The value of the WTO: An economic viewpoint in six instalments Part II: Why MFN matters

The most-favoured-nation (MFN) principle is one of the cornerstones of the multilateral trading system founded by the General Agreement on Tariffs and Trade (GATT) and embodied by the WTO. It is often described simply as a non-discrimination rule, requiring that any tariff granted to one trading partner be extended to all WTO members. But its economic significance runs much deeper.

This is not to say that the principle is beyond reproach. There are problems with the application of this principle that are well known to economists and that I will feature in my next post. Also, some members have questioned whether MFN is still fit for purpose in today's environment. In an earlier post, I described my impressions from the reform discussions at the 14th Ministerial Conference earlier this year in Yaounde, Cameroon, on MFN and other issues. As background for these reform discussions, it is useful to revisit the role that MFN has played in the world trading system that we have today.

In my previous post, I argued that the central purpose of the GATT/WTO is to provide a forum where governments can "negotiate away" the cross-border spillovers created by their unilateral tariff policies. Viewed through this lens, MFN is far more than a fairness principle. It is a key institutional feature that has made multilateral tariff negotiations possible.

MFN performs two fundamental economic functions. First, it dramatically simplifies the tariff bargaining problem that governments face by reducing both the complexity and the transaction costs of negotiating market access. Second, it helps preserve the value of negotiated concessions by preventing them from being undermined through subsequent discriminatory agreements. Together, and especially when combined with the principle of reciprocity (i.e. a balanced exchange of market access commitments), these functions transform what could otherwise be a fragmented and highly strategic bargaining process into a stable framework for international cooperation.

The historical record illustrates just how important these institutional features have been. Evidence from the early GATT negotiating rounds suggests that they fundamentally changed the way governments bargained, replacing many of the strategic tactics that had characterized earlier tariff negotiations with a more transparent and cooperative process. The result was the extensive network of tariff commitments and market access concessions that continues to underpin the multilateral trading system today.

In this post, I develop three main ideas. First, MFN simplifies tariff bargaining by reducing the complexity of the negotiating problem for members, a result that follows from the fact that the international spillovers created by tariffs are transmitted through markets and prices. Second, MFN helps preserve the value of negotiated market access commitments, whereas weakening it risks reintroducing instability and complexity into the trading system. Finally, historical evidence from the early GATT rounds shows how these institutional features shaped bargaining behaviour in practice.

The first role of MFN: Simplifying tariff bargaining

The first critical function of MFN is that it simplifies tariff negotiations.

As pointed out in the previous post, a fundamental feature of tariff bargaining in the GATT/WTO has been to address the cross-border effects, or "pecuniary" externalities, that governments impose on one another when they set tariffs unilaterally. Without MFN, this bargaining problem would quickly become extraordinarily complex.

If for any individual product, governments were free to apply different tariffs to different trading partners on that product, each tariff change would affect every exporter differently. Every bilateral trading relationship regarding that product would generate its own cross-border spillover. Governments would therefore need to negotiate separately with each trading partner over every product, creating an enormous number of individual bargaining problems. The complexity of such negotiations would rapidly become unmanageable.

MFN largely eliminates this problem. When the same tariff must be applied to every trading partner, a tariff change affects all exporters of a given product in the same way on a per-unit basis. Instead of confronting a separate bargaining problem with every trading partner, governments face a single bargaining problem for each product.

The negotiating process was simplified even further through the GATT's "principal supplier" rule. Rather than negotiating simultaneously with every exporter, governments can focus their discussions on the main ("principal") suppliers of each product, while the resulting concession is automatically extended to all other members through MFN. This greatly reduces the number of negotiations needed to achieve a multilateral outcome. This insight is especially relevant, as renegotiating discriminatory trade arrangements risks reintroducing the complexity problem that MFN helped to solve.¹

What makes MFN particularly remarkable is that its ability to simplify tariff bargaining stems directly from the nature of the cross-border effect that governments are trying to address. The international spillovers created by tariffs operate through prices and markets. When a government raises a tariff, it can affect not only domestic prices and the quantity of imports but also the prices received by foreign exporters. Because these are "pecuniary" spillovers, a rule requiring non-discrimination can fundamentally change their features, and hence change the structure of the bargaining problem itself. By contrast, no comparable institutional rule exists in areas such as climate policy. Carbon emissions generate a global environmental spillover that does not operate through market prices, and that cannot be reshaped through bargaining rules in the way MFN reshapes tariff bargaining.²

Seen in this light, MFN occupies a unique place within the multilateral trading system. This is not to say that the principle admits no exceptions. From the very beginning, the GATT allowed carefully defined departures from MFN, most notably through Article XXIV, which permits the formation of customs unions and free-trade areas. But these departures have always been understood precisely as exceptions, and the economic logic outlined above helps explain why. Weakening or abandoning MFN inevitably comes at a cost, reintroducing complexity into the bargaining process and undermining one of the institutional innovations that has made multilateral tariff negotiations workable.

The second role of MFN: Preserving the value of trade bargains

If MFN makes agreements easier to reach, it also plays an equally important role after negotiations have concluded. The second critical function of MFN is that it helps ensure that negotiated market access commitments are secure and durable.

A central objective of the GATT/WTO system is to provide a forum in which governments can exchange market access concessions, reducing tariffs on imports in return for improved market access abroad, while being confident that the value of those concessions will not be eroded over time.

MFN helps preserve the value of market access concessions by requiring that any tariff concession granted to one member be extended to all. In doing so, it ensures that negotiated market access is multilateralized and therefore protected from erosion through later discriminatory deals. This matters because trade negotiations do not occur in isolation. Governments continue to negotiate with other partners. Without MFN, the value of concessions secured in one negotiation could later be eroded by agreements reached elsewhere. When combined with the principle of reciprocity, which anchors negotiations in the exchange of balanced concessions, this mechanism helps stabilize the value of negotiated agreements over time, without the need for continual renegotiation in response to the tariff bargaining of others.

By contrast, if economies were free to negotiate unrestricted discriminatory tariff changes on a product-by-product basis, even when aimed at market opening, the stability of earlier bargains would be put at risk. A concession secured in one negotiation could be partially or fully neutralized by subsequent agreements between other parties.

How have MFN and other GATT rules shaped tariff bargaining?

The importance of MFN becomes clearer when one considers how tariff bargaining has actually taken place under GATT rules.

The MFN tariff commitments that exist today are the outcome of eight multilateral negotiating rounds conducted under the GATT, culminating in the Uruguay Round and the creation of the WTO in 1995. As Table 1 shows, a key feature of many of these rounds was the use of simultaneous bilateral negotiations, in which governments exchanged market access requests and offers to reduce tariffs across multiple bargaining partners at the same time. The results of these bilateral bargains were then multilateralized and extended to all members through MFN.

Table 1: Negotiation methods used in successive GATT negotiating rounds

Year	Conference / Round	Methodology	Members
1947	Geneva Tariff Conference	Reciprocal "request-and-offer" bilateral talks. Product-by-product negotiations based on principal supplier rule.	23
1949	Annecy Tariff Conference		13
1950-51	Torquay Tariff Conference		38
1956	Geneva Tariff Conference		26
1960-61	Geneva Dillon Round	Primarily "request-and-offer" negotiations, with limited use of linear tariff-cut proposals.	26
1964-67	Geneva Kennedy Round	Linear Formula*	62
1973-79	Tokyo Round	Swiss Formula* with exceptions	102
1986-94	Uruguay Round	Mixed approach: Formula(s) and "request-and-offer"	123

^*Note: The Kennedy Round linear formula applied a uniform proportional tariff cut to all tariffs. The Tokyo Round Swiss formula applied deeper proportional cuts to higher tariffs than to lower tariffs, thereby compressing tariff dispersion.

This institutional structure allowed tariff negotiations to become more transparent and less strategic. Rather than engaging in complex bargaining over each bilateral relationship, governments could focus on offering and requesting market access in a way that was effectively anchored by the reciprocity norm. In this environment, the "price" of market access became more predictable, and negotiations resembled a structured exchange rather than open-ended bargaining.

Economists have described this outcome as transforming tariff negotiations into something that could be characterized as a "retail store for market access," where governments arrived at the negotiating table with clearly defined tariff concessions that they were ready to offer, and sought counterparties willing to exchange equivalent concessions. In this setting, the scope for strategic delay or manipulation is reduced, and the focus shifts toward identifying mutually beneficial exchanges.

Contemporary accounts of early GATT rounds suggest that this characterization is not purely theoretical. Curzon (1965) explains why "lowball" opening offers, where minimal initial tariff reductions are offered to begin negotiations, made little sense in this setting. Because governments expected non-strategic behaviour from their negotiating partners, an initial offer was generally taken at face value rather than treated as a starting point for haggling. He notes that, in the Torquay Round, this non-strategic behaviour was most pronounced among governments that had already bargained under GATT rules at the earlier Geneva (1947) and Annecy (1949) rounds and had by then learned the new bargaining technique encouraged by GATT rules.

Supporting this view are the findings of Bagwell, Staiger and Yurukoglu (2020), summarized in Table 2, who examine the GATT bargaining records for the Torquay Round (1950-51).

Table 2: "Old-timers" and newcomers in the Torquay Round

Country group	Initial tariff offer as a % of existing tariff	Final agreed concession as a % of existing tariff
	(Total number of tariff offers)	(Total number of tariff concessions)
GATT old-timers	80.8% (11,964)	80.6% (7,997)
GATT newcomers	85.5% (5,243)	81.9% (2,939)

Source: The table is drawn from Table A3 and Table A2 of the Online Appendix of Bagwell, Staiger and Yurukoglu (2020) and is based on the GATT bargaining records posted on the WTO website.

"Old-timer" governments, those that had participated in earlier GATT rounds, arrived in Torquay with offers that already reflected a substantial degree of market opening, and that would have reduced average tariffs to 80.8% of existing levels. These initial offers were very close to the final agreed outcomes (80.6%): there was relatively little strategic adjustment during the tariff bargaining process itself. The negotiations were not characterized by large revisions or tactical "lowballing," but rather by the search for counterparties willing to reciprocate pre-existing offers.

By contrast, "newcomer" governments behaved quite differently. They entered the negotiations with relatively cautious offers, proposing to reduce tariffs only to 85.5% of existing levels, and expecting that bargaining would involve gradual convergence toward mutually acceptable outcomes. As Curzon (1965, p.74) observes, that expectation proved misplaced:

"Several newcomers to GATT unaware of this new technique and starting with low offers found that in the course of negotiations they were unable to reach the level of requests they aimed for. Their initially low offers were taken as proof of their intentions and they either had to go home with a tariff higher than expected or had to increase their offers in the course of the negotiations."

By the close of the round, newcomers had agreed to bindings cutting average tariffs to 81.9% of existing levels, a much larger move from their opening position. Having started with lowball offers, these governments ended up revealing to their bargaining partners that they were willing to accept deeper cuts than they had first suggested.

As Ronald Coase argued, this is precisely the kind of strategic behaviour that can lead to costly delay or even bargaining failure. The old-timers, more experienced in the GATT's bargaining technique, had largely given up this kind of behaviour.

This contrast is also reflected in the success rates of initial offers. Old-timer governments succeeded in concluding agreements on 67% (7,997 of 11,964) of the products for which they made initial offers, while newcomers succeeded on only 56% (2,939 of 5,243). This difference is consistent with the idea that experienced participants had internalized the institutional logic of GATT bargaining, while newcomers were still adapting to it.

Main takeaway

MFN is often described as a simple non-discrimination rule, but its economic role within the multilateral trading system is much more fundamental.

By simplifying the tariff bargaining problem and stabilizing the value of negotiated concessions, MFN has helped make multilateral tariff negotiations feasible and durable. The historical evidence from the early GATT rounds further suggests that these institutional features were not merely formal constraints, but actively shaped how governments behaved in practice.

Seen from this perspective, MFN is one of the institutional innovations that has made the multilateral trading system a durable framework for reciprocal cooperation.

---

1. This is separate from the complexity that discriminatory tariffs introduce for the governments that must administer them and the firms that must navigate them. This complexity can create additional costs in terms of lost income and diminished consumer choice. back to text
2. Carbon border adjustments are aimed at the pecuniary international price effects of the policies that economies adopt to address climate change (e.g., carbon taxes), but carbon border adjustments do not change the features of the global non-pecuniary spillover created by carbon emissions. back to text

References

Bagwell, Kyle, Robert W. Staiger, and Ali Yurukoglu (2020). "Multilateral Trade Bargaining: A First Look at the GATT Bargaining Records." American Economic Journal: Applied 12(3): 72-105.

Curzon, Gerard (1965). "Multilateral Commercial Diplomacy: The General Agreement on Tariffs and Trade and Its Impact on National Commercial Policies and Techniques". London: Michael Joseph.