International law … may not contain, and generally does not contain, express rules decisive of particular cases; but the function of jurisprudence is to resolve the conflict of opposing rights and interests by applying, in default of any specific provisions of law, the corollaries of general principles, and so to find … the solution of the problem.Footnote 1
I. THE CONCEPT OF INTERNATIONAL INVESTMENT LAW
A. Has Lex Specialis Become the General Law?
Delivering its recent judgment in the Diallo Case, the International Court of Justice noted that:
… in contemporary international law, the protection of the rights of companies and the rights of their shareholders, and the settlement of the associated disputes, are essentially governed by bilateral or multilateral agreements … In that context, the role of diplomatic protection somewhat faded …Footnote 2
The point is illustrated starkly by the docket of the International Court itself. Diallo is the first occasion on which the customary international law on the treatment of the property of an alien investor has reached the International Court since Barcelona Traction in 1970.Footnote 3 The International Law Commission also drew attention to the greater importance of treaty law in this field when it concluded its work on diplomatic protection in 2006.Footnote 4 Is it then the case, as the Tribunal in CMS v Argentina suggested, that ‘[t]he fact is that lex specialis in this respect is so prevalent that it can now be considered the general rule.’?Footnote 5
The enormous proliferation of bilateral investment treaties (BITs), and of arbitrations under them, has masked a more fundamental issue as to the relationship between the primary substantive protections in investment treaties and general international law.Footnote 6 This question has two aspects. First, to what extent may or must the arbitral tribunal in an investment treaty dispute refer to general international law in interpreting the treaty before it? Secondly, to what extent has the elaboration and application of investment treaties itself contributed to the development of general international law?
It cannot be said that there is as yet any clear consensus on the matter in arbitral practice. Some tribunals continue to take as their point of departure the autonomous position of each BIT, pointing out, in the terms of the International Tribunal for the Law of the Sea, that:
… the application of international law rules on interpretation of treaties to identical or similar provisions of different treaties may not yield the same results, having regard to, inter alia, differences in the respective contexts, objects and purposes, subsequent practice of parties and travaux préparatoires.Footnote 7
From this, they draw the inference that:
Each tribunal is sovereign, and may retain … a different solution for resolving the same problem …Footnote 8
The consequent licence to decide each case differently has notoriously led to differently constituted tribunals reaching diametrically opposite conclusions on the same facts.Footnote 9 It has also led to tribunals coming to diametrically opposite conclusions on the same issue, even where they share members in common.Footnote 10
One of the most serious divisions of view between tribunals to emerge to date crystallized in September 2007 with the publication of the Annulment Committee's decision in CMS v Argentina.Footnote 11 The Committee found that the Tribunal had applied the treaty to Argentina's defence of necessity ‘cryptically and defectively,’Footnote 12 but there was no manifest excess of powers, and therefore no basis on which it could intervene. In the result, therefore, wholly inconsistent awards against the same State on the same issue remain undisturbed. In the process, the Committee may have lit a touch-paper under the current operation of the investment arbitration system as a whole.
Quite apart from the practical urgency of the issue, there are other reasons why an evaluation at this stage of the place of investment treaty law within the larger frame of general international law may be both possible and timely. It is 20 years since the first arbitration claim under an investment treaty was registered.Footnote 13 The substantial volume of litigation since then has given rise to a ‘first wave’ of investment-treaty jurisprudence, making it possible to begin to analyse and distill the common principles which emerge from various fact patterns confronted in the modern case law, and to restate them.Footnote 14
At the same time, responding to a wider concern about the fragmentation of international law, the International Law Commission delivered in 2006 its seminal report offering both a theoretical construct and a set of practical techniques aimed at harmonizing the disparate parts of international law.Footnote 15 A key element of that report was the adoption and elaboration of a principle of ‘systemic integration’ in treaty interpretation.Footnote 16 This report (in which the author collaborated) provides a framework of legal reasoning, which may be applied to the particular context of investment law.
This article advances three insights into the relationship between the primary substantive rules of investment treaties and general international law. The first point is that the relationship is symbiotic. The obligations owed by States as a matter of custom are not necessarily identical to the related obligation undertaken by treaty, but nor is the treaty to be regarded as inhabiting its own watertight compartment. Rather the content of the treaty obligation may be informed by general international law. In turn, the promulgation of the treaty obligation, and its application by arbitral tribunals, may inform the progressive development of general international law. Thus, it is here contended that the current debate about the need for a doctrine of precedent in investment arbitration is misguided.Footnote 17 The question is not one of a binding obligation to achieve consistency of awards inter se. Rather it is that of the proper relationship between the treaty obligation under consideration and general international law. This relationship is not static. It is part of the dynamic process of the evolution of international law.
Secondly, merely to identify the law applicable to an obligation as international law will not suffice. Rather, the extent and manner in which general international law is applied in the context of a treaty obligation will be determined by a structured process of treaty interpretation. Thirdly, the main route through which practice under investment treaties contributes to the development of general international law is through the elucidation of general principles of international law, which illuminate the province of international law in its supervision of the conduct of national State officials. The consequence is to dress that rather neglected third source of international law, ‘general principles of law common to civilised nations’ in the new garb of general principles of international law.
In order to make good each of these propositions, it will be necessary to do three things: (a) to explore and apply the principle of systemic integration to the interpretation of the substantive obligations in investment treaties (Part I); (b) to examine the evidence for the emergence of general principles in the specific context of the key fair and equitable treatment standard and to demonstrate the fallacy of an approach based solely upon applicable law rather than interpretation, by reference to the recent ‘doctrine of necessity cases (Part II); and (c) to consider the wider contribution of investment treaty practice by reference to the classical sources of international law (Part III). But first it is necessary to say a little more about the development of international investment law, since that provides an essential context for what is to follow.
B. Dissent in the Development of Investment Law
The historical context reveals three deep-rooted problems in the relationship between investment treaties and general international law. First, the essentially contested nature of so many of the rights in custom, even at their pre-War High Noon; secondly, the fact that States often entered into investment treaties precisely in order to remedy perceived limitations in the protections afforded by custom; and thirdly, the fact that the overwhelming majority of State practice in this field in the last few decades has been through the medium of treaty-making, starving custom of independent progressive development.
Until the advent after the Second World War of the modern BIT, with its rights of direct recourse for investors in investor-State arbitration, the primary means by which the protection of the property and investments of aliens was achieved was through inter-State claims made upon the exercise of diplomatic protection. In this, the origin of the rights claimed is found in general international law, namely in the:
… general requirements of customary international law, such as those which impose on a state international responsibility for denial of justice to aliens, or which require it to observe in its treatment of aliens certain minimum international standards.Footnote 18
In the latter part of the 19th century and the first half of the 20th century, a very extensive body of jurisprudence and doctrine built up concerning the minimum standard for the treatment of aliens generally, and the standard of denial of justice in particular.Footnote 19 The high-water mark of this analysis may be found in the Harvard Draft of 1929 on ‘The Law of Responsibility of States for Damage done in their Territory to the Person or Property of Foreigners.’Footnote 20
The evolution of the standard through diplomatic protection had the consequence that it could only be successfully invoked by the home State if the affected person had himself first exhausted local remedies,Footnote 21 and then under strict rules as to nationality of claims. In practice, save in cases of a refusal to investigate or prosecute, the cases on the international minimum standard and denial of justice were almost always concerned with alleged failures in the judicial system of the host State. Any failures in administrative decision-making would not give rise themselves to an international claim, since they would first have had to be tested by the investor in the local courts.
In any event, even at the time this standard was being developed and refined through the mixed claims commissions of the inter-war period, it was attracting increasing dissent, notably from Latin American States who opposed the imposition of a standard beyond that of national treatment,Footnote 22 and who would have limited a denial of justice to the refusal of access to a court.Footnote 23
Since the Second World War, it has proved even more difficult to achieve multilateral consensus on the content of these principles as a matter of custom. In part this is as a result of the extent of the disagreement between States which arose in the specific context of investment law. But it is also a reflection of the general difficulty of reaching a consensus on the primary rules of the delictual responsibility of States. This led to their exclusion from the International Law Commission's work on Draft Articles on State Responsibility. The Commission abandoned the earlier ‘enormously ambitious’Footnote 24 draft code of Garcia Amador on the protection of aliens as unlikely to achieve a sufficient degree of agreement by States. Similarly, the more recent work of the International Law Commission on diplomatic protection has been expressly limited to secondary rules.Footnote 25
However, these limitations on the development of custom have always been expressed to be without prejudice to the right of States to accept binding standards by treaty. Thus the International Law Commission's draft articles on both State responsibilityFootnote 26 and diplomatic protection are expressed to be without prejudice to the right of States to develop special rules of international law in particular areas. These include, in the case of diplomatic protection, treaty provisions concerning the protection of investments.Footnote 27 Despite, or perhaps rather because of, the huge rise in BITs, the customary international law rules relating to the non-contingent standards of treatment of investments have not developed in the post-War period to the same extent through the practice of diplomatic protection.
No better progress has been achieved with any attempt to achieve multilateral agreement by treaty on the content of such standards.Footnote 28 The best that could be done at Havana in 1948 was a draft agreement to make recommendations for subsequent multilateral or bilateral agreements for just and equitable treatment of investments. The Havana Charter itself never came into force. The General Agreement on Tariffs and Trade 1947 (the GATT), which then formed the cornerstone of world trade law, never extended to investments, until the (very limited) Agreement on Trade-Related Investment Measures (or ‘TRIMS’) was concluded some 50 years later in 1994.Footnote 29
It was against this background that Hermann Abs and Lord Shawcross developed in 1959 the template for a bilateral investment treaty, which came to have so much impact on subsequent treaty practice.Footnote 30 In so doing, the drafters sought to steer a careful path on the substantive standards. They proposed a text which might be acceptable as derived from the consensus represented by customary international law, but clothed in language which avoided reigniting some of the pre-War controversies by offering new scope for flexibility. Nowhere is this more apparent than in the new language of the non-contingent standard. Gone is the old language of minimum standard of treatment and denial of justice. In its place is found:
Each Party shall at all times ensure fair and equitable treatment to the property of the nationals of the other Parties. Such property shall be accorded the most constant protection and security within the territories of the other Parties and the management, use and enjoyment thereof shall not in any way be impaired by unreasonable or discriminatory measures.Footnote 31
The result, as Schwarzenberger observed at the time, ‘presents an imaginative attempt to combine the minimum standard with the standard of equitable treatment.’Footnote 32
But attempts to translate these formulations into multilateral agreement were still-born. The OECD draft Convention on the Protection of Foreign Property of 1967Footnote 33 contained, in Article 1(a), substantially the same text. The drafters of this Convention were in no doubt that:
The phrase ‘fair and equitable treatment’, customary in relevant bilateral agreements, indicates the standard set by international law for the treatment due by each State with regard to the property of foreign nationals … The standard required conforms in effect to the ‘minimum standard’ which forms part of customary international law.Footnote 34
However, the Convention was never opened for signature, and came to be seen as embodying the perspective of capital-exporting countries. Two decades later, the Draft United Nations Code of Conduct on Transnational Corporations reached no consensus on the content of the non-contingent standards. The last negotiating text of 1983 contained more un-agreed text in square brackets than finalized text.Footnote 35
The Multilateral Agreement on Investment (MAI), proposed by the OECD in 1995, would have provided a general protection of ‘fair and equitable treatment and full and constant security. In no case shall a Contracting Party accord treatment less favourable than that required by international law.’ The MAI was seen by the OECD as largely a harmonization exercise, and an attempt to address the fragmented nature of investment protection through BITs. However, it provoked intense opposition from NGOs on the grounds that it would weaken the regulatory capacity of host States in favour of investor protections.Footnote 36 Efforts to conclude it collapsed in 1998.
This picture of the failure to reach multilateral agreement on an acceptable content of investors' rights is to some extent qualified by the emergence of significant pluri-lateral agreements, involving both capital-exporting and capital-importing States, which do contain such protections. Prominent amongst these are NAFTAFootnote 37 and the Energy Charter Treaty.Footnote 38 The Lomé IV Treaty of 1990, between ACP and European Union States,Footnote 39 and the ASEAN Treaty of 1987Footnote 40 also contain mutual guarantees of fair and equitable treatment. The World Bank adopted such a standard in its Guidelines on the Treatment of Foreign Direct Investment of 1992.Footnote 41
Nevertheless, bilateral treaties remain the overwhelmingly dominant mode of investor protection. The extent of State practice represented by such treaties is both connected to, and a development from, the earlier position reached in custom. Today, at least, if there is to be a meaningful consideration of international law in relation to the treatment of aliens, it can only be done by reference to a continuum of development from the pre-War custom into post-War treaty negotiation and application. So it remains of vital importance to consider how such treaties fit, and fall to be interpreted, within the general framework of international law.
C. Systemic Interpretation of Investment Treaties
This article starts from the proposition that investment treaties are not self-contained regimes.Footnote 42 International law is a legal system, and investment treaties are creatures of it and governed by it.Footnote 43 The Tribunal in the first ICSID treaty arbitration, Asian Agricultural Products Ltd v Republic of Sri Lanka,Footnote 44 explained that a BIT:
… is not a self-contained closed legal system limited to provide for substantive material rules of direct applicability, but it has to be envisaged within a wider juridical context in which rules from other sources are integrated through implied incorporation methods, or by direct reference to certain supplementary rules, whether of international law character or of domestic law nature.Footnote 45
At least in the context of investment law, however, it is not sufficient simply to determine international law as the applicable law. There are difficult issues of choice of law in investment cases, as between host State law, the law applicable to the parties' private law obligations, and public international law. Article 42 of the ICSID ConventionFootnote 46 notoriously defers these difficult questions by requiring the tribunal, in the absence of express choice, to apply host State law ‘and such rules of international law as may be applicable.’ But, where the obligations in question are those created under international law, as in the case of treaty obligations, there is no doubt that the applicable law is potentially international law as a whole. As the Annulment Committee in MTD v Chile put it: ‘ … the Tribunal had to apply international law as a whole to the claim, and not the provisions of the BIT in isolation.’Footnote 47
Thus, investment law stands at the outset as in a different position from trade law, where the hermetically sealed nature of the system prior to the Marrakesh Agreements made it important to make the case for a broader set of potentially applicable norms.Footnote 48 To be sure, the distinction between jurisdiction and applicable law (a key distinction in both private and public international law) still remains important in investment arbitration. Investment tribunals, like many public international arbitral tribunals,Footnote 49 do not have plenary jurisdiction. They only have as much jurisdiction as has been vouchsafed to them by the parties. Where the source of that consent on the part of the host State is an investment treaty, it will be the dispute resolution clause in the treaty itself which will delimit the extent of the matters which the tribunal is competent to decide. But that does not of course proscribe the law which the tribunal may apply to determine those issues. The requirement of Article 42 to have regard to the whole of international law thus operates at the applicable law stage, whatever may be the limitation on the tribunal's jurisdiction.
But international law must be introduced into the analysis of a claim under an investment treaty in the first place through the medium of treaty interpretation. The framework for such an exercise is provided by Articles 31 and 32 of the Vienna Convention on the Law of Treaties 1969.Footnote 50 These articles state rules of customary international law on treaty interpretation.Footnote 51 They have also been repeatedly accepted by investment arbitral tribunals as constituting rules of interpretation which are binding on them in the interpretation of investment treaties, whether by virtue of being directly binding on the parties to the BIT as treaty rules, or as customary international law.Footnote 52 They of course include, in Article 31(3)(c), a requirement to refer to ‘relevant rules of international law applicable in the relations between the parties.’
Nevertheless, the starting point of any analysis must be the ordinary meaning of the terms, as required by Article 31(1). Sometimes this may indicate a particular formulation as lex specialis, which requires direct application.Footnote 53 But, in the case of the general test for fair and equitable treatment, it may result in little more than an exchange of synonyms. Thus, the Tribunal in MTD began by quoting the Oxford Concise English Dictionary and observed: ‘In their ordinary meaning, the terms “fair” and “equitable” … mean “just”, “even-handed”, “unbiased”, “legitimate”.’Footnote 54
Determination of object and purpose may be deceptive. Investment treaties commonly contain preambular statements that their purpose is to promote investment. But this cannot simply be conflated with a general preference for the interests of the investor over those of the host State.Footnote 55 The overall objective in fact requires a balanced approach since:
… an interpretation which exaggerates the protection to be accorded to foreign investments may serve to dissuade host States from admitting foreign investments and so undermine the overall aim of extending and intensifying the parties' mutual economic relations.Footnote 56
Reference to travaux préparatoires under Article 32 is unlikely to assist in this field. It is rare for bilateral negotiations to produce the kind of explanatory reports, or official records of plenary debates which are characteristic of multilateral negotiation.Footnote 57 When the confidential working papers on the negotiation of NAFTA were produced in Pope & Talbot, they did little to illuminate the matters in dispute.Footnote 58 The transmittal statements by which domestic approval for such treaties is sought by the signatory States, to which reference is sometimes made,Footnote 59 are merely unilateral statements and not travuax préparatoires.Footnote 60 They are therefore more likely to be relevant as evidence of State practice for the purposes of a customary rule, than as to the meaning of a treaty text.Footnote 61
The parties to investment treaties rarely create a mechanism through which they may make binding subsequent interpretative decisions.Footnote 62 Further, the hybrid character of investment treaties, which gives the prime role in enforcement to private investors, leaves little room for subsequent inter-State practice in their application.Footnote 63
Thus the assistance to be gleaned from the requirement in Article 31(3)(c) to take into account ‘other rules of international law applicable in the relations between the parties’ becomes of some importance. On one level, this requirement states little more than a truism—no treaty can exist in isolation from general international law.Footnote 64 But, in the case of BITs, there may be particularly compelling reasons to refer to general international law in interpretation. The treaty framers often consciously sought not to go beyond obligations which were thought to reflect the current state of international law.Footnote 65 Their purpose was to enhance the mechanisms for the protection of rights, rather than to extend the rights themselves.
Article 31(3)(c), as the author has submitted previously in the Quarterly,Footnote 66 embodies a principle of systemic integration in treaty interpretation, which may be articulated as a presumption with both positive and negative aspects:Footnote 67
(a) positively that the parties are to taken ‘to refer to general principles of international law for all questions which [the treaty] does not itself resolve in express terms or in a different way’;Footnote 68 and
(b) negatively that, in entering into treaty obligations, the parties intend not to act inconsistently with generally recognised principles of international law or with previous treaty obligations towards third states.Footnote 69
Most of the cases in the investment field where tribunals have looked beyond the four corners of the treaty have involved the application of principles of general international law. This may also be seen as an application of the notion of ‘special meaning’ in Article 31(4). As Judge Higgins observed in Oil Platforms, ‘the key terms “fair and equitable treatment to nationals and companies” … are legal terms of art well known in the field of overseas investment protection.’Footnote 70 In this context, a special meaning attributed to the term, which the parties can be taken to have intended to employ, would itself derive from a general international law practice, unless it could be established that the parties to the particular treaty had a specific and different meaning in mind.
The importance of general principles of international law in this process is not because of their overriding character, since international law reserves for overriding customary rules the special category of jus cogens. Otherwise, it must be accepted that a treaty can of course derogate from custom, provided that it does so expressly. Rather, the significance of such rules is that they perform a systemic or constitutional function in describing the operation of the international legal order, and in establishing a common set of underlying principles which inform it.
Further, BITs normally say little about the secondary rules as to the conduct of States which are traditionally the province of the law of State responsibility: attribution; the nature and breach of an international obligation; and circumstances which may preclude wrongfulness. Investment treaties are typically restricted to a short catalogue of primary rules of investment protection, together with the provision for dispute resolution through arbitration. Part One of the ILC Draft Articles on State Responsibility applies generally to the international responsibility of a State, whether owed to other States, or where the primary beneficiary is not a State. Its provisions may therefore apply to breaches of investment treaties, even where claimed at the suit of a private investor.Footnote 71 In contrast, then, to a complex multilateral treaty in other spheres of international law,Footnote 72 BITs appear more than usually dependant upon their wider context.
The tribunal's task in this regard is to establish the contemporary content of international law, which may therefore be influenced by both older doctrine and more recent developments.Footnote 73 Given the relatively recent conclusion of most investment treaties, it has not been necessary in this field to deal to the same extent with the problem of inter-temporality in treaty interpretation.Footnote 74 However, the central concepts in investment treaties are mobile, in the sense that, by using the generally adopted terms, the parties' intention is to submit to the evolving meaning in international law, and not to confine themselves to their own idiosyncratic definition, fixed at the time of conclusion of the treaty.
To what extent, then, is a coherent jurisprudence interpreting the primary rules of investment treaty law actually emerging? It is to this issue that this article now turns.
GENERAL INTERNATIONAL LAW IN INVESTMENT ARBITRATION
In examining the impact of custom on the application of investment treaties, this paper draws primarily on the construction of the key non-contingent standard of ‘fair and equitable treatment’. A similar investigation could be undertaken in relation to expropriation, where the treaty language commonly follows closely the international law standard.Footnote 75 But fair and equitable treatment offers some particular insights for the theme of this paper. As shown above, the term is undoubtedly the product of the bilateral investment treaty process, adopting consciously new language to what had gone before. But its function has, from the start, been linked to the pre-existing general international law. The almost universal adoption of the standard in investment treaties, and its application to a wide range of claims in arbitration in the last ten years, makes it an object lesson in the interplay between treaty and custom.
A. Emergence of a Body of Concordant Practice
1. The process of contemporary jurisprudence
The practical application of the right to fair and equitable treatment under BITs is a very modern story. As recently as 1999, Vasciannie concluded an exhaustive study of this treaty standard by bemoaning ‘the paucity of jurisprudence’.Footnote 76 BITs, he observed, ‘are yet to generate a substantial flow of international litigation, and even where litigation has occurred, the fair and equitable standard has not been decisive in the proceedings.’Footnote 77 In these circumstances, the most that could be said on content of the standard was that it barred ‘host countries from treating foreign investors unfairly and inequitably.’Footnote 78
The experience of the last decade has turned this position on its head. Not only has the standard been frequently the subject of litigation, it has also been outcome-decisive in many such cases, eclipsing in the result the more established right of protection against expropriation.Footnote 79 The cases decided so far fall into two broad categories: (a) treatment of investors by the court of the host State (denial of justice cases); and (b) a much more numerous set of cases dealing with administrative decision-making.
The ‘denial of justice’ cases display an explicit debt to custom.Footnote 80 All three of the cases dealing with the conduct of the host State courts have referred back to formulations of the concept developed in earlier doctrine and jurisprudence and codifications.Footnote 81 Indeed, the formulations of denial of justice found in the Harvard drafts on the responsibilities of States for injuries to aliens proved particularly influential in all three cases. On one level, this is unsurprising. As suggested earlier, the development of custom in this field was shaped particularly by complaints in relation to judicial process, because diplomatic protection was available as a remedy only once local remedies had been exhausted. Therefore the treatment accorded in the course of seeking such remedies became itself the prime focus of complaint once the case had reached the international level. The case books thus provide a rich source of jurisprudence on the working out of the concept.
To some extent this heritage is valuable. It has shown that a ‘governmental authority cannot be faulted for acting in a manner validated by its courts, unless the courts themselves are disavowed at the international level.’Footnote 82 If a national court is to be found to have breached the international standard, it must be for a lack of due process, and not merely for a substantive error in the way in which the law has been applied.Footnote 83 As Paulsson puts it: ‘In international law, denial of justice is about due process, nothing else—and that is plenty.’Footnote 84 But, in other respects, the reference in the cases to pre-Second World War custom, developed in the context of diplomatic protection, has proved curiously inhibiting. There has been a reluctance to refer to international human rights law developed since 1945.Footnote 85 This is despite the fact that the civil and political rights standards speak directly to the concept of fair trial, and have themselves a good claim to customary law status. Moreover, the Tribunal in Loewen was content to include in the substantive standard a requirement to exhaust local remedies,Footnote 86 which it is submitted makes no sense outside the context of diplomatic protection.Footnote 87
In the field of review of administrative decision-making, tribunals have had to develop the scope of the standard without specific guidance from the old law on the treatment of aliens, since claims of this kind could not previously have been brought directly before an international tribunal. Thus far, two types of factors have influenced tribunals. The first set of cases has been concerned with the protection of an investor's legitimate expectations by reference to the law of the host State at the time of investment, together with any specific assurances which the investor received from the agencies of the host State upon which he relied in deciding to invest. The second set of cases is concerned directly with due process in decision-making.
Where does the concept of legitimate expectations come from? The most frequently cited authority for it is a formulation in Tecmed v Mexico.Footnote 88 The dictum imposes a lengthy list of desiderata including that:
The foreign investor expects the host State to act in a consistent manner, free from ambiguity and totally transparently in its relations with the foreign investor, so that it may know beforehand any and all rules and regulations that will govern its investments … to be able to plan its investment …Footnote 89
As one commentator points out:
The Tecmed ‘standard’ is actually not a standard at all; it is rather a description of perfect public regulation in a perfect world, to which all states should aspire but very few (if any) will ever attain. But in the aftermath of the tribunal's correct finding of liability in Tecmed, the quoted obiter dictum in that award, unsupported by any authority, is now frequently cited by tribunals as the only and therefore definitive authority for the requirements of fair and equitable treatment.Footnote 90
The notion of legitimate expectations has, of course, had a particular life within the administrative law of Common Law countries.Footnote 91 It is important to ensure that Anglo-American doctrine alone does not form the basis for a rule of international law.Footnote 92 But the Tribunal in Tecmed was composed of civilian lawyers. On further analysis, the doctrine may in fact express a less extravagant idea than the dictum may suggest. As the Annulment Committee recently reminded us in MTD v Chile,Footnote 93 the concept of legitimate expectations is not an independent concept. It is relevant only to the extent that it applies to the investment treaty's guarantee of fair and equitable treatment. Seen in this way, the idea of legitimate expectations does not in fact impose its own substantive requirement. In fact, it supports the application of host State law, and the liberty of the host State to determine the content of that law.Footnote 94 Rather, it is concerned with due process in administrative decision-making—ensuring the consistent application of the law and enforcing representations by the host State only where these were made specifically enough to the particular investor to justify reliance.Footnote 95
The central concept of due process may also be seen at work in the cases which have applied fair and equitable treatment to other aspects of administrative decision-making:Footnote 96 in prohibiting discrimination between foreign and local investors;Footnote 97 the use of powers for improper purposes;Footnote 98 inconsistency of treatment by different government agencies;Footnote 99 coercion and harassment by State authorities;Footnote 100 and bad faith;Footnote 101 and in requiring some degree of transparency.Footnote 102
None of these factors have been treated by tribunals as a sine qua non. On the contrary, their presence or absence will take on greater or lesser relevance, depending on the facts of the particular case. In this way, the operationalization of the general test in its application to the modern fact patterns presented to tribunals may be compared to the development of the law of civil wrongs, whether in the Common Law after Donoghue v Stevenson Footnote 103 or under the very general provisions of the civil codes.Footnote 104 No one doubts that the legal rule remains that which is stated shortly in terms of great generality. But the case law sheds the considerable light of experience upon how the general rule applies, by isolating the factors which may be relevant in different fact situations. In this way, the law is not simply applied; it is elucidated in ways which assist the work of subsequent tribunals.
By the same token, the cases have begun to identify factors where balancing equitably the legitimate public interests of the State may indicate that the standard has not been breached: in preserving a legitimate scope for regulatory flexibility;Footnote 105 in denying claims where the alleged right is not found in either host State law or international law;Footnote 106 and in upholding administrative decisions which have an objective basisFootnote 107 and do not have a disproportionate impact on the foreign investor.Footnote 108 Tribunals have recognized that they must take the investor's conduct into account as well: BITs ‘are not insurance policies against bad business decisions’;Footnote 109 and the investor has its own duty to investigate the host State's applicable law.Footnote 110
Thus far, the jurisprudence of investment protection may be envisaged as an elaborative enterprise: the working out of the application of the general test in the myriad of different specific contexts which the cases have presented to tribunals. The development of such a common law of a legislative text is a familiar one to lawyers—albeit that in the case of BITs the concern is rather with the common language of numerous separate texts. This is not to say that every award is reconcilable, as the doctrine of necessity cases discussed below demonstrate. But treaty arbitration has, on the whole, resisted the temptation of extreme compartmentalization. Instead, tribunals have regularly borrowed reasoning and interpretations adopted by tribunals hearing cases under different treaties, but with similar language. Schreuer observes: ‘Fortunately the problem of inconsistency is not pervasive. Most tribunals carefully examine earlier decisions and accept these as authority most of the time.’Footnote 111 As a result, an increasingly detailed body of jurisprudence is developing.
What larger significance is to be given to this phenomenon? The tribunal in AES v Argentina gave this explanation:
Each tribunal remains sovereign and may retain, as it is confirmed by ICSID practice, a different solution for resolving the same problem; but decisions on jurisdiction dealing with the same or very similar issues may at least indicate some lines of reasoning of real interest; this Tribunal may consider them in order to compare its own position with those already adopted by its predecessors and, if it shares the views already expressed by one or more of these tribunals on a specific point of law, it is free to adopt the same solution … precedents may also be rightly considered, at least as a matter of comparison and, if so considered by the Tribunal, of inspiration.Footnote 112
One tribunal has recently gone further and suggested that arbitrators have:
… a duty to adopt solutions established in a consistent series of cases … . [and] a duty to seek to contribute to the harmonious development of investment law and thereby to meet the legitimate expectations of the community of States and investors towards the certainty of the rule of law.Footnote 113
The extensive exchange of ideas between tribunals has been facilitated by the wide publication of awards, as well as by scholarly journalsFootnote 114 and committees.Footnote 115 Further, it is no accident that this new jurisprudence has developed in the era of the internet, which has rapidly built a global community of scholars, practitioners and arbitrators exchanging ideas about current developments in the field as they arise.Footnote 116
2. The significance of underlying principle
Are there, nevertheless, general principles of international law which may properly be said to animate and guide the application of the fair and equitable treatment rule, taking the exercise beyond a mere comparative review of solutions adopted by other arbitral tribunals? At this fundamental level, the awards are in disarray. In one view, ‘[t]he terms are to be understood and applied independently and autonomously’Footnote 117 and ‘envisage conduct which goes far beyond the minimum standard’Footnote 118 of the treatment of aliens at customary international law level. Even where the treaty language expressly refers to ‘fair and equitable treatment in conformity with the principles of international law,’ one recent tribunal has found it possible to maintain that this merely sets a floor and not a ceiling on the applicable treaty standard.Footnote 119
The alternative view is that fair and equitable treatment is synonymous with a customary concept of the minimum standard. Sometimes this is made express in the treaty language. Thus, for example, the French prototype guarantees: ‘traitement juste et équitable, conformément aux principes du Droit international’.Footnote 120 The 2004 US model BIT enshrines ‘the customary international law minimum standard of treatment of aliens’,Footnote 121 defining customary international law as resulting from ‘a general and consistent practice of States that they follow from a sense of legal obligation’.Footnote 122 NAFTA Article 1105 requires that: ‘Each Party shall accord to investments of investors of another Party treatment in accordance with international law, including fair and equitable treatment and full protection and security.’ Controversy amongst arbitral tribunals as to whether that form of words left room for a more expansive reading of ‘fair and equitable treatment’Footnote 123 was settled by the NAFTA Free Trade Commission, which issued an interpretation of the article, providing that it ‘prescribes the customary international law minimum standard of treatment of aliens’ and not more.Footnote 124 A similar view has also been taken by States as to the interpretation of other BITs which do not have an express reference to an external international law standard.Footnote 125
However, the symbiotic relationship between investment treaties and general international law does not require a black-and-white choice between complete discretion on the one hand, and the application of a conception of customary international law ‘frozen in amber’Footnote 126 at some time in the past. It is submitted that it is both possible and necessary to reconcile the particular treaty language with general international law. This is not simply because, as it was put in one recent award, ‘ … the difference between the Treaty standard … and the customary minimum standard, when applied to the specific facts of a case, may well be more apparent than real’.Footnote 127 More fundamentally, it is because the legal protection afforded by the guarantee of fair and equitable treatment cannot be understood without a conception of the proper function of international law in assessing the standards of justice achieved by national systems of law and administration. It is this function which the protection shares with the minimum standard.
Elihu Root, speaking on the appropriate treatment to be accorded to aliens in 1910, concluded that:
There is a standard of justice, very simple, very fundamental, and of such general acceptance by all civilized countries as to form a part of the international law of the world. The condition upon which any country is entitled to measure the justice due from it to an alien by the justice which it accords to its own citizens is that its system of law and administration shall conform to this general standard.Footnote 128
Seen in this light, the fair and equitable standard gives modern expression to a general principle of due process in its application to the treatment of investors.Footnote 129 The foundation of this principle is that, by agreeing to extend such treatment to nationals of a reciprocating country, States have accepted that there is an objective standard of treatment by which their own legal and administrative system may be judged. The overarching concept is that of the rule of law. A Chamber of the International Court of Justice expressed this idea by contrasting due process with arbitrariness:
Arbitrariness is not so much something opposed to a rule of law, as something opposed to the rule of law. This idea was expressed by the Court in the Asylum case, when it spoke of ‘arbitrary action’ being ‘substituted for the rule of law’ (Asylum Judgment, ICJ Reports 1950, p 284). It is a wilful disregard of due process of law …Footnote 130
Reference to ‘general principles of law common to civilized nations’Footnote 131 will assist in giving content to that concept. Here, modern international law presents a substantial advantage over the position in the early 20th century as a result of the elaboration of relevant international human rights standards, and of the gradual emergence of global administrative law.Footnote 132
Importantly, however, and unlike the earlier customary international law standard, the protection is not simply concerned with ‘fair’ or ‘just’ treatment to investors. Appropriate weight and meaning must also be given to the requirement of ‘equitable treatment.’ This additional element consciously takes the minimum standard in a different direction to that of the earlier law. Schwarzenberger observed on the first emergence of this formula, this addition:
… is well justified on two grounds. The experiences of the last forty years suggest that whenever, in fact, an agreed settlement has been reached, the creditors have consented to temper the application of the minimum standard by the introduction of an equitable element in the form of considerable concessions on their part. Thus, it appears wise to anticipate—and limit—such contingencies. Moreover, in relations between heterogeneous communities—in varying stages of technological advancement, social structure and political organization—and in an age of rapid change, the standard of equitable treatment provides equality on a footing of commendable elasticity.Footnote 133
The concept of equitable treatment thus provides not merely a means of doing equity as between different classes of investors, whether natives or equally favoured foreigners. That is something which the concomitant rights of national treatment and most-favoured-nation treatment will secure in any event. The inclusion of the reference to equitable treatment also provides a means by which an appropriate balance may be struck between the protection of the investor and the public interest which the host State may properly seek to protect in the light of the particular circumstances then prevailing. The editors of Oppenheim put it in this way: ‘The requirements of international law in this field … represent an attempt at accommodation between the conflicting interests involved.’Footnote 134 This process involves, as was observed in the Saluka award, ‘a weighing of the [investor's] legitimate and reasonable expectations on the one hand and the [host State's] legitimate regulatory interests on the other.’Footnote 135
Having examined the lines of convergence in the application of investment treaty standards, it is now necessary to look at where the specific solutions adopted by treaty may diverge from custom and require a different approach. Here, the arbitral practice provides some salient case studies of what happens where the application of general international law is not considered through the lens of the rules of treaty interpretation.
B. Restoration of the Place of the Treaty
1. The curative function of treaties
In considering the potential impact of custom as a source of interpretation of treaties, it should not be forgotten that the blind transposition of customary international law rules into treaty cases may in fact subvert some of the very purposes for which States negotiated specific treaty arrangements. There is a paradox at the heart of this field: despite their reliance on general international law in framing the substantive provisions of investment treaties, States had to negotiate on a predominantly bilateral basis in the face of widely perceived shortcomings in method by which custom protected such rights, namely diplomatic protection.
This point is well made by the International Law Commission itself in its report on Diplomatic Protection in discussing the advantages presented by investment treaties as contrasted with the customary rules of diplomatic protection:
Such treaties abandon or relax the conditions relating to the exercise of diplomatic protection, particularly the rules relating to the nationality of claims and the exhaustion of local remedies … The dispute settlement procedures provided for in BITs and ICSID offer greater advantages to the investor than the customary international law system of diplomatic protection, as they give the investor direct access to international arbitration, avoid the political uncertainty inherent in the very nature of diplomatic protection and dispense with the conditions necessary for the exercise of diplomatic protection.Footnote 136
The obverse of this point was made by the International Court in Diallo when it stated, in considering the nationality requirements for the exercise of diplomatic protection:
The fact invoked by Guinea that various international agreements, such as agreements for the promotion and protection of foreign investments and the Washington Convention, have established special legal regimes governing investment protection … is not sufficient to show that there has been a change in the customary rules of diplomatic protection; it could equally show the contrary.Footnote 137
It was a failure to consider the impact of the respects in which investment treaties consciously depart from the strictures of the old law which led the Tribunal in Loewen Footnote 138 into error in applying the fair and equitable standard to the treatment accorded to a Canadian investor by the Mississippi court system. The Tribunal in effect imposed a requirement to exhaust local remedies, not as a procedural bar, but as a part of the substantive requirement for breach of the standard of fair and equitable treatment. In this way, a very serious miscarriage of justice against a foreign national was dismissed as a local error, which could not be remedied by the international tribunal.
An international tribunal is of course concerned to see whether there has been a failure in the operation of the judicial system as a whole. As a general proposition, ‘an aberrant decision by an official lower in the hierarchy, which is capable of being reconsidered, does not of itself amount to an unlawful act.’Footnote 139 But it is quite another matter to import the full rigour of the local-remedies rule into investment arbitration on the ground that, in the absence of evidence of full exhaustion, there could be no breach of the treaty standard. Where the responsibility of a State vis-à-vis another State is engaged as a result of the treatment of an alien in the State's courts, it is only to be expected that the individual should first exhaust local remedies before the dispute is elevated to that of an international wrong between States. It was in the context of just such an inter-State claim in ELSI that a Chamber of the ICJ emphasized the continuing importance of the rule. But to insist on a strict application of this requirement in investor-State arbitration is simply inconsistent with the creation of a right of arbitration by investors directly, as an alternative to host State court litigation. As Bennouna observed in his preliminary report to the International Law Commission on Diplomatic Protection:
In consenting to arbitration, the parties to a dispute waive all other remedies. In this way, both the demand of the host State that local remedies be exhausted and the exercise of diplomatic protection by the State of nationality are put aside.Footnote 140
All of this suggests that an uncritical application of custom may be as damaging as its exclusion. The guiding principles of interpretation remain those set by the template of Article 31 of the Vienna Convention. This requires the Tribunal to begin by considering the treaty's terms ‘in their context and in the light of their object and purpose’. The treaty's terms or object may well militate against incorporation of particular rules of international law or indicate their own direction in the construction of the treaty.Footnote 141 It was a failure to start with the treaty text, and to work out from it to the relevant rules of international law, which led the tribunals in CMS, Enron and Sempra into error in their consideration of the proper place of the doctrine of necessity in investment treaty law. It is to an examination of that issue to which this article must now turn.
2. The case of the doctrine of necessity
A series of cases, decided under the US–Argentina BIT,Footnote 142 have considered the question whether the Argentine fiscal and economic crisis of 2001–2 could give rise to a defence of necessity to a claim of breach of treaty. The reasoning adopted in these cases is potentially of great significance to the subject of the present article, since, in various ways, the tribunals all considered the doctrine of necessity as a matter of customary international law. This is now conveniently codified in Article 25 of the ILC Draft Articles on State Responsibility, as one of a number of circumstances precluding wrongfulness.Footnote 143
Unfortunately, however, the tribunals which have so far considered the matter have come to very different conclusions on the application of the defence. In CMS,Footnote 144Enron,Footnote 145 and Sempra Footnote 146 the tribunals (which all had the same President) considered that the defence was not available. In LG&E,Footnote 147 however, the tribunal decided that ‘between 1 December 2001 and 26 April 2003, Argentina was in a state of necessity, for which reason it shall be exempted from the payment of compensation for damages incurred during that period.’ Two members of the tribunal had also sat on tribunals which reached the contrary conclusion. Despite this, neither LG&E nor Enron refer to the opposing authority, and so it is not clear whether the arbitrators considered that it was to be distinguished, or merely not followed. In Sempra, the difference was said to be as to the respective tribunals' appreciation of the facts as to the severity of the crisis.Footnote 148
The award in CMS was the subject of annulment proceedings. On 25 September 2007, the Annulment Committee delivered its decision.Footnote 149 It found manifest errors of law in the tribunal's treatment of necessity, but it declined to annul on this ground, holding that the errors did not amount to a manifest excess of powers or lack of reasoning, as would have been required for annulment.Footnote 150 The situation is one of considerable disarray, both for Argentina as judgment creditor on all of these awards, but also for the development of the doctrine. The resulting confusion appears to call into question the very utility of reference to general international law in investment treaty arbitration. In order to see what went wrong, it is necessary to look a little more closely at the reasoning in the awards and annulment decision.
The facts underlying each of the awards are substantially the same. In order to put an end to its economic and fiscal crisis of the 1980s, Argentina embarked in the early 1990s on a recovery programme, which involved both the privatization of a number of State-owned industries, and the pegging of the Argentine peso to the US dollar. The gas industry in particular was the subject of privatization under long-term supply contracts, pursuant to which the price at which gas would be purchased was calculated in US dollars in accordance with the US Producer Price Index (PPI) and then expressed in pesos. Towards the end of the 1990s another serious economic crisis hit Argentina: there were several deferrals of recalculation of the gas prices. On 6 January 2002, Argentina adopted a law declaring a public emergency, which terminated the right of public utilities to calculate tariffs by reference to the US PPI and to calculate tariffs in US dollars. A new government was elected in Argentina in April 2003, and a new period of economic and fiscal stabilization began.
All of the cases cited above concerned alleged losses suffered by American investors in the Argentine gas industry as a result of the changes to the calculation of gas prices made from the late 1990s and thereafter. In all of the cases, the investors alleged, inter alia, that they had not received fair and equitable treatment, and the tribunal agreed. In all of the cases, however, Argentina alleged that it should be wholly or partly excused from liability by reason of the pendency of a state of emergency or necessity. In so doing, Argentina relied on both customary international law and on Article XI of the BIT which provides:
This treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfilment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.Footnote 151
In CMS, the Tribunal approached this defence by holding that it raised ‘one fundamental issue.’Footnote 152 It addressed that issue by considering, first, whether each of the requirements imposed by Article 25 of the ILC Draft Articles on State Responsibility had been met. It found that they were not. It did not conduct a separate review of the requirements of Article XI of the BIT.Footnote 153 The tribunal then concluded, invoking Article 27 of the ILC Draft Articles, that: ‘[e]ven if the plea of necessity were accepted, compliance with the obligation would reemerge as soon as the circumstance precluding wrongfulness no longer existed, which is the case at present.’Footnote 154 Thus, such a plea would not obviate the obligation to pay compensation.
In Enron, the Tribunal also considered the application of Article 25 of the ILC Draft Articles first, and only subsequently Article XI of the BIT itself. On this occasion, Argentina's experts specifically opined that the treaty regime was different and separate from customary law as lex specialis. The Tribunal disposed of this argument in the following way:
This is no doubt correct in terms that a treaty regime specifically dealing with a given matter will prevail over more general rules of customary law. Had this been the case here the Tribunal would have started out its considerations on the basis of the Treaty provisions and would have resorted to the Articles on State Responsibility only as a supplementary means. But the problem is that the Treaty itself did not deal with these elements. The Treaty thus becomes inseparable from the customary law standard insofar as the conditions for the operation of state of necessity are concerned.Footnote 155
The Enron Tribunal maintained the same view as to the effect of any state of necessity on the requirement to pay compensation as had been adopted in CMS.Footnote 156
In Sempra, the Tribunal returned to the theme of the relationship between Article XI of the BIT and custom, holding that:
The problem here, however, is that the Treaty itself did not deal with the legal elements necessary for the legitimate invocation of a state of necessity. The rule governing such questions will thus be found under customary law … International law is not a fragmented body of law as far as basic principles are concerned and necessity is no doubt one such basic principle.Footnote 157
The approach of the Tribunal in LG&E stands in marked contrast to this line of reasoning. For this Tribunal, Article XI was the essential starting-point for its analysis:
The Tribunal underscores that the claims and defenses mentioned derive from the Treaty and that, to the extent required for the interpretation and application of its provisions, the general international law shall be applied.Footnote 158 (emphasis added)
Like the other tribunals, the LG&E Tribunal rejected Argentina's argument that the test in Article XI was ‘self-judging’. The Tribunal considered that it was entitled to conduct its own substantive review of whether the conditions for the application of Article XI had been met, and it did so. The Tribunal found that the conditions specified by Article XI had been met, since the measures taken were on the facts ‘necessary for the maintenance of public order’ and for the protection of Argentina's ‘essential security interests’. The Tribunal then referred to Article 25 of the ILC Draft Articles, but only as additional support to a conclusion which it had already reached on the application of Article XI of the BIT.Footnote 159 Once this point had been reached, there could be no question of compensation for the period during which the necessary measures applied, as the application of Article XI precluded any finding of breach of treaty.Footnote 160
This approach to the problem was in substance adopted by the Annulment Committee in CMS in finding that the Tribunal in that case was in error of law in its treatment of the state-of-necessity defence. The Committee held that both the function and the substantive conditions of BIT Article XI and of ILC Article 25 are different. On function, it pointed out that:
… Article XI is a threshold requirement: if it applies, the substantive obligations under the Treaty do not apply. By contrast, Article 25 is an excuse which is only relevant once it has been decided that there has otherwise been a breach of those substantive obligations.Footnote 161
On substance, it said:
Furthermore, Article XI and Article 25 are substantively different. The first covers measures necessary for the maintenance of public order or the protection of each Party's own essential security interests, without qualifying such measures. The second subordinates the state of necessity to four conditions. It requires for instance that the action taken ‘does not seriously impair an essential interest of the State or States towards which the obligation exists, or of the international community as a whole’, a condition which is foreign to Article XI.Footnote 162
The Committee concluded:
Those two texts having a different operation and content, it was necessary for the Tribunal to take a position on their relationship and to decide whether they were both applicable in the present case. The Tribunal did not enter into such an analysis, simply assuming that Article XI and Article 25 are on the same footing.Footnote 163
This led to a second error of law, namely the failure to consider whether each such rule was a primary or a secondary rule of international law. The Committee's view was that Article XI was a primary rule, in that, if it applied, there would have been no breach of the BIT.Footnote 164 In contrast, Article 25 was a secondary rule, and would only apply in the event that there had been a breach of a primary rule. The Committee concluded on the approach which the Tribunal ought to have taken:
Only if it had concluded that there was conduct not in conformity with the Treaty would it have had to consider whether Argentina's responsibility could be precluded in whole or in part under customary international law.
These two errors made by the Tribunal could have had a decisive impact on the operative part of the Award.Footnote 165
Finally, the Tribunal found that a similar failure to put the Treaty first had led the Tribunal in its consideration of the question whether compensation would be due for the period of any state of necessity:
The answer to that question is clear enough: Article XI, of and for so long as it applied, excluded the operation of the substantive provisions of the BIT. That being so, there could be no possibility of compensation being payable during that period.Footnote 166
In formal terms, the decision of an annulment committee has no greater precedential effect than an award.Footnote 167 Nevertheless, the very opportunity of a second tier of review; the narrowly circumscribed limits of that review; and the eminent experience in public international law of the Committee,Footnote 168 suggest that great weight should be given to the Committee's categorical views on the central issues confronted in these cases. Their significance for the present study lies in two central tenets of the approach here advocated as to the relationship of general international law to investment treaties.
First, the mere identification of the applicable law as being international law as a whole (pursuant to Article 42 of the ICSID Convention) is insufficient to provide guidance to a tribunal as to how it should integrate the wider provisions of international law with the specific terms of the Treaty. Indeed, characterizing the question as being simply one of applicable law may, as the cases discussed above demonstrate, be positively dangerous. Treatment of the issue in this way does not imply any particular order for the consideration of the relevant sources of law. It was taking the customary doctrine first, and then conflating its test with that of the Treaty, without close consideration of the differences, which contributed to the errors of the CMS Tribunal, and those which followed it.
Secondly, once the issue is properly approached as one of treaty interpretation, the very structure of Articles 31–32 of the Vienna Convention suggests an orderly method of legal reasoning which would place the Treaty text in the centre of the enquiry. General international law would then be applied to assist in the interpretation of that text.Footnote 169 This is not to exclude the possible application of the secondary rules of State responsibility to provide an independent defence to liability under the Treaty. Necessity, as a circumstance precluding wrongfulness, is found within Part One of the ILC Draft Articles on State Responsibility. As such, it is potentially applicable to any breach of the responsibility of a State, whether or not the claimant is also a State.Footnote 170 But such rules would only be applicable to the extent to which they are have not been by implication necessarily excluded by the express words of the Treaty. Where, as here, the customary rule lays down a stricter test than the treaty language, it is unlikely that there will be a need for separate resort to custom.
Applied in this way, general international law in investment treaty cases does not become the juridical equivalent of a bag of liquorice allsorts, in which the Tribunal may pick and choose at will those doctrines which suit its decision. Rather, its primary role is the progressive illumination of the parties' intentions, as expressed in their treaty text; or its application to issues not expressly addressed in the Treaty in a different way.
III. IMPLICATIONS FOR THE MAKING OF INTERNATIONAL LAW
A. Arbitral Awards as an Elucidation of the Law
What, then, may be the implications of the investment treaty phenomenon for more general theories of the way in which international law is created and developed? The issue here is to what extent the practice by or under particular international conventions contributes to the development of general international law under the other heads of Article 38 of the Statute of the ICJ: custom ‘as evidence of a general practice accepted as law’; ‘general principles of law common to civilised nations’; and judicial decisions ‘as subsidiary means for the determination of rules of law’.Footnote 171
This question is not simply to be disposed of as a matter of the weight to be accorded to arbitral awards themselves. The jurisprudence of international tribunals is a means of the elucidation of the law, not as a source of law in its own right. Thus, in the debates in 1920, which led to the formulation of what is now Article 38(1)(d) of the ICJ Statute, Baron Descamps said:
Doctrine and jurisprudence no doubt do not create law; but they assist in determining rules which exist. A judge should make use of both jurisprudence and doctrine, but they should serve only as elucidation.Footnote 172
For that reason, judgments of the International Court were given no binding effect, save upon the disputing parties.
What is true of the judgments of the International Court of Justice is true a fortiori of arbitral awards. As Brownlie puts it:
The literature of the law contains frequent reference to decisions of arbitral tribunals. The quality of arbitral tribunals has varied considerably, but there have been a number of awards which contain notable contributions to the development of the law by eminent jurists sitting as arbitrators …Footnote 173
To some extent, of course, the distinction between elucidation of norms found elsewhere and the ability to make law is a legal fiction maintained to preserve the control of States over the development of the international legal system. As Paulsson rightly observes:
Just as jealous sovereigns may be averse to any suggestion that compacts other than those to which they have consented may be invoked against them, so too, they are often disinclined to submit to the elaboration of international law by anything resembling the accretion of binding precedents known as common law.Footnote 174
The formal distinction is belied by practice in the extent to which international tribunals do refer to the previous decisions of other tribunals as authority for propositions of law, and are engaged in the process of developing the law.Footnote 175 Nevertheless, the distinction is an important one. The extent to which any particular tribunal decision deserves currency as a precedent depends on whether it correctly expounds international law, and not upon any a priori weight to be attached to the decision per se.
B. Custom
The potential contribution of investment treaties and awards to custom is more problematic. Standard approaches to the formation of custom would give a largely negative account of the potential contribution of bilateral treaties. Thus, the influential report of the ILA Committee on the Formation of Customary (General) International LawFootnote 176 adopted a principle that: ‘there is no presumption that a succession of similar treaty provisions gives rise to a new customary rule with the same content.’Footnote 177 The Commentary deals specifically with the case of bilateral investment treaties, stating:
Some have argued that provisions of bilateral investment protection treaties (especially the arrangements about compensation or damages for expropriation) are declaratory of, or have come to constitute, customary law. But … there seems to be no special reason to assume that this is the case, unless it can be shown that these provisions demonstrate a widespread acceptance of the rules set out in these treaties outside the treaty framework.Footnote 178
It is of course correct that no easy assumption can be made as to the binding character in custom of the treaty obligations. State practice in concluding such treaties is widespread. The number of such treaties is impressive, and the similarity in their language is striking. But, nevertheless, the total tally falls considerably short of anything approaching universality.Footnote 179 Some important States have concluded few of them. There are relatively few such agreements between developed countries (with the important exceptions of the Energy Charter Treaty, NAFTA and recent free trade agreements). There are numerous important differences between the treaty texts as a result of bilateral negotiation.
Further, the mere conclusion of numerous similarly worded treaties does not necessarily provide an indication of opinio juris, namely that the State considers that the standard is binding upon it as a matter of the general law, irrespective of its agreement by treaty. As it was put by the Tribunal in UPS v Canada:
… the many bilateral treaties for the protection of investment … vary in their substantive obligations; while they are large in number, their coverage is limited; and in terms of opinio juris there is no indication that they reflect a general sense of obligation.Footnote 180
FA Mann gave an initial response to the proposition that the investment treaty practice could not contribute to the development of custom some 25 years ago. Whilst accepting the general point made by the International Court of Justice in the North Sea Continental Shelf Cases,Footnote 181 that it is difficult to deduce a rule of international law from a treaty, he continued that the significance of that decision should not be overrated, at least in the context of investment treaties:
There is, in the first place, the very large number of treaties the scope of which is increased by the operation of the most-favoured-nation clause. There is, secondly, the fact that many States which have purported to reject the traditional conceptions and standards included in these treaties have accepted them, when (if the colloquial phrase may be permitted) it came to the crunch. There is, thirdly, the most important fact that these treaties establish and accept and thus enlarge the force of traditional conceptions. Is it possible for a State to reject the rule according to which alien property may be expropriated only on certain terms long believed to be required by customary international law, yet to accept it for the purpose of these treaties? The paramount duty of States imposed by international law is to observe and act in accordance with the requirements of good faith. From this point of view it follows that, where these treaties express a duty which customary international law imposes or is widely believed to impose, they give very strong support to the existence of such a duty and preclude the Contracting States from denying its existence.Footnote 182
Mann's point may be taken a step further in the light of the practice adopted in many States of promulgating model investment treaties.Footnote 183 This practice is not confined to traditional capital-exporting States, but includes many developing countries as well. The publication of a model treaty is of course a piece of State practice. It is a statement of what that State considers to be an acceptable basis for bilateral negotiations. Nevertheless, the practice may have no greater significance in the case of those obligations which can only operate reciprocally between two States when given conventional force, such as national treatment and most-favoured-nation treatment.
But the inclusion in such prototype treaties of the non-contingent standards, such as expropriation and fair and equitable treatment, has wider implications. As has been seen, there is a remarkable uniformity in the treaty language used to enshrine these standards. There is also a prior history of non-contingent standards being regarded as applicable by custom. Their inclusion in model treaties ‘represents the set of norms that the relevant state holds out to be both reasonable and acceptable as a legal basis for the protection of foreign investment in its own economy.’Footnote 184
In the case of expropriation, it is perhaps possible still to interpret such a statement as merely an offer to provide compensation for State takings of property to the nationals of reciprocating States. The fair and equitable treatment standard, on the other hand, applies generally to decision-making by all manner of State officials: judges, government ministers and bureaucrats. It is not possible to imagine the offer of the application of such a standard being made unless the State making the offer was satisfied that its national legal system already met such a standard, and that it would continue to be applied in the making of decisions regarding investments. By its nature such a standard cannot be applied selectively only to investors from particular foreign States. Rather, it is a general undertaking of due process and the rule of law. Seen in this light, its standing inclusion in a model treaty does represent something more than a mere offer to treat. Rather, it is submitted that it indicates the acceptance of a general obligation to accord treatment, judged according to such a standard. The inclusion of such a standard in the model treaties of many States in different parts of the world, with vastly differing legal systems and economic conditions, lends considerable weight to its candidacy as the modern expression of a rule of custom.
The result is a convergence, on these issues, between treaty practice and custom, in which the modern understanding of the content of the customary right is being elaborated primarily through the treaty jurisprudence. Indeed, an application of the classic test for the formation of a rule of custom in this area would have little meaning, given the paucity of any State practice outside the treaties' reach.
This process of cross-fertilization in the development of the customary standards through the treaty jurisprudence saves general international law from being cast in aspic at some earlier point in time; and saves treaty tribunals from isolation and inconsistency. It reflects the fact that the general standards are in their nature evolutionary.Footnote 185 The Tribunal in Mondev Footnote 186 described this process in the following way. It was responding to a submission advanced by Canada that the customary international law standard incorporated into NAFTA Article 1105 was to be determined by reference to Claims Commission awards of the inter-war years, in particular the Neer case.Footnote 187 It held:
On a remarkably widespread basis, States have repeatedly obliged themselves to accord foreign investment such treatment. In the Tribunal's view, such a body of concordant practice will necessarily have influenced the content of rules governing the treatment of foreign investment in current international law.Footnote 188
One may ask the question rhetorically: what is the alternative for custom? If the extensive practice under treaties is to be excluded, a court considering the matter would have only two other possible alternatives. The first would be to treat custom as frozen in time circa 1945 before State practice shifted to the elaboration of BITs. This would, in any event, face the considerable problem of the degree of dissent over the basic rules which then existed. The second would be simply to find that custom afforded no minimum standard of protection to foreign investors at all. This question will have to be confronted by the International Court when it reaches the merits in Diallo,Footnote 189 where the binding character of such rules outside the treaty context will take centre stage. However, for the most part in the investment field, this problem is the exception rather than the rule. In most cases the question will be one of content rather than binding character.
C. General Principles of International Law
The elaboration of norms in this field may share more with the way in which ‘general principles of law common to civilized nations’ are used as a source of law in international law.Footnote 190 It has been seen above that the content of the principle of ‘fair and equitable treatment’ is to be derived in part by reference to such general principles. An international tribunal has to be sure that the standards for judicial or administrative decision-making which it exacts from municipal courts and administrators are general principles of law common to civilized nations, and not rules of law specific only to some national legal systems, but rejected by others.
Yet here too, the analogy, though valuable, is inexact. The original idea animating Article 38(1)(c) of the Statute was to avoid the risk of a non liquet by permitting the International Court to borrow principles from domestic law, provided that they were sufficiently established as to command common acceptance.Footnote 191 In practice, the category is perhaps the least well-understood in the catalogue of sources, and examples of its express use by the International Court are rare.
Reference to general principles of national law in the investment context may indeed be required to supply interstitial rules which are not referred to in, or required by, the treaty, but which may be regarded by a tribunal as essential for the proper operation of the investment treaty system as a whole. An example of this would be the consideration of the general principles of res judicata and lis pendens, where the parallel claims meet the conditions for the application of the doctrines.Footnote 192 But the reference to general principles of law in the investment context more commonly serves a rather different function, namely to inform the content of an existing, but open-textured treaty norm.
There is, however, a certain incongruity in the seeking to fit the development of common international principles within a category designed for the adaptation for international use of national law principles. Whilst tribunals have occasionally accepted that one of the relevant developments in international law has been in ‘the substantive and procedural rights of the individual’,Footnote 193 it is regrettable that, to date, no notable effort has been made in the investment arbitration jurisprudence to link the concept of ‘fair and equitable treatment’ with the specific standards of international human rights law. Reference to these sources would facilitate the application of standards which may be seen as genuinely common to civilized nations. It is not, as has sometimes been suggested, that the emergence of international human rights norms would subsume and render obsolete both the international minimum standard and the standard of national treatment.Footnote 194 Rather, some elements of human rights law may furnish a source of general principle from which the obligation of fair and equitable treatment may be given contemporary content.Footnote 195 In this way, investment law may meet, in modern clothes, the objective which Elihu Root identified, of simply providing an effective avenue of redress for a set of principles common to humanity.
So the final possibility is that the emergence of this ‘body of concordant practice’, which serves as the central, and defining, feature of modern investment law is in fact a process which is not fully explained by a list of sources of international law originally developed in 1920 for the Permanent Court of International Justice. Indeed, the way in which international law develops has changed utterly since that time. One consequence of the relentless rise in the use of treaties as a means for ordering international civil societyFootnote 196 is that the dynamic process of the development of international law now takes place in no small measure through the continuous progressive development of treaties. Thus, for example, in the Myers case,Footnote 197 the arbitral tribunal was invited to consider numerous international instruments in the field of environmental protection—each one building upon those that had come before.
A similar process may be observed in the framing of bilateral investment treaties themselves. Each State brings to the negotiating table a lexicon which is derived from prior treaties (bilateral or multilateral) into which it has entered with other States. The resulting text in each case may be different. It is, after all, the product of a specific negotiation. But it will inevitably share common elements with what has gone before, and, as Mann observed, the almost invariable addition of a most-favoured-nation clause reinforces this already strong tendency towards homogeneity.
In making this observation about the nature of the modern treaty-making process, it is not always necessary to go so far as to contend that such common elements may point to the emergence of a binding rule of customary international law. The important point is that this everyday reality in the practice of foreign ministries has the inevitable consequence that treaties are developed in an iterative process in which many normative elements are shared. From having been a series of distinct conversations in separate rooms, the process of treaty-making is now better seen as akin to a continuous dialogue within an open-plan office.Footnote 198 This is no mere matter of drafting convenience. Rather, it is suggested here that, at least in the field of investment, the negotiation of treaties has been animated by a set of general principles which inform the content of particular norms.
The emergence of over-arching common principles as the unifying feature of international investment law finds parallels in other areas of international law as well: in international criminal law;Footnote 199 and in environmental law.Footnote 200
If an observable phenomenon in the actual practice of States on such a scale emerges, which is not adequately explained by the existing legal categories, it may be time to ask whether those categories ought to be developed. It is, of course, still of vital importance to be able to distinguish between law and non-law. All legal systems need formal mechanisms to enable them to determine the binding character of norms. But, equally, all legal systems contain norms at different levels of generality, and some of the most important such norms are stated at the highest level of abstraction. It is these which then qualify for constitutional significance within the system.Footnote 201 International law is no exception.Footnote 202
International tribunals frequently rely on such principles. As the ILC Study Group puts it:
… fragmentation takes place against the background and often by express reference to not only the VCLT but to something called ‘general international law’. However, there is no well-articulated or uniform understanding of what this might mean. ‘General international law’ clearly refers to general customary law as well as ‘general principles of law recognized by civilized nations’ … But it might also refer to principles of international law proper … In the practice of international tribunals … reference is constantly made to various kinds of ‘principles’ sometimes drawn from domestic law, sometimes from international practice but often in a way that leaves their authority unspecified.Footnote 203
D. Conclusions
The results of the enquiry in this article may be expressed in the following ten conclusions:
(1) The investment treaty as a creature of international law: The primary obligations assumed by States towards foreign investors and enshrined in investment treaties are governed by public international law. This flows from the elementary fact that such treaties are themselves creatures of international law. This general proposition is particularly significant in the case of investment treaties, since their terms and structure make them particularly dependant upon the wider context of general international law.
(2) International law as a whole applicable to treaty obligations: Thus, when, by virtue of Article 42 of the ICSID Convention or otherwise, tribunals are directed that the applicable law to the particular issue before them is international law, that is a reference to the whole of international law, and not merely the specific treaty before them. (That does not of course exclude the possibility of the application of national law, and in particular host-State law, to other issues in the case. Article 42 so provides, and this would follow in any event from the hybrid character of investment arbitration.)
(3) Treaty interpretation as the process for its application: The identification of the international legal system as a whole as the applicable law is not the end of the enquiry. The means by which the applicable norms of international law are determined is primarily supplied by the process of treaty interpretation encapsulated in Articles 31 and 32 of the VCLT. This process envisages a specific role for other rules of international law applicable in the relations between the parties (Article 31(3)(c)) and for shared understandings between the parties as to the meanings of specific terms (Article 31(4)).
(4) Central place of the treaty text: The application of such other rules is subject to the general rule of interpretation, which starts with the ordinary meaning of the words and their object and purpose (Article 31(1)). If the specific words of the treaty supply an answer to the problem which is different to that otherwise applicable in general international law, that solution must prevail. For this reason, the new hybrid procedures of the investment treaties render the restrictions of the general law of diplomatic protection largely inapplicable. Further, as in the doctrine of necessity cases, the treaty may supply a primary rule which takes precedence over a different test provided in custom.
(5) Application of custom through interpretation: Nevertheless, custom may be more directly applicable:
(a) to determine the content of a primary treaty rule where the language, purpose or context of the particular treaty provision indicates that the parties intended to invoke the customary rule, and where there is a developed body of customary law to which reference may be made. Examples of this include expropriation, and that part of the fair and equitable treatment standard which in concerned with denial of justice; and
(b) where the treaty is silent as to the issue, and the answer is supplied by other rules of international law. This has been characterized here as a positive presumption of interpretation. But it may equally be seen simply as a consequence of international law being the applicable law. Thus, for example, general secondary rules of State responsibility relating to attribution and circumstances precluding wrongfulness may be applied if the treaty language, properly interpreted, permits.
(6) Guidance through general principles: Where the investment treaty poses new questions not previously addressed by custom (as, for example, in the application of the fair and equitable treatment standard to administrative decision-making), reference to general international law may still illuminate the rule to the extent that it reveals general principles which guide the application of the rule.
(7) General principles applicable to fair and equitable treatment: In the case of fair and equitable treatment, the general principles informing the standard flow from its character as an instrument of supervision at the international law level of the processes of governmental decision-making at national level. The standard is concerned with due process in decision-making, and not with substantive outcomes. It requires the application of fundamental rule-of-law values in decision-making: predictability; accessibility; impartiality; and natural justice, as contrasted with arbitrary action. The standard shares these concerns with international human rights law. The addition of the concept of equity also requires due weight to be given to the proper public purposes of the host State—performing a similar function to that of proportionality in the application of human rights standards.
(8) Arbitral awards as elucidation of international law not precedent: The contribution of the treaty practice to general international law is not a matter of the weight to be accorded to arbitral awards as precedents. Like all other judicial decisions in international law, investment arbitral awards can only serve as an elucidation of the law, not as a binding source of it. The weight to be accorded to such awards is rather a matter of their consistency within the general framework of international law.
(9) Treaty rules as custom: The mere prevalence of similarly worded treaty language, however numerous, will not, without more, give rise to a binding obligation in custom. Many obligations contained in investment treaties can only ever operate as conventional obligations differing from the general law (national treatment, most-favoured-nation treatment, and indeed the provision for investor-State arbitration itself). Fair and equitable treatment, however, may stand in a distinct position in view of the substantial body of evidence for the prior existence of a minimum standard of treatment as a matter of custom; the almost universal inclusion of the standard in modern times in very similar terms; and its inclusion in many model treaties published by states. The last point supports opinio juris connoting the State's view that such a standard is internationally acceptable; and that it is one which its legal system already meets, and should continue to meet in its treatment of investors generally.
(10) The organic and dynamic character of general principles of international law: The relationship between general international law and investment treaties is properly characterized as conducted at the level of general principles of international law. Such principles inform the content of the specific treaty norm and of States' obligations more generally. The consequence of this is that, at this level, the relationship is symbiotic. Developments in custom may, indeed should, influence the content of the treaty rule (an example being the development of international human rights standards). By the same token, the developments in the treaty rule (as with the addition of equitable treatment) may influence custom.
As the Eastern Extension Tribunal put it in the dictum with which the article opened,Footnote 204 such a reference to general principles may indeed be integral to the very ‘function of jurisprudence’, in international law as much as in any domestic legal system.
If the practice of investment arbitral tribunals helps to explain the operation of the larger system of international law better, it may be because it exposes to view the essential dependence of the system as a process of reasoning upon a set of norms which are not fully explained as either custom or general principles of domestic law. Rather, they are general principles of international law. The ongoing challenge, then, is to articulate those principles in a way which both connects investment treaty law to the vast hinterland of international law on the treatment of aliens, but which also recognizes the progressive nature of the treaty-making programme itself, and the dynamic qualities of general international law as a ‘living and expanding code.’Footnote 205 In this way, investment treaty law may finally come of age, and take its proper place within the modern international legal system.
