Norms, self-interest and effectiveness: explaining double standards in EU reactions to violations of democratic principles in Sub-Saharan Africa

The promotion of democracy has become a key objective of the European Union (EU) in sub-Saharan Africa. One of the ways in which this objective is pursued is by reacting to violations of democratic principles using negative measures: naming and shaming strategies or economic/diplomatic sanctions. Yet the application of negative measures has been criticised as being characterised by „double standards‟, meaning that similar violations of democratic principles have led to a different response from the EU. This dissertation searches for explanations for these double standards. In-depth comparative studies on the motivations for double standards in the application of negative measures in sub-Saharan Africa in the post-2000 period have been lacking. While double standards have mostly been attributed to the prevalence of self-interested objectives of the EU, this dissertation considers two additional factors: (1) a potential conflict in the EU‟s normative objectives (democracy, development and stability) and (2) expectations about the effectiveness of negative measures. 
These three explanatory factors (norms, self-interest and effectiveness) are investigated for ten case studies in sub-Saharan Africa: Eritrea, Ethiopia, Kenya, Chad, Niger, Nigeria, Rwanda, Guinea, Cote d‟Ivoire and Zimbabwe. It is shown that there have been double standards in the EU‟s reaction to violations of democratic principles in these countries: similar violations of democratic principles have led to different reactions from the EU (positive measures, low-cost negative measures and sanctions). The motivations for these double standards are then investigated by studying the potential impact of norms, self-interest and effectiveness. 
On the one hand, previous studies emphasising self-interest are confirmed. Historical interests related to the desire of member states to maintain their sphere of influence have in some cases led to double standards (Chad, Cote d‟Ivoire, Zimbabwe). Moreover, the growing concern for self-interested security objectives such as the war on terror has led to increased tolerance for governments that are seen as crucial allies in fighting these security concerns (Ethiopia, Nigeria, Chad), while sanctions were more easily adopted where this was not the case (Zimbabwe, Guinea, Cote d‟Ivoire). On the other hand, the study also found evidence that runs counter to the importance of self-interest: (1) commercial interests were found to have little explanatory value, (2) some countries that can also be considered allies in the security sphere were not shielded from negative measures (Kenya), and (3) the EU also avoided negatives measures in countries that were not allies in the pursuit of security objectives (Eritrea, Rwanda). Furthermore, the EU‟s other normative objectives (stability, development) were found to be equally important. Negative measures were more easily adopted in countries where development performance had been weak (Zimbabwe, Guinea, Cote d‟Ivoire, Niger) or where internal stability was under threat (Kenya, Cote d‟Ivoire, Niger). In contrast, negative measures were avoided in the case of „development success stories‟ (Rwanda, Ethiopia). Lastly, the study revealed that expectations about effectiveness were equally taken into account by the EU. Sanctions were more often adopted in case of a coordinated action with the main other actors, and when the domestic position of the government was already weakened. Furthermore, sanctions were avoided where the EU was found to have no leverage (Eritrea, Ethiopia, Rwanda, Ethiopia).


Introduction
Democracy promotion is often criticized for its double standards: in some countries, negative measures (public criticism, sanctions) are used, while in other countries there is a preference for positive measures (incentives, dialogue and capacity building).While such double standards have also been observed in European Union (EU) policies (Brummer, 2009;Warkotsch, 2008), few studies have thoroughly investigated the reasons for these double standards, particularly in Sub-Saharan Africa.
Double standards in democracy promotion are generally attributed to a conflict between democratization and self-interest (Crawford, 2001;Tomasevski, 2000;Gillies, 1996).For the EU, it has been argued that double standards are caused by concerns over maintaining cooperation on security on energy supply, trade interests and a sphere of influence in former colonies (Warkotsch, 2006;Crawford, 2008;Gillespie & Whitehead, 2002;Youngs, 2008;Algieri, 2007).This then runs counter to the representation of the EU as a 'Normative Power' (Manners, 2002(Manners, & 2008) ) or a 'force for good' in the world (Aggestam, 2008).
These studies have mostly focused on strategically or commercially crucial regions such as the Middle East, North Africa, Central and South East Asia, while only a few studies have focused on Sub-Saharan Africa.What is so particular about Sub-Saharan Africa, however, is the dichotomy between norms and self-interest.On the one hand, Sub-Saharan Africa is often perceived from a developmental or humanitarian point of view, referring to its needs in terms of poverty, diseases, conflict, humanitarian aid, etc. (Engel & Olsen, 2005).On the other hand, the strategic role of Sub-Saharan Africa has increased drastically in the 1 This is the report of PhD research carried out at Ghent University, under the supervision of Jan Orbie.
[ 110 ] new millennium, when compared to the post-Cold War era.First of all, there has been a growing interest from the international community in security issues, especially since the Global War on Terror (Verhoeven, 2009).Secondly, commercial interests are also becoming increasingly important.A scramble for Africa's mineral resources is taking place, with old powers (US, EU member states) taking a renewed interest and newly emerging economic powers (Brazil, India, China, Russia) entering the stage (Mangala, 2010).Moreover, there is a growing awareness of Africa's economic potential (Radelet, 2010).This then raises the question about the EU's main motivations to engage in Sub-Saharan Africa in the post-2000 period, and how these motivations may cause double standards in democracy promotion.Scholarly literature suggests that EU-Africa policies are driven by both norms and self-interest.On the one hand, the EU tries to appear as a force for the good in Africa by providing substantial amounts of aid and aid effectiveness (Birchfield, 2011;Carbone, 2007Carbone, & 2011) ) or by focusing on conflict prevention and increasing 'human security' (Sicurelli, 2010;Martin, 2011).The elaborate procedure for imposing sanctions in the case of violations of the essential elements (human rights, democracy, the rule of law) in Article 96 of the Cotonou Agreement has also been considered as an example of EU's Normative Power (Manners, 2002;Bonaglia et al., 2005).On the other hand, it has been claimed that EU self-interest informs EU-Africa policies.For example, the EU's aid allocation was found to reflect EU interests rather than recipient needs (Zanger, 2000;Del Biondo et al., 2012;Neumayer, 2003).Moreover, EU security interventions in Africa are often driven by member states that want to maintain their sphere of influence in their former colonies (Gegout, 2009;Olsen, 2009).
While there is thus uncertainty about whether norms or self-interest are dominating in EU-Africa policies, it is believed that double standards in EU democracy promotion in Sub-Saharan Africa are mainly driven by self-interest (Olsen, 2000;Crawford, 2005;Brüne, 2007).However, these studies have either focused on the pre-2000 period (Olsen, 2000) or have only considered one case study (Crawford, 2005;Brüne, 2007).

Case studies and methodology
In order to fill the above-mentioned research gap, a comparative case study analysis was conducted of ten country cases.In these ten countries, there have either been serious problems with elections 2 and/or frequent violations of political rights in the period from 2000 until mid-2011. 3Despite this being the typical kind of violations that often lead to negative measures, 4 the EU's reaction has varied substantially.In Chad, Rwanda and Nigeria, the EU mainly relied on dialogue, capacity-building and occasional public criticism.In Ethiopia and Eritrea, there was only a brief suspension of development assistance whereas in Ivory Coast, 2 Meaning elections have not occurred, were perceived as unfair by the international community or were illegitimate because of the changing of the constitution to favour or exclude certain candidates.
3 Attacks on the freedom of the press, expression or association, in the extreme case a severe and violent government crackdown on dissent.

4
Public reactions, suspension of development aid/other economic benefits or targeted sanctions (travel sanctions, freezing of assets, etc.).
afrika focus -2012-12 [ 111 ] Reports -Rapports consultations under Article 96 of the Cotonou Agreement were held after the controversial elections of November 2000, but there was no suspension of development assistance.Furthermore, in 2004, a reopening of Article 96 consultations was blocked in the EU Council.In other countries, there was no such reluctance to resort to negative measures.In Zimbabwe (2002) and Guinea (2009), the EU deployed its whole arsenal of sanctions.The position of the EU against President Gbagbo of Ivory Coast also changed completely when he refused to acknowledge his opponent's victory in November 2010 and very severe sanctions were imposed at EU level.In Niger (2009), development aid was suspended and in Kenya (2007), the EU used strong diplomatic pressure, including the threat of sanctions.In these countries, the EU followed up closely on progress made, and strict conditions were set for sanctions to be lifted, which might explain why some of these countries are still subject to sanctions today (Zimbabwe, Guinea).

Kenya
Table 1.Case studies Hence, there have been double standards in EU democracy promotion in the ten countries listed above.The aim of the present research was to explain these double standards.The research mainly relied on qualitative data, although quantitative data were also used in some cases (see infra).As for qualitative data, both primary (official documents) and secondary data (news and academic articles, reports) were used.In addition, 46 non-structured expert interviews were conducted with officials from the European Commission and the European External Action Service, Belgian diplomats and academic experts, most of these interviews were held face-to-face in Brussels in January-March 2012.

Analytical framework, hypotheses and operationalization
As an analytical framework, we departed from the above-mentioned norms versus selfinterest debate, and added one additional factor: expectations about effectiveness.Norms refer to objectives that are predominantly other-based, meant to respond to various needs in the country.These norms include democracy, development and stability (see infra).Selfinterest, however, refers to objectives that are predominantly EU-focused, meant to respond to national interests of member states or common interests of all EU member states, such as security and commercial interests and the desire to maintain a sphere of influence in Africa.Both normative and self-interested behaviour may have 'good' or 'bad' consequences, but the focus of the research was on the motivations rather than on the results of the action.Expectations about the effectiveness of negative measures formed the third factor.Scholarly literature has suggested that the EU avoids negative measures when it does not expect them to be successful, for example where it does not have any leverage (Fischer, 2007;Echagüe & Youngs, 2007).Smith (2008) has argued that this is one of the reasons for double standards in EU democracy promotion.Expectations about effectiveness were seen as a separate factor from norms and self-interest, given that it is not about what the EU pursues, but about how this is pursued.

Norms
The EU's normative objectives in Sub-Saharan Africa can be summarized in three categories: democracy, development and stability.This appears clearly from EU policy documents.In a 1996 Communication on conflict prevention in Africa, the European Commission states that 'the EU policy aims concerning Africa might be summarised as helping to foster peace and stability, development, democracy and the respect for human rights' (emphasis added).The problematic aspect of EU discourse lies in the assumption that democracy, development and stability are mutually reinforcing and interrelated.This can be derived from holistic terms encompassing democracy, development and stability, including 'structural stability' (European Commission, 1996) and 'democratic governance' (European Commission, 2006).Despite this rather idealistic discourse, however, academic research has found that democracy, development and stability do not necessarily go together.For example, so-called developing states combine strong economic growth and political stability with authoritarian rule (Leftwich, 2000).This points to trade-offs between democracy, development and stability in some situations.Double standards can then be the result of a trade-off between (1) democracy and development, or (2) democracy and stability.
The following hypotheses were therefore tested: H.1.The EU is less likely to impose negative measures against countries that are perceived as development success stories, and more likely to impose negative measures against countries that are not perceived as development success stories.
It can be expected that the EU perceives as 'development success stories' countries with high economic growth, good administrative governance and substantial progress regarding poverty reduction and attaining the Millennium Development Goals.Hence, we considered afrika focus -2012-12 [ 113 ] Reports -Rapports GDP growth rates; World Bank governance indicators on the control of corruption and government effectiveness; and the Human Development Index.These quantitative indicators were complemented with official EU documents, leaked US embassy cables, academic and news articles, as well as experts interviews providing information on how donors, and the EU in particular, perceive development progress.
H.2. The EU is less likely to impose negative measures against governments that are perceived as contributing to internal stability, and more likely to impose negative measures against governments that are not perceived as contributing to internal stability.
We looked at ethnic and religious tensions and how these are accommodated by governments, as well at the states' monopoly on violence.Information on these indicators was found in academic articles and reports by the Bertelsmann Foundation and the International Crisis Group.
H.3.The EU is less likely to impose negative measures against governments that are perceived as contributing to regional stability, and more likely to impose negative measures against governments that are not perceived as contributing to regional stability.Contributing to regional stability was operationalized as: providing a safe harbour for refugees, participating in peacekeeping operations, mediating in neighbouring conflicts and respecting peace agreements.Foreign policies of the countries were therefore studied on the basis of reports by the Bertelsmann Foundation and the International Crisis Group, the UN website on peacekeeping and leaked cables from the US embassy.Information on refugee flows was found in the Migration Profiles attached to the Country Strategy Papers and in the UNHCR statistical yearbook.

Self-interest
Self-interest can largely be divided into three categories: historical, commercial and security interests.Historical interests relate to the desire of EU member states to maintain a sphere of influence in Sub-Saharan Africa.This can be linked to colonial ties or to historical links on language-based grounds.Undemocratic rulers who help maintain such links are more likely to be exempted from negative measures.Many studies have argued that French Africa policies are still very much influenced by such a post-colonial sphere-of-influence way of thinking (Bourmaud, 2011;Moncrieff, 2012;Mengara, 2010), while UK Africa policies are more influenced by humanitarian concerns and security interests (Cumming, 2004;Williams, 2010).Commercial interests of the EU in Sub-Saharan Africa have increased in the past decade.Although EU's trade relations with most African countries are marginal, Sub-Saharan Africa is home to many strategic commodities that are becoming scarce, including coltan, diamonds, gold, uranium, casiterite and wolframite (Williams, 2010).Moreover, energy security is becoming more important in the Africa policies of the great powers, and the US in particular.At the same time, EU's energy policies in Sub-Saharan Africa have often focused on the developmental aspect of energy, namely broadening access to energy within Africa (Youngs, 2009).Lastly, security interests have gained in importance in the new millennium, and particularly after the terrorist attacks of September 11, 2001.Compared to the 1990s, there has been a growing awareness that security problems in Africa, including organized crime, illegal migration and terrorism can equally affect security in Europe.This has led to a growing interest from the international community to cooperate with African governments on these issues (Picciotto, 2004;Kraxberger, 2005).
Hence, the following three hypotheses were tested: H.4. The EU is less likely to impose negative measures against governments that protect the sphere of influence of member states, and more likely to impose negative measures against governments that do not protect the sphere of influence of member states.
On the basis of secondary data (academic articles, news articles), we investigated whether a country belonged to a 'sphere of influence' of one of the EU member states.Current and former government-to-government relations were investigated to find out whether governments can be seen as preserving that sphere of influence.

H.5. The EU is less likely to impose negative measures in countries where EU commercial interests are important and where governments serve EU commercial interests, and more likely to impose negative measures against countries where the EU has only minor commercial interests or where the government does not serve EU commercial interests.
Using data on trade and direct foreign investment from the Eurostat online database, we calculated which EU member states were the main commercial partners of the ten country cases in 2000-2010.As a second step, we investigated whether the governments could be seen as securing these commercial interests, e.g. by favouring European companies.Lastly, it was investigated whether commercial interests have led to a more reluctant position of 'interested member states' to use negative measures.
H.6.The EU is less likely to impose negative measures against governments that are perceived as important allies in the fight against security problems affecting the EU, and more likely to impose negative measures against governments that are not perceived as important allies in the fight against security problems affecting the EU.
Information was sought on cooperation in: (1) international terrorism, (2) piracy, (3) illegal migration, (4) organised crime and (5) security of energy supply and production.We consulted leaked US embassy cables and information on projects at the international (International Organisation for Migration, United Nations Office on Drugs and Crime) or EU level on these issues.

Effectiveness
The last set of hypotheses suggests that the EU will avoid negative measures in situations where effectiveness is unlikely.This is especially relevant for the EU as a foreign policy actor that focuses strongly on dialogue and persuasion rather than on sanctions (Manners, 2008;Carbone, 2010).The importance of considerations of effectiveness should also be seen in the light of the reduced leverage of democracy promoters because of the rise of new pow- Reports -Rapports ers, including China and Russia, which are opposed to conditionality and in some cases even promote autocracy (Burnell & Schlumberger, 2010).The literature on sanctions and conditionality suggests that effectiveness is likely to be greater: (1) when sanctions are taken as an internationally coordinated action, (2) when the sender of the sanctions has substantial leverage, and (3) when the domestic position of the government is weak (Crawford, 2001;Emmanuel, 2010;Hufbauer et al., 2007;Stokke, 1995).
Following this literature, the following hypotheses were formulated: H.7. The EU is less likely to impose negative measures when there are no or only a few other international actors prepared to impose negative measures, while it is more likely to impose negative measures in case of a coordinated action by international actors.
We looked at the political and economic support provided by the main non-EU international actors.On the basis of the Organisation for Economic Co-operation and Development's 'Aid at a glance' profiles for 2009-2010, the five main donors in each country were identified.The individual reaction of these main partners to violations of democratic principles was then assessed.
H.8. The EU is less likely to impose negative measures in countries where it lacks leverage, while it is more likely to impose negative measures when it has substantial leverage.The EU's economic leverage was measured by looking at the dependence of each country on EU development assistance. 5To determine political leverage, we looked at the reaction of the governments to the EU's implementation of instruments to promote democratisation.
H.9. The EU is less likely to impose negative measures when the domestic position of the government is strong, while it is more likely to impose negative measures when the domestic position of the government is weak.
It was predicted that the position of the government was weak in case of a potential military coup or armed insurgencies, the presence of pro-democratic forces, anti-government protest and a deteriorating economic situation.Data were mostly retrieved from reports by the International Crisis Group and Bertelsmann Foundation.

Conclusions
The results of the analysis are summarized in table 2. The pluses and minuses indicate whether the hypotheses were confirmed, refuted, or partly confirmed, partly refuted.

Norms
Self

Norms
Despite the focus in literature on self-interest, norms were found to be at least as important.The EU deferred from using negative measures when there was success in the developmental sphere (Rwanda, Ethiopia), while countries with poor development records (Zimbabwe, Guinea) and/or poor administrative governance (Niger, Ivory Coast, Kenya) were more often sanctioned.However, the EU was reluctant to use negative measures against Nigeria, Eritrea and Chad, despite poor performance in the economic sector.
Internal stability also played a role in a number of cases.In Kenya andIvory Coast (2010-2011), the main reason for the EU's tough approach was the fear of destabilisation rather than violations of democratic principles.Similarly, the unwillingness of President Tandja of Niger to cope with Tuareg rebellions made the EU less reluctant to use negative measures.Moreover, in Rwanda, Eritrea and Ethiopia, the relative internal stability under undemocratic rule can explain why the EU did not use strong negative measures.However, a number Reports -Rapports of other governments that were not successful in stabilising their countries (Nigeria, Ivory Coast 2000-2005, Chad) were equally shielded from negative measures.There was no strong evidence of the importance of regional stability.In only three cases was the hypothesis confirmed: Nigeria's role as a peacemaker in West Africa resulted in a more lenient approach towards violations of democratic principles whereas the destabilizing regional role of the governments of Zimbabwe and Ivory Coast made it easier for the EU to use negative measures.However, in the majority of cases, the hypothesis was only partially confirmed.This was because the regional role of most governments was almost never unequivocally good or bad.

Self-interest
We found historical and security interests play a role in explaining double standards, while commercial interests do not.The influence of historical interests is surprising, given that many African countries broke away from their former colonizers, while the interest of the latter in ex-colonies has diminished.However, historical interests still seem to play strongly in some former French colonies, including Chad and Ivory Coast.Historical interests can lead to a stronger or weaker preference for sanctions: in Ivory Coast, France was the main driver behind the sanctions imposed on President Gbagbo after he refused to cede power to Alassane Ouattara following the November 2010 elections.This should be seen in the light of the deterioration of relations between Gbagbo and France in the preceding years.A similar pattern can be observed in Zimbabwe, where tense relations over land reform resulted in the UK's insistence on strong sanctions in 2002.By way of contrast, earlier intentions by the EU to impose sanctions on Ivory Coast as a reaction to violations of democratic principles were blocked by the French, who hoped to improve relations with President Gbagbo during the first years of his rule.Equally, in Chad, sanctions were `off the table´ because of the close relations between President Déby and France.
Commercial interests, on the other hand, were not a good predictor for EU's preference for positive or negative measures.This could be due to the fact that commercial interests were of limited importance in most cases.Furthermore, when commercial interests were important, for example in Nigeria and Niger, they did not conflict with democracy promotion.In Nigeria, EU's preference for positive measures even ran counter to its commercial interests as Presidents Obasanjo and Jonathan were increasingly displaying nationalistic policies.In Niger, a new exploitation contract had been granted to France, but this did not prevent the EU from imposing sanctions a few months later (Grégoire, 2011).
Security interests were found to play a role.In most countries where the EU took negative measures, except for Kenya, the EU received scant cooperation on security issues.Furthermore, Ethiopia and Nigeria, where the EU was reluctant to use negative measures, were crucial allies of the EU in security affairs.Equally, however, there were cases where the EU did not use negative measures despite the lack of security cooperation (Eritrea) or where the EU threatened crucial security allies (Kenya) with negative measures.

Effectiveness
Finally, we found strong evidence of the importance of the position of other actors and the domestic position of the government, but not for EU leverage.Indeed, the EU is more likely to take negative measures when other actors adopt a similar stance, and is reluctant to 'go it alone', even if it could have some individual leverage, such as in the case of Chad or Ivory Coast.Moreover, the EU is more likely to take negative measures when domestic conditions are conducive to democratic change and international pressure can help 'tip the balance' in favour of democratisation.Such situations take place when elections are strongly contested (e.g.Zimbabwe, Ivory Coast 2010-2011) or when there is protest from civil society (e.g.Niger, Guinea).When such opposition is absent, sometimes as a result of repression (e.g.Ethiopia, Rwanda and Eritrea), there are fewer possibilities for the EU to use negative measures to encourage democratisation.On the other hand, the governments of Chad and Nigeria were weak, yet the EU preferred dialogue to negative measures.