Historic development of the WTO dispute settlement system

The (WTO) dispute settlement system is often praised as one of the most important innovations of the Uruguay Round. This should not, however, be misunderstood to mean that the WTO dispute settlement system was a total innovation and that the previous multilateral trading system based on GATT 1947 did not have a dispute settlement system.

On the contrary, there was a dispute settlement system under GATT 1947 that evolved quite remarkably over nearly 50 years on the basis of Articles XXII and XXIII of GATT 1947. Several of the principles and practices that evolved in the GATT dispute settlement system were, over the years, codified in decisions and understandings of the contracting parties to GATT 1947. The current WTO system builds on, and adheres to, the principles for the management of disputes applied under Articles XXII and XXIII of GATT 1947 (Article 3.1 of the DSU). Of course, the Uruguay Round brought important modifications and elaborations to the previous system, which will be mentioned later.1 This chapter provides a brief overview of the historic roots of the current dispute settlement system.

Click the + to open an item.

2.1 The system under GATT 1947 and its evolution over the years

show help page

Articles XXII and XXIII and emerging practices

The rudimentary rules in Article XXIII:2 of GATT 1947 provided that the contracting parties themselves, acting jointly, had to deal with any dispute between individual contracting parties. Accordingly, disputes in the very early years of GATT 1947 were decided by rulings of the Chairman of the GATT Council. Later, they were referred to working parties composed of representatives from all interested contracting parties, including the parties to the dispute. These working parties adopted their reports by consensus decisions. They were soon replaced by panels made up of three or five independent experts who were unrelated to the parties of the dispute. These panels wrote independent reports with recommendations and rulings for resolving the dispute, and referred them to the GATT Council. Only upon approval by the GATT Council did these reports become legally binding on the parties to the dispute. The GATT panels thus built up a body of jurisprudence, which remains important today, and followed an increasingly rules-based approach and juridical style of reasoning in their reports.

The contracting parties to GATT 1947 progressively codified and sometimes also modified the emerging procedural dispute settlement practices. The most important pre-Uruguay Round decisions and understandings were:

  • The Decision of 5 April 1966 on Procedures under Article XXIII;
  • The Understanding on Notification, Consultation, Dispute Settlement and Surveillance, adopted on 28 November 19792;
  • The Decision on Dispute Settlement, contained in the Ministerial Declaration of 29 November 19823;
  • The Decision on Dispute Settlement of 30 November 1984.4


Weaknesses of the GATT dispute settlement system back to top

Some key principles, however, remained unchanged up to the Uruguay Round, the most important being the rule of positive consensus that existed under GATT 1947. For example, there needed to be a positive consensus in the GATT Council in order to refer a dispute to a panel. Positive consensus meant that there had to be no objection from any contracting party to the decision. Importantly, the parties to the dispute were not excluded from participation in this decision-making process. In other words, the respondent could block the establishment of a panel. Moreover, the adoption of the panel report also required a positive consensus, and so did the authorization of countermeasures against a non-implementing respondent. Such actions could also be blocked by the respondent.

One might think that such a system could not possibly have worked. Why would a respondent not use its right to block the establishment of a panel if it thought that it might lose the case? Why would the losing party not block the adoption of the panel report? How could a party refrain from using its veto against the authorization of countermeasures, from which it would suffer economically? If domestic judicial systems were to operate on the basis of such a consensus rule, they would probably fail in most instances.

Quite surprisingly, this was generally not the experience of the dispute settlement system of GATT 1947. Individual respondent contracting parties mostly refrained from blocking consensus decisions and allowed disputes in which they were involved to proceed, even if this was to their short-term detriment. They did so because they had a long-term systemic interest and knew that excessive use of the veto right would result in a response in kind by the others. Accordingly, panels were established and their reports frequently adopted, albeit often with delays (even though the authorization of countermeasures was only granted once).

On the basis of empirical research, it has been concluded that the GATT 1947 dispute settlement system brought about solutions satisfying the parties in a large majority of the cases. However, it must be noted that, by their nature, such statistics can only cover complaints that were actually brought. Certainly, there were a significant number of disputes that were never brought before the GATT because the complainant suspected that the respondent would exercise its veto. Thus, the risk of a veto also weakened the GATT dispute settlement system. In addition, such vetoes actually occurred, especially in economically important or politically sensitive areas such as anti-dumping. Finally, there was a deterioration of the system in the 1980s as contracting parties increasingly blocked the establishment of panels and the adoption of panel reports.

Even when panel reports were adopted, the risk of one party blocking adoption must often have influenced the panels’rulings. The three panelists knew that their report had also to be accepted by the losing party in order to be adopted. Accordingly, there was an incentive to rule not solely on the basis of the legal merits of a complaint, but to aim for a somewhat “diplomatic” solution by crafting a compromise that would be acceptable to both sides.

Hence, the structural weaknesses of the old GATT dispute settlement system were significant even though many disputes were ultimately resolved. As noted in the late 1980s, when the Uruguay Round negotiations were ongoing, the situation deteriorated, especially in politically sensitive areas or because some contracting parties attempted to achieve trade-offs between ongoing disputes and matters being negotiated. This resulted in a decreasing confidence by the contracting parties in the ability of the GATT dispute settlement system to resolve the difficult cases. In turn, this also led to more unilateral action by individual contracting parties, who, instead of invoking the GATT dispute settlement system, would take direct action against other parties in order to enforce their rights.5


Dispute settlement under the Tokyo Round “codes” back to top

Several of the plurilateral agreements emerging from the Tokyo Round of Multilateral Trade Negotiations, the so-called “Tokyo Round Codes”, for example the one on Anti-Dumping, contained code-specific dispute settlement procedures. Like the codes as a whole, these specific dispute settlement procedures were applicable only to the signatories of the codes, and only with regard to the specific subject matter. If the multilateral trading system before the establishment of the WTO was often referred to as a “GATT Ó la carte”, this also applied to dispute settlement. In some instances, where rules pertaining to a specific subject-matter existed both in GATT 1947 and in one of the Tokyo Round Codes, a complainant also had some leeway for “forum-shopping” and “forum-duplication”, i.e. choosing the agreement and the dispute settlement mechanism that promised to be the most beneficial to its interests, or launching two separate disputes under different agreements on the same matter.

In terms of how satisfactorily the dispute settlement system under these codes functioned, the record was less favourable than it was for GATT 1947, i.e. consensus was blocked quite frequently.


The Uruguay Round and the Decision of 1989 back to top

As the inherent problems in the GATT dispute settlement system led to increasing problems in the 1980s, many contracting parties to GATT 1947, both developing and developed countries, felt that the system needed improving and strengthening. Negotiations on dispute settlement were accordingly included and given high priority on the agenda of the Uruguay Round negotiations.

By 1989, midway through the Uruguay Round negotiations, the contracting parties were ready to implement some preliminary results of the negotiations (“early harvest”) on certain issues and accordingly adopted the Decision of 12 April 1989 on Improvements to the GATT Dispute Settlement Rules and Procedures.6 The decision was to apply on a trial basis until the end of the Uruguay Round and already contained many of the rules later embodied in the DSU, such as a right to a panel and strict time-frames for panel proceedings. However, there was no agreement yet on the important issue of the procedure to be used for the adoption of panel reports. Nor was appellate review foreseen at that stage.



1. See section on Major changes in the Uruguay Round  back to text

2. BISD 26S/210  back to text

3. BISD 29S/13  back to text

4. BISD 31S/9  back to text

5. This is now excluded by Article 23.1 DSU. See the section on the Prohibition against unilateral determinations  back to text

6. BISD 36S/61  back to text



show previous page show next page

This interactive training module is based on the “Handbook on the WTO Dispute Settlement System” published in 2004. The second edition of this handbook published in 2017 can be found at here.

Chapters done:

show previous page show next page