Jump to Content Jump to Main Navigation
Max Planck Encyclopedia of Public International Law [MPEPIL]

Most-Favoured-Nation Clause

Robin Geiβ, Meinhard Hilf

From: Oxford Public International Law (http://opil.ouplaw.com). (c) Oxford University Press, 2015. All Rights Reserved.date: 20 June 2019

Subject(s):
Most-favoured-nation treatment (MFN) — Taxation — Developing countries — BITs (Bilateral Investment Treaties) — Peace treaties — Treaties, effect for third states — Vienna Convention on the Law of Treaties — Ejusdem generis rule

Published under the auspices of the Max Planck Foundation for International Peace and the Rule of Law under the direction of Rüdiger Wolfrum.

A.  Notion

1.  Definition

A most-favoured-nation clause (commonly abbreviated as ‘MFN’) is a treaty provision whereby a State (the granting State) undertakes an obligation towards another State (the beneficiary State) to accord most-favoured-nation treatment in an agreed sphere of relations. Most-favoured-nation treatment is treatment accorded by the granting State to the beneficiary State, or to persons or things in a determined relationship with that State not less favourable than treatment extended by the granting State to a third State, or to persons or things in the same relationship with that third State. Establishing the juridical meaning of the most-favoured-nation clause by way of reference to most-favoured-nation treatment takes into account that while State practice has led to the evolution of a sufficiently defined most-favoured-nation standard, the clauses containing this standard may be drafted in the most diverse ways. Indeed, it has rightly been pointed out that ‘although it is customary to speak of the most-favoured-nation clause, there are many forms of the clause, so that any attempt to generalize upon the meaning and effect of such clauses must be made, and accepted, with caution’ (MacNair 273). The International Law Commission (ILC) heeded this warning in its elaboration of 30 draft articles on the topic in 1978. The definition provided in Arts 4 and 5 (ILC Draft Articles on Most-Favoured-Nation Clauses, Report of the ILC on the Work of its Thirtieth Session 33, 41) is adhered to in the present article. Consequently, a clause constitutes a most-favoured-nation clause if it accords most-favoured-nation treatment. Whether or not this is the case is a question of interpretation in accordance with the general rules of the law of treaties.

2.  Function and Effects

The clause’s history of over 800 years evidences that the most-favoured-nation clause serves fundamental State interests. The clause promotes the equality of States by diminishing differential treatment (States, Sovereign Equality). Indeed, in the words of the International Court of Justice (ICJ) it is the very intention of the clause ‘to establish and to maintain at all times fundamental equality without discrimination among all of the countries concerned’ (United States Nationals in Morocco Case [Judgment of 27 August 1952] 192). The most-favoured-nation clause insures the beneficiary against discrimination and establishes equality of opportunity on the highest possible plane: the minimum of discrimination and the maximum of favours conceded to any third State (Schwarzenberger [1945] 99).

The principle of non-discrimination does not imply the generalization of most-favoured-nation treatment (States, Equal Treatment and Non-Discrimination). Under this principle the relevant benchmark is the general standard of treatment of other States whereas it is the effect of a most-favoured-nation clause to establish the most favourable treatment as the relevant tertium comparationis. The principle of non-discrimination thus only secures a State general non-discriminatory treatment. At the same time, the principle does not prohibit a State from granting treatment more favourable than the general standard to particular States. It follows that a State cannot invoke the principle of non-discrimination in order to be put on a par with a State that is placed in a position more favourable than the general treatment. This latter claim can only be advanced on the basis of an explicit commitment of the State granting the favours, namely in the form of a most-favoured-nation clause.

The most-favoured-nation clause, in addition to the abovementioned levelling effect, also has a harmonizing effect. Tensions in international relations, as well as the likelihood of trade disputes, are significantly reduced. Due to the clause’s automatism, ie the fact that favours granted to a third State automatically apply to the beneficiary State, countries within a most-favoured-nation arrangement can always be content to enjoy treatment equal to that of the most-favoured country. Anybody’s advantage accrues to everybody’s profit (Schwarzenberger [1945] 99). Moreover, the clause’s automatism renders the conclusion of new individual agreements superfluous and thus creates a tendency to multilateralism (Unilateralism/Multilateralism).

In the realm of world trade, most-favoured-nation clauses have an overall liberalizing effect by automatically harmonizing trade relations between States and levelling trade barriers in the abovementioned sense. As a consequence most-favoured-nation clauses reduce obstacles to free trade and create favourable conditions for the development of mutual commercial relations between States. Economically, the most-favoured-nation clause facilitates the exports of those goods in the production of which a country enjoys a comparative advantage, and promotes imports of those goods in which a country suffers a comparative disadvantage.

3.  Fields of Application

The most-favoured-nation clause is used in various areas of international law, most importantly, in the realm of international trade law. It is this area from which the clause originates, and in which it is most regularly employed. However, States are free to grant each other most-favoured-nation treatment in all fields which lend themselves to such an agreement. Hence, most-favoured-nation clauses can be found in a rather varied spectrum of areas, in bilateral as well as in multilateral treaties. Traditionally, they have often featured in bilateral Treaties of Friendship, Commerce and Navigation. An example is the Treaty of Peace and Commerce between Great Britain and Sweden ([signed 11 April 1654] 1 BSP 691) which provided: ‘The people, subjects and inhabitants of both confederates shall have, and enjoy in each other’s kingdoms, countries, lands and dominions, as large and ample privileges, relations, liberties and immunities, as any other foreigner at present doth and hereafter shall enjoy’. Most-favoured-nation clauses were also often included in bilateral treaties regarding diplomatic and consular relations, for example in the Consular Convention between the Kingdom of Italy and the Turkish Republic ([signed 9 September 1929, entered into force 13 April 1932] 129 LNTS 195; Consular Treaties). In this particular area, the clause commonly provided guarantees with respect to the maintenance of diplomatic and consular premises, and with respect to the privileges granted to diplomatic and consular personnel. However, today, multilateral conventions on diplomatic and consular relations prescribe a catalogue of rights that are directly applicable, and there is thus no longer a need to prevent discrimination in this particular area by virtue of MFN clauses.

The General Agreement on Tariffs and Trade (1947 and 1994) (‘GATT’), the General Agreement on Trade in Services (1994) (‘GATS’), and the Agreement on Trade-Related Aspects of Intellectual Property Rights (1994) (‘TRIPS’) are the most prominent examples of multilateral treaties containing a most-favoured-nation clause in the trade area. However, the Convention relating to the Status of Refugees ([signed 28 July 1951, entered into force 22 April 1954] 189 UNTS 150) and the Convention relating to the Status of Stateless Persons ([done 28 September 1954, entered into force 6 June 1960] 360 UNTS 130) evidence that the clause is employed in a far wider spectrum (Aliens; Refugees; Stateless Persons). More recently, Bilateral Investment Treaties (‘BITs’) containing a most-favoured-nation clause have seen a rather vast proliferation (Investments, Bilateral Treaties).

On the basis of the list elaborated by the Special Rapporteur of the ILC a more general, albeit non-exhaustive overview of areas in which most-favoured-nation clauses are employed can be given as follows:

  1. a)  International regulation of trade and payments, eg exports, imports, customs tariffs (World Trade, Principles)

  2. b)  International investment protection (Investments, International Protection).

  3. c)  Transport in general and treatment of foreign means of transport, in particular ships, aircraft, trains, motor vehicles (Traffic and Transport, International Regulation).

  4. d)  Establishment of foreign physical and juridical persons, their personal rights and obligations.

  5. e)  Establishment of diplomatic, consular and other missions, their privileges and immunities, and treatment in general.

  6. f)  International taxation (Taxation, International).

  7. g)  Protection of intellectual and industrial property (Industrial Property, International Protection).

  8. h)  Recognition of foreign legislative and administrative acts.

  9. i)  Administration of justice, eg access to courts and tribunals (Recognition and Enforcement of Foreign Arbitral Awards; Recognition and Enforcement of Foreign Judgments).

Clauses relating to most-favoured-treatment are also applicable to subjects of law other than States. For example, the clause entailed in Art. VIII (4) Constitution of the Food and Agriculture Organization of the United Nations (FAO) ([adopted 16 October 1945, entered into force 16 October 1945] 145 BSP 910) could be referred to as a ‘most-favoured international organization’ clause. Art. VIII obligates members of the FAO, inter alia, to accord to staff members of the FAO other than the Director-General and the senior staff the immunities and facilities which they accord to equivalent staff members of other public international organizations. The purpose of such a provision is to establish and benefit from a general scheme to be accorded to the staff of all international organizations facing similar difficulties while discharging their duties.

B.  Historical Development

10  The evolution of the most-favoured-nation clause shows how altering versions of the clause have served different economic and political purposes at different times. Nevertheless, the very length of the clause’s history reflects that its egalitarian function constitutes a common denominator on which States of the most diverse political, social, or economic structures can settle.

1.  Early History

11  Embryonic versions of the most-favoured-nation clause have been traced as far back as the Middle Ages. At that time, the merchants of Italian, French, and Spanish trading cities employed unilateral most-favoured-nation clauses to secure themselves a monopoly in the exploitation of foreign markets. When, with the development of commerce, their efforts to exclude their competitors from these markets failed, they aimed to secure trading opportunities at least equal to those of their rivals. Reciprocal versions of the clause did not appear before the 15th century and, coinciding with the gradual growth of world commerce, only became common in the 17th century when the clause was discovered as a convenient shorthand to incorporate by way of reference advantages previously granted in other treaties (Jackson 250; Reciprocity). With the gradual decline of mercantilism the trend towards a wider use of reciprocal most-favoured-nation clauses continued.

12  However, throughout this time unilateral most-favoured-nation clauses remained a constant feature of peace treaties and capitulatory regimes, eg in the capitulations elicited by European rulers from the Ottoman Empire, from South Asian powers and from China. In the context of peace treaties the clause served as a tool for the victorious power to secure its participation in any new concessions other victorious powers would extort from the granting State in future conflicts. It was not until the French-Turkish Peace Treaty of 1802 that a Western Power concluded a treaty with the Turkish Emperor which contained a reciprocity clause (Turkish Act of Accession to the Definitive Treaty of Peace between France, Great Britain, Spain, and the Batavian Republic [signed 27 March 1802] (1801–03) 56 CTS 299).

2.  From the 18th Century until World War I: The Conditional and the Unconditional Most-Favoured-Nation Clause

13  The phrase ‘most-favoured nation’ found its way into the commercial treaties of the 18th century when political and commercial treaties became more clearly differentiated. Until the late 18th century the clause was exclusively drafted in an unconditional manner, ie the right to most-favoured-nation treatment would inure automatically as soon as a third State was more favourably treated than the beneficiary State with regard to the subject-matter comprised by the clause.

14  However, the economic policies of States did not always match with the equalizing dynamic entailed in this automatism. With the Treaty of Amity and Commerce between the United States and France ([done 6 February 1778] 8 Statutes at Large 12 [1848]) a new variant of the clause emerged, the conditional most-favoured-nation clause (Conditionality). This type of clause lacks the automatism entailed in the unconditional version as the most-favoured-nation treatment is only extended on the condition that the beneficiary State agrees to grant compensation equivalent to that given by the third State. Technically, a conditional most-favoured-nation clause thus constitutes a pactum de contrahendo. The beneficiary and the granting State merely undertake to enter into negotiations to grant each other certain advantages similar or correlative to those previously granted to third countries. The unconditional type on the other hand assumes that compensation has been given once and for all, commonly in the form of a reciprocal most-favoured-nation pledge, in the original treaty in which the most-favoured-nation clause is contained. The conditional most-favoured-nation clause particularly served the economic purposes of the United States which aimed to protect its growing industry. Indeed, the wording of Art. II of the US-French Treaty of 1778, according to which the respective beneficiary should enjoy the favours granted to other nations ‘freely, if the Concession was freely made, or on allowing the same Compensation, if the Concession was Conditional’, became the model for practically all commercial treaties of the United States until 1923. After the Napoleonic period, in the years 1830–60, the conditional version of the clause was transiently prevalent also in Europe.

15  However, the so-called Chevalier-Cobden Treaty (Treaty of Commerce between Great Britain and France [signed 23 January 1860, entered into force 4 February 1860] 50 BSP 13), a bilateral treaty of commerce in which Great Britain and France granted each other unconditionally the status of a most-favoured nation, virtually marked the end of the use of the conditional type in Europe. Henceforth a wave of liberal economic sentiment, transitionally interrupted merely by the depression following the Franco–Prussian War, elevated the unconditional most-favoured-nation clause to become the almost universal basis of a vast system of commercial treaties (Snyder 239). This ascendancy continued up until the World War I, and it is from this time that the famous designation of the clause as ‘the corner-stone of all modern commercial treaties’ originates (Hornbeck 395).

3.  The Modern Version of the Clause: Attempts at Codification

16  The unilateral version of the most-favoured-nation clause once more made its reappearance on the international stage in the peace treaties concluding World War I, namely in Art. 267 Versailles Peace Treaty (1919) (Treaty of Peace between the Allied and Associated Powers and Germany (1919) 225 CTS 188); in Art. 220 St Germain Peace Treaty (1919) (Treaty of Peace between the Allied and Associated Powers and Austria 112 BSP 317); in Art. 150 Neuilly Peace Treaty (1919) (Treaty of Peace between the Allied and Associated Powers and Bulgaria (1919) 226 CTS 332), and in Art. 203 Trianon Peace Treaty (1920) (Treaty of Peace between the Allied and Associated Powers and Hungary 6 LNTS 187). In the case of Germany, the victorious Allies were unilaterally granted unconditional most-favoured-nation treatment for five years, and in the case of Austria, Bulgaria, and Hungary for three years.

17  Subsequently, in the immediate aftermath of World War I, the destructed European economy as well as the ensuing depression of 1921–24 necessitated the adoption of a more restrictive economic policy and led to a temporary decline in the use of the most-favoured-nation clause. However, with gradual economic recovery and increasing stabilization, the clause again became a common feature of commercial treaties. Moreover the United States, in light of its altered position in world economy, realigned its economic policy and, aiming at the penetration of international markets, abandoned the use of the conditional version of the most-favoured-nation clause in 1923. The conditional version of the clause henceforth fell into disuse, whereas the unconditional most-favoured-nation clause was a device many believed could help to solve the imminent problems of the time, namely political tensions resulting from trade disputes. Indeed, this view had already been endorsed by President Wilson who in the third of his ‘Fourteen Points’ had argued in favour of the removal of trade barriers and the establishment of equality in trade conditions (Address of the President of the United States Delivered at a Joint Session of the Two Houses of Congress [8 January 1918] in United States Department of State [ed], Papers relating to the Foreign Relations of the United States 1918 Supp 1 The World War [United States Government Printing Office Washington 1933] 12; Fourteen Points of Wilson [1918]).

18  The widely shared perception of the clause’s importance triggered the in-depth study of the rules pertaining to the most-favoured-nation clause as it appeared in commercial treaties and prompted first attempts at codification (Codification and Progressive Development of International Law). However, the Committee of Experts for the Progressive Codification of International Law of the League of Nations in 1927 arrived at the conclusion that international regulation of rules relating to the most-favoured-nation clause by way of a general convention, even if desirable, would encounter serious obstacles. The study of the subject continued throughout the inter-war period, particularly under the auspices of the Economic Committee of the League of Nations and the international economic conferences. In 1936 the Institut de Droit international, intending to promote an unofficial codification of the topic, adopted a resolution on the effects of the most-favoured-nation clause in matters of commerce and navigation proposing certain principles for the legal construction of most-favoured-nation clauses, for example that most-favoured-nation clauses should have an unconditional character unless provided otherwise.

19  Even though no formal agreement could be reached at the time, the unconditional most-favoured-nation clause was widely endorsed and in 1947 was incorporated into Art. I GATT which provides that:

With respect to customs duties and charges of any kind imposed on or in connection with importation or exportation … any advantage, favour, privilege or immunity granted by any contracting party to any product … shall be accorded immediately and unconditionally to the like product … of all other contracting parties.

The incorporation into Art. I GATT aimed to underline the importance of the most-favoured-nation principle which has rightly been described as ‘a cornerstone of the GATT and … one of the pillars of the world trading system’ (WTO Canada—Certain Measures Affecting the Automotive Industry para. 69).

20  The desirability of conducting international trade on the basis of unconditional most-favoured-nation treatment has been pointed out ever since, eg in Art. 8 General Principles adopted by the first United Nations Conference on Trade and Development (UNCTAD) (UNCTAD ‘General and Special Principles’ in ‘Final Act’ [15 June 1964] UN Doc E/CONF.46/141 vol 1, 18), Art. 26 Charter of Economic Rights and Duties of States (1974) (UNGA Res 3281 [XXIX] [12 December 1974] GAOR 29th Session Supp 31 vol 1, 50), in the Helsinki Final Act (1975) (Conference for Security and Co-operation in Europe ‘Final Act’ [adopted 1 August 1975] (1975) 14 ILM 1292), in Art. II GATS in 1994, and in Art. 1 TRIPS in 1994. Today, despite lately being challenged by the proliferation of free trade areas and customs unions, the overwhelming majority of international trade is conducted on the basis of unconditional most-favoured-nation clauses. In 1969 the Institut de Droit international issued a brief resolution on the most-favoured-nation clause in multilateral treaties and codification efforts gained new momentum when, in 1978, the ILC submitted 30 draft articles on most-favoured-nation clauses to the UN General Assembly (‘UNGA’) in order to serve as a basis for the conclusion of a convention on most-favoured-nation clauses. It was not before 9 December 1991 that the UNGA decided to bring these draft articles to the attention of Member States and intergovernmental organizations ‘for their consideration in such cases and to such extent as they deem appropriate’ (UNGA ‘Provisional Verbatim Record of the 67th Meeting’ [23 December 1991] UN Doc A/46/PV.67; UNGA ‘Consideration of the Draft Articles on Most-Favoured-Nation Clauses: Report of the Sixth Committee’ [15 November 1991] UN Doc A/46/655). To this day, the draft articles have not found their way into a formal convention. However, in 2007, following the establishment of a new working group, the ILC decided to refer the topic of most-favoured-nation clauses in the field of investment agreements to the Planning Group in order to include the topic in the ILC’s long-term programme of work.

C.  Current Legal Situation

1.  General Rules of Application

21  Most-favoured-nation treatment is not a rule of customary international law nor can a right to such treatment be derived from the general principle of non-discrimination. Rather, most-favoured-nation treatment is an optional standard which States are free to either adopt or reject (Schwarzenberger [1948] 408). Still, in the absence of exceptional circumstances it has been argued that it constitutes an unfriendly act if a State refuses, without serious reasons, another State’s offer to base their reciprocal trade on a most-favoured-nation commitment (see E Ustor, ‘Most-Favoured-Nation Clause’ in R Bernhardt [ed], Encyclopedia of Public International Law vol 3 [Elsevier Amsterdam 1997] 468–73, 470). Beyond, it could potentially be argued that the principle of most-favoured-nation treatment has consolidated into a rule of customary international law. However, it seems more than doubtful whether State practice indeed supports such a point of view. Rather, in view of States’ general freedom to enter into treaties and to negotiate their content as well as in view of the various exceptions from MFN treatment in international treaties, it would seem that there is neither a customary law obligation to accord most-favoured-nation treatment nor could it be considered an unfriendly act if a State refused to do so.

22  As a treaty provision the most-favoured-nation clause is subject to the general rules of treaty interpretation. Historically, interpretational problems have often pertained to the conditionality of a clause (eg Treaty between the United States of America and the French Republic [done 30 April 1803] 8 Statutes at Large 200 [1848], the so-called ‘Louisiana Purchase’). Today, the prevalence of the unconditional version generally militates in favour of an unconditional clause.

23  Unilateral most-favoured-nation clauses, historically a common feature of capitulatory regimes and peace treaties, have largely fallen into disuse. They are resorted to only if exceptional circumstances so demand, eg when most-favoured-nation treatment is accorded to the ships of a land-locked State which is not in a position to offer equal treatment in return.

24  A State Party to a treaty including a most-favoured-nation clause may be a granting State and a beneficiary State at the same time. This is the case if, by the same clause, a State has undertaken to accord to another State most-favoured-nation treatment and that other State has undertaken to accord it the same treatment. Today, most-favoured-nation clauses are commonly drafted in such a synallagmatic fashion. Moreover, they are often combined with a national treatment clause (National Treatment, Principle). While the most-favoured-nation clause, as a renvoi to another treaty prohibits discrimination in relation to third countries, a national treatment clause, as a renvoi to municipal law, prohibits discrimination in relation to the contracting parties’ own nationals and products.

(a)  Source of the Right to Most-Favoured-Nation Treatment

25  The right of the beneficiary State to receive from the granting State most-favoured-nation treatment is anchored in the most-favoured-nation clause. The direct beneficiary is the beneficiary State whereas the persons or things in a particular relationship with that State enjoy the treatment stipulated by the granting State only indirectly through the beneficiary State. The treaty containing the clause is the basic treaty (acte règle) from which the right to most-favoured-nation treatment is derived (Anglo-Iranian Oil Company Case [Judgment of 22 July 1952] 110). The third-party treaty merely constitutes the acte condition and remains a res inter alios acta for the beneficiary. The most-favoured-nation clause thus does not constitute an exception to the rule that treaties produce effects only as between the contracting parties (pacta tertiis nec nocent nec prosunt; see Art. 34 Vienna Convention on the Law of Treaties [1969] 1155 UNTS 331; Treaties, Third-Party Effect).

(b)  The Clause’s Scope Ratione Temporis

26  The initiation as well as the termination of the clause’s effects hinge on the acte condition. The clause begins to operate, ie the right to more favourable treatment inures automatically as soon as the third State is more favourably treated than the beneficiary (United States Nationals in Morocco Case [Judgment of 27 August 1952] 187). The mere fact of more favourable treatment—be it based upon a treaty, a unilateral, legislative, or other act—is sufficient to set in motion the operation of the clause. In the most common case that such treatment is extended by virtue of a treaty the right of the beneficiary accrues with the entitlement of the third State, irrespective of whether the latter avails itself of the benefits due to it or not. However, in the absence of any such undertaking vis-à-vis a third State no such right arises and in this case the clause remains but an ‘empty shell’ (Schwarzenberger [1945] 96). In multilateral treaties the relations created by a most-favoured-nation clause are essentially bilateral, ie every party to the treaty may demand from any other party to accord to it treatment equal to that extended to any third State, irrespective of whether that third State is a party to the treaty or not.

27  The clause ceases to operate, ie the right of the beneficiary State to most-favoured-nation treatment is terminated or suspended at the moment when the extension of the relevant treatment by the granting State to the third State is terminated or suspended (United States Nationals in Morocco Case [Judgment of 27 August 1952] 191). The beneficiary State by virtue of the most-favoured-nation clause has no right to claim the perpetuation of the favours extended to the third State, nor can it object to the termination of the third-party treaty (Jaenicke 498). Moreover, as a general rule the rights derived form a most-favoured-nation clause may also be brought to an end by virtue of the termination or suspension of the treaty containing the clause on the basis of the provisions laid out in Sec. 3 Vienna Convention on the Law of Treaties.

(c)  The Clause’s Scope Ratione Materiae

28  The extent of the rights to which the beneficiary of a most-favoured-nation clause may lay claim is confined by two factors. According to the ejusdem generis rule, no other rights can be claimed than those falling within the limits of the subject-matter of the clause. Secondly, the extent of the favours due to the beneficiary is determined by the actual benefits extended by the granting State to the third State. The beneficiary is thus entitled to the same treatment as extended by the granting State to a third State in respect of persons and things which belong to the same category and have the same relationship to the beneficiary State as those persons and things related to the third State.

(d)  Conditions and Exceptions

29  Unconditional reciprocally agreed most-favoured-nation clauses can be risky. They ensure equal rights but not necessarily an equality of material advantages. Over time the aspired balancing of the reciprocal trade interests could become considerably distorted. The employment of unconditional clauses thus carries a significant degree of uncertainty for the contracting parties. State practice, in response, has generated an abundant array of safeguards, notably the combination of the unconditional clause with certain restrictions, exceptions or ‘conditions’, eg a time-limit, in order to countermand the detrimental effects of a purely unconditional clause. In exercise of their contractual freedom States are free to shape and condition most-favoured-nation clauses in accordance with their needs. The above-mentioned conditional most-favoured-nation clause has fallen into disuse because it no longer serves any such needs. The condition that favours granted to a third State in return for a compensation accrue to the beneficiary only in consideration of a concession equivalent to that given by the third State complicated commercial relations, instead of simplifying them. In order to forestall misconceptions it bears mentioning that only this particular condition would qualify a clause as a conditional most-favoured-nation clause in the historic sense. Outside this category the ‘unconditional’ clause may be, and in practice commonly is, made subject to conditions or certain exceptions.

30  According to Art. IX (3) Marrakesh Agreement Establishing the World Trade Organization ([adopted 15 April 1994, entered into force 1 January 1995] 1867 UNTS 154) any obligation—including the most-favoured-nation obligation—imposed on a member by the establishing agreement itself or any of its incorporated agreements can under exceptional circumstances be subject to a waiver to be granted by the ministerial conference of the World Trade Organization (WTO). Aside this general exception, the GATT entails a mosaic of other exceptions, exemplary of the objectives and interests States regularly perceive as requiring a safeguard within a most-favoured-nation arrangement. For example, according to Art. XIV GATT the contracting parties may set aside their most-favoured-nation obligation to safeguard their balance of payments or waive the obligation under exceptional circumstances in accordance with Art. XXV (5) GATT. Art. XXIV (3) (a) GATT allows exceptions regarding frontier traffic with adjacent countries and Art. XX GATT entails a general exception for measures necessary to protect public morals, life and health while Art. XI GATT allows deviation from most-favoured-nation treatment for national security reasons. The two most significant, and at the same time most controversial, exceptions are the exception relating to customs unions and free trade areas entailed in Art. XXIV GATT as well as the implementation of a ‘generalized, non-reciprocal, non-discriminatory system of preferences’ (‘GSP’) under the so-called Enabling Clause (see paras 3537 below).

2.  Special Legal Problems

(a)  The Ejusdem Generis Principle Applied in Practice

31  In the absence of distinct specifications within a clause, the ascertainment of what subject-matters, persons, or things fall within the same category has proven to be rather difficult as it hinges on the interpretation of the intended scope of the clause. International jurisprudence contains ample evidence of the difficulties appertaining to the application of the clause in practice. In the recent past such problems can be seen in particular in the context of modern investment treaties, which nowadays regularly include most-favoured-nation clauses. Here, a rather mechanical application of the clause disregarding the parties’ original intentions poses the risk of altering treaty arrangements specifically made by the parties and extending most-favoured-nation treatment to areas which were not meant to be included (Dolzer and Schreuer 186–87).

32  However, the ICJ in the United States Nationals in Morocco Case, albeit inconclusively, touched upon the question whether a clause contained in a treaty of commerce could be understood to cover consular jurisdiction ([Judgment of 27 August 1952] 191) while the Commission of Arbitration in the Ambatielos Case adopted a rather broad understanding of the clause’s scope holding that ‘the “administration of justice”, when viewed in isolation, is a subject-matter other than “commerce and navigation”, but this is not necessarily so when it is viewed in connection with the protection of the rights of traders’ (Ambatielos Arbitration [Greece v United Kingdom] [6 March 1956] 107). Largely on the basis of this decision an Arbitration Tribunal acting under the auspices of the International Centre for Settlement of Investment Disputes (ICSID) in a more recent case held that ‘there are good reasons to conclude that today dispute settlement arrangements are inextricably related to the protection of foreign investors, as they are also related to the protection of rights of traders under treaties of commerce’ (Maffezini v Spain Case [Decision of the Tribunal on Objections to Jurisdiction of 25 January 2000] para. 54). Subsequently, however, the ICSID Tribunal in the Tecmed Case adopted a more restrictive understanding by excluding from the clause’s scope such provisions in third-party treaties that refer to ‘the core of matters that must be deemed to be specifically negotiated by the contracting parties’ (Técnicas Medioambientales Tecmed SA v United Mexican States [Award of 29 May 2003] para. 69). This more restrictive interpretation has been affirmed by ICSID Tribunals in two recent cases, one of which held that: ‘[A]n MFN provision in a basic treaty does not incorporate by reference dispute settlement provisions in whole or in part set forth in another treaty, unless the MFN provision in the basic treaty leaves no doubt that the contracting parties intended to incorporate them’ (Plama Consortium Limited v Republic of Bulgaria [Decision on Jurisdiction of 8 February 2005] para. 223; see also Salini Costruttori SpA and Italstrade SpA. v Hashemite Kingdom of Jordan [Decision of the Tribunal on Jurisdiction of 29 November 2004] paras 102–19). In light of the number of investment treaties and the wide spectrum of differently formulated most-favoured-nation clauses, disputes relating to the scope of the clause in the investment area are likely to reoccur in the near future. In particular, the application of MFN clauses to dispute settlement provisions remains an unsettled question in investment treaty arbitration. In several cases, States used MFN clauses trying to import dispute settlement clauses from other BITs in order to bypass procedural pre-conditions to arbitration such as the exhaustion of domestic remedies or certain waiting periods (Impregilo SpA v Argentina [Award of 21 June 2011] ICSID Case No ARB/07/17; Hochtief AG v Argentina [Decision on Jurisdiction of 24 October 2011] ICSID Case No ARB/07/31; Teinver SA v Argentina [Decision on Jurisdiction of 21 December 2012] ICSID Case No ARB/09/1). In Impregilo, Brigitte Stern disagreed with the majority decision and warned in her detailed dissenting opinion of the ‘great dangers’ of allowing claimants to bypass a treaty’s jurisdictional requirements by invoking MFN clauses (see also ICS Inspection and Control Services Ltd [United Kingdom] v Argentine Republic [Award on Jurisdiction of 10 February 2012] PCA Case No 2010-9; Daimler Financial Services AG v Argentina [Award of 22 August 2012] ICSID Case No ARB/05/1; Kiliç İnşaat İthalat İhracat Sanayi ve Ticaret Anonim Şirketi v Turkmenistan [Award of 2 July 2013] ICSID Case No ARB/10/01). In Garanti Koza v Turkmenistan ([Decision on Jurisdiction of 3 July 2013] ICSID Case No ARB/11/20) the tribunal went even further by allowing a claimant to use an MFN clause to ‘import’ a State’s consent to a particular arbitral forum found in another BIT (see De Brabandere; Dissenting Opinion by Arbitrator Boisson de Chazournes).

33  The negotiating history of the Dominican Republic–Central America Free Trade Agreement (2004) (‘CAFTA-DR’), reflects how such ambiguities in the interpretation of MFN clauses can be avoided by way of careful drafting. The parties to the CAFTA-DR, referring explicitly to the Maffezini Case, included a footnote in the negotiating history to reflect their shared interpretation, providing that the most-favoured-nation clause entailed in the investment chapter does not encompass international dispute resolution mechanisms ‘and therefore could not reasonably lead to a conclusion similar to that of the Maffezini Case’.

(b)  Free Trade Areas and Customs Unions

34  A free trade area or customs union confers relative advantages on its members and corresponding disadvantages on all countries that are not members. Art. XXIV GATT therefore authorizes free trade areas only if certain conditions are met. Historically, these conditions represent an effort to reconcile the objectives of encouraging a European customs union while insisting on most-favoured-nation treatment. However, despite the creation of a Committee on Regional Trade Agreements in 1996, mandated to determine whether the conditions set out in Art. XXIV GATT have been met, regional, and in fact bilateral, free trade areas have proliferated vastly (Cone 564). According to WTO statistics, 382 RTAs are currently in force (<http://rtais.wto.org/UI/publicsummarytable.aspx> [5 June 2014]). This process undermines the principle of most-favoured-nation treatment. On this basis some authors have gone so far as to question the future of the most-favoured-nation rule as a fundamental principle of international trade agreements (ibid 564, 583).

(c)  Preferences for Developing States

35  The trade needs of a developing economy are substantially different from those of a developed economy (Developing Countries). The question how to accommodate this asymmetry within one coherent trade system has accompanied the GATT/WTO system ever since its inception. The most-favoured-nation clause, if universally applied to all Member States regardless of their development, would involve implicit discrimination against the weaker Member States. Accordingly, developing States demanded preferential conditions of market access in order to gain a foothold in the highly competitive markets of the developed countries. In response to these demands a GSP was agreed in principle at the second session of UNCTAD in 1968. In 1971 the GATT contracting parties, following the lead of UNCTAD, decided to waive the provisions of Art. I GATT for a period of 10 years, permitting developed countries to grant preferential tariff treatment to the exports of developing countries. In 1979 on the basis of the ‘Decision on Differential and More Favourable Treatment, Reciprocity, and Fuller Participation of Developing Countries’ this system was extended beyond its initial 10-year period. This decision, commonly known as the ‘Enabling Clause’, gave formal embodiment and therefore a stronger legal basis to the GSP concept of special and differential treatment of developing and least developed countries within the GATT.

36  Whether the GSP constitutes a wise departure from the obligation of most-favoured-nation treatment remains a matter of continuous controversy. Originally conceived as a system that would be generalized and non-discriminatory, GSP as implemented under the Enabling Clause consists of a series of individual national schemes which differ in many respects, eg as to product coverage, tariff cuts, safeguards (escape clause), and rules of origin. The extension of preferential treatment is voluntary, it may be revoked at any time, and donor countries reserve their right to self-selection, ie to grant preferences to certain but not necessarily to all developing countries. Moreover, in practice donor countries often condition the granting of a preference on the developing country’s attainment of certain non-trade-related goals eg labour, environmental, or drug-related issues. Recently, the United States conditioned the granting of a preference on the beneficiary country’s recognition and enforcement of arbitral awards in favor of United States nationals. Thus, the Obama Administration announced on 26 March 2012 that Argentina’s GSP beneficiary designation would be suspended ‘because it has not acted in good faith in enforcing arbitral awards in favor of United States citizens’ (‘Presidential Proclamation—To Modify Duty-free Treatment Under the Generalized System of Preferences and for Other Purposes’ para. 2; see also Alford 50 et seq).

37  This question of conditioning the granting of a preference also came to the fore in a trade dispute between India and the European (Economic) Community regarding the conditions under which the EC accords tariff preferences to developing countries. For the first time the WTO Dispute Settlement Body had the opportunity to analyse the nature and conditions of the Enabling Clause in detail (World Trade Organization, Dispute Settlement). The Appellate Body, in its decision of April 2004, concurred with the Panel’s finding that the Enabling Clause does not exclude the applicability of Art. I (1) GATT 1994, but merely constitutes an ‘exception’ thereto (WTO EC—Conditions for the Granting of Tariff Preferences to Developing Countries—Report of the Appellate Body para. 190), ie in the words of the Panel a ‘limited authorization of derogation’ (WTO EC—Conditions for the Granting of Tariff Preferences to Developing Countries—Report of the Panel para. 7.38). However, the Appellate Body reversed the Panel’s rather rigid interpretation according to which the Enabling Clause—apart from two defined exceptions—requires identical tariff preferences to be accorded to all developing countries without differentiation (ibid paras 7.116, 7.161). Instead, the Appellate Body held that the requirement for non-discriminatory treatment entailed in the Enabling Clause does not presuppose formally identical treatment, but requires identical treatment merely of similarly-situated beneficiaries (Conditions for the Granting of Tariff Preferences to Developing Countries—Report of the Appellate Body para. 153). According to the Appellate Body distinctions among beneficiaries must respond positively to the needs of developing countries (ibid paras 162, 165) and the types of needs to which a response is envisaged are limited to development, financial, and trade needs to be assessed according to an objective standard (ibid para. 163). The Appellate Body thus did not rule out conditional GSP programs per se. However, it is to be expected that the scope of the requirements for differential treatment drawn up by the Appellate Body and the conformance of the conditions contemporarily employed will be tested in future litigation (Schmahl 406).

D.  Evaluation

38  The most-favoured-nation clause is an age-old device that continues to serve constant needs of the international society. The clause has flourished the most in the era of free trade. Apart from its egalitarian function, the automatism it entails very suitably responds to State demands in times of globalization and a historically unprecedented proliferation of treaties in the trade and investment area. The clause is a flexible instrument, adopted at will, it can be tailored according to individual interests and even adjusted to correct the inherent injustice of treating equally States, which economically are highly unequal. The overriding aim to preserve this flexibility may be one reason why the ILC draft articles—albeit they provide most useful guidance on the topic and even though they are without prejudice to any provision on which States may otherwise agree—never found their way into a formal convention. The beneficial functions and effects of the clause have secured its survival for more than 800 years. Thus, even though the on-going proliferation of bilateral and regional free trade areas carries the potential to challenge the clause’s position as a cornerstone of world trade, over time, when disadvantages of such a fragmented trade order again become evident, the most-favoured-nation clause will once more prevail.

39  In the area of world trade, the clause’s long-time position as a cornerstone of world trade is now challenged by the on-going proliferation of bilateral and regional free trade areas. Yet, more recently, a new domain of the clause has emerged: the field of investment agreements. It is in view of this development that the ILC once again decided to focus its work on the meaning and effect of the clause. On 1 June 2007, the ILC established an open-ended working group, which recommended that the topic of the most-favoured-nation clause be included in the long-term programme of work of the ILC with a view to prepare commentaries on model most-favoured-nation clauses developed from a consideration of State practice and jurisprudence since the conclusion of the ILC’s work on the subject in 1978. The proliferation of MFN provisions in the field of investment agreements and particularly the different approaches taken by various investment tribunals have created new challenges with respect to MFN provisions. Most importantly, in the absence of a clear and consistent interpretation of MFN provisions, granting States may not be able to reliably determine in advance what kind of obligations they are undertaking by agreeing to a MFN clause in an investment agreement. The ILC could thus provide authoritative guidance for the coherent interpretation of MFN clauses in investment agreements. In view of these developments, it seems save to say that the finding of one League of Nations expert, still holds true: ‘nations do not seem to be able to escape the use of the clause’ (see Ustor 472).

Select Bibliography

Select Documents