VoxEU Column International trade

Clubs and the WTO post-Nairobi: What is feasible? What is desirable?

The 2015 Nairobi WTO Ministerial unshackled governments from the deadlocked Doha Development Agenda and opened the door for new issues and new approaches. The Ministerial Declaration calls on new initiatives to be agreed by consensus. This column argues that WTO procedures permit ‘clubs’ of countries to agree on additional policy disciplines if the benefits extend on a non-discriminatory basis to all WTO members. Consensus is not needed for such clubs. 

A staple of WTO negotiations on new rules of the game for trade has been the notion of a ‘single undertaking’ – countries must agree on a package deal embodying issue linkages and tradeoffs across different policy areas. This was a core feature of the Doha round, as it was in the Uruguay Round (1986-1993). As Doha talks became deadlocked, WTO members relaxed the single undertaking constraint to permit deals to be struck on elements of the negotiating agenda. The foremost example was the 2013 Trade Facilitation Agreement – the first multilateral agreement negotiated under WTO auspices. Other examples of such pragmatism include the Nairobi Ministerial agreement among 53 countries to extend the product coverage of the Information Technology Agreement (ITA). The ITA involves a club of WTO members agreeing to eliminate tariffs and extending the benefits of this to all WTO members in accordance with the most-favoured-nation (MFN) rule. It is an example of an ‘MFN club’ (Saggi and Sengul 2009).

In recent years, analysts and governments have suggested many issues in addition to those on the agenda of the Doha round for deliberation in the WTO. Looking forward, governments will need to determine not only what subjects call for multilateral policy disciplines, but also whether such disciplines should apply to all WTO members. In the case of some issues it may not be necessary, desirable, or feasible that new rules of the game apply to all WTO members.  An important legal/technical question that arises in this regard is whether a subset of countries can form agreements on new WTO+ rules pertaining to non-tariff policies in the absence of consensus. In the case of tariff reductions, clubs can move ahead because any WTO member can amend its schedule of tariff concessions and reduce its tariff bindings to zero. Indeed, GATT schedules of concessions effectively are comprised of tariff bindings. As long as unilateral decisions to lower bindings are extended to all WTO members – as is required by the MFN rule – there is no constraint on a country doing so, or, as in the ITA, on a group of countries acting in concert.

Whether such MFN clubs can address non-tariff measures (NTMs) is less obvious. The presumption of many analysts and government representatives seems to be that GATT schedules of concessions are limited to tariff bindings for traded goods. These are complemented by specific commitments pertaining to policies affecting trade in a range of services that are annexed to the GATS. The GATS differs from the GATT by including a provision – Article XVIII GATS – allowing members to make additional commitments that complement their specific market access and national treatment commitments for services sectors and sub-sectors. The GATT has no such provision. The inference that is often is often drawn is that if one or more WTO members wants to make new binding commitments on a non-tariff policy that affects its trade in goods, there is no obvious way to do so. Any group would have to convince the membership as a whole to agree to apply the proposed rules, or they would have to obtain the approval of the whole membership for a club to form a new plurilateral agreement that applies only to signatories.1 In both cases, consensus is required.

Scheduling NTM commitments under the GATT

In a new paper (Hoekman and Mavroidis 2016), we argue that MFN clubs seeking to deepen commitments on NTMs can schedule these in the same way they list tariff bindings. Art. II:1(a) GATT states that a contracting party “shall accord to the commerce of the other contracting parties treatment no less favourable than that provided for in the appropriate Part of the appropriate Schedule annexed to this Agreement.” There is no mention of tariffs, suggesting a broader set of policies can be bound.  In fact, the WTO membership formally accepted scheduling of NTMs during the Uruguay round negotiations. The agreed format of Uruguay round schedules comprised of four parts:2

  • Part I: tariff bindings and concessions on tariff rate quotas;
  • Part II: historic preferential tariffs from the early GATT days which have since become defunct;
  • Part III: commitments on nontariff measures; and 
  • Part IV: non-tariff commitments on farm goods.

All WTO members have structured their schedules according to this format; acquiescence with this 1993 document is uncontested. The format permits WTO members to list whatever NTM commitments they deem appropriate to schedule under Part III – the content of what can be listed is not at all prejudged. As discussed in detail in Hoekman and Mavroidis (2016), the only constraint is that whatever is scheduled conforms to the disciplines of the GATT/WTO, most notably MFN. This means that MFN clubs can act in a concerted fashion and schedule additional commitments for specific NTMs – non-participants cannot block other WTO members from doing so. In this respect, while the GATT differs from the GATS in that the latter makes life much simpler by including a provision under which WTO members can make additional commitments (Art. XVIII GATS), in practice the absence of such a provision in GATT does not preclude members from scheduling additional commitments. 

Scheduling additional commitments that go beyond the current WTO rules does not in itself confer legality. WTO members can challenge any scheduled items, with the legal benchmark for deciding on complaints being the WTO public order. An implication is that multilateral scrutiny cannot be used as a mechanism to ‘approve’ or reject WTO+ commitments. The legality of such commitments is ultimately a matter for dispute settlement – the Appellate Body.

This does not mean scheduling additional GATT commitments is a panacea. For example, in practice new issues may affect both goods and services. In principle, the best way forward for WTO members with an interest in considering the pursuit of MFN clubs for new policy disciplines affecting trade in goods would be to add a provision to the GATT that would provide an analogous function to Art. XVIII GATS. This would be the ‘cleanest’ way of creating a mechanism through which WTO members could incorporate MFN clubs pertaining to policies affecting trade in goods as well as services. While we believe that this would be beneficial in providing a structured framework for club-based cooperation looking forward, WTO members need not wait for such a process to be initiated and concluded. Club-based discussions on potential new disciplines can be launched already under current rules and provisions. Indeed, doing so may generate the political attention and engagement that will be needed for WTO members to consider if and how an Art. XVIII analogue could be included into the GATT, thereby provide a supporting framework for new MFN clubs.


The final sentence of the Nairobi WTO Ministerial Declaration states that “[a]ny decision to launch negotiations multilaterally on such [new] issues would need to be agreed by all members” (WTO 2015). The discussion above suggests that consensus is not needed for new agreements among MFN clubs. Of course, many policy areas that have been suggested for rule-making in recent years are best pursued on a universal basis. The Bali Trade Facilitation Agreement illustrates that it is possible to agree on rules that apply to all WTO members and to address differences in implementation capacities. But it must also be recognised that clubs are a fundamental part of the international landscape – as revealed by the even expanding number of PTAs. MFN clubs, while not necessarily first-best, are likely to be superior in terms of global welfare to PTAs.

The binding constraint on MFN clubs is political—a critical mass of countries must see it in their self-interest to work on a multilateral basis as opposed to a preferential, discriminatory one. The prospects for club-based multilateral initiatives may be better than is often perceived by observers of the WTO. The successful outcome of the 2015 COP21 climate change conference in Paris, which was premised on a bottom-up, club-based approach, may motivate countries to do the same in the trade context. In any event, current WTO rules do not prevent such approaches from being pursued.


Hoekman, B and P C Mavroidis (2015), “Embracing Diversity: Plurilateral Agreements and the Trading System”, World Trade Review 14: 101-116.

Hoekman, B and P C Mavroidis (2016), “MFN Clubs and Scheduling Additional Commitments in the GATT: Learning from the GATS”, EUI Working Paper RSCAS 2016/06.   

Saggi, K and F Sengul (2009), “On the Emergence of an MFN Club, Equal Treatment in an Unequal World”, Canadian Journal of Economics 42: 267-299.

WTO (2015), Nairobi Ministerial Declaration, 19 December.


1 The plurilateral agreement option is discussed in Hoekman and Mavroidis (2015).

2 GATT Doc. MTN.GNG/MA/W/25 of December 25, 1993.

1,365 Reads