Putting the MFN Genie Back in the Bottle

This essay underscores the importance of background understandings in general international law for interpreting brief, open-ended clauses such as most favored nation (MFN) clauses. Contrary to Simon Batifort and J. Benton Heath's claim, I suggest that often interpreters of MFN clauses cannot limit themselves to the text, context, and preparatory materials of a specific MFN clause. A common international negotiating technique, including for investment treaties, is to rely on the general background understanding of what a clause typically means in international law—its default meaning. I also argue that MFN clauses have played a surprisingly limited role in the international investment regime to date. In the main, they have functioned as a stepping stone for procedural and substantive guarantees found in third-party investment treaties. This use, and the limited role of MFN clauses in investment treaty awards, stands in sharp contrast to MFN clauses in the trade regime.

This essay first examines whether an exclusive focus on the specific text of a single MFN clause is possible. It emphasizes the importance of the typical meaning of an MFN clause in general international law, alongside the specific clause at issue. It then turns to examine the role that MFN has played in the international investment regime to date, contrasting it with the trade regime. Batifort and Heath's proposal could further limit the role of MFN clauses in the investment treaty regime.

The Default Meaning of MFN Clauses in General International Law
Batifort and Heath contend that there is no such thing as the MFN clause. Indeed, there is more variation across investment treaties in how MFN clauses are formulated than is often assumed-for example, whether MFN clauses expressly refer to "like circumstances." Their claim's logical corollary is that there is no default meaning of MFN clauses in international investment law. Any such default understanding of what MFN clauses mean would, like the orthodox approach, adopt a "top-down" approach-which Batifort and Heath reject. In keeping with their "bottom-up" approach, they suggest that the meaning of a given MFN clause depends entirely on its specific terms.
Yet hardly any investment tribunal or writer favors-at least explicitly-a "top-down" approach that rides roughshod over the text of the applicable investment treaty. Investment tribunals commonly interpret investment treaty provisions-whether they are fair and equitable treatment clauses, MFN clauses, or dispute settlement provisions-bilateral investment treaty (BIT)-by-BIT. While emphasizing the importance of specific treaty terms, however, most tribunals and commentators also agree on the need to anchor the interpretation of MFN clauses in general international law. Some appeal to the principle of systemic integration for this purpose. 3 Indeed, relying exclusively on the specific terms of an MFN clause is not conducive to predictability in the investment treaty regime. The result could be a more fragmented jurisprudence-a result that at least some actors in the investment treaty regime may prefer to a uniform interpretation of the MFN clause. However, such fragmentation is unlikely to be good news for investors due to the lack of predictability of how future investment tribunals will apply MFN clauses.
The real disagreement centers on the default meaning of MFN clauses in general international law. What is the default position against the background of which an investment treaty tribunal ought to interpret a given MFN clause? As the International Law Commission (ILC) emphasized, this question is not only a narrow question of interpretation, but a far broader question of the role and function of MFN clauses in international law. 4 Contrary to Batifort and Heath's claim, it is extremely difficult to ignore all background understandings of what a clause in a treaty typically means-and even more so for MFN clauses that are frequently drafted in an openended, brief manner. A common international negotiating technique, including for investment treaties, is to rely on the general background understanding of what this type of clause typically means in international law. Unless the negotiating states opt for significantly higher precision and detail in future investment treaties, or expressly contract out of the background understanding, this understanding will continue to play a crucial role in investment treaty interpretation.
Batifort and Heath's call for a "bottom-up" approach provides rhetorical flourish for their normative claim about the appropriate interpretation of MFN clauses, given the generally negative connotation of "top-down" approaches. 5 But by itself, this label does not provide an answer to the central interpretive question for MFN clauses that are silent on whether they apply to substantive protections: does the MFN clause in question cover substantive protections in third-party investment treaties?
For MFN clauses, treaty interpreters will look in vain at the text, or the context, to answer this question. Moreover, travaux préparatoires are unlikely to shed much light on what the states drafting the investment treaty intended. Travaux remain relatively uncommon for investment treaties, in part because of the frequent use of model investment treaties in their negotiation. And even when travaux exist, they do not typically contain any discussion of what the parties intended to achieve with the MFN clause. Subsequent practice may assist in this interpretive task, to a degree. But it too is rare. By necessity, the default position in general international law is thus part of the interpretive exercise.
The real aim of Batifort and Heath's proposal-which they should have acknowledged in their article-appears to be to replace the existing default understanding of MFN clauses with another default position: either that investors, as a rule, cannot use MFN clauses to invoke substantive standards in third-party investment treaties, or that each tribunal has unfettered discretion to decide this central issue, without much guidance from the treaty text, because no default understanding applies.

Differences in How MFN Clauses Operate in the Investment and Trade Regimes
On the surface, MFN clauses are a central pillar of the investment treaty regime, being found in more than 95 percent of all investment treaties in force. 6 Despite their frequent inclusion in treaties, however, MFN clauses have thus far played only a subsidiary role in investment treaty awards, except as a stepping stone to procedural and substantive guarantees found in other investment treaties. The low salience of MFN clauses in the investment treaty regime contrasts with the World Trade Organization (WTO), where the MFN obligation is the "quintessential GATT [General Agreement on Tariffs and Trade]/WTO discipline." 7 MFN clauses are less important to the investment treaty regime than is often assumed. No investment tribunal in a publicly available award to date has found that a host state breached an MFN clause by treating a foreign investor from state A worse than an investor from state B. 8 Thus far, MFN clauses have exclusively functioned as an enabling device that allows investors to rely on more favorable provisions from a third-party investment treaty.
The common invocation of substantive provisions found in third-party investment treaties also contrasts with how MFN clauses operate in the trade regime. In this respect, the bilateral architecture of the investment treaty regime is an important factor. While investment treaties often include similar substantive guarantees for investors, these protections vary in the details, and sometimes even at a basic level. For example, the basic treaty may not include an umbrella clause, whereas the comparator treaty does. It is this variation that has opened the door to arguments that investors should benefit from the most favorable standards found in any of the host state's investment treaties. By and large, tribunals have been receptive to these arguments.
Unlike MFN clauses in the investment regime, the MFN obligation in the trade regime operates "within the relatively confined framework of the WTO covered agreements." 9 The comparison with the WTO shows that the ability to rely on more favorable provisions in third-party investment treaties deserves closer scrutiny-a key contribution of Batifort and Heath's article.
In particular, states do not rely on the MFN obligation to invoke better substantive and procedural guarantees in other trade agreements. 10 The MFN obligation in the GATT/WTO is the quintessential club good that benefits only members (cf. the reference to "any other country" in GATT Article I). The aim of the MFN obligation in the GATT/WTO is intimately linked to the WTO's multilateral architecture: it prevents GATT/WTO members from offering better competitive conditions to nonmembers. 11 The focus of the GATT's MFN clause is not on trade advantages by virtue of another trade-in-goods agreement, but on border measures, particularly tariffs, and nontariff barriers behind the border.
GATT Article 1 covers four categories of measures: customs duties and charges of any kind imposed on or in connection with importation and exportation; any method for calculating these; rules and formalities in connection with importation and exportation; and internal measures. Only if another agreement on trade in goods provides for "advantages, favours, privileges or immunities" in respect to these four types of measures is the measure within the scope of the GATT's MFN clause. By contrast, this MFN obligation does not cover a more extensive discipline that may be included in a third-party trade agreement. 12 In other words, the focus of the MFN clause in the GATT is not on advantages extended in treaties between a WTO member and a non-WTO member, but on advantages extended tout court.
This analysis suggests that more work on the similarities and differences of how MFN clauses operate across various areas of international law would be beneficial. Thus far, there is little research on how the operation of MFN clauses in one issue area (e.g., taxation) compares to another (e.g., investment). It would not be surprising if MFN clauses operated differently at least in some respects, given different regime objectives. At the risk of oversimplifying, these differing regime objectives are investment protection with investment treaties; liberalization under trade agreements; financial stability in the monetary sphere; and the avoidance of double taxation with tax treaties.

Conclusion
If future investment tribunals follow Batifort and Heath's proposal, MFN clauses' already limited role in the investment treaty regime will be further reduced. Even in the one area where MFN clauses have done real work in investment arbitrations over the last decade-the invocation of substantive and procedural provisions in thirdparty investment agreements-investors would likely find it much more difficult to invoke MFN clauses successfully. Some actors in the investment treaty regime-particularly states that are keen to put the genie of the MFN back into the bottle-are likely to welcome this development. Others will regard it as a regressive step for international investment law.  19 (2012). 10 But cf. Schill, supra note 3, at 925 ("MFN clauses in international trade law, including under the WTO, without doubt apply to better treatment extended by the granting state through international agreements."). 11 MAVROIDIS, supra note 7, at 206. 12 The MFN in the General Agreement on Trade in Services is likewise limited to "any measure covered by this Agreement."