Free-trade agreements: the other side of liberalization: What is developing countries' room to manoeuvre?
The degree and pace of liberalization necessary for a free-trade agreement (FTA) to comply with WTO rules (especially Article XXIV of GATT) remains an important discussion point in EPA negotiations. This article helps clarify the different interpretations of Article XXIV by analysing some 40 free-trade agreements notified to the World Trade Organization, including interim EPAs.
The number of bilateral FTAs negotiated between developed countries and developing countries has increased dramatically since 2000, in contrast to the WTO negotiations, which have tended to stall. Out of 158 FTAs notified to the WTO, 86 were concluded between developed countries and developing ones (far fewer involved Least Developed Countries). The liberalization commitments implemented in these agreements tends to go beyond the multilateral framework, since they often apply to areas which are still under WTO negotiation, and in some cases involve LDCs which are not required to liberalize their trade at the WTO. If we consider Economic Partnership Agreements (EPAs) as primarily development tools for the Africa, Caribbean and Pacific (ACP) countries, a greater flexibility might be needed, both in terms of the pace and scope of liberalization in FTAs. This requires maximizing the asymmetry between developed and developing countries authorized under WTO rules.
Article XXIV of GATT covers FTAs involving at least one developed country. It states that FTAs must cover substantially all trade and be implemented within a reasonable length of time. The Understanding on the Interpretation of Article XXIV adopted in 1994 specifies that a reasonable length of time should exceed 10 years only in exceptional cases. In the framework of the EPA negotiations, the European Commission (EC) considers that the transition period should not exceed 15 years and that substantially all trade should equal at least 90 percent of the total value of trade. If the EU opens 100 percent of its market, the EC argues that under the principle of asymmetry, ACP partners should liberalize at least 80 percent of their imports from the EU.
Our study of about 40 FTAs shows that, despite the Understanding, some existing FTAs demonstrate significant flexibility in the interpretation of Article XXIV, and have not been subject to dispute at the WTO. These count as useful precedents when negotiating FTAs.
A long and often asymmetrical transition period
A number of agreements (20 out of the 41 examined) have defined a transition period for liberalization that exceeds 10 years, even over 15 years for five of them. While most liberalization is reached within 10 years, extending the transition period enables some very sensitive sectors to adapt to increasing competition. Asymmetry in the transition period is often a flexibility granted among countries with different levels of development.
Flexibilities according to the sensitivity of products
The liberalization schedule often offers progressive liberalization according to the sensitivity of products. Sometimes, a moratorium is implemented to allow sensitive sectors additional time to adapt. In some FTAs, rendezvous clauses also enable countries to negotiate liberalization in stages (i.e., the liberalization goal indicated in the agreement only affects the first phase and can be less than 80 percent of tariff lines.)
Two examples illustrate this point:
In the Pakistan-China trade agreement notified under Article XXIV of GATT, the first phase is quite short (five years) and only concerns 36.4 percent of China's tariff lines (44.4 percent of imports in value over 2004-06) and 35.4 percent of Pakistan's tariff lines (30.3 percent of imports in value). No deadline has been settled for the second phase, although it should be discussed later and could lead to a liberalization of at least 90 percent of products.
The liberalization schedule for industrial goods in the EU-Tunisian Association Agreement is particularly complex: liberalization is immediate for a first list of products annexed to the agreement; then, liberalization is implemented over a five-year period for a second list of goods; for the third list, the transition period is 12 years and for the last list, covering the most sensitive products, liberalization will begin five years after the agreement has entered into force.
The degree of liberalization is not always high
Not all FTAs liberalize 90 percent of their value of trade, or tariff lines. In 30 per cent of the cases examined, less than 90 percent of trade is liberalized, and in 13 percent, it is less than 80 percent, with an asymmetry between developing and developed countries. The India-Singapore trade agreement is an example. India liberalized only 23.6 percent of tariff lines, representing 75 percent of its trade, allowing the country to shield a significant number of sectors it considers a priority.
Special treatment for the agricultural sector
Most agreements acknowledge the sensitivity of the agricultural sector, in particular for developing countries, and offer it a larger flexibility: lower degree of liberalization, longer transition period, and special safeguards. Some agreements partially or completely exclude agricultural products, since the latter are covered by specific agreements (not notified to the WTO) or are subjected to a more or less detailed rendezvous clause. For instance, in the EU-Mexico trade agreement, while the EC and Mexico respectively benefit from a three-year and seven-year period for industrial goods, it goes up to 10 years for agricultural products. Furthermore, Mexico liberalized only 29 percent of its agricultural tariff lines. It annexed an important list of agricultural products not liberalized; although they should have been negotiated for additional opening in 2003 (this deadline has not been respected).
Bilateral safeguard clauses
Various safeguards measures can be used, temporarily, to correct the negative effects of liberalization, such as threats to infant industries, food security (found in some Interim EPAs) or sensitive agricultural products. The last case is interesting for developing countries that have bound their tariffs and hence have no access to the special safeguard clause of the WTO Agreement on Agriculture. Nevertheless, these measures are generally only applied during the transition period.
Other possible flexibilities
The following flexibilities can also be used:
- More binding review clauses linked to benchmarks (for instance the number of persons living under the poverty line);
- Exceptions to national treatment to apply lower taxation on imported products;
- Increased protection on imported products benefiting from subsidies;
- Asymmetrical rules of origin more favourable to developing countries;
- A development component as an integral part of FTAs.
To conclude, it seems necessary to once again question Article XXIV. Developing countries can make proposals for flexibilities in the FTAs they negotiate with developed countries. The concept of asymmetry justifies this approach, and these flexibilities are an important means for adjusting to liberalization that goes beyond WTO requirements.
Two WTO legal texts can be used as a basis: on the one hand, the enabling clause which states that "contracting parties may accord differential and more favourable treatment to developing countries, without according such treatment to other contracting parties", and, on the other hand, the General Agreement on Trade in Services (GATS), which allow some flexibility to developing countries depending on their global and sectorial and subsectorial development level.
At a time when the West Africa and Central Africa EPA negotiations have stalled over provisions that would provide more flexibility, the analysis of the FTAs notified to the WTO reveals that there is room to manoeuvre. Indeed, the precedent set in some FTAs is a basis for understanding and accepting the ACP's request to liberalize 60 percent - and not 80 percent - of their market or to benefit from a 25-year transition period.
This article is based on a study * carried out by Gret with the support of the French Development Agency. Based on the inventory of 158 FTAs in the WTO database on Regional Trade Agreements, information for 41 of them was immediately available and these agreements were thus analysed. The full report is available on the Gret Website: http://www.gret.org. The views expressed in this article are those of the authors.
*Rolland J.-P., Lagandre D., Alpha A., Étude comparative des accords de libre échange impliquant des PED ou des PMA (Comparative study of Free-Trade Agreements involving déveloping countries and LDCs), October 2009.
See Lui, D. and S. Bilal (2009), "Contentious issues in interim EPAs", ECDPM Discussion Paper 89, www.ecdpm.org/dp89 and the recent El Hadji Diouf's article, TNI 8(7), September 2009, www.acp-eu-trade.org/tni.