Arbitration under Investment Treaties



A. Introduction

[8.01] Historically, an individual or a corporation who wished to assert a claim against a foreign state for breach of customary international law could not do so directly. Instead, the individual or corporation had to petition its government to take up, or ‘espouse’, the claim on its behalf. In the course of the nineteenth century, influential individuals or corporations would convince their government to send a small contingent of warships to moor off the coast of the offending state until reparation was forthcoming. This form of ‘gunboat diplomacy’ was exercised frequently by European powers on behalf of their subjects in the not-so-distant past. For example, when faced with Venezuela’s default on its sovereign debt in 1902, the governments of Great Britain, Germany, and Italy sent warships to the Venezuelan coast to demand reparation for the losses incurred by their nationals.

[8.02] Argentine jurist and diplomat Carlos Calvo fought for the right of newly independent states to be free of such intervention by foreign powers, promoting the so-called Calvo doctrine, whereby foreign investors should be in no better position than local investors, with their rights and obligations to be determined through the exclusive jurisdiction of the courts of that state.1 His position was adopted by the First International Conference of American States in 1889, the ad hoc Commission on International Law of which (without the support of the United States) concluded:

Foreigners are entitled to enjoy all the civil rights enjoyed by natives and they shall be accorded all the benefits of said rights in all that is essential as well as in the form or procedure, and the legal remedies incident thereto, absolutely in like manner as said natives. A nation has not, nor recognizes in favour to foreigners, any other obligations or responsibilities than those which in favour of the natives are established in like cases by the constitution and the laws.2

[8.03] The doctrine was incorporated into the forerunner of the modern investment treaty, the ‘treaty of friendship, commerce and navigation’ (FCN treaty). For example, Article 21 of the FCN treaty between Italy and Colombia of 1894 stated as follows:

The Contracting Parties express their desire to avoid all types of dispute which might affect their cordial relations and agree that, in connection with disputes which involve individuals arising out of criminal, civil or administrative matters, their diplomatic agents will abstain from intervening except in cases of denial of justice or extraordinary or unlawful delay in the administration of justice.3

[8.04] Gunboat diplomacy as a means of asserting rights of nationals was finally laid to rest at the Second International Peace Conference of The Hague in 1907, when the Convention on the Peaceful Resolution of International Disputes was signed. The Convention provided the framework for the conclusion of bilateral arbitration treaties. In accordance with these treaties, in the event of a dispute between two states arising out of the particular interests of a national of the other state, an independent arbitral tribunal would be formed. In effect, a state could espouse the claim of its national (the so-called right of diplomatic protection) by means of a horizontal inter-state procedure. There was no direct cause of action by the foreign national whose interests had been harmed.

[8.05] The legal basis of the right of ‘diplomatic protection’, in the words of the Permanent Court of International Justice (PCIJ) in the Panevezys-Saldutiskis Railway case, was that:

[I]‌n taking up the case of one of its nationals, by resorting to diplomatic action or international judicial proceedings on his behalf, a state is in reality asserting its own right, the right to ensure in the person of its nationals respect for the rules of international law. This right is necessarily limited to the intervention on behalf of its own nationals because, in the absence of a special agreement, it is the bond of nationality between the state and the individual which alone confers upon the state the right of diplomatic protection, and it is as a part of the function of diplomatic protection that the right to take up a claim and to ensure respect for the rules of international law must be envisaged.4

[8.06] As one learned commentator stated, the procedure was unsatisfactory from the individual claimant’s point of view:

He has no remedy of his own, and the state to which he belongs may be unwilling to take up his case for reasons which have nothing to do with its merits; and even if it is willing to do so, there may be interminable delays before, if ever, the defendant state can be induced to let the matter go to arbitration. … It has been suggested that a solution might be found by allowing individuals access in their own right to some form of international tribunal for the purpose, and if proper safeguards against merely frivolous or vexatious claims could be devised, that is a possible reform which deserves to be considered. For the time being, however, the prospect of states accepting such a change is not very great.5

[8.07] Since that text was written in 1963, the situation has changed dramatically and what Professor Brierley thought unlikely has become a commonplace reality. The validity of his concerns, and the inevitable ‘politicisation’ of disputes ‘leaving investors, particularly small and medium-sized enterprises, with little recourse save what their government cares to give them after weighing the diplomatic pros and cons of bringing any particular claim’,6 led to a radical reform in the dispute-settlement provisions of many BITs.

[8.08] This reform was made possible by the conclusion of the Convention on the Settlement of Investment Disputes between States and Nationals of Other States of 1965 (the ICSID Convention).7 The Convention was aimed primarily at creating a new arbitral forum for the resolution of disputes between investors and states by means of the inclusion of arbitration clauses in state contracts. Nevertheless, the travaux préparatoires of the Convention also made clear that the consent of the state to arbitration could be established through the provisions of an investment law.8 Following the 1959 Abs-Shawcross Draft Convention on Investments Abroad and the 1967 Organisation for Economic Co-operation and Development (OECD) Draft Convention on the Protection of Foreign Property,9 many states had begun a programme of bilateral treaties for the promotion and protection of investments (the so-called BITs) that set out explicit protections in favour of foreign investment.10 They were a natural successor to the FCN treaties of the early part of the twentieth century, but still suffered from the limitations imposed by diplomatic protection. Once the ICSID Convention was in place, treaty drafters from signatory states quickly seized upon the possibility of using this specialist forum for the resolution of treaty disputes between states and investors, and did so by incorporating a clause establishing the consent of the state to arbitrate with covered investors. Professor Brierley’s vision of a diagonal clause, permitting investors to claim directly under a treaty against the state in which the investment was made (the ‘host state’), thus became a reality. Switzerland, for example, inserted a diagonal clause for the first time in its 1981 BIT with Sri Lanka,11 and has done so systematically ever since.12

[8.09] This right of direct recourse ensures that the investor’s claim is not subject to the political considerations inherent in diplomatic protection. Even if there is no agreement between the investor and the host state, the investor may usually commence arbitration directly against the host state.13 Foreign investors were nevertheless slow to take up their newfound rights: the first case brought by an investor under the investment protections of a BIT was not decided until 1990.14

[8.10] In light of the dramatic increase in the number of BITs15 and the emergence of clearer legal principles through case law, the number of investor–state arbitrations has mushroomed. In 2014, ICSID registered thirty-eight new arbitration cases—nearly seven times the number of cases registered during the whole of ICSID’s first ten years of existence.16 By the end of 2014, the number of known treaty-based investor–state cases had reached 608—approximately ten times the figure as it stood at 2000.17

[8.11] The dramatic growth of BITs since the mid-1980s has led to the adoption of similar provisions in the ‘investment chapters’, or collateral agreements, to multilateral economic cooperation treaties. These include the Association of Southeast Asian Nations (ASEAN) Comprehensive Investment Agreement,18 the North American Free Trade Agreement (NAFTA),19 the Energy Charter Treaty (ECT),20 and the Dominican Republic and Central America–United States Free Trade Agreement (DR-CAFTA).21 Similar provisions have found their way into bilateral free trade agreements (FTAs) such as the United States–Chile FTA22 and the 2010 Canada–Panama FTA.23

[8.12] As the number of investment treaty arbitrations has grown, concerns over the investment treaty system have arisen. These concerns include a perceived deficit of legitimacy given that states are being judged on their conduct by private non-elected individuals. Concerns have also arisen in respect of inconsistent arbitral awards, the independence and impartiality of arbitrators, and the delays and costs of arbitral procedures.24 These complaints have resonated in some scholarly publications25 and popular media outlets.26 At the same time, between 2007 and 2012, a small group of Latin American countries defending multiple claims—Bolivia, Ecuador, and Venezuela—denounced the ICSID Convention and certain BITs.27 These concerns and isolated denunciations are not symptomatic of an exodus from the investment treaty system. According to the United Nations Conference on Trade and Development (UNCTAD), 330 new investment treaties were concluded between 2010 and 2014,28 including more than two dozen in Latin America alone. During that same period, ICSID gained seven new member states.29 Systemic reforms are being considered30 and implemented, including the introduction of new transparency provisions.31

[8.13] As a result of the growth in investment treaties, many foreign investments are protected by investment treaties. The question is whether an investor can rely on one or more of these investment treaties to vindicate its legal rights in a particular case. This raises fundamental issues relating to the scope and application of those treaties, which issues are addressed next.

B. Jurisdictional Issues

(a) Existence of an applicable treaty

[8.14] To determine whether an investor enjoys investment treaty protection, an applicable treaty between the state in which the investment was made and the home state of the investor must be identified.32 It is easy to identify multilateral investment treaties, because they are sufficiently notorious. It is, however, more difficult to detect applicable BITs, considering their number and the absence of a comprehensive list.33 Although the UNCTAD list is helpful, the only accurate means of verifying the existence of a BIT, and whether it is in force,34 is by contacting the treaty section of the relevant government or embassy.

[8.15] Most BITs contain provisions with respect to their effective date and duration. An issue may arise as to whether investments made prior to the date on which the BIT came into effect are eligible for protection under it. Tribunals have generally taken the position that prior investments are afforded protection and some treaties are explicit in this regard. The Argentina–United States BIT, for example, provides that it shall apply to investments existing at the time of entry into force, as well as to investments made or acquired thereafter.35 A distinction should be drawn between application of a BIT to investments made prior to its entry in force and its application to alleged breaches that occurred prior to that date. In Técnicas Medioambientales Tecmed SA v The United Mexican States,36 the tribunal held that while the concerned investment was eligible for protection under the BIT, the BIT could not have retrospective application to actions by the host state prior to its entry into force.

[8.16] Bilateral investment treaties also commonly include provisions regarding the legal status of investments after the termination or expiry of the particular BIT. Generally, such provisions indicate that investments that were otherwise covered by the treaty whilst in force will continue to benefit from the same protection for a specified ‘sunset’ period, usually of between ten and fifteen years after termination or expiry.37

(b) Protected investors

[8.17] Once a potentially applicable treaty has been identified, the relevant treaty provisions defining the eligible ‘investors’, or ‘nationals’, should be reviewed. Although treaties may vary substantially in this respect, the following provision of the Switzerland–Pakistan BIT is representative:

For the purposes of this Agreement:

(1) The term ‘investor’ refers with regard to either Contracting Party to:

(a) natural persons who, according to the law of that Contracting Party, are considered to be its nationals;

(b) legal entities, including companies, corporations, business associations and other organisations, which are constituted or otherwise duly organised under the law of that Contracting Party and have their seat, together with real economic activities, in the territory of that same Contracting Party;

(c) legal entities established under the law of any country which are, directly or indirectly, controlled by nationals of that Contracting Party or by legal entities having their seat, together with real economic activities, in the territory of that same Contracting Party.38

Investors covered by protection of investment treaties can thus be divided into natural persons and legal entities.

(i) Natural persons

[8.18] Most BIT provisions establish the nationality of a natural person by reference to the domestic laws of the respective contracting states. This is consistent with the concept of state sovereignty in deciding the criteria for identifying its nationals. Certain BITs may contain an additional requirement of residence,39 or domicile.40

[8.19] Difficulties may arise where the purported investor is a national of both state parties to the BIT. The ICSID Convention precludes individuals from suing any state of which they are nationals.41 Tribunals have favoured formal nationality, rather than the test of effective (or dominant) nationality, in determining whether an individual qualifies as a ‘national of another [ICSID] Contracting State’ under the Convention.42 This preclusion of an investor who also holds the nationality of the host state does not apply where other arbitration mechanisms, such as the UNCITRAL arbitration rules, are used, provided that there is no separate express prohibition on dual nationals in the BIT itself.43

(ii) Legal entities

[8.20] All investment treaties extend the benefit of their protection to legal entities such as companies. Many BITs simply require that the entity be incorporated or constituted under the laws of one of the contracting parties. Some treaties add other requirements, such as the need actually to carry out business in the home state.44

[8.21] Where no such additional requirements have been stipulated, tribunals generally conduct a review limited to determining whether the legal entity satisfies the formal definition of investor under the treaty and refuse to incorporate additional requirements that the treaty drafters did not include. For instance, in Yukos Universal Ltd (Isle of Man) v Russian Federation,45 Russia argued that the claimant should not qualify as an investor under the ECT (which defines ‘investors’ based on the law under which an entity is organised), because it was a shell company that was owned and controlled by Russian nationals. In rejecting this argument, the tribunal held that it knew of ‘no general principles of international law that would require investigating how a company or another organization operates when the applicable treaty simply requires it to be organized in accordance with the laws of a Contracting Party’ and refused to ‘write new, additional requirements—which the drafters did not include—into a treaty, no matter how auspicious or appropriate they may appear’.46

[8.22] Similarly, in Tokios Tokelės v Ukraine,47 a majority of the arbitrators held that an investor incorporated in Lithuania qualified as an investor under the Lithuania– Ukraine BIT,48 even though 99 per cent of its shares were owned by Ukrainian nationals. By contrast, in TSA Spectrum de Argentina SA v Argentine Republic,49 a majority denied jurisdiction under the Netherlands–Argentina BIT50 over a claim against Argentina brought by an Argentine company claiming to be ‘controlled’ by its Dutch parent (which owned 100 per cent of its shares). The tribunal looked beyond the claimant’s immediate (Dutch) parent company and found that its ‘ultimate owner’ was an Argentine national. The tribunal then held that Article 25(2)(b) of the ICSID Convention’s ‘foreign control’ requirement for juridical entities having the nationality of the state party to the arbitration had not been met.

[8.23] The natural consequence of the formalistic language of most treaties is that adopting a particular corporate structure for the purposes of attracting the protection of an investment treaty is wholly legitimate (as it is in the case of tax structuring). In Aguas del Tunari SA v Republic of Bolivia,51 the tribunal rejected Bolivia’s objection that the ‘availability of the BIT was the result of strategic changes in the corporate structure’, noting that:

[I]‌t is not uncommon in practice, and – absent a particular limitation – not illegal to locate one’s operations in a jurisdiction perceived to provide a beneficial regulatory and legal environment in terms, for examples, of taxation or the substantive law of the jurisdiction, including the availability of a BIT.52

[8.24] Some tribunals have nevertheless drawn the line where the restructuring of an investment has been undertaken solely to gain BIT protections in relation to an earlier dispute, holding that this would amount to a lack of good faith and an ‘abusive manipulation of the system’.53 Thus, in Mobil Corporation Venezuela Holdings BV v Bolivarian Republic of Venezuela,54 the tribunal held that it had jurisdiction over claims relating to nationalisation measures taken by Venezuela after the corporate restructuring in question had taken place, but not claims relating to disputes over royalty and tax rates that had arisen prior to the restructuring.

[8.25] Some treaties seek to limit the scope of protection to protected investors by means of treaty clauses allowing the state parties to deny treaty benefits to investors that do not have substantial business activities in their home state and which are controlled by entities or persons of a third state (known as ‘denial of benefits’ clauses).55 While it remains a matter of debate whether these clauses can be applied retroactively (that is, after an investment is made),56 tribunals have held that respondent states bear the burden of proving that the requisite elements have been satisfied.57