Image may be NSFW.
Clik here to view.Dr Anastasios Gourgourinis is Lecturer in Public International Law at the National and Kapodistrian University of Athens Faculty of Law, and Research Fellow at the Academy of Athens.
Martins Paparinskis’ EJIL Article argues that the conceptual challenges faced by contemporary investment treaty arbitration can be effectively addressed if the regime is not viewed in isolation from its progeny, i.e. international human rights and consular law, the law of treaties and the law of diplomatic protection. The discussion in Paparinskis’ piece is essentially centred on the debate regarding the character and nature of investors’ rights under international investment agreements (IIAs), i.e. either as direct or derivative rights. Paparinskis address the nature of investors’ rights under IIAs from the terminologically different, but contextually similar, lens of the models of direct rights, beneficiary rights, and agency. He explicitly declines to take a firm position regarding which of these models is the most plausible one, but he appears, at least in my eyes, more prone to side with the direct rights model, especially in view of the analogies with human rights law (see also his analysis here). This is where we will have to part ways, and in this post I take issue with this, largely drawing from my chapter on the direct/derivative rights debate.
Below, I take issue with the idea that an investor may be considered as a holder of direct rights akin to human rights. Although Paparinskis does a tremendous job in drawing normative parallels between the two regimes, I remain of the view that the very different nature of obligations derived from human rights instruments and IIAs cautions against such an approach, and rather supports the derivative model of investors’ rights. The debate over the direct or derivative nature of investors’ rights received full treatment in the seminal BYBIL article by Zachary Douglas and evidence that it is ongoing and growing, is manifested by the fact that during the recent 2013 ILA Regional Conference in Sounion, Greece, three papers touching upon these and similar issues were presented (see here, here and here).
Existing state practice favours the derivative model
The direct/derivative rights debate has mainly developed in the context of North American Free Trade Agreement (NAFTA) Chapter Eleven Disputes. Initial indications of the debate appeared in the Loewen v. USA Award. The most controversial instances where the nature of rights enjoyed by NAFTA investors were put to the test were the awards in the ADM v. Mexico, Corn Products v. Mexico and Cargill v. Mexico investment disputes under NAFTA, where Mexico unsuccessfully sought to justify the breach of its Chapter Eleven obligations as a lawful non-forcible countermeasure taken in response to an alleged previous violation by the United States (the home-state of the three investors) of its obligations under Chapters Three and Seven as well as the frustration of the inter-State dispute settlement mechanism under Chapter Twenty (see here).
Any relevant practice by NAFTA parties is stricto sensu limited to and relevant for the four corners of NAFTA. Nevertheless, I take the view that state practice (as evidenced in pleadings and submissions in the context of Chapter Eleven disputes) on the nature of investors’ rights under NAFTA, qua treaty, may indeed provide a first indication regarding the prevalence of the derivative model in general.
Under Article 1128, NAFTA Parties may present their submissions to arbitral tribunals on a question of interpretation of NAFTA. In Loewen v. USA the Tribunal stated that NAFTA investment arbitration is but an instance ‘where claimants are permitted for convenience to enforce what are in origin the rights of Party states [to the NAFTA]’ (para. 233). This statement should not be seen in isolation. In the course of the proceedings, Mexico, as a non disputing NAFTA party, filed its Third Article 1128 NAFTA submission where it unequivocally stressed that NAFTA rights ‘exist at the international plane between the States inter se’ (para. 28), and that the rights and obligations under Chapter Eleven NAFTA do not ‘differ from the rest of the Treaty (which, in general, is subject only to State-to-State dispute settlement)’ and are held by the Parties alone (para. 29). The United States, Respondent in that case, agreed with Mexico in its Response [p. 8]. Even more categorical appears to be the position of Canada regarding the obligations under NAFTA Chapter Eleven: ‘They are not owed directly to individual investors. Nor do investors derive any rights from obligations owed to the Party of which they are nationals.’ (Methanex v USA, Second NAFTA Article 1128 Submission of Canada, para. 9)
The ADM Tribunal (para. 176) did consider the above and similar statements when siding with the derivative rights model, and it is difficult to understand how and why the Corn Products and Cargill Awards declined to take them into account. Be that as it may, the fact remains that the above statements by the NAFTA Parties may indeed qualify as subsequent practice in the application of the NAFTA which establishes the agreement of the NAFTA parties regarding its authentic interpretation under Article 31(3)(b) of the 1969 Vienna Convention on the Law of Treaties. Even if relevant state practice outside NAFTA appears to be largely lacking, the view that investors are to be considered as derivative rights’ holders indeed finds firm support, at least in the NAFTA context.
Turning from the nature of rights to the nature of obligations under IIAs
One of the aspects of Paparinskis’ analysis I am reluctant to accept is the idea that an investor may be considered as a holder of direct rights akin to human rights. However, the nature of obligations derived from human rights instruments and IIAs are very different and those differences tend to support the derivative model of investors’ rights.
Sir Gerald Fitzmaurice, in its Third Report on the Law of Treaties (UN Doc. A/CN.4/115, YILC (1958), Vol. II, Commentary to Article 19, at 27) famously drew the distinction between multilateral treaty obligations which are not of the mutually reciprocating type, but which are either of the interdependent or the integral type. Human right treaties are hence considered as establishing obligations of ‘integral’ character, not dependent on a corresponding performance. i.e. without requiring reciprocity. It was in this spirit that the International Court of Justice in the Reservations Advisory Opinion referred to the 1948 Convention on the Prevention and Punishment of the Crime of Genocide as a (human rights) instrument under which ‘contracting States do not have any interests of their own’ ([1951] ICJ Rep 15, at 23); or, that the European Court of Human Rights held that ‘[u]nlike international treaties of the classic kind, the Convention comprises more than mere reciprocal engagements between Contracting States. It creates, over and above a network of mutual, bilateral undertakings, objective obligations’ (Ireland v. United Kingdom, EHRR 2 (1979–80): 25, para. 239).
Nevertheless, the picture is substantially different in the investment law context. In contradistinction to human rights instruments, IIAs are far from creating integral obligations and their performance is rather underlined by reciprocity. For instance, Article 1115 NAFTA prescribes the settlement of disputes under Chapter Eleven ‘in accordance with the principle of international reciprocity’, while investment arbitral tribunals have emphasized the reciprocal character of obligations derived under IIAs, often indicated in their very preambles; as the Daimler v. Argentina Tribunal stated in so many words, ‘‘[i]t would be an error to start from the assumption that the bilateral and synallagmatic dimension of [bilateral investment treaties] is of a mere rhetorical nature’ (para. 162). I hence find myself hesitant to ground the direct right model for investors’ rights upon the human rights paradigm, exactly due to the different nature of obligations involved.
A pertinent illustration of the derivative model: HICEE v Slovakia
Paparinskis does not see much in HICEE v Slovakia Partial Award that could advocate for the derivative nature of investors’ rights. This is another point of our inter se disagreement.
In HICEE, the Tribunal examined various ways so as to take into account the Dutch Explanatory Notes when interpreting the BIT between the Netherlands and the Czech and Slovak Federal Republic; it eventually qualified the Notes as supplementary material falling under Article 32 VCLT. That was crucial, since these Explanatory Notes excluded sub-subsidiaries from the BIT’s protective scope. In order to support this conclusion, the Tribunal made the hypothesis of the same dispute arising at the interstate level (between Netherlands and Slovakia), to then observe that the Notes would have surely operated so as to bar the Netherlands from adopting a contrary position vis-à-vis Slovakia, i.e. in terms of application of estoppel. Extrapolating this conclusion at the investor-State level, the Tribunal eloquently reasoned:
‘As the Tribunal has already pointed out, the present question of interpretation could have arisen in inter-State proceedings under Article 10. If it had done so, it would have attracted the consequence that the decision of the tribunal “shall be final and binding on both Contracting Parties”. That cannot possibly mean that the arbitral decision would be binding on the States only, but without effect on an investor claiming derivatively through the rights procured for it by one of those States.’ (para. 139)
The above statement can hardly be reconciled with the direct rights model, which isolates investors’ rights from the conduct, obligations and rights of their home State. One can easily identify the futility of advancing a similar argument in the human rights context, exactly due to the integral character of obligation under human rights treaties. The potential effects of interstate estoppel indirectly accruing to investors, as highlighted in HICEE, hence demonstrate that investors’ rights under BITs are not direct and there exist good (normative) reasons not to be considered such.