Select Committee on Agriculture Appendices to the Minutes of Evidence


Memorandum submitted by the Overseas Development Institute


  On 20 September 2000, the European Commission issued a press statement, followed some days later by a draft amendment to the GSP Regulation, proposing that Least Developed Countries (an official UN designation, not a description) have duty- and quota-free access for all products except arms, with immediate effect (once approved by the Council) for most goods and a three year phase-in period for bananas, sugar and rice. The only restrictions were the usual ones on rules of origin and provisions against fraud, plus provision for possible suspension in case of "massive" increases in imports "in relation to their usual levels of production and export capacity". The formal proposal included the statements (required for EC proposals) of the impact on business, with special reference to small and medium-sized enterprises, and of the financial effect; the latter looked only at customs revenue.

  In informal discussions immediately after the proposal, now apparently moving to formal discussions, it has been suggested that the three-year phase-in could be extended to five years and that the anti-surge provision be made more automatic. (The original proposal was for a large increase only to be a "possible" reason for imposing limits.) Timing of the "immediate" provisions remains uncertain: the Council has the opportunity to consider the proposal in December, but if it wants to see studies of the impact on developing countries (which could in turn modify the expected internal and financial impacts), the decision is likely to be delayed into 2001.



  The GATT Enabling Clause of 1979 allowed not only special preferences for all developing countries, but additional preferences for the Least Developed Countries. These are defined by the UN. Countries are measured against quantitative economic and social criteria every three years, and recommendations are made to the General Assembly for countries to be added to or removed from the list. Under these provisions, the European Generalised System of Preferences has offered better access to Least Developed for at least 15 years. Since 1997 the EU has proposed extending this. The first proposal (presented as a policy objective, not an official document) was for "all goods", but in all official documents until September 2000 it has been phrased as "essentially all".

  The EU 1997 proposal was part of a WTO initiative (initiated by the then Director General, Ruggiero). Following the Uruguay Round of trade negotiations, analysis indicated that most of the gains to developing countries had gone to the more advanced (in particular the clothing exporters of Asia and some natural resource exporters in Latin America), while some poor countries, including many Least Developed, had actually lost (because of the indirect effect of gains made by others). This led to disillusionment with the trade regime on the part of poor countries and those concerned with their interests, and various efforts both to counter the possible economic damage and to demonstrate that the international trading system could offer benefits. A conference in November 1997 on the Least Developed encouraged both developed countries and the more advanced developing to make improved offers to the Least Developed on trade and also to target technical assistance to improve export capacity on these countries. As with "normal" GSP, the actual offers were made unilaterally by individual countries (the EU acting together), not negotiated. The US and Japan made offers with lists of exceptions; some Asian developing countries made offers on specific products. There was a proposal to include negotiations on this in the draft agenda for the aborted Round of WTO trade negotiations; in this the EU insisted on including the word "essentially" to modify "all" in order to permit exceptions. After the Seattle collapse of negotiations, the new WTO Director General was concerned to keep Least Developed Countries committed to the multilateral system through improving and implementing the proposed concessions, and the EU, as a strong supporter of a new Round, had a clear interest in trying to secure support from all groups of countries for this.

EU-ACP relations

  At the same time as this revival of interest in preferential arrangements for the Least Developed at multilateral level, the EU was negotiating with the ACP to find a successor to the trade provisions of the Lome Agreement. Starting with the Green Paper of 1997, the EU made it clear that it wanted to shift from unilateral concessions to a negotiated agreement. While many ACP countries doubted the economic advantages of this, especially as it required them to "negotiate" to preserve their current access, in return for giving the EU greater access, it did appear to have one clear advantage: that an agreement would be contractual, requiring both sides to observe it. The EU also indicated very strongly that it supported regional agreements among the ACP countries to strengthen their trading positions. During the pre-negotiation phase, studies of the impact of reciprocal trading agreements on the ACP countries noted the possible inconsistency between improved concessional arrangements for the Least Developed and uniform agreements with the members of regional groups. Every potential regional group of ACP countries includes both Least Developed and non-Least Developed countries. While the Least Developed would gain nothing from an additional trade agreement with the EU, the others would gain; both the Least Developed and the others would lose from any splitting of regions by level of development. The EU therefore modified its original proposal (that Least Developed in a region would lose their Least Developed access), and the Cotonou Agreement needed to include a mention of the plans for the Least Developed: it recognised that the EU intended to grant access for all Least Developed Countries for "essentially all" goods. While neither "at most" nor "at least" appeared as a modifier of "essentially all", it was clear from the previous EU use of the term (in the multilateral negotiations) that the intention was to limit access for some goods. The proposal to offer "all" three months later was, therefore, a unilateral change in what had been negotiated. (In WTO terms, it might have been considered a `nullification or impairment' of an existing agreement, and therefore challengeable.) The Cotonou Agreement contained provisions for the ACP countries to be consulted through an ACP-EU Ministerial Trade meeting prior to any proposals to change the agreement; there had been no consultations before September 2000.

Status of Proposal

  The proposal made in September 2000 was a proposal for formal implementation of an existing policy objective, not a new initiative in an unexplored area. It was reasonable to expect that its consequences, direct and indirect, had been considered before the draft was submitted (in formal terms, that the required impact assessments had been done), and that it would take account of existing policy commitments. There may be circumstantial explanations for the failure to take official account (through consultations or an impact assessment) of either the interests of all other developing countries or the consequences for the existing commitments under the Cotonou Agreement. The moving of formal responsibility for trade with the ACP countries from the Development DG to Trade may have reduced appreciation of the development implications of trade measures. There has been an apparent reduction in Commission enthusiasm for regional agreements in general and those proposed with the ACP in particular. Securing support for a new WTO Round has become the priority for the Trade DG. These considerations, however, should not affect either formal obligations or the commitment to help all developing countries.


  The proposal to reform the EU sugar regime has the same timetable as the Least Developed proposal (in its original three-year form), with no proposals yet for the post 2003 regime. The quota regime for the ACP countries plus others, including India, continues unchanged for at least these three years. Under this, the countries are obliged to offer and the EU is obliged to buy fixed quantities at the European price (substantially above the world price). Legally, the Sugar Protocol is not part of the other arrangements with the ACP countries, but an agreement of "permanent" duration. Therefore, increased sugar imports under the Least Developed proposal could not displace imports under the sugar quota: the EU remains committed to purchase these indefinitely at above market prices, even if their market is reduced by imports at much lower prices from the Least Developed. This could imply high budgetary costs (to subsidise re-exporting or otherwise disposing of the quota sugar). These would come at a time when the agricultural policy and the budget were under pressure from production in the accession countries. This has caused fears on the part of sugar producers that the Protocol may not be as less long-lasting as foreseen in previous commitments. These fears are strengthened by the reaction of the EU (and informally of the UK government) to their complaints about the short duration of the transition for sugar under the Least Developed proposals: that the ACP countries with sugar quotas need to adapt to a changing international regime. Expectations about policy are clearly important in determining how investors both in Least Developed and in other countries will respond to the new opportunities. Wrong expectations that the Protocol would continue could lead to preserving capacity that will not be used. Wrong expectations that it will end could lead to underinvestment. Both Least Developed Countries and quota-holders need certainty.


  The effect of removing trade barriers to one group of countries, leaving others unchanged, is to increase imports from the favoured group; some of these will replace production in the EU (if it is now more efficient to import); others will replace imports from other countries (if the removal of the barrier creates competitive advantage). The first, "trade creation", is a clear benefit to both the EU (lower prices) and the Least Developed (higher income). The second is a benefit to the Least Developed, a cost to the non-Least Developed, and, depending on assumptions about how traders price and whether the other countries pay tariffs or not, a potential cost or benefit to the EU. It is recognised under WTO rules that free trade areas or customs unions among members can damage the trading interests of non-members through trade diversion. Therefore, although FTAs or CUs are permitted, they are required to offer compensation to non-members for any anticipated trade diversion. While the calculations are necessarily arbitrary and subject to error, the provision has been followed, for example with each enlargement of the EU. There is no similar requirement when preferences are granted. The original concept was that preferences were offered by all developed countries for all developing. Therefore any diversion would be from other developed, and these had agreed to bear any costs for the benefit of developing. When differentiation for the Least Developed was introduced, the implicit assumption was that it was harmless to offer them higher preferences because they would be unable to increase their exports very much, and therefore that any diversion would be negligible. The question of whether it was right for individual developed countries to be able to decide to impose diversion on some developing for the benefit of a different class of developing country was not considered. At the time, developing country participation in trade negotiations was very limited.

Effects on the Least Developed

  How quickly and effectively will they respond to the new opportunities? If the proposal is modified or restricted before it is implemented, and particularly if the implicit threats to restrict imports are activated, then the failure of Least Developed to take advantage of their opportunities will be a self-fulfilling prophecy. This anti-surge provision could, of course, prevent any Least Developed country from making effective use of the new access to increase their export and production capacity, and could be inconsistent with the recognition in the proposal of the need for assistance to the Least Developed to improve their capacity to export. It appears to be intended to avoid damage to EU producer interests, not to prevent damage either to EU budgetary costs or to the exports of existing suppliers to the EU. But if the proposal is not restricted, previous experience suggests that some countries will increase some exports very strongly. Bangladesh, exempted as Least Developed from controls on clothing and textiles by the EU (but not by other developed countries, so its experience is a very close parallel with the current proposal) saw an increase in the share of clothing in its exports from 0.01 per cent in 1977 to 51 per cent in 1991. It was able to increase its exports partly because both producers and purchasers with experience in the restricted countries encouraged the development of the Bangladeshi industry. The increase did not, therefore, rely on existing production or exporting capacity in the country. Some sugar producers in quota countries are already involved in Least Developed (for example Mauritius in Mozambique), so that the links to markets can be quickly established. If Least Developed are able to obtain a price higher than the world price (even if lower than that on quota sugar), then they will have the incentive to import supplies for domestic consumption and export their total production to the EU.[1] The greater the uncertainty over the future of the Sugar Protocol, the greater the incentive for purchasers and investors to move into the Least Developed countries. The size of the effect (and its distribution between the Least Developed countries and outside investors and purchasers) cannot be calculated simply by looking at existing sales (as is done in the financial impact assessment in the Commission proposal). It is a basic error to assume that the level of a restricted import will not rise when the restriction is removed. It is necessary to look at the natural conditions, including transport and other infrastructure for trading and at the potential to change these.

Effect on other ACP and other developing countries

  Until the potential for shifting trade to the Least Developed has been assessed, it is not possible to assess the effect on ACP countries. But, particularly in Africa, there are many cases of neighbouring countries (Malawi and Zambia with Zimbabwe; Uganda and Tanzania with Kenya) where the exports of the Least Developed are likely to compete directly with those of the ACP.

  For those products where there are still some restrictions on ACP exporters (including sugar, but also some other CAP products, horticultural products, etc.), the Least Developed may gain immediately. (See Box for some indications of the effect on Caribbean countries.) After 2008, if the non-Least Developed do not sign an agreement which renews their access to the EU under the Cotonou Agreement, the Least Developed will gain on all products on which developing countries still pay tariffs. They will gain on these immediately compared to non-ACP, non-Least Developed countries. The gains in sugar may represent trade creation and an increase in efficiency (although the convoluted nature of the sugar market will make it impossible to be certain, and make it unlikely that EU consumers or quota-holders will benefit from the increase in efficiency). Where the gains come because the competitors of the Least Developed face restrictions, there will be trade diversion, and loss of tariff revenue without a corresponding gain in efficiency in the EU.

Effects on Caribbean

  Sugar is probably barely viable long-term without the Sugar Protocol—Caribbean countries could live with the gradual decrease in European prices but not with Least Developed competition from Asia. This will force St Kitts,Trinidad and Tobago and Jamaica out of sugar entirely. St Kitts will lose all production:, like Antigua but without the mass tourism or airline hub.

  Trinidad and Tobago and Barbados will probably benefit and further diversify their economies. In Jamaica the risk of social unrest and political instability is acute. Unlike Belize and Guyana, where there are prospects (and existing investments) to modernise the sugar industry and make it viable regionally, Jamaica cannot really produce cane sugar economically given its costs. Even in Guyana, where the Indo-Guyanese community supply the labour and depend on sugar and rice exports, the three years does not leave enough time to adjust.

  On rice, Guyana is getting more efficient and can compete with Vietnam for regional if not world markets. But if the Asian Least Developed get effective preference, it won't break into the world and won't hold the regional market.

  Rum is quite detached from the sugar issue (except historically) because the Caribbean industry (fragmented between the islands and Guyana, mostly with old plant) imports the molasses rather than using unreliable local sources. Because of the original EU protectionism in Lome I to help French DOMs, they took the route of bulk rum exports while the only successful cases went for brands (Cuba, then Bermuda/Bahamas). Barbados now has a little branding in Int. Distillers' Malibu but it is mainly too little too late.

  Already Bangladesh is exporting bulk rum and eating into that world (and EU) market. Ever since the 1996/97 zero for zero white spirits deal, on which the ACP were again not consulted, the EU have recognised they hurt the Caribbean ACP. These producers are now making common cause with the French to extract a few extra subsidies out of the EC. This can be done, though not on the sort of proportions which would be required for sugar.

  But Guyana possesses some fantastic potential sugar brands eg Demerara (and speciality sugars within them) which have never been exploited. EU money might be better spent on marketing, design, market research, and strategy to turn commodities into products which feed off the rest of the Caribbean's USPs eg tourism, entertainment and youth, than in producer subsidies.

Calculating the impact

  It is a truism to say that it is difficult to forecast the future. But both governments and investors have expectations about the consequences of their actions, and where these are diffuse (many products, affecting many countries, at different times, in different directions), it is useful to project in a systematic way. Even if large areas of uncertainty remain, the direction of effects, and in some cases their likely magnitude, can be specified, and where there seem to be either serious risks or large uncertainties, monitoring can be put in place to check the effects as the policy proceeds. It is for this reason that many governments establish procedures (which are recognised to be imperfect) for requiring "impact assessments" of any policy proposal, at a minimum of the likely financial implications and the effects on identified vulnerable groups (the poor, small business, the environment), In the UK, of course, these also apply for domestic implementation of European measures. The EU and UK provisions do not include a requirement to consider the impact on developing countries, and the WTO does not require it for preferential agreements, but it was precisely because it was recognised that liberalisation to other countries by the EU had already affected the ACP countries and might do so in the future that provision for consultation after assessment was included in the Cotonou Agreement. The difficulties of doing it may seem large, but are no greater than those for any regional arrangements (they will be required when the accession countries join the EU) or than those done for the proposed regional partnerships with the ACP countries which were done in 1998-99. That the Commission has now said that it is willing to do the studies indicates that it does not consider them impossible. As the policy proposal is already three years old, it is not clear why it has now become so urgent that it is impossible to take the time to do the studies.

  The impact assessments on finance and business which were included with the proposal implicitly make forecasts about the impact on the Least Developed: effectively very small, except for some unquantified words about the "dynamic, long-term effect on spurring trade on new products for the LLDCs". The effects on other countries will therefore be correspondingly small. Giving an explicit assessment of the effects on Least Developed and other countries, however, would require spelling out the assumptions that must have been made. If the analysis of the impact on Least Developed was not done with sufficient care, then the assessment of the financial and SME effects, which is legally required, must be deficient.


  The justifications for giving special preferences for the Least Developed countries (over and above those for other developing) are that their needs for special treatment to gain access to markets are greater and their greater poverty means that any "costs" (in trade bargaining terms) to developed countries of preferences should be concentrated on them. Both arguments need examination because the current definition of Least Developed countries is not designed as a direct measure of either poverty or trade disadvantage. The UN definition excludes countries with a population above 75 million (except Bangladesh which was already included when this limit was imposed) and has three components, income per capita, human resources[2], and economic vulnerability[3]. A Least Developed country must be below the cut off points on all three, or on income plus one of the others plus special factors. The income level, at US$900, ensures that all least developed countries are poor, but the population limit and the economic and human resource components exclude some poor countries (notably India, Pakistan, and Kenya) from the classification. Countries which meet the standard of a low share of advanced products in their exports are often the least likely to face trade barriers, while the UN index naturally does not distinguish between specialisation in primary products vulnerable to EU protection for CAP goods and specialisation in tropical agriculture or minerals, which are largely free of barriers. Offering special access to the Least Developed therefore excludes some countries which a poverty-focused policy would want to target, and may damage them if they suffer from trade diversion to Least Developed competitors.


  The EC initiative shared one of the purposes of the WTO initiatives for Least Developed, to counter Least Developed Countries' dissatisfaction with the multilateral system. There is a risk, however, that it will increase the dissatisfaction of other developing countries, either because they consider the discrimination in favour of Least Developed "unfair": they also are poor or feel that they suffer some other serious disadvantage (island, landlocked, vulnerable to natural disasters . . . ), or because it appears to go against an existing agreement, the Cotonou Agreement, and therefore to weaken the role of law in the international system. One of the few advantages of the Uruguay Round perceived for all developing countries was the strengthening of international rules and the mechanisms for enforcing them. A predictable and enforceable international system is most important for the weakest countries (as is true of national rule-of-law). Thus even the least developed countries have an interest in preserving correct procedures. It will be more difficult for the EU to negotiate with the non-Least Developed countries in the eight years of the Cotonou negotiations if they feel that the June 2000 agreement has already been violated, and it may be more difficult to have their support in the WTO. On a practical negotiating level, the developing countries which have been treated as negotiating leaders both by other developing countries and by the developed are all non-Least Developed, for example India, South Africa, Brazil, Egypt, but also, for the smaller countries, Mauritius. These were included in the "Green Room" procedures in Seattle; they have mobilised support for continuing negotiations since then. If they believe that the initiative is economically damaging (trade diversion) and fails to respect existing agreements, negotiating with them will be more difficult.


  If the initiative is expected to have important developmental effects on the Least Developed Countries, both they and other developing countries need clear analysis of what these will be. Even if the initiative is considered to be entirely within the control of the EC, such an assessment is essential. Without this, business and consumers in Europe, who have the legal right to such analysis, cannot have a credible assessment of impact.

  Such an analysis will require assumptions or forecasts of other EU policy: if all trade barriers are expected to fall within period of 10 years (sometimes argued by DG Trade), then any preferences may give only a temporary advantage: a new producer has compensation for the costs of setting up, but an uncompetitive one does not have a long-term advantage. This implies one pattern of reactions in Least Developed Countries and in other developing countries. If barriers are assumed to be long-term, this implies different reactions. Does including sugar in the initiative indicate that the European market for sugar and all CAP products will be completely uncontrolled after 2003, or will there be the current system plus Least Developed exports?

  In the context of both the multilateral trade regime and the bilateral relations of the EU with the ACP countries, it is now necessary to clarify what types of special treatment for developing countries can be treated as purely concessional, a benefit to the recipient with "essentially" no costs except to the donor (aid, at one extreme, but also benefits given equally to all developing countries), and types that affect other countries, whether directly, through significant trade diversion, or indirectly, by damaging confidence in the predictability of international rules. Non-Least Developed countries also have legitimate trading interests and risk being alienated by actual or perceived diversion of preferences to the least developed.

  In the context of EU-ACP negotiations, regardless of the legal advice to the Commission on the exact obligations of the Cotonou Agreement, the Commission needs to reconsider its interpretation of what measures require consultation and to accept that most observers interpret consultation as meaning ex ante consultation, not ex post.

1   The argument sometimes used that consideration of food security would prevent this is unlikely to hold: sugar is not normally considered a staple food commodity and producers from non-Least Developed countries will be able to offer a complete service of supplying the local market and obtaining good prices for exports. Back

2   Including indicators for nourishment, education, and mortality. Back

3   Including indicators of export concentration, industrial development, instability of exports and production and population size. Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2001
Prepared 20 February 2001