Jump to Content Jump to Main Navigation


Forgotten your password?

The Coordination of Multiple Proceedings in Investment Treaty Arbitration by Wehland, Hanno (2013-06-01)

1 Introduction and Delimitation of the Subject

From: The Coordination of Multiple Proceedings in Investment Treaty Arbitration

Hanno Wehland

European Court of Human Rights (ECtHR) — Arbitral tribunal

(p. 1) Introduction and Delimitation of the Subject

  1. A.  Introduction 1.01

  2. B.  Terminology 1.14

    1. 1.  The notions of claim and cause of action 1.15

    2. 2.  The notion of multiple proceedings 1.19

    3. 3.  The notion of investment treaty arbitration and related notions 1.24

  3. C.  Problems Associated with Uncoordinated Multiple Proceedings 1.31

  4. D.  Methodology and Scope 1.43

    1. 1.  Overview of the different chapters 1.44

    2. 2.  Issues not addressed 1.52

A.  Introduction

1.01  The number of investor-state arbitrations adjudicating claims under international investment treaties has grown significantly since the rendering of an award in the first such dispute twenty years ago.1 With this growth in numbers has come the realization that in many instances the facts underlying these disputes can give rise not only to investment treaty arbitrations, but may also lead to additional sets of proceedings between the same or closely related parties regarding largely identical issues. Such additional proceedings can take place before other arbitral tribunals, the domestic courts of host states, and even specialized international forums such as the European Court of Human Rights.

1.02  The potential for multiple proceedings involving investment treaty tribunals is greatly increased by a number of specific characteristics reflecting the complexity of the international investment framework. Investment treaty disputes tend to be complex in that they frequently relate to claims arising under different legal instruments, they often involve formally separate yet economically related entities, and it is typically possible to think of more than one forum for their resolution. This complexity translates into a particular risk of having multiple proceedings through essentially three phenomena:

1.03  First, the fact that separate legal instruments may refer a dispute to a number of different forums increases the potential for disagreement between parties as to which forum should ultimately resolve their dispute. Such a disagreement can lead to a situation where one party initiates a claim for performance before one forum, while the other party requests a different forum to declare that, to the contrary, there is no basis for the relevant claim. This phenomenon has sometimes been referred to as ‘jurisdictional competition’.2(p. 2)

1.04  Second, the fact that substantive rights of investors typically arise out of a number of different instruments can lead to a fragmentation of adjudicatory competences, potentially forcing investors to initiate proceedings in a number of forums in order to ensure that all of their rights will ultimately be taken into account. This phenomenon of ‘jurisdictional fragmentation’3 is compounded by the fact that investment transactions frequently involve more than just one entity on either the investor’s or on the host state’s side, thus leading to formally separate, yet substantially related, procedural relationships. Where disputes arising within these different relationships cannot all be referred to the same forum, the parties may yet again have to initiate several actions in order to resolve a dispute in its entirety.

1.05  Third, the complexity in investment treaty disputes also presents a potential for abuse by investors in that claimants may seek to have several ‘shots’ at an issue even where a dispute could be adjudicated fully in a single forum, in a way that could not have been intended by the signatories to IIAs. All these characteristics make multiple proceedings in investment disputes a conspicuously frequent phenomenon.

1.06  There is general agreement that the resulting duplication is fundamentally undesirable and that multiple proceedings should ideally be avoided, or, where that is not possible, at least coordinated. At the same time, there appears to be considerable uncertainty with regard to the rules that forums must apply where a different set of proceedings has already been brought elsewhere and is either still pending or has already resulted in a final decision.4 In fact, this uncertainty has several aspects to it.

1.07  First, at the level of jurisdiction, there can be doubts as to when a specific forum is competent to decide a dispute and to what questions exactly its competence extends.5 Addressing this issue is of particular importance in cases where parties disagree as to the appropriate forum for resolving their dispute, since if it can be established that there is in reality only one forum with jurisdiction, this should in principle prevent multiple proceedings from the outset.6

1.08  Second, once an investment treaty tribunal has found that it has jurisdiction, the situation is equally unclear as to which mechanisms it should apply to address the fact that a second set of proceedings is pending, or has already been implemented, in another forum. A number of different coordination methods—ranging from the consolidation of proceedings, review mechanisms, and the theories of lis pendens and res judicata to general principles of comity and the prohibition of abuse of process—would in principle seem to be available in this regard. However, the question of how exactly these different mechanisms relate to each other (p. 3) and under what circumstances one (rather than the other) of them should be applied by a tribunal remains subject to considerable uncertainty.7

1.09  Third, even where the applicability of a specific coordination mechanism is recognized in principle, doubts may nonetheless exist as to the applicable law and the specifics of the mechanism’s application.

1.10  Achieving a higher degree of consensus with regard to these issues seems desirable, to assist the decision-making of arbitral tribunals and thus increase the traceability of relevant decisions for investors and host states. In addition, where the relevant rules are known to parties in advance, this may effectively obviate the need for tribunals to decide these issues, by discouraging the artificial splitting of disputes and facilitating an agreed coordination between the parties.

1.11  In particular, a potential claimant will typically ask himself the following questions before initiating any judicial proceedings: Does the envisaged forum have jurisdiction? Could a different choice of forum by his opponent nonetheless frustrate the successful implementation of proceedings? Might claims be inadmissible because of proceedings already implemented or still pending elsewhere? If the forum proceeded to render a decision on the merits, would it be bound by the outcome in a different set of proceedings? And finally, assuming that the forum ultimately rendered a decision in his favour, would that decision still be subject to review elsewhere? Depending on the answers to these questions, a potential claimant might well refrain from initiating proceedings or decide to bring only a single action before the forum offering him the best prospects.

1.12  Once proceedings have been initiated, a respondent will be faced with very similar issues. If the rules regarding the determination of jurisdiction and the coordination of proceedings are sufficiently clear, he can, in principle, be expected to reach the same conclusions and give up any idea of seeking to undermine the claimant’s forum choice by initiating another set of proceedings elsewhere. What is more, where a respondent concludes that a claimant is (or would be) entitled to bring multiple proceedings in several forums, he may well decide that it is in his interest to offer to extend the jurisdiction of the claimant’s preferred forum or to otherwise suggest the coordination of proceedings.8

1.13  While questions of jurisdiction and the coordination of proceedings have frequently been addressed in arbitral jurisprudence and academic commentary, the reasoning with regard to these issues has typically been sparse and the discussion somewhat fragmentary. As a consequence, a systematic analysis of the different coordination mechanisms and their interrelationship still appears to be missing. The present study aims to fill this gap by comprehensively addressing the coordination of multiple proceedings before investment treaty tribunals. In so doing, the study not only seeks to identify clear, predictable, and sensible (p. 4) coordination rules, but also to suggest an application of these mechanisms that reduces the occurrence of the three phenomena that tend to be at the origin of multiple proceedings: jurisdictional fragmentation, jurisdictional competition, and the abuse of the complexities of the system.

B.  Terminology

1.14  As a preliminary matter, it will be necessary to clarify the terminology used throughout this study. In particular, the central notions of claim and cause of action (1), multiple proceedings (2), and investment treaty arbitration (3) require some explanation.

1.  The notions of claim and cause of action

1.15  Two related terms of central significance for the present study are the notions of claim and of cause of action. Both terms can be used in either a substantive or a procedural sense, and different meanings will be ascribed to them, depending on the context in which they are used.

1.16  In a substantive sense, a claim can be defined as a person’s right to request a specific act or omission from another person on the basis of a specific set of facts and a specific ground of substantive law.9 By contrast, in a procedural sense, a claim appears as the actual request of an act or omission based on a specific set of facts, presented against another person in proceedings before a judicial forum.10 In this latter sense, a claim is thus primarily defined by merely two elements: the requested relief, and the facts on which the request is based. Contrary to the notion of claim in a substantive sense, the idea of a procedural claim does not per se require the specification of any particular legal basis—it is rather accepted that a claim (or action) can simultaneously be based on several legal grounds or rights existing as a matter of substantive law.

1.17  A similar ambiguity exists with regard to the notion of cause of action, which is sometimes used interchangeably with the notion of claim.11 Here again, the term can be used in either a substantive or a procedural sense. Where cause of action is being referred to as a notion of substantive law, its meaning can indeed be the same as that of claim (in a substantive sense).12 By contrast, where the term is used in a procedural sense, it typically contains no reference to the requested relief, instead referring more narrowly to the specific aspect of a claim (in a procedural sense) that is also frequently designated with the Latin expression causa petendi.13

1.18  As a consequence, an undifferentiated use of these terms has a considerable potential for causing terminological confusion. To avoid ambiguity, the present study will, unless otherwise indicated, use the notion of claim in a substantive sense, which is not directly linked to the (p. 5) idea of judicial proceedings. By contrast, the notion of cause of action will be used exclusively in a procedural sense synonymously with a dispute’s causa petendi.

2.  The notion of multiple proceedings

1.19  The study deals with multiple judicial proceedings. The notion of judicial proceedings itself does not present any major ambiguity and will be used in the sense of ‘[a]ny procedural means for seeking redress from a tribunal’.14 It should, however, be pointed out that the study only deals with proceedings of a judicial character, meaning that decisions of the relevant forum must in principle be final and binding within the field of its competence.15 Such a judicial character is recognized in particular for court and arbitration proceedings, but not, for instance, for review proceedings before administrative authorities.16

1.20  The term multiple proceedings will then be used to refer to situations in which a certain set of facts gives rise to judicial proceedings before different forums, involving substantially related parties and substantially identical requests for relief.17 This definition is deliberately chosen to be broad, so as to include any situation in which proceedings are merely arguably identical. In particular, the present study is not limited to scenarios where the parties to two sets of proceedings are formally identical, but also extends to situations where parties are substantially related. Such a substantial relation will be assumed where two formally distinct legal entities are affiliated to each other so that their interests become fully or partly congruent. On the investor’s side, this will be the case where one entity is a shareholder of the other entity. On the side of a host state, substantial identity will be assumed where one entity is fully or partly owned by or constitutes a sub-entity of the other.

1.21  Similarly, the procedural positions of the parties must not necessarily be identical. Rather than having one party acting as claimant and the other as respondent before both forums, their roles can also be inverted in the two sets of proceedings. One can thus effectively distinguish between two fundamentally different categories of multiple proceedings.

1.22  First, there is the situation where the same claimant initiates (or substantially related claimants initiate) essentially identical proceedings against the same respondent (or substantially related respondents) before two different forums. Here, multiple proceedings reflect the claimant’s attempt to maximize his chances and secure a successful outcome for himself even if one of the actions fails.18(p. 6)

1.23  Second, there is the situation where the respondent in a first action brings identical proceedings against the claimant in a different forum. In this case, the claim of one party is typically for performance, whereas the other party requests a negative declaration that no such claim for performance exists. Since the requests in the two sets of proceedings mutually exclude one another and a decision on one of them also implicitly decides on the other one, the requests must again be seen as substantially identical. However, the reason for multiple proceedings in this scenario is not a claimant’s attempt to insure himself against an unsuccessful outcome before the forum first chosen, but rather a disagreement between the parties as to which of the forums should adjudicate their dispute.19

3.  The notion of investment treaty arbitration and related notions

1.24  Another notion of central importance for the present study is that of investment treaty arbitration. To explain that term, it first seems useful to clarify the meaning of international investment arbitration and international investment proceedings more generally (a). Once this is done, the more specific term of investment treaty arbitration shall be addressed (b).

a.  International investment arbitration and international investment proceedings

1.25  The term of international investment arbitration has been used in a variety of contexts, casting some doubt as to its exact meaning.20 Some commentators have limited its use to arbitration based on the generic consent given by a host state in its domestic legislation or an international treaty without regard to any specific investment. Thus one practitioner explains that in ‘investment arbitration, the consent of the state is exhibited through an open offer to arbitrate disputes…in relation to defined classes of investment’.21 Following such a definition, investment-related proceedings based on the arbitration clause in an investor-state contract would not appear to constitute investment arbitration.

1.26  By contrast, others have defined investment arbitration as ‘arbitration [that] refers to disputes between a host state and a private investor concerning the investment of the latter as protected by an investor-state contract, an investment law or a bi- or multilateral investment treaty’.22 This definition is different from the previous one in that it (a) refers to the substance of the dispute rather than to the mechanism through which consent to arbitration is granted and (b) includes investment-related disputes concerning the adjudication of contractual rights.

1.27  The present study adopts the latter understanding, defining investment arbitration by essentially two elements. First, investment arbitration relates to an investment, as opposed to any ordinary commercial transaction.23Second, investment arbitration concerns a dispute between an investor and a host state.24 Since the two elements apply cumulatively, the (p. 7) presence of a state party to arbitral proceedings is not sufficient to make these proceedings an investment arbitration.25 Similarly, investment-related state-to-state proceedings do not fall under the notion of investment arbitration adopted here.26

1.28  Investment disputes do not always take place before arbitral tribunals, but occasionally involve other forums, in particular the courts of a host state or a specialized human rights forum such as the European Court of Human Rights.27 Where the present study does not specifically refer to arbitral proceedings, but more generally to any form of dispute resolution between host states and investors, it will treat the terms investment proceedings or investment disputes as including, but not being limited to, investment arbitration.

b.  Investment treaty arbitration and treaty-based arbitration

1.29  In contrast to the terms just defined, the notion of ‘investment treaty arbitration’ will be given a more limited meaning, referring only to those investment arbitration proceedings the substance of which is at least partly based on an IIA.28 The defining criterion is thus seen in the source of the substantive rights asserted rather than the mechanism through which consent to arbitration is granted.29 While in the typical scenario the assertion of treaty rights goes hand in hand with the invocation of the treaty’s dispute resolution mechanism,30 there may be exceptions where this will not necessarily be the case. As will be shown, non-treaty claims can sometimes be asserted in proceedings based on an investor’s acceptance of a host state’s arbitration offer made in an IIA, while treaty claims can in principle also be brought in arbitrations initiated under a contractual mechanism.31

1.30  If this study defines the ‘treaty character’ of arbitral proceedings by reference to the source of the asserted substantive rights rather than any underlying procedural mechanism, it is because this source seems more relevant for the position of an arbitral tribunal vis-à-vis the position of other forums than the manner in which the arbitration agreement has been formed.32 Where the study exceptionally seeks to highlight the mechanism underlying the arbitration agreement (rather than the source of the asserted claims), it will refer to an arbitration as being based on a particular legal instrument.33

(p. 8) C.  Problems Associated with Uncoordinated Multiple Proceedings

1.31  Before addressing the scope and methodology of the present study, it may also be useful to explain its general rationale, by outlining the problems typically associated with uncoordinated multiple proceedings. The adjudication of substantially identical disputes in more than one forum is generally regarded as undesirable for a number of reasons.34

1.32  First, where claimants manage to secure decisions granting them the same type of relief in several proceedings there is always a danger of double recovery.35 This risk is particularly acute where separate proceedings with regard to the same investment are initiated by formally separate entities situated at different levels of an investment structure.36 While taking into account the damages previously awarded to a claimant when calculating the compensation due in a second set of proceedings might in principle seem fairly straightforward,37 things get considerably more complicated where one successful claimant is a shareholder of the claimant in another action.38

1.33  In particular, where a shareholder in a company has already been paid damages in a first set of proceedings, it is not clear how such a payment can be taken into account if damages are subsequently awarded to the company itself.39 Since the company will typically not have (p. 9) benefited from the payment, deducting it when calculating the damages to be awarded would indirectly prejudice other shareholders and the creditors of the company.40 It appears that there is no entirely satisfactory general solution to the resulting difficulties.41

1.34  Second, where multiple proceedings are initiated by the same claimant, the very fact that the dispute is adjudicated before a number of forums may distort the chances of success in the claimant’s favour.42 In particular, assuming that a negative outcome in a first set of proceedings would not affect the claimant’s case in a second action, an investor could hedge his bets by repeatedly asserting the same claims before different forums.43 Coordinating multiple proceedings may thus be necessary not only to avert a potential double recovery, but also to prevent a claimant from effectively relitigating a claim that has already failed.44

1.35  Third, multiple proceedings addressing the same facts always create the possibility of conflicting outcomes.45 Even where two sets of proceedings relate to formally distinct claims, conflicts may arise where the issues to be adjudicated are materially identical. The seriousness of such conflicting outcomes depends on whether they are limited to the forums’ factual assessments and reasoning or also extend to their decisions on the requested relief.46 Thus in some instances two forums will merely assess the facts of the case differently or adopt differing reasonings, but ultimately still reach decisions that can be reconciled with one another. This kind of discrepancy might already affect the parties’ perception of the credibility (p. 10) of the adjudication process.47 In addition to that, however, the dispositifs of the decisions might also contradict each other.48 This type of conflict seems particularly serious, as it has the potential to effectively prevent any final resolution of the dispute.49

1.36  Fourth, even in the absence of actually conflicting decisions, the very possibility that the same matter might be readjudicated with a different outcome may effectively deprive a successful party of the benefits of a decision, by failing to provide it with the required legal security.50

1.37  Fifth, irrespective of the outcome of multiple proceedings, it seems clear that the litigation of a dispute in more than one forum typically increases costs and thus constitutes a waste of resources.51

1.38  Many of these difficulties can be illustrated by the outcome in the parallel CME and Lauder arbitrations, which are typically seen as the epitomy of the failure to coordinate proceedings. Lauder, an American businessman, had invested in a Czech television services company via the Dutch holding company CME, which he controlled. When the investment relationship turned sour, first Lauder and then CME initiated UNCITRAL proceedings against the Czech Republic, the former invoking his protection as a US national under the Czech Republic-US BIT, while the latter claimed to be protected as a Dutch investor under the Czech Republic-Netherlands BIT.52 Both sets of proceedings were based on the same facts and essentially concerned the same requests for relief.(p. 11)

1.39  The claimants offered several methods for coordinating the two arbitrations,53 but all of them were refused by the respondent. In fact, the Czech Republic took the view that if the claimants were ‘concerned by duplicative proceedings, clearly the proper course for them to take would be to discontinue’ either set of proceedings.54 In the instance, both proceedings continued and the tribunals issued awards on liability that came to diametrically opposite conclusions. While the tribunal in Lauder only found a minor treaty breach that had not caused the claimant any damage, the majority in CME held that the respondent had breached several of its treaty obligations and proceeded to assess the damage caused to the investor.55 In the latter proceedings the Czech Republic was ultimately ordered to pay almost US$270 million in damages to CME.56

1.40  While this example clearly shows the danger of uncoordinated multiple proceedings from a general policy perspective, it equally illustrates the fact that individual litigants may well have a strategic interest in adjudicating a dispute in more than one forum.57 In particular, even though potential claimants might ideally want to avoid the additional expenses associated with multiple proceedings,58 they may ultimately consider them money well spent, if they increase their overall chances of success. After all, as rightly pointed out by one prominent commentator, ‘every rational person with basic strategic thinking will prefer two shots to one shot’.59

1.41  The potential discrepancy between the undesirability of multiple proceedings from a general policy viewpoint and the different perspective of potential claimants cannot be without consequences for the application of coordination mechanisms. Multiple proceedings of the sort observed in CME and Lauder have so far remained a rare occurrence, potentially indicating a tacit consensus that there are unwritten limits regarding claimants’ procedural rights.60 However, these limits still remain barely defined—with considerable legal uncertainty for all the participants as a consequence.61(p. 12)

1.42  In particular, there is clearly a risk that investors will ‘seek advantage from the absence of hierarchy and coordination among the various types of international tribunals’.62 At the same time, if claimants can never entirely be sure whether they will be allowed a second set of proceedings, they may in some situations effectively have to abandon rights that they should really be able to enforce.63 One of the aims of the present study is thus to propose a reading of coordination mechanisms which ensures that claimants cannot abuse their procedural possibilities—but also allows them to assert all of their rights in a manner compatible with the function of the judicial process.

D.  Methodology and Scope

1.43  Based on the preceding discussion, it is now possible to outline the methodology and scope of the present study. The analysis will primarily be based on a review of the relevant international investment agreements and arbitration rules as well as international arbitral decisions and commentary. Where appropriate, general public international law will also be referred to, either as part of the applicable law or as an indication of recognized legal principles. By contrast, reference to national legal rules and judicial decisions will be kept to a minimum, so as to maintain the general validity of the argumentation.

1.  Overview of the different chapters

1.44  This chapter introduces the subject and the terminology used throughout the study. It also sets out the study’s general rationale, methodology, and scope.

1.45  Chapter 2 looks into a number of characteristics that are specific to the international investment framework. The purpose of this chapter is to illustrate the particular challenge one faces when addressing the coordination of multiple proceedings in an investment context. At the same time, by describing the fundamental traits of the typical investor-state relationship, Chapter 2 also lays the foundation for the debates on determining the jurisdiction of a forum and applicable coordination mechanisms in the later chapters.

1.46  Chapter 3 then addresses the issue of determining the jurisdictions of those forums that are most frequently called upon to decide investment disputes, namely arbitral tribunals and the domestic courts of a host state. This logically precedes any questions regarding the coordination of proceedings, in that a forum can only apply a coordinative mechanism after having verified its own jurisdiction. In addition, in many instances of multiple proceedings, the analysis will show that jurisdiction with regard to a dispute is actually limited to a single forum. As a consequence, a rigorous approach to issues of jurisdiction must in itself be seen as a step of first priority when addressing situations of multiple proceedings.64(p. 13)

1.47  Chapter 4 goes on to provide an overview of the different mechanisms that may be applied by investment treaty tribunals when coordinating multiple proceedings. It first addresses the situation where parties have explicitly agreed to coordinate proceedings either by consolidating them or through one of several possible arrangements frequently referred to as ‘quasi-consolidation’. Where there is an agreement on coordination between the parties, this must take precedence over any other mechanisms under the principle of party autonomy. In the absence of such an agreement, a number of ‘default’ mechanisms can be envisaged, partly depending on whether the different proceedings are situated at the same level of judicial hierarchy. At the same time, it will be argued that there is a link between these ‘default’ mechanisms and the willingness of parties to cooperate with regard to the coordination of their proceedings.

1.48  Chapter 5 looks at the role of the hierarchy of forums for coordinating multiple proceedings, by assessing the relationship between investment treaty tribunals and other types of forums that may be involved in the resolution of investment disputes. The analysis shows that, while different treaty tribunals are in principle hierarchically equivalent, they normally also have the power to review the decisions of non-treaty forums. As a consequence, it will be argued that, when it comes to coordinating treaty proceedings on the one hand and non-treaty proceedings on the other, hierarchical coordination mechanisms should apply.

1.49  Chapter 6 deals specifically with the coordination mechanisms applicable in the absence of any hierarchy between two forums, namely the principles of lis pendens and res judicata. As will be shown, these mechanisms actually play a more limited role in investment treaty arbitration than is frequently assumed. In particular, a lis pendens principle (in the sense of a strict ‘first-in-time’ rule) typically does not apply in proceedings before treaty tribunals. Similarly, the concept of res judicata appears to be inappropriate for dealing with many of the situations of multiple proceedings that may arise in an investment context, in particular where the parties to subsequent proceedings are not formally identical.

1.50  Chapter 7 addresses two more general coordination mechanisms that apply irrespective of the existence of hierarchy between two forums, which are the principle of comity and the prohibition of abuse of process. Given the limited applicability of lis pendens and res judicata, these more flexible mechanisms gain considerable relevance for the coordination of multiple proceedings in investment disputes. While the open-ended character of these mechanisms makes it impossible to address them in a conclusive manner, the study seeks to establish a number of principles that should guide their application in situations of multiple proceedings before investment treaty tribunals.

1.51  Finally, Chapter 85 summarizes the study’s findings and concludes with an outlook on potential future developments.

2.  Issues not addressed

1.52  It should be pointed out that a number of issues will not specifically be addressed in the present study.

1.53  First, it bears emphasis that the coordination of proceedings will in principle only be assessed from the perspective of treaty tribunals. The analysis relates to what a treaty tribunal should do when faced with a second set of proceedings before a different forum, but does not address the question of how the latter should act when faced with an additional set of proceedings (p. 14) before a treaty forum. Coordination mechanisms available to forums other than treaty tribunals are rather beyond the scope of the present study.65

1.54  Second, the present study is limited to coordination mechanisms with a serious claim to international application irrespective of any country-specific context, leaving out in particular the concept of forum non conveniens and the use of anti-suit injunctions. Differently from the lis pendens principle (which has repeatedly been invoked in the practice of investment treaty tribunals), there appears to be agreement that the doctrine of forum non conveniens does not constitute a generally applicable legal concept.66 The same is true for anti-suit injunctions, the use of which is not generally recognized in international arbitral proceedings.67

1.55  Third, the study will not discuss the relationship between investment treaty arbitrations and state-to-state dispute resolution proceedings.68 Clearly, foreign investment issues can give rise to inter-state disputes before a number of forums, including arbitral tribunals, international courts, and special dispute settlement mechanisms.69 Such inter-state disputes can in principle relate to facts and legal provisions that are also the subject of related treaty disputes70 and might thus in theory lead to conflicting decisions.71 At the same time, however, (p. 15) the threat posed by such a possibility should not be overstated. In particular, where state-to-state arbitration under an IIA results in a final award,72 the latter would seem to constitute an authoritative determination of the relevant treaty provisions that has to be taken into account in subsequent investor-state proceedings under Article 31 VCLT.73

1.56  The bigger concern with regard to state-to-state proceedings might actually be seen in their potential use by respondent states to derail investor-state arbitrations brought against them.74 However, as long as treaty tribunals do not consider the mere initiation of a state-to-state dispute as a sufficient reason for staying a pending investor-state arbitration, the potential for abuse of this mechanism seems clearly limited.75

1.57  Similarly, the risk of overlap between an investment treaty arbitration and state-to-state proceedings relating to the diplomatic protection of an investor seems virtually negligible.76 There is general agreement that the home state of an investor cannot initiate diplomatic protection proceedings once the investor has asserted treaty claims, unless the investor-state proceedings are being frustrated or an award rendered therein is not complied with.77 As a consequence, there seems no real need to further coordinate state-to-state disputes and investor-state proceedings.

1.58  Fourth, considering the present study’s purpose of providing practical guidance to parties and tribunals in treaty arbitrations, the subsequent analysis will be limited to the existing legal framework, without looking into coordination mechanisms that might only be envisaged (p. 16) de lege ferenda.78 The study’s aim will rather be to suggest an application of the relevant legal mechanisms that should provide a satisfactory solution to the problems associated with multiple proceedings under the existing legal framework.


1  See Asian Agricultural Products v. Sri Lanka (ICSID), Award of 27 June 1990.

2  See Ch. 2, para. 2.67.

3  See Ch. 2, para. 2.69.

4  See also M.W. Friedman, in International Arbitration 2006, 564.

5  The terms of competence and jurisdiction will be used synonymously for purposes of the present study, see the definition of ‘competence’ in Black’s Law Dictionary, at 322, as ‘[t]he capacity of an official body to do something’, and the definition of ‘jurisdiction’, at 927, as ‘[a] court’s power to decide a case or issue a decree’. See also Chorzów Factory, Jurisdiction (Germany v. Poland), No. 8 (1927), PCIJ (Ser. A), 30. The English and Spanish versions of the ICSID Convention appear to distinguish between the ‘jurisdiction’ (jurisdicción) of the Centre and the ‘competence’ (competencia) of a tribunal, while the French version indistinctly refers to ‘compétence’, see C. Schreuer, The ICSID Convention, Art. 25 para. 17. Commentators have noted that in arbitral practice ‘[t]he two terms are frequently used interchangeably’ and appear to agree that ‘the distinction is of little consequence’, see Schreuer, Art. 41 para. 57.

6  In addition, the question of which forums have jurisdiction can also be relevant for the application of coordination mechanisms such as 1res judicata and the prohibition of abuse of process, see Ch. 6, para. 6.112 and Ch. 7, para. 7.42.

7  Thus some authors have claimed that ‘the unanimous opinion in international legal circles is that the doctrines of res judicata and lis pendens also apply within the scope of international investment disputes’, see M. Dimsey, The Resolution of International Investment Disputes, 81. See also S. Wolfgram, Investitionsschiedsgerichtsbarkeit, 126. By contrast, others have taken the view that, where a treaty tribunal has jurisdiction, it cannot be prevented from ruling on the dispute before it through either a lis pendens or a res judicata rule unless there is an explicit treaty provision to that effect, see P. Glavinis, in Studia in Honorem Pelayia Yessiou-Faltsi, 246. See also SPP v. Egypt (ICSID), Decision on Jurisdiction of 14 April 1988, paras 84, 121; B. Hanotiau, in Liber amicorum Lucien Simont, 309; A. Koutoglidou, in Multiple Party Actions, 257.

8  See Ch. 7, para. 7.42.

9  See also the entry for ‘claim’ in Black’s Law Dictionary, at 281, referring to ‘any right’.

10  See also the entry for ‘claim’ in Black’s Law Dictionary, at 281, referring to ‘[t]he assertion of an existing right’ (emphasis added). In this sense, the notion of claim is synonymous with the notions of ‘suit’ or ‘action’.

11  See e.g. Sheppard, noting that ‘[f]or “cause of action”, one can also read “claim”’, A. Sheppard, in Parallel State and Arbitral Procedures, 224.

12  For a use of the notion of cause of action in a substantive sense, see e.g. Vivendi v. Argentina (ICSID), Decision on Annulment of 3 July 2002, para. 113; B.M. Cremades and D.J.A Cairns, in Parallel State and Arbitral Procedures, 14; G.S. Tawil, in International Arbitration 2006, 493.

13  The notion of claim in a procedural sense can be said to be composed of the notion of cause of action (in a procedural sense) and the request or petitum of the action, see Ch. 4, para. 4.82.

14  See the entry for ‘proceeding’ in Black’s Law Dictionary, at 1324. The term ‘dispute’ will be used synonymously as referring to the proceedings in which a disagreement between the parties is adjudicated.

15  See Effect of Awards of Compensation Made by the UN Administrative Tribunal Advisory Opinion, ICJ Reports (1954), 53.

16  See M.E. Schneider, 6(2) Arb.Int’l (1990), 100. While the ECtHR clearly constitutes a judicial authority, the situation is different with regard to the Human Rights Committee established under Part IV of the ICCPR, since most commentators assume that the views of the latter are non-binding, see Y. Shany, Regulating Jurisdictional Relations, 163; J.S. Davidson, New Zealand L.Rev. (2001), 142. Given a number of similarities between the relevant procedures, reference will nonetheless be made to proceedings before both forums.

17  See also Söderlund, defining multiple proceedings as proceedings having ‘some communality of interest’, C. Söderlund, in Investment Arbitration and the ECT, 237. See also J. Shookman, in 27(4) J.Int’l Arb. (2010), 362; R.F. Hansen, in 73(4) Mod.L.Rev. (2010), 528.

18  See Lauder v. Czech Republic (UNCITRAL), Award of 3 September 2001; CME v. Czech Republic (UNCITRAL), Partial Award of 13 September 2001.

19  See C. McLachlan, Lis Pendens in International Litigation, 38.

20  See J.D.M Lew, in Parallel State and Arbitral Procedures, 308.

21  N. Blackaby, in Pervasive Problems, 219. See also G. Van Harten and M. Loughlin, 17(1) E.J.I.L. (2006), 140.

22  See S. Wilske, M. Raible, and L. Markert, 1(2) Cont. Asia Arb.J. (2008), 215. It goes without saying that while ICSID proceedings are necessarily included in this definition of investment arbitration (Art. 25(1) ICSID requiring that disputes arise ‘directly out of an investment’), the definition is not limited to ICSID proceedings. Indeed, many investment arbitrations take place outside the ICSID framework, see Van Harten and Loughlin (n. 21) 127.

23  See also G. Sacerdoti, 19 ICSID Rev. (2004), 13.

24  As a consequence, investment arbitration appears as a subcategory of what is often referred to as ‘mixed’ or ‘transnational’ arbitration, see W. Ben Hamida, L’arbitrage transnational unilatéral, 6.

25  See also Wilske, Raible, and Markert (n. 22) 224.

26  For a different understanding, see Glavinis (n. 7) 235.

27  See Ch. 2, para. 2.65.

28  See also the definition of ‘Treaty Dispute’ as a ‘dispute arising from an alleged breach of an investment- related treaty’, Glavinis (n. 7) 235.

29  By contrast, Douglas defines ‘treaty tribunals’ and ‘contractual tribunals’ by reference to the instrument pursuant to which a tribunal is established, see Z. Douglas, 74 B.Y.I.L. (2003), 155.

30  As a consequence, commentators frequently do not specify whether they consider the basis of the substantive claims or the origin of the arbitration agreement as the decisive criterion for qualifying proceedings as a treaty arbitration, see e.g. B.M. Cremades, in Parallel State and Arbitral Procedures, 8; B.M. Cremades and I. Madalena, 24(4) Arb.Int’l (2008), 509. See also Glavinis, who elsewhere refers to ‘Treaty Arbitration’ as an ‘arbitration, conducted pursuant to the dispute settlement provisions contained in an investment-related treaty’, Glavinis (n. 7) 235.

31  See Ch. 3, paras 3.53, 3.73.

32  See Ch. 5, para. 5.72.

33  As a consequence, the notions of ‘treaty arbitration’ and ‘treaty-based arbitration’ are not used synonymously by the present study. The former refers to the fact that an arbitration deals (at least partially) with claims arising under an investment treaty, while the latter indicates that the host state’s consent to arbitration has been granted through the dispute resolution mechanism in an IIA.

34  See Y. Shany, 99 A.J.I.L. (2005), 838. That multiple proceedings are seen as an inherently negative phenomenon is also confirmed by the fact that domestic legal systems typically seek to prevent them, see Y. Shany, Competing Jurisdictions (2003), 161.

35  See CME v. Czech Republic (UNCITRAL), Partial Award of 13 September 2001, paras 419, 525; Wena Hotels v. Egypt (ICSID), Decision on Annulment of 28 January 2002, para. 49; Azurix v. Argentina (ICSID), Decision on Jurisdiction of 8 December 2003, para. 101; Camuzzi International v. Argentina (ICSID), Decision on Objections to Jurisdiction of 11 May 2005, para. 91; Sempra Energy v. Argentina (ICSID), Decision on Objections to Jurisdiction of 11 May 2005, para. 102; Bayindir v. Pakistan (ICSID), Decision on Jurisdiction of 14 November 2005, para. 270; Getma International v. Guinea (ICSID), Decision on Jurisdiction of 29 December 2012, para. 108; K. Hobér, in Parallel State and Arbitral Procedures, 246; Shany (n. 16) 19; H. Bagner, 5(1) J.W.I.T. (2004), 34; T. Wiwen-Nilsson, Stockholm Arb.Rep. (2003), 257; E. Wu, 6(2) Asian Arb. J. (2010), 134; J.O. Voss, The Impact of Investment Treaties on Investor-State Contracts, 280.

36  See Sedelmayer v. Russia (SCC), Dissenting Opinion of I.S. Zykin to Award of 7 July 1998, 3; Z. Douglas, The International Law of Investment Claims, 417; Dimsey (n. 7) 99; Wu (n. 35) 134.

37  See Camuzzi International v. Argentina (ICSID), Decision on Objections to Jurisdiction of 11 May 2005, para. 91; Sempra Energy v. Argentina (ICSID), Decision on Objections to Jurisdiction of 11 May 2005, para. 102; Suez and Interaguas v. Argentina (ICSID), Decision on Jurisdiction of 16 May 2006, para. 51; Suez and Vivendi v. Argentina (ICSID), Decision on Jurisdiction of 3 August 2006, para. 51. By contrast, Sacerdoti suggests that the risk of double-compensation should not be taken into account by a second forum when awarding damages, and should only become relevant at the level of enforcement, where it would be ‘for the country where recovery would be sought…to determine under domestic law whether any previous recovery…should be taken into account as a deduction’, G. Sacerdoti, 2(5) T.D.M. (2005), 127. Against this possibility see Wiwen-Nilsson (n. 35) 257. Also sceptical is Hansen (n. 17) 537. As another means to avoid the possibility of double recovery, tribunals have occasionally envisaged an order directed at a successful claimant to desist from any further actions seeking the same redress, see Occidental Exploration v. Ecuador (LCIA, UNCITRAL), Award of 1 July 2004, para. 216.

38  See Pan American Energy v. Argentina (ICSID), Decision on Preliminary Objections of 27 July 2006, para. 219. The difficulties related to the compensation of separate entities situated at different levels of an investment structure were also clearly anticipated by the ICJ in its Barcelona Traction decision, see Barcelona Traction (Belgium v. Spain), ICJ Reports (1970), 49–50.

39  See GAMI Investments v. Mexico (UNCITRAL), Award of 15 November 2004, paras 116–121; Hobér (n. 35) 246. By contrast, in the reverse situation where the company has been awarded damages first, such payments can be taken into account when calculating the damage suffered by the shareholder, see Lauder v. Czech Republic (UNCITRAL), Award of 3 September 2001, para. 172; Daimler Financial Services v. Argentina (ICSID), Award of 22 August 2012, para. 155; Hobér (n. 35) 246.

40  See also Douglas (n. 36) 417. For an apparently different view see Renta 4 v. Russia (SCC), Award of 20 July 2012, para. 34.

41  One might consider the possibility of awarding the full amount of damages to the company, while at the same time giving the respondent a claim for unjust enrichment against the shareholder-claimant in the first proceedings. However, since the forum in the second proceedings would normally not have jurisdiction with regard to this latter claim, it could only be asserted in a third set of proceedings. A different solution is envisaged by Wu, who suggests a ‘tiered approach’, where indirect investors could bring claims only once the more direct investors had either already brought theirs or waived their rights to do so, see Wu (n. 35) 134, 143. See also Douglas (n. 36) 455; J. Crawford, in The Art of Advocacy, 312.

42  See Söderlund (n. 17) 245. See also Cremades and Madalena (n. 30) 508; Schreuer (n. 5) Art. 26, para. 109; P. Mayer, 136(1) J.Droit Int’l (2009), 78; Shookman (n. 17) 363.

43  See Lew (n. 20) 309. Since different forums may appreciate the same set of circumstances differently, unless a case is either completely straightforward (thus entailing a probability of success of 100%) or entirely hopeless (thus having a probability of success of 0%), an investor’s chances might increase with the number of proceedings in which the same claims are being asserted. To illustrate this idea, one may assume that the strength of an investor’s case provides him with a 50% chance of success before any given forum. If the investor can bring this dispute before two different forums, which do not take into account each other’s decision, the investor’s overall chance of being successful in at least one forum is already 75%. If the investor can bring the same dispute before a third forum, this raises his overall chance of success to 87.5%—a significant increase on the investor’s original prospects of securing a successful outcome.

44  See McLachlan (n. 19) 289; C. McLachlan, L. Shore, and M. Weiniger, International Investment Arbitration, 123. This rationale would seem to have been ignored by the tribunal in SPP when stating that ‘[s]o long as the arbitral and judicial processes do not produce more than one enforceable remedy…it matters not how many different paths are pursued in an effort to obtain that remedy’, SPP v. Egypt (ICSID), Decision on Jurisdiction of 27 November 1985, para. 61.

45  See Bayindir v. Pakistan (ICSID), Decision on Jurisdiction of 14 November 2005, para. 270; G. Delaume, in The Changing World of International Law, 175; Douglas (n. 29) 257; G. Kaufmann-Kohler, in Annulment of ICSID Awards, 202; A. Reinisch, in The Backlash Against Investment Arbitration, 115; E. Gaillard, 130 J.Droit Int’l (2003), 236; Hobér (n. 35) 246; G. Cuniberti, 21 ICSID Rev. (2006), 395; V. Lowe, 8 A.J.I.C.L. (1996), 48; Schreuer (n. 5) Art. 26, para. 109; S. Lemaire, Rev. arb. (2006), 392; Shany (2003) (n. 34) 80; Y. Shany (n. 45) 36; Hansen (n. 17) 528; Voss (n. 35) 280.

46  See Cuniberti (n. 45) 416. Reinisch distinguishes between divergent views on legal issues, the divergent assessment of identical facts and conflicting results in the same dispute, Reinisch (n. 45) 115–117; A. Reinisch, in International Investment Law for the 21st Century, 906. See also Dimsey (n. 7) 74.

47  See Cuniberti (n. 45) 418. A number of commentators have suggested that where the conflicting decisions belong to different legal systems, a conflict may be less of a concern, since in that case decisions should not be presumed to be coherent in the first place, see Cuniberti (n. 45) 395; Shany (n. 16) 132; Shany (2003) (n. 34) 82. At the same time, in particular where proceedings encounter a significant degree of public interest and decisions are published, any divergences in the factual findings or the reasoning of different forums may have a damaging effect on the general perception of the system’s legitimacy, see J. Gill, 2(2) T.D.M. (2005), 13; Shany (2003) (n. 34) 80; Shany (n. 45) 36; Shany (n. 16) 18; Mayer (n. 42) 78; S.D. Franck, 73 Fordham L.Rev. (2005), 1583; Reinisch (n. 45) 115; C. Schreuer and A. Reinisch, Legal Opinion in Czech Republic v. CME (Svea Court of Appeal), 107; C. Schreuer and A. Reinisch, Legal Opinion in CME v. Czech Republic, 41; J. Werner, 4(5) J.W.I. (2003), 786; Shookman (n. 17) 362; C.N. Brower, 36 Vand.J.Transnat’l L. (2000), 52.

48  The most obvious example would be the situation where one forum explicitly states that the consequences of the decision of another forum should be reversed, see R. Dolzer and C. Schreuer, Principles of International Investment Law, 220.

49  See Cuniberti (n. 45) 418. Given the possible interaction between different legal orders, it would not seem to matter whether or not the conflicting decisions were originally rendered within the same legal order, see Cuniberti (n. 45) 419–420; S. Brekoulakis, 16 Am.Rev.Int’l Arb. (2005), 179. In particular, a respondent might be able to prevent the enforcement of a decision granting the requested remedy in a given jurisdiction by obtaining the prior recognition of a decision denying that remedy. As a consequence, there would be a risk that parties might engage in a ‘race to enforcement’ in the relevant jurisdictions. For that reason the suggestion that conflicts between decisions should be resolved at the level of enforcement in accordance with each state’s national laws (see G. Sacerdoti, in Appeals Mechanism in International Investment Disputes, 136) does not seem practical.

50  See Dolzer and Schreuer (n. 48) 220; C. Schreuer, 4 L.P.I.C.T. (2005), 12; Douglas, (n. 29) 257; Hansen (n. 17) 529.

51  See Schreuer (n. 5) Art. 26, para. 109; Schreuer (n. 50) 12; C. Schreuer, in Investment Treaty Law, 162; Hansen (n. 17) 529; Dolzer and Schreuer (n. 48) 220; Douglas (n. 29) 257; Reinisch (n. 45) 115; Kaufmann-Kohler (n. 45) 202; Cuniberti (n. 45) 395; Lowe (n. 45) 48; Shookman (n. 17) 362; Voss (n. 35) 281; I.P. Amaza, 6(2) Disp.Resol.Int’l (2012), 153. This is true both with regard to the parties’ resources and, at least as far as court litigation is concerned, to the resources of the judicial system involved, see Shany (n. 16) 17, 155; Shany (2003) (n. 34) 80; Shany (n. 45) 36. Some authors take the view that the cost of multiple proceedings should be less of an issue in investment disputes, as both investors and host states would have sufficient means to afford them, see Cuniberti (n. 45) 414; Brekoulakis (n. 49) 179.

52  The reason for the invocation of the two different treaties appears to have been that the claimants wanted to avail themselves of the protection under the Czech Republic-US BIT ‘for political reasons’, while at the same time avoiding potential jurisdictional problems with regard to that treaty, which they anticipated would not exist under the Czech Republic-Netherlands BIT, see J. Carver, 5(1) J.W.I.T. (2004), 26.

53  These included the full consolidation of the proceedings into one arbitration, the appointment of the same arbitrators in both disputes, the agreement that the second tribunal would be bound by the determinations on the existence of a treaty breach by the first tribunal, and the postponement of the hearing in the second arbitration until after an award in the first proceedings was issued, see CME v. Czech Republic (UNCITRAL), Final Award of 14 March 2003, para. 427.

54  See CME v. Czech Republic (UNCITRAL), Final Award of 14 March 2003, para. 428.

55  See Lauder v. Czech Republic (UNCITRAL), Award of 3 September 2001, para. 319; CME v. Czech Republic (UNCITRAL), Partial Award of 13 September 2001, para. 624.

56  See CME v. Czech Republic (UNCITRAL), Final Award of 14 March 2003, para. 650.

57  See Lew (n. 20) 309; A. Crivellaro, in Parallel State and Arbitral Procedures, 112; Hobér (n. 35) 242. See also para. 1.34 and Ch. 4, para. 4.100.

58  See also Söderlund (n. 17) 246.

59  T.W. Wälde, 5(1) J.W.I.T. (2004), 43. Another commentator formulated ironically that ‘[t]he practical lesson to be learnt from the CME case must lead to carefully structured investments to ensure that the ultimate investor can take advantage of as many BITs (and BIT arbitrations) as possible, thereby ensuring multiple bites at the cherry until success is ensured’, V.V. Veeder, in Complex Arbitrations, ICC Special Supplement 2003, 78. As a consequence, a number of practitioners have predicted that the CME case will ‘not remain an isolated incident’ (A. Reinisch, 3 L.P.I.C.T. (2004), 41), and that one may well have to expect not only ‘twin’ inconsistent decisions, but even ‘triplets and quadruplets’, see Franck (n. 47) 1583.

60  This would appear to be the likely explanation for the ‘almost surprising’ observation that the ‘dangers associated with the proliferation of dispute settlement, such as forum shopping and multiplication of proceedings, have materialized only to a limited degree’, see Reinisch (n. 45) 115.

61  The hope expressed by some commentators that common sense and courtesy might prevent conflicts in this regard (see N. Gallagher, in Pervasive Problems, 356; Bagner (n. 35) 35; Dimsey (n. 7) 85; Ben Hamida (n. 24) 415; P. Turner, in International Arbitration 2006, 470) must arguably be viewed with some scepticism. It would certainly seem preferable to identify formal coordination rules that would make multiple proceedings more predictable for the parties, see also T.W. Wälde, 2(2) T.D.M. (2005), 76; Shookman (n. 17) 378.

62  See Douglas (n. 29) 236.

63  Thus Cremades and Cairns take the view that ‘it will usually be prudent for an investor to confine its legal action to a single forum’, Cremades and Cairns (n. 12) 18. The same authors assume that investors will typically prefer the enforcement of their treaty rights over the enforcement of contractual rights, see Cremades and Cairns (n. 12) 18; Cremades and Madalena (n. 30) 527. At the same time, an investor’s choice in this regard is likely to depend on the respective strength of his contractual and treaty claims, see R. Happ, in International Investment Protection and Arbitration, 64.

64  This is true in particular where the roles of claimant and respondent in two sets of proceedings are inverted, see para. 1.23. In addition, as will be shown, in some cases the application of a coordination mechanism may depend on the question of whether or not the forum in the other set of proceedings has jurisdiction over the claims before it (see Ch. 6, para. 6.112 and Ch. 7, para. 7.42), thus making it necessary to address this as a preliminary issue.

65  Nevertheless certain conclusions may arguably be drawn in this regard from the analysis of the relationship between treaty tribunals and other types of forums, see Ch. 5, para. 5.77.

66  See Shany (2003) (n. 34) 239; V. Lowe, 20 Aust.Y.B.I.L. (1999), 201. By contrast, a principle of lis pendens has recently been incorporated in a number of international initiatives relating to the coordination of proceedings, see P. Mayer, 327 Recueil des cours (2007), 295. In particular, Art. 21 of the Preliminary Draft Convention on Jurisdiction and Foreign Judgments in Civil and Commercial Matters and Principle 2.6 of the ALI/UNIDROIT 2004 Draft Principles of Transnational Civil Procedure contain provisions on lis pendens, see also McLachlan (n. 19) 414, 418.

67  See e.g. L. Lévy, in Anti-Suit Injunctions in International Arbitration, 128; Hansen (n. 17) 532; Hobér (n. 35) 258.

68  The relationship between investment treaty arbitration and state-to-state disputes must be distinguished from the different issue of the extent to which statements and interpretations of the text of an IIA rendered by the signatories in the course of ongoing investor-state proceedings can be relevant for interpretation purposes. With regard to the latter issue, Art. 31(3)(a) VCLT in principle provides that ‘any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions’ must be taken into account together with the context of a provision, see also M. Kinnear, in International Arbitration 2006, 436. Nevertheless, it has been claimed that ‘a mechanism whereby a party to a dispute is able to influence the outcome of judicial proceedings, by issuing official interpretation[s] to the detriment of the other party, is incompatible with principles of a fair procedure’, see C. Schreuer, 3(2) T.D.M. (2006), 20. See also Ben Hamida (n. 24) 468. This allegation of procedural unfairness seems exaggerated where the treaty’s signatories limit themselves to an interpretation (rather than an amendment) of the agreement, as even an interpretation unfavourable to an investor should hardly ever come as a complete surprise, see also B. Stern, in International Arbitration 2006, 572. In accordance with such an understanding, the tribunal in CME agreed to consider the interpretation provided by the governments of the Netherlands and the Czech Republic regarding certain issues arising under their BIT following the rendering of its partial award as ‘conclusive and binding’ for purposes of its final award, see CME v. Czech Republic (UNCITRAL), Final Award of 14 March 2003, para. 217. In the case of NAFTA, Art. 1131(2) institutionalizes an interpretative mechanism by providing that ‘[a]n interpretation by the [NAFTA Free Trade] Commission of a provision of this Agreement shall be binding on a Tribunal established under this Section’. The Free Trade Commission used this power to issue an interpretation of Art. 1105 NAFTA, which the tribunal in Methanex took into account as a ‘subsequent agreement’ on interpretation in accordance with Art. 31(3)(a) of the Vienna Convention, see Methanex v. US (UNCITRAL), Award of 3 August 2005, para. 21; Kinnear, in International Arbitration 2006, 437, 438. The need to distinguish between an interpretation and an amendment for the purposes of Art. 1131(2) NAFTA was stressed by the tribunal in Pope and Talbot v. Canada (UNCITRAL), Award of 31 May 2002, para. 24.

69  These include the possibility of proceedings before the WTO, see D.M. Price, in Parallel State and Arbitral Procedures, 75; W. Ben Hamida, 3(2) T.D.M. (2006), 20.

70  See Reinisch (n. 59) 42; Douglas (n. 29) 237.

71  See Price (n. 69) 73; A.R. Parra, 12 ICSID Rev. (1997), 335.

72  IIAs typically provide for a mechanism for the settlement of disputes between signatories in relation to the interpretation or implementation of their agreement, see e.g. Art. 27 ECT.

73  See Price (n. 69) 73. As a consequence, this type of state-to-state arbitration has occasionally been mentioned as a possible means to avoid conflicting outcomes in different investor-state proceedings. The same technique has been used before the Iran-US Claims Tribunal, see Iran v. US (Dual Nationality), Decision of 6 April 1984, 5 Iran-US C.T.R., 252. At the same time, such a concept would seem to be limited in that it can only apply with regard to investors protected under the same treaty, see G. Kaufmann-Kohler, L. Boisson de Chazournes, V. Bonnin, and M.M. Mbengue, 21 ICSID Rev. (2006), 77.

74  See C. Schreuer, in The Law of International Relations, 350. Occasionally it has been suggested that an inter-state tribunal should consider the initiation of proceedings by a host state with the purpose of avoiding the obligations accepted vis-à-vis an investor as an abuse of rights that would justify its declining of jurisdiction, see F.O. Vicuña, in Parallel State and Arbitral Procedures, 213. By contrast, most commentators seem to agree that Art. 27 ICSID or a similar provision in an IIA does not generally prevent the initiation of a dispute relating to the interpretation of a treaty, even where such an interpretation might affect the outcome of pending investor-state proceedings, see Kaufmann-Kohler, Boisson de Chazournes, Bonnin, and Mbengue (n. 73) 76; J. Kokott, in ILA, Report of the 70th Conference, 265; Ben Hamida (n. 69) 17.

75  This can be exemplified by the Industria Nacional de Alimentos v. Peru proceedings, where Peru (being sued by the investor before ICSID) initiated state-to-state arbitration proceedings under the Chile-Peru BIT only to request a stay of the ICSID arbitration, see Industria Nacional de Alimentos v. Peru (ICSID), Award of 7 February 2005, para. 7. The ICSID tribunal rejected the request and Peru did not move forward with the proceedings under the state-to-state mechanism, see Price (n. 69) 74; R. Dolzer and C. Schreuer, Principles of International Investment Law, 214; Schreuer (n. 74) 351. Obviously, if the mere initiation of state-to-state disputes were considered a sufficient ground for ordering a stay by investment treaty tribunals, this would provide respondent states with a readily available means of obstructing proceedings brought against them, see Price (n. 69) 73; Schreuer (n. 74) 351.

76  See B. Juratowitch, 23 ICSID Rev. (2008), 34; Douglas (n. 29) 166.

77  See Italy v. Cuba (ad hoc arbitration), Decision on Preliminary Objections of 15 March 2005, para. 65. While some authors see this as the consequence of a general opt-out of the inter-state secondary rules on international responsibility with regard to investors’ rights (see Douglas (n. 29) 190), others appear to suggest that home states should be ‘estopped’ from exercising diplomatic protection under such circumstances, see Juratowitch (n. 76) 22. Art. 27 ICSID and a number of IIAs specifically provide that the exercise of diplomatic protection by an investor’s home state is excluded during the pendency of investor-state proceedings, see also Ch. 3, para. 3.104.

78  Suggested approaches to the problem of conflicting decisions in the broadest sense that would require a significant modification of the international investment framework include preliminary rulings similar to the procedure contained in Art. 267 of the Treaty on the Functioning of the European Union (see Kaufmann-Kohler (n. 45) 221; G. Kaufmann-Kohler, 23(2) Arb.Int’l (2007), 378; Schreuer (n. 68) 23; Reinisch (n. 45) 118) and an appellate mechanism for arbitral awards, see e.g. W.H. Knull and N.D. Rubins, 11 Am.Rev.Int’l Arb. (2000), 531–565; Schreuer (n. 68) 20; Reinisch (n. 45) 118; Reinisch (n. 46) 910. For a sceptical assessment of the extent to which an appellate body could prevent conflicting outcomes in different proceedings see W. Kühn, 5(1) J.W.I.T. (2004), 17; I. Laird and R. Askew, 7(2) J.App.Pract.Proc. (2005), 299; B. Legum, in Appeals Mechanism in International Investment Disputes, 237; Gallagher (n. 61) 354. See also J. Clapham, 26(3) J.Int’l Arb. (2009), 437. A different approach, aiming to restrict the number of possible claimants through an international convention is suggested by Kühn, 5(1) J.W.I.T. (2004), 17. See also Bagner (n. 35) 34; Stern (n. 68) 576; Hansen (n. 17) 542.