Keywords

Introduction

State politics and its transformation in Africa, as elsewhere in developing regions, have been primarily nurtured by its relations with international partners and, most importantly, by how individual states commit to accruing agreements and policy frameworks. The European Union (EU) is one of the most important international partners of Eastern African countries. The EU and its member states are not only the major source of development aid to the region, but they also have strong cultural, trade and strategic relations with Eastern African countries—not to forget the colonial past. This, together with aspirations to build coherence between different policy sectors and interests, reflects a unique relationship and a challenging one. A significant element in the EU co-operation with Eastern Africa relates to the EU’s aspirations to support democracy worldwide. It also lies within Agenda 2063’s aspirations of enhancing strong global partnerships, building stronger institutions and establishing democratic norms. This chapter’s focus deals mainly with Aspiration 3 that seeks to realise an Africa of Good Governance, Democracy, Respect for Human Rights, Justice and the Rule of Law. Like stipulated in other global partnerships or conventions, this aspiration frames fundamental principles for creating a participating African society or realising national integration. The Treaty of Nice of the EU (2001), for example, explicitly stipulated that promoting human rights and freedoms should be part of all EU development and other co-operation with third countries. This principle gained prominence and an explicit codification in the Cotonou Agreement between the EU and the African, Caribbean and Pacific states (ACP) between 2000 and 2022.

The year 2022, when a new agreement is to be signed, is a good time to look back at the application of these principles. Has the EU been able to support democracy in Eastern Africa? To this end, it is useful to pay attention to both positive and negative democratisation instruments the EU employs to prevent democratic backsliding in partner countries in Eastern Africa. As far as the positive instruments are concerned, assistance to the conduct of democratic elections and the election observation missions (EOM) in particular appears to be the public image of EU democracy support before everything else. Economic sanctions, particularly smart sanctions and suspension of development aid, are the most important negative instruments. Altogether, these instruments have also become an increasingly important part of the African Union (AU) and EU’s attempts to prevent the erosion of democratic institutions in its partner countries.

The proceeding sections of this chapter discuss these democratisation instruments and frameworks that undergird them, their consequences for democracy and challenges, mainly in the Eastern African region, albeit we also draw additional illustrations from other African countries. The next section gives an overview of election observations, sanctions and democracy. This is followed by a section on democratic performance in Eastern Africa in light of the Electoral democracy Index. A section after this discusses different components of the EU election observation in Eastern Africa, particularly focusing on the positive and negative democratisation instruments. The last section concludes with a general overview of key issues covered in the chapter.

Election Observation, Sanctions and Democracy

As outlined in several EU policy documents like the Initiatives for Democracy and Human rights (2000–2006 and 2007–2013), the Strategic Framework on Human Rights and Democracy (2012) and the European Instrument for Democracy and Human Rights (EIDHR 2014), promotions of free and fair elections including election observation are important ingredients of the EU’s democracy support. The number of EU election observation missions worldwide has been approximately 12 per year since 2000 (Election observation missions (EUEOMs) 2021). The literature on election observation confirms the generally positive impact it can have on the quality of elections but also the downsides. Inviting election observers has become almost a norm for governments legitimising the use of public power through elections (Hyde 2007). These governments include authoritarian ones. According to Nicholas Cheeseman and Brian Klaas, it seemed easier for authoritarian governments to remain in power by arranging elections than without them (Cheeseman and Klaas 2018).

Consequently, at least potentially, election observation can be used to erode rather than support democracy. Thus, researchers have paid attention to the quality of election observation, particularly as far as the length of the electoral process is concerned. Emphasis is given to the need to extend observation beyond the actual polling day to the preparatory phases like primary elections, voters’ registration and campaign, and the period after, that is, vote counting and announcement and implementation of the election results (Carothers 1997, pp. 22–23). And indeed, the EU has tried to send the first observers about five weeks before the elections, and the final report is compiled only months after.

Another important theme in the literature concerns the criteria of free and fair elections used in elections observation (Bishop and Hoeffler 2016). United Nations criteria are the most preferred ones and have been endorsed by the African Union and the European Union, among others, in 2005. According to these criteria, international election observation is ‘the systematic, comprehensive and accurate gathering of information concerning the laws, processes and institutions related to the conduct of elections and other factors concerning the overall electoral environment’, ‘impartial and professional analysis of such information’, and it should evaluate ‘pre-election, election-day and post-election periods through comprehensive, long-term observation, employing a variety of techniques’ and ‘offer recommendations for improving the integrity and effectiveness of electoral and related processes’ (The Declaration of Principles for International Election Observation and the Code of Conduct for International Election Observers 2005, pp. 2–3). The EU handbook of election observation describes the methodology and aims more in detail, including assessing the political context, human rights issues and civil society’s role. Previous electoral events and the ability of non-governmental organisations to contribute to the electoral process, for instance, are mentioned as issues the observers should consider (Handbook for European Union election observation: third edition 2016).

Yet double standards seem to prevail. Susan Dodsworth, for instance, has shown that observers are less likely to report fraud in an election in sub-Saharan Africa than elsewhere. She notes that between 1991 and 2001, during the first multiparty elections after the post-Cold War transitions, this could be due to ‘progress bias’, that is, observed improvements compared to the previous elections. Since then, according to her, more plausible reasons are related to the objectives to support political stability and relationships with strategic partners. Factors making reporting fraud more likely included allegations of fraud in previous elections and higher levels of foreign aid (Dodsworth 2019).

Sanctions and their usefulness as a tool in international politics have been the object of extensive analysis. Although intended to be most effective as threats to international partners violating jointly agreed on principles, and in that sense to function as a preventive measure, the credibility of such threats requires that they be also implemented against non-compliant partners. Although there are individual cases where sanctions have succeeded, a common criticism is that their initial objectives often become even more difficult to achieve when applied. Negative impacts of economic sanctions on the economy have been frequently observed (Neuenkirch and Neumeier 2015). Peksen and Drury’s (2010) analysis shows that sanctions may also constrain political institutionalisation by worsening the level of democracy, and Wood (Wood 2008) notes that sanctions increase state-sponsored repression. Continuing intimidation of opposition in Zimbabwe, while the EU extended in 2022 its targeted sanctions against individuals and companies, and the escalation of Ethiopia’s Tigray conflict despite new sanctions regimes by the United States in September 2021 are practical examples of the difficulties in reaching positive results (see Nyathi 22/02/2022; Felbab-Brown 1/10/2021). The same can also be stated regarding targeted sanctions on Sudan and South Sudan. A further concern raised in the literature relates to the poor stipulation of exit, that is, when and how the relations will be normalised again (Attia and Grauvogel 2019). However, although weak in terms of their overall effectiveness, targeted sanctions seem to cause less harm to the whole population (Mack and Khan 2000).

EU states that it applies sanctions as per the UN Security Council Resolutions or to further its own foreign policy objectives and nation-building efforts in Eastern Africa. These include ‘promoting international peace and security; preventing conflicts; supporting democracy, the rule of law and human rights; and defending the principles of international law’ (Overview of sanctions and related tools n.d.). Clara Portela, who investigated the effectiveness of EU sanctions policy between 1993 and 2009, notes that suspensions of development aid under the Cotonou Agreement were successful. In contrast, the sanctions imposed under the Common Foreign and Security Policy (CFSP) achieved their objectives only sometimes, and informal sanctions tended to be abandoned without any progress. This was due to the stricter legal framework and management of the sanctions in the Cotonou framework (Portela 2012).

Francesco Giumelli, Fabian Hoffmann and Anna Książczaková, whose study extended to 2010, observed that the number of parallel EU sanction regimes had increased from a handful to over 20 since 2010 and that also the content had become more varied from arms embargoes to asset freezes and travel bans (Giumelli et al. 2021). EU’s subsequent rounds of economic and individual sanctions on Russia over its war against Ukraine in 2022 represented the most comprehensive set of measures with which it intended to influence the behaviour of a third country. Before looking at the performance of EU policy in the 11 countries in Eastern Africa, it is useful to have an overview of the development of democracy in them.

Developments in Eastern Africa

A common understanding of democracy relates to elections, their regularity, quality and universal voting rights. In their discussion of the measurement of electoral democracy, Teorell et al. (2019) argue for including liberal freedoms like freedom of expression and assembly. The electoral democracy index of Varieties of Democracy database covering these elements gives thus a comprehensive picture of the trends in levels of democracy in Eastern Africa. While the region’s overall performance has been relatively stable, country-level differences and volatility are remarkable, as shown in Fig. 20.1 below. We see Eritrea in the bottom and until the end of the decade Mauritius is well above all the others as the extreme cases (see Fig. 20.2).

Fig. 20.1
A graph of ratings versus years depicts the east or horn of Africa, the world, Burundi, Kenya, and Malawi. Malawi is on an increasing trend high at 0.65, Burundi denotes a fluctuation, and East Africa and World remain unchanged.

Country-level differences and volatility in democratic performance. (Source: V-Dem (Coppedge et al. 2022))

Full size image
Fig. 20.2
A graph titled electoral democracy index of rating versus year depicts Burundi, Ethiopia, Rwanda, South Sudan, Zanzibar, Comoros, Kenya, Seychelles, Sudan, Djibouti, Malawi, Somalia, Tanzania, Eritrea, Mauritius, Somaliland, and Uganda.

Country-level differences and volatility in democratic performance. (Source: V-Dem (Coppedge et al. 2022))

Full size image

The figures above compare Eastern Africa to the world average and show the very different trends in Burundi, Malawi and Kenya. Malawi and Kenya have witnessed gradual improvement in the quality of elections and are now at the top of the democratisation levels in the region. In both countries, the role of civil society and its international support and the organisational capacity of the opposition have been critical for democratic consolidation and systems and norms essential to an inclusive state (Mati 2020; Leininger and Nowack 2022). The most dramatic erosion of electoral democracy in the region, in turn, has happened in Burundi. After the Arusha Peace and Reconciliation Agreement and years of civil war, the seemingly successful transition has not abolished anti-democratic political mobilisation, where violence and militarism have remained central (Reyntjens 2020). These three are illustrative examples of the EU’s responses to its partners’ evolving political situations.

EU Election Observation in Eastern Africa

Eastern African countries frequently feature in the list of the EU Election observation missions (EOM) and smaller Election Experts Missions (EEM), and Election Follow-up Missions. Between 2000 and 2020, there were altogether 25 EU missions to the region (Election observation missions (EUEOMs) 2021). With four EOMs and two follow-up missions, Malawi appears to be the most popular country in the region, followed by Kenya with four EOMs and Uganda with three EOMs. Ethiopia, Tanzania and Rwanda have welcomed two EOMs, Sudan one EOM with a follow-up mission, and Somalia one Election Experts mission. Burundi in 2015 is the only case where the EU has decided to discontinue its observation mission. In 2005 and 2010, the EU fully observed the elections in Burundi. As noted above, Burundi, Malawi and Kenya are examples of different political development and institutionalisation in their levels of democracy. It is thus useful to look more in detail at the history of EU election observation in these three countries.

The general elections in Malawi in 2019 were the sixth multiparty elections since 1994, and the fourth time the EU deployed an EOM there, following EOMs in 2004, 2009 and 2014 and Election Follow-up missions in 2013 and 2017. The final report of the EOM from 2019 notes that the Malawian government had addressed more than half, namely 16 of the 27 recommendations made by the EOM in 2014. Improvements were observed in voter registration, a revised election calendar and the creation of constituency tally centres. However, the government had addressed none of the previously suggested electoral legal reforms. These included transparency and accountability of campaign financing and the management of disputes. The report of 2019 also noted the need for a more independent regulatory body and impartial state media (Election Observation Mission, Malawi 2019, 2020).

The 2017 elections in Kenya, the second under the 2010 Constitution, raised again concerns of ethnic divisions and violence exacerbated by electoral problems (Election Observation Mission, Kenya General Elections 2017 2018). Of the 22 recommendations listed in the 2013 EU EOM final report, only two, unified voter registers and defining the campaign duration, were fully implemented. Other recommendations were only partially addressed or wholly ignored. However, the 2013 elections held after the constitutional reform had marked major improvements compared to previous ones. For example, after the 2007 elections, EU EOM recommendations included elections for the President, Senators, Governors, Members of Parliament and County Assembly Representatives in one general election (Election Observation Mission to Kenya 2013). In the preceding rounds, the EOMs had noted only limited amendments addressing the shortfalls of the electoral framework since the 2002 elections. The 2002 EU EOM final report mentions legal shortfalls like the variety and inconsistency of applicable laws (Election Observation Mission, Kenya, 27 December 2007 Final Report on the General Elections 2008, p. 11).

In its first final report of the 2005 elections in Burundi, the EU EOM noted isolated violent incidents but characterised the actual polling as calm and orderly. In the 2010 elections, EOM raised serious insecurity concerns and politically motivated arrests, but the 2015 elections became a turning point. The EU had already sent long-term observers to Burundi but withdrew after the government’s violent response to protests against President Pierre Nkurunziza’s unconstitutional plan to run for a third mandate, intimidation of the opposition, media shutdowns and delay of the elections. African Union, too, concluded that the conditions were not supportive of free and fair elections. However, the East African Community and the UN decided to send observers to the elections. Their report was critical, stating that the electoral process was not credible (Jones and Wittig 2016, pp. 207–208). Within this turmoil, the EU reconsidered its development co-operation with the government of Burundi. Successful elections appeared critical in this consideration.

The Malawian, Kenyan and Burundian cases show that the EU, with its EOMs, build on past mission experiences. It was eager to send observers to subsequent elections and considered the reports of the previous rounds. While this might have resulted in ‘path-dependency’ and regional double standards, in the sense of observers failing to call out problems they had not noted in the past (see Dodsworth 2019, p. 392), Malawi and Kenya give evidence that addressing problems that had been noted with recommendations by the earlier EOMs were duly noted and commended. In Burundi, the EOM’s satisfaction with the credibility of the elections in 2005 did not prevent the 2010 EOM from ending up with a more critical statement and a total withdrawal from the elections in 2015.

One indication of the reliability of the EOMs in this regard is the reluctance of the governments in countries that have become more authoritarian or where democratisation has stagnated to welcome EU EOMs. Tanzania in 2020 and Uganda in 2021 are cases in point. Even in Kenya in 2021, the government referred to ‘previous experiences’ of ‘interference by foreign missions in other nations’ internal affairs’ and asked diplomats and international organisations not to support elections or any activities without government permission (Otieno and Anyango 2021).

Aid Conditionality

Political conditionality was regarded as an innovation in the Cotonou Partnership Agreement (2000) between the EU and 78 African, Caribbean and Pacific states (ACP) (Mackie and Zinke 2005). Its evolution, however, can be traced back to Idi Amin’s military rule, when the-then European Communities (EC) froze its development co-operation with the government of Uganda. Human rights violations were obvious reasons for the decision, but officially, it was justified by Uganda’s inability to implement development programmes. EC’s co-operation agreement with the ACP at that time, the Lomé Convention (1975–2000), did not stipulate political conditions for the co-operation.

On the contrary, the agreement was based on respecting the sovereignty of the EC’s partner governments in managing the co-operation resources. In practice, this meant that the EC did not interfere in the political developments in the ACP countries. Interesting to note that the very first arms embargo of the EU, which was established by the Maastricht treaty of 1992, making European integration more political, was imposed on Sudan in 1994. This arms embargo occurred under the CFSP, not development co-operation and the Lomé Convention. The reason was the civil war in the southern part of the country (Council Decision, 94/165/CFSP 1994).

Only the last revision of the Lomé Convention in 1995 made an explicit reference to democratic principles, the violation of which could then lead to consultations and possibly suspension of European aid—the only case in Eastern Africa where this legal base was utilised as a military coup d’etat in Comoros in 1999. For the ACP side, the EU’s demand to respect democracy was an erosion of the idea of mutual sovereignty and partnership (Vanheukelom et al. 2007). This ACP concern that the EU wanted to decide for its partners’ internal political institutionalisation developments became central in the negotiations for the new agreement (Green Paper on relations between the European Union and the ACP countries on the eve of the twenty-first century 1997; Raffer 2002).

In the final formulation of the new Cotonou Agreement, the principles of human rights, democracy, the rule of law and good governance were emphasised as shared aims. Violation of these principles should trigger consultations between the partners to find a jointly acceptable solution. Only if these consultations did not succeed was the EU, by Article 96 of the agreement, able to take appropriate measures. In practice, this meant withdrawing development funding from the partner country’s government.

Even before any official introduction of consultations, dialogue formed an essential part of the mechanism. The purpose of the dialogue was ‘to exchange information, to foster mutual understanding, and to facilitate the establishment of agreed priorities and shared agendas’ as well as ‘preventing situations arising in which one party might deem it necessary to have recourse to the non-execution clause’ (Article 8). Vanheukelom et al. (2007) noted that the ACP partners were already concerned about the possible lack of consistency and transparency in the transition from dialogue into official consultations. In the later revisions of the agreement, this process phase was defined more in detail. In practice, the EU has implemented a case-by-case approach, reflecting the political history of its partners, lessons learned from the earlier cases and the EU member states’ different interests. For instance, Gordon Crawford and Simonida Kacarska (2019) noted inconsistency in the policies and the importance of security interests in the post-2000 period sanctions policy of the EU.

Furthermore, it is noteworthy that the political dimensions of the Cotonou Agreement, and the tensions related to it, were not restricted to Article 96. Most importantly, the 2005 revision of the Cotonou Agreement encouraged the parties to join the Rome Statute of the International Criminal Court (Article 11), making it the first binding international legal instrument with such a clause. Sudan, for instance, refused to ratify the revision as the ICC issued an arrest warrant for Omar Hassan Ahmad Al Bashir in 2009 over war crimes. An obvious challenge for the EU was that withdrawal of development aid would limit its co-operation with governments that, undemocratic by definition, were least likely to change their behaviour due to popular or international pressure (Morrissey 2004). It is thus not surprising that the EU has also declared its commitment to engage in ‘difficult partnerships’ (Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions 2003). For instance, despite its very authoritarian government, Eritrea has never been invited to consultations under Article 96.

Use of the Article 96

Until 2022, Article 96 had been invoked 17 times with 13 countries: Haiti (2000), Fiji (2000 and 2007), Côte d’Ivoire (2001), Liberia (2001), Zimbabwe (2002), Central African Republic (2003), Guinea-Bissau (2004 and 2011), Togo (2004), Guinea-Conakry (2004 and 2009), Mauritania (2005 and 2008), Madagascar (2010), Niger (2009) and Burundi (2015) (Consultation procedure (article 96) 2021). Most of the cases were from the early years of the Cotonou Agreement. And indeed, early on, the EU had become cautious about proceeding with official consultations and sanctions. In particular, the difficult process in Zimbabwe had instigated a view of Article 96 as an instrument of ‘last resort’ to be used in a reactive manner when all other attempts to support democracy and human rights had been used instead of one to prevent major political crises.

Experience of the use of Article 96 showed that the process, somewhat paradoxically but not surprisingly, was most likely to lead to positive results when the government of the ACP partner was co-operative. That was usually the case after a coup d’état. For the coup-makers, the new power holders, consultations with the EU provided a window to be recognised in the international community, be a partner for the EU, and normalise their international relations to the extent possible. Without exception, the jointly agreed appropriate measures then were positive. Most important among these was the organisation of elections and EU support for these to be free and fair, including their observation. Electoral politics appeared central to Article 96 (Laakso et al. 2007).

It is thus interesting to look at cases where Article 96 has not been used despite violations of the democratic principle. The highly centralised, authoritarian, but strategically important Eritrea was already mentioned. The lack of political reform in Eritrea led the EU in 2020 to reallocate funds reserved for Eritrean infrastructure programmes to Sudan (Vilmer 2021, pp. 31–32). In Sudan, the EU has concentrated on rehabilitation and humanitarian aid after the 2005 Comprehensive Peace Agreement (Ahmed 2009). As Sudan did not ratify the revised version of the Cotonou Agreement, due to the clause of the International Criminal Court, the EU was not providing support directly to the government. Development co-operation was implemented through NGOs and international organisations. This, of course, meant that the government could neither be invited to official consultations over co-operation nor punished by withdrawing this co-operation.

Also, in Somalia, where instead of a sudden coup d’état, the context has been one of state collapse and prevailing interregnum governance, the priority of the EU has been state-building, humanitarian aid and co-operation through the UN organisations and NGOs. Regular development co-operation has been constrained by logistical difficulties related to violent conflicts. A fourth important example is Ethiopia, one of the major development co-operation partners of the EU. The reports of the EU EOMs in 2005 and 2010 of ‘a narrowing of political space and an uneven playing field’ never led to consultations (Election Observation Mission—Ethiopia 2010, PRELIMINARY STATEMENT 2010). At the same time, civil society and democratic governance have appeared on the co-operation agenda of the EU in Ethiopia. Emphasis on regional security and countering terrorism required the continuation of good relations with the government.

The only Eastern African country where the EU used Article 96 was Burundi. This happened after the political crisis of 2015 and the EU’s withdrawal from monitoring the elections. The consultation process ended without a solution, and in March 2016, the EU suspended direct budgetary support to the government of Burundi. Burundi became the last country where sanctions were applied under the Cotonou Agreement. The sudden death of President Pierre Nkurunziza in 2020, the subsequent election of Évariste Ndayishimiye, the formation of a new government, revitalisation of Burundi’s co-operation with neighbouring countries and voluntary return of refugees created a momentum for the EU to lift sanctions (COM (2021) 833 final). However, a list of issues in human rights, good governance and the rule of law needing improvement was still long. The EU thus referred to its ongoing political dialogue with the government (Burundi: EU lifts existing restrictions under Article 96 of the ACP-EU Partnership Agreement n.d.).

The reluctance of the EU to impose sanctions upon its partners in Eastern Africa confirms the general observation that it has been patient regarding countries with continuing problems with democratic principles and countries that have strategic importance. A sudden erosion of the situation in a smaller country, Burundi, led to action. Burundi is also an example of an exit that exploited an opportunity that accidentally emerged than an outcome of a successful sanction policy. Throughout the implementation of sanctions, exit had been a difficult balancing act for the EU between its credibility and relations with the recipient state.

Conclusions

While it is difficult to assess the overall effectiveness of the political dimension in EU’s relations with Eastern Africa, that is, the actual influence that the EU’s policy to support democracy and prevent democratic backsliding has had, there is no doubt that this dimension has been and remains significant in discerning political institutionalisation of Eastern African countries. The EU’s aspiration to move from development aid-centred co-operation to a co-operation between equal partners, emphasised in its Africa strategy of 2020, only accentuates that dimension precisely. The strategy defines the promotion of nation-building policies bordering on human rights, democracy and the rule of law as well as ‘engaging together on the global scene to strengthen the multilateral rules-based order’ as the mutual interest of the ‘twin continents’ (towards a comprehensive strategy with Africa 2020, p. 1).

The emphasis on shared values and aims is key to understanding the EU’s strategy towards Eastern Africa in the broader context of international relations and international development co-operation. Combined with Africa’s own aspiration to become politically more united and also influential in international politics, ‘reinforcing Africa’s Global Voice’ along with the priorities of the African Union (Continental Priorities | African Union n.d.), commitment to the same values and aims could indeed be a basis for major force in the international fora. For the EU, this entails continued co-ordination of its policies with so-called like-minded partners, Western countries outside of the EU, which in the post-Brexit era also includes the United Kingdom, the major European partner of Eastern Africa. Most importantly, however, this entails responses to the potentially non-democratic influence of other international powers present in the region and eager to capitalise on the troubled history of African countries with their former colonial masters.

Competing and conflicting external interests are a challenge to Africa’s unity. One example is the division of the voting positions of African states in multilateral fora. In 2022, the UN General Assembly resolution called for Russia to be suspended from the Human Rights Council. This was due to reports of war crimes committed by Russian forces in Ukraine. To the disappointment of the EU, only four Eastern African countries: Comoros, Malawi, Mauritius and Seychelles, supported the resolution. Burundi, Eritrea and Ethiopia voted with China and Russia against the resolution, while the rest abstained like India.

Looking back to the EU’s democracy support in Eastern Africa since 2000, it seems obvious that there cannot be one fit for all forms. Case-by-case approach and difficult balancing between different aims and available instruments are likely to be part of that policy also in the future. What remains to be seen are ways in which governments and civil societies in the region will use that support for the benefit of their society. As the experience of Article 96 of the Cotonou Agreement has shown, such instruments are most effective if used in a co-operative manner and through joint decision-making.