The scope of Most Favoured Nation (MFN) clauses in bilateral investment treaties (BITs) has been a source of rich debate for many years. In sum, the debate centres around whether MFN “treatment” includes only substantive rules for the protection of investments, or if it also extends to procedural protections such as dispute resolution. There have been conflicting decisions over the years, but the topic is particularly newsworthy at the moment following French academic and arbitrator Brigitte Stern using her dissenting opinion in an ICSID award in Impregilo SpA v Argentina (ICSID Case No ARB/07/17 – June 21, 2011) to warn of the “great dangers” of allowing claimants to bypass a BIT’s jurisdictional requirements by invoking a MFN clause. This blog considers both sides of the debate in relation to the scope of MFN clauses.

There is no pro forma for a MFN clause, although most link BITs by ensuring that parties to one BIT provide treatment no less favourable than the treatment they provide investors of any third nationality, which includes treatment guaranteed under BITs with other states. However, most BITs are very general in their wording and leave considerable scope to argue competing interpretations. In particular, most BITs are silent on whether MFN “treatment” includes only substantive rules for the protection of investments or if it also extends to procedural protections, like dispute resolution. The BIT in Impregilo (the Argentina-Italy BIT) used similar generic language and, importantly, provided that it extended to not only “investments”, but “all other matters regulated by this Agreement”.

One side of the argument (which was the view of the majority in Impregilo – Judge Danelius and Judge Bower) is that MFN clauses should be interpreted broadly. The underlying rationale for this side of the argument is that the term “treatment” in MFN clauses is in itself wide enough to be applicable to procedural matters such as dispute settlement. Furthermore, dispute resolution provisions are essential to the protections envisaged under BITs and therefore the MFN clause should be interpreted as giving the claimant the benefit of the wider dispute resolution clause in another BIT. Moreover, as regards “all matters” MFN clauses (such as the MFN clause in Impregilo), such a phrase should not be read narrowly as referring to “all similar matters” or “all other matters of the same kind” such that it excludes procedural matters.

There is a significant volume of case law to support this position. The leading decision is Maffezini v Spain (ICSID Case No. ARB/97/7 – January 25, 2000), although other examples include: Gas Natural v. Argentina (ICSID Case No. ARB/03/10 – June 17, 2005); Suez and Vivendi v. Argentina (ICSID Case No. ARB/03/19 – August 3, 2006); and Camuzzi International v. Argentina, ICSID Case No. ARB/03/2 – May 11, 2005). It should, however, be noted that the above cases all concerned “all matters” MFN clauses. Nonetheless, there have also been decisions where less comprehensive MFN clauses which only provided for MFN treatment of investors and investments were found to be sufficient to cover dispute settlement. Examples of such cases are: Siemens v. Argentina (ICSID Case No. ARB/02/8 – August 3, 2004); National Grid v. Argentina (UNCITRAL – June 20, 2006); and RosInvest v. Russian Federation (SCC Case No. V079/2005 – October 2007).

The other side of the argument is that MFN clauses relate to the substantial protections afforded to investors and investments and that, therefore, their reach should not extend to procedural issues such as dispute resolution. In particular, procedural issues affect how the protections in the relevant BIT operate and are enforced – that is fundamentally different to ensuring investors receive most favourable treatment.

There are decisions to support this side of the argument, examples of which include: Salini v. Jordan (ICSID Case No. ARB/02/13 – November 9, 2004); Plama v. Bulgaria (ICSID Case No. ARB/03/24) – February 8, 2005); and Telenor v Hungary (ICSID Case No. ARB/04/15 – September 13, 2006). However, it should be noted that none of the above decisions relate to “all matters” MFN clauses. Indeed, there is only one decision to support the position that an “all matters” MFN clause should not extend to dispute settlement (Berschader v. Russian Federation (SCC Case No. 080/2004 – April 21, 2006)), and even then one of the arbitrators did strongly dissent.

In Impregilo, Professor Stern introduced her dissent by stating that she hoped it “will contribute in a modest and constructive manner to the ongoing debate on the way MFN clauses should be applied”. Firstly, Professor Stern took issue with the majority’s portrayal of the case law as weighted in favour of their position. She argued that if one looks at the number of arbitrators who are in favour of applying MFN clauses to dispute resolution rather than at the number of awards then the picture looks almost balanced because many of the same arbitrators have been involved in the same cases. In any event, she stated that it is not a legally convincing argument to rely on former cases as if they were binding precedents.

Professor Stern compared the dispute resolution clauses in the Argentina-Italy BIT and the Argentina-US BIT (which was the dispute resolution clause Impregilo sought to import). While the Argentina-Italy BIT requires the investor to exhaust local remedies over 18 months, the Argentina-US BIT provides that the investor may choose to submit the dispute for resolution to the domestic courts or administrative tribunals, or deal with it in accordance with previously agreed dispute settlement procedures, or, after six months from the date on which the dispute arose, to submit it to international arbitration. In such circumstances, importing just a time limit from one mechanism into the other does not really make any sense as it cannot be based on a serious comparison between two clauses with completely different underlying rationales. She concluded that Impregilo had been granted an inexistent favourable treatment that did not correspond to any real situation under any treaty. Indeed, she stated that such an interpretation effectively allowed the claimant to de-structure jurisdictional requirements and pick and choose from a menu of treaty options.

Professor Stern’s dissent is one of the most thorough articulations to date of the case against extending MFN treatment to dispute resolution. In particular, it highlights one of the central issues with this debate – namely, rights on the one hand, and conditions for access to the rights on the other. In Professor Stern’s view, a MFN clause can only concern the rights that an investor can enjoy – it cannot modify the fundamental conditions for the enjoyment of such rights. In particular, a conditional right to ICSID cannot “magically” be transformed into an unconditional right by the grace of a MFN clause. Indeed, as Professor Stern notes, if the majority position is taken to its logical conclusion, applying the scope of a MFN clause to dispute resolution provisions would theoretically allow the importation of an ICSID clause into a treaty that does not provide at all for international arbitration – this could be of particular concern to the Australian Government that announced earlier this year that it will no longer pursue investor-state arbitration provisions in future international economic agreements with developing countries.

Overall, while most tribunals that have considered this issue have ultimately based their conclusions upon the precise wording and scope of the MFN clauses before them, there can be no doubt that there appears to be a creep towards tribunals endorsing the view that MFN can – and should – extend to dispute resolution (although the issue of whether there is precedent in investment treaty arbitration and whether tribunal can – or should – refer to previous decisions is a matter for another blog). Nonetheless, as Professor Stern’s dissent highlights, this is one area of investment treaty application/interpretation where there is no sign that practitioners or tribunals will reach a consensus any time soon.

Mike McClure
Herbert Smith Moscow


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Profile Navigator and Relationship Indicator
Includes 7,300+ profiles of arbitrators, expert witnesses, counsels & 13,500+ relationships to uncover potential conflicts of interest.

Learn how Kluwer Arbitration can support you.

Kluwer Arbitration
This page as PDF


  1. Dear Mike

    You have given a very nice summary of the both sides of the arguments re: MFN.

    I tend to agree with the opinion of Professor Stern. She has cuationed the investment arbitration community at a very right time. Perhaps it is the right time to have fresh look at the aims, objects and purpose of MFN in the context of Invetsment treaty and investment arbitration.
    For example should MFN is used to pick and choose the favourable things from different BITs relating to the states out of context. Isn’t it strnage that onlu number of days could be picked up from other treaty whilst that number of days was negotiated by two different states in a different context.

    I hope that Professor Stern’s dissenting opinion will sooner become that majority opinion among the investment arbitration community.

    Rajesh Sharma
    Assistant Professor
    School of Law
    City University of Hong Kong

  2. I would concur with the opinion presented by Mr. Sharma. Nevertheless, in my humble opinion we shall draw the attention to distinction – rights or conditions. I realy admire the passage in which Prof. Stern is pointing out that MFN clause grants rights to an investor, not conditions.

    However, still we may find cases in which right to arbitration granted to an investor would be so limited that using MFN wouldn’t mean a detour of conditions. Thus, I think that decision in Maffezini’s case was right one.

    Wojciech Bazan
    student, Jagiellonian University in Cracow

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.