In international investment law, some MFN claims mirror the use of MFN in international trade law. Other MFN claims are more expansive and controversial.
MFN clauses in trade agreements are typically invoked in connection with benefits, such as preferential tariffs, that have a concrete impact on the day-to-day flow of goods and services among economic actors. Investors sometimes invoke MFN clauses in investment treaties in a similar way: in the ICSID arbitration Parkerings v. Lithuania, for example, a Norwegian operator of parking garages complained that Lithuania had denied it permission to build its facility on a given site whereas it had authorized a Dutch investor in like circumstances to build on the same site. The notion that investors can invoke MFN clauses in connection with such concrete benefits resulting from state measures is not controversial, although it is not always easy to demonstrate that other investments in like circumstances in fact received better treatment.
But claimants in investment treaty arbitration have also invoked MFN clauses in an entirely different manner, arguing that MFN allows them to “import” or “incorporate” into the applicable treaty any provisions contained in other investment treaties entered into by the respondent state. Thus, investors have argued that MFN clauses allow them to disregard the dispute settlement provisions of the applicable treaty, disregard the provisions defining its scope of application (e.g. the definition of “investor”), ignore the absence of a given standard of treatment or any limitations placed on such a standard in the applicable treaty, compose what is effectively a “fantasy treaty” by mixing and matching any jurisdictional provisions and standards of treatment they like in the universe of treaties entered into by the respondent state , and then invoke that “fantasy treaty” to claim damages from the state in arbitration proceedings. This is not an isolated strategy—it has become a standard argument advanced by claimants in almost every investment treaty arbitration.
Unlike the first use of MFN clauses mentioned above, this second use is very controversial. The controversy has for a long time focused on the so-called “Maffezini issue” of whether claimants should be entitled to use MFN clauses to import jurisdictional provisions, as distinguished from substantive standards of treatment. In 2000, the tribunal in Maffezini v. Spain allowed the claimant to brush aside the dispute settlement provisions of the applicable investment treaty and rely instead on allegedly more favorable dispute settlement provisions contained in a third-party treaty. Since then, numerous tribunals have rendered conflicting decisions on this issue, and many commentators have voiced divergent opinions on the appropriateness of this use of MFN.
Meanwhile, the use of MFN clauses to import substantive standards of treatment has generally been treated as unremarkable. Many arbitral tribunals have allowed claimants to use MFN to import standards of treatment, such as fair and equitable treatment (FET), from one investment treaty to another. However, they have generally provided very little or no justification for this practice, and have appeared to rely on an assumption that all MFN clauses allow such importation. Moreover, very few scholars have analyzed this issue, and the commentaries that do touch upon it are generally limited to noting that the importation of substantive provisions is uncontroversial.
However, as my co-author and I argued in an article published in the American Journal of International Law in 2018, the conventional wisdom that all MFN clauses permit the importation of substantive standards of treatment is not justified.Footnote 1 We recalled that since MFN is a treaty-based standard and not a customary international law norm, it is necessary to verify in each case that the MFN clause in the applicable treaty permits this type of importation, in accordance with the normal rules of treaty interpretation. We also pointed out that MFN clauses in investment treaties come in a wide variety, and they rarely if ever provide expressly that investors are entitled to invoke standards of treatment contained in other investment treaties. Instead, their scope is usually defined through terms such as “treatment” and “in like circumstances.” Like national treatment provisions, the scope of such MFN clauses may well be limited to addressing actual state measures affecting concrete economic interests—i.e., the first use of MFN clauses discussed above—rather than importation of treaty provisions, whether substantive or jurisdictional. In support of our view, we also pointed to three sets of materials that have often been overlooked: (1) the first published arbitral award, rendered in 2016 in İçkale v. Turkmenistan, finding that an MFN clause could not be used to import any substantive provisions from other treaties; (2) the repeated opposition of the three NAFTA states to the claimants’ attempts in specific investment arbitrations to use MFN to import standards of treatment, as set forth in the states’ briefs and non-disputing party submissions; and (3) the wording of several new treaties, such as the EU-Canada CETA, that clarify “for greater certainty” that importation of substantive or procedural provisions is impermissible. Based on the foregoing, we argued that MFN importation of substantive standards can no longer be treated as uncontroversial, and we called for a fundamental reconsideration of this issue.
Our call for a new debate coincided with recent efforts by international organizations to alert states about the MFN importation issue. UNCTAD has characterized the MFN clause as “a crucial provision for IIA reform” and has laid out various options available to states to address it, including omitting the MFN clause from investment treaties altogether, or limiting its use to nationality-based discrimination resulting from state measures.Footnote 2 The OECD Secretariat has also placed the issue of “MFN shopping”—as it calls it—on its agenda, and organized a conference for states and other stakeholders in March 2018 to discuss the policy implications of this practice.Footnote 3 As I explained at that conference, it is easy to understand why states are reacting against “MFN shopping.” Apart from the significant cost of litigating this issue in case after case in investment treaty arbitration, the alleged benefits of MFN importation are dubious, and seem to boil down to the position that rules that favor investors are always better, all the time, and in all circumstances. Indeed, proponents of MFN importation usually argue that this practice is beneficial because it results in harmonization of investment protection. But in this context, “harmonization” is a euphemism for the maximization of investment protection, which few states (if any) view as the overriding objective of the international investment regime. That issue is particularly salient with respect to new treaties: why would the EU and Canada negotiate detailed dispute resolution provisions and calibrated standards of treatment in the CETA, and then allow investors to disregard this carefully crafted package by invoking MFN? Moreover, there is no reason to analogize the disparate standards of treatment in investment treaties to discriminatory tariffs in the trade context, as proponents of MFN importation sometimes do. While discriminatory tariffs can have serious ex ante effects on trade, the availability or absence of a given standard of treatment in an investment treaty—whose application to a particular case is known only ex post, after a dispute arises—is different in nature.
To conclude, the real question today regarding MFN clauses in investment treaties is whether states, tribunals, and scholars will engage in a fundamental reconsideration of the issue of MFN importation. I believe there are clear signs that we are going in that direction.
It seems as though some states are catching on that there is a problem here—with using MFN to bootstrap extraneous treaty protections into a BIT. However, states have seemed to differ in how far to go toward addressing it. Can you elaborate on some of the recent approaches to constraining MFN in recent marquee regional treaties?
There are very interesting differences in how the “tailored” MFN clause is drafted in the JEEPA and the CETA. Both the second sentence of CETA paragraph 4 of Article 8.7 (MFN) and JEEPA paragraph 5 of Article [x4 (MFN)] as of today (April 5, 2018)Footnote 1 are related to importing substantive protections from the other IA to the CETA or the JEEPA, and both clauses articulate conditions on that importation as well. That condition of the CETA MFN is considered stricter than that of the JEEPA MFN. The reason is as follows.
What is remarkable on the CETA MFN is the last line, which reads “absent measures adopted or maintained by a Party pursuant to those obligations,” which holds that substantive protections from the other IA can be imported in so far as the actual measures to the investments and/or investors of non-party (who are the investors of the party under the other IA) are adopted or maintained, in other words, the actual measures and actual investments and/or investors under the other IAs must exist when importing substantive protections from the other IAs to the CETA.
On the other hand, when it comes to the JEEPA MFN, what is remarkable is the footnote 4, which reads: “For greater certainty, the entrepreneurs (if in other IAs, ‘investors’)  or covered enterprises (if in the other IAs, ‘covered investments’) would be entitled to receive such treatment (in other words, substantive protections of the other IAs), even in the absence of enterprises (investments) when the comparison is made,” which holds that substantive protections from the other IA can be imported even though the actual measures and actual investments and/or investors under the other IAs do not exist when importing substantive protections from the other IAs to the JEEPA.
I would point to a couple of other notable examples in recent practice, starting with India's approach to MFN clauses in investment treaties. In the White Industries arbitration, India was forced to pay damages based on a controversial interpretation of a standard of treatment—“effective means”—that did not appear in the applicable treaty, but which the tribunal had allowed the claimant to import through MFN. India subsequently omitted MFN from its revised model investment treaty. India also addressed MFN clauses in past treaties by proposing a draft interpretive statement to its treaty partners, which in some cases led to the conclusion of joint interpretive statements. India's draft interpretive statement clarifies that MFN clauses cannot be used to import either procedural or substantive provisions, and that such clauses are limited to addressing state measures that discriminate among investments in like circumstances.
Another interesting example is the MFN clause in the CPTPP. It clarifies that MFN cannot be used to import more favorable dispute settlement provisions, but does not expressly address the issue of importation of substantive standards of treatment. However, this does not necessarily mean that it allows such importation. Indeed, the MFN clause in the CPTPP is almost identical to the MFN clauses in NAFTA and in the recent CETA, and the states party to those treaties—which include states that are also party to the CPTPP (e.g. Canada)—often have firmly opposed importation of standards of treatment via MFN. It is likely that at least some of those states (and probably others) do not view the MFN clause in the CPTPP as allowing any form of “MFN shopping,” regardless of the lack of an express clarification.Footnote 1
What do you think accounts for the different evolutions of the non-discrimination doctrine in trade and investment law? Is it a matter of mentality? Or do you find that fundamental differences in the trade and investment regimes may account for the different approaches/problems cropping up with non-discrimination across these fields?
The member states of the WTO established the international trade regime to advance their collective, macroeconomic welfare through the elimination of both tariff and non-tariff barriers to market access for goods and services. Trade disputes are inherently political—governments decide which ones to bring after weighing the interests of competing domestic constituencies, as well as when to come into compliance with an adverse decision from the WTO. Accordingly, elected officials from the member states are frequently engaged in the WTO's dispute resolution process at every phase of the proceedings.
In contrast, governments created the international investment regime to address microeconomic problems experienced by individual investors seeking to mitigate risk when they invest abroad, as well as to signal that their economies are hospitable to foreign investment. Governments designed the regime with a view to depoliticizing investment claims and eliminating the need for their espousal. For this reason, governments empowered ISDS tribunals with the authority to issue only certain kinds of remedies and retained for themselves the right to refuse to modify measures deemed non-compliant with their commitments.
For these reasons, governments may have contemplated greater scrutiny of domestic measures by the WTO than of comparable measures by ISDS arbitration panels simply because they envisioned additional political checks on the WTO dispute-resolution process. Notwithstanding this fact, persistent criticism of the level of deference accorded to governments when adopting domestic public welfare measures in both dispute resolution mechanisms remains.