1 Introduction
The Articles on State Responsibility for Internationally Wrongful Acts (‘ARSIWA’) constitute an experiment in international law-making. Unlike other successful projects of the International Law Commission (‘ILC’), such as its work on the law of treaties and diplomatic and consular relations, the ARSIWA have not yet led to the adoption of a multilateral treaty.Footnote 1 Yet, their text is cited commonly as the authoritative statement of the law on State responsibility with investment tribunals being by far their most frequent users. To put this into perspective, in a 2017 report to the UN General Assembly, the UN Secretariat identified 392 publicly available decisions of various bodies which make reference to the ARSIWA including those of the ICJ, the ICC, the WTO, international and hybrid criminal tribunals, and human rights courts and treaty bodies.Footnote 2 Although the report does not provide a specific number for investment arbitrations, it records 264 arbitral decisions referencing ARSIWA (72.5%) with investment tribunals accounting for the majority of these references.Footnote 3 At the same time, the interpretation and/or application of ARSIWA is one of the most common issues arising in investment arbitration. In numerical terms, 444 cases have led to a decision since 2000 including cases in which no issues of responsibility arose or in which the reasoning of the decision is not public.Footnote 4 The present study has traced at least 200 decisions issued in the same period citing ARSIWA or its previous versions. The extent of the practice attests to the central importance of the formally unwritten law of State responsibility and its interpretation in investment proceedings. Indeed, investment tribunals refer to this body of law to determine a variety of key issues including whether the acts forming the basis of the claim belong to the respondent state, what are the consequences of a finding against the respondent state, or whether there are circumstances precluding a finding against the respondent state or calling for mitigation. The sheer volume of the practice also raises broader questions about the ways in which investment tribunals engage with the identification and development of unwritten international law, that is, customary international law and general principles of law.
This chapter examines and also critically assesses the methods which investment tribunals explicitly or implicitly employ when using the ARSIWA in order to identify rules of general international law on State responsibility and determine their content. It does so by building upon, and adding to, the work of the UN Secretariat.Footnote 5 Section 2 introduces the problem of ARSIWA’s lack of formal status and its implications for the use of ARSIWA in international adjudication. Section 3 surveys whether and how investment tribunals justify their reliance on ARSIWA. Section 4 highlights the variety of ways in which ARSIWA are used in the process of the determination of the content of applicable rules of law. Section 5 discusses the outcomes of this survey against the analytical backdrop of the unity of the law on State responsibility and the law relating to sources of international law. The chapter argues in favour of a principled use of ARSIWA in investment proceedings based on the distinction between the ascertainment of the legal status of a normative proposition contained therein and the determination of the content of a normative proposition whose status is undisputed.
2 Formal Status of ARSIWA and Its Methodological Corollaries
From a ‘formal’ perspective, the ARSIWA, as a document originating from the International Law Commission, does not possess any binding force.Footnote 6 In terms of the Statute of the ICJ, the ARSIWA constitutes ‘teachings of the most qualified publicists’ that can be used as ‘subsidiary means for the determination of rules of law’.Footnote 7 Yet, in fact, international courts and tribunals tend to attach to ARSIWA much more weight than the label of ‘teachings’ would normally suggest and often treat them as uncontroversial statements of applicable rules of law.Footnote 8 This section briefly maps out the available justifications and methodologies for the use of ARSIWA in investment arbitration and points out why the practice of investment tribunals matters.
To start, investment tribunals are constituted for the settlement of a dispute between a State and a national of another State arising out of an investment which the parties have consented to submit to arbitration.Footnote 9 Whether investment tribunals can apply in this context also other rules of international law including customary international law and general principles of law is a procedural issue.Footnote 10 Relevant procedural rules commonly uphold in the first place the autonomy of the parties to determine the applicable law,Footnote 11 but provide residually for the application of ‘applicable rules of international law’ in the absence of an agreement or when the tribunal determines such law to be appropriate.Footnote 12 Besides, even when the rules of general international law are not deemed directly applicable to a specific issue, a tribunal might decide to take them into account as relevant rules for the interpretation of the applicable IIA.Footnote 13 Questions of identification of the international law on State responsibility arise against this procedural background. By implication, investment tribunals turn to ARSIWA for the purposes of identifying rules of international law external to the IIA in question or otherwise applicable to the case. Accordingly, investment tribunals tend to justify their reliance on ARSIWA by using the terminology of ‘formal’ sources of international law, albeit with varying degrees of sophistication.
In the first place, the well-recorded discrepancy between the ‘formal’ status of ARSIWA as a source of law and their effective ‘authority’ in the context of investment arbitration calls for a more detailed survey into how investment tribunals justify their use of ARSIWA.Footnote 14 With respect to customary international law, the ILC has opined – in a somewhat self-aggrandising manner – that ‘a determination by the Commission affirming the existence and content of a rule of customary international law may have particular value, as may a conclusion by it that no such rule exists’.Footnote 15 This is so notwithstanding the fact that the ILC outputs cannot constitute evidence of State practice or opinio juris in and of themselves, as they do not originate directly, or even indirectly, from States.Footnote 16 Moreover, as the late Judge Crawford recounts, the ARSIWA ‘have been derived from cases, from practice, and from often unarticulated instantiations of general legal ideas’.Footnote 17 This implies that the ARSIWA are not necessarily a monolith from the perspective of sources of international law. Even assuming that some normative propositions included in the ARSIWA do not strictly adhere to the standards of identification of customary international law, tribunals might still have recourse to them as articulations of underlying general principles of law.Footnote 18 In a nutshell, there are multiple justifications available to investment tribunals for their use of ARSIWA.
Besides, apart from why investment tribunals rely so much on ARSIWA, a closely related question is also how they make use of ARSIWA. In their final form, the ARSIWA consist of provisions articulated in prescriptive terms much like the draft of a treaty.Footnote 19 It could be argued that an international court or tribunal having recourse to them should approach them in a way akin to any document having binding effect.Footnote 20 In so doing, the court or tribunal should follow a methodology that builds upon the rules of treaty interpretation but also accounts for the fact that the ARSIWA do not originate directly from States but from a technical body of the UN.Footnote 21 However, the traditional perception of ILC outputs as ‘teachings’ is inimical to according any particular value to the views of the ILC as such, the ILC being a body of legal experts not of representatives of States.Footnote 22 Rather, a decision-maker should focus on the evidence that the ILC adduces for the existence of a rule and reconstruct the content of the rule in question on the basis of that evidence.Footnote 23 In more practical terms, the ILC outputs on the topic not only encompass the ARSIWA and their accompanying commentaries, but also a multitude of documents including the previous reports of the Commission, comments by governments, the summary records of discussions within the plenary including the reports of drafting committees, and the reports of the special rapporteurs.Footnote 24 A combined reading of these documents reveals ‘titanic disagreements’ on virtually all issues, which are imprinted in the carefully balanced language of the final text.Footnote 25 In this respect, the traditional label of ‘teachings’ provides little guidance as to how to navigate through all these materials in determining applicable rules of law, as it treats all these materials indistinctly.Footnote 26
In this context of contestation, the premise of this study is that the practice of investment tribunals, as the most frequent users of ARSIWA, can further elucidate the connections between the ARSIWA and the framework of sources of international law, and shed light on the methodology for their proper use. As a preliminary point, the practice of investment tribunals regarding ARSIWA has been paralleled to ‘a drowning man …grab[bing] a stick in the sea in the hope of having certainty’.Footnote 27 More fundamentally, it has been argued that the power of the international judge to resolve an international dispute necessarily entails a certain degree of discretion as to the identification of applicable rules and the determination of their content.Footnote 28 Hence, the tendency of investment to rely on ARSIWA could be dismissed as an instantiation or corollary of such discretion or ‘expediency’ in international judicial decision-making.Footnote 29 Yet, whilst it is arguable that international judges enjoy more leeway than domestic ones in identifying and determining the content of applicable rules of law, they are not entirely uninhibited by any rules or principles.Footnote 30 In resolving competing claims as to the identification and determination of the content of rules of State responsibility, investment tribunals have the incentive –and often do– justify thoroughly their legal findings.Footnote 31 Apart from dispelling any impression of bias or arbitrariness, investment tribunals have the strong incentive to pre-empt the annulment of the award or to prevent any domestic obstacles in the implementation of their decision.Footnote 32 Therefore, the practice of investment tribunals engaging with ARSIWA cannot be reduced easily to mere expediency, but could constitute evidence of existing or emerging rules or principles, or at least, good practices, for the identification and interpretation of applicable rules of unwritten international law.
3 Patterns of Justification for the Use of ARSIWA in International Investment Arbitration
At first glance, a majority of investment tribunals give specific reasons for relying on ARSIWA, albeit a significant number of tribunals are silent on the matter. In purely quantitative terms, out of a sample of 205 decisions surveyed for the purposes of this study, 144 provide separately a justification for their reliance on ARSIWA or its previous versions. Conversely, 61 decisions cite the ILC Articles without explicitly providing a reason for doing so. On the one hand, the significant number of decisions that lack any justification for reliance on ARSIWA give support to the argument that the distinction between identification and determination of content is not water-tight not only in theory but also in practice.Footnote 33 However, a closer look at the reasoning of the tribunals depicts a much more diverse and nuanced picture than a numerical presentation suggests. In some cases, the tribunals’ stance towards the nexus between ARSIWA and the sources of international law becomes apparent from the context of the reference or the decision notwithstanding the lack of an expressis verbis or clear-cut justification. On the other hand, the tribunals provide a wide variety of justifications which do not always consist of a clear link between the ARSIWA and the ‘formal’ source of the applicable rule. Besides, the same decision might follow different lines of reasoning with respect to different provisions of ARSIWA or rely on ARSIWA only partly for the identification of applicable rules on State responsibility.Footnote 34 Mindful of these difficulties, the present subsection attempts to flag up common patterns in the ways in which investment tribunals justify their reliance on ARSIWA, whereas the following subsection focuses on how they use ARSIWA for the determination of the content of the applicable rule.
From the outset, tribunals recognise that the ARSIWA have no formally binding status as such either implicitly or, less often, explicitly.Footnote 35 However, remarkably, this survey has found only one award that explicitly characterises the ARSIWA – and, particularly, the ILC’s Commentary – as ‘works of highly qualified writers’,Footnote 36 despite this characterisation being relatively uncontroversial in theory with respect to works of the ILC.Footnote 37 The underlying reason for this discrepancy seems to be the ensuing discrepancy between the relative value which such sources are to be accorded generally in the determination of applicable rules according to the ICJ Statute and the effective authority of ARSIWA in the context of investment arbitration.
In the first place, tribunals use the ARSIWA in the process of identifying rules stemming from customary international law. In most cases, investment tribunals affirm the applicability of ARSIWA, because they ‘reflect’,Footnote 38 ‘codify’,Footnote 39 ‘state’,Footnote 40 ‘restate’,Footnote 41 ‘express’,Footnote 42 ‘formulate’,Footnote 43 ‘articulate’,Footnote 44 ‘represent’,Footnote 45 or ‘are declaratory of’Footnote 46 rules or principles of customary international law on State responsibility. Very often, these findings are couched in axiomatic terms without any further explanation or are reasoned in such vague terms so as to amount to little more than assertions.Footnote 47 When they do reason such findings, tribunals tend to uphold the authority of ARSIWA because of the evidence they rely upon,Footnote 48 their particular drafting process,Footnote 49 or their subsequent reception in practice.Footnote 50 In this latter respect, several tribunals have taken note of the fact that the General Assembly has annexed ARSIWA to a resolution and commended them to the consideration of States.Footnote 51 Other tribunals refer to the pronouncements of other international courts or investment tribunals finding that certain provision of ARSIWA reflect customary international law.Footnote 52 Whatever the specific line of reasoning, the common thread between these decisions is the finding that ARSIWA or a specific provision has decisive value for the identification of customary international law on this matter. In practical terms, the normative propositions contained in ARSIWA are treated as having the status of – or, more precisely, as materially identical with – rules of customary international law.
Second, a significant number of decisions use ARSIWA as means to identify rules of general application without explicitly utilising the terminology of customary international law. On the one hand, there are decisions mentioning ARSIWA in the context of identification of general principles of law. In this respect, tribunals usually declare that certain provision of ARSIWA or a statement in the Commentary is generally recognised in domestic legal systems without engaging in any detailed comparative examination or independently assessing its transposability in international law.Footnote 53 On the other hand, quite often, tribunals turn to ARSIWA as means for the identification of international law without deciding or clarifying what is the particular status of the underlying rules. In most cases, it is impossible to discern whether there are any legal reasons for such ambiguity or whether it is just a product of idiosyncratic drafting.Footnote 54 However, in some cases, the context of the decision reveals deeper concerns about the applicability of the rules on State responsibility reflected in ARSIWA in the investor-State context.Footnote 55 So, for instance, in the Jan de Nul Award, the Tribunal declared that the General Assembly resolution, to which ARSIWA are annexed, ‘is considered as a statement of customary international law on the question of attribution for purposes of asserting the responsibility of a State towards another State’.Footnote 56 Despite this finding, it held that ARSIWA was ‘applicable by analogy to the responsibility of States towards private parties’.Footnote 57 A similar issue arose in Vestey where Venezuela argued that the principle of full reparation reflected in Art 31 ARSIWA was inapplicable in the case of unlawful expropriation.Footnote 58 In response, the Tribunal held that ‘while the ILC Articles govern a State responsibility vis-à-vis another State and not a private person, it is generally accepted that the key provisions of the ILC, such as Article 31(1) can be transposed in the context of the investor-State disputes.’Footnote 59 Relying on judicial decisions applying the principle of full reparation in the context of expropriation,Footnote 60 the Tribunal concluded that ‘Venezuela must provide full reparation under customary international law’.Footnote 61 In all these cases, tribunals treat ARSIWA as definitive statements of applicable rules of law, albeit with some ambiguity as to the precise source of such rules. Apart from general principles of law common to domestic laws, it is not apparent whether tribunals apply customary international law or refer to an altogether different category of general principles of law emanating from within international law.Footnote 62
Another strand of decisions seems to take into account the parties’ mutual reliance on ARSIWA in their submissions so as to affirm their applicability in the proceedings. Notably, some tribunals appear to treat the parties’ agreement as the sole basis for the application of a provision of ARSIWA in the proceedings without any finding on whether such provision actually reflects a rule stemming from a ‘formal’ source of international law.Footnote 63 So, for instance, in EDF, the Tribunal found it unnecessary to take a position on ‘the theoretical question of how far the various aspects of ILC Article 25 codifi[ed] customary defenses related to necessity’.Footnote 64 The Tribunal took note that ‘both sides in this arbitration stipulate[d] that the Tribunal‘s analysis should take as applicable legal norms the State of Necessity defen[c]e presented by the contours articulated in ILC Article 25’ and found that ‘the standards urged by both sides, as providing applicable norms’.Footnote 65 In these cases, the emphasis of tribunals seems to be on the relevance of party autonomy for the determination of applicable law rather than its identification on the basis of the relevant rules of international law. Yet, it could be argued that the parties’ agreement has only evidentiary value as to the status of the rule reflected in certain provision of ARSIWA so as to obviate the need for further independent analysis.Footnote 66 So, for instance, the annulment committee in Continental Casualty upheld the legal findings of the Tribunal on applicable law on the basis that ‘it was not disputed by either party that Article 25 of the ILC Articles codified the customary international law principles, and the Tribunal proceeded on this basis’.Footnote 67
This overview shows that investment tribunals tend to justify their reliance on ARSIWA by using the terminology of ‘formal’ sources of international law. However, although the vocabulary of ‘formal’ sources is used or alluded to, it can be concluded that investment tribunals do not seem to treat ARSIWA as a monolith in the process of identification of applicable rules. In most cases, tribunals accord ARSIWA decisive value in the process of identification of rules of customary international law on State responsibility. Other tribunals evoke the terminology of general principles of law or are ambiguous in this respect. What is more, tribunals usually avoid wide-ranging conclusions and focus on whether a specific provision of ARSIWA reflects an applicable rule. That said, a noteworthy insight gained from the empirical analysis is that these variations are not limited to certain parts or provisions of ARSIWA. Rather, they exist with respect to a variety of issues such as the rules on attribution of conduct, the breach of an international obligation, the circumstances precluding wrongfulness, or the provisions relating to reparation and implementation of responsibility.
4 Patterns of Interpretation in the Use of ARSIWA in International Investment Arbitration
4.1 Content-Determination Pursuant to the ‘Text’ of ARSIWA
Whether they provide a justification or not, tribunals rely on ARSIWA primarily as means to determine the content of applicable rules on State responsibility. In the case of treaties, the determination whether a text or statement has the formal hallmark of a treaty entailing binding obligations, on the one hand, and the determination of the meaning of a binding treaty provision, on the other, clearly involves different considerations so much so that it is possible to speak of two distinct juridical operations governed by different rules.Footnote 68 However, the ARSIWA are not a treaty. As we have seen, tribunals justify their applicability in investment arbitration as expressions of unwritten international law, that is, customary international law or, less often, general principles of law. In this respect, it has been maintained in theory that it is impossible to identify a rule of unwritten international law without, at the same time, determining its content.Footnote 69 Conversely, rules of unwritten law are not amenable to interpretation, this operation presupposing the existence of a text.Footnote 70 As a corollary, the determination of the content of a rule depends on the very same means as the identification of a rule and requires the establishment of State practice and opinio juris or of recognition and transposability, as the case may be.Footnote 71 This section (and the subsections that comprise it) shows that these theoretical considerations can explain the practice of investment tribunals relating to ARSIWA only partially. Although decisions differ to a large extent as to how they use ARSIWA in the determination of the content of applicable rules, it is possible to discern certain trends. One discernible trend that is not easily reconcilable with these considerations is investment tribunal’s emphasis on a ‘textualist’ approach when using ARSIWA to determine the content of applicable rules on State responsibility.
To start with, the role of interpretation in determining the content of applicable rules of State responsibility is not usually pronounced and intertwines with the ways in which tribunals use ARSIWA in this process. According to the classification laid down in Art 38(1) ICJ Statute, ILC works constitute ‘subsidiary means for the determination of rules of law’.Footnote 72 Albeit investment tribunals never refer to this categorisation explicitly (but for one singular exception),Footnote 73 some of them effectively use ARSIWA as such by merely citing them to support a determination that certain normative proposition found in judicial pronouncements or other sources reflects a rule of international law.Footnote 74 These tribunals seem to treat ARSIWA indistinctly as secondary evidence in the determination of unwritten international law without any methodological explanation as to the steps followed in this process.
However, in many cases, the ways in which tribunals engage with ARSIWA in the process of determination of applicable rules goes beyond indirect reliance. Most notably, there is an abundance of decisions in which tribunals proceed to apply provisions of ARSIWA as self-explanatory to the facts of the case. Thus, for instance, many tribunals quote Article 4 ARSIWA as a representation of applicable law and continue to determine whether a person or an entity is a State organ or not relying on the characterisation provided in the domestic law of the relevant State.Footnote 75 Similarly, tribunals have applied without any discussion a variety of provisions of ARSIWA including those on attribution of conduct,Footnote 76 on the time of the breach,Footnote 77 on circumstances precluding wrongfulness,Footnote 78 and on reparation.Footnote 79 In the same vein, tribunals often identify in a provision of ARSIWA an applicable rule of law and then refer to judicial pronouncements as means to determine the meaning of that provision. To illustrate this point, in the Jan de Nul award on the merits, the Tribunal considered ILC’s provision that a given conduct is considered an act of State ‘if the person or group of persons is in fact acting on the instructions of, or under the direction or control of, that State in carrying out the conduct’.Footnote 80 After characterising ARSIWA ‘a statement of customary international law’, it proceeded to further clarify the meaning of the provision.Footnote 81 The Tribunal held that ‘[i]nternational jurisprudence is very demanding in order to attribute the act of a person or entity to a State, as it requires both a general control of the State over the person or entity and a specific control of the State over the act the attribution of which is at stake; this is known as the “effective control” test’, citing in support the findings of the ICJ in Nicaragua.Footnote 82 Subsequent awards reproduce the Jan de Nul formula more or less verbatim,Footnote 83 even though the ‘general control’ prong of the test does not feature explicitly in the text of Article 8 ARSIWA, its Commentary or the pronouncements of the ICJ after Nicaragua.Footnote 84
In these two groups of cases, it is difficult to discern what precise juridical operation is at play, but two alternatives are conceivable from an analytical perspective. On the one hand, the lack of any separate analysis on the content of the applicable rule is suggestive of the absence of an intermediate step between identification of a rule of customary international law or general principle of law and its application.Footnote 85 Similarly, the reliance on judicial pronouncements can be construed as an extension of the determination of State practice/opinio juris or recognition/transposability, as the case may be, albeit implicitly and on the basis of secondary evidence.Footnote 86 After all, judicial decisions, much like ARSIWA, constitute ‘subsidiary means’ for the determination of applicable rules.Footnote 87 On the other hand, these tribunals not even purport to determine the content of the applicable rule on State responsibility through an independent analysis of State practice and opinio juris or a comparative survey. Rather, they proceed to apply the formulations of the ILC to the facts of the case as if they were a binding text.
In this respect, the conciseness of analysis can also be construed as an emanation of a textual approach towards ARSIWA in a way that parallels known approaches of treaty interpretation. In other words, the tribunals’ line of reasoning consists conceivably of the application of the terms of a provision whose source of legal validity (CIL or general principle of law) has already been determined, because they deem its ordinary meaning sufficiently clear.Footnote 88 To illustrate this point, in Tulip, the Tribunal examined whether the termination of a contract by a company owned predominantly by a State agency was attributable to Turkey. The Tribunal accepted that ‘the ILC Articles constitute a codification of customary international law with respect to the issue of attribution of conduct to the State’.Footnote 89 Turning to Article 8 ARSIWA, the Tribunal focused on its text and decided that ‘[p]lainly, the words “instructions”, “direction” and “control” are to be read disjunctively’.Footnote 90 The fact that a State agency owned the majority share of the company in question entailed that the company was under the control of the Turkish State in the sense that Turkey was capable of exercising a degree of control to implement governmental policies.Footnote 91 Nonetheless, the Tribunal held that such control was insufficient for attribution, because, according to the ILC’s Commentary, the State must ‘us[e] its ownership interest in or control of a corporation in order to achieve a particular result’.Footnote 92 In the subsequent annulment decision in Tulip, the committee clarified that ‘the tribunal, in interpreting Article 8, took into account the ILC Commentary’ and upheld the analysis of the Tribunal finding that ‘[it] correctly interpreted Article 8’.Footnote 93 Such textual analysis of ARSIWA with reference to the ILC Commentary as an authoritative interpretative aid is most common in the practice of investment tribunals, even if they are not always as explicit as the Tulip committee when justifying their methodological choices.Footnote 94 Therefore, there is evidence to suggest that such ‘textualist’ approach constitutes essentially an interpretative operation.
Along similar lines, it is possible to argue that tribunals frame their reliance on previous case law in terms of interpretation because such previous pronouncements only purported to interpret rules whose existence was undisputed. So, for instance in El Paso, Argentina argued that the Tribunal exceeded its powers by relying on case law to determine the applicable standard of reparation under the law on State responsibility, despite judicial decisions’ lack of binding status beyond the confines of a specific case.Footnote 95 The annulment committee dismissed this claim on the basis that ‘[a]rbitral tribunals must resort to different methods of interpretation to decide the dispute’ before them and, in the event, the tribunal relied on previous case law only ‘to be helped in its interpretation’.Footnote 96 More generally, tribunals seem to single out a provision of ARSIWA as the embodiment of an unwritten rule, whilst the reference to international jurisprudence only comes after this determination without justification, as if it were only an interpretative aid.
4.2 Content-Determination through the Application of Means Akin to Treaty Interpretation
Apart from such textual approach, it is not uncommon for tribunals to invoke or evoke other interpretative principles commonly used in the process of treaty interpretation. First, a few tribunals have recourse to materials produced in the long preparatory process of ARSIWA as another means to determine the content of the applicable rules on State responsibility. Second, investment tribunals often refer to means of interpretation applied to treaties focusing on the context or object and purpose of ARSIWA.
References to the discussions within the ILC leading up to the adoption of ARSIWA are not particularly frequent in investment decisions. In some cases, tribunals rely on the record of discussions in order to determine whether the silence of ARSIWA also implies a determination by the ILC that certain concept or proposition does not form part of international law.Footnote 97 Thus, in Alghanim, the investor invoked the distinction between ‘obligations of conduct’ (ie those that prescribe or proscribe a specific conduct) and ‘obligations of result’ (ie those that require the achievement of a specific result irrespective of the conduct adopted) that appeared in previous drafts of ARSIWA.Footnote 98 The Tribunal took note of the critical stance of the last Special Rapporteur and the deletion of the distinction from the final draft of ARSIWA and concluded that the distinction did not form part of customary international law.Footnote 99 Furthermore, in other cases, tribunals purport to rely on the preparatory works of ARSIWA as a means to interpret a provision of ARSIWA. For instance, in the LG&E decision, the Tribunal started from the determination that Article 25 ARSIWA reflects the standard of necessity in international law.Footnote 100 It, then, proceeded to discuss each of the elements of Article 25 ARSIWA referring exclusively to the opinions of the ILC Special Rapporteurs and other individual members of the ILC.Footnote 101 From a methodological perspective, it is possible to maintain that these tribunals merely examine all available secondary evidence without endorsing any firm distinction between identification and determination of the content of the applicable rule.Footnote 102 However, in cases like the LG&E, the analysis clearly emphasises the intention of the ILC in adopting certain provision rather than the evidence upon which the ILC relied. For this reason, these findings are not easily explicable under the mainstream view on the identification of customary international law or general principles of law.Footnote 103 Rather, these tribunals seem to resort to the record of discussions of the ILC more as an interpretative aid in a way that parallels the use of travaux préparatoires in the context of treaty interpretation.
More conspicuously, investment tribunals often refer to interpretative principles which pertain to the ‘the spirit, purpose and context of the clause or instrument in which the words are contained’.Footnote 104 The Tribunal’s approach in Devas relating to the attribution of conduct of a State-owned company to India provides a very illustrative example. In this case, the Tribunal noted that the text of Article 8 ARSIWA only mentioned ‘persons or group of persons’, but made no reference to ‘entities’ like, for instance, Article 5 ARSIWA establishing also a rule of attribution of conduct.Footnote 105 Nonetheless, the Tribunal considered that ‘it is generally recognized in modern legal systems that “person” includes not only a natural person but also a legal person’ and that several Institutes of Internal Auditors (IIAs) included corporations in their definition of ‘persons’.Footnote 106 In methodological terms, the Tribunal referred to other rules of international law, which it deemed relevant for the interpretation of the rule reflected in ARSIWA, in a way akin to the context of a treaty.Footnote 107 Furthermore, the Tribunal remarked that ‘it would make no sense to impose a restrictive interpretation that would allow a State to circumvent the rules of attribution by sending its direction or instruction to a corporate entity rather than a physical person or group of physical persons’.Footnote 108 Instead, it opted for a different interpretation noting that even in the case of corporations the instructions or direction would be received and acted upon by natural persons (ie the directors and agents of the corporation).Footnote 109 From a doctrinal viewpoint, the Tribunal chose out of two available interpretations the one that gave full effect to Article 8 ARSIWA in what appears to be a straightforward application of the interpretative principle of effectiveness (ut res magis valeat quam pereat or effet utile).Footnote 110
Moreover, several tribunals often proceed to construct provisions of ARSIWA on the basis of broader considerations, which they deem as cross-cutting in the law of State responsibility. For instance, several tribunals invoke the stability of international obligations as a stepping stone for a restrictive interpretation of the customary defence of necessity.Footnote 111 Another set of illustrative decisions declare that the purpose of an award of interest under Article 38 ARSIWA is to ‘ensure full reparation’ and proceed to award compound interest.Footnote 112 This is so notwithstanding the fact that Article 38 ARSIWA is silent on the matter and the ILC Commentary clearly favours the award of simple interest.Footnote 113 These findings seem to evoke the object and purpose or the ratio of ARSIWA or of specific provisions in order to determine the meaning of the applicable rule in a way that parallels known approaches to treaty interpretation.Footnote 114 The common thread that binds all these pronouncements together is the blending of a literal reading of ARSIWA with contextual or teleological considerations that mirrors the process dictated by the rules of treaty interpretation.
4.3 Interpretation of the Law on State Responsibility and Its ‘Rule-ness’ in International Investment Arbitration
It has been shown so far that it is not uncommon for investment tribunals to hold explicitly or implicitly, through the use of relevant methods, that the determination of the content of the applicable rules of State responsibility also requires a process of interpretation notwithstanding its unwritten character.Footnote 115 A further question that arises is whether the practice of investment tribunals evidences the existence of rules or principles governing this process.
Remarkably, some decisions not only accept that the determination of the content of rules of State responsibility also involves interpretation, but they also clearly suggest that such process is governed by relevant rules or principles of international law. The award on jurisdiction in ST-AD is telling as to this general point. In this case, the Tribunal enunciated that ‘every rule … of international law must be interpreted in good faith’.Footnote 116 Applying this rule of interpretation to the requirement of exhaustion of local remedies under customary international law,Footnote 117 the Tribunal found that ‘[t]his rule is interpreted to mean that applicants are only required to exhaust domestic remedies that are available and effective’.Footnote 118 Similarly, several decisions relating to the customary defence of necessity, which is articulated in Article 25 ARSIWA, raise explicitly the issue of interpretation.Footnote 119 Most notably, in Enron, the annulment committee censured parts of the award discussing whether measures taken by Argentina were the only way to address the economic crisis constituting the situation of necessity and whether Argentina contributed to that crisis.Footnote 120 According to the committee, the ‘only way’ and ‘non-contribution’ requirements spelled out in Articles 25(1)(a) and 25(2)(b) ARSIWA, respectively, were ‘capable of more than one interpretation’.Footnote 121 The committee held that the Tribunal ‘was necessarily required, either expressly or sub silentio, to decide or assume the correct interpretation in order to apply the provision to the facts of the case’.Footnote 122 It, thus, concluded that the Tribunal committed an annullable error by not laying down its own interpretation of these requirements.Footnote 123 Inversely, in EDF, the annulment committee admitted as a matter of principle that the ‘the concept of “only means” is open to more than one interpretation’.Footnote 124 It held that ‘[i]n the light of the principle that necessity is an exceptional plea which must be strictly applied (a principle expressly stated in paragraph 1171 of the Award), … “only” means “only”; it is not enough if another lawful means is more expensive or less convenient’.Footnote 125 Although the committee held that failure to elaborate on the issue of interpretation did not constitute an annullable error, it nonetheless recognised the application of a principle or rule of interpretation to the customary rule of necessity according to which ‘exceptions to general principles are to be interpreted restrictively’.Footnote 126
By contrast, some decisions eschew a precise determination as to the content of the applicable rule of State responsibility by limiting themselves to taking note of the parties’ stance in the proceedings. In some cases, the lack of an independent analysis on the content of the applicable rule seems to be elicited by factual or evidentiary considerations. So, in Inmaris, the Tribunal identified the applicable rule in Article 31(1) ARSIWA that provides for ‘the obligation to make full reparation for the injury caused by the internationally wrongful act’.Footnote 127 As to the issue of causation, Ukraine proposed two tests, namely, one based on proximity and another based on foreseeability.Footnote 128 The Tribunal declined to determine the applicable test finding that the relevant acts caused harm to the investor ‘under either standard discussed by the respondent’.Footnote 129 However, in other decisions, the agnostic stance towards interpretation appears to stem from legal considerations. To give an illustrative example, in Suez annulment decision, the crucial issue was whether the Tribunal failed to apply the proper law, in the event, Article 25 ARSIWA on the state of necessity. The committee conceded as a matter of principle that the ‘only way’ and ‘non-contribution’ requirements appearing in Article 25 ARSIWA ‘are indeed susceptible to a certain degree of interpretation’.Footnote 130 However, it emphasised that no party raised any issue of interpretation in the proceedings before the Tribunal. On this basis, the committee concluded that ‘an interpretation issue that was not raised by the Parties cannot be considered “outcome-determinative” with the consequence that a failure to address such issue would amount to a manifest excess of powers under Article 52(1)(b)’.Footnote 131
The committee’s reasoning in Suez echoes other annulment decisions that distinguish between ‘disregarding the proper law’, which constitutes an annullable error, from ‘misapplication of the proper law’, which does not.Footnote 132 In the case of applicable treaty provisions, annulment committees also occasionally examine whether tribunals disregarded any applicable rules of interpretation, despite allowing them ample deference as to the application of such rules in the specific case.Footnote 133 The Suez case raises the question whether and how this distinction can be applied with respect to applicable rules of unwritten international law, such as those under the law of State responsibility. In this respect, as has been shown, the Enron annulment decision clearly suggests that a tribunal must pay some regard, either explicitly or sub silentio, to the principles upon which it bases its determination of the content of the applicable rules of unwritten international law.Footnote 134 What is more, the EDF committee traced back the interpretative principle applied by the Tribunal despite finding that this was beyond the scope of annulment review.Footnote 135 By contrast, whilst acknowledging a distinction between identification of applicable rules of unwritten law and their interpretation, the Suez committee relied exclusively on the parties’ stance in the underlying proceedings to decide whether an issue of interpretation was ‘outcome-determinative’. In this way, it remained entirely agnostic as to the existence of rules or principles of interpretation of unwritten international law.
To summarise, investment tribunals employ an array of methods in the course of determining applicable rules on unwritten international law on State responsibility. At the crux of these diverse approaches seems to lay a latent, or sometimes explicit, distinction at an analytical level between the identification of applicable rules on State responsibility and their interpretation with ARSIWA operating as the focal point. This distinction is not merely a theoretical one, but it has both substantive and procedural implications. First, the use of ARSIWA as an object of interpretation has allowed investment tribunals to develop the law on State responsibility in ways that contradict, or cannot be inferred from, the evidence adduced by the ILC. The award of compound interest in investment arbitration is a case in point. Second, a comparative survey between investment awards reveals that new divisions have spun out as to the choice of the proper means of interpretation. The cases on the content of the necessity defence are paradigmatic. Third, the casting of disputes on the content of the law on State responsibility in terms of interpretation by reference to ARSIWA has also played a role in the context of annulment proceedings. In this respect, the division between annulment committees is not so much whether customary international law or general principles of law on State responsibility can be interpreted; annulment committees readily accept this point. Rather, the main point of contention seems to be the existence and interplay of applicable interpretative rules or principles so as to enable annulment review, delineate its scope, and determine its operation.
5 Interpretation as a Balancing Exercise between Centrifugal Forces in International Investment Arbitration and the Unity of Law on State Responsibility
The previous section has identified a variety of ways in which investment tribunals use ARSIWA in determining applicable rules on State responsibility. In the main, investment tribunals favour a formalist approach: they tend to justify their reliance on ARSIWA as means to determine the applicable rules on State responsibility on the basis of the rules on the identification of international law. Moreover, many investment tribunals go even further and proceed to apply, explicitly or implicitly, (meta-)rules or (meta-)principles of interpretation on ARSIWA in order to determine the content of the applicable rule. At the same time, some decisions remain agnostic with regards to the issue of interpretation. This section argues that the fundamental consideration of the unity of the law on State responsibility militates in favour of its uniform application and against a case-by-case approach. The theory of sources of international law provides several footholds for the interpretation of rules of unwritten international law.
To start, the fundamental premise of the law on State responsibility is its unity: ‘every internationally wrongful act of a state entails the international responsibility of that state’.Footnote 136 Yet, an agnostic approach as to the methodology for the determination of the content of the applicable rule can severely distort its content. To illustrate this point, in Bayindir, the Tribunal referred to ‘the international rules of attribution reflected in Articles 4, 5 and 8 [ARSIWA]’, which it found applicable ‘as expressing current customary international law’.Footnote 137 In its analysis on Article 8 ARSIWA, the Tribunal held:
the Tribunal is aware that the levels of control required for a finding of attribution under Article 8 in other factual contexts, such as foreign armed intervention or international criminal responsibility, may be different. It believes, however, that the approach developed in such areas of international law is not always adapted to the realities of international economic law and that they should not prevent a finding of attribution if the specific facts of an investment dispute so warrant.Footnote 138
In the specific case of Article 8 ARSIWA, the ICJ has ruled that the ‘effective control’ test cannot be displaced in the absence of an applicable lex specialis.Footnote 139 Yet, at first glance, the Bayindir panel was not obligated to follow the ICJ’s ruling and an inductive examination of the available primary and secondary evidence at the time would have led to conflicting results.Footnote 140 What is striking in this award is that the Tribunal put aside the much debated ‘effective control’ test purporting that it is a factual issue pertaining to the application of the customary rule. This enabled the Tribunal to infuse its analysis of the content of the applicable rule of responsibility with asides about the speciality of international economic law in a way that runs counter to the fundamental consideration of the unity of the law on State responsibility.Footnote 141 The Bayindir award stands out as an outlier, investment tribunals accepting this premise virtually unanimously. Accepting the premise of the unity of the law on State responsibility, which the ICJ has upheld as a rule of customary international law, militates against a case-by-case approach and reveals the need for a consistent methodology for the determination of the content of rules on State responsibility.Footnote 142
Consistently with the mainstream approach concerning the sources of international law, the determination of the applicable rules of the law on State responsibility must take place in principle on the basis of an inductive examination of all available evidence.Footnote 143 However, at the same time, a purely inductive approach towards the identification of unwritten international law has certain limitations. So, with respect to the identification of customary international law, the ILC has concluded that ‘the two-elements approach does not preclude an element of deduction as an aid’ particularly ‘when considering possible rules of customary international law that operate against the backdrop of rules framed in more general terms that themselves derive from and reflect a general practice accepted as law or when concluding that possible rules of international law form part of an “indivisible regime”’.Footnote 144 More emphatically, the current special rapporteur on general principles of law has opined that ‘deduction is …the main criterion to establish the existence of a legal principle that has a general scope’.Footnote 145 The fact that the ILC perceives deduction as part and parcel of the process of identification of a rule of unwritten international law is somewhat less relevant. The key point is the acknowledgment of a juridical operation analytically distinct from induction that consists of the determination of the content of the applicable rule on the basis of an inference from a normative proposition whose status is undisputed.
To illustrate this point, according to the ILC, the provisions of ARSIWA relating to the content of State responsibility 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.’Footnote 146 Nonetheless, investment tribunals commonly apply the rules on reparation reflected in ARSIWA, even if this involves an element of interpretation as an intermediate step.Footnote 147 Notably, in Quiborax, the issue arose whether investment tribunals can issue a declaratory award as a form of reparation. The Tribunal referred to Articles 34 and 37 ARSIWA and enunciated that ARSIWA ‘restate customary international law and its rules on reparation have served as guidance to many tribunals in investor-State disputes’.Footnote 148 It specified that ‘the remedies outlined by the ILC Articles may apply in investor-State arbitration depending on the nature of the remedy and of the injury which it is meant to repair.’Footnote 149 In this respect, it cautioned that ‘some types of satisfaction as a remedy are not transposable to investor-State disputes’.Footnote 150 In particular, it held that ‘the type of satisfaction which is meant to redress harm caused to the dignity, hono[u]r and prestige of a State, is not applicable in investor-State disputes.’Footnote 151 The Tribunal concluded, thus, that ‘[t]he fact that some types of satisfaction are not available does not mean that the Tribunal cannot make a declaratory judgment as a means of satisfaction under Article 37 [ARSIWA], if appropriate’.Footnote 152 In practical terms, the tribunal essentially engaged in the interpretation of Article 37 ARSIWA as a rule of customary international law referring expressly to its wording and its object and purpose. Whether it did so in order to determine its transposability in the context of investor-State arbitration or whether it interpreted it as a directly applicable rule of customary international law is less relevant for practical purposes.
The important point is that the determination of the applicable rules of State responsibility is composed of two analytically distinct operations with ARSIWA being at the crux of the analysis. On the one hand, the determination of the status of a normative proposition contained in ARSIWA involves an inductive analysis of evidence.Footnote 153 In this respect, the practice of investment tribunals is an unambiguous attestation that ARSIWA enjoy a ‘présomption de positivité’.Footnote 154 In practical terms, challenging the status of most normative propositions contained in ARSIWA would require evidence indicating lack of generality of practice/opinio juris or recognition, as the case might be. On the other hand, the determination of the content of a normative proposition contained in ARSIWA also involves an element of interpretation. Specifically, it involves criteria that parallel the process of treaty interpretation. The key difference is that the intention of the drafters of ARSIWA – that is, the ILC – is less relevant in this process because the formal foundation of the normative propositions contained therein is only the consuetudo of States.Footnote 155 In practical terms, this is reflected in the findings of investment tribunals emphasising the text, context, and object and purpose of ARSIWA or a specific provision at the expense of the multitude of materials leading up to their adoption or even, in case of inconsistency, the ILC Commentary.Footnote 156 Therefore, the process of interpretation is not an unprincipled process, even if the relevant principles and their interaction are still in a process of elaboration and refinement (not unlike the rules of treaty interpretation before the adoption of the Vienna Convention on the Law of Treaties (VCLT)).
To recap, the premise of the law on State responsibility is inimical to an agnostic approach with respect to the identification and interpretation of the rules on State responsibility. The crucial normative concern is the unmaking of the law on State responsibility through its blending with considerations special to specific sub-systems, in this case, international investment law. In this respect, adherence to ARSIWA is only a starting point. Rather, the key point is that the determination of the applicable rules on State responsibility on the basis of ARSIWA is also a principled process to which interpretation is an inextricable part.
6 Conclusion
Undeniably, investment tribunals spearhead the consolidation of the general law of State responsibility through their widespread endorsement of ARSIWA. This comparative analysis of the use of ARSIWA in investment arbitration has shown that the discrepancy between the current form of ARSIWA and their effective authority should not be overstated. In the main, investment tribunals do justify their reliance on ARSIWA using the vocabulary of the sources of international law. In fact, the formal status of most normative propositions contained in ARSIWA has been tested and analysed from every possible angle by hundreds of investment panels. Whilst admittedly this has led to divergent views as to theoretical points, the preponderance of decisions converges into treating ARSIWA as authoritative statements on the law on State responsibility.
What emerges as a new challenge is ensuring the uniformity of the law on State responsibility within this context of pervasive use. The focus of States, investors, and tribunals seems to have moved away from ‘grand questions of principle’ towards ‘the boring small print’ of responsibility rules like the meaning of control, contribution, injury, causation, or damage.Footnote 157 In this respect, the traditional approaches on the identification of customary international law and general principles of law have certain limitations. This survey has shown that investment tribunals are increasingly aware that, through their use and elaboration of ARSIWA, they are engaging in the interpretation of the formally unwritten law on State responsibility. The realisation that interpretation is also a principled process guided by international law will ensure the unity and consistent development of the law on State responsibility even within the specialised context of investment arbitration.
In fact, this survey has shown that elements of such a principled methodology are already present in the practice of investment tribunals, even if in incipient form. In the first place, the determination that a normative proposition contained in ARSIWA reflects a rule of customary international law or a general principle of law is only a starting point. This determination entails that the ordinary meaning of the terms of ARSIWA is important in specifying the content of the formally unwritten rule. In this respect, tribunals seem mindful that this is essentially an interpretative endeavour guided by considerations of good faith. Second, another overarching consideration is that rules on State responsibility do not exist in a vacuum but form part of an internally coherent law of State responsibility that should be deemed compatible with the rules and principles of international law which this law is meant to make operational. As a corollary, whilst tribunals tend to confirm the status of a normative proposition contained in ARSIWA in a piecemeal fashion (ie for each provision separately), this does not constitute the end of the analysis. Tribunals also take into account the immediate and broader context of each provision alongside the object and purpose of the entirety of ARSIWA and of each provision within the broader system of ARSIWA when clarifying its content. Third, tribunals seem also to appreciate that the anticipated outcome of this innately interpretative exercise is not the determination of the intention of the drafters of ARSIWA, but the concretisation of rules and principles emanating from the legally relevant conduct of States. Whilst they seem to accord particular value to an ILC’s determination denying binding status to a certain normative proposition, they use materials produced in the run-up to the adoption of ARSIWA only exceptionally and in a supplementary fashion in the context of the interpretation of a normative proposition contained in the ARSIWA.
1 Introduction
The Articles on the Responsibility of States for Internationally Wrongful Acts (ARSIWA) were intended as the International Law Commission’s (ILC) ‘permanent contribution to general international law’.Footnote 1 As such, the provisions enshrined therein, including the ‘circumstances precluding wrongfulness’ under Part I, Chapter 5, are in principle binding on, and applicable to, all StatesFootnote 2 (generality ratione personae) and apply to the whole field of international obligations of States,Footnote 3 regardless of their content or sourceFootnote 4 (generality ratione materiae). This chapter discusses whether, despite this prima facie general applicability, a responding State may be precluded from invoking the customary defence of countermeasures in the context of an investment dispute. It identifies and assesses four factors potentially affecting the availability of countermeasures in investment arbitration: the jurisdictional constraints of the arbitral tribunal, limitations in the law applicable to the dispute, the interpretation of the investment protection treaty in question, and potential limitations to the scope of application of the defence under customary international law.
The jurisdiction of an arbitral tribunal and the law applicable to the dispute are defined by the claim and the treaty in question. The tribunal’s jurisdiction extends only to rulings on the matters raised by the claimant (non ultra petita), based on interpretation and application of the treaty at hand.Footnote 5 Nonetheless, many investment treaties authorise the tribunal to decide the issues in dispute, not only on the basis of the treaty itself, but also of ‘applicable rules of international law’.Footnote 6
The first task for the tribunal is to determine whether countermeasures are such ‘applicable rules’. This is an interpretative task. The tribunal through interpretation of the investment treaty in question will need to ascertain whether countermeasures as a defence are, explicitly or implicitly, displaced in the context of disputes arising thereunder. For example, the treaty in question may contain rules that constitute lex specialis to the defence of countermeasures under general international law. In such cases, the special provisions prevail over the general, to the extent that they regulate the same subject matter in a different manner.Footnote 7 Moreover, it has been argued that the conferral of substantive, direct rights to investors through investment protection treaties implies the non-applicability of the defence of countermeasures. If countermeasures are taken in response to a prior internationally wrongful act of the investor’s home State, then the defence can only preclude the wrongfulness of a temporary non-performance of obligations owed to such State and not of obligations owed directly to the investor, which would, essentially, be a third party to such dispute. In this context, the tribunal shall also take into consideration any relevant rules of international law applicable between the parties,Footnote 8 such as the rules under customary international law on the protection of aliens and the institution of diplomatic protection, which may inform the content of relevant treaty provisions.
If the defence is, following this interpretative exercise, found to be applicable to an investment dispute, then the arbitral tribunal is further faced with two jurisdictional considerations. Firstly, the application of countermeasures may require examination of rights and obligations that fall, prima facie, outside its subject matter jurisdiction as established in that dispute, or even outside its jurisdictional field in general.Footnote 9 The tribunal would have to determine whether a prior internationally wrongful act has been committed, which would probably constitute a breach of an obligation outside the investment treaty under consideration, and thus, outside the jurisdictional limits of the arbitral tribunal. Secondly, the prior internationally wrongful act would be committed by the State of nationality of the investor and not the investor itself. This raises a Monetary Gold-likeFootnote 10 consideration regarding the personal jurisdiction of the arbitral tribunal, as the author of the act would not be a party to the proceedings.
This chapter addresses the considerations above in two substantive sections. Section 2 provides an overview of the tribunals’ reasoning and the parties’ arguments in the relevant investment case law. All tribunals that have been seized of this matter rejected the arguments of the respondent on the defence of countermeasures but diverged significantly in terms of reasoning. Thus, arbitral tribunals so far have not provided a definite and consistent answer to the applicability of the defence. Section 3 critically assesses the approach of the tribunals to the application of the lex specialis principle, to the nature of investors’ rights under investment protection treaties and to the limits of their jurisdiction ratione materiae and personae. This chapter argues that the defence of countermeasures under customary international law is indeed applicable in the context of investment disputes and that arbitral tribunals have the power to examine all the requirements of the defence in order to rule on its applicability in a given case.
2 Countermeasures in Investment Case Law: The US–Mexico Sugar War and the Ambivalent Arbitral Practice
The defence of countermeasures in the context of investment arbitration was primarily discussed in three North American Free Trade Agreement (NAFTA)-basedFootnote 11 disputes: Archer Daniels Midland Company and Tate & Lyle Ingredients Americas, Inc v Mexico (ADM), Corn Products International, Inc v Mexico (CPI) and Cargill, Inc v Mexico (Cargill).Footnote 12 All three disputes arose in the context of the same broader situation of tension between the US and Mexico, which involved proceedings not only before these arbitral tribunals, discussed in this chapter, but also before the WTO Dispute Settlement SystemFootnote 13 and NAFTA.Footnote 14 The disagreement started with the US imposing measures, which restricted the access of Mexico’s surplus sugar produce to the US market. It further escalated by the subsequent failure of the US to take part in the NAFTA dispute settlement proceedings initiated by Mexico with respect to the market restrictions, by blocking the appointment of panelists contrary to NAFTA Chapter Twenty. Mexico proceeded with a series of measures aiming to protect its domestic sugar industry, including a 20% tax on soft drinks using sweeteners which were primarily used by US companies. In response to the US claims that the tax was in breach of its NAFTA obligations, Mexico invoked the defence of countermeasures under general international law, claiming that the tax was lawfully introduced in response to prior violations by the US of its NAFTA obligations regarding the access of Mexican-produced sugar to the US market and dispute settlement proceedings. Three (redacted) awards were issued on this matter by tribunals constituted under the NAFTA dispute settlement provisions.
Even though all three disputes were based on the same factual matrix and the same legal instrument, the three awards have significant differences in terms of reasoning. The proceedings for all three cases took place largely in parallel. The Tribunals in ADM and CPI were unaware of each other’s findings and there are no cross-references between the two awards or engagement with each other’s reasoning. The Tribunal in Cargill received the ADM award during its deliberations along with comments from both sides.Footnote 15 Thus, the Cargill Tribunal had the opportunity to take into consideration the reasoning of the Tribunal in ADM and engage with its findings on the defence of countermeasures.Footnote 16
All disputes were submitted to arbitration pursuant to NAFTA Chapter Eleven,Footnote 17 for alleged violations of, inter alia, the national treatment and expropriation provisions of NAFTA by Mexico.Footnote 18 The kind of claims that may be submitted to investor-State arbitration under Article 1120 NAFTA are specified in Articles 1116 and 1117, which accordingly, along with the application of the claimant, establish the scope of jurisdiction of the arbitral tribunals.Footnote 19 Moreover, the tribunals were constituted under the auspices of the International Centre for Settlement of Investment Disputes (ICSID). Under Article 42(1) ICSID Convention, a tribunal should ‘decide a dispute in accordance with such rules of law as may be agreed by the parties.’Footnote 20 In turn, Article 1131(1) NAFTA provides that ‘[a] Tribunal established under this Section shall decide the issues in dispute in accordance with this Agreement and applicable rules of international law.’ Thus, the parties to NAFTA have agreed that general international law is, in principle, within the law that a NAFTA tribunal may apply in the determination of a dispute before it. The provisions on jurisdiction and applicable law were relevant to the discussion of the tribunals regarding the availability of countermeasures to investment arbitration and their power to examine the customary requirements of the defence.
The present section discusses the relevant analysis of the tribunals in the cases of ADM, CPI and Cargill. It outlines the tribunals’ approach to the application of the lex specialis principle, the customary requirements for successful invocation of countermeasures, the nature of investors’ rights under investment treaties and the limits of their jurisdiction ratione materiae and ratione personae.
2.1 Countermeasures under General International Law and the lex specialis Principle
An arbitral tribunal would, first of all, have to decide whether the customary defence of countermeasures is an ‘applicable rule … of international law’ within the meaning of Article 1131(1) NAFTA. Even if NAFTA empowers, in principle, arbitral tribunals to apply rules of international law that are not explicitly included in the text of NAFTA itself, tribunals must still investigate whether NAFTA contains any rules that constitute lex specialis, and thus, apply to the exclusion of the relevant rules under general international law. The challenge is, first, to identify whether there is indeed a relationship of lex specialis/lex generalis between the two rules and, second, to determine the extent to which the two rules are co-extensive and mutually exclusive. According to the ILC, for the lex specialis to displace the relevant lex generalis, it is not enough that the same subject matter is covered by the two rules, but there must be an actual inconsistency between the two rules – a genuine conflict of norms.Footnote 21 The general rule is displaced only to the extent of the inconsistency with the treaty specific rule, whilst other aspects continue to operate in a residual fashion.Footnote 22 The tribunal in ADM was the only one of the three that discussed the NAFTA applicable law clause, explored the interaction of the defence of countermeasures under general international law with the NAFTA provisions and examined the application of the lex specialis principle in order to discern whether countermeasures can be invoked in the context of a NAFTA Chapter Eleven dispute.
The Tribunal began its analysis by reference to Article 1131(1) NAFTA. Although this provision pertains to the law applicable to the dispute, the Tribunal examined its content to determine whether it has ‘jurisdiction to decide on the validity of the defense’.Footnote 23 This shows the close relationship between the jurisdiction of an adjudicative body and the law that it may apply in the exercise of such jurisdiction. The Tribunal confirmed that Article 1131(1) NAFTA ‘includes the application of rules of customary international law with respect to claimed breaches [of NAFTA provisions]’.Footnote 24 This finding confirms that the Tribunal has, indeed, the power to examine and apply customary international law, including the defences under the general international law on State responsibility, but only in the context of deciding the specific claims of breaches that are properly brought before it on the basis of the relevant jurisdictional clause.
Having established that customary defences are prima facie applicable to the dispute at hand in accordance with the NAFTA applicable law clause, the Tribunal proceeded to examine whether NAFTA otherwise excludes the application of countermeasures. The claimants have argued that NAFTA includes certain provisions that constitute lex specialis to the customary defence of countermeasures, which is, therefore, excluded in accordance with Article 55 ARSIWA.
According to the claimants, NAFTA
Chapters Nineteen and Twenty establish the regime for dispute resolution that governs the ‘existence of an internationally wrongful act’ and the ‘content’ of the international responsibility of the Parties in the event of a breach of their obligations under the NAFTA … In other words, by signing the NAFTA, the Parties have deliberately forgone the residual right to take countermeasures under customary law.Footnote 25
In response to this argument, the Tribunal acknowledged that the NAFTA indeed offers a form of lex specialis to supplement the standards of customary international law on a number of issues such as the treatment of aliens and diplomatic protection.Footnote 26 This ‘express content’Footnote 27 of the NAFTA, which clearly deviates from – or rather, goes beyond – the relevant rules of customary international law, constitutes lex specialis and is applicable to the exclusion of the relevant customary rules. However, according to the Tribunal, customary international law ‘continues to govern all matters not covered’ by the NAFTA provisions. To this end, it found that ‘Chapter Eleven [NAFTA] neither provides nor specifically prohibits the use of countermeasures. Therefore, the question of whether the countermeasures defence is available to the Respondent is not a question of lex specialis, but of customary international law.’Footnote 28 The Tribunal confirmed that the only reference to countermeasures in the NAFTA was as a means of penalty for non-compliance with a decision rendered in a Chapter Twenty State-to-State arbitration.Footnote 29 In that case, no such decision has been rendered. Accordingly, it found that countermeasures can be invoked as a defence in a Chapter Eleven dispute subject to the conditions specified in general international law.Footnote 30 In other words, the silence of Chapter Eleven on the issue of countermeasures was interpreted as implicit acceptance of the relevant rules of general international law.
2.2 The Nature of Investors’ Rights under Investment Protection Treaties
The arbitral tribunals were faced with an additional question in deciding the applicability of the customary defence of countermeasures to disputes under Chapter Eleven NAFTA. A responding State would invoke the defence of countermeasure as a response to a prior internationally wrongful act by the State of nationality of the investor. Nonetheless, that State is not a party to the dispute. It is rather the investor, a private individual, that takes part in the proceedings under Chapter Eleven. The tribunals discussed the nature of investors’ rights under Chapter Eleven and whether such nature affects the applicability of the defence of countermeasures under general international law to investor-State disputes.
The claimants argued that investors under Chapter Eleven NAFTA
are vested with direct independent rights and … are immune from the legal relationship between the Member States. The investor’s cause of action is grounded upon substantive investment obligations which are owed to it directly. A breach of these obligations does not therefore amount to a breach of an inter-state obligation; thus the general rules of state responsibility – including those regarding the circumstances precluding wrongfulness – cannot be presumed.Footnote 31
In other words, countermeasures cannot be invoked as a defence to justify non-performance of obligations under Chapter Eleven, because such obligations are owed to the investors directly and not their State of nationality. According to this argument, the investor is a third party to the dispute between its State of nationality and the responding host State.
Mexico, on the other hand, argued that obligations under NAFTA, including Chapter Eleven, are owed only to the State of the investors’ nationality. According to this line of argument, investors are either ‘stepping into the shoes and asserting the rights of their home State’ when initiating arbitration (derivative theory) or are ‘vested only with an exceptional procedural right to claim State responsibility’ (intermediate theory).Footnote 32 The relevant substantive obligations remain always inter-State, and thus, the defence of countermeasures can be properly invoked against the State of nationality for prior internationally wrongful acts that it has committed.
Indeed, under customary international law, countermeasures ‘must be directed against’ a State which is the author of the internationally wrongful act.Footnote 33 Accordingly, the wrongfulness of an act taken as countermeasure is precluded only with respect to obligations owed to the responsible State and not obligations owed to another party.Footnote 34 On the other hand, as stipulated in the ILC Commentary to Article 49 ARSIWA, a countermeasure can affect the interests of third parties as an ‘indirect or collateral effect’.Footnote 35 Thus, it must be determined whether an investor under NAFTA ‘has rights of its own, distinct from those of the State of its nationality, or merely interests.’Footnote 36
In ADM, the Tribunal sided with the view that investment rights are derivative and thus countermeasures can indeed be invoked as a defence by the responding State as they do not affect ‘individual substantive rights’.Footnote 37 It found that
the proper interpretation of the NAFTA does not substantiate that investors have individual rights as alleged by the Claimants. Nor is the nature of investors’ rights under Chapter Eleven comparable with the protections conferred by human rights treaties. Chapter Eleven may share … with human rights treaties the possibility of granting to non-State actors a procedural right to invoke the responsibility of a sovereign State before an international dispute settlement body. But the fundamental difference between Chapter Eleven of the NAFTA and human rights treaties in this regard is … that Chapter Eleven does not provide individual substantive rights for investors, but rather complements the promotion and protection standards of the rules regarding the protection of aliens under customary international law.Footnote 38
According to the Tribunal, the substantive obligations under Chapter Eleven remain inter-State providing the standards by which the conduct of the NAFTA Party towards the investor will be assessed in the arbitration.Footnote 39 These substantive obligations cannot be waived by the investors.Footnote 40 They are binding by virtue of the agreement between the State parties to the treaty. Moreover, according to the Tribunal, these obligations are complemented by customary international law to the extent that it is not displaced by the lex specialis of the treaty.Footnote 41 On the contrary, the right of investors to trigger arbitration against the host State is a purely procedural one.Footnote 42 Investors are given the ‘exceptional right of action through arbitration that would not otherwise exist under international law’Footnote 43 and it is a right that they may choose to exercise or waive, at their own discretion. When an investor files a request for arbitration, it accepts a standing offer by the host State and creates a direct legal relationship in the form of an arbitration agreement.Footnote 44 This is the only direct relationship under international law. The relationship between the State of nationality and the host State remains unchanged.
By contrast, the tribunals in CPI and Cargill took a very different approach to this issue. They found that rights under Chapter Eleven NAFTA are not merely procedural.Footnote 45 Rather ‘NAFTA confers upon investors substantive rights separate and distinct from those of the State of which they are nationals.’Footnote 46 The Tribunal in CPI put emphasis on the interpretation of the relevant treaty provisions. It held that individuals and corporations can also hold rights under international law and that when such rights are said to be derived from a treaty ‘the question will be whether the text of the treaty reveals an intention to confer rights not only upon the Parties thereto but also upon individuals and/or corporations.’Footnote 47 In the case of NAFTA, the Tribunal found that the parties’ intention was to confer substantive rights directly upon investors.Footnote 48 This was, according to the Tribunal, evident from the language of the treaty and from the fact that procedural rights are also conferred upon the investors.Footnote 49 The Tribunal observed that ‘[t]he notion that Chapter XI conferred upon investors a right, in their own name and for their own benefit, to institute proceedings to enforce rights which were not theirs but were solely the property of the State of their nationality is counterintuitive.’Footnote 50
Further to this textual interpretation of NAFTA, the Tribunal in CPI referred to the rights of aliens under customary international law. It argued that ‘the notion that in diplomatic protection cases the State was asserting a right of its own’ was just a fiction, which was only necessary because procedurally individuals could not bring an international claim.Footnote 51 It further argued that this fiction did not reflect the substantive reality, something that is evident ‘not only in the juristic writing but also in various rules of law surrounding diplomatic protection claims’ such as the rule on exhaustion of local remedies and the doctrine of continuing nationality.Footnote 52 According to the Tribunal, if the rights were rights of the State itself, then there wouldn’t be a requirement to bring a case first before domestic courts, as is the case with other international obligations owed directly to the State. Similarly, if an injury to the national is a violation of the rights of the State, then the victim’s nationality after the date of the injury would not be of legal relevance. Thus, the Tribunal implies that even under customary international law, the substantive rights of investors are essentially direct. Accordingly, the investor ‘is a third party in any dispute between its own State and another NAFTA Party and a countermeasure taken by that other State against the State of nationality of the investor cannot deprive that investor of its rights.’Footnote 53
The Tribunal in Cargill provided a different reasoning in support of its finding that investors’ rights under NAFTA are substantive. It drew a distinction between those rights and diplomatic protection under general international law. According to the Tribunal, ‘in the case of diplomatic espousal … the claim is owned by the espousing State and the espousing State is the named party. Moreover, the operative paragraph of the resulting award reciting the decision of the tribunal names the espousing State, and not the national’.Footnote 54 Conversely, in the case of investment arbitration under NAFTA Chapter Eleven, the investor itself initiates proceedings, is the named party to such proceedings and is named in the dispositive of the award.Footnote 55 The Tribunal held that the origin of the rights should not be confused with the holder of the rights: ‘That the origin of individual rights may be found in the act of a sovereign, or in the joint act of sovereigns, does not negate the existence of the rights conferred’.Footnote 56 It thus concluded that countermeasures cannot afford a defence in respect of a claim asserted under Chapter Eleven by a national of the allegedly offending State.Footnote 57
2.3 The Customary Requirements for a Lawful Countermeasure and the Jurisdictional Limits of Arbitral Tribunals
The tribunals in their analysis regarded Articles 22 and 49–53 ARSIWA ‘as an authoritative statement of customary international law on countermeasures’.Footnote 58 Article 49 ARSIWA provides that an injured State may take countermeasures, in the form of limited and temporary non-performance of international obligations, against a State which is responsible for an internationally wrongful act, in order to induce compliance of that State with its international obligations. Article 51 ARSIWA further provides the requirement of proportionality, ie, that the effects of the countermeasure must be commensurate with the injury suffered taking into account the gravity of the internationally wrongful act and the rights in question. The tribunals also referred to the findings of the International Court of Justice (ICJ) in Gabčikovo-Nagymaros,Footnote 59 which establishes the same criteria as those codified by the ILC.Footnote 60 The analysis of the tribunals implies acceptance of these criteria as part of the customary law on countermeasures.Footnote 61
The first requirement under customary international law, ie the existence of a prior internationally wrongful act, is also the most interesting in the context of investor-State dispute settlement (ISDS). The arbitral tribunals would have to determine whether such a prior breach has taken place in order to rule on the applicability of the defence. This raises important jurisdictional questions in cases where this prior breach is outside the limits of the tribunal’s subject matter jurisdiction.
First of all, as confirmed by the Tribunal in CPI, ‘the requirement of a prior violation of international law is an absolute precondition of the right to take countermeasures’.Footnote 62 Mexico had argued that to succeed in its countermeasures defence under customary international law, it did not need to prove that the US had indeed violated its obligations but rather that at the time of taking the countermeasure and while it was in place, it had ‘a genuine belief that it had a reasonable prospect of succeeding in establishing that the United States was in breach, should that question come before a competent tribunal.’Footnote 63 Accordingly, the Tribunal would not need exceed the limits of its jurisdiction as it could decide on whether Mexico had such genuine belief without examination of the actions of the US per se. The Tribunal rejected this argument as an attempt by Mexico to ‘square the circle’, confirming that the jurisdictional concern cannot be easily avoided.Footnote 64
The Tribunal in ADM found on this matter that it ‘has no jurisdiction to decide whether the United States breached any of its international obligations under Chapter Three or Chapter Twenty of the NAFTA.’Footnote 65 The Tribunal recalled that it was established under Chapter Eleven for the settlement of an investment dispute, comprising allegations of violations by the respondent of Articles 1102, 1106 and 1110 NAFTA. Thus, it had no jurisdiction to decide whether the US breached any of its international obligations complained by Mexico, since those were prescribed under NAFTA Chapters Three, Seven and Twenty.Footnote 66 Mexico argued that the jurisdiction of the Tribunal entailed the power to examine its argument on countermeasures as this was invoked as a defence precluding its international responsibility,Footnote 67 but the Tribunal rejected this line of reasoning.
Interestingly, the Tribunal found that it had jurisdiction to examine the rest of the customary requirements of countermeasures. It stipulated that ‘[b]oth parties agree that the Tribunal has jurisdiction to decide any defense under Chapter Eleven, including a countermeasures defense’.Footnote 68 Thus, the Tribunal acknowledged the nature of countermeasures as a defence and its jurisdiction to rule on applicable defences but still found that its power does not extend to all the customary requirements of such defence. Accordingly, the Tribunal proceeded to examine whether Mexico’s tax measure was taken in response to the alleged US violations, aimed at inducing the US to comply with the NAFTA obligations that it has allegedly breached and was proportionate to such aim, without first establishing that a prior internationally wrongful act has taken place. It found that Mexico’s measure did not meet these requirements and rejected the countermeasures defence on that basis, thereby circumventing the jurisdictional considerations.
The Tribunal in CPI reached a similar conclusion regarding the limits of its jurisdiction. In an obiter dictum,Footnote 69 it held that it would not have the jurisdiction to determine whether Mexico’s allegations against the US were well-founded or not, because ‘the United States is not a party to these proceedings and the Tribunal does not have jurisdiction to determine whether any provision of the NAFTA falling outside Chapter XI has been violated’.Footnote 70 The Tribunal here raises not only the issue of jurisdiction ratione materiae, discussed in ADM, but also a concern regarding the limits of its personal jurisdiction. It makes a Monetary Gold-like argument,Footnote 71 implying that it may not determine the legal interests of a non-party to the dispute (the US) without its consent.Footnote 72
3 Evaluating the Tribunals’ Approach: Treaty Interpretation, Customary Law and General Principles of International Adjudication
It becomes evident from the analysis above that tribunals have adopted opposing views on issues crucial to the applicability of the customary defence of countermeasures to investor-State disputes. Despite the inconsistencies, through their reasoning and analysis we can identify the main questions and opposing arguments regarding the applicability of defence. This section evaluates the tribunals’ reasoning and discusses the availability of countermeasures to responding States in ISDS in view of the customary rules of treaty interpretation, the lex specialis principle and general principles of international adjudication.
3.1 The Lex Specialis Principle: Revisiting the Interaction of Countermeasures with Investment Protection Treaties
As discussed in Section 2, only the Tribunal in ADM examined how the defence of countermeasures under customary international law interacts with the NAFTA provisions and whether it is displaced by a lex specialis. The analysis in ADM confirms, in the first place, the applicability by default of the general international law on State responsibility, including the circumstances precluding wrongfulness, to treaty-based claims. The Tribunal’s analysis on the issue of lex specialis suggests that the absence of explicit derogation must be regarded as a continuation, or rather, an implicit acceptance of the existing rules under general international law. The silence of NAFTA on the issue of countermeasures as a defence was interpreted as implicit acceptance of its applicability to disputes arising thereunder.
This approach is in line with the general ‘presumption against normative conflict’ in international law.Footnote 73 As explained by the Arbitral Tribunal in the Georges Pinson case:
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.Footnote 74
The ICJ has also adopted the same approach on the applicability of general international law to treaty claims. In the case of ELSI, the ICJ Chamber discussed whether the customary rule of exhaustion of local remedies for the exercise of diplomatic protection was applicable, even though the compromissory clause in question made no reference to such prerequisite for submission of a dispute to the Court. It held that it was ‘unable to accept that an important principle of customary international law should be held to have been tacitly dispensed with, in the absence of any words making clear an intention to do so’.Footnote 75 Nonetheless, the Court admitted that the parties could indeed deviate from the customary rule. As Sir Frank Berman points out ‘[a]s a matter of abstract logic, it is perfectly conceivable that a pair of Contracting States might wish to displace the general law in their mutual relations; very often that is the whole purpose of a bilateral treaty.’Footnote 76 The Court suggested, however, through its reasoning, that this should be clearly discernible from the text of the treaty and cannot be presumed. The Tribunal in ADM seems to adopt the same approach. General international law, including the defence of countermeasures, applies by default to all international disputes, subject to the application of the lex specialis principle, which dictates that a special provision shall prevail, to the extent that it clearly derogates from general international law. Thus, the premise of the Tribunal’s reasoning in ADM is consistent with the customary rules of interpretation and the lex specialis principle. Chapter Eleven NAFTA does not seem to preclude the applicability of countermeasures to disputes arising thereunder.
Nevertheless, Chapter Twenty NAFTA establishes a lex specialis that displaces at least some countermeasures: those in response to violations of NAFTA itself. As evidenced also by the Tribunal’s reasoning in ADM, countermeasures in international law have a ‘dual’ character, which is already reflected in their double appearance in both Articles 22 and 49 to 54 ARSIWA: they constitute a means of implementing international responsibility and at the same time a defence against a claim of breach.Footnote 77 NAFTA indeed specifies the means by which international responsibility may be implemented for breaches of its own provisions by one of its parties. In such cases the parties may resort to the dispute settlement provisions of the NAFTA as well as the provisions on suspension of benefits (‘countermeasures’ under NAFTA aiming to induce compliance with a panel decision).Footnote 78 These provisions constitute lex specialis with respect to some countermeasures: those in response to a prior breach of the same treaty. The do not, however, preclude all countermeasures.Footnote 79 If, for example, the respondent argued that non-performance of its NAFTA obligations was in response to prior violations of another State party in the field of human rights or environmental protection, this scenario would not be covered by the dispute settlement or the suspension of benefits provisions of the NAFTA, and thus, it wouldn’t be governed by the lex specialis. Lex specialis prevails over the relevant lex generalis only to the extent that the two rules are co-extensive and in genuine conflict.Footnote 80 Aspects of the lex generalis that are not derogated from continue to apply by default in the relations between the parties.
The Tribunal referred to the provisions on countermeasures under Chapter Twenty and stipulated that since there has been no panel decision under Chapter Twenty the provisions on suspension of benefits were inapplicable. But it did not acknowledge that the dispute settlement arrangements under Article 2004 NAFTA establish a compulsory procedure that displaces other unilateral means for the implementation of responsibility with respect to NAFTA violations. Article 2004 provides that ‘the dispute settlement provisions of this Chapter shall apply with respect to the avoidance or settlement of all disputes between the Parties regarding the interpretation or application of this Agreement or wherever a Party considers that an actual or proposed measure of another Party is or would be inconsistent with the obligations of this Agreement …’.Footnote 81 Unilateral countermeasures in the form of suspension of obligations under another NAFTA chapter in order to induce compliance with NAFTA Chapter Twenty are incompatible with this provision.
In ADM, Mexico invoked the defence of countermeasures under general international law with respect to the US breaches of NAFTA itself. In principle, this situation would indeed be regulated by the lex specialis, ie, the NAFTA provisions on State-to-State dispute settlement. The Tribunal introduced an artificial distinction between different chapters of the same instrument and stipulated that only in Chapter Twenty we find any reference to countermeasures, whilst the dispute at hand was brought before it under Chapter Eleven.Footnote 82 However, this argument, on its own, is not convincing. In principle, any grievances against the State of nationality of the investor for alleged breaches of the NAFTA are subject to the dispute settlement provisions of Chapter Twenty and cannot be presented before a Chapter Eleven tribunal through the defence of countermeasures. This approach is in line with the principle of effective interpretation:Footnote 83 the tribunal should not read a treaty in a manner that leads its provisions (in this case, the dispute settlement provisions under Chapter Twenty) to redundancy or inefficiency.Footnote 84
The Tribunal’s findings in ADM were probably motivated, albeit implicitly, by another factual aspect of the dispute. As the respondent submitted before the Tribunal, the means described in NAFTA Chapter Twenty for implementing the international responsibility of the US were not available to Mexico, as the US had blocked Mexico’s access to a NAFTA panel.Footnote 85 This very fact constituted one of the breaches of NAFTA complained of by Mexico. Therefore, neither the dispute settlement nor the countermeasures’ provisions under NAFTA were actually available to Mexico in that case. In the words of Mexico, ‘a State party cannot be bound by a lex specialis that has proved impossible to invoke’.Footnote 86 The Tribunal, through its findings, seems to implicitly side with this argument. This approach can also be justified on the basis of general principles of international law. The principle of law expressed in the maxim ex iniuria jus non oritur dictates that unlawful conduct cannot modify the law applicable in the relations between the parties, or, in other words, that States should not benefit from their own wrong.Footnote 87
3.2 The Question of Independent Rights: Revisiting the Power of Host States to Suspend Investors’ Rights in Response to Conduct of Their Home State
Regarding the nature of the rights of investors, the analysis of the tribunals, discussed above, is twofold. In the first instance it seems that they interpret the NAFTA to discern whether the rights enshrined therein are directly ascribed to the investor or derivative of the inter-State relationship. Through this first limb of analysis, they seem to suggest that the issue of the nature of investors’ rights is not a theoretical question to be determined in abstracto. As it has been pointed out by Paparinskis,
the nature of the rights is not an abstract and irrebutable a priori proposition, and as a rule of jus dispositivum is open to amendment or reinterpretation, in particular through subsequent agreement and practice. … The nature of investors’ rights may at least in principle differ in different investment treaty rules, having important consequences for the applicability of countermeasures.Footnote 88
However, the second part of their reasoning traces the nature of investors’ rights back to customary international law. For example, the analysis in ADM suggests that investment treaties are simply an extension of the customary rules regarding the protection of aliens, which are already binding on all States. According to the Tribunal, although the institution of diplomatic protection is displaced by the establishment of a different dispute settlement mechanism for the implementation of responsibility under investment treaties, the nature of the substantive obligations regarding the protection of aliens, that already existed under customary international law, remains unaffected by the addition of further obligations in investment treaties. Although the Tribunal did not expressly characterise it as such, its reasoning was in essence an interpretation of the obligations under NAFTA in view of relevant rules of customary international law. The Tribunal in CPI, although it reached a different conclusion, followed a similar approach. First, it offered an interpretation of the NAFTA and argued that it confers direct rights to the investors. Then, it further buttressed this argument by recourse to customary international law. According to the Tribunal, evidence suggests that even under the traditional institution of diplomatic protection, where the State is the one bringing the investor’s claim to the international plane, the rights are in reality owned by the investor. Accordingly, the NAFTA provisions, read in light of customary international law, suggest that investors’ rights are independent from their home State. In Cargill, the Tribunal had recourse to the customary law on diplomatic protection in the context of an a contrario argument. It argued that diplomatic protection is diametrically different to contemporary investment arbitration and that such differences lead to the conclusion that today investors’ rights under investment treaties are direct, whereas under customary international law and the rule on diplomatic protection, they remained tied to the State of nationality.
It emerges from the above that all three tribunals are using the same evidence relating to the protection of aliens and the institution of diplomatic protection under customary international law, but reach very different interpretative results. It becomes evident that there is a lack of methodology in the deductive reasoning of tribunals which leads to conflicting conclusions and inconsistencies in terms of reasoning.
In evaluating the evidence and arguments raised in the three cases above, there is nothing to suggest that the rights of investors are detached from their State of nationality. The default applicability of customary international law, discussed in the previous sub-section, suggests that substantive obligations assumed by States under investment treaties are simply in addition to the existing customary obligations and they do not alter their scope and nature if such change is not explicitly provided for. Thus, only the procedural aspects of implementation of international responsibility are affected by current investment treaties that provide for investor-State arbitration without the involvement of the State of nationality.
Moreover, under customary international law, obligations assumed by States which have individuals are beneficiaries and are considered non-derogable are listed in Article 50 ARSIWA as obligations that cannot be affected by countermeasures. The tribunals failed to provide evidence that such list under customary international law has been expanded and also failed to take this list into consideration in interpreting the provisions of the investment treaty in question.
Lastly, other provisions that are typically included in investment protection treaties further suggest that the protection of investors is not immune from changes in the relationship between the host State and the State of nationality. Most pertinently, the ‘security exception’, which is included in a number of bilateral and multilateral investment protection treaties, including NAFTA,Footnote 89 often stipulates, among others, that nothing in the text of the agreement should be construed ‘to prevent any Party from taking any actions that it considers necessary for the protection of its essential security interests … taken in time of war or other emergency in international relations’.Footnote 90 It becomes evident that under this clause, an emergency in the relations between the home State of the investor and the host State would justify the taking of measures that are prima facie incompatible with the obligations of the host State under the investment agreement. In other words, through no fault of its own, the investor may have to suffer the consequences of the deterioration of the relations between the two States to the extent that the essential security interests of the host state are at risk. Countermeasures taken in response to a prior internationally wrongful act of the State of nationality of an investor may affect the interests of the latter in a similar fashion.
3.3 Countermeasures as a Response to a Prior Internationally Wrongful Act: Revisiting the Limits of Arbitral Tribunals’ Subject Matter and Personal Jurisdiction
As seen in the analysis of the case law above, the arbitral tribunals that have been seized with the issue of the applicability of countermeasures in investment disputes so far have taken a restrictive approach to the limits of their subject matter and personal jurisdiction. They require that the indispensable incidental matter which arises in the context of an applicable defence, ie, the existence of a prior internationally wrongful act in the case of countermeasures, would also fall within their scope of jurisdiction if it was brought before them as an independent claim, in order to proceed to its examination. This essentially means that very few countermeasures, if any at all, could properly be examined by an arbitral tribunal: those that consist of the non-performance of obligations arising under the same investment treaty.
The WTO adjudicative bodies have also taken a similar approach to this issue. In the case of Mexico – Soft Drinks, which was adjudicated in the context of the same broader US-Mexico dispute, the panel and Appellate Body argued that their jurisdiction ratione materiae does not entail the power to examine a prior internationally wrongful act that constitutes violation of rules other than the WTO covered agreements. According to the AB, WTO adjudicative bodies cannot ‘become adjudicators of non-WTO disputes’, as this is not their function as intended by the DSU.Footnote 91
This approach introduces a Monetary Gold-like consideration with regards to jurisdiction ratione materiae.Footnote 92 The arbitral tribunals examined above, as well as the WTO adjudicative bodies, seem to suggest that ‘to adjudicate upon the international responsibility of [a State] without its consent [on that particular matter] would run counter to a well-established principle of international law … namely, that [an adjudicative body] can only exercise jurisdiction over a State with its consent.’Footnote 93
In the case of ISDS, we are also faced with a more traditional Monetary Gold-like consideration: the investor’s State of nationality, who is the author of the alleged internationally wrongful act, is absent from the proceedings. This raises a problem regarding the scope of the arbitral tribunal’s jurisdiction ratione personae, as it is called to discuss the international responsibility of a non-participating State.
3.3.1 The Case for an Expansive Approach to the Jurisdiction of Arbitral Tribunals: The Approach of the ICJ
In its 2020 International Civil Aviation Organisation (ICAO) Council judgments,Footnote 94 the ICJ expressly endorsed an expansive approach to the jurisdiction of adjudicative bodies in the context of applicable defences. The applicants (Saudi Arabia, Bahrain, United Arab Emirates and Egypt, also known as ‘the Quartet’) challenged the decision of the ICAO Council to uphold its jurisdiction over Qatar’s claims relating to the Quartet’s aviation restrictions against Qatar-registered aircrafts. The Quartet had argued that the ‘ICAO Council lacked jurisdiction under the Chicago Convention since the real issue in dispute between the Parties involved matters extending beyond the scope of that instrument, including whether the aviation restrictions could be characterised as lawful countermeasures under international law’.Footnote 95 The Council rejected this preliminary objection. The respondents instituted an appeal from the Council’s decision on jurisdiction before the ICJ under Article 84 of the Chicago Convention. The ICJ, in its judgment, found that the ICAO Council did not err in rejecting the Quartet’s assertion and rejected the applicant’s grounds of appeal.
In its judgment, the Court stipulated that ‘the integrity of the Council’s dispute settlement function would not be affected if the Council examined issues outside matters of civil aviation for the exclusive purpose of deciding a dispute which falls within its jurisdiction … Therefore, a possible need for the ICAO Council to consider issues falling outside the scope of the Chicago Convention solely in order to settle a disagreement relating to the interpretation or application of the Chicago Convention would not render the application submitting that disagreement to it inadmissible.’Footnote 96 In other words, the ICJ confirmed that the ICAO Council’s jurisdiction entails the power to decide an indispensable incidental matter in the context of the defence of countermeasures in order to discharge its function under the Chicago Convention.
The findings of the Court in 2020 are in line with previous jurisprudence. The PCIJ, already in 1927, famously pronounced in Factory at Chorzów that ‘reparation … is the indispensable complement of a failure to apply a convention … Differences relating to reparations … are consequently differences relating to its application’.Footnote 97 The power of an adjudicative body to examine a defence along with all issues indispensable to determine its applicability in the case before it follows from the ‘principle that jurisdiction to determine a breach implies jurisdiction to award compensation’,Footnote 98 or more generally, to rule on the content of a State’s international responsibility. The applicability of a defence under the law of State responsibility determines whether the responding State bears international responsibility and the ensuing consequences. Thus, incidental findings in the context of a respondent’s defence are indispensable for the adjudicative body to exercise its function.
This approach is also in line with the principle elaborated by the ICJ in the Nuclear Tests case. The Court confirmed that an adjudicative body ‘is fully empowered to make whatever findings may be necessary’ in order ‘to ensure that the exercise of its jurisdiction over the merits, if and when established, shall not be frustrated’ and ‘to provide for the orderly settlement of all matters in dispute’.Footnote 99 Accordingly, the competence of an adjudicative body to examine all indispensable incidental issues in the context of an applicable defence is part of its inherent powers deriving from its ‘mere existence … as a judicial organ established by the consent of States, and is conferred upon it in order that its basic judicial functions may be safeguarded.’Footnote 100
Importantly, the approach of the ICJ suggests that the limited or specialised jurisdictional field of an adjudicative body, ie, the general class of cases in respect of which it exercises and is entitled to exercise its functions,Footnote 101 does not affect the power to rule on such indispensable incidental issues in the context of defences. The function of an international adjudicative body, regardless of its character as general (such as the ICJ, which may, in principle and subject to the parties’ consent, exercise jurisdiction over all issues of international law, as outlined in Article 36 its Statute) or specialised (such as an investment arbitral tribunal or the WTO panels and Appellate Body), is to decide the case that is brought before it in accordance with its statute and rules of procedure. To discharge this function, it may need to examine rights and obligations that are necessarily implicated by the main claim before it. The findings of the ICJ in the 2020 ICAO Council case discussed above clearly support this argument. The ICJ confirmed that the ICAO Council, a specialised dispute settlement body, has the power to examine indispensable incidental issues for the purpose of deciding a case properly brought before it.Footnote 102
Moreover, the argument that courts of general competence can have an expansive understanding of their jurisdiction whereas specialised courts should exercise self-restraint seems to imply, essentially, a difference in the competency, adequacy or suitability of such court to determine a wider spectrum of issues under international law. Naturally, the competency of the court can only be assessed by reference to its members. In other words, this argument seems to imply that the judges (and not the court) are in one case more ‘competent’ to determine all issues of international law than in the other. In international dispute settlement, as we know it today, this assumption cannot be easily substantiated. Although the professional and academic backgrounds of arbitrators vary, we cannot disregard the fact that several of them are highly specialised in general public international law and have even served as judges in other international adjudicative fora.Footnote 103 When broader matters of international law are implicated in the dispute in question, the arbitral panel should comprise individuals that possess the necessary knowledge. This is a matter that relates to the selection of arbitrators in any given case.
3.3.2 Countermeasures vs Counterclaims
The tribunals’ narrow approach to jurisdiction ratione materiae brings to mind the requirements for the admissibility of counter-claims.Footnote 104 Article 46 of the ICSID Convention provides that ‘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.’
Indeed, there is a reasonable analogy to be drawn between the requirements on the admissibility of a counterclaim and an adjudicative body’s jurisdiction to examine the defence of countermeasures. Given that countermeasures are a lawful reaction to a prior internationally wrongful act, a responding State can always claim that its action was in response to the wrongful act of the applicant regardless of whether the court or tribunal would be able to rule on the wrongfulness of such act as an independent claim or counterclaim. This could constitute a back door for entertaining claims that would otherwise be inadmissible.
However, in the case of countermeasures, the respondent will not have an independent verdict on the actions of the applicant. The court’s finding of a prior breach will only constitute part of its reasoning to determine the case brought before it by the applicant. This was clearly stipulated by the ICJ in the 2020 ICAO Council judgment: incidental findings are made by an adjudicative body ‘for the exclusive purpose of deciding a dispute which falls within its jurisdiction’.Footnote 105 As Judge Higgins conceded in her, otherwise very critical, separate opinion to the Oil Platforms judgment, the non ultra petita rule ‘does not operate to preclude the Court from dealing with certain other matters “in the reasoning of its Judgment, should it deem this necessary or desirable”’.Footnote 106
Moreover, an incidental finding would normally not appear in the judgment’s dispositif.Footnote 107 The findings of an adjudicative body may still be an authoritative affirmation of the applicant’s wrongdoing but incidental findings dependent on the principal claim in the dispute cannot generate any rights to remedies. This is a key difference which suggests that Monetary Gold-like considerations and analogies with the rules on counterclaims are misplaced.
4 Concluding Remarks
Economic restrictions, or the threat thereof, have always been a core foreign policy tool,Footnote 108 used to enforce international rules, react to illegality, prevent conflict, respond to emerging or current crises or exert pressure towards a change in policy or activity. To the extent that such restrictions are, prima facie, inconsistent with a State’s international obligations, and not lawful (yet unfriendly) acts of retorsion, their imposition must be justified on the basis of an applicable defence, or it will prompt the consequences of international responsibility.Footnote 109 It is thus essential to clarify whether and under what conditions, restrictions that affect the rights of foreign investors are allowed under the provisions of current investment protection treaties and customary international law.
Given the importance of foreign investment for both the destination and the origin State in terms of economic growth and productivity, and the prime role of investment protection commitments in the bilateral relations of States, the imposition of restrictions affecting the rights of foreign investors can be a powerful tool for the enforcement of the international rule of law. The defence of countermeasures under customary international law recognises the right of States to react to a breach of international obligations by temporarily suspending its own obligations towards the responsible State in order to induce compliance with international law. This chapter discussed whether this customary defence can be invoked in the context of relevant arbitral proceedings in order to preclude the wrongfulness of restrictions that are prima facie contrary to a State’s investment protection obligations. In other words, it aimed to clarify whether suspension of investment protection can be used as a tool to exert pressure on a State that violates international law.
This chapter demonstrated that the defence of countermeasures under general international law is applicable by default to investment disputes, unless interpretation of the investment treaty in question suggests otherwise. On the nature of investors’ rights, on which much ink has been spilled in the academic literature, it was argued that they are not detached from the inter-State relations of the home and the host State. Customary international law on the protection of aliens and the institution of diplomatic protection informs the content of current investment treaties and suggests that the protections enshrined therein are an extension of such customary law, complemented by a procedural right to initiate proceedings without the involvement of the State of nationality. The text and context of investment treaties provides further support to this argument. Lastly, this chapter demonstrated that the power of an adjudicative body to rule on all indispensable incidental matters that arise in the context of an applicable defence is inherent. Findings on such incidental matters form part of the reasoning of the tribunal and have no independent legal force. Thus, the limited scope of an arbitral tribunal’s subject matter jurisdiction and the absence of the State of nationality of the investor from the proceedings do not preclude the examination of the defence of countermeasures.
In view of the above, this chapter concludes that the customary defence of countermeasures is indeed available to responding States in international investment proceedings. A host State is entitled under general international law to react to another State’s breach of an international obligation by temporarily suspending its protection of the latter’s investors within its territory, in accordance with the requirements for a lawful countermeasure under customary international law. Such countermeasures can be a powerful arrow in the quiver of States that can be used to ensure the effective implementation of international responsibility.
1 Introduction
According to Article 27(b) of the Articles on the Responsibility of States for Internationally Wrongful Acts (ARSIWA), the successful invocation of a defence is ‘without prejudice … to the question of compensation for any material loss caused by the act in question’.Footnote 1 The Commentary to this provision clarifies that this compensation is not part of the framework of reparation: it is not, in short, one of the obligations arising out of the wrongful act.Footnote 2 It concerns, instead, the question whether a State invoking a defence ‘should nonetheless be expected to make good any material loss suffered by the State directly affected’.Footnote 3 Material loss, the Commentary continues, is a narrower concept than damage as it concerns only the adjustment of losses that may occur when a party relies on a defence.
Beyond this, the Commentary gives little guidance as to when such a duty could be owed. It states that in certain situations a duty of compensation ‘is a proper condition’ for allowing reliance on a defence, as otherwise a State might shift the burden of protecting its own interests onto other ‘innocent’ third parties.Footnote 4 By way of example, it notes that Hungary accepted this principle when relying on the plea of necessity in Gabčíkovo-Nagymaros.Footnote 5 But the Commentary does not say in respect of which defences, and in what circumstances, compensation will be due. Indeed, it does not say that compensation is due in cases of necessity; it only notes that Hungary offered compensation when invoking necessity. Indeed, the Commentary clarifies that it ‘does not attempt to specify in what circumstances compensation should be payable’.Footnote 6 The most that it offers, by way of guidance, is that it will be for the parties involved to agree on any possible compensation.Footnote 7
It is not unusual for parties who benefit from the plea of necessity to offer compensation to affected parties. Hungary, as noted by the ILC, did it in Gabčíkovo-Nagymaros and, more recently, Bolivia may be seen as having made a similar offer in an investment arbitration.Footnote 8 But such instances are certainly not the norm. In most instances, no offer of compensation will be forthcoming and, consequently, there will be no agreement between the parties. What happens, then, when the parties do not agree on any possible compensation for material loss? What happens when, as has most often been the case in practice, the State invoking a defence rejects that it owes compensation to the affected party? It is precisely here that the question whether compensation is owed, as a matter of obligation, arises.
For the most part, investment tribunals have had to address situations in which offers of, or agreements on, compensation for material loss caused by acts adopted in a state of necessity have not been forthcoming. At least four different States have relied on the defence of necessity to justify,Footnote 9 and thus, render lawful conduct incompatible with their obligations under bilateral investment treaties.Footnote 10 In all cases the claimants have appealed to ARSIWA Article 27(b) and argued that, notwithstanding the necessity defence, respondents were required to compensate them for the material loss suffered as a result of the allegedly justified measures. All States denied owing such compensation.
Are these States required to compensate claimants for the loss caused by their justified, and therefore lawful, conduct? Absent the parties’ agreement, an obligation to make compensation in these circumstances requires a basis in positive law. This compensation, as the ILC Commentary to Article 27 clarifies,Footnote 11 is not a form of reparation; after all, there has been no wrongful act. So, it cannot be based on the obligation to make reparation that arises for States as a consequence of a wrongful act. It must have a discrete legal source. Investment tribunals, deciding in accordance with international law needed, therefore, to identify a positive law basis for the respondent’s duty to compensate material loss resulting from acts justified by necessity.
A duty of compensation could be found in the relevant applicable treaty: the treaty may specifically provide for this.Footnote 12 Investment tribunals have also interpreted such a duty from the purpose of the Bilateral Investment Treaties (BITs) themselves. Thus, the Tribunal in BG Group found that a duty of compensation in these circumstances could be required by the UK–Argentina BIT.Footnote 13 But this is relatively rare. Most treaties do not make provision for compensation in relation to emergency measures and, when they do, they often provide for compensation in only a limited range of cases. In all other cases, therefore, a tribunal will need to look to other sources of international law: customary law or general principles of law. The focus of this chapter is the tribunals’ engagement with customary international law in their assessment of the existence of an obligation to make compensation for material loss caused by acts justified under the plea of necessity. As will be seen, tribunals have reached opposite conclusions on the existence of a duty of compensation. However, they all share in common one feature: an omission to engage with the method of customary law identification. Some tribunals assert the existence of the duty, others derive it from the elements of the customary defence of necessity, yet others still simply name-check precedents and general principles of law. But none of these awards provided any evidence of practice and opinio juris in relation to this duty.
In addition to this introduction, the chapter proceeds in three steps. Section 2 provides an overview of the investment case law. It will review the range of conclusions reached by different investment tribunals and, in so doing, focus on the reasoning deployed to reach those conclusions. It will show that, whether they accept or reject the existence of a duty of compensation, their reasoning does not involve any engagement with the elements of customary law. At most, tribunals offer vague references to previous precedents and case law, and never once to State practice and opinio juris. Section 3 will then focus on analysis: it reviews the available practice and opinio juris, limited as it is, and assesses the precedents invoked in support of the duty of compensation by these tribunals. As will be seen, the practice is scant and inconsistent and the precedents invoked are at best equivocal as to the existence of a duty of compensation. If States have expressed any opinio in this regard this is an opinio non juris: there is no customary obligation to make compensation in cases of necessity. Section 4 concludes.
Two clarifications are necessary before proceeding. First, this chapter takes an orthodox approach to the identification of customary law, in line with the so-called ‘two element theory’ followed by the ILC in its recent work on customary law and supported by States in connection with that work.Footnote 14 In light of this, it will focus first and foremost on identifying existing practice and opinio juris of States in respect of the duty of compensation. The article will also take into account the case law of international tribunals. This is because while international courts and tribunals do not have a formal role in the development of international law, in practice, decisions of international tribunals can influence the development of international law, including the law of State responsibility and the law of investment protection.Footnote 15 Second, this chapter will take necessity as a justification, namely, as a defence which renders conduct lawful, and not as an excuse, namely, as a defence which excludes the consequences of a wrongful act.Footnote 16 The reasons for this choice are that the majority of States who support this defence at international law classify it as a justification, and that States have invoked it in international courts and tribunals as such.Footnote 17
2 Investment Tribunals and the Duty of Compensation
Several States have invoked the plea of necessity in investment treaty arbitration. In most (if not all) of these instances, the parties have addressed the question of compensation in the event that the State’s plea of necessity was successful. Likewise, in most instances, tribunals have addressed the duty of compensation in obiter only: respondents’ plea of necessity having been unsuccessful on other, often multiple, grounds. Many tribunals have not addressed the question of compensation at all: having rejected the plea of necessity on other grounds, there was no need to consider this issue.Footnote 18
Tribunals’ approaches to the duty of compensation have varied significantly, covering the full range of possibilities: some have accepted the existence of this duty, others have denied it, and others still have not taken a position. They have all had in common, however, minimal or no engagement with the evidence of the (potential) positive law source of this duty. In particular, none of these tribunals have applied the orthodox method (or any other method, for that matter) for the identification of customary law: tribunals have instead resorted to simple assertions, deductions, and denials. The next four sections review the different approaches taken by investment tribunals.
2.1 Assertions
Some tribunals have asserted the existence of a duty of compensation. When a tribunal asserts a rule, it provides no reasoning (inductive or deductive as it may be) in support of the stated rule.Footnote 19 To use Stefan Talmon’s words, asserting customary rules is like pulling rabbits out of a hat.Footnote 20 To be sure, assertion is not a method for the identification of customary rules: it ‘is a way of stating a conclusion.’Footnote 21 For the most part, investment tribunals have not simply ‘pulled’ the duty of compensation from out of a hat. But their reasoning in support of this duty is often unsuitable, and where available it is so thin as to provide no support at all.
The Tribunal in CMS v Argentina held that Article 27(b) ‘establishe[d] the appropriate rule of international law on this issue’ and that it was ‘the meaning of international law or the principles governing most domestic legal systems’ that a party invoking necessity owed compensation.Footnote 22 These seem to be references to customary law and general principles of law as the potential source of the duty of compensation. Reliance on each of these two sources is, however, insufficient. As to general principles, these are referred to in two paragraphs,Footnote 23 and in neither case are references provided. As to customary law, the tribunal provided no evidence of practice or opinio juris, and simply mentioned the cases of Orr and Laubenheimer,Footnote 24 General Company of the Orinoco River,Footnote 25 Bulgarian Property,Footnote 26 and Gabčíkovo-NagymarosFootnote 27 as precedents. To be sure, tribunals can (and do) contribute to the development of the law: but once these cases are considered in detail, it will become apparent that these decisions do not unequivocally support a duty of compensation. The award was indeed subsequently annulled, among others, due to a manifest error of law in relation to Article 27(b): as the Annulment Committee explained, this provision did not impose a duty of compensation; it was simply a without prejudice clause.Footnote 28
The Tribunals in Enron and Sempra both noted that Article 27(b) was vague and, in line with the Commentary to this provision, that whether compensation was due in these circumstances was a matter that must be decided by the parties. Both Tribunals added that absent agreement between the parties ‘this determination is to be made by the Tribunal to which the dispute has been submitted.’Footnote 29 The reasoning is not entirely clear, but it is plausible to read these awards as endorsing a duty of compensation in cases of necessity: compensation is either agreed between the parties or decided by the tribunal. The tribunals do not clarify, however, what is the source of their power to determine (and impose the payment of) compensation. Neither tribunal ultimately went on to make the determination since both rejected the Argentine defence.
The Tribunal in South American Silver v Bolivia was even briefer on this point. In a dispute concerning the payment of compensation for an expropriation, the Tribunal stated:
It is clear that Bolivia’s state-of-necessity [sic] defense was not designed to excuse the non-payment of compensation for the expropriation, nor could it, since the invocation of this defense does not preclude the payment of compensation by the State for the damage effectively resulting from acts attributable to it.Footnote 30
There is some ambiguity in this statement. The passage uses the language of Article 27(b) (compensation is not precluded), seemingly going no further than this provision: the successful invocation of a defence is without prejudice to the question of compensation. But the passage can also be interpreted as going beyond Article 27(b); supporting a duty of compensation in (at least some) cases of necessity. The Tribunal first states that the plea of necessity cannot apply to deny compensation for expropriation: this is not what the plea was ‘designed’ to do. The plea of necessity cannot be invoked when the relevant primary rule excludes its invocation, implicitly or explicitly.Footnote 31 Arguably, the situation of necessity is already catered to by the primary rule on expropriation, as it is necessity that justifies the taking of the property. Necessity cannot do double work, as it were, it cannot justify the taking and justify the denial of compensation. The plea cannot thus be invoked as Bolivia has done. So far, so plausible. The tribunal then appears to go further. Even if necessity were applicable to this situation, it says, the plea could not deny compensation for expropriation. In short, a successful invocation of the plea would still involve an obligation to pay compensation for the expropriation. If this were the correct interpretation of the Tribunal’s statement, it would be no more than an assertion that compensation was due even in cases of successful invocation of the plea – at least, in some of these cases (expropriation). The Tribunal provides no evidence of a positive law source, let alone of customary law, for this duty in the award.
2.2 Deductions
Other tribunals have obviated the need to provide a positive law basis to the duty of compensation, by grafting this obligation to the customary rule of necessity itself. They have done this by interpreting the rule on compensation as including a duty of compensation as one of its requirements. If the plea of necessity is recognised in customary law, and the duty of compensation is inherent in the plea, then it follows that the duty of compensation is also part of customary law. This is the case of the awards of the Tribunal and annulment committee in EDF v Argentina.
The EDF Tribunal stated that to succeed in its invocation, Argentina had to demonstrate:
three key elements of ILC Articles 25 and 27: (i) that the wrongful act was the only way to safeguard Argentina’s essential interest under Article 25 (1)(a); (ii) that Respondent did not contribute to the situation of necessity; and (iii) that Respondent did not return to the pre-necessity status quo when possible, or compensate Claimants for damage suffered as a result of the relevant measures.Footnote 32
It further explained that even if Argentina were successful in invoking the plea, this did not ‘per se preclude payment of compensation to the injured investor for any damage suffered as a result of the necessity measures enacted by the State.’Footnote 33 Having contributed to the situation of necessity, and having failed to re-establish the status quo, Argentina’s defence failed.Footnote 34
In annulment proceedings, Argentina claimed that the Tribunal had ‘invented’ this additional element.Footnote 35 The Annulment Committee, however, endorsed the Tribunal’s finding. According to the Committee, this requirement had not been invented by the Tribunal, but was rather reflective of ‘what is inherent in the very concept of necessity’.Footnote 36 By this, the Committee meant its temporary character: ‘If a departure from a legal obligation can be justified by a state of necessity, it can be justified for only so long as that state of necessity exists’Footnote 37 – an argument also adduced by the Tribunal in CMS. In short, since the plea of necessity is only temporary, therefore, compensation is due.
Panos Merkouris has argued that deductive methods may be applied to the interpretation of customary rules the existence of which has already been established.Footnote 38 The Committee’s approach could be viewed in this light, as proceeding either teleologically or by necessary implication to deduce the existence of a duty of compensation from the (established) rule of necessity. Leaving aside the doctrinal question whether interpretation of customary rules differs from their identification, it seems a step too far to accept that additional obligations may be inferred, by deduction, from established customary rules: especially where the practice supporting that rule does not provide evidence in respect of that specific obligation (as will be seen below). At any rate, even if this method were found to be in line with the generally accepted method for customary law identification, the conclusion the Committee reaches is a non sequitur. The temporariness of the plea (which is only a contingent feature)Footnote 39 concerns compliance with the obligation. In principle, defences do not strike down the rule at issue, which remains in force throughout the period that the defence subsists.Footnote 40 But the State’s obligation to comply with the rule is set aside throughout the period in which the facts giving rise to the defence continue to exist. Once these facts come to an end, the obligation is ‘restored’, as it were, and the State must resume compliance with it.Footnote 41 If it does not, then the State will be pro tanto responsible for the violation of the obligation from the moment when the defence ceased.Footnote 42 Take the following example. Due to a situation of necessity which arose on 10 February 2020, State A was unable to comply with its treaty obligation to deliver 10 tonnes of rice on the first day of every month to State B. Say, then, that the situation of necessity ended on 15 July 2020. State A would be required on 1 August to deliver 10 tonnes of rice to State B. If it failed to deliver those on 1 August, then State A would be responsible as from 1 August for the failure to comply with its obligation to State B. The defence is temporary in that it can only justify State A’s failure to comply with its obligation for the five months of March, April, May, June, and July. Indeed, throughout this time, State A’s obligation to deliver is in abeyance due to the situation of necessity. But the plea’s temporary character, which concerns the return to compliance after the defence has ended, has nothing to do with the question of compensation for material loss, which concerns the allocation of the losses generated during the situation of necessity (ie, during the period when the State was justified in not complying with the obligation). In the example above, a duty of compensation would relate to the loss caused to State B as a result of A’s failure to deliver the required amount of rice for the five months between the start (on 10 February) and the end (15 July) of the situation of necessity.
The reasoning of the Tribunal – that compensation is due because of the inherent character of the plea – confuses, or fuses, resumption of compliance with compensation. In short, it confuses, or fuses, the provisions in ARSIWA Article 27(a) and Article 27(b). Resumption of compliance with the underlying obligation at the end of the situation of necessity and compensation for material loss during the period of necessity relate to two different obligations. The former, resumption of compliance, is just a consequence of the underlying obligation no longer being in abeyance. The latter is a different – new – obligation of the State invoking necessity. This is an obligation that arises as a result of the loss caused by the act of necessity: in our example above, the loss caused by the failure to deliver the rice. To say that the invoking State must resume compliance with the underlying obligation – which is the consequence of the defence’s temporary character – has no bearing on whether the invoking State is now burdened by a new obligation to pay compensation for losses caused during the defence.
This conclusion does not change by saying that payment of compensation can only occur after the necessity has ended. We still need to find a basis in positive law for this obligation to pay compensation. The former is a question of the performance of the duty (when it falls due), the latter one of its existence. The underlying obligation cannot – itself – sustain this duty. In the example above, the obligation is to deliver rice: it is not ‘to deliver rice or pay compensation’. It is also not a duty derived from responsibility – namely, one of forms of reparation – because there has been no wrongful act: the failure to deliver rice, in our example, was justified by necessity. What is, then, the positive law source of this obligation to pay compensation? After all, loss occasioned by a permitted or lawful act is not typically one that requires compensation. In other words, liability for the injurious consequences of lawful acts is not the norm. Such liability is exceptional, and needs to be grounded on a positive law rule.Footnote 43
No positive law source – customary or otherwise – was identified by either the Tribunal or the Annulment Committee to ground the duty of compensation. The Tribunal noted and set aside the question of the customary character of the defence of necessity, arguing that the parties agreed on the application of ARSIWA Article 25.Footnote 44 Notably, the Tribunal does not mention whether the parties agreed on the application of Article 27(b) and, more importantly, whether they agreed on the question of compensation. The Committee, in turn, only referred to the correspondence in the Caroline incident which, indeed, supports the proposition that necessity is only temporary but, as will be seen, does not support a duty to pay compensation for material loss caused by an act adopted in circumstances of necessity.
2.3 Denials
Other tribunals have denied the existence of a duty of compensation in cases of necessity. This is clearly the case of the LG&E Tribunal. In its Liability decision, the Tribunal noted that Article 27(b) was a without prejudice clause, and that it did not ‘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.’Footnote 45 Whether compensation was due, said the Tribunal, depended on the interpretation of the defence in question. The Tribunal focused on Article XI of the BIT and found that no compensation was due since this provision ‘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.’Footnote 46 The Tribunal’s decision is grounded on Article XI of the BIT, but to the extent that the Tribunal equated Article XI to the customary plea of necessity its conclusion can be extended to the latter as well. In line with this reasoning, the LG&E Tribunal – the only one to have accepted Argentina’s plea – eventually excluded compensation for the period covered by the necessity defence.Footnote 47 The reasoning is circular: the circumstance precluding wrongfulness of necessity does not attract a duty of compensation because it is a circumstance precluding wrongfulness. But the whole point is whether compensation should be due even if something is a circumstance precluding wrongfulness. In short, the Tribunal chose where to allocate the loss (the investor) but failed to provide a reasoned argument or any evidence of a source in positive law for this conclusion.
The award in Urbaser could be read as a denial of the duty of compensation, but it is a more equivocal precedent. The Tribunal found that Argentina’s necessity plea was satisfied,Footnote 48 but it denied the payment of damages to the claimants. However, the ratio of this decision seems to have rested on the fact that the failure of the investment was primarily attributable to claimants themselves,Footnote 49 and not on the non-existence of a duty of compensation in cases of necessity. Indeed, the Tribunal made no comments on the existence of this, despite the fact that the parties presented arguments in this regard.Footnote 50
2.4 Agnosticism
Finally, other tribunals have taken a more agnostic stance. The Annulment Committee in CMS simply noted that Article 27(b) is a without prejudice clause and not a stipulation, and that it did not attempt to ‘specify in which circumstances compensation could be due, notwithstanding the state of necessity.’Footnote 51 The Annulment Committee in Sempra, in addressing the difference between state of necessity and the BIT’s non-precluded-measures clause, noted that no compensation was due in the latter case but that the question of compensation ‘was not precluded’ in the former. The Tribunal thus acknowledges the possibility that compensation could be due, without taking sides in the debate.Footnote 52
3 Doing the Homework: What Evidence for a Customary Duty of Compensation?
Tribunals’ divided opinions on this point are not unique. Far from it. Scholars are equally divided on the existence of a duty of compensation in cases of necessity. Thus, some scholars have argued that international law recognises an obligation to provide compensation in these cases,Footnote 53 whereas others have expressed doubts as to the existence of this duty.Footnote 54 Many of these scholars, however, regardless of their views on the existence, in positive law, of this duty, agree that compensation would be fair in such circumstances.Footnote 55
In the case law and the literature on this topic, arguments as to the existence of a customary duty of compensation usually rely, as evidence, on the ILC’s drafting of, and Commentary to, Article 27(b), and the case law.Footnote 56 As will be seen in the next two sections, however, the evidence in support of this duty is far from clear. Regardless of how one interprets the ILC’s work on, and the Commentary to, Article 27(b), only a handful of States commenting on the draft expressly supported a duty of compensation generally, or in cases of necessity specifically. Indeed, the evidence of practice and opinio juris in favour of this duty is scant and vague (Section 3.1), and the precedents relied upon by tribunals and scholars alike to evidence the existence of the duty of compensation are equivocal at best (Section 3.2).
3.1 Missing Practice
According to Article 38(1)(b) of the ICJ Statute, customary international law is evidenced by the existence of a ‘general practice accepted as law’.Footnote 57 The practice must be general in the sense that it is ‘sufficiently widespread and representative, as well as consistent.’Footnote 58 It is not necessary for all States in the international community to engage in the practice, nor is it needed for the practice to be absolutely uniform. The threshold required for the identification of any given rule of customary law may vary by reference to the context.Footnote 59 Thus, it is arguable that in the case of very exceptional circumstances, like those that trigger the plea of necessity, the threshold is lower as there will be fewer opportunities for States to engage in the relevant practice. This practice must be accompanied by opinio juris, namely, evidence that States engage in the relevant practice out of a sense of legal obligation (or legal entitlement).Footnote 60 As will be seen, there is almost no support in the practice and opinio juris of States for the duty of compensation in cases of necessity.
It seems fair to read the ILC’s work on Article 27(b) as generally supportive of a duty of compensation in cases of necessity.Footnote 61 Special Rapporteurs Ago and Crawford supported it, as well as several members of the Commission. None of them wished to take too exacting a position on this matter, however, because of the scarcity of practice and of the variety of cases in which this duty might arise.Footnote 62 In this regard, the ILC’s work mirrors that of scholars – a strong sense that it would be fair for compensation to be due.
Nevertheless, States’ views on this duty have been much more mixed and, often, negative. Only three of the States commenting on the ILC’s drafts, Germany,Footnote 63 RussiaFootnote 64 and the UK,Footnote 65 explicitly accepted the possibility of the duty arising in situations of necessity. Other States like Denmark (speaking on behalf of the Nordic countries),Footnote 66 the NetherlandsFootnote 67 and PolandFootnote 68 generally endorsed Article 27 (or its predecessor). Their statements, however, fall short of endorsing the actual existence of a duty of compensation following invocations of necessity.Footnote 69 Austria did not outrightly reject the possibility that a duty of compensation may arise in situations of necessity, but it warned that the provision required a more specific formulation since ‘it would otherwise lead to the danger of possibly undercutting the effect of circumstances precluding wrongfulness.’Footnote 70 Other States were more negative. France rejected altogether the idea that compensation may arise in the event of a successful invocation of a defence,Footnote 71 and Chile rejected it in respect of a state of necessity in particular.Footnote 72
Furthermore, in the context of judicial or arbitral proceedings, HungaryFootnote 73 and SlovakiaFootnote 74 have acknowledged the existence of the duty, while ArgentinaFootnote 75 and ZimbabweFootnote 76 have rejected it. Bolivia also addressed this duty in arbitral proceedings against a foreign investor, though its position is not entirely clear. It offered compensation to the investor for the taking of property, which it justified under the plea of necessity and argued that in this way it respected the ‘hypothetical interests’ of the United Kingdom (the other party to the BIT) and of the international community as a whole.Footnote 77 However, the case involved an expropriation and the primary rules on expropriation themselves require compensation.
Overall, as this review shows, just over a dozen States (out of nearly 200) have expressed views on the existence of a duty of compensation in cases of necessity. As Fernando Bordin has noted, few (if any) customary rules ‘even those long viewed as established, can survive the brutal scrutiny of the magnifying glass’.Footnote 78 The two-element approach to the identification of customary law, endorsed by the ILC Conclusions, must be applied with flexibility.Footnote 79 Even with this caveat in mind, however, it seems clear that the practice available at present is insufficient and is, moreover, is inconsistent as the broad range of views shows. This makes it difficult to draw any conclusions as to the existence of a customary duty of compensation. While there seems to be a trend towards favouring the recognition of this duty in the case law and scholarship, to date, such trend has not been followed by States in their practice: the evidence available at present falls far short of the requirement of generality necessary to identify a rule of customary law.
3.2 Equivocal Precedents
Whether international tribunals can make or develop international law, in addition to just applying the law to specific facts, is a persistent and thorny question in international law.Footnote 80 It is also a question which eschews simple answers.Footnote 81 One thing, however, is clear: as a matter of the formal sources of international law, judicial decisions are, as stated in Article 38(1)(d) of the ICJ Statute, subsidiary means for the determination of rules of law. International courts and tribunals do not make law, as recently reaffirmed by the ILC work on the Identification of Customary Law.Footnote 82 But this is not to say that courts and tribunals cannot act as agents of legal development.Footnote 83 As observed by Rosalyn Higgins, former President of the ICJ, ‘the very determination of specific disputes, and the provision of specific advice, does develop international law’.Footnote 84 To be sure, there are ‘decisions and decisions’, to paraphrase Jan Paulsson.Footnote 85 Some decisions will exert an influence in legal development and, again in Paulsson’s words, ‘become ever brighter beacons’, while others ‘flicker and die near-instant deaths’.Footnote 86 Judicial development of international law relies on the interactions with the decisions by other actors in this process: whether the decision is endorsed by States in their practice,Footnote 87 or it is followed by other tribunals. In turn, these interactions depend on a variety of factors such as whether the field is receptive to judicial development;Footnote 88 and whether the decision showcases certain attributes (including the authority of the tribunal, the composition of the tribunal, the context of the decisions, the size of the majority, and the quality of the reasoning).Footnote 89
It thus seems worth examining the case law relied upon by the investment awards discussed earlier. As will be seen, the precedents invoked are, at best, equivocal on this point. In most of these cases, the existence of compensation can be explained on other, more plausible, legal bases. As such, they cannot be relied upon as evidence of the existence of a duty of compensation. It is not surprising, then, that none of these cases has become a ‘bright beacon’ on this point, as evidenced by how few States have endorsed the existence of the duty of compensation. The list below is not intended to be exhaustive, but merely to assess those cases that are usually cited by investment tribunals and by scholars in their analyses of the duty of compensation.
3.2.1 The Neptune (1797)
During the Napoleonic wars, The Neptune,Footnote 90 an American vessel on voyage from Charleston to Bordeaux, carrying rice among other things, was stopped and seized by the British navy in April 1795.Footnote 91 The Admiralty Court of London ordered the sale of the Neptune’s cargo to the British Government at the invoice price plus 10% profit. The owner claimed that it was owed the commercial price at which the articles would have sold in Bordeaux.Footnote 92 Before the Commission established under the Jay Treaty,Footnote 93 the British rejected the claim arguing that the seizure was lawful as the merchandise constituted contraband and, in the alternative, the seizure was a lawful preemptive purchase to provide for a threatened famine.Footnote 94 On this latter claim, agents for the British Crown asserted that the ‘capture was made under such circumstances of distress as rendered the act lawful against the neutral’.Footnote 95
The British claim of pre-emptive purchase was understood by the Commissioners as a plea of necessity.Footnote 96 Deciding by majority, the Commissioners rejected the British argumentFootnote 97 as the conditions of the plea were not met in fact.Footnote 98 Nevertheless, in his consideration of this plea, Commissioner Pinkney endorsed a duty of compensation in the following cases: ‘Great Britain might be able to say to neutrals “You shall sell to us”, but it does not follow that she could also say “You shall sell to us upon worse terms than you would have procured elsewhere in the lawful prosecution of your commerce”’.Footnote 99
While the Neptune is often cited as evidence of the existence of a duty of compensation,Footnote 100 three important factors detract from the weight and precedential value of this case. First, the applicable law by the Commission included ‘justice, equity and the law of nations’,Footnote 101 such that very little can be inferred from this case as to the positive law between States at the time.Footnote 102 Second, only one of the five Commissioners upheld the existence of this duty. Lastly, the Commission rejected the plea of necessity so Pinkney’s statement was only obiter.
3.2.2 The Caroline Incident (1837)
In 1837, Canadian rebels were attempting to declare, and establish, an independent Republic of Canada in the British colony of Upper Canada (now Ontario). The US steamer the Caroline supplied Canadian rebels and their US recruits on Navy island, within Ontario, from the US shore of the Niagara river. On 29 December 1837, British forces entered US territory and apprehended and destroyed the Caroline, which was moored off Fort Schlosser in the American bank of the river.Footnote 103 The incident led to a protracted diplomatic correspondence between the two States, in which the notions of self-preservation, self-defence, and necessity were invoked. And it is indeed in relation to self-defence that the incident is renowned: the so-called ‘Webster formula’ of self-defence, still invoked today,Footnote 104 was articulated by the US Secretary of State, Daniel Webster, in a letter to his British counterpart.Footnote 105
The US demanded redress for Britain’s wrong, including compensation for the value of the destroyed property, which it estimated at $5000 US dollars.Footnote 106 Britain disputed the illegality of its actions claiming to have acted in self-preservation and self-defence,Footnote 107 thus rejecting the claim for reparation. The parties eventually settled the dispute with a (feeble) apology from the UK, which nevertheless insisted on the permissibility of its actions. The issue of compensation did not ultimately play a role in the settlement of the dispute. But there is here an interesting twist. The letter sent by Lord Ashburton, on behalf of the UK, to Webster included the following paragraph:
If the Boat which was destroyed could by any fair construction of the case have been considered as the private property of a citizen bona fide and innocently employed by him as a passage vessel, compensation for its loss might perhaps have been admitted, but it is notorious that it was part and parcel of the armament of the insurgent force, and I have reason to know, that the property in part, if not wholly, was in British subjects. Under such circumstances no question of compensation could be entertained or expected.Footnote 108
A copy of this letter, with these words crossed over but still legible, was kept in the Public Record Office, at the Foreign Office in London.Footnote 109 This original letter was subsequently withdrawn at the request of Lord Ashburton and replaced with another letter, amended by agreement of the parties. This second letter did not include the paragraph just quoted on compensation. As Lord Ashburton explained to Lord Aberdeen in this connection, on subsequent consideration he had thought it ‘expedient to suppress’ this paragraph from his original note.Footnote 110
The Caroline incident was referred to by the Annulment Committee in EDF, in considering the duty of compensation. But it is doubtful that this case actually supports a duty of compensation. First, Britain did not accept the principle that compensation was payable to the owner of The Caroline: even if the latter had been innocent, Ashburton only says that compensation ‘might perhaps have been admitted’. Second, such a statement was not communicated to the US and compensation was not actually paid to the owner of the steamer. Third, the legal principle at issue in this dispute remains contested: while some argue that the parties relied on the plea of necessity,Footnote 111 others have argued that the legal principle at issue was that of self-defence.Footnote 112
3.2.3 Orr and Laubenheimer (1900)
Orr and Laubenheimer, two US citizens, were engaged in the banana trade, importing bananas to the United States from the Nicaraguan port of Bluefields, on the mouth of the Rama River.Footnote 113 Bananas grew in plantations along the banks of the river and its tributaries, and were transported to the port by tugboats. In 1894, in the course of suppressing an insurrection in Bluefield, a Nicaraguan general seized two of Orr and Laubenheimer’s tugboats to transport troops down the Rama river to Bluefields. Orr and Laubenheimer subsequently claimed indemnity for damages sustained as a result of Nicaragua’s alleged seizure and detention of the tugboats, and the matter was submitted to arbitration by agreement of the governments of the US and Nicaragua. In its decision, the Arbitrator stated that the ‘rights incident to a state of war … justify the use by any Government, in an emergency, of any private property found available.’Footnote 114 It went on: ‘Full compensation, however, for all damage suffered by private parties must afterwards be made. But the obligation rests upon every party damaged to do all in his power to reduce his losses to a minimum. That is the law the world over…’Footnote 115
This award was referred to by the Tribunal in CMS in support of the proposition that acts of necessity generate a duty of compensation for material loss. But that is reading too much into this short decision, for three reasons. First, as per the parties’ agreement, Nicaragua ‘waive[d] its denial of liability … and agree[d] that said arbitrator may award such sum as he believe[d] said Orr and Laubenheimer … to be justly entitled to’.Footnote 116 Second, as a result of the waiver on the question of liability, the arbitrator did not need to, and did not, apply international law to the dispute: its task was to decide the amount of just compensation due. Finally, if there is a legal basis for Nicaragua’s obligation to compensate, this is the right of angary. Pursuant to this right, as explained by Oppenheim, States engaged in hostilities are entitled to use the property of neutrals ‘provided the articles concerned are serviceable to military ends and wants’, and so long as, in every case, ‘the neutral owner [is] fully indemnified.’Footnote 117
3.2.4 Company General of the Orinoco (1905)
The case involved the rescission of concession contracts between Venezuela and a French Company, signed in the late 1880s. The contract was for the exploitation of vegetable and mineral resources on territory that Venezuela believed to be under its sovereignty.Footnote 118 Following protests by Colombia,Footnote 119 Venezuela rescinded the contract with the French company. Venezuela subsequently found that most of the territory in the concession was under the sovereignty of Colombia.Footnote 120 The company claimed compensation from Venezuela, and the matter came before the Franco-Venezuelan Mixed Commission. Umpire Plumley upheld the rescission but ordered the payment of compensation to the company. In his reasoning, he framed the question as one of necessity.Footnote 121 In his view:
As the Government of Venezuela, whose duty of self-preservation rose superior to any question of contract, it had the power to abrogate the contract in whole or in part. It exercised that power and canceled [sic] the provision of unrestricted assignment. It considered the peril superior to the obligation and substituted therefor [sic] the duty of compensation.Footnote 122
The peril, as the Umpire explained, came from multiple sources. It came from the Colombian government, which claimed sovereignty over much of the area under concession, and which threatened force to recover the territory, but also from the local population and businessmen who were dissatisfied by the monopoly granted to the company and who, with the support of the local government, revolted sometimes violently.Footnote 123
As with the Neptune, there are a number of factors which may diminish the weight and precedential value of this decision in respect of the duty of compensation.Footnote 124 First, there is uncertainty as to the law actually applied by the Umpire to decide the case.Footnote 125 The Umpire was competent to take into account, in reaching his decision as to the need for compensation, ‘the ethical precepts of international law, equity and good conscience’.Footnote 126 On the specifics of the case, the Umpire held that ‘if there were aught of wrong towards the Company General of the Orinoco done or permitted by the respondent Government’, then he may award ‘damages if justice and equity so permit and so require.’Footnote 127 Ultimately, in his view, there was ‘no inequity’ in apportioning some of the loss caused to the company by the rescission on the Government.Footnote 128 The decision was thus apparently based on equitable considerations.
Second, even if by application of international law, the award of compensation can be explained on other legal bases. It could be explained as a case of compensation for wrongful conduct: the compensation paid was not for the damage caused by the rescission of the contract itself, but rather for the breach of the contract before its termination.Footnote 129 Or it can be explained as involving the application ‘of the rule that compensation must be paid when foreign-owned property is expropriated in the public interest.’Footnote 130 This explanation is more convincing than the former, as it can account for the necessity-like reasoning of the Umpire.Footnote 131 Being able to account for this reasoning of the Umpire is particularly important for two reasons: first, because it is this aspect of the reasoning that scholars seize upon to provide support for the existence of a duty of compensation; and, second, because there are considerable doubts as to the recognition of a defence of necessity in the positive law of the time.Footnote 132 As Sarah Heathcote explains, there exist in international law primary rules ‘in the image of necessity’: these are substantive rules of international law that cater to a specific (factual) situation of necessity. A State’s right to expropriate property is precisely one of these rules, as it can only be exercised in situations of public necessity. As the Umpire noted in the award, a situation of public necessity existed in Venezuela at the time, as a result of the external (from Colombia) and internal (local population) threats that the country was facing.Footnote 133 In such circumstances, the rescission of the concession was an expropriation of foreign-owned property due to necessity. The payment of compensation in this case was, therefore, a matter of the primary rule in question (expropriation) rather than one of the applications of the plea of necessity under the law of State responsibility.
3.2.5 Properties of Bulgarian Minorities in Greece (1926)
Following the exchange of minorities provisions in the post-World War I settlements, foreign refugees of Greek origin were transferred from Turkey to Greece. In order to house them, the Greek Government forced Bulgarian minorities to move out of their homes in Greece. The matter was considered by a League of Nations’ Commission of Enquiry.Footnote 134 By the time the Commission issued its report, the Bulgarian minorities had left Greece and the Greek refugees were already settled in the homes. The Commission allowed that the take-over of Bulgarian property by Greece had been the result of a situation of what it termed ‘force majeure’. Indeed, according to the report, to remove the Greek refugees to allow the return of the owners would have been impossible in these circumstances, as well as undesirable.Footnote 135 Nevertheless, the Commission argued that it could not be expected that the Bulgarian minorities would simply renounce their right to the homes, so it was just that they receive compensation for the value of their property.Footnote 136
The argument for a duty of compensation is more plausible in this instance, but it is not clear cut. The situation certainly seems to be one that could fit within the plea of necessity (as currently formulated): in order to protect one interest (housing Greek refugees), Greece infringed the rights of others (Bulgarian minorities). But here too there are a number of factors which may weaken this argument. To begin with, the Commission spoke of force majeure and not of necessity. Much of contemporary doctrine has tended to assume that force majeure and ‘necessity’ were used interchangeably at the time, but this view requires some nuance. The concepts are (and were) indeed different, and I have argued elsewhere that reference was made to ‘force majeure’ during this period to address situations of necessity because international law did not recognise a rule of necessity at the time.Footnote 137 At any rate, even if this had been a case decided on a plea of necessity (at least in substance, if not expressly), it does not seem that the requirements of the plea were met. As Heathcote has argued, this was not a case of protecting a superior interest as against an inferior one: in this case, the interests were equal for ‘why should Bulgarian minorities, who … had only been in Greece for a decade or so … be moved out of their homes to house refugees of Greek origins – the Smirna “Greeks” [who] had been in Turkey for centuries?’Footnote 138
Once more, the Commission’s decision is better explained on other legal bases: either as a situation of reparation for wrongful conduct or, as in the Orinoco Company case, as a case of expropriation for public necessity.Footnote 139 In any event, there are doubts as to whether the basis of the Commission’s recommendation was premised on law at all. While its mandate was to ‘establish the facts enabling the responsibility to be fixed, and supply the necessary material for the determination of any indemnities or reparation which may be considered appropriate’, the Council did not specify the basis (legal or otherwise) upon which such ‘responsibility’ and ‘indemnities’ ought to be decided. Perhaps for this reason, Michael Akehurst has interpreted the Commission’s finding as a political compromise.Footnote 140
3.2.6 Gabčíkovo-Nagymaros (1997)
The dispute between Slovakia and Hungary concerned the unilateral termination of the Treaty of 1977, which envisaged a joint project between the two States for the construction of a system of locks in the Danube, by Hungary. Among other things, Hungary invoked the plea of necessity to justify its unilateral termination of the Treaty. At the time of unilateral termination, both parties had commenced the works already. Slovakia had completed a section of works in one of the sectors, so the question as to whether any compensation was due to Slovakia as a result of the unilateral termination emerged. The ICJ rejected Hungary’s plea of necessity, but it noted that ‘Hungary [had] pointed out’ that a duty to compensate Slovakia for the works undertaken existed.Footnote 141 The Court’s statement was obiter and is not a direct endorsement by the Court of the duty: it is merely a description of Hungary’s position. As such, not much weight can be given to the judgment itself.
More pertinent are, at any rate, the statements made by Hungary during the proceedings. Hungary raised the point multiple times during the oral phase of the proceedings.Footnote 142 In very clear terms, Hungary stated that
Hungary recognizes that in modern international law the plea of necessity can only be admitted on a limited and strictly defined basis. ‘Necessity’ allows the sovereign State to commit what would otherwise be an unlawful act while avoiding international responsibility – though not the requirement to make appropriate compensation.
Slovakia’s own views on the matter were less assertive. It recognised that the duty of compensation was required as a matter of fairnessFootnote 143 and common sense, but it warned of the risk of States ‘buying’ their ‘way out of [their] breaches of its international obligations’.Footnote 144
4 Assessment
Investment tribunals have tackled the question of compensation in cases of necessity in numerous cases. Their conclusions on the point (almost always in obiter) are as varied as the reasoning behind them. A common thread among them is that they have, for the most part, failed to assess the positive law basis for the existence (or non-existence) of the duty of compensation. Only a handful of decisions name-check some precedents and cases, but none of them in any way refer to State practice or opinio juris.
To be sure, the question of compensation in cases of necessity is a philosophically and theoretically difficult one, having troubled legal scholars and theorists for many centuries.Footnote 145 It is a question that elicits intuitive and often strongly held opinions: it would be unfair for the affected party to bear the burden of the protection of others’ interests. In short, it would be unfair to let the loss lie where it falls. And yet, the action that causes the loss is a permitted one, it is lawful behaviour, and under normal circumstances, we would not expect those acting lawfully to compensate losses caused by their actions. In the absence of a wrong, losses do lie where they fall: herein lies the dilemma at the heart of the duty of compensation. And yet, necessitated acts seem different from other lawful acts that cause loss. Indeed, necessitated acts have a baggage that other lawful acts do not: they evoke moral hazard and, in the history of international law, they evoke abusive behaviour by powerful States.
This baggage explains the intuitive perceptions of unfairness at the allocation of loss onto the affected party, and the support for a re-distribution of the loss onto the agent. It may also explain, at least partly, why investment tribunals have been sympathetic to the idea that States invoking necessity owe a duty of compensation to the affected parties. But this is no justification for these tribunals’ omission to engage with the methods of law ascertainment: aside from the fairness and justness of the duty of compensation, is there evidence that this is required by positive law; is there evidence, in particular, of practice and opinio juris about the existence of this duty? As shown, there is not only limited practice and opinio juris on this duty, but the few precedents cited in investment decisions do not support, nor do they provide evidence of, the existence of a duty of compensation. The nobility of the sentiment is no substitute for the absence of positive law on the existence of this duty. Indeed, in asserting or deducing the duty of compensation in this manner, investment tribunals are closer to deciding the matter ex aequo et bono, for which they would need specific consent by the parties, than by application of the rules of international law, as they are mandated to do.
1 Introduction
It is perhaps trite to say that the principle of ‘full reparation’, enunciated by the Permanent Court of International Justice (PCIJ) in the Chorzów Factory case, has been widely recognised as the customary rule governing the reparation of internationally wrongful acts.Footnote 1 According to that judgment, where restitution is unavailable or insufficient, customary law requires the payment of compensation in the form of ‘a sum corresponding to the value which a restitution in kind would bear [and] the award, if need be, of damages for loss sustained which would not be covered by restitution in kind or payment in place of it’.Footnote 2 Whilst the rights of private entities are ‘on a different plane’ to those belonging to States, the damage suffered by an individual affords ‘a convenient scale for the calculation of the reparation due to the State’,Footnote 3 so the extent of the individual injury affords the metric for the calculation of damages at the inter-State level.
The International Law Commission (ILC) took Chorzów Factory as the basis for the elaboration of the rules governing the consequences of wrongful acts in the 2001 Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA).Footnote 4 In the Commentary to Article 31, the ILC explained that ‘[t]he responsible State’s obligation to make full reparation relates to the “injury caused by the internationally wrongful act”’.Footnote 5 If restitution is unavailable or insufficient, ‘[t]he role of compensation is to fill in any gaps so as to ensure full reparation for damage suffered.’Footnote 6 Article 36 ARSIWA then states that ‘[t]he State responsible for an internationally wrongful act is under an obligation to compensate for the damage caused thereby, insofar as such damage is not made good by restitution’, damage being understood as ‘any financially assessable damage including loss of profits’.Footnote 7 Given that most Bilateral Investment Treaties (BITs) are silent as to the remedies applicable in case of their violation, investment tribunals have referred to Article 36 ARSIWA as reflecting the applicable standard of compensation,Footnote 8 and have recognised Chorzów Factory as an ‘authoritative description’ of customary law on the subject.Footnote 9
In line with the Chorzów Factory standard, the determination of compensation seems to operate within three governing parameters.Footnote 10 First, the identification of the extent of the damage (material or moral) as a question of fact.Footnote 11 Second, the establishment of a sufficiently direct and certain causal nexus between the damage and the internationally wrongful act.Footnote 12 Third, the quantification, in monetary terms, of any ‘financially assessable’ damage through the application of an appropriate valuation methodology.Footnote 13 The final amount of compensation will vary depending on permutations of these factors. Conversely, factual or legal considerations beyond these parameters are generally deemed irrelevant to quantum.Footnote 14
Within this conceptual framework, strongly influenced by private-law analogies from municipal tort law,Footnote 15 compensation has acquired a strong, ‘damage-centric’ focus, in the sense that it depends primarily – if not exclusively – on the demonstration of a financially assessable damage and a ‘sufficiently direct and certain causal nexus’ between the damage and the wrongful act.Footnote 16 Thus, at the final stages of its codification efforts the ILC decided to omit from the text of ARSIWA any provision allowing for extraneous factors to be taken into account in the determination of compensation beyond damage and causality, such as aggravating or mitigating circumstances, the gravity of the act, or limitations relating to proportionality.Footnote 17
A perusal of international jurisprudence, however, paints quite a different picture. In fact, early arbitral commissions,Footnote 18 the International Court of Justice (ICJ),Footnote 19 the Iran-US Claims Tribunal,Footnote 20 ad hoc inter-State tribunals,Footnote 21 as well as regional human rights courts,Footnote 22 have referred to equity (or, interchangeably, ‘equitable considerations’Footnote 23) as a normative proposition capable of affecting the determination of damages in ways not expressly contemplated in Chorzów Factory. Investment tribunals have followed a similar path, invoking equitable considerations for the determination of compensation due for violations of BIT provisions.Footnote 24 What this means in practice is unclear: despite the frequent invocation of equity for the purposes of determining compensation, international courts and tribunals have made little effort to explain the legal basis of these considerations or their underlying methodology.
Arguably, the integration of equitable considerations in quantum analysis presents some significant advantages. From a procedural point of view, it allows for some flexibility in the fact-gathering process and enables the tribunal to award compensation even when objective circumstances preclude the injured party from producing sufficient evidence to substantiate its loss. The ICJ, for example, has invoked equitable considerations for the determination of compensation where the evidence was insufficient to enable a precise quantification, for ‘it would be a perversion of fundamental principles of justice to deny all relief to the injured person, and thereby relieve the wrongdoer from making any amend for his acts.’Footnote 25 From a substantive point of view, the rigid or mechanical application of customary rules governing compensation might also lead to a juridical outcome that places too strong an emphasis on the extent of the injury caused by the wrongful act, in a manner disconnected from the context in which the injury arose, the nature of the unlawful act, or the respective interests and conduct of the parties. Thus, the application of equity to compensation enables the tribunal to ‘infuse’ elements of reasonableness and ‘individualized justice’ in its reasoning,Footnote 26 and arrive at a balanced outcome that accommodates the interests of both parties.
Be that as it may, equitable considerations may also give rise to complications in practice. In fact, an unprincipled application of equity to compensation may have serious repercussions for the legitimacy of the dispute settlement procedure. It might also affect the procedural rights of the parties and, ultimately, undermine the integrity of the decision itself. This complexity is exemplified in the International Centre for Settlement of Investment Disputes (ICSID) framework where ad hoc annulment committees have stated that if a tribunal misapplies the legal rules in favour of a settlement based on ‘general equity’, the award might be subject to annulment for manifest excess of power or a failure to state adequate reasons, within the meaning of Article 52(1)(b) and (e) of the Washington Convention.Footnote 27
Outside the ICSID framework, the application of equitable considerations may also give rise to challenges to recognition and enforcement of arbitral awards. For instance, States have challenged the validity of arbitral awards relying upon equitable considerations before domestic courts, portraying them as attempts at awarding punitive damages in a manner contrary to international law.Footnote 28 Conversely, some municipal courts have attempted at re-opening certain arbitral awards, especially with regard to questions of compensation, invoking ‘equity’, ‘fairness’ or ‘proportionality’ as the legal basis for their judicial review.Footnote 29 This development may have serious implications for the finality of arbitral awards: a broadly-construed conception of ‘equity’ for quantum purposes may in fact be used as the trojan horse to re-open arbitral proceedings and substitute a tribunal’s decision for the views of domestic courts, especially where large monetary awards are at stake.
In light of these challenges and the risk of protracted proceedings, it is imperative to develop an analytical framework for the operation of equity in the determination of compensation by investment tribunals. Against this background, this chapter argues that while investment arbitral tribunals are entitled to apply equitable considerations when determining compensation as a general principle of international law, this possibility is restrained by certain limitations beyond which the award might result in a legal error or an excess of powers. To that end, Section 2 will distinguish between the different forms that equity may take and examine the interpretative function of equity in the framework of compensation. Section 3 will examine the interpretative function of equity in the framework of customary norms of State responsibility, whereas Section 4 will argue that recourse to equity is subject to intrinsic and extrinsic limitations, emanating either from the nature of equity as an interpretative canon or from the procedural framework in which tribunals are bound to operate, respectively.
A few words are in order on the scope of this chapter. For analytical purposes, the term ‘compensation’ should be understood as a pecuniary remedy for the reparation of injury caused by an internationally wrongful act within the meaning of Articles 31 and 36 of ARSIWA. The relationship between equity and other forms of remedies, such as restitution or satisfaction, fall outside the scope of the analysis. In the same vein, equity may also have a bearing on the determination of ‘compensation’ that is due upon liability for injurious, yet lawful acts. Indeed, there are numerous treaty provisions that require the payment of ‘fair’, ‘equitable’ or ‘just’ compensation for acts that are not prohibited by international law, such as the expropriation of foreign investmentsFootnote 30 or the civil liability of economic operators for the harm caused to persons or the environment by hazardous or ultrahazardous activities.Footnote 31 Nevertheless, the interpretation of these treaty-specific provisions is a question of primary, not secondary, norms, which are subject to distinct rationales and present structural legal differences when compared to State responsibility. Thus, the meaning of equity within these treaty-specific regimes falls to be determined by reference to their distinctive teleologies and contextual specificities. Finally, even though equitable considerations are frequently integrated in computational models proposed by valuation expertsFootnote 32 and the methodology employed by tribunals,Footnote 33 this chapter will only address the legal function of equity, as opposed to the use of equity in the process of valuation methodologies.
2 The Legal Basis for the Application of Equity to Compensation
Doctrinal analysis of the concept of equity typically begins with some preliminary questions regarding the normative character of equity and its functions in general international law. It is not, however, the purpose of this chapter to revisit the doctrinal debate surrounding the normativity of equity.Footnote 34 Suffice to say that, throughout the twentieth century, the development of international law has transformed equity from a non-legal conceptFootnote 35 to a general principle of law within the meaning of Article 38(1)(c) of the PCIJ and later ICJ Statute.Footnote 36 As early as 1920, the Advisory Committee of Jurists, tasked with the preparation of the draft Statute for the PCIJ, understood equity as an integral part of international law to be applied by the World Court,Footnote 37 a point subsequently endorsed by the overwhelming majority of scholars and jurisprudence.Footnote 38
Nevertheless, investment tribunals have not clearly articulated the legal basis for the application of equitable considerations to the assessment of damages. In LIAMCO, for example, the tribunal confusingly referred to equity as a ‘general principle of law’ under Article 38(2) of the ICJ Statute, instead of Article 38(1)(c).Footnote 39 In the same vein, the Aminoil Tribunal stated that ‘redress will be ensured ex aequo et bono’ without ‘depart[ing] from principles of law’, in plain contradiction to the terms of Article 38(2) of the ICJ Statute.Footnote 40
It is, however, clear, that the application of equity as a general principle of law should not be confused with a decision ex aequo et bono. In the North Sea Continental Shelf cases, the ICJ drew a distinction between the power of the Court to settle disputes ex aequo et bono and equity as an integral part of international law (equity intra legem).Footnote 41 In Tunisia/Libya, the Court explained that ‘the legal concept of equity is a general principle directly applicable as law.’Footnote 42 The Court, ‘whose task is by definition to administer justice is bound to apply it.’Footnote 43 By contrast, dispute-settlement ex aequo et bono entails that a tribunal may act as an amiable compositeur for the ‘adjustment of the respective interests’ of the parties.Footnote 44 While the latter function requires express agreement by the parties (Article 42(3) of ICSID),Footnote 45 the former is not simply ‘a matter of abstract justice’, but a rule of law capable of generating legal obligations between States.Footnote 46
Within the framework of intra legem equity, the ICJ has drawn a further distinction between equity praeter and infra legem.Footnote 47 Equity praeter legem acquires an autonomous normative function in case of lacunae, ‘in order to remedy the insufficiencies of international law and fill in its logical lacunae.’Footnote 48 Conversely, infra legem equity consists in ‘a method of interpretation of the law in force, and is one of its attributes.’Footnote 49 Leaving aside doctrinal objections against the traditional typology of equity,Footnote 50 this analysis will not focus on the praeter legem of equity: to the extent that the Chorzów Factory standard has received wide-spread acceptance in State practice and jurisprudence as reflecting customary law, it seems untenable to speak of a general ‘gap’ in State responsibility to which praeter legem equity could apply,Footnote 51 although it may always be possible to identify smaller gaps to which praeter legem equity may be of relevance.
Rather, it is the interpretative function of equity that is most pertinent to the customary rules governing the determination of compensation.Footnote 52 Thus, in Amco v Indonesia the ICSID annulment committee dismissed the idea that any mention of ‘equitable considerations’ in the award would necessarily amount to a decision ex aequo et bono, and accepted that equitable considerations may ‘form part of the law to be applied by the Tribunal’ for the purposes of compensation.Footnote 53 In Dogan v Turkmenistan the annulment committee also stated that equitable considerations were ‘inherent (…) in the interpretation of the law applied by the Tribunal.’Footnote 54 The committee in MTD v Chile developed this point further, stating that a tribunal is entitled to ‘tak[e] into account considerations of fairness in applying the law’, given that ‘individual rules of law will often require fairness or a balancing of interests to be taken to account.’Footnote 55
These propositions align with the general understanding of the hermeneutical function of equity as intimately linked with the requirements of good faith and reasonableness. Following a long tradition of jurists,Footnote 56 Schwarzenberger postulates that equity demands ‘reasonableness and good faith in the interpretation and application of treaties’.Footnote 57 For ‘[e]ven in a relatively static environment, the need arises sooner or later to soften the harshness of jus strictum by the infusion of elements of equity and elasticity.’Footnote 58 In the same vein, certain authors have argued that equity may ‘soften’ or ‘temper’ the strict application of positive rules, by ‘infusing elements of reasonableness and ‘individualised’ justice in their interpretation, whenever the applicable law leaves a margin of discretion’,Footnote 59 or as a ‘a normative flexifier [sic] mitigating the rigidity of application of positive international law’.Footnote 60
Insofar as infra legem equity has been mostly theorised at the level of treaty interpretation, the question arises whether its interpretative function extends to customary rules. In this regard, academic authors have raised a series of doctrinal objections against the possibility of interpreting customary norms as such, pointing towards the absence of a written text that could be analysed through the ordinary means of interpretation.Footnote 61 Without further delving into this wider doctrinal debate, it is sufficient to note that the applicability of infra legem equity to compensation has been relatively uncontested in practice: thus, in the Armed Activities case the ICJ made several references to equitable considerations at the reparations stage, and several judges acknowledged that recourse to equitable considerations in determining compensation ‘is an application of equity infra legem’.Footnote 62 In Total v Argentina, the tribunal also noted that ‘[e]quitable considerations in the application of the law, including in performing calculation of damages, pertain to aequitas infra legem (…) and not aequitas praeter legem to use a Latin expression (equity within what the law admits)’.Footnote 63 Indeed, if infra legem equity can affect the interpretation of treaty-based rules, it stands to reason that it can also affect the interpretation of customary norms, which are framed at such level of generality that a further deductive process is required to particularise their content and meaning to the circumstances of each case.Footnote 64 The question, therefore, is not about whether equitable considerations may apply to the interpretation of customary law in abstracto, but rather about how that process comes to bear.
3 Lost and Found: Equitable Considerations in the Law of State Responsibility
Heretofore, academic authors have approached the principle of equity through the lens of primary rules governing inter-State relations, ranging from maritime boundary delimitationFootnote 65 to the exploitation and management of natural resources,Footnote 66 the ‘fair and equitable treatment’ standard in BITs,Footnote 67 or the law of the WTO.Footnote 68 By contrast, much less attention has been paid to the question whether – and if so, how – equity may affect the interpretation of secondary norms governing the consequences of wrongful acts,Footnote 69 even less so compensation.Footnote 70 As noted by Milano, the function of equity in the identification of remedies for wrongful acts
is an aspect of the general principle of equity which has been under- investigated in the literature and one where the relationship between equity itself and the application of the ordinary rules of State responsibility, presumably of a customary nature, becomes crucial.Footnote 71
However, recourse to equitable considerations for compensation purposes is not new.Footnote 72 States have instructed arbitral tribunals and mixed claims commissions to apply ‘equity’ to the assessment of damages arising from foreign claims as early as the 1794 Jay Treaty, where the US and Great Britain mandated the umpire to settle their claims to compensation on the basis of ‘justice, equity and the law of nations’. But even where equity was not expressly mentioned in the arbitral agreement, this did not prevent arbitrators from invoking equity proprio motu for the determination of compensation.Footnote 73 Even though some of these early decisions were rendered ex aequo et bono,Footnote 74 some other tribunals invoked equity within the framework of legal reasoning, either as the normative basis for allowing a claim (praeter legem),Footnote 75 or as a legal principle capable of influencing the interpretation of customary law (infra legem).Footnote 76
Eventually, the concept of equity found its way into the codification efforts of the ILC on the law of State responsibility. Originally, the 1930 Preparatory Committee of the Hague Conference stated in its ‘Basis for Discussion No. 29’ that ‘[r]esponsibility involves (…) an obligation to make good the damage suffered in so far as it results from failure to comply with the international obligation.’Footnote 77 This standard echoes the conventional understanding of the Chorzów Factory judgment that had been rendered by the PCIJ just the previous year. As early as 1956, however, Special Rapporteur Garcia-Amador recognised that, apart from the remedial function of compensation, there may also be some ‘attenuating’, ‘extenuating’ or ‘aggravating’ circumstances that can affect the extent to which a State is bound to compensate for injury caused to aliens in its territory.Footnote 78 This position, which signalled a departure from Chorzów Factory, became clearer in his subsequent reports, where he noted that, while
the basic and at the same time general criterion, is that the reparation should be commensurate with the nature or extent of the actual injury (…) the reparation is not always strictly in keeping with the true nature or extent of the injury. Other factors generally come into play, such as the circumstances in which the injury occurred, the gravity, in special situations, of the act or omission imputable to the respondent State and, on occasion, factors justifying a reduction in the amount of the reparation.Footnote 79
Special Rapporteur Ago did not submit a report on the consequences of internationally wrongful acts before his election to the ICJ. However, when discussing his eighth report on the circumstances precluding wrongfulness, several ILC members observed that, while certain circumstances may not preclude the wrongfulness of an act qua, they may nonetheless operate as mitigating factors for the purposes of reparation.Footnote 80 Special Rapporteur Ago acknowledged this pointFootnote 81 and the commentary to draft Chapter V stated that circumstances precluding wrongfulness ‘must not be confused with other circumstances which might have the effect not of precluding the wrongfulness of the act of the State but of attenuating or aggravating the responsibility entailed by that act’, with regard to the content, form and degree of responsibility.Footnote 82 Even though neither Rapporteur expressly referred to ‘equity’ as the legal basis, they both acknowledged that it was possible for compensation to take into account not only the extent of the injury caused by the wrongful act, but additional factors as well.
The following year, Special Rapporteur Riphagen argued in favour of a ‘qualitative’ and ‘quantitative’ degree of proportionality between the characteristics of the unlawful conduct and the consequences in response thereto, including the level and amount of compensation.Footnote 83 Notably, since the North Sea Continental Shelf cases the ICJ has drawn a connection between the application of equitable principles and a ‘reasonable degree of proportionality’ to be observed by the respective decision-maker.Footnote 84 In the course of the debate on Riphagen’s report, the principle of proportionality was expressly endorsed by some ILC membersFootnote 85 (notably in the context of compensationFootnote 86) but met with scepticism from others.Footnote 87
Equitable considerations resurfaced with greater force at the last stages of the codification process. In his first report to the ILC, Special Rapporteur Arangio-Ruiz described the Chorzów Factory principle as too vague or sweeping a proposition, which does not settle all of potential legal issues involved, such as the ‘relevance of the injured State’s conduct’, of the ‘gravity of the wrongful act’ or the ‘degree of fault of the offending State’.Footnote 88 The following year, however, he explained that ‘[h]e had omitted express references to equity from the report because, as experience showed, such references were apt to be unhelpful. Needless to say, however, equity was implied in all legal rules and formed an essential and integral part of law.’Footnote 89 In his opinion, equity was intimately linked to the relevance of fault, wilful intent or negligence for the purposes of compensation.Footnote 90 In the same vein, other ILC members raised the question of the onerousness of the financial obligation upon the obligor State as an equitable consideration that could justify a proportional reduction of damages in some cases.Footnote 91 The Special Rapporteur expressly acknowledged the role of equity in the assessment of damages, but noted that ‘it might be dangerous to refer expressly to [equity], since it was part and parcel of law and of any legal decision’.Footnote 92 Still, the original Commentary to Article 6bis stated that:
There may be other equitable considerations that militate against full reparation, particularly in cases involving an author State with limited financial resources, but only to the extent that such considerations can be reconciled with the principle of the equality of all States before the law and the corresponding equality of the legal obligations of all States.Footnote 93
It was, therefore, understood that the customary standard of full reparation could be balanced against equitable considerations, which could reduce the extent of reparation, including the amount of compensation. In his third report to the ILC, Special Rapporteur Crawford observed that international jurisprudence reflected
the wide variety of factual situations, the influence of particular primary obligations, evaluations of the respective behaviour of the parties (both in terms of the gravity of the breach and their subsequent conduct), and, more generally, a concern to reach an equitable and acceptable outcome.Footnote 94
When international judges are making a complex judgment such as one regarding the amount of compensation, he observed, ‘equitable considerations will inevitably be taken into account, whether acknowledged or not.’Footnote 95 He warned, however, that, ‘while illustrations can be given of the operation of equitable considerations and of proportionality in international law, the attempt to specify them in detail is likely to fail.’Footnote 96 Given that the ILC was anxious to conclude its codification work before 2001, the Special Rapporteur made no effort to define ‘equitable considerations’ in detail. But the general proposition made it to the final commentary to Article 36 of ARSIWA (albeit with diluted wording), stating that:
As to (…) the principles of assessment to be applied in quantification, these will vary, depending upon the content of particular primary obligations, an evaluation of the respective behaviour of the parties and, more generally, a concern to reach an equitable and acceptable outcome.Footnote 97
It follows that the ILC understood that the customary standard of ‘full reparation’ would not always provide satisfactory solutions and that international tribunals could have recourse to equity, as a general principle of law, for the adjustment of compensation in such a way as to reflect additional factors, such as the parties’ conduct and situation, the content of the primary norm breached, or an evaluation of their respective interests. While most Special Rapporteurs acknowledged the relevance of equity for compensation (including under the rubric of proportionality), it was agreed not to insert an express qualification to the text of the draft articles. However, far from discounting the relevance of equity in favour of a mechanical approach to quantum, the ILC suggested that infra legem equity forms part-and-parcel of the secondary rules of State responsibility and may have a bearing on the level of compensation. This proposition, reflected in the final commentary of ARSIWA after more than 50 years of discussions and buttressed by the contemporaneous and subsequent practice of courts and tribunals, is of key import for the interpretation of the rules governing compensation.Footnote 98
4 Equitable Considerations in Investment Arbitration: Is There a Limit?
In line with the preceding analysis, investment tribunals have, expressly or impliedly, applied equitable considerations to the determination of compensation payable in case of unlawful acts. As early as 1982, for example, the Aminoil Tribunal observed that ‘any estimate in money terms of amounts intended to express the value of an asset, of an undertaking, of a contract, or of services rendered, must take equitable principles into account’.Footnote 99 In Amco v Indonesia, the committee further stated that ‘a tribunal applying international law may take account of equitable considerations in non-maritime boundaries cases’, such as compensation.Footnote 100
While this proposition is generally accepted in the literature,Footnote 101 there seems to be no consensus as to what these ‘equitable considerations’ might be. Commentators, practitioners and tribunals alike have proffered a wide range of factors capable of influencing quantum (well beyond the Chorzów formula), albeit not always invoking equity as the relevant legal basis. On the one hand, it has been suggested that equitable considerations may justify an aggravated amount of compensation, in order to reflect the ‘seriousness’ or ‘gravity’ of the unlawful act,Footnote 102 or the subjective intent (or fault) of the wrongdoer State.Footnote 103 Similar considerations, however, evoked serious objections within the ILC during the codification process that led to the ARSIWA: despite the original proposals to enable damages reflecting the gravity of the breach,Footnote 104 the ILC eventually rejected the idea that compensation be used as a vehicle for the introduction of punitive damages.Footnote 105 The ICJ has also rejected the availability of punitive or exemplary damages, even in those cases involving the most serious violations of international human rights and humanitarian law.Footnote 106
On the other hand, it has been suggested that equitable considerations may warrant an adjustment (or proportionate reduction) of compensation in order to accommodate additional, countervailing considerations, that relate either to the injured party or the wrongdoer State. As regards to the first category, some tribunals have sought to limit the amount of recoverable compensation by reference to the injured party’s conduct that either precedes or follows the wrongful act. For example, in Himpurna v PLN the Tribunal applied the doctrine of abuse of right in favour of the respondent, in order to prevent the claimant’s contractual rights from being extended ‘beyond tolerable norms’, also taking into account PLN’s status ‘as an arm of governmental policy acting in pursuit of the public welfare’. On that basis, it lowered the amount of compensation due referring to ‘equitable principles’.Footnote 107
Conversely, international tribunals have endorsed the proposition that the amount of compensation must be reduced when the claimant has failed to take ‘reasonable steps’ to mitigate the injury caused.Footnote 108 Failure to mitigate damages has been understood as a matter that is notionally distinct from the contributory fault/negligence of the injured party in the occurrence of the wrongful conduct and the emergence of the harm, which is a matter related to the existence of causal nexus.Footnote 109 Insofar as tribunals have based the existence of a ‘duty to mitigate damages’ upon considerations of fairness,Footnote 110 this proposition may be interpreted as a specific form of equitable considerations that comes into play after the occurrence of the harm. In the same vein, the Arbitral Tribunal in the Loan Agreement case emphasised the ‘equitable character’ of the customary norms of compensation, which led the Tribunal to consider not only the ‘technical’ provisions of the treaty and loan agreements, but also the ‘overall circumstances of the case, including the causes of delay, the misunderstandings (…) and generally the specific situation and conduct of both Parties, as well as the totality of the relations of amity and co-operation.’Footnote 111
With respect to equitable considerations relating to the wrongdoer State, it is important to note that recent scholars have argued for the reconceptualisation of investment arbitration from a private-law-type arbitration into a form of ‘public-law adjudication’, which takes into consideration the public functions of the host State vis-à-vis its population in furtherance of the public interest and allows for some flexibility to the host State concerned. The public-law paradigm has found expression in investment arbitration in several ways, such as the principle of proportionality, legitimate expectations and the applicable standard of review.Footnote 112 Similar considerations have found their way into the assessment of damages.Footnote 113 For instance, it has been suggested that investment tribunals should either weigh the level of compensation against legitimate ‘public interest’ considerations motivating the unlawful conduct of the host StateFootnote 114 or consider the circumstances surrounding the wrongful act, such as the occurrence of an armed conflict in the host State’s territory.Footnote 115
Within that context, the potentially ‘crippling’ effect that large sums of compensation may have for a host State’s financial subsistence has been suggested as a potential equitable consideration that is relevant to quantum. In his separate opinion in the quantum phase of CME v Czech Republic, Sir Ian opined that the principles of compensation must be read within the framework of the BIT and noted that ‘[i]t would be strange indeed, if the outcome of a [BIT] took the form of liabilities “likely to entail catastrophic repercussions for the livelihood and economic well-being of the population”’.Footnote 116 Along similar lines, Paparinskis has argued in favour of an exception to the principle of full compensation, with a view to ensuring that ‘[r]emedies serve social as well as individual needs’: to the extent that the bilateralist precepts of corrective justice that underlain Chorzów Factory have gradually evolved in a more communitarian direction, Paparinskis posits that the standard of compensation ‘can be changed in line with the broader structural shifts in modern international law’.Footnote 117 Indeed, in the Armed Activities case the ICJ took note of Uganda’s plea that a large amount of compensation would exceed its capacity to pay and seems implicitly to have endorsed the relevance of this legal ground by reference to the award by the Eritrea-Ethiopia Claims Commission (EECC).Footnote 118 Ultimately, however, the Court was ‘satisfied that the total sum awarded (…) remain[ed] within the capacity of Uganda to pay’, although it ordered the payment of the sum in annual instalments, presumably in order to Uganda’s ability to meet its people’s basic needs.Footnote 119
The foregoing remarks serve to show that equitable considerations are not a monolithic concept but are used as an umbrella term to denote a wide variety of factors and circumstances which have an influence upon quantum. To be sure, a detailed analysis of each these considerations would exceed the limited purpose of this chapter, but a key point stands out: whatever these ‘equitable considerations’ may be, it is suggested that investment tribunals do not have a carte blanche to subvert the customary principle of full reparation on the basis of ‘abstract equity’.Footnote 120 Indeed, it is well settled that an investment tribunal ‘must base its decision on objective and rational considerations which must be stated.’Footnote 121 The US’ strong objections to the application of equitable considerations in the determination of compensation arising from the Norwegian Shipowners case aptly illustrates how an unprincipled application of equity to damages may have serious repercussions on the validity of the arbitral award and the integrity of the process.Footnote 122 As Judge Yusuf stated in the Armed Activities case,
Equitable considerations (…) should be understood within the legal framework governing the judicial function of the Court. They cannot serve as the basis to dispense with the applicable rules altogether, or not to provide reasons for their applicability. The Court should have made an attempt at explaining how it intends to apply equity within the general framework of State responsibility and the procedural framework governing the fact-finding procedure before it’.Footnote 123
On that basis, it is submitted that, whilst infra legem equity may have a bearing on the interpretation of customary norms governing compensation, it is by its nature subject to certain limitations that may either be intrinsic to the ‘general framework of State responsibility’ or extrinsic to it, appertaining to the ‘procedural framework governing the fact-finding procedure’ before the respective court or tribunal. We shall examine these two kinds of limitations in turn.
4.1 Intrinsic Limitations to infra legem Equity
Intrinsic limitations emanate from the nature of equity as a canon of interpretation. As the Institut de Droit International noted in 1937, an international judge may be called upon to consider equitable considerations in the interpretation of norms, ‘to the extent consistent with respect for the applicable law.’Footnote 124 If equity is supposed to operate within the limits of the law, its hermeneutical function must be understood by reference to the limitations applicable to any rule of interpretation. As a cognitive exercise, the process of interpretation is ordinarily restrained by the rule to which it relates and its possible meanings.Footnote 125 Essentially, any method of interpretation involves the selection of a meaning amongst a spectrum of possible meanings within a conceptual radius defined by the widest possible meaning.Footnote 126
Within that hermeneutical process, infra legem equity assists the interpreter in both identifying the outer limits of the norm in question, and in selecting ‘among several possible interpretations of the law the one which appears, in the light of the circumstances of the case, to be closest to the requirements of justice.’Footnote 127 While these analytical choices will not always be clear-cut, it is suggested that a tribunal cannot exceed the conceptual radius of a norm and select a meaning beyond its range on the basis of equitable considerations – for this would result in extending or altering the norm’s content into something else.Footnote 128 Consequently, in the South West Africa cases the ICJ rejected the applicants’ contention that ‘humanitarian considerations [we]re sufficient in themselves to generate legal rights and obligations’ from the applicable treaties.Footnote 129 To do so would exceed the process of interpretation and result in rectification or revision of the treaty, whereas a court of law ‘can take account of moral principles only in so far as these are given a sufficient expression in legal form.’Footnote 130
The interpretative function of equity becomes much more complex in the realm of customary law, precisely because there is no authoritative text to be interpreted.Footnote 131 Depending on the availability and specificity of State practice and opinio juris, customary rules tend to be much more vague and flexible, leaving some scope for debate regarding the precise limits and content of the rule.Footnote 132, Footnote 133 As Merkouris points out, customary rules exist at such level of abstraction that a further deductive process is required to particularise their meaning to the facts of each case.Footnote 134 Thus, customary law affords the decision-maker a wide margin of discretion in interpreting and applying abstract rules to the circumstances of each case. In the absence of textual limitations, it is here that infra legem equity has a key role to play by defining the contours of customary law or providing the basis from which to infer potential qualifications.
This does not mean that the output of the interpretative process will vary along with the proverbial foot of the Chancellor.Footnote 135 Nor should the application of equitable considerations to customary rules be understood as being ‘freed from the moorings of international law (…) drifting towards elusive subjectivism with little room left for the necessary guarantee of the objectivity and predictability of the law.’Footnote 136 While infra legem equity to some extent involves the exercise of discretion,Footnote 137 it forms part of the applicable law and must, therefore, display a minimum degree of consistency. As Jennings observes, no reasonable litigant expects the decision of a court to be predictable; but the range of considerations used for a decision and the procedures or their application should certainly be predictable.Footnote 138
In determining these potential limitations, useful lessons can be drawn from the practice of equitable considerations in other areas of international law. In the North Sea Continental Shelf cases, the ICJ explained that the interpretative function of equity was limited by the object and purpose of the customary principles governing the continental shelf and could not result in the de novo apportionment of maritime areas on the basis of distributive justice.Footnote 139 To hold otherwise would contravene the ‘most fundamental of all the rules of law relating to the continental shelf’.Footnote 140 In Libya/Malta, the Court further stressed the need for consistency and identified potential limitations to the application of equity to maritime delimitation:
Th[e] justice of which equity is an emanation, is not abstract justice but justice according to the rule of law; which is to say that its application should display consistency and a degree of predictability; even though it looks with particularity to the peculiar circumstances of an instant case, it also looks beyond it to principles of more general application. This is precisely why the courts have, from the beginning, elaborated equitable principles as being, at the same time, means to an equitable result in a particular case, yet also having a more general validity and hence expressible in general terms.Footnote 141
On that basis, the Court distinguished between those equitable considerations which are ‘pertinent to the institution of the continental shelf as it has developed within the law’ and may, therefore, qualify for inclusion in the rule (ie, circumstances of a geographical nature), and those ‘which are strange to its nature’ and cannot be used ‘fundamentally’ to alter its character.Footnote 142 Thus, the Court seems to have recognised certain limitations to the interpretative function of equity as a canon by reference to the content of the rule and its teleology.
The ICJ affirmed this proposition in Barcelona Traction, where it rejected Belgium’s contention that the customary rule of diplomatic protection should have been extended to a company’s shareholders on the basis of ‘equitable considerations’.Footnote 143 In so doing, the ICJ emphasised the nature of the customary right to exercise diplomatic protection and explained that any different interpretation would render inoperable the ‘original’ right of the State and severely undermine ‘the stability which it is the object of international law to establish in international relations.’Footnote 144 Thus, the Court implicitly confirmed that equitable considerations cannot be used to justify any possible exception within the interpretation of a customary norm in the framework of State responsibility and it is subject to certain limitations stemming from the rule’s object and purpose.
Extending this logic to compensation for internationally wrongful acts, it is suggested that equitable considerations cannot contradict the object and purpose of the rule, which is to ‘wipe out’, as far as possible, ‘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’.Footnote 145 The implications of this point are two-fold. On the one hand, equity cannot circumvent the inherently remedial function of compensation: thus, the invocation of the ‘seriousness’ of the breach, the ‘magnitude’ of the damage, or the ‘wilful intent’ of the host State cannot serve as bases for awarding damages beyond and above the amount of injury suffered by the claimant. Such damages, essentially of an afflictive, punitive or exemplary character are foreclosed under modern international lawFootnote 146 and an award to that effect might be contra legem.Footnote 147 At the same time, however, equitable considerations cannot obviate claims for damages altogether: as the PCIJ explained in Chorzów Factory, compensation constitutes an ‘essential principle contained in the actual notion of an illegal act’. Even if compensation may be adjusted to accommodate a balancing exercise between competing interests at stake, this cannot detract from the core principle that ‘the breach of an engagement involves an obligation to make reparation in an adequate form.’Footnote 148 A different understanding would essentially transform equity into a circumstance precluding wrongfulness, without a clear legal basis.
Within these two extremes, it is quite difficult to pinpoint the extent to which equity may affect quantum in the abstract. Each factor requires an independent legal analysis and a careful application of the customary principle to the facts of each case. But this does not mean that equity is a matter of chance.Footnote 149 Whilst Radi emphasises the ‘casuistic normativity’ of equity for the proposition that it is impossible to define infra legem equity in general legal terms,Footnote 150 international tribunals are required to state the reasons for their decisions, including on their interpretation and application of equity. As the committee stated in Rumeli v Kazakhstan,
It is highly desirable that tribunals should minimise to the greatest extent possible the element of estimation in their quantification of damages and maximise the specifics of the ratiocination explaining how the ultimate figure was arrived at.Footnote 151
Within that wider normative process of trial and error, equity may perform its essential role by introducing elements of flexibility and reasonableness into the law, without departing from the object and purpose of the rule being interpreted. As tribunals continue to expand and rationalise the ways in which equity affects their interpretation of the customary standard of compensation, these judicial pronouncements (and the attitude of States to these interpretations) will tend to harden into rules and legal principles, to the effect that ‘the freedom to frame arguments and frame the reasoning leading to decisions in the court will be correspondingly reduced.’Footnote 152
4.2 Extrinsic Limitations to Equitable Considerations: Between Alchemy and Science
The duty to state reasons brings us to the next point, which is the extrinsic limitations to the interpretative function of equity. Contrary to intrinsic limitations, extrinsic limitations do not relate from the content of the rule being interpreted but rather emanate from the tribunal’s adjudicative authority and the procedural framework in which it is bound to operate. When a tribunal resorts to equity, it can neither exceed the scope of its jurisdiction nor can it disregard certain fundamental rules of procedure upon which its function is conditioned (cf. Article 52(1)(d) ICSID) – at least not without the validity of its decision being impinged. As a result, in Klöckner v Cameroon, the ICSID annulment committee annulled an arbitral tribunal’s award for manifest excess of power inter alia on the basis that the tribunal had reached an ‘equitable estimate’ of damages, using ‘approximate equivalents’, and had, therefore, failed to state reasons as required by Articles 52(1)(e) and 48(3) of the ICSID Convention.Footnote 153
The obligation to state reasons is of key import in the framework of compensation, where equity is typically associated with the lack of sufficient evidence to ascertain the precise extent of the injury.Footnote 154 The valuation of injury caused to long-term investments and commercial undertakings is a complex task, that often involves conflicting methodologies and inadequate evidence.Footnote 155 While it may be feasible to gauge the amount of lucrum cessans on the basis of business records from ongoing concerns, investment disputes frequently arise from still-born projects or failed contracts, where the value of profits can hardly be determined at all. As noted in Santa Elena v Costa Rica, investment tribunals enjoy a wide measure of discretion in making an approximation, ‘taking into account all relevant circumstances (…) including equitable considerations’.Footnote 156
Nevertheless, equitable considerations in the interpretation of the law must not be confused with the ordinary exercise of arbitral discretion in the appreciation of the facts.Footnote 157 As noted in ADM v Mexico, ‘the assessment of damages for lost profits is not a precise science.’Footnote 158 Thus, the discretion of an arbitral tribunal in the calculation of damages arises from the uncertainty of the inquiry into lost profits, involving an inquiry with a counterfactual premise, namely, the consideration of the profits that would have been made if an illegal act – which did in fact occur – had not occurred.Footnote 159 In order to make these complex factual determinations, the tribunal will have to consider documentary evidence, witness testimony, admissions against interest, as well as expert reports and shall be the judge of the admissibility of any evidence adduced and of its probative value.Footnote 160 Inevitably, the assessment of the facts lies within the arbitrator’s free evaluation of evidence (also known as the conviction intime du juge).Footnote 161 While this should be guided by a spirit of fairness, it is difficult to lay down precise legal rules of international law, upon which infra legem equity may come to bear as an interpretative canon.
By contrast, equitable considerations might affect the interpretation of the procedural rules governing the fact-finding process,Footnote 162 especially since investment tribunals consider themselves not to be bound to adhere to strict judicial rules of evidence.Footnote 163 In line with international case-law,Footnote 164 tribunals have relied upon equity to suggest that the fact that damages cannot be settled with certainty is no reason not to award damages when a loss has incurred.Footnote 165
In this regard, tribunals have drawn a distinction between the allocation of the burden of proof and the standard of proof. It is well established that the claimant bears the onus to establish the conditions required by substantive law to corroborate a claim for damages.Footnote 166 The claimant must not only bring evidence in support of its allegations but must also convince the tribunal of their truth, lest they be disregarded for want, or insufficiency, of proof.Footnote 167 However, as the ICJ pointed out in Diallo, this rule may be applied flexibly in certain cases, especially where the respondent may be in a better position to establish certain facts.Footnote 168 Given that tribunals have a relative freedom as to the allocation of the burden of proof,Footnote 169 equitable considerations may therefore warrant a shifting of the burden of proof, where the establishment of certain facts lies within the power of the respondent,Footnote 170 or where a party has impeded access to material evidence.
The Archer Daniels Tribunal took this reasoning a step further, noting that ‘failure of a claimant to prove its damages with certainty, or to establish its right to the full damages claimed, does not relieve the tribunal of its duty to assess damages as best it can on the evidence available’ (emphasis added).Footnote 171 This ‘duty’, endorsed by subsequent investment tribunals in compensation proceedings,Footnote 172 implies a gradual departure from the adversarial model of arbitration into a more inquisitorial system, a proposition that may have far-reaching implications for investor-State dispute settlement.Footnote 173 As Lord Neuberger has observed, ‘an increase in arbitral powers must be accompanied by an increased responsibility’.Footnote 174 As mega-awards intimating claims over USD one billion plus continue to emerge, it seems likely investment tribunals will assert more control over the assessment of claims that may have serious impact on tax-payers’ resources.Footnote 175 The progressive recognition of investment arbitration as a form of public law adjudication might require tribunals to take a more active role in quantum, inquiring into possible mitigating circumstances for the assessment of compensation. Thus, when faced with a (relatively modest) claim of about USD 22 million against war-torn Zaire, the AMT Tribunal took the formal step of appointing a former World Bank official as an independent expert to evaluate damages suffered by the claimant, invoking ‘its discretionary and sovereign power to determine the quantum of compensation’.Footnote 176 Such recourse to external expertise may provide as a legitimate alternative when the evidence on the record is not sufficient to justify a precise amount of compensation, as shown in the Armed Activities case, where the ICJ relied upon the analyses of two Court-appointed for the purposes of determining compensation for the loss of life and natural resource. However, it may also give rise to objections as unfairly interfering with the allocation of the burden of proof and tilting the balance in favour of one Party to the detriment of the other, contrary to the principles of a fair hearing and equality of arms and outsourcing the tribunal’s function to the experts.Footnote 177
On the other hand, investment tribunals have lowered the standard of proof invoking equitable considerations as the basic justification. In Crystallex v Venezuela, for example, the Tribunal distinguished between the existence of damage as a fact that must be proven ‘with certainty’ and the precise quantification of that damage which is not subject to the same degree of certainty, ‘because any future damage is inherently difficult to prove.’Footnote 178 Similarly, the Tribunal in Lemire v Ukraine observed that ‘less certainty is required in proof of the actual amount of damages; for this latter determination claimant only needs to provide a basis upon which the tribunal can, with reasonable confidence, estimate the extent of the loss.’Footnote 179 In Impregilo, the Tribunal also held that it would be unreasonable to require precise proof of the extent of the damage caused. Instead, reasonable probabilities and estimates would suffice as a basis for compensation.Footnote 180 The same principle was expressed by the EECC, which applied a differentiated standard of proof between the merits and the reparation phase:
The Commission has required clear and convincing evidence to establish that damage occurred, within the liability parameters of the Partial Awards. However, for purposes of quantification, it has required less rigorous proof. The considerations dictating the ‘clear and convincing standard’ are much less compelling for the less politically and emotively charged matters involved in assessing the monetary extent of injury.Footnote 181
Again, however, a balance must be struck between competing interests. The function of procedural equity is limited by the requirements of due process from which a tribunal may not detract. As the Iran-US Claims Tribunal recalled in Amoco, ‘[o]ne of the best settled rules of the law on international responsibility of States is that no reparation for speculative or uncertain damage can be awarded.’Footnote 182 In the Tribunal’s view, international law does not permit the use of a method which yields uncertain figures for the valuation of damages, even if the existence of damages is certain.Footnote 183 Thus, investment tribunals have generally rejected possible but contingent and indeterminate damages in the absence of evidence.Footnote 184 In Diallo, the ICJ also affirmed that, whilst an award of compensation relating to loss of future earnings inevitably involves some uncertainty, ‘such a claim cannot be purely speculative’.Footnote 185 Nevertheless, as the Tribunal held in Achmea I, the requirement of proof must not be impossible to discharge. Nor must the requirement for reasonable precision in the assessment of the quantum be carried so far that the search for exactness in the quantification of losses becomes disproportionately onerous when compared with the margin of error.Footnote 186 Indeed, compensation matters are not capable of precise quantification because they depend on the exercise of judgmental factors that are better expressed in approximations or ranges.Footnote 187 That is particularly so where the absence of evidence is a result of the behaviour of the author of the damageFootnote 188 or results from a failure of the claimant to present its case. These are equitable considerations that require a further analysis, depending on the procedural framework governing the arbitral process.
5 Conclusion
In 1939, Winston Churchill described Russia’s foreign policy during World War II as ‘a riddle wrapped in a mystery inside an enigma’.Footnote 189 There might be no better phrase to describe the use of equitable considerations by investment arbitral tribunals in the process of determining damages for violations of international law. This chapter may have been unable fully to unravel that mystery, but some useful lessons can be drawn. By now, the Chorzów Factory standard has been widely recognised as reflecting the customary standard governing reparation, but customary rules do not operate in a legal vacuum. International courts and tribunals have progressively acknowledged the general principle of equity as normative proposition capable of affecting the interpretation of customary norms, including secondary rules governing State responsibility. Even though the ILC avoided to articulate an express rule to that effect, it was well understood that equitable considerations had a role in the determination of compensation either as ‘aggravating’ or ‘mitigating’ circumstances.
Yet, infra legem equity is not unbound: it is subject to certain limitations that may either be intrinsic to its nature as an interpretative canon, or stemming from the procedural framework governing the function of the tribunal. Within these parameters, equity is not a magic spell that elides rational conceptualisation; to hold otherwise could have serious implications to the interpretative process and undermine the integrity of the procedure itself. And while it may be true that the assessment of damages is not always a precise science,Footnote 190 the opposite also holds true: ‘[e]quitable principles should not be used to make good the shortcomings in a claimant’s case by being substituted for evidence which could have been produced if it actually existed: equity is not alchemy.’Footnote 191
1 Introduction
There are several provisions in the Vienna Convention of the Law of Treaties (VCLT), reflecting customary law. Already in 1969, the VCLT was considered to be a partial codification of customary international law (CIL). Other rules of the VCLT constituted then developed into customary law.Footnote 1 Indeed, according to the commentaries to the VCLT the large majority of its provisions currently reflects custom.Footnote 2 As international investment law is a branch of international law based primarily on treatiesFootnote 3 it is not surprising that the provisions of the VCLT and corresponding customary rules have been often interpreted and applied by investment tribunals. Considering that the total number of investor-State dispute settlement (ISDS) cases had reached 1,190 by the end of 2021,Footnote 4 investment arbitrationFootnote 5 has the potential to significantly influence the interpretation of the law of treaties, both in the rules contained in the VCLT as well as those confirmed by CIL.
The situation of parallel existence between treaty and customary rules demonstrates that it is not only a treaty’s rules, but also the customary ones, that in practice can be, and in fact are, interpreted.Footnote 6 Furthermore, this scenario in particular proves that the content determination of customary rules can be, and is, accomplished through a different approach than the ascertainment of two classical elements of custom, that is, State practice and opinio juris.Footnote 7
This chapter will focus on two issues discussed broadly by investment tribunals: rules on conflicts of treaty norms (Article 30 of the VCLT and corresponding CIL) and rules relating to subsequent agreements in relation to the interpretation of treaties (Article 31(3)(a) of the VCLT and corresponding CIL). Emphasis on these two examples of the interpretation of customary law in investment arbitration seems to be particularly pertinent as, in these two areas, tribunal’s decisions seem to diverge from the approach, as reflected in the works of the International Law Commission (ILC), traditionally taken in general international law. Why do investment tribunals deviate from agreed understanding of rules on conflicts of norms? How they define the scope of these rules? And with respect to subsequent agreements, can they influence the CIL concerning the interpretation of treaties which envisage rights for individuals? These issues certainly call for a study on the matter.
The aim of this chapter is not to provide an extensive and exhaustive list of all such cases where the interpretation of the customary rules codified in Articles 30 and 31(3)(a) of the VCLT has occurred but rather to highlight some most interesting examples which demonstrate the general approach of investment tribunals to these norms.
2 Interpreting Rules on Conflicts of Treaty Norms in Investment Arbitration
The issue of resolving conflicts of norms originating from different legal acts is central to every legal order.Footnote 8 In international law it is regulated in Article 30 of the VCLT and corresponding CIL. From the perspective of international investment law, the main issue that has arisen in the interpretation of this legal norm relates to the material scope of the entirety of Article 30, ie the reference in the title and paragraph 1 of this provision to treaties ‘relating to the same subject matter’.Footnote 9
The jurisprudence of investment arbitral tribunals has mainly referred to Article 30 VCLT in cases concerning the relationship between intra-EU investment treaties, or the intra-EU application of the Energy Charter TreatyFootnote 10 in relation to the Treaty on the Functioning of the European UnionFootnote 11 (TFEU, or its predecessor – the Treaty Establishing the European CommunityFootnote 12). In the former case, Article 30 of the VCLT was invoked by the respondent, most often a Central-Eastern European State, as an argument emphasising the priority of the EU Treaties – subsequent treaties, over investment treaties – earlier treaties, ie dating back to the 1990s. At times, arbitral tribunals have also commented on the concept of the ‘same subject matter’ against the background of the applicability of Article 59 VCLT, which is also considered to reflect a customary rule, and then referred their conclusions to the applicability of Article 30 of the VCLT, even if they noticed differences between the purposes of these provisions.Footnote 13
As the ILC indicated in its report on fragmentation, adopting a narrow interpretation of this formulation could result in a number of potential treaty conflicts not being covered at all by this provision.Footnote 14 The Commission emphasised that:
If conflict were to exist only between rules that deal with the “same” subject-matter, then the way a treaty is applied would become crucially dependent on how it would classify under some (presumably) pre-existing classification scheme of different subjects. But there are no such classification schemes.Footnote 15
The ILC, therefore, opted for a flexible approach to the formulation of ‘same subject matter’. Support for this position can be found in the Commission’s commentary to the 1966 Draft Articles on the Law of Treaties,Footnote 16 where the view was expressed that this formulation was intended to broadly cover cases of incompatibility between treaty norms. This issue has rarely been a subject of consideration by international courts and tribunals. Similar reasoning to the position of the ILC can be found in GATT/WTO decisions. In EC – Imposition of Anti-Dumping Duties on Imports of Cotton Yarn from Brazil the panel stated that on the basis of analysis of a single of provision of the Multifibre Arrangement (MFA), the Anti-Dumping Agreement and the MFA were treaties ‘relating to the same subject-matter’.Footnote 17 In the China – Measures Related to the Exportation of Rare Earths, Tungsten, and Molybdenum the Appellate Body acknowledged that all multilateral trade agreements annexed to the Marrakesh Agreement relate to the same subject matter without detailed examination.Footnote 18
Arbitration decisions, however, have adopted a narrower understanding of Article 30 VCLT and, consequently, of the customary norm that this provision codifies. As was stated by the Eastern Sugar Tribunal, ‘[w]hile it is true that European Union law deals with intra-EU cross border investment, say between the Netherlands and the Czech Republic, as does the BIT, the two regulations do not cover the same precise subject-matter’.Footnote 19 The Tribunal underlined inter alia the existence of fair and equitable standard,Footnote 20 as well as the possibility for an investor to sue the host-State directly, as grounds to reject the ‘equivalence argument’.Footnote 21 This approach has been upheld in many subsequent arbitration awards relating to intra-EU bilateral investment treaties (BITs).Footnote 22 In EURAM, the Tribunal rejected the interpretation that the ‘same subject matter’ can be equated to being applicable to ‘the same facts’, or having ‘the same goal’.Footnote 23 The decision then subsequently elaborated that ‘[t]he subject matter of a treaty, in the Tribunal’s understanding, therefore differs both from the concrete situations in which it will be applicable and from its goal’.Footnote 24 For this Tribunal, the crucial argument was that prior to the Lisbon Treaty the EU had no competence in relation to direct investment.Footnote 25 A narrow interpretation was also presented by the Strabag Tribunal which ‘under[stood] the precondition of “same subject matter” as requiring the subject matters of the two treaties in question to be “identical”’.Footnote 26 Again, a similar position was expressed by the Juvel Tribunal: ‘[t]wo different treaties may apply simultaneously to the same set of facts without them having the same subject-matter. Further, if two treaties have the same goal but approach the achievement of that goal from two different perspectives, the treaties do not have the same subject-matter’.Footnote 27 The negative position of these tribunals, towards the applicability of the conflict of norms provisions, was primarily related to the far-reaching consequences that States could derive from the arguments in this regard, ie lack of jurisdiction of those tribunals to examine the case.Footnote 28
It is significant that in most of the aforementioned decisions, the arbitral tribunals did de facto interpret CIL, as the VCLT was inapplicable. In the overwhelming majority of these cases, tribunals seemed to be unaware of this legal situation, as evidenced by the awards, which clearly indicate that they were applying the VCLT. This issue typically did not explicitly appear in the proceedings, as usually both the respondents and claimants seemed to also presume that the VCLT did apply after all.Footnote 29 This was the case in disputes against Slovakia or the Czech Republic brought on the basis of 1990 BIT with the UKFootnote 30 (A11Y, Anglia, Busta), the 1990 BIT with GermanyFootnote 31 (Binder, JSW Solar), or the 1992 BIT with the NetherlandsFootnote 32 (Eastern Sugar, Oostergetel, Achmea), as the Czech and Slovak Republics were only bound by the VCLT in 1993. The same issue related to the disputes against Poland (which acceded to the VCLT in 1990) on the basis of the 1988 BIT with AustriaFootnote 33 (Strabag) and 1987 BIT with Belgium and LuxembourgFootnote 34 (GPFFootnote 35).
Conversely, the only cognizant approach in this respect in the cases involving intra-EU BITs was applied by the tribunal in EURAM on the basis of the 1990 Czechoslovakia-Austria BIT.Footnote 36 Considering the argument of Slovakia that the VCLT was not applicable and that only the corresponding CIL, with respect to Articles 30 and 59 VCLT (but not Article 65), being applicable, the Tribunal referred to the exchange of diplomatic notes between the States concerned confirming the succession of the BIT in 1994. This assertion allowed it to recognise that the BIT was concluded in 1994.Footnote 37 It is to be noted that the approach formulated by the tribunals on the same subject matter issue, where they decide incognizantly on the basis of CIL (such as Eastern Sugar, Oostergetel, Binder, or Achmea (I)), has been recognised and followed by arbitral tribunals directly adjudicating under the VCLT.Footnote 38
When reaching the conclusion that the conflict of norms provisions do not apply as prerequisite of ‘the same subject matter’ is not fulfilled – arbitral tribunals most often did not offer any suggestions on how to solve the problem of a conflict of such norms. Initially, this approach was a consequence of not perceiving existence of such a conflict at all. Tribunals have also continued to maintain such a position after the judgment of the Court of Justice of the European Union (CJEU) in the Achmea case, despite the fact that the CJEU unequivocally confirmed the existence of such a conflict.Footnote 39 Regardless, it is noticeable that since this judgment, arguments concerning the different material scope of the treaties has been increasingly and more clearly evoked by tribunals.Footnote 40 Significantly, arbitral tribunals have pointed to the lack of ISDS in the EU Treaties as the main argument for the difference, while the CJEU explicitly stated that ISDS is contrary to the EU Treaties. A culmination of this legal reasoning was the 2019 ruling in Magyar Farming Company v Hungary in which the Tribunal, while noting that Article 30 VCLT was inapplicable due to the treaties’ differing subject matter,Footnote 41 determined:
[The] [t]ribunal is not aware of the existence of, provisions in the VCLT or of norms of customary international law that would govern the resolution of possible conflicts between successive treaties that do not share the same subject matter.Footnote 42
In conclusion, the case law of arbitral tribunals makes it necessary to analyse Article 30 VCLT’s ‘same subject matter’ formulae in more detail. As it turns out, an element of this provision, overlooked even in the commentaries to the Convention,Footnote 43 may be a key argument for rejecting application of this rule. This position, if comprehensively applied, could prevent any practical application of Article 30 VCLT and its corresponding CIL. Thus, from a systemic perspective, this line of interpretation should not be upheld,Footnote 44 as it is difficult to assert that this potential interpretation of CIL presented by the arbitral tribunal would be followed by States or other international court and tribunals. The fact that approach of investment arbitration was clearly linked with the defense of its own jurisdiction by the tribunals has created circular or, perhaps, opportunistic lines of argumentation. This position, however, has been poorly embedded in general international law, which leads to a conclusion that its significance outside the framework of international investment law is limited. An alternative position would be recognising that this line of interpretation has led to significant gaps in conflict of rules under general international law.Footnote 45
3 Subsequent Agreements in Relation to the Interpretation of Treaties
The second point under scrutiny concerns the subsequent agreements in relation to the interpretation of treaties regulated by Articles 31 (3)(a) VCLT. This issue was recently elaborated by the ILC in its 2018 Draft Conclusions on Subsequent Agreements and Subsequent Practice in Relation to the Interpretation of Treaties.Footnote 46 Articles 31(3)(a) VCLT provide:
There shall be taken into account, together with the context:
any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions.Footnote 47
As has been stated by the ILC,Footnote 48 and previously by international courts,Footnote 49 tribunals,Footnote 50 and investment tribunals,Footnote 51 the abovementioned rules also apply under customary law. Thus, these norms regarding interpretation could also be the object of interpretation as customary rules.Footnote 52 Such an approach has been applied by the WTO, where Article 31(3)(c) VCLT was considered to be custom in the Measures Affecting the Production and Sale of Clove Cigarettes case.Footnote 53 Concerning the most-favoured nation (MFN) clause, the whole of Article 31 VLCT as a customary rule has also been scrutinised in investment arbitration jurisprudence.Footnote 54
Already in 1966, the ILC had confirmed that the joint intention of parties, which underpins the conclusion of a treaty, has a particular authority when identifying the meaning of that treaty, even after its conclusion.Footnote 55 This was in line with the ICJ’s jurisprudence.Footnote 56 Regarding the legal effects of subsequent agreements, it is worth noting, first of all, Conclusion 7(1) adopted by the ILC, which states that
Subsequent agreements and subsequent practice under article 31, paragraph 3, contribute, in their interaction with other means of interpretation, to the clarification of the meaning of a treaty. This may result in narrowing, widening, or otherwise determining the range of possible interpretations, including any scope for the exercise of discretion which the treaty accords to the parties.Footnote 57
In the commentary to this provision, the Commission recognised, inter alia, the possibility for the parties to depart, by a subsequent agreement, from the ordinary meaning of the words of a treaty and to give them special meaning within the context of Article 31(4) VCLT. A departure from the ordinary meaning of words may be particularly justified if it is considered that the parties, in concluding the treaty, intentionally wished to give them an evolving meaning or content, by using general expressions, to take account of developments in international law.Footnote 58
The ILC’s interpretation of Article 31(3)(a)-(b) VCLT can be compared to the position of arbitral tribunals on declarations or joint interpretations formulated by the parties to the treaty. One of the most famous examples of a subsequent agreement in international investment law is the interpretative note by the North American Free Trade Agreement (NAFTA) Commission,Footnote 59 which narrowed the scope of Article 1105 NAFTA. In particular, the arbitral tribunal in Pope & Talbot v Canada stated that ‘were the Tribunal required to make a determination whether the Commission’s action is an interpretation or an amendment, it would choose the latter’.Footnote 60 However, the tribunal found that this issue was not relevant as, regardless of the legal qualification of the note, its previous decision was consistent with it.Footnote 61 In the Methanex case, by contrast, the arbitral tribunal cited Oppenheim’s position according to which authentic interpretation of the parties to the treaty ‘overrides the ordinary principles of interpretation’.Footnote 62
Besides the NAFTA Tribunal’s approach, which was based on an explicit clause in NAFTA, the issue of joint interpretation as a subsequent agreement has featured significantly in disputes concerning intra-EU BITs and intra-EU applications of the ECT.Footnote 63 What is striking in the practice of tribunals in intra-EU disputes is that the readiness to accept any significance of such an interpretation is very limited, which differs from the position of the NAFTA tribunals. In the fundamental Eskosol decision, the Tribunal evaluated whether the Declaration 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 of 22 EU-Member States (EU Declaration)Footnote 64 was a subsequent agreement under Article 31(3)(a) VCLT.Footnote 65
The Tribunal criticised the lack of detailed justification in the EU Declaration, although it did not explain its source for such a requirement.Footnote 66 It referred to the ILC’s works, although, surprisingly, not to those related to subsequent agreements, but instead its work related to reservations.Footnote 67 Consequently, the Tribunal decided that the EU Declaration cannot be considered as a subsequent agreement as it does not refer to any particular provision of the ECT.Footnote 68 Following the judgment of Singapore Court of Appeal in the Sanum case,Footnote 69 the Eskosol Tribunal cited an excerpt from the 1966 Commentary of the Draft Articles on the Law of Treaties to assert that subsequent agreements are only intended to clarify the findings of the treaty negotiations.Footnote 70 This element requires closer inspection as it has been repeated on several occasions by arbitral tribunals interpreting Article 31(3)(a) VCLT.Footnote 71
As a matter of fact, a broader citation of the ILC’s commentary in this respect does not in any way prejudge the accuracy of this position but rather (contrary to the conclusions drawn by the arbitral tribunals) emphasises in particular the significant nature of the subsequent agreements.Footnote 72 Furthermore, it is worth noting another extract from this decision: ‘VCLT Article 31(3)(a) is not, however, a trump card to allow States to offer new interpretations of old treaty language, simply to override unpopular treaty interpretations based on the plain meaning of the terms actually used’.Footnote 73 This interpretation was subsequently cited by arbitral tribunals as rationale of their position in this respect, even when evaluating different legal situations.Footnote 74 Thus, in the view of the Tribunal in Eskosol, a subsequent agreement ‘may “corroborate” or “support an interpretation that has already been determined by other methods,” such as “the objective elements listed in articles 31 and 32 of the Vienna Convention,” but it cannot override the application of those elements’.Footnote 75 Finally, when the rights of individuals are impacted purportedly on the basis of general principles of law or CIL, the Eskosol decision supports the need for limitations upon the interpretative influence of subsequent agreements.Footnote 76 This position had already been presented by the Enron tribunal; according to which, no new interpretation of an International Investment Agreement (IIA) can ‘affect rights acquired under the Treaty by investors or other beneficiaries’.Footnote 77
Following the Eskosol decision, this restrained approach to subsequent agreements concerning the EU question (or so-called intra-EU objection) has been presented by other tribunals. According to the Addiko Bank Tribunal, EU member States do not have the right to interpret the TFEU, as this competence has been entrusted exclusively to the Court of Justice of the EU.Footnote 78 The Strabag Tribunal stated that ‘[f]rom the text of Article 31(3) VCLT, it is evident that such “extrinsic” elements, while informative to the context of a treaty, cannot be used to rewrite the ordinary meaning of the text of the treaty under interpretation’.Footnote 79
In Muszynianka, subsequent agreements relating to interpretation, although recognised as CIL,Footnote 80 were considered as ‘merely one element’ that ‘[is] not an exclusive and dispositive method of treaty interpretation’.Footnote 81 Thus, the influence of subsequent agreements upon interpretation could be ruled out when, in the opinion of tribunal, it explicitly flows from the ordinary meaning.Footnote 82 The Muszynianka decision is extraordinary in this context as it totally ignored the bilateral declaration of the BIT’s State Parties, which the Tribunal justified by the fact the declaration was linked to the EU Declaration. Analysis of the latter was sufficient according to the tribunal.Footnote 83
With regard to the relationship between the rights of individuals and interpretation on the basis of Articles 31 (3)(a) VCLT, the Tribunal in Green Power noted that the competence of States in this respect can induce inequality between parties of an arbitral proceeding, as one of them could change – with retroactive effect – the text of the ECT. Nevertheless, in this monumental decision, the Tribunal acknowledged that no such situation arises with respect to the EU Declaration in the context of the relationship between ECT and the EU law.Footnote 84 Thus, what was crucial for the Green Power Tribunal was the interpretations made by the respondent and the investor’s home-State, and not that this interpretation was not confirmed by all the parties of both applicable multilateral treaties.Footnote 85 Furthermore, the Tribunal, on one hand, seemed to support the position that subsequent agreements could not be applied retroactively if they related to the rights of individuals, but, on the other hand, stated that ‘Spain’s offer to arbitrate under the ECT is not applicable in intra-EU relations’ without indicating any concrete moment in time when this offer’s non-applicability took effect.Footnote 86
As in the case of Article 30 VCLT – investment arbitration, in several cases, has interpreted Articles 31 (3)(a) VCLT and its customary counterpart. At least in one case (Strabag), only CIL was the subject of interpretation, as the VCLT was not applicable in the case. In this context it is to be noted that investment arbitration informs possible interpretations of Articles 31 (3)(a) in at least two areas – retroactive application and, connected with it, the impact on the rights of individuals.Footnote 87 Thus, although ‘the parties to a treaty own the treaty and can interpret it’,Footnote 88 their freedom in respect to international investment law is met with significant scrutiny and reticence from the arbitral tribunals. In fact: ‘[t]he only situations in which tribunals have been bound to follow the interpretive directions of the state parties are where the relevant treaty goes beyond the general VCLT rules and specifically provides for joint interpretations to be binding’.Footnote 89
Such an approach tends to look at IIAs not only as a transaction between the parties,Footnote 90 which have full discretion as to how to apply the mutually agreed upon provisions, but as an agreement which relates also to other beneficiaries. As Anthea Roberts puts it:
Instead of privileging the rights and powers of states and state-to-state tribunals (as in the first era) or investors and investor-state tribunals (as in the second era), we should move into a third era based on the ideas that investment treaty rights are granted to investors and home states on an interdependent basis.Footnote 91
The problem with this line of reasoning is twofold. First, under international law, the State parties have the principal competence in the interpretation of treaties. This is not in any way changed by the fact that the interpretation can have a positive influence on their position in a dispute. It is still ‘their’ treaty.Footnote 92 Second, certainly, there is an accepted domestic practice of retroactive change in the interpretation of statutesFootnote 93 or retroactive application of new precedents. The latter point has been evaluated by domestic constitutional and international courts and was, in principle, not found as violating the rule of law.Footnote 94 Thus, domestic law does not have to justify its departure from the VCLT or customary law rules on interpretation. The powers of State parties to interpret ‘their’ treaties is well-represented in the China–Australia Free Trade Agreement (ChAFTA)Footnote 95 which allows parties, in certain situations, to agree on whether the case pending before arbitral tribunal can be adjudicated due to the treaty constraints in this respect.Footnote 96 As a result, there is an evident tension between the right of State parties regarding treaty interpretation under general international law and the emerging tendency for this competence to be limited with respect to rights of investors.
4 Conclusions
The two mentioned rules, on the conflict of norms and the influence of subsequent agreements regarding the interpretation of treaties, exemplify that investment arbitration can not only impact the interpretation of general international law concerning the responsibility of States, but they also have the potential to influence the law of treaties.
With respect to the first issue, it is to be noted that there exists a coherent line of tribunals decisions concerning relations between international investment agreements and EU treaties, which narrowly interpret the issue of ‘the same subject matter’ for possible conflicting treaty norms. Such an approach is particularly remarkable as it differs from the position taken by the ILC both in its work on the law of the treaties as well as in its study on the fragmentation of international law. Such a position differs also from the interpretations contained in two GATT/WTO decisions. Nevertheless, as this issue has not generally been the subject of scrutiny in international courts and tribunals in inter-State cases, investment awards cannot be disregarded as they represent the most extensive jurisprudence in this area. Still, from a systemic perspective, investment tribunals’ line of argumentation leads clearly to fragmentation as they approach investment treaties as almost self-contained regimes, which are resilient to a conflict of rules analysis unless the two applicable treaties are almost identical. It is remarkable that even a year after publication of the ILCs report on fragmentation, whose main message was ‘that the emergence of special treaty-regimes (which should not be called “self-contained”) has not seriously undermined legal security, predictability or the equality of legal subjects’,Footnote 97 investment tribunals, at least with respect to the application of conflict of norms rules, have initiated a process leading towards the opposite direction. In this respect, in my view, investment awards which interpret not only rules of the VCLT, but also their corresponding CIL rules, seem to have pro futuro limited value and should rather be considered to have incidental significance for general international law.Footnote 98 It seems that this approach taken by investment tribunals has been largely dictated by the need to defend their jurisdiction. What is striking is the number of cases in which tribunals were convinced that they could apply the VCLT when, in fact, only CIL was applicable in the respective case. This approach can also prove that at least with respect to rules with double force (treaty – customary) under international law, the method of interpretation can be similar to a large extent.
Regarding the VCLT and CIL rules on subsequent agreements several observations can be made. First, the ease with which arbitral tribunals reject interpretations formulated by States through subsequent agreements is striking. The analysis of these agreements is often rudimentary, absent, or the tribunal simply refers to other rulings – indicating that it considers these interpretations as their own – without sometimes paying attention to the difference in legal situations between the cases whose jurisprudence it cites. Second, tribunals have set out detailed requirements for the formulation of a declaration, without explaining the source for these requirements. Third, there is a considerable difference between tribunals referring to joint interpretations of State Parties, as envisaged by the treaty as binding (such as with NAFTA), and the tribunal assessing such interpretations through the lens of the VCLT or corresponding customary rules. With respect to the latter, tribunals tend to diminish the role of subsequent agreements by underlining the ‘taken into account’ formulae of Article 31(3) VCLT. Due to the laconic nature of the provisions of most of the investment agreementsFootnote 99 (in particular so called ‘old-generation’ treaties),Footnote 100 this approach enables investment tribunals, through the interpretation of the ‘ordinary meaning’ of terms and the ‘object and purpose’ of a treaty, to maintain its far-reaching interpretative power.
What follows from these observations is, to my mind, the necessity of bringing consistency between the case law of international investment arbitration tribunals and general international law, both as regards conflict of norms as well as of the role subsequent agreement and subsequent practices. Thus, there is a continuous need for more grounding of investment jurisprudence in the realm of general international law.Footnote 101
Still, investment arbitration can inform general international law in at least in two areas, which were not explicitly articulated by the ILC in its 2018 Conclusions, the retroactive application of interpretations and, connected with it, the impact on the rights of individuals. Certainly, in this respect, investment arbitration leans towards interpretations which restrain the rights of State parties as the ‘masters’ of treaties.Footnote 102 However, the line between rights of State parties and rights of the individual beneficiaries of the treaties (ie investors) has not yet crystallised. Thus, this trend coming from investment arbitration should be juxtaposed with the approach of other international tribunals and domestic constitutional courts’ jurisprudence.