| 1 March 1999 |
 | |
|
 Discussion Paper 3: The Impact Studies on the Effects of REPAs between the ACP and the EU
| |
This paper should be cited as: McQueen, M. 1999. The Impact Studies on the Effects of REPAs between the ACP and the EU. (ECDPM Discussion Paper 3). Maastricht: ECDPM.
Acronyms
ACP Africa, Caribbean, Pacific
BLNS Botswana, Lesotho, Namibia, Swaziland
CAP Common Agricultural Policy
CARICOM Caribbean Community
CEMAC Communauté Economique et Monétaire de l’Afrique Centrale
CET Common External Tariff
COMESA Common Market for Eastern and Southern Africa
EAC East African Cooperation
ECOWAS Economic Community of West African States
EC European Commission
EU European Union
FTA Free Trade Area
GSP Generalised System of Preferences
IOC Indian Ocean Commission
LDC Least-Developed Country
REPA Regional Economic Partnership Agreement
SADC Southern African Development Community
SPARTECA South Pacific Regional Trade and Economic Co-operation Agreement
UDEAC Union Douanière et Economique de l’Afrique Centrale
UEMOA Union Economique et Monétaire Ouest Africaine
WTO World Trade Organisation
Executive Summary
In November 1998, the European Commission released five studies on the impact on ACP countries of its proposed Regional Economic Partnership Agreements. A sixth one on Francophone West Africa was made available in February 1999. This paper presents the main results of these studies. The studies do not make a clear case for or against REPAs, because:
- the results of the quantitative assessments depend on the many assumptions that were made, many of which were necessarily a simplification of reality (e.g. in assessing the degree of substitution between local goods and imports from Europe, or between imports from Europe and from somewhere else);
- The studies were mainly desk-based, often relying on data that was not readily available or of poor quality, therefore making any quantitative assessment at best a rough estimation of trends;
- the six studies were not easy to compare as they used different methodologies and sometimes different assumptions.
Despite these limitations, some general conclusions can be drawn:
- in most cases, LDCs have little to gain from REPAs. They can keep non-reciprocal trade preferences anyway;
- the replacement of non-reciprocal tariff preferences with the GSP would adversely affect some products exported by some non-LDC ACP countries, but most ACP exports would be barely affected;
- by contrast, the direct or indirect effects of not renewing the commodity protocols could dramatically affect the exports of some ACP countries. However, none of the studies estimated these effects as this would have required separate studies;
- the negative impact on customs revenues varies considerably, but could be substantial for some countries, which may thus claim for adequate financial support;
- lower import taxes would benefit customers as well as importers of capital goods, but it is difficult to say to what extent these welfare gains would offset the losses mentioned above.
Some questions that the studies could not address remain unanswered. For instance:
- it was impossible to measure the eventual growth-enhancing dynamic effects of REPAs on ACP economies (the "lock-in" effect on economic reforms in the ACP, the positive "stabilisation" impact on investment, etc.).
- the testing of alternative trade liberalisation scenarios, such as a gradual opening of ACP markets on a multilateral basis, towards Europe and elsewhere, was not required in the terms of reference.
Altogether, these impact studies are an important contribution to the debate on future EU-ACP trade relations. Building on their results, the ACP may now want to carry out their own impact studies, with different approaches. The usefulness of any additional studies will heavily depend on the quality of the work and the capacity of experts who do them, but most of all on the degree of ownership that ACP governments and regional bodies can exert, from the conception of the studies until the dissemination of the final results.
The Controversy about REPAs
To replace existing non-reciprocal trade preferences, the European Union (EU) proposes that "Regional Economic Partnership Agreements" (REPAs) be signed with different ACP regions or countries. REPAs represent a form of free trade agreements in which there would be "more than free trade" (they would include provisions for economic cooperation in other fields). ACP partners in such agreements would retain their current preferential access to European markets, but would have to reciprocate by progressively opening their own markets to imports from Europe on a preferential basis. These agreements would be put in place starting in 2005. A lively debate on the potential impact of such agreements on ACP economies is ongoing.
Optimists stress that REPAs would bring prices down in the ACP - benefiting consumers and importers -, foster trade and other economic reforms, and make the economic environment more conducive to domestic and foreign investment. REPA sceptics fear that they would lead to the closure of many companies in ACP countries, a rise in unemployment and a fall in revenues from import duties.
In fact, ACP countries would encounter all of these effects, to various extents. The question is whether REPAs would, overall, be beneficial or detrimental to ACP economies. So far, we have no examples of REPAs - between the EU and groups of developing countries - from which to draw lessons. The closest examples are the free trade agreements between the EU and some countries in North Africa and the Middle East (Lebanon, Morocco, Tunisia), but it is still too early to measure their economic impacts.
In 1998, the European Commission (EC) contracted six studies on the economic impact of its proposed Regional Economic Partnership Agreements between the EU and CARICOM (plus the Dominican Republic), EAC, the Pacific, SADC, UDEAC-CEMAC, and UEMOA (plus Ghana).
It should be emphasised that the studues were not comprehensive and did not cover all ACP countries and regions. For instance, the SADC study excluded South Africa. Other groupings were not considered (such as COMESA, ECOWAS, and the IOC). One of the six groups that was studied (the Pacific) is not legally, nor technically, a trade region. Finally, Ghana, Nigeria and sixteen least-developed African ACP countries do not formally belong to any of the regions or trade groupings that were studied (see box).
ACP regional groupings in the impact studies
|
CARICOM (Caribbean Community): Antigua and Barbuda, Barbados, Belize, Dominica, Grenada, Guyana, Haiti, Jamaica, St Kitts & Nevis,
St Lucia, St Vincent & Grenadines, Suriname, Trinidad & Tobago. [The Bahamas are members of the Community, but not the common market.
The Dominican Republic is negotiating a FTA with CARICOM].
EAC (East African Cooperation): Kenya, Tanzania, Uganda.
Pacific (no regional organisation): Papua New Guinea, Fiji, Kiribati, Samoa, Solomon Islands, Tonga, Tuvalu, Vanuatu.
SADC (Southern African Development Community): Angola, Botswana, DR Congo, Lesotho, Malawi, Mauritius, Mozambique, Namibia,
Seychelles, South Africa, Swaziland, Tanzania, Zambia, Zimbabwe.
UDEAC-CEMAC (Union Douanière et Economique de l’Afrique Centrale – Communauté Economique et Monétaire de l’Afrique Centrale):
Cameroon, Central African Republic, Congo, Gabon, Chad, Equatorial Guinea.
UEMOA (Union Economique et Monétaire Ouest Africaine): Burkina Faso, Mali, Niger, Côte d’Ivoire, Senegal, Guinea-Bissau, Togo.
Not in regions above: Burundi, Cape Verde, Comoros, Djibouti, Eritrea, Ethiopia, Gambia, Ghana, Guinea, Liberia, Madagascar, Mauritania, Nigeria, Rwanda, Sao Tome & Principe, Sierra Leone, Somalia, Sudan.
|
The terms of reference did not require a comparison with existing agreements. Instead, the studies aimed to:
i. present a description of the present situation with respect to regional integration, trade policies and trade structure;
ii. identify one or more liberalisation plans for the post-Lomé period;
iii. and in particular, assess the economic impact of the proposed liberalisation plans.
In assessing the economic impact, the consultants were asked to compare a "base level" scenario —without a REPA — with the REPA scenario. The base level scenario assumed that after 2005, the ACP countries concerned would obtain EU preferences only from the GSP (which, for the least developed countries, would be Lomé-equivalent for all products not subject to tariff quotas) and that the special preferences for Protocol products (sugar, bananas, beef and veal) would cease. The consultants were also advised to assume that the GSP would be improved so that the current gap between the GSP and Lomé rates would be reduced by 25 percent across the board, but for the same list of GSP products. The REPA scenario assumed that it would be compatible with Article XXIV of GATT and would therefore cover ‘substantially all trade’, with ‘no complete sector excluded’ and completed in a period exceeding 10 years ‘only in exceptional cases’. ‘Substantially all trade’ was assumed to cover 80 percent of actual trade in 2005 and 90 percent of actual trade flows by the end of the transition period.
Estimating the general effects of a REPA
One of the key tasks for the studies was to assess the extent to which the abolition, by ACP countries, of duties on imports from the EU would lead to a rise in imports from the EU, and the size of the costs and benefits which would rise from this. It is therefore important to appreciate both the strengths and weaknesses of the models used to assess these effects.
- The elimination of tariffs by an ACP country on most goods imported from the EU creates the opportunity to decrease the domestic price of these goods and of locally produced competing goods. This decrease benefits consumers of finished goods and producers who purchase imported (or competing locally produced) intermediate goods used in production and investment goods. Against these welfare gains must be set the costs to the economy in terms of the loss of tariff revenue to the Government and the adjustment costs both of re-allocating resources from producers displaced by lower cost imports (the ‘trade creation’ effects) and the development of alternative sources of Government revenue.
The extent to which prices actually fall depends on the competitive structure of the domestic market and the pricing behaviour of EU exporters. In common with most empirical analysis in this field, the studies have little or nothing to say about the extent to which the goods markets in the countries studied are competitive or not (although it might be assumed that being small markets they could often be subject to domination by a small number of firms). They confine themselves either to producing alternative estimates based on the presence or absence of increasing supply costs from EU exporters, or assume a given proportion of the decrease in tariffs is passed on to purchasers (70 percent in the case of the UDEAC study).
- ‘Trade creation’ gains arising from the displacement of inefficient local production depends on the extent to which imports can substitute for local production. Most of the studies - apart from the one on UEMOA - assume perfect substitutability in goods defined at generally high levels of aggregation, in which case they will tend to exaggerate the trade creation effects (and adjustment costs) of a REPA if locally produced goods, defined more precisely, are non-competing products.
The alternative approach adopted in the UDEAC study, is to argue that imports from the EU and locally produced goods do not compete, in which case there is no trade creation effect. In all of the studies very general assumptions were made (because of the absence of data) on the responsiveness of import demand to a change in price, yet the results are highly sensitive to the values selected for this parameter.
- A REPA can also lead to ‘trade diversion’ costs to the importing country, if the EU is not a globally-efficient producer, and the abolition of duties on goods from the EU (but not the rest of the world) leads to existing trade from more efficient producers in the rest of the world being switched to the higher cost suppliers in the EU. Again, the lack of adequate data forced researchers to make very general assumptions about the magnitude of this potential substitution effect and these assumptions differ between the studies. Some studies (SADC, Pacific, CARICOM/DR) assume general values (such as -1 or -3) for the ‘elasticity of substitution’. The EAC study assumes no trade diversion where the EU is the dominant supplier (and therefore assumed to be world competitive) and complete trade diversion where the EU is the minority supplier. The UDEAC study argues that, because of the preponderance of EU imports, it is likely that goods from the rest of the world are either non-competitive with EU imports, or have a very strong competitive advantage or are exempt from duties, so that there is little or no trade diversion effect. As before, these strong assumptions affected the results.
- A fourth effect of a REPA could be to substitute imports from the EU for regional production. This effect is only seriously considered by the EAC and UEMOA studies. The other studies either do not analyse this (SADC and CARICOM/DR) or dismiss this possibility as irrelevant because of the very low volume of regional trade (UDEAC, Pacific).
On all the trade effects mentioned above - trade creation and trade diversion - it should be noted that the UEMOA study assesses the impact of a REPA based on two sets of assumptions: one with perfect substitutability between imports from the region, from the EU and from the rest of the world, and one with imperfect substitutability. The authors consider the latter as the most realistic approach, but they do provide quantitative estimations for both sets of assumptions.
- All studies emphasise the adverse effects of a REPA on Government revenues. Again, however, care must be taken in interpreting this information. General import liberalisation is occurring in any case (although to varying degrees) in all of the ACP countries studied and this will entail a shift away from a dependence on import taxes in sources of government revenues whether or not a REPA is concluded with the EU. Also, not all studies use actual tariff rates (as collected) for their calculation, but instead rely on official rates, which can be quite different from the former.
- The adverse effects on ACP exports of not concluding a REPA are largely confined to the non-least-developed ACP countries. Indeed, the EU has improved its GSP offer for all LDCs to be Lomé-equivalent for all products not subject to tariff quotas, although Madagascar would lose its preferences for sugar, beef and veal, Malawi, Tanzania and Uganda would lose preferences for sugar. None of the studies have estimated the direct and indirect effects on the relevant ACP countries of not renewing the Protocols, as the effects on the major exporters would be so wide ranging as to require separate studies. The effects on other products have been estimated either in terms of the decrease in demand in the EU market consequential on the rise in the price of ACP exports as they move to higher GSP tariffs (EAC, CARICOM/DR studies), or (less satisfactorily) as an estimate of the ‘value of access to the EU market’ calculated on the basis of assumed GSP tariffs in 2015 (SADC studies). For the CEMAC and Pacific studies, non-protocol exports have either a zero or very small margin of preferences under Lomé and its removal would not affect their exports to the EU. The UEMOA study does not analyse the consequences for the UEMOA countries and Ghana of not signing a REPA.
These substantial differences between the studies mean that they are not directly comparable (because they use different methodologies) and that the quantitative effects of the studies should be interpreted as broad orders of magnitude and not precise estimates of effects. Further detailed work will be required in individual ACP countries to identify, as far as possible, the effects of REPA and non-REPA scenarios on particular sectors, intra- and inter-regional trade, and any alternative regional groupings that the ACP may prefer.
Each of the studies is now presented in a summarised form, with emphasis given to the feasibility and likely impact of a REPA on the region concerned.
SADC
(prepared by Imani Development International Ltd, Mauritius)
SADC comprises 14 countries. Seven are classified as least-developed (Angola, DR Congo, Lesotho, Malawi, Mozambique, Tanzania and Zambia). One (South Africa) is an industrialised country and is currently negotiating a bilateral free trade agreement with the EU. The remaining developing countries are Botswana, Mauritius, Namibia, Seychelles, Swaziland and Zimbabwe.
The report emphasised that while, in the long run, SADC would be a natural partner for the EU in a REPA, institutional, political and economic constraints make it difficult, if not impossible, for SADC to conclude a REPA in the time scale envisaged by the EU. For example, the Trade Protocol signed by eleven member states in August 1996 to establish a free trade area within eight years has only been ratified so far by five countries. It therefore seems unlikely that the FTA will be established before 2008, while the Trade Protocol makes no mention of a movement towards a common external tariff (CET). In addition, although many countries have liberalised their trade and foreign exchange regimes, this has been undermined by weak fiscal polices. There is also a weak institutional structure, which would make it very difficult to negotiate a REPA. Much of the technical work on a REPA would have to be carried out at the sectoral level, but sector responsibilities are allocated to different SADC countries. For example, the Industry and Trade Sectoral Co-ordinating Unit is in Tanzania, while the Finance and Investment Unit is located in South Africa. The resources available to these Units are highly variable, and national considerations often take precedence over regional issues. The SADC Secretariat is small with no specialists in trade negotiations and it lacks strategic responsibilities.
In view of these constraints and the very diverse nature of the SADC countries in terms of their levels of economic development, size and politics, the report suggested the following. First, the EU-South African FTA be extended to cover all of the SACU countries (i.e. Botswana, Namibia, Lesotho and Swaziland). Second, that the LDCs would be better off without a REPA since they already have Lomé-equivalent access under the GSP (except for products subject to tariff quotas). Third, that a REPA be concluded (either individually or collectively) with Mauritius, Seychelles and Zimbabwe. A SADC-wide REPA might then be possible commencing in 2010 and fully implemented between 2020 and 2025. The study could, however, point out that this proposal would weaken moves towards regional integration as imports from the EU would enter SADC countries under different trade regimes, and thus rules of origin and border controls would have to be tightened to preserve the integrity of individual tariff structures. It would also weaken incentives to investors to locate in the SADC region since the regional market would be more easily served from the EU, the common denominator in these increasingly complex trade arrangements.
The modelling of the effect on SADC imports assumed that the SADC countries would reduce their current MFN tariffs by 30 percent by 2015 and postulated two scenarios. The first assumed that imports from the EU and the rest of the world were supplied at constant costs, and the second assumed a rising EU supply price proportionate to the decline in the tariff (unitary elasticity of supply). No allowance was made for the exclusion of sensitive products from the REPA and so the estimated increase in imports is probably a ‘worse case’ scenario. The modelling of SADC exports to the EU were based on the additional assumption that EU tariffs (and therefore the preference margins) are reduced by 30 percent from their current levels. One strong assumption, which appears to run contrary to the Commission’s guidelines, was to assume that the Protocols would continue whether or not there was a REPA (page 112). As a result, there would be no difference in the value of access to the EU market for the LDCs, whether or not they concluded a REPA. For the non-LDCs, the value of access under the non-REPA scenario was computed by comparing EU-improved MFN with the improved GSP rates, while under the REPA scenario, the value of access was the difference between duty-free access and the improved GSP rates. Apart from not estimating the impact of losing Protocol preferences, this ‘value of access’ approach also assumed that all of the tariff revenue foregone by the EU under preferences (GSP or REPA) accrues to SADC exporters, whereas it is much more likely to be shared with EU importers, distributors and (possibly) EU consumers. It does not quantify the decrease in demand for SADC exports as a result of the loss of preferences by non-LDCs in the non-REPA case.
Trade creation gains are very small, ranging from zero for five countries to $3.7m (0.02% of GDP) for Angola. Trade diversion losses are, in all cases, substantially greater, especially for the Seychelles (2.0% of GDP) and Mauritius (1.7% of GDP). The substantial excess of trade diversion loss over gains from trade creation also supports the conclusion that the LDCs would, at least for the present, be better off without a REPA. The impact of the lower cost imports on SADC producers of manufactures was difficult to quantify since opinions differed on whether EU goods competed with SADC goods, while EU goods would have to compete with highly competitive imports from Asia. The major impact was thought to lie in the agricultural sector where there is a fear that subsidised EU products entering duty free would displace local production and undermine regional trade. There is also a fear that improved access of South African exports into the EU under a bilateral FTA would erode the preferences of other SADC countries. Given the importance of the agriculture sector and agricultural exports to these countries, further research is clearly needed on these and Protocol issues. Estimated losses in Government revenues varied between countries, ranging from negligible amounts for the BLNS countries (probably understated) to losses of 9 percent and 8 percent for Mauritius and Tanzania and 30 percent for the Seychelles.
In the longer term, the report sees positive dynamic benefits in a SADC-wide REPA since it would ‘lock’ the SADC countries into their own free trade area and underpin regional integration. More liberal access to the wider SADC market would enable greater economies of scale and induce EU manufacturers to establish branch plants in SADC countries either for finished goods or as sub-contractors for component inputs for SADC producers. A flight of labour-intensive industry from South Africa to neighbouring countries is also envisaged, providing an investment boom similar to that experienced by Spain and Portugal when they joined the EU.
EAC
(prepared by CREDIT, School of Economics, University of Nottingham)
As the prospects for an EAC region with zero internal tariffs on most goods and a CET by 2005 are good, this study is optimistic about the feasibility of a REPA being implemented during the period 2005-2015.
Substantial progress has been made in trade liberalisation in recent years in the EAC as a result of different factors, including aid conditionality and the Cross Border Initiative (CBI). A REPA is seen to be useful to ‘lock in’ trade liberalisation, especially perhaps in Kenya, where import restrictions have been introduced and relaxed, not in accordance with any trade strategy, but in response to internal and external imbalances.
The report does not discuss the institutional capacity of the EAC to negotiate an agreement, and emphasises that a REPA could increase tension between Kenya and its two partner countries. It is also recognised that a separate REPA with SADC could create difficulties for Tanzania as it (unlike Uganda and Tanzania) is a member of both regional organisations.
Since Tanzania and Uganda are LDCs, the study assumed that they would not suffer a loss of preferences on their exports in the non-REPA scenario. The report therefore concentrates on the effect of a REPA on imports, with a computation of the ‘GSP-only’ effect on Kenya’s exports. Trade statistics at the two digit HS level and tariff data were only available for Uganda and Tanzania. Therefore, while estimates of trade, revenue and welfare effects could be made for the two countries, only rough estimates could be made for Kenya on the basis that its patterns of trade with the EU and the rest of the world were similar to the other two countries. Information from Uganda was also used both to select tariff rates of 15 percent and 7 percent for the CET by 2005, and to choose the ‘sensitive’ products which would be excluded both from EAC and REPA trade liberalisation. It was also assumed that a tariff rate of 12 percent would apply to trade in these sensitive products both with the EU and within the EAC, although it is recognised that the EAC countries may not wish to give the EU a preference over the rest of the world on these sensitive products.
In those sectors where the EU accounted for more than half of EAC imports, and where the study assumed that only welfare increasing trade creation would therefore occur, imports from the EU increased by 16 percent for Tanzania and 23 percent for Uganda (the increase in net welfare is less because of the associated loss of tariff revenue). In sectors where the EU was not the dominant supplier, welfare increasing trade creation was overwhelmed by trade diversion (because the authors assume all imports from the rest of the world is replaced by higher cost EU imports). The third potential effect, which the study refers to as ‘trade deflection’, is where EAC imports switch from less efficient EAC to more efficient EU suppliers as a result of a REPA (i.e. decreasing regional trade). This effect was particularly significant for Uganda, given its heavy dependence on Kenya for imports in some sectors. The overall effect was that, although consumer and national welfare increased as a result of trade creation and deflection, the potentially large costs of trade diversion from more efficient non-EU sources meant that net welfare could fall in both countries as a result of REPA, although this loss was small relative to GDP. The direct impact on Government revenues was likely to be large and estimated at 20 percent for Tanzania and 16 percent for Uganda. The rough estimates for Kenya also suggested a net welfare loss on trade with the EU, and to this must be added the potential loss of Kenyan exports to the EAC from trade deflection which, at its maximum, could offset the gains to Kenya’s exports to the EU from a REPA compared to GSP treatment. Loss of tariff revenue could also be significant at 12.3 percent of total tax revenue.
All three trade effects were concentrated in the manufacturing sectors (unlike the SADC study, which emphasised the impact of EU imports on agriculture) and the adjustment problems were expected to be modest. First, because this sector only accounted for less than 5 percent of output and employment in all three countries. Second, the substantial decrease in effective rates of protection for import-competing manufacturing activities implies a corresponding increased incentive to produce exportables, and since the latter are likely to be more labour intensive than the former, the REPA could increase both employment and investment. More generally, the study demonstrated that the revealed comparative advantage of all three countries lies in agriculture, agro-business and relatively low value added activities using local materials.
UDEAC-CEMAC
(prepared by Groupe Planistat, Belgium)
To an even greater extent than the SADC study, the report emphasised two fundamental problems in regional development. First, the limits to common regional interests within CEMAC and, second, the lack of a technical capacity to design and implement effective reforms to meet national and sectoral needs. The result is a wide variety of barriers and constraints to CEMAC’s internal and external trade. In this context, the process of moving towards a REPA is regarded as beneficial, as it would create an impetus for trade policy reforms. However, doubts were cast on the capacity of CEMAC to negotiate a REPA and, even if negotiated, of implementing it at the regional, national and local levels.
In terms of the impact of a REPA on exports, only Cameroon obtained significant benefits, largely because exports of bananas would probably cease under the GSP-only scenario (while it is considered that they would increase in a REPA). To a lesser extent, exports of aluminium, cocoa, cotton fabrics, beans and plywood could also be adversely affected. For the other countries, exports are subject to zero, or very low, tariffs under the GSP and would be unaffected.
The impact of a REPA on imports was analysed by identifying the sectors with the highest tariffs which accounted for 10 percent of imports from the EU, and excluding these sensitive products from this analysis. The increase in imports from the EU from the remaining 90 percent was then calculated on the basis of 70 percent of the decrease in tariff being passed on to consumers as a fall in price and an elasticity (response of demand to a change in price) of import demand of unity. Under the assumptions outlined above, there were no trade diversion or trade creation effects. Any loss of domestic production was expected to be offset by consumers increasing their demand for locally produced goods as their real income increased with falling import prices. The results of this analysis indicated that imports to the region from the EU would increase by just over 5 percent.
The study emphasised the impact of REPA on Government tax revenues on two groups of countries. The first group of countries - Cameroon, Gabon and Equatorial Guinea - facing a decrease in government revenues of a maximum of 8.2 percent were expected to be able to adjust their revenue systems over time, though this would be slow and difficult to achieve. The second group of countries - Congo, Central African Republic and Chad - faced a minimum decrease in Government revenues of 14.1 percent. These are poor countries with low growth rates and severe problems of government finance. The report concluded that, given their very limited capacity to design and implement fiscal reforms, it was entirely possible that alternative tax polices consequent on a REPA could create greater distortions than the tariffs they replaced. For these countries, a REPA was regarded as ‘at best a very low priority and at worst counter-productive’. Concluding a REPA with the first group of countries but not the second would, however, necessarily increase barriers to regional trade.
CARICOM/Dominican Republic
(prepared by the Institute of Development Studies, University of Sussex)
The sixteen member states of CARICOM have pursued a staged approach to integration. Haiti became a formal member in 1997 and the Dominican Republic is at an advanced stage of negotiating an FTA. Currently, CARICOM is negotiating the entry into force of eight Protocols which seek to create a single market by 1999, and is scheduled to introduce a maximum CET of 20 percent. CARICOM is, therefore, probably the most integrated of the ACP sub-regions and has institutional mechanisms capable of negotiating with the EU.
The most important constraint on introducing a REPA identified by the report was the need to coordinate this with hemispheric trade negotiations. CARICOM already obtains special preferential access to the US and Canadian markets under agreements covered by WTO waivers, and negotiations are under way, or are planned, with Colombia, Venezuela, Costa Rica, Mexico and ultimately with the rest of the hemisphere under the Free Trade Area for the America (FTAA). To avoid major and unnecessary costs, both in negotiating agreements and to CARICOM’s trade, it is essential that any negotiation of a REPA takes place within the framework of CARICOM’s other trade negotiations and not necessarily within the 2005-2015 timetable.
CARICOM/DR exports to the EU are heavily concentrated on sugar and bananas, where preferential access to the EU depends on the maintenance of the Protocols, since large parts of their production are generally regarded as uncompetitive. Modelling the effects of the removal of the Protocols would require detailed fieldwork, but the report indicated that the magnitude of the effects on sugar exports could be of the order of ECU 132 million (7.4 percent of exports to the EU). This can be compared to the modelled difference between REPA and GSP covered exports for non-Protocol exports of ECU 2 million. The likely treatment of these products within and outside a REPA, both in terms of EU trade policy and WTO disciplines, therefore deserves further research.
In contrast to the previous studies mentioned, the CARICOM/DR report matched the trade data for imports from the EU as closely as possible to the tariff data. This was achieved by focusing on the 91, 8-digit items on which imports from the EU exceeded ECU 1 million, after excluding sensitive (largely agricultural items) which comprised less than 10 percent of trade. The results indicated that trade creation would substantially exceed trade diversion, although the former was estimated at only 10 percent of CARICOM/DR imports. Also the extent to which EU goods could displace North American goods may have been underestimated in the model. The sectors most affected by trade creation (and therefore involving potential adjustment costs) were ships and boats, machinery and mechanical appliances; with iron and steel and articles of iron and steel affected to a much smaller extent in absolute terms.
The trade creation and diversion effects depend upon prices falling as a result of duty-free entry. To test whether this was likely, the study assumed that a price reduction was more likely for products for which the EU was able to increase supply competitively, as shown by rapid growth rates from substantial initial values. Most products (excluding sensitive items) passed this test and the report concluded that it seems reasonable to believe that there would be good commercial reasons for decreasing prices to increase market shares.
The CARICOM/DR countries rely heavily on import taxes for revenue, and a detailed analysis indicated significant losses from a REPA, especially for the smaller Eastern Caribbean states. Since a REPA would probably be introduced along with hemispheric trade liberalisation, this particular adjustment cost will need to be given special attention by aid donors, including the EU.
Pacific
(prepared by the Netherlands Economic Institute)
The Pacific ACP countries (PACP) are not, as yet, organised as a regional group in the same way as the other sub-regional groups selected by the Commission. The report suggests that the most suitable body for negotiations would be the Melanesian Spearhead Group (MSG): Papua New Guinea, Fiji, Solomon Islands and Vanuatu. An FTA could be negotiated with these four countries, with a more general Partnership Agreement covering all 8 countries (although it is unclear how this would operate). To achieve this, the lack of institutional capacity of the MSG would have to be addressed. An FTA would also need to take into account the trade interests of the other 8 member states of the South Pacific Forum, notably Australia and New Zealand. These two countries, in particular, would probably require the same access to the PACP countries as granted to the EU as a condition of continued duty-free access to their markets under the present SPARTECA preferential agreement.
The termination of the Sugar Protocol and the preferences for canned tuna would have large implications for Fiji. Sugar accounts for 34.5 percent of Fiji’s export revenues and 11.5 percent of employment, and the EU accounts for 68 percent of export earnings from sugar. High EU sugar prices have encouraged excessive expansion and inefficient production. A withdrawal of preferential access to the EU market would lead to a substantial decrease in output and an even greater fall in earnings; with all that this would entail in terms of direct and indirect effects on income, output and employment in the economy and budgetary effects for the Government. Even with the continuation of the Protocol, Fiji will have to adjust to a more competitive EU and world market. The report therefore calls for a thorough investigation into the sustainability and viability of the industry over the next 25 years. Canned tuna benefits from a 24 percent margin of preferences and a derogation from the rules of origin, but the canning industry in Fiji is an inefficient, high cost and loss making public enterprise. Nevertheless, the fish industry is regarded as a sector where the PACP countries have a potential for sustainable processing, but requires financial and technical assistance. For the other 7 PACP countries, the termination of Lomé preferences would have no significant effect. Five countries are eligible for the EU’s enhanced GSP, while Papua New Guinea and Tonga only receive a small margin of preferences under Lomé.
The effects of a REPA on PACP imports are also likely to be small. The EU accounts for only a small proportion of imports, concentrated in below average tariff rate categories, and in products essential for local production and not manufactured by local producers. This equally implies that the efficiency and dynamic gains from a REPA are also likely to be very small.
The report therefore envisages that the absence of a negotiating capacity by the PACP will seriously delay the negotiation of a REPA, but once an FTA between the MSG and the EU has been agreed, then there are no particular reasons for introducing a lengthy phasing period.
UEMOA/Ghana
(Prepared by the Centre d'Etudes et de Recherches sur le Développement International,
Université d'Auvergne, Clermont Ferrand)
UEMOA is a monetary union and aims to become a customs union in 2000. It already has regional institutions to coordinate economic policies, mainly through the adoption of budgetary convergence criteria. The region comprises two non-LDCs (Côte d’Ivoire and Senegal) and six LDCs (Benin, Burkina Faso, Guinea Bissau, Mali, Niger and Togo). Together with Ghana and Nigeria, all belong to the larger ECOWAS, a grouping that has achieved very little success so far in implementing its economic integration agenda. The terms of reference of the study provided for assessing the impact of a complementary REPA between the EU and Ghana, a non-LDC whose neighbours are all UEMOA countries.
The study does not provide a detailed analysis of the feasibility of a REPA. However, while assuming that UEMOA will become a customs union on January 1, 2000 as planned, it argues that this process - a prerequisite for a REPA - may be more difficult than implementing the REPA itself. The main problems are the remaining disparities in protection levels and the budgetary costs of removing intra-regional tariffs. Tax losses would be particularly heavy for the LDCs, who trade most with other UEMOA countries. On the whole, customs revenues make up 40 percent of UEMOA governments' aggregate revenues. As for Ghana, it has lower tariff levels than UEMOA countries and depends on customs revenues for 20 percent of its total government revenues.
Some elements in the study suggest, but only indirectly, that UEMOA may be better suited than other ACP regions for implementing a REPA with the EU along the proposed timetable. In particular, unlike UDEAC - and while keeping in mind the difficulties mentioned above - UEMOA has managed to build on its common currency to increase economic cooperation. The structure of agricultural production in UEMOA countries is largely dissimilar to that of the EU – by contrast with many SADC countries — which makes them less vulnerable to EU subsidised exports. There would therefore be less sensitive issues in the negotiations than between the EU and SADC.
The study does not assess the impact of the non-REPA scenario (where Côte d'Ivoire and Senegal would be transferred into the GSP). For the REPA scenario, it assumed the agreement would be established between 2005-2017, with a gradual reduction in tariffs (of about 8 per cent per year across the board, instead of back-loading the most sensitive liberalisation steps) and the exclusion of "strategic" consumer goods. Welfare gains and losses were measured using a partial equilibrium model, based on the assumption that UEMOA succeeds in completing its customs union by 2000. Ghana is not included in the calculations. Substitutability between goods imported from UEMOA countries, from Europe and from the rest of the world is assumed to be imperfect (although the perfect substitutability assumption was tested as well). The lack of adequate data did not allow the consultants to take account of revenue effects or the impact of change in trade policy, but they did control for the difference between official tariff rates and actually collected tariffs. Two alternative sets of assumptions were therefore used: the "high hypothesis" (perfectly substitutable goods, calculations based on official tariff rates) and the "low hypothesis" (imperfectly substitutable goods, calculations based on actually collected tariffs).
The authors considered the latter as the most realistic. The results of the model show that:
- Substantial trade diversion will occur, with imports from Europe displacing imports from the rest of the world and in some cases from the UEMOA countries.
- Efficiency gains for consumers (in terms of lower prices) vary between 0.29 to 1.35 percent of GDP for each country (0.81 to 2.52 percent in the high hypothesis). Senegal benefits most because Europe already makes up a substantial share of its imports.
- Fiscal losses from 2017 onwards vary between -0.20 and -0.84 percent of GDP across countries (-0.59 to -2.65 percent in high hypothesis). Here again Senegal is most affected for the same reason as above. For UEMOA as a whole, this represents an average 44 billion FCFA per year over the transition period 2005-2017 (129 billion FCFA in high hypothesis) and 81 billion FCFA per year beyond 2017 (235 billion FCFA in high hypothesis).
The authors warned that these results must be taken with great caution, because (i) the partial equilibrium model, where budgetary constraints are not taken into account, will systematically exaggerate gains and losses, (ii) some customs data are not totally reliable and (iii) the length of the period studied makes such results indications and not precise measurements. They also stressed that, over ten years, UEMOA countries would have to radically reform the structure of their fiscal regime, which they have not managed in almost 20 years of structural adjustment. Finally, the study emphasised that financial measures by the EU would be necessary to offset economic and social costs and to ease the budgetary pressure. Fiscal losses induced by a REPA, while relatively high for UEMOA countries, would be small in absolute terms for the donor.
Future Studies
The EU’s negotiating mandate makes provision for the ACP to carry out their own impact studies, and these could usefully build on the present studies. For example:
- the ACP will probably have different ideas on appropriate sub-regional groupings, in particular, including Nigeria in a West African group.
- The Commission’s terms of reference excluded an examination of the possible adverse effects of REPAs on trade between sub-regional groups and measures which would be required to minimise these effects.
- The resources available for the studies meant that most of the research was desk-based in Europe and, as indicated in this paper, based on strong simplifying assumptions. Much more fieldwork in ACP countries is needed to assist ACP negotiators assess the potential impact of alternative forms of a REPA on individual ACP countries and sectors to ensure the maximum benefits and minimum costs of an FTA.
- It should be recognised that the net static effects estimated in the studies are (excluding the effects on Protocol products) usually a small proportion of GDP. The potentially more important effects are the growth enhancing dynamic effects, and these could be maximised by using an FTA to go beyond WTO disciplines.
- Additional studies could include areas such as competition policy, enhanced guarantees of access to the EU market through the elimination of anti-dumping and safeguard measures, agreements regarding the implementation of technical, sanitary and phytosanitary standards which can act as barriers to trade, and enhanced disputes procedures on the application of the agreement (for example as incorporated in the EU agreement with Morocco).
- Detailed studies will also be required to identify the measures and EU assistance which will be required both to implement the REPA and to enable ACP countries to smoothly adjust the structure of their economies and Government revenues to the new economic environment.
- The strong assumption by the Commission that the Protocols will be terminated if a REPA is not signed also needs to be challenged, and studies carried out on the status of the Protocols and their future in the world trading system.
|
|