Against the foregoing, it is appropriate to turn to the principle of proportionality in the investor-state arbitration context. Of interest here is examining the role proportionality has played in the awards in which it has been invoked, paying particular attention to the justifications offered for its use and the manner in which tribunals have addressed the issues of interest identi- fication, weight, commensurability, and standards of review.
1. a ReVIeW OF THe PRINCIPal aWaRDs
a. S.D. Myers and Pope & Talbot v. Canada
The earliest reference to proportionality, directly or implicitly, in any publicly available investment-treaty decision appears to have been in 2000 in S.D. Myers v. Canada, an award of
involving claims of expropriation. For a detailed analysis of Lithgow, see Maurice Mendelson, ‘The United Kingdom nationalization cases and the European Convention on Human Rights’ (1986) 57 British Yearbook of International Law 33.
127. This point has been made with particular clarity by Stephen Ratner. As Ratner has highlighted, the pur- poses and institutional design of the European human rights project have been central to the development of the Court’s approach to claims arising under Protocol No. 1 and an understanding of those purposes and that design is central to understanding the Court’s development of its Protocol No. 1 case law, especially the adop- tion of the principle of proportionality and the introduction of a margin of appreciation to guide the Court’s standard of review. See Stephen R. Ratner, ‘Regulatory takings in institutional context: beyond the fear of fragmented international law’ (2008) 102 American Journal of International Law 475. It also has been sug- gested that approach taken by the Court may reflect a long term institutional strategy, whereby it has sought broad acceptance of the European human rights mechanism through what may be described as a ‘softly-softly’
an ad hoc tribunal established under Chapter 11 of the North American Free Trade Agreement (NAFTA).128 There, the tribunal invoked the principle of proportionality in its consideration of whether Canada had breached Article 1102 of the NAFTA.129 In so doing, the tribunal asked, inter alia, whether the practical effect of the relevant measure—to ban the export of PCBs from Canada—had been to create a disproportionate benefit for nationals over non-nationals.130 The tribunal did not elaborate on why it believed proportionality was relevant in the context of the national treatment standard of the NAFTA. Instead, in rather sparse reasoning, the tribunal accepted that Canada’s aim (to maintain the ability to process PCBs within Canada in the future) was a legitimate one, but concluded nevertheless that “[t] here were a number of legitimate ways by which Canada could have achieved [this goal], but preventing SDMI from exporting PCBs for processing in the USA … was not one of them.”131 Consequently, the tribunal determined that there had been a violation of NAFTA Article 1102.
In S.D. Myers the principle of proportionality was invoked by the tribunal as a standard against which to evaluate the state’s argument that discriminatory treatment should be excused as an unactionable by-product of measures taken in pursuit environmental aims. It was only after determining that there had been discriminatory treatment that the tribunal turned to the principle of proportionality. Proportionality thus underscored the tribunal’s (consequentialist) conception of the investor’s right to national treatment and also signaled the tribunal’s tacit con- clusions about the commensurability of the investor’s right to national treatment and the state’s justification of its measures to protect the environment. The tribunal offered no explicit justifica- tion for its adoption of proportionality as an analytical framework, although its heavy reliance on World Trade Organization (WTO) decisions and its reference to the General Agreement on Tariffs and Trade (GATT) (and Article XX in particular) is suggestive of a source for its inspi- ration.132 The NAFTA, however, contains no analog to Article XX of the GATT applicable to Chapter 11 investment claims and the tribunal’s decision to “cross-pollinate” from WTO juris- prudence is of doubtful authority.133 Notably, in applying its principle of proportionality, the tribunal provided no discussion of the standard under which its review should proceed, that is, was Canada entitled to any deference in its determination that an export ban was appropriate?
The use of proportionality in S.D. Myers was not met with uniform endorsement by other NAFTA tribunals. In Pope & Talbot v. Canada, decided shortly after S.D. Myers, Canada attempted to build upon the use of proportionality in S.D. Myers by arguing that a finding of
approach of interpretation and application. See Brigitte Stern, ‘Le droit de propriété, l’expropriation et la nationalisation dans la Convention européenne des droits de l’homme’ (1991) 17 Droit et Practique du Commerce International 394, 403 (describing the Court’s ‘least common dominator’ approach to the protec- tion of private property (‘le plus petit commun dénominateur européen de protection de la propriété privée’)).
128. S.D. Myers v Canada (Partial Award, 2001) ad hoc UNCITRAL Trib.
129. ‘1. Each Party shall accord to investors of another Party treatment no less favorable than that it accords, in like circumstances, to its own investors with respect to the establishment, acquisition, expansion, manage- ment, conduct, operation, and sale or other disposition of investments.’ NAFTA 1102(1). Stone Sweet gives the impression that the invocation of proportionality in S.D. Myers was in connection with the tribunal’s interpretation of NAFTA art 1105, addressing the international minimum standard, see Stone Sweet (n 54) 64–65, but this is not so.
130. S.D. Myers [252].
131. ibid [255].
132. ibid [246].
133. See Andrea K. Bjorklund and Sophie Nappert, ‘Beyond fragmentation’ in Todd Weiler and Freya Baetens (eds), New Directions in International Economic Law (Brill 2011).
de facto discrimination could only be shown where the claimant was able to demonstrate the disproportionate effect of the measure under challenge.134 According to Canada, in order to establish a de facto violation of Article 1102, the tribunal should employ a two-step analysis.
First, the tribunal must establish whether any Canadian-owned investments received the same treatment as the foreign-owned investment. Second, the tribunal must compare the size of that group of Canadian-owned investments to the size of the group of Canadian-owned invest- ments receiving more favorable treatment. Only if the size of the disadvantaged Canadian group was smaller than the size of the group of advantaged Canadian-owned investments would a finding of a violation of Article 1102 be appropriate.
Canada’s argument effectively would have created a proportionality defense to claims of de facto violations of Article 1102. On Canada’s argument, states would have been entitled to dis- criminate in favor of national investments so long as the aggregate of discrimination in favor of nationals did not outweigh the aggregate number of nationals not receiving the advantageous treatment. The Pope & Talbot tribunal flatly rejected the Canadian approach and Canada’s purported reliance on S.D. Myers:
Once the Myers tribunal found that the claimant and its Canadian competitors were in “like circumstances,” the finding of a denial of national treatment was a foregone conclusion. That is, the situation at that point was that two Canadian companies were free to operate, while their American competitor was effectively out of business. Weighing of proportionate advantages and disadvantages was not required. …135
The use of proportionality advanced by Canada in Pope & Talbot stems from the same con- sequentialist position as all proportionality arguments do: a right is not a trump but merely a claim for interest maximization. But Canada’s argument in Pope & Talbot also suggested a novel twist on the commensurability and weight concerns. For while concerns over commensurability arise when one considers balancing the substantive rights of an individual against the right of the state to advance a collective interest, Canada’s argument would have required the investor to aggregate the assertion of its rights with those of similarly disadvantaged Canadian investors in order to assess the weight of the investor’s claim. On Canada’s argument, the individual inves- tor’s right under Article 1102 would have been transformed into little more than a data point in a larger exercise to determine a collective degree of disadvantage. The weight of the investor’s right would have become contingent. So long as a sufficient number of Canadian investors were dis- advantaged in the same way as the foreign investor, no Article 1102 claim could succeed unless
“the differing treatment between the class of [foreign] investors and their Canadian competitors in like circumstances is ‘disproportionately’ in favour of the domestic investments… .”136
134. Pope & Talbot Inc. v Canada (Award on the Merits of Phase 2, 2001) Ad hoc UNCITRAL Trib., [65].
The claimant made a similar argument in Methanex Corp. v United States (Final Award on Jurisdiction and Merits, 2005) Ad hoc UNCITRAL Trib., Part IV.B.2.iii, [9] . In that case, the tribunal rejected the claim with- out reaching the issue. ibid [38]. In Enron Corp. v Argentina, the Argentine Government invoked the propor- tionality of its measures in response to Enron’s claims of discrimination in violation of Article II(2)(b) of the Argentina-United States bilateral investment treaty. Enron Corp. v Argentina (Award, 2007) ICSID Case No ARB/01/3, [279]. The tribunal rejected the claim of discrimination without particular reference to the issue of the proportionality of the measures, conflating Enron’s claim of a violation of the protection against discrimi- natory measures with its claim for violation of the protection against arbitrary measures. ibid [282].
135. Pope & Talbot [66].
136. ibid [71].
Even if the tribunal had been willing to accept the basic premise of the argument, it would have needed to confront the question of how to measure and weigh these aggregations of advan- tage and disadvantage and determine whether the balance struck was disproportionate. The Pope & Talbot tribunal was clearly skeptical that this could be done in a principled way, placing the term “disproportionately” in quotation marks and wondering aloud “whatever that might mean.”137 The tribunal’s doubts about using the principle of proportionality as a yardstick for adjudicative decision-making reaffirms the critique made above concerning the assertion that the proportionality principle is capable of application without reference to larger claims about social and political values.138 Where are the norms, the tribunal seems to be asking, that will enable us to make a principled determination on this question? Facts do not weigh themselves.
b. Tecmed and Its Progeny
Significant treatment of proportionality has occurred as well in the context of claims alleg- ing indirect expropriation. The earliest (and most influential) invocation in the expropria- tion context is in Técnicas Medioambientales Tecmed SA v. Mexico, a claim arising under the Spain-Mexico investment treaty.139 There, the Spanish investor alleged that an expropriation occurred when the Mexican authorities refused to renew a permit necessary for the investor to continue operating a landfill. In response, Mexico acknowledged that the permit had not been renewed, but argued that the denial had occurred pursuant to a decision designed to maintain ecological balance and to respond to community concerns. Accordingly, Mexico claimed that the denial of the renewal could not constitute an expropriation or a “measure with similar characteristics or effects to expropriation”140 because it came about through Mexico’s exercise of its “police powers.”141
In addressing the parties’ conflicting claims, the tribunal took the view that in order to determine whether a regulatory measure amounted to a compensable expropriation under the Spain-Mexico investment treaty, it was not enough to consider only the effects of the measure on the investment but that it must also determine whether the effects of the measure were disproportionate to the ends sought to be achieved by the measure. Relying directly upon case law from the European Court of Human Rights applying Protocol No. 1, the tribunal stated:
After establishing that regulatory actions and measures will not be initially excluded from the definition of expropriatory actions, in addition to the negative financial impact of [the expro- priatory acts] or measures, the Arbitral Tribunal will consider, in order to determine if they are to be characterised as expropriatory, whether such actions or measures are proportional to the public interest presumably protected thereby and to the protection legally granted to invest- ments, taking into account that the significance of such impact has a key role upon deciding the proportionality.142
137. ibid 138. See Part B.3.
139. ARB(AF)/00/2 (Award, 2003).
140. Spain-Mexico Agreement for the Reciprocal Promotion and Protection of Investments (1996) art 5(1).
141. Tecmed [97].
142. ibid [122] (citing James v United Kingdom, (1986) 8 E.H.R.R. 123).
The Tecmed tribunal offered no support for its reliance on the decisions of the European Court of Human Rights, and nor did it otherwise justify the invocation of the principle of propor- tionality in its formulation of a test to determine the existence of an expropriation. Rather, the tribunal simply borrowed the European jurisprudence and placed proportionality at the center of its indirect expropriation analysis.
As laid out by the Tecmed tribunal, a claim of indirect expropriation based upon the application of a regulatory measure is to be addressed in two parts. In the first instance, the tribunal must look to the effects that the regulation has had upon the investment.143 If those effects rise to the level of interference necessary for a finding of expropriation—radical deprivation of the economic use and enjoyment of the investment144—then the tribunal must proceed to consider whether notwithstanding the effects of the measures, the mea- sures might nevertheless be excused from classification as a compensable expropriation by reference to their purpose,145 specifically “whether such actions or measures are proportional to the public interest presumably protected thereby and to the protection legally granted to investments.”146
Having established its two-step test, the Tecmed tribunal declared its intention to give “due deference” to Mexico’s determination of “the issues that affect its public policy or the interests of society as a whole, as well as the actions that will be implemented to protect such values.”147 Rhetorical flourishes about deference, however, did not translate into the actual application of deference in the tribunal’s analysis. For while the tribunal does appear to have been willing to extend deference to Mexico with respect to its choice of the objectives to be pursued in the name of the public good and in the exercise of its police powers, the tribunal took an altogether more robust approach to its examination of the motivations and grounds for Mexico’s regula- tion. Thus the tribunal conducted a searching review to determine whether, indeed, Mexico had adopted the regulation for environmental purposes, as it had claimed, or whether it had adopted the regulation in response to political pressure from the local community, as Mexico had suggested was merely a peripheral concern for its actions.148 Moreover, once the tribunal had concluded that the real motivation for Mexico’s refusal to renew the investor’s permit was
143. ibid [116].
144. ‘It must be first determined if the Claimant, due to the Resolution, was radically deprived of the eco- nomical use and enjoyment of its investments, as if the rights related thereto—such as the income or benefits related to the Landfill or to its exploitation—had ceased to exist. In other words, if due to the actions of the Respondent, the assets involved have lost their value or economic use for their holder and the extent of the loss.’ ibid [116].
145. ibid [119]: ‘The principle that the State’s exercise of its sovereign powers within the framework of its police power may cause economic damage to those subject to its powers as administrator without entitling them to any compensation whatsoever is undisputable.’
146. ibid 122. This is a considerable expansion of the concept of ‘police powers’ recognized in customary inter- national law. Rather than any limitation of the police powers concept to the protection of public health, safety, morals or welfare or pursuant to a system of bona fide tax regulation, as some academics have suggested, see e.g., George Christie, ‘What constitutes a taking of property under international law?’ (1962) British Yearbook of International Law 38, the tribunal appears to have replaced these categories with a generalized balancing test. This approach reflects the consequentialist treatment of rights essential to a proportionality analysis and seems open to the kind of horizontal normativity (and consequential ad hoc decision-making) proposed by Beatty.
147. Although the tribunal did not explain the basis for its invocation of deference, it seems again that the tribunal was inspired by statements in the European Court of Human Rights’ jurisprudence.
148. ibid [123]–[133].
to assuage community opposition, the tribunal proceeded to conduct a de novo inquiry into whether such an aim could justify the decision to refuse the permit and effectively shut down the investor’s business. Following a somewhat meandering recitation of factual findings sur- rounding the decision to refuse the license renewal, the tribunal concluded that no justifica- tion was possible and that, accordingly, the measure was disproportionate and, therefore, an indirect expropriation.149
The use of proportionality in Tecmed is problematic for a number of reasons, not least of which is the tribunal’s unexplained and extensive adoption of European Court of Human Rights jurisprudence as applicable in a case not arising under the European Convention on Human Rights, involving a non-European state (Mexico), and without any attempt to show that the principles applied reflect either customary international law or general principles of law. As described earlier, the jurisprudence of the European Court of Human Rights repre- sents that court’s articulation of the unique constitutional settlement established under the European Convention and the specialized phrasing of Protocol No. 1.150 The European settle- ment on these issues is not a universal settlement by any means and the Tecmed tribunal made no effort either in logic or doctrine to explain its applicability.
Moreover, the tribunal’s invocation of proportionality reveals the inherent limitations of the principle of proportionality as a juridical tool. As in Pope & Talbot, the Tecmed tribunal was willing to treat the investor’s rights as commensurable with the state’s asserted interests, although this issue passed without examination by the tribunal. Unlike Pope & Talbot, how- ever, the Tecmed tribunal seems to have considered that there might exist a hierarchy of aims that a state may address through its regulatory power. Thus, for the tribunal, the determina- tion as to whether Mexico was acting to address an ecological concern or was acting to address local political pressure was of pivotal importance for its inquiry. While a state may legiti- mately act toward both ends, the tribunal intimated, the pursuit of the environmental goal is of greater normative weight that the aim of addressing grassroots political pressure. Why and how this should be the case is neither raised nor addressed by the tribunal, but the differing amount of weight given by the tribunal seems in the end to be determinative. This comes as no surprise, of course, given proportionality’s need for the input of external values. What is sur- prising, however, is the tribunal’s unexamined approach to the decision on weight: ipse dixit pronouncements substitute for reasoning and evidence.
One may accept that within the range of aims that a state may pursue there are vary- ing degrees of importance to the public interest. But recognizing this truism does not put one closer to establishing criteria by which to establish a normative hierarchy of state aims, nor indeed does it assist in establishing normative priority between identified state aims. To reach such conclusions in a principled way requires the injection of values external to the proportionality analysis. In Tecmed the tribunal appears to have formed a tacit view about
149. ibid [129]–[151]. It is worth comparing the approach of the Tecmed tribunal in this respect with the approach taken by the International Court of Justice in the ELSI case. In ELSI, the Mayor of Palermo issued an order requisitioning the U.S. investors’ factory, which was later determined by the Italian courts to have been ultra vires of his legal authority. The International Court of Justice (ICJ), however, found that no violation of international law had occurred, reasoning in part that ‘[i] t was of course understandable that the Mayor, as a public official, should have made his order, in some measure, as a response to local public pressures;
and the Chamber does not see … any ground on which it might be suggested that the order was arbitrary [as a result].’ Case Concerning Elettronica Sicula S.p.A. (ELSI) (United States v Italy) (Judgment of 20 July 1989) I.C.J. Rep. 75, [126].
150. See Part C.