ACP-EU Trade and Aid Co-operation Post-Lomé IV:
Chapter 3. Options for Trade Co-operation
Libreville, Gabon, 6-7 November 1997
Renewal of the Lomé Convention
The Generalised System of Preferences
Viable Options for Negotiation
The Commission Green Paper exaggerates both the difficulty of obtaining a renewed WTO waiver for the Lomé Convention after 2000 and the possibility of challenge from non-ACP countries. At the same time, the ACP States need to carefully consider whether or not it is in their interests to extend Lomé trade preferences on industrial products to a wider group of developing countries to enable the Convention to be more easily defended in the WTO. The advantages and disadvantages of a renewal of the Lomé Convention and the alternative of a Free Trade Agreement with the EU are considered. The analysis emphasises that an FTA is inferior to trade liberalisation bound in the WTO by the ACP countries unless the FTA enables the ACP States to go beyond WTO disciplines and obtain concessions from the EU not otherwise obtainable and sufficient to compensate for the costs associated with an FTA. Possible additional concessions are discussed and considered to be difficult to obtain in the light of the EUs 'Europe Agreements and free trade agreements with the Mediterranean countries. The Green Paper proposal for regional free trade agreements is considered unworkable given the heterogeneity of existing regional groups. It would also weaken the negotiating position of the ACP Group and, by creating a 'hub and spoke system of trade, create barriers to trade between ACP countries. Reliance on the GSP and the enhanced GSP (for the least developed, Andean and Central American countries) would represent a significant loss of security of preferences and, especially in the case of the standard GSP, a reduction in the margin of preferences compared to the Lomé Convention. The conclusion is that the ACP should concentrate either on a renewal of the Lomé Convention, but focused on a limited range of objectives and in particular on increasing the export capacity of the ACP States and with improved terms of access; or an 'umbrella agreement which enabled ACP countries to select the appropriate form of trade agreement with the EU, combined with a focused and well integrated package of aid and technical assistance. Variations in trade arrangements within the umbrella agreement, however, should be kept to a minimum, otherwise additional barrier to trade will be created between ACP countries.
There are basically five options available to the ACP countries in relation to trade with the EU:
Renewing Lomé as a non- reciprocal agreement.
Negotiate free trade agreement(s) with the EU.
Limit trade co-operation to the GSP (because of incompatibility of Lomé with GATT rules) and have a separate agreement covering technical and financial co-operation.
A combination of any or all of the previous options under an 'umbrella Convention.
Trade liberalisation by ACP States and bound in the WTO (included because it affects the evaluation of the other options)
3. 2 Renewal of the Lomé Convention
The challenge to the Lomé Convention
In order to discuss this question we have to set out the reasons why, after twenty-two years of operation, the future existence of the Convention after the year 2000 is being questioned. The contention is that it is incompatible with the basic principles of the General Agreement on Tariffs and Trade (GATT) since it discriminates between developing countries (violating Article I, the principle of non-discrimination) and offers non-reciprocal preferences (violating Article XXIV, on free trade areas and customs unions).
First, it is important to remember the origins of Lomé. Before joining the EC, the UK offered Commonwealth preferences to its ex-colonies while the EC offered preferences to the ex-colonies of the six original member states in Sub-Saharan Africa under the Yaoundé Conventions I and II. The latter preferences were on a reciprocal basis with the EC. The accession of the UK in 1973 required a new agreement. The US had previously objected to reciprocal preferences being required by the EC because of their potential to exclude the US from important markets and this had resulted in the Casey-Soames(1) understanding. Under this, the US would not challenge EC trade agreements while the EC would not seek reverse preferences with the developing countries nor negotiate preferential trade agreements outside of the African Caribbean and Pacific group of countries and the Mediterranean countries.
In presenting the first Lomé Convention to the GATT(2), the EC claimed that Lomé I had not been made a reciprocal agreement because of the development needs of the ACP countries, and for this reason was compatible with the principles of Part IV of the GATT. Part IV of the GATT on Trade and Development was added in 1965 (just after the establishment of UNCTAD in 1964) and contains no legally binding obligations on members but sets out principles and objectives, commitments, and areas of joint action, in relation to trade and development. For example, Part IV Article 36.8 states that 'the developed contracting parties do not expect reciprocity for commitment made by them in trade negotiations to reduce or remove tariffs and other barriers to the trade of less developed contracting parties. The EC also argued that Lomé should be considered in the light of Article XXIV (on Free Trade Areas and Customs Unions) and Article XXXVI (the principles and objectives of Part IV) taken in conjunction.
Other members of the GATT working party set up to consider Lomé I and subsequent Conventions have questioned this view and pointed out that Part IV referred to special treatment for all developing countries. In addition, the working party on Lomé IV reported the view of several members that the Convention would only be GATT compatible if a waiver were obtained to Article I (principle of non-discrimination) under Article XXV, which states that:
'In exceptional circumstances not elsewhere provided for in this Agreement the Contracting Parties may waive an obligation imposed upon a contracting party by this Agreement; Provided that any such decision shall be approved by a two-thirds (now three-quarters) of the votes cast and that such majority shall comprise more than half of the contracting parties. The Contracting Parties may also by such vote ..... (ii) prescribe such criteria as may be necessary for the application of this paragraph (Art.XXV.5).
Previous working parties had also failed to reach agreement on the Lomé Convention but this opinion was given additional impetus both by the objective of reinforcing the GATT as the basis for a rule based system of international trade and by the two disputes panels set up in 1993 to consider the bananas issue (for further details, see Chapter 12). This arose because high cost ACP exporters of bananas (relative to Latin American producers) required and obtained preferential access to their traditional markets in the UK, France and Italy under a special Protocol of the Lomé Convention, and the application of Article 115 of the Treaty of Rome (permitting trade restrictions by individual EC member states) by these three member states. The completion of the single European market in 1992 meant that Article 115 could no longer be applied and it was therefore impossible to segment the EC market in this way. To implement the Protocol on bananas, the EU instituted Community-wide import restrictions procedures in July 1993, and this resulted in a request to the GATT in February 1993 by Colombia, Costa Rica, Nicaragua, and Venezuela to establish a disputes panel to examine the EU import regime for bananas policy. The EU sought to justify the banana import regime on the basis of its obligations under the Lomé Convention, and so this inevitably led the panel to consider the broader issue of the compatibility of Lomé with the GATT, as well as specific issues concerning the import regime (which the disputes panels in 1993 and 1997 also considered incompatible with the GATT and GATS). The GATT panel found that the arrangement for bananas was not GATT consistent, firstly, because Part IV was not intended to subtract from other GATT obligations (for example Article I), and secondly, the arrangement was not covered by Article XXIV, or Article XXIV in conjunction with Part IV.
It should be remembered (see Chapter 1) that the findings of the WTO disputes panels now have much more authority than in the past. If member countries involved in the dispute cannot agree with the findings of the panel, the dispute goes to a final appeal body whose decisions are not open to challenge and whose report is automatically adopted unless the Disputes Settlements Body (DSB), which administers the agreement on disputes rules and procedures, decides by consensus not to adopt the Appellate Body Report. Since, in practice, this would undermine the recent strengthening of the authority of the WTO and its disciplines, this is unlikely to occur and so the findings of the disputes procedures have very much greater weight than in the past.
In the light of these events and the change in attitudes in the WTO towards strengthening the GATT as the foundations for a rule-based system of international trade, the EU sought and obtained in December 1994 a waiver from Article I.1 under the provisions of Article XXV until 29th February, 2000.
Implications for the renewal of the Lomé Convention
These events form the basis for the view in the Commissions Green Paper which calls into question the future existence of Lomé. It can be argued, however, that this pessimistic view exaggerates and misinterprets the purpose and significance of the dispute in the WTO. The EU defence of Lomé was undermined in 1993 because one of the important advances of the Uruguay Round was to strengthen the General Agreement as a means of providing a rule-based system of international trade. The EU defence of Lomé on the basis of the combination of Article XXIV with Part IV was not credible. Article XXIV applies to customs unions and free trade areas and since Lomé was a non-reciprocal agreement, it clearly fails to qualify under this heading. Equally, Part IV applies to all developing countries and cannot be used in a discriminatory fashion. The Convention had to conform to GATT rules by obtaining a waiver under Article XXV.5 from the provisions of Article I in the same way in which the US obtained waivers for its preferential agreements with the Caribbean and the Andean countries and the Former Trust Territory of the Pacific Islands, and Canada obtained for its preferential agreement with the Caribbean countries. Equally, there would appear to be good reasons to expect the EU to obtain a further waiver for the Convention for the period 2000 to 2010 if it so wished. In this context it is worth noting the history of the US Caribbean Basin Initiative.
The Caribbean Basin Initiative (CBI) originated in the Caribbean Basin Economic Recovery Act of 1984 and was implemented by the US under a waiver to Article 1 of the GATT on 15th February, 1985. The Act allows the US President to grant duty-free treatment for certain goods imported from particular countries and territories in the Caribbean Basin. The CBI was intended to be a temporary programme, set to expire in 1995, but due to the success of programme, it was made permanent in 1990 and in November 1995 a further waiver was granted until December 2005. This waiver allows the US to continue to provide duty-free treatment for imports of a list of products from the Caribbean Basin countries without being required to extend the same duty free treatment to similar products from any other WTO member. The waiver, among other conditions, provides for consultation to take place upon the request of a WTO member with respect to any difficulty or matter that may arise as a result of the implementation of the trade related provisions of the Caribbean Basin Economic Recovery Act. (WTO Annual Report 1996).
The CBI demonstrates that the objections to the Lomé Convention were not against its discriminatory and non-reciprocal provisions, per se, but against the way in which the EU had tried to justify them in the WTO. Indeed, the recent disputes panel on bananas has accepted that the waiver for the Lomé Convention allows the EU to provide preferential treatment for the products originating in the ACP States as required by the relevant provisions of the Fourth Lomé Convention and their objections related to specific aspects of the import regime for bananas. In order to strengthen the rule bound systems of international trade, the EU had to conform to WTO disciplines in the same way as the US and Canada had to conform for their non-reciprocal preferences, namely, by obtaining a waiver from the provisions of Article I. Once this was obtained, the issue was settled.
Advantages of renewal of Lomé
First, unlike the GSP, Lomé is a contractual relationship lasting for ten years under Lomé IV and based on the principle of a partnership. This is given effect by the institutional arrangements for the joint management of the Convention, the policy dialogue between the EU and the ACP States, and the mechanisms available to negotiate improvements and extensions to the Convention.
Second, the trade arrangements are, in total, superior to any of the other preferential agreements of the EU. All industrial products are allowed into the EU 'free of customs duties and measures having equivalent effect (Art.168) and in practice (apart from a short-lived attempt by Britain and France to impose VERs on Mauritius in 1979) the ACP countries have not been subject to the use of, or threats of, non-tariff measures restricting imports of industrial goods. The trade provisions of the Convention, in conjunction with the Protocols, also provide the most generous concessions on CAP products. Rules of Origin are also superior in Lomé in that they allow cumulation of origin between all of the ACP States and provide for derogations from the rules of origin (though only under onerous conditions which mean that they are rarely used).
Third, the financial and technical provisions are the subject of greater negotiation and discussion both in their design and implementation than other aid programmes and include features such as STABEX , and joint institutional arrangements such as the Centre for Development of Industry and the Trade Development Project, which are unique to the Convention.
Fourth, renewal of Lomé would preserve as much as possible of the 'acquis obtained in successive negotiations of the Convention from Lomé 1 Lomé IV, both in terms of its contractual relationships and in terms of the individual components of the twelve areas of co-operation in the convention.
Fifth, being non- reciprocal, the renewal of Lomé would leave the ACP free to pursue their own trade and industrialisation strategy on a multilateral basis without any trade distortions or other adverse effects arising from granting preferences to the EU.
Sixth, renewal of Lomé would maintain the momentum for building the cohesion of the ACP as a group, contribute towards reducing tensions between the member countries, and enable them to most effectively negotiate with the EU and other developed countries, as well as within international bodies such as the WTO.
Disadvantages of renewing Lomé
First, the European Commission's Green Paper states that 'one of the shortcomings of this option is that being differentiated (ie not offered to all developing countries), it will continue to require a WTO waiver which needs to be reviewed every year. As a result, the security of the preferences, one of the main assets of the Lomé trade package, will be severely undermined (p.40). This overstates the situation. Once a WTO waiver is granted, it lasts for the period agreed (in the case of the Caribbean Basin Initiative this will last for the agreed 10 year period 1995 to 2005) and the annual reviews which are required as part of the waiver process are simply a reporting formality. The Green Paper may be confusing the formality of the annual reviews of a waiver with the one-off 'Extension of the Waiver which was obtained in October 1996. This 'extension was required as a result of the 'Uruguay Round Understanding in Respect of Waivers of Obligations under the General Agreement on Tariffs and Trade 1994 which meant that all waivers in effect on entry into force of the Marrakesh Agreement lapsed two years thereafter (ie on 1 January, 1997).
Renewal of Lomé could of course, be challenged by a WTO member country, but which countries would gain by such a process? The US is most unlikely to challenge, since they have relied on such a waiver for the CBI since 1984. This will expire in 2005 but President Clinton has already proposed improvements to the CBI as partial compensation for the reduction of EU banana preference as a result of the WTO panel ruling. Objection to the renewal of the Convention would also push the ACP countries towards a free trade area with the EU, and the US has long opposed the EU requiring reverse preferences, since these would discriminate against US exports to the ACP. Similarly, by voting against the renewal of the waiver for Lomé, other industrialised countries would also effectively be pushing the ACP countries towards a WTO compatible free trade area under Article XXIV and this would inevitably produce discrimination against their exports to the ACP. The success of the Latin American banana exporters in challenging the EU may, it is thought, encourage these countries to challenge a renewal of Lomé in order to obtain compensatory trade measures from the EU. But the damage to their exports would have to be proved and (unlike the banana case), it is difficult to see how such a case could be substantiated. Also, as discussed by Hallam and Preston (1997), the bananas challenge had much more to do with the global strategy of a small group of multinational companies than the trade interests of the particular Latin American countries concerned. A challenge by the Latin American countries to a Lomé waiver could also be viewed unfavourably by the US and act as an obstacle to their growing trade with the US (for the reasons given above) and with the EU, while producing little or no offsetting gain in their exports either to the ACP or to the EU.
Second, the Green Paper also emphasises the adverse effects of what it assumes would be the diminished security of preferences of a renewed Lomé Convention on foreign direct investment (FDI) in the ACP countries. The above arguments, however, cast doubt on the assumption of diminished security of preferences. If, however, the counter-factual is a free trade agreement with the EU then, provided this was compatible with Article XXIV (see the following section), such an agreement could provide greater security of access than Lomé and thus make the ACP a more attractive destination for foreign direct investment (through the investment creating and investment diversion effects of preferences). Article XXIV, however, defines a free trade area as covering 'substantially all trade yet the EUs free trade agreements invariably exclude or severely limit concessions in agricultural products in order to maintain the integrity of the CAP. As explained in section 3.3, and in chapter 7, the Uruguay Round has radically changed the likelihood of the EU being able to maintain such agreements without being successfully challenged in the WTO, in which case a free trade agreement would be no more secure (and perhaps less secure) than a renewed Lomé Convention, especially if the trade provisions of the latter covered a wider group of developing counties than the existing ACP and was therefore less discriminatory between developing countries. Also, as discussed in Chapter 6, the destination of FDI is generally determined by much more important factors than reciprocal or non-reciprocal preferences.
Third, Article 174.2(a) of the Lomé Convention states that 'in their trade with the community, the ACP state shall not discriminate among the member states and shall grant to the community treatment no less favourable than most favoured nation treatment. This raises the issue of how the EU would react to an ACP country or group of countries entering into a customs union or free trade arrangement with a developed country (eg Caribbean countries with the US and Canada; Pacific countries with Australia, New Zealand, or Japan. The use of the term 'most favoured nation suggests that the agreement need only treat EU goods in accordance with Art. I.1 of the General Agreement, but as stated in the Green Paper, at a political level it seems likely that the EU would challenge the continued free entry for imports from an ACP country which offered preferential access for other industrialised countries exports, while denying such access for imports from the EU. Such ACP countries would then have to choose between remaining within Lomé, or leaving the Convention and relying on GSP preferences in order to join another regional arrangement involving reciprocal preferences with a non-EU developed country.
Fourth, the most important question is whether a continuation of the Convention would enable the ACP countries to participate more fully in the world economy. This report has emphasised the need for ACP countries not only to diversify by product but also by market, particularly towards the rapidly expanding market in South and South-East Asia, Latin America, and the Central and East European countries. The continuation of the Lomé arrangements could be regarded as a 'soft option, continuing with familiar relationships but at the expense of maintaining a dependence on a slowly growing EU market and a continued level of product and market concentration, which would be detrimental to the growth and diversification of ACP exports. The 'soft option of renewal of the Convention may be even more detrimental to the ACP if it induced them to neglect necessary trade liberalisation and structural adjustment of their economies.
Fifth, the negotiation of successive Conventions is generally recognised to have led to a proliferation of policy instruments at the expense of development strategies (see Chapter 6). This has reduced the effectiveness of technical and financial assistance and created a lack of co-ordination between different aspects of the Convention. In the context of this study, it has led to an absence of strategies aimed at increasing the export capacity of ACP countries and enabling them to take advantage of the preferences offered. The renewal of the Convention in its present form would perpetuate these potential problems. In this sense the status quo is an inferior option, but there is no reason why the existing (or improved) market access provisions of the Lomé Convention could not be combined with much more focused and co-ordinated aid provisions.
The trade creation and trade diversion effects of an FTA
Multilateral or unilateral trade liberalisation offers the possibility of efficiency gains, especially where it involves the removal of indiscriminate protection. A free trade area, by removing barriers to imports from the partner country (EU) also offers efficiency gains (trade creation), but by removing barriers to trade against the partner country (EU) but not against third countries, it also leads to the replacement of imports from cheaper third countries with dearer imports from the partner country (trade diversion). In this important sense, an FTA is always inferior to unilateral or multilateral trade liberalisation. The size of these trade creation and trade diversion effects on an ACP State critically depends on the structure of the economy and the patterns of trade, and few valid generalisations can be made about any country or group of countries. What we can do is outline the fundamental factors which would determine these effects in an EU-ACP free trade agreement.
The trade creation rise in imports from the EU will be larger the greater:
(a) the initial level of imports from the EU;
(b) the pre-agreement level of tariffs plus the tariff equivalent of the non-tariff barriers to trade removed as a result of the agreement;
(c) the extent to which imports from the EU substitute for ACP produced goods in production and consumption. For some (small) ACP countries at higher levels of economic development (eg Mauritius?) this could be quite high, whilst for other ACP countries most goods may be of a price, quality and design specific to a low income and less industrialised economy, or may be protected by poor channels for marketing and distribution, and so substitution between domestic production and imports is low.
(d) the preference for imported goods from the EU in production and consumption.
It will be appreciated that the estimation of (b) and particularly (c) is a matter of judgement rather than precise measurement, yet the interaction of the three parameters will substantially alter the magnitude of the trade creation rise in imports.
Table 3.1ACP Imports from the EU
(a) Greater than 40% of total imports
Angola, Burundi, Cameroon, Cape Verde, C. African Republic, Chad, Congo, Cote dIvoire, Ethiopia, Gabon, Gambia, Ghana, Guinea-Bissau, Madagascar, Mauritania, Niger, Nigeria, Senegal, Sierra-Leone, (S. Africa), Sudan, Uganda, Tanzania, Zaire.
(b) 20%-40% of total imports
Bermuda, Burkina-Faso, Dominica, Grenada, Kenya, Kiribati, Liberia, Mali, Mauritius, Rwanda, St. Kitts-Nevis, St. Lucia, St. Vincent and Grenadines, Seychelles, Somalia, Surinam, Togo, Trinidad and Tobago, Vanuatu, Zambia, Zimbabwe.
(c) Less than 20% of total imports
Bahamas, Barbados, Belize, Dominican Republic, Fiji, Guyana, Haiti, Jamaica, Papua New Guinea, Samoa, Solomon Islands, Tonga.
Source: UNCTAD 'Handbook 1995
Table 3.1 summarises the substantial differences in the share of imports from the EU in total imports, even for countries in the same regional grouping. The countries in category (b) import around one-quarter to one-third of their goods from the EU while those in category (c) import more than twice this proportion from the EU.
Table 3.2Simulations of trade creation increases in imports
| Initial share of imports from the EU |
| 20% | 40% | 60% | |
| e = -0.5, t = 10% | 0.9% | 1.8% | 2.8% |
| e = -1.5, t = 20% | 5.0% | 10.0% | 15.1% |
| e = -3.0, t = 30% | 13.9% | 27.7% | 41.6% |
e = price elasticity of demand for imports
t = trade weighted average tariff + tariff equivalent of non tariff barriers removed by FTA
The simulations in Table 3.2 suggest that even with a reasonable degree of substitutability of imports for domestic production and preference for EU imports (e = -1.5) and protection (t = 20%), the once and for all trade creation rise in imports is quite modest even with a high initial share of imports from the EU. But, with a higher level of import protection (t = 30%) and a doubling of the import elasticity of demand (e = -3.0), imports increase substantially, depending on the initial share of the EU in imports.
The increase in imports from the EU as a result of lower cost imports from third countries being replaced by higher cost imports from the EU (trade diversion) depends on the extent to which ACP importers are willing and able to substitute in favour of EU suppliers. We could expect this to be lower for Caribbean and Pacific countries than for African countries, given their respective patterns of trade and distance from the EU. Even in the case of Sub-Saharan Africa, however, the small size of the domestic market in many ACP countries, together with the costs of servicing the market, can also be expected to reduce the extent to which EU exporters seek to displace exporters from the rest of the world. Also, EU goods may not compete with imports from the rest of the world, especially where these consist of imports from other developing or newly industrialising countries.
The conclusions from the analysis are:
Even where a decrease in import prices leads to a more than proportional rise in the volume of imports (e = -1.5), levels of protection are significant (t = 20%) and there is a high degree of dependence on the EU for imports (60%), the once and for all trade creation rise in imports would appear to be manageable, both in terms of the necessary re-allocation of domestic resources (adjustment in the face of increased competition from the EU) and in the balance of payments.
Increased imports from the EU as a result of trade creation only appear to be substantial where import barriers are high (30% or more) and EU goods compete to a significant degree with local production (e = -3.0 or more).
Given the dissimilarity of the economies of the EU and the ACP, increased imports from the EU as a result of trade diversion will probably be greater than those resulting from trade creation. This does not create an adjustment cost to the ACP countries but does create a loss of welfare, since a lower cost source of supply (rest of the world) is being replaced by a higher cost source (the EU). The magnitude of the trade diversion effect fundamentally depends upon the elasticity of substitution and this can be expected to be lower the greater the proportion of imports from other developing countries, the smaller the size of the ACP market and higher the costs of servicing the ACP market.
The increase in imports from the EU as a result of trade creation and trade diversion will lower Government revenues from import duties which, for ACP countries such as Benin, Madagascar and Mauritius account for 37% to 42% of Government revenues and for a number of other countries account for 20% or more of total revenues. Compensating for the loss of revenue as a result of a FTA with the EU by increasing duties on imports from the rest of the world will not only increase distortions in the allocation of resources in the economy and lower growth rates, but will exacerbate the problem because it will raise the trade diversion component of the increase in imports from the EU. It may appear counter-intuitive, but lowering tariffs and other duties on imports may be expected to increase revenues since it will reduce the extent of trade diversion and increase dutiable imports and, if import elasticities of demand are greater than unity, further increase revenues.
The case for a free trade area with the EU
Given the costs of a free trade area with the EU, such an agreement has to offer advantages not otherwise obtainable. These are largely potentially strategic or dynamic advantages.
First, as discussed in Chapter 1, the Sub-Saharan African countries have generally declared bound tariffs at very high levels and in many cases well above actual applied levels. A free trade area, it is argued, would help to overcome this resistance to essential trade liberalisation and complement structural adjustment policies acting on the supply side of the economy. For example, a free trade area would enable governments to overcome the resistance of special interest groups seeking to maintain existing levels of protection. It is also argued that by 'locking in structural adjustment reforms, an FTA with the EU would enable the ACP to obtain the same benefits as multilateral trade liberalisation. However, this assumes that the FTA leads to imports entering the ACP at world prices and if in fact EU exporters use their market dominance to charge substantially above world prices, (Yeats, 1990, shows that EU exporters charge 20% to 30% above world prices) then an FTA will largely transfer import revenues to EU exporters, with little in the way of efficiency gains to the ACP countries. Also, if trade diversion costs out-weigh the trade creation gains, then there will be a net loss of welfare. Taken together, these factors could decrease the credibility of the reforms, while the negotiation and implementation of an FTA could very well divert attention and resources from the more important area of general trade liberalisation.
Second, a free trade agreement could cover areas not covered by the WTO or where WTO disciplines are relatively weak for example:
• Guaranteeing market access against contingent protection (anti-dumping, use of safeguards, etc);
Offering a wider coverage of protected products which, compared to the Lomé Convention, would be agricultural products;
'National treatment of foreign investment and business services;
EU Investment promotion schemes, for example similar to the AL-Invest, and EC Investment Partners (ECIP) schemes for Latin America, and the investment promotion schemes in the Europe Agreements. EU Investment guarantees on foreign direct investment could be given in exchange, for example, for the right of establishment of EU firms and 'national treatment for EU firms;
Additional assistance in complying with the regulatory and administrative requirements of the EU, for example, health and safety regulations, phyto-sanitary regulations, industrial standards, etc, combined with guarantees (underwritten by the option of independent arbitration binding on both parties in cases of dispute) that compliance will ensure that these regulations will not be used as a non-tariff barrier;
Agreement covering the liberalisation of trade in cross-border services enabling the ACP both to obtain these services on more favourable terms and to regulate the activities of EU service providers;
Competition measures which could seek to prevent anti-competitive behaviour by EU firms.
Third, the free trade agreements invariably include financial and technical resources specifically aimed at increasing export capacity and export promotion opportunities. The ACP could seek to utilise these resources, not only for export promotion in the EU market, but also in markets in which the EU already has preferential agreements, for example, the Central and East European Countries (CEEC).
Fourth, the guarantee of access both to the ACP and EU market may induce investment creation and investment diversion by EU and non-EU firms. For example, Mauritius has benefited from FDI from firms in the EU, Hong Kong and a variety of other countries. Some Caribbean countries have attracted significant amounts of US investment (investment creation) attracted by the US, and to a lesser extent, EU preferences.
Fifth, the Uruguay Round undoubtedly strengthened the rule of law in international trade relations, but a proliferation of regional integration arrangements could still undermine the multilateral trading system. A recent example which has concerned the EU has been the case of Mexico which, faced with a financial crisis and the impossibility of increasing tariffs on imports from the US because of the NAFTA agreement, increased duties on imports from third countries, including the EU. As a result of this and the deepening of preferences between the ASEAN countries, the EU has requested the WTO to review the content and operations of all free trade agreements. An FTA could act as an 'insurance against a reversal of the move to a more liberal multilateral trading system.
Sixth, NAFTA and the general Enterprise of the Americas Initiative may pose a particularly acute policy dilemma for a number of the Caribbean countries where exports to the US account for 50% or more of total exports. Estimates of the static trade diversion effects of NAFTA on the Caribbean countries produce relatively small amounts (Davenport 1995) with much larger effects coming from loss of preferences as a result of the Uruguay Round. Much more important are the unquantifiable long term effects in terms of future market access and foreign direct investment. The CBI is a unilateral offer by the US under a GATT waiver (see section 3.2) and there is no guarantee as to how long this will be continued after 2005, while even if continued, there is always the possibility of protectionist measures by the US. NAFTA also allows for safeguard measures to be applied on imports, but the existence of the agreement makes their imposition subject to closer scrutiny while, subject to these measures, access to the US market is guaranteed indefinitely. This guarantee of market access can be expected to attract foreign direct investment to utilise the lower costs of production in those countries with stable economic and political systems, and transparent and relatively straightforward regulations. Such investment is a crucial element in diversifying the exports of small economies. FDI brings a package of benefits (as well as costs) to host countries, including access to importers, knowledge, and technology, If the Caribbean countries stay out of NAFTA or are not offered 'NAFTA-parity in their trade relations with the US, then at least for some countries there could be investment diversion towards NAFTA members, affecting not only current industries such as clothing but a whole range of future industries. As discussed previously, NAFTA membership may well mean that these Caribbean countries would also have to offer an FTA with the EU as well. Having done that, they will have effectively engaged in multilateral trade liberalisation and permanently removed the power of protectionist lobbies in their own countries.
Potential areas of difficulty in negotiating a free trade agreement with the EU
A free trade agreement would have to be in conformity with Article XXIV of the General Agreement which, inter alia, requires the removal of 'tariffs on the regulations of commerce for a substantial part of the trade (Article XXIV.2) and for 'substantially all the trade (Article XXIV.8 (b)) between the countries concerned. This raises the first potential area of difficulty for an EU-ACP free trade agreement.
The EU has consistently limited its offer on sensitive products, most notably in agricultural products, by means of product exclusions and limited access for products through partial reductions in tariffs, tariff quotas, calendar restrictions and high minimum entry prices. The Europe Agreement (association agreements) for example, exclude agricultural products, the Mediterranean FTAs provide only limited concessions on restrictions on agricultural products, and the proposed trade agreement with S. Africa excluded 40% of its farm exports and was referred back to the EU for further negotiation. As explained in Chapter 7, the widespread use of exclusions by the Commission in EU free trade agreements has led the Council of Ministers to questioned the ability of the EU to maintain such agreements within WTO rules.
More generally, these product exclusions illustrate one of the fundamental weaknesses of Article XXIV. The EU has argued in the past that 'substantially all trade should be interpreted as quantitative requirements applying to existing import protected (and therefore distorted) flows, in which case a free trade agreement which does not include such protected products would still be compatible with Article XXIV. The EU has also argued that 'substantial all should be interpreted as meaning at least 80% of actual trade. It could equally be argued, however, that this is special pleading and stretches the language of Article XXIV beyond its original intention, which was clearly directed at ensuring that no major sectors were excluded from such agreements. GATT working panel examinations of EU agreements have not, in the past, limited the EUs freedom of manoeuvre, despite their repeated criticisms of the exclusion of whole sectors. Our reading of the UR 'Understanding or the Interpretation of Article XXIV of GATT 1994, however, strongly suggests that this freedom of action will be much more limited in the future. The fourth introductory paragraph of the Understanding states that the contribution of free trade areas to the expansion of world trade 'is increased by the elimination between the constituent territories of duties and other restrictive regulations of commerce extends to all trade, and diminished if any major sector of trade is excluded. The Understanding on the evaluation under Article XXIV:5 of the general incidence of the duties and other regulations of commerce applicable before and after the formation of a customs union or free trade area states that it should be based upon an overall assessment (our italics) of the weighted average tariff rates and customs duties. The Understanding also states that the Dispute Settlement procedures can be invoked with respect to any matters arising from the application of the provisions of Article XXIV, including interim agreements leading to free trade areas. It should be recalled that the change in the UR from a 'consensus to accept to a 'consensus to reject means that the EU has effectively lost any power of veto which it previously had over an adverse disputes panel decision.
Taken together, this strongly suggests that the EUs interpretation of Article XXIV would be at least as much open to challenge in the WTO as a renewed Lomé Convention and undermines the assertion in the Green Paper that a free trade agreement would enhance the security and predictability of preferences and lend credibility to ACP trade policies. In order to be confident that an ACP-EU Free Trade Agreement would be WTO compatible, then all sectors of trade, including sensitive sectors such as agriculture would have to be included in the FTA. This is unlikely to be acceptable to the EU and may also be unacceptable to the ACP countries.
The second area of potential difficulty in negotiating a free trade agreement with the EU concerns the length of the transitional period. Under Article XXIV.5(c) this has to be achieved 'within a reasonable length of time and this was clarified in the GATT 94 Understanding as 'exceeding 10 years only in exceptional cases. In one sense, the low level of industrialisation of most of the ACP countries implies that there are relatively few industries competing with EU imports and so the adjustment cost to most of the ACP of liberalising trade with the EU would be correspondingly small. Secondly, the neo-classical (World Bank) 'shock approach emphasises that the quicker an economy is opened up to international trade, the smaller are the adjustment costs. The reasoning behind this is that it gets the 'pain over quickly, helps to overcome vested interest in protection, and enables consumers to benefit sooner rather than later from a wider range of goods and at lower prices, thus establishing the credibility and therefore support for the programme. Chapters 4 and 5 of this report, however, emphasise that the ACP are infant economies which require government intervention not simply in the World Bank sense of neutral 'market friendly intervention, but also strategic intervention in order to create a learning environment between related activities. If the logic of this argument is accepted, then a transitional period to free trade with the EU which would be WTO compatible would probably be too short a period of time for many ACP countries.
Thirdly, the EU agreements with Morocco and Tunisia (which could be taken as a model for the trade provisions of an ACP-EU agreement) which are to be implemented over a twelve year period, do not provide a good example of a free trade agreement going substantially beyond WTO disciplines and deepening free trade relations. Only a very limited range of agricultural products are included in the agreements and concessions are limited (see chapter 7) despite the importance of the agricultural sector to these countries, particularly as a source of employment. The EU has also insisted on safeguard clauses in addition to the right to impose anti-dumping and countervailing duties in ways which go beyond those permitted in the General Agreement, allowing 'substantial opportunity to intervene if the political willingness to do so exists (Hockman and Djankov, 1996). The authors also point out that 'anti- dumping remains applicable to trade flows between the partners despite the agreement of Morocco and Tunisia to apply EU competition disciplines(p.399). These and other factors lead the authors to conclude that 'the European Mediterranean agreements do not go significantly beyond existing multilateral (WTO) disciplines.
With regard to financial co-operation in the Mediterranean Agreements, although the volume of aid is expected to increase, no commitments are given in terms of real flows. It is also worth noting that the previous protocols on financial co-operation have not been renewed, and instead, there is a total sum available to all Mediterranean countries, with individual applications depending, at least in part, on country performance (as determined solely by the EU). This is interesting as it would seem to come close to the proposals discussed in Chapter 6 of the Green Paper.
If the Mediterranean Agreements are taken as representative of EU free trade agreement, then it is difficult to escape the conclusion that the motivations of the EU for such agreement has more to do with 'locking in countries into the EU sphere of influence and capturing markets for EU exports at the expense of third countries through the trade diversion effects of such agreements than it is to do with a widening and deepening of WTO trade disciplines.
3.4 The Generalised System of Preferences
The third possible option is to separate the trade provisions of Lomé from other areas of co-operation and to replace these trade provisions by the EUs generalised system of preferences.
All industrial and agricultural products imported from the least developed countries (as defined by the UN) covered by the GSP enter the EU duty free and thus similarly defined least developed countries in the ACP group would receive identical treatment to Lomé for these products. This 'enhanced GSP applies to imports of industrial products from the Andean and Central American States, with some exceptions in the case of the Central American States and with concessions on agricultural products (for further details see chapter 7).
For non-least developed ACP countries a reliance on the GSP would mean a significant erosion of their preferences, as guaranteed market access for industrial products free of 'duties and measures having equivalent effect, would be replaced by, what the EU refers to as, 'tariff modulation. Very sensitive goods (all textiles, clothing, ferro-alloys) only receive a 15% decrease in tariffs, sensitive products (such as footwear) only receive a 30% decrease, semi-sensitive products a 65% decrease, and only non-sensitive products enter duty free.
The small reduction in tariffs on these products, combined with the decrease in post-Uruguay Round tariff rates (albeit, small cuts in the case of sensitive products), means that for most products of potential relevance to the ACP, the margin of preference is so low (for example, 5.1% for footwear on the final Uruguay Round rate) that it would not be worth the administrative cost of applying for preferences.
In the case of agricultural products there are concessions on the tariff rates on some temperate products such as fruit, flowers, peas and beans, but in most cases the reduction in the MFN rate in the standard GSP is only 15% or 30%, giving GSP rates of 6.8% for peas, through to 10.2% for most flowers in the peak season and 12.2% for frozen beans.
The rules of origin in the EUs new GSP were improved to include as 'originating products imports from the EU of intermediate products used in the production of the exported good ('donor country content provision), but there is no general 'cumulation of origin among beneficiaries and only limited cumulation allowed among certain regional groupings.
The GSP is also a unilateral offer given by the EU and is not negotiable, nor is it contractual and this, combined with the limitations of the scheme, suggest that it will have little effect in stimulating domestic or foreign investment in recipient countries.
The safeguard clauses in the new GSP are also more restrictive than in the old GSP. For example, the previous scheme referred only to the possible implementation of the safeguard clause where there was any difficulty to the Community as a whole or to one of its regions. Now, the safeguard clause can be implemented if there is simply an economic difficulty to an EU producer. In addition, the least developed countries are no longer exempt from the re-imposition of tariff under these provisions.
The GSP also includes a complex formula for the exclusion of product/country combinations which are deemed to be competitive and no longer requiring preferences. Further analysis would be required to determine the effects of this formula on individual ACP countries, but it would seem unlikely that it would apply to any except, perhaps, some exports of clothing from certain countries such as Mauritius, Jamaica and Zimbabwe (and perhaps not even in these cases).
The GSP would seem to be of some value to the least developed ACP countries, provided there were improvements to the rules of origin and to the safeguard clauses to upgrade them to Lomé standards, although there would still not be the same degree of certainty and contractual relationship of the Lomé Convention. For the remaining ACP countries, the replacement of Lomé by the GSP would represent a decrease both in their level of preferences and in their security of access to the EU market.
3.5 Choosing between the Options
The permutations of 'variable geometry
The options of a renewed Lomé Convention; integration into the GSP; and a Free Trade Agreement; can be combined in various permutations negotiated with individual ACP countries, regional groupings, and ACP countries defined by different levels of economic development (with the added modification of 'vulnerability defined in terms of small size of the economy, specialisation in a very few exports, geographical isolation from markets, etc). The Green Paper presents some of what are an infinite series of possibilities. No one, however, can imagine how such complex variations could be negotiated, and an assessment of their impact on the ACPs concerned, other ACP countries, third countries, with such endless possibilities is an impossible task. The permutations of 'variable geometry may be an interesting intellectual exercise, but it is difficult to see how they could form the basis for practical policies. On the other hand, the possibly divergent interests of some ACP countries need to be recognised. Countries in the Caribbean and Pacific may see their future increasingly involving closer integration into regional trade, and if they were unable to achieve this through other means, for example NAFTA-parity in the case of the Caribbean countries, then a simple renewal of Lomé may conflict with such an objective. Also, ACP countries vary widely in their levels of development, from the poorest countries in the world like Rwanda (per capita GNP $80) to Trinidad and Tobago, and Gabon (with per capita GNPs of just under $4000).
Guidelines for deciding among the options
This conundrum is best approached using guidelines for decision taking.
First, ACP preferences are now of relatively minor and diminishing importance compared to 1975 when Lomé began. Preference margins have been eroded by tariff cuts and as a result of the Uruguay Round the EUs trade weighted GSP tariff average in tropical agricultural products has decreased from 15% to 9.4% (tropical non-agricultural products are 0%); natural resource based products have decreased from 3.4% to 3.2%, textiles and clothing are 9.3%(3) and leather and leather footwear has declined from 0.2% to 0.1%, while other industrial products have declined from 5.1% to 3.5% (Greenaway and Milner, 1995). There are probably further opportunities for utilising preferential access to the EU for some temperate agricultural products and fisheries, but the possibilities here seem limited. In the case of textiles and clothing, the market is highly competitive, already well developed and non-price factors are more important than the margin of preference (although they may be useful at the margin when importers are choosing between competing sources of supply). Import quotas under the MFA (from which the ACP were exempt) are scheduled to disappear by 2005, and although it is questioned whether this timetable will be adhered to (given that most sensitive categories of clothing are scheduled to have quotas removed in that year, as well as the possibility of using contingent protection measures), ACP countries can only rely in the future on a margin of preference equal to exemption from duties (10% to 12% on GSP/MFN rates, but 0% with respect to the country/product coverage of the enhanced GSP) and will have to compete in a free (or relatively free) market with low cost producers in Asia. In addition, ACP preference margins have been eroded by the proliferation of the EUs special preferences for countries in Europe and in the Mediterranean region.
This leads us to two important conclusions. First, the ACP countries could consider opening Lomé tariff preferences to an appropriate group of non-ACP countries, since the costs in terms of further erosion of existing preferences would be small and the benefits could be a greater degree of acceptance of Lomé in the WTO (this is explored in more detail in section 3.6). Second, the benefits in terms of exemption from tariffs of an FTA with the EU are likely to be small and insufficient in themselves to outweigh the costs of granting free entry for imports from the EU (although, as we have discussed, there may be other intangible benefits).
Secondly, the aid and technical co-operation provisions between the EU and the ACP may well be separable from the trade co-operation provisions of a new agreement so that if an ACP country did not wish to obtain EU preferences under Lomé or an FTA, it could still participate in the non-trade provisions of a new ACP-EU agreement.
Third, free trade agreements are inferior to MFN trade liberalisation unless they include benefits which could not otherwise be obtained through other means. These may be intangible or tangible benefits. Intangible benefits, include a fear that the multilateral world trading system could break down into competing regional groups. This is always possible, but it seems unlikely as it would lead to a return to the situation similar to the depression of the 1930s and it was this disastrous experience of 'beggar my neighbour policies which led to the General Agreement on Tariffs and Trade being established to avoid a repetition of this disaster.
Tangible benefit could include exemption from the use of 'contingent measures of protection and other non-tariff barriers to trade, not at present covered by Lomé. For example, anti-dumping investigations; countervailing measures; the application of safeguards, standards (technical, health, safety, etc) to imports; and government policies which directly or indirectly adversely affect trade. Preferential Agreements could also include additional trade and investment measures available only to members of the preferential area.
Fourth, the analysis of the EU trade agreements with other groups of countries suggests that the EU operates a 'hub and spoke system through the use of regional trade agreements, both to capture as much as possible of the market and to minimise the adjustment costs to itself of reciprocal preferences. Sensitive products are omitted from the trade agreements, safeguard provisions are more onerous than those under the WTO, and compliance with EU competition rules and protection of EU investment is demanded without compensatory concessions, such as exemptions from contingent protection measures.
Fifth, all of the above suggests that great skill will be required in the negotiations both to maximise the benefits from a trade agreement (and the additional benefits required if a free trade agreement with the EU is included in the trade options). This strongly points to the need for the ACP countries to pool resources and to negotiate as far as possible as a group.
Sixth, the number of alternative arrangements forming a new agreement ('variable geometry) should be kept to the absolute minimum. This is essential, not only to maintain the common objectives and cohesion of the ACP group, but also because the greater the variation in the content of ACP-EU trade provisions, the stronger will be ACP-EU ties, and the weaker will be the incentives for ACP regional trade. A range of different ACP-EU trade agreements will make it much easier and less costly for either an EU or third country to service an ACP market from the EU than from an individual ACP country. This arises, firstly, because where duties have to be paid on imports of intermediate products from the EU in an ACP country, manufacturers will be at a competitive disadvantage compared to manufacturers in a neighbouring ACP with a free trade agreement with the EU, and will resist moves towards regional free trade. Secondly, if the product coverage of agreements between the ACP states and the EU differs, then EU rules of origin under different agreements will be strictly enforced to maintain the integrity and differentiation of the two agreements and this will inhibit regional trade.
3.6 Viable Options for Negotiation
Of all the many possibilities open to the ACP regarding trade co-operation with the EU, only the following seem viable (not ranked in order of merit):
1. Renewed non-reciprocal Lomé preferences under a WTO waiver with a more focused range of policy instruments and objectives (as discussed in Chapter 6).
2. Non-discriminatory trade liberalisation in the ACP countries with a more focused ACP-EU development co-operation agreement.
3. A free trade agreement subject to:
(a) Obtaining benefits not otherwise obtainable and sufficient to outweigh the costs of such agreements.
(b) Minimising the variations of such agreements in terms of product coverage, lengths of transitional period (long transitional periods limit the usefulness of such agreements as an 'anchor for economic reforms, decrease the credibility of these reforms for domestic and foreign investors, and encourage protectionist pressure groups) and rules of origin.
(c) Reducing the cost of free trade agreements by simultaneously decreasing barriers to trade with third countries, and binding reduced tariffs in the WTO.
(d) Obtaining a new and more focused development co-operation agreement.
Strengthening the acceptability in the WTO of a renewed Convention
We have argued that the Lomé Convention could be renewed on the basis of a further waiver from Article I in the WTO if there is sufficient political will to do so on the part of the EU. Nevertheless, a further waiver would probably be subject to closer scrutiny in the WTO than the present waiver, given the change in attitudes of the Contracting Parties reflected in the Uruguay Round negotiations and agreements towards arrangements which did not conform to the GATT and which clearly discriminated between member countries. Discrimination between developing countries, in particular, combined with the greatly strengthened Disputes Settlement procedures also, as the bananas case has demonstrated, substantially increases the possibility of challenge and panel litigation even when a waiver is granted; and this reduces the confidence of exporters and investors in receiving preferential treatment. It is therefore in the interests of the ACP and the EU to consider placing Lomé trade preferences on a less arbitrary and discriminatory basis than the present regional grouping. As we have pointed out, the substantial erosion of Lomé preferences since 1975 implies that the extension of trade preferences to a wider group of comparable developing countries would impose very little additional costs to the ACP countries in terms of a further reduction in their margin of preferences. A generally applicable criteria for Lomé preferences would also imply that such preferences were 'time bound and therefore of additional compatibility with WTO procedures.
The precise form of such criteria is clearly a sensitive issue which would have to be acceptable both to the ACP countries and to the member states of the EU. One obvious extension of Lomé Preferences on trade would be to extend them to the least developed countries (UN definition). This would be consistent with the poverty-oriented focus of EU development co-operation, as stated in the Maastricht Treaty, and with the aid polices of a number of the EU member states. The least developed countries already have benefited from the EUs enhanced GSP and the only potential cost in terms of preference erosion for the ACP would be in certain sensitive textiles and clothing goods exported by Bangladesh. Extension of preferences could also include the Andean and Central American countries under the terms of preferential entry covered by the enhanced GSP. Table 3.3 lists these three groups of countries.
Table 3.3Countries Covered by the EUs Enhanced GSP
Least Developed
Yemen, the Maldives, Myrimar (Burma), Bangladesh, Nepal, Bhutan, Afghanistan, Laos, Cambodia
Andean Countries
Bolivia, Ecuador, Colombia, Peru
Central American Countries
Panama, Costa Rica, Nicaragua, El Salvador, Honduras, Guatemala
The Andean countries have tariff-free entry for industrial products and the Central American countries have tariff-free entry for non-sensitive industrial products and additional preferences over the standard GSP for sensitive products. Agricultural products are subject to a more complex regime in which there are five lists of agricultural products (for further details, see chapter 7) in which the least developed countries have duty-free access for the products in all five lists. The other two groups of countries only have duty-free access for the products in the fifth list, with tariff concessions of varying depth on the products in the other four lists. Given the sensitivity of agricultural products in the EU, the most that could be expected would be an extension of Lomé preferences on industrial goods to these countries with the consolidation of enhanced GSP agricultural preferences into Lomé. The advantages to these non-ACP countries would be to place their preferences on a more secure contractual and negotiable basis than the EUs unilateral GSP offer and to extend to them the benefits of regional cumulation of Lomés rules of origin. The costs to the ACP in terms of erosion of preferences would (with perhaps the exception of certain sections of clothing from Bangladesh, and this would have to be quantified) be minimal.
This enlarged regional grouping to include non-ACP States would still be arbitrary and therefore open to the charge of unjustified discrimination against non-member developing countries. A bolder step would be to place the EUs special preferences, at least for industrial products, on a more rational and therefore more easily defended basis, which could be recognised by the WTO. The basis for such a scheme could be founded on the fundamental rationale for preferences, namely that developing countries have infant industries which require preferential access to a protected market to enable them to 'learn-by-doing and achieve minimum efficient scale of production. The authors of this report would enlarge on this traditional analysis by emphasising that the analysis should be extended beyond the level of the firm and recognise the importance of externalities, the absence or limitations of which create infant economies. The limitations or absence of internal and external economies of scale particularly apply to developing countries with a relatively small size of domestic market and this suggests a criteria which combines low per/capita income levels with levels of GDP.
Table 3.4Excluded Developing Countries if Lomé Preferences Applied
Countries with a GDP less than $50,000 million (1994) and to countries eligible for the EUs Enhanced GSP (in rank order of per/capita GNP, 1994)
| India
Pakistan China Indonesia Philippines |
Thailand
Venezuela Iran Brazil Malaysia |
Mexico
Argentina plus: Rep. of Korea Singapore (Hong Kong) |
Table 3.4 should be regarded only as illustrative of one result of such an exercise based on the simple criteria of Lomé preferences being granted to developing countries having a GDP of less than $50,000. The countries excluded under such a criteria are intuitively those which, given the large size of their domestic markets and/or their level of economic development, seem least in need of preferences beyond the GSP. The major advantage of this proposal is that it would not only place the EUs special preferences on more rational basis than the present complex hierarchy of preferences, but also simplify the hierarchy so that, for example, there would be a simple unified scheme for industrial products covering the non-candidate (for entry into the EU) Mediterranean countries as well as countries in Asia and Latin America comparable to the ACP countries. As we have emphasised, given the EUs existing structure of tariffs, preferences, and imports from these countries, we would not expect the costs to the ACP (in terms of erosion of preferences) and to the EU to be significant (though this requires more detailed analysis). Equally we would not expect this, or a similar proposal, to be able to be agreed by 2000. A commitment to liberalise trade in this way, however, would fulfil the WTOs objective of regional trade agreements acting as a stepping stone for future global liberalisation, overcome the pessimism of those who regard regional integration as acting as a force opposed to multilateral liberalisation, and automatically generate a large measure of support in the WTO for a further waiver for a new Convention. Special preferences based on an objective criteria would also mean that they were time bound and recipients would, as their economies developed, move from Lomé preferences, to GSP, and finally graduate out of the GSP to receiving MFN treatment. Within this framework, countries could, if they so wished, opt out of special preferences and negotiate WTO-compatible free trade arrangements.
Accommodating the divergent interests of the ACP Group - An Umbrella Agreement
The ACP countries comprise a heterogeneous group with the ten poorest countries in the world with a per/capita GNP of $200 or less, compared to over $3,000 for Gabon, Mauritius, Trinidad and Tobago. Similarly, shares of agriculture in GDP range from over 50% for the poorest countries to under 10% for the richest, while shares of manufacturing range from under 5% for some of the poorest countries to up to one-quarter for the more developed. These large structural variations could be accommodated within an 'umbrella agreement.
This could take the form of the least developed ACPs renewing the Lomé Convention (this would make it easier to obtain a WTO waiver and would be less subject to subsequent challenge), while the remaining ACP countries would negotiate a free trade agreement with the EU but with varying transitional periods, depending on their ability to adjust to free trade with the EU and reduced barriers to trade with the rest of the world.
Alternatively, individual ACP countries could elect either to join a renewed Lomé type arrangement or to join in the negotiation of a free trade agreement, having decided for themselves the costs and benefits all these two alternatives.
The third possibility is that the least developed ACPs could enter into a non-reciprocal Lomé-type agreement; the more economically advanced ACPs could (possibly self-elected) negotiate a free trade agreement with the EU, while the less economically advanced ACP countries (excluding the least developed) could obtain their trade preferences from the GSP. For the reasons given when discussing the option of the GSP, we do not regard the latter as a useful option. GSP preferences are of limited value and unlikely to have the potential benefits for export or foreign direct investment either of Lomé or a free trade agreement and to try and improve the GSP solely for a group of ACP countries would, almost certainly, be successfully challenged in the WTO.
The practicalities of negotiating an 'umbrella agreement and its detailed content require discussion between the ACP countries, but it could broadly take the form of the free trade agreement incorporating a general timetable for trade liberalisation. The Community would offer free access to the EU market at the beginning of the agreement, with the ACP countries concerned agreeing to liberalise trade over a period of somewhere between 10 to 15 years (the maximum probably acceptable to the WTO). A commitment would be made to liberalise x% (eg one-third, or one half) of tariff headings by the mid-point of the agreement and all tariff headings by the end of the transitional period, and each ACP country would undertake to submit a binding schedule of trade liberalisation measures to the EU by an agreed period of time (eg three years) after signatory of the agreement. This time period would be necessary because the drawing up of the schedule would be very demanding on the limited resources of many ACP countries and they would probably require technical assistance. In view of this, it would probably be necessary to obtain a waiver covering all ACP countries for an extension of Lomé past the year 2000 until the year set for the beginning of the transitional period of the free trade agreement. Aid and technical assistance provisions would be closely linked to the phasing of the transitional period.
For those ACP countries not joining the free trade arrangement, Lomé would be renewed broadly on its existing terms, but with improved coverage for agricultural products of interest to the ACP.
For both a renewed Lomé and a free trade agreement, the ACP could seek to improve on the present, Lomé, rules of origin. The stringent process rules require ACP countries to add value to non-originating (ie non-ACP or EU) inputs which are often beyond their manufacturing capabilities. The exclusion of 'simple processing and 'simple assembly discriminates against the least developed countries, while rules on fisheries are prohibitive (see McQueen, 1982). Some minor concessions have been made over the years, but the rules remain highly restrictive and discriminate in favour of EU as against non-EU, suppliers. Unfortunately, improvements are even less likely in the future as the Green Paper indicates that rules of origin for all preference schemes are to be harmonised and this will make it more difficult to liberalise the rules for the ACP.
All ACP countries would participate in a Financial and Technical Assistance Agreement focusing on particular areas, notably building export capacity, with different instruments tailored to the needs of ACP countries at different levels of development (this is explored more fully in chapter 6).
An alternative structure for negotiation, and one which would appear to be favoured by the Commission (following the pattern set for the CEEC and Mediterranean countries) is to negotiate new agreements on a regional basis. The difficulty is that the ACP countries do not conform to homogeneous geographical regions. In Sub-Saharan Africa, for example, the most obvious regional groupings are the 23 ACP members of COMESA (Common Market for East and South Africa), the 7 members of CEAO (Communanté Economique de lAfrique de lOuest), the 3 members of the MRU (Mano River Union), and the 6 members of the UDEAC (the Customs and Economic Union of Central Africa). This still leaves a number of countries not covered by these regional groupings - Ghana, Nigeria, Togo, Cape Verde, Gambia, Guinea Bissau. In theory, these countries, together with the members of the CEAO, and the MRU, could be covered under ECOWAS (the Economic Community of Western African States) but in practice, the economic, political and cultural differences between the member states has meant that the ECOWAS Treaty is 'a dead letter (Faroutan, 1993). These groupings also omit Zaire, Sao Tomé and Principe, and the Seychelles. Levels of economic development, structure of the economies and future development strategies all differ substantially within these groupings and it would therefore be extremely difficult, if not counter productive, to arrive at a common negotiating position with the EU. The Caribbean countries also exhibit substantial differences in per/capita incomes, size and structure of the economy and future development strategies (particularly the vulnerable Eastern Caribbean countries). The economic interests and levels of development of the Pacific countries also show large variations and therefore substantial differences in their objectives in a new agreement. An 'umbrella agreement negotiated between the EU and the ACP would avoid the need to accommodate the conflicting objectives of countries within a regional grouping and the problem of how to cover ACP countries outside regional groups, and would most effectively utilise the negotiating capacity of the ACP.
In conclusion we would emphasise the following. First, we believe that the Lomé Convention could be renewed with a WTO waiver in the same way that the US renewed the Caribbean Initiative, provided the political will is present on the part of the EU. Second, a free trade arrangement is inferior to unilateral and multilateral trade liberalisation unless it enables the ACP to obtain EU concessions and assistance not otherwise available. In any case, most ACP countries would have to engage in trade liberalisation if they are to minimise the potential costs of a FTA. Third, the most important policy objective for the ACP is to integrate their economies much more closely into the world economy, and preferences are now of relatively minor importance (and may be counter-productive) in accelerating this process. Much more important are the domestic policies of the ACP countries which rationalise and reduce barriers to imports in a non-discriminatory manner, as between trading partners, and policies which enhance the export capacity of the country. A new agreement could be useful in providing preferences in 'sensitive goods and providing some degree of guarantee of unhindered market access, but its most valuable potential role is in supporting and reinforcing domestic policies in the ACP States.
References
Davenport, M., Impact of the Uruguay Round and NAFTA on Commonwealth Caribbean Countries, with Special Reference to Jamaica, London, Commonwealth Secretariat, 1995
Greenaway, D., and Milner, C., The Uruguay Round and Commonwealth Developing Countries: An Assessment, London, Commonwealth Secretariat, 1995
Hallam, D., and Preston, M., The Political Economy of Europes Banana Trade, University of Reading, Department of Agricultural and Food Economics, Occasional Paper No.5., 1997
Hockman, B., and Djankov, S., 'The European Unions Mediterranean Free Trade Initiative, The World Economy, 1996, pp.382-406
McQueen, M., Lomé and the Protective Effect of Rules of Origin, J of World Trade Law, Vol.16, 1982, pp.119-132
Yeats, A., 'Do African countries pay more for imports? Yes in World Bank Economics Review, Vol.4., No.1%.
1. 1 I am grateful to Dickson Yeboah of the WTO for this information.
2. 2 I am grateful to Edwini Kessie of the WTO for guidance on this section.
Any errors are, of course, entirely the responsibility of the author.
3. 3 Stated as 0% by Greenaway and Milner but this appears to be incorrect and so we have calculated
the average GSP rate.
Developer's Note: These pages were developed for use on the Netscape browser. Please address comments to Alec Singh
ACP Home Page
Updated on November 3, 1997