The economic partnership agreements and regional integration in Africa : a comparative case study of the East African Community and the Southern African Development Community

(1)

UNIVERSITY OF AMSTERDAM

Word count: 24,020

Sophie van Huut

10363262 13 June 2017

Master Thesis Political Science: International Relations Supervisor: Dhr. dr. S. (Sebastian) Krapohl

Second reader: Dhr. dr. J.G.W. (Jasper) Blom

THE ECONOMIC PARTNERSHIP AGREEMENTS

AND REGIONAL INTEGRATION IN AFRICA

A comparative case study of the East African Community

and the Southern African Development Community

(2)

Picture on Cover:

Cargo ship outside Dar es Salaam Port Trafigura Images

(3)

Preface

This thesis is the result of the intensive but rewarding process of five months of research. Although the research area of regionalism was new for me, I have quickly developed a fascination and a vast

interest for the regional aspects of political economy. The master thesis project has been a challenging task in which I have developed varying skills: from accurate planning to logical reasoning and clear, attractive writing. Moreover, although the project has greatly appealed to my perseverance, I can say that I have enjoyed the research process and the rewarding results of extensively analysing the topic. There are a few persons I would like to express my gratitude to. First of all, my thesis

supervisor, Sebastian Krapohl, who was willing to take me on board as a research intern. If it weren’t for his flexibility I would not have been able to learn as much as I did about regionalism in the Global South. Sebastian has sparked my fascination for regionalism in sub-Saharan Africa and kept me on the right track during my moments of confusions in the process. Moreover, I would like to thank my dad for his critical and helpful comments, and my mom, for giving me confidence in what I was doing. Lastly, I would like to thank my sister, who did a great job in designing the cover page of my thesis.

(4)

Abstract

More and more countries are organising themselves in regional organisations. In sub-Saharan Africa, the emergence of several regional organisations has led to a patchwork of overlapping regional economic communities (RECs). The EU claims to be promoting and enforcing regional integration in Africa by negotiating the Economic Partnership Agreements (EPAs) with (groups of) African, Caribbean and Pacific (ACP) countries. However, it is questionable whether this is indeed the case as overlapping memberships problematize this effect and African states are often more motivated by extra-regional trade than by intraregional trade. This paper examines the effect of the EPAs on regional integration in sub-Saharan Africa, using an expended interpretation of the ‘extra-regional logic of regional integration’, developed by Krapohl (2017). Based on a comparative case study of the East African Community (EAC) and the Southern African Development Community (SADC) we claim that extra-regional privileges, in combination with the overlap of RECs on the African

continent, led states to negotiate EPAs in groups that do not follow the lines of the existing RECs. The extra-regional privileges can be obtained in (at least) two ways: by states having the position of a regional power and by states having least developed country (LDC) status. The fragmentation of RECs in different EPA groups negatively affects further regional integration. EPAs require a common external policy towards the EU, and if states within one REC have different policies towards the EU (as a result of different EPAs) further integration within the REC is problematized. This effect is especially strong, since the EU is an important trade partner to most African states.

(7)

List of Abbreviations

AU African Union

ACP African Caribbean Pacific CET common external tariff

COMESA Common Market for Eastern and Southern Africa

CU customs union

DRC Democratic Republic of Congo EAC East African Community

EACSO East African Common Services Organisation EAHC East African High Commission

EBA everything-but-arms

ECCAS Economic Community of Central African States ECOWAS Economic Community of West African States EPA Economic Partnership Agreement

ESA Eastern and Southern Africa

EU European Union

EVI Economic Vulnerability Index FDI foreign direct investment

FDLR Democratic Forces for the Liberation of Rwanda FTA free trade agreement

GDP gross domestic product GNI gross national income

GSP Generalise Scheme of Preferences HAI Human Assets Index

IGAD Intergovernmental Authority on Development IPE international political economy

ISM international support measure ITC International Trade Centre LDC least developed country NTB non-tariff barrier

REC regional economic community

RISDP Regional Indicative Strategic Development Plan SACU Southern African Customs Union

SADC Southern African Development Community

SADCC Southern African Development Coordination Conference TFTA Tripartite Free Trade Area Agreement

(8)

TDCA Trade Development and Cooperation Agreement UAE United Arab Emirates

UN United Nations

UNHDI United Nations Human Development Index USA United States of America

(9)

List of Figures

Fig. 3.1 Points of Interest 13

Fig. 3.2 Causal Argument 15

Fig. 5.1 EAC Trade Network 25

Fig. 5.2 Overlapping Membership EAC 29

Fig. 5.3 EAC EPA Timeline 29

Fig. 6.1 SADC Trade Network 43

Fig. 6.2 Overlapping Membership SADC 46

Fig. 6.3 SADC EPA Timeline 46

Fig. 6.4 The SADC and its EPA groups 48

List of Tables

Table 5.1 History of the EAC 20

Table 5.2 EAC: Intra- vs. Extra-regional Trade 22

Table 5.3 EAC: Most Important Trade Partners 23

Table 5.4 EAC: Total extra- and intraregional import and export values 24 Table 5.5 EAC LDC states and the LDC criteria and thresholds 27

Table 6.1 SADC: Intra- vs. Extra-regional Trade 40

Table 6.2 SADC: Most Important Trade Partners 41

(10)

Maps

Map 1: The East African Community

Source: http://www.dw.com/en/east-‐african-‐states-‐delay-‐signing-‐of-‐eu-‐trade-‐deal/a-‐19537689

Map 2: The Southern African Development Community

(11)

1. Introduction

All over the world countries are organising themselves in regional groups to strengthen their political, economical and security position. Especially since the 1980s, the number of regional organisations has increased (Schirm, 2002: 1). This ‘trend’ has also reached Africa, which has several (overlapping) regional organisations, creating a ‘patchwork’ of regional organisations. At this moment, the African continent has eight regional economic communities (RECs) recognised by the African Union (AU), of which the East African Community (EAC), the Southern African Development Community (SADC) and the Economic Community of West African States (ECOWAS) are the most integrated (African Union Commission et al., 2016). Research has been done on the motivations behind regional integration and concerns have been raised about the effectiveness and utility of the regional organisations. An useful new interest lies with the influence of external actors and factors on

initiatives of regional integration (Krapohl, 2017; Buzdugan, 2013). A very important external actor to Africa is the European Union (EU), as a trade partner, provider of foreign aid and for historical reasons (Braude, 2008). The EU is thus in the position to influence the process of regional integration in Africa in multiple ways (Buzdugan, 2013). Importantly, the EU claims to be a supporter of regional integration in Africa, positioning itself as the ‘good example’. The EU shows this support for example by financing regional organisations like the SADC (Muntschick, 2017). Another way in which the EU expresses its commitment to regional integration in Africa, is through its trade policy. EU-Africa trade falls within the policy on the Africa, Caribbean and Pacific (ACP) trade relations of the EU, which is regulated under the Cotonou Agreement since 2000. The EU has been negotiating Economic

Partnership Agreements (EPAs) since 2002, which were supposed to replace the trade component of the Cotonou Agreement by 2007. These EPAs are presented by the EU as a contribution to

development and to “regional economic integration by joining up smaller markets in larger EPA regions that were established by the ACP countries themselves” (European Commission, 2017a: 2). However, whether the EPAs do indeed contribute to regional economic integration in Africa is questionable. The ACP countries are free to form their EPA negotiation blocks as they wish. The overlapping memberships to RECs on the African continent, different trade interests and varying treatments by the EU are just some of the factors that might complicate the EPA process and deeper regional integration. The main question of our research is therefore:

What is the effect of the Economic Partnership Agreements (EPAs) on regional

integration in Africa?

This question is of relevance not only because of the EU’s claims of promoting regional integration but also because of the role regional integration plays, or can play in the economic development in Africa and its inclusion in the world economy.

(12)

question. In the third chapter we will link the theories to our research question and formulate two subquestions and hypotheses. After this, in chapter four, we will present our research design and methods, including possible limitations of our design. Chapter five will focus on our first case: the East African Community. The second case will be discussed in a similar way in chapter six: the Southern African Development Community. Finally, in chapter seven we will formulate out conclusions based on our hypotheses and comparative case study. Here we will also present some possible directions for further research on the topic.

(13)

2. Relevant Theories and Literature

The literature on regionalism deals with multiple debates. Questions revolve around how regional cooperation and integration can be defined, what the motivations are behind regional cooperation and regional integration and also what effects it has on multiple issues. This also brings along questions about the actors that are involved and their motivations. Before we move on to discuss different theoretical views on these questions (in paragraph 2.2 and 2.3) we will first explain the relevant concepts (paragraph 2.1). In paragraph 2.4 we assess the work on extra-regional involvement in regionalism, especially from Muntschick (2012) and Krapohl (2017). Their logic forms the theoretical framework of our research. However, we suggest some additions to their logic, in light of the

Economic Partnership Agreements.

2.1 Defining relevant concepts

Different concepts play a role in the study of regions, with shifting meanings and understandings. In international relations, regions should be understood as macro-regions which can be defined as “supranational subsystems within the international system, whose constituents are states that are geographically close and share some degree of interdependence” (Muntschick, 2012: 4). Besides regions, four other important relevant concepts should be discussed: regional coopertaion, regional intergration, regionalism and regionalisation. These concepts differ in their definition and complexity. Regional cooperation generally refers to “joint efforts by states to solve specific problems” (Hettne, 2005: 545). Regional integration is more complex, and should, according to Joseph Nye, be broken down into three categories: economic integration, social integration and political integration (Nye, 1971 as discussed in Hettne, 2005: 544-545). Regional integration is more complex, as it generally implies “some change in terms of sovereignty” (Hettne, 2005: 545), whereas regional cooperation does not. Thus, “the study of regional integration is concerned with explaining how and why states cease to be wholly sovereign” (Haas, 1970: 610). Regional cooperation is thus a more non-binding form of regional integration.

Moreover, a distinction has been made between regionalism and regionalisation. The former refers to “a tendency and a political commitment to organise the world in terms of regions; more narrowly, the concept refers to a specific regional project” (Hettne, 2005: 545). Muntschick follows Bach in defining regionalism and describes it as “planned, multilateral, and state-led organisation of interdependence within a confined regional space that manifests in various specific regional projects and accompanying institutions” (Bach, 2003: 22 as discussed in Muntschick, 2012: 4). Hettne & Söderbaum describe regionalism as an ideology and programme (2000: 458). Regionalisation is more complex, and refers to “the processes of forming regions”, which should be studied empirically (Hettne, 2005: 545; Hettne & Söderbaum, 2000: 458). Regionalism thus generally refers to a project at

(14)

a certain point in time, while regionalisation refers to the process of this project.

2.2 Old Regionalism

A common way to look at regionalism (after the Second World War) is by distinguishing two waves: ‘old regionalism’, which began in the 1950s and stagnated in the 1970s and ‘new regionalism’ which emerged in the 1980s (Hettne, 2005: 546; Hettne & Söderbaum, 2000: 457; Manfield & Milner, 1999: 590; Bhagwati, 1993). The old regionalism was mainly a European phenomenon and was initially mainly theorised from functionalist theories. The functionalist approach took ‘form’ as subordinate to ‘function’, thus arguing that international cooperation should be based on functional activities like transportation, trade, production and welfare (Hettne, 2005: 546). Later, the neofunctionalists argued that these functional activities could not be separated from politics, and that cooperation in one field would eventually lead to cooperation on a political level. The neofunctionalist approach was mainly used to explain the European integration. These studies generally concluded that “increased economic flows among members of the EU have changed the preference of domestic actors, leading them to press for policies and institutions that promote deeper integration” (Mansfield & Milner, 1999: 602). These theories claim that domestic preferences and spillover effects influence preferences for regional integration. Important neofunctionalist explanations for regional integration have been worked out by Deutsch et al. (1957), Haas (1958; 1970) and Nye (1971).

2.3 New Regionalism

Around the end of the 1980s and beginning of 1990s a new wave of regional cooperation emerged. The international political context changed after the end of the Cold War which meant states changed their foreign policies. The usefulness of neofunctionalism in the debate on regionalism started to be questioned as replications of the ‘example’ of the EU failed in other regions (Warleigh-Lack & Rosamond, 2010). These failures of regional integration were hard to explain from the existing theoretical work. Neofunctionalism lacked a specification of “the conditions under which societal demands for integration became accepted at the national level”, which resulted in a rather “naive view” of regionalism (Mattli, 1999: 5). Thus scholars of the ‘new regionalism’ left neofunctionalism behind and used neoliberalism and the process of globalisation to explain regional integration (Hettne, 2005). For the neoliberalists, the new regionalism could be seen as “a trade promotion policy, building on regional arrangements rather than a multilateral framework” (Hettne, 2005: 549). However, studies of international political economy (IPE) did not hold free trade at the centre of regionalism, but also saw cooperation on other levels (e.g. security and environment) as a possible beneficial contribution of regionalism (idem). Although it has been argued that scholars should not reinforce the divide between studies on European integration and studies on the ‘new regionalism’ (Warleigh-Lack & Rosamond, 2010), most literature on the ‘new regionalism’ takes a different approach and seeks to explain new forms of regionalism.

(15)

An important debate in the literature revolves around the driving forces of regionalism: what drives the emergence, evolution, functioning and effects of regional integration? A division exists between a top-down and a bottom-up approach. Some authors argue that external factors like globalisation have forced states to organise themselves in regions, while others focus on internal forces that drive regional integration.

Bottom-up approach

Mattli (1999) proposes an approach to regional integration that includes the demand side and the supply conditions. For the demand side conditions, he uses ideas of institutional theories to explain demands for regional integration. He points at profits that can only be gained if the institutional arrangements are reformed which means that if the possibility of these profits grows (e.g. by technological development) the demand for institutional reform will grow (Mattli, 1999: 10). The possible gains are specified using theories of international trade and investment, pointing at economies of scale (increasing production lowers average costs) and the comparative advantage (the benefits of specialization).

However, Mattli (1999) argues that the demand conditions alone are not enough to make regional integration succeed. The supply conditions also play a crucial role. He defines supply conditions as “the conditions under which political leaders are willing and able to accommodate demands for functional integration”. Mattli states that the willingness greatly depends on political leaders’ ‘economic success’. There will be no incentive for leaders to seek regional integration if “their expected marginal benefit from integration in terms of improved re-election chances is minimal and thus not worth the cost of integration” (Mattli, 1999: 12). The situation changes if political leaders find themselves in economic difficulties and feel the need to improve the economy to improve their position. Besides willingness, Mattli also points at collective action problems as a potential obstacle at the supply side. He discusses the prisoners’ dilemma and coordination games, which can be solved by strong institutions and the presence of a regional leader (Mattli, 1999: 14-15).

Mansfield & Milner’s (1999) conclusion that flows from their overview of literature on regionalism relates to Mattli’s view. They conclude that “the preferences and political influence of different societal groups are likely to affect whether a state enters a preferential trading arrangement” (Mansfield & Milner, 1999: 619). This is comparable to what Mattli calls ‘demand conditions’. Manfield & Milner also acknowledge the supply side by pointing at the effect of the preferences of government officials on efforts of regional integration.

Top-down approach

Hettne (2005) argues that the ‘new’ initiatives of regional integration should be seen in the light of globalisation and the unequal world order it has resulted in. He states that: “much of today’s

(16)

the WTO and globalisation” (Hettne, 2005: 551), thus focussing on trade agreements. These trade agreements can be initiated to enhance development in a region. Hettne argues that development requires “a regional approach, whereby trade integration is coupled with other forms of economic and factor market integration, as well as various types of economic cooperation in specified sectors” (2005: 552).

Schirm (2002) also takes a top-down approach. He sees the global economy as the driving force that makes states cooperate in regions. According to Schirm, states shift their focus from national policies to transnational cooperation because globalization and the emergence of a global market makes this necessary. As Schirm states: “global markets imply a deterritorialization and denationalization of economic activity because their operational logic transcends the functional logic of states” (2002: 12). Thus the logic of the global markets changes the way in which states are best governed and increases the profits of world market-oriented liberalization. States still have to make decisions on opening their markets and participating in regional cooperation, but they are motivated by global markets, according to Schirm (2002). In some way, Schirms approach can be linked to Mattli’s (1999) idea of the ‘demand side’; the demands within a state change as globalisation and global markets gain influence. Schirm thus adds an explanation of the changing domestic demands, by pointing at global markets as the explanatory factor.

Other theoretical contributions

While the drivers of regionalism can be explained in a vertical perspective, claiming for either a bottom-up or a top-down approach, this can also be done in a horizontal perspective in which regional organisations influence each other and are interdependent (Jetschke & Lenz, 2013). Jetschke & Lenz argue for a diffusion perspective which explains the clustering of regional organisations and their similarities (2013: 634). In this sense, the EU takes a central position in regionalism as its example has been ‘copied’ by other regions, which can be explained by the diffusion argument.

Lenz (2012) contributed to the work on the influence of the EU on efforts of regional integration in different parts of the world. This work analyses different processes through which the EU may influence regionalism outside Europe (e.g. through learning, emulation and persuasion) (Lenz, 2012: 156). Lenz maps these contributions and highlights two main factors that play a role in the EU spreading its influence on regional integration outside Europe. He argues that regionalism follows the example of the EU when the demands for institutional change exist and when domestic actors actively call for this change to be ‘EU-style’ (Lenz, 2012: 169-170). This leads Lenz to conclude in a vision of so-called ‘spurred emulation’.

Hettne & Söderbaum (2000) also do an attempt to lay the basis for a theory of the new regionalism, which is a combination of both top-down and bottom-up approaches. They depart from global social theory, social constructivism and comparative regional studies, and use these theoretical frames as their buildings blocks. Their aim is to form a theory that integrates macro- and micro

(17)

perspectives on regionalism and that sees regions as social constructions. Hettne & Söderbaums see regions as entities in process; regions have a certain degree of ‘regionness’ which can increase and decrease. They describe five generalised levels of regionness, which represent five phases of the process of becoming a region. Their focus is not on states and their existence in a certain region, but on the process of regionness, which transcends state-centrism (Hettne & Söderbaum, 2000: 462). The five levels they distinguish are: regional space, regional complex, regional society, regional

community and region-state. These should be seen as successive phases representing the degree of regionness, but Hettne & Söderbaum do not whish to imply that this is the (only) path every region walks. Instead they argue: “many regionalisation processes and experiments are initiated from quite different starting points in terms of regionness” (Hettne & Söderbaum, 2000: 470). Moreover, they deal with the observation of failed regional cooperation by pointing out that regions can be

deconstructed by the same forces that construct them. Hettne & Söderbaum stress that their approach merely aims to form a framework or starting point for further theoretical work, and their discussion is indeed still rather simplistic and unsatisfying. It does not give a suggestion on the circumstances or conditions that underlie the different levels or stages.

Besides the drivers behind regional integration, attention has also been raised for the effects of regionalism. A current point of interest in regionalism is how regional integration affects the global trade order. This resulted in a debate on whether trade blocs are ‘building or stumbling blocs’ for the world trading system (Baldwin & Seghezza, 2010; Bhagwati, 2008). As explained by Baldwin & Seghezza: “In this debate, trade blocs are stumbling blocs if they prevent or slow multilateral tariff cutting, while they are building blocs if they accelerate or at least do not hinder multilateralism” (2010: 277). Both points of view can be convincingly argued and which one prevails is still open for debate. However, it seems that more authors accept the view that regional trade agreements help, instead of hinder global trade liberalisation (Krapohl, 2017; Baldwin, 2006).

2.4 Extra-regional involvement

Many cases of the ‘new regionalism’ took a significantly different path than the process of regional integration in Europe. One important observation is that the intra-regional trade is of much less importance to developing regions, and instead the focus lies on extra-regional trade. In developed regions, like the EU, the intra-regional trade is dominant and makes states dependent on trade with their regional partners. This makes theoretical approaches focussed on developed regions less useful for cases of regionalism in developing regions. Moreover, most theoretical work discussed above, focuses on actors within the region or on the forces of globalisation and global markets. However, states also engage in extra-regional dynamics, both as a region and as individual member states. Especially if intra-regional trade is not significant, regions might be receptive to extra-regional influences.

(18)

several authors (Krapohl & Muntschick, 2009; Muntschick, 2012; Buzdugan, 2013; Krapohl, 2017). They point at the inadequacy of existing theories to explain efforts of regional integration in the South. Buzdugan (2013) argues that regionalism is not truly a states-led strategy, but instead a process that is directly influenced by extra-regional actors, especially in sub-Saharan Africa. He pledges for “greater attention to the role of international actors” in the emergence of regions (Buzdugan, 2013: 940). In his work, Buzdugan claims that “in the case of sub-Saharan Africa, there is a strong external dimension to sovereignty” (2013: 927). This makes theories focussing on intra-regional actors unsuitable to explain regional integration in sub-Saharan Africa.

Buzdugan also criticizes the recent literature on ‘interregionalism’, because this implies an internal dimension of sovereignty, which, according to Buzdugan, does not exist in regions in sub-Saharan Africa. Literature on interregionalism sees the relations between regions as comparable to bilateral relations between countries (Söderbaum & Van Langenhove, 2005: 259). Its focus lies mainly with the relations that the EU, as a region and actor, holds with other regions. This is a useful

development in the literature on regionalism in the sense that it acknowledges the influences regions can have on each other. However, the extra-regional actors relevant to regionalism, do not necessarily have to be regions themselves. A comprehensive theory on regionalism that includes extra-regional actors has been developed by Muntschick and Krapohl.

The extra-regional logic of regional integration

Krapohl and Muntschick (2009) have developed an ‘extra-regional logic of regional integration’, deduced from the new regionalism literature. They focus on economic integration and on trade relations. Their theory is based on the idea that “member states of regional organisations cooperate in order to achieve gains in their interaction with extra-regional actors” (Krapohl, 2017: 13). It is thus the interaction with other world regions that determine the extent to which regional integration is desirable and beneficial. To some extent, they build further on the rational-choice and collective action

problems approach also taken by Mattli (1999). However, instead of just looking at demand and supply within a certain region, Muntschick and Krapohl involve the extra-regional conditions to explain successes and failures of regional integration.

Much of the benefits states can gain from integration in the region are extra-regional benefits, instead of intraregional. When states integrate successfully, they can gain from “increased investment inflows from other world regions” and “better market access to other world regions” (Krapohl, 2017: 14). These benefits stem from the enlarged market size as a result of regional integration, and

improved political and economic stability of integrated regions. These arguments can be linked to the influence of the global market on benefits of regional integration (discussed by Schirm, 2002). States can benefit more from these resources if they integrate in the region, but whether they have an interest in this depends on the existing intra-regional and extra-regional relations. The competition over these extra-regional benefits in a region can lead to two different situations: a ‘battle of the sexes’ or a

(19)

‘Rambo situations’. Whether the one or the other situation emerges depends on the behaviour of extra-regional actors. The extra-extra-regional actors can do two things: 1) they can “systematically reward regional integration”, thus increasing investment in, and granting market access to cooperative member states of the regional organisation, or 2) they can not systematically reward regional integration and instead treat member states of the regional organisation on other bases than their commitment to regional organisations (Krapohl, 2017: 46). The first extra-regional response leads to a ‘battle of the sexes’, while the second leads to a ‘Rambo situation’. In a battle of the sexes, all member states have equal interest in regional integration, which makes regional integration more likely. In contrast, in a Rambo situation, one, or more, member states are privileged by extra-regional actors, which makes regional integration harder. Privileged member states might lose interest in regional cooperation if they gain more from their privilege than from regional benefits, and would have to give this privilege up to integrate in the region (Krapohl, 2017: 57).

This implies that states in a region have different positions in bargaining on cooperation and integration projects, depending on the “availability of attractive unilateral policy alternatives and exit-options” (Muntschick, 2012: 7). Whether they succeed in regional cooperation thus depends on “the relative power-position of the negotiators involved” (idem).

In sum, to assess the possible success of regional integration, one should first analyse the trade relations within a region and with its extra-regional partners, to determine the existence of regional power. Secondly, one should analyse the treatment of the region’s states by extra-regional partners. In this way, the existence of “attractive alternative exit options for regional actors”, that problematize regional integration are brought to the foreground (Muntschick, 2012: 11).

Adjustment and addition to ‘the extra-regional logic of regional integration’

As our research focuses on the influence of the EPAs on regional integration in sub-Saharan Africa, we need to include considerations on extra-regional influences. In a way, our contribution can be seen as a case of ‘interregionalism’. However, we do not aim to focus on the EU’s foreign and interregional policy and their motivations behind the EPAs, explaining “why and how the EU promotes

interregionalism” (Söderbaum et al, 2005: 378). Instead, our interest lies with the impact of the EPA on regionalism in sub-Saharan Africa. We aim to explain what happens in Africa as a result of this form of interregionalism with the EU, instead of the nature of, and motivations behind this

interregional phenomenon. We see the EPAs as a process that lays blank the extra-regional influences on regional integration in sub-Saharan Africa. It shows the rationalities of states’ considerations in negotiating the EPAs, based on their position within the region, and their relations with extra-regional actors.

To a large extent, the extra-regional logic of regional integration developed by Krapohl (2017) forms a useful framework that can explain the process of regionalism in the global South. However, we feel that this theory should be expended to explain the process of the EPAs and

(20)

regionalism. We are not convinced that states within the region that receive extra-regional privileges, necessarily have to be regional powers. States can receive privileges from extra-regional actors, without being a regional power. This is illustrated by the Least Developed Countries (LDCs) that receive preferential market access to the EU. These states are the least developed, and least powerful states in economic terms, but do receive extra-regional privileges that may influence their bargaining position in the process of (further) regional integration. This means that any state in a region can receive extra-regional privileges that conflict with further regional integration, which leads states to defect from regional integration. This is not to say that the economic power of states is irrelevant; a regional power will easier defect from regional integration because of extra-regional privileges, because the other regional states do not have leverage on the regional power. A minor power state in a region that enjoys extra-regional privileges (by e.g. its LCD status) can be ‘threatened’ by other states by border closures, or ‘bribed’ by side payments to make the minor power state cooperate in regional integration. In other words, other states can make defection more costly for the minor power enjoying extra-regional privileges, which makes regional cooperation the preferred option for the minor power. In case a regional power enjoys extra-regional privileges, it will be harder for the other regional states to make regional cooperation the most beneficial option, as they are economically weaker than, and often dependent on the regional power. Thus, although economic power is relevant to extra-regional privileges, it is not a necessary condition to receive extra-regional privileges.

Besides this adjustment, we also propose an additional variable that should be included in the extra-regional logic of regional integration. This is the factor of overlapping membership, which is particularly relevant to regionalism on the African continent, where great overlap exists in

memberships to regional organisations (De Melo & Tsikato, 2015; Hartzenberg, 2011; Braude, 2008; Sindzingre, 2008; Draper et al., 2007; Jakobeit et al., 2005). While almost every African state is a member of more than one REC, not much theoretical attention has been given to the implications of this phenomenon. Instead, most studies focus on the empirical implications of this membership overlap.

In light of the extra-regional logic of regional integration, we can state that the large number of RECs in Africa can be attributed to the focus on extra-regional benefits of regional integration. If states were mostly interested in improving conditions of intraregional trade, one big regional

organisation would be more beneficial. Instead, African states, and especially the regional powers, are more interested in extra-regional trade. In this sense, they benefit more from joining several regional organisations, which makes them an attractive trading partner for extra-regional actors. Unlike in Europe, the intra-regional trade flows are not strong enough, to ‘drag’ all (or most) states to one, central REC.

This fragmentation in multiple, overlapping RECs forms a stumbling block for the formation of EPA groups, as states can only be part of one EPA. This leads to competition between the different regional organisations, as an REC that does not form one EPA together will have trouble to integrate

(21)

further. The variable of overlapping REC memberships can thus be integrated in the extra-regional logic of regional integration because extra-regional forces likely caused the existing overlap. In turn, this overlap is of significant relevance to regional integration as it problematizes the process of deeper integration. We see the membership overlap as another independent variable that problematizes the formation of an EPA group. RECs can compete with each other, when trying to form EPA groups, on how deep and successful their integration already is and on grounds of the power and relevance of their members. Thus, membership overlap and presence of a state that enjoys extra-regional privileges simultaneously problematize the formation of an EPA group.

In sum, we use the extra-regional logic of regional integration as the basis of our theoretical framework. We make one adjustment and one addition to the theory. Firstly, we reconsider the importance of economic power in relation to extra-regional privileges. We claim that states can enjoy extra-regional privileges that influence their intraregional interests without being a regional power. Secondly, we emphasize the context of membership overlap as an additional variable in the process of regional integration. In the following chapter, we will discuss our subquestions and hypotheses based on this logic.

(22)

3. Subquestions and Hypotheses

Subquestions

As mentioned in the introduction, the main question we aim to answer is: what is the effect of the

Economic Partnership Agreements (EPAs) on regional integration in Africa? To determine the

influence of EPAs on regional integration in Africa we have to distinguish two steps in the process of the EPAs. First, countries need to organise themselves in negotiation blocks to negotiate EPAs with the EU. All ACP countries got complete freedom from the EU to determine the group in which they want to negotiate an EPA with the EU. It is therefore worth exploring what factors played a role in the formation of these ‘EPA groups’. If these EPA groups follow the same lines as the already existing RECs, this probably eases the process of regional integration. However, especially in Africa the EPA groups do not always match the existing RECs. Several countries in Africa have negotiated an EPA individually with the EU, and other EPA groups consist of only a few countries within one REC. Moreover, there is a great overlap in membership of regional organisations in Africa, creating a patchwork of regional organisations. Most African countries are a member of more than one REC, which gives them several options in choosing an EPA group. But what explains the fact that some regional organisations negotiate on one EPA together, while others split up in several EPA groups? This is important to consider, because if the EPA group is not in line with the existing REC, the EPA will likely complicate regional integration. An important aspect of regional integration, a customs union, requires a common external trade policy, which is (almost) impossible if member states do not belong to the same EPA group. Countries can have several reasons to negotiate an EPA together, and negotiating one EPA together might be easier under certain circumstances. Our first sub question is therefore:

Subquestion 1: Under what circumstances do regional organisations form one ‘EPA group’

together?

Secondly, when an EPA group has been created and an EPA agreed upon, we can look at the effects this agreement has on integration in the region. Does the EPA have certain spillover effects on regional integration? As mentioned, we expect EPAs to have a positive influence on regional

integration within the countries of one EPA, because they share one common external trade policy, but is this really the case? Our second sub question is thus:

Subquestion 2: To what extent, and under what circumstances do EPAs contribute to

(23)

This second sub question might be harder to answer, because most EPAs with African countries are either just signed and in the process of implementation, or still negotiated upon. However, we believe it is useful to already look at the possible effects of the agreements in this phase of the process. Figure 1 shows an illustration of our points of interest.

Hypotheses

Now we have formulated our subquestions, we will continue by explaining our hypotheses based on our expanded understanding of the extra-regional logic of regional integration. For both subquestions we will determine our expectations.

The first subquestion focuses on the formation of EPA groups in Africa. What factors

determine which countries negotiate one EPA together? A starting point is that we expect EPA blocks to (roughly) follow the same lines as the existing regional organisations, because this eases the process. However, several factors might problematize this, and lead to different outcomes.

Based on the theory of Krapohl (2017) we expect that the presence or absence of states that enjoy extra-regional privileges to be the most important factor determining whether one regional organisation also creates one EPA group. In this context, especially extra-regional privileges from the EU are important, not only because the EPAs are negotiated with the EU, but also because the EU is the most important trade partner to most states we analyse. As described in the previous paragraph, the extra-regional logic of regional integration shows that states that enjoy privileges from extra-regional partners do not have the same motivation to form one regional block as less privileged countries in the same region. In such situations, the region ends up in a ‘Rambo situation’. If all states enjoy the same treatment from extra-regional actors, they find themselves in a ‘battle of the sexes’. The latter situation makes regional cooperation and integration more likely. This trend might also be visible in the

forming of EPA groups. Only if countries have the same, or comparable motivations to negotiate an EPA with the EU, they will successfully create an EPA group together. It might be possible that a state already enjoys trade privileges from the EU, which gives the state an alternative to the EPA and

(24)

changes its position in the negotiations. This effect of privileges can emerge from (at least) two situations; states can enjoy extra-regional privileges because they are a regional power, or because they are a least developed country. In the first scenario (privileged regional power), the less powerful states will not have the resources and position to ‘force’ the privileged regional power into regional cooperation. In contrast, in the second scenario (privileged least developed country), it might be easier for the other regional states to ‘force’ the privileged least developed country into regional cooperation. The extra-regional privileges of a state can thus be a result of its powerful position within the region, or instead of its least developed country status.

On the other hand, if countries are comparable in their economic power and position, they might gain a lot from negotiating an EPA together. They get a better negotiation position as the size of their market increases and because the stability of their region improves, as countries become

dependent on each other’s cooperation. However, these benefits are not worth much to states that enjoy extra-regional privileges, especially if they are economically powerful. Our first hypothesis is therefore:

Hypothesis 1a: The existence of a state that enjoys extra-regional privileges

makes it unlikely that a REC forms one EPA group, especially, but not only, if

this ‘privileged’ state is a regional power.

As discussed above, we distinguish another factor that problematizes the formation of EPA groups: the overlap of different RECs on the African continent. As most African states are a member of more than one REC, and they can only join one EPA group, competition will exists between different RECs to negotiate one EPA including all the member states. Having multiple memberships, gives states several options in the EPA process. At the same time, this inevitably means that not all RECs can negotiate one inclusive EPA together. Some RECs will thus experience the (negative) effects of a division in more than one EPA group. How states decide which EPA group to join, will depend on the existence of extra-regional privileges and the (a)symmetry in economic power, which is the topic of hypothesis 1a. However, if there were no membership overlap in RECs in Africa, the process of forming EPA groups would probably be much less complex. Hypothesis 1b is thus:

Hypothesis 1b: Overlapping memberships in RECs problematizes the process of forming one

EPA group together.

Subquestion 2 relates to a different phase of the EPAs. If an EPA is agreed upon, between a group of African states and the EU, what does this mean for regional integration? Our expectation is that EPAs positively influence regional integration, that is, that states will become more integrated if they agree on one EPA together. Agreeing on an EPA with the EU, means that states share the same external

(25)

trade policies towards the EU, and enjoy the same trade treatment from the EU. As the EU forms an important trade partner for most African countries, this eases the process of regional integration. For example, to form a customs union, states need to have the same external trade policy; an EPA makes this easier. These considerations lead us to the following second hypothesis:

Hypothesis 2: Different EPA groups problematize the process of regional integration

of the REC

In sum, based on the extra-regional logic of regional integration we thus believe that the

presence/absence of a state that enjoys extra-regional privileges is the most determining factor in the formation of EPA groups. Moreover, overlapping memberships give states alternative options in the EPAs, which problematizes the process of forming one EPA group. As for regional integration, we expect that an EPA will ease this process among the countries of the same EPA group. Figure 3.2 shows a schematic overview of our causal argument.

Fig. 3.2: Causal Argument

(26)

4. Research Design

Comparative case study

Within our hypotheses we distinguish two dependent variables. First of all: ‘the forming of different EPA groups’ (y1), and secondly: ‘deeper regional integration’ (y2). For y1 we identify two independent variables: ‘presence of regional state that enjoys extra-regional privileges’ (x1) and ‘overlapping memberships in RECs’ (x2). For y2 the independent variable is ‘forming of different EPA groups’ (x3). Note that the first dependent variable (y1) turns into an independent variable in the second hypothesis. In sum:

Hypothesis 1a: The presence of a state that enjoys extra-regional privileges (x1) →

Different EPA groups (y1)

Hypothesis 1b: Overlapping memberships in RECs (x2) à Different EPA

groups (y1)

Hypothesis 2: Different EPA groups (x3) → No deeper regional integration (y2)

To test our hypotheses we will perform a comparative case study with two cases. We follow “Mill’s Method of Difference” (Mill, 1843, as discussed in Ragin, 1987) to select our cases. This means that we will discuss two cases that exist in a similar context but show different outcomes, so that we can comment on the possible explanations for the difference in outcome. Thus we select two cases of regional organisations in sub-Saharan Africa that are similar in most aspects, but differ on the independent variables. For the independent variables ‘presence of state that enjoys extra-regional

privileges’ (x1) and ‘different EPA groups’ (x3) we can find variety among the RECs on the African

continent. However, the independent variable ‘overlapping memberships in RECs’ (x2) counts for (almost) all RECs in Africa, as all important RECs overlap with other RECs. There does exist some variety as to how divided the member states are over different RECs. This will be discussed further in the next sections. Opting for a comparative case study means we will use qualitative research methods to analyse the cases.

Case selection

To select our cases we first have to define ‘regional organisations’, to determine which cases qualify as such. We will follow the definition formulated by Krapohl, who defines regional organisations as: “Agreements between more than two neighbouring states that set up at least

(27)

rudimentary forms of organisational structures (…) and grant some kind of preferential market access for each other.”

(Krapohl, 2017, 66).

The African Union (AU) recognises eight such regional organisations in Africa, which can be supplemented by another three, recognised by the WTO, which brings us on a total of 11 regional organisations in Africa. We will select our cases from the ones that are recognised by the AU. Within these eight cases, we select two cases that are similar on most variables, but differ on the variables that we see as explanatory to the dependent variables. This brings us to selecting the East African

Community (EAC) and the Southern African Development Community (SADC).

The East African Community

The EAC consists of six member states (Tanzania, Kenya, Uganda, Burundi, Rwanda and South Sudan, who joined in 2016), all of which are least developed countries (LDCs), except Kenya.

Although Kenya has the highest GDP, and is not considered a LDC, it can hardly be seen as a regional power, as the other member states do not seem to be dependent on their trade with Kenya. Further research is necessary to confidently claim that the EAC does not have a regional power that enjoys extra-regional privileges, but we expect the member states to be relatively equal to each other. As five of the six member states are LDCs, there are states that receive extra-regional privileges. However, the nature of these privileges are different from the privileges that play a role in the SADC, where a regional power can be recognised.

As mentioned, almost all African RECs overlap with other RECs. EAC, which has six

member states, overlaps with four other RECs. As for the last independent variable (x3): five of six the EAC member states (leaving out South Sudan) finalised negotiations on an EPA with the EU in 2014, meaning the EAC scores negative on this independent variable (forming different EPA groups).It should however be noted that although the EAC finalised negotiations on an EPA, the agreement has not been signed and ratified by all EAC states. This will be further discussed in chapter five.

The Southern African Development Community

The SADC consists of fifteen member states (Angola, Botswana, Democratic Republic of the Congo, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Seychelles, South Africa,

Swaziland, United Republic of Tanzania, Zambia and Zimbabwe). Tanzania is both a member of the EAC and the SADC. Judging from the work done on the SADC, South Africa can be seen as a regional power that enjoys extra-regional privileges (Krapohl & Muntschick, 2009; Krapohl, 2017). SADC thus scores positively on our first independent variable, as it has a regional power that enjoys extra-regional privileges.

(28)

many member states, but overlaps with as much with other RECs as EAC; with four other RECs. The SADC did not form one EPA group. The SADC-EPA group consists of only six SADC member states (Botswana, Lesotho, Mozambique, Namibia, South Africa and Swaziland, possibly seven if Angola joins in the future), while the other eight member states are negotiating EPAs as part of other groups. In contrast to the EAC, the SADC thus scores positively on the second independent variable, as barely half of the regional group negotiated an EPA together, and they thus negotiated on different EPAs.

The dependent variables

The dependent variables we distinguish are: forming different EPA groups (y1) and no deeper regional integration (y2). Y1 is easy to determine as we can simply look at the EPA negotiation groups to see if they follow the same lines as the existing regional organisations. The second dependent variable is harder to measure, as it is disputable what ‘regional integration’ exactly means. For the aim of this research we follow Krapohl’s definition of regional integration: “a process of iterated cooperation that leads to an ever-increasing web of substantive and procedural institutions at the regional level” (2017: 5). Following this definition, we can measure regional integration by looking at the degree of regional cooperation. This can be determined by assessing how far the region is on the ‘integration ladder’; from a free trade area to a customs union, to a common market to a monetary union, and so on.

Methods

As mentioned, we will use qualitative research methods to analyse our cases. To make a fair comparison we will collect the same information on both cases, using similar sources. The methods will be desk research. We will use policy papers, secondary-literature analysis and media reports to form a description of both cases. Furthermore we collect data on trade flows from Trape Map. We focus on the regions’ intra-regional trade regimes, their extra-regional trade regimes, their (trade) relation with the EU, their EPA negotiation process and their regional integration process.

Possible limitations

Opting for a case study to find answers to our research question automatically brings along questions about the generalizability of our results. Will our observations be typical for all regional organisations in Africa? And perhaps even beyond the African continent? The latter will be harder to argue, but we believe that our study will be useful in the sense that if our hypotheses are confirmed in these cases, this also tells us something about other cases of regional integration in Africa, because they are similar on the most important aspects. In order to make reasonable claims about this, we will have to make sure to place our cases within the broader context of other cases.

Another possible limitation relates to the internal validity. It is impossible for us to involve all possible explanatory variables in the case study. However, to make the chance of missing important and relevant observations as small as possible, we will analyse our cases as extensively as is

(29)

reasonable for this study. This means doing an in-case analysis, to see if the theory likely holds within the case. We need to form an extensive picture of both cases, comparing them on several variables, like organisational structure, intraregional economic and development diversity and intraregional relations.

A third possible limitation is that the two cases vary considerably in size (6 and 15 member states), while we tried to find similar cases. However, Krapohl (2017) has shown that this does not play a considerable role in the theory he develops; the ‘extra-regional logic of regional integration’ works the same in both smaller and bigger regions. We can thus confidently use two regional organisations that vary in size.

(30)

5. The East African Community

5.1 Introduction

The East African Community (EAC) is one of the oldest of the regional organisations in Africa. The ‘first’ EAC originates from 1967, but lasted only until 1977. Even before this, Kenya and Uganda had been working together under one customs union from 1917 onwards, which was joined by Tanganyika (now part of Tanzania) in 1927. This customs union was replaced by the East African High

Commission (EAHC), which lasted from 1948 to 1961. The EAHC was then followed up by the East African Common Services Organization (EACSO), which lasted until 1967 when the (first) EAC was agreed upon. However, the EAC was dissolved in 1977 due to a lack of political will from the

memberstates (Hazlewood, 1979: 55) and revived more than 20 years later with the signing of the Treaty of the Establishment of the East African Community in 1999 (see table 5.1). This treaty entered into force in 2000, after being ratified by Kenya, Tanzania and Uganda; the original partner states of the first EAC (EAC, n.d.). Since the signing of the treaty, the most important achievements of regional integration have been the establishment of a customs union in March 2004 (which commenced in January 2005) and the establishment of a common market in 2010.

One could state that the regional integration in Eastern Africa has been diminishing until the revival of the EAC in 1999 and has been slowly increasing again since then. The customs union of 1919 and the East African High Commission of 1948 were highly integrated organisations, having established a free trade area, a customs union with a common external tariff and a common currency (Hazlewood, 1979: 42). However, these forms of regional integration were led by the colonial powers of that time and thus could not continue in the same form after the independence of the countries.

Table 5.1: History of the EAC1

Years Organisation

1917 -‐ 1948 Customs Union between Kenya and Uganda (joined by Tanganyika in 1927)

1948 – 1961 East African High Commission (EAHC)

1961 – 1967 East African Common Services Organization (EACSO)

1967 – 1977 EAC (I)

1999 – now EAC (II) (Joined by Rwanda and Burundi in 2009, and by South Sudan in 2016)

The current EAC has six member states. The original member states Kenya, Uganda and Tanzania were joined by Rwanda and Burundi in 2009, and by South Sudan in 2016. The EAC has formulated four integration pillars that it aims to reach: a customs union, a common market, a monetary union and a political federation. At this moment in time the EAC has a customs union and a common market. In

(31)

2013, the Protocol for the Establishment of the EAC Monetary Union has been signed, which aims to create a monetary union by 2024. The political federation can be seen as the ultimate goal of the EAC, and the final stage of regional integration. However, it remains to be seen if, and when, the necessary steps to achieve this will be taken. A relatively new initiative has somewhat changed the plans of the EAC: the negotiations about a Tripartite Free Trade Area Agreement (TFTA) with the SADC and the Common Market for Eastern and Southern Africa (COMESA). The negotiations are initiated to look for possibilities to establish one free trade area amongst these three regional organisations (the SADC, the COMESA and the EAC).

In this chapter we will test our hypotheses on the case of the EAC. In the next paragraph (5.2) we will analyse the EAC by looking at its economic structure. We shall focus on its relation with extra-regional trade partners and the existence of one or more states that enjoys extra-regional privileges, and the strength of these possible privileges. In paragraph 5.3 the extra-regional privileges stemming from LDC status will be discussed. These variables are expected to be of influence on the forming of an EPA group, according to hypothesis 1a. Moreover, we will also look at the structure of membership overlap of the EAC states, to analyse the effect of this overlap on the process of forming an EPA group (hypothesis 1b) in paragraph 5.4. The second hypothesis will be central to paragraph 5.5 as we will discuss two possible scenarios for further integration of the EAC. In paragraph 5.6 we will analyse the possible involvement of other factors in the EAC-EPA process. Finally, in the conclusion (paragraph 5.7) we will discuss our findings and connect them to our hypotheses and theoretical argument.

5.2 Economic structure of the East African Community

An important variable in our research is the existence of a regional power that enjoys extra-regional privileges, as we expect this to influence the forming of the EPA groups. To determine whether the EAC has such a regional power, we analyse the economic structure of the EAC. This will be done using trade data, thus looking at the trade relations between the member states and extra-regional partners. Another possible variable that tells us more about the economic structure of the region is the foreign direct investment (FDI) flows in region. However, data on FDI is very limitedly available which leads us to choose for trade data. To collect data on trade flows we use Trade Map, which is a data source developed by the International Trade Centre (ITC) and uses data from the UN Comtrade database. The most recent data available from Trade Map on all EAC countries is from 2014, so this is the year we focus on. However, as the EPA negotiations started around 2002, we will also collect data on the years 2001 (right before the negotiations started) and 2007 (when the negotiations officially should have finished). As South Sudan has only joined the EAC in 2016, and has barely any data available, we will exclude South Sudan from this overview. We believe this is justified as South Sudan currently plays a marginal role in intra-EAC and extra-regional trade, and was not an EAC member in the years the EPA’s were negotiated on. South Sudan is struggling with a civil war since

(32)

2013 and is dealing with big problems like famine and refugees (Quinn, 2017).

5.2.1 Intra- vs. extra-regional trade

Before moving on to the specific trade relations between the EAC members and their partners, we first need to determine to what extent the EAC member states are dependent on intra- or on extra-regional trade in general. In table 5.2 we have laid out the total numbers of imports and exports within and outside the EAC region, as a percentage of the GDP of each country, and as a percentage of total trade.

Table 5.2: EAC: Intra-‐ vs. Extra-‐regional trade23

GDP in

billion US dollars

Intraregional trade as % of GDP

Extra-‐regional trade as % of GDP

Total trade in million dollars

Intraregional trade as % of total trade

Extra-‐regional trade as % of total trade Burundi 3.09 4.9 21.4 814.08 18.6 81.4 Kenya 61.40 3.0 36.9 24,498.09 7.5 92.5 Rwanda 7.91 11.2 21.7 2,608.58 34.0 66.0 Tanzania 48.20 2.7 35.5 18,395.76 7.1 92.9 Uganda 27.76 4.8 25.2 8,335.49 15.9 84.1 Total EAC 148.36 3.7 33.12 54,656 10.1 89.9

Table 5.2 shows that for every EAC member state, the percentage of extra-regional trade is much higher than the percentage of intraregional trade. This indicates that the economies of the EAC member states (Rwanda to a lesser extent) are mainly focussed and dependent on trade with extra-regional actors. The EAC is thus more dependent on trade with extra-extra-regional partners, than it is on trade within the region. This is often the case in developing countries, and justifies our use of the ‘extra-regional logic of regional integration’. However, we should be careful with this observation, as obviously the number of extra-regional countries is much bigger than the number of intra-regional countries. We will specify our focus on extra-regional actors in the following paragraph, so we can make more specific claims about the trade relations of the EAC.

5.2.2 The EAC and its extra-regional trade partners

Now we have briefly looked at the intra- and extra-regional trade balance of the EAC, we will dive deeper into the economic structure of the EAC. We aim to determine whether the EAC has a member state that can be classified as a regional power, and if this comes along with extra-regional privileges. In the next paragraph we will focus on extra-regional privileges based on other grounds than regional power. In this paragraph we analyse the trade flows of the member states. This allows us to see which trade relations exist within the EAC and with extra-regional partners. We want to find out which

2_{GDP
in
2014,
from
Worldbank
(http://data.worldbank.org/indicator/NY.GDP.MKTP.CD)} 3_{Trade
numbers
as
the
sum
of
import
and
export,
from
Trade
Map
(http://www.trademap.org/)}

(33)

economic relations are most important. If we find economic imbalances, for example because one member state has closer ties with external partners than the others, the interests in regional integration will be unequal. This will likely problematize further integration.

Following Krapohl (2017), we narrow down the external trade partners to the three most important partners to make trade interests visible. In 2001 EACs most important extra-regional trade partners were the EU, the United Arab Emirates (UAE) and the United States of America (USA). In 2007, India took over the USA’s position as one of the EAC’s three most important extra-regional partners. And in 2014, the UAE was replaced by China. In 2014, the EAC’s most important extra-regional trade partners were thus China (15.7% of total trade), the EU (13.8%) and India (12.6%). As can be found in table 5.3 below, in 2014, for almost every EAC member, the EU, China and India are their most important trade partner, either for export or import.

Table 5.3: Most important trade partners in 2014 (percentage of total imports/exports)4

First Second Third

Burundi Imports EU (18.3%) India (11.2%) China (10.8%)

Exports DRC (20.1%) UAE (17.6%) EU (16.9%)

Kenya Imports India (16.4%) China (15.3%) EU (13.9%)

Exports EU (22.4%) Uganda (11.3%) Tanzania (8%)

Rwanda Imports China (18.7%) EU (17.9%) Uganda (13.7%)

Exports Tanzania (27.8%) DRC (23.5%) Uganda (11.9%)

Tanzania Imports India (19.3%) China (16.1%) EU (10.4%)

Exports India (22.0%) South Africa (12.1%) China (12.0%)

Uganda Imports India (24.5%) China (12.2%) EU (10.4%)

Exports EU (21.2%) Sudan (Sudan and South Sudan) (17.0%)

Kenya (13.1%)

EAC total Imports India (18.1%) China (15.2%) EU (12.5%)

Exports EU (17.2%) India (9.3%) China (5.6%)

To get a better idea of the economic structure of the EAC, and the balance between intraregional and extra-regional trade we thus include the three most important extra-regional trade partners of the EAC in the years 2001, 2007 and 2014, and look at their relation with EAC members. Naturally, we also look at the intraregional trade relations of the EAC.

To collect data on import and export values between states, we use the Trade Map database, focussing on trade in goods, thus excluding services. For four of the EAC members, the total import from the three most important extra-regional trade partners is bigger, than the total import from the other four member states of the EAC, leaving out only Rwanda (see table 5.4). For the exports, this is a little bit different, as only two member states (Kenya and Tanzania) export more to China, the EU and India combined than within the EAC (see table 5). The difference is especially big for Rwanda, who trades more within the region. However, if we sum up the imports and exports of the EAC