The ACP Advantage: Interpreting GATT Article XXIV and Market Access Implications for EPAs
The Economic Partnership Agreements between ACP countries and the European Commission must comply with Article XXIV of the General Agreement on Tariffs and Trade (GATT), which governs regional trade agreements concluded by WTO members. Article XXIV stipulates that ‘free trade areas' must eliminate duties on "substantially all the trade" within a "reasonable length of time," yet these terms remain loosely defined in the WTO. West Africa and the European Commission hold divergent views on the definition of these two terms. This article highlights the legal arguments in support of West Africa's position.
One of the most important and contentious issues in ACP-EU EPA negotiations relates to the interpretation of Article XXIV of the General Agreement on Tariffs and Trade (GATT) and determining the market access offers of the parties. The EU stands by its position of a minimal trade coverage rate of 80 percent in ACP regions, over a 15-year liberalisation schedule. Some regions, such as Western Africa, argue that a 60 percent coverage rate along with a 25-year implementation period would not be WTO-incompatible. The following legal analysis attempts to prove it.
The term "substantially all the trade" in WTO/GATT law and practice
Pursuant to Article 3.2 of the Memorandum of Understanding on the WTO's Dispute Settlement Understanding (DSU), the clarification of existing rules of the WTO agreements should follow the customary rules of interpretation of public international law. These rules are explained in Articles 31 and 32 of the Vienna Convention on the Law of Treaties of 1969. They make reference to good faith, ordinary meaning, context, with subsequent agreements and practices, and preparatory work. Most of these tools have been used for a proper understanding of Article XXIV of GATT in vain.
Interpretation tools for international treaties noted earlier have been applied to Article XXIV of GATT in vain. While WTO jurisprudence has tried to follow the Vienna Convention instructions recurring to the preparatory work, the results are not satisfactory:
"We have... analyzed in detail Article XXIV negotiations history. We observe that this article's title is not perfectly clear and has been subject to many divergent - sometimes conflicting - views between the signing parties, the Members and the authors.
The WTO has also tried to question the members on their practice. But no additional information was obtained due to the absence of an agreed upon practice between members.
"Examining WTO/GATT practice clearly shows that... there has been no consensus nor agreed practice in relation to Article XXIV of GATT"
Furthermore, the notion of "substantially all the trade" referred to in Article XXIV.8 is not clearly defined and the meaning of the word substantial - which is intended to define the liberalisation threshold of regional agreements - has never been made explicit. The WTO Appellate Body itself came to the following conclusion:
"Neither the GATT Contracting Parties nor the WTO Members have ever reached an agreement on the interpretation of the term "substantially" in this provision."
All things considered, the only rule of interpretation used by jurisprudence and which seems to apply to Article XXIV.8 is the implicit reference to the context and circumstances of the treaty's conclusion. The Special Group opened a breach for a new reading of the article in a different context.
"We know... that the economic and political realities that existed at the time of Article XXIV drafting have changed and that the scope of regional agreements is now much wider than it was in 1948"
By using a-contrario reasoning, the Special Group suggests to read and apply Article XXIV by taking today's economic and political realities into account. This stance includes the new dimension of RTAs between developed and developing countries and posits an interpretation which legitimises an important asymmetry in EPAs.
After all, one of the few consensuses in interpreting the concept of substantially all the trade is the primacy given to the quantitative approach. But this primacy does not solve the problem by itself. Thresholds must be defined. Nevertheless, WTO/GATT texts give no figure, nor do the jurisprudence or the practices agreed in the system. Moreover, jurisprudence recently said the following on that question.
"It is clear that 'substantially all the trade‘ is not the same as all the trade, and also that 'substantially all the trade' is something considerably more than merely some of the trade."
If the text refers to neither all the trade nor just a part of it, then what does substantially all trade actually mean? Members' practice indicates that a regional trade agreement liberalising 80 percent of goods would not have compatibility problems with WTO law. Therefore, assuming that liberalisation of 80 percent of trade has been accepted by the EC in the WTO context, this implies that the same percentage would be acceptable to both the EC and the WTO membership in the context of the EPAs. The question remains: how to make up 80 percent of the trade in EPAs. Two interpretations are possible. The first one is that each party shall liberalise at least 80 percent of its goods. In this case, this threshold would be a consolidated limit above which the EPA would no longer be WTO-compatible. This interpretation seems to be incorrect. Article XXIV.8 clearly states that substantially all trade to be liberalised relates to "products originating from the constituent territories of the free-trade area".
There is clearly no attempt to share the burden of liberalisation between parties. And the numerous regional agreements signed under the aegis of Article XXIV almost never imply liberalisation evenly shared between the parties. Hence, the agreed 80 percent will be considered as a weighted average incorporating the parties' efforts. The fact is the EU is going to liberalise up to 100 percent of its trade. From that point, to reach the weighted average of 80 percent of liberalised trade, nothing but their own unilateral political will should force African countries to go beyond a 60 percent market opening. This stance would not infringe any WTO rule nor would it be an obstacle to the legality of the EPA. We cannot reasonably say that 80 percent of trade liberalised in a regional agreement does not go beyond "merely some of the trade".
Reasonable length of time and exceptional circumstances in WTO law
In the same vein, the concept of a reasonable length of time allowed for regional agreements' implementation period is not explicit in Article XXIV of GATT.It is stated that the reasonable length of time for regional agreements should exceed 10 years only in exceptional cases and that in cases where member parties to an interim agreement believe that 10 years would be insufficient they shall provide a full explanation to the Council for Trade in Goods of the need for a longer period. As well as for the substantial part of the trade, it is the concept of exceptional cases that can be subject to interpretation.
The interpretation of Article XXIV in favour of exceeding a 10-year period only in exceptional cases has been followed by a legalism period before the members adopted more permissive stances regarding calculation of time limits on regional agreements.
"Regarding the RTAs which entered into force in the latter half of the 1990s, only in rare cases did transition periods exceed ten years." On the contrary, regarding the many RTAs recently concluded, we observe that transition periods greatly exceed 10 years. These cases become the rule rather than the exception"
In spite of this, all regional agreements concluded with an implementation period beyond the rule have been considered WTO-compatible. The idea of conformity with WTO law no longer guides regional agreements, but rather the consideration of the parties' interests. As members have not reached agreement on this, practice provides many interesting cases and orientations, as the table below shows:
It appears that a large majority of free-trade areas between developed and developing countries are asymmetrical and that a systematic time limit of 20 years is not exceptional (e.g., Thailand's agreements with Australia and New-Zealand). Moreover, even agreements between developed countries such as the Australia/US FTA define an 18-year implementation period without exceptional circumstances. Finally, we observe that the longest time period belongs to Morocco in its agreement with the US. It's a 24-year period and it is practically similar to the 25-year period asked for by African countries for their EPA. Yet, this agreement has been notified to the WTO. From that example, what can justify such a long time period, which is obviously a special and differential treatment for developed countries and would not be applicable in the case of an EPA involving some of the poorest Least Developed Countries in the world? Clearly, this alone constitutes an exceptional case which justifies that African countries may benefit from a 25-year time period, just one year longer than the Morocco/US agreement.
Dr. El Hadji A. Diouf is the EPA and Regionalism Programme Manager at the International Centre for Trade and Sustainable Development (ICTSD).
Report of the Special Group, Paragraph 9.97
Ibid. Paragraph 9.166
Turkey - Restrictions on Imports of Textile and Clothing Products. Report of the Appellate Body, 22 October 1999 (WT/DS34/AB/R), Paragraph 48
Turkey/Textiles, Special Group, Paragraph 9.97
Turkey/Textiles, Appellate Body, 22 October 1999 (WT/DS34/AB/R), Paragraph 48
Report of the WTO Secretary, 2002 (WT/REG/W/46, page 22