Revisiting the ACP-EU Economic Partnership Agreements – The Role of Complementary Trade and Investment Policies

Negotiations on the Economic Partnership Agreements between African, Caribbean and Pacific countries and the European Union have not been making much progress recently. How should the Partnership Agreements be structured to ensure a pro-development outcome and how could the negotiations gain more momentum in the months and years ahead?

The Economic Partnership Agreements (EPAs) between the European Union (EU) and the African, Caribbean and Pacific (ACP) group of countries are intended to establish a comprehensive new framework for bilateral economic relations between the EU and the ACP countries. As part of the Cotonou Agreement, the EPAs aim to promote economic growth and development as well as the smooth and gradual integration of ACP states into the world economy. From the perspective of the EU, two main objectives stand out. First, the EU was looking for new trading arrangements with the ACP states that would ensure compatibility with the regulations of the World Trade Organisation (WTO). The non-reciprocal trade preferences established under the Lomé Conventions required a WTO waiver, as these preferences were neither restricted to least developed countries nor granted to all developing countries. The new agreements would provide for a shift from the system of non-reciprocal trade preferences to EPAs, which are in effect bilateral free trade agreements. This implies that ACP states would have to open up their markets for EU products within a twelve-year period, scheduled between 2008 and 2020.

The EU acknowledged that ACP countries might have difficulties in achieving the potential gains from reciprocal trade agreements and therefore broadened the Cotonou Agreement to include a perspective that combines politics, trade and development. In fact, the EPAs aimed not only to provide improved market access for ACP countries to EU markets, to enhance trade in services and to increase co-operation in trade-related areas like competition and investment. Rather, the Cotonou Agreement intended to go beyond these standard features of a free trade agreement by enhancing the political dimension, explicitly addressing corruption, promoting participatory approaches, and refocusing development policies on poverty reduction.

The main argument for this second objective was relatively obvious, since the export performance of ACP countries has been far from satisfactory in recent decades. Despite non-reciprocal trade preferences for products originating in ACP countries as part of the predecessors of the Cotonou Agreement (the Lomé I to IV Conventions), ACP countries’ share of the EU market declined from 6.7 per cent in 1976 to 2.3 per cent in 2008.1 Moreover, about 65 per cent of total exports consist of raw materials and some 60 per cent are concentrated in only ten products. Additional market access preferences alone are, therefore, not very likely to benefit ACP countries in the future.

Current EPA Agreements

The EPAs went into force in January 2008. Yet the only comprehensive EPA that has been agreed on is that with the CARIFORUM group of countries from the Caribbean region. Negotiated between 2004 and 2007, this EPA includes chapters on trade in goods, trade in services, investment, competition, innovation, intellectual property, public procurement, and development aid. Hence, much of the original EPA agenda was incorporated into this agreement. Importantly, there are considerable time-lags involved before the Caribbean countries have to open up their markets to EU products duty free. While the EU removed all tariffs and quotas for Caribbean producers immediately (apart from sugar and rice, for which longer periods have been agreed on), there is a gradual opening of Caribbean markets for EU products over a period of 25 years.2

In the other five EPA regions, the EU and the regional organisations could not agree on comprehensive agreements as intended in the Cotonou Agreement.3 Rather, the EU and ACP regions in Africa and the Pacific signed so-called interim EPAs that have replaced the previous trade preferences. The EU signed bilateral interim agreements with a few countries, such as Côte d’Ivoire, Ghana and Cameroon, to avoid any disruption in trade. Indeed, unlike the large majority of (mainly African) ACP countries that could switch from ACP trade preferences to those under the EU’s Everything But Arms Initiative, these countries are not least developed countries. Consequently, they would have fallen back to the less generous trade preferences under the EU’s Generalised System of Trade Preferences, which they preferred to avoid. Hence, they concluded bilateral interim EPAs that are scheduled to be replaced by regional EPAs at a later stage.

So far, none of the interim agreements have been replaced by full agreements. Most of the recent interim agreements have yet to be reported to the WTO and implemented. Moreover, several countries have continued to postpone signing their interim agreements with the EU. One important reason for this outcome is the fact that a number of disputes have yet to be resolved. Depending on the particular EPA group, there were either disagreements on tariff liberalisation schedules or on a joint list of products that were to be excluded from the liberalisation process. Apart from sometimes conflicting interests within the EPA regions, it became clear during the negotiations that most regional organisations (and their member countries) lack the bureaucratic and political capacities to negotiate complicated trade agreements with a partner such as the European Commission. Unfortunately, changing this would very likely take a considerable number of years, which is beyond current negotiating timelines.

Another possible reason for the slow progress is the enormous trust gap in the negotiations. Governments of ACP countries fear that they might not be able to cope with the consequences of comprehensive EPAs that could change their societies and economies to a substantial degree.4 Local companies might not be in a position to compete with more efficient EU producers, ACP governments might lose tariff revenue that they need to finance public spending, or the agreements might be so binding that they do not allow sufficient policy flexibility to develop their countries. Moreover, EPA countries fear that in the years ahead the EU will not provide the financial and technical assistance that they offered in the past.

Potential Benefits of EPAs

In addition to the lack of trust in the negotiations, there is also an underappreciation of the potential development benefits of an EPA in ACP countries. Based on economic theory, we could expect beneficial effects from lowering trade barriers for ACP countries, as nations benefit from the well-known gains from exchange and specialisation through trade. The static gains from trade can be divided into trade creation and trade diversion effects. The former category is defined as the change in imports from beneficiaries’ countries and consists of the consumption effect, i.e. the increase in overall consumption due to lower prices, and the displacement of domestic production.5 Similarly, trade diversion is defined as the substitution of preferred imports for non-preferred ones due to the preferential tariff elimination. While trade creation is welfare increasing, trade diversion reduces welfare levels. In addition to these basic effects, preferential trade liberalisation leads to a reduction of government revenues as imports from preferred sources, i.e. from the EU, enter ACP countries duty free.

A number of studies have examined the potential trade and budget effects of the EPAs. For example, Busse and Grossmann analyse the impact of ACP/EU EPAs on trade and government revenue for ECOWAS in West Africa.6 Using a simple partial equilibrium model for an estimate of the static trade effects, their main results can be summarised as follows. First, an EPA between the EU and ECOWAS will lead to overall trade effects (trade creation and trade diversion) for most West African countries in the medium range of some 5 to 10 per cent. Second, much larger effects can be expected for a small number of products at a highly disaggregated level, such as textiles and clothing and certain agricultural goods. Third, trade creation exceeds trade diversion in all West African countries, thus increasing welfare. Based on these results, trade could have a beneficial impact on growth in ECOWAS countries if these countries agree on an EPA with the EU.

Yet the study also points out that West African governments would encounter a considerable fall in customs revenues due to the preferential tariff elimination, which could represent up to 20 per cent of total government revenue, making effective changes in the tax regime essential. Moreover, an effective competition policy is required to ensure, for instance, that “pricing to market” by EU exporters to ACP markets can be reduced. Otherwise, ACP countries are less likely to achieve welfare gains from trade liberalisation. However, the partial equilibrium model used in the study is built on a number of standard assumptions in quantitative analysis of international trade, such as perfect competition and constant returns to scale, which presumes that large firms either do not exist or cannot take advantage of their market power.

These so-called dynamic gains from trade are not only quite important sources for a boost in growth rates in the medium to long run. They are also rather difficult to model and examine empirically for developing countries, such as the African ACP countries. Most importantly, appropriate data for sophisticated models that take the specific economic and political situation in ACP countries into account is hard to find. Accordingly, most researchers focus on the traditional static gains from trade that lack these more advanced effects.

Still, the potential dynamic effects of the EPAs include not only an increase in trade but also in foreign direct investment (FDI) in ACP countries. First of all, the EPAs could lead to the provision of much needed new capital in ACP countries from European (or other foreign) investors. This could boost growth rates in the host economies. Likewise, FDI is generally accepted as a means to incorporate new knowledge from abroad. The theory of the multinational firm proposes that multinational corporations have a technological advantage over local firms that outweighs the cost of doing business in external markets.7 The inflow of new knowledge may benefit domestic firms through imitation and learning, increased competition in local markets, facilitation of human capital mobility among firms, and vertical linkages, thereby increasing productivity and sustaining a higher growth rate.

Despite these potential benefits resulting from FDI inflows, the FDI track record of ACP countries has been rather poor to date. While total FDI stocks have increased over the last 30 years, the relative performance is below average. The percentage of total inward FDI stocks of all sub-Saharan African countries as a share of world FDI inward stocks declined from an already low 2 per cent in the early 1980s to 1.2 per cent in 2008. For the CARIFORUM group, the corresponding share remained below 0.5 per cent. These figures indicate that the large majority of multinational enterprises set up their factories in other developing (and developed) countries, but not in ACP regions. Thus, the potential benefits of FDI are largely unexplored.

Harnessing the Benefits of Trade Liberalisation and Foreign Direct Investment

While it has been acknowledged by the EU and ACP governments that the static and dynamic trade and FDI effects are fundamental for the long-term development of ACP countries, there is less agreement on how to ensure that trade and FDI (and, thus, the EPAs) are indeed beneficial for ACP countries. In fact, over the last couple of years a number of studies have analysed the preconditions for a successful trade-growth and FDI-growth nexus. While the results of these studies are highly relevant for policy makers, it is astonishing that governments in both the EU and ACP countries have not included these findings in the design of the EPAs so far.

With regards to the trade-growth nexus, three studies have examined this relationship. First, both Borrmann et al. and Freund and Bolaky show that good governance and high-quality institutions are necessary to reap the potential benefits of trade liberalisation.8 These include, for example, the quality of the bureaucracy, law and order, and control of corruption. In fact, without good governance, the costs of the reallocation of factor resources due to trade liberalisation could be much higher than the potential welfare gains, resulting in a net welfare loss. For example, if government regulations impede the start of a business (in the booming export sector) or hinder the transfer of labour from the import-competing sector to expanding sectors, high reallocation costs of capital and labour are likely to occur.

Chang et al. broadened the analytical framework and examined various further policy complementarities for the trade-growth nexus.9 Apart from good government regulations, they found that high human capital depth, well-developed financial markets, a good public infrastructure and sufficient labour market flexibility all play important roles in the relationship between trade and economic growth rates. The paper concludes that the growth effects of openness may be significantly improved if certain complementary reforms are undertaken.

As for the FDI-growth nexus, a number of papers with comparable identification strategies came to quite similar conclusions. For example, Borensztein et al. and Alfaro et al. found that the growth effect of FDI is significantly positive only when the host country has sufficiently high human capital and financial depth.10 Likewise, Busse and Groizard showed that government regulations are also a prerequisite for developing host countries to harness the benefits from incoming FDI flows.11 They argue that countries may only benefit from foreign investment inflows if appropriate local government regulations and institutions are in place. Above all, excessive regulations are likely to restrict growth through FDI if human and capital resources are prevented from reallocating. For example, if starting a business is hindered by extensive and costly government regulations which involve many bureaucratic procedures demanding entrepreneurs’ time and resources, then capital flows are prevented from being reallocated to the most productive sectors.

Likewise, if restrictive employment laws for the hiring and firing of employees lead to lower labour market turnover, technology spillovers from the multinational enterprise to domestic firms are less likely to occur. A similar argument can be made for other forms of government regulations, like those that pertain to closing a business, such as protecting foreign and domestic investors by ensuring creditor rights and enforcing contracts. Both of these are difficult tasks involving high uncertainty, considerable time and very large expenses. Hence, multinationals would reduce forward and backward linkages with the local economy, thereby affecting the likelihood of horizontal or vertical spillovers taking place.

To sum up, improving performance in these various policy areas could not only ensure better growth performance based on existing trade flows and FDI (partly in the form of increased technology spillovers). Indeed, it would also help to attract more of the type of FDI (and, thus, exports) that is closely related to a pro-development impact. As pointed out by Moran, foreign investment in the form of full-scale plants with cutting-edge technology and management practices, strong export orientation and integration into the supply chain of the multinational enterprise is much more likely to benefit the host economy than sub-scale plants that produce for the local market and that are protected from international competition.12

Benchmarking for ACP Countries

After identifying important determinants for a positive trade-growth and/or FDI-growth nexus, we subsequently perform a benchmarking for ACP countries. Since all the studies mentioned above included a considerable number of ACP countries in their empirical analysis, we can compare their absolute performance as well as their relative performance with other developing (and developed) countries.

We start with a set of six good governance indicators published by the World Bank.13 Researchers at the World Bank grouped various aspects of good governance into six dimensions: voice and accountability, political stability, government effectiveness, regulatory quality, rule of law and control of corruption. Based on different sources, they computed averages for almost all countries in the world and set up standardised indicators ranging from -2.5 to +2.5 with a mean of zero, where a higher figure indicates better governance.

Unfortunately, most sub-Saharan African countries have rather low governance scores. In fact, sub-Saharan Africa usually has the worst scores among all regions for measures of good governance (Table 1). Even worse, African political leaders and interest groups have considerable power in their countries, as democratic accountability, including several checks and balances, is not always ensured. This might in turn lead to abuses of power and can negatively affect other government areas, for example, government effectiveness or law and order. Likewise, excessive government regulations may lead to corruption due to complicated and costly bureaucratic procedures. As a consequence, Africa is a very difficult place to do business for entrepreneurs, and firms require a relatively high risk premium, which reduces investment and growth rates.

The other ACP regions in the Caribbean and the Pacific have – on average – better governance scores, though this depends on the particular indicator concerned. For example, while ACP Pacific countries score on average relatively high on voice and accountability, political stability and the rule of law, they have below average scores for government effectiveness, regulatory quality, and control of corruption (Table 1). On the other hand, Caribbean countries have scores near the mean for all governance areas. Yet these averages conceal enormous differences within the region, as countries such as Haiti score much worse than other CARIFORUM countries.

Regarding the second main policy area, educational attainment levels, again we observe that sub-Saharan African countries score relatively low. While African countries have improved their educational performance in recent years, they are still at the bottom in a regional comparison (Table 2). In fact, as of 2008, adults 25 years and older in sub-Saharan Africa attained on average 4.4 years of total schooling, much below the equivalent averages for the Caribbean (7.8 years) or Pacific countries (8.2) or for all developing countries (6.7).

With respect to one of the key financial indicators used by the studies cited above, namely domestic credit provided to the private sector in per cent of GDP in 2008, the picture is rather mixed (Table 2). While the equivalent figures for the three ACP regions are at, or somewhat above, the average for all developing countries (45 per cent of GDP), this figure is rather low in comparison to high-income countries (158 per cent). Hence, the relative performance is still unsatisfactory in ACP countries, as credit is much less widely available, which is partly the result of less well-established financial markets and inadequate financial regulations.

Finally, the last indicator of interest is the quality of public infrastructure, as measured by the number of telephone lines per 100 people in 2008.14 Here we find very poor performance in sub-Saharan African countries, where only two out of 100 people have (fixed) telephone lines. The equivalent figures for the Caribbean and the Pacific countries are much better with 25 and 12 lines per 100 people respectively.

Summing up, while the absolute and relative performance of Caribbean and ACP Pacific countries is rather mixed and depends on the particular country concerned, most of the sub-Saharan African countries belong to the group of nations for which the various empirical studies found that the FDI-growth or trade-growth nexus is not positive (and sometimes even negative). While it is clearly beyond the scope of this paper to explore the reasons for the underdevelopment of the educational, financial and infrastructure sector as well as the low governance quality in African countries, we can still draw the conclusion that most of these countries are not well equipped for trade liberalisation as part of the EPAs. In fact, sub-Saharan Africa scores much worse in almost all indicators in comparison to other income groups or regions. Accordingly, it is less likely to harness the benefits of trade and FDI.

Policy Implications

One of the most important policy implications of the studies mentioned above is the fact that a country has to reach a particular development level first before it can benefit from trade openness and/or FDI. Hence, the most pressing concern is that EPAs have to be designed with a pro-development orientation to ensure that ACP countries benefit from them. On the other hand, countries that are already well governed and that have a decent educational, financial and public infrastructure will benefit much more from trade liberalisation and FDI, but – intuitively – do not need the EPAs as much as those countries at lower development stages.

Regarding those ACP countries at lower development stages, it remains an open question whether signing an EPA itself will provide the expected benefits. Rather, it would be much more promising if the EPA trade liberalisation timeline were tied to progress made in the policy areas indicated above. Once a large number of countries within a regional EPA have achieved considerable progress in a given set of development indicators, then these countries could progress with the scheduled trade liberalisation. The European Commission should further support ACP countries to reach the agreed targets with financial and technical assistance. In this way, governments in ACP countries might be more convinced about the EPA process, have more trust in the EU as a negotiation partner, and have an incentive to further improve important policy areas without being overburdened by timeframes that are too demanding or completely unrealistic.

Overall, it should not be forgotten in the EPA negotiations that some ACP regions, in particular those in sub-Saharan Africa, include some of the poorest countries in the world with the worst preconditions for an ambitious EPA agenda. Whether or not the EPAs are suitable for changing this is highly ambiguous. The strict timeframe and the logic of trade negotiations, however, might not be adequate to deal with the needs of these countries.

Matthias Busse, Ruhr-Universität Bochum and Hamburg Institute of International Economics (HWWI), Germany.

• 1 European Commission: EU Trade Statistics: ACP Trade, Internet Posting: http://ec.europa.eu/trade/creating-opportunities/bilateral-relations/regions/africa-caribbean-pacific/.
• 2 Due to its special situation, Haiti has been excluded from the EPA and any tariff liberalisation so far.
• 3 Apart from the CARIFORUM group, which represents a group of 15 Caribbean countries, the other five EPA groupings include the Economic Community of West African States (ECOWAS), the Southern African Development group (SADC), the Eastern African Community (EAC), the Eastern and Southern African group (ESA), and the ACP Pacific region. The latter group consists of 14 (mostly small) islands in the Pacific.
• 4 D. Marti: Updating the Economic Partnership Agreements to Match Today’s Global challenges, in: Trade Negotiations Insights, Vol. 9, No. 1, February 2010, pp. 8-9.
• 5 The basic concepts of trade creation and trade diversion have been established in the literature by the seminal contribution of Jacob Viner: The Customs Union Issue, New York 1950, Carnegie Endowment for International Peace.
• 6 M. Busse, H. Grossmann: Assessing the Impact of ACP/EU Economic Partnership Agreements on West African Countries, in: Journal of Development Studies, Vol. 43, No. 5, 2007, pp. 787-811. Other studies include: M. Busse, S. Lüthje: Should Caribbean Countries Sign an Economic Partnership Agreement With the EU? Expected Challenges and Strategic Options, in: Journal of Economic Integration, Vol. 22, No. 3, 2007, pp. 598-618; A. Borrmann, M. Busse, M. de la Rocha: Consequences of Economic Partnership Agreements Between East and Southern African Countries and the EU for Inter- and Intra-regional Integration, in: International Economic Journal, Vol. 21, No. 2, 2007, pp. 233-254; S. Karingi, R. Lang, N. Oulmane, R. Perez, M. Sadni Jallab, H. Ben Hammouda: Economic and Welfare Impacts of the EU-Africa Economic Partnership Agreements. United Nations Economic Commission for Africa Working Paper ECA/TRID/06/05, 2005.
• 7 See R. Caves: Multinational Enterprises and Economic Analysis, Third Edition, Cambridge 2007, Cambridge University Press.
• 8 A. Borrmann, M. Busse, S. Neuhaus: Institutional Quality and the Gains from Trade, in: Kyklos, Vol. 59, No. 3, 2006, pp. 345-368; C. Freund, B. Bolaky: Trade, Regulations, and Income, in: Journal of Development Economics, Vol. 87, No. 2, 2008, pp. 309-321.
• 9 R. Chang, L. Kaltani, N. Loayza: Openness Can be Good for Growth: The Role of Policy Complementarities, in: Journal of Development Economics, Vol. 90, No. 1, 2009, pp. 33-49.
• 10 E. Borensztein, J. De Gregorio, J. Lee: How Does Foreign Direct Investment Affect Economic Growth?, in: Journal of International Economics, Vol. 45, No. 1, 1998, pp. 115-135; L. Alfaro, A. Chanda, S. Kalemli-Ozcan, S. Sayek: Does Foreign Direct Investment Promote Growth? Exploring the Role of Financial Markets on Linkages, in: Journal of Development Economics, Vol. 91, No. 2, 2010, pp. 242-256.
• 11 M. Busse, J. L. Groizard: Foreign Direct Investment, Regulations and Growth, in: World Economy, Vol. 31, No. 7, 2008, pp. 861-886.
• 12 T. Moran: Harnessing Foreign Direct Investment for Development – Policies for Developed and Developing Countries, Washington DC 2006, Center for Global Development.
• 13 Source: World Bank: Good Governance Dataset, 2009. Internet Posting: http://info.worldbank.org/governance/wgi/index.asp.
• 14 While this indicator excludes mobile phones and thus does not capture all telecommunication areas, it is a good proxy for overall infrastructure quality and widely available in developing countries.