Introduction to the Symposium on Simon Batifort and J. Benton Heath, “The New Debate on the Interpretation of MFN Clauses in Investment Treaties: Putting the Brakes on Multilateralization”

Simon Batifort and J. Benton Heath have advanced the radical view that most favored nation (MFN) clauses may not be effective in importing substantive standards of treatment from other treaties.¹ While states, tribunals, and academics have been debating whether MFN clauses can be used to incorporate procedural provisions from other treaties, Batifort and Heath now launch a new debate on the hitherto uncontroversial assumption that MFN clauses can in fact import substantive provisions. The participants in this symposium critically explore that thesis by raising questions of treaty interpretation; exploring whether there is any foundation in state practice for Batifort and Heath's view; and debating the role and objectives of MFN clauses, including the role they play in the multilateralization of international investment law.

Central to Batifort and Heath's analysis is the award in Íçkale v. Turkmenistan where the tribunal refused to apply an MFN clause, thus preventing the claimants from relying on the substantive protections of another bilateral investment treaty (BIT).² Batifort and Heath argue that after Íçkale, states, litigants, and scholars can no longer automatically assume that substantive standards of treatment can be imported by way of an MFN clause. Past decisions, they say, have been based on a “top-down” approach to interpretation under which decision-makers have imposed presumptions about the nature of MFN clauses on the wording of the treaty. A “bottom-up” approach that focuses on the terms of the treaty itself is now required.³ Thus, the MFN clause in each treaty must be interpreted on its own terms with no assumption that MFN clauses by their very nature incorporate substantive standards of treatment. The authors also refer to provisions in more recently negotiated treaties, which purport to place limitations on the use of MFN clauses. In the light of all these factors, they claim “the time has come to engage in a serious debate regarding the use of MFN clauses to import standards of treatment from one investment treaty to another.”⁴

Perhaps in recognition of the disruptive potential of this thesis, the editors of the American Journal of International Law published in the same issue a response by Stephan Schill.⁵ In contradicting the Batifort and Heath thesis, Schill's response serves to reassure investment lawyers and arbitrators that their received wisdom is still intact. While welcoming the redirection of the MFN debate from the “stale and stalemated” question of whether MFN clauses import procedure and jurisdiction from other treaties to the question of whether MFN clauses import substantive provisions, Schill challenges the fundamental premises of Batifort and Heath. He argues that interpretation of MFN clauses has always been on the basis of a “bottom-up,” treaty-by-treaty approach. Moreover, he argues, Batifort and Heath neglect the fact that the current interpretative approach “builds on the foundations of MFN clauses in general international law” and that MFN clauses are “essential pieces of the multilateral structures that underlie international investment law.”⁶ In considering the Batifort and Heath article, the contributors to this symposium also comment on and raise questions about the Schill response.

Andrea Bjorklund queries whether the “top-down” and “bottom-up” approaches to treaty interpretation that Batifort and Heath identify are really consistent with “ordinary treaty analysis.”⁷ Like Schill, she queries whether the Vienna Convention on the Law of Treaties (VCLT)⁸ regime is adequate to deal with the interpretive challenges of investment treaties, and she invokes the role of interpretive communities in influencing interpretation. However, she is concerned that Schill's broader approach may expand the concept of context too much. Bjorklund argues that intent is not a proper basis for interpretation, notwithstanding the frequency with which it is found in the decisions of investment tribunals. She also considers ways in which states have sought to clarify interpretation through official interpretative statements. Thus, she focuses on the NAFTA Interpretative Notes and the problems to which they have given rise, and she demonstrates the difficulty of finding state intent in attempted modifications of treaty provisions.

Martins Paparinskis also comes to the symposium from the perspective of the law of treaties.⁹ He agrees with Batifort and Heath that much more attention has to be given to the question of the use of MFN clauses to incorporate substantive standards of treatment, as compared to the debate about the extent to which MFN clauses import procedural provisions. Paparinskis argues that when such state practice is considered, Batifort and Heath overstate their argument that the “conventional wisdom” as applied by successive tribunals is incorrect. In particular, he gathers states’ comments in debates in the Sixth Committee of the General Assembly, in dispute settlement, and in their treaty-making practices as evidence that they have accepted the use of MFN clauses to incorporate substantive protections guaranteed by other treaties. The change that Batifort and Heath purport to identify, he reminds us, will only come about through the normal modes of international lawmaking, which is traditionally and primarily based on the practice of states.

Facundo Pérez-Aznar emphasizes the close link between MFN and National Treatment clauses and considers that the interpretation of MFN provisions should take this into account.¹⁰ MFN clauses are both primary rules that impose obligations as well as provisions that allow for importation from other treaties. In his view, the nature of MFN clauses is as important as the specific text in the interpretation of MFN clauses. Nonetheless, he points out, even tribunals that take a treaty-by-treaty approach in fact are influenced by presumptions about the scope and effect of MFN clauses. However, Pérez-Aznar criticizes the multilateralization thesis. Multilateralism, in his view, comes through the cooperation of states and not through the decisions of arbitral tribunals. He also challenges Schill's critique of the Batifort and Heath thesis as not ideologically neutral; Schill's approach, he contends, simply replaces one ideology with another.

Michael Waibel queries the corollary of the Batifort and Heath thesis that there is no default meaning of MFN clauses in international law.¹¹ He points out that most tribunals and commentators see a need to anchor MFN clauses in general international law. Waibel argues that there is a need to have common understandings of the meaning of MFN clauses, since recourse to the text of an agreement will often provide little guidance on the meaning of an MFN clause. In his view, states and scholars should give more attention to the functioning of MFN clauses across both trade and investment regimes, pointing to the similarities and differences between the way MFN clauses function under the Agreements of the World Trade Organization (WTO) and how they function under BITs. Ultimately, in his view, the Batifort and Heath thesis, if correct, would lead to a reduction in the successful invocation of MFN clauses in respect of both procedural and substantive provisions.

The Core of the Debate

There seems to be a consensus among the commentators that Batifort and Heath have sparked an interesting and timely debate, but at the same time the commentators share doubts about whether the details of the authors’ analysis are persuasive. In short, they are reluctant to accept that the conventional wisdom—that MFN clauses can be used to import substantive standards of treatment—is wrong.

But Batifort and Heath are perhaps saying something less than the commentaries ascribe to them. Their view that each MFN clause must be interpreted on its own terms and in accordance with the normal rules for treaty interpretation (as embodied in VCLT Articles 31 and 32) is hardly controversial. It is the corollary that seems to be more radical and the interpretation more questionable: that when properly interpreted, MFN clauses may not allow for the importation of substantive standards of treatment from another treaty. That view comes up against the conventional wisdom that the importation of substantive standards of treatment is precisely what MFN clauses are designed to do. The concern about the implications of the Batifort and Heath thesis is heightened when we remember that the MFN clause in Íçkale, the case that prompted this debate, was a relatively conventional MFN clause.¹² If the MFN clause in Íçkale did not incorporate substantive standards of treatment, then perhaps no MFN clause will.

In this regard, it is useful to look more closely at the MFN clause in Íçkale and what the tribunal actually did. The relevant part of that provision, Article II(2) of the Turkey-Turkmenistan BIT, which dealt with both national treatment and MFN treatment, provided that each party shall accord “treatment no less favourable than that accorded in similar situations to investments of its investors or to investments of investors of any third country.” The term “in similar situations” in the context of national treatment, the tribunal said, requires a factual comparison of the treatment of investments of investors of the home state with the treatment of investments of investors of third parties. The mere fact that they are both investing in the same state does not mean they are in a “similar situation.”¹³ The tribunal, then, was insisting on more rigorous analysis of the situation of the litigating investor and the third-country investor whose BIT it is alleged provides more favourable treatment.

In other words, the real contribution of Íçkale is that it gives content to the term “in similar situations,” a term found in some but not all MFN clauses. Thus, contrary to Schill's view,¹⁴ the reasoning in Íçkale cannot be discarded on the ground that it applies only to that case. In this respect, Bjorklund's reference to the NAFTA experience is apposite. The MFN clause in NAFTA includes the comparable term “in like circumstances,” a term the meaning of which has been the subject of controversy. But it makes the NAFTA parallel more relevant. What Íçkale has done is to suggest that such terms do have content—and that content could have implications for the application of the MFN clause.

The Issue of Treaty Interpretation

Íçkale does not really resolve the debate about whether MFN clauses can generally be used to import substantive provisions of treatment. The answer in part, as all of the commentators accept, depends on treaty interpretation. And, in a sense, this is the problem. While everyone accepts that investment treaties are treaties governed by the principles of public international law embodied in the VCLT, the challenge of the “New Debate” is that there is no accepted interpretative theory or practice for the application of the VCLT rules.

This is evident in the difference between Batifort and Heath and Schill. Batifort and Heath refer to the Article 31 rule that interpretation is based on the trilogy of ordinary meaning, context, and object and purpose. They then mention other specific principles of interpretation, but do not go further and discuss how the interpretative rules are to be applied in fact.¹⁵

Schill does not focus on the VCLT rule but refers instead to “the general international law framework governing the interpretation of MFN clauses in any bilateral treaty.”¹⁶ This is a prelude to Schill's thesis that interpretation cannot be limited just to considering each treaty on a treaty-by-treaty basis, but that the interpretation of investment treaties must follow “multilateral rationales.”¹⁷ Rather than being based on the VCLT, this approach to interpretation seems to locate the interpretation of MFN clauses in the assumption that the overriding purpose of MFN clauses is the multilateralization of treaty commitments. Schill's argument is rich and evocative, reminiscent of the teleological approach to treaty interpretation that was part of the debate about treaty interpretation during the negotiation of the VCLT. But VCLT Articles 31 and 32 provided a compromise. These articles endorsed neither the literal nor the teleological approach, though both concepts are implicit in the trilogy of ordinary meaning, context, and object and purpose.

The underlying issue in this debate is how investment tribunals should engage in the interpretative process. There is no doubt that tribunals have to interpret investment agreements, but there is less clarity on how they should go about doing so. Neither in this symposium nor in the literature nor in the reasoning of investment tribunals has a coherent approach emerged that would provide a common basis for understanding how investment treaties are to be interpreted and applied.

The VCLT rules are generally understood as the proper framework for interpretation. However, investment tribunals often treat the VCLT rules as a smorgasbord from which they can choose whichever approach supports the conclusions that the tribunal wishes to reach. Paying lip service to the VCLT rules and then doing something that has no basis in those rules is hardly a principled approach to interpretation. If there is no methodological consistency in the application of the VCLT rules, then interpretation of investment treaties becomes little more than the ad hoc preferences of arbitrators dressed up as an application of VCLT Articles 31 and 32.

The Economic Issue–Investment and Trade

It matters to individual claimants whether MFN clauses can be used to incorporate substantive rights, but does it matter to the international investment system as a whole? MFN clauses originated in treaties of friendship, commerce, and navigation (FCN treaties). They were designed to ensure that the nationals of one state were not treated less favorably than the nationals of another state. In the field of trade, goods produced by the nationals of one foreign state were not to be put at a competitive disadvantage in the territory of the grantor state as compared with the goods produced by nationals of other states. And national treatment had the same objective but with a different comparator. In short, MFN clauses in the days of FCN treaties were provisions of nondiscrimination that had an economic objective and consequence.

One question to ask is whether the routine application of MFN clauses to substantive provisions of other investment treaties (the “conventional wisdom”) finds justification on economic grounds. In other words, do limits on the application of MFN clauses to substantive provisions produce negative economic consequences, as they would in the case of trade in goods? Do MFN clauses in investment treaties, like MFN clauses in trade treaties, serve an economic function?

If investment protection was solely about the protection of the “rights” of investors, as Batifort and Heath and Pérez-Aznar imply, then the broader economic implications of allowing the incorporation of substantive provisions from other investment treaties would not be of great significance and certainly not an overriding consideration in determining the scope of an MFN provision. An interpretation that did not admit the incorporation of substantive provisions would hardly be denying effet utile to the clause. But if MFN clauses serve an economic function, as is the case for MFN under the General Agreement on Tariffs and Trade (GATT), then we should assess the economic implications of either the “conventional wisdom” or the new approach of Batifort and Heath. And surely MFN clauses do have an economic function beyond protecting the rights of investors. Equality of opportunity, or avoiding giving a competitive advantage to some investors, encourages investment, consistent with the objective of investment agreements. In that sense, MFN clauses in BITs have the same economic objective as they have under GATT. The economic implications for the international investment system of permitting or restricting the incorporation of substantive provisions from other treaties thus must be part of the “New Debate” over the scope of MFN clauses.

There are, however, limits to drawing parallels between MFN clauses under GATT and MFN clauses in BITs, particularly on the question of the multilaterizing function of MFN clauses. Multilateralization occurred in the field of trade not because of the use of MFN clauses—it occurred because states included such a clause as the core obligation in a multilateral agreement. The fact that GATT Article I applies to all WTO Members is what multilateralizes that obligation; it is not because its MFN clause as such was designed to effect that multilateralization.

Moreover, as Michael Waibel points out, the breadth of scope of the wording of GATT Article I—“any advantage, favour, privilege or immunity”—and the fact that it applies to specific areas suggests that our experience with MFN clauses in the field of trade might not readily apply in the field of investment. While it may be doubtful whether MFN clauses in the field of investment were designed to ensure that the benefits of other treaties would automatically be provided to the beneficiaries of MFN clauses, can the same be said of MFN clauses in trade? Schill invokes the WTO treatment of MFN clauses in support of the argument that MFN clauses can apply to more favorable treatment extended by the granting state on the basis of international treaties.¹⁸ However, the issue has yet to arise under GATT Article I, although there is some suggestion that MFN under Article II of the General Agreement on Trade in Services would be broad enough to cover benefits under other treaties.¹⁹

Conclusion

Batifort and Heath have not only raised a question about the use of MFN clauses to incorporate substantive provisions from other treaties, but they have also prompted a broader debate about the role and function of MFN clauses and their interpretation and application. The various commentaries in this symposium attest to the timeliness of the discussion and the significance of the issues raised. The issues are unlikely to be readily resolved. International investment law resembles in some respects the English common law in the early days before hierarchies were developed amongst courts and before authoritative forms of reporting emerged, with individual decisions lacking consistency and coherence. Will problems of consistency and coherence in treaty interpretation be dealt with by investment tribunals on a case-by-case basis; will a unifying investment court with appellate functions take root; or will investor-state arbitration slowly atrophy and diplomatic protection reemerge? Batifort and Heath have raised questions that are in reality just a microcosm of a much broader contemporary debate.