sovereigns – is said to result in regulatory chill.’ 57 ibid 58 See,

The CJEU Saga 523 to formal threat to EU autonomy. The fact that a CETA Tribunal rules on the treatment of a specific investor rather than the legality of the EU measure itself, becomes irrelevant if the EU must repeal or adjust legislation to avoid being repeatedly compelled to pay damages to claimant investors.59 This effect could therefore result in a regulatory chill to the EU institutions’ operation. The Court responded, however, by emphasising that CETA explicitly preserves the parties’ rights to protect public health and environment among others, and that Article 8.9.2 provides that ‘the mere fact that a Party regulates, including through a modification to its laws, in a manner which negatively affects [investments] does not amount to a breach of an obligation under this Section’.60 The Court also notes that, according to Point 1(d) and Point 2 of the Joint Interpretative Instrument, CETA ‘will … not lower [the standards and regulations of each Party] related to food safety, product safety, consumer protection, health, environment or labour protection’, that ‘imported goods, service suppliers and investors must continue to respect domestic requirements, including rules and regulations’, and that CETA ‘preserves the ability of the European Union and its Member States and Canada to adopt and apply their own laws and regulations that regulate economic activity in the public interest’.61 These explicit safeguards are not common in older investment protection agreements which, at best, may contain a general saver of the right to regulate of states parties.62 The Court therefore approved the delicate balance that CETA walks between facilitating both international investment and regulatory autonomy of the EU institutions and the Member States. Given that all measures falling outside the CETA public interest exclusions remain subject to the same threat identified above, the Court here effectively rubber stamps the vision of the CETA negotiators. The ambit of the specific exclusions on public interest grounds is held to be broad enough to ensure the EU’s complete independence in certain areas of strategic importance, beyond which the need to facilitate investment may take precedence.63

F. The Compatibility of the Envisaged ISDS Mechanism with the General Principle of Equal Treatment The CJEU was also required to consider whether the envisaged system was compatible with the equal treatment principle guaranteed in Article 20 of the Charter of Fundamental Rights of the EU (the ‘Charter’). This issue was raised in Achmea from the perspective that Member States parties to an intra-EU BIT were given different protections than other EU Member States and the latter were therefore discriminated against. In Achmea, the Court declined to answer this question in light of the response to the first two preliminary questions.

59 Opinion

1/17, ECLI:EU:C:2019:341, para 148 to 150. para 154. 61 ibid para 155. 62 Newer International Investment Agreements (IIAs) have started to strengthen and introduce explicit exceptions as regards the right to regulate. It is unclear as of yet the extent to which these clauses will be effective. See further, p 8 of the ‘UNCTAD IIAs Issues Note’, https://perma.cc/EJ8W-WJBU. 63 Opinion 1/17, ECLI:EU:C:2019:341, para 158. This reasoning is mirrored in the Court’s approach to covered investments where it states that ‘It must be added that the jurisdiction of the CETA Tribunal to find infringements of the obligation, laid down in Article 8.10 of the CETA, to accord ‘fair and equitable treatment’ to covered investments is specifically circumscribed, since Article 8.10.2 lists exhaustively the situations in which such a finding can be made.’ 60 ibid

524 Gillian Cahill and Daniel Sarmiento In CETA, the difference in treatment referred to by the CJEU stemmed from the fact that enterprises and natural persons of Member States that invest within the EU and that are subject to EU law will not be able to challenge EU measures before the CETA tribunals. By contrast, Canadian enterprises and natural persons that invest within the same commercial or industrial sector of the EU internal market would be able to challenge those measures before the CETA tribunals. For the Court, however, such Canadian investors were not comparable to Member States investors.64 The special status of Canadian investors as ‘foreign investors’ under CETA enabled them to access a specific legal remedy against EU measures. Other investors of Member States in the EU cannot be considered ‘foreign’ and are therefore not permitted to have the same legal remedy. This finding is not in and of itself controversial as these ideas forms a key tenant of all international investment agreements. In all events, such an award is conceivable solely in a scenario where the decision imposing the fine were to be vitiated by one of the defects specified in Article 8.10.2 of the CETA or were to deprive the investor of the fundamental attributes of property in its investment, including the right to use, enjoy and dispose of this investment.65 The CJEU echoes its approach to public policy, observing that 8.10.2 CETA provisions tightly proscribe behaviour that constitutes a breach of fair and equitable treatment, limiting it to, among others, acts of ‘manifest arbitrariness’.66 In circumstances where an EU competition regulator behaved in such a way, EU investors would have parallel recourse to annulment of a fine before Union courts. This reasoning glosses over the category difference between the Investment Court and EU courts, as well as the tendency of competition rulings to impact an investor’s ‘right to use, enjoy and dispose of its investment’,67 especially in relation to Article 102 TFEU. The thrust of the approach, so different to the formalism that characterised Achmea, appears driven by a desire to ensure the EU’s place on the stage of international investment. Yet absent recourse to the principle of mutual trust as the distinguishing factor, the Court’s reasoning sits ill with the all-encompassing vision of the autonomy of the EU legal order as expounded in Achmea.

G. Compatibility with the Principle of Effectiveness: Right of Access to an Independent Tribunal One of the other key findings in the CETA opinion was the CJEU’s approval that the CETA tribunals fulfilled the EU law principle of effectiveness. Having examined their nature, the Court determined that without there being any need to ascertain whether the Parties will formally classify those tribunals as ‘judicial bodies’ or whether their Members, as suggested by Statement No 36, will be called ‘judges’, it follows from the foregoing factors that those tribunals will, in essence, exercise judicial functions.68

The CJEU then examined the criteria that it normally applies to determine whether or not a body is a court or tribunal of a Member State pursuant to Article 267 TFEU69 finding first

64 ibid

para 180. the meaning of point 1(b) of Annex 8-A to CETA. 66 CETA, Art 18.10.1 (C). 67 CETA, Art 18.10.2, to be read within the meaning of Annex 8-A. 68 Opinion 1/17, ECLI:EU:C:2019:341, para 197. 69 See further criteria endorsed by CJEU case law referred to in footnote 33. 65 Within

The CJEU Saga 525 that the CETA tribunals ‘will be permanent and they will be established by law in the form of the acts approving the CETA adopted by the Parties’, will also ‘apply, following an adversarial procedure, rules of law, will be required to exercise their functions wholly autonomously and will issue decisions that are final and binding’.70 Finally, the Court looked at whether the jurisdiction of the CETA tribunals is compulsory. Jurisdiction under a BIT is, generally, asymmetric. The investor retains a choice of remedies (usually including an investment arbitration or domestic courts) whereas the state being sued is required to accept the investor’s choice with no corresponding choice of its own. The state, by signing the BIT, has effectively consented in advance to being a potential Respondent in an unknown amount of investment arbitrations. Whether or not this asymmetrical approach creates sufficient mandatory jurisdiction is arguably one of the reasons why investment arbitral tribunals have trouble fulfilling the criteria to be considered a court or tribunal of a Member State pursuant to Article 267 TFEU. However, for the CETA tribunals, the Court had no issue in validating the CETA approach. The CJEU held that ‘(…) jurisdiction is mandatory not only for the respondent, who will have to accept it under Article 8.25 of the CETA, but also for the claimant investor, in the event that the latter seeks to rely directly on the provisions of the CETA’.71 Accordingly, once an investor chooses to rely on CETA, the asymmetry is resolved. For the CJEU, since Article 30.6 of the CETA deprives investors of the possibility of relying directly on the CETA before the domestic courts and tribunals of the Parties, any action directly based on the provisions of that agreement will have to be brought before the CETA Tribunal. Thereafter, any appeal against the decision of that Tribunal will have to be brought before the CETA Appellate Tribunal.72

In essence, the fact that CETA could only be challenged before a CETA Tribunal was sufficient for the CJEU to find that the CETA Tribunal’s jurisdiction is compulsory, in so far as no other body. Finally, the Court examined its classic criteria of independence and impartiality, holding on both counts that the CETA Tribunals fulfilled these requirements. As regards the level of accessibility and independence that those Tribunals must achieve in order that the Investment Court could be considered to be compatible with Article 47 of the Charter, the Court held that: [T]he purpose of inserting in the CETA provisions concerning non-discriminatory treatment and protection of investments, and the creation of tribunals that stand outside the judicial systems of the Parties to ensure compliance with those provisions, is to give complete confidence to the enterprises and natural persons of a Party that they will be treated, with respect to their investments in the territory of the other Party, on an equal footing with the enterprises and natural persons of that other Party, and that their investments in the territory of that other Party will be secure.73

The Court went on to hold that the creation of the investment court mechanism standing outside the judicial systems of the Parties, (…) is (…) to ensure that the confidence of foreign investors extends to the body that has jurisdiction to declare infringements, by the host State with respect to their investments (…). It is apparent, therefore, that the independence of the envisaged

70 See

further case law referred to in footnote 33. 1/17, ECLI:EU:C:2019:341, para 198.

71 Opinion 72 ibid. 73 ibid

para 199.

526 Gillian Cahill and Daniel Sarmiento tribunals from the host State and the access to those tribunals for foreign investors are inextricably linked to the objective of free and fair trade that is stated in Article 3(5) TEU and that is pursued by the CETA.74

Having set out the specifics of the Investment Court, the CJEU examined the criteria of accessibility and independence separately affirming that the Investment Court complied with both. A few points of note arise that are worth commenting on here. First, in respect of small and medium sized investors, whose accessibility to a tribunal had been a focal point of controversy in the TTIP negotiations, the Court noted that section F of the CETA does not contain any legally binding commitment from the CETA Joint Committee to improve financial accessibility for such investors who may not have the resources to launch expensive proceedings. However, the CJEU held that the accessibility was still satisfied by Statement No 36, which commits the Commission to ‘propose appropriate measures of (co)-financing of actions of small and medium-sized enterprises before that Court’,75 irrespective of the outcome of the discussions within the CETA Joint Committee. Once again, this is a highly practical approach for the Court derogates its own ability to make a factual assessment of whether these measures, once formulated, sufficiently facilitate access to an independent judiciary even prior these measures have been formulated. This is also a significant concession from the CJEU. Second, Belgium’s concerns relating to the independence of the Tribunal pertained in large part to the powers of the CETA Joint Committee to adopt interpretations of the agreement that will be binding on the Tribunal.76 In response, the CJEU noted that the CETA Joint Committee is composed of representatives of both parties, and that the interpretative acts of that Committee have legal effects that stem from, inter alia, Article 31(3) of the Vienna Convention on the Law of Treaties, given it ‘states that account is to be taken, for the interpretation of a treaty, of ‘any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions’.77 As the Court observes, the Union’s participation in interpretive decisions of the Joint Committee is a safeguard against their retroactive or future effect. The Union’s consent to any decision made by the Joint Committee must comply with EU primary law, thereby encapsulating the right to an effective remedy enshrined in Article 47 of the Charter. The Court also accepts fulfilment of the independence criterion despite the Joint Committees’ powers to appoint Members, increase and reduce their numbers and adjust their remuneration, on the basis of the Investment Court’s uniquely evolving nature, whose final iteration is intended to be an established court composed of full-time members. In relation to remuneration and impartiality of ICS Members, the Court holds the IBA guidelines CETA incorporates to be sufficient safeguards.78

V. CONCLUSION

The CJEU’s acceptance of this interplay between the norms of arbitration and judicial protection is testament to the novelty of CETA’s context. Both the structure of the Investment 74 ibid

para 200. Statement No 36. 76 CETA, Art 8.27. 77 Opinion 1/17, ECLI:EU:C:2019:341, para 234. 78 ibid para 238. 75 CETA,

The CJEU Saga 527 Court, and the CJEU’s approbation of it, is driven by the EU’s desire to create a tribunal that satisfies the competing demands of inviting investment, and political opposition to the typical ISDS arbitral model. As underscored by Advocate General Bot in his opinion, ‘what is at issue here is the definition of a model which is consistent with the structural principles of the EU legal order and which, at the same time, may be applied in all commercial agreements between the European Union and third States’.79 To approve this model and avoid a situation where the EU’s Investment Court is stymied by its own courts, the principled formalism of the CJEU’s approach in Achmea gives way to a much more pragmatic analysis in the CETA Opinion. But it is not without its compromises. Opinion 1/17 makes some significant sacrifices particularly to the principle of the autonomy of the EU legal order, in order to reconcile the effect of the Investment Court with the regulatory autonomy of the parties. By rubber stamping the Investment Court’s international free trade agreements with the compatibility of EU law, the Court gives expression to the Europeanisation of international trade objectives in the 2009 Lisbon Treaty, an approach first registered in Opinion 2/15. It also provides a ‘sort-of ’ future for the ill easy company that investment arbitration and EU law had had up to this point. But the company has been forever changed as the Investment Court is more heavily shaped by and towards EU principles. The relief with which Opinion 1/17 has been received is therefore perhaps testament to the exhaustion felt, following the CJEU along this winding path to relative resolution.

VI. POST SCRIPT – AN UPDATE ON THE MICULA SAGA

A final case, or more accurately series of litigations, that is worth mentioning in the context of the CJEU saga is that of the Miculas’ case. The background to this saga is lengthy and has involved actions in four EU countries and the US. In essence, that case relates to a dispute over an ICSID arbitral award made against Romania, in the Miculas’ favour, and the issue of illegal state aid under EU rules. In 2013, an ICSID arbitral award, arising out of a dispute under an intra-EU BIT that Romania and Sweden had signed in 2002, granted the Miculas approximately €70 million. The Arbitral Tribunal made its findings on the grounds, inter alia, that the Miculas had a legitimate expectation that a Romanian aid regime (Emergency Government Ordinance No 24/1998 – ‘EGO 24’) applicable to their investments in Romania would have continued. Romania had repealed all but one of the incentives provided for in EGO 24 prior to its accession to the EU on grounds that, inter alia, EU law required EGO 24 be repealed, as the regime amounted to illegal state aid pursuant to EU law rules. In the ICSID arbitral proceedings both Romania and the Commission (as amicus curiae) had raised the state aid issue arguing that if compensation to the Miculas was granted and paid, payment of any such award would amount to illegal state aid and be in breach of Article 107(1) TFEU. The Arbitral Tribunal declined to address the EU law issues raised. But once the award was rendered, these hypothetical EU law issues became an immediate reality setting up a direct clash of the EU and international law legal order. The Commission ultimately injuncted Romania so as to prevent any payment being made under the award, and made a final decision that the payment of the award by Romania constituted state aid under Article 107(1) TFEU and was incompatible

79 Opinion

of AG Bot, Opinion 1/17, ECLI:EU:C:2019:72, para 86.

528 Gillian Cahill and Daniel Sarmiento with the internal market. The Commission accordingly prohibited Romania from making any payment of such state aid to the Claimants and demanded that Romania recover any payments already made under the award. Romania was therefore left between a rock and a hard place: it could not comply with the Arbitral Tribunal’s award without breaching its EU law obligations. And at an international level by not paying the award, Romania was arguably in breach of its obligations under the ICSID Convention. In return, the Miculas sought to annul the Commission’s final decision and on 18 June 2019, the General Court of the Court of Justice found in favour of the Miculas on the ground that the Commission had purported to apply its powers retroactively to events pre-dating Romania’s accession to the EU.80 The Commission has applied to appeal this decision and this appeal is, at the time of writing, pending before the CJEU.81 Meanwhile, the Miculas sought to have the award enforced in various other countries, including the US, the UK and Belgium. On 20 February 2020, the UK Supreme Court lifted the stay on enforcement of the Micula award. Once the Micula award takes the next step and has been translated into a judgment pursuant to the English Arbitration Act 1996, a full range of options will be available to the Miculas to try and extract payment from the award (including third party debt orders, charging orders and freezing injunctions). In the meantime, the Commission’s appeal against the GCEU’s order will make its way through the CJEU’s appeal system leaving open the possibility for a further clash of the international and EU legal orders. The question that remains is what legal recourse there may be if the CJEU sides with the Commission and overturns the GCEU’s decision, but in the meantime the award has been enforced in the UK and payment made? In those hypothetical circumstances what real effect will the CJEU’s decision now have? And could the Commission still consider Romania in breach of its EU law state aid obligations? This case leaves open the possibility for a further twist in the investment arbitration and EU law tale.

80 Micula

& Ors v Commission, Cases T-624/15, T-694/15 and T-704/15; ECLI:EU:T:2019:423. v Micula & Ors, Case, 683/19P.

81 Commission

30 The Legacy of the SADC Tribunal in International Investment Law HENOK ASMELASH*

I. INTRODUCTION

T

HE NOW DEFUNCT Tribunal of the Southern African Development Community (SADC) was called upon a few times to decide cases related to the protection of (foreign) property rights during its short lifespan of less than five years. The most controversial of these cases was Mike Campbell (Pvt) Limited and others versus the Republic of Zimbabwe (the Campbell case), which led to the eventual suspension of the Tribunal. At the heart of the case was the legality of a postcolonial land reform program and the classic tension in international investment law between the promotion of public interest and the protection of individual property rights. The Tribunal stirred controversy by condemning the land reform program without due regard to its underlying public policy rationale. Much of the scholarly writing on the Campbell saga has focused on Zimbabwe’s hostile reaction and the impact of the Tribunal’s demise on the protection of human rights in the region.1 Relatively little attention has been given to its implications for the SADC investment protection regime. This chapter sets out to examine the investment jurisprudence and legacy of the Tribunal focusing on the Campbell decision. This landmark decision is largely similar to the decisions of the two arbitral tribunals established under the auspices of the International Centre for Settlement of Investment Disputes (ICSID) to try essentially the same subject matter.2 All three tribunals ruled against Zimbabwe’s expropriation of white-owned commercial farms using different legal bases. While the SADC Tribunal drew upon international human rights treaties and instruments, the ICSID

* Henok Asmelash is a Lecturer at the Birmingham Law School, University of Birmingham. 1 KJ Alter, JT Gathii and LR Helfer, ‘Backlash Against International Courts in West, East and Southern Africa: Causes and Consequences’ (2015) 27 European Journal of International Law 293; O Jonas, ‘Neutering the SADC Tribunal by Blocking Individuals’ Access to the Tribunal’ (2013) 2 International Human Rights Law Review 294; E de Wet, ‘The Rise and Fall of the Tribunal of the Southern African Development Community: Implications for Dispute Settlement in Southern Africa’ (2013) 28 ICSID Review 45; W Scholtz and G Ferreira, ‘Much Ado About Nothing? The SADC Tribunal’s Quest for the Rule of Law Pursuant to Regional Integration’ (2011) 71 Heidelberg International Law Journal 331; ET Achiume, ‘The SADC Tribunal: Sociopolitical Dissonance and the Authority of International Courts’ in KJ Alter, LR Helfer and MR Madsen (eds), International Court Authority (Oxford, Oxford University Press, 2018). 2 See Bernardus Henricus Funnekotter and Others v Republic of Zimbabwe, Award (22 April 2009) ICSID Case No ARB/05/6, 52; Bernhard von Pezold and Others v Republic of Zimbabwe, ICSID Case No ARB/10/15.

530 Henok Asmelash tribunals relied on Bilateral Investment Treaties (BITs). The divergent legal bases provide an interesting context for reflection on the convergence of international human rights and investment law on the protection of property rights. The interplay between these two international law regimes has received increasing attention in academic literature over the last few years.3 This chapter contributes to this emerging literature by analysing the investment case law of a de facto human rights court. Section II of this chapter explores the establishment, composition and competence of the SADC Tribunal in light of Zimbabwe’s challenge to its human rights jurisdiction. Section 3 provides a brief overview of the historical context underpinning the land reform program. The purpose here is not to provide a detailed historical account but to place the program in its proper historical context. Section IV offers a detailed overview and analysis of the case law with a particular focus on whether and, if so, how the tribunals took into account the underlying public policy objectives of the land reform program. Section V examines the disbandment of the Tribunal and the post-Campbell developments in the SADC investment protection regime. Section VI concludes this chapter.

II. ESTABLISHMENT AND MANDATE OF THE SADC TRIBUNAL

The SADC Treaty established the Tribunal as the judicial organs of the SADC in 1992.4 The Treaty left important matters governing the Tribunal such as composition, functions and procedures for the Summit of SADC Heads of State or Government (the Summit) to prescribe in a subsequent protocol.5 However it took the Summit eight years to adopt the Protocol on the Tribunal (the Protocol) and another five years to operationalise the Tribunal. The legal grounds upon which the Tribunal was founded (Article 16 of the SADC Treaty and the Protocol) were also subjected to a series of amendments even before the Tribunal saw the light of the day.6 Having weathered these early hiccups, the Tribunal was officially inaugurated in November 2005 and finally heard its first case in September 2007.7 The Campbell case was the second and most popular of the 18 cases that made it to the Tribunal’s docket. The Tribunal had an extremely broad personal and subject matter jurisdiction. The personal jurisdiction of the Tribunal extended to disputes between: Member States, Member States and natural or legal persons, natural or legal persons and the SADC, Member States and the SADC, and SADC employees and the SADC.8 What is interesting here is that access to the Tribunal was not limited to SADC nationals or residents as in other regional courts and tribunals; both domestic and foreign investors had access to the Tribunal. We will see later in this chapter how the suspension of the Tribunal led to an unconventional international

3 See P-M Dupuy, E-U Petersmann and F Francioni (eds), Human Rights in International Investment Law and Arbitration (Oxford, Oxford University Press, 2009); M Krajewski, ‘Human Rights in International Investment Law: Recent Trends in Arbitration and Treaty-Making Practice’ in L Sachs, L Johnson and J Coleman (eds), Yearbook on International Investment Law & Policy 2017 (Oxford, Oxford University Press, 2019); T Leary, ‘Non-Disputing Parties and Human Rights in Investor-State Arbitration’ (2017) 18 Journal of World Investment & Trade 1062. 4 Arts 9(1) and 16, Treaty of the Southern African Development Community (signed 17 August 1992, entered into force 5 October 1993) 32 International Legal Materials 120. 5 ibid, Art 16(2). 6 The ambiguities in these amendments later led to questions as to whether the Tribunal was legally established. See HB Asmelash, ‘Southern African Development Community Tribunal’ in H Ruiz Fabri (ed), Max Planck Encyclopedia of International Procedural Law (Oxford, Oxford University Press, 2019), paras 43–48. 7 Ernest Francis Mtingwi v SADC Secretariat, Judgment (27 May 2008) SADC Case No 01/2007, 16. 8 ibid, Arts 15, 17–19.

The Legacy of the SADC Tribunal in International Investment Law 531 investment claim in Swissbourgh and others v Lesotho.9 The Swissbourgh claimants brought an investor-state dispute settlement (ISDS) claim against Lesotho for its alleged contribution to, or facilitation of, the ‘shuttering’ of the SADC Tribunal. The broad subject matter jurisdiction allowed the Tribunal to wear different hats for different cases. It adopted the role of a human rights court in cases such as Campbell and Tembani and that of an investment tribunal in the Bach’s Transport and Swissbourgh cases. The case that raised controversy and eventually brought about the Tribunal’s demise was, however, its assertion of a human rights jurisdiction in Campbel. The subject matter jurisdiction of the Tribunal is defined in Article 16(1) of the SADC Treaty and Article 14 of the Protocol. These provisions made it clear that the primary function of the Tribunal was to ensure adherence to the interpretation and application of the SADC Treaty and its implementing protocols. Numerous such protocols were adopted over the years on issues ranging from energy and corruption to trade and tourism, but not on human rights or agrarian reform.10 A protocol on finance and investment was adopted but had not yet entered into force during the Campbell proceedings.11 The absence of a human rights specific protocol coupled with the fact that SADC is primarily an ‘economic regional block’12 established to promote regional economic cooperation raised questions as to whether the Tribunal had human rights jurisdiction to hear the claims in that case. It is important to note here that the Tribunal had little to do with the framing of the claims in human rights language.13 In the absence of a human rights or investment protocol, the principles and objectives of the SADC set out in Articles 4 and 6 of the SADC Treaty were the only legal hook that the claimants could find to challenge the legality of the land reform program. While Article 6(2) enjoined SADC Member States not to discriminate against any person on grounds such as race and ethnic origin, Article 4(c) required the SADC and its Member States to ‘act in accordance with … human rights, democracy and the rule of law’. Zimbabwe argued that these were very general principles that required detailed implementing provisions. It insisted that ‘there should first be a protocol on human rights in order to give effect to the principles set out in the Treaty’.14 The Tribunal rejected this argument explaining that Article 4(c) imposed a collective and individual legal obligation upon Member States to respect and protect human rights and to ensure that there was democracy and the rule of law within the region.15 For the Tribunal, the alleged infringement of property rights in the Campbell case was a matter that required the interpretation and application of the SADC Treaty and thus squarely fell within its jurisdiction.16 The Tribunal also invoked Article 21(b) of the Protocol, which mandated it to develop its own community jurisprudence ‘having regard to applicable treaties, general principles and rules of public international law’, to justify its human rights jurisdiction.17 9 See Swissbourgh Diamond Mines (Pty) Limited, Josias Van Zyl, The Josias Van Zyl Family Trust and others v The Kingdom of Lesotho, PCA Case No 2013-29. 10 On past efforts to adopt a SADC human rights instrument, see F Viljoen, ‘The Realisation of Human Rights through Sub-Regional Institutions’ (1999) 7 African Yearbook of International Law Online 185. 11 Protocol on Finance and Investment (adopted 18 August 2006, entered into force 16 April 2010). 12 See the Mtingwi case, above (n 7). 13 For a sharp criticism of the Tribunal’s decision to assume a human rights jurisdiction, see B Chigara, ‘What Should a Re-Constituted Southern African Development Community (SADC) Tribunal Be Mindful of to Succeed?’ (2012) 81 Nordic Journal of International Law 341. 14 Mike Campbell (Pvt) Limited and Others v The Republic of Zimbabwe, Judgment (28 November 2008) SADC Case No 2/2007. 15 Mike Campbell (Pvt) Limited and Others v The Republic of Zimbabwe, Ruling (13 December 2007) SADC Case No 2/07. 16 ibid. 17 The Campbell case, above (n 14) 24.

532 Henok Asmelash This remarkably broad authority also made it easier for the Tribunal to extensively rely on international human rights treaties and instruments to find answers where the Treaty was silent on the content and scope of the general principles in question. SADC Member States may not have intended to create a human rights court, but the legal instruments they adopted undoubtedly allowed the Tribunal to act as one.18 The important question is therefore not whether the Tribunal had a human rights jurisdiction, but whether it properly applied the human rights treaties in question.

III. BACKGROUND: ZIMBABWE’S LAND REFORM PROGRAM

None of the three tribunals mentioned above paid much attention to the historical context underpinning the land reform program, but the past and the present are inextricably linked when it comes to postcolonial land reform programs.19 This section starts with a brief sketch of the historical context to provide a better perspective and complete understanding of the land reform program. The roots of the program were to be found in the unjust land distribution system of the colonial era carried out ruthlessly by law and force.20 The colonial regime enacted a series of legislation to facilitate and consolidate the forced eviction of blacks from their ancestral land with no compensation or due process. The 1930 Land Apportionment Act introduced the first racial allocation of land and divided the country into four sections with separate white and black areas. The Land Tenure Act of 1969 replaced and strengthened the 1930 Act by unfairly dividing the country into white and black areas with 45 million acres of land each.21 These laws allocated the most fertile and economically important land to the white minority and left the black majority with arid land.22 This unjust colonial system of land distribution created extreme inequality that persists to this day. It also sowed the seeds of the liberation struggle,23 which culminated in the independence of the country in 1980. However, independence alone proved inadequate to redress the unjust land ownership pattern inherited from the colonial era. The independence Constitution agreed at Lancaster House in 1979 overlooked the land imbalances and introduced a ‘strong and robust legal framework for the protection of property rights’.24 The Constitution in effect enjoined the postcolonial government to carry out land redistribution via the ‘willing buyer, willing seller’ approach for the first ten years of independence.25 It limited compulsory acquisition only to ‘underutilised’ land at full market 18 On the various human rights-related provisions of the SADC legal instruments, see OC Ruppel, ‘The Southern African Development Community (SADC) and Its Tribunal: Reflexions on a Regional Economic Communities’ Potential Impact on Human Rights Protection’ (2009) 42 Verfassung und Recht in Übersee 173. 19 Sir Shridath Ramphal captured this well when he noted that ‘it was about land in the beginning; it was about land during the struggle; it has remained about land today. The land issue in Zimbabwe (Rhodesia) is not ancient history’. See S Ramphal, Glimpses of a Global Life (Dundurn, Dundurn Press, 2014) 388. 20 For a comprehensive account of the unjust land distribution system of the colonial era, see International Committee of Jurists, ‘Racial Discrimination and Repression in Southern Rhodesia’ (1976) ICJ Report. 21 ibid. 22 ibid 11 (noting that the objective of these colonial legislations was to ‘strengthen white dominion over the most fertile and economically important land and maintain the African population as a labouring class’). 23 G Linington and Legal Resources Foundation (Zimbabwe), Constitutional Law of Zimbabwe (Legal Resources Foundation 2001) (noting that the liberation struggle was ‘rooted in the issues of land redistribution’). 24 See AT Magaisa, ‘The Land Question in Zimbabwe: The Judiciary as an Instrument of Recovery?’ in B Chigara (ed), Southern African Development Community Land Issues: Towards a New Sustainable Land Relations Policy (Oxfordshire, Routledge, 2013) (arguing that in doing so the Constitution came to ‘acknowledge, legalise and attempt to legitimise white hegemony whose moral legitimacy remained intensely contested’). 25 The moratorium was not specific to the provisions on property rights but part of a general commitment not to alter the constitution for the first 10 years agreed at the Lancaster House Conference in 1979.

The Legacy of the SADC Tribunal in International Investment Law 533 price compensation (in foreign currency). Immediately after the expiry of the ten-year de facto moratorium in 1990, the government introduced amendments to the property rights provisions of the Constitution. The 1990 amendment removed the constitutional limitation on the type of land that could be compulsorily acquired.26 A subsequent amendment further strengthened the government’s hand by replacing the constitutional requirement of ‘promptly adequate compensation’ with fair payment within a reasonable time. It also prohibited judicial challenges against the fairness of compensation.27 These two constitutional amendments were accompanied by the enactment of the Land Acquisition Act of 1992, which empowered the government to compulsorily acquire land for resettlement and other public purposes subject to the designation of the land for such purposes, preliminary notice, and the payment of fair compensation within a reasonable time.28 However, the protracted compulsory acquisition procedure and subsequent litigations undermined the ability of the Act to expedite the land reform process.29 The government therefore remained reliant on the voluntary ‘willing buyer, willing seller’ principle. However, the deteriorating economic conditions in the country and the disagreement with the UK over the responsibility to pay compensation led to the eventual failure of the ‘willing buyer, willing seller’ land reform program.30 While there was an initial understanding for the UK to provide the necessary funds, the latter unilaterally changed its position in 1997 and stopped funding the program.31 This infuriated Robert Mugabe and his ruling ZANU-PF party and marked the end of the market-based voluntary land redistribution program in Zimbabwe. The limited progress that had been achieved almost two decades after independence caused much concern and frustration particularly among veterans of the liberation struggle. Faced with mounting political pressure and in an attempt to gain political advantage over the opposition, the government placed the land question at the forefront of its 2000 election platform.32 It called for a referendum on a draft constitution that would authorise compulsory land acquisition for resettlement in February 2000. Having lost the referendum but won the parliamentary election by a narrow victory (ahead of the presidential election in 2002), the Mugabe Government embarked on its flagship fast track land reform program (FTLRP) in July 2000. With a radical move to settle the land question ‘once and for all’, the government passed Constitutional Amendment Act No 16 in 2000. This ruled out the payment of compensation for the compulsory acquisition of agricultural land except for improvements effected on such land before it was acquired.33 Amendment No 16 was further strengthened by Amendment No 17, which removed the right to challenge the compulsory acquisition of agricultural land and criminalised the continued possession or occupation of land expropriated for resettlement purposes.34 The amendment effectively ousted the jurisdiction of

26 Constitution

of Zimbabwe Amendment (No 11) Act 1990, s 16(1)(a)(i). of Zimbabwe Amendment (No 12) Act 1993, ss 16(1)(c) and 16(2)(b). 28 Land Acquisition (No 3) Act 1992, ss 5 and 16. 29 ibid. The Act contained rather complex notice and compensation assessment procedures. 30 On the failure of ‘willing-buyer, willing seller’, see S Moyo, ‘A Failed Land Reform Strategy in Zimbabwe. The Willing Buyer Willing Seller’ (2014) 2 Public Policy and Administration Review 67. 31 ibid. 32 ZANU-PF used the slogan ‘Land is the economy and the economy is land’ for the 2000 election. See A Hellum and B Derman, ‘Land Reform and Human Rights in Contemporary Zimbabwe: Balancing Individual and Social Justice Through an Integrated Human Rights Framework’ (2004) 32 World Development 1785. 33 Constitution of Zimbabwe Amendment (No 16) Act 2000, Art 16B(2)(c). Section 16(B)(1) of the Act shifted the burden of compensating landowners for their compulsorily acquired land to the former colonial power (ie UK) and absolved the Zimbabwean Government from paying any compensation for land. 34 Constitution of Zimbabwe Amendment (No 17) Act 2005, s 16B(3)(a)&(b). 27 Constitution

534 Henok Asmelash domestic courts by stating that ‘a person having any right or interest in the land shall not apply to a court to challenge the acquisition of the land by the state, and no court shall entertain any such challenge’.35 It also retrospectively legalised previous illegal acquisitions of agricultural land by the government. Having sidestepped the legal and financial constraints, the government went on to vigorously implement the FTLRP. However, the process led to violent invasions of white-owned commercial farms, which the government tolerated, if not encouraged. The process was also plagued by corruption and ended up unduly benefiting war veterans and political adherents of ZANU-PF. It was within this complex historical context that the tribunals were asked to determine the legality or otherwise of the fast track land reform program. There are two contradictions at play here. The first is between the policy objective of the program and its implementation. On the one hand, the need for postcolonial land reform was unassailable. Land expropriation and redistribution being a prima facie first order priority in any policy to redress the colonial legacy of inequitable land distribution. On the other hand, the implementation of the program seriously undermined its legitimate public policy rationale to the extent that the expropriated land was not properly redistributed to the landless and historically marginalised community. It was also enforced through violence that threatened the life and safety of the white commercial farmers. The inherent and profound contradiction was between the need to right historical wrongs and to protect individual property rights (over commercial farms acquired at the behest of colonial and apartheid authorities).

IV. THE LAND REFORM PROGRAM ON TRIAL: CAMPBELL, FUNNEKOTTER AND VON PEZOLD

These three cases were brought by different claimants and adjudicated based on different bodies of law but arose from substantially the same factual circumstances. All the claimants were white commercial farmers whose farms were expropriated by the Zimbabwean authorities, without compensation, pursuant to Amendment 17. Some of the claimants sustained physical injury and were forcefully evicted from their farms, while others were still in full or partial possession of their expropriated farms during the proceedings. In the Campbell case, for example, only three of the 79 claimants had been evicted from their land when the SADC Tribunal issued its final judgment on 28 November 2008. This section presents and examines the arguments of the parties and the findings of the SADC Tribunal in the Campbell case in comparison with the findings of the two ICSID tribunals (the Von Pezold and Funnekotter tribunals). On 11 October 2007, Mike Campbell (Pvt) Limited and William Michael Campbell filed an application with the Tribunal challenging the compulsory acquisition of their commercial farms in central Zimbabwe.36 Before approaching the Tribunal, they challenged the constitutionality of Amendment 17 at the Zimbabwean High Court and then at the Supreme Court on appeal.37 As the Supreme Court was yet to deliver its judgment, Zimbabwe argued that the claimants had not exhausted local remedies as per Article 15(2) of the Protocol. Although the Supreme Court dismissed the case for lack of jurisdiction in January 2008, the Tribunal went on to explain the prevalence of the exhaustion of local remedies requirement in international 35 ibid

s 16(3)(a). Campbell case, above (n 14), claimants were later joined by 77 other white commercial farmers. Court of Zimbabwe, Mike Campbell (Pty) Ltd v Minister of National Security Responsible for Land, Land Reform and Resettlement, Judgment (22 January 2008) Case No SC 49/07. 36 The

37 Supreme

The Legacy of the SADC Tribunal in International Investment Law 535 human rights law and how the rationale behind this requirement was not only to enable the well placed local courts to deal with the matter first but also to ensure that ‘the international tribunal does not deal with cases which could easily have been disposed of by national courts’.38 However having noted that Amendment 17 ousted the jurisdiction of Zimbabwean courts to rule on the lawfulness of the land reform program, the Tribunal concluded that the claimants had no local remedies to exhaust.39 In contrast to the numerous international human rights treaties cited by the Tribunal, neither the ICSID Convention nor the applicable BITs in Von Pezold and Funnekotter required the prior exhaustion of local remedies. The requirement to exhaust local remedies was one area where the two regimes tended to differ. In international investment law host states often agree in advance not to request exhaustion in return for home states agreeing not to grant diplomatic protection to their nationals.40 There are only a few BITs that require the exhaustion of local remedies.41 In international human rights law, on the other hand, the requirement to exhaust local remedies is the rule rather than exception.42 The absence of the exhaustion requirement in the respective BITs allowed the Von Pezold and Funnekotter claimants43 to bypass the Zimbabwean court system – a right unavailable to the Campbell claimants. Having established its jurisdiction, the Tribunal then proceeded to consider the substantive questions as to whether Zimbabwe had breached its obligations under the SADC Treaty: (i) to act in accordance with the principles of human rights, democracy and the rule of law (Article 4(c)); and (ii) not to discriminate against any person on grounds such as race and ethnic origin (Article 6(2)). On the first issue, the claimants alleged that – by ousting the jurisdiction of the courts to hear challenges to the legality of the compulsory acquisition of agricultural land – Amendment 17 obliterated their right to a fair hearing before an independent and impartial court under Article 4(c). The Tribunal started by noting that it was settled law that the concept of the rule of law embraces ‘twin fundamental rights’: the right of access to the courts and the right to a fair hearing.44 It then quoted numerous international human rights treaties and courts to support the view that these two rights were well-recognised and entrenched in international human rights law.45 The Tribunal construed this to mean that Article 4(c) obligated Zimbabwe to ensure access to courts and a fair hearing. The Supreme Court had earlier confirmed that Amendment 17 took away the claimants’ right of access to the courts, but Zimbabwe insisted that the claimants could have sought judicial review under Amendment 17. The Tribunal rejected this argument noting that no judicial review was available for the compulsory acquisition of any land designated as ‘agricultural land required for resettlement purposes’ before 14 September 2005 – including that of the claimants.46 38 See

the Campbell case, above (n 14) 20. 22. C Cutler, ‘Human Rights Promotion through Transnational Investment Regimes: An International Political Economy Approach’ (2013) 1 Politics and Governance 16, 20. 41 Even those BITs often contain exit clauses that allow bypassing the exhaustion requirement. See ibid. On the exhaustion requirement in international investment and human rights law, see MD Brauch, ‘Exhaustion of Local Remedies in International Investment Law’ [2017] IISD Best Practices Series. 42 All the major human rights regimes require the exhaustion of local remedies. See ibid. 43 In the Von Pezold case, above (n 2) claimants were of German and Swiss nationality and brought their claims under the 1995 German-Zimbabwe and the 1996 Switzerland – Zimbabwe BITs. Whereas all the Funnekotter case, above (n 2) claimants had Dutch nationality and brought their claims under the 1996 Netherlands-Zimbabwe BIT. 44 The Campbell case, above (n 14) 27. 45 ibid 26–37. 46 Juridical review was available for land acquired under s 16(B)(2)(a)(iii) (designated as agricultural land for resettlement purposes after 14 September 2005), but not for land acquired under s 16(B)(2)(a)(i)&(ii). See Constitution of Zimbabwe Amendment (No 17) Act, 2005; and the Campbell case, above (n 14) 39–40. 39 ibid 40 See

536 Henok Asmelash The Tribunal accordingly unanimously concluded that the claimants had been denied their rights of access to justice in breach of Article 4(c) of the SADC Treaty. Access to justice is an area of convergence between international human rights and investment law. Both domestic and foreign investors enjoy the fundamental right of access to justice under most national constitutions and international human rights treaties. International investment law also guarantees access to (domestic) justice through the fair and equitable treatment standard and the due process condition of lawful expropriation. Perhaps the major difference is that BIT-era international investment law also guarantees direct access to international justice.47 This leads to fewer denial of access to justice claims in international investment arbitration than in international human rights adjudication. Neither the Von Pezold nor the Funnekotter claimants, for example, made a direct denial of access claim unlike their counterparts in the Campbell case. The only instance in which the ouster clause in Amendment 17 was raised in these cases was where the Von Pezold48 tribunal considered, in arguendo, whether the land acquisition met the due process requirement of lawful expropriation having already ruled that the expropriation was unlawful (see below). By far the most controversial issue in the Campbell case concerned the question of whether Amendment 17 discriminated against the claimants on the basis of race. The claimants alleged that the decision regarding whether or not to expropriate agricultural land under Amendment 17 was primarily determined based on consideration of race or country of origin.49 Although Amendment 17 makes no reference to race, they argued, there was a clear legislative intent directed only at white farmers.50 Zimbabwe argued in response that the land reform program affected the claimants not because it was targeted at white farmers but rather because they happened to own most of the land suitable for agriculture. It insisted that ‘such expropriation under the program cannot be attributed to racism but circumstances brought about by colonial history’.51 It also alleged that the land reform program affected not only white farmers but also the few black farmers who possessed large tracts of land. The Tribunal sidestepped the historical context of the land reform program and framed the issue narrowly as whether or not Amendment 17 discriminated against white farmers and as such violated the prohibition against racial discrimination under Article 6(2) of the SADC Treaty.52 It then went on to make an extensive reference to international human rights treaties and instruments to establish that discrimination of whatever nature is outlawed in international law.53 Having noted the absence of an explicit mention of the word ‘race’ in Amendment 17, the Tribunal then turned the question into one of whether or not the land reform program was de facto discriminatory. The issue was no longer whether Amendment 17 in itself was discriminatory but whether its effect made it racially discriminatory. The Tribunal agreed with the claimants that even if Amendment 17 did not explicitly refer to race, the effect of its implementation had an ‘unjustifiable and disproportionate’ impact on white farmers.54 It, therefore, concluded by a majority of four to one that Zimbabwe indirectly discriminated against the claimants on

47 F Francioni, ‘Access to Justice, Denial of Justice and International Investment Law’ (2009) 20 European Journal of International Law 729. 48 The Von Pezold case, above (n 2) paras 499–500. 49 Campbell, above (n 14) 42. 50 ibid 43. 51 ibid 44. 52 ibid 45. 53 ibid 45–48. 54 ibid 53.

The Legacy of the SADC Tribunal in International Investment Law 537 the basis of race and thereby violated its obligation under Article 6(2) of the SADC Treaty.55 Justice Onkemetse Tshosa wrote a brief dissenting opinion in which he argued that the land reform program affected white farmers not because they were of white origin but because the agricultural land required for resettlement purposes was in their hands.56 He was of the view that since Amendment 17 targeted agricultural land (not people of a particular racial group) its implementation was always going to affect those in possession of the land in question be they of white, black or other racial background.57 He argued that the mere fact that the effects of the amendment were mostly felt by white farmers was therefore insufficient to conclude that the amendment discriminated against white farmers directly or indirectly. He also criticised the majority for its unsubstantiated assumption that there were no black farmers whose land was compulsorily acquired under Amendment 17.58 It is noteworthy that the claimants did not challenge Zimbabwe’s assertion that black-owned farms were also compulsorily acquired in either the written or oral proceedings. The latest amendment to the Constitution at that time also confirmed this by providing a compensation scheme for black farmers whose land was compulsorily acquired for resettlement purposes under Amendment 17. The real question here was not in fact whether the land reform program was discriminatory. Given the unjust pattern of land ownership in Zimbabwe, it was inevitable that any land reform program in that country disproportionately affected white farmers. The real question was whether the underlying public policy rationale justified the indirect discrimination. The Tribunal’s brief and simplistic answer to this question is the most troublesome part of the decision. It first noted that the rationale ‘might be legitimate if and when all lands under the program were indeed distributed to poor, landless and other disadvantaged and marginalized individuals or groups’.59 It then went onto state that the land reform program would not have constituted racial discrimination if: (i) it was based on reasonable and objective criteria; (ii) fair compensation was paid; (iii) and the expropriated land was distributed to poor, landless and other disadvantaged and marginalised individuals and groups.60 The Tribunal made no effort whatsoever to apply these criteria to the land reform program – it simply quoted an earlier decision of the Zimbabwean Supreme Court in which the latter found that the expropriation under Amendment 16 was unfair discrimination because it ‘awarded the spoils of the expropriation primarily to ruling party adherents’.61 This is however self-contradictory given that the Tribunal had already determined that the land reform program was racially discriminatory (albeit indirectly) because of its disproportionate impact on white farmers. The Tribunal was not saying the indirect discrimination was justifiable under these three conditions, but rather that it might not even be discriminatory at all had it met these three conditions. These are separate issues. The existence or otherwise of indirect racial discrimination is entirely different from the justifiability or otherwise of the discrimination. The Tribunal erroneously conflated the existence of indirect discrimination with its policy justification. If these three conditions were definitional requirements, then they should have been applied to determine whether the land reform program was racially 55 ibid. 56 Mike Campbell (Pvt) Limited and Others v The Republic of Zimbabwe, Dissenting Opinion of Justice Onkemetse B Tshosa (28 November 2008) SADC Case No 2/2007. 57 ibid. 58 ibid. 59 The Campbell case, above (n 14) 53. 60 ibid. 61 ibid 54–55.

538 Henok Asmelash discriminatory rather than determining the justifiability of the already determined indirect racial discrimination. Part of the problem stems from the fact that this was a last-minute decision. The dissenting judge revealed that it was ‘only a day before the judgment was to be delivered that the majority were inclined to hold that Amendment 17 indirectly discriminated against the claimants’.62 Perhaps this was also the reason why a tribunal that cited countless international human rights treaties and courts at every opportunity offered no legal basis whatsoever for its criteria to determine the most sensitive issue in the case. The reference to the Zimbabwean Supreme Court was simply an attempt to protect itself from the inevitable backlash over its narrow and simplistic approach to determining the justifiability of the discrimination. In fact, the Tribunal tied its own hands from the outset by ruling out the relevance of the historical context.63 It is difficult to fully understand the aim of the land reform program without its proper historical context. The Tribunal also failed to distinguish between the purpose and implementation of the program. Its failure to do so meant that it relied solely upon the implementation problem (ie the redistribution of expropriated land primarily to war veterans) to erroneously conclude that the purpose of the land reform program was not legitimate. Recognising the legitimate public policy rationale of the program and condemning its poor implementation are not mutually exclusive and could have better protected the Tribunal from the ensuing backlash. Racial discrimination and policy justification were also at issue in Von Pezzold and Funnekotter in two separate instances. The first concerns the expropriation claim. The claimants in both cases alleged that Zimbabwe expropriated their farms without compensation in violation of several provisions of the relevant BITs. The three BITs in question prohibit expropriation unless it was carried out for public purpose, in a non-discriminatory manner, with due process and accompanied by compensation (full compensation in Funnekotter and ‘prompt, adequate and effective compensation’ in Von Pezold).64 Both tribunals held that these requirements were cumulative and hence all four requirements had to be met for the expropriation to be considered lawful.65 Having then noted that no compensation was paid, both tribunals quickly concluded that the expropriation of land under Amendment 17 was unlawful. The non-payment of compensation meant that the tribunals found no need to consider whether the expropriation met the non-discrimination and public purpose requirements of lawful expropriation.66 The Von Pezold tribunal nevertheless briefly considered whether the land reform program met the other requirements of lawful expropriation simply because ‘the parties pleaded so extensively on these matters’.67 It took only a single paragraph for the tribunal to conclude that the land reform program was racially discriminatory.68 Although it acknowledged that a small number of black-owned farms were also expropriated, it relied on information obtained from witness cross examination to conclude that the general policy was not to expropriate black-owned farms. The tribunal did not even attempt to distinguish between direct and indirect discrimination. Its consideration of whether the land reform program was carried out for public purpose is equally brief and narrow. In another single paragraph, it

62 The

Campbell Case, Dissenting Opinion, above (n 56). Tribunal held early on that it took note of the long history of land acquisition in Zimbabwe but would confine itself only to acquisitions carried out under Amendment 17. See the Campbell case, above (n 14) 8. 64 Art 6(1) of the Switzerland-Zimbabwe BIT and Art 4(2) of the German-Zimbabwe BIT; and Art 6 of the Netherlands-Zimbabwe BIT. See the Von Pezold case, above (n 2). 65 The Funnekotter case, above (n 2) para 98; the Von Pezold case, above (n 2) para 496. 66 ibid para 498. 67 ibid para 499. 68 ibid para 501. 63 The

The Legacy of the SADC Tribunal in International Investment Law 539 dismissed Zimbabwe’s argument that the land reform program was necessary to right historical wrongs simply because ‘there appears to be a clear trend towards politically-motivated allocations of land’.69 Here the tribunal followed the line taken by the SADC Tribunal in the Campbell case (albeit implicitly) and conflated the policy purpose and implementation of the land reform program. The second instance in which the ICSID tribunals considered discrimination and the public purpose behind the land reform program was in relation to Zimbabwe’s ‘state of necessity’ defence under Article 7 of the Netherlands–Zimbabwe BIT (in Funnekotter) and Article 25 of the ILC Draft Articles on State Responsibility (in Von Pezold). The Funnekotter tribunal quickly rejected this defence saying Zimbabwe had not explained why the state of necessity prevented it from paying the compensation due to the claimants.70 In contrast, the Von Pezold tribunal discussed the necessity defence at length.71 To successfully invoke the necessity defence under Article 25 of the Draft Articles, Zimbabwe needed to establish that the land reform program was the only way to safeguard an ‘essential interest against a grave and imminent peril’ and it ‘does not seriously impair an essential interest of the State or States towards which the obligation exists, or of the international community as a whole’.72 Zimbabwe’s argument here was that the violent land invasions of the early 2000s (the land occupation movement) presented a ‘grave danger to the existence of the State itself, to its political and economic survival’ and the implementation of Amendment 17 was the only way to safeguard this essential interest.73 The tribunal dismissed this argument on several grounds: it found that the land occupation movement threatened the existence of the incumbent political party but not that of the state itself, that the fast track land reform program was not the only option available to address this alleged threat, and that the program impaired an essential interest of the international community. The tribunal considered whether Amendment 17 was racially discriminatory and, if so, whether racial discrimination was justifiable in the context of this last finding. It took the view that determining whether Amendment 17 impaired the essential interest of the international community required determining whether its implementation constituted ‘racial discrimination’, which it referred to as ‘undisputedly obligations erga omnes’.74 Having quickly concluded that Amendment 17 ‘undoubtedly distinguished between persons based on race’ and was ‘prima facie racially discriminatory’,75 the tribunal then turned to the question of whether this racial discrimination was justifiable. Here it acknowledged that racial discrimination could be justified under certain circumstances such as affirmative actions to redress historical inequalities. The tribunal was at pains to stress that it was ‘not unsympathetic to national aspirations to correct colonial wrongs’.76 It also held that ‘the discrimination in favour of indigenous Zimbabweans was necessary in order to remedy the unconscionable anti-aboriginal policies implemented throughout the Rhodesian era’.77 It however found that Amendment 17 was not among the ‘legitimate policies that justifiably discriminate by race in order to address

69 The

Campbell case, above (n 14) para 502. Funnekotter case, above (n 2) para 102. 71 The Von Pezold case, above (n 2) paras 610–668. 72 ILC (53rd Session), ‘Draft Articles on Responsibility of States for Internationally Wrongful Acts, with Commentaries’ (2001) 2(2) Yearbook of the International Law Commission Art 25(1). 73 The Von Pezold case, above (n 2) para 610. 74 ibid paras 648. 75 ibid paras 651. 76 ibid paras 652. 77 ibid. 70 The

540 Henok Asmelash historical wrongs’.78 The Tribunal drew a distinction between the pre- and post-2000 land reform programs to explain that the former was ‘adequately founded and justifiable’ while the latter was ‘enacted in response to political pressures’, rather than an underlying need to remedy historical land imbalances.79 The fact that the sluggish process under the market-based willing buyer, willing seller approach of the pre-2000 land reform program led to frustration and the land occupation movement of the early 2000s was an irony lost on the tribunal. The SADC Tribunal in Campbell and the Von Pezold tribunal rejected the argument that the land reform program was in the public interest primarily because the expropriated land was not redistributed to the landless and historically marginalised population. Their ultimate findings seem to suggest that the land reform program would have been in the public interest if the expropriated land had in fact been redistributed to the landless population. However, despite recognising the importance of discriminatory expropriation to overcome the persisting legacy of colonial land dispossession, none of the tribunals left any room for Zimbabwe to rectify the implementation issues or ensure that the expropriated land would be redistributed to the people under whose name the expropriation took place. They went on to award remedies that placed the individual property rights of the white commercial farmers ahead of the historically disadvantaged population and thereby perpetuated the historical injustices. This partly highlights the colonial legacy of international law itself and the limitations of investment arbitration to properly take the interests of the public into account. The tribunals awarded substantially similar remedies. In Campbell, Zimbabwe did not dispute the fact that the claimants deserved compensation but argued that the responsibility to pay the compensation rested with its former colonial power. The Tribunal found it ‘difficult’ to ‘understand’ this argument given the clear position in international law that it is the responsibility of the expropriating state to pay compensation.80 It may not have changed the outcome but even a cursory look into the history and provisions of Amendment 17 would at least have helped the Tribunal understand the argument. The Tribunal dodged this noting that Zimbabwe could not in any case rely on Amendment 17 to avoid paying its international obligation to pay compensation for the expropriated farms.81 It ordered Zimbabwe to pay fair compensation (without calculating the exact amount) to only the three claimants who had already been evicted from their farms and ‘to take all appropriate measures’ to ensure that the other 76 claimants were not evicted from their farms.82 In Funnekotter, the claimants requested compensation for their farms plus reparation for the disturbance caused by the expropriation and for moral damages. The tribunal rejected the moral damage claim for late submission but awarded the claimants €7.2 million for the farms and €20,000 for the disturbance of each claimant.83 In contrast, the Von Pezold Tribunal opted for what are rarely used remedies in investment arbitration – restitution and moral damages. It reasoned that restitution is rarely used simply because of ‘pragmatic concerns’.84 It argued that ICSID tribunals have the mandate to award restitution insofar as it is neither materially impossible nor disproportionate.85 The tribunal found that nothing prevented Zimbabwe from returning the

78 ibid

paras 653. paras 656. Campbell case, above (n 14) 56. 81 ibid 57. 82 ibid 59. 83 See the Funnekotter case, above (n 2) paras 132 and 138. 84 The Von Pezold case, above (n 2) para 699. 85 ibid 681–700. 79 ibid

80 The

The Legacy of the SADC Tribunal in International Investment Law 541 farms to the claimants as it simply needed to issue legal title to unlawfully expropriated lands and this was not disproportionate.86 The tribunal also awarded the claimants compensation of around US$64 million for lost value and US$1 million for moral damages. If Zimbabwe failed to effect restitution within 90 days, the total amount of compensation would rise to US$196 million. It mattered little to the tribunal whether and how Zimbabwe would ever be able to redress the historical land imbalances in the country paying this much compensation purely for the Von Pezold claimants. The tribunal even went out of its way to consider the potential impact of its restitution order on third parties – not the landless and historically marginalised population, but those who held minority shares in the companies owned by the Von Pezold claimants.87 This was a further indication of the system of investment arbitration’s lack of public interest orientation at that time.88 Despite expressing concern about the impact of its decision on third-party minority shareholders, the tribunal did not even allow four indigenous Zimbabweans, who were directly and substantially affected by the outcome of its decision, to submit amicus curiae.89 It denied their request to make amicus curie submissions on two grounds. The first was that the rights of indigenous communities under international human rights law was outside the scope of the dispute.90 It also noted that accepting their submission required determining whether they constituted ‘indigenous peoples’ under international human rights law but the tribunal was neither the appropriate arbiter nor had the mandate to make such a decision. The second ground was its reservation as to the independence and neutrality of the indigenous Zimbabweans.91 Given the outcome of the proceeding would have an impact on their interest, it was worried that their participation may ‘unfairly prejudice the claimants’.92 Interestingly, the tribunal also rejected another request for amicus curie submission by the European Centre for Constitutional and Human Rights (ECCHR) on the ground that the latter lacked ‘significant interest in the proceedings’.93 One had too much interest, while the other not enough. This clearly illustrates how the tribunal was bent on not taking into account the interest of indigenous Zimbabweans.

V. POST-CAMPBELL SADC INVESTMENT PROTECTION REGIME

The SADC Tribunal served as a regional forum for the resolution of investment disputes during its short lifespan – in effect if not by design. Its broad personal jurisdiction allowed both domestic and foreign investors to file direct and indirect investment claims against SADC Member States. In Bach’s Transport, for example, the Tribunal awarded compensation to a transportation company from Botswana whose truck and trailer were unlawfully expropriated by the Democratic Republic of Congo (DRC).94 In Cimexpan, the Tribunal found the claims by

86 Here Zimbabwe argued that restitution would cause instability but the tribunal insisted that most of the land was still in the hands of the claimants and lost damage was reparable. ibid paras 725–735. 87 ibid 739–743. 88 See D Schneiderman, Constitutionalizing Economic Globalization: Investment Rules and Democracy’s Promise (Cambridge, Cambridge University Press, 2008); M Sornarajah, Resistance and Change in the International Law on Foreign Investment (Cambridge, Cambridge University Press, 2015). 89 See Bernhard von Pezold and Others v Zimbabwe, ICSID ARB/10/15, and Border Timbers Ltd and Others v Zimbabwe, Procedural Order No 2 (26 June 2012) ICSID ARB/10/25. 90 ibid para 60. 91 ibid para 56. 92 ibid para 62. 93 ibid para 61. 94 Bach’s Transport (PTY) LTD v the Democratic Republic of Congo, Judgment (11 June 2010) SADC Case No 14/2008.

542 Henok Asmelash a Mauritian company and its owner against Tanzania inadmissible for non-exhaustion of local remedies.95 In Swissbourgh, a case involving an expropriation claim brought by foreign investors against Lesotho, it assumed jurisdiction but was suspended before ruling on the merits.96 The number of investment cases filed with the Tribunal indicated the need and potential of the Tribunal to act as a regional investment court. However, the Tribunal did not last long after its controversial decision in Campbell and this potential never materialised. The tussle between the Tribunal and Zimbabwe started early on when the latter refused to comply with the former’s interim orders to refrain from taking any action to evict the Campbell claimants from their farms pending its decision on the merits. The Tribunal repeatedly reported Zimbabwe to the Summit for contempt pursuant to Article 32(5) of the Protocol. However, the latter was reluctant to take action against Zimbabwe due to legal constraints and/or out of political solidarity.97 Once the Tribunal pronounced its final decision on the merits, Zimbabwe launched a concerted campaign to discredit and dismantle the Tribunal. It bore fruit almost immediately when in August 2010 the Summit refused to reappoint or replace judges whose terms of office were about to expire, and suspended the Tribunal from receiving any new case or holding hearings in pending cases until its role, functions an terms of references were reviewed.98 A subsequent Summit decision in August 2011 officially suspended the Tribunal and set in motion the process for the amendment of the Protocol99 resulting in the adoption of a new Protocol on the Tribunal on 18 August 2014. The new Protocol, which was signed by 9 of the 15 SADC Member States but yet to enter into force, limited the jurisdiction of the Tribunal to the interpretation and application of the SADC Treaty and Protocols in inter-state disputes.100 This severe curtailment of personal jurisdiction blocked individual access to the Tribunal and left both investors and victims of human rights violations with no access to regional justice. The legality of the decision to suspend the operation of the Tribunal and the subsequent curtailment of its personal jurisdiction led to litigations before domestic courts in South Africa, Tanzania, the African Commission on Human and Peoples’ Rights (ACHPR) and an investment arbitration. The South African Constitutional Court found the participation of the South African President in the process ‘unconstitutional, unlawful and irrational’ and directed the President to withdraw his signature from the new Protocol.101 The Tanzanian High Court took a similar position and declared the participation of the relevant government

95 The United Republic of Tanzania v Cimexpan (Mauritius) Limited and Others, Ruling (11 June 2010) SADC Case No 01/2009. 96 The Swissbourgh case, above (n 9). 97 The source of the legal constraint stems from Article 10(9) of the SADC Treaty which enjoins the Summit to take decisions by consensus and thereby gives Zimbabwe a veto over any Summit decision. Some scholars disagree with such interpretation noting that the principle of effective interpretation necessitates an implied exception to the consensus rule. However, there is past practice in international organisations whereby the consensus requirement applied to similar circumstances. The GATT applied a consensus rule whereby even the losing contracting party had to agree for a panel report to be adopted. See L Bartels, WTIA Review of the Role, Responsibilities and Terms of Reference of The SADC Tribunal – Final Report (March 2011) 47, www.scribd.com/doc/115660010/WTIA-Review-of-the-RoleResponsibilities-and-Terms-of-Reference-of-the-SADC-Tribunal-Final-Report. 98 ‘Communiqué of the 30th Jubilee Summit of SADC Heads of State and Government’ (Windhoek, Namibia, 16-17 August 2010), para 32. 99 ‘Communiqué of Extraordinary Summit of SADC Heads of State and Government’ (Windhoek, Namibia, 20 May 2011), paras 7–8. 100 Protocol on Tribunal in Southern African Development Community (signed on 18 August 2014, not yet entered into force), Art 33. For a comparison between the 2000 and 2014 Protocol, see Asmelash, above (n 6). 101 Constitutional Court of South Africa, Law Society of South Africa and Others v President of the Republic of South Africa and Others, Order (11 December 2018) Case CCT 67/18.

The Legacy of the SADC Tribunal in International Investment Law 543 authorities in the suspension of the Tribunal ‘inimical to the rule of law’.102 ACPHR rejected the request from two of the Campbell claimants to refer the matter to the African Court on Human and Peoples’ Right primarily on the ground that the African Charter on Human and Peoples’ Rights did not guarantee access to regional justice.103 The Swissbourgh claimants also brought a denial of access to justice claim before an investment arbitration tribunal established under the auspices of the Permanent Court of Arbitration (PCA) pursuant to the ISDS clause of the SADC Protocol on Finance and Investment (FIP). Invoking the fair and equitable treatment clause of the FIP, they contended that Lesotho denied them any alternative means to settle their expropriation claim by contributing to and facilitating the dissolution of the SADC Tribunal. Interestingly, the PCA tribunal seated in Singapore ruled in favour of the Swissbourgh claimants and ordered the establishment of a new tribunal with the same jurisdiction as the SADC Tribunal to hear the original expropriation claim.104 However, this award was subsequently set aside by the Singapore Court of Appeals, which upheld the High Court’s decision that the PCA tribunal did not have jurisdiction (for, among other, non-exhaustion of local remedies).105 SADC Member States reacted to the Swissbourgh saga by amending the FIP to remove its ISDS clause. The amendment limited investors’ access to justice in domestic courts and tribunals.106 Together with the suspension of the SADC Tribunal, the removal of the ISDS clause from the FIP left investors in the SADC region wishing to pursue their claims with only domestic courts and ad hoc ISDS tribunals established under BITs signed between their home States and SADC Member States.

VI. CONCLUSION

The SADC Tribunal may not have been designed as a human rights or investment tribunal but its broad jurisdiction allowed it to hear a number of human rights and investment cases before its suspension in 2010. Its influence and legacy in the area of international investment law is however largely defined by its decision in the Campbell case – a conventional investment dispute couched in human rights language. The highly emotive nature of the land question not only in Zimbabwe but also in much of the SADC region meant that the Campbell decision had significant implications for Southern Africa. The Tribunal’s sweeping condemnation of the land reform program without due regard to its underpinning historical context did not help the Tribunal fend off Zimbabwe’s attack against its legality and legitimacy, which eventually brought about its demise. There was little appetite among the other SADC Member States to rescue the Tribunal partly stemming from what has been described as the ‘sociopolitical dissonance’ of the Campbell decision.107 The way in which Zimbabwe implemented the program undoubtedly undermined the justifiability of the program itself. However, the reluctance of the Tribunal to acknowledge the public policy rationale behind the program ran counter to the widely shared understanding in the region of the importance of race-based policy measures

102 High Court of Tanzania, Tanganyika Law Society Minister of Foreign Affairs and International Cooperation and Attorney General of the United Republic of Tanzania, 4 June 2019. 103 Tembani and Freeth (represented by Tjombe) v Angola and Thirteen Others, Merits (30 April 2014) African Commission on Human and Peoples’ Rights, Banjul, Communication No 409/12. 104 The Swissbourgh case, above (n 9). 105 Court of Appeal of the Republic of Singapore, Swissbourgh Diamond Mines (Pty) and Others v Kingdom of Lesotho, Judgment (27 November 2018) SGCA 81, Civil Appeal No 149 of 2017. 106 Agreement Amending Annex 1 of the Protocol on Finance and Investment (31 August 2016). 107 Achiume, above (n 1). See also Chigara, above (n 13).

544 Henok Asmelash to overcome the colonial legacy of highly unequal wealth distribution. The ICSID tribunal in Von Pezold followed the same line taken by the SADC Tribunal in conflating the policy justification and implementation of the land reform program. Both tribunals reached the same conclusion that the land reform program lacked a legitimate public purpose primarily because the expropriated land was not redistributed to poor, landless, disadvantaged and marginalised indigenous Zimbabweans. While the Von Pezold tribunal was at pains to convey its understanding of the historical context, the SADC tribunal made a last minute, face-saving attempt by stating that the land reform program would not even have constituted racial discrimination had the expropriated land been distributed to the landless poor. Yet both tribunals went on to order the restitution of the expropriated farms and the payment of hefty compensation to the white commercial farmers without any regard for the implications of their decisions for the indigenous Zimbabweans left landless by the unjust land distribution policies of the colonial and apartheid eras. The Von Pezold tribunals even rejected the request of four indigenous communities to submit amicus curie, ironically for having substantial interest in the outcome of the proceedings. The decisions show the limitations of investment adjudication in taking account of the interests and perspectives of indigenous communities as well as the challenges extant international investment and human rights law pose for countries with limited resources to pay compensation at fair market value to carry out postcolonial land reforms to redress historical injustices and inequalities.

31 Accountability Mechanisms and Investment Arbitration: Parallel Dynamics EDOUARD FROMAGEAU*

I

NTERNATIONAL INVESTMENT LAW and development law have often been seen as two distinct fields of international economic law, with little or no interaction between the two. This could be explained by their long ‘history of ignorance and mistrust’.1 The purpose of this chapter is to highlight the dynamics that emerged in the past decade in the case law of dispute settlement mechanisms in development law, more precisely independent accountability mechanisms (IAMs), that have recently begun to develop in investment arbitration. IAMs as an umbrella concept covers dispute settlement bodies, most of which can be seen as quasi-judicial bodies2 that aim to provide recourse for citizens and communities adversely affected by International Financial Institutions (IFIs)-funded projects, particularly in instances when IFIs are alleged to have failed to follow their own social and environmental safeguard policies, guidelines, standards, or procedures.3

In other words, IAMs contribute to holding these institutions accountable by providing those affected with access to remedy and redress. These mechanisms allow public access to a means of redress for populations affected by development projects that have a social or environmental impact. They are also independent of the management of the given IFI they are part of. This independence is ‘fundamental in providing confidence to the public with complaints that their concerns will be considered fairly’.4 The World Bank started to take account of the impact of its investment due to social unrest around the Sardar Sarovar Dam project on the Narmada River in India.5 In order to

* Edouard Fromageau is a Lecturer in International Economic Law at the University of Aberdeen. 1 SW Schill, CJ Tams and R Hofmann, ‘International Investment Law and Development: Friends or Foes?’ in SW Schill, CJ Tams and R Hofmann (eds), International Investment Law and Development: Bridging the Gap (Cheltenham, Edward Elgar, 2015) 3. 2 For more details, see E Fromageau, ‘Quasi-judicial bodies’ in H Ruiz Fabri (ed), Max Planck Encyclopedia of International Procedural Law (Oxford, Oxford University Press, 2020). 3 K Lewis, Citizen-Driven Accountability for Sustainable Development: Giving Affected People a Greater Voice – 20 Years On (Independent Accountability Mechanisms Network, 2012), https://perma.cc/9LFK-PNMT. 4 RE Bissell, ‘Origin and Evolution of International Accountability Mechanism’, in O McIntyre and S Nanwani (eds), The Practice of Independent Accountability Mechanisms: Towards Good Governance in Development Finance (Leiden, Brill Nijhoff 2019) 7. 5 TR Berger, ‘The World Bank’s Independent Review of India’s Sardar Sarovar Projects’ (1993) 9(1) American University International Law Review 33.

546 Edouard Fromageau respond to domestic and international concerns about the project, the Bank commissioned an independent review, the Morse Commission, to assess its role. The 1992 Morse Commission report established that the Bank had clearly violated its own policies and largely disregarded its social and environmental policies. Furthermore, the follow-up report on the Morse Commission report found that the violations identified in the Narmada River case ‘were not an aberration, but a systemic part of the Bank’s culture’.6 In response to these reports, on 1 September 1993, the Bank’s Board of Executive Directors passed two separate resolutions creating the World Bank Inspection Panel (WBIP) to investigate claims, made by two or more affected claimants, of noncompliance with the bank’s policies in the design and implementation of its projects. Since then, most, if not all, of the other IFIs have created their own IAMs taking the WBIP as a model.7 At the time of writing, at least 16 procedures are achieving a mission similar to that of the WBIP at the level of regional8 and sub-regional9 development banks, and at the level of multilateral financial institutions.10 The panel’s mandate extends only to projects undertaken by the International Bank for Reconstruction and Development (IBRD) and the International Development Association (IDA). It does not extend to the activities of the private sector lending arms of the World Bank Group; that is, the International Finance Corporation (IFC) and the Multilateral International Guarantee Agency (MIGA). Complaints regarding IFC and MIGA are dealt with by the Compliance Advisor Ombudsman (CAO).11 Since 1994 the WBIP has rendered decisions in 147 cases.12 The process followed consists of seven steps: 1. It receives a request and registers it, unless it is ineligible; 2. Management of the Bank (persons responsible for the design, appraisal, planning, and implementation of projects) responds to the claim by disputing the alleged violations or propose appropriate measures; 3. The panel reports to the Bank’s Board of Directors on eligibility and recommends either the initiation of an investigation of all or part of the request or the rejection of the request. The Panel’s Operating Procedures establish clearly that the panel may decide not to recommend an investigation even if it confirms that the technical eligibility criteria for an investigation are met, particularly with regard to the seriousness of the harm alleged; 4. The Board decides whether to request an investigation by the panel; 5. If an investigation is initiated, the panel investigates and reports to the Board and Management; 6. Management proposes appropriate measures in response to the investigation findings; 7. The Board makes a final decision on whether to take action, and both the panel’s report and management’s response are made public.13

6 Y Wong, ‘Inspection Panel: World Bank’ in H Ruiz Fabri (ed), Max Planck Encyclopedia of International Procedural Law (Oxford, Oxford University Press, 2019) 2. 7 See L Boisson de Chazournes, ‘Les panels d’inspection’ in SFDI, Droit international et développement (Paris, Pedone, 2015). 8 The AfDB Independent Review Mechanism, the ADB Accountability Mechanism, the AIIB Complaints Handling Mechanism and Integrity Unit, the CAF-GEF Accountability Mechanism, the EBRD Project Complaint Mechanism, the EIB Complaints Mechanism, the IADB Independent Consultation and Investigation Mechanism, and the IsDB Group Integrity Office. 9 The BSTDB Complaint Mechanism, the CABEI Reporting Channel, the CDB Projects Complaint Mechanism, and the ETDB Internal Audit. 10 The IFAD Office of Audit and Oversight, the IIB Complaint Procedure, the FMO Speak Up Policy, and the NIB Office of Chief Compliance Officer. 11 For more details on CAO, see B Arp, ‘Integrating Human Rights into the Work of the World Bank Group: The International Finance Corporation’s Compliance Advisor/Ombudsman’ in C Jiménez Piernas and AM Aronovitz (eds), New Trends in International Economic Law (Zurich, Schulthess, 2018). 12 As of July 2020. The decisions are available at www.inspectionpanel.org/panel-cases. 13 Y Wong and B Mayer, ‘The World Bank’s Inspection Panel: A Tool for Accountability’ in J Wouters and others (eds), The World Bank Legal Review Volume 6: Improving Delivery in Development – The Role of Voice, Social Contract and Accountability (Washington, World Bank, 2015).

Accountability Mechanisms and Investment Arbitration: Parallel Dynamics

547

In this chapter, I aim to identify the dynamics in the WBIP’s case law that may provide some insight into the current crisis of legitimacy faced by investment arbitration. In the first section, I will focus on one of the several causes of this crisis, the inability of arbitral tribunals to properly engage with affected communities, and I will argue that the creation of the panel has served as a means for the bank to respond to its very own crisis of legitimacy and accountability (section I). In the second and third sections of the chapter, I will look at various cases that resonate with the current dynamics of investment arbitration: the impact on local communities of mining projects (section II) (Ecuador: Mining Development, and Mongolia: Mining Infrastructure Investment Support and Mining Infrastructure Support – Additional Financing), and the impact of investment on human rights (section III) (Chad: Petroleum Development and Pipeline Project and Honduras: Land Administration Project). Section IV will conclude.

I. ENGAGING WITH AFFECTED COMMUNITIES: WHAT CAN WE LEARN FROM THE INSPECTION PANEL?

Investment arbitration is generally understood as a mechanism that hardly engages with local communities. Investments can however have a far-reaching impact on the rights and interests of affected communities. As pointed out by Lorenzo Cotula and Mika Schroeder, two main types of community dimension are nowadays present in investment arbitration: 1) where states took action, at least in part or in rhetoric, in response to community concerns, or more generally to the situation, triggering the investor’s arbitration claim; and 2) where states failed to take action they may have been legally obliged to take.14 Although affected communities’ interests are still largely ignored in arbitral awards, a quite recent dynamic has emerged that takes into account to a limited extent the local communities impacted by the investment. In his dissenting opinion in Bear Creek Mining Corp v Republic of Peru (ICSID Case No ARB/14/21), Philippe Sands called for such an inclusive approach when he suggested that a possible explanation for the adverse responses of local communities to Bear Creek’s mining project could be that ‘the investor did not engage the trust of all potentially affected communities, and that even if the investor was on notice of those numerous communities, failed to take the appropriate steps to address the concerns of those communities’.15 In Copper Mesa Mining Corporation v Republic of Ecuador (PCA Case 2012-2), the arbitral tribunal seemed to take the interests of the local communities seriously by devoting a significant part of the award to the involvement of the investor’s local representatives in violent actions towards communities opposing the project.16 Amicus curiae submissions from NGOs representing communities’ interests were also accepted by a few arbitral tribunals, but had a limited impact on the outcomes of these proceedings.17 All in all, investor-state arbitration is still ill-equipped to respond to community concerns, even in cases in which these concerns led to a troubled relationship between the investor and the state. One of the main consequences resulting from this inability is the ‘asymmetric access to international dispute settlement’,18 while private investors can seek remedy at the international level affected communities cannot. 14 L Cotula and M Schroeder, Community Perspectives in Investor-State Arbitration (London, International Institute for Environment and Development, 2017) 14. 15 D Paez-Salgado, ‘Four Key Takeaways of the Decision in Bear Creek Mining Corp v Republic of Peru’ (Kluwer Arbitration Blog, 16 December 2017). The Partial Dissenting Opinion is available at https://perma.cc/F7VL-8YFE. 16 Cotula and Schroeder, above (n 14); Copper Mesa Mining Corp. v Republic of Ecuador, PCA No 2012-2, Award (15 March 2016). 17 ibid. 18 SW Schill and V Djanic, ‘Wherefore Art Thou? Towards a Public Interest-Based Justification of International Investment Law’ (2018) 33 ICSID Review 55.

548 Edouard Fromageau One possible way forward is to reform the investor-state arbitration system. Such calls for systemic reform propose different paths to allow for the better representation of affected communities, such as granting third parties the right to join proceedings as interveners if they have a reasonable claim of potential injury from an existing dispute.19 In the context of the UNCITRAL reform process, several countries, including South Africa and Ecuador, have put forward in their submissions their desire to have an ISDS that includes all affected parties in proceedings.20 The means through which this objective would be achieved have not yet been proposed in clear terms.21 NGOs have also proposed to duplicate commercial arbitration thirdparty intervention rules, such as Article 37.1 of the Netherlands Arbitration Institute Rules.22 What can we learn from the case law of the WBIP in that context? Arguably, investment arbitration and the inspection panel are two distinct procedures that differ in almost all of their features. The purpose of the inspection panel process is not reparation. It aims at determining whether or not the World Bank has violated its own policies and procedures, and whether or not the rights or interests of individuals have been affected. In practical terms, a successful claim means that the Bank would cancel or withdraw its financial support to a given project. The practical impact is of course significant as this cancellation or withdrawal can eventually put an end to a harmful project in environmental or social terms. The impact of the work of the panel is also highly symbolic. When in the Arun III Hydroelectric project,23 its very first case, the panel withdrew the bank’s financial support for the project, it sent several messages: one to the local communities that their voices had been heard, one to the bank that it is accountable for its actions and, finally, one to borrowing states that their investment projects financed by the bank have a social and environmental impact that is under scrutiny. It would however be naive to think that by creating the panel, the Bank acted in a purely philanthropic manner. The creation of the panel has helped to ease pressures on the bank, notably in terms of reputation and legitimacy. The panel itself, as has been demonstrated, has evolved according to various internal and external pressures.24 Without a doubt, the same can also be said when it comes to investment arbitration as a system. In the context of investment impact on third parties, the pressure exercised by affected communities is first directed towards the state. The reaction to those pressures can take different forms. As explained by Zoe Phillips Williams, ‘when faced with this pressure from “third parties” policymakers have two options: respond to domestic pressure to prevent or mitigate negative impacts of an investment, or maintain high levels of investment protection to avoid a dispute’.25 In that regard, accountability mechanisms and investment arbitration are allegories of two opposite ways to respond to

19 P Wieland, ‘Why the amicus curia institution is ill-suited to address indigenous peoples’ rights before investorstate arbitration tribunals: Glamis Gold and the right of intervention’ (2011) 3 Trade, Law and Development 334. 20 ‘It should be noted that the point of this proposal is not for third parties to be included in all arbitral processes. Rather, with the agreement of the tribunal and the parties, and depending on the circumstances, provision should be made to include parties that, aside from having a legitimate interest in a dispute, could also be directly affected by the arbitral award.’ Submission from the Government of Ecuador, A/CN.9/WGIII/WP.175, para 25. 21 A Roberts and T St. John, ‘UNCITRAL and ISDS Reforms: Agenda-Widening and Paradigm-Shifting’ (EJIL Talk, 20 September 2019). 22 https://perma.cc/MFW3-S75C. 23 For an overview of the impact of the project on local communities see L Udall, ‘The Arun III Dam: A Test Case in World Bank Accountability’ [1994] Bulletin of Concerned Asian Scholars 82. 24 BK Sovacool, A Naude Fourie and M Tan-Mullins, ‘Disequilibrium in Development Finance: The Contested Politics of Institutional Accountability and Transparency at the World Bank Inspection Panel’ [2018] Development and Change 1. 25 Z Phillips Williams, ‘How Does Investment Arbitration Affect “Third Parties”’ in Investment Claims (Oxford, Oxford University Press, 2018), https://perma.cc/8NWT-95W2.

Accountability Mechanisms and Investment Arbitration: Parallel Dynamics

549

these pressures. So far, the symbolic message sent by investment tribunals to affected communities is that what matters most in that context is to ease the pressures existing between the investor and the state, and not to respond to theirs. That may be acceptable in cases in which affected communities do not have a role to play in the relationship between the investor and the state, and in which the award rendered by the Tribunal has no impact on these communities. In cases in which local communities are an integral part of the dispute however, this message is not only unacceptable, it also threatens the legitimacy of the Tribunal and of the system as a whole. Reforming the system by allowing affected parties to intervene in the proceedings, instead of merely submitting amicus curiae briefs, could be one of the ways forward. Studying the case law of the inspection panel may help to determine the way in which this organ allows affected parties to participate, and how it mitigates the negative effects of investments. In that regard its case law on mining projects is illustrative.

II. THE CASE LAW OF THE INSPECTION PANEL ON MINING PROJECTS

Investments in extractive industries can particularly affect local communities as they often contribute to environmental degradation and depletion of water resources that are vital for local communities.26 Mining projects can also have direct and indirect adverse health effects on these populations. For example, the Canadian-owned Marlin Mine in Guatemala has been linked not only to various environmental damage, water depletion – as it uses 45,000 litres of water per hour – and forced dispossession, but also to heavy metal poisoning, increased poverty and food insecurity on the Mayan communities.27 In investment law, a considerable number of cases that have an impact on local communities are related to mining projects, with claims either challenging actions of the state taken in response to the mobilisation of an affected community, or challenging the inaction of the state in cases when it is obliged to act. These disputes can involve competing rights and interests where, for example, affected communities challenge before national courts a government agency’s granting of a concession, arguing that consultation processes were inadequate, and the investor brings an ISDS claim to challenge a court injunction that stopped continuation of the project until consultation was complemented.28

The other stream of arbitration relates to disputes in which an investor that has a competing claim over rights to a piece of land with an indigenous community, seeks an award to secure their title.29 The first case of the inspection panel relating to mining projects was received by the panel in December 1999 (Ecuador: Mining Development). It involved a project named the Ecuador Mining Development and Environmental Control Technical Assistance (PRODEMINCA),

26 ibid. 27 A-E Birn, L Shipton and T Schrecker, ‘Canadian mining and ill health in Latin America: a call for action’ (2018) 109(5–6) Canadian Journal of Public Health 786. 28 Third Party Rights in Investor-State Dispute Settlement: Options for Reform, Submission to UNCITRAL Working Group III on ISDS Reform, contributed by Columbia Center on Sustainable Investment (CCSI), International Institute for Environment and Development (IIED), and International Institute for Sustainable Development (IISD), 15 July 2019, 5, https://perma.cc/6SXW-SJTR. 29 ibid.

550 Edouard Fromageau which received a US$14 million loan from the IBRD in 1993.30 The aim of this project was, inter alia, to attract new private mining investment and support the systematic development of increased but environmentally sound mineral production, arrest mining-related environmental degradation and mitigate the damage resulting from the use of primitive and inadequate technology by informal miners.31 Among the different components of this project, the geoinformation subcomponent was dedicated to mapping all the mineral data of specific regions of Ecuador. An Ecuadorian NGO named Defensa y Conservacion Ecologica de Intag (DECOIN), a grassroots environmental organisation, submitted a request to the panel on behalf of the people living in the Intag area, a region of Ecuador marked by a long history of conflict with mining companies.32 In the request for inspection, DECOIN argued that the communities of the Intag valley were likely to suffer material harm as a result of the public release of maps with mineral data, which could potentially attract mining companies and informal miners.33 The NGO identified several violations on the part of the Bank, notably a violation of Operational Directive (OD) 4.01 on Environmental Assessment,34 a violation which came about because the environmental assessment for PRODEMINCA was drawn up without the participation of the communities and NGOs involved and without taking their opinions into account. In its Investigation Report, the panel found that the Bank for the most part complied with the provisions of OD 4.01. However, it also found that the Bank was in violation of certain provisions of OD 4.01 relating to processing, geographical scope, baseline data, and consultation during project preparation. Regarding the consultation requirement, the investigation showed that the Bank organised a series of ‘information meetings’ in the areas concerned before initiating the mapping process, but the panel also found that these meetings did not commence until March 1998, which was five years after the Environment Assessment was completed.35 During the on-site visit, members of the panel were also confronted with the fact that part of the information on which the locals based their opinions was, in fact, misinformation: [S]ome people, for example, had the impression that PRODEMINCA was a mining company. Some did not know that parts of the Intag Valley would not be mapped. Others were “convinced” that PRODEMINCA was producing detailed maps that easily could lead to an ‘invasion’ of illegal and artisanal miners. No one, of course, had seen the maps, nor had the CD-ROMs properly explained to them.36

As a result, the members of the panel were convinced that a program of consultation undertaken shortly after the draft Environmental Assessment had been prepared ‘could have addressed the legitimate needs of potentially affected people for information on the Project.’37 Three years after this decision, the mining company, Ascendant Copper (later renamed Copper Mesa), started its activity in the Intag valley, proving that the fears expressed by DECOIN were legitimate. Due to the company’s failure to get the Ministry of Mines’ approval

30 World

Bank, The Inspection Panel Annual Report 2000–2001 (Washington, World Bank, 2001) 19.

31 ibid. 32 See

the documentary movie Under Rich Earth. 2008, directed by M Rogge.

33 https://perma.cc/5EWQ-AGN9. 34 For

more details on OD 4.01 see P Sands and P Galizzi (eds), Documents in International Environmental Law (Cambridge, Cambridge University Press, 2004) 1097. 35 Investigation Report on Ecuador Mining Development and Environmental Control Technical Assistance Project, Report No 21870, 23 February 2001, para 52. 36 ibid para 56. 37 ibid para 57.

Accountability Mechanisms and Investment Arbitration: Parallel Dynamics

551

for its environmental impact study and its failure to consult the affected local communities, its concession was terminated in 2008. In 2011, Copper Mesa sued Ecuador and started a process that led to the PCA award of March 2016.38 This award is seen by some observers as progress as it recognised that the investor’s violent behaviour towards the local communities had an impact on the unviability of the project, and eventually led to a 30 per cent reduction of the compensation. However, as pointed out by Nicolas M Perrone, ‘the outcome of this approach is that even when foreign investor actions lead to social unrest, the investors are rarely held responsible and might even receive some compensation if the project is later cancelled’.39 Other interesting cases dealt with by the panel are Mongolia: Mining Infrastructure Investment Support (2015) and Mining Infrastructure Support – Additional Financing (2018). These four requests (Case No 102, received in February 2015, and Case Nos 123, 124 and 125 received in April 2018) are related to the Mining Infrastructure Investment Support Project (MINIS), a technical assistance project financed by a US$25 million IBRD loan, which aimed at facilitating infrastructure investments to support mining operations in Mongolia.40 MINIS helped to finance assessment studies for various subprojects, among them the Shuren Hydropower Plant, a dam which would supply the electricity to support the mining sector,41 and the Orkhon Gobi Water Diversion project, a pipeline diverting water to supply dry mines. These projects were to be located respectively on the Selenge and the Orkhon Rivers, two of the main sources of freshwater into Lake Baikal. In the first request filed by community representatives and local civil society organisations in Mongolia and Russia, the requesters argued that these two subprojects might have potential irreversible environmental, social and economic impact on the Selenge River, the surrounding areas, and on Lake Baikal, which is a UNESCO World Heritage Site.42 The requesters also argued that these subprojects were not selected following a proper procedure, with insufficient scientific assessment of risks, and inadequate consultations with communities and civil society groups.43 This was the start of an intense two-year process during which the members of the panel held several meetings with the management of the Bank and with the requesters, and attended conferences in Russia, with the goal of assessing whether or not the Bank’s actions were effective, specifically regarding the quality of the consultation process and the progress and scope of the environmental and social impact assessment. The management of the Bank held several rounds of consultations both in Russia and Mongolia, proceeded to adjustments after the consultations, and created a Panel of Experts and a MINIS Grievance Redress Mechanism that was in charge, inter alia, of identifying the concerns of potentially affected stakeholders.44 The panel fostered inclusiveness by allowing all the parties involved, including the World Heritage Program, to communicate, and helped to recognise ‘the relevance of transboundary issues, and the greater importance given to ensuring meaningful consultations with both Russian and Mongolian stakeholders’.45 Given the progress made by the Bank, the panel eventually

38 Copper

Mesa Mining Corporation v Republic of Ecuador (PCA No 2012-2) Award (15 March 2016).

39 NM Perrone, ‘The ‘Invisible’ Local Communities: Foreign Investor Obligations, Inclusiveness, and the International

Investment Regime’ [2019] AJIL Unbound 19. 40 World Bank, The Inspection Panel Annual Report 2016–2017 (Washington, World Bank, 2017) 20. 41 http://minis.mn/en/p/shuren-hydropower-plant-project. 42 World Bank, above (n 40) 20. 43 https://perma.cc/G4PR-J5TA. 44 Mining Infrastructure Investment Support Project (P118109) and Mining Infrastructure Investment Support Project – Additional Financing (P145439), Third and Final Report and Recommendation, 13 July 2017, 4. 45 World Bank, above (n 40) 21.

552 Edouard Fromageau recommended there was no need to investigate the complaint but emphasised the need for the Bank to remain in close contact with the affected communities.46 In April 2018, the panel received three other requests related to MINIS. The first was submitted by Oyu Tolgoi Watch, a Mongolia-based NGO, and local NGOs alleging potential harm from the feasibility study and impact assessment for the Baganuur mine expansion.47 The second was also filed by Oyu Tolgoi Watch and alleged potential harm from the Altain-Uvur Gobiin River Basin Administration.48 The third was filed by Rivers without Boundaries and claimed potential harm from the same subprojects as the 2015 complaint. In this third complaint, the requesters argued that ‘the measures recommended by the Inspection Panel in July 2017 to bring the project into compliance with World Bank policies have not been implemented’, and that new circumstances could justify a new request.49 The panel eventually decided not to register the complaint as it did not consider that the information communicated by both the requesters and the management constituted new evidence or circumstances.50 The panel examined the first and second requests jointly. After traveling to Mongolia to hear from the affected communities, it recommended no investigation was necessary given management’s efforts in these projects, and its commitment to work on the shortcomings identified by the requesters in the future.51 These cases are illustrative of the panel’s quite recent tendency to focus on an ‘early solution’52 of the dispute, notably by fostering interaction between management and the requesters rather than recommending a full investigation of the case. While this approach, similar in many aspects to a conciliation process, allows for faster resolution of the dispute, it also has several shortcomings.53 When it comes to investment arbitration, these cases might provide some guidance to future tribunals when faced with investment projects threatening protected natural sites, as was the case for Lake Titicaca in the Bear Creek v Peru case or the Aymara and Quechua communities’ sacred lakes in South American Silver v Bolivia case.

III. THE INSPECTION PANEL AND HUMAN RIGHTS

Taking into account the impact on human rights of investments on local communities is also one of the recent developments in investment arbitration, as illustrated by cases such as Urbaser v Argentina, or Burlington v Ecuador.54 An interesting case which resonates with this recent trend is Chad: Petroleum Development and Pipeline Project (2001). In the request for inspection received by the panel in March 2001, the requesters argued that their rights and interests had been directly harmed by the Bank’s actions in the design, appraisal, and supervision of pipeline projects.55 Among the various arguments presented by

46 ibid. 47 https://perma.cc/3JKN-PKWK. 48 ibid. 49 https://perma.cc/F5FZ-MQM9. 50 https://perma.cc/BPV5-VH79,

para 19. Bank, above (n 40) 21. 52 Sovacool and others, above (n 24) 883. 53 ‘The avoidance of an investigation could mean that serious policy noncompliance would remain unchecked and might indeed result in missed opportunities for institutional learning’. ibid. See also N Bugalski, ‘An Evaluation of the Inspection Panel’s Early Solutions Pilot in Lagos, Nigeria’ (Amsterdam, SOMO/IDI, 2016). 54 See E de Brabandere, ‘Human Rights and International Investment Law’, in M Krajewski and RT Hoffmann (eds), Research Handbook on Foreign Direct Investment (Cheltenham, Edward Elgar, 2019). 55 World Bank, The Inspection Panel Annual Report 2002–2003 (Washington, World Bank, 2003) 109. 51 World

Accountability Mechanisms and Investment Arbitration: Parallel Dynamics

553

the requesters, they claimed that their human rights were violated as a result of the Bank’s actions, specifically their right to life as a consequence of the ‘bad governance reflected in the recent misappropriation of a premium of US$ 25 million and its use for the purchase of weapons’.56 In its response, management recalled the longstanding position of the World Bank on human rights, ‘that is, that the Bank’s mandate, as per its Articles of Agreement, only extends to social, economic and cultural rights, not civil and political rights’.57 Management did not however completely close the door. More precisely, it claimed that: [T]he Bank is concerned about violations of human rights in Chad as elsewhere while respecting the Bank’s Articles of Agreement which require the Bank to focus on economic considerations and not on political or other non-economic influences as the basis for its decisions. In evaluating the economic aspects of any project, human rights issues may be relevant to the Bank’s work if they may have a significant direct economic effect on the project. Having carefully considered all aspects of this issue, management’s conclusion is that the project can achieve its developmental objectives.58

In its investigation report the panel called into question management’s claim that human rights violations in this case did not have a direct effect on the project, and it ‘felt obliged to examine whether the issues of proper governance or human rights violations in Chad were such as to impede the implementation of the Project in a manner compatible with the Bank’s policies’.59 Given the multiple human rights violations that its investigation revealed, the panel eventually observed that the situation in Chad raised questions about compliance with bank policies, ‘in particular those that relate to informed and open consultation’.60 In that respect, Chad: Petroleum Development and Pipeline Project (2001) is without doubt a landmark case in the panel’s history as it is the first to consider that human rights fall within the boundaries of its jurisdiction.61 Following this, the panel reaffirmed its policy to include human rights instruments in its inspection process. In Honduras: Land Administration Project (2006) the requesters invoked several human rights treaties, such as ILO Convention 169 on Indigenous and Tribal Peoples, through the invocation of the Bank’s policies.62 Management rejected this claim, arguing that: [W]hile this issue may be appropriate to raise within the jurisdictional context of other fora, such as Inter-American Human Rights tribunals, the World Bank’s obligation in this project is to ensure compliance with the World Bank’s applicable policies, including the Operational Directive on Indigenous Peoples. As noted elsewhere, it is management’s view that the project has fully complied with this policy.63

56 https://perma.cc/LH7T-RBC7. 57 Sovacool

and others, above (n 24) 880. Management Response to Request for Inspection Panel Review of the Chad-Cameroon Petroleum Development and Pipeline Project, Chad Petroleum Sector Management Capacity Building Project, and Chad Management of the Petroleum Economy Project, 10 May 2001, para 151. 59 Investigation Report – Chad-Cameroon Petroleum Development and Pipeline Project, Chad Petroleum Sector Management Capacity Building Project, and Chad Management of the Petroleum Economy Project, 17 July 2002, para 215. 60 ibid para 217. 61 In that sense, see Sovacool and others, above (n 24) 880. For signs of this evolution in earlier cases, see GF Sinclair, ‘Beyond Accountability? Human Rights, Global Governance, and the World Bank Inspection Panel’ (21 May 2019), https://ssrn.com/abstract=3391646. 62 Request for Inspection, Honduras: Land Administration Project, 3 January 2006, para 3. 63 Management Response, Honduras: Land Administration Project, 9 February 2006, fn 11. 58 Bank

554 Edouard Fromageau In its investigation report the panel reminded the bank that its operational policies, and especially those applicable to the project, must ensure that the Project Plan is consistent with the terms of international agreements related to a country’s environment and the health and well-being of its citizens.64 It went on to raise serious concern about the statement made by the General Counsel that this obligation only refers to ‘agreements of essentially an environmental nature’.65 Ultimately, the panel explicitly affirmed that the bank should make sure that the Project Plan was consistent with existing human rights treaties.66

IV. CONCLUDING REMARKS

The picture this chapter has attempted to present is one of two parallel universes that seem to evolve in a similar direction. Although still quite limited, an emerging trend in international investment law is materialising due to the willingness of some tribunals to engage with affected communities. The experience of the World Bank Inspection Panel, a body specifically created to give a voice to affected communities, can offer some guidance in that regard. Investment projects are multi-dimensional, and consequently can lead to the emergence of multiple disputes and can involve a wide array of actors. Due to their limited role accountability mechanisms are not a reasonable alternative to arbitration for communities affected by investment projects. Although political support is still limited, current reform processes may lead to procedural changes allowing affected communities, when appropriate, to become intervening parties in future disputes. Lessons from the panel’s procedures can also be useful for policymakers, especially when it comes to representation of affected communities, protection from the risks associated with legal proceedings, the conduct of on-site visits, the proper assessment of consultation processes, or the inclusion of human rights.

64 Investigation

Report, Honduras: Land Administration Project, 12 June 2006, para 253. para 255. 66 ibid para 258. 65 ibid

32 International Investment Law in Indian Courts RIDHI KABRA*

I. INTRODUCTION

I

N 2005, JUDGE Stephen Schwebel commented that ‘the anti-suit enjoinder of international arbitration is a phenomenon that has generated too little consideration, still less confrontation, and still less cure’.1 Since that seminal piece of writing, anti-suit enjoinder of international arbitration (anti-arbitration injunction) has advanced, especially in the field of investment treaty arbitration. In recent times, jurisprudence on this subject has been developed by Indian courts, as India has repeatedly resorted to its domestic courts to thwart investors’ attempts to pursue investment treaty proceedings against it. These decisions highlight not only the jurisdictional conflicts between domestic courts and arbitral tribunals. As this chapter shows, they also implicate issues that are of fundamental systemic concern for the international investment law regime, including the nature of investment treaty arbitration and the extent to which issuing such injunctions can result in a breach of an investment treaty obligation. Accordingly, this chapter examines the contribution to international investment law that Indian courts have made, or are likely to make, through their recent decisions on anti-arbitration injunctions.2 Anti-arbitration injunctions are a species of anti-suit injunctions. Created in the common law world,3 they are a device by which courts restrain a party from commencing or continuing arbitration.4 Although domestic courts interact with international investment tribunals in a

* Ridhi Kabra is an Associate at Three Crowns LLP, London. Views expressed in this chapter are solely of the author. 1 SM Schwebel, ‘Anti-Suit Injunctions in International Arbitration – An Overview’ in E Gaillard (ed), Anti-Suit Injunctions in International Arbitration (New York, Juris Publishing, 2005) 5. 2 Indian courts have contributed to the development of international investment law in other (indirect) ways. Most notably, the first investment treaty case against India, White Industries v India, arose from the Indian courts’ inordinate delays in enforcing a commercial arbitration award. An UNCITRAL tribunal held that the delays did not amount to a denial of justice. However, the tribunal used the Australia-India BIT’s most-favoured nation clause to import the obligation to provide ‘effective means of asserting claims and enforcing rights’ from the India-Kuwait BIT. Having so imported, the tribunal concluded that the delays amounted to a breach of the ‘effective means’ standard. See White Industries Australia Limited v The Republic of India, UNCITRAL, Final Award (30 November 2011). 3 For the origins of anti-suit injunctions in common law, see: JRC Arkins, ‘Borderline Legal: Anti-suit Injunctions in Common law Jurisdictions’ (2001) 18(6) Journal of International Arbitration 603. 4 See, eg, GB Born, International Commercial Arbitration, 2nd edn (Alphen aan den Rijn, Wolters Kluwer, 2014) 1306.

558 Ridhi Kabra number of ways,5 many of which are conciliatory in nature, anti-arbitration injunctions come inbuilt with a presumption of hostility towards such proceedings. This chapter charts India’s contribution to investment treaty arbitration through an examination of decisions relating to injunctions against investment treaty arbitrations. By building on a study of these cases, it further considers whether anti-arbitration injunctions can indeed be viewed as a new ground of contestation between states and arbitral tribunals. The remaining sections are arranged as follows: Section II undertakes a review of three recent decisions in which Indian courts have examined their ability to issue injunctions to enjoin investment treaty arbitrations. Section III examines three discrete issues relating to a court’s interference in arbitral proceedings. These include – (III.A) the source and extent of a domestic court’s power to injunct investment treaty proceedings; (III.B) the practical impact of such injunctions on investment treaty proceedings; and (III.C) the potential for such injunctions to engage the international responsibility of states. Section IV concludes.

II. INDIAN COURTS AND ANTI-ARBITRATION INJUNCTIONS

This section examines three recent decisions issued by Indian courts on the subject of injuncting investment treaty arbitrations. These include: (i) Board of Trustees of the Port of Kolkata v Louis Dreyfus Armatures SAS (Port Trust v Louis Dreyfus Armatures);6 (ii) Union of India v Vodafone Group Plc (UoI v Vodafone);7 and (iii) Union of India v Khaitan Holdings (Mauritius) Limited (UoI v Khaitan Holdings).8 The trajectory of these decisions demonstrates: (i) the evolution of the Indian courts’ approach towards their jurisdiction to issue injunctions against investment treaty arbitrations; and (ii) the sustained recognition of the need to show restraint in issuing such injunctions.

A. Port Trust v Louis Dreyfus Armatures The Indian courts’ first encounter with an application seeking an injunction against an investment treaty arbitration occurred in 2014, when the Board of Trustees of the Port of Kolkata (the Port Trust) sought to restrain Louis Dreyfus Armatuers SAS (LDA) from pursuing arbitration under the France-India Bilateral Investment Treaty (BIT). i. Facts of the Case The underlying dispute concerned a contract between the Port Trust and an Indian joint venture company to supply, operate and maintain cargo handling equipment at a dock complex. LDA was an indirect minority shareholder in the joint venture. Following several

5 See, eg, C Schreuer, ‘Interaction of International Tribunals and Domestic Courts in Investment Law’ in AW Rovine (ed), Contemporary Issues in International Arbitration and Mediation – The Fordham Papers 2010 (Leiden, Brill, 2010). 6 The Board of Port of Trustees of the Port of Kolkata v Louis Dreyfus Armatures SAS and others, 2014 SCC Online Cal 17695. 7 Union of India v Vodafone Group Plc, CS(OS) 383/2017 & IA No 9460/2017, Delhi High Court, Judgment (7 May 2018). 8 Union of India v Khaitan Holdings (Mauritius) Limited, CS (OS) 46/2019, IAs 1235 and 1238/2019, Delhi High Court, Judgment (29 January 2019).

International Investment Law in Indian Courts

559

problems in performing the contract, the joint venture terminated the contract and commenced commercial arbitration against the Port Trust. Simultaneously, LDA commenced investment treaty arbitration against the Indian Republic and the Port Trust, alleging indirect expropriation of its investments, and breach of fair and equitable treatment and full protection and security.9 The Port Trust, in turn, approached the Calcutta High Court for an injunction against LDA on two grounds. The Port Trust first argued that it was not a party to the arbitration agreement under the France-India BIT and could not validly be served a notice of claim by LDA. Second, the Port Trust contended that LDA’s treaty arbitration was oppressive, vexatious and amounted to an abuse of process since: (i) LDA was not a qualified investor under the BIT and lacked jurisdiction to commence the arbitration; and (ii) the treaty arbitration overlapped with the commercial arbitration, thereby resulting in multiplicity of proceedings and creating a risk of conflicting decisions. ii. The Court’s Jurisdiction to Order Anti-Arbitration Injunctions The Calcutta High Court began its analysis with a consideration of its jurisdiction to award anti-arbitration injunctions. LDA had challenged the Court’s power to grant an anti-arbitration injunction on the grounds that that such a power had not been made available to the Court under the Indian Arbitration Act 1996 (the Arbitration Act); to the contrary, Section 5 of the Arbitration Act provided for minimal judicial intervention.10 The Port Trust, for its part, contested the applicability of Section 5 of the Arbitration Act, arguing instead that a court’s power to injunct arbitrations was ‘well-recognised’ and that Section 45 of the Arbitration Act granted courts the power to find arbitration agreements ‘null and void, inoperative or incapable of being performed’.11 Importantly, neither party questioned the applicability of the Arbitration Act to an investment treaty arbitration seated outside India. The Court also failed to address the issue of the Arbitration Act’s applicability to the proceedings at hand. Proceeding on the assumption that the Arbitration Act applied not only to commercial arbitrations, but also to treaty arbitrations seated outside India, the Court held that: he (sic) jurisdiction of the Court to interfere in such a situation is not completely obliterated as one could found (sic) that in Sec. 45 powers have been given to the Court to refuse reference in case it is found that the said agreement is null and void, inoperative or incapable of being performed.12

However, the Court cautioned that it would only exercise its jurisdiction to stay a foreign arbitration if ‘the facts and circumstances of a particular case demonstrate that continuation of such foreign arbitration would cause a demonstrable injustice’,13 such as when: i. no valid agreement exists between the parties; ii. the arbitration agreement is null and void, inoperative and incapable of being performed; or 9 See

Louis Dreyfus Armateurs SAS (France) v The Republic of India, PCA Case No 2014-26 (2018). Indian Arbitration Act 1996, s 5 (‘Notwithstanding anything contained in any other law for the time being in force, in matters governed by this Part, no judicial authority shall intervene except where so provided in this Part’). 11 See, Indian Arbitration Act 1996, s 45 (‘… a judicial authority, when seized of an action in a matter in respect of which the parties have made an agreement referred to in section 44, shall, at the request of one of the parties or any person claiming through or under him, refer the parties to arbitration, unless it finds that the said agreement is null and void, inoperative or incapable of being performed’). 12 Port Trust v Louis Dreyfus Armatures, above (n 6) para 76. 13 ibid para 82. 10 See,

560 Ridhi Kabra iii. continuation of the arbitration proceedings might be oppressive or vexatious or unconscionable.14 iii. The Oppressive Nature of the Proceedings On the substance of the Port Trust’s application, the Court adopted such a cautious approach and deferred to the Tribunal’s jurisdiction, holding that whether LDA qualifies as an investor under the France-India BIT is for the arbitral Tribunal to determine. Similarly, the Court acknowledged the possibility of overlap between the treaty arbitration and the commercial arbitration, but expressed ‘no doubt’ that the Tribunal hearing the investment treaty claims would stay its proceedings in favour of the commercial arbitration if the risk of inconsistent judgments arose. Ultimately, the Court granted the Port Trust’s application and restrained LDA from proceeding with arbitration against the Port Trust on the limited ground that an arbitration against the Port Trust would be oppressive since the Post Trust was not a party to the BIT.

B. UoI v Vodafone If there is one decision that can qualify as ‘landmark’ on the subject of anti-arbitration injunctions in India, it is the 2018 decision of the Delhi High Court in UoI v Vodafone as it offers the most comprehensive analysis to date by an Indian court of anti-arbitration injunctions against investment treaty arbitrations. i. Facts of the Case The case arose out of the retrospective imposition of taxes on Vodafone Group Plc’s (Vodafone) Dutch subsidiary, Vodafone International Holdings Limited (VIHBV) by the Indian Government. VIBHV was subject to a tax demand by Indian tax authorities in relation to its acquisition of certain Indian assets. VIHBV challenged the tax demand and received a decision in its favour from the Indian Supreme Court which discharged VIHBV of its tax liabilities. However, in 2012, the Indian Parliament amended the country’s income tax laws with retrospective effect, thereby enabling the re-imposition of tax on VIHBV. This gave rise to two arbitrations by affiliates of the Vodafone group. The first to commence arbitration was Netherlands-based VIHBV in 2014 under the Netherlands-India BIT.15 In 2017, VIHBV’s parent company, Vodafone, followed suit commencing arbitration under the UK-India BIT.16 It is the pursuit of this second arbitration by Vodafone that India sought to permanently restrain, alleging that the parallel pursuit of arbitrations concerning the same dispute by affiliate entities in the same vertical corporate chain amounted to an ‘abuse of law’. In a short-lived victory for India, the Delhi High Court ordered an interim stay of the proceedings in 2017 on the grounds of ‘abuse of process of law’, finding that the parties and the dispute in the two arbitrations were prima facie identical and that there was a risk of

14 ibid 15 See 16 See

(2017).

para 84. Vodafone International Holdings BV v Government of India, PCA Case No 2016-35. Vodafone Group Plc and Vodafone Consolidated Holdings Limited v Government of India (II), UNCITRAL

International Investment Law in Indian Courts

561

parallel proceedings and inconsistent decisions.17 In 2018, the Delhi High Court overturned the interim injunction and declined India’s request for a permanent injunction. ii. Interaction between the Indian Arbitration Act and Investment Treaty Arbitration Early on in its decision, the Court sought to dispel the assumption on which the decision of the Calcutta High Court in Port Trust v Louis Dreyfus Armatures was founded – that the Arbitration Act could apply to investment treaty arbitrations seated outside India. The Court clarified that being creatures of public international law, investment treaty arbitrations are neither ‘international commercial arbitration[s]’ nor ‘domestic arbitration[s]’:18 Investment Arbitration disputes are fundamentally different from commercial disputes as the cause of action (whether contractual or not) is grounded on State guarantees and assurances (and are not commercial in nature). The roots of Investment Arbitrations are in public international law, obligations of State and administrative law.19

On this basis, the Court significantly departed from the decision in Port Trust v Louis Dreyfus Armatures, and held that: The Act, 1996 including Sections 5 and 45 thereof, do not apply proprio vigore to a BIPA. Section 5 does not apply as this is not a Part I arbitration and Section 45 does not apply as Section 44 makes it clear that Part II of the Act, 1996 will apply to an arbitration considered to be commercial under the Indian law.20

iii. The Court’s Jurisdiction to Order Anti-Arbitration Injunctions The Court commenced its jurisdictional analysis by making short shrift of Vodafone’s argument that national courts inherently lack jurisdiction over investment treaty disputes.21 The Court based this conclusion on two reasons: First, it declared that an Indian civil court’s jurisdiction is ‘all embracing except to the extent it is excluded by an explicit provision of law or by clear intendment arising from such law’.22 In order to find that its jurisdiction to restrain investment treaty arbitrations was not ousted, the Court took refuge in the difference between investment treaty arbitrations under the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (the ICSID Convention) and non-ICSID Convention arbitrations. Noting that exclusivity of the arbitral process is fundamental only to arbitrations under the ICSID Convention, the Court concluded that India’s refusal to become a party to the ICSID Convention indicated that it did not intend to remove domestic court jurisdiction over matters regulated by BITs.23

17 Union

of India v Vodafone Group PLC United Kingdom and Ors, Delhi High Court, Judgment (22 August 2017). para 89. 19 ibid para 91. 20 ibid para 90. 21 Note: The Court also examined its personal jurisdiction over national courts. This discussion is excluded from this chapter for lack of space. A summary of the Court’s analysis of its personal jurisdiction is available here: C Salonidis and S Roy, ‘Union of India v Vodafone Group plc: One Step Closer to Reconciling the Jurisdictional Competence of Domestic Courts and Investment Treaty Tribunals?’ (2019) 34(3) ICSID Review 585. 22 UoI v Vodafone, above (n 17) para 76. 23 ibid paras 76–78. 18 ibid

562 Ridhi Kabra Second, the Court drew a distinction between inter-state arbitration and investor-state arbitration in response to Vodafone’s argument that ‘the obligations of a State, under a bilateral or multilateral international treaty, are owed by a Sovereign State to one or more other Sovereign States … and the remedy for [breach of treaty obligations] had to be found in international law’.24 The Court reasoned that the agreement that materialises when an investor accepts a state’s offer of arbitration is not itself a treaty but ‘falls in a sui generis category’.25 This sui generis instrument involves ‘a contractual obligation and a contractual right’. The mere incorporation of a state’s offer to arbitrate with an investor in a treaty does not mean that the arbitration agreement itself is a treaty as classifying it as such ‘would amount to “lifting the status” of the private investor to the “pedestal of a foreign State”’.26 Not being a treaty itself, the arbitration agreement could not bar the Court’s right to issue an injunction. The Court then proceeded to analyse the source of its power to issue anti-arbitration injunctions. For the Court, this power was founded in its inherent jurisdiction to regulate proceedings that are ‘oppressive, vexatious, inequitable or constitute an abuse of the legal process’:27 the concepts of ‘oppression’, ‘vexation’, ‘inequity’ and ‘abuse of process’ have been known to the common law and equity for centuries, being the primary theories used by the court to regulate its process pursuant to its inherent jurisdiction.

While finding in favour of its inherent jurisdiction, the Court advised for the power to be exercised ‘with caution’ and ‘only with extreme hesitation’,28 noting that: as a matter of self-restraint, a National Court would generally not exercise jurisdiction where the subject matter of the dispute would be governed by an investment treaty having its own dispute resolution mechanism, except if there are compelling circumstances and the Court has been approached in good faith and there is no alternative efficacious remedy available.29

The Court found the need for such restraint in the principle of kompetenz-kompetenz which the Court considered to be a ‘fundamental feature of international arbitration’ and one that is reflected in the 1976 UNCITRAL Arbitration Rules (the Rules governing the arbitration between Vodafone and India).30 Curiously, in reaching this conclusion the Court rejected Vodafone’s argument that issuing such an injunction could itself amount to a breach of India’s obligation to provide fair and equitable treatment under the UK-India BIT.31 The Court’s conclusion was based on a misguided reliance on the difference between the international and domestic definitions of a ‘State’. Whilst agreeing that the judiciary constitutes an organ of the state under international law, the Court reasoned that the Indian Constitution excludes the judiciary from the definition of a state.32 iv. Parallel Proceedings and Abuse of Process Ultimately, the Court found no abuse of process in the case. It concluded that parallel proceedings by affiliates that are part of the same vertical chain of companies in relation to the same 24 ibid

para 12. para 83. 26 ibid para 81. 27 ibid paras 104–105. 28 ibid paras 114–15. 29 ibid para 119. 30 ibid para 116. 31 ibid para 13. 32 ibid paras 112–14. 25 ibid

International Investment Law in Indian Courts

563

dispute are not ‘per se vexatious’.33 Only ‘absurd’ proceedings – such as proceedings brought sequentially after an investor has lost the first arbitration over the same dispute and not simultaneously as in the facts of this case – could be considered vexatious.34 The Court was also comforted by Vodafone’s offer to consolidate the two arbitrations.35 Importantly, the Court again placed weight on the kompetenz-kompetenz principle and affirmed that the arbitral Tribunal was the correct forum to assess the likelihood of abuse of process arising from the parallel proceedings.36

C. UoI v Khaitan Holdings The third and (until now) final foray by Indian courts into the subject of anti-arbitration injunctions occurred in UoI v Khaitan Holdings. This decision reinforced the findings of UoI v Vodafone. i. Facts of the Case India’s application for an injunction in this case arose out of a dispute that has its genesis in the cancellation of various telecom licences by the Indian Supreme Court in 2012 on account of irregularities in the licence granting process. Loop Telecom, an Indian company, was one such company that found its telecom licences cancelled by the Supreme Court. Khaitan Holdings (Mauritius) Limited (Khaitan Holdings), a Mauritius-based company with shareholdings in Loop Telecom, commenced investment treaty arbitration against India under the MauritiusIndia BIT, alleging inter alia that the cancellation of the licences constituted an expropriation of its investment.37 On this occasion India sought to permanently restrain the proceedings for the reason that Khaitan Holdings did not qualify as a bonafide investor under the Mauritius-India BIT as its ultimate controlling shareholders were nationals of India.38 ii. The Court’s Jurisdiction to Order Anti-Arbitration Injunctions Despite having the advantage of the decision in UoI v Vodafone, the Court in this case made little effort to expound on its authority to order anti-arbitration injunctions. Instead, it stated cursorily that ‘the Republic of India is well within its rights to invoke the jurisdiction of domestic Courts’.39 In reaching this conclusion however the Court reinforced the decision in UoI v Vodafone that ‘BITs are sui generis in nature’ and that ‘arbitration proceedings under BITs are not governed by the Arbitration and Conciliation Act, 1996 as they are not commercial arbitrations’,40 seemingly settling the debate as to the applicability of the Arbitration Act to investment treaty arbitration in the Indian context.

33 ibid

para 120. para 122. paras 126–27. 36 ibid paras 134–40. 37 See, Khaitan Holdings (Mauritius) Limited v Republic of India, PCA Case No 2018-50. 38 ibid para 10. 39 ibid para 28. 40 ibid paras 23, 29. 34 ibid 35 ibid

564 Ridhi Kabra The Court also agreed with the need to show restraint in interfering with investment treaty arbitrations. Teleological and practical considerations drove this conclusion: Interference with the BIT dispute resolution mechanism in the case of a genuine investor dispute could lead to erosion of investor confidence and also dislodge the fundamental precincts on which BITs are based.41

Thus the Court held that interference by domestic courts would be justifiable ‘only in “compelling circumstances”, in “rare cases”’.42 However, demonstrating an evolution in Indian courts’ understanding of what can constitute ‘compelling circumstances’ and ‘rare cases’, the Court concluded that injunctions could be ordered for two reasons – first, if the continuation of the investment treaty arbitration would be ‘oppressive, vexatious or an abuse of process’;43 and second, if allowing the arbitration to continue would be ‘contrary to public policy’.44 Regrettably, the Court failed to address the question of what would constitute a breach of ‘public policy’ in the context of an investment treaty arbitration, noting simply that: It is a principle of public policy that the Government has to honour its commitments including bilateral ones … The adherence to treaties is therefore not just a contractual stipulation but a solemn commitment by a sovereign nation. Thus, the continuation of arbitral proceedings is the rule and not the exception.45

iii. Arbitral Tribunals have First Right to Decide Matters of Jurisdiction In line with previous decisions, the Court relied on the kompetenz-kompetenz principle to show deference to an arbitral tribunal’s jurisdiction to decide questions regarding its own jurisdiction.46 Specifically, on the facts of the case, the Court held that the issue of whether Khaitan Holdings is a genuine investor under the terms of the BIT fell within the arbitral Tribunal’s purview to decide: the above grounds, are those that can be that with and decided by the Arbitral Tribunal. The arbitration having been invoked in 2013 and the Tribunal having been constituted and being seized of the dispute, it is not for this Court to adjudicate on these issues. The above issues ought to be raised by the Republic of India before the Arbitral Tribunal, which under Article 21 [of the 1976 UNCITRAL Arbitration Rules], would rule upon the same.47

III. ANALYSIS

This section undertakes a comprehensive review of case law relating to injunctions against investment treaty arbitration with a view to deciphering the common principles on a domestic court’s authority to injunct investment treaty arbitrations (section III.A). It then considers the practical impact of such injunctions on an arbitral tribunal’s ability to continue

41 ibid

para 23. para 1. 43 ibid para 54. 44 ibid para 47. 45 ibid. 46 ibid paras 46, 52–54. 47 ibid para 54. 42 ibid

International Investment Law in Indian Courts

565

the proceedings (section III.B). Finally, this section offers preliminary views on whether antiarbitration injunctions can cause states to be in breach of their investment treaty obligations (section III.C).

A. Domestic Courts’ Authority to Injunct Investment Treaty Arbitrations Although a creation of the common law, courts in both common law and civil law jurisdictions have upheld their ability to order injunctions against international commercial arbitration on several occasions.48 However, common law courts have thus far been the only fora for seeking injunctions against investment treaty arbitration. The credit for extending this mechanism to investment treaty arbitrations does not rest with India. While injunctions are yet to become a popular means of resisting investment treaty arbitrations even in the common law world, they have been considered, even issued, by courts across the common law world since the turn of the millennium. Until its consideration in UoI v Vodafone, the most notable instance of a domestic court injuncting an investment treaty arbitration occurred when the Supreme Court of Pakistan restrained SGS from pursuing arbitration against Pakistan.49 The origins of the dispute in SGS v Pakistan lay in a pre-shipment inspection agreement that SGS had concluded with the Government of Pakistan. When Pakistan terminated the agreement, SGS first resorted to litigation before Swiss courts. Having lost in the Swiss proceedings, SGS then commenced investment treaty arbitration under the Pakistan-Switzerland BIT.50 Pakistan, in turn, sought to have the dispute resolved in accordance with the agreement’s arbitration provision. Both parties took their battle before the courts of Pakistan; SGS sought an injunction against the commercial arbitration commenced by Pakistan, while Pakistan requested an injunction against the investment treaty arbitration commenced by SGS. The Supreme Court of Pakistan granted Pakistan’s request, restraining SGS from pursuing or participating in investment treaty arbitration. In deciding which arbitration – commercial or investment – should be allowed to proceed, the Supreme Court concluded that the BIT had not been incorporated into domestic law and could not be enforced as such over the terms of the parties’ contractual arrangement. Undertaking a substantive review of matters of jurisdiction, the Supreme Court also found that SGS did not qualify as an investor under the BIT, nor did the agreement qualify as an investment. Since Pakistan’s example, courts have shown greater restraint in issuing injunctions against investment treaty arbitration. In the United States (US), for instance, the Second Circuit Court of Appeals rejected Ecuador’s bid to restrain Chevron from proceeding to investment treaty arbitration.51 Ecuador’s reasons for seeking an injunction arose from claims pursued

48 See, eg, Himpurna California Energy Ltd (Bermuda) v Republic of Indonesia, UNCITRAL, Interim Award (26 September 1999), Final Award (16 October 1999), (2000) 25 YB Comm Arb 11, 157; Hub Power Co Ltd v Pakistan Water and Power Development Authority, Judgment, Supreme Court of Pakistan (20 June 2000), (2000) 16(4) Arb Int’l 439; Attorney-General of New Zealand v Mobil Oil of New Zealand, Judgment, High Court, Wellington (1 July 1987) [1989] 2 NZLR 649; Salini v Ethiopia, ICC Case No 10623 (7 December 2001), (2003) 21 ASA Bull 59; Societe Générale de Surveillance, SA v Raytheon European Mgt & Sys. Co., 643 F.2d 863, 868 (1st Cir. 1981); Excalibur Ventures LLC v Texas Keystone Inc. [2011] EWHC 1624. 49 SGS Société Générale de Surveillance SA v Pakistan, ICSID Case No ARB/01/13, Judgment, Supreme Court of Pakistan (Appellate Jurisdiction) (3 July 2002), (2003) 19 Arb Int’l 181. 50 SGS Société Générale de Surveillance SA v Islamic Republic of Pakistan, ICSID Case No ARB/01/13. 51 Republic of Ecuador v Chevron Corporation, 683 F.3d 384 (2d Cir 2011) (Ecuador v Chevron).

566 Ridhi Kabra by Ecuadorian nationals in US courts against Texaco Petroleum (Texaco) for environmental damage. In exchange for a dismissal of the claim, Texaco promised the Court that it would participate in the resolution of the dispute before Ecuadorian courts. When Chevron later commenced investment treaty arbitration,52 Ecuador sought an injunction arguing that Texaco’s promise to litigate the disputes in Ecuador estopped Chevron from commencing investment treaty proceedings. The Court rejected Ecuador’s request for the reason that matters of estoppel fell within the jurisdiction of an arbitral tribunal. As a matter of policy, the Court opined that it would approach Ecuador’s claim ‘with a healthy regard for the federal policy favoring arbitration’.53 Equally, the Court did not see any reason to injunct the proceedings on the grounds that litigation on the same issue was already pending before Ecuadorian courts: there is no inherent conflict between BIT arbitration and the Lago Agrio litigation. Those two proceedings involve different parties and distinct claims … The existence of those parallel proceedings – one in which Chevron asserts wrongdoing on the part of Ecuador and another in which Plaintiffs assert wrongdoing on the part of Chevron – makes clear that the Lago Agrio litigation can coexist with BIT arbitration.54

Caribbean courts have similarly espoused the view that injunctions against investment treaty arbitration should be granted exceptionally. In British Caribbean Bank Limited v The Attorney General of Belize (BCB v Belize), the Caribbean Court of Justice (CCJ) held that: The Court exercises heightened vigilance when asked to restrain international arbitration because the parties have contracted to arbitrate their dispute … The approach to modern arbitration agreements contained in investment treaties is for the court to support, so far as possible, the bargain for international arbitration. Belizean cases have affirmed that it is ‘only with extreme hesitation’ that the court will interfere with the process of arbitration.55

While the emerging agreement to show restraint in issuing anti-arbitration injunctions is a step in the right direction, international consensus on the grounds on which a court may order an anti-arbitration injunction remains to be achieved. There appears to be some agreement that injunctions will be ordered only against inequitable arbitrations, ie arbitrations that are oppressive, vexatious, or amount to an abuse of process. However, oppression, vexation, and abuse of process, are undefined notions that a domestic court is free to interpret as it chooses based on the circumstances of a case. As the Caribbean Court of Appeal held in BCB v Belize: Thus far, the courts have been careful not to restrict the notions of vexation and oppression by attempting a comprehensive definition and have emphasized that what amounts to vexation and oppression must vary with the circumstances of each case.56

This is undoubtedly a double-edged sword. Retaining flexibility to determine situations of inequity is not only inherent to the sovereign powers of a domestic court but also allows the court to assess each application for an injunction on its merits. On the other hand, it risks

52 Chevron Corporation and Texaco Petroleum Corporation v The Republic of Ecuador, UNCITRAL, PCA Case No 2009-23. 53 Ecuador v Chevron, above (n 51) 393. 54 ibid 396. 55 British Caribbean Bank Limited v The Attorney General of Belize [2013] CCJ 4 (AJ) (Caribbean Court of Justice), paras 37–38. See also, British Caribbean Bank Limited v The Attorney General of Belize, Court of Appeal of Belize, Judgment (3 August 2012), para 99; Dunkeld International Investment Limited v The Attorney General (Dunkeld v AG), Court of Appeal of Belize, Judgment (1 November 2013) para 135. 56 BCB v Belize (Caribbean Court of Appeal), above (n 55) para 101.

International Investment Law in Indian Courts

567

fragmentation and prevents the creation of universal rules on the subject. That risk is most evident in notions such as ‘public policy’ – notoriously considered an ‘unruly horse’57 – which the Delhi High Court in UoI v Khaitan Holdings recognised as a legitimate ground to injunct investment treaty arbitrations. Another notable theme emerging from decisions on granting injunctions against investment treaty arbitration, is the courts’ hesitance to review their jurisdiction to order such relief. Neither the Supreme Court of Pakistan nor the US Second Circuit of Appeal considered their power to issue such injunctions. While the Pakistani Supreme Court failed even to consider this issue, the US Second Circuit of Appeal deliberately avoided a decision on the point, noting that: Because we conclude that a stay is unnecessary in this case, we need not resolve the question of whether federal courts have the power to stay arbitration under the FAA (or any other authority) in an appropriate case.58

Aside from UoI v Vodafone, only the Caribbean courts have undertaken a detailed review of their jurisdiction to restrain investment treaty arbitration in the twin cases of BCB v Belize and Dunkeld v AG. In both of those cases the courts had the express power to issue anti-arbitration injunctions under Article 106A(8) of the Supreme Court of Judicature Act of Belize on the grounds that the arbitration would be ‘oppressive, vexatious, inequitable or would constitute an abuse of the legal or arbitral process’.59 While BCB v Belize was underway before the Caribbean Court of Appeal, Article 106A(8) had been struck down as unconstitutional. Notwithstanding, the Caribbean Court of Appeal held that it had the inherent power, emanating from its equitable jurisdiction, to order such injunctions: courts of equity have long accepted the jurisdiction to restrain foreign arbitral proceedings on the grounds of vexation and oppression.60

By the time the case was appealed before the CCJ, the constitutionality of Article 106A(8) had been upheld. The CCJ affirmed the courts’ authority to issue such injunctions, noting that it would have the inherent jurisdiction to issue the injunction even in the absence of a provision such as Article 106A(8): [t]he concepts of ‘oppression’, ‘vexation’, ‘inequity’ and ‘abuse of process’ have been known to the common law and equity for centuries, being the primary theories used by the court to regulate its process pursuant to its inherent jurisdiction … Under the doctrine of kompetenz-kompetenz the arbitrators are competent to determine their jurisdiction although the effective exercise of that jurisdiction remains subject to the inherent competence of the court to decide, in relation to an injunction to restrain international arbitration, whether a particular dispute falls within the scope [of] the arbitration agreement.61

Current jurisprudence, thus, leaves much to be desired regarding clarity on the source of a domestic court’s jurisdiction to injunct investment treaty arbitration. One line of jurisprudence – comprising the Indian cases of UoI v Vodafone and UoI v Khaitan Holdings, and the Caribbean case of BCB v Belize – finds such a power in a court’s inherent jurisdiction. Another line of jurisprudence – comprising the Indian case of Port Trust v Louis Dreyfus Armatures

57 Richardson

v Mellish (1824) 2 Bing 229, 252. v Chevron, above (n 51) 391. 59 BCB v Belize (CCJ), above (n 55) para 30. 60 BCB v Belize (Caribbean Court of Appeal), above (n 55) para 98. 61 BCB v Belize (CCJ), above (n 55) paras 32, 53. 58 Ecuador

568 Ridhi Kabra and the US case of Ecuador v Chevron – suggests that such a power emanates from the domestic arbitration laws of a country. The accuracy of this second line of jurisprudence was questioned in UoI v Vodafone and UoI v Khaitan Holdings, which held that investment treaty arbitrations are not commercial in nature and cannot be subject to injunctions based on domestic arbitration laws.62 It is incorrect to suggest that domestic arbitration laws are inapplicable to investment treaty arbitration, since such laws do apply to non-ICSID investment arbitrations.63 However whether the simple fact that domestic arbitration laws apply to non-ICSID investment treaty arbitrations can be taken to mean that the source for injuncting investment arbitrations lies in such laws remains to be settled.

B. The Practical Effect of an Anti-Arbitration Injunction The extent to which an injunction from a domestic court is likely to prevent an ultimate investment treaty arbitration is suspect. In ICSID arbitration the exclusivity of the arbitral process is expressly preserved in Article 26 of the ICSID Convention.64 As Christoph Schreuer, explains: A faithful application of Art. 26 would have required an automatic deference to the impending decision of the Arbitral Tribunal on its jurisdiction. Any attempt to block ICSID arbitration through injunctions by domestic courts is at variance with the clear mandate of the Convention and may well be in violation of that State’s international obligations under the Convention.65

Even in non-ICSID Convention arbitrations, the conflict between a court issuing an antiarbitration injunction, and an arbitral tribunal’s power to continue arbitral proceedings notwithstanding the injunction, is inherent to the nature of international arbitral jurisdiction. As the Tribunal held in Salini v Ethiopia, ‘[t]he primary source of the Tribunal’s powers is the parties’ agreement to arbitrate. An important consequence of this is that the Tribunal has a duty vis à vis the parties to ensure that their arbitration agreement is not frustrated’.66 In fulfilling such a duty, ‘it may be necessary [for a tribunal] to decline to comply with an order issued by a court’ that prevents an arbitral proceeding from continuing.67 The only known instance of such an injunction, SGS v Pakistan, did not prevent the arbitral tribunal from exercising its jurisdiction, which held that a domestic court’s order restraining an investor from pursuing investment treaty arbitration is not binding upon a tribunal: although the Supreme Court Judgment … is final as a matter of the law of Pakistan, as a matter of international law, it does not in any way bind this Tribunal. We have already adverted to the requirement of Article 41 of the ICSID Convention that this Tribunal determine whether it has the jurisdiction to consider the claims that have been advanced and that we cannot decline to do so.68

62 See

section II above. also P Nair and S Singhal, ‘Enjoining treaty arbitrations under Indian law’ (Global Arbitration Review, 4 February 2019), https://globalarbitrationreview.com/article/1179907/enjoining-treaty-arbitrations-under-indian-law. 64 Convention on the Settlement of Investment Disputes between States and Nationals of Other States (opened for signature 18 March 1965, entered into force 14 October 1966) (ICSID Convention) Art 26 (‘Consent of the parties to arbitration under this Convention shall, unless otherwise stated, be deemed consent to such arbitration to the exclusion of any other remedy’). 65 C Schreuer and others, The ICSID Convention: A Commentary, 2nd edn (Cambridge, Cambridge University Press, 2009) 26–157. See also R Garnett, ‘National Court Intervention in Arbitration as an Investment Treaty Claim’ (2011) 60(2) ICLQ 485, 493. 66 Salini v Ethiopia, ICC Case No 10623, Award regarding the Suspension of the Proceedings and Jurisdiction of 7 December 2001, in ASA Bull 1/2003, para 128. 67 ibid. 68 SGS Société Générale de Surveillance SA v Islamic Republic of Pakistan, ICSID Case No ARB/01/13, Procedural Order No 2 (16 October 2002), (2003) 18(1) ICSID Review 293, 299. 63 See

International Investment Law in Indian Courts

569

The Tribunal further emphasised the need to protect access to investment treaty arbitration. Noting that ‘SGS has a prima facie right to seek international adjudication under the ICSID Convention’, the Tribunal declared that: It is essential for the proper operation of both the BIT and the ICSID Convention that the right of access to international adjudication be maintained. In the Tribunal’s view, it has a duty to protect this right of access and should exercise such powers as are vested in it under Article 47 of the ICSID Convention in furtherance of that duty.69

C. Anti-Arbitration Injunctions and Breaches of Investment Treaties Thus far, this chapter has considered a domestic court’s jurisdiction to order anti-arbitration injunctions, and the practical impact, if any, of such injunctions on a tribunal’s ability to continue arbitral proceedings. Another important issue that merits consideration is whether anti-arbitration injunctions can themselves give rise to investment treaty claims. While the Delhi High Court in UoI v Vodafone suggested that ‘all actions and orders passed by National Courts are not per se violative of the fair and equitable treatment guaranteed by the BIPA’,70 Justice Pollard’s dissenting opinion in the Caribbean Court of Appeal’s decision in BCB v Belize recognised that: the grant of the injunction was tantamount to a denial of the right of the Appellant to resort to arbitration under the Treaty. As an important organ of the State, the Court, in granting of the antiarbitration injunction, by the Court deprived the Appellant of its right under the Treaty and clearly engaged the international responsibility of Belize.71

The reason anti-arbitration injunctions can constitute a state’s breach of its obligations under an investment treaty is because the right to arbitration itself qualifies as a distinct investment under an investment treaty,72 or at the very least, forms part of a ‘bundle of rights’ that together constitute an investment.73 Already in ATA v Jordan, an ICSID Tribunal found Jordan’s extinguishment of an investor’s right to (commercial) arbitration to be a breach of the Jordan-Turkey BIT.74 While the Tribunal in ATA v Jordan did not clarify which obligation(s) under an investment treaty is breached when a state refuses to honour a right to arbitration, what follows identifies, without attempting to analyse in detail, three potential obligations: (i) the obligation to provide fair and equitable treatment; (ii) the obligation against expropriation without compensation; and (iii) the obligation not to frustrate the arbitration agreement.

69 ibid

300. v Vodafone, above (n 17) para 113. v Belize (Caribbean Court of Appeal), above (n 55) para 178. 72 See ATA Construction, Industrial and Trading Company v the Hashemite Kingdom of Jordan, Award (18 May 2010) ICSID Case No ARB 08/2, para 117 (noting that ‘the right to arbitration is a distinct “investment” within the meaning of the BIT because Article I(2)(a)(ii) defines an investment inter alia as “claims to […] any other rights to legitimate performance having financial value related to an investment”’). 73 See Swissbourgh Diamond Mines (Pty) Limited and others v Kingdom of Lesotho, Singapore Court of Appeal (27 November 2018), [2018] SGCA 81, paras 124, 144 (noting that ‘an “investment” comprises a “bundle of rights”, which in turn may in principle include the secondary (procedural) right to bring a claim in order to vindicate the primary (substantive) right to enjoy and exploit the commercial benefit of the investment’). 74 ATA v Jordan, above (n 72) paras 121–29. 70 UoI

71 BCB

570 Ridhi Kabra i. Fair and Equitable Treatment The obligation to provide fair and equitable treatment includes the obligation against denial of justice.75 A traditional denial of justice claim involves allegations that a host state has failed to provide justice through its domestic courts.76 The notion that denial of justice can also encompass claims relating to a state’s interference with access to an international arbitral tribunal was first conceived by FA Mann, who wrote that: Whatever the position may be in regard to contractual obligations in general, the repudiation of an arbitration clause has a distinct and special character in that it involves the denial of access to the only tribunal which has jurisdiction and upon which the parties have agreed. The failure to afford access to tribunals has traditionally been treated as a peculiar and particularly grave instance of State responsibility. It is submitted, therefore, that it would be in line with the accepted tendency of international law, sound doctrine and the demands of justice to hold that a State which repudiates an arbitration clause denies justice. In the past, it is true, denial of justice in the strict and narrow sense of the term implied the failure to afford access to the tribunals of the respondent State itself. But there is no reason of logic or justice why the doctrine of denial of justice should not be so interpreted as to comprise the relatively modern case of the repudiation of an arbitration clause.77

This idea was later championed by Judge Schwebel, who stated in the context of anti-arbitration injunctions that: In customary international law, it has long been accepted that a State that refuses an alien person or company access to its courts commits a denial of justice. That principle equally embraces the situation in which a State refuses access not to its courts but to the arbitral process for which a contract provides … when a domestic court, an organ of the State in the eyes of international law, blocks access to arbitration through issuance of an anti-suit injunction, that too constitutes a denial of justice for which the State of which the court is part (whether or not the judicial branch be independent) is internationally responsible.78

Arbitral tribunals have taken the lead from these eminent scholars, giving rise to a growing line of jurisprudence that recognises disruptions to the right to arbitration as a denial of justice. Thus, the Tribunal in Swissbourgh v Lesotho upheld (by majority) that Lesotho’s shuttering of the investors’ right to access the Tribunal constituted under the South African Development Community Treaty amounted to a denial of justice, reasoning that it is now ‘generally accepted’ that states commit a denial of justice when they interfere with the international arbitral process.79 Similarly, several commercial arbitral tribunals have recognised that anti-arbitration injunctions issued by domestic courts cause a denial of justice giving rise to the international responsibility of a state. For instance, the Tribunal in Himpurna v Indonesia found Indonesia responsible for a denial of justice, noting that: it is a denial of justice for the courts of a State to prevent a foreign party from pursuing its remedies before a forum to the authority of which the State consented, and on the availability of which the

75 For a commentary on denial of justice, see: J Paulsson, Denial of Justice in International Law (Cambridge, Cambridge University Press, 2005). 76 SM Schwebel and others, International Arbitration: Three Salient Problems, 2nd edn (Cambridge, Cambridge University Press, 2020) 67. 77 FA Mann, ‘State Contracts and International Arbitration’ (1967) 42 British Yearbook of International Law 1, 27–28. 78 Schwebel, above (n 1) 12–13. See also, Schwebel and others, above (n 75) ch II; Paulsson, above (n 74) 149–57. 79 Swissbourgh Diamond Mines (Pty) Limited and others v The Kingdom of Lesotho, PCA Case No 2013-29, confidential award summarised by Investment Arbitration Reporter (IAReporter, 14 July 2016), www.iareporter.com/ articles/arbitrators-hold-state-liable-for-a-denial-of-justice-occurring-in-relation-to-actions-taken-in-internationalforums-rule-of-law-treaty-obligation-also-breached/.

International Investment Law in Indian Courts

571

foreigner relied in making investments explicitly envisaged by that State. … [A] state is responsible for the actions of its courts, and one of the areas of state liability in this connection is precisely that of denial of justice.80

ii. Expropriation A domestic court’s interference with an investor’s right to arbitrate through an anti-arbitration injunction may also give rise to a claim of expropriation.81 Belize faced such a claim in Dunkeld v Belize.82 Dunkeld, a company registered in the Turks and Caicos, commenced investment treaty proceedings against Belize on account of Belize’s compulsory acquisition of Dunkeld’s investment in a Belizean telecommunications company (Dunkeld v Belize I).83 The Supreme Court of Belize injuncted Dunkeld from pursuing this arbitration, which led Dunkeld to commence a second investment treaty arbitration on the grounds that Belize had expropriated its right to arbitration (Dunkeld v Belize II).84 The arbitration was terminated before the merits of Dunkeld’s claim could be tested because the Belize Court of Appeal lifted the injunction. The merits of an expropriation claim arising from an anti-arbitration injunction has been addressed most clearly, albeit indirectly, in the ICSID case of Saipem v Bangladesh. The dispute related to a Bangladesh-seated International Chamber of Commerce (ICC) arbitration between Saipem and a Bangladeshi state-owned company, Petrobangla. During the pendency of the ICC arbitration, courts in Bangladesh issued anti-arbitration injunctions enjoining Saipem from pursuing the ICC arbitration. The Bangladeshi courts also revoked the tribunal’s authority and ultimately found the award issued in the ICC arbitration to be a nullity, that could neither be set-aside nor enforced. Aggrieved by the interference of the Bangladeshi courts, Saipem instituted investment treaty arbitration against Bangladesh, alleging inter alia that Bangladesh’s interference with its right to arbitration and the revocation of the tribunal’s authority constituted an illegal expropriation. The Tribunal upheld Saipem’s claim of expropriation as regards the revocation of the tribunal’s authority, but did not deliver a binding pronouncement on the impact of the anti-arbitration injunction. In obiter dicta, however, the Tribunal recognised the possibility that anti-arbitration injunctions can result in an expropriation of an investor’s right to arbitration, noting that ‘the issuance of an anti-arbitration injunction can amount to a violation of the principle [of recognition of arbitration agreements] embedded in Article II of the New York Convention’.85 iii. Right to Arbitrate as a Substantive Right Uniquely Dunkeld v Belize I has created the opportunity for states to be held in breach of an investment treaty’s dispute resolution provision, for injunctions issued by their courts against

80 Himpurna California Energy Ltd v Indonesia, UNCITRAL, Interim Award and Final Award (26 September 1999 and 16 October 1999) (2000) XXV Yearbook Commercial Arbitration 109, para 184. See also Salini v Ethiopia, paras 143, 146; ICC Case No 4695, Interim Award (November 1984) (1986) XI Yearbook Commercial Arbitration 149, at 158. 81 See Garnett, above (n 64) 492–93. 82 Dunkeld International Investment Ltd v The Government of Belize (Number 2), PCA Case No 2010-21. 83 See, Dunkeld International Investment Ltd. v The Government of Belize (Number 1), PCA Case No 2010-13. 84 See Le Peterson, ‘Belize manages to stall trio of treaty arbitrations by foreign investors’ (IAReporter, 4 January 2012), www.iareporter.com/articles/belize-manages-to-stall-trio-of-treaty-arbitrations-by-foreign-investors/. 85 Saipem SpA v The People’s Republic of Bangladesh, Award (30 June 2009) ICSID Case No ARB/05/7, paras 167–68.

572 Ridhi Kabra investment treaty arbitrations. As discussed above, Dunkeld v Belize I concerned a dispute relating to Belize’s expropriation of Dunkeld’s investment. When this arbitration was injuncted by the Belizean Supreme Court, Dunkeld commenced Dunkeld v Belize II seeking relief for the alleged expropriation of its right to arbitration. However, the Belizean Supreme Court also injuncted this second arbitration. Ultimately, both injunctions were lifted by the Belizean Court of Appeal, thus, removing the need for Dunkeld to continue its claim in Dunkeld v Belize II. In Dunkeld v Belize I, Dunkeld sought ‘compensation for legal and other costs’ that it had spent on the second arbitration and in the court proceedings relating to the injunctions. Instead of requesting only the recovery of costs, Dunkeld instead claimed damages from the Belizean Government on the grounds that its expenses amounted to ‘losses … either flowing from the Government’s breach of the agreement to arbitrate set out in Article 8 of the Treaty and/or on the basis that they were necessary to bring and pursue these proceedings’.86 Belize, for its part, conceded both that the injunction constituted a breach of the UK-Belize BIT’s dispute resolution provision, and that the Tribunal had jurisdiction to award the requested costs.87 The Tribunal too muddied the waters between an order for costs and an award for damages. On the one hand, the Tribunal indicated that the right to dispute resolution under an investment treaty is a substantive right, pronouncing that a breach of an investment treaty’s dispute resolution provision entitles an investor to claim damages under the customary international law standard of full reparation: the Parties’ agreement that the Government’s efforts to enjoin the present proceedings constitute a violation of Article 8 of the Treaty – and that the violation of that provision entails a damages remedy – leads to a different result … Article 8 specifies no particular standard of damages for a breach of the obligation to arbitrate. The applicable standard in respect of such a breach is accordingly that existing in customary international law.88

On this basis, the Tribunal held that full reparation would cover the costs incurred by Dunkeld in both the domestic litigations and the investment treaty arbitration.89 However, on the other hand, the Tribunal created ambiguity when it refrained from stating clearly whether it was ordering costs or awarding damages: Whether viewed as a matter of costs or as compensation for the breach of Article 8, the Tribunal therefore considers that its role is limited to determining whether the amounts claimed by Dunkeld are reasonable.90

IV. CONCLUSION

Anti-arbitration injunctions have faced severe criticism by scholars who consider it objectionable for domestic courts to interfere with the workings of an international dispute resolution process.91 Meanwhile, the judgments of the Indian (and even the Belizean) courts betray the 86 Dunkeld

v Belize II, above (n 82) para 332. para 334. para 336. 89 ibid para 337. 90 ibid para 338. 91 Schwebel, above (n 1) 5 (noting that ‘[t]he threats to and breaches of the efficacy, the integrity, and in some cases the very viability of international arbitration are profound’); JDM Lew, ‘Anti-Suit Injunctions Issues by National Courts to Prevent Arbitration Proceedings’ in E Gaillard (ed), Anti-Suit Injunctions in International Arbitration 87 ibid 88 ibid

International Investment Law in Indian Courts

573

fact that underpinning their decision to uphold their jurisdiction to injunct investment treaty arbitrations was the concern that investment treaty tribunals should not be the sole arbiter of issues impinging on a state’s regulatory and public interest. This jurisdictional battle has dominated the narrative on anti-arbitration injunctions to date. However, as this chapter has shown, the relevance of anti-arbitration injunction extends beyond this jurisdictional battle. Such injunctions have the potential to give rise to investment treaty claims, defying the very purpose for which the state sought the injunctions in the first place. Furthermore, it is possible to view anti-arbitration injunctions through an alternative lens as being an effective tool that can ensure early dismissal of frivolous claims, which have been identified as one of the issues affecting the legitimacy of investment treaty arbitration.92 Current mechanisms for early dismissal of frivolous claims are far from comprehensive; they are available only in arbitrations conducted under the ICSID Convention,93 or pursuant to certain investment treaties.94 As policy-makers consider further reforms to create a ‘more predictable framework’ to address frivolous claims,95 domestic courts could present themselves as a useful fora for weeding out frivolous claims. Such a partnership with domestic courts to prevent frivolous claims at an early stage could reduce the overall costs of arbitration and help enhance the overall legitimacy of investment arbitration.

(New York, Juris Publishing, 2005) 25 (noting that ‘[a]nti-suit injunctions cause many difficulties for all those who are involved in an arbitration, destabilising the parties’ dispute resolution environment’); Born, above (n 3) 1306 (noting that ‘[i]n many cases, antiarbitration injunctions are part of deliberately obstructionist tactics, typically pursued in sympathetic local courts, aimed at disrupting the parties’ agreed arbitral mechanism’); M Scherer and T Giovannini, ‘Anti-Arbitration and Anti-Suit Injunctions in International Arbitration: Some Remarks Following a Recent Judgment of the Geneva Court’ (2005) 3 Stockholm International Arbitration Review 201, 205 (noting that ‘[a]s a matter of fact, an anti-suit (or rather an anti-arbitration) injunction would clearly violate the internationally recognised principle of Kompetenz-Kompetenz, which requires that the arbitral tribunal has priority to decide on its jurisdiction’). 92 For instance, the UNCITRAL Working Group III on ‘Investor-State Dispute Settlement Reform’ has identified frivolous claims as an issue requiring procedural reform. See, eg, Report of the Working Group III (Investor-State Dispute Settlement Reform) on the work of its thirty-fourth session (Vienna, 27 November – 1 December 2017)’, UN Doc A/CN.9/930/Rev.1, para 59. 93 ICSID Rules of Procedure for Arbitration Proceedings (ICSID Arbitration Rules) (April 2006) r 41(5). 94 See, eg, Dominican Republic-Central America-United States Free Trade Agreement (CAFTA), Arts 10.2.4 and 10.2.5; EU-Canada Comprehensive Economic and Trade Agreement, Art 8.33. 95 UNCITRAL Working Group III, Possible Reform of investor-state dispute settlement (ISDS): Security for cost and frivolous claims, UN Doc A/CN.9/WG.III/WP/192 (16 January 2020) para 27.

33 International Investment Law in US Courts JAVIER GARCÍA OLMEDO*

I. INTRODUCTION

R

ECENT US JURISPRUDENCE has clarified the scope of judicial review of investment treaty awards and their recognition and enforcement procedures in the investment treaty system, especially the execution of awards rendered under the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (ICSID Convention).1 This chapter examines this jurisprudence on these two post-award stages and shows that American courts are providing respondent states with defences against investment treaty awards through setting aside and enforcement proceedings. To that end, section I examines how American courts have interpreted the grounds in the Federal Arbitration Act (FAA)2 for setting aside investment treaty awards, with a focus on the standard of review adopted by these courts when deciding challenges to non-ICSID awards. The question that arises here is whether courts or arbitrators in the US should decide issues of substantive arbitrability, that is, ‘whether the disputing parties had agreed to arbitrate, not simply when they had agreed to do so’.3 The recent decisions of the US Supreme Court in BG Group v Argentina and Henry Schein v Archer & White are pertinent. In BG Group v Argentina, considered as the first case in which the Supreme Court ruled on the standard of review of investment treaty awards, the Court held that a question concerning the existence, scope or validity of an arbitration agreement should be subject to a de novo review by domestic courts in setting aside proceedings.4 Albeit in the context of a commercial arbitration, the Supreme Court confirmed this standard of review in Henry Schein v Archer & White. In so doing, the Supreme Court may have provided respondent states with a new of type of defence to overturn non-ICSID awards. Section II examines the position of US courts on the interplay between the Foreign Sovereign Immunities Act (FSIA) and the provisions for the recognition and enforcement of

* Postdoctoral Researcher, University of Luxembourg and Lecturer, Distance Learning LL.M. in International Dispute Resolution, Queen Mary University of London. 1 Convention on the Settlement of Investment Disputes Between States and Nationals of Other States, 18 March 1965, 17 UST 1270, 575 UNTS 159 (hereinafter ICSID Convention). 2 9 U.S.C. §§ 1–14 (2012). The FAA, which governs domestic and international arbitral agreements, was passed in 1925. It was amended in 1970 to incorporate the New York Convention. 9 U.S.C. §§ 201–08. 3 A Roberts and C Trahanas, ‘Judicial Review of Investment Treaty Awards: BG Group v Argentina’ (2014) 108(4) American Journal of International Law 753 (emphasis added). 4 S Breyer, The Court and the World: American Law and the New Global Realities (New York, Random House, 2016).

576 Javier García Olmedo awards under the ICSID Convention. The focus here is on the question whether the jurisdictional requirements and the ‘immunity from execution’ rule found in the FSIA are conditions for the recognition and enforcement of ICSID awards in the United States. The recent decisions from the US Court of Appeals for the Second Circuit in Mobil v Venezuela and Micula v Romania will be examined in the context of the FSIA’s jurisdictional requirements. With respect to immunity from execution, this chapter will consider the ruling in LETCO v Liberia, also decided by the Second Circuit. This case law allows a conclusion that, within limits, US courts exert more control over ICSID awards, providing remedies that can either invalidate these awards or frustrate their enforcement.

II. SETTING ASIDE AND THE STANDARD OF REVIEW OF INVESTMENT TREATY AWARDS

As Roberts observes, ‘[s]tandards of review are a crucial issue in the investment treaty system because investor-state awards are not currently subject to a full-blown appellate review mechanism’.5 The standards of review for ICSID and non-ICSID awards vary significantly. For ICSID arbitration, states can only challenge an arbitral award through annulment proceedings on the grounds in Article 52(1) ICSID Convention, which include that the Tribunal ‘manifestly exceeded its powers’.6 An application for annulment will be decided by an ad hoc committee and, if rejected, Articles 53(1) and 54(1) ICSID Convention mandates that the respondent state must enforce the pecuniary obligations imposed by the award as if it was a final judgment of its own courts. The Convention thus creates a self-contained and delocalised award review regime, meaning that a national court is not entitled to determine the validity of ICSID awards. Attempts to annul or set aside ICSID awards before national courts have consequently been rejected.7 Investment treaty awards rendered outside the legal framework of the ICSID Convention are, however, subject to judicial review by national courts. The respondent state can seek to set aside the award in the seat of arbitration under the grounds for annulment of the laws of that state. The respondent state can equally resist recognition and enforcement of the award on the grounds established under the Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York Convention).8 The grounds to refuse enforcement and to set aside awards are normally similar and ‘typically include that the award deals with a dispute or matters outside the scope of the submission to arbitration’.9 A national arbitration law that provides for this ‘arbitrability’ ground for annulment is the FAA. Section 10 FAA outlines four interpretative provisions that qualify as grounds for vacating an arbitral award, including those rendered under investment treaties, when the seat of the arbitration is in the US. The FAA provides that the fourth ground for vacating an arbitration award can be invoked ‘where the arbitrators exceeded their powers, or so imperfectly executed them that a mutual, final, and definite award upon the subject matter submitted was not made’. This ground was at the heart of the decision by the US Supreme Court in BG Group v Argentina. 5 ibid

750–763. Convention, Art 52(1)(b). 7 See, eg, Tembec Inc. et al v United States of America 570 F.Supp.2d 137 (2008) (Court held petition for vacatur was barred because of res judicata and collateral estoppel). 8 Convention on the Recognition and Enforcement of Foreign Arbitral Awards, Arts III, V, June 10, 1958, 21 UST 2517, 330 UNTS 3 (hereinafter New York Convention). 9 Roberts above (n 3) 2. 6 ICSID

International Investment Law in US Courts

577

A. BG Group v Argentina The case arose out of emergency actions (tariffs laws) taken by Argentina in the wake of its economic meltdown in 2001. These actions adversely affected BG Group, a company from the UK with investments in the Argentinean gas industry. BG group initiated arbitration proceedings pursuant to the United Nations Commission on International Trade Law (UNCITRAL) Arbitration Rules and under Argentina-United Kingdom Bilateral Investment Treaty (BIT).10 BG Group claimed that Argentina’s gas tariff laws breached the expropriation and fair and equitable treatment provisions of the BIT.11 BG Group succeeded in the arbitration seated in Washington, D.C., and Argentina sought to vacate the award before US courts on the basis that under section 10 FAA ‘the arbitrators exceeded their powers’ by permitting BG Group to arbitrate without complying with the local litigation requirement of the BIT. Articles 8(1) and (2)(a)(i) BIT required that an investor litigate its claim for 18 months at the courts of the host state before submitting the dispute to an arbitral tribunal. BG Group commenced arbitration without first attempting recourse in the Argentinian courts. Argentina objected to the jurisdiction of the Tribunal, arguing that BG Group did not respect the local litigation requirement in Article 8(2)(a)(i).12 In response, BG Group argued that the local litigation requirement was senseless because there was no chance that a decision in a case of this nature could be rendered within 18 months.13 The arbitral tribunal ultimately rejected Argentina’s objection, finding that Article 8(2)(a)(i) should not prevent BG Group from arbitrating the dispute, given that, by its own actions, Argentina had unilaterally hindered recourse to its domestic judiciary.14 On the merits, the Tribunal ordered Argentina to pay BG Group US$185 million for violating the BIT’s fair and equitable treatment obligations.15 In setting aside proceedings, the US District Court for the District of Columbia denied Argentina’s challenge and confirmed the award.16 The Court first noted that ‘judicial review of arbitral awards is extremely limited’ and that ‘careful scrutiny of an arbitrator’s decision would frustrate the FAA’s “emphatic federal policy in favor of arbitral dispute resolution”’.17 The Court then ruled that, since the Tribunal has not ‘stray[ed] from interpretation and application of the agreement [ie the BIT]’,18 it had not exceeded its powers, thereby accepting the Tribunal’s decision to treat the local litigation requirement as a non-impediment to arbitration. On appeal, the US Court of Appeals for the District of Columbia Circuit reversed the District Court’s judgment and vacated the award.19 The Court of Appeals reasoned that it could review de novo the jurisdictional basis of the resulting award if, as in the present case, the matter before the Tribunal concerned a challenge on the existence and formation of an arbitration agreement. This all the more so, the Court added, when the parties did not agree on whether the courts or arbitrators should decide that matter.20 On that basis, the Court found

10 Agreement for the Promotion and Protection of Investments, UK-Arg., Dec. 11, 1990, 1993 UKTS No 41, 1765 UNTS 33. 11 BG Group Plc v The Republic of Argentina, UNCITRAL, Final Award, 24 December 2017, para 141. 12 ibid para 141. 13 ibid para 142. 14 ibid paras 155–57. 15 ibid para 467. 16 BG Grp. PLC v Republic of Arg., 715 F.Supp.2d 108 (D.D.C. 2010); 764 F.Supp.2d 21 (D.D.C. 2011). 17 ibid paras 121–22. 18 ibid. 19 BG Grp. PLC v Republic of Arg., 665 F.3d 1363 (D.C. Cir. 2012). 20 ibid 1371.

578 Javier García Olmedo that, by disregarding the local litigation requirement, the Tribunal had ‘rendered a decision wholly based on outside legal sources and without regard to the contracting parties’ agreement’.21 In the Court’s view, for the arbitration agreement to materialise, BG Group must resort to the Argentine courts and wait 18 months before initiating arbitration. The case went to the US Supreme Court,22 which had to decide for the first time the standard of review of an investment treaty award. The core question before the Court was who, court or arbitrator, must determine the application and interpretation of the local litigation requirement in the BIT.23 The answer to this question depended in turn on whether the requirement was considered merely as a procedural prerequisite to arbitration (in which case deference should be given to the arbitrators’ decision) or whether it constitutes a substantive condition on Argentina’s consent to arbitration (reviewable de novo by courts). This analytical framework on procedural and substantive arbitrability derives from previous jurisprudence, albeit in the context of commercial arbitration proceedings. American courts distinguish between ‘gateway’ issues and ‘non-gateway’ issues of arbitrability. Gateway issues of arbitrability (also known as substantive arbitrability) concern ‘the existence, validity and scope of an arbitration agreement’. Non-gateway issues of arbitrability (also known as procedural arbitrability) ‘include whether a condition precedent to arbitration has been fulfilled, and whether a claim sought to be resolved by arbitration is timely’.24 In First Options v Kaplan, the Supreme Court held that courts will decide gateway issues of arbitrability unless the parties ‘have “clearly and unmistakably” agreed that such issues be delegated for arbitral resolution’.25 However, some courts have recognised an exception to the First Option rule in cases ‘where the defendant’s argument in favour of arbitration was “wholly groundless”’.26 As explained in more detail below, the ‘wholly groundless’ exception was recently abolished by the Supreme Court in Henry Schein v Archer & White for being inconsistent with the FAA. This notwithstanding, under US law, issues concerning consent to arbitration should, in principle, be decided by courts, since, as one author puts it, this ‘avoids the risk of forcing the parties to arbitrate a matter that they may not have not agreed to arbitrate’.27 In BG Group v Argentina, the majority of the Supreme Court acknowledged this jurisprudence, noting that, if the contract or arbitration agreement does not specify whether a particular matter is primarily for arbitrators or for courts to decide, courts will decide gateway or substantive questions of arbitrability, whereas arbitrators will determine non-gateway or procedural questions of arbitrability.28 The majority first noted that nothing in the BIT ‘gives substantive weight to the local court’s determinations on the matter at issue between the parties’ and ‘provides that “only the arbitration decision shall be final and binding on both parties”’.29 It then held that the application and interpretation of the local litigation requirement was an issue of procedural arbitrability because ‘[i]t determines when the contractual

21 ibid

1365–66. Grp. PLC v Republic of Arg., 133 S.Ct. 2795 (2013). 23 ibid 1204. 24 EJ Lee, ‘The US Supreme Court’s decision in Henry Schein v Archer & White: an affirmation of first principles and of the need for clarity over delegated gateway issues’ (Arbitration Blog, 24 January 2019), https://perma. cc/83QV-MQ5N. 25 ibid. See also First Options of Chi., Inc. v Kaplan, 514 S.Ct. 938, 945 (1995). 26 ibid. 27 J Wong, ‘BG Group v Republic of Argentina: A Supreme Misunderstanding of Investment Treaty Arbitration’ (2016) 43(5) Pepperdine Law Review 556, quoting Howsam v Dean Witter Reynolds, Inc., 537 S.Ct. 79 (2002). 28 BG Grp. PLC v Republic of Arg., 134 S.Ct. 1198 (2014), 1206–07. The majority opinion was written by Justice Breyer, and joined by Justices Scalia, Thomas, Ginsburg, Alito, and Kagan, with Justice Sotomayor concurring in part. 29 ibid 1204, citing the United Kingdom-Argentina BIT, Art 8(3). 22 BG

International Investment Law in US Courts

579

duty to arbitrate arises, not whether there is a contractual duty to arbitrate’.30 Accordingly, the majority presumed that the parties intended for arbitrators, not the courts, to decide the issue and held that the award could not be reviewed de novo. By looking at domestic case law developed under the FAA, the majority decided to treat the BIT ‘as if it were an ordinary contract between private parties’ providing for commercial arbitration.31 Joined by Justice Kennedy, Chief Justice Roberts dissented and held that the question concerning the application and interpretation of the local litigation requirement should be decided by the court because such requirement was a ‘condition [precedent] to the formation of an agreement, not simply a matter of performing an existing agreement’.32 He further criticised the majority for disregarding the public international law nature of the BIT. In his view, the BIT should not be treated as an ‘ordinary contract’ but rather as a ‘treaty between two sovereign nations’ to which the investor is not a party.33 For the dissenting Justices, the public international law character of the BIT meant that courts are entitled to an independent assessment of not only conditions to arbitration expressed as such, but also prerequisites to arbitration like the local litigation requirement. Leaving the dissent and its supporters aside, the majority decision in BG Group v Argentina has provided a welcome opportunity for the US Supreme Court to elucidate the scope of judicial review of investment treaty awards in non-ICSID arbitrations. What emerges plainly from the Court’s reasoning is that, perhaps inadvertently, the Court has provided for a potential remedy against these awards. Indeed, by relying on its previous case law, the Court accepted that deference is not owed to the arbitrators’ decision where one of the parties raises doubts as to the validity, existence or scope of an arbitration agreement, in which case the award can be reviewed de novo. In other words, if setting aside proceedings is triggered by a gateway issue of arbitrability, there is a possibility that the competent court at the seat overturns the award in favour of the respondent state.

B. Henry Schein v Archer & White The standard of review established by the Supreme Court in BG Group v Argentina was recently confirmed in Henry Schein v Archer & White.34 This case involved an antitrust dispute arising out of a distributorship agreement, which in turn contained a dispute resolution clause providing for arbitration under the arbitration rules of the American Arbitration Association (AAA Rules). The AAA Rules contain the typical kompetenz-kompetenz provision found in the rules of most arbitral institutions and national arbitration laws. Rule 7 AAA Rules provides: ‘[T]he arbitrator shall have the power to rule on his or her own jurisdiction, including any objections with respect to the existence, scope or validity of the arbitration agreement or to the arbitrability of any claim or counterclaim’.35 In an attempt to circumvent the arbitration agreement, Archer & White initiated court litigation proceedings against Henry Schein in the US District Court for the Eastern District of

30 ibid

1207–08 (emphasis added).

31 ibid. 32 ibid

1216, 1221–22, Roberts, C.J., dissenting. 1215. 34 Henry Schein, Inc., et al v Archer & White Sales, Inc., 139 S. Ct. 524 (2019). 35 American Bar Association, Commercial Arbitration Rules and Mediation Procedures, Including Procedures for Large, Complex Commercial Disputes Arbitration, Rules Amended and Effective as of 1 October 2013. 33 ibid

580 Javier García Olmedo Texas, alleging violations of state antitrust law and seeking monetary damages and injunctive relief. Henry Schein requested that the District Court compel Archer & White arbitrate the dispute. The question for the Court was who, court or arbitrator, should decide the gateway issue of whether the parties agreed to arbitrate antitrust claims. Henry Schein argued that, by agreeing to arbitration under the AAA Rules, which contain the kompetenz-kompetenz provision, the parties have delegated the substantive arbitrability question to the arbitrators. Henry Schein relied on the First Options judgment described above, where the Supreme Court held that, unless the parties ‘have “clearly and unmistakably” agreed’ to the contrary, gateway issues are to be determined by the court.36 In response, Archer & White invoked the ‘wholly groundless’ exception to the First Options rule and argued that antitrust matters are clearly non-arbitrable in the US.37 Both the District Court and the Court of Appeal for the Fifth Circuit agreed with Archer & White that a ‘wholly groundless’ exception existed, and denied Henry Schein’s motion to compel arbitration on that basis. The Supreme Court took a different view. It found that the ‘wholly groundless’ exception was incompatible with the FAA. The Court reasoned that, in accordance with the text of the FAA, ‘arbitration is a matter of contract, and courts must enforce arbitration contracts according to their terms’.38 With that in mind, and relying upon First Options, the Court stated that ‘parties may agree to have an arbitrator decide not only the merits of a particular dispute’ but also ‘gateway questions of arbitrability, such as whether their agreement covers a particular controversy’.39 Interestingly however, though it abolished the ‘wholly groundless’ exception, the Court expressed ‘no view’ on whether ‘the contract at issue in this case in fact delegated the arbitrability question to an arbitrator’.40 Instead, the Court restated its holding in First Options that ‘courts should not assume that the parties agreed to arbitrate arbitrability unless there is clear and unmistakable evidence that they did so’.41 Thus, the Court expressly failed to answer the question raised by Henry Schein, namely whether the kompetenz-kompetenz provision implied that the parties have ‘clearly’ and ‘unmistakably’ delegated the substantive arbitrability question to an arbitrator. The Court considered that it was for the Court of Appeals to answer this question. Lee explains that, ‘[w] hile seemingly innocuous’, the Court’s position was ‘likely prompted by an important amicus curiae brief submitted by Professor George A. Berman’, in which he states that: Although a majority of courts have found the in-corporation of rules containing such a provision [ie kompetenz-kompetenz] to satisfy First Options’ ‘clear and unmistakable’ evidence test, the [American Law Institute] ALI’s Restatement of the U.S. Law of International Commercial and Investor-State Arbitration has concluded, after extended debate, that these cases were incorrectly decided because incorporation of such rules cannot be regarded as manifesting the ‘clear and unmistakable’ intention that First Options requires. This case presents an opportunity to settle the meaning and application of the First Options test and thereby preserve the proper balance under federal law between the roles of courts and arbitrators in determining arbitrability.42

36 First

Options v Kaplan, above (n 26). above n (24) 1. Schein v Archer & White Sales, above (n 34) 529.

37 Lee,

38 Henry 39 ibid. 40 ibid. 41 ibid

530.

42 Henry Schein v Archer & White Sales, above (n 34), Brief

Respondent, 2 (emphasis added).

of Amicus Curiae, Professor G.A. Berman in Support of

International Investment Law in US Courts

581

In leaving the issue to the Court of Appeals on remand, the Supreme Court seemed to have agreed with Professor Berman’s opinion that a kompetenz-kompetenz provision does not deprive courts of their authority to decide gateway issues of arbitrability, an opinion that also applies to investor-state arbitration. Practically, this means that if an investment treaty does not contain an express and additional (delegation) clause that reserves for arbitrator’s gateway questions of arbitrability, courts in the US will retain the power to review the award in setting aside proceedings. Investment treaties, such as the Argentina-United Kingdom BIT, rarely contain a delegation clause. The dispute resolution provisions of these instruments merely offer access to arbitration in accordance with different arbitral rules, such as the UNCITRAL Rules. However, the UNCITRAL Rules, like the AAA Rules and most arbitral rules, contain a standard kompetenz-kompetenz provision. For obvious reasons, it is also improbable that a respondent state would expressly and separately agree to confer exclusive jurisdiction over arbitrators to decide issues of substantive arbitrability, in which case the concurrent jurisdiction of national courts will remain intact. In short, the Supreme Court in BG Group v Argentina and Henry Schein v Archer & White confirms that, unless the parties have ‘clearly’ and ‘unmistakably’ agreed otherwise (eg through a delegation clause), questions of substantive arbitrability arising in an investor-state dispute are to be determined by courts and not arbitrators. Put another way, insofar as arbitration clauses contained in investment treaties are interpreted according to the same principles as private arbitration agreements subject to the FAA, respondent states may enjoy a new remedy to set aside investment treaty awards in arbitration proceedings seated in the US. Wong rightly observes, in this context, that: The uncertain and malleable classification of an arbitrability question as substantive or procedural, and the ability of the relevant court to review the former de novo, can only encourage a respondent state to seek a second bite at the apple in the vacatur of the award.43

At this stage, the question arises as to how to distinguish substantive arbitrability from procedural arbitrability. As previously explained, gateway issues of (substantive) arbitrability include the existence, validity and scope of an arbitration agreement, whereas non-gateway issues of (procedural) arbitrability include whether a condition precedent to arbitration has been fulfilled. Yet, as Lee notes, ‘[t]hese lines are often blurry, and the characterisation of a jurisdictional issue as either gateway or non-gateway is not always straightforward’.44 The decision in BG Group v Argentina provides some guidance in the context of investment treaty arbitration. The Court made clear that compliance with a local litigation requirement under an investment treaty is not an issue of substantive arbitrability because it simply concerned when arbitration could commence, and not whether the parties have agreed to arbitrate. Accordingly, the Court concluded that it owed deference to the arbitrator’s interpretation of Article 8 within the United Kingdom-Argentina BIT. This does not, however, mean that investment treaties cannot give rise to other jurisdictional issues that may classify as substantive arbitrarily. Consider, for instance, nationality requirements. Under virtually all investment treaties, the host state’s consent to arbitrate depends on the claimant-investor qualifying as an ‘investor’ or ‘national’ of the home state party. Compliance with nationality requirements, therefore, confers jurisdiction ratione personae upon the Tribunal to decide the dispute.

43 Wong, 44 Lee,

above (n 27) 572. above n (24) 1.

582 Javier García Olmedo These requirements, however, are vaguely drafted in many investment treaties, which has resulted in the practice of nationality planning by individuals and corporations. Nationality planning has proven a fertile area for jurisdictional objections in an increasing number of cases.45 Questions on the application and interpretation of nationality definitions clearly constitute an issue of substantive arbitrability, as they concern whether the parties have agreed to arbitrate. Assuming that the arbitral seat is in the US, and that the Tribunal rejects the respondent state’s challenge based on the nationality of the investor, the respondent state may have a compelling argument that the court should review the award de novo. To give another example, think of fork-in-the-road provisions preventing the investor from initiating arbitration proceedings if the investor has previously resorted to the national courts of the host state. A number of investment treaties contain these provisions,46 which have also been subject to jurisdictional objections by respondent states in several cases.47 Non-compliance with a fork-in-the-road provision also qualifies as a question of substantive arbitrability, since consent to arbitration cannot materialise if the investor has chosen court litigation as the mechanism to seek damages.48 In addition to providing a de novo judicial resolution of jurisdictional matters in setting aside proceedings, US courts have also rendered decisions that may frustrate the recognition and enforcement of investment treaty awards in the US. The next section examines how US courts have responded to the invocation of sovereign immunity defences by a respondent against the enforcement of ICSID awards.

III. ICSID AWARDS AND SOVEREIGN IMMUNITY DEFENCES

We have seen that, unlike awards issued under the New York Convention, ICSID awards are in principle self-contained, meaning that these awards can only be challenged before an ad hoc committee within ICSID itself. However, the self-contained regime of ICSID arbitration ends at the stage of enforcement, for the competent court may intervene on grounds of state immunity.49 This proposition finds support in Articles 54(3) and 55 ICSID Convention. Article 54(3) provides that: ‘[e]xecution of the award shall be governed by the law concerning the execution of judgments in force in the State in whose territories such execution is sought’. Article 55 provides that: ‘nothing in Article 54 shall be construed as derogating from the law in force in any Contracting State relating to immunity of that State or of any foreign State from execution’. When read together, these provisions afford a defence against forced execution in the event that the assets of a foreign sovereign are immune under the law of the enforcement forum. This is a logical consequence of ‘the equalization of ICSID awards to national judgments’ as envisaged in Article 54(1).50 As Bjorklund puts it: ‘[although] the holder of an unpaid 45 For a recent analysis of this issue, see J García Olmedo, ‘Narrowing Personal Jurisdiction: An Overlooked yet Indispensable Tool to Recalibrate the International Investment Regime’ International and Comparative Law Quarterly (forthcoming). 46 See, eg, Mexico-Greece BIT (2000) and China-Argentina BIT (2001). 47 S Lee and M Phua, ‘Supervisión y Control v Costa Rica: Developing the Pantechniki v Albania Standard for “Fork in the Road” Provisions in Investment Treaties’ (2019) 34(1) ICSID Review – Foreign Investment Law Journal 203. 48 Wong, above (n 27) 573. 49 J García Olmedo, ‘Sovereign Immunity as a Ground to Refuse Compliance with Investor-State Awards: Past Experience and Future Developments’ in K Fach Gómez and M López Rodríguez (eds), 60 Years of the New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards: Key Issues and Future Challenges (New York, Kluwer Law International, 2019). 50 I Uchkunova and O Temnikov, ‘Enforcement of Awards under the ICSID Convention – What Solutions to the Problem of State Immunity?’ (2014) 29(1) ICSID Review – Foreign Investment Law Journal 187, 191.

International Investment Law in US Courts

583

ICSID award can seek enforcement in the courts of any ICSID Convention country […] its ability to recover will be limited by municipal laws on sovereign immunity’.51 Reliance on sovereign immunity laws against recognition and enforcement may occur at two different stages: first, at the jurisdiction stage (immunity from the competence of the domestic courts where recognition is sought) and second, at the execution stage (immunity from the attachment or seizure of state property).

A. Immunity from Jurisdiction: A Procedural Defence to the Enforcement? The US has become a prime jurisdiction for the recognition and enforcement of foreign arbitral awards. In 1976, Congress passed the Foreign Sovereign Immunities Act (FSIA), which governs all claims brought in any courts in the US, whether state or federal, against states.52 As in many jurisdictions, the FSIA treats immunity from execution as a separate matter from immunity from jurisdiction. In 1988, Congress amended the FSIA to address the issue of jurisdictional immunity from recognition. Section 1605(a)(6) of the amended text provides that the state’s agreement to arbitrate an investment dispute constitutes an implicit waiver of its immunity from proceedings concerning the recognition or confirmation of an award: A foreign State shall not be immune from the jurisdiction of courts of the United States … in any case … in which the action is brought … to confirm an award made pursuant to … an agreement to arbitrate, if … the agreement or award is or may be governed by a treaty or other international agreement in force for the United States calling for the recognition and enforcement of arbitral awards.53

The treaty, or international agreement, referred to in this section includes the New York and ICSID Conventions, to which the US is a party.54 US courts have consistently held that consent to arbitration under the New York Convention constitutes a waiver of state immunity from jurisdiction in the US.55 The same holds true for the ICSID Convention.56 As a result, assuming an agreement to arbitrate exists, the FSIA does not shield a foreign sovereign from the jurisdiction of courts over a civil action to enforce ICSID awards. This ‘arbitration exception’ to immunity is also exhibited in what is known as the ICSID Enabling Statute, which was enacted ten years before the FSIA to implement the ICSID Convention.57 This instrument states that ‘[t]he pecuniary obligations imposed by … an [ICSID] award shall be enforced and shall be given the same full faith and credit as if the award were a final judgment of a court of general jurisdiction of one of the several states’.58

51 AK Bjorklund, ‘Sovereign Immunity as a Barrier to the Enforcement of Investor-State Arbitral Awards: The Re-politicization of International Investment Disputes’ (2010) 21 American Review of International Arbitration 212, 217. 52 Foreign Sovereign Immunities Act, 28 U.S.C., s 1610 (1976). 53 ibid s 1605(a)(6). 54 H Fox and P Webb, The Law of State Immunity, 3rd edn (Oxford, Oxford University Press, 2016) 258. 55 S & R Davis International v The Republic of Yemen, 218 F.3d 1292 at 1301–131302 (11th Cir. 2000), Creighton Limited v Qatar 181 F.3d 118, 123–124 (D.C. Cir. 1999), Cargill International S.A. v M/T Pavel Dybenko 991 F.2d 1012, 1018 (2nd Cir. 1993). See, more recently Pao Tatneft v Ukraine, a decision of 28 May 2019 from the US Court of Appeals for the District of Columbia, discussed in JP Duffy and D Avila, ‘Does Signing an International Treaty Impliedly Waive Sovereign Immunity in the U.S. under the FSIA?’ (Kluwer Arbitration Blog, 2 November 2019), https://perma.cc/BH7Z-BW63. 56 Blue Ridge Investments, LLC v Argentina, 10 Civ. 153 (SDNY. 2012). 57 22 U.S. Code § 1650a – Arbitration awards under the Convention. 58 ibid.

584 Javier García Olmedo However, as one author observes, ‘[s]eeking to alleviate the “diplomatic pressures” that the U.S. government faced from foreign governments seeking immunity from suit in U.S. courts’, the FSIA ‘sought to provide certain requirements for obtaining in personam jurisdiction over a foreign state’.59 These requirements include service of process upon the state debtor and venue. The question arises as to whether such requirements apply to the recognition and enforcement of ICSID awards. Until very recently, the Southern District of New York had adopted a practice of ex parte recognition of ICSID awards without regard to the requirements of the FSIA, thereby allowing investors to avoid the ‘expensive and time-consuming process of a plenary proceeding’.60 The court has done so by relying on the ICSID Enabling Statute, which presumably permits this practice. An illustrative example includes the enforcement proceedings initiated in Siag v Egypt.61 In that case, after Egypt refused to comply with the award, the investor filed an ex parte application with a proposed order and judgment of US$133 million. In ruling on the application, the court noted that the ICSID Enabling Statute and Article 54(1) ICSID Convention allows courts to automatically convert ICSID awards into final judgments. In this view, the court considered that it was entitled to apply the New York Civil Practice Law and Rules (CPLR), which allow ex parte confirmation of the judgment (award) without the need for a separate plenary action. In subsequent years, judges in the Southern District continued to grant ex parte applications. This, however, changed after the Court of Appeals for the Second Circuit issued a much-awaited decision in Mobil v Venezuela, a decision that signaled a turning point for ICSID award creditors in the US. i. Mobil v Venezuela To provide some background, in an award dated 10 October 2014, an ICSID tribunal ordered Venezuela to pay ExxonMobil over US$1.6 billion plus interest.62 Venezuela denied payment and the investor filed an ex parte petition in the Southern District of New York for recognition of the award under New York’s CPLR.63 The court granted the requested relief the same day and Venezuela moved to vacate the judgment. The state argued that nothing in Article 54 or the ICSID Enabling Statute permitted the issuance of a judgment on an ICSID award on an ex parte basis. Venezuela contended that the FSIA took precedence over any other instrument, meaning that the state was entitled inter alia to service of (and response to) all pleadings.64 Both arguments were rejected. In line with its previous practice, the court reasoned that ‘Congress’s use of “full faith and credit” in the ICSID enabling statute’ reassured the automatic enforceability (hence recognition) of ICSID awards.65 In this respect, the court feared that the FSIA’s procedural and jurisdictional requirements might provide foreign respondents with an avenue for delay and non-compliance.66 59 MD Slater and others, ‘Jurisdictional and Forum Requirements for ICSID Award Recognition Against Foreign Sovereigns: Recent Developments and Debates’ (2017) 32 Mealey’s International Arbitration Report, citing H.R. Rep. No 94-1487, 7 (1976), reprinted in 1976 U.S.C.C.A.N. 6605. 60 Micula v Government of Romania, No 15 Misc. 107 (Part I), 2015 WL 4643180, 3 (SDNY Aug. 5, 2015). 61 Siag v Arab Republic of Egypt, No M-82, 2009 WL 1834562 (SDNY June 19, 2009). 62 Mobil Cerro Negro Holding, Ltd., Mobil Cerro Negro, Ltd., Mobil Corporation and others v Bolivarian Republic of Venezuela, ICSID Case No ARB/07/27, Award, 9 October 2014. 63 Mobil Cerro Negro Ltd. v Bolivarian Republic of Venezuela, 87 F. Supp. 3d 573 (SDNY Feb. 13, 2015). 64 ibid 586. 65 ibid 597. 66 ibid 600.

International Investment Law in US Courts

585

The enforcement battle did not end there. Venezuela appealed the decision before the Second Circuit and argued that the lower court impermissibly treated the ICSID Enabling Statute as ‘an alternative … basis to the FSIA for federal court subject matter jurisdiction’.67 Venezuela insisted that the ICSID Convention does not purport to limit its sovereign immunity and thus the application of the FSIA’s service and notice requirements.68 Upon request to the Second Circuit, the US Government submitted an amicus curiae,69 taking the position that the ‘FSIA is the sole source of [personal] jurisdiction over an action to enforce an ICSID award against a foreign sovereign and its rules must be followed’. Interestingly, citing Article 55 ICSID Convention, the government opined that the ‘the ICSID statute’ mandated reliance on the immunity provisions of the FSIA because the ICSID Convention itself forbids derogation ‘from the law in force in any Contracting State relating to immunity of … any foreign State from execution’. Thus, according to the government, since the FSIA does not permit ‘a federal court to “borrow” procedures from state law that permit an ex parte proceeding’, the lower court erred in granting Mobil’s petition. By a decision of 11 July 2017, the Second Circuit sided with the opinion of the US Government that the FSIA should not be averted. The Second District concluded that ‘the FSIA provides the sole source of jurisdiction – subject matter and personal – for federal courts over actions brought to enforce ICSID awards against foreign sovereigns’. Consequently, the judgment of the Southern District of New York was vacated. ii. Micula v Romania A few months later, on 23 October 2017, a different Second Circuit panel followed the Mobil holding in Micula v Romania. In that case, an ICSID tribunal rendered an award in favour of Swedish investors Viorel and Ioan Micula, holding that Romania had breached the protections of the Sweden-Romania BIT by revoking the incentives that had led them to invest in Romania. The Tribunal awarded the claimants damages equating to US$116,317,868. After the award was rendered, the European Commission (EC) ordered Romania not to comply with the ICSID award, since doing so would violate state aid rules under European Union (EU) law. The decision went so far as to order the Micula claimants to repay Romania any funds they had received that qualified as such state aid.70 The EC so decided despite the fact that Romania acceded to the EU in 2007, that is, two years after the commencement of the arbitration. Against this backdrop, the Micula brothers sought an ex parte judgment on the ICSID award in the District Court, which denied the application. Dissatisfied with the decision, the award creditors took a chance before the Southern District of New York, which, in line with its previous practice, allowed the plaintiffs to proceed with their petition under the expedited procedures of New York’s CPLR. Romania appealed the decision before the Second Circuit, which reiterated the Mobil holding that ‘the FSIA [and its personal jurisdiction requirements] provides the sole basis for jurisdiction over [a foreign state] and sets forth the exclusive procedures for the recognition of the ICSID Award’. After years of unsuccessful efforts, the Micula claimants finally obtained confirmation of the much controversial award. They went back to the District Court and submitted a new 67 Br. for Resp’t-Appellant at 1, Mobil Cerro Negro Ltd. v Bolivarian Republic of Venezuela, No 15-707 (2d Cir. 2017), ECF No 31. 68 ibid. 69 Br. of United States of America as Amicus Curiae at 1, Mobil Cerro Negro Ltd. v Bolivarian Republic of Venezuela, No 15-707 (2d Cir. 2017), ECF No 87. 70 Commission Decision (EU) No 2015/1470 of 30 March 2015, Arts 1-2, [2015] OJ L232/43, 69.

586 Javier García Olmedo petition, with Romania participating in the process this time. In a decision of 11 September 2019, the District Court granted their petition for recognition and enforcement of the award.71 During that process, Romania again opposed the confirmation of the award, joined by the EC appearing as amicus curiae. Both relied heavily on the well-known Achmea judgment in which the Court of Justice of the European Union (CJEU) held that investor-state arbitration provisions in intra-EU BITs are incompatible with EU Law. The District Court rejected this argument, noting that the investment dispute between Romania and the Miculas had arisen prior to Romania’s accession to the EU. Moreover, the District Court relied on a ruling of 18 June 2019 from the General Court of the EU, which annulled the EC’s decision to prevent Romania from complying with the award also on the basis of Romania’s ‘late’ accession to the EU. This is the first time a US court had decided on the recognition and enforcement of an intra-EU investment treaty award since the Achmea case. On 9 October 2019, Romania appealed before the US Court of Appeals for the District of Columbia Circuit and, as of January 2019, these proceedings are pending. Given the inability of European investors to enforce intra-EU treaty awards in Member States, and the importance of the US as an enforcement jurisdiction, a decision by a US court finding that the Achmea judgment precludes enforcement of these intra-EU treaty awards would have serious consequences for these investors. There are at least six other pending cases for the enforcement of intra-EU treaty awards in the US, all against Spain and involving claims brought under the Energy Charter Treaty (ECT).72 Most of these awards were rendered under the ICSID Convention. It should also be noted that, as opposed to Romania, Spain accessed to the EU long before the initiation of these arbitrations. One can only speculate what other US courts, including the US Court of Appeals for the District of Columbia Circuit, will decide with respect to the impact of the Achema judgment on the enforcement of intra-EU treaty awards. What is certain for now, however, is that the decisions by the Second Circuit in Mobil v Venezuela and Micula v Romania seem to have put an end to the Southern District’s longstanding tradition of granting ex parte enforcement of ICSID awards. In fact, on 13 November 2017, that very same court vacated an order for the ex parte recognition of a €128 million award73 rendered against Spain in one of the abovementioned ECT arbitrations.74 The order was initially issued in favour of Eiser Infrastructure and its subsidiary. The Southern District’s reversal followed a letter from Spain of 24 October 2017 addressed to Lewis A Kaplan, the judge who had previously granted the ex parte petition. In the letter, Spain averred that ‘[i]n light of the Second Circuit’s controlling decisions in Mobil and Micula, the ex parte judgment entered against the Kingdom of Spain should be vacated and this proceeding should be dismissed’. 71 Micula

v Romania, Case No 17-cv-02332 (D.D.C. Sept. 11, 2019). Infrastructure Limited, et al v The Kingdom of Spain, Civ. No 18-cv-01686 (Jul. 19, 2018); Novenergia II – Energy & Environment (SCA) v The Kingdom of Spain, Civ. No 18-cv-01148 (May 16, 2018); Infrastructure Services Luxembourg S.A.R.L. et al v Kingdom of Spain, Civ. No 18-cv-01753 (Jul. 27, 2018); Masdar Solar & Wind Cooperatief U.A. v Kingdom of Spain, Civ. No 18-cv-02254 (Sept. 28, 2018); Nextera Energy Global Holdings B.V. et al v Kingdom of Spain, Civ. No 1:19-cv-01618 (June 3, 2019); 9REN Holdings S.A.R.L. v. Spain, Civ. No 1:19-cv-01871 (June 25, 2019). One proceeding (Novenergia) is an UNCITRAL award and is being enforced pursuant to the New York Convention, and two (Infrastructure and Masdar) have been stayed pending ICSID annulment proceedings. Memorandum Opinion and Order, Infrastructure Services Luxembourg S.A.R.L. et al v Kingdom of Spain, Civ. No 18-cv-1753 (D.D.C. Aug. 28, 2019); Memorandum Opinion, Masdar Solar & Wind Cooperatief U.A. v Kingdom of Spain, Civ. No 18-cv-02254 (D.D.C. Sept. 18, 2019). 73 Eiser Infrastructure Ltd. and Energía Solar Luxembourg S.à.r.l v Kingdom of Spain, ICSID Case No ARB/13/36, Award, 4 May 2017. 74 EISER Infrastructure Ltd. and Energía Solar Luxembourg S.à.r.l v Kingdom of Spain, No 17 Civ. 3808 (LAK) (SDNY 2017), ECF No 2. 72 Eiser

International Investment Law in US Courts

587

The legal and practical consequences of this jurisprudence are straightforward: the specific preservation of immunity from jurisdiction in the FSIA does not disable debtor states from relying on the FSIA’s jurisdictional requirements. Put another way, establishing jurisdiction over foreign sovereigns is now an additional hurdle for investors that seek to enforce investorstate awards. Needless to say, this hurdle also affects awards rendered under the framework of the New York Convention like the one in BG Group v Argentina. The struggle over compliance with the FSIA’s jurisdictional requirements is not, however, the only obstacle faced by foreign creditors of ICSID and non-ICSID awards. iii. Immunity from Execution: The Final Burden Once an award has been successfully converted into a local judgment, the investor creditor will hold an enforcement title. This represents the end of the confirmation process, but the start of another: execution, or the attachment of sovereign assets, to satisfy the judgment. To that end, the investor needs to obtain an attachment order, also known as a writ of execution, from a US Court. As previously explained, the FSIA reforms aimed at ensuring that jurisdictional immunity rules were more closely adapted to execution immunity rules. Slater explains that ‘[w]ithout these reforms’: Congress feared that ‘disparate treatment of cases involving foreign governments may have adverse foreign relations consequences’. This was particularly true in the case of execution immunity because Congress was concerned about how other countries would treat execution attempts against the U.S. government’s assets abroad. The FSIA thus mandates that a foreign sovereign’s claim to immunity ‘should … be decided … in conformity with the principles set forth in [the FSIA].’75

The FSIA begins with the presumption that the property of foreign sovereigns in the US is immune from execution.76 This presumption follows the doctrine of restrictive immunity, according to which the implicit waiver of state immunity from jurisdiction provided by an arbitration agreement does not extend to execution. A waiver of state immunity from execution of an award will only exist if the debtor state has expressly consented to such a waiver.77 Under Section 1610(a)(1) FSIA, if a waiver is not expressly provided, execution against a state’s assets to satisfy the debt imposed by an award may proceed only if such assets are ‘used for a commercial activity in the United States’. The exception of a property used for commercial purposes is further qualified by a requirement of a connection between that activity and the underlying claim.78 Accordingly, in the event of a default in payment by the losing state, an investor must locate the sovereign’s assets and establish that they are commercial in nature. Execution immunity was invoked against the enforcement of an ICSID award in LETCO v Liberia.79 The Southern District of New York recognised the award and entered judgment expeditiously. It also issued a writ of execution on Liberia’s property. Liberia moved to vacate the confirmation judgment or, in the alternative, the execution of that judgment on

75 Slater,

above (n 59). Procedures for Recognition and Enforcement of International Arbitration Awards Rendered Under the ICSID Convention, July 2012. 77 Birch Shipping Corporation v The Embassy of the United Republic of Tanzania, Judgment of the United States District Court for the District of Columbia, 18 November 1980, 507 F. Supp. 311 (1980) 312; Creighton Ltd. v Government of the State of Qatar, 181 F. 3d 118 (DC Cir. 1999). 78 FSIA, above (n 52) s 1610(a)(2). 79 Liberian Eastern Timber Corporation v Republic of Liberia, ICSID Case No ARB/83/2, Award, 30 March 1986. 76 Recommended

588 Javier García Olmedo its property. The second motion was successful. Liberia argued that the assets against which LETCO had levied execution (tonnage fees, registration fees and other taxes due the government) were sovereign and not commercial assets.80 Liberia relied on Section 1610 FSIA, which, as just explained, protects sovereign assets from attachment. The court agreed and appropriately preserved Liberia’s rights under the FSIA by vacating the writ of execution.81 The Second Circuit affirmed the decision on appeal.82 This finding, which was consistent with Article 55 ICSID Convention, has been upheld in subsequent cases.83 We should expect states like Venezuela, Romania and Spain to rely on the FSIA’s rule on immunity from execution to continue to avoid the payment of awards. Should they do so, the Miculas and the other investors seeking enforcement in the US would thus be compelled to start the long and costly process of identifying state assets that are used for commercial activities. The enforcement battle is not over yet.

IV. CONCLUSIONS

This chapter has shown that US courts are becoming increasingly more skeptical about investorstate dispute settlement. In BG Group v Argentina, the Supreme Court clarified the standard of review of non-ICSID awards, providing respondent states with a new remedy whereby the arbitrators’ decisions can be reviewed de novo for challenges on substantive arbitrability. This standard was later confirmed in Henry Schein v Archer & White, where the Court innocuously added that even the kompetenz-kompetenz doctrine should not deprive courts of their authority to review issues of substantive arbitrability. In Mobil v Venezuela and Micula v Romania, the Court of Appeals for the Second Circuit put an end to the Southern District’s longstanding tradition of granting ex parte enforcement of ICSID awards. Investors seeking the confirmation, and eventual enforcement, of investment treaty awards in the US must now satisfy the expensive and time-consuming process required by the FSIA’s jurisdictional requirements. The same Court clarified in LETCO v Liberia that, even if those requirements are satisfied, investors may be prevented from collecting payment on the grounds of sovereign immunity from execution where the assets sought are not for a commercial purpose. With these considerations in mind, a conclusion may be drawn that US courts will ultimately determine the significance and finality of investment treaty awards rendered under both the New York and the ICSID Conventions. This was surely not the intention of the drafters of the Conventions, both of which were established to restrict to the maximum extent possible the control of national courts over arbitral awards and to facilitate their recognition and enforcement. This objective is all the more present in the ICSID Convention, whose enforcement mechanism was supposed to be ‘self-contained’ and ‘automatic’. Ironically, however, the Convention itself introduces defences to recognition and enforcement by allowing states to apply their sovereign immunity laws, thereby failing to shelter awards from the scrutiny of national courts. It is thus safe to say that, as has been noted, ‘actions under the ICSID Convention are now operating on a similar timeline to enforcement actions under the New

80 Liber.

E. Timber Corp. v Liberia (LETCO), 650 F. Supp. 73, 7 (SDNY 1986).

81 ibid. 82 Liber. 83 See,

E. Timber Corp. v Liberia, 854 F.2d 1314 (2d Cir. 1987). eg, Trans Commodities, Inv. v Kazakhstan Trading House, S.A. Capital (SDNY 1997).

International Investment Law in US Courts

589

York Convention’, having the effect of ‘a diminution in value of an ICSID award over a nonICSID award’.84 The increasing scepticism of US courts towards investor-state dispute settlement may not be that surprising after all. The US has already shown its discontent with the system, by eliminating investor-state arbitration between it and Canada from the 2018 United StatesMexico-Canada Agreement. Time will tell whether the US Government adopts a similar approach with other investment treaties. In the meantime, when choosing the seat of the arbitration and where to enforce their awards, investors should ask themselves whether the US is still today the same supporter of investment arbitration as it was 50 years ago.

84 J Rotstein, ‘Is the D.D.C. Becoming a Specialized Enforcement Court?’ (Kluwer Arbitration Blog, 6 June 2019), https://perma.cc/XT2B-23G2.

34 Investment Cases in the Mexican Legal System: Willingness to Compensate, Federalism Issues, and Parallel Litigation GUILLERMO J GARCIA SANCHEZ*

T

HIS CHAPTER DESCRIBES the major decisions that defined the way the Mexican system interacts with international investment awards.1 It is one example of the way the investment regime interacts with domestic courts in Latin America. The case of Mexico, however, is exceptional in that it has been highly influenced by the 1994 North American Free Trade Agreement (NAFTA) cases and its unique ‘waiver’ to initiate domestic proceedings provision.2 Other domestic judiciaries in Latin America, such as Colombia, have had the opportunity to engage in meaningful discussions on the constitutionality of investment treaties.3 As will be explained below, the substantive discussion on the constitutionality of investment agreements or investment chapters in trade agreements has not been discussed in Mexico. To address the way the Mexican legal system engages with the investment regime the chapter reviews two points of contact. First, investment awards generated domestic legal challenges among different branches of government to allocate responsibility for harms stemming from the breach of international treaties. When arbitral tribunals issue awards involving State or municipal authorities that require Mexico to compensate foreign investors, the central government must establish which authority has to sign the check and disburse the payment. Such an inquiry in a federal system of government raises questions about the power of the central government to retain funds belonging to States or municipalities in order to compensate an investor for actions taken at the local level. Domestic courts in Mexico have had to resolve disputes surrounding the disbursement of payments and the retention of funds belonging to local authorities.

* Guillermo J. Garcia Sanchez is an Associate Professor at Texas A&M University School of Law. 1 The research, cases, and analysis in this chapter includes available information up to February 2020. It does not include any cases reported after that period. 2 As this chapter was being written the United States Mexico Canada Agreement has ratified. The author is aware that the NAFTA 1994 was substituted by the USMCA but no analysis of the treaty or any cases that might arise under its provisions are included in this chapter. 3 G Prieto, ‘The Colombian Constitutional Court Judgment C-252/19: A New Frontier for Reform in International Investment Law’ (EJIL Talk!, 29 July 2019), https://perma.cc/3JBW-SWL5.

592 Guillermo J Garcia Sanchez The second point of contact of local courts with the arbitral investment regime is when litigation is brought within domestic courts either before reaching the international fora or in tandem with the arbitral proceedings. Depending on the treaty requirement on the exhaustion of domestic remedies, local courts can serve either as gatekeepers or as the source of investment claims. Under NAFTA, investors may pursue domestic claims seeking other types of relief not connected to the payment of damages.4 As such, courts had to resolve claims based on Mexican law that have an impact on how investment tribunals resolve their cases. As will be explained below, this is not direct interaction among courts and tribunals, but rather a parallel relationship that under concrete circumstances influences each other. Finally, in one case, KBR v Mexico, a federal court’s decision to set aside a commercial arbitral award became the main ground in a NAFTA case. In other words, the federal judiciary became the primary actor accused of breaching international investment protection duties. The chapter is divided into two subsections that address these circumstances: the enforcement of the award at the domestic level and federalism questions surrounding the responsibility of State actors to compensate; and the parallel litigation of investors claims at the domestic level, including the cases where the judiciary becomes the main governmental actor in the claim. On both of these fronts, Mexico has valuable lessons to share from significant cases litigated against actions taken by the Mexican authorities. A note of caution is warranted. The focus of the chapter is to address the way the domestic legal system has dealt with cases and not on how investment tribunals have interpreted the domestic legal proceedings. As such, the latter will be described when warranted, but the chapter does not in any exhaustive way analyse the development of investment doctrine on the matter. Many of these issues are covered more broadly by the contributions made by other authors in this book.

I. DEBO NO NIEGO: MEXICAN WILLINGNESS TO COMPLY WITH AWARDS AND FEDERALISM QUESTIONS SURROUNDING THE PAYMENT OF DAMAGES

In Mexican folklore there is a phrase that describes the debtors’ willingness to recognise a debt but their incapacity to provide for payment: Debo no niego. Pago, no tengo (I don’t deny that I am in debt with you, but I don’t have the payment due to you). The phrase captures the attitudes of many local authorities towards foreign investors after losing the arbitral proceedings. However, as will be explained below, the federal government has always been willing to compensate, even if the case involves actions taken by other government actors. Up until February 2020, in all of the cases recorded by the Ministry of the Economy, Mexico honoured its duties to compensate foreign investors after losing an arbitral proceeding.5 Consequently, investors have not requested the judiciary to force the State to compensate them in accordance with an issued award. This is surprising considering the fact that, until 2018, Mexico had not ratified the International Centre for Settlement of Investment Disputes (ICSID) Convention and investors could only enforce the awards through the 1958 New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards.6 4 North

American Free Trade Agreement, US-Can-Mex, 17 December 1992, 32 ILM 605, 702. de Economía, ‘Comercio Exterior / Solución de Controversias’ (gob.mx), www.gob.mx/se/acciones-yprogramas/comercio-exterior-solucion-de-controversias. 6 ICSID, ‘Member States’, https://icsid.worldbank.org/en/Pages/about/Member-States.aspx; Convention on the Settlement of Investment Disputes between States and Nationals of Other States, Mar. 18, 1965, 4 I.L.M. 524; Convention on the Recognition and Enforcement of Foreign Arbitral Awards, June 10, 1958, 21 U.S.T. 2517, 330 U.N.T.S. 3. 5 Secretaría

Investment Cases in the Mexican Legal System 593 Since the signing of NAFTA, in almost every case, Mexico has compensated the investor after losing the arbitration proceeding without even seeking the annulment of the award at the seat of arbitration. In only three of the early NAFTA-related cases, did the Mexican authorities try to annul the award in the seat of arbitration (Canada) and, after losing the annulment, the government paid without the investors having to seek enforcement of the award in another jurisdiction. In the five remaining cases where Mexico lost, the government has always paid after the Tribunal announced its final decision. With the expectation of Cargill and CPI (the sweeteners cases), on average investors received payment for damages within the following nine months after the Tribunal issued the award or the Canadian court decided on the vacatur motion. Consequently, there are no major decisions by local judges that specifically deal with the enforcement of international investment decisions against the State. As of February 2020, investors have initiated 36 investment cases against the Mexican Government: 11 of those are pending, seven are in the early stage of notification, three discontinued, eight decided in favour of Mexico, one settled, and eight decided in favour of the investor.7 Below is the list of the concluded cases.

Authorities involved in the measure

Case

Measure litigated in domestic courts

Treaty

State found liable

Appeal of the award at seat of arbitration

Time elapsed between award or recognition of the award and payment

1. Metalcad

Municipal

yes

NAFTA

yes

yes

6 months

2. Robert Azinian et. Al

Municipal & Judiciary

yes

NAFTA

no

N/A

N/A

3. Marvin R. Feldman

Fed Executive

yes

NAFTA

yes

yes

6 months

4. Waste Management I

Municipal & Fed Judicial

yes

NAFTA

no

N/A

N/A

5. Waste Management II

Municipal & Fed Judicial

yes

NAFTA

no

N/A

N/A

6. Fireman ’s Fund

Fed Executive

yes

NAFTA

no

N/A

N/A

7. GAMI

Fed Executive

yes

NAFTA

no

N/A

N/A

8. Thunderbird

Fed Executive

yes

NAFTA

no

N/A

N/A

9. Corn Products Intl

Fed Legislative

yes

NAFTA

yes

no

20 months

10. ADM/TLIA Legislative

yes

NAFTA

yes

no

Data Not Available

11. Cargill

Legislative & Executive

yes

NAFTA

yes

yes

40 months

12. Bayview Et Al.

Executive

no

NAFTA

no

yes, by investors

N/A (continued)

7 Secretaría

de Economía, above (n 5).

594 Guillermo J Garcia Sanchez (Continued)

Case

Authorities involved in the measure

Measure litigated in domestic courts

Treaty

State found liable

Appeal of the award at seat of arbitration

Time elapsed between award or recognition of the award and payment

13. TECMED

Fed Executive, no State, Municipal

BIT Spain-mx8

yes

no

9 months

14. Gemplus

Fed Executive

yes

BIT yes Argentina-Mx

no

9 months

15. Abengoa

Fed Executive & Municipal

no

BIT Spain-Mx yes

no

10 months

16. KBR (‘Comissa’)

Fed Judiciary

yes

NAFTA

N/A

N/A

17. Telefonica

Fed Executive

no

BIT Spain-Mx settled N/A

N/A

no

The only case found in the local judicial docket that resolved questions involving the enforcement of an international investment decision in Mexico was Metalclad.9 Metalcald arose out of a violation of NAFTA’s Article 1105 by municipal authorities in the state of San Luis Potosi. Metalclad initially received approval by the federal government to build a hazardous waste matters facility in the municipality of Guadalzacar, but local state authorities cancelled the building permits due to social protests. The arbitral panel awarded Metalclad $16.5 million as compensation for the actions committed by the State and municipal authorities. As with other nations with a federal system of government, such as Mexico, the federal government is the official defendant in any cases, notwithstanding the fact that other authorities might have breached international obligations.10 Consequently, the Mexican central government, through the Ministry of Treasury, signed an agreement with Metalclad to pay the awarded damages. However, it also sent an official request to the state of San Luis Potosi for reimbursement of the payment made by the federal government.11 If the state denied the petition, the federal government would withhold federal funds from the fiscal coordination agreements signed with San Luis Potosi. The state of San Luis Potosi brought a constitutional challenge, controversia constitucional, against the Ministry of Treasury’s decision. The controversia constitucional proceeding allows different branches of government to challenge acts from other branches directly to the Supreme Court on the grounds that the challenged measure breaches the Constitution.12 The state

8 The Spain-Mexico BIT did contain a fork in the road provision. Investors could not bring claims internationally if they had initiated domestic proceedings. 9 Metalclad Corp v United Mexican States (30 August 2000) Case No ARB(AF)/97/1. 10 GJ Garcia Sanchez, ‘To Speak with One Voice: The Political Effects of Centralizing the Legal Defense of the State’ (2017) 34 Arizona Journal of International & Comparative Law 557. 11 T Gutiérrez-Haces, ‘Entre la Protección a la Inversión Extranjera Directa y la Disciplina sobre los Gobiernos’ in M Gambrill (ed), Diez años del TLCAN en México (Mexico, Universidad Nacional Autónoma de México, 2006) 411–12. 12 Sentencia relativa a la controversia constitucional 32/2002, promovida por la Camara de Diputados del Congreso de la Union en contra del Titular del poder Ejecutivo Federal, 17 de julio de 2002 [hereinafter Supreme Court Suspension Decision].

Investment Cases in the Mexican Legal System 595 argued that the Ministry of Treasury lacked the powers to use funds that belonged to the local authorities to compensate the investors.13 Both the Constitution and the federal fiscal coordination agreements were silent on the issue, and as such the Court would have to interpret the law in a restrictive way: only those powers recognised in the Constitution belong to the federal government.14 The Ministry of Treasury argued that the actions that affected the investors concerned a local authority and that the federal government’s only involvement was to defend the case in the international fora. There was no reason to assume that the federal government had to pay for the actions of another authority. By withholding federal funds, the Ministry of Treasury was trying to make those officials who had made the original decision accountable. In 2004, however, the Supreme Court of Mexico decided that the Ministry of Treasury lacked the powers to use funds that belong to local authorities to pay for awards in which they were involved; such action would violate the federal fiscal pact of the Mexican Constitution.15 Since then, the federal government has had to move funds within the budget approved by Congress to pay for all the pending investment cases that could potentially end in damages, regardless of the level of government where the actions originated.

II. INTERACTIONS BETWEEN THE MEXICAN JUDICIARY AND INTERNATIONAL INVESTMENT TRIBUNALS: PARALLEL LITIGATION AND THE JUDICIARY AS THE SOURCE OF THE BREACH OF INTERNATIONAL DUTIES

As stated above, there are no cases where the Mexican judiciary explicitly addressed whether investment arbitral awards or investment treaties are unconstitutional, nor whether they could be enforced in the domestic system.16 However, the judiciary has on more than one occasion dealt with NAFTA related disputes against the Mexican Government brought by investors through the amparo proceeding which eventually resulted in investment claims. That is, cases that involve an administrative action that affects companies that are controlled by a foreign investor, and through domestic judicial proceedings, the companies concerned seek relief. The claims are mainly brought arguing a violation of domestic law, but they arise out of the same governmental action that also gives rise to the treaty breach argued in international fora. The amparo proceeding in the Mexican legal system allows petitioners to bring claims to federal courts for violations of constitutional rights and international treaties.17 In the process, the petitioner can request provisional remedies, including the suspension of the governmental action or enforcement of the alleged unconstitutional law.18 The Mexican Supreme Court can exercise jurisdiction over the claim if it has a salient constitutional question or if there are contradictory constitutional interpretations among circuit courts. Under NAFTA’s Chapter 11, investors are forced to waive their right to initiate or continue domestic judicial proceedings concerning the measure involving the alleged breach of the treaty’s investor protection provisions.19 As such, this is not the traditional exhaustion of domestic

13 ibid. 14 Gutiérrez-Haces,

above (n 12) 411–12.

15 ibid. 16 As

mentioned above, the case of Colombia departs in this sense. See Prieto, above (n 3). Avalos, The Mexican Legal System : A Comprehensive Research Guide (Getzville, William S Hein & Co, 2013). 18 ibid. 19 A Bjorklund, ‘Waiver and the Exhaustion of Local Remedies Rule in NAFTA Jurisprudence’ in T Weiler (ed), NAFTA Investment Law and Arbitration: Past Issues, Current Practice, Future Prospects (Leiden, Brill, 2004). 17 F

596 Guillermo J Garcia Sanchez remedies precondition for bringing international claims against the State.20 The jurisdictional ‘waiver’ in Article 1121 of NAFTA forces investors to discontinue proceedings that had been initiated in federal tribunals in order for the investors to bring the claims to an international forum.21 The only exception are ‘proceedings for injunctive, declaratory or other extraordinary relief, not involving the payment of damages, before an administrative tribunal or court under the law of the disputing Party’.22 In sum, the 1994 NAFTA did not require investors to exhaust local remedies first, and in certain circumstances, when the investors seek relief not involving the payment of damages, the affected parties could bring simultaneous or subsequent claims in administrative or federal tribunals. Since 77 per cent of the concluded cases involving Mexico are NAFTA-related cases, the unique ‘waiver’ has impacted the way local judiciaries interact with investor-related claims and how investment tribunals interpret domestic proceedings as a prerequisite for international proceedings. The most notorious cases that reached the Supreme Court of Mexico were those involving the sweeteners industry (CPI, ADM/TLIA, and Cargill).23 These cases stand aside from the rest in that they were litigated in domestic proceedings to the highest court, in investment tribunals, and in international trade panels.24 The disputes arose out of the Presidential Decree issued in 2001 by Vicente Fox, then President of Mexico, which expropriated 27 sugar mills. The action was motivated by efforts to protect the Mexican sugar industry from unfair international competition and avoid the sector’s collapse.25 Four months later, the Mexican Congress approved a tax on the use of sweeteners (HFCS) in soft drinks but exempted those that used Mexican sugar cane. The tax discriminated against HFCS producers and distributors, who happened to be almost exclusively US investors.26

A. The Influence of NAFTA in the Constitutional Interpretation of Expropriation: The GAMI Case The sweeteners industry case reached the Supreme Court of Mexico through the amparo proceeding brought by the sugar mill owners. In their petition to the highest court of Mexico, the mill owners alleged that the government had violated Article 27 of the Mexican Constitution that regulated the protection of private property and the procedure for expropriations. In sum, the petitioners alleged that Article 27 had to be interpreted in conjunction with the Mexican Government’s duties under international law. As far as the petitioners were concerned, both Article 27 and the treaties required the government to grant the mill owners a hearing prior to the expropriation act. The Supreme Court of Mexico agreed with the petitioners, invalidated the decree, and restored the property to the mill owners.

20 ibid. 21 ibid. 22 See

NAFTA, Art 1121, subs (1)(b) and (2)(b). Products International, Inc v United Mexican States, ICSID Case No ARB(AF)/04/1, www.gob.mx/cms/ uploads/attachment/file/121777/Ficha_tecnica_Corn_Products_International.pdf; Archer Daniels Midland Company and Tate& Lyle Ingredients Americas, Inc v United Mexican States, ICSID Case No Arb(AF)/04/5), www.gob.mx/cms/ uploads/attachment/file/41974/Ficha_tecnica_Archer_Daniels_Midland_Co.pdf; Cargill, Inc v united Mexican States (2005) ICSID Case No ARB(AF)/05/2, www.gob.mx/cms/uploads/attachment/file/114770/Cargill_Inc._URL_2.pdf. 24 S Puig, ‘Investor-State Tribunals and Constitutional Courts: The Mexican Sweeteners Saga’ (2013) 5 Mexican Law Review 199. 25 ibid 224. 26 ibid. 23 Corn

Investment Cases in the Mexican Legal System 597 Notwithstanding the fact that the Constitution only requires the government to exercise expropriation acts for public order purposes and subject to adequate compensation, the international treaties, in the Court’s interpretation, also required the government to provide investors with a prior hearing where they could challenge the rationale behind the expropriation. In its decision, the Court examined the texts of NAFTA and the free trade agreements with Japan and Argentina. In the Supreme Court’s view, the investment treaties awarded foreign investors rights to a prior hearing as part of the ‘due process’ of expropriation rights.27 As such, the Court concluded that if under the treaties foreigners were granted the right to a prior hearing, there was no reason to deny the same right to Mexican nationals.28 Any interpretation to the contrary would be discriminatory to Mexican nationals. The Supreme Court’s decision modified its jurisprudence to constitutionalise the right to a prior hearing when the State decides to exercise its power of eminent domain.29 The decision did not, however, discuss the constitutionality of the international treaties – instead it gave a ‘systemic’ interpretation of the treaties and the Constitution. That is, in construing both sources, it privileged the interpretation that avoided any contradiction between the legal sources. In a certain sense, the Court constitutionalised NAFTA as a source to interpret rights contained in the Constitution.30 Before this decision, the Supreme Court’s jurisprudence only allowed petitioners to challenge the amount of compensation to be paid to the affected parties, but not the decision to expropriate. With the inclusion of the prior hearing, the new interpretation opens the way to challenge the grounds on which the government bases its decision to expropriate an asset.31 The Supreme Court’s decision impacted the proceedings brought by the investors to the international fora.32 The US-incorporated GAMI Investments Inc. (GAMI) was a minority shareholder of GAM, one of the sugar mill owners who benefited from the domestic proceedings in the Mexican Supreme Court. As the case was being litigated domestically, GAMI brought a claim under NAFTA against the Mexican Government.33 The US based company alleged breaches of the minimum standard of treatment (Article 1105), the expropriation requirements in NAFTA (Article 1110), and breach of the national treatment standard (Article 1102).34 GAMI contended that Mexico’s decree indirectly expropriated its share value in GAM.35 Moreover, it argued that the sugar regime that benefited the Mexican sector treated GAMI and its investments in a less favourable way, and its arbitrary implementation breached the minimum standard of treatment of aliens.36 For the NAFTA breaches, GAMI requested US$42 million as monetary compensation, including applicable interest, fees, and expenses.37 When Mexican courts announced their decision rendering the expropriation decree unconstitutional, the arbitral tribunal had to decide how the restoration of the three mills owned

27 C Mayer-Sierra and LM Pérez de Acha, ‘¿Un Nuevo Derecho o El Debilitamiento Del Estado? Garantía de Audiencia Previsa En La Expropiación’ (2009) 21 Cuestiones Contitucionales Revista Mexicana de Derecho Constitucional 99, https://perma.cc/76CE-RBFQ. 28 ibid. 29 ibid. 30 ibid. 31 ibid. 32 GAMI Investments, Inc v Mexico, Final Award, Ad Hoc-UNCITRAL Arbitration Rules; IIC 109 (2004), signed 15 November 2004 [hereinafter GAMI Award]. 33 GAMI Investments, Inc v United Mexican States, UNCITRAL Case (2002), Memorial, www.gob.mx/cms/ uploads/attachment/file/1640/Escrito_de_demanda_GAMI.pdf [hereinafter GAMI Memorial]. 34 ibid para 7. 35 ibid paras 132–46. 36 ibid paras 74–107, 108–31. 37 ibid paras 149–50.

598 Guillermo J Garcia Sanchez by GAM impacted GAMI’s indirect expropriation claims. The Mexican Government tried to move to dismiss the whole NAFTA proceedings but failed. The Tribunal ultimately acknowledged GAMI’s right to bring independent NAFTA claims to determine whether the actions by the Mexican Government affected the investments in a way that was ‘tantamount to expropriation’.38 However, the fact that the decree was inconsistent with the ‘norms of NAFTA’ was insufficient to consider it a treaty breach. The government’s act would have to ‘affect interests protected by NAFTA’.39 Consequently, in order for the government’s action to be considered a breach of international obligations, it would have to impact the value of GAMI’s shareholding to such an extent that it would be tantamount to expropriation.40 Since the mills were reinstated to the investors and the claimants did not provide any additional evidence on the effects that the measure had on the value of GAMI’s shares in GAM, the Tribunal dismissed the expropriation claims.41 The Tribunal did mention the difficulty in assessing the case by having two simultaneous proceedings dealing with the same acts of the State. It noted that ‘[t]he overwhelming implausibility of a simultaneous resolution of the problem by national and international jurisdictions impels consideration of the practically certain scenario of unsynchronized resolution’.42 The Tribunal also dismissed the claims based on national treatment and minimum standard brought against Mexico.

B. The Tax Law as a Breach of NAFTA’s Chapter 11 Another group of investors, mainly Corn Products International, brought claims against the government’s tax regime both in the domestic legal system and in international tribunals. As mentioned above, the Mexican Congress approved a tax on the soft drink producers/distributors that use HFCS. The soft drink producers initiated an amparo proceeding in Mexican federal courts against the tax. They based their claims solely on Mexican law, alleging that the tax was discriminatory and violated the principles of tax equality and proportionality contained in the fiscal code. Moreover, they argued that the tax had monopolistic effects in favour of the national sugar industry and, as such, were also a violation of antitrust law. The case increased its complexity when President Fox suspended the tax in an effort to prevent damage to what he considered an important economic sector. In response, the Chamber of Deputies filed a controversia consitucional before the Supreme Court of Justice. The Chamber of Deputies believed that the executive had exceeded its powers by basing its decision on a rarely employed exception in the Tax code. As a remedy, the Chamber requested the re-establishment of the tax. The HFCS producers were unsuccessful in their amparo proceedings. Since the tax was imposed on the producers/distributors of soft drinks and not on the producers of HFCS, the justices of the Supreme Court dismissed their claims. Under Mexican law at that time, for an amparo proceeding against a tax regime to succeed, the claimant must be directly affected by the tax. The claimants must be those that bear the main economic burden of the unconstitutional fiscal measure. To the Supreme Court, the producers of HFCS, such as CPI, were not directly impacted by the tax and consequently had no standing in the amparo proceeding. 38 GAMI

Award, above (n 33) para 123. para 129. 40 ibid paras 128–33. 41 ibid. 42 ibid para 119 (emphasis in original). 39 ibid

Investment Cases in the Mexican Legal System 599 The producers/distributors of soft drinks also lost the legal challenge because the Supreme Court considered that the tax established different standards of treatment but it was done for valid legal reasons. However, the Chamber of Deputies won its controversia constitutional claim. The Court ruled that the President’s decision ignored the valid non-fiscal related purposes of the tax.43 As such, it had exceeded its constitutional powers by disregarding the extra-fiscal objective of protecting the domestic sugar industry.44 In addition to CPI, three other HFCS producers, Archer Daniels Midland Company (ADM), Tate & Lyle Ingredients Americas, Inc (TLIA), and Cargill, Inc., brought claims against Mexico under NAFTA’s Chapter 11.45 The Mexican Government tried to consolidate the proceedings but the efforts failed. All of the companies argued that the tax regime breached the national treatment, performance requirements, and expropriation provisions of NAFTA. Cargill also claimed that the measures violated the most-favoured-nation obligation and the minimum standard of treatment of aliens. The sum of damages requested amounted to US$575 million. The Mexican defence to the tax-related claims was based on international trade law: the tax regime was a legitimate countermeasure in response to US violations of NAFTA. The tax was intended to be a temporary and proportionate measure adopted in reaction to a pending trade dispute between the US and Mexico involving Chapter 3 and NAFTA’s side-letters on sugar.46 It is beyond the scope of this chapter to explain in detail how the investment tribunals analysed the questions on countermeasures as a defence in investment claims, but suffice it to say that the Tribunals found them uncompelling.47 In contrast with the domestic legal proceedings, the Tribunals did find that the tax regime was discriminatory and a violation of Article 1102. There was no question, as the government recognised, that the discrimination was based on nationality. The government intended to pressure North American companies as a countermeasure for policies adopted by the US Government. In other words, the countermeasure defence was a recognition of a discriminatory policy. The Tribunal in ADM/TLIA even analysed the Congressional debates preceding the implementation of the tax and recognised that the regime imposed was ‘afford[ing] protection to the production of cane sugar, which is in line with [other] measures taken by Mexico before the imposition of the Tax’.48 Chapter 11 protections on national treatment were in place precisely to avoid discrimination based on the investors’ nationality.

C. The Mexican Judiciary’s Acts as the Main Grounds for a NAFTA Claim: The KBR (Commisa) Case The case of Commisa departs from the previously analysed cases in that the main action alleged as a breach of international investment duties was a decision of the domestic judiciary

43 ibid

para 100.

44 ibid. 45 Archer

Daniels Midland Company and Tate& Lyle Ingredients Americas, Inc v United Mexican States, ICSID Case No ARB(AF)/04/5 [hereinafter ADM/TLIA]; Cargill, Inc v United Mexican States (ISCIS Case No ARB(AF)/05/2) [hereinafter Cargill Memorial]. 46 Archer Daniels midland Company v United Mexican States, Final Award (21 November 2007) [hereinafter ADM/TLIA Final Award] para 106. 47 S Puig, ‘Investor-State Tribunals and Constitutional Courts: The Mexican Sweeteners Saga’ (2012) 5(2) Mexican Law Review 199. The US also brought claims against Mexico for the action in the WTO and Mexico lost the case, www.wto.org/english/tratop_e/dispu_e/cases_e/ds308_e.htm. 48 ADM/TLIA Final Award, para 212.

600 Guillermo J Garcia Sanchez to annul a commercial arbitration award. Hence, it serves as an example of a case that deals with relations between the investment regime, international commercial arbitration and the national judiciary. In 1997 an oil and gas service provider, Commisa, signed a contract with PEMEX, Mexico’s State-owned company, to build and install two offshore natural gas platforms off the Gulf of Mexico.49 Commisa is the Mexican subsidiary of the US based company KBR. After signing the contract, for two years Commisa and PEMEX had difficulties in complying with the timelines specified in the contract.50 Both accused each other of breaching contractual obligations that prevented them from executing the works. In 2004 Commisa notified the State-owned company that it would not continue working on the project and PEMEX decided to activate its right under the contract and Mexican legislation to administratively rescind the contract.51 The contract was regulated by Mexican administrative law and specified that PEMEX could rescind the contract if the private company failed to comply with its duties. The contract also contained an arbitration clause under the rules of the International Chamber of Commerce (ICC) with Mexico as the seat of arbitration.52 Under NAFTA, any contract signed with Mexican Stateowned entities involved in the energy sector were not considered investments covered under Chapter 11. Under the 1994 NAFTA, the Mexican energy sector, including the production, extraction, and distribution of hydrocarbons, was considered a national strategic sector, which under the Constitution did not allow any foreign direct investment.53 After PEMEX rescinded the contract with Commisa, in December 2004, the company initiated arbitral proceedings in Mexico. Before the ICC arbitral tribunal issued its final decision, the Mexican Congress passed two laws that affected the merits of the dispute. The regulations provided for the exclusive jurisdiction of the Mexican administrative courts over all disputes related to public contracts and excluded from arbitration any dispute arising out of the State-owned companies’ right to exercise ‘administrative rescission’ over service contracts with private parties. Shortly after the law passed, the ICC arbitral panel issued an award in favour of Commisa for nearly US$300 million. With the new legislation in place, PEMEX sought to vacate the ICC award in Mexico. In September 2011, the Tenth Mexican federal circuit court ruled that Commisa’s claims were not arbitrable under the new legislation and, as such, vacated the award.54 KBR initiated a parallel proceeding. It sought to enforce the arbitral award in Luxemburg and US courts, notwithstanding the fact that it had been vacated by the local court of the arbitral seat, and initiated a NAFTA proceeding against Mexico.55 The proceeding in the US had several twists, including the ‘resurrection’ and granting of the enforcement of the award. Due to limited space, this chapter will not cover the legal implications of the decisions in US courts. What is essential for the purposes of the interaction of the investment regime and the Mexican system is to review how the NAFTA tribunal interpreted the parallel proceeding initiated by KBR in US courts. 49 KBR, Inc y Corporacion Mexicana de Mantenimiento Integral v los Estados Unidos Mexicanos, Ficha Técnica, Secretaría de Economía, available at ‘KBR, Inc y Corporacion Mexicana de Mantenimiento Integral v Estados Unidos Mexicanos’, www.gob.mx/cms/uploads/attachment/file/30095/Ficha_tecnica_KBR.pdf. 50 KBR, Inc. y Corporacion Mexicana de Mantenimiento Integral v. Estados Unidos Mexicanos, Jurisdictional Objection Brief, ICSID Case No UNCT/14/1, 9–10, www.gob.mx/cms/uploads/attachment/file/1443/7_mexico_ primer_escrito_sobre_cuestion_preliminar_KBR.pdf [hereinafter KBR Jurisdictional Objection brief]. 51 Articulo 40 de la Ley de Adquisiciones y Obras Publicas (Diario Oficial de la Federacion (DOF) 30/12/1993). 52 KBR, Inc y Corporacion Mexicana de Mantenimiento Integral v Estados Unidos Mexicanos, above (n 50). 53 GJ Garcia Sanchez, ‘The Mexican Petroleum License of 2013’ in T Soliman Hunter et al (eds), The Character of Petroleum Licenses (Edward Elgar, 2020) 207–233. 54 Amparo en Revisión 358/2010 (25 de agosto de 2011); see also KBR Jurisdictional Objection Brief para 16. 55 KBR, Inc y Corporacion Mexicana de Mantenimiento Integral v Estados Unidos Mexicanos, above (n 50).

Investment Cases in the Mexican Legal System 601 In its NAFTA arbitration proceeding, KBR alleged that the annulment decision by Mexican courts breached Chapter 11’s provisions on national treatment, most favoured nation, the minimum standard of treatment, and expropriation.56 KBR requested US$400 million in compensation.57 Mexico’s defence to KBR’s claims rested on the fact that KBR failed to waive the domestic proceedings it had initiated in the US and Luxembourg as required by Article 1121 of NAFTA. On 30 April 2015 the Tribunal agreed with Mexico: KBR and COMMISA’s claims requesting the enforcement of the award in the US and Luxembourg courts implied a claim involving monetary compensations. As mentioned above, Article 1121 only allows parallel litigation at the domestic level when the relief sought is different from monetary compensation.58

III. CONCLUSION

The major decisions in the Mexican legal system involving investment law have not led to a direct confrontation with arbitral tribunals. Rather, the interaction between the two systems can be characterised as a synchronised effort to address investors complaints. As explained in previous paragraphs, one of the main facts allowing this relationship to continue without tension, is the fact that Mexico has been a good faith participant of the investment regime. Of the 17 concluded cases studied for this chapter, Mexico was found liable in only eight of them and has paid compensation in all of them without the need for investors to seek enforcement of the award. This willingness to pay, reduces the opportunities of having the domestic judiciary decide issues on the constitutionality of investment treaties, of the investment award, and the process of enforcement of arbitration related-duties. The only case in which the domestic courts had to decide challenges involving the payment of damages to an investor, was brought before the courts as a challenge to the federal government’s power to withhold funds from local authorities in order to compensate investors. In other words, it was not a case that challenged the enforceability of investment obligations at the domestic level, rather a question of allocating responsibility among branches of government to compensate the foreign investor. Other cases worth mentioning for the purpose of this chapter were those connected to domestic procedures that tangentially affected the investment proceedings. These cases involved investors that brought domestic legal claims through amparo proceedings, and they impacted the way international arbitral tribunals interpreted the ‘waiver’ to initiate domestic proceedings in 1994 NAFTA’s Chapter 11 provisions. This chapter also described a case in which the judiciary became the main actor accused of breaching international legal obligations. In the case of KBR the judiciary had vacated an ICC award, and served as the ground for an investment claim in a NAFTA tribunal. Ultimately, the investor’s claim failed and the case did not set a precedent at the domestic level that could impact the future interaction between the Mexican legal system and the investment regime. As a concluding remark, it is worth noting that the tendency in Mexico is to expand the rights of corporations and equate them to the human rights recognised in international treaties.59 The 2009 reform that constitutionalised all the ‘rights’ recognised in international agreements

56 ibid. 57 ibid. 58 ibid. 59 ‘Las empresas y sus derechos humanos según la Corte’ (El Juego de la Suprema Corte), https://perma.cc/ H9EQ-Z4JB.

602 Guillermo J Garcia Sanchez as constitutional rights has devolved into an effort from corporations to use all recognised rights in treaties as also being applicable to them.60 The latest examples include cases where the Mexican Supreme Court recognised the right to protect honour and the right to privacy as applicable to corporations.61 The claimants successfully argued that the Constitution does not distinguish between legal person and physical persons, and as such, all rights recognised in treaties should also apply to corporations.62 The conclusion of this precedent leaves the way open for arguing that rights contained in international investment treaties have the same constitutional hierarchy as internationally recognised human rights. One could argue that in Mexico, foreign corporations have the same constitutional protections as individuals when it comes to actions taken by State authorities.

60 ibid. 61 ibid; Derechos Fundamentales. Su Vigencia en las Relaciones Entre Particulares. Principio Pro Persona. Criterio de Selección de la Norma de Derecho Fundamental Aplicable. Amparo Directo 8/2012. Arrendadora Ocean Mexicana, SA de CV y otros. 4 de julio de 2012. Mayoría de Cuatro Votos. Disidente: Guillermo I Ortiz Mayagoitia Ponente: Arturo Zaldívar Lelo de Larrea. Novena Época. Seminario Judicial de la Federación y su Gaceta No. 23866, Primera Sala, Libro XIII, Octubre 2012 Tomo 2, página 732; See also ‘Derecho Fundametnal al honor de las personas juridicas’ Tesis aislada 1a. XXI/2011 (10a.), registro de IUS 2000082, publicada en el Semanario Judicial de la Federación y su Gaceta, Décima Época, Libro IV, enero de 2012, Tomo 3, página 2905; Véase las discusiones en el Pleno de la SCJN en el asunto 56/2011 Contradiccion de tesis suscitada entre la Primera y la Segunda Salas. Discutido los días 23, 27 y 30 de mayo de 2013. 62 ibid.

35 International Investment Law in African Courts MAKANE MOÏSE MBENGUE AND STEFANIE SCHACHERER

I. INTRODUCTION

I

N AN ARTICLE published in 1989, Augustus Agyemang, a practising barrister in Ghana, affirmed that ‘for a number of reasons African courts are unsuitable for settling investment disputes and, therefore, the role of African courts in this area should, as far as possible, be minimised’.1 His main arguments were the absence of strong traditions of judicial independence in African states and the fact that foreign investment in Africa mainly involves the exploitation of natural resources, which would jeopardise the objectivity of national courts because of the very high national interests that are at stake.2 Mr Agyemang’s point of view obviously reflects the prevailing – rather Eurocentric – narrative on the advantage of international investment arbitration in bypassing the domestic courts of developing countries, including African courts.3 In the last 30 years, the great majority of disputes between a foreign investor and an African state have indeed been settled by international arbitration, either on the basis of investment contracts or bilateral investment treaties (BITs). And one should not forget that African states were key drivers for the creation and development of the International Centre for Settlement of Investment Disputes (ICSID) dispute resolution system.4 Therefore, analysing international investment law in African courts comes with the caveat that judicial decisions on foreign investment matters are rare. It is also important to mention the practical difficulties of conducting research on this topic. As Africa has over 50 different

* Makane Moïse Mbengue is Professor of Public International Law at the University of Geneva and Affiliated Professor at Science Po Paris (School of Law); Stefanie Schacherer is a Postdoctoral Fellow at the National University of Singapore, Centre for International Law. 1 AA Agyemang, ‘African Courts, the Settlement of Investment Disputes and the Enforcement of Awards’ (1989) 33 Journal of African Law 31, 33. 2 ibid. See also B-O Bryde, The Politics and Sociology of African Legal Development (Hamburg, Metzner, 1976) 67. 3 C Schreuer, ‘Interaction of International Tribunals and Domestic Courts in Investment Law’ in AW Rovine (ed), Contemporary Issues in International Arbitration and Mediation: The Fordham Papers (2010) (Leiden, Brill | Nijhoff, 2011) 71. 4 P-J Le Cannu, ‘Foundation and Innovation: The Participation of African States in the ICSID Dispute Resolution System’ (2018) 33 ICSID Review – Foreign Investment Law Journal 456, 457. MM Mbengue, ‘Africa’s Voice in the Formation, Shaping and Redesign of International Investment Law’, (2019) 34 ICSID Review – Foreign Investment Law Journal, 455–481.

604 Makane Moïse Mbengue and Stefanie Schacherer jurisdictions, the selection of cases dealing with investment disputes is to some extent a subjective undertaking. The selection is also not the result of a systematic research of all African courts and tribunals because gaining access to court decisions is for most African jurisdictions extremely difficult if not impossible since there are in general no accessible online databases. Against this backdrop, the methodology of this chapter has been guided less by finding major decisions and more by trying to identify major categories of situations, events and institutions in which or through which questions of international investment law have been dealt with in the African context. The chapter is thus selective in nature and aims at showing tendencies instead of exhaustivity. A noteworthy tendency that becomes apparent through the chapter’s analysis is that the prevailing narrative of bypassing African courts and tribunals seems to be declining in importance as an increasing number of national investment law instruments foresee the use of African judicial and arbitral institutions for the settlement of investment disputes.

II. AFRICAN NATIONAL COURTS

Section II dealing with African national courts, first looks at major interactions, such as the situation in which international investment tribunals review prior African national proceedings (section II.A.) and second, where African national courts made attempts to interfere in international arbitral proceedings (section II.B). This part also considers major historical events leading to national court decisions, such as the judicial order to allow the seizure of the Argentinean warship Ara Libertad by Ghanaian authorities (section II.C.), judgments of the Zimbabwean constitutional court enforcing the country’s highly controversial land reform (section II.D.) and lastly, judgments of the South African constitutional court interpreting the constitutional guarantee against expropriation in light of the country’s apartheid history (section II.E).

A. Prior Proceedings in African Courts and Subsequent Review Thereof by International Investment Tribunals A typical form of interaction between international arbitral tribunals and domestic courts in investment law occurs when the national court proceedings become subject to review in investment arbitration.5 In the arbitral proceedings opposing the investors Jan de Nul and Dredging International to Egypt, a judgment of an Egyptian lower court in the district was under scrutiny.6 The dispute arose out of a misunderstanding about the terms and conditions of a contract on dredging activities in the Suez Canal. The investor alleged that the Suez Canal Authority (SCA) had misrepresented the size of the tasks. The investor contended having to encounter

5 Schreuer, above (n 3). There are in fact a number of such cases against Egypt, see Middle East Cement Shipping v Egypt, ICSID Case No ARB/99/; Siag v Egypt, ICSID Case No ARB/05/15; H&H Enterprises v Egypt, ICSID Case No ARB 09/15. For more details on these cases, see H Wehland, ‘Domestic Courts and Investment Treaty Tribunals: The Effect of Local Recourse Against Administrative Measures on the Breach of Investment Protection Standards’ [2019] Journal of International Arbitration 207. 6 Jan de Nul NV and Dredging International NV v Arab Republic of Egypt, Award (6 November 2008) ICSID Case No ARB/04/13, paras 191, 255–261.

International Investment Law in African Courts

605

conditions during the performance of the dredging works that were not mentioned at the tender stage, namely a lesser volume to be dredged, an imbalance between the deepening and widening operations, and a higher proportion of rock.7 Therefore, the claimant requested compensation of additional costs, which was denied by SCA and thus led to two contract claims before Egyptian national courts. The first was an action to declare the dredging contract null and void due to error and fraud. With the second action, the investors sought relief for a series of deductions made by SCA from the amounts paid under the dredging contract.8 The Administrative Court of Ismaïlia rendered the decisions for the two contractual disputes.9 In substance on the first contractual case, the Court declined to annul the contract for fraudulent misrepresentation or error and dismissed the claim for extra compensation because the investors ‘had failed to make the necessary investigations and had undertaken to perform at the price agreed regardless of the dredging conditions’.10 Concerning the second contractual case, the Court awarded the investors US$1,087,997.64 and LE216,045, which constituted around one-third of the deductions claimed.11 Disagreeing with the outcome of the case, Jan de Nul appealed the judgment on both cases before the High Administrative Court of Egypt. SCA also appealed, but only against the second contractual claim for undue deductions. In addition, the investors submitted a bilateral investment treaty (BIT) claim to the ICSID.12 Throughout the arbitral proceeding, the national appeal proceedings remained pending.13 Before the investment tribunal, the investor claimed, relying on the fair and equitable treatment standard, that a denial of justice had occurred through the ‘unfair’ decision taken by the Administrative Court of Ismaïlia.14 The claimant’s concern with the Egyptian judgment was its outcome, which did not recognise the fraudulent misrepresentation in the contract as SCA retained technical information about the size of the project. The arbitral tribunal did not follow the arguments of the claimant. First, the Tribunal underlined that the investor did not ‘complain of the failure of the Egyptian legal system as such, but merely of the conduct of the Ismaïlia Court […]. This is not sufficient to justify a claim for denial of justice, let it be through the fair and equitable claim.’15 Moreover, the appellate proceedings were ongoing and did not appear to be ‘in any manner dysfunctional’.16 As a result, the Tribunal explicitly denied that ‘an unjust judgement of a lower court may per se constitute unfair and inequitable treatment and, therefore, denial of justice’.17 Finally, all claims based on the BIT were dismissed by the ICSID tribunal. According to available information, the investor has dropped both actions for appeal before the Egyptian national courts.18 The appeal initiated by the SCA was rendered on 24 January 2017. The High Administrative Court of Egypt nullified the 2013 decision of the Administrative Court of Ismaïlia, which thus resulted in the nullification of any compensation for the investor for the alleged undue deductions made by the SCA.19

7 ibid

para 76. para 81. 9 ibid para 103. 10 ibid para 104. 11 ibid para 106. 12 ibid para 107. 13 ibid para 260. 14 ibid para 259. 15 ibid para 260; see also paras 258–259. 16 ibid para 260. 17 ibid para 259. 18 Probably as a consequence of the unsuccessful ICSID claim. 19 Egyptian High Administrative Court, Appeal No 11120/49 and 12400/49, Judgment of 24 January 2017. 8 ibid

606 Makane Moïse Mbengue and Stefanie Schacherer The dispute between Jan de Nul and Egypt shows that investment tribunals are not easily satisfied with a denial of justice claim made by an investor. To the contrary, the case underlines that investors are to make a reasonable attempt in domestic courts to obtain redress before a claim for violation of the international protection standard can be filed against the host country. The Jan de Nul tribunal thus reserved the role of the Egyptian national courts to decide upon the contract claim that was submitted to them.

B. African National Courts Interfering in Investment Arbitration It is not uncommon in international investment arbitration for a respondent to challenge the jurisdiction of a tribunal through action in domestic courts. In the African context, especially for the early ICSID cases, international arbitral tribunals had to assert on several occasions their jurisdiction in the face of attempts by the respondent state to seize national courts.20 Occasionally, this has led national courts to order temporary injunctions to suspend the proceedings before arbitral tribunals.21 A recent dispute opposing a US investor against Ghana is illustrative of this type of interaction between national courts and international investment law. Right after the start of the arbitral proceeding of Balkan Energy v Ghana,22 the respondent filed an interlocutory injunction against the arbitration, which was granted by the High Court of Justice in Accra.23 The injunction restrained the investor from, inter alia, taking any further steps in the arbitration proceedings. A few weeks later, the Ghanaian High Court confirmed the injunction.24 The central argument of the respondent, in its suit before the High Court, was that the power purchase agreement (PPA) between the investor and Ghana, including its arbitration clause, was invalid because the national Parliament did not approve the deal, which would have been necessary according to the country’s Constitution.25 In a next action, the respondent made an expedited reference to the Supreme Court of Ghana concerning the constitutional validity of the PPA.26 In its judgment, the Supreme Court had to determine whether the PPA fell under Article 181 of the country’s Constitution, which requires approval by the national Parliament to authorise the government to enter into an ‘international business or economic transaction to which the Government is a party’.27 20 Holiday Inns v Morocco, see P Lalive, ‘The First ‘World Bank’ Arbitration (Holiday Inns v Morocco) – Some Legal Problems’ (1980) 51 British Yearbook of International Law 123, 160; See also Benvenuti & Bonfant v Congo, Award (15 August 1980) 1 ICSID Reports 335, paras 1.12–1.14; LETCO v Liberia, Award (31 March 1986) 2 ICSID Reports 356, 378; MINE v Guinea, Award (6 January 1988) 4 ICSID Reports 69. 21 See Salini Costruttori v Federal Democratic Republic of Ethiopia, Award regarding the Suspension of the Proceedings (7 December 2001) ICC Arbitration No 10623/AER/ACS, paras 77–78. In this case, the Tribunal issued provisional measures enjoining parties from pursuing related claims in domestic courts. See for a more recent case, Salini Costruttori v Federal Democratic Republic of Ethiopia, Award regarding the Suspension of the Proceedings (7 December 2001) ICC Arbitration No 10623/AER/ACS, para 60. 22 Balkan Energy Ltd v the Republic of Ghana, PCA Case No 2010-7, Award on the Merits, 1 April 2014. 23 ibid, para 327. See also Order for Interlocutory Injunction (25 June 2010) Ghana High Court of Justice (Commercial Division). 24 Ruling, 6 Sep 2010, Ghana High Court of Justice (Commercial Division). 25 Balkan Energy Ltd v the Republic of Ghana, above (n 22) para 327. 26 ibid para 328, making reference to Supreme Court of Justice, Ghana, Rulings, 2 Nov 2011 and 16 May 2012. 27 Paragraphs 1 and 5 of Art 181 read in conjunction. See Art 181(1): ‘Parliament may, be a resolution supported by the votes of a majority of all the members of Parliament, authorise the Government to enter into an agreement for the granting of a loan out of any public fund or public account.’; and 181(5): ‘This article shall, with the necessary modifications by Parliament, apply to an international business or economic transaction to which the Government is a party as it applies to a loan.’

International Investment Law in African Courts

607

The Supreme Court first acknowledged that the provision was very vague as it does not formulate clear criteria that allow distinguishing between international transactions that require parliamentary approval and those that do not. To give meaning and sense to the provision, the Court found that ‘there is a need to imply that only major international business or economic transaction are to be subject to its provision’.28 The Supreme Court then analysed the nature of the PPA and concluded that not only was it of international character but also that given its scope, it had to be qualified as a major economic transaction. For the PPA in question, parliamentary approval would have been necessary. As a consequence, the PPA was invalid according to the Supreme Court.29 Interestingly, the arbitral tribunal gave a different interpretation to Article 181 of the Constitution of Ghana. Like the Supreme Court, the arbitral tribunal also first underlined that the constitutional basis was very vague and therefore, whether the PPA fell under it or not, were both ‘correct’ legal interpretations.30 On the other hand, and this was the Tribunal’s main reason for not following the Supreme Court, Ghana invoked the invalidity of the PPA based on the absence of parliamentary approval only once the dispute had already occurred. In other words, the Tribunal upheld the claimant’s argument by which the constitutionality of a contractual arrangement should not be subject to an ‘after-the-event scrutiny’.31 Based on its own interpretation, the arbitral tribunal upheld its jurisdiction under the PPA.32 The Balkan Energy v Ghana case confirms the position of arbitral tribunals not to submit to an attempt by national courts (even the highest courts) to curtail or negate their jurisdiction. The case also shows that national courts and investment arbitrators dealing with the same national legal object can find extremely different, even opposite, solutions.

C. Enforcement of International Awards by African Courts: The Saga of Ara Libertad Another classic interaction between international investment law and national courts is at the enforcement stage of an award, where the prevailing party seeks to obtain payment of the compensation before a national court. One major hurdle that can prevent the execution of an award is state immunity. A well-known case in this respect occurred in Ghana and related to the detention of the Argentinean warship Ara Libertad at the port of Tema close to Accra. The case attracted significantly international attention given that Argentina requested the International Tribunal for the Law of the Sea (ITLOS) to order provisional measures.33 In this section, the history of this unique case shall be traced back with a focus on the national court decisions. The events that led to the detention of the Ara Libertad originated from a dispute relating to a fiscal agency agreement (FAA) that Argentina entered into in 1994 with the Bankers Trust Company, a New York bank. Under the FAA, Argentina issued a series of bonds. NML Capital Ltd (NML) purchased two of them. As Argentina defaulted on the bonds, NML sued

28 Balkan

Energy Ltd v the Republic of Ghana, above (n 22) para 338. Emphasis in original. para 342. 30 ibid para 377. 31 ibid para 390. 32 ibid paras 374–397. 33 ARA Libertad (Argentina v Ghana), ITLOS Provisional Measures, Order of 15 Dec 2012, ITLOS Reports 2012, 332. 29 ibid

608 Makane Moïse Mbengue and Stefanie Schacherer Argentina in a New York court and obtained judgment.34 Subsequently, Argentina failed to settle the judgment debt, which led NML to start enforcement proceedings in the UK. Contrary to the argument of Argentina objecting to the jurisdiction of the British courts on grounds of state immunity, the UK Supreme Court found that Argentina did not enjoy state immunity in relation to the suit because it had waived its immunity under the FAA.35 A consent order was made against it in favour of NML. Yet again, Argentina failed to pay the judgment debt. When the Argentinean warship Ara Libertad was on its way to Ghana, NML issued a writ in the High Court of Ghana seeking enforcement of the judgment debt. The High Court ordered the arrest and detention of the warship at Tema.36 The High Court based its decision on the explicit waiver of state immunity contained in the bonds and referred back to decisions taken by British courts finding that state immunity did not apply in commercial or private transactions entered into between a state and a private commercial entity.37 After several unsuccessful negotiations between Argentinian diplomats and Ghanaian authorities, Argentina brought an action against Ghana before ITLOS arguing that Ghana had breached international law, namely violating state immunity for warships.38 Pending the decision of the Tribunal, Argentina requested the prescription of provisional measures from ITLOS ordering Ghana to release the ship immediately. On 15 December 2012, ITLOS granted provisional measures and the ARA Libertad was released shortly after.39 To allow Ghana to comply with the ITLOS order, the Attorney-General of Ghana sought the annulment of the decision of the High Court before the Supreme Court.40 According to the Attorney-General, the High Court had erred in international law as it found that the immunity of warships could be waived. In its judgment, the Supreme Court agreed with the AttorneyGeneral and held that the High Court had made a fundamentally and patently wrong decision by holding that Argentina’s contractual waiver of immunity, in so far as it related to the seizure of a military asset, should be given effect.41 The Supreme Court, moreover, found that the 34 NML Capital, Ltd v Argentina, No 08 Civ 6978 (TPG) (SDNY) 23 February 2012. NML also sought enforcement in France. In 2013, the French Supreme Court rendered three fundamental decisions setting aside enforcement measures carried out by NML Capital Ltd against the Republic of Argentina, see Arrêt n° 394 du 28 mars 2013 (10-25.938) – Cour de cassation, FR:CCASS:2013:C100394; Arrêt n° 396 du 28 mars 2013 (11-13.323), Cour de cassation, FR:CCASS:2013:C100396; Arrêt n° 395 du 28 mars 2013 (11-10.450), Cour de cassation, FR:CCASS:2013:C100395. 35 NML Capital Limited (Appellant) v Republic of Argentina (Respondent) [2011] UKSC31, On appeal from [2010] EWCA Civ 41. 36 Order for Interlocutory Injunction and Interim Preservation of the ‘ARA Libertad’, High Court of Justice (Commercial Division), Accra, 2 October 2012. 37 ibid. 38 Argentina initiated arbitration under Annex VII of the 1982 United Nations Convention on the Law of the Sea (UNCLOS). For more details, see J Kraska, ‘The ‘ARA Libertad’ (Argentina v Ghana)’ (2013) 107 American Journal of International Law 404. 39 ARA Libertad (Argentina v Ghana), above (n 33) 332, para 108: ‘(…) THE TRIBUNAL, (1) Unanimously, Prescribes, pending a decision by the Annex VII arbitral tribunal, the following provisional measures under article 290, paragraph 5, of the Convention: Ghana shall forthwith and unconditionally release the frigate ARA Libertad, shall ensure that the frigate ARA Libertad, its Commander and crew are able to leave the port of Tema and the maritime areas under the jurisdiction of Ghana, and shall ensure that the frigate ARA Libertad is resupplied to that end.’ 40 See Supreme Court of Ghana, Civil Motion No J5/10/2013, judgment, 20 June 2013. More concretely, the motion brought by the Attorney-General of Ghana was for certiorari and prohibition against the High Court. In the common law system, ‘certiorari’ is a court process to seek judicial review of a decision of a lower court or administrative agency. 41 Supreme Court of Ghana, Civil Motion, para 56: The ‘learned trial High Court judge, in deciding this case of first impression in this jurisdiction, made a fundamental error which is patent on the face of the record, by failing to respond to the significance of the military nature of the asset sought to be attached. The order to attach a military vessel was on its face palpably and fundamentally wrong in law and principle’. The paragraph numbering has been added by OUP, see D Dagbanja, ‘ARA Libertad Case, Ghana and NML Capital Limited (joining) v Attorney-General and Argentina (joining), Ruling, Civil Motion No J5/10/2013, ILDC 2547 (GH 2013)’ (2014) 108(1) AJIL 73; 20th June 2013, Ghana; Supreme Court, Oxford Reports on International Law, https://perma.cc/9LKL-DTX2.

International Investment Law in African Courts

609

courts of Ghana should not have promoted conditions leading to a possible military conflict when they had the judicial discretion to follow an alternative path. This public policy consideration should have informed the court that waiver of sovereign state immunity over military assets should not be recognised under Ghanaian common law.42

D. African Courts Enforcing National Law Conflicting with International Investment Law: The Land Reform in Zimbabwe Since Zimbabwe’s independence in 1980, the Zimbabwean Government under the rule of Robert Mugabe, embarked on a programme of land and agrarian reform to establish more equitable access to land for black farmers.43 In 1990, Section 16 of the Zimbabwean Constitution, which guarantees the right to property, was amended to facilitate the expropriation of land owned in majority by white farmers.44 The constitutional amendment was complemented in 1992 by a legislative act, the Land Acquisition Act, which allowed the acquisition of land for the purpose of resettlement of black communities.45 Any expropriation was to result in fair or equitable compensation within a reasonable time under the Land Acquisition Act. At first, the Government of Zimbabwe took the precaution of ensuring that an amendment was introduced into the Constitution (Section 16A) in order not to violate its international obligations, in particular, those guarantees granted to foreign investors under international investment agreements that it had concluded with third countries.46 Due to the slowness in the effective implementation of the reform, the Zimbabwean Government decided in 2000 to carry out another reform, called the Fast Track Land Reform Programme.47 The new reform was to amend the Constitution (new Articles 16A and 16B) to allow for the expropriation of land without compensation from the Zimbabwean Government.48 Specifically, Article 16B(1) of the Constitution allows for the expropriation without compensation of all those agricultural lands that have been acquired and listed for resettlement purposes within the Land Acquisition Act. In the investment arbitration case of Von Pezold v Zimbabwe, the Tribunal found that the measures taken by Zimbabwe were contrary to the applicable BITs.49 The ICSID award did not lead to any change of legislation. To the contrary, the national courts of Zimbabwe further enforced the country’s land reform. This is highlighted by the case Nyahondo Farm Ltd v Brigadier General Tapfumaneyi, in which the Zimbabwean Supreme Court found that Article 16B prevailed over the general Article 16,

42 ibid

para 61. Mbengue, ‘La portée globale d’une lutte locale (Mike Campbell c Zimbabwe)’ in H Muir Watt and others (eds), Tournant Global en Droit international privé (Paris, Pedone, 2020). 44 For more details, see GJ Nadli, ‘Land Reform in Zimbabwe: Some Legal Aspects’ [1993] Journal of Modern African Studies 585, 587–589. 45 In 1996, the Supreme Court of Zimbabwe ruled for the first time that the acquisition of land for social equity and distributive justice purposes by compliance with the Land Acquisition Act was constitutional, see Davies and others v Minister of Lands, Agriculture and Water Development 1996 (1) ZLR 681. 46 See Bernhard von Pezold and Others v Republic of Zimbabwe, ICSID Case No ARB/10/15, Award 28 July 2005, para 101. 47 Mbengue, above (n 43). 48 See 2005 Constitution of Zimbabwe, Arts 16A and 16B, https://perma.cc/NRD4-AZ72. See especially von Pezold v Zimbabwe (n 46) para 184: ‘The effect of the Constitutional Amendment was to expropriate the farms of nearly every white farmer in Zimbabwe (of the 4,500 white farmers farming in 2000, today there are less than 200 whose farms have not been expropriated)’. 49 ibid paras 488–521. 43 MM

610 Makane Moïse Mbengue and Stefanie Schacherer which foresees compensation for expropriation.50 Subsequent national court decisions have systematically relied on the Nyhondo case.51 The successor of Robert Mugabe, Emmerson Mnangagwa, indicated that the land reform was irreversible. He promised that former white farmers would be compensated, although not for the land itself but rather for the improvements they had made on their former farms.52

E. African Courts Applying National Investment Law: The Case of South Africa The South African Investment Promotion and Protection Bill was adopted in 2015 and came into force in 2018.53 One of the characteristic features of this legislation is that disputes between foreign investors and South Africa will be dealt with in South African national courts.54 In terms of its substantive standards, the instrument is largely pegged to the South African Constitution and based on the extension to foreign investors of the protection granted to nationals. To date, foreign investor claims based on the Bill have not yet occurred but a closer look at the case law of the South African Constitutional Court bears interesting insights for foreign investors in South Africa. In particular for the guarantees against expropriation, the Promotion and Protection Bill explicitly refers to the country’s Constitution stating that ‘Investors have the right to property in terms of Section 25 of the Constitution.’55 An interesting decision on how the South African Constitutional Court interprets this Section is the Agri South Africa v Minister for Minerals and Energy case which merits further discussion.56 At the beginning of the early 2000s, South Arica enacted the Mineral and Petroleum Resources Development Act (Mineral Act), which was to reform the mineral rights regime. Under the Act, the state became the custodian of the country’s mineral resources acting for the benefit of all South Africans. It may grant, refuse and administer mineral rights.57 Agri South Africa, a union of South African landowners brought an application against the Minister for

50 Nyahondo Farm (Private) Limited v Brigadier General AW Tapfumaneyi & Ors, Case No SC 176/08, High Court of Zimbabwe, Harare, 7 July 2008, aff ’d on appeal to the Supreme Court of Zimbabwe, Harare, 6 November 2008, CLEX-91 and CLEX-92. Judgment No CCZ 5/17 Constitutional Appeal No SC 176/08. 51 See, eg, the High Court’s judgment in Route Toute v Minister of National Security Responsible for Land, Land Reform and Resettlement: ‘I am bound by the contrary position recently adopted by the Supreme Court in [Nyahondo] to the effect that agricultural land covered by investment protection agreements under section 16(9b) is susceptible to acquisition in terms of section 16B” (CLEX-93, 19)’, quoted in von Pezold v Zimbabwe (n 46) para 529, fn 7. 52 G Mkodzongi and P Lawrence, ‘The Fast-Track Land Reform and Agrarian Change in Zimbabwe’ (2019) 46 Review of African Political Economy 1. 53 Promotion and Protection of Investment Bill of South Africa (adopted 2015, entered into force on 13 July 2018), https://perma.cc/CGP6-7TAJ. 54 Promotion and Protection of Investment Bill of South Africa, s 13: ‘4) Subject to applicable legislation, an investor, upon becoming aware of a dispute as referred to in subsection (1), is not precluded from approaching any competent court, independent tribunal or statutory body within the Republic for the resolution of a dispute relating to an investment. 5) The government may consent to international arbitration in respect of investments covered by this Act, subject to the exhaustion of domestic remedies. The consideration of a request for international arbitration will be subject to the administrative processes set out in section 6. Such arbitration will be conducted between the Republic and the home state of the applicable investor.’ 55 Promotion and Protection of Investment Bill of South Africa, s 10. Legal protection of investment. 56 Agri South Africa and Minister for Minerals and Energy, Constitutional Court of South Africa, Case CCT 51/12 [2013] ZACC 9, Judgment of 18 April 2013 (hereafter Agri South Africa). Similar cases are Shoprite Checkers (PTY) Limited and others v Affairs and Tourism Eastern Cape and others, Constitutional Court of the Republic of South Africa, Judgment, 30 June 2015 [46]; Thiagraj Soobramoney v Minister of Health (Kwayulu-Natal), Constitutional Court of the Republic of South Africa, Judgment, 27 Nov 1997, [36]. 57 B Winks, ‘Expropriation – a minefield?’ (1 July 2013) De Rebus 44.

International Investment Law in African Courts

611

Minerals and Energy given that the implementation of the Mineral Act interfered with their mineral ownership. Agri South Africa held coal rights in respect to some of its farms and land. These coal rights had not yet been used in the sense that no prospecting or mining had been conducted. As a result of the Mineral Act, the application of Agri South Africa to obtain new rights for the exploitation of the coal was denied.58 For the claimant, this amounted to an expropriation of their mineral rights, yet no compensation was paid by the state. The first instance tribunal, the North Gauteng High Court, found that the claimant’s mineral rights had been expropriated and therefore compensation was due. The Minister for Minerals and Energy subsequently appealed to the Supreme Court, which, in return, denied the existence of any expropriation. As a consequence, Agri South Africa took recourse before the Constitutional Court, invoking Section 25 of the 1996 South African Constitution. Section 25 of the Constitution states, on the one hand, that ‘[n]o one may be deprived of property except in terms of law of general application, and no law may permit arbitrary deprivation of property’.59 On the other hand, Section 25 foresees that property can be expropriated only when based on a law of general application and only if that law has been enacted for the public purpose and that compensation has been paid.60 In deciding the case, the Court had to tackle two key questions. First, the question of whether or not mineral rights can be considered as property. Second, the question of whether the implementation of the Mineral Act constituted a deprivation, which amounted to expropriation as alleged by Agri South Africa.61 As regard to the first question, the Court found that mineral rights are indeed property and unfold in two aspects: the ownership of the minerals and the right to exploit them.62 As regards the second question, the Court distinguished the notions of deprivation and expropriation. Depriving property relates, according to the Constitutional Court, to the sacrifices that holders of private property may have to make or the extinguishing of such property rights without the right to receive compensation. Expropriation, however, entails state acquisition of the property in the public interest and must always be accompanied by compensation.63 In its analysis, the Court accepted that the Mineral Act had an effect of deprivation for the claimant but the state measure was in conformity with Section 25 of the Constitution given that the Mineral Act was of general application and, moreover, the deprivation was not arbitrarily based on the object of the law as well as the transitional arrangements. According to the Court, there could be no expropriation where deprivation does not result in property being acquired by the state. As mentioned before, under the Mineral Act, the state becomes the custodian of the mineral rights, ie property, but does not acquire the property. For the Court, this finding was corroborated by the fact that under the Mineral Act, the state will not be a co-contender with citizens for the rights to exploit national minerals. The state acts as a ‘facilitator’ so that broader and more equitable access to mineral and petroleum resources can be realised. As a result, the deprivation of property in question did not amount to expropriation.64

58 Agri

South Africa, above (n 56) paras 14–15. African Constitution, s 25(1). 60 ibid s 25(2). 61 Agri South Africa, above (n 56) para 46. 62 ibid paras 37ff. 63 ibid para 48. 64 ibid para 69: ‘[…] An assertion by Agri SA that the state has in terms of the correct interpretation of section 25 expropriated the mineral rights, is an overly liberal one. It disregards the public interest and constitutional imperative to transform and facilitate equitable access to our mineral and natural resources, to which courts are enjoined to have regard when construing section 25.’ 59 South

612 Makane Moïse Mbengue and Stefanie Schacherer Hence, the Constitutional Court did not easily accept the presence of expropriation, which would require the state to pay compensation. In its interpretation of Section 25(2) of the Constitution, the Court reminded the disputing parties of the underlying issues and specificities of property in South Africa. The Court stressed the role that Section 25 plays in ‘facilitating the fulfilment of [the] country’s nation-building and reconciliation responsibilities, by recognising the need to open up economic opportunities to all South Africans’.65 The Court went on by highlighting the history of the country, which would warrant against a ‘nearabsolute’ status of individual property rights.66 Indeed, the way in which the Constitutional Court weighed the private interest against the public interest is telling. It reveals the overriding aim expressed in the South African Constitution to bring about a fundamental transformation of a society that suffered political and economic injustice in the past. Therefore, and in the words of the Court, Section 25 must be interpreted ‘with due regard to the gross inequality in relation to wealth and land distribution’.67

III. AFRICAN SUPRANATIONAL COURTS

Section III of the present chapter deals with African regional courts and institutions given that they play an increasing role in the judicial landscape of Africa. Major institutions, such as the OHADA Common Court of Justice and Arbitration (section III.A), the East African Court of Justice (section III.B) and the ECOWAS Community Court of Justice (section III.C) will be analysed.68

A. OHADA Common Court of Justice and Arbitration (CCJA) The CCJA was established by the OHADA69 Treaty. The Court’s main responsibility is to ensure and to monitor the application and uniform interpretation of OHADA law in all Member States.70 The CCJA also serves as the region’s arbitration centre administrating and supervising the arbitral proceedings.71 In addition, the CCJA has the competence to annul arbitral decisions.72

65 ibid

para 60. para 62: ‘[…] It must always be remembered that our history does not permit a near-absolute status to be given to individual property rights to the detriment of the equally important duty of the state to ensure that all South Africans partake of the benefits flowing from our mineral and petroleum resources.’ 67 ibid para 60. 68 The present chapter deliberately excludes the Tribunal of the South African Development Community (SADC), which is dealt with in H Asmelash, ‘The Legacy of the SADC Tribunal in International Investment Law’, ch 30 in this volume. 69 OHADA is the Organisation for Harmonisation of Business Law in Africa (OHADA – the French acronym for Organisation pour l’Harmonisation en Afrique du Droit des Affaires) with its current 17 Member States is the largest harmonisation project in Africa. It is interesting to note that OHADA is not listed as one of the RECs but OHADA represents a unique case of the ‘use of law to promote regional development in Africa’, J Bashi Rudahindwa, Regional Development Through International Law – Establishing an African Economic Community (Oxfordshire, Routledge, 2018) 175. 70 OHADA Treaty, Art 1: ‘The main objective of having conformity with OHADA law is to ensure that local and foreign investors enjoy protection and reliable judicial framework and thereby to generally improve the business environment in order to attract more investment.’ See also Bashi Rudahindwa, above (n 69) 175. 71 OHADA Treaty, Art 21. 72 OHADA Arbitration Rules, Art 29. 66 ibid

International Investment Law in African Courts

613

In a recent judgment of 19 November 2015, the CCJA annulled an OHADA award in the case Getma v Guinea.73 The original award, rendered in April 2014, held that Guinea’s termination of a port and railway concession contract breached the contract, which led to a finding of €38 million plus interest in compensation. Shortly after, Guinea submitted an annulment action before the CCJA. The private fee agreement between the parties became the central issue in the annulment judgment.74 More concretely, Guinea alleged that the Tribunal had failed to fulfil its mandate by ignoring provisions that prohibited private fee agreements between the parties and the arbitrators. According to Guinea, the three arbitrators in the OHADA arbitration were bound by a fee schedule approved in advance by the CCJA itself, which sets the arbitrators’ fees at a total of around 40 million CFA francs (approximately €60,000). Guinea revealed that the president of the Tribunal had requested on two occasions that the amount should be increased, yet the CCJA denied both requests. Nevertheless, the President of the Tribunal, Ibrahim Fadlallah, actively engaged in negotiating a sperate fee agreement with Getma and Guinea. Under the private fee agreement, the total arbitrators’ fee amounted to €450,000.75 For its part, Getma contested this view of the Tribunal’s conduct, arguing that OHADA law permitted parties to derogate from general rules on fees by mutual agreement. In any case, Getma contended, a tribunal’s violation of fee rules could not lead to the annulment of the whole award. The CCJA did not follow the arguments of Getma finding good reasons for the CCJA provisions setting fees in advance. Such reasons include encouraging parties in OHADA states to submit claims to arbitration and to ensure that arbitral fees were foreseeable and proportional to the amount in dispute.76 Furthermore, the court held that the provisions of OHADA law cited by Getma did not apply to the original arbitration, which was instead governed by the more specific CCJA rules. Given that the Tribunal’s requests to approve a higher fee were rejected twice, and that the Tribunal had nonetheless made its own arrangements with the parties, the CCJA held that the Tribunal had exceeded its mandate. Consequently, the award was annulled. For Getma the annulment of the CCJA has had clear consequences. The company failed to enforce the award in US courts.77 Furthermore, it is interesting to note that in a parallel ICSID proceeding based on the US-Guinea BIT, Getma argued that the annulment of the OHADA award constituted a denial of justice.78 The ICSID tribunal found, however, that no denial of justice could occur in circumstances where Getma could simply present its claims again to a new OHADA tribunal.79 The claimant’s nominated arbitrator, Bernardo Cremades, issued a

73 Getma International v Republic of Guinea, Case 139/2015, CCJA Annulment, 19 November 2015. The underlying arbitration: Getma International v Republic of Guinea [I], CCJA Case 1:14-cv-01616-RBW, Award, 22 September 2014 (originally written in French); For the parallel BIT claim, see Getma International and others v Republic of Guinea [II], Award (16 August 2016) ICSID Case No ARB/11/29. 74 It goes beyond the present contribution to comment on the pros and benefits of the CCJA’s reasoning. For a critical comment, see T Jones, ‘Attempt to enforce Guinea award in the US continues after “repugnant set aside”’ (Global Arbitration Review, 3 May 2016). For a more positive account, see R Soopramanien and S Soopramanien, ‘Spotlight on Africa: Problem of Legitimacy and Inclusivity in International Arbitration’ (2016) 4 Transnational Dispute Management 1. 75 Getma International v Republic of Guinea, above (n 73). 76 ibid pt 5: ‘Attendu que ces disposition ont pour objet de garantir aux parties qui ont décidé de soumettre leur litige à l’arbitrage de la Cour, le paiement d’honoraires prévisibles, proportionnels à la valeur réelle du litige et déterminées selon un barème connu à l’avance.’ 77 On 3 November 2015, the US District Court for the District of Columbia ruled that Getma’s enforcement action should be stayed. After, in the judgment of the CCJA, the enforcement was denied by order of 9 June 2016. 78 Getma International and others v Republic of Guinea, above (n 73) [II]. 79 ibid paras 348ff.

614 Makane Moïse Mbengue and Stefanie Schacherer five-page dissenting opinion on this precise question expressing the view that this was not a viable option since the claimants had meanwhile lost confidence in the OHADA forum.80

B. The East African Court of Justice (EACJ) The EACJ is the main judicial organ of the East African Community (EAC). Since its establishment, the Court has been very active in supporting the objective of the EAC to foster the process of integration. According to Article 30(1) of the EAC Treaty, any person who is resident in a Partner State may refer for determination by the Court, the legality of any Act, regulation, directive, decision or action of a Partner State or an institution of the Community on the grounds that such Act, regulation, directive, decision or action is unlawful or is an infringement of the provisions of this Treaty.

This provision allows corporations to directly bring a claim against an EAC Member State when the latter adopts a measure that infringes any of the EAC Treaty provisions including the guarantees for economic operators.81 An illustrative case is British American Tobacco Ltd v The Attorney General of Uganda.82 Under scrutiny, in this case, was the 2014 Excise Duty Act of Uganda that aimed to introduce an excise duty on cigarettes applicable in a non-discriminatory manner to all such goods in the EAC region. Yet in 2017, the Act was amended to make a distinction between locally (ie EAC) manufactured cigarettes and those imported from outside the EAC. British American Tobacco, a company with headquarters in the UK, challenged the legality of the 2017 amendment, arguing that it violated certain provisions of the EAC Treaty as well as the Customs Union and the Common Market Protocols.83 In its judgment, the Court made an interesting statement on those provisions of the EAC Treaty framework dealing with investment as the Court examined whether the impugned amendment of the Ugandan legislation ‘violate[d] and/ or infringe[d]’ Article 80(1)(f).84 According to the Court: Article 80(1)(f) imposes the obligation to harmonize and rationalise investment incentives including those relating to taxation of industries particularly those that use local materials and labour with a view to promoting the Community as a single investment area. The gist of these legal provisions is to impress it upon Partner States to establish an export oriented economic dispensation in the EAC region and pursue such investment policies as would entrench the EAC single investment area.’85

In the overall outcome of the case, the Court found that Uganda had acted in a manner that was likely to jeopardise the objectives of the EAC Treaty as it would ‘roll back’ the gains of the Customs Union and the Common Market established thus far in the EAC region. Concerning the question of whether or not Uganda had violated Article 80(1)(f) of the EAC Treaty, the Court ultimately found that the applicant had failed to prove its case.86 Despite this latter circumstance, the British American Tobacco case is interesting in that it highlights that the EACJ is keen to promote the single investment area of the EAC and that in fact a foreign

80 ibid,

Dissenting Opinion of Bernardo Cremades, para 11. eg, Simon Peter Ochiebg & Another v Attorney General of Uganda, EACJ, Decisions, Ref No 11 of 2013. American Tobacco Ltd v The Attorney General of Uganda, EACJ, Decision, Ref No 7 of 2017, 26 March 2019. 83 See also Bashi Rudahindwa, above (n 69) 157. 84 British American Tobacco Ltd v The Attorney General of Uganda, above (n 82) para 47. 85 ibid para 73. Emphasis in original. 86 ibid para 79. 81 See,

82 British

International Investment Law in African Courts

615

company incorporated in an EAC country can bring a claim against such country before the EACJ.

C. ECOWAS Court of Justice The possibility for foreign investors within ECOWAS87 to use the ECOWAS Community Court of Justice (CCJ) has been considered and promoted.88 Yet as the law currently stands, there are a number of obstacles and legal uncertainties. The latest amendment of 2005 extended CCJ jurisdiction to human rights cases and expanded the admissibility rules to disputes between individuals and their Member State of residence.89 As of today, the CCJ has become ‘four courts in one’: an administrative tribunal for ECOWAS, a human rights court, a court of arbitration, and an inter-state dispute resolution tribunal. A priori, the constellation when the CCJ acts in its capacity of an arbitral tribunal seems to be relevant for investment disputes. However, the jurisdiction must be established through consent in an investment treaty or contract.90 And in fact, the 2008 ECOWAS Supplementary Act on Investments allows possible recourse to the CCJ only in exceptional circumstances.91 The other route by which investors can bring a claim before the CCJ, is to file a claim based on an alleged violation of the investor’s human rights. In this respect, the case law of the Court is, however, not conclusive. In CNDD v Côte d’Ivoire,92 the Court found that legal persons can file human right complaints. Yet the Court later rejected this in Ocean King Nigeria Ltd v Senegal.93 In Ocean King Nigeria, however, the Court held that the right to a fair trial was not to be treated as a human right and found that it must examine the right of corporations to a fair trial under the ECOWAS Treaty.94 This short overview of the case law of the ECOWAS Court already reveals certain ambiguities. In sum, except for CNDD v Côte d’Ivoire, corporations cannot submit violations of human rights to the CCJ; as far as the right of a corporation to a fair trial is concerned, investors can bring a claim against the state before the Court. At the same time, it is important to mention that the importance of the CCJ might grow in the future. The new ECOWAS Investment Code that was finalised in 2018 includes the arbitration division of the CCJ as one of the fora for the settlement of investment disputes.95 87 Economic

Community of West African States / Communauté économique des États de l’Afrique de l’Ouest. M Happold and R Radović, ‘The ECOWAS Court of Justice as an Investment Tribunal’ (2018) 19(1) The Journal of World Investment & Trade 95. 89 Revised Treaty of ECOWAS, 1995 Revised ECOWAS Treaty, 1991 Protocol creating the CCJ, A/P1/7/91; 2005 Supplementary Protocol, A/SP1/01/05. 90 See Ocean King Nigeria Ltd v Senegal, ECW/CCJIJUD/07/11, 8 July 2011, para 47: ‘[…] corporate bodies such as the plaintiff herein can access the Court only where there is a prior agreement between the parties to a particular transaction that disputes arising out of that transaction shall bet settled by the Court […]’. 91 See ECOWAS Supplementary Act on Investments, Art 33(6): ‘Any dispute between a host Member State and an Investor, as envisaged under this Article that is not amicably settled through mutual discussions may be submitted to arbitration as follows: (a) a national court; (b) any national machinery for the settlement of investment disputes; (c) the relevant national courts of the Member States.’ Art 33(7) thereof then states: ‘Where in respect of any dispute envisaged under this Article, there is disagreement as to the method of dispute settlement to be adopted; the dispute shall be referred to the ECOWAS Court of Justice.’ 92 The National Co-ordinating Group of Departmental Representatives of the Cocoa-Coffee Sector (CNDD) v Côte d’Ivoire, ECW/CCJ/APP/02/09, 17 Dec 2009, paras 20–30. 93 Ocean King Nigeria Ltd v Senegal, ECW/CCJIJUD/07/11, 8 July 2011, para 50. See for more details Happold and Radovic, above (n 88) 111–113. 94 One needs to note that from the Court’s reasoning, it is not clear why and on what basis the Court considers itself to be under such obligation; presumably it is a matter for the CCJ of inherent jurisdiction. 95 The new 2018 ECOWAS (the new Act is on file with the authors), Art 54(2): ‘Where recourse is made to arbitration, the arbitration may be conducted at any established public or private alternative dispute resolution centre or the 88 See

616 Makane Moïse Mbengue and Stefanie Schacherer IV. CONCLUDING REMARKS

Based on the discussed cases, African courts are an important example that show how national judges are seen by investment arbitration and how they look at the arbitral systems, and how these two visions can diverge. The important aspect to take away from a discussion on international investment law in African courts is that there seems to exist a trend in Africa towards using African national and regional courts for settling investment disputes. Evidence of the trend is a certain push for regionalisation of international investment law and the emergence of an African vision of international investment law.96 South Africa is a frontrunner in this respect. Tanzania also enacted the new Sovereign Act in 2017, which stipulates that permanent sovereignty over natural resources requires that any disputes relating to resource extraction be adjudicated in judicial bodies or other organs established in Tanzania.97 Other African states, such as the Ivory Coast, now promote African arbitral centres instead of ICSID or other international institutions. The Ivory Coast’s 2018 Investment Code states that disputing parties can agree to submit their dispute to the CCJA. The clause is not obligatory but mentioning the CCJA as the sole arbitral centre is a clear indication of promoting this institution. A very similar approach has been taken in the 2018 ECOWAS Investment Code, which, in turn, suggests the ECOWAS Court as the forum for the settlement of investment disputes falling under the new code. All these examples show a nascent trend. To some extent, they seem to suggest a certain loss of confidence in the ability of international arbitrators to understand the specific circumstances of Africa, which inter alia encompass issues to do with the redistribution of national resources and the mitigation of historical inequalities. The trend of returning to African courts, tribunals and arbitration centres is highly interesting and these developments should further be monitored.

arbitration division of the ECOWAS Court of Justice. Member States and investors are encouraged to utilise regional and national alternative dispute settlement institutions.’ 96 MM Mbengue and S Schacherer, ‘Evolution of International Investment Agreements in Africa: Features and Challenges of Investment Law “Africanization”’ in J Chaisse, L Choukroune and S Jusoh (eds), Handbook of International Investment Law and Policy (Berlin, Springer International Publishing, 2019), https://link.springer.com/ referenceworkentry/10.1007/978-981-13-5744-2_77-1. 97 See Tanzania, The Natural Wealth and Resources Act 2017 (Sovereignty Act), s 11(2).

36 International Investment Law in European Courts DOMINIQUE HASCHER

I. INTRODUCTION

S

UBJECT TO CONSTITUTIONAL requirements for the ratification of investment treaties,1 the legal protection of investors can be afforded by the courts of the host state.2 With the exception of the French Conseil d’État which decided that investment treaties only address inter-state obligations,3 Bilateral Investment Treaties (BITs) are considered to give rise to individual rights and actions as recognised by the English Court of Appeal4 and the German Constitutional Court (BVG).5 The BVG also considered the protection of the legitimate expectations in the stability of the investor’s legal situation in the constitutional review of legislation on the nuclear energy phase-out.6 In addition to the above topics (constitutional review, nature of BIT rights, etc …), European courts have also preoccupied themselves with the interaction between investment treaties and the European Union legal order.7 Investment disputes are, more often than not, submitted to international arbitral tribunals which postpones the intervention of state courts after making the award. The present chapter however focuses on the judicial function of national judges

1 For CETA, see Decision of the Conseil Constitutionnel (France) of 31 July 2017, Conseil Constitutionnel, No 2017-749 DC and Decision of 13 October 2016, BVG, 2BvR 1368/16, 2BvR 1444/16, 2BvR 1823/16, 2BvR 1482/16, 2BvE 3/16. 2 M Paparinskis, ‘Substantive Standards of Investment Protection under EU Law and International Investment Law’ in H Ruiz Fabri and E Gaillard (eds), EU Law and International Investment Arbitration (New York, JurisNet, 2018) 225. 3 Decision of the Conseil d’État (France) of 21 December 2007, CE, No 280264. 4 Judgment of the English Court of Appeals of 9 September 2005 [2005] EWCA Civ 1116, 19–22. 5 Decision of the BVG of 8 May 2007, BVG, 2 BvM 1/03- Rn (1-95), para 54. 6 Decision of the BVG of 6 December 2016, BVG, 1BvR 2821/11, 1 BvR 321/12, 1 BvR 1456/12, 334–346. 7 Judgment of the Bundesverfassungsgericht (Germany) of 3 March 2016, BVG, I ZB 2/15 referral on the compatibility of the arbitration clause in the Netherlands-Czechoslovakia BIT of 1991 with EU Law. Before the General Court judgment of 18 June 2019, joined cases T-624/15, 694/15, 704/15, which annulled the EU Commission’s decision on State aid in Micula, the English Commercial Court considered a reference inappropriate ([2017] EWHC 31 (Comm) judgment of 20/01/2017 para 181. See also UK Supreme Court judgment of 19 February 2020 [2020] UKSC 5), but not the Judgment of the Brussels Court of Appeal (Belgium) of 12 March 2019. The Svea Court of Appeal refused to make a request for preliminary question regarding the compatibility of Art 26 of the ECT with European Union primary legislation in the annulment proceedings of an SCC award (Judgment of the Svea Court of Appeal (Seden) of 25 April 2019, Svea Court of Appeal, case No T 4658-18 (Novenergia II v Kingdom of Spain)).

618 Dominique Hascher facing an investment award, which is at the core of the problem of the relations between legal systems.

II. JUDICIAL RESTRICTIONS

The supervisory power of national courts over investment arbitration may be excluded by the sovereign immunity and non-justiciability doctrines. It may be curbed by the existence of an estoppel.

A. Sovereign Immunity Consent to arbitration is generally not regarded as a waiver of sovereign immunity of execution. Applying Article 55 of the International Centre for Settlement of Investment Disputes (ICSID) Convention,8 the English Commercial Court viewed the London assets of the National Fund of Kazakhstan, created to assist in the management of the economy and government revenues in the short and long term, as intended for sovereign use.9 Sovereign immunity of jurisdiction has been an issue in English and German cases which discussed the interpretation of BIT arbitration provisions. In Gold Reserve Inc v Bolivarian Republic of Venezuela10 and PAO Tatneft v Ukraine,11 sovereign immunity defences were made by Venezuela and Ukraine against an order to enforce the award in the English Courts. Both states argued that they had not lost immunity inasmuch as they had not consented to arbitrate the claims decided in the award. The Kingdom of Thailand submitted to the German Bundesgerichtshof (BGH) that an award made under the Germany-Thailand BIT of 24 June 2002 dealt with sovereign activities which are immune of the jurisdiction of other states.12 Examination of sovereign immunity arguments often comes down to a determination of the arbitrators’ competence under the BIT arbitration provisions even when, as in Tatneft, the award on jurisdiction was not directly impugned.

B. The Doctrine of Non-Justiciability The Republic of Ecuador v Occidental judgment of 29 April 2005,13 a first reported English decision to deal with a BIT award, considered the Common-Law doctrine of non-justiciability, following which, courts will not adjudicate on the transactions made between sovereign states on the level of international law.14 Occidental argued that Ecuador’s challenge on count of jurisdiction would depend upon the interpretation of the US-Ecuador BIT of 27 August 1993 to which the UK was not a party. The Commercial Court of England and Wales retorted that Occidental and Ecuador had agreed that rights originating in international law will be

8 ‘Nothing in Article 54 shall be construed as derogating from the law in force in any Contracting State relating to immunity of that State or any foreign State from execution’. 9 Judgment of the English Commercial Court of 20 October 2005 [2005] EWHC 2239 (Comm) 92; H Fox, The Law of State Immunity, 3rd edn (Oxford, Oxford University Press, 2013) 524. 10 Judgment of the High Court of England and Wales of 2 February 2016 [2016] EWHC 153 (Comm). 11 Judgment of the High Court of England and Wales of 13 July 2018 [2018] EWHC 1797 (Comm). 12 Judgment of the German Bundesgerichtshof of 6 October 2016, BGH, I ZB 13/15 (Thailand-Germany BIT). 13 Judgment of the High Court of England and Wales of 29 April 2005 [2005] EWHC 774 (Comm) 67–86. 14 Fox, above (n 9) 59, 63.

International Investment Law in European Courts 619 considered by a tribunal seated in London, whose procedure is subject to domestic law, including section 67 of the English Arbitration Act 1996 for challenging an award on substantive jurisdiction.15 The judgment illustrates the ‘foothold’ in domestic law exception to non-justiciability when consideration of a treaty not part of English Law becomes necessary to enforce private law rights, such as challenging an award.16 The Court of Appeal of England and Wales held that such ‘foothold’ was sufficient although the question had to be answered by looking more widely than at section 67 alone: The case is not concerned with an attempt to invoke at a national level a Treaty which operates only at an international level. It concerns a Treaty intended to give rise to rights in favour of private investors capable of enforcement, to an extent specified by the Treaty wording, in consensual arbitration against one or other of its signatory States.17

The Court of Appeal remarked that the BIT involves a deliberate attempt to ensure for private investors the benefits and protection of consensual arbitration; and this is an aim to which national courts should, in an internationalist spirit and because it has been agreed between States at an international level, aspire to give effect.18

C. Estoppel The challenge of an award is generally regarded as permissible on grounds that have not knowingly been abandoned before the arbitral tribunal.19 The application of estoppel depends on the nature of the court’s review over the ground for attack. The English Commercial Court remarked that: [I]t is difficult to see how a waiver could arise in circumstances where it is well established that there can be a re-hearing under section 67, a fact parties are taken to know, and in the context of no restriction being set out in section 67 itself restricting what arguments may be re-run, no question of any loss of a right to advance particular arguments (…) on a re-hearing under section 67 can arise.20

Jurisdiction is one field where estoppel is frequently discussed. The moment and extent of review of arbitral jurisdiction by state courts are open to several solutions.21 The English Commercial Court recognised in Republic of Ecuador v Occidental that it is entitled to rehear and re-evaluate all issues of fact as well as law: It is now well established that a challenge to the jurisdiction of an arbitration panel under section 67 proceeds by way of a re-hearing of the matters before the arbitrators. The test for the court is: was the Tribunal correct in its decision on jurisdiction? The test is not: was the Tribunal entitled to reach the decision it did.22 15 ‘A party to arbitral proceedings may (upon notice to the other party and the arbitral tribunal) apply to the court (a) for challenging any award of the arbitral tribunal as to its substantive jurisdiction; or (b) for an order declaring an award made by the tribunal on the merits to be no effect, in whole or in part, because the tribunal did not have substantive jurisdiction’ (M Mustill and S Boyd, Commercial Arbitration, 2nd edn (New York, Lexis Nexis, 2001). 16 Judgment of the High Court of England and Wales of 29 April 2005, above (n 13) 32–35, 73, 75–77. 17 Judgment of the English Court of Appeals of 9 September 2005, above (n 4) 37. 18 ibid 32. 19 See judgment of the Svea Court of Appeal (Sweden) 9 December 2016, No T 2675-14, 5.3.3, and judgment of the Svea Court of Appeal (Sweden) of 15 May 2003, case No T8735-01. 20 Judgment of the High Court of England and Wales of 2 March 2018 [2018] EWHC 409 (Comm) 72. 21 A Rau, The Allocation of Power between Arbitral Tribunals and State Courts (Leiden, Brill Nijhoff, 2018). 22 Judgment of the High Court of England and Wales of 2 March 2006 [2006] EWHC 345 (Comm) 7; see also judgment of the High Court of England and Wales of 5 December 2007 [2007] EWHC 2851 (Comm) 13. As opposed to a

620 Dominique Hascher It follows that the arbitral tribunal’s conclusions may have an interest but have no legal or evidential weight.23 The Paris Court of Appeal decided in the case of The Czech Republic v Nreka: ‘the Court, in assessing the meaning and scope of the arbitration agreement […], reviews de novo the parties’ grounds and arguments in fact and law’.24 In contrast, the Swiss Federal Tribunal freely reviews the legal issues pertaining to jurisdiction, but it must make its decision on the basis of the facts found in the award under attack because an application to set aside does not turn the Tribunal into an appeal court.25 The English Commercial Court drew a distinction in PAO Tatneft between an immunity plea against the Court’s jurisdiction to hear an application to enforce the award and the absence of a jurisdictional objection before the arbitral tribunal: [T]here is nothing in the SIA which suggests that there can be a foreclosure of the points which the State may raise as to the applicability of the immunity afforded by the SIA by reason of what may have occurred in front of an arbitral tribunal in a way similar to that provided for by the Arbitration Act.26

The possibility that a waiver of immunity could be made in the course of an investment arbitration cannot be excluded and in the English Commercial Court’s opinion, what would at least be required would be conduct which clearly indicated that the State was foregoing reliance on a particular point not just for the purposes of the arbitration but for wider purposes including any subsequent issues as to state immunity before a national court.27

The admissibility of jurisdictional objections different from those which were made before the arbitral tribunal has been considered in different manners. The English Commercial Court declared in GFP GP v The Republic of Poland that: [I]t is for the Court under section 67 to consider whether jurisdiction does or does not exist, unfettered by the reasoning of the arbitrators or indeed the precise manner in which arguments were advanced before the arbitrators. Ultimately jurisdiction either is, or is not, conferred on the true construction of the arbitration agreement, and that ought not to be fettered by how arguments were advanced below […]. Indeed, experience shows that the arguments on challenge can be, and are, often presented in fresh and different ways.28

When a party has raised an objection to jurisdiction, it has also reserved itself new arguments and new evidence. The Paris Court of Appeal decided in the case of The Czech Republic v Nreka: [C]ontrary to what the Czech Republic or Mr. Nreka argued with regards to the applicability of the rule of estoppel, it is of little importance whether the arguments raised in these setting aside proceedings regarding the lack of an arbitration agreement were raised before the arbitrators, provided an objection to jurisdiction was raised before during the arbitration procedure, as the fact that an

rehearing type appeal, under a review type appeal, the appeal court can only make a different conclusion on the facts if no reasonable court, on the evidence available to it, could have reached that conclusion. 23 Judgment of the High Court of England and Wales of 13 October 2017 [2017] EWHC 2539 (Comm) 42. 24 Judgment of the Paris Court of Appeal (France) 25 September 2008, Mealey’s International Arbitration Report, Vol 24, 2 February 2009, 12; T Clay and P Pinsolle, The French International Arbitration Law Reports, Case (2008) n 25, p 22 (JurisNet 2012). 25 Judgment of the Swiss Federal Tribunal of 20 September 2016, TF, No 4A 616/2015, 3.1.1, 3.1.2. 26 Judgment of the High Court of England and Wales of 13 July 2018, above (n 9) 35. 27 ibid 37. In contrast to sovereign immunity, application of the principle of non-justiciability does not depend on the consent of the state (Fox, above (n 9) 70). 28 Judgment of 2 March 2018 [2018] EWHC 409 (Comm) para 70.

International Investment Law in European Courts 621 argument was not previously raised before the arbitrators does not imply that the applicant accepted their jurisdiction.29

Failure to object to an irregularity is not a failure to object to all arguments in support of the irregularity that has been waived.30 Consistent with its limited review stance over arbitral jurisdiction, the Swiss Federal Tribunal took a reverse approach with regard to a jurisdictional defence raised in the annulment proceedings of an award rendered in Geneva in a PCA administered United Nations Commission on International Trade Law (UNCITRAL) arbitration. The Swiss Court ruled that India was estopped from challenging the competence of the arbitral tribunal on a basis which had not been discussed before the arbitral tribunal.31

III. JUDICIAL CONTROL

The ordinary standards of review of the arbitration law apply to investment awards.32 The Svea Court of Appeal noted with regard to an arbitration under the Netherlands-Czech Republic BIT, that the existence of an international dimension has no consequence other than that the issue of whether the arbitral award should be declared invalid or set aside shall primarily be determined in accordance with Swedish law, international law being a part of Swedish law.33 The Washington Convention of 18 March 1965 (ICSID Convention) does not single out jurisdiction as a particular ground of its review process.34 The sameness of the standards of review in the ICSID control system for jurisdiction and substance contrasts with the review of jurisdiction before national courts35 and the narrow exception of public policy regarding substance.36 ICSID arbitration and awards rarely give rise to litigation because the ICSID Convention creates an international self-contained framework.37 There is no room for discussing the validity of the award before national courts as would be the case under the New York Convention of 10 June 1958 on the Recognition and Enforcement of Foreign Arbitral Awards.38 Article 54(1)

29 Judgment of the Paris Court of Appeal (France) of 25 September 2008, above (n 24). See also, judgment of the Cour de cassation (France) of 2 December 2020, Civ. 1st No 1915396. 30 New arguments in support of the Tribunal’s decision on jurisdiction may be raised by the defendant in setting aside proceedings, The Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, paras 4.4.3–4.4.5 31 Judgment of the Swiss Federal Tribunal of 11 December 2018, TF, No 4A_65/2018 para 3.2.3.3.1 relying on good faith as a basis for estoppel. 32 Judgment of the Paris Court of Appeal (France) of 18 November 2010, CA Paris, Rev arb 2011.695, commentary S Lemaire. 33 Judgment of the Svea Court of Appeal (Sweden) 15 May 2003, Case No T8735-01 (UNCITRAL award made in Stockholm under the arbitration provisions of the Netherlands-Czech Republic BIT). 34 WM Reisman, ‘Reflections on the control mechanism of the ICSID system’ in E Gaillard (ed), The Review of International Awards (New York, JurisNet, 2010) 197. 35 Judgment of the Swiss Federal Tribunal of 20 September 2016, above (n 25) paras 3.1.1, 3.1.2.; Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 4.4.1; Judgment of the Svea Court of Appeal (Sweden) of 29 October 2002, Svea Court of Appeal, No T 7192-01. 36 Judgment of the Swiss Federal Tribunal of 14 December 2017, TF, No 4A_157/2017 on an FET breach and violation of public policy (para 3.1). Judgment of the Paris Court of Appeals (France) of 21 February 2017, JDI 2017.1361, commentary by E Gaillard. 37 P Bernardini, The Enforcement of Awards Under the ICSID Convention, Festschrift Ahmed Sadek El Kosheri (Alphen aan den Rijn, Kluwer, 2015) 249. Regarding state court ordered provisional measures, see Judgment of 16 November 1986, Cour de cassation, Civ 1ère, Rev arb 1987.315, commentary G Flécheux; Judgment of 23 July 2018, Corte Suprema di Cassazione (Italy-Argentine BIT of 22 May 1990). 38 Judgment of the Cour de cassation (France) of 11 June 1991, Civ 1ère, Rev arb 1991, commentary A Broches.

622 Dominique Hascher of the Convention requires the Contracting Sates to recognise and enforce the pecuniary obligations imposed by an award as if it were a final judgment of a court in that state. In the Micula case, the Court of Appeal of England and Wales identified two approaches to Article 54(1). According to the minority view, an ICSID award is the equivalent for all purposes to a judgment of the High Court given in domestic proceedings, including grounds that would justify staying enforcement of a domestic judgment such as inconsistency with EU law. According to the majority view, a stay would only be permissible if temporary and consistent with the Convention.39 The UK Supreme Court admitted that in any event the proper interpretation is given by principles of international law applicable to all Convention Contracting States and cannot be affected by EU law.40 BIT interpretation is treated by state courts under the ground of jurisdiction as a matter of review under the New York Convention as well as under the arbitration laws of the courts whose decisions are reported. Jurisdiction and Admissibility must be distinguished. Having noted the absence of uniform treatment of the cooling off period in arbitral jurisprudence, the Svea Court of Appeal considered that the object of such provision of encouraging an amicable settlement in the ECT can be achieved during the arbitration with a stay, and is not a condition for the Tribunal’s jurisdiction.41 With regard to an ICSID Additional Facility award the French Cour de cassation held that the three-year limitation period the investor has for bringing a claim under the Venezuela-Canada BIT was a question of admissibility and not one of jurisdiction.42

A. Protected Investment and Protected Investor The French Cour de cassation decided that conditions of the ratione materiae and ratione personae scope of the BIT must be cumulatively met.43 Notwithstanding the lack of a unanimous definition of investment which was remarked by the Swiss Federal Tribunal,44 the presence of an investment has generally been defined in broad and extensive terms.45 The Svea Court of Appeal46 and the French Cour de cassation acknowledged the wide scope of a protected investment in the Energy Charter Treaty (ECT).47 The ECT provides an illustrative list of cases considered as investment. It does not set out the criteria for the definition of an investment. In Gold Reserve, the issue turned on the definition of investor under the Canada-Venezuela BIT of 1 July 1996. Venezuela, which resisted enforcement of the ICSID Additional Facility

39 [2018]

EWCA Civ. 1801, judgment of 27 July 2018. UKSC 5, judgment of 19 February 2020, paras 83–87. 41 Judgment of the Svea Court of Appeal (Sweden) 9 December 2016, above (n 19) para 5.3.2. 42 Judgment of the French Cour de cassation of 31 March 2021, Civ 1st, No 1911551. JC Witemberg, ‘La recevabilité des réclamations devant les juridictions internatioanles’ (1932) 41 Collected Courses of the Hague Academy of International Law. 43 Judgment of the Cour de cassation (France) of 13 February 2019, Civ 1st, No 1725851 (Spain-Venezuela BIT of 2 November 1995). 44 Judgment of the Swiss Federal Tribunal of 20 September 2016, above (n 25) para 3.4; Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 3.2.1.2.3; E Gaillard and Y Banifatemi, ‘The Long March towards a Jurisprudence Constante on the Notion of Investment’ in M Kinnear and others (eds), Building International Investment Law: The First 50 Years of ICSID (Alphen aan den Rijn, Kluwer, 2015) 97. 45 Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 3.1.1.1. 46 Judgment of the Svea Court of Appeal (Sweden) of 19 January 2007, Svea Court of Appeal, No T 5298-05. 47 Judgment of the French Cour de cassation of 28 March 2018, Civ 1st, No 1616568, JDI 2019.6, commentary E Gaillard. On a preliminary reference on the interpretation of an investment under the ECT (CA Paris, judgment of 24 September 2019, No RG 18/14721), see: ECJ Judgment of 2 September 2021 C-741/19. 40 [2020]

International Investment Law in European Courts 623 award made in Paris, submitted that Gold Reserve Inc (GRI), a Canadian company, was not an investor within the meaning of the BIT as it did not make any investment in the territory of the Republic but merely acquired an indirect ownership of shares and mining rights. The English Commercial Court replied: Were the definition of an investor, ‘one who makes the investment’, restricted to making the investment which created or acquired the asset the category of investments which are promoted and protected by the BIT would exclude investments which took the form of funding the development of assets in Venezuela where such investments were made by a person who, although the indirect owner or controller of such assets, had not paid to create or acquire such assets. This would not […] be consistent with the object and purpose of the BIT.48

As the Commercial Court further remarked, ‘Venezuelan assets such as rights to exploit mineral resources may need foreign investment not only to create or acquire them but also to develop and maintain them.’49 In Paris in the annulment proceedings of the same award, the definition of investor was also at issue with the Canadian nationality of GRI. The French Court held that GRI was not the vehicle used by its US affiliate, Gold Reserve Corp, in light of the fact that the majority of the latter’s shareholders were Canadian nationals notwithstanding an American management, and because of the Canadian origin of the funds which served to finance the Venezuelan project.50 In support of its application in Tatneft to set aside the partial award on jurisdiction before the Paris Court of Appeal, Ukraine argued that several ministers of the Government of the Republic of Tatarstan sat on the board of directors of Tatneft which was presided over by the Prime Minister and that the Republic was the major shareholder of the company which was financing the public economic policies of Tatarstan. Ukraine contended that the BIT investorstate arbitration agreement could not be used to settle differences between the Contracting State Parties themselves. The Court found that Tatneft was a commercial joint stock company publicly traded on the London and Moscow stock exchanges, with 64 per cent of shares owned by more than 50,000 private shareholders and the rest by the Republic of Tatarstan, which also had a golden share resulting in veto rights and appointment privileges. Under these circumstances, the Court concluded that Tatneft was not an emanation of the Republic of Tatarstan, itself a subdivision of the Russian Federation. Unlike the arbitral tribunal, the Court declined to refer to ILC Articles 8 and 9 on state responsibility which attribute to the state the conduct of a state owned or controlled entity when it exercises governmental authority.51 As remarked by the Paris Court of Appeal, state responsibility was not an issue. However, the two criteria put forward, confusion of assets with the state and lack of independent decisional power, used in French law for immunity of enforcement matters, are questionable in the context.52

48 Judgment

of the High Court of England and Wales of 2 February 2016, above (n 10) para 41. para 42. of the Paris Court of Appeals (France) of 7 February 2017, CA Paris, Rev arb 2017.566, commentary by J Fouret and A Reynaud. Companies controlled by nationals of the host state are eligible for investment protection under the ECT, The Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, para 5.1.8.10. 51 J Crawford, The International Law Commission ’s Articles on State Responsibility (Cambridge, Cambridge University Press, 2005). 52 Judgment of the Paris Court of Appeal (France) 29 November 2016, CA Paris, Rev arb 2017.500, commentary by L Franc-Menget. As remarked by the commentator, the choice made by the French Court distances itself from the ‘Broches test’ of acting as an agent of the government or discharging an essentially governmental function applied by investment arbitral jurisprudence in the context of a qualifying investor. 49 ibid

50 Judgment

624 Dominique Hascher Enforcement of the said award in the UK turned on the definition of a protected investment within the terms of the Ukraine-Russia BIT of 27 November 1998 for determining jurisdiction of the arbitral tribunal over Tatneft’s arbitration claims. According to Ukraine, Tatneft, which acquired shares of a Swiss and a US company, in each case from a shareholder which was a Seychellois company, made no investment within the territory of Ukraine. An investment may be made through an acquisition: The construction contended for by Ukraine would mean that if an investor from one Contracting State acquired what would in ordinary language be described as an investment which is located in the territory of the other Contracting State from someone other than a natural or legal person operating within the host state, then the new investor would have no ‘investment’ in the host state for the purposes of the BIT […]. If protection were confined to the original acquirer of the asset, and subsequent purchasers had none, this would be unlikely to promote the flows of capital.53

In a case where a German investor had channelled funds to an Indian company through its affiliate in Singapore, the Swiss Federal Tribunal held that ‘every kind of asset invested in the Germany-India BIT of 10 July 1995 does not require the investor to directly hold the assets in the host State’.54 Courts have thus refrained from giving a definition of an investment in the abstract and it is unsurprising therefore that contribution to the development of the host state economy has not been singled out as a criterion for the presence of an investment. The Cour de cassation55 as well as The Hague Court of Appeal56 rejected contribution as a condition of a protected investment under the ECT. The Swiss Federal Tribunal likely regarded contribution as an additional condition to the France-Vietnam BIT of 10 August 1994.57 True, the English Court in Gold Reserve required an active relationship between the investor and the investment in relation to the term ‘makes the investment’ in the Venezuela-Canada BIT of 1 July 1996,58 however, it conceded in PAO Tatneft that Gold Reserve was concerned with the meaning of investor and not that of an investment.59

B. Non-Precluded Measures Clause The Swiss Federal Tribunal considered the Non-Precluded Measures (NPM) clause which permits the adoption of measures otherwise precluded to the extent necessary for the protection of essential security interests as a substantive defence and not as a jurisdictional matter.60 The same solution would be achieved with a fair necessity defence under Article 25 ILC which justifies otherwise wrongful conduct.

53 Judgment

of the High Court of England and Wales of 13 July 2018, above (n 11) para 69. the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 3.2.1.2.3; Judgment of 7 February 2020, TF, No 4A_2018, para 4.6.2: ‘Often, one or more third parties come in between the investor and the investments; in such cases, the investment is an indirect investment’. 55 Judgment of the Cour de cassation (France) of 28 March 2018, No 1616568, JDI 2019.6, commentary E Gaillard; I Fadlallah, ‘La notion d’investissement: vers une restriction à la compétence du CIRDI ?’ in G Aksen (ed), Global Reflections on International Law, Commerce and Dispute Resolution: Liber amicorum in honoiur of Robert Briner (Paris, ICC Publishing, 2005) 259. 56 The Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, para 5.1.9.5. 57 Judgment of the Swiss Federal Tribunal of 20 September 2016, above (n 25) para 3.4.2. 58 Judgment of the High Court of England and Wales of 2 February 2016, above (n 10) para 37. 59 Judgment of the High Court of England and Wales of 13 July 2018, above (n 11) paras 77–81. 60 Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) paras 3.2.3.3.2–3.2.3.4. 54 Judgment of

International Investment Law in European Courts 625 C. FET Standard and MFN Provision The Fair and Equitable Treatment (FET) standard in international investment jurisprudence covers the decision-making process affecting the investor either before the courts of the host state or by state officials.61 Without giving reasons the Paris Court of Appeal affirmed in PAO Tatneft that competence over an FET claim is an issue on the merits and not one of jurisdiction.62 This isolated decision notwithstanding, competence is granted for any dispute relating to an investment and not according to a particular cause of action. In its efforts to resist enforcement of the same award in the UK, Ukraine alleged that, because of the deliberate omission of an FET obligation in the Ukraine-Russia BIT, it could not have agreed to arbitrate the breach of that standard. The English Commercial Court held that Ukraine agreed to arbitrate, not a particular type of disputes, but any dispute in connection with an investment, such as what type of substantive protections, like FET, are accorded by the BIT.63 In Ecuador v Occidental, the English Commercial Court had similarly regarded as a matter of jurisdiction whether the refusal to refund VAT to the investor was a breach of the Most Favoured Nation (MFN) clause, FET treatment and expropriation obligations.64 GPF GP v Poland concerned the challenge of a Stockholm Chamber of Commerce (SCC) award made in London under the arbitration provisions of the Poland-Belgium and Luxembourg BIT of 19 May 1987. The investor submitted that the arbitral tribunal’s jurisdiction extended to its claims for breach of the FET standard. The English Commercial Court agreed that ‘an FET claim based on measures involving a deprivation or restriction of property rights and which leads to/causes consequences similar to expropriation does fall within the scope of disputes that can be submitted to arbitration’.65 The Svea Court of Appeal considered that the MFN treatment in the USSR-Spain BIT of 26 October 1990 only related to the FET standard and could not grant the arbitral tribunal jurisdiction beyond the confines of the matters eligible for arbitration on the more favourable dispute resolution provisions of another investment protection treaty. The Swedish Court added that the FET standard does not include a right of access to an international arbitral tribunal for investors.66 There is no unanimous solution in investment jurisprudence where the MFN clause has been held a valid basis for granting access to the dispute resolution mechanism of a third-party treaty.67

D. Abuse of Rights for Claiming Protection Before the English Commercial Court, Ukraine contended that Tatneft only acquired interest in the minority shareholders of the Ukrainian joint-venture in order to gain the protection of 61 C McLachlan, L Shore and M Weiniger, International Investment Arbitration (Oxford, Oxford University Press, 2010) No 7.76, 7.182, 7.188. 62 Judgment of the Paris Court of Appeals (France) of 29 November 2016, above (n 52), commentary by L Franc-Menget; I Fadlallah, ‘Les composantes de l’incompétence arbitrale’ (2015) 2 Les Cahiers de l’Arbitrage 343. 63 Judgment of the High Court of England and Wales of 13 July 2018, above (n 11) paras 42–53. See s 9 SIA 1978 (‘Where a State has agreed in writing to submit a dispute which has arisen, or may arise, to arbitration, the State is not immune as respects proceedings in the courts of United Kingdom which relate to arbitration.’). 64 Judgment of the High Court of England and Wales of 2 March 2006, above (n 22) para 96. 65 Judgment of the High Court of England and Wales of 2 March 2018, above (n 20) para 98. 66 Judgment of the Svea Court of Appeal (Sweden) of 18 January 2016, Svea Court of Appeal, No T 9128-14, citing the two oppositely decided RosInvest Co v Russian Federation, Award on Jurisdiction (October 2007) and Berschader v Russian Federation Award on Jurisdiction (April 2006). 67 E Gaillard, La jurisprudence du CIRDI, Vol II (Paris, Pedone, 2010) 300–307. M Paparinskis, The International Minimum Standard and Fair and Equitable Treatment (Oxford, Oxford University Press, 2013) 181; C Crépet

626 Dominique Hascher the Ukraine-Russia BIT. The Court, however, read the offer to arbitrate in the BIT without restriction to the abusive nature of a claim: ‘[i]ssues of jurisdiction go to the existence or otherwise of a tribunal’s power to adjudge the merits of a dispute; issues of admissibility go to whether the tribunal will exercise that power in relation to the claims submitted to it’.68 The Paris Court of Appeal took the opposite approach in the annulment proceedings of the same award69 as well as in Gold Reserve70 and considered abuse of rights as a jurisdictional issue which could be reviewed. Although the Court left its decision unexplained, justification could be discovered in the existence of a fraud with a view to creating an artificial jurisdiction.71

E. Conformity and Compliance Clauses An investment falls under an investment treaty when it meets all the criteria laid down for a protected investment, including conformity with the host state legislation or acceptance by the host state’s authorities. The BGH in interpreting the concept of an ‘approved investment’ in the Germany-Thailand BIT of 13 December 1961 ruled that an acceptance regime is a matter of jurisdiction.72 The Paris Court of Appeal interpreted compliance with the law of the host state as relating to the host state law on investments. The alleged violations of the local legislation on lease which happen during the performance of the transaction are not a condition of the investment’s existence.73 As remarked by Professor Fadlallah, the scope of the treaty must not be unduly restricted. Once the investment squares with the conditions of the treaty, arguments regarding conformity are no longer an issue of jurisdiction but one on the merits, which address the qualitative requirement of the investment with an impact on the amount of damages.74 The Swiss Federal Tribunal on the one hand regarded conformity with the host state legislation as a requirement of the offer to arbitrate conditioning the existence of the arbitrator’s competence and, on the other hand, decided that the lack of delivery of a government licence to the investor may affect the quantum of reparation but does not exclude protection of an investment which is no longer in the entry phase.75 In the French annulment proceedings in PAO Tatneft, Ukraine challenged the legality of the investment with judgments that had annulled the investor’s participation in the Ukrainian joint venture, which Tatneft claimed to have been taken under the influence of the Ukrainian Government at the request of the Prosecutor General. The circumstances of the case may explain why the French court held illegality of the investment to be an issue on the merits, which escapes review as a jurisdictional matter.76

Daigremont, ‘Traitement national et traitement de la nation la plus favorisée dans la jurisprudence arbitrale récente relative à l’investissement international’ in C Leben, Le contentieux arbitral relatif à l’investissement (Paris, Librairie générale de droit et de jurisprudence, 2006) 107. 68 Judgment of the High Court of England and Wales of 13 July 2018, above (n 11) para 97. 69 Judgment of the Paris Court of Appeal (France) of 29 November 2016, above (n 52). 70 Judgment of the Paris Court of Appeal (France) of 7 February 2017, above (n 50). 71 Judgment of the Swiss Federal Tribunal of 7 February 2020, TF No 4A_80/2018 (treaty shopping practice). 72 Judgment of the German Bundesgerichtshof of 6 October 2016, above (n 12). Gaillard, above (n 67) 447–448. 73 Judgment of the Paris Court of Appeal (France) of 25 September 2008, above (n 24). See The Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, para 5.1.11.2 finding that, in the absence of a legality requirement in the ECT, there is no generally accepted principle of law following which the Arbitral Tribunal should decline jurisdiction, para 5.1.11.5. 74 I Fadlallah, ‘Retour sur investissement’ in L Lévy and Y Derains (eds), Liber amicorum en l’honneur de Serge Lazareff (Paris Pedone, 2011) 267. 75 Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) paras 4.4.1, 3.2.2.2.2. 76 Judgment of the Paris Court of Appeal (France) 29 November 2016, above (n 50) 267.

International Investment Law in European Courts 627 IV. THE TOOLS OF INTERPRETATION

The significant issues of investment law involved in the control of national courts over arbitral awards makes an inquiry into the means of interpretation of investment protection treaties implemented by the courts all the more necessary in the search for a common approach.

A. The 1969 Vienna Convention on the Law of Treaties It is an accepted view that the arbitration provisions of investment treaties are subject to Articles 31 and 32 of the Vienna Convention on the Law of Treaties (VCLT) for their interpretation.77 Even when they are not conventional rules between the BIT Contracting Parties, because one of them is not a Contracting State to the VCLT, Articles 31 and 32 are regarded as codifying international custom.78 Courts have submitted under their private international law rules, BIT arbitration provisions, which are not a treaty but an agreement between an investor and a state, to public international law rather than to the national law of the state against which an investor is arbitrating.79 The Swiss Federal Tribunal determined on the basis of Article 178 of the Swiss Private International Law that the ECT arbitration provisions are governed by Swiss law, which designates the ECT as an integral part of Swiss Law following its ratification.80 The English Commercial Court’s remark in Ecuador v Occidental that a BIT is not drafted with either the Common Law or Civil Law systems in mind is all the more pertinent in the interpretative context of an international covenant.81 In Gold Reserve, the English Commercial Court opposed the approach favouring the jurisdiction of the arbitral tribunal in commercial arbitration to treaty interpretation according to the VCLT.82 However debatable the accuracy of this pronouncement might be, the proposed rigor in BIT interpretation is largely counterbalanced when the object and purpose of the BIT is found in broad objectives such as the promotion of investment or favourable conditions for mutual investments,83 the conferment of rights on an investor, including the right to arbitrate and allowing the resolution of uncertainties of interpretation in favour of the investor.84 77 M Villiger, Commentary on the 1969 Vienna Convention on the Law of Treaties (Leiden, Martinus Nijhoff, 2009). B Conforti, ‘The Activities of National Judges and the International Relations of States’ (1993) 65(1) Annuaire de l’Institut de droit international 328. Judgment of the Court of Appeal (UK) of 4 July 2007 [2007] EWCA Civ 656, para 25; Judgment of the Paris Court of Appeal (France) of 18 November 2010, above (n 32); Judgment of the Paris Court of Appeal (France) of 29 November 2016, above (n 52); Judgment of the Paris Court of Appeals (France) of 7 February 2017, above (n 50). Judgment of the Swiss Federal Tribunal of 6 October 2015 No 4A_34/2015, para 3.5.1; Judgment of the Swiss Federal Tribunal of 16 October 2018, n° 4A_396/2017 (territorial and temporal scope of the Russia-Ukraine BIT of 27 November 1998) and of 2 November 2020, n° 4A_461/2019 (temporal scope of the Turkey-Libya BIT of 25 November 2009). Judgment of the Hague District Court (The Netherlands) of 20 April 2016, C/09/477160/HA ZA 15-1, C/09/477162/HA ZA 15-2, C/09/481619/HA ZA 15-112, para 5.9; The Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, para 4.2.1. 78 Judgment of the German Bundesgerichtshof of 6 October 2016, above (n 12) paras 25, 31, 32 (Thailand-Germany BIT); Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 2.4.1 (India-Germany BIT); Judgment of 25 March 2020, TF, No 4A_306/2019, para 3.4.1 (Spain-Venezuela BIT). 79 Judgment of the High Court of England and Wales of 29 April 2005, above (n 13) para 63; Judgment of the High Court of England and Wales of 9 September 2005, above (n 4) paras 33–41. 80 Judgment of the Swiss Federal Tribunal of 6 October 2015, above (n 77) paras 3.4.2. (with additional justification that Swiss law also applies as the law of the seat). 81 Judgment of the High Court of England and Wales of 2 March 2006, above (n 22) para 101. 82 Judgment of the High Court of England and Wales of 2 February 2016, above (n 10) para 22. 83 ibid para 42; Judgment of the High Court of England and Wales of 13 July 2018, above (n 11) para 69; Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 3.1.1.1. 84 Judgment of the High Court of England and Wales of 5 December 2007, above (n 22) para 23. See also Judgment the Court of Appeal (UK) of 4 July 2007, above (n 76) para 28.

628 Dominique Hascher The ordinary meaning of treaty terms must be ascertained in context. In Ecuador v Occidental, the English Commercial Court, after consideration of the US-Ecuador BIT Articles on taxation and on eligible disputes, held that all BIT provisions, and not only those under arbitration, applied in the context of the three exceptions relating to tax matters which could be the subject of disputes.85 The Svea Court of Appeal took the view that the terms in the USSR-Spain BIT of 26 October 1990 ‘any dispute’ and ‘relating to’ the amount or method of compensation for expropriation considered in isolation would indicate that any other disputes could be covered. The Swedish Court read the arbitration provisions together with the BIT provisions on expropriation to conclude that jurisdiction does not cover disputes concerning the conditions and existence of expropriation.86 As contrasted, the interpretation of the Czech-Belgium-Luxembourg Economic Union BIT of 24 April 1989 by the English Commercial Court in European Media Ventures focused on the meaning of the words in the context in which they came to be agreed. The Court concluded from a cross-reference of the BIT provisions on the protected investment and on jurisdiction that arbitrable matters are not limited to the single issue of the amount of compensation.87 This view is endorsed in arbitral jurisprudence.88 Article 31 VCLT does not direct the interpreter to reach a conclusion about the meaning and then check if it finds support in the context or the object and purpose of the treaty. It is one single operation.89 The Svea Court of Appeal stressed that the object and purpose are part of the interpretative process and not an independent means of interpretation.90 The object and purpose of a treaty, which can be sought in the preambular provisions, are regularly resorted to in the interpretative process.91 In GPF v Poland, the English Commercial Court considered the object and purpose of the BIT to ensure an ‘effet utile’ to the words of the treaty.92 In Gold Reserve, the English and French courts examined the object and purpose of the BIT of promoting investments for determining the presence of an investment ‘in the territory of Venezuela’.93 The French court remarked that such aim would be defeated if the assets acquired by Canadian investors from a third party which is neither Canadian nor Venezuelan were deprived of the BIT protection. Subsequent practice after conclusion of the treaty mentioned at Article 31(3)(b) VCLT is illustrated with the importance attached by the BGH to a note verbale of the Thai Ministry of

85 Judgment of the High Court of England and Wales of 2 March 2006, above (n 22) paras 91–96. On tax matters, see also Judgment of the Swiss Federal Tribunal of 7 February 2020, above (n 54) para 3. 86 Judgment of the Svea Court of Appeal (Sweden) of 18 January 2016, above (n 66). 87 Judgment of the High Court of England and Wales of 5 December 2007, above (n 22) paras 16, 36, 51. See also Judgment of the High Court of England and Wales of 2 March 2006, above (n 22) para 101. 88 Tza Yap Shum v Peru, Decision on Jurisdiction (19 June 2009) ICSID case No ARB/07/6. 89 Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 2.4.2. 90 Judgment of the Svea Court of Appeal (Sweden) of 9 December 2016, above (n 19) para 5.3.2. 91 Judgment of the High Court of England and Wales of 5 December 2007, above (n 22) para 20; Judgment of the High Court of England and Wales of 2 February 2016, above (n 10) para 41; Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 3.2.1.2.5; Judgment of the Swiss Federal Tribunal of 16 October 2018, n° 4A_396/2017, para 4.4.3. Judgment of the Hague District Court (The Netherlands) of 20 April 2016, above (n 77) para 5.19; The Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, para 4.5.15. 92 Judgment of the High Court of England and Wales of 2 March 2018, above (n 20) paras 53–55, 99. See also on ‘effet utile’ judgment of the High Court of England and Wales of 5 December 2007, above (n 22) para 37; Judgment of the Swiss Federal Tribunal of 6 October 2015, above (n 77) para 3.5.3.1 (distinction between Treaty claims and contract claims). 93 Judgment of the High Court of England and Wales of 2 February 2016, above (n 10) paras 30–41; Judgment of the Paris Court of Appeal (France) of 7 February 2017, above (n 50).

International Investment Law in European Courts 629 Foreign Affairs in interpreting the concept of ‘approved capital investment’ in the ThailandGermany BIT of 196194 and more recently in the Yukos case by The Hague Court of Appeal regarding the provisional application of the ECT by Russia.95 Article 32 VCLT should not come into play at the outset of interpretation. When the ordinary meaning stands without recourse to the supplementary means of interpretation of Article 32 VCLT,96 the preparatory works of the Treaty are merely used for confirmation.97 Comparative treaty practice is superfluous if interpretation succeeds under Article 31 VCLT.98 The English Commercial Court in European Media Ventures stressed that its task was to interpret the treaty rather than to interpret supplementary means of interpretation such as the other BITs signed by Czechoslovakia before and following the fall of the Communist power, or the documents of the Czech administration pre-dating the conclusion of the BIT.99 The documents of the ratification process were considered as supplementary means of interpretation although the ratification expresses the unilateral consent of the state to be bound by the treaty and is not controlling on the scope of the BIT.100 The English Commercial Court considered that the differences in translation in the authentic French and English versions of the Poland-Belgium and Luxembourg BIT did not result in different interpretations of the arbitrable matters under Article 31 VCLT.101 While these circumstances explain the oversight of Article 33 VCLT on interpretation of treaties authenticated in several languages, it is doubtful that the court, which mentioned expert evidence in case of disagreement on the translation, would have turned to Article 33 at all.102

B. National Investment Laws The challenge of an UNCITRAL award rendered in London in The Kyrgyz Republic v Stans Energy and Kutisay Mining turned on the meaning of ‘arising in the course of realization of investments’ in the Kyrgyz Investment Law of 27 March 2003. The interpretation of qualifying investment and of the dispute resolution provision required the interpretation of the Kyrgyz statute which, being an issue of foreign law, is treated as an issue of fact in English law. The

94 Judgment 95 The

of the German Bundesgerichtshof of 6 October 2016, above (n 12) para 32. Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, paras 4.5.28 and

5.1.8.11. 96 Judgment of the High Court of England and Wales of 2 March 2018, above (n 20) para 103; Judgment of the Hague District Court (The Netherlands) 20 April 2016, above (n 77) para 5.22. 97 Judgment of the Svea Court of Appeal (Sweden) of 18 January 2016, above (n 66); Judgment of the Swiss Federal Tribunal of 16 October 2018, n° 4A_396/2017, para 4.4.5; The Hague Court of Appeal, judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, paras 4.2.5, 4.5.34, 4.5.40. 98 Judgment of the Swiss Federal Tribunal of 11 December 2018, above (n 31) para 3.2.1.2.5. 99 Judgment of the High Court of England and Wales of 5 December 2007, above (n 22) paras 25, 26, 29–31. 100 Judgment of the Court of Appeal (UK) of 4 July 2007, above (n 77) para 27 (Department of State Letter of Submittal to the US Senate BIT Ecuador-US); Judgment of the High Court of England and Wales of 5 December 2007, above (n 22) para 27 (records of the Parliamentary ratification sessions of the Belgian-Luxembourg -Czech Republic BIT). Judgment of the Swiss Federal Tribunal of 20 September 2016, above (n 25) paras 3.2.2., 3.4.2. (French travaux parlementaires for the ratification of the Vietnam-France BIT). 101 Judgment of the High Court of England and Wales of 2 March 2018, above (n 20) paras 80–84. See Judgment of the Court of Appeal (UK) of 4 July 2007, above (n 77) para 29 with both linguistic versions of the BIT for ascertaining the meaning of the ordinary words of the Ecuador-US BIT. 102 Expert evidence for translation purposes, see Judgment of the High Court of England and Wales of 13 October 2017, above (n 23) paras 13–16. See also on the different linguistic versions of the ECT, The Hague Court of Appeal, Judgment of 18 February 2020, ECLI: NL: GHDHA: 2020: 234, para 5.2.21.

630 Dominique Hascher Commercial Court applied Kyrgyz principles of statutory interpretation with the assistance of expert evidence. Identification of the intention of the draftsman: is done, by identifying the plain (or literal) meaning of words as they are to be understood in their surrounding statutory context and having regard to the statutory purpose, which involves having regard not only to the individual words in isolation, but the sentence, the article and the law (statute).103

Having regard to the entirety of the definition of investment in that law and to the Preamble, as part of the statute even if not binding, the Court concluded from the breadth of the investments covered and the purpose of the law in question that all disputes connected to the implementation of investments are covered.104

V. CONCLUSION

According to the Swiss Federal Tribunal the lack of homogeneous solutions to investment law issues is explicable by the restricted control of state courts over arbitral awards and by their residual role on the scene due to the ICSID control system.105 ICSID ad hoc committees have divergent views as to the extent of their control and have not achieved any uniform interpretation of investment law, which would by far exceed their remit under the ICSID Convention.106 The fragmentation and contradictory trends of arbitral jurisprudence cannot be reproached to the courts. Arbitral jurisprudence is cited and discussed in English or Swiss judgments and we may not conclude from the more stripped judicial style of other jurisdictions that it has not come to the attention of the court. Scrutiny of awards for annulment or enforcement is not the type of action that would allow any supervisory body, including courts, to unify investment law. The narrow limits of award control notwithstanding, a comparative approach is useful in the search for some consensus.107 The submission of the arbitration clause to international law, rather than to the law of the host state, greatly favours the emergence of compatible solutions. The VCLT could be a unifying factor as a common interpretative instrument. This is no longer the case when investment laws are involved, although such laws regulate international investment relations. Procedurally treated as a factual matter,108 with no appeals on law, interpretation of national investment legislation leaves room for a wider diversity of opinions than investment treaties which enjoy the interpretative protection of the VCLT. A robust comparative law analysis should work towards the harmonisation of concepts used in investment laws and treaties in light of their similarity.

103 ibid

para 89. paras 94, 96, 104, 113, 117(6), 124, 131, 146. of the Swiss Federal Tribunal of 11 December 2018, above (n 30) para 3.2.2.2.1. 106 WM Reisman, Systems of Control in International Adjudication and Arbitration (Durham, Duke University Press, 1992). 107 E Gaillard, ‘Comparative Law in International Arbitration’ (2020) 1 Ius Comparatum 1. 108 Swiss Institute of Comparative Law, ‘The Application of Foreign Law in Civil Matters in the EU Member States and its Perspectives for the Future’ (Lausanne, Swiss Institute of Comparative Law, 2011). 104 ibid

105 Judgment

Index A Abuse of rights 233, 625–626 Achmea case 154, 208, 308, 469, 508–509, 511–512, 513–524, 527, 586 Accountability see Independent accountability mechanisms (IAMs) Africa national courts attempts to interfere in international arbitral proceedings 606–607 historical events leading to national court decisions 607–609 importance 616 rarity of investment decisions 603–604 South African constitutional guarantee against expropriation 610–612 tribunals review prior to African national proceedings 604–606 Zimbabwe’s land reform claims 609–610 regional courts and tribunals Common Court of Justice and Arbitration (CCJA) 612–614 East African Court of Justice (EACJ) 614–615 ECOWAS Court of Justice 615 SADC Tribunal establishment and mandate 530–532 landmark decisions 534–541 limitations of investment adjudication 543–544 Model Bilateral Investment Treaty Template 406 overview of cases 529–530 post-Cambell protection regime 541–543 three cases dealing with expropriation 534–541 Zimbabwe’s land reforms 532–534 Aliens see Protection of aliens Amici curiae interventions acceptance and underlying rationale 194–195 conditions for and limits on interventions Biwater Gauff v Tanzania 202 von Pezold v zimbabwe 200–201 contribution of IUSCT to interpretation of UNCITRAL procedural rules 68 ‘EU saga’ Electrabel v Hungary (2008) 206–208 importance 205–206 more recent cases 208–209 identity and legal nature of amici Glamis Gold v United States of America (2005) 203

overview 202–203 Phillip Morris v Uruguay 203–204 implications 193 increased acceptance in international dispute settlement 209 landmark decisions Aguas del Tunari v Bolivia (2005) 197–198 Biwater Gauff v Tanzania 202 Electrabel v Hungary (2008) 206–208 Glamis Gold v United States of America (2005) 203 Methanex v United States (2001) 195–196 Phillip Morris v Uruguay 203–204 Suez v Argentina (2005) 198–199 von Pezold v zimbabwe 200–201 overview 193–194 Ancient Greece dispute settlement 11 ‘private international law’ approach 11 protection of aliens’ property rights 10 treaties 10–11 two fundamental factors 10 Anti-arbitration injunctions advantages and disadvantages 572–573 common principles on domestic court’s authority 565–568 decisions issued by Indian courts Port Trust v Louis Dreyfus Armatures 558–560 UoI v Khaitan Holdings 563–564 UoI v Vodafone 560–563 developing jurisprudence 557 effect 557–558 effect on investment treaty obligations expropriation 571 fair and equitable treatment 570–571 overview 569 right to arbitrate 571–572 Port Trust v Louis Dreyfus Armatures facts of the case 558–559 grant of application 560 jurisdiction to make order 559–560 practical impact of injunctions 568–569 UoI v Khaitan Holdings 563–564 deference to an arbitral tribunal’s jurisdiction 564 facts of the case 563 jurisdiction to make order 563–564 UoI v Vodafone facts of the case 560–561

632 Index interaction of Indian Arbitration Act 561 jurisdiction to make order 561–562 parallel proceedings and abuse of process 562–563 Anti-suit injunctions see Anti-arbitration injunctions Applicable law choice of law 293–296 complexity 292 conclusions 309 contentious topic 291–292 counterclaims 281–283 EU law 307–308 general international law general principles 296–298 relationship between international and national law 300–304 history of investment law implicit rejection of municipal legal system in Concession agreements 29–31 international law as a corrective for insufficiently developed municipal law 27–31 Lena Goldfields arbitration 25 nationalisation of oil concessions in Libya and Kuwait 32–35 importance 291 national law general principles 304–306 relationship between international and national law 300–304 non-investment instruments 298–300 overview 292–293 shift from contract to treaty basis – SPP and AAPL 93–96 Arbitrariness CETA provisions 524 ECtHR 494 Eli Lilly v Canada 331 establishing an investment treaty breach 341 Glamis Gold v United States 331 ICJ influence on standards contained in IIAs impact on investment arbitration 51 reception in investment jurisprudence 49–51 relevant case 49 significance 48 Metalclad v Mexico 110 NAFTA rules 374 Tecmed v Mexico 366 Arbitrators see Challenges procedures Argentinian crisis cases background to economic crisis 120–121 breaches of BITS expropriation 124 fair and equitable treatment 124–125 umbrella clauses 125 importance 133 jurisdictional challenges

general economic measures emanate from the sovereign power 121–122 procedural objections to mass claims 122–123 justified violations 125–131 lack of compliance with awards 131–133 overview 119–120 Attribution AAPL case 84 contribution of IUSCT to substantive law 74–76 ILC’s articles 286 umbrella clauses 388, 391 B Bifurcated decisions central problem 180 discretionary process 179 fairness 178–179 landmark decisions Emmis v Hungary 186–188 Gavazzi v Romania 188–190 Glamis Gold Limited v United States 181–183 Philip Morris v Australia 183–186 overreaching principles 190 preliminary questions 179 rationale 178 regular use 177–178 relevance 190 ways in which the proceedings can be divided jurisdiction and merits 179 liability and quantum 179 Bilateral investment treaties (BITs) CJEU – preliminary ruling Slowakische Republik v Achmea BV background to judgement 513 Court’s decision 513–518 death-knell for intra-EU investment arbitration 512 seismic impact on investment arbitration 518–519 CJEU – Singapore Opinion background to request for Opinion 509 careful line between competing forces 512 CJEU Opinion 1/15 510 foreign non-direct investment 511–512 consent to arbitration 261–263 dispute resolution provisions of intra-EU BITs 469 EU standards of review 622 ‘investments’ defined 237–241 Mexican approach to constitutionality of investment agreements overview 591–592 parallel proceedings 595–601 synchronised effort to address investors complaints 601–602 willingness to comply with awards 592–595 protected investors corporate restructuring 255–257

Index dual nationality 249 minority shareholders 254 nationality requirement 243–246 shift from contract to treaty basis – SPP and AAPL consent as cornerstone of arbitral settlement 82–87 determination of applicable law 93–96 impact on domestic investment statutes 92–93 interpretation of IPPAs 96–97 overview 79–81 pioneering commentators and key players 87–92 twentieth century legal milestone 97 US standards of review BG Group v Argentina 577–579 Henry Schein v Archer & White 579–582 overview 576 violations in Argentinian crisis cases expropriation 124 fair and equitable treatment 124–125 umbrella clauses 125 C Challenges procedures applicable standards of review 212–214 dangers and disadvantages 211 essential control mechanism 211 importance 224–225 independence and impartiality behaviour of arbitrators 224 double hatting 220–222 familiarity with another participant 219–220 issue conflict 222–224 multiple and repeat appointments 216–219 overview 215–216 Commercial arbitration see International commercial arbitration Common Court of Justice and Arbitration (CCJA) 612–614 Companies corporate restructuring 255–257 minority shareholders 253–255 nationality 250 piercing the corporate veil 250–253 shareholder protection European Court of Human Rights 483–488 minority shareholders 253–255 role of World Court 40–43 umbrella clauses 391–394 Compensation see Reparations Compliance and enforcement African national courts 607–609 Argentinian crisis cases 131–133 benefits of bringing related claims together 285 decisions in EU courts 626–627

633

Mexican approach to constitutionality of investment agreements 592–595 sovereign immunity applicable law 475 French rules 151 matter governed by national law 149 relevance of counterclaims 286 US approach to sovereign immunity ‘arbitration exception’ to immunity 583–584 increasing scepticism about ISDS 588–589 Micula v Romania 585–588 Mobil v Venezuela 584–585 overview 582–583 Yukos cases 147–149 Consent to arbitration acceptance of offer, conditional consent 263–264 counterclaims overview 275–276 party consent to hearing counterclaims 276–280 denunciation of ICSID Convention 272–273 direct agreement of parties 260 indispensible condition 259 landmark decisions Ambiente Ufficio v Argentina 267–270 Amco v Indonesia 260 Churchill Mining v Indonesia and Planet Mining v Indonesia 261–263 Fábrica de Vidrios v Venezuela, 272–273 Generation Ukraine v Ukraine 263–264 Millicom v Senegal 270–271 Ping An v Belgium 264–265 Rumeli v Kazakhstan 271–272 SGS v Paraguay 266–267 SPP v Egypt 260–261 past events and disputes 264–265 scope of consent 266–267 shift from contract to treaty basis general consequences 259 SPP and AAPL 82–87 subject to procedural conditions 267–270 termination of contract containing consent 270–271 termination of legislation providing consent 271–272 through host state legislation 260–261 through treaties 261–263 waiver of state immunity 583 Constitutionality anti-arbitration injunctions 567 exhaustion of local remedies ‘futility exception’ 270 Zimbabwe 534 Mexican approach to constitutionality of investment agreements overview 591–592 parallel proceedings 595–601

634 Index synchronised effort to address investors complaints 601–602 willingness to comply with awards 592–595 ‘Rosatti Doctrine’ 131 South African constitutional guarantee against expropriation 610–612 Counterclaims advantages and disadvantages 287–288 benefits of bringing related claims together correction of asymmetry in investor-state dispute settlement i 286–287 procedural observations 284–286 contribution of IUSCT to interpretation of UNCITRAL procedural rules 66–67 importance of applicable law 281–283 overview 275 tribunal authority to apply counterclaims ‘close connection’ requirement 280–281 party consent to hearing counterclaims 276–280 Court of Justice of the European Union (CJEU) see also EU law CETA Opinion attempts to deal with competing demands 526–527 change of approach 519–520 compatibility of CETA with EU legal order 521–522 compatibility with effectiveness and right of access 524–526 compatibility with equal treatment 523–524 investment court system 520 no jurisdiction to interpret EU law 520–521 request for Opinion 520 threat to EU institutions 522–523 dispute resolution provisions of intra-EU Bits 469 landmark decisions CETA Opinion 519–526 Micula case 527–529 Singapore Opinion 509–512 Slowakische Republik v Achmea BV 512–519 possibility of future twist in investment arbitration 527–529 preliminary ruling on Slowakische Republik v Achmea BV background to judgement 513 Court’s decision 513–518 death-knell for intra-EU investment arbitration 512 seismic impact on investment arbitration 518–519 rapid and significant developments in the area of investment arbitration 507–508 Singapore Opinion background to request for Opinion 509 careful line between competing forces 512 CJEU Opinion 510 foreign non-direct investment 511–512 request for Opinion 509–510 Courts see International courts and tribunals; National courts and judges

D Damages see Reparations Denial of benefits clauses importance of decisions 421 interpretation 397–398 landmark decisions Amto v. Ukraine 412 Gran Colombia Gold Corp v, Republic of Colombia 415–416 Pac Rim v. El Salvador 404–407, 412–414 Plama Consortium Limited v Republic of Bulgaria 399–404 table of cases 421–425 Ulysseas, Inc. v. The Republic of Ecuador 404–407 nature of objections 397–398 benefit of jurisdictional treaty provisions 404–407 benefit of substantive treaty provisions 399–404 key distinctions 398–399 protected investors 255 substantive requirements for the exercise of a clause lack of ‘substantial business activities’ 410–416 overview 407–408 ownership or control by a national or third state 416–421 preliminary matters 408–410 table of cases 421–425 treaty-based tool 397 Yukos Oil cases 146 Diplomatic protection model alternative approaches 21 BITs as alternative 95 continuing prominence 132–133 current use 21 exhaustion of local remedies rule 270, 535 ILC Draft Articles 40 importance of defining nationality 243 issue of dual nationals 69, 247 limitations on shareholder protection under general international law 40–43 piercing of the corporate veil 488 procedural and substantive inadequacies 44 unfavourable towards claims brought on behalf of foreign investors 43–44 world Court’s jurisprudence 38 Dispute settlement see Investor-state dispute settlement (ISDS) E East African Court of Justice (EACJ) 614–615 Emergency clauses Argentinian crisis cases conflicting legal assessments 126–129 relationship with necessity 126–129 US standards of review 577–579 Enforcement see Compliance and enforcement Environmental protection applicable law 298–299 expropriation 356–358

Index interpretation of IPPAs 96 key principles 103 NAFTA Ch 11 cases Ethyl v Canada 102 Feldman v Mexico 112–114 Metalclad v Mexico 110–112 Methanex v United States 114–117 Pope & Talbot v Canada 105–109 remaining discretion and indeterminacy 117 S.D. Myers v Canada 102–105 observance of the investor’s treaty rights 292 World Bank Inspection Panel (WBIP) 554 Equal treatment theory compatibility of ISDS mechanism 523–524 fundamental to the idea of justice 465 Guiso Gallisay v Italy (ECtHR) 503 ‘mixed commissions’ 18 Equality of arms principle 465–466 Estoppel 240, 272, 378, 566, 619–620 Ethics behaviour of arbitrators 224 CJEU 508 commonalities with international commercial arbitration 467–468 European Court of Human Rights (ECtHR) see also Human rights cases intersecting with investor protection compensation 499–501 competing jurisdiction and parallel proceedings 482–483 just satisfaction 502–504 legal standing of shareholders 483–488 licences of a foreign investment 488–489 tax collection 495–499 tobacco retail licences 490–491 trademarks 491–494 disadvantages for investor protection 504–505 landmark decisions Anheuser-Busch Inc v Portugal 491–494 Bimer SA v Moldova 488–489 Guiso Gallisay v Italy 503–504 James v United Kingdom 499–501 Le Bridge v Moldova 482–483 Papamichalopoulos v Greece 499–501 Sovtransavto Holding v Ukraine 483–488 Vékony v Hungary 490–491 Yukos v Russia 495–499 margin of appreciation doctrine 331–332 overlap with standards of protection 481 reparations 447 European Union see also European Court of Human Rights (ECtHR) Court of Justice of the European Union (CJEU) CETA Opinion 519–526 possibility of future twist in investment arbitration 527–529 preliminary ruling on Slowakische Republik v Achmea BV 512–519

635

rapid and significant developments in the area of investment arbitration 507–508 Singapore Opinion 509–512 EU law applicable law 307–308 interpretation 307 RE ECT arbitrations 154 national courts and judges abuse of rights 625–626 conformity and compliance clauses 626–627 doctrine of non-justiciability 618–619 estoppel 619–620 fair and equitable treatment 625 ‘investments’ 621 lack of homogeneous solutions to investment law 630 Most-Favoured-Nation Treatment (MFN) clauses 625 Non-Precluded Measures (NPM) clauses 624 overview 617–618 protected investors 621–624 sovereign immunity 618 standards of review 621–622 tools of interpretation 627–630 Exhaustion of local remedies European Court of Human Rights 487, 504 ‘futility exception’ 270 MFN clauses 434 NAFTA rules 592, 595–596 reform proposals 476 removal of the condition 92 SADC Tribunal 534–535 South African courts 610 state consent 270 treaty interpretation 315 Yukos Oil cases 138 Zimbabwe 534 Expropriation see also Reparations African national courts South African constitutional guarantee against expropriation 610–612 Zimbabwe’s land reform claims 609–610 applicable law 297 Argentinian crisis cases 124 common elements of an expropriation clause and their significance 344–347 compensation damages for treaty breaches distinguished 54 general principles 53 contribution of IUSCT to substantive law 70–72 direct and indirect expropriation distinguished 348–352 effect of anti-arbitration injunctions 571 European Court of Human Rights compensation 499–501 just satisfaction 502–504 importance 342–344

636 Index landmark decisions ADC Affiliate Limited and ADC & ADMC Management Limited v The Republic of Hungary 344–347 Metalclad Corporation v United Mexican States 348–352 Methanex v United States of America 352–355 Philip Morris Brands SARL, Philip Morris Products SA, and Abal Hermanos SA v The Oriental Republic of Uruguay 352–355 Waste Management Inc v United Mexican States 352–355 Mexican approach to constitutionality of investment agreements 596–598 NAFTA Ch 11 cases Feldman v Mexico 112–114 Metalclad v Mexico 110 Methanex v United States 114–115 S.D. Myers v Canada 103 source of legitimacy crisis 99 property which may be expropriated 352–355 SADC Tribunal overview of cases 529–530 post-Cambell protection regime 541–543 three cases dealing with expropriation 534–541 Yukos cases Quasar De Valores and others v The Russian Federation 139–141 RosinvestCo v The Russian Federation 138–139 Zimbabwe’s land reforms African national courts 609–610 historical context 532–534 ICSID rules 534–541 F Fair and equitable treatment (FET) ‘a legal term of art’ 360 African national courts 605 Argentinian crisis cases 124–125 controlling arbitrators as law-makers Glamis Gold v United States 373–376 International Thunderbird Gaming v Mexico 376–379 L.F.H. Neer and Pauline Neer (U.S.A.) v United Mexican States 371–372 overview 370–371 core standard 359 decisions in EU courts 625 effect of anti-arbitration injunctions 570–571 effect of normative vagueness 379 empowering arbitrators as law-makers overview 362 Saluka v Czech Republic 368–370 Tecmed v Mexico 363–366 Waste Management II v Mexico 366–368 interpretation 360–361

landmark decisions Glamis Gold v United States 373–376 International Thunderbird Gaming v Mexico 376–379 L.F.H. Neer and Pauline Neer (U.S.A.) v United Mexican States 371–372 Saluka v Czech Republic 368–370 Tecmed v Mexico 363–366 Waste Management II v Mexico 366–368 NAFTA Ch 11 cases Metalclad v Mexico 111 source of legitimacy crisis 99 Renewable Energy ECT cases Charanne v Spain (2016) 152–156 Eiser v Spain (2017) 160–162 Masdar v Spain (2018) 162–164 RREEF v Spain (2019) 164–167 shaping of normative contours 361–362 significance of RE ECT cases content of FET 167–171 methodology for determining FET 168–169 G General international law applicable law 296–298 history of investment law as a corrective for insufficiently developed municipal law 27–31 implicit rejection of municipal legal system in Concession agreements 29–31 Lena Goldfields arbitration 25–26 nationalisation of oil concessions in Libya and Kuwait 32–35 significance of RE ECT cases 172–173 Greece see Ancient Greece H Health protection CETA provisions 522–523 expropriation 356–357 FET standard 332 NAFTA Ch 11 cases Ethyl v Canada 102 Feldman v Mexico 112–114 Metalclad v Mexico 110–112 Methanex v United States 114–117 Pope & Talbot v Canada 105–109 remaining discretion and indeterminacy 117 S.D. Myers v Canada 102–105 observance of the investor’s treaty rights 292 petitions to intervene as amicus 204 tobacco regulations adopted by Uruguay 299 Vékony v Hungary 490–491 World Bank Inspection Panel (WBIP) 549, 554

Index History of investment law Ancient Greece dispute settlement 11 ‘private international law’ approach 11 protection of aliens’ property rights 10 treaties 10–11 Argentinian crisis cases background to economic crisis 120–121 breaches of BITS 124–125 importance 133 jurisdictional challenges 121–123 justified violations 125–131 lack of compliance with awards 131–133 overview 119–120 ICJ influence on standards contained in IIAs defining concept of ‘arbitrariness’ 48–51 most-favoured nation clauses 45–48 importance 20 Iran-United States Claims Tribunal contribution to interpretation of UNCITRAL procedural rules 65–68 contributions to substantive law of international investment 68–76 institutional nature of Tribunal 62–65 origins 59–62 pioneering work in formulating legal principles and tests 76–77 ‘Mixed commissions’ contribution to the development of international investment law 14–15 four fundamental features 15–16 individual autonomous standing 16–17 interpretation of treaties 18–20 Jay Treaty 1794 14 procedure 16–17 strong colonial discourse 15 substantive rights 17–18 NAFTA Ch 11 cases Ethyl v Canada 102 exceptional influence of early cases 101–102 Feldman v Mexico 112–114 Metalclad v Mexico 110–112 Methanex v United States 114–117 Pope & Talbot v Canada 105–109 remaining discretion and indeterminacy 117 S.D. Myers v Canada 102–105 source of legitimacy crisis 99–101 overview 9 remedies defining feature of investment disputes 51–52 general principles of reparations 52–53 impact on investment arbitration 56 reception in investment arbitration 53–56 Renewable Energy ECT cases Charanne v Spain (2016) 156–160

637

effect of Spain’s state regulatory measures 152–156 Eiser v Spain (2017) 160–162 Masdar v Spain (2018) 162–164 overview 153–154 RREEF v Spain (2019) 164–167 shift in arbitral decisions away from the unworkable thresholds of earlier decisions 173–174 significance for ECT 167–171 significance for investment law 171–172 significance for public international law 172–173 role of arbitration during the Middle Ages fundamental trends 13 inter-state arbitrations 12 private arbitration 12–13 two different and opposite tendencies 12 role of World Court investment protection via internationalised contracts 38–39 limitations on shareholder protection under general international law 40–43 overview 37–38 varied in scope and over time 57 Rome’s approach to international law 11 shift from contract to treaty basis – SPP and AAPL consent as cornerstone of arbitral settlement 82–87 determination of applicable law 93–96 impact on domestic investment statutes 92–93 interpretation of IPPAs 96–97 overview 79–81 pioneering commentators and key players 87–92 twentieth century legal milestone 97 State contracts early arbitrations concerning petroleum concessions 26–31 legal nature and specific legal regime 22 Lena Goldfields arbitration 22–26 nationalisation of oil concessions in Libya and Kuwait 31–35 origins 21–22 use of sovereign powers to avoid obligations 35 Yukos Oil cases enforcement saga 147–149 expropriation without compensation 138–141 fork-in-the-road provision of article 26(3)(b)(i) ECT 141–145 impetus for reform 149–152 new confidential ‘second-wave’ trilogy 145–146 overview 135–136 possible inconsistencies from concurrent jurisdiction 147 violations of human rights 137–138

638 Index Human rights see also European Court of Human Rights (ECtHR) AAPL v Sri Lanka 96 amici curiae interventions 200–201 applicable law 298–300 Argentinian crisis cases 120 ECOWAS Court of Justice 615 interpretation of IPPAs 96 relevance to investment disputes 4 role of World Court 41 SADC Tribunal jurisdiction and mandate 531 overview of cases 530 post-Cambell protection regime 542–543 World Bank Inspection Panel (WBIP) 552–554 Yukos Oil cases biggest divide 147 compliance and enforcement 147–151 Hulley Enterprises Limited (Cyprus), Yukos Universal Limited (isle of man), and Veteran Petroleum Limited (Cyprus) 143 OAO Neftyanaya Kompaniya Yukos v Russia 137–138 RosInvestCo UK Ltd v The Russian Federation 138–139 I ICSID rules amici curiae interventions acceptance and underlying rationale 194 Biwater Gauff v Tanzania 202 identity and legal nature of amici 203 Phillip Morris v Uruguay 203–204 Suez v Argentina (2005) 198–199 von Pezold v zimbabwe 200–201 applicable law 294 Argentinian crisis cases 131–133 balance between investor protection and human rights 4 challenges procedures applicable standards of review 212–214 behaviour of arbitrators 224 double hatting 221–222 issue conflict 222–224 multiple and repeat appointments 216 who makes the decision 214–215 consent to arbitration 272–273 contribution of IUSCT 64 counterclaims benefits of bringing related claims together 287 ‘close connection’ requirement 280 party consent to hearing counterclaims 277–278 differences between international commercial and investment arbitration 474 first precedent for submission to ICSID jurisdiction 83 independence and impartiality of arbitrator behaviour of arbitrators 224–225

double hatting 220–222 familiarity with another participant 219–220 importance 215–216 issue conflict 222–224 multiple and repeat appointments 216–219 ‘investment’ notion exclusion of commercial transactions 232–233 good faith element 233–234 key requirements 227–231 lack of homogeneous solutions to investment law 630 mass claims 121–123 protected investors companies 250–253 corporate restructuring 256 dual nationality 247 minority shareholders 254 nationality requirement 243–246 Renewable Energy ECT cases 163 shift from contract to treaty basis – SPP and AAPL 79–81 sovereign immunity 618 specific remedies 31 Zimbabwe’s land reform claims 534–541 Independence and impartiality of arbitrators behaviour of arbitrators 224–225 commonalities with international commercial arbitration 466–467 double hatting 220–222 familiarity with another participant 219–220 importance 215–216 issue conflict 222–224 multiple and repeat appointments 216–219 Independent accountability mechanisms (IAMs) means of redress 545 morse Commission report 1992 546 umbrella concept 545 World Bank Inspection Panel (WBIP) see World Bank Inspection Panel (WBIP) mandate 546 overview of decisions since 1994 546–547 India anti-arbitration injunctions common principles on domestic court’s authority 565–568 effect 557–558 effect on investment treaty obligations 569–572 practical impact of injunctions 568–569 landmark decisions Port Trust v Louis Dreyfus Armatures 558–560 UoI v Khaitan Holdings 563–564 UoI v Vodafone 560–563 Injunctions see Anti-arbitration injunctions Inter-state arbitrations investor-state arbitration distinguished 562 role of arbitration during the Middle Ages 12

Index International Centre for Settlement of Investment Disputes (ICSID) see ICSID rules International commercial arbitration applicable law 309 bifurcated decisions 177 commonalities with investment arbitration appointment of and challenges to arbitrator 466–467 equality of arms principle 465–466 Kompetenz-Kompetenz 470–471 professional ethics 467–468 seat of arbitration 468–470 comparative approaches England 627 India 559–560, 566 Mexico 600 United States 578–579 counterclaims 280 divergencies from investment arbitration methodology 471–476 procedure 476–480 impact of AAPL and SPP cases 88, 91 importance 463–464 independence and impartiality of arbitrator 219 investment arbitration distinguished 517 procedural genesis for IIL 215 shortcomings of enforcement regime 148 substantial influence on investment arbitration 481 sui generis normative framework of FET 377–378 third-party intervention rules 548 International Court of Justice (ICJ) see also World Court ‘close connection’ requirement for counterclaims 280 influence on standards contained in IIAs defining concept of ‘arbitrariness’ 48–51 most-favoured nation clauses 45–48 overview 44–45 Judgement on Bolivia v Chile 172–173 limitations on shareholder protection under general international law influence on contemporary IIL 42–43 most fundamental principle 40–42 International courts and tribunals see also National courts and judges African regional courts and tribunals Common Court of Justice and Arbitration (CCJA) 612–614 East African Court of Justice (EACJ) 614–615 ECOWAS Court of Justice 615 Court of Justice of the European Union see also EU law CETA Opinion 519–526 dispute resolution provisions of intra-EU Bits 469 possibility of future twist in investment arbitration 527–529

639

preliminary ruling on Slowakische Republik v Achmea BV 512–519 rapid and significant developments in the area of investment arbitration 507–508 Singapore Opinion 509–512 European Court of Human Rights see also Human rights compensation 499–501 competing jurisdiction and parallel proceedings 482–483 disadvantages for investor protection 504–505 just satisfaction 502–504 legal standing of shareholders 483–488 licences of a foreign investment 488–489 margin of appreciation doctrine 331–332 overlap with standards of protection 481 reparations 447 tax collection 495–499 tobacco retail licences 490–491 trademarks 491–494 ICJ ‘close connection’ requirement for counterclaims 280 defining concept of ‘arbitrariness’ 48–51 influence on standards contained in IIAs 44–45 Judgement on Bolivia v Chile 172–173 limitations on shareholder protection under general international law 40–43 most-favoured nation clauses 45–48 SADC Tribunal establishment and mandate 530–532 historical context – Zimbabwe’s land reforms 532–534 limitations of investment adjudication 543–544 Model Bilateral Investment Treaty Template 406 overview of cases 529–530 post-Cambell protection regime 541–543 three cases dealing with expropriation 534–541 World Bank Inspection Panel (WBIP) case law on mining projects 549–552 current reform processes 554 establishment 546 human rights impacts 552–554 inclusive approach to local communities 547–549 investment arbitration distinguished 548 mandate 546 overview of decisions since 1994 546–547 World Court investment protection via internationalised contracts 38–39 limitations on shareholder protection under general international law 40–43 overview 37–38 remedies 51–56 varied impact in scope and over time 57 International investment law (IIL) commonalities with international commercial arbitration

640 Index appointment of and challenges to arbitrator 466–467 equality of arms principle 465–466 Kompetenz-Kompetenz 470–471 professional ethics 467–468 seat of arbitration 468–470 contribution of IUSCT to substantive law attribution 74–76 compensation 72–74 expropriation 70–72 nationality 68–70 divergencies from international commercial arbitration methodology 471–476 procedure 476–480 emergence of contemporary IIL through World Court 38–39 history see History of international investment law significance of RE ECT cases 171–172 Interpretation contribution of IUSCT to interpretation of UNCITRAL procedural rules 65–68 denial of benefits clauses 397–398 difficult systemic questions of treaty law 108–109 EU law 307 EU standards of review 622 fair and equitable treatment 360–361 Lena Goldfields arbitration 25–26 Mexican approach to constitutionality of investment agreements 596–598 ‘mixed commissions’ 18–20 shift from contract to treaty basis – SPP and AAPL 96–97 significance of RE ECT cases 169–171 tools used by European courts national investment laws 629–630 VCLT 1969 627–629 treaty interpretation dual function ascribed to treaties 325–326 interaction with jurisdiction and annulment 320–323 interpretation as a legal dispute per se 323–325 interpretation as a shared power 319–320 overview 311–312 VCLT framework 312–318 UNCITRAL rules 64 VCLT framework interpretative methods 317–318 relationships between primary and secondary means of interpretation 312–317 ‘Investment’ decisions in EU courts 621 under the ICSID Convention exclusion of commercial transactions 232–233 good faith element 233–234 key requirements 227–231 IIA provisions 237–241 overview 227

Investor-state dispute settlement (ISDS) benefits of bringing related claims together 286–287 bifurcated decisions Emmis v Hungary 186–188 Gavazzi v Romania 188–190 CJEU – CETA Opinion attempts to deal with competing demands 526–527 change of approach 519–520 compatibility of CETA with EU legal order 521–522 compatibility with effectiveness and right of access 524–526 compatibility with equal treatment 523–524 investment court system 520 no jurisdiction to interpret EU law 520–521 possibility of future twist in investment arbitration 527–529 request for Opinion 520 threat to EU institutions 522–523 CJEU – preliminary ruling Slowakische Republik v Achmea BV 512–519 CJEU – Singapore Opinion 511–512 impact of Yukos Oil cases 149–152 increasing US scepticism 588–589 Most-Favoured-Nation Treatment (MFN) clauses 434–436 NAFTA Ch 11 cases 101–102 Ethyl v Canada 102 exceptional influence of early cases 101–102 Feldman v Mexico 112–114 Metalclad v Mexico 110–112 Methanex v United States 114–117 Pope & Talbot v Canada 105–109 remaining discretion and indeterminacy 117 S.D. Myers v Canada 102–105 source of legitimacy crisis 99–101 shift from contract to treaty basis – SPP and AAPL consent as cornerstone of arbitral settlement 82–87 determination of applicable law 93–96 impact on domestic investment statutes 92–93 interpretation of IPPAs 96–97 overview 79–81 pioneering commentators and key players 87–92 twentieth century legal milestone 97 State contracts see State contracts Iran-United States Claims Tribunal contribution to interpretation of UNCITRAL procedural rules bold and ingenious choice of rules 65–66 counterclaims 66–67 participation of amici curiae 68 contributions to substantive law of international investment attribution 74–76 compensation 72–74

Index expropriation 70–72 nationality 68–70 institutional nature of Tribunal 62–65 origins 59–62 pioneering work in formulating legal principles and tests 76–77 J Judges see National courts and judges Judicial review see Standards of review Jurisdiction African national courts 606 anti-arbitration injunctions common principles on domestic court’s authority 565–568 Port Trust v Louis Dreyfus Armatures 559–560 UoI v Khaitan Holdings 563–564 UoI v Vodafone 561–562 bifurcated decisions Emmis v Hungary 186–188 Glamis Gold Limited v United States 181–183 Philip Morris v Australia 183–186 ways in which the proceedings can be divided 179 challenges in Argentinian crisis cases general economic measures emanate from the sovereign power 121–122 procedural objections to mass claims 122–123 consent to arbitration see Consent to arbitration denial of benefits clauses 404–407 European Court of Human Rights 482–483 function of consuls 12–13, 20 ICSID 83–90 interaction with treaty interpretation 320–323 ‘investments’ – ICSID exclusion of commercial transactions 232–233 good faith element 233–234 key requirements 227–231 Iran-United States Claims Tribunal 61–62 Lena Goldfields arbitration 24–25 MFN clauses 46–47 oil development contracts 39, 43 Petroleum Development v Abu Dhabi 27 possible inconsistencies from concurrent jurisdiction 147 SADC Tribunal 530–532 tribunal authority to apply counterclaims ‘close connection’ requirement 280–281 party consent to hearing counterclaims 276–280 umbrella clauses 383–384 US Nationals in Morocco case 46 Yukos Oil case 141–143 K Kompetenz-Kompetenz anti-arbitration injunctions UoI v Khaitan Holdings 564 UoI v Vodafone 563–564

641

impact of international commercial arbitration on ILL 470–471 Lena Goldfields arbitration 24–25 US standards of review 579–582 L Landmark decisions amici curiae interventions Aguas del Tunari v Bolivia (2005) 197–198 Biwater Gauff v Tanzania 202 Electrabel v Hungary (2008) 206–208 Glamis Gold v United States of America (2005) 203 Methanex v United States (2001) 195–196 Phillip Morris v Uruguay 203–204 Suez v Argentina (2005) 198–199 von Pezold v Zimbabwe 200–201 Ancient Greece 11 anti-arbitration injunctions Port Trust v Louis Dreyfus Armatures 558–560 UoI v Khaitan Holdings 563–564 UoI v Vodafone 560–563 Argentinian crisis cases see Argentinian crisis cases bifurcated decisions Emmis v Hungary 186–188 Gavazzi v Romania 188–190 Glamis Gold Limited v United States 181–183 Philip Morris v Australia 183–186 consent to arbitration Ambiente Ufficio v Argentina 267–270 Amco v Indonesia 260 Churchill Mining v Indonesia and Planet Mining v Indonesia 261–263 Fábrica de Vidrios v Venezuela, 272–273 Generation Ukraine v Ukraine 263–264 Millicom v Senegal 270–271 Ping An v Belgium 264–265 Rumeli v Kazakhstan 271–272 SGS v Paraguay 266–267 SPP v Egypt 260–261 counterclaims 277 Court of Justice of European Union CETA Opinion 519–526 Micula case 527–529 Singapore Opinion 509–512 Slowakische Republik v Achmea BV 512–519 defined 2 denial of benefits clauses Amto v. Ukraine 412 Gran Colombia Gold Corp v, Republic of Colombia 415–416 Pac Rim v. El Salvador 404–407, 412–414 Plama Consortium Limited v Republic of Bulgaria 399–404 table of cases 421–425 Ulysseas, Inc. v. The Republic of Ecuador 404–407

642 Index early arbitrations concerning petroleum concessions Petroleum Development v Abu Dhabi 27–28 Sapphire International v NIOC 29–31 Saudi Arabia and ARAMCO 28–29 European Court of Human Rights Anheuser-Busch Inc v Portugal 491–494 Bimer SA v Moldova 488–489 Guiso Gallisay v Italy 503–504 James v United Kingdom 499–501 Le Bridge v Moldova 482–483 Papamichalopoulos v Greece 499–501 Sovtransavto Holding v Ukraine 483–488 Vékony v Hungary 490–491 Yukos v Russia 495–499 expropriation ADC Affiliate Limited and ADC & ADMC Management Limited v The Republic of Hungary 344–347 Metalclad Corporation v United Mexican States 348–352 Methanex v United States of America 352–355 Philip Morris Brands SARL, Philip Morris Products SA, and Abal Hermanos SA v The Oriental Republic of Uruguay 352–355 Waste Management Inc v United Mexican States 352–355 fair and equitable treatment Glamis Gold v United States 373–376 International Thunderbird Gaming v Mexico 376–379 L.F.H. Neer and Pauline Neer (U.S.A.) v United Mexican States 371–372 Saluka v Czech Republic 368–370 Tecmed v Mexico 363–366 Waste Management II v Mexico 366–368 Lena Goldfields arbitration arbitration proceedings and outcome 24–26 the Concession agreement 23–24 early transnational arbitration 22 Mexican approach to constitutionality of investment agreements GAMI Case 596–598 KBR (Commisa) Case 599–601 Most-Favoured-Nation Treatment (MFN) clauses Bayindir v Pakistan 432 CME v Czech Republic 433 Maffezzini v Spain 434–435 Magdenli v Kazakstan 432 Occidental v Ecuador 429–431 Parkerings v Lithuania 431 Plama v Bulgaria 435–436 Pope & Talbot Inc. v Canada 433 NAFTA Ch 11 cases Ethyl v Canada 102 Feldman v Mexico 112–114 Metalclad v Mexico 110–112 Methanex v United States 114–117

Pope & Talbot v Canada 105–109 remaining discretion and indeterminacy 117 S.D. Myers v Canada 102–105 National Treatment (NT) clauses ADF Group v Buy America 440 AES v Hungary 439–440 Cargill v Mexico 441 Mercer v Canada 440 Methanex v U.S.A. 437, 441–442 Occidental v Ecuador 438 Pope & Talbot Inc. v Canada 438, 439, 442 SD Myers v Canada 438–439, 442 Renewable Energy ECT cases Charanne v Spain (2016) 152–156 Eiser v Spain (2017) 160–162 Masdar v Spain (2018) 162–164 RREEF v Spain (2019) 164–167 reparations ADC v Hungary 444–449 LG&E v Argentina 454–457 Tidewater v Venezuela 457–459 Yukos v Russia 449–453 role of arbitration during the Middle Ages Honoré de Cuges and Joseph Vaccon of Pisa 1615 13–14 Nau dels Barquers case 1417–1422 13 SADC Tribunal 534–541 shift from contract to treaty basis – SPP and AAPL consent as cornerstone of arbitral settlement 82–87 determination of applicable law 93–96 impact on domestic investment statutes 92–93 interpretation of IPPAs 96–97 overview 79–81 pioneering commentators and key players 87–92 twentieth century legal milestone 97 sovereign immunity Micula v Romania 585–588 Mobil v Venezuela 584–585 standards of review Bilcon v Canada 334–336 Eli Lilly v Canada 339–340 Glamis Gold v United States 330–331 Marfin v Cyprus 333–334 OAO Tatneft v Ukraine 338–339 Philip Morris v Uruguay 331–333 SD Myers v Canada 330 treaty interpretation Daimler Financial services AG v Argentine Republic 315–317 Hrvatska Elektroprivreda D.D v Republic of Slovenia 313–315 Malaysian Historical Salvors, SDN, BHD v Malaysia 320–323 Mondev International Ltd. v United States of America 317–318

Index Pope & Talbot v Government of Canada 319–320 Republic of Ecuador v United States of America 323–325 umbrella clauses BIVAC v Paraguay and SGS v Paraguay 388–391 Burlington v Ecuador 391–394 SGS v Philippines 385–388 Vivendi v Argentina (I) 382–384 US standards of review BG Group v Argentina 577–579 Henry Schein v Archer & White 579–582 World Bank Inspection Panel (WBIP) Bear Creek Mining Corp v Republic of Peru 547 Chad: Petroleum Development and Pipeline Project 552–553 Copper Mesa Mining Corporation v Republic of Ecuador 547 Defensa y Conservacion Ecologica de Intag (DECOIN)) 550–551 Honduras: Land Administration Project 553–554 Mining Infrastructure Support – Additional Financing 551–552 Mongolia: Mining Infrastructure Investment Support 551–552 Urbaser v Argentina, or Burlington v Ecuador 552–553 Yukos Oil cases see Yukos Oil cases Legitimate expectations see also Fair and equitable treatment (FET) Argentinian crisis cases 124–125 ICJ Judgement on Bolivia v Chile 172–173 NAFTA Ch 11 cases 111 significance of RE ECT cases 154, 171–172 Yukos Oil cases 144 M Mexico approach to constitutionality of investment agreements 591–592 NAFTA Ch 11 cases Feldman v Mexico 112–114 Metalclad v Mexico 110–112 parallel proceedings 595–601 synchronised effort to address investors complaints 601–602 willingness to comply with awards 592–595 Minimum standards see Standards of investment protection ‘Mixed commissions’ contribution to the development of international investment law 14–15 individual autonomous standing 16–17 strong colonial discourse 15 Most-Favoured-Nation Treatment (MFN) clauses application to establish more favourable treatment application to ISDS 434–436

643

as basis for claiming general treatment 431–432 as basis for disregarding limitations to treatment 433 invoked to benefit from better specific treatment 429–431 decisions in EU courts 625 ICJ influence on standards contained in IIAs cautious take on internationalised contracts 45–46 decreased demand for an ICJ pronouncement 48 impact on investment arbitration 46–48 landmark decisions Bayindir v Pakistan 432 CME v Czech Republic 433 Maffezzini v Spain 434–435 Magdenli v Kazakstan 432 Occidental v Ecuador 429–431 Parkerings v Lithuania 431 Plama v Bulgaria 435–436 Pope & Talbot Inc. v Canada 433 reception in investment arbitration 46–47 underlying rationale 427–429 Yukos Oil cases 138 N NAFTA rules amici curiae interventions Aguas del Tunari v Bolivia (2005) 197–198 Glamis Gold v United States of America (2005) 203 Methanex v United States (2001) 195–196 applicable law choice of law 294 Ch 11 cases Ethyl v Canada 102 exceptional influence of early cases 101–102 Feldman v Mexico 112–114 Metalclad v Mexico 110–112 Methanex v United States 114–117 Pope & Talbot v Canada 105–109 remaining discretion and indeterminacy 117 S.D. Myers v Canada 102–105 source of legitimacy crisis 99–101 challenges procedures 221 counterclaims benefits of bringing related claims together 286 party consent to hearing counterclaims 278 Mexican approach to constitutionality of investment agreements parallel proceedings 595–601 willingness to comply with awards 593–594 National courts and judges see also International courts and tribunals African regional courts and tribunals Common Court of Justice and Arbitration (CCJA) 612–614

644 Index East African Court of Justice (EACJ) 614–615 ECOWAS Court of Justice 615 anti-arbitration injunctions advantages and disadvantages 572–573 common principles on domestic court’s authority 565–568 developing jurisprudence 557 effect 557–558 effect on investment treaty obligations 569–572 Port Trust v Louis Dreyfus Armatures 558–560 practical impact of injunctions 568–569 UoI v Khaitan Holdings 563–564 UoI v Vodafone 560–563 applicable law 304–306 diversity of perspectives 4 EU courts abuse of rights 625–626 conformity and compliance clauses 626–627 doctrine of non-justiciability 618–619 estoppel 619–620 fair and equitable treatment 625 ‘investments’ 621 lack of homogeneous solutions to investment law 630 Most-Favoured-Nation Treatment (MFN) clauses 625 Non-Precluded Measures (NPM) clauses 624 overview 617–618 protected investors 621–624 sovereign immunity 618 standards of review 621–622 tools of interpretation 627–630 importance 5 Mexican approach to constitutionality of investment agreements overview 591–592 parallel proceedings 595–601 synchronised effort to address investors complaints 601–602 willingness to comply with awards 592–595 relationship between international and national law 300–304 standards of review determinations of national law 334–336 domestic determinations involving non-legal expertise or discretion 330 domestic judicial decisions 336–340 overarching themes 340–341 US approach to sovereign immunity ‘arbitration exception’ to immunity 583–584 increasing scepticism about ISDS 588–589 Mobil v Venezuela 584–585 overview 582–583 US standards of review BG Group v Argentina 577–579 Henry Schein v Archer & White 579–582 overview 576

National Treatment (NT) clauses application to establish more favourable treatment 440 landmark decisions ADF Group v Buy America 440 AES v Hungary 439–440 Cargill v Mexico 441 Mercer v Canada 440 Methanex v U.S.A. 437, 441–442 Occidental v Ecuador 438 Pope & Talbot Inc. v Canada 438, 439, 442 SD Myers v Canada 438–439, 442 less favourable treatment in case law 439–440 likeness in the case law 437–439 major trends and conflicts 442 NAFTA Ch 11 cases Feldman v Mexico 113–114 Methanex v United States 115–116 Pope & Talbot v Canada 105–109 S.D. Myers v Canada 103 source of legitimacy crisis 99–101 overview 436–437 regulatory purpose and host state justification 441–442 underlying rationale 427–429 Nationality companies 250 contribution of IUSCT to substantive law 68–70 natural persons dual nationality 246 key requirement 243–246 non-ICSID arbitration 248–249 Necessity Argentinian crisis cases conflicting legal assessments 126–129 damages 130–131 relationship with emergency clauses 129–130 reparations 454–457 Non-discrimination clauses see Most-FavouredNation Treatment (MFN) clauses; National Treatment (NT) clauses Non-justiciability 618–619 Non-Precluded Measures (NPM) clauses Argentinian cases 119, 454 circumstances precluding wrongfulness distinguished 460 decisions in EU courts 624 North American Free Trade Agreement (NAFTA) see NAFTA rules O OHADA Common Court of Justice and Arbitration (CCJA) 612–614 P Parallel proceedings anti-arbitration injunctions

562–563

Index European Court of Human Rights 482–483 Mexican approach to constitutionality of investment agreements 595–601 Permanent Court of International Justice (PCIJ) see World Court Procedure amici curiae interventions acceptance and underlying rationale 194–195 Aguas del Tunari v Bolivia (2005) 197–198 conditions for and limits on interventions 199–202 ‘EU saga’ 205–209 identity and legal nature of amici 202–204 implications 193 increased acceptance in international dispute settlement 209 Methanex v United States (2001) 195–196 overview 193–194 Suez v Argentina (2005) 198–199 bifurcated decisions central problem 180 discretionary process 179 Emmis v Hungary 186–188 fairness 178–179 Gavazzi v Romania 188–190 Glamis Gold Limited v United States 181–183 jurisdiction and merits 179 liability and quantum 179 overreaching principles 190 Philip Morris v Australia 183–186 preliminary questions 179 rationale 178 regular use 177–178 relevance 190 challenges procedures applicable standards of review 212–214 dangers and disadvantages 211 essential control mechanism 211 importance 224–225 independence and impartiality 215–224 who makes the decision 214–215 consent to arbitration acceptance of offer, conditional consent 263–264 denunciation of ICSID Convention 272–273 direct agreement of parties 260 indispensible condition 259 past events and disputes 264–265 scope of consent 266–267 shift from contract-based to treaty-based arbitration 259 subject to procedural conditions 267–270 termination of contract containing consent 270–271 termination of legislation providing consent 271–272 through host state legislation 260–261 through treaties 261–263

645

contribution of IUSCT to interpretation of UNCITRAL procedural rules 65–68 counterclaims advantages and disadvantages 287–288 benefits of bringing related claims together 283–287 effect of investor conduct on liability and remedies 283 importance of applicable law 281–283 overview 275–276 tribunal authority to apply counterclaims 276–281 fundamental features of dispute settlement mechanism 15–17 inadequacies of diplomatic protection model 44 ‘investments’ defined under the ICSID Convention 227–234 IIA provisions 237–241 overview 227 ‘mixed commissions’ 16–17 protected investors companies 250–253 corporate restructuring 255–257 minority shareholders 253–255 natural persons 243–249 significance 243 trifurcated decisions common phases 180 defined 177 Professional ethics behaviour of arbitrators 224 CJEU 508 commonalities with international commercial arbitration 467–468 Protected investors companies corporate restructuring 255–257 nationality 250 piercing the corporate veil 250–253 decisions in EU courts 621–624 minority shareholders 253–255 natural persons dual nationality 246 ICSID arbitrations 247 nationality requirement 243–246 non-ICSID arbitration 248–249 significance 243 Protection of aliens Ancient Greece 10 equal treatment theory see Equal treatment theory ‘mixed commissions’ 17–18 R Remedies see also Compliance and enforcement; Reparations counterclaims 283 history of investment law

646 Index defining feature of investment disputes 51–52 general principles of reparations 52–53 impact on investment arbitration 56 reception in investment arbitration 53–56 WTO and ISDS distinguished 101 Renewable Energy ECT cases effect of Spain’s state regulatory measures 152–156 landmark decisions Charanne v Spain (2016) 152–156 Eiser v Spain (2017) 160–162 Masdar v Spain (2018) 162–164 RREEF v Spain (2019) 164–167 overview 153–154 shift in arbitral decisions away from the unworkable thresholds of earlier decisions 173–174 significance for ECT content of FET 167–168 interpretation 169–171 methodology for determining FET 168–169 significance for investment law 171–172 significance for public international law 172–173 Reparations see also Compliance and enforcement applicable standard principles for calculation of damages 444–449 assessing country risk 457–459 bifurcated decisions Gavazzi v Romania 188–190 ways in which the proceedings can be divided 179 contribution of IUSCT to substantive law 72–74 counterclaims 283 enforcement see Compliance and enforcement European Court of Human Rights compensation 499–501 just satisfaction 502–504 general principle of law 443 history of investment law Argentinian crisis cases 130–131 establishment of general principles 31 Lena Goldfields arbitration 26 importance 459–460 landmark decisions ADC v Hungary 444–449 LG&E v Argentina 454–457 Tidewater v Venezuela 457–459 Yukos v Russia 449–453 largest amount ever awarded 449–453 Mexican approach to constitutionality of investment agreements 592–595 necessity and assessment of damages 454–457 role of World Court damages for treaty breaches distinguished 54 defining feature of investment disputes 51–52 general principles 53 general principles of reparations 52–53 impact on investment arbitration 56 reception in investment arbitration 53–56

Yukos Oil cases Quasar De Valores and others v The Russian Federation 139–141 Rosinvestco v The Russian Federation 138–139 Review see Standards of review S SADC Tribunal establishment and mandate 530–532 landmark decisions 534–541 limitations of investment adjudication 543–544 Model Bilateral Investment Treaty Template 406 overview of cases 529–530 post-Cambell protection regime 541–543 three cases dealing with expropriation 534–541 Zimbabwe’s land reforms historical context 532–534 Seat of arbitration impact of international commercial arbitration on ILL 468–470 judicial review by national courts 576 Mexico 593, 600 Shareholder protection European Court of Human Rights 483–488 minority shareholders 253–255 role of World Court 40–43 umbrella clauses 391–394 South African Development Community Tribunal see SADC Tribunal Sovereign immunity applicable law 475 Argentinian crisis cases 121–122 EU courts 618 French rules 151 ICSID rules 618 matter governed by national law 149 relevance of counterclaims 286 US approach ‘arbitration exception’ to immunity 583–584 increasing scepticism about ISDS 588–589 Micula v Romania 585–588 Mobil v Venezuela 584–585 overview 582–583 Standards of investment protection fair and equitable treatment ‘a legal term of art’ 360 Argentinian crisis cases 124–125 controlling arbitrators as law-makers 370–379 core standard 359 effect of normative vagueness 379 empowering arbitrators as law-makers 362–370 interpretation 360–361 NAFTA Ch 11 cases 99, 111 Renewable Energy ECT cases 152–167 shaping of normative contours 361–362 significance of RE ECT cases 167–171 ICJ influence on standards contained in IIAs

Index defining concept of ‘arbitrariness’ 48–51 most-favoured nation clauses 45–48 overview 44–45 ‘mixed commissions’ 17–18 Most-Favoured-Nation Treatment (MFN) clauses as basis for claiming general treatment 431–432 as basis for disregarding limitations to treatment 433 ICJ influence on standards contained in IIAs 45–48 invoked to benefit from better specific treatment 429–431 underlying rationale 427–429 Yukos Oil cases 138 NAFTA Ch 11 cases Ethyl v Canada 102 Feldman v Mexico 112–114 Metalclad v Mexico 110–112 Methanex v United States 114–117 Pope & Talbot v Canada 105–109 remaining discretion and indeterminacy 117 S.D. Myers v Canada 102–105 source of legitimacy crisis 99–101 National Treatment (NT) clauses less favourable treatment in case law 439–440 likeness in the case law 437–439 major trends and conflicts 442 NAFTA Ch 11 cases 99–116 overview 436–437 regulatory purpose and host state justification 441–442 underlying rationale 427–429 overlap with ECtHR 481 South African national courts 610–612 Standards of review challenges procedures 212–214 decisions in EU courts 621–622 domestic determinations involving non-legal expertise or discretion Glamis Gold v United States 330–331 Marfin v Cyprus 333–334 Philip Morris v Uruguay 331–333 SD Myers v Canada 330 domestic judicial decisions Eli Lilly v Canada 339–340 OAO Tatneft v Ukraine 338–339 overview 336–338 ICSID and non-ICSID awards distinguished 576 national treatment rulings 100 overarching themes 340–341 overview 327–329 treaty changes reflected in USMCA 117 tribunals review prior to African national proceedings 604–606 US standards of review BG Group v Argentina 577–579 overview 576

647

Standing ‘mixed commissions’ 16–17 natural persons 244 State consent see Consent to arbitration State contracts early arbitrations concerning petroleum concessions importance 26–27 Petroleum Development v Abu Dhabi 27–28 Sapphire International v NIOC 29–31 Saudi Arabia and ARAMCO 28–29 legal nature and specific legal regime 22 Lena Goldfields arbitration arbitration proceedings and outcome 24–26 the Concession agreement 23–24 early transnational arbitration 22 nationalisation of oil concessions in Libya and Kuwait applicable law 32–35 importance 31 overview 31–32 three libyan arbitrations 31–32 origins 21–22 State immunity see Sovereign immunity Substantive matters applicable law choice of law 293–296 complexity 292 conclusions 309 contentious topic 291–292 EU law 307–308 general international law 296–298 importance 291 national law 304–306 non-investment instruments 298–300 overview 292–293 relationship between international and national law 300–304 consent to arbitration 265–267 contribution of IUSCT to substantive law attribution 74–76 compensation 72–74 expropriation 70–72 nationality 68–70 countermeasures 277 denial of benefits clauses importance of decisions 421 interpretation 397–398 nature of objections 398–407 protected investors 255 substantive requirements for the exercise of a clause 407–421 table of cases 421–425 treaty-based tool 397 Yukos Oil cases 146 expropriation common elements of an expropriation clause and their significance 344–347

648 Index direct and indirect expropriation distinguished 348–352 importance 342–344 inherent state powers and right to regulate 355–358 property which may be expropriated 352–355 fair and equitable treatment ‘a legal term of art’ 360 controlling arbitrators as law-makers 370–379 core standard 359 effect of normative vagueness 379 empowering arbitrators as law-makers 362–370 interpretation 360–361 shaping of normative contours 361–362 fundamental features of dispute settlement mechanism 15–17 inadequacies of diplomatic protection model 44 independence and impartiality of arbitrator behaviour of arbitrators 224–225 double hatting 220–222 familiarity with another participant 219–220 importance 215–216 issue conflict 222–224 multiple and repeat appointments 216–219 ‘mixed commissions’ 17–18 Most-Favoured-Nation Treatment (MFN) clauses as basis for claiming general treatment 431–432 as basis for disregarding limitations to treatment 433 ICJ influence on standards contained in IIAs 45–48 invoked to benefit from better specific treatment 429–431 reception in investment arbitration 46–47 underlying rationale 427–429 Yukos Oil cases 138 National Treatment (NT) clauses less favourable treatment in case law 439–440 likeness in the case law 437–439 major trends and conflicts 442 overview 436–437 regulatory purpose and host state justification 441–442 reparations applicable standard principles for calculation of damages 444–449 assessing country risk 457–459 general principle of law 443 importance 459–460 largest amount ever awarded 449–453 necessity and assessment of damages 454–457 standards of review determinations of national law 334–336 domestic determinations involving non-legal expertise or discretion 330–334 domestic judicial decisions 336–340

overarching themes 340–341 overview 327–329 treaty interpretation dual function ascribed to treaties 325–326 interaction with jurisdiction and annulment 320–323 interpretation as a legal dispute per se 323–325 interpretation as a shared power 319–320 overview 311–312 VCLT framework 312–318 umbrella clauses complexities 381–382 effect 385–388 lessening of unresolved issues 394–395 nature of the investor’s claim 388–391 relationship between treaty and contract 382–384 shareholders claims 391–394 T Treaties consent to arbitration 261–263 interpretation dual function ascribed to treaties 325–326 interaction with jurisdiction and annulment 320–323 interpretation as a legal dispute per se 323–325 interpretation as a shared power 319–320 overview 311–312 VCLT framework 312–318 ‘investments’ defined 237–241 origin of ‘mixed commissions’ – Jay Treaty 1794 14 shift from contract to treaty basis – SPP and AAPL consent as cornerstone of arbitral settlement 82–87 determination of applicable law 93–96 impact on domestic investment statutes 92–93 interpretation of IPPAs 96–97 overview 79–81 pioneering commentators and key players 87–92 twentieth century legal milestone 97 symbolai treaties of Ancient Greece 10–11 Tribunals see International courts and tribunals Trifurcated decisions common phases 180 defined 177 U Umbrella clauses Argentinian crisis cases 125 complexities 381–382 effect 385–388 investor protection 35 landmark decisions BIVAC v Paraguay and SGS v Paraguay 388–391 Burlington v Ecuador 391–394

Index SGS v Philippines 385–388 Vivendi v Argentina (I) 382–384 lessening of unresolved issues 394–395 nature of the investor’s claim 388–391 relationship between treaty and contract jurisdictional aspects of treaty/contract claims 383–384 substantial aspect of treaty/contract claims 382–383 shareholders claims 391–394 UNCITRAL rules amici curiae interventions Glamis Gold v United States of America (2005) 203 overview 193–195 Suez v Argentina (2005) 198–199 applicable law choice of law 293 bifurcated decisions absence of clear legal standards 178 Glamis Gold Limited v United States 181–183 Philip Morris v Australia 183–186 challenges procedures applicable standards of review 213 double hatting 220–222 who makes the decision 214 counterclaims ‘close connection’ requirement 280 party consent to hearing counterclaims 277–279 differences between international commercial and investment arbitration 474 impact of Yukos Oil cases 149–152 independence and impartiality of arbitrator double hatting 220–222 issue conflict 222–224 Iran-United States Claims Tribunal contribution to interpretation of UNCITRAL procedural rules 65–68 overview 63–64 protected investors 249 United Nations Commission on International Trade Law see UNCITRAL rules United States approach to sovereign immunity ‘arbitration exception’ to immunity 583–584 increasing scepticism about ISDS 588–589 Mobil v Venezuela 584–585 overview 582–583 bifurcated decisions 181–183 expropriation 352–355 fair and equitable treatment 373–376 Iran-United States Claims Tribunal attribution 74–76 bold and ingenious choice of rules 65–66 compensation 72–74

649

counterclaims 66–67 expropriation 70–72 institutional nature of Tribunal 62–65 nationality 68–70 participation of amici curiae 68 pioneering work in formulating legal principles and tests 76–77 NAFTA Ch 11 cases 114–117 standards of review BG Group v Argentina 577–579 Glamis Gold v United States 330–331 Henry Schein v Archer & White 579–582 overview 576 treaty interpretation Mondev International Ltd. v United States of America 317–318 Republic of Ecuador v United States of America 323–325 W World Bank Inspection Panel (WBIP) case law on mining projects 549–552 current reform processes 554 establishment 546 human rights impacts 552–554 inclusive approach to local communities 547–549 landmark decisions Bear Creek Mining Corp v Republic of Peru 547 Chad: Petroleum Development and Pipeline Project 552–553 Copper Mesa Mining Corporation v Republic of Ecuador 547 Defensa y Conservacion Ecologica de Intag (DECOIN)) 550–551 Honduras: Land Administration Project 553–554 Mining Infrastructure Support – Additional Financing 551–552 Mongolia: Mining Infrastructure Investment Support 551–552 Urbaser v Argentina, or Burlington v Ecuador 552–553 World Court emergence of contemporary IIL investment protection via internationalised contracts 38–39 limitations on shareholder protection under general international law 40–43 overview 37–38 remedies defining feature of investment disputes 51–52 general principles of reparations 52–53 impact on investment arbitration 56 reception in investment arbitration 53–56 varied impact in scope and over time 57

650 Index World Trade Organization (WTO) burden of proof 114 national treatment standard 106, 116–117 WTO and ISDS remedies distinguished 101 Y Yukos Oil cases enforcement saga 147–149 expropriation without compensation

138–139

fork-in-the-road provision of article 26(3)(b)(i) ECT 141–143 impetus for reform 149–152 new confidential ‘second-wave’ trilogy 145–146 overview 135–136 possible inconsistencies from concurrent jurisdiction 147 violations of human rights 137–138