Development Policy Review, 2017, 35 (4): 531-548
Make Europe happen on the ground? Enabling and constraining factors for European Union aid coordination in Africa
Maurizio Carbone*
This article investigates the extent to which the collective commitments on aid coordination made in the EU context trigger any changes in the aid practices of EU member states in Africa. By exploring the trajectory of joint programming, it demonstrates that the development policies of EU members states are affected by EU membership, yet the impact of Europe is, surprisingly, less pronounced on the ground than at the headquarter level. This decoupling of norms from practices can be attributed not only (and necessarily) to the attempt of EU donors to pursue their national goals, but also to the resistance of national aid bureaucracies on the ground, the increased scepticism of developing countries, and the growing complexity of the development architecture.
Key words: aid effectiveness, donor coordination, Europeanization, EU development policy, joint programming
1 Introduction
It is widely agreed that in the first decade of the new century aid effectiveness was a central concern for donors. It is also generally accepted that the European Union (EU) has significantly affected the outcomes of some prominent summits on international development, most notably the 2005 Paris Declaration on Aid Effectiveness, the 2008 Accra Agenda for Action and the 2011 Busan Partnership for Effective Development Cooperation (Carbone, 2007; Holland and Doidge, 2012; Carbone, 2013; Ganzle et al., 2013). Surprisingly, its impact on the development policies of the EU member states has been largely overlooked. An emerging strand of the literature has, tentatively, concluded that any exogenous pressure has resulted, at best, in shallow Europeanization (Lightfoot and Szent-Ivanyi, 2014; Orbie and Carbone, 2016; Orbie and Lightfoot, 2017), yet it cannot be dismissed that the increased level and nature of interactions between member states through the EU does elicit some (forms of) adaptation to Europe. One of
:Professor of International Relations and Development; Jean Monnet Professor of EU External Policies, University of Glasgow. The author wishes to thank Mark Furness, Martin Holland, Stephan Klingebiel, Jan Orbie, Balazs Szent-Ivanyi, and the anonymous reviewers for their helpful comments, and the 82 policy-makers in Europe and Africa who have accepted to be interviewed and have shared unpublished documents.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
532    Maurizio Carbone
the areas in which such interactions have been most evident concerns joint programming (JP) - that is, the process of formulating a collective European strategy for programming of aid in specific developing countries (EC, 2015: 202). JP was officially proposed in the context of the Busan summit and then launched in a handful of developing countries. Importantly, it has been explicitly mentioned, and in some cases integrated, in(to) the bilateral policies of several EU member states. In the space of a few years, according to official discourse, JP has emerged 'as a norm' in EU external relations, so as to promote donor coordination as well as to 'make Europe happen on the ground, translating shared European values and policies on issues such as fundamental rights and good governance into coherent, targeted action in partner countries' (EC and EE AS, 2015a: 4).
The situation on the ground, however, does not match these expectations. By mid-2016, JP had been considered in 55 developing countries, with outcomes varying from mere exchanges of information to adoptions of joint strategies through mapping of donor activities and shared analyses - not to mention the fact that in a large number of countries a JP process has not been set in motion (EC and EEAS, 2015b; Helly et al., 2015). This variation begs the following research question: why do the collective commitments made at the EU level translate so differently on the ground? To answer this question, this article is divided into two parts. The first part proposes a conceptual framework. Drawing on the Europeanization and foreign aid literatures, it identifies a series of factors that enable or constrain aid coordination on the ground. The second part focuses on the trajectory of JP, from its inception in Europe to its application in sub-Saharan Africa. This region is where aid fragmentation is highest, owing to the simultaneous presence of a large number of EU member states, other traditional donors and various new development actors. Furthermore, the enhanced agency of African countries makes this region even more interesting (Aldasoro et al., 2010; Sumner and Mallett, 2012).
Considering that JP processes have attracted little attention, if any at all, this article relies on a meticulous analysis of primary sources, including unpublished documents, and three rounds of elite interviews with a total of 82 policy-makers. The first round took place in May-June 2013 in Europe, with officials working in the European Commission (EC), the European External Action Service (EEAS), several EU member states and some non-governmental organizations (NGOs). The second round was conducted in June-July 2015 in five countries in sub-Saharan Africa (Ethiopia, Mozambique, Namibia, Zambia, Zimbabwe) and included representatives from the EU Delegations, the embassies and aid agencies of various EU member states and some non-EU countries (Norway, USA, Japan), the foreign affairs and economy/finance ministries of the five African countries, various international organizations (African Development Bank, International Monetary Fund, United Nations, World Bank) and some civil society organizations. A third round was completed in Brussels in January-February 2016 with officials working in EU institutions and member states and concentrated on five additional countries (Burkina Faso, Ghana, Mali, Rwanda and Senegal) with the view to gaining a more comprehensive understanding of
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 533
the sub-Saharan Africa region.2 Overall, the ten countries constitute a balanced mix in terms of geographical location, aid dependence, presence of EU donors and JP outcomes.
2   Conceptual framework: explaining aid coordination in the EU
In order to unravel the dynamics surrounding EU aid coordination, this article draws on - and contributes to - two different strands of the academic literature. A first group of scholars, in the area of development studies, has focused on the contentious issue of aid effectiveness. More specifically, they have identified costs and benefits of donor coordination, and have highlighted the influencing role of some international organizations in shaping country decisions, particularly the World Bank and the Development Assistance Committee, rather than others, for instance the European Union. A second group of scholars, in the area of EU studies, has focused on the concept of Europeanization - which for the purpose of this article is defined as 'a process of incorporation of EU norms, practices and procedures into the domestic level' (Moumoutzis, 2011: 608) - and its differential impact across EU member states, seeking to isolate its effects from other external pressures. In this case, the role of the EU is, often, over-emphasized. This section delimits the perimeter of the research and proposes a conceptual framework so as to show how the impact of the EU is mediated by a series of intervening factors, thus generating different outcomes across different member states.
2.1 Aid effectiveness and Europeanization
One of the most researched areas in international development is aid effectiveness. For a long time, the blame for the allegedly limited impact of aid was largely attributed to the pernicious behaviour of recipients, but since the turn of the new century donors have accepted some responsibility for it. Thus, in the context of various forums on aid effectiveness (that is, Paris, Accra, Busan), they have committed to coordinating their efforts. Nevertheless, empirical evidence shows that aid proliferation and aid fragmentation have increased since the adoption of the 2005 Paris Declaration (Aldasoro et al., 2010; Nunnenkamp et al., 2013). This decoupling of norms from practice, which is a well-known phenomenon in studies of norm compliance (Brown and Swiss, 2013), can be attributed not solely to the donor use of aid to pursue their political and commercial interests, but also to the increased resistance of recipients themselves. While aid coordination generates evident benefits in terms of reduced transaction costs and, most likely, maximized development impact, it also creates costs for both donors, in terms of loss of sovereignty in aid allocations, and recipients, in terms of reduced leverage vis-a-vis potentially more intrusive donors (Bigsten and
2. In order to preserve the anonymity of interviewees, the following system is used: MS, for officials from EU Member States (both headquarters and ground); EU, for officials from European institutions (both headquarters and ground); IO, for officials from international (non-European) organizations and third countries; AC, for officials from African countries.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
534    Maurizio Carbone
Tengstam, 2015; Bourguignon and Platteau, 2015; Steinwand, 2015). Moreover, it can be expected that actors/organizations participate in the generation of global norms with the aim of enhancing or maintaining legitimacy and then may not (be able to) fulfil the commitment of which they are rhetorically supportive: 'Especially in the case of norms with weak enforcement mechanisms or with minimal consequences for non-compliance, the likelihood of decoupling increases' (Brown and Swiss, 2013: 752).
To be sure, there are different degrees of donor coordination, ranging from simple exchange of information to fully integrated approaches through harmonized strategies and arrangements (Klingebiel et al., 2016). Setting an optimal level of coordination, as convincingly argued by the political economy of aid literature, depends on the value that both donors and recipients attach to national autonomy and aid effectiveness and the congruence of their goals (Bourguignon and Platteau, 2015; Bigsten and Tengstam, 2015; Steinwand, 2015). The consequence is that, on the one hand, owing to the free-rider problem alimented by the presence of multiple development actors, coordination 'will not automatically guarantee the sub-optimality of aid provisions unless there is a fundamental change in the incentives for aid provisions' (Rahman and Sawada, 2012: 611). On the other hand, if donors are genuinely interested in coordination they can either deliver more of their aid through multilateral channels or accept tighter coordination by a supranational organization. While the first option would represent a radical change in existing practices, 'what can be achieved in the short-term is probably a strengthening of the processes of joint programming and policy coordination' (Bigsten and Tengstam, 2015: 83). The latter is the path chosen within the EU (which has also attempted to change the incentive structure for member states), as we will see in the empirical section, but to fully ascertain the impact of Europe on national development policies it is necessary to refer to the Europeanization literature.
The primary concern for Europeanization scholars is the degree of change generated by the EU on member states. From a top-down perspective, the shared consensus is that the impact of the EU would be high in areas of direct EU competence and low in areas of no direct EU competence. The predominance of soft law measures and the lack of a clear chain of command, in fact, create a permissive context for member states, which incur no evident costs in case of failure to adapt. Member states, however, are not passive takers of EU demands, and often they proactively seek to shape EU policies. A bottom-up perspective, therefore, would see member states attempting to upload their preferences, generally because they want to reduce the costs of adaptation, but also because they think they may achieve more collectively than unilaterally or, in some cases, they believe in the virtue of their choice. In all cases, the more successful they are in uploading their preferences to the EU level, the fewer problems they would face in adapting. These two phases, thus, should not be treated as two separate processes, but as a sequel. Importantly, Europeanization can also involve horizontal policy exchanges between countries without necessarily passing through the EU: countries can learn directly from other countries (Wong and Hill, 2011; Alecu de Flers and Müller, 2012; Ladrech, 2012).
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 535
The conventional departure of existing Europeanization studies is the existence of a misfit between European and national policies. Such a gap generates pressure for adaptation to Europe, to which member states respond differently (Borzel and Risse, 2007; Graziano and Vink, 2007). Europeanization, however, is not only a process of adaptation, but also of social learning, which would require adjusting the preferences of the actors and to a degree even their beliefs, which may start from EU-level interactions or rhetoric (Exadaktylos, 2012). Generally, Europeanization scholars categorise the impact of the EU not in binary terms, but distinguish between degrees along a continuum, from inertia (no change to existing practices) to transformation (fundamental change to existing practices), through absorption (low incorporation of EU demands into existing policies) and accommodation (substantial adaptation of existing policies without changing core features) (Borzel and Risse, 2007; Ladrech, 2012). Other studies have distinguished between different forms of change: formal, behavioural or discursive adoption of rules, depending on whether rules lead to actual change or whether there is only a mere lip service paid to them (Sedelmeier, 2012). This article, however, is not so much interested in any 'arithmetic of convergence', but more in the drivers and limits of Europeanization of development policy (Hill and Wong, 2011). More relevant for the purpose of the article, in fact, is one of the conundrums faced by the Europeanization literature, that is the extent to which change, when it occurs, can be attributed to Europe, rather than to domestic or global factors, with the EU acting as an intervening variable at best (Graziano and Vink, 2007; Ladrech, 2012). More specifically, the EU is not the only actor that has attempted to tackle the issue of aid effectiveness -other players, such as the World Bank and the DAC, or clusters of states, notably the like-minded group, have actively pursued this area (Sumner and Mallett, 2012; Smith, 2016) - so the key concern is whether certain norms are incorporated into the policies of EU member states because they are universally accepted or because the course of action they prescribe is considered appropriate within the EU context (Moumoutzis, 2011).
2.2 From pressure to change: drivers and brakes
The pressure that the EU generates on national development policies does not result in automatic adaptation, but translates differently in different countries. Whether change occurs certainly depends on the strength of the pressure and the degree of the misfit, but, more importantly, depends on the presence (or absence) of different mediating factors. Existing studies derive these factors from the debate between rationalists and constructivists in international relations and comparative politics, whereby two different logics can (simultaneously) be at play: the logic of consequentialism, which emphasizes the importance of incentives and efficiency calculations, and the logic of appropriateness, which highlights the importance of social learning and socialization. The focus of these studies is mostly on domestic factors (Vink and Graziano, 2007; Ladrech, 2012), yet a number of external factors can also have a significant role - and here the literatures on EU foreign policy and
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
536    Maurizio Carbone
on foreign aid are of great help. All these intervening factors are not mutually exclusive, and do not necessarily pull in the same direction.
A first group of mediating factors concerns actors and factors that, allegedly, are expected to enable Europeanization - which in our case means more aid coordination orchestrated by the EU (not necessarily by the European Commission). The process of Europeanization is facilitated, first and foremost, by formal institutions, which provide national actors with material and ideational resources necessary to promote domestic adaptation, and informal (behavioural) norms, which have emerged over the years thanks to the increased interaction between policymakers (Borzel and Risse, 2007). Importantly, the EU literature on foreign policy refers to the presence of a sort of 'coordination reflex' - whereby EU member states seek to coordinate with their EU peers before any ensuing action is implemented (Wong and Hill, 2011; Pomorska and Wright, 2012). The role of formal institutions could be supplemented by norm entrepreneurs, who seek to mobilize support and refine the interests and identities of others in light of new norms and rules (Borzel and Risse, 2007). Taking a rational approach, some member states could seek to benefit from the multiplier effect of political cooperation or could try to hide behind the EU to promote their national goals (Moumoutzis, 2011; Alecu de Flers and Miiller, 2012). From a constructivist perspective, the EU's impact would not depend only on material reasons. In particular, the perceived legitimacy of EU rules can even counterbalance material costs. Such legitimacy depends on the properties of the rules that are promoted or the process through which such rules are created (Sedelmeier, 2012. In EU foreign policy, the expectation is that the 'big three' (France, Germany, UK) would influence policy outputs in most areas, whereas smaller countries would seek to upload their preferences in selected areas (Alecu de Flers and Miiller, 2012; Pomorska and Wright, 2012). Yet, some countries, to different degrees, seek to strengthen the role of the EU in the international arena as a way to project a specific set of values globally (Wong and Hill, 2011). Drawing on the aid literature, it could be confidently stated that larger and smaller donors incur different costs when they seek to participate in (or stay away from) the coordination game: 'Big countries tend to assign much greater weight to considerations of political sovereignty and control than smaller ones, if only because they are more able to design strategies that serve their own national interests' (Bourguignon and Platteau, 2015: 90).
A second group of factors is expected to hinder Europeanization - and again in our case aid coordination. Certainly, different types of development legacies and cultures co-exist in the EU, which may not necessarily positively resonate with EU-sponsored norms (Smith, 2016). Resistance to Europeanization may also result from an ideological hostility to European integration. The EU foreign policy literature has underlined the fact that some member states may decide to ring-fence issues or relations with special countries to prevent EU interference in areas that are of primary importance for their national interest. In such cases, the member states would either try to 'keep the EU out' or would try to upload their preferences to the EU level (Wong and Hill, 2011; Pomorska and Wright, 2012). Similarly, the foreign aid literature points out that donors find it difficult to coordinate and justify a reduced presence in countries which are strategic from a political point of view,
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 537
constitute important export markets, or are resource-abundant (Bigsten and Tengstam, 2015; Fuchs et al., 2015). Then there are also the potential losers of Europeanization, which would oppose any adjustment to EU demands. For instance, bureaucracies can act as de facto veto players, rejecting or obstructing change, because they would be afraid of losing power, or because they lack incentives to promote coordination (Easterly, 2003; Carbone, 2013). Other potential veto points are related to the pressure exercised by vested interests, particularly large or dominant economic actors or by societal mobilization (Bretherton and Mannin, 2013; Sedelmeier, 2012.
Finally, a third group of factors, linked to the external context, can be expected to both enable and hinder Europeanization - and again in our case aid coordination - depending on the circumstance. Firstly, traditional big powers, notably the USA and Russia, have at times played on internal EU divisions and at other times acted as 'external federators' (Wong and Hill, 2011). Emerging development actors, for instance China, or developing countries themselves, can behave similarly. In the case of foreign aid, the USA, traditionally recalcitrant regarding donor coordination, has been joined by a group of new development actors that has questioned an aid effectiveness regime which they have not contributed to forging and have offered developing countries an incentive to reject yet another agenda imposed by traditional donors. At the same time, it may be possible that other poles of attraction could be available and be more appealing to specific member states (e.g. like-minded countries) than the European Union. As for developing countries, they would be expected to support aid coordination, essentially because aid fragmentation leads to an ineffective use of scarce resources. Yet, they could extract a larger portion of aid by playing off different donors and/or could have less bargaining power when confronted with a group of well-coordinated donors: greater donor coordination could therefore result in a further circumscribing of recipient ownership over development processes (Carbone, 2010; Sumner and Mallett, 2012; Bigsten and Tengstam, 2015; Bourguignon and Platteau, 2015; Janus et al., 2015).
3   Empirical analysis: the trajectory of EU joint programming
Development policy in the European Union has evolved around two tracks. On the one hand, member states have, gradually, transferred larger resources to the supranational programme - though this trend has stabilized since the late 1990s. On the other hand, they have asserted their right to conduct their bilateral policies autonomously, rejecting any attempted inference by the European Commission (Carbone, 2007; Holland and Doidge, 2012). This special feature of EU development policy has often been cited by scholars (and invoked by member states) to justify the dearth of impact of the EU on national choices. The concept of Europeanization was for a long time used solely, and inappropriately, to explore the extent to which different member states managed to shape the EU supranational aid programme. It was only in the early 2000s, when the European Commission launched a number of initiatives  aimed  at projecting  a  common  European vision  of international
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
538    Maurizio Carbone
development - and, some have even argued, manifested an overt ambition to 'federate' the development policies of the member states (Delputte and Orbie, 2014; Orbie and Lightfoot, 2017) - that the concept of Europeanization could be fully applied to development policy.3 The generation of a new body of legislation was accompanied by the creation of a number of formal and informal venues in which European development actors could engage, at ministerial and bureaucratic levels. These changes have attracted some attention among scholars, who share the view that Europeanization would at best be shallow and superficial and could occur only through socialization and social learning (Bretherton, 2013; Lightfoot and Szent-Ivanyi, 2014; Orbie and Carbone, 2016; Orbie and Lightfoot, 2017). These studies, however, have mostly concentrated on strategic choices, made by headquarters. This article goes further, looking not only at the construction of the JP norm but also at its implementation in a series of developing countries: to ascertain the transformative power of Europe, particularly in development policy, what happens on the ground is of paramount importance.
3.1 The construction of the JP norm
The European Union has tackled the issue of donor coordination through a number of initiatives. In particular, the 2007 code of conduct on complementarity and division of labour called on European donors to reduce the number of sectors and countries in which they actively operate. Moreover, some studies, funded by the European Commission, have documented the significant savings (in the order of several billion euros) accrued from additional donor coordination (Carbone, 2010; Carbone, 2013; Klingebiel et al., 2016). The limited effect of these initiatives led the European Commission to rethink its approach, concluding that it was 'necessary to go beyond previous technocratic approaches, and add a political dimension to donor coordination' (Interview with EU official, June 2013). The opportunity came with the preparation of the EU's common position for the fourth high-level forum on aid effectiveness in Busan. In that context, the EU officially agreed on a new initiative on joint programming, 'in order to show leadership in Busan and beyond and as a response to the increased [aid] fragmentation and proliferation' (Council of the European Union, 2011: 6). Essentially, JP consists of a joint analysis and a joint response, to be completed on the ground, by the EU Delegations together with the member states, taking into account the national development plans of recipient countries as well as the activities of other donors. The joint analysis is meant to provide an overview of the recipient country situation (political, social, economic and environmental) as well as a mapping of donor presence and existing general coordination mechanisms. The joint response is meant to outline the overall rationale and the funding allocations of all EU (supranational and national) resources, including a tentative division of labour between all actors (EC and EEAS, 2015a).
3. These commitments involve three broad areas: quantity of aid, specifically on the 0.7% target, sub-Saharan Africa, and low-income countries; quality of aid, specifically on aid coordination, aid untying and budget support; policy coherence for development.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 539
The construction of the JP norm was not easy (Interviews with EU and MS officials, May-June 2013). On one side, France and Germany strongly promoted the idea of JP, but for different reasons. For France, the motivation was more political, broadly in line with the view that the EU should strengthen its role in the global arena and affirm its identity vis-a-vis third countries and actors (see also Bretherton, 2013). For Germany, the reason was related to the developmental nature of the initiative: JP was primarily seen as a technical exercise to improve aid effectiveness. These two rationales also underpinned the behaviour of, respectively, Belgium and Spain on the one hand, and the Netherlands on the other hand. On the other side, the UK reluctantly accepted the commitment to JP in return for support of an initiative to increase aid transparency and with the proviso that any JP process would be 'conducted at partner country level in order to respond to the specific needs and the situation on the ground' (Council of the European Union, 2011: 13). Interestingly, the decision on JP was endorsed by Nordic countries, which have traditionally manifested reservations on EU collective action on international development issues and resisted any European interference in their bilateral policies. As admitted by a representative of one of these countries: 'We do support the idea of JP, but only because it coincides with our agenda on aid effectiveness. For us, it is more the leverage aspect: achieving better results by playing the collective card. We are not like the French or the Spaniards: we couldn't care less about the EU's impact or visibility' (Interview with MS official, June 2013).
The European Commission not only welcomed the JP commitment by all member states so as to allow the EU to play a leading role at the Busan summit, in line with previous development summits (Interviews with EU officials, May-June 2013; see also Carbone, 2013), but also 'went on a roadshow through the EU capitals' (CONCORD, 2012: 38) and convened technical seminars in Brussels (Galeazzi et al., 2013) to continue reassuring reluctant member states that JP would not impinge on their national development policies (Interviews with EU officials, May-June 2013). JP was then given preliminary testing in a handful of countries, before being launched in about 50 countries in which the prospects of a successful outcome were deemed high. Importantly, a range of EU member states (e.g. Belgium, Denmark, Finland, Germany, Italy, Luxembourg, Netherlands, Spain) have issued detailed guidelines to inform practices on the ground, whereas others have committed to participating in all JP processes (e.g. France, Sweden, Austria), or on a case-by-case basis (e.g. Estonia, Poland, Slovakia, United Kingdom) (EC, 2014; 2015). In spite of these commitments, progress on the ground has been (far) below expectations: by mid-2016, about 20 countries had actually completed a JP strategy, while the rest of the countries 'seem stuck in endless discussions', with some of the initial champions (e.g. Rwanda, Ghana) manifesting signs of disillusionment (Interviews with EU officials, January-February 2016). The reasons for such a variation are discussed in the remainder of this section, starting from EU-related factors and then focusing on context-related factors.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
540    Maurizio Carbone
3.2 JP on the ground: EU-related factors
The responsibility to launch JP processes falls under the remit of EU Delegations, whose role in terms of coordination has been substantially strengthened by the Treaty of Lisbon (Delputte and Orbie, 2014). In general, an active EU Delegation has facilitated the initiation, and in most cases the conclusion, of a JP document (Interviews with EU and MS officials, June-July 2015, January-February 2016; see also Galeazzi et al., 2013; Helly et al., 2015). However, in some cases (e.g. Mozambique, Tanzania, Zambia), EU Delegations have lacked leadership or have suffered from the unresolved division of labour between the EC (DG DEVCO) and the EEAS: heads of mission seem more prone to augmenting the EU's visibility even at the expense of aid effectiveness, whereas heads of cooperation are generally keener on enhancing the impact of foreign aid (Interviews with EU and MS officials, June-July 2015, January-February 2016; see also CONCORD, 2012; Bretherton, 2013). Importantly, JP processes are eased when one or a small group of member states - as has happened in Ethiopia and Namibia - act as facilitators, drafting documents and seeking to persuade diffident member states. In some cases, specific member states can be more entrepreneurial than the EU Delegation itself - for instance in Burkina Faso or Ghana (Interviews with EU and MS officials, June-July 2015, January-February 2016; see also Furness and Vollmer, 2013; Galeazzi et al., 2013; Helly et al., 2015). More generally, the responses of EU member states can vary significantly.
A first group, including both large and small donors, has complied overall with the JP norm. France and Germany, which had shaped the construction of the JP norm, has not found it too difficult to adapt; yet, French officials have been very active in West Africa and more passive in other areas of Africa, whereas German officials have been hesitant in cases in which there were no major gains in terms of transaction costs from the adoption of a JP document (e.g. Zambia) (Interviews with MS officials, June-July 2015, January-February 2016). As for other member states, the general expectation from headquarters that 'small EU member states would see more added value in JP' (Interview with EU official, June 2013; see also Bourguignon and Platteau, 2015) was broadly confirmed across sub-Saharan Africa: some with a view to promoting the aid effectiveness agenda (e.g., Netherlands) and others with the view to projecting the EU's values and identity vis-a-vis developing countries and other donors (e.g., Belgium, Italy, Spain). Nevertheless, on a few occasions an eagerness to preserve special relationships (e.g. Belgium in Burundi; Italy in Ethiopia; Portugal in Mozambique) has prevailed, even at the risk of undermining JP processes (Interviews with EU and MS officials, June-July 2015, January-February 2016).
A second group has been hesitant towards, and in some cases openly resisted, JP processes, yet again for different reasons. For Nordic donors, despite their general commitment to a European approach at the rhetorical level, there is still the obstacle of established cultural and normative structures on the ground, where they have preferred a multilateral approach to aid coordination and thus have resisted the idea of a collective EU identity in international development (Interviews with
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 541
EU and MS officials, June-July 2015, January-February 2016; see also Delputte and Orbie, 2014; Smith, 2016). An official in one of these countries has admitted that his colleagues in headquarters 'realized that, in order to push aid effectiveness principles forward, it was necessary to work with and through the EU. But the implementation level, in specific developing countries, does not necessarily respond to the same logic' (Interview with MS official, June 2015). If for Nordic countries resistance to JP processes is in the name of aid effectiveness, in the case of the UK it has been caused, to a large degree, by an ideological opposition to further European integration, to the pooling of sovereignty in yet another policy area. Moreover, the UK perceives itself as a big donor, performing better than most European donors, so it has not seen much added value from cooperating with other EU member states (Interviews with EU and MS officials, June-July 2015, January-February 2016). Importantly, numerous policy-makers from various African countries have noted that not only is 'the UK a donor of its own, doing things differently from any other EU member states', but also that 'it is no longer a shining example in the promotion of aid effectiveness and ownership, and now everything must be on their terms' (Interviews with AC officials, June-July 2015). It is therefore not surprising that numerous European policy-makers have openly stated that 'progress on JP strategies is easier in those countries in which the UK is not particularly active as a donor' (Interviews with EU and MS officials, June-July 2015, January-February 2016).
These discrepancies in JP outcomes serve to qualify the role of socialization in causing change in national policies. The initiatives on aid effectiveness have created new opportunities for member states as a group to interact on the ground. The EU Delegations generally convene meetings at least once a month; moreover, several workshops with the aim of facilitating the sharing of good practices were organized in different regions in 2014-15 (Interviews with EU and MS officials, June-July 2015).4 Nevertheless, it appears that EU socialization effects on the ground have been less pronounced than expected. Instead, national aid bureaucrats have resisted JP processes, fearing a potential loss of power (and of course jobs) and complaining about the lack of incentives coming from their capitals to engage in joint approaches (Interviews with MS officials, June-July 2015; see also Carbone, 2013; EC, 2014). A large number of them has claimed that JP, like any other coordination initiative, 'is burdensome and not rewarded by headquarters, which continuously and insistently ask for more tangible results' (Interviews with MS officials, June-July 2015). In Tanzania, for instance, member states seem to have suffered from a sort of 'consultation overdose' and have therefore been reluctant to engage in JP processes (Helly et al., 2015). It should not also be forgotten that other groupings (e.g. like-minded donors, Nordic group) can perform important socialization functions in the area of development (Smith, 2016). Another interesting finding is that, contrary to expectations (Alecu de Flers and M tiller, 2012), diplomatic missions with larger degrees of autonomy from capitals are less prone to Europeanization pressures (Interviews with MS officials, June-July 2015; see also Delputte and Orbie, 2014).
4. These workshops were jointly organized by DG DEVCO and the EEAS, with the support of some EU Member States: France in West Africa; Belgium and the Netherlands in East and Southern Africa; Spain in Latin America; and Germany in Asia.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
542    Maurizio Carbone
Paradoxically, some policy-makers on the ground have even lamented a 'contradiction between the high level political declarations on joint programming and the flexible and voluntary character of the process. Although there are clearly advantages to this informal character of the process, it also leads to ambiguity' (Interview with EU official, January 2016).
3.3 JP on the ground: context-related factors
While so far analysis has concentrated on EU-related factors, the context in which JP processes take place is equally important. The expectation, particularly in EU headquarters, was that aid recipients would be supportive of any endeavour aimed at reducing aid fragmentation. Empirical evidence projects a slightly different picture. Some African policy-makers have questioned the added value, and the legitimacy, of yet another initiative on aid effectiveness launched by the EU - in fact, the EU is perceived as not being able 'to live up to the many commitments it has made over the years' (Interview with AC official, June 2015). Other African policy-makers have associated JP with other donor coordination initiatives, which in their view would likely lead to aid retrenchments and more intrusive conditionality (Interviews with AC officials, June-July 2015; see also Carbone, 2013; Bigsten and Tengstam, 2015). The response of the government in Ethiopia well exemplifies these sentiments: the initial active resistance, owing to the fear of confronting a larger front of donors, has eventually turned into passive acceptance: 'we accept the process for as long as it does not add to our administrative burdens, but we do not participate in it' (Interview with EU official, January 2016).5 In general, the support of aid recipients may not be so decisive to set JP processes in motion (Helly et al., 2015), yet it may still play a crucial role: in Zimbabwe, for instance, the political environment was deemed not conducive - and a similar situation emerges from Mali, where the process was delayed as a result of civil unrest (Interviews with EU officials, June-July 2015, January-February 2016). The support of aid recipients, certainly, is fundamental in more advanced stages of JP processes, for instance in the case of joint responses or joint implementation (e.g. Senegal). In this respect, 'the example of Rwanda demonstrates that governments already taking the lead in donor coordination are more eager to reap the potential benefits of this initiative' (Galeazzi et al., 2013, p. 7).
The attitude of recipient countries towards JP processes has also been partly conditioned by the presence of donor-wide coordination mechanisms. When such mechanisms are perceived as functioning, as argued by various African policymakers, 'EU joint programming represents a step back from where we are at the moment, with all donors sitting at the same table' (Interviews with AC officials, June-July 2015). When such mechanisms are absent, or are perceived as not
5. Resistance to JP processes may also be the indirect consequence of a potential redistribution of power within governments in recipient countries: the initiative on JP, and more generally EU donor coordination, assigns a more prominent role to central ministries (mainly finance) at the expense of sector ministries (such as education, health, agriculture), which in fact 'seem to prefer a decentralized approach' (European Commission, 2015: 139; Interviews with EU and MS officials, June-July 2015).
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 543
effective, EU pressures for coordination are likely to affect national decisions: in the view of various European officials, 'if the EU acts collectively, we can have a louder message and a stronger impact' (Interview with EU official, June 2015). Interestingly, the initial intention was that of making JP as inclusive as possible, taking into account what other donors were doing so as to avoid waste of resources. But it gradually became evident that 'JP could be sustained only if certain political foundations exist. If we open it to other non-European donors, we would lose an important driver: this process can fly only if it maintains this European spirit' (Interview with EU official, June 2015). It is not surprising that the signing of JP documents has often been accompanied by a public launch involving senior representatives of African governments (as has been the case in Ghana or Ethiopia). More surprising is the view that the significance of JP in Rwanda, which has involved both EU and non-EU donors, has been 'somehow diluted, so much so that it is not actually possible to recognize the EU driver in it' (Interview with EU official, June 2013).
Another important element to consider is the development status of the recipient country. Specifically, aid dependence does not seem to have a direct effect on JP outcomes (Helly et al., 2015). Moreover, it is not necessarily true that 'the incentive to coordinate aid efforts increases ... when the size of the coordinating club of donor countries is smaller' (Bourguignon and Platteau, 2015: 92). In Namibia and Zambia, for instance, the number of active European donors is small but the JP outcome has been different. In Tanzania and Ethiopia the number of donors is high, but again the outcome has been different. What is noteworthy, however, is the fact that JP processes have consistently been more difficult in middle-income countries and in countries experiencing an economic boom or endowed with natural resources - for instance Ghana, Mozambique, Zambia (Interviews with EU and MS officials, June-July 2015, January-February 2016). In these contexts, 'foreign aid is perceived as a gateway to the government and the state; JP processes, and more generally EU donor coordination, have met with remarkable resistance or have even been boycotted by some EU member states' (Interviews with EU officials, June-July 2015). Interestingly, some European officials have appeared 'annoyed by a new initiative on aid effectiveness', especially at a time in which other non-EU actors have been more actively pursuing their commercial interests, and wished 'the European Union and the Commission to move beyond the aid effectiveness decade' (Interviews with MS officials, June-July 2015).
This last point leads to a final consideration on how the changing development architecture has altered dynamics between European actors. Existing literature concurs that the USA has shied away from aid coordination (Bourguignon and Platteau, 2015; Steinwand, 2015), yet over the past decade it has emerged as a new pole of attraction for some EU member states, at least in specific sectors (Interviews with MS and IO officials, June-July 2015). The fact that 'the UK shares similar principles and practices with the USA, more than it actually does with countries in southern Europe' (Interview with MS official, June 2015) has been confirmed by several US officials, so much so that one of them has admitted that 'it is a joy to work closely with the UK', whereas another one has revealed how the US 'pragmatic approach to development is in line with that of some Nordic countries'
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
544    Maurizio Carbone
(Interviews with IO officials, June 2015). As for China's rise as a development actor, officials in European headquarters appeared convinced that once their colleagues in Africa 'felt the rising competition of new donors, they would naturally seek to act collectively through the EU' (Interview with EU official, June 2013). The situation on the ground, according to various non-EU officials, has turned out to be different: 'European donors are wary of the increased influence of China in the region, but there is no common front, no collective approach' (Interviews with IO officials, June-July 2015). Paradoxically, the increased prominence of China has provided an ulterior motive not to engage in JP processes and other forms of donor coordination, as indirectly admitted by a European policy-maker: 'China looks like a lost battle; other European countries have become our competitors' (Interview with MS official, June 2015).
4 Conclusion
The role of the European Union in shaping the development policies of the EU member states has largely been overlooked by scholars in both EU and development studies. On the one hand, the prevailing literature on international development has concluded that foreign aid decisions are largely determined by national processes -and only minimally, if at all, by exogenous pressures. On the other hand, the consensus in EU studies is that, in areas of no direct EU competence, Europeanization tends to be light, working mostly through socialization and social learning. This article has added to these two literatures by showing that EU membership does affect the development policies of the member states, in some cases more than in others, and not necessarily only through social learning and socialization, but also as a result of strategic calculations. Moreover, it has pointed to a significant decoupling of norms from practice. Specifically, the initiative on EU joint programming, which is the EU's latest response to the problem of aid fragmentation, has revealed a disconnect between what is decided at headquarter level and what actually happens on the ground. In fact, in the construction of the JP norm two different national positions reflecting two different rationales have been uploaded to the EU level. On the one hand, there is a view that EU donor coordination would significantly contribute to enhancing the impact of foreign aid. On the other hand, there is the idea (if not the hope) that JP processes would contribute to boosting the EU's impact and identity in international politics and development.
The translation of this vision into action has been enabled, or hindered, by various intervening factors, which help explain change in some member states and not others. Some of these factors pertain to the EU context, others to the external context. In particular, the initiatives undertaken by EU institutions, the attempt of some member states to pursue the aid effectiveness agenda through the EU or the desire of others to strengthen/legitimize the EU's international development actorness can be included among the enabling factors. By contrast, the pursuit of political and commercial interests by several member states (especially in fast growing economies), the resistance of aid bureaucrats who have sought to preserve
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 545
their privileged positions in developing countries or have lacked adequate incentives to engage in coordination initiatives can be included among the constraining factors. Socialization effects have not kicked in as expected, or at least have been less pronounced on the ground than in headquarters. Moreover, contrary to what has happened at headquarter level, differences between those member states that have sought to emphasize the aid effectiveness aspect of the JP initiative and those that have stressed the EU's increased political leverage certainly have slowed down progress - and similar tensions have, though in only a few cases, been experienced within EU Delegations.
But besides the EU level, resistance to JP has come, albeit indirectly, from external (fjactors. In particular, aid recipients have seen very little benefit in an initiative which they have barely understood or bothered to get involved in. As for other donors, they certainly have not acted as external federators - the rise of China in Africa has, in fact, generated increased competition between European donors - and on some occasions have acted as an alternative pole of attraction for some EU member states (e.g. the USA with the UK and Nordic countries). To this changing development landscape, the EU has reacted too slowly, so much so that if one of the primary objectives of JP is that of boosting the profile of the EU in international development, then the result is not very positive. It may be an exaggeration to claim that the aid effectiveness era has come to an end, and with it the ambition of the EU to play a leadership role in international development (discourses). It is however fair to state that the translation of grandiose political commitments into concrete actions has turned out to be, once again, a problem for the European Union, and that, at least in the specific case of joint programming and more generally on aid coordination, there may be more Europeanization at national and supranational levels than in specific contexts in Africa: this, ultimately, means that Europe is still far from happening on the ground.
first submitted November 2015 final revision accepted July 2016
References
Aldasoro, I., Nunnenkamp, P. and Thiele, R. (2010) 'Less Aid Proliferation and
More Donor Coordination? The wide gap between words and deeds', Journal of
International Development 22(7): 920—40. Alecu de Flers, N. and Müller, P. (2012) 'Dimensions and Mechanisms of the
Europeanization of Member State Foreign Policy: State of the art and new
research avenues', Journal of European integration 34(1): 19-35. Bigsten,   A.   and  Tengstam,   S.   (2015)   'International   Coordination  and the
Effectiveness of Aid', World Development 69(1): 75-85. Börzel, T. and Risse, T. (2007) 'Europeanization: The domestic impact of European
Union polities', in K. E. Jorgensen, M. Pollack and B. Rosamond (eds),
Handbook of European Union Politics. London: Sage.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
546    Maurizio Carbone
Bourguignon,   F.   and  Platteau,  J.-P.   (2015)   'The  Hard  Challenge  of Aid
Coordination', World Development 69(1): 86-97. Bretherton, C. and Mannin, M.L. (eds) (2013) The Europeanization of European
Politics. New York, NY: Palgrave Macmillan. Bretherton, C. (2013) 'Development Policy', in C. Bretherton and M. L. Mannin
(eds) The Europeanization of European Politics. New York, NY: Palgrave
Macmillan.
Brown, S. and Swiss, L. (2013) 'The Hollow Ring of Donor Commitment: Country
concentration and the decoupling of aid-effectiveness norms from donor
practice', Development Policy Review 31(6): 737-55. Carbone, M. (2013) 'Between EU Actorness and Aid Effectiveness: The logics of EU
aid to sub-Saharan Africa', International Relations 27(3): 341-55. Carbone, M. (2010) 'The European Union, Good Governance and Aid Co-Ordination', Third World Quarterly 31(1): 13-29. Carbone, M. (2007) The European Union and International Development: The politics
of foreign aid. Abingdon: Routledge. CONCORD (2012) Making Sense of EU Development Cooperation Effectiveness
Special report on the post-Busan development effectiveness agenda. Brussels:
CONCORD AidWatch. Council of the European Union (2011) EU Common Position for the Fourth High
Level Forum on Aid Effectiveness (Busan, 29 November-1 December 2011),
Council Conclusions, 3124th Foreign Affairs Development Council meeting,
Brussels, 14 November. Delputte, S. and Orbie, J. (2014) 'The EU and Donor Coordination on the Ground:
Perspectives from Tanzania and Zambia', European Journal of Development
Research 26(5): 676-91. Easterly, W. (2003) 'The Cartel of Good Intentions: The problem of bureaucracy in
foreign aid', Journal of Policy Reform 5(4): 223-50. EC (2015) EU Accountability Report 2015 on Financing for Development: Review of
progress by the EU and its Member States, SWD (2015) 128, 22 June. Brussels:
European Commission. EC (2014) EU Accountability Report 2014 on Financing for Development: Review of
progress by the EU and its Member States, SWD (2014) 235, 3 July. Brussels:
European Commission. EC and EEAS (2015a) EU Joint Programming Guidance Pack, April. Brussels:
European Commission and European External Action Service EC and EEAS (2015b) Joint Programming: State of Play, October. Brussels:
European Commission and European External Action Service. Exadaktylos, T. (2012) 'Europeanization of Foreign Policy beyond the Common
Foreign and Security Policy', in T. Exadaktylos and C. M. Radaelli (eds)
Research Design in European Studies: Establishing causality in Europeanization.
Basingstoke: Palgrave Macmillan. Fuchs, A., Nunnenkamp, P. and Ohler, H. (2015) 'Why Donors of Foreign Aid Do
Not Coordinate: The role of competition for export markets and political
support', The World Economy 38(2): 255-85.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
European Union aid coordination in Africa 547
Furness, M. and Vollmer, F. (2013) 'EU Joint Programming: Lessons from South
Sudan for EU aid coordination', DIE Briefing Paper 18/2013. Bonn: German
Development Institute. Galeazzi, G., Helly, D. and Krätke, F. (2013) 'All for One and Free-for-All? Early
experiences in EU joint programming', ECDPM Briefing Note 50. Maastricht:
European Centre for Development Policy Management. Gänzle, S., Makhan, D. and Grimm, S. (eds) (2012) The European Union and Global
Development: An 'enlightened superpower' in the making? Basingstoke: Palgrave
Macmillan.
Graziano, P. and Vink, M. (eds) (2007) Europeanization: New research agendas.
Basingstoke: Palgrave Macmillan. Helly, D.; Galeazzi, G.; Parshotam, A.; Gregersen, C; Kokolo, W. and Sherriff, A.
(2015) Stepping Up? Best practice in joint programming and prospects for EU joint
cooperation strategies. ECDPM Discussion Paper 183. Maastricht: ECDPM. Holland, M. and Doidge, M. (2012) Development Policy of the European Union.
Basingstoke: Palgrave Macmillan. Janus, H., Klingebiel, S. and Paulo, S. (2015) 'Beyond Aid: A conceptual perspective
on the transformation of development cooperation', Journal of International
Development 27(2): 155-69. Klingebiel, S., Negre, M., and Morazän, P. (2016) 'Costs, Benefits and the Political
Economy of Aid Coordination: The case of the European Union', European
Journal of Development Research doi:10.1057/ejdr.2015.84. Ladrech, R. (2010) Europeanization and National Politics. Basingstoke: Palgrave
Macmillan.
Lightfoot, S. and Szent-Ivänyi, B. (2014) 'Reluctant Donors? The Europeanization
of international development policies in the new Member States', JCMS:
Journal of Common Market Studies 52(6): 1257-72. Moumoutzis, K. (2011) 'Still Fashionable Yet Useless? Addressing problems with
research on the Europeanization of foreign policy', JCMS: Journal of Common
Market Studies 49(3): 607-29. Nunnenkamp, P., Ohler, H. and Thiele, R. (2013) 'Donor Coordination and
Specialization: Did the Paris Declaration make a difference?', Review of World
Economics 149(3): 537-63. Orbie,   J.   and  Carbone,   M.   (2016)   'Introduction:   The   Europeanisation of
development policy', European Politics and Society 17(1): 1-11. Orbie, J. and Lightfoot, S. (2017) 'Development', in A. Hadfield, I. Manners and R.
W.  Whitman (eds),  Foreign  Policies of EU Member  States. Abingdon:
Routledge.
Pomorska, K. and Wright, N. (2013) 'Europeanization and the Common Foreign and Security Policy', in C. Bretherton and M. L. Mannin (eds) The Europeanization of European Politics. New York, NY: Palgrave Macmillan.
Rahman, A. and Sawada, Y. (2012) 'Can donor coordination solve the aid proliferation problem?, Economics Letters 116(3): 609-612.
Sedelmeier, U. (2012) 'Europeanization', in E. Jones, A. Menon and S. Weatherill (eds) Oxford Handbook of the European Union. Oxford: Oxford University Press.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)
548    Maurizio Carbone
Smith, M. (2016) 'Conclusions: Europeanisation, globalisation or (re)nationalisation?
Revisiting development policy in the European Union', European Politics and
Society 17(1): 136-41. Steinwand, M. C. (2015) 'Compete or Coordinate? Aid fragmentation and lead
donorship', International Organization 69(2): 443-72. Sumner, A. and Mallett, R. (2012) The Future of Foreign Aid: Development
cooperation and the new geography of global poverty. Basingstoke: Palgrave
Macmillan.
Wong, R. and Hill, C. (2011) National and European Foreign Policies: Towards Europeanization. Abingdon: Routledge.
© The Authors 2016. Development Policy Review © 2016 Overseas Development Institute. Development Policy Review 35 (4)