The European Commission’s implementation of budget support and the Governance Incentive Tranche in Ethiopia: democracy promoter or developmental donor?

Abstract

The complex relationship between democracy and development has been extensively discussed in academic literature. However, we do not have much knowledge of how this translates into donor practices. How does the European Commission (ec) deal with tensions arising from promoting democracy and development? To answer this question, this article operationalises the distinction between ‘democracy promoters’ and ‘developmental donors’, focusing specifically on budget support and governance incentive tranches. Empirically we examine the implementation of the ec’s budget support and Governance Incentive Tranche in Ethiopia (2005–10), a case where the dilemma between democracy promotion and development cooperation is particularly strong. Investigating the position of the ec along the democracy promoter versus developmental donor continuum, we conclude that the ec’s position lies between these extremes. However, in the case of budget support, a shift has been made away from the Commission being a democracy promoter and towards the role of developmental donor.

Keywords:

• European Union
• Ethiopia
• conditionality
• democracy
• budget support
• incentive tranche

Introduction

This article examines the inclusion of democracy in the European Commission’s (ec) development policy. Research on EU democracy promotion in developing countries has primarily focused on competing interests, and historical, economic and geopolitical interests have been found to be a more important concern for the EU than democracy.¹ Similarly those analysing the EU’s aid allocations have mostly juxtaposed normative goals (development, democracy, human rights) against donor interests. Such studies have found multilateral EU assistance to be responsive to the development needs of its partners,² but more correlated with strategic interests than with democracy and human rights.³ While the studies have mostly investigated whether norms or interests inspired EU policies, the literature has not sufficiently recognised the tension between objectives such as democracy and development.

The complicated relationship between democracy and development has been amply researched. However, the translation of insights from this research into donor practices has only been addressed to a limited extent. Whereas the ec has always had substantial relations with developing countries, and has strengthened its profile in development since the 2000s, it has also been increasingly active in the field of democracy promotion over the past 20 years. How does the ec deal with the tensions arising from promoting both democracy and development? In order to examine this research question, the article first elaborates on the ideal types of ‘democracy promoters’ and ‘developmental donors’. Democracy promoters prioritise democracy in their relations with developing countries and thus integrate strict selectivity and political conditionality in aid policies. Developmental donors prioritise development and are reluctant to let democracy interfere with aid policies. Subsequently, we operationalise the ‘democracy promoter’ and ‘developmental donor’ ideal types, by looking specifically at the implementation of budget support and governance incentive tranches. These instruments are particularly relevant to the analysis of the development–democracy nexus as they are relatively unclear on whether they have democracy, development or good governance as their objectives, and on which should prevail in the case of conflicting objectives. In the case of budget support, there is a divide between the ec’s official position, which was until recently that budget support was a development instrument, and policy practice that included suspensions for purely political reasons. In the case of the Governance Incentive Tranche (git), ‘democratic governance’ is the main purpose, the definition of which vacillates between democracy and development. Whereas research on both instruments has mostly focused on the level of policy formulation,⁴ this study makes an empirical contribution by focusing on a concrete case study: the ec’s implementation of budget support and the git in Ethiopia. The latter is a relevant case study for three reasons. First, Ethiopia plays a central role in the ec’s development policy: In 2005–10 it was on average the eighth largest recipient of ec aid.⁵ Second, the divide between economic progress and democratic regression between 2005 and 2010 has created a clear dilemma for donors, who do not want to jeopardise their productive relations on developmental issues by raising problems in the field of democracy.⁶ Third, Ethiopia is a strategically important country, although the impact of this strategic position on EU democracy promotion is unclear. While some have argued that the EU has prioritised security over democracy,⁷ others emphasise that its lack of economic interests in the country has enabled it to focus on democracy.⁸

The article is structured in two main parts. The first section briefly reviews the debate on the relationship between democracy and development and elaborates on how this has translated into donor practices. In doing so, we focus on the ideal-typical distinction between ‘democracy promoters’ and ‘developmental donors’ and how this can be operationalised by examining budget support and ex-post conditionality. We then situate the ec’s position in this debate. The second section examines the ec’s implementation of budget support and the git in Ethiopia in 2005–10.

The paper is based on an analysis of policy documents, quantitative and qualitative data, and 45 expert interviews conducted by one of the authors in Brussels and Addis Ababa. The interviews in Brussels were held in 2009 and 2012 with five ec officials and six representatives of EU member states.⁹ In Addis Ababa 34 interviews were conducted in January 2011 with representatives from EU member state embassies and aid agencies, the World Bank, United Nations Development Programme (undp), the Canadian embassy, the EU Delegation, and relevant Ethiopian authorities and civil society organisations.¹⁰ All interviews were conducted in confidentiality and the names and specific functions of the interviewees were withheld by mutual agreement.

Translating the democracy–development nexus into donor policy practice

Linking democracy and development

‘Democracy’ and ‘development’ are mostly viewed as analytically distinct concepts, although it is acknowledged that they are interrelated in a complex way.¹¹ Nonetheless, research on the relationship between democracy and development has led to conflicting views.¹² Those who believe in a positive relationship argue that competitive multiparty elections make rulers accountable to citizens, leading to more development-oriented policies.¹³ In contrast, some believe that democracy has a negative effect on development. Democracy entails a number of characteristics that can be dysfunctional for development, including the tendency for short-term planning, immediate spending and the influence of interest groups.¹⁴ Authoritarian regimes, on the other hand, foster political stability and insulation from interest groups. Examples of this are the Southeast Asian ‘developmental states’, including Taiwan, Thailand and Singapore, which had spectacular growth rates in the 1980s and 1990s.¹⁵ Some authors have transcended these two opposing views by noting the developmental value of elements that can be achieved under both democracy and authoritarian regimes: political stability and good governance.¹⁶ Evidence on the positive effect of political stability on development has led to the ‘sequential’ argument that donors should focus on state building before moving on to democracy.¹⁷ Similarly there is evidence that good governance characteristics such as property rights, the rule of law, a good regulatory framework and a non-corrupt administration are essential for development.¹⁸ However, the latter argument is based on a narrow, technocratic definition of good governance. A broader definition presumes that democracy, human rights and the rule of law are essential elements of ‘democratic governance’.¹⁹ To distinguish both dimensions of governance, we will henceforth refer to the former as ‘technocratic governance’, and to the latter as ‘political governance’.

Democracy promoters and developmental donors

The literature on how the debate on democracy and development has been reflected in donor practices is less evolved, basically centring on two opposing ideal types.²⁰ ‘Democracy promoters’ believe that democracy is conducive to development and that democracy should be promoted for its intrinsic value.²¹ ‘Developmental donors’, on the other hand, are reluctant about pursuing democracy, questioning its development dividend. In developmental states they avoid making criticism of the quality of democracy. In politically unstable countries they prefer a ‘sequential’ approach (see above).²² When it comes to ‘governance’, the former would promote ‘political governance’, while the latter focus on ‘technocratic governance’.²³

Neither of these ideal types has been clearly operationalised in the literature. We aim to contribute to filling this gap by looking specifically at the ec’s implementation of budget support and the git. Both instruments were primarily meant to enhance aid effectiveness, but have subsequently been used as democracy promotion instruments. Budget support, which is the transfer of financial assistance into the budget of recipient countries, is appreciated by developmental donors as the best aid modality to meet the principles of the Paris Declaration on Aid Effectiveness: ownership, alignment, harmonisation, managing for results and mutual accountability.²⁴ Indeed, budget support improves the predictability and stability of aid, the respect for national agendas and donor harmonisation, a results-based approach, and the notion of partnership and mutual accountability.²⁵ However, budget support is controversial in authoritarian countries as it is perceived as symbolising trust in the recipient government.²⁶ This is especially the case with General Budget Support (gbs), though less so with Sector Budget Support (sbs). Budget support, and gbs in particular, carries a higher fiduciary risk (the risk that funds are diverted for unknown and possibly unauthorised purposes, eg corruption, military expenses, vote buying).²⁷ As a result, budget support is more difficult to defend to taxpayers than other aid modalities.²⁸ To avoid public criticism and a misuse of donor funds by authoritarian countries, budget support is increasingly being given with political strings attached.²⁹

The emergence of governance incentive tranches should be seen in the context of the move from ex-ante conditionality (whereby criteria are set beforehand and then tested on their implementation) to ex-post conditionality (whereby conditionality depends on past performance).³⁰ Following the 1998 World Bank report Assessing Aid: What Works, What Doesn’t and Why, donors started to focus on countries with good past performance in the use of aid.³¹ Most donors thus established a methodology to identify good performers, but the implications for aid allocation varied from donor to donor. Some limited partner countries to those considered ‘good performers’, while others set up an incentive scheme to reward them.³² Donors diverge on whether political governance should be included in the criteria. Some donors include both political and technocratic governance, others only include technocratic governance.³³

Hence, there are diverging views on budget support and ex-post conditionality along the democracy promoter versus developmental donor continuum. In the remainder of this section we will further operationalise how ideal-typical ‘democracy promoters’ and ‘developmental donors’ would act in both cases, thereby elaborating some indicators that can be used for empirical research. A democracy promoter sees budget support as an instrument with which to induce democratic reform. Hence, political governance is included in the eligibility criteria for budget support, which is suspended in the event of regression on political governance. Moreover, democracy promoters discuss political governance in the budget support policy dialogue.³⁴ Democracy promoters include political governance in the aid allocation criteria, in the hope that this will motivate governments to democratise and lead to dialogue on political governance. Such indicators can be found in Dutch aid policies. In 2002 it was decided to limit Dutch assistance to 36 partner countries. Political governance was an important criterion in this selection. Moreover, political governance became a condition for receiving budget support and was included in the budget support dialogue.³⁵

At the other end of the continuum developmental donors provide aid on the basis of purely developmental criteria, such as growth and poverty reduction. This includes political criteria, whose relationship with development is not questioned, such as political stability and technocratic governance. Developmental donors only suspend budget support when development performance criteria are not respected. Reversals of democracy only lead to suspension when political stability is challenged.³⁶ Dialogue between developmental donors and recipient governments is restricted to development; there is no ‘political dialogue’. The World Bank falls into this category, as its mandate is limited to economic issues. The Bank’s Country Policy and Institutional Assessment, which is used for aid allocation, does not include political governance.³⁷

Of course, most donors are positioned between these extremes. A common perspective is to see budget support as a development instrument, but to have political conditionality for all aid. In this case political governance is not included in the eligibility criteria nor discussed in the policy dialogue, but can be discussed in the political dialogue.³⁸ This is the approach recommended by the Organisation for Economic Cooperation and Development’s Development Assistance Committee (oecd-dac),³⁹ which has also been followed by the UK’s Department for International Development (dfid).⁴⁰ Donors may take a position in which political governance is included in the criteria for aid allocation, but is not a necessary condition. The practice of governance incentive tranches also fits into this approach, as it entails a less strict form of conditionality that rewards good performance but does not punish bad performance. For example, Belgium provides incentive tranches on the basis of governance performance, including political governance.⁴¹ Hence, positions on budget support and ex-post conditionality can be placed on a continuum from democracy promoters to developmental donors. In the following section we will look at the position the ec has taken on this continuum.

The democracy–development nexus in ec policy practice

It is unclear where the ec is positioned on the democracy promoter to developmental donor continuum. On the one hand, the ec is expected to lean towards the developmental side. Development policy has always been at the heart of the European project, whereas democracy promotion only goes back to the 1990s.⁴² Since the early 2000s the ec has presented itself as a frontrunner in aid effectiveness and a staunch supporter of poverty reduction and the Millennium Development Goals (s).⁴³ Development principles have also increasingly defined its democracy assistance.⁴⁴ The ec has enjoyed a relatively autonomous position, with limited control from the European Parliament and only indirect control from member states’ parliaments.⁴⁵ Carothers has also suggested that – like most European donors – the ec is a developmental donor, although he has not systematically investigated this thesis.⁴⁶ On the other hand, the ec may be moving in the direction of a democracy promoter. Since the early 1990s its external agreements have increasingly incorporated political conditionality.⁴⁷ Democracy has become a ‘core norm’ of ec development policies.⁴⁸ For example, the Cotonou Partnership Agreement (cpa, 2000) foresees political dialogue and a conditionality clause.

From the available evidence on the design and implementation of ec budget support, the ec’s position is unclear. The official position until 2011–12 was to keep political governance outside budget support, following the oecd-dac. Eligibility criteria were restricted to technocratic governance, in terms of: (1) transparent, effective and reliable public finance management; (2) well defined macroeconomic or sector policies; and (3) open and transparent public procurement (cpa, Article 61). Political conditionality applied to the overall aid envelope. Article 96 of the cpa enabled the EU to suspend development assistance in the case that one of the essential elements described in Article 9 (democracy, human rights, the rule of law) was violated. These issues were also discussed in political dialogue, as foreseen in Article 8 of the cpa. In contrast, the budget support dialogue was limited to development.⁴⁹ In practice, however, there are instances where the ec has acted as a democracy promoter. In Ethiopia (2005) and Nicaragua (2008) budget support was suspended as a reaction to flawed elections. In Niger (2009) the ec froze budget support in response to the referendum that abolished presidential term limits. Moreover, the distinction between policy dialogue and political dialogue is not entirely clear. As former Commissioner for Development Louis Michel suggested, ‘the influence the Commission may have […] on a country’s budgetary choices calls for an overall vision of all the country’s political priorities and thus provides a framework for broader dialogue encompassing other forms of aid’.⁵⁰ Moreover, policy dialogue includes all budget support donors, including those with a democracy promoter’s perspective.⁵¹ For example, the electoral process was addressed in the budget support dialogue with Mozambique,⁵² and, following recent proposals by the Commission, gbs will be substituted by ‘Good Governance and Development Contracts’, which will be restricted to countries that respect the EU’s fundamental values: democracy, the rule of law and human rights. These issues will also be discussed in the policy dialogue, which ‘should be consistent with the EU’s overall political dialogue’.⁵³

Similarly, in the case of the git, the ec’s position is unclear. Political governance was included in the criteria for the git, but only two of its nine criteria concern political governance, the others being primarily developmental or related to technocratic governance and political stability.⁵⁴ The git only determined €2.7 billion of the total allocation of the 10th European Development Fund (edf). Governance was not taken into account in the main allocation of the latter.⁵⁵ Furthermore, the size of the incentive tranche was based on commitments to governance reform, rather than on past performance. The allocation of the git was done as follows. First, the ec and the member states, through their delegations, prepared a Governance Profile, identifying the main governance shortcomings. Recipient governments then drew up Governance Action Plans (gaps) to address these shortcomings. Subsequently the ec and the member states assessed the gaps on relevance, ambition and credibility. Four possible tranches were granted, depending on the quality of the gap: 10%, 20%, 25% or 30% on top of the initial amount granted under the 10th edf.⁵⁶ The git’s focus on ownership corresponded to the ec’s aim to ‘apply the Paris Declaration principles to governance’.⁵⁷ Nonetheless, certain elements of the git leaned closer to a democracy promoter’s approach. First, failure to implement the gap could lead to aid reductions. Indeed, the gap was used to assess governance performance in the Mid-Term Review (mtr), which is carried out after two-and-a-half years to identify how far funds must be adapted according to needs and performance in four areas: governance, the economic situation, the poverty and social situation, and performance in the implementation of ec cooperation.⁵⁸ Second, the gaps would be discussed in political dialogue: ‘The tools developed for analysis and dialogue in the programming process [the git] should in their turn inform political dialogue’.⁵⁹ Whether the ec can be considered a democracy promoter or developmental donor through budget support and the git will now be analysed by looking at its policies towards Ethiopia.

The ec’s implementation of budget support and the git in Ethiopia

In the period 2005–10 Ethiopia saw great development progress but at the same time slid backwards in the democratisation process. The next section illustrates this democracy–development gap by measuring the situation of development, technocratic governance, political stability and political governance, using available qualitative (interviews, academic articles, reports by intergovernmental and nongovernmental organisations) and quantitative (the World Bank’s Governance Indicators, Freedom House) data. Then the ec’s implementation of budget support and the git is investigated.

The democracy–development gap in Ethiopia

Most academics and donors perceive Ethiopia as a development success story. It is counted among the 17 emerging African countries,⁶⁰ as well as the ‘Lion Kings’– the fastest growing African economies (The Economist, January 6, 2011). This perception is primarily based on Ethiopia’s gdp growth figures. According to the imf, real gdp growth was 11.8% on average between 2004 and 2008. Moreover, progress in social indicators has been noticeable. Despite its scarcity of primary resources, Ethiopia is most likely to meet most mdgs;⁶¹ it made the second largest improvement in the Human Development Index over the period 2000-2010.⁶² Service delivery systems function relatively well compared with most low-income African countries. Government effectiveness has significantly improved since 2002 (Table 1). While such a positive evolution cannot be found in the control of corruption, corruption in Ethiopia is perceived as quite low compared with the rest of sub-Saharan Africa.⁶³ Ethiopia’s growth has been strongly led by the government. Inspired by the Southeast Asian ‘developmental states’,⁶⁴ three five-year development plans have been adopted since 2002, each of which has been based on strong agricultural growth and a state-centred view of development. The most recent plan, the Growth and Transformation Plan (gtp, 2010–15), aims to maintain a growth path of 11% on average and envisages full attainment of the mdgs by 2015 and middle-income status by 2020.⁶⁵

Table 1. Ethiopia, worldwide governance indicators.

	2002	2004	2006	2008	2010
Political stability	-1.28	-1.26	-1.78	-1.78	-1.72
Government effectiveness	-0.93	-0.73	-0.59	-0.41	-0.35
Regulatory quality	-1.24	-0.96	-0.94	-0.82	-0.85
Control of corruption	-0.73	-0.72	-0.62	-0.67	-0.73

Source: Data retrieved from the World Bank Worldwide Governance Indicators website.

This is not to say that there are no reasons for concern. The high population growth offsets efforts to reduce the number of poor people. Despite growth, Ethiopia continues to suffer from droughts and food shortages.⁶⁶ Its regulatory quality has decreased since 2004, as shown in the table. Furthermore, the cost of living has increased as a result of the high inflation rate, estimated at 18% in 2006–09.⁶⁷ The trend towards political stability has also been negative. According to Foreign Policy’s ‘Failed States Index’, Ethiopia can be considered a ‘failed state’.⁶⁸ Nonetheless, donors generally have a positive impression of Ethiopia’s development process and, as a result, it is a ‘donor darling’ (The Economist, November 1, 2007; bbc News, June 1, 2010). This should be seen in relation to Ethiopia’s conflict-ridden (Sudan, Somalia) and/or economically less successful (Eritrea) neighbours, and in the light of its history of civil war.⁶⁹ Donors give the ruling Ethiopian People’s Revolutionary Democratic Front (eprdf) credit for the introduction of ethnic federalism, which they see as the best option for stability in the ethnically divided country.⁷⁰

In the field of political governance, however, there is unanimity that the process went backwards between 2005 and 2010. The 2005 elections ranked as the most competitive ever, as previous elections were largely dominated by the eprdf. The opposition enlarged its share in the parliament from 12 to 172 seats. However, election observers, including the EU Election Observation Mission (eu-eom), found serious evidence of fraud. This led to demonstrations, which were met by a crackdown by security forces in June and November 2005, leading to almost 200 deaths.⁷¹ Subsequently the eprdf multiplied its membership by pressuring officials and their families to become party members, while several laws stifled independent voices. Under the 2008–09 media and anti-terrorism laws press freedom and the freedom of opposition can be restricted to protect national security. The 2009 Charities and Societies Proclamation prohibited civil society organisations that receive more than 10% in foreign funding from working on advocacy.⁷² Together with the weakening of the opposition, this may explain why the eprdf won all but a few seats in the 2008 local and 2010 parliamentary elections.⁷³ Table 2 shows Ethiopia’s scores on indicators of political governance. Freedom House’s indicators on political rights and civil liberties worsened in 2010 (a higher score means less freedom), bringing Ethiopia into the group of ‘Not Free’ countries. The country’s Voice and Accountability rating also decreased between 2004 and 2010.

Table 2. Ethiopia, indicators on political governance.

	2002	2004	2006	2008	2010
wgi Voice and Accountability	-1.22	-1.14	-1.17	-1.29	-1.32
Freedom House political rights	5	5	5	5	6
Freedom House civil liberties	5	5	5	5	6

Source: Data retrieved from Freedom House and World Bank Worldwide Governance Indicators (wgi) websites.

The implementation of budget support

The ec provided €139 million in gbs to Ethiopia’s Sustainable Development and Poverty Reduction Programme (2002–05). However, gbs came under pressure after the crackdown on demonstrations in November 2005. The European Parliament was especially vocal and condemned the violence in two resolutions in July and September. The Development Assistance Group (dag), a group of mainly oecd donors, decided to suspend budget support.⁷⁴ The ec blocked its funds earmarked for gbs and took off the agenda the Financing Proposal for €155 million in sbs for roads.⁷⁵ While the official reason for this suspension was ‘political instability’ (Agence France Presse, December 28, 2005), the decision was also based on ‘political governance grounds’.⁷⁶ However, rather than using the conditionality clause foreseen in Articles 9 and 96 of the cpa, political conditionality was only applied to budget support. In this sense, the ec took a democracy promoter’s approach to budget support. At the same time Article 8 political dialogue took place and Commissioner Louis Michel travelled to Ethiopia to encourage the release of political prisoners.⁷⁷

This position changed from 2006 onwards, when the ec moved more in the direction of a developmental donor. This shift has often been attributed to Ethiopia’s role in fighting the Islamic Courts Union in Somalia.⁷⁸ Nevertheless, developmental motivations were more crucial. First, policy dialogue was affected by the suspension of budget support. The minister of finance called the ec’s decision unacceptable (Agence France Presse, December 30, 2005) and criticised donors for failing to fulfil their commitment to provide predictable support based on an objective assessment of government performance.⁷⁹ Second, the ec was confronted with a proactive World Bank, which worked with the minister of finance on a new programme to recommit the frozen funds.⁸⁰ The Protection of Basic Services (pbs) programme, approved in May 2006, transfers budget support in a block grant to the federal government, which then disperses it to the regional and district level. It is specifically directed to poverty reduction sectors: health, education, water, agriculture and roads. Certain provisions aim to reduce the fiduciary risk of pbs funds, including technical assistance for budget transparency and a social accountability component.⁸¹ Despite these safeguards, Human Rights Watch found in 2010 that, in the distribution of services financed by the pbs, local officials discriminated against political opponents.⁸² However, the dag continued with the pbs, arguing that its own assessments contradicted these findings.⁸³

Despite the reluctance of the Ethiopian government to release political prisoners, the ec unfroze sbs in the road sector in 2006.⁸⁴ It also joined the pbs in December 2007 and has since then committed about €255 million to the programme.⁸⁵ The ec’s position thus changed from that of a democracy promoter in the direction of a developmental donor. Although the pbs is technically not budget support, it is not that different from gbs, which was also transferred to lower levels under Ethiopia’s decentralised system. Several donor representatives have therefore described it as ‘budget support in disguise’.⁸⁶ In addition, rather than re-evaluating its contribution to the pbs in 2010, given the evidence of aid politicisation, the ec was in favour of going back to gbs. The dag had endorsed the gtp and the ec had €100 million left from its macroeconomic envelope. By keeping this amount together and disbursing it as gbs to the gtp, the ec hoped to have more leverage in the policy dialogue in the context of Ethiopia’s industrialisation process and to ‘play with the big ones’, namely the imf and the World Bank.⁸⁷ The ec thus ordered a study, which found that Ethiopia had fulfilled all the technical conditions for gbs. However, the member states took a democracy promoter’s position by opposing the resumption of gbs, in the light of the situation of political governance. This runs counter to the thesis that Ethiopia’s status as a donor darling is merely inspired by security interests:⁸⁸ one would expect it to be the member states which would prioritise security cooperation, rather than the ec. Following the opposition of the member states, the ec divided the €100 million and increased its contribution to the pbs (€50 million), created a Transformation Triggering Facility (€30 million) for the private sector, and increased sbs for roads (€20 million). Even with the new proposals on budget support the ec’s aid package to Ethiopia will not be affected. Indeed, its contribution to the pbs is technically not gbs, but a contribution to the World Bank. Nor would sbs be affected, given that the eligibility criteria for it will not change.⁸⁹ A return to gbs would, however, be impossible with the current conditions for budget support.⁹⁰

The implementation of the git

The git was discussed in the aftermath of the 2005 elections, amid tensions over the critical report by the eu-eom. Most probably for this reason discussions were held in the policy dialogue, rather than in the political dialogue.⁹¹ Ethiopia’s Governance Action Plan was based on the Plan for Accelerated and Sustainable Development to End Poverty (2005–10), which included a matrix with benchmarks for political and technocratic governance. In this sense ownership and alignment were respected. However, these principles were problematic from the perspective of a democracy promoter. The policy matrix was vague and omitted crucial issues such as the independence of the electoral board, lessons learned from the 2005 elections, modalities for political party financing and the role of civil society.⁹² Regarding the electoral process, for example, the gap planned ‘free and fair elections in accordance with the electoral law, regulations and procedures of the National Electoral Board of Ethiopia’.⁹³ However, the eu-eom had called for a revision of the electoral law and questioned the independence of the electoral board.⁹⁴ Nonetheless, the ec was reluctant to act as a democracy promoter by letting these political governance problems overshadow Ethiopia’s rather promising commitments on socioeconomic governance. It approved the gap and granted a middle-size tranche,⁹⁵ which was perceived to be quite large given the difficulties in drafting the plan.⁹⁶

The mtr was undertaken in the run-up to the 2010 elections, when political space had seriously narrowed. The implementation of several commitments in the gap had proved problematic in practice. For example, the revision of the press law, as provided in the gap, allowed prosecutors to stop any publication under the pretext of ‘national security’ and increased punishment for defamation.⁹⁷ However, although the ec and member states agreed that the situation of political governance was bad, it was not considered bad enough to overshadow the good marks in other areas.⁹⁸ Hence, Ethiopia received a rather negative rating on governance and also on the economic situation (a C on a scale from A to D), but this was compensated by more positive scores for its social and poverty situation and for performance in the implementation of ec assistance.⁹⁹ As a result, the mtr did not lead to an adaptation of the ec’s development package in Ethiopia. The ec’s approach was thus neither purely developmental, as political governance problems were taken into account, nor in line with the democracy promoter ideal type, as better scores in other areas compensated for the deterioration in political governance.

Conclusions

This article started from the observation that democracy and development do not necessarily go hand in hand, which may result in a democracy promoter or developmental donor approach to development policy. Different positions on the democracy promoter to developmental donor spectrum were defined for budget support and ex-post conditionality, and then applied to the ec’s implementation of budget support and the Governance Incentive Tranche in Ethiopia. It was found that, in the implementation of budget support, the ec made a shift from being a democracy promoter in 2005 towards being a developmental donor in 2006–10. In the case of the git the ec is situated between the ideal types.

We have found that the ec is not a pure democracy promoter, as development considerations have played a role, particularly in its decisions on budget support and less so on the git. Regarding budget support the ec has followed the position of the World Bank that the loss of budget support should be compensated by the pbs, and even considered a return to General budget support while political governance was sliding backwards. Where the git is concerned, development performance compensated for flaws in political governance. One might argue that other instruments may be used for democracy promotion purposes, but even the use of Articles 8 and 96 of the cpa was limited. However, the ec is not a pure developmental donor either. It cannot be said that democracy played no role in the implementation of budget support and the git. The former was suspended in 2005 for political reasons and, in the case of the latter, political governance was one of four criteria in the Mid-Term Review. The ec’s position midway between a democracy promoter and developmental donor can be partly explained by its sui generis character. It is not a purely bilateral donor, such as the USA or its own member states, and hence does not have a direct link with taxpayers. At the same time it is not a purely multilateral donor, such as the World Bank, as its member states have more influence and there is indirect pressure from the European Parliament.¹⁰⁰ Moreover, unlike the World Bank, the ec’s mandate is broader than that of a purely developmental donor.

The shift from being a democracy promoter to a more developmental position on budget support points to flexibility in the implementation of this instrument. The ec’s diversion from the original set-up of budget support can be explained as follows. First, all budget support donors were in favour of suspension, creating peer pressure and the hope of having some effect through coordinated action. Second, the European Parliament was strongly involved in 2005, and less so in subsequent years. Third, as noted on the website of the Bureau of Investigative Journalism on August 4, 2011, the head of the ec delegation at the time, Tim Clarke, was personally concerned about the political situation in Ethiopia and called upon the EU to ‘act firmly and resolutely’.

Nonetheless, flexibility in the use of budget support creates uncertainty about the ec’s position. The new proposals on budget support further add to this confusion, especially since they do not spell out the threshold on political governance necessary to qualify for gbs or the conditions for its suspension. Similarly the long list of governance issues in the git makes the ec’s position more diffuse and there is no clear definition of when incentives are granted and aid is decreased. As a result, almost every country received a middle-sized tranche, even if their gap was of low quality.¹⁰¹ A recent evaluation found that, as in the case of Ethiopia, the git has not led to any adjustment of aid in most African, Pacific or Caribbean countries, nor has it often been used in political dialogue.¹⁰²

Notes on Contributors

Karen Del Biondo is a postdoctoral fellow at the Research College ‘The Transformative Power of Europe’ at the Free University of Berlin. She spent a year as a postdoctoral fellow at the Center on Democracy, Development and the Rule of Law, Stanford University and in 2008–12 held a doctoral fellowship from the Fund for Scientific Research at the Centre for EU Studies, Ghent University. Her publications include “Security and Development in EU External Relations: Converging, but in Which Direction?” (with Stefan Oltsch and Jan Orbie), in Handbook of European Union Security, edited by Biscop and Whitman (2012); “Democracy Promotion Meets Development Cooperation: The EU as a Promoter of Democratic Governance in Sub-Saharan Africa”, European Foreign Affairs Review 16, no.5 (2011); and “EU Aid Conditionality in acp Countries: Explaining Inconsistency in EU Sanctions Practice”, Journal of Contemporary European Research 7, no.3 (2011).

Jan Orbie is a Professor with the rank of senior lecturer in the Department of Political Science, and co-director of the Centre for EU Studies, both at Ghent University. His research focuses on the ‘soft’ (trade, development, social) dimensions of EU external relations and on the EU’s role as a normative power. His publications include “The European Commission’s Neoliberal Trade Discourse since the Crisis: Legitimizing Continuity through Subtle Discursive Change” (with Ferdi De Ville), British Journal of Politics & International Relations, May 2013; “EU Trade Policy and the European Security Strategy: Speaking the Same Language?” (with Fabienne Bossuyt and Lotte Drieghe), Studia Diplomatica 65, no. 3 (2012); and “The European Union’s Trade Policy Response to the Crisis: Paradigm Lost or Reinforced?” (with Ferdi De Ville), European Integration Online Papers 15, no. 2 (2011). He has also edited two books on EU trade and development politics, one on European external policies and one on Europe’s global social policies and has been co-editor of special issues of the European Foreign Affairs Review (2009; 2011), Journal of Contemporary European Research (2013) and the Cambridge Review of International Affairs (forthcoming 2014).

Acknowledgements

This article was possible thanks to a PhD fellowship and a travel grant from the Fund for Scientific Research (fwo) and postdoctoral grants from the Belgian–American Educational Foundation, Fulbright and the kfg ‘Transformative Power of Europe’. An earlier version of the article was presented at the eadi/dsa General Conference in York, September 2011. The paper has benefited from comments by Jonathan Fisher, Bregt Saenen, Hrant Konstanyan, Peter Stevens, Petra Debusscher and Jan Van Heukelom. All remaining errors are ours.

Notes

1. Crawford, “The European Union and Democracy Promotion in Africa”; Olsen, “Promotion of Democracy”; Brüne, “Testfall Äthiopien”; and Khakee, EU Democracy Promotion in Nigeria.

2. Grilli and Riess, “ec Aid to Associated Countries”; and Dearden, “An Analysis.”

3. Zanger, “Good Governance and European Aid”; and Neumayer, The Pattern of Aid Giving.

4. Molenaers and Nijs, “From the Theory of Aid Effectiveness”; Molenaers and Nijs, “Why the European Commission Fails”; and Schmidt, “Budget Support.”

5. Calculated on the basis of the oecd-dac’s online Creditor Reporting System.

6. Carothers, “The Elusive Synthesis,” 24.

7. Brüne, “Testfall Äthiopien”; and Borchgrevink, “Limits to Donor Influence.”

8. Börzel and Hackenesch, “Small Carrots, Few Sticks.”

9. Three of them were working for a Permanent Representation, one was a former Cabinet member and two were officials working for an aid agency.

10. The Ministry of Finance and Economic Development, Ministry of Foreign Affairs and Ethiopian Human Rights Commission.

11. Dahl, On Democracy; Diamond, Developing Democracy; Merkel, Embedded and Defective Democracies; and Levy and Fukuyama, Development Strategies.

12. Sirowy and Inkeles, “The Effects of Democracy”; and Doucouliagos and Ulubaşoğlu, “Democracy and Economic Growth.”

13. Collier and Chauvet, “Elections and Economic Policy”; and Sen, Development as Freedom.

14. Przeworski and Limongi, “Political Regimes”; and Haggard, Pathways from the Periphery.

15. Evans, Embedded Autonomy; and Leftwich, States of Development.

16. Norris, Making Democratic Governance Work; and Fukuyama, “What is Governance?”

17. Mansfield and Snyder, Electing to Fight; and Fukuyama, “The Imperative of State-Building.”

18. North, Institutions; and Keefer and Knack, “Why don’t Poor Countries Catch Up?”

19. Norris, Making Democratic Governance Work; and Fukuyama, “What is Governance?”

20. Carothers, “Democracy Assistance”; and Carothers, “The Elusive Synthesis.”

21. Carothers, “The Elusive Synthesis,” 14.

22. Ibid., 20; and Unsworth, “What’s Politics got to Do with It?,” 887–890.

23. Carothers, “Democracy Assistance,” 7–9.

24. oecd-dac, The Paris Declaration.

25. Koeberle and Stavreski, “Budget Support”; Hayman, “Budget Support and Democracy”; Molenaers, “The Great Divide?”; and Unwin, “Beyond Budgetary Support.”

26. Molenaers et al., “Budget Support and Policy/Political Dialogue,” 13; Unwin, “Beyond Budgetary Support,” 1516; and Hayman, “Budget Support and Democracy,” 680.

27. Morrissey, “Fungibility”; and Shand, “Managing Fiduciary Issues,” 27–29.

28. Knack, When Do Donors Trust?, 5–6.

29. Hayman, “Budget Support and Democracy.”

30. Hout, The Politics of Aid Selectivity, 96; Molenaers, “The Great Divide,” 796; and Molenaers and Nijs, “Why the European Commission Fails,” 411–412.

31. Doornbos, “Good Governance,” 103.

32. Carbone, “The European Union.”

33. Doornbos, “Good Governance”; and Hout, The Politics of Aid Selectivity.

34. Molenaers et al., “Budget Support and Policy/Political Dialogue,” 18–19; and Molenaers, “The Great Divide,” 799–780.

35. Ministerie van Buitenlandse Zaken [Ministry of Foreign Affairs], Begrotingssteun, 52–54.

36. Molenaers et al., “Budget Support and Policy/Political Dialogue,” 14–15; and Molenaers, “The Great Divide,” 793–801.

37. Hout, The Politics of Aid Selectivity, 25–32.

38. Molenaers et al., “Budget Support and Policy Dialogue,” 17–18; and Molenaers, “The Great Divide,” 799.

39. oecd-dac, “Harmonising Donor Practices.”

40. dfid, Poverty Reduction and Budget Support, 4.

41. oecd-dac, “Belgium Peer Review,” 71.

42. Baracani, “US and EU Strategies,” 304.

43. Fraser and Whitfield, “Understanding Contemporary Aid Relationships,” 83; and Carbone, The European Union.

44. Del Biondo, “Democracy Promotion meets Development Cooperation.”

45. Molenaers and Nijs, “Why the European Commission Fails,” 420; and Martens et al., “Some Policy Conclusions.”

46. Carothers, “Democracy Assistance,” 16–18.

47. Börzel and Risse, “Venus approaching Mars?,” 37–40.

48. Vogt, “Coping with Historic Responsibility.”

49. European Commission, “Community Support for Development Programmes,” 9.

50. Michel, “Budget Support,” 19 (emphasis added).

51. Molenaers, “The Great Divide,” 799–801.

52. Interview, member state aid official, Brussels, January 2012.

53. European Commission, “The Future Approach”; and Council of the EU, “The Future Approach.”

54. European Commission, “Supporting Democratic Governance,” 6.

55. European Commission, “Commission Decision Regarding the Adoption of Aid Allocation Criteria,” 5–9.

56. Molenaers and Nijs, “From the Theory of Aid Effectiveness,” 422.

57. European Commission, “Governance in the European Consensus,” 10.

58. European Commission, “Guidelines for the 10th edf,” 11–12, A8–10.

59. European Commission, “Governance in the European Consensus,” 11.

60. Radelet, “Success Stories.”

61. Ethiopia will most probably meet mdgs 2–7, and could also achieve mdgs 1 and 8 if changes are made. “mdg Monitor.” Accessed October 19, 2012. http://www.mdgmonitor.org/country_progress.cfm?c=ETH&cd=231.

62. undp, Human Development Report 2010, 3.

63. Furtado and Smith, “Ethiopia,” 132.

64. Zenawi, “States and Markets.”

65. Federal Democratic Republic of Ethiopia, Growth and Transformation Plan.

66. European Community and Ethiopia, “Country Strategy Paper,” 9–10, 24.

67. Authors’ calculations, on the basis of the imf World Economic Outlook 2011.

68. Foreign Policy, “Failed States Index.” Accessed December 20, 2012.

http://www.foreignpolicy.com/failed_states_index_2012_interactive.

69. Abbink, “The Ethiopian Second Republic.”

70. Clapham, “Post-war Ethiopia,” 182.

71. Abbink, “Discomfiture of Democracy?,” 185–186; Tronvoll, “Ambiguous Elections,” 458.

72. Aalen and Tronvoll, “The End of Democracy?,” 203.

73. Tronvoll, “The Ethiopian Federal and Regional Elections.”

74. Human Rights Watch, Development without Freedom, 19.

75. European Community and Ethiopia, “Country Strategy Paper,” 33–34.

76. Ibid., 34–35.

77. Ibid., 6.

78. Borchgrevink, “Limits to Donor Influence”; and Brüne, “Testfall Äthiopien.”

79. Furtado and Smith, “Ethiopia,” 137.

80. Ibid., 138.

81. World Bank, “Protection of Basic Services.”

82. Human Rights Watch, Development without Freedom.

83. Ibid., Annex 2.

84. European Community and Ethiopia, “Country Strategy Paper,” 33.

85. According to data from the oecd-dac Creditor Reporting System (converted from US$, constant 2011 prices).

86. Anonymous interview, Addis Ababa, January 2011.

87. Anonymous interview, Addis Ababa, January 2011.

88. Borchgrevink, “Limits to Donor Influence”; and Brüne, “Testfall Äthiopien.”

89. Anonymous interview, Brussels, January 2012.

90. Anonymous interview, Brussels, October 2013.

91. Hackenesch, European Good Governance Policies, 28.

92. dag, “Ethiopia,” 10–11.

93. European Community and Ethiopia, “Country Strategy Paper,” Annex 9.

94. eu-eom Ethiopia, “Final Report,” 3–4.

95. The size of the incentive tranche is not published; however, several interviews confirmed that Ethiopia received a middle-size tranche.

96. Hackenesch, “European Good Governance Policies,” 27.

97. Human Rights Watch, One Hundred Ways, 50; and Aalen and Tronvoll, “The End of Democracy,” 200.

98. Anonymous interview, Addis Ababa, January 2011.

99. Anonymous interview, Addis Ababa, January 2011.

100. Martens et al., “Some Policy Conclusions.”

101. Molenaers and Nijs, “From the Theory of Aid Effectiveness”; and ecdpm and agi, “Support Study,” 12.

102. ecdpm and agi, “Support Study,” 24–26.