The EU’s Economic Partnership Agreements sideshow is coming to a close

Jaime de Melo, Julie Regolo

17 September 2014



At the concluding days of the Doha negotiations in November 2001, WTO members signed a waiver extending the Cotonou Partnership Agreement to allow the Cotonou trade regime to be extended provided that it became WTO-compatible, that is through reciprocal Free Trade Areas.  Negotiations were to be concluded by December 2007 else the non-Least Developed Countries (non-LDCs) among the Africa, Caribbean and Pacific (ACP) former colonies would no longer benefit from preferential access in the EU market beyond the preferences granted under the Generalised System of Preferences (GSP). The Cotonou Partnership Agreement set out the following four ambitious core elements around which to build Economic Partnership Agreements (EPAs):

  • Differentiation: Keep differential and special treatment taking into account the level of development using asymmetry to benefit especially vulnerable, landlocked and small island economies;
  • Reciprocity: ensuring WTO-compatibility represents a radical departure from  previous EU-ACP relations whose rationale is to liberalise ACP markets, foster competition, better resource allocation and enhanced investment, both foreign and domestic;
  • Regionalism: (two-layer objective) only in exceptional circumstances would negotiations be envisaged with individual countries, the conviction being that regional integration for ACPs is the stepping stone towards a successful integration in the world trading system;
  • Development: EPAs are to be “economically meaningful, politically sustainable, and socially acceptable”.

Disappointments with the CA-EU EPA

By late 2007 only one ‘full’ EPA was signed between the EU and 15 CARIFORUM countries of the Caribbean, the CA-EU EPA. Though hailed as a “deep” agreement at the time, six years on, progress at implementation has been slow. Inherent difficulties in satisfying the regulatory and policy changes across a disparate group proved to be greater than expected. On the EU side, the hoped-for financing and technical assistance has fallen short (most CARIFORUM members will see a reduction in European Development Funds of 20% or more under the 11th EDF). And the anticipated benefits at the time of signing are not likely to materialise as the CA-EU EPA not only does not allow for the movement of unskilled workers but entertainers and cultural practitioners without a university degree will require certification to perform. In sum, so far, the deal has turned out to be ‘access’ without entry. Moreover, the EU has suggested that the necessary mutual recognition agreements necessary to access the EU markets in services be first negotiated among the 15 CARIFORUM members. The most-favoured-nation clause in the will also narrow the options for the Caribbean countries as they enter negotiations to renew bilateral deals with Canada and the US.  As of end 2013, only 6 members had ratified the EPA and only two had started tariff reductions.  (South Centre 2013).  This outcome supports the view that the EPAs were a ‘sideshow’ in EU trade policy all along as the EU pursued too many objectives and did not make any serious concessions to negotiating partners (Evenett 2010).

Since 2002, African LDCs benefitted from duty-free-quota-free market access under the Everything But Arms EU initiative so they had little interest in entering into an EPA, were it not for their belonging to a regional grouping including non-LDCs (the case for all regional groupings negotiating EPAs). Interim EPAs were signed in December 2007 preserving duty-free-quota-free access to the EU for non-LDC ACP countries, roughly amounting to €100 million per year of saved duties relative to GSP status. Probably the most significant market access obtained from the negotiations was the relaxation of some rules-of-origin requirements.1

Finalising the EAC-EU EPA

The sideshow is about to come to a close (the EU has set a 1 October 2014 deadline to avoid revoking preferential access to non-EBA members). On 14 July, Nigeria lifted its opposition, allowing 16 ECOWAS (Economic Community of West African States) members to finalise a full ECOWAS-EU EPA, albeit with a 5-year moratorium prior to start cutting tariffs.2 The next day, news came that a SADC-EU EPA is about to be signed and, hopefully, ratified by the 28 member bloc, preserving preferential access (beyond GSP) for fish, beef and sugar exports for Botswana, Namibia and Swaziland. Both are likely to run into implementation difficulties as the agreements call for reforms and regulatory changes beyond the implementation capabilities of the disparate group of beneficiaries, much as it turned out to be the case for the CA-EU EPA. The result could then be continued deterioration in Africa-EU political and trade relations to the benefit of China as predicted at the signature of the interim EPAs (Draper 2008).

This leaves the EAC, also about to conclude shortly a final deal, an interesting case since it is the most integrated regional grouping in Africa with a reasonably functioning customs union since 2005 (and a common market since 2010) much along the two-layer regional objective set out by the EU at the launch of the EPAs.  Has the closer degree of integration among members helped this grouping negotiate a more ambitious EPA? Our recent research suggests not.

First, is the unfounded fear that the negotiated tariff reduction on imports from the EU would lead to substantial revenue loss (these reductions are to take place over 19 years starting in 2014 for 80% of imports from the EU due to the exclusion list). This is because actual tariff revenue losses from tariff reductions are much less when estimated from customs data, approximately half those resulting from estimates based on statutory schedules (customs data was only available for Uganda and Rwanda but the same patterns would hold for the other EAC members and the other countries negotiating EPAs since they followed the same negotiation strategy of excluding high-tariff products which are those where exemption rates are highest). Thus, the time expended at making up the exclusion lists so that 80% of imports from the EU would enter with zero duties was misplaced as revenue losses would have been small even with full liberalisation (table 1 rows 2 and 3). A corollary is that welfare and efficiency gain estimates are also small (table 1 row 4).

Table 1. Revenue and welfare estimates from a full EPA with and without exclusion lists

Source: Melo and Regolo (table 4 and figure 2) from Customs data: * 2011 for Uganda, 2012 for Rwanda. Notes: Figures reported here are results for total imports; total tariff revenue and total tax revenue from all partners (including the EU) estimated over the entire period.

Second, the final stages of the negotiations show that 372 products are under negotiation for product-specific rules or origin and while the EAC-EU protocol (July 2013) is 78 pages long, the corresponding list of proposed rules of origin has 180 pages! Product-specific rules of origin are tougher for the EU for 70 products and only for 4 products for the EAC with disagreement only for 5 products. If rules of origin are necessary to prevent trade-deflection they should be kept simple, i.e. they should be ‘business-friendly’ rather than ‘business-owned’ since, in the end, the objective of granting market-access is to favour the partner.  One way to interpret this pattern is to conclude that the EAC has potentially obtained extra protection from these more stringent rules for 70 products. From an efficiency point of view, however, it is possible that rules of origin for imports from the EU will be sufficiently restrictive as to reduce the already low import response estimates (table 1, row 1). A more ambitious agenda would have pushed for simpler rules (e.g. a single economy-wide rule on the percentage of non-originating input in the final price).

Third, a more ambitious agenda would have included negotiating liberalisation in services even though the CA-EU EPA has run into implementation difficulties. In fact, in spite of more advanced integration, the EAC has followed the African linear regional integration paradigm with stepwise integration of goods, labour and capital markets.  In spite of deepening integration, trade in services has been largely missing in the EAC’s regional integration agenda, at least until recently. This is ironic since, in a world where the production chain is increasingly delocalised, a well-functioning regulatory environment and a relatively open services sector is required to attract the investment needed to provide the backbone services to compete in goods markets and participate in regional production chains.

Non-developmental EPAs

African EPAs will have focused on ‘shallow’ goods markets integration as the ‘rendez-vous clause’ in the interim EPAs calling for an extension of negotiations  beyond goods trade to cover services and other trade-related matters will not have been taken up. Rather, negotiation efforts will have revolved around the inclusion of a most-favoured-nation clause and the exclusion of export taxes in bilateral trade with the EU, both narrowing the policy space for African EPA members.

In sum, twelve years of negotiations will have ended up preserving EU market access for non-LDC African members, preferences worth about 1 percentage point after adjusting for the preferences the EU accords to other trading partners (WTO 2011).  Minimal reduction in tariffs on goods trade accompanied by extremely long time tables (often exceeding twenty years), exceptions lists excluding goods with high protection, and generous temporary protection lend little support to the objectives of the initial Cotonou Partnership Agreement objective of contributing to the deepening of regional integration in Africa along ‘development-friendly’ lines.


Draper, P (2008) “Africa-EU Trade Relations: Round number two!”,

Evenett, S (2010) “ A Future Agenda for EU Trade Policy as if the real world really mattered”,

Melo, J de, and A Portugal-Perez (2014). “Preferences and Rules of Origin in Apparel: African Exports to the US and to the EU”, World Bank Economic Review, 28(1), 74-98.

Melo, J de and J Regolo (2014) “Reflections on Finalizing and Economic Partnership Agreement: The EAC with the EU”, FERDI  WP 99

South Centre (2013) “The EU-CARIFORUM EPA: Regulatory and Policy Changes and Lessons for other ACP Countries”.

WTO World Trade Report (2011). The WTO and preferential Trade Agreements: From co-existence to coherence, Geneva.


1 See Melo and Regolo (2014). Rules of origin were relaxed for fisheries and for textiles and apparel as the EU followed in the footsteps of the Africa Growth Opportunity Act and adopted the single transformation rule for textiles and apparel (see Melo and Portugal Perez 2014). 

2 Nigeria had insisted on renegotiating key provisions including requiring that 181 tariff lines be reclassified within different categories of the offer. In the end, to obtain Nigeria’s support it was accepted that the Agreement would be revised every five years on the basis of results of an economic impact study and that the generous safeguard measures for ECOWAS’ common external tariff (external tariff up to 70% for a five year period since the start of the implementation thus giving each country plenty of room to protect domestic industries. See here.



Topics:  Development International trade

Tags:  EU, Economic Partnership Agreements, Africa

Senior Fellow, FERDI; Professor, University of Geneva; CEPR Research Fellow

Research Associate, Universidade Nova de Lisboa