WPS4627 Policy ReseaRch WoRking PaPeR 4627 Economic Partnership Agreements and the Export Competitiveness of Africa Paul Brenton Mombert Hoppe Richard Newfarmer The World Bank Poverty Reduction and Economic Management Network International Trade Department May 2008 Policy ReseaRch WoRking PaPeR 4627 Abstract Trade can be a key driver of growth for African to regional integration, including both tariff and non- countries, as it has been for those countries, particularly tariff barriers; improve trade facilitation; and define in East Asia, that have experienced high and sustained appropriate most favored nation services liberalization. rates of growth. Economic partnership agreements At the same time, African countries will need to reduce with the European Union could be instrumental in a external tariff peak barriers on a most favored nation basis competitiveness framework, but to do so they would to ensure that when preferences for the European Union have to be designed carefully in a way that supports are implemented after transitional periods, they do not integration into the global economy and is consistent lead to substantial losses from trade diversion. This with national development strategies. Interim agreements entails an ambitious agenda of policy reform that must be have focused on reciprocal tariff removal and less backed up by development assistance in the form of "aid restrictive rules of origin. To be fully effective, economic for trade." partnership agreements will have to address constraints This paper--a product of the InternationalTrade Department, Poverty Reduction and Economic Management Network-- is part of a larger effort in the department to contribute to the discussion of trade policy, competitiveness and export growth. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at pbrenton@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Economic Partnership Agreements and the Export Competitiveness of Africa1 Paul Brenton, Mombert Hoppe and Richard Newfarmer International Trade Department The World Bank MSN MC2-201 1818 H Street St. NW Washington DC 20433 Tel: (202) 473-4255, Fax: (202) 522-7551 E-mail:pbrenton@worldbank.org, mombert.hoppe@gmail.com, rnewfarmer@worldbank.org 1The views expressed in this paper are those of the authors only and should not in any way be attributed to the World Bank. We are very grateful for the comments of Shuby Andriamananjara, Gerrit Faber, Larry Hinkle, Elke Kreuzweiser and Jan Orbie. 1 1. Introduction Trade has been a key driver of growth in countries that have been successful in achieving high rates of growth over the past three decades, such as those in East Asia. If they are to accelerate their growth, trade will have to perform the same role for African countries. Designing policies that promote trade and trade competitiveness must therefore be at the heart of growth strategies for African countries. Economic partnership agreements (EPAs) with the EU could be instrumental in a competitiveness framework, but to do so they have to be designed carefully. Any successful EPA will need to take into account critical features of the new global economy: · intense new competition emanating from large developing countries such as China, India and Brazil; · simultaneous rapid growth of these economies opening up new dynamic market opportunities; · explosive growth of services trade, creating new opportunities in the global market and offering new avenues for diversification away from primary commodities; · the increasing importance of domestic institutions, policies and infrastructure in affecting productivity and the ability of a country's firms to compete in international market. In this new global context, the EPAs create an opportunity to undertake domestic reforms that support global competitiveness. A broad template for such reforms would include (i) reforms that improve the incentive framework especially by reducing tariff and non-tariff barriers to remove bias against exporting, (ii) increase access to and lower the costs of backbone services, such as telecoms, transport, energy and finance and (iii) address non- tariff barriers and weak trade supporting institutions, including standards, customs, and trade promotion agencies and design effective mechanisms to assist the restructuring of domestic firms with the potential to be globally competitive. The EU can contribute by helping to design and support African countries in effectively implementing the provisions of a pro-development EPA. In this context "Aid for trade" associated with EPAs (and the multilateral discussions) can help build infrastructure and the institutional capacity that is necessary to remove NTBs, facilitate trade, regulate services and support trade promotion and firm restructuring. The first order of business is to reduce barriers to regional integration and allow African firms the opportunity to exploit nearby markets as a springboard to the global market. Typically these barriers also directly limit access to the global market. EPAs can play an important role in leveraging coherent regional reforms and in putting in place mechanisms that allow for coordinated development support from EU countries. The priority of delivering effective regional integration was recognized in the initial objectives of the EPA negotiations and was strongly identified as such by the EU Trade Commissioner in several speeches. In a context of wanting to have in place WTO consistent agreements when the Cotonou Agreement expired, the EU pushed hard in 2007 to define interim agreements that focus 2 primarily on reciprocal tariff reductions to satisfy GATT Article XXIV requirements.2 The interim agreements also include an important relaxation of the rules of origin for certain products, but especially clothing, which had previously restricted preferential access to the EU market for African exporters. A number of African countries signed these interim agreements either collectively (as with the EAC) or individually. Other countries are expected to sign shortly, while other countries such as Senegal and Nigeria have been strongly critical of the interim agreements and have expressed reluctance to signing. If EPAs are to realize their development potential, it is important that these interim agreements do not become de facto EPAs -- otherwise the opportunity for such agreements to address constraints to competitiveness and integration into global markets and to be a key mechanism in supporting development in Africa will be lost. Bilateral reductions of tariffs is unlikely to be effective in driving rapid growth to the extent they leverage only minimally the domestic and regional reforms that are necessary to overcome the barriers that limit integration into the global economy. Many African countries have seen a very weak supply response to preferences in the past and this is unlikely to change unless these supply constraints are ameliorated. Worse, there is a real risk that removal of tariffs against the EU without broader tariff reform and in the presence of substantial NTBs to regional trade will lead to economic losses for African countries. Further, there are real concerns that the current patchwork of agreements will undermine regional integration. With the exception of the EAC, countries in the same regional bloc have signed agreements with the EU that have different product exclusion lists which will necessitate strict controls on the movement of EU products within regional groupings to ensure that a product exclusion in one country is not undermined by preferences for the same product in a partner country.3 The EU Commission has asserted that these WTO-consistent market access agreements will be a stepping stone to deeper agreements in the future. However, whether the second stage of such a process will actually materialize remains to be seen as African countries may well lose the bargaining power domestically to obtain progress on competitiveness issues that came from market access concessions.4 This paper elaborates on ways EPAs could be designed to drive growth through effective regional integration and integration into the global market. A first section highlights the trade challenge that African countries face and makes the point that preferences have not been effective in preventing the substantial decline in world market share of African countries. A second section describes opportunities that the new wave of globalization is opening up to African countries. A third section briefly discusses what this means in terms of a strategy to attain global competitiveness and suggests design features of an EPA that would complement and support competitiveness reform programs. 2These require that the parties to a free trade agreement remove tariffs on "substantially all" (interpreted by the EU to mean 80% of mutual trade) trade under a defined and reasonable timetable. 3See, Stevens et al (2008) 4See, for example, Francesco Rampa, "Love is blind"? A two-stage agreement risks being an `EPA tight' rather than an `EPA light', ECPDM, http://www.acp-eu-trade.org/library/files/Rampa_EN_1107_ECDPM- ICTSD_Love-is-blind-A-two-stage-agreem%E2%80%A6.pdf 3 2. The trade challenge facing African countries The trade preferences that have been granted to the ACP countries by the European Communities under the Yaoundé, Lomé, and Cotonou Agreements for the last 30 years will expire in 2008. Negotiations for a new WTO consistent trade agreement have been taking place for the last 5 years. To limit the demands of negotiating agreements with each ACP country the EU has pursued negotiations to establish six Free Trade Areas (FTAs). To date, negotiations have advanced only slowly and there is considerable debate as to how EPAs should be designed and what elements and commitments they should contain. Before proceeding to these issues it is important to reflect upon the impact of current and previous EU development agreements. We focus our description of past trends on Eastern and Southern Africa (the ESA-region in the EPA negotiations). The trade challenges that we highlight for ESA are probably broadly similar to those facing most of the other EPA regions. The average share of global merchandise exports of the ESA countries was small 25 years ago -- but has by now fallen to half its original level (Figure 1). Most of the decline occurred in the 1980s with a leveling off in the 1990s but little subsequent rebound. The performance of the average ESA country stands in stark comparison with the average of a group of 16 fast growing countries5. It is the trade and growth performance of these countries that African countries will have to replicate over the next 20 years if the targets that they have set for poverty reduction are to be achieved. On average, the fast-growing countries of the past two decades have seen their share of global exports rise rapidly (a 3- fold increase over the past 25 years), fuelling the sustained growth that these countries have enjoyed. The figure highlights the importance of trade for the fast growing countries and shows the lacklustre export performance of the average country in the ESA region. Given the small (economic) size of most countries in Sub-Saharan Africa, the role that trade can play for the achievement of robust and sustained economic growth is particularly important. The figure indicates the enormous potential for sustained growth of exports from the ESA region. 5This group contains non-oil exporting countries that have grown at an average annual rate of growth of 4.5 per cent of more over the past 25 years. The averages are unweighted so that country size does not influence the measure. The sixteen countries are Botswana, Sri Lanka, Chile, Indonesia, Pakistan, Mauritius, Uganda, Burkina Faso, India, Thailand, Malaysia, Taiwan, Cambodia, Singapore, Korea, Rep., and China. 4 Figure 1: In contrast to the fast growing countries of the past 20 years, Figu cont he grow count he ESA countries have lost share of the global market for merchandise count hav ost of he global or chandi Average share of world merchandise exports (% of US$ current) chandi ur Average HP 16 Averag Average ESA 0.04% 1.2% Average High Performing 16 er gh 16 0.04% 1.0% 0.03% 0.8% 0.03% 0.02% 0.6% 0.02% 0.4% 0.01% Average ESA 0.2% 0.01% 0.0% 0.00% 80 82 90 92 98 00 04 19 19 1984 1986 1988 19 19 1994 1996 19 20 2002 20 Source: Bank staff calculations, based on data from IMF W orld Economic Outlook, Ethiopia and Eritrea counted as one country Figure 2 shows a similar story for exports of services. The average ESA country has had a very low and declining share of the global market for services. This is in contrast to the fast growing countries, which have, on average, almost tripled their share of the world market over the past 25 years. Hence, the challenge for ESA and other ACP countries is to provide a trade policy climate in which to attain the sustained increases in exports of both goods and services and rising global market shares that have driven growth in the high performing countries. These figures also suggest a lack of effectiveness of current and previous agreements with the EU in supporting sustained growth of exports or in dealing with the constraints that have undermined the share of ESA countries in the world market. In other words, this weak performance occurred despite preferential access to the EU and other markets. Preferences alone have not helped to strongly integrate ESA into the global economy. With decreasing margins of preference due to continued multilateral liberalization, they are even less likely to help in achieving this objective in the future. 5 Figure 2: W hilst fast growers have used trade in services to drive Fi st gr ave vi growth, ESA countries have been less successful h, ve been ess succes ul Exports of service as share of world services trade por of service shar of or ces ade Average HP 16 Averag (% of US current dollars) of urrent do ars Average ESA er 0.9% 0.04% 0.8% Average for HP 16 er or 16 0.03% 0.7% 0.03% 0.6% 0.5% 0.02% 0.4% Average for ESA age 0.02% 0.3% 0.01% 0.2% 0.01% 0.1% 0.0% 0.00% 80 82 84 86 88 0 19 19 19 19 19 199 1992 94 96 98 00 02 04 19 19 19 20 20 20 Source: Banks staff calculations, based on data from W orld Economic Outlook, Ethiopia and Eritrea counted as one country There are three key reasons why preferences have not been effective in stimulating a significant export response in ACP countries.6 First, preferences have not dealt with, and may have distracted attention away from, the key supply side constraints that limit access to all markets. Brenton and Hoppe (2007), for example, find that whilst preferences for clothing have supported exports to the EU and US they have not overcome the negative impact of weak governance in Africa on the sourcing decisions of global buyers. The conclusions from a review of the range of diagnostic trade integration studies7 that have been undertaken in least developed countries concludes that the principal constraints to trade are typically those that limit access to and raise the costs of the key backbone services that are critical for competitiveness (transport and logistics, energy, finance, telecommunications). These problems are compounded by poorly designed structures of incentives that constrain the flow of resources into their most productive uses and weak and ineffectual trade supporting institutions, such as customs, standards, export promotion agencies, trade ministries themselves, and the often existing fragmentation of authority with regard to trade related issues between a number of ministries, coupled with a lack of communication between stakeholders. Second, and related, preferences margins for many of products currently exported by African countries are typically small and have, and will continue, to be eroded, by multilateral trade liberalization and the EU's predilection to sign free trade agreements. It should be pointed out that 46 percent of exports from Africa to the EU are concentrated in 6 Ollareaga and Ozden (2005) suggest that preferences may actually have hindered integration of poor countries into global markets, showing that those countries that received preferences tended to have liberalised their own trade policies less than those developing countries that had not received preferential access to OECD markets. 7 See Biggs (2007) 6 oil products. Of the remaining non-oil exports, 58 percent enter the EU under a zero MFN tariff rate (this percentage is similar for both LDCs and non-LDCs). The value of preferences that many in individual ACP countries perceive is often far in excess of reality. For example, looking at the 13 non-LDCs which may lose some of their preferential access to the EU if they do not sign an EPA and revert to the GSP, the value of preferences under Cotonou amounted to only 3.9 percent of their exports to the EU or EUR 782 million in 2005.8 The value of EU preferences for African LDCs is less at 2.1 percent of their exports to the EU.9 (The appendix provides a more detailed discussion of the value of EU preferences under Cotonou and the GSP.) Third, for products where margins of preference have been substantial, such as clothing, market access has been severely limited by restrictive rules of origin imposed by the EU. Prior to the end of the Agreement on Textiles and Clothing, exports of apparel from African least developed countries (LDCs) to the EU stagnated despite preferences, whilst exports to the US under AGOA grew very strongly (Figure 3). Exports of apparel from African LDCs to the EU and US were almost equal in 2000, but the value of exports to the US in 2004 was almost four times greater than the value of exports to the EU. Source: USITC and Eurostat The key factor explaining why exports to the US grew much faster than to the EU is the rules of origin. EU rules stipulated production from yarn. This entails that a double transformation process must take place in the beneficiary with the yarn being woven into fabric and then the fabric cut and made-up into apparel. These rules prohibit the use of imported fabric, although cumulation provisions allow for the use of inputs produced in the 8Under the GSP the value of their preferences would fall to 0.5 percent of their exports or EUR 103 million, a loss of EUR 679. 9The value of ACP/GSP preferences for all African countries (excluding South Africa) equals 2.6 (3.3) percent of their respective exports. South Africa accounts for about a third of all SSA exports to the EU. 7 EU and other ACP countries. To obtain preferences, apparel producers had to use local, EU or ACP fabrics. They could not use fabrics from the main fabric-producing countries in Asia and still qualify for EU preferences--a binding restriction, since few countries in Africa have competitive fabric industries. Until the recent EPA agreements, the EU rules did not allow producers in low-income African countries the flexibility they currently have had under AGOA to source fabrics globally. It is worth remembering that the EU granted preferences to African countries for apparel subject to these strict rules of origin for more than 20 years under the Lome and then Cotonou agreements. However, these strict rules did little to encourage the development of an efficient fabric industry in Africa, the main justification for their imposition,10 and are likely to have severely constrained the impact of preferences in stimulating the clothing industry.11 This is because a competitive and thriving clothing sector is the most important driver for a local textile industry. By limiting the ability of the clothing firms in Africa to source inputs globally, EU rules undermined the competitiveness of the sector and additionally compromised opportunities for the emergence of efficient textile firms. At the same time, the lack of regional infrastructure and market size for textile products have hindered the exploitation of economies of scale and scope in competitively producing high quality textiles and specialized textile products. An important aspect of the interim agreements that have been signed is that these rules for clothing have been relaxed to allow for a single transformation requirement similar to that under AGOA. 3. Opportunities and Challenges in the Changing Global Economy The negotiation of EPAs is taking place in a rapidly changing world economy. EPAs will be successful if they leverage the reforms and supporting aid for trade that are essential if African countries are to exploit the opportunities that are available to them in the global market whilst effectively addressing the challenges that come from an increasingly competitive global economy. Opening domestic markets to EU firms driven by the perceived need to lock-in access to the EU for preference dependent exports will not be a recipe for success. This section briefly discusses the emerging opportunities for ACP countries in the world economy, with a focus on Africa. It highlights the growing importance of East Asia in the global economy and the surge in trade in services that has accompanied declining telecommunications costs. 3.1 The rising importance of developing countries in the global economy The pace of global integration is likely to intensify and will be powered increasingly by developing countries. World Bank (2006) suggests that the size of the global economy will rise from $35 trillion in 2005 to $72 trillion (at constant market exchange rates and prices) in 2030, an average annual increase of more than 3 percent--2.5 percent for high-income countries and 4.2 percent for developing countries. Developing countries, once considered the periphery of the global economy, will become main drivers. The share of developing 10See Brenton (2006) for a discussion of why restrictive rules of origin are inappropriate as a development tool and are more likely to reflect protectionist interests in the preference granting country. 11See Brenton and Ozden (2005) for a more detailed analysis of the impact of the EBA and AGOA on apparel exports from African LDCs and the role of the rules of origin. 8 countries in global output will increase steadily and China's output will exceed that of the European Union and that of the United States sometime around 2015 (in purchasing power parity terms). Global integration is likely to enter a new phase. In virtually every growing economy the importance of trade--captured by the ratio of trade to GDP--will rise, continuing the trend of the past two decades. Global trade in goods and services, growing faster than output, is likely to rise more than threefold to $27 trillion in 2030. Roughly half that increase will come through developing countries, with the share of developing countries in global exports rising from 32 percent now to 45 percent in 2030 (Figure 4). This means that a growing share of global production of goods and services will be performed in those developing countries able to take advantage of new opportunities and respond to the emerging challenges. Whether countries exceed projections--or fall short--depends heavily on the policies they adopt over this long period. Figure 4: Developing countries share of world trade will rise as global Figur oping count shar of ade obal integration intensifies te Exports from developing and developed countries, 2005-2030 evel ed US$2001 trln. 30 $27 trln 25 High-income countries 20 15 10 Developing countries 45% 5 0 22% 32% 1980 2005 2030 Source: W orld Bank Global Economic Prospects 2007 ce: Econom Pr 2007 The new wave of globalization brings important challenges for developing countries. Developing countries with wages higher than those in China and India risk losing ground following the entry of these countries into the global marketplace. The sheer size of China and India may also preclude the diversification of the poorest countries into manufactures and so close of a route to growth and development (Cline (2006)). However, what matters is productivity. In middle income countries is the wage gap with China greater than the difference in productivity and can productivity differentials be maintained? Similarly, least developed countries in Africa that have lower wages than China and India, will be able to compete in the global market if levels of productivity are close to those in India and China. 9 Another problem with the view that the global market will be swamped by products from China and India is that the law of comparative advantage entails that there will always be opportunities for other countries to export even though India and China will come to dominate certain sectors. In general, as the global demand for say Chinese manufactured products increases, (dollar denominated) wages in China will tend to increase, through higher wage demands from Chinese workers (especially if the rural and urban labor markets remain partially segmented) and from the inevitable additional pressure on the yuan to rise. There is evidence that this process has already begun (Figure 5). Figure 5: China's surge ­ and that of other developing countries ­ ge represents more opportunity than threat han hr Exporting requires importing por ng por Surging growth means rising wages Su ng ow wages Internationally comparable average wage rates, indexed, 1998=100 nt comparabl aver ge age es, exed, 230 C h in a 180 M ex ic o 130 S o u th A fric a In d ia 80 P h ilip p in es 30 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Based on scenarios from World Bank Economic Prospects 2007 ased scenar om or ank ono spect 2007 Thus, Chinese development will not preclude the lowest income countries being able to export low skilled intensive products. However, this will only arise if these countries put in place a business climate that supports investment and trade. For example, in Africa, competitiveness from low cost labor is undermined by high indirect costs, with the main barriers being corruption, crime and inadequate infrastructure (Eifert et al 2005). The poor business environment leads to lower returns to labor and capital in production which depresses investment, labor demand and real wages. The entry of these large economic entities into the global market offers opportunities as enormous as the challenges posed for developing countries. The large markets in India and China have changed the dynamic of south-south trade (Figure 6) and offer developing countries a route to decreased dependence on rich countries, demand in which for products produced in the poorest countries has been growing slowly for a long period. 10 Figure 6: C hina's surge ­ and that of other developing countries ­ and hat her devel count es represents m ore opportunity than threat Exporting requires im porting, and C hina's im ports are rising rapidly qui ap Im ports in U S D billions Im ts 140 120 A fric a S o u th A s ia 100 M id d le E a s t a n d N o rth A fric a 80 L a tin Am . E u ro p e a n d 60 C e n tra l A s ia 40 E a s t-As ia P a c ific 20 0 1985 1990 1995 2000 2005 Source: W orld Bank G lobal Econom ic P rospects 2007 Nevertheless, at the beginning of the 21st century, African countries still ship nearly two thirds of their total exports to EU and the US. There has been, however, a recent shift in export patterns. Demand in Asia, and primarily in India and China, has been the main source of the accelerated growth of African exports since 1990 and the growth of such exports has intensified in the 2000s (Broadman (2007)). Table 1: Destination of Sub-Saharan (non-oil) Exports Destination 2004 2030 Asia 18.3% 31.3% Eastern Europe and Central Asia 8.2% 8.6% European Union 53.3% 38.7% The Americas 8.0% 5.7% o.w. USA 6.2% 4.3% o.w. Brazil 0.3% 0.2% Middle East, North Africa 2.0% 2.5% Oceania 0.3% 0.2% Sub-Saharan Africa 9.9% 12.9% Total 100.0% 100.0% source: Linkage model, COMTRADE, own calculations; non-oil trade only In table 1 we show, using a very simple simulation based on the gravity model and scenarios of growth in the different regions drawn from Global Economic Prospects 2007, that the importance of China and other East Asian countries as a destination for Africa's 11 exports should increase substantially over the next 25 years12. The main adjustment that takes place is a shift away from Europe. If bilateral trade flows continue to be driven by incomes and location then Europe will cease to be the largest destination market for SSA exports. Exports to the EU market will remain critical for many exporters and important new markets and niches will clearly arise. However, if EPAs are narrowly focused on bilateral trade between the EU and sub-Saharan Africa and fail to address the key constraints to global competitiveness then they will fail to assist the ACP countries in exploiting these new opportunities in the global market. 3.2 Opportunities from the Global Sourcing of Services The global competition that has been underway in goods for decades is now visible in services, as falling telecommunications costs and greater openness to FDI enable different parts of the services value chain to be performed in different locations around the globe-- global sourcing of services. Global sourcing has increased competition in services markets for a wide variety of activities, from low-skilled such as data entry, word processing and call centers to higher skilled activities such as software development, consultancy, medical services and R&D. A range of services that were previously thought to be non-tradable are now being provided electronically over large distances. The potential number of service jobs that may migrate from rich to poor countries is uncertain but could be large (World Bank (2006)). This phenomenon is creating considerable opportunities for development in poor low wage countries, both in terms of export possibilities as well as access through imports to cheaper services inputs that raise productivity when used in other sectors. Global sourcing is providing important new employment. In the relatively low-value segments such as call centers, wage costs are important determinants of location (along with language skills) and competition is fierce among developing countries. At the higher skill end, global sourcing of services may be reducing incentives for skilled migration by creating new opportunities at home. It is also important to note that a large number of those employed as a result of global sourcing are women, offering a different route to development than that based on growth of agriculture and manufacturing. The services revolution and global sourcing are offering opportunities for new exports and for attracting services-related foreign investment for a range of poor countries. IT and global sourcing offer new and alternative drivers of development that circumvent some of the key constraints to growth driven by the expansion of exports of agricultural and manufactured goods. This is most apparent for landlocked countries and small (often island) economies that face very high costs for physical transportation (by air, road and sea). For example, development in Rwanda has to confront an extremely adverse location, one of the highest population densities and a high population growth rate. While increasing the quality of quantity of exports of traditional agricultural exports (coffee) and minerals is crucial to increases in incomes for the poor in the short to medium term, the government of Rwanda has identified the provision of IT intensive services, both locally and abroad, as a 12The growth rates for each region are based on simple projections of average performance over the past 15 years. The approach is based on simply applying the coefficients on exporter and importer GDP from a standard gravity model to the changes in income over the next 25 years. 12 base for growth in the long run, to provide for employment and to turn the country's high, but very young, population into a driver of development rather than a constraint. While India and China will come to dominate the market for global sourcing of services, comparative advantage will ensure that there are opportunities for many developing countries. Small island economies in the Caribbean, for example, have been able to attract certain back office activities from the United States, such as data entry. While there are important new opportunities for developing countries there are also considerable challenges regarding the provision of necessary infrastructure, the design and implementation of appropriate regulation and effective education strategies to increase the supply of human capital. While telecommunications infrastructure is clearly crucial, access to relatively cheap and reliable electricity is also necessary (a critical problem for many poor countries). Providing access to high-quality telecommunications infrastructure is not sufficient. It is also necessary to establish a competitive framework for the provision of telecommunications services. Liberalization of the trade and investment regime can be central to ensuring the efficient and competitive provision of the backbone telecommunications services complemented by an appropriate and effective regulatory environment. Considering the difficulties in establishing such a framework, it is in this sector that Aid for Trade and technical assistance, similar to the concept of "twinning", are needed. This concept has been used very successfully in EU accession countries where administrations and agencies of member and accession countries exchanged personnel to transfer knowledge and procedures. 4. Trade and Competitiveness Reform Programs It is important that developing countries make trade central to their development strategies to enable them to exploit the opportunities that global integration offers to attain and sustain high rates of growth. This calls for more attention to policies that facilitate trade and improve competitiveness ­ supported where necessary by Aid for Trade. In this context it is useful to have a framework in which to assess the range of issues that affect countries' ability to compete in international markets. Experience suggests that a practical approach to improving competitiveness will be built on a framework containing three critical complementary elements:13 1. The incentives regime. A key challenge for policy makers is to ensure that domestic resources are channeled to their most productive activities. This requires a careful analysis of the structure of incentives in the economy to ensure that land, labor, capital and technology are moving to a) sectors in which the country has a long-term capacity to compete and b) to the most productive firms within sectors. In turn, this necessitates a clear understanding of how trade, tax, the business environment and labor market policies interact to affect investment, output and trade decisions. In many small low income countries the economy tends to be dominated by a small number of sectors so that many of the key issues regarding the allocation of resources can be unearthed by analysis that focuses on these sectors. This method, however, is likely to miss growth potential in 13For more details see http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/TRADE/EXTEXPCOMNET/0,,contentMDK:212 98368~menuPK:2534115~pagePK:64168427~piPK:64168435~theSitePK:2463594,00.html 13 sectors that are only on the verge of becoming successful as they do not figure in the policy makers' mindsets. There would hence be a bias towards already established sectors, missing the potentially large growth potential in non-traditional sectors. 2. Lowering the costs of backbone services. Of great importance in today's globalized economy is that domestic firms have access to efficiently produced critical backbone services inputs. Firms that have to pay more than their competitors for energy, telecommunications, customs services, transport and logistics, finance and security will find it hard to compete in both the domestic and overseas markets. Competition and regulatory oversight in these services industries lie at the heart of the policy challenge. In many developing countries lack of infrastructure is a critical constraint on the availability and cost of backbone services. Other critical services are those related to education and training that are necessary to ensure supply of the type of labor required by the more productive expanding sectors in the economy and to foster a process by which value is increasingly added to the products and services produced in the country. 3. Pro-active policies to support trade. It is important to address a range of market and government failures that tend to afflict countries as they seek to expand exports and growth. In many cases these constraints to competitiveness require specific interventions and institutions. These are likely to include export and investment promotion agencies, standards bodies, agencies to support innovation and clustering. In addition, governments will need to address the political economy constraints to reform and to limit the costs of adjustment, especially when they fall disproportionately on particular groups. In general, such a strategy will be more effective if focused on protecting workers rather than protecting jobs (World Bank Global Economic Prospects 2007). This means that effective safety nets for those workers losing their jobs due to increased competition should be put in place. In tackling government failures and weak capacity for policy formulation and implementation, an effective mechanism can be to establish an empowered and dedicated trade and competitiveness policy unit within government that brings together stakeholders and coordinates policies on export processing zones, duty refund schemes, and other initiatives. It is important that these initiatives are brought together within a strategy for competitiveness rather than as a series of ad hoc interventions. In isolation these agencies tend to be rather weak and ineffective, and by potentially duplicating efforts will most likely waste scarce skilled resources. Export diversification is a critical element in achieving high and sustained growth in commodity dependent countries in Africa. This framework for competitiveness will be appropriate for supporting diversification. Brenton, Newfarmer and Walkenhorst (2007) show that the global economy offers a range of opportunities for diversification: from new products, from raising the quality of existing goods, from increasing the penetration of overseas markets for existing exports and by exploiting possibilities to export services. Successful diversification by definition requires that resources move into new activities and are not bottled up in old low productivity operations. This necessitates a neutral set of incentives. Export diversification will tend to increase the importance of activities that require the more intensive use of backbone services than traditional activities. Exports of services rely heavily on the use of other services as inputs. For example, 14 telecommunications are a critical input into call centers and other business processing activities. Transport is vital to tourism. Finally, higher quality and differentiated products tend to face higher information barriers in accessing overseas markets and are often more dependent on institutions such as those relating to testing and certifying quality and compliance with standards in overseas markets. 5. The role of EPAs At the end of 2007, having fallen short of successfully negotiating full Economic Partnership Agreements and given a strict deadline by the EU to sign WTO consistent agreements at the point of expiry of the Cotonou Agreement, interim agreements were initialed by 19 Africa countries. Most African non-LDCs, with the exception of Nigeria, Republic of Congo, Gabon and South Africa, concluded interim agreements. South Africa will continue to receive market access in the EU under its TDCA agreement, whilst the others will receive access to the EU market according to the GSP and hence loose preferences relative to Cotonou. Interim agreements were agreed with the EAC and the ESA and SADC groups. No agreement was achieved with Central or West Africa. The LDCs in these regions will receive access to the EU under the EBA. The interim agreements call for the conclusion of more comprehensive EPAs (indicating but not determining some of the issues to be included) "as soon as possible", meaning by the end of 2008. However, ECOWAS has already decided to request an extension of the deadline until the end of 2009. The standard partial equilibrium analysis of the trade and welfare effects of reciprocal trade agreements, such as the interim agreements that the EU has signed is presented in Milner et al. (2005). Several studies have provided empirical estimates of the impact of reciprocal tariff reductions under EPAs on trade, revenues and welfare, for example, Karingi et al (2005) and Keck and Piermartini (2008).14 In this paper, whilst cognizant of these impacts, we look at the opportunities of the EPAs in the broader policy objective of seeking to increase competitiveness and integrate into the global economy. One of the fundamental challenges that the EPA process has faced is that the negotiations have been undertaken between the EU and the secretariats of regional groupings while to be successful, agreements must be consistent with the trade and development strategies of individual ACP countries. Trade relations with the EU and regional partners are just one element of trade policy that should also encompass unilateral, multilateral and bilateral agreements with other countries. To improve competitiveness, African countries need to invest in significant domestic reforms and upgrading of infrastructure. But because many of the African countries are small in population and economic size and/or are landlocked, many competitiveness issues are best addressed at the regional level. For example, it is important that the interests of land-locked countries are properly taken into account in motivating reform of ports, customs and transport in the countries that provide access to the global market. EPAs can play an important role in strengthening and moving forward the process of regional 14The ODI has provided a short summary of the main results of empirical studies of the potential impacts of EPAs: http://www.odi.org.uk/Publications/briefing/bp_june06_epa_bp2.pdf 15 integration. Well-designed EPAs offer an opportunity to harness the power of trade negotiations to further internal regional development and spur integration into the global economy as well as improving access to the EU market and leveraging aid for trade to support reform programs. In contrast, a poorly designed EPA could undermine regional integration and seriously hinder development (Stevens (2005)). As each negotiating region consists of countries with different economic interests and political conditions, the challenge for the EU and African countries will be to structure EPAs in a way that offers mechanisms and incentives to strengthen the process of regional integration.15 Box 1 suggests and summarizes key elements of a pro-development EPA. The interim agreements do make some progress in guaranteeing duty and quota free access to the EU and in reducing the restrictiveness of the EU's rules of origin. However, if pro- development EPAs are to be achieved in 2008 there will have to be substantial progress in the other key areas outlines in the box. The following sections briefly elaborate on how EPAs can support reform in the three areas that are crucial for global competitiveness. 5.1 The incentive regime Reducing the dispersion and then the level of tariffs will be a critical element of a competitiveness strategy in many of African countries in terms of reducing the impact of distortions in the structure of incentives. Lower external tariffs are necessary to avoid potentially damaging trade diversion from the removal of tariffs against EU suppliers in an EPA. Trade liberalization in the EPA regions should be a staged process with intra regional barriers removed first to ensure effectively functioning free trade areas. Then peak external tariffs should be capped at the regional level, in a coordinated manner, to remove the largest distortions in the trade regime. In the same coordinated manner, but not necessarily under a CET, the average of external tariffs should be lowered to pursue global competitiveness and integration into global market. Such a strategy was an important feature of the development of the fast growing countries of the past 20 years. Lower tariffs allow domestic firms access to needed inputs at closer to world market prices and reduce bias against exports and the bottling up of resources into low productivity activities. Only subsequently, should tariffs against imports from the EU be removed. It is essential that preferences for EU suppliers are only implemented after the reduction of MFN tariffs since lack of competition among EU suppliers in small African markets could lead to unchanged prices and simply to a transfer of revenue from the customs authority in the African country to EU suppliers. 5.2 Lowering the costs of backbone services and trading Trade-related services throughout Africa are less efficient than in other regions. In Sub- Saharan Africa, 52 percent of firms report unreliable electricity supply as a constraint. This compares to 42 percent in South Asia and 24 percent in East Asia and Latin America (World Development Report 2005) The Diagnostic Trade Integration Studies undertaken under the Integrated Framework16 for most countries in Africa, highlight that electricity, 15How this should best be done in the agreements, is still an open issue. The CARIFORIM-EC EPA makes reference to coordination in most chapters of the text but CARIFORUM reserves their right to sovereignty of related decisions, showing the reluctance of these ACP countries to commit to a predetermined pace of regional integration. 16www.integratedframework.org 16 telecommunications, transport, and water charges are often higher in low-income countries and that access to these and other services, such as financial services, are extremely limited. Regional integration can be an important mechanism to address these problems and will be effective when incorporated into a broad strategy encompassing appropriate multilateral and unilateral reform. The domestic markets of many African countries are too small to realize full economies of scale in network services and their regulation. In addition, regional approaches can led to more effective regulatory frameworks that allow for greater competition, with positive effects on prices and efficiency. EPAs offer the opportunity to bring all relevant stakeholders together and to lock in domestic reforms and regional integration with regard to the regulation and supply of crucial telecommunication, electricity, water, and financial services needed for exporting. Regional integration must coupled with phased programs of services liberalization on an MFN basis to avoid the enlargement and entrenchment of existing monopolies or their transfer to local or EU suppliers. In practice, the positive economy wide impacts on employment that emerge as firms elsewhere in the economy expand as the costs of their services inputs decline and their competitiveness improves will considerably outweigh the employment impact on incumbent service providers as markets are opened up to competition. The potential for African countries to expand regional and global exports by lowering trade costs other than policy border barriers is enormous. The costs of institutional obstacles, informal barriers and sub-optimal regulatory scales are often high and, unlike tariffs, do not generate revenues but simply waste economic resources and dampen productivity. Regional agreements can precipitate effective cooperation to lower these trading costs by raising the level of policy salience, dispersing information, ensuring better coordination between the institutions of different countries, providing an institutional home for joint initiatives and by facilitating both informal and formal dispute resolution. Trade in Africa remains dogged by lack of coordination and weak harmonization of documentation and regulations. Finally, negotiating these backbone services in a regional forum can act to depoliticize issues. Again, aid for trade will be needed to address the large investment demands that will address problems in regional infrastructure networks within each region. Reforms that will increase access to crucial backbone services should also aim at establishing simple but functioning, region-wide, approached to competition policy. Linking EPAs to small but firm commitments on establishing a supra-national competition policy within each EPA region could help to advance policy reforms. Due to the extensive experience of the EU with the establishment of a supra-national competition policy, the EU should consider offering aid for trade in this area, in particular in the form of technical assistance and twinning. 5.3 Pro-active policies to support trade While integrating into the world economy requires that import taxes are low and relatively uniform, for the least developed countries they are often a key source of revenue relative to VAT and sales taxes (Figure 7). High-income countries are able to recover revenues lost from trade liberalization from other sources: on average, middle income countries recover 17 45-60 percent of lost tariff revenues while least developed countries recover less than 30 percent of lost tariff revenues (Baunsgaard and Keen 2005). Concern over this revenue issue is an important factor constraining low income countries in reforming external tariffs to reduce bias against exporting and a better structure of incentives for global integration. Figure 7: Duties and Taxes as share of GDP 7 VAT, sales, or turnover taxes 6 import duties PD 5 Gfotnecr 4 3 Pe2 1 0 Low Low er middle Upper middle High Income level Source: Bank staff calculations But many countries collect far less in tariff revenue than the applied tax rates would suggest, due to the widespread (discretionary) granting of exemptions. Analysis for a number of ESA countries suggests that a large share of statutory customs revenues is not actually collected, such that the revenue impact of tariff reform will be considerably less than would be expected on the basis of statutory tariff rates (Brenton, Hoppe and von Uexkull (2007)). For example, a simulation of the impact of an EPA for Ethiopia suggests a revenue loss of 3.4 percent of total tax revenues when actually applied tariffs are used but a loss of 6.4 percent if the statutory tariff rates are used. If Ethiopia were to remove tariff exemptions as tariffs against the EU were removed then the simulated revenue loss would amount to 0.8 percent of total tax revenues. Exemptions make the tax regime opaque and difficult to administer and can lead to a distorted incentive structure that discriminates against small firms with less influence. Further, there is little evidence that exemptions have a significant impact on investment, their primary justification. Many countries could substantially reduce applied tariffs while maintaining or even increasing revenue if exemptions were removed and collection improved. However, it is still necessary to address the development challenge highlighted in Figure 7 of moving from easy-to-collect trade taxes to harder-to-collect consumption and income taxes. Simply implementing a VAT is not sufficient; a high degree of collection efficiency (the ratio of actual to potential revenues) is needed. To meet this development challenge and obtain the necessary funds to finance relevant projects requires an increase in the tax base for the "difficult to collect taxes" which will be achieved through economic growth, simplification, in terms of fewer rates and less exemptions, an increase size of formal relative to informal sector and through improved 18 revenue collection. A key objective for many of the African countries is customs reform to increase the efficiency of collection of taxes applied at the border. This in turn requires initiatives to raise the capacity of customs.17 Regional reform and harmonization of tax administration and structures will promote regional integration and help replace lost tariff revenue. It will furthermore simplify and streamline the regional incentive regime if done properly. Exporters in African countries tend to face common problems in accessing new markets and so coordinated responses at the regional level facilitated by an EPA could be instrumental in addressing the associated market and policy failures. For example, weak institutions regarding testing and conformity assessment may hamper development and export of new products or higher quality variants of existing exports. The ability of would- be exporters to comply with mandatory health and safety standards, as well as market- driven voluntary standards in overseas markets, is a major factor determining access to those markets. Upgrading testing facilities and measuring equipment is essential for reducing the costs of conformity assessments. EPAs can contribute to alleviating these problems by leveraging regional approaches that provide more efficient and effective standards and conformity assessment systems and thereby providing a framework through which aid for trade can be delivered to support improved competitiveness. Box 1: Elements of a Pro-Development EPA What does a pro-development EPA look like? Here are 8 elements that would collectively constitute central tenants of a pro-development EPA. A guiding principle of these measures is that all 8 are in the fundamental interest of all countries as part of their development strategies--and of the EU in its objective of promoting development. These would include: 1. A program of phased and variable geometry for MFN reductions in external tariffs, consistent with regional development programs. For example: Phases 1 and 2: Eliminate all internal tariff barriers in CU/FTA to promote regional trade. Phases 1 and 2: Bring down MFN peak tariffs to average to promote intra-African and other efficient trade with third parties. Phases 2 and 3: Bring down MFN average levels toward East Asian levels in EPA group. Phase 3: Allow preferential for the EU; taken as a final step, the adverse risks of trade diversion and hub-and-spokes development would be lessened. 2. Nonrestrictive rules of origin for ACP access to the EU market (choice of satisfying either a 10% value-added requirement or change of tariff heading); if the value-added requirement is higher, cumulation should cover all developing countries to allow ACP producers maximum access to the world's lowest-cost inputs and to avoid putting regional suppliers outside the EPA group at a disadvantage. 3. A program of gradual but purposeful reforms of tax administration and intra-regional tax policy to harmonize tax structures so as to promote regional integration and replace lost tariff revenues. This can complement tariff and customs reforms at the regional level. 17The EU could support ACP countries by including a mechanism for temporary staff exchange and a knowledge transfer program. These types of initiatives could also be useful in providing incentives for certain of the LDCs to come to the negotiating table. 19 4. A phased, region-specific program of services liberalization designed to promote wider access to lower cost trade-related services, such as efficient telecommunications, electricity, and transport; this should be undertaken on an MFN basis, so as to avoid entrenching monopolies from one or another country. 5. A program of trade facilitation measures--for example, improvements in customs, ports, border posts -- to reduce costs, linked to intra-regional programs to lower costs of trading, with special attention to lowering transit costs of land-locked countries, with specific benchmarks for implementation; 6. A program, however small, of temporary movement of persons (mode 4); while not currently in the negotiations, this may be easier to achieve in regional arrangements than in multilateral talks. 7. Agreements could include new IPR rules and rules on investment, but these should be back-loaded and phased in accordance to a region's capacity to implement ­ and to benefit from them. Agreements could include adoption of competition policies consistent with national development strategies and in accordance with a region's capacity to implement and benefit from them. 8. "Aid for trade" in the form of a program of technical and financial assistance to support competitiveness reform programs through funding projects that improve trade facilitation, conformity with SPS and TBT measures, alleviate supply side constraints (e.g., infrastructure) and increase administrative capacity in key policy areas (such as tax administration), as well as funding programs that address the adjustment costs that arise from trade reform. Derived from Hinkle, Hoppe, and Newfarmer (2005) Successful implementation of the bold reform programs required for African countries to be able to integrate effectively into the global economy will typically require measures to limit the adjustment costs that will arise. Assistance to firms and workers can both speed the adjustment process and reduce adjustment costs.18 It is important to get the balance right between assisting firms in previously protected sectors that have long-term viability and supporting resources moving into new high growth/high productivity internationally competitive sectors. Inter-sectoral adjustment is inevitable and necessary to allow new sectors to grow. For labor, costs arise from (i) finding and taking up a new job ­ which can be reduced by projects that improve the availability of information on new jobs, support job fairs, and more generally build a bridge between employers and the unemployed; (ii) loss of income during the period of transition ­ which can be addressed through a social safety net, for example, in the form of social security and unemployment benefits; (iii) the costs of retraining and preparing for work in new sectors ­ which can be addressed through subsidized training programs, apprenticeship schemes and so on. In general these adjustment costs tend to be lower in more flexibility labor markets. Similarly, reforms that aim at improving the education infrastructure and quality, in particular at the vocational level, should be undertaken as part of the pro-active policies to support trade reform. For two reasons, it will be important to focus on portable skills that 18For more details on the costs associated with trade reform see Bacchetta and Jansen (2003) and Cordoba et al (2005). 20 are easily transferable between jobs; a) to prevent in investment of knowledge that is easily "destroyed" in a dynamic economic environment with high job turnover, and b) because private companies are less likely to invest in portable skills of their workers--as compared to specific skills--due to the fear of the workers being competed away. Well-designed and productivity improving support measures that address the vocational education infrastructure could possibly form part of the comprehensive reforms driven by EPAs with some financial support from the EU and other development partners. Helping selected firms ­ notably the most productive -- in the declining sectors to increase competitiveness may allow them to survive after liberalization, hence maintaining some output and employment in the sector. The costs of such intra-sectoral adjustment tend to be smaller than those that arise when resources need to move into new sectors reducing overall cost of trade reform. This often requires investment in new production techniques and repositioning of the firm's product on the market through upgrading quality and exploit specific market niches. Initiatives to support such adjustment by existing firms has to first address the issue of why viable firms have not already made the investments. It may be that trade protection and/or government regulation (including price controls) may have dissuaded investment in the most efficient technologies, so hat trade reform itself is the key part of the upgrading process. Nevertheless, there are also likely to be specific market failures that limit adjustment possibilities, such as, lack of a functioning credit market, lack of a market in technical advice and in support for marketing. When long-term survival is based upon exporting, adjusting firms will need to make new (and different) investments and incur start-up costs relating to issues such as market information and product redesign to satisfy the standards of different markets. Aid for trade under the EPAs can address the institutional weaknesses and financial constraints that limit the ability of African countries to address these adjustment costs. The focus here has been on the ways in which African countries can leverage an EPA to address the critical constraints to competitiveness and growth. Many of these issues need to be dealt with regardless of whether an EPA is signed or not. Indeed, it is essential that an EPA be integrated into a broad strategy towards competitiveness that motivates unilateral actions for reform as well as those that are locked in through regional and multilateral agreements. For example, as discussed earlier, it is vital that African countries reduce their external tariffs since open regions have performed much better than those closed to the global economy. Preferential trade agreements are only effective for developing countries if implemented in conjunction with more comprehensive domestic reforms. At the same time, a successful trade agreement will contribute to the overall economic impact of that reform program. In Europe, the eight Central and Eastern European countries that joined the EU in 2005 experienced strong growth in trade and investment inflows during the 1990s; yet two countries in the region, Bulgaria and Romania, having almost identical trade agreements with the EU but initially much less extensive domestic reform programs, saw a much weaker trade and investment response. Regional integration initiatives in Latin America in the 1990s have been much more effective than early efforts, reflecting broad and credible reforms in many countries (Devlin and French-Davis (1999)). Successful countries in the region such as Chile have effectively pursued improved market access in the major overseas market, the US, as well 21 as regional agreements, while strongly implementing unilateral reductions in external tariffs. 6. Conclusions Both the EU and the ACP countries will have to work hard to build upon the initial momentum generated by the interim agreements to achieve pro-development EPAs. Key issues must be addressed to avoid the risks that come from the interim agreements that have been initialed by many African countries. Because external tariffs are relatively high and internal barriers within groupings still prevalent, enacting the provisions of the interim agreements on reciprocal tariff reduction in African countries without prior action on these issues could result in a hub and spoke pattern of trade integration, trade diversion, and the possibility of net losses of income. Without action on external and internal barriers, giving EU firms preferential access could well divert trade to EU-producers from more efficient producers based in the rest of the world and even from those based in the regional grouping if the internal non-tariff barriers exceed the external tariff. As a result, trade liberalization vis-à-vis the EU might not lead to lower consumer prices but rather to a transfer of tariff revenue to EU producers. This problem is aggravated by the fact that many national markets in Africa are small and transport costs to these markets can be high. Markets are often monopolized by one external supplier and competition among EU producers is not likely to increase after the implementation of the EPAs. Thus, an EPA focused on preferential tariff removal for EU suppliers that is implemented in the absence of a strategy to raise competitiveness in Africa through reductions in internal and external barriers to trade, attention to supply side constraints such as lack of infrastructure, and active policies to support trade and address trade related adjustment costs, will likely undermine development. To achieve pro-development EPAs, the EU should go further in reducing the restrictiveness of its rules of origin, defer its own preferences in ACP markets until important MFN liberalization has occurred in the regions and to support ACP countries implement their reform programs through "aid for trade". African countries should focus on developing regional strategies for integration that harness the potential power of the EPA process to move forward on reforms that will promote their competitiveness, integration, and growth and be willing to lock-in a program of implementation consistent with their own development priorities and pace of reform. The new phase of the EPA process that will follow the initialing of the interim agreements provides an opportunity, through a more realistic approach to variable geometry, to connect the negotiations more closely to national trade and development strategies. 22 References Bacchetta, M and M Jansen (2003) `Adjusting to Trade Liberalisation: The Role of Policy, Institutions and WTO Disciplines', WTO Special Study 7, WTO, Geneva Baunsgaard, T and Keen, M (2005), "Tax Revenue and (or?) Trade Liberalization", IMF Working Paper WP/05/112, International Monetary Fund, Washington DC. Brenton, P and Ikezuki, T (2005) `The Value of Trade Preferences for Africa', in Newfarmer, R (ed) Trade, Doha and Development: A Window on the Issues, World Bank, Washington DC. Brenton, P and Hoppe, M (2007), "Clothing and Export Diversification: Still a Route to Growth for Low Income Countries?", Policy Research Working Paper 4343, World Bank. Brenton, P and Ozden, C (2007) `The Effectiveness of the EU and US Trade Preferences for Least Developed Countries with a Focus on the Rules of Origin in the Clothing Sector', in Gerrit Faber and Jan Orbie (eds) European Union Trade Politics and Development: `Everything but Arms' Unravelled, Routledge, London Brenton, P, Hoppe, M and E von Uexkull (2007) `Evaluating the Revenue Effects of Trade Policy Options for COMESA Countries: the Impacts of a Customs Union and an EPA with the European Union', World Bank Washington Brenton, P, Newfarmer, R and P. Walkenhorst (2007) `Export Diversification: A Policy Portfolio Approach', paper presented to the Growth Commission Conference on Development, Yale University, mimeo, World Bank Broadman, H (2007) Africa's Silk Road, Washington, World Bank. Busse, M., Borrmann, A., and H. Großmann (2004), The Impact of ACP/EU Economic Partnership Agreements on ECOWAS Countries: An Empirical Analysis of the Trade and Budget Effects, Final Report, HWWA, Hamburg. Cline, W (2006), "Exports of Manufactures and Economic Growth: The Fallacy of Composition Revisited." Institute for International Economics, Washington DC. Cordoba, S., S. Laird and J.M Serena (2005) `Adjusting to Trade Reforms', UNCTAD, Geneva Devlin, R and R. French-Davis (1999) "Towards an Evaluation of Regional Integration in Latin America in the 1990s', The World Economy, 22, 261-90 Eifert, Ben, A. Gelb, and V. Ramachandran, 2005. "Business Environment and Comparative Advantage in Africa: Evidence from the Investment Climate Data", The World Bank, Washington DC. 23 Hinkle, L.. M. Hoppe, and R. Newfarmer 2005 "Beyond Cotonou: Economic Partnership Agreements in Trade , Doha, and Development: A Window into the Issues R. Newfarmer (ed) Washington: World Bank. Hoppe, M (2007), Economic Partnership Agreements: Does Preferential Access of Non- LDC African Countries Increase?, Trade Note No. 32, World Bank, Washington, DC. Karingi, S., R. Lang, N. Oulmane, R. Perez, M. Sadni Jallab and H. Ben Hammouda (2005) `Economic and welfare impacts of the EU-Africa Economic Partnership Agreement', African Trade Policy Centre (ATPC), work in progress n 10, United Nations Economic Commission for Africa (UNECA) Keck A. and R. Piermartini (2008) `The Impact of Economic Partnership Agreements in Countries of the Southern African Development Community', Journal of African Economies 2008 17(1):85-130 Milner, C., O. Morrissey and A McKay (2005) `Some Simple Analytics of the Trade and Welfare Effects of Economic Partnership Agreements', Journal of African Economies,14, 327­58 Ozden, C. and Olarreaga, M. (2005) "AGOA and Apparel: Who Captures the Tariff Rent in the Presence of Preferential Market Access?", World Economy 28:63-77. Stevens, Chris (2005) `Economic Partnership Agreements and African integration: a help or a hindrance?', ODI, London, http://www.odi.org.uk/iedg/Projects/chris_stevens/CSEPAs_SSA-Web.pdf Stevens, C, M. Meyn, J. Kennan, S. Bilal, C. Braun-Munzinger, F. Jerosch, D. Makhan and F. Rampa (2008) ,The New EPAs: comparative analysis of their content and the challenges for 2008, joint ODI-ECDPM report, http://www.odi.org.uk/iedg/Projects/0708010_The_new_EPAs.html World Bank, Global Economic Prospects, 2007: Managing the Next Wave of Globalisation Washington: World Bank, 2006. World Development Report 2005, World Bank, Washington DC. 24 Appendix19 The current value of preferences in the EU market EU trade preferences for the 13 non-LDCs in Africa amounted to only 3.9 percent of their exports to the EU or EUR 782 million in 2005.20 The value of preferences for all African LDCs is much less at 2.1 percent of total exports, although this is less relevant to our discussions of the EPAs since these countries will continue to have duty and quota free access to the EU market under Everything but Arms. However, market access under EBA is less secure than would be provided by an EPA and the opportunity to obtain less restrictive rules of origin than the GSP would be lost (EBA is a special form of GSP).21 One of the problems that one faces when talking about the value of preferences is that this is an artificial concept. It summarises the amount of import duties that exporters did not pay because of preferential tariff rates; it is the volume of trade multiplied by the difference between MFN and preferential tariffs, taking utilization rates into account. Its calculation is based on actually observed trade flows and omits the fact that some of these trade flows might only take place because of the preferences. In that case, the whole export value would be additional value that is transferred to the exporter because preferences are granted. At the same time, the data do not allow us to identify the economic agents that actually benefit from the preferential rates. The issue is that tariffs create a wedge between the producer price in the exporting country (including transport costs to the export market) and the consumer price in the importing country. Here, we assume that the exporter actually receives a net transfer of the value of preferences. However, if there is little competition amongst the importers, the price received by the producers in the exporting country may not change and the importer or intermediary may obtain the rents generated by the preferences. The `value of preferences' is hence a high estimate for the value that is transferred to the exporting country given the observed trade patterns (Olarreaga and Ozden 2005). Given the increasing dominance of large supermarket chains in certain sectors it is probably fair to assume that for a range of products a large share of the value of preferences will accrue to the importers rather than to the relatively large number of small producers in developing countries. The question then remains whether the preferences actually create trade relationships that otherwise would not have taken place. This more global value of preferences is impossible to estimate without specific knowledge of export prices from a larger number of suppliers and the final sales prices. 19This appendix builds on Hoppe (2007), Economic Partnership Agreements: Does Preferential Access of Non-LDC African Countries Increase?, Trade Note No. 32, World Bank, Washington, DC. 20We use a new dataset the EU has recently made available. Assuming that Cotonou preferences expire and no new preferences will be in place by that time, 2008, it is possible that these countries would only have access to preferences under the GSP. Other things equal, this would reduce the value of their preferences to 0.5 percent of their exports or EUR 103 million, a loss of EUR 679. 21The value of EBA preferences for all LDCs equals 3.8 percent of their exports. The value of ACP/GSP preferences for all African countries (excluding South Africa) equals 2.6 (3.3) percent of their respective exports. South Africa accounts for about a third of all SSA exports to the EU. 25 Figure1: Preferences as share of total exports Botsw ana Cameroon Congo Gabon Ghana Ivory Coast Kenya Mauritius Namibia Nigeria Seychelles Sw aziland Zimbabw e Total 0% 10% 20% 30% 40% 50% 60% GSP Cotonou EPA note: all values in EUR1000, calculation of preferences takes current utilization rates into account Source: COMEXT, EU preference data, TRAINS, own calculations Overall, actually claimed preferences for all 13 sub-Saharan non-LDCs22 represent 3.9 percent of their exports to the EU. If only GSP preferences were available, this value would fall to 0.5 percent of total exports or EUR 103 million, a loss of EUR 679. The fall would be substantial for all countries but particularly high in the case of Mauritius, the Seychelles, and Swaziland where the value of preferences would fall from about 23, 16, and 50 percent respectively to nearly zero. The actual value of preferences under an EPA, however, would depend on the rules of origin that were to be agreed. If EPAs with Cotonou-type rules of origin were to be in place by 2008, the value of preferences for the 13 non-LDC countries could increase to 4.5 percent of exports, a notional gain of EUR 107 million, assuming identical utilization rates of preferences as under the current regime. Inclusion of non-restrictive rules of origin in EPAs could increase that value to 5.2 percent assuming constant trade flows. This is a simple estimate calculated on the assumption that current utilization rates for existing exports would increase under less restrictive rules of origin. However, this is likely to 22South Africa is not included in the analysis as it is only an ACP country since 1998 and is not a signatory to the commercial provisions of the Cotonou Agreement, which foresees the negotiation of EPAs. While South Africa has signed an FTA with the EU in 1999, its partners in the customs union (SACU) are supposed to sign an EPA. Both arrangements are likely to create barriers within SACU. South Africa's demand to join the SADC EPA has long been rejected by the EU but South Africa is now taking part in the negotiations. 26 substantially understate the positive impact on trade of more development friendly rules of origin. Non-restrictive rules of origin such as the 10% value-added rule proposed by the Blair commission or a simple change of tariff heading rule would likely open up a range of new export opportunities for beneficiary countries, such as in clothing. The importance of preferences also differs strongly by country, mainly due to differences in the composition of exports to the EU. For example, the ratio of actually claimed preferences to exports ranges from 0.1 percent for Nigeria, which mainly exports oil to the EU, to 49.5 percent for Swaziland, which mainly exports sugar to the EU, a product subject to very large margins of preference (see Figure A1). These products are in particular sugar (36 percent of preference value), edible fruits and nuts (12 percent), and preparations of meat, fish or crustaceans (12 percent). Continuing multilateral liberalization from a successful Doha round would reduce MFN tariffs and reduce the value of preferences on exports to the EU. A rule of thumb puts the costs of proving conformity with the rules of origin at three percent of the goods value. If one deducts these costs from the value of preferences calculated, the value of preferences will be even smaller. When preference margins fall below three percent, usage of preferences is no longer likely to be worthwhile. 27