The effect of EU conditionality on the democratic development in the Western Balkans and the European Eastern Neighborhood: The cases of Serbia, North Macedonia, Albania, Georgia and Moldova

(1)

The effect of EU conditionality on the

democratic development in the

Western Balkans and the European

Eastern Neighborhood

The cases of Serbia, North Macedonia, Albania, Georgia and

Moldova

Master Thesis

Bozhana Bozhilova s2093367

Thesis supervisor: Dr. Rik de Ruiter Second reader: Dr. Natascha van der Zwan

MSc Public Administration, Public Management and Leadership Leiden University

(2)

2

List of Abbreviations

AA Association Agreement

BTI Bertelsmann Stiftung’s Trasformation Index CEE Central and Eastern Europe

CEEC Central and Eastern European Country DCFTA Deep and Comprehensive Free Trade Area DG Georgian Dream-Democratic Georgia DPM Democratic Party of Moldova

DUI Democratic Union for Integration of Macedonia EEA European Economic Area

ENP European Neighborhood Policy EU European Union

FH Freedom House

IPA I Pre-accession assistance 2007-2013 IPA II Pre-accession assistance 2014-2020 NATO North Atlantic Treaty Organization PD Democratic Party of Albania PS Socialist Party of Albania

PSRM Party of Socialists of the Republic of Moldova SAP Stabilization and Association Process

SDSM Social Democratic Union of Macedonia

SNS Serbian Progressive Party (Srpska Napredna Stranka) SRS Serbian Radical Party (Srpska Radikalna Stranka) UN United Nations

UNM United National Movement of Georgia WTO World Trade Organization

(4)

4

1. Introduction

Since the end of the Cold War, the European Union, among other organizations, has become a prominent promoter of peace, liberal democracy and human rights, as well as the peaceful resolution of international and ethnic conflicts in Eastern Europe (Schimmelfennig, 2004, p. 2). Evidently, the creation of the Enlargement policy as a means for EU’s external political integration has been often cited as most effective in inducing democratic change (e.g. Schimmelfennig & Scholtz, 2008). However, the recent decline in the quality of democracy in some of the CEE members and narrow or reversible patterns of fulfilment of EU criteria in the Western Balkans have brought to greater criticism and uncertainties about the effectiveness of EU conditionality (e.g. Boerzel & Schimmelfennig, 2017; Elbasani & Šabić, 2018). At the same time, the EU has been promoting democracy and other EU values even in states which are not aspiring to join the Union or that are denied opportunities for application. This happens through various institutional arrangements, such as bilateral agreements, the European Economic Area, the Barcelona Process or the European Neighborhood Policy (ENP) (Schimmelfennig, 2010). Hereby, the latter is often described as being modelled or imitated on the EU’s Enlargement policy and as such having a similar structure. Nonetheless, compared to the Enlargement policy, the ENP has gained even greater disapproval by scholars for not being able to transmit democratic values amongst the Neighborhood countries (e.g. Gawrich et al., 2010; Nilsson & Silander, 2016; Theuns, 2017).

Researchers argue that the key components of the pre-accession conditionality were replicated in the ENP (e.g. Kelley, 2006; Kochenov, 2008; Schimmelfennig, 2009). However, the main difference between the two is that the ultimate reward or carrot of EU membership in the Enlargement policy is not offered in the ENP. According to Boerzel & Schimmelfennig (2017), specifically the size and credibility of the reward make for the effectiveness of EU conditionality. In this sense, EU membership is regarded as the most sizable reward, i.e. the strongest external driver for countries to adopt EU rules and to increase their level of democracy. Also, states that perceive EU membership as a credible outcome are considered to experience more extensive democratic changes. On the contrary, Boerzel & Schimmelfenning (2017) maintain that countries which only have partner agreements and associations with the EU without a credible EU perspective will experience democratic advancements in a slower pace to a much lesser extent. Similarly, other authors contend that precisely without the perception of potential membership as an outcome, ENP countries

(5)

5 cannot be motivated to adapt their laws to the EU, and thus to become more democratic (Schimmelfenning & Scholtz, 2008, Smith, 2005).

This makes the countries being subject to EU conditionality in the Enlargement and the Neighborhood policy a noteworthy case for comparison. While the first are subject to clear and specific incentives and enforcement procedures, the latter depend on vaguer and lighter in its structure conditionality. Furthermore, it is interesting to test whether the hypothesis made by Boerzel and Schimmelfennig (2017), that countries without a perspective for EU membership as a reward cannot undergo as quick and effective democratic progress. For this purpose, the thesis will study particular current candidate states and countries from the ENP. The countries that are focus of this study are Serbia, North Macedonia, Albania, Georgia and Moldova. Consequently, the effectiveness of EU conditionality on the progress in democracy will be tested on the aforementioned case countries. The key question of the research is: What is the effect of EU conditionality on the progress in democratic development in the Western Balkan countries of Serbia, North Macedonia and Albania, and the countries of the European Eastern Neighborhood, Georgia and Moldova? While attempting to provide a sufficient answer to this question, additional sub-questions have been developed aiming at facilitating the study’s investigation and analysis. These are: 1) How does EU conditionality work? 2) What was the progress in democratic development for the case countries in the period 2009-2018? 3) In which country was conditionality most effective in terms of progress in democratic development?

The research will analyze each case country’s progress in democratic development and the effectiveness of EU conditionality in the country related to size of reward, domestic costs, governmental and societal commitment and economic interdependence. In order to do so, the thesis will rely on a number of sources, such as European Commission progress and association reports, statistics on governance, democracy and corruption from the World Bank, Freedom House, Bertelsmann Stiftung and Transparency International, official web sites of international organizations, in particular the UN, NATO and WTO, and political parties’ web sites, as well as survey data on citizens’ perceptions about the EU. The results and analysis section indicate that there is a positive relationship between EU conditionality and progress in the democratic developments in the case countries. Nonetheless, conditions for this relationship prove to be low or moderate domestic costs, a strong political identification with the EU and positive societal commitment towards the Union.

(6)

6 There is extensive literature on the EU conditionality and its effectiveness. On the one hand, the Western Balkan countries are commonly selected as case countries from researchers studying EU conditionality. On the other, the most advanced ENP countries, namely Georgia, Moldova and Ukraine are also often examined in terms of EU’s external policy integration. Nonetheless, the literature on relating the Western Balkan countries which are given membership perspective to economically and politically similar countries from the ENP which are denied such perspective is scarce. This study aims to contribute to the existing literature on the effectiveness of conditionality by considering the theory by Schimmelfennig and Sedelmeier (2004, and 2005) and the aforementioned assumption by Boerzel and Schimmelfennig (2017). Moreover, since the end of Communism, both regions under investigation have been experiencing particular difficulties in the transformation of their government structures into full democracies. Hence, the comparison between the case countries can provide a better understanding to the mechanisms and approaches behind the compliance of countries with EU rules and their liberal and democratic development. Importantly, the research is socially relevant for both the EU and domestic actors. Domestic actors can be stirred to take into consideration best practices from other countries, while the EU can become more open towards states which are currently denied EU membership but show considerate improvements in human rights, rule of law and democracy. This on the other hand, can help for improving the development of policies on EU integration.

Following this section, the second chapter of the thesis will provide a background information on the European Enlargement and Neighborhood policy for better understanding and distinguishing them. The third chapter will discuss the relevant existing literature, the theoretical framework and the derived from it expectations which will be analyzed later in the study. The fourth chapter will explain the methodology of the research, including the selection of cases and the operationalization of concepts. Chapter five and six will provide information on the findings and the analysis of the case countries respectively. The last chapter will summarize the main points of the thesis and its limitations, while also considering possible grounds for future research.

(7)

7

2. The EU Enlargement and Neighborhood policy clarified

This section will briefly explain the EU Enlargement policy and the European Neighborhood policy (ENP). This is necessary in order to better understand the similarities and differences between the two policies and the effects they have on the case countries.

2.1 Enlargement policy

The policy is targeted at countries that are aspiring to become EU members and at current potential candidates. The main objectives of the Enlargement policy are focused on fostering peace and stability in states close to the EU, increasing prosperity, assisting and monitoring countries attempting to bring their legislation in line with the EU values, assisting in the advancing of the citizens’ quality of life through integration and enhanced cooperation (European Commission, 2014). Additionally, the pillars of rule of law, economic governance and public administration reform have been of key importance for deciding when states are fully prepared to become members (European Commission, 2014).

Countries that wish to become part of the EU need to fulfil a number of requirements. These requirements are referred to as the Copenhagen criteria since they were agreed upon at the European Council in Copenhagen in 1993 (European Commission, 2016a). The criteria include countries to ensure democracy, respect the rule of law and the protection of human rights, to ensure a functioning market economy and to be able successfully undertake the obligations of membership (European Commission, 2016a). Hereby, it is important to note that the countries from the Western Balkans have to comply with additional conditions related to reconciliation, regional cooperation and good neighborly relations which are outlined in the so-called “Stabilization and Association Process” (SAP). Additionally, countries applying for accession to the EU, negotiate the conditions and timing of complying with all EU rules separated in different policy areas and set out in 35 chapters in the “acquis communautaire” (European Commission, 2016a). The EU rules themselves cannot be negotiated by the candidate country. Only the timing and the conditions under which the country adopts the rules can be discussed considering the individual political and economic situation of each candidate. Throughout the talks, the European Commission is the institutional body which monitors and oversees the progress of the candidate countries. The Commission also guides the countries into meeting the required criteria and submits reports

(8)

8 for the Parliament and the EU Council on the current state of the candidates (European Commission, 2016a).

The process of becoming a member of the EU consists of three main stages, namely providing an official candidate status of the country that is applying, opening of the chapters and starting formal negotiations, the final completion of all negotiations and requirements, and the satisfaction of all parties followed by a ratification of the signed agreement and the official joining of the country to the EU (European Commission, 2016b). As already mentioned above, the Western Balkan countries have to comply with additional criteria set out in the SAP. This process aims at further stabilizing the countries in the region, assisting them in the transition towards market economy and fostering regional cooperation with an end objective of prospective membership.

The EU assists in the progress of the countries in financial and technical ways through the instrument for Pre-accession Assistance 2014-2020 (IPA II). The contributions intend to help the beneficiaries in the implementation of the new reforms and to prepare them for their future obligations as EU members. The IPA’s objective is to equalize the standards in the candidate countries with those of the Union in order to achieve a smooth economic and political transition. As communicated by the European Commission, the IPA II makes for greater ownership by the recipients through incorporating their own reforms and agendas. The IPA II also includes new indicators for better assessment of what has been achieved (European Commission, 2019b).

Countries that aspire EU membership are given the status of “potential candidates”. When these countries are ready, they receive a status of “candidate countries”. The route between the potential candidates and the candidates helps countries to gather knowledge and competence needed for the implementation of EU legislation. This is created through free trade agreements, financial and technical assistance, stabilization and association agreements, all of which bring the country closer to the EU. The progress of each country is evaluated in the annual progress reports announced by the Commission (European Commission, 2016b). At the present, candidate countries are Albania, North Macedonia, Montenegro, Serbia and Turkey. Negotiations have been opened only with Serbia, Montenegro and Turkey. Potential candidates are Bosnia and Herzegovina and Kosovo (European Commission, 2019a).

(9)

9 2.2 European Neighborhood Policy (ENP)

The European Neighborhood Policy was adopted by the EU in 2004 as a result of the EU 2004 Enlargement and the followed changes in the EU external borders (Delcour & i Lecha, 2017). With this, the EU aimed at fostering security, stability and prosperity in its neighborhood as well as strengthening its relations with the closest countries (Schimmelfennig, 2015; European Commission, 2016c). The policy is founded on EU values of democracy, rule of law and respect for human rights (European Commission, 2016c). Initially, it was focused on the Eastern European neighboring countries of the EU. However, subsequently it was extended to include North African and Middle Eastern EU neighbors of the Euro-Mediterranean Partnership, also known as the Barcelona Process, and the Caucasus (Schimmelfennig, 2015). Overall, the countries of the ENP are divided into two categories, namely Eastern partnership neighbors (Armenia, Azerbaijan, Belarus, Georgia, Moldova and Ukraine) and Southern partnership neighbors (Algeria, Morocco, Egypt, Israel, Jordan, Lebanon, Libya, Syria, Tunisia, and the Palestinian Authority) (European Commission, 2016c).

The two main principles of the ENP are differentiation and ownership (European Commission, 2016c). Differentiation means that the EU utilizes a differentiated approach towards its neighbors by understanding the diverse aspirations in progressing their relationships with the EU. In the 2004 Strategy Paper on the ENP, it is indicated that the degree to which relationships between the EU and the neighbors are being developed will consider the extent to which the values of democracy, rule of law, respect for human rights among others are successfully shared (European Commission, 2004, p. 13). Also, the policy rather implies a positive conditionality and bases itself on the more for more principle, meaning that the more the country adopts EU-like reforms and brings itself closer to the Union, the more it will receive in return (Delcour & I Lecha, 2017). Ownership refers to the process through which priorities are set, namely the fact that the priorities are outlined together with the partner countries which brings for a greater feeling of ownership on the part of neighbors. This ensures that the priorities are based on common interests and their accomplishment is expected to lead to closer relations between the neighboring partners and the EU (European Commission, 2004; European Commission, 2016c).

(10)

10 Several scholars have reported about the similarities of the ENP with the EU Enlargement policy stating that the ENP was created as an alternative of the Enlargement policy with the main difference of the absence of EU membership perspective (e.g. Kelley, 2006; Schimmelfenning, 2015; Schimmelfennig, 2017). Kelley (2006) explains that the ENP is the outcome of path-dependent institutional choices, policy learning and adaptations of the Enlargement policy. She argues that both the ENP and the Enlargement policy comprise of socialization and conditionality practices. Aside the distinction about membership within the ENP, differences also arise around the principles of differentiation and joint ownership. For instance, as already mentioned the ENP utilized a differentiated approach to the partner countries depending on their progress whereby countries are not grouped or treated uniformly as when accession negotiations are opened within a particular group of countries (Kelley, 2006, p. 34). Furthermore, mutual ownership also differentiates the ENP from the Enlargement policy since partner countries are allowed to develop a more tailored approach on how to tackle issues whereby control from the EU is more distant (Korosteleva, 2017). Additionally, neighboring countries are permitted to approximate their laws to the European, while membership candidates are subject to more hierarchical model of governance and legal harmonization (Schimmelfenning, 2017).

2.2.1 The Eastern Partnership

The Eastern Partnership (EaP) was developed as a reaction of the European Commission to the vulnerability and issues of security in the region, particularly the conflict in Georgia in 2008, and the willingness of the partnership countries to strengthen their relations with the EU (European Commission, n.d.). The initiative was launched in 2009 between the EU, the EU member states and six Eastern and Southern neighbors of the EU, namely Armenia, Azerbaijan, Belarus, Georgia, Ukraine and Moldova (Council of the European Union, 2019). The EaP is specific element of the ENP which provides base for further cooperation and aims at realizing closer political correlation and economic integration. Here, the differentiation approach is preserved, however, bilateral relations are intensified and multilateral cooperation is additionally encouraged, meaning that when issues are shared cooperation and information exchanges are of great importance. Bilateral relations are achieved through Association Agreements (AA) providing strengthening of political association with the EU and improved political interchanges (Council of the European Union, 2019). The EaP motivates countries to pursue democratic reforms which are in line with the EU by concretely

(11)

11 offering visa facilitations, free trade agreements (DCFTA) and other contractual relations in return (Council of the European Union, 2019). The European Union has established free trade agreements with three of the countries of the EaP, namely Georgia, Moldova and Ukraine. This means that the three countries and the EU have allowed access to their markets in particular areas, whereby investors on each territory are conditioned upon the same regulations (e.g. Ministry of Economy and Sustainable Development of Georgia (MoESD), 2016; Hellyer & Pyatnisky, 2013; van der Loo, 2017; European Commission, Moldova, 2018).

(12)

12

3. Theoretical framework

This chapter will firstly conceptualize EU conditionality, democracy and democratic development, focusing on providing definitions of the concepts, what they entail and under what conditions they are applicable. Secondly, the thesis will discuss the relevant literature on the effectiveness of the EU’s political conditionality, whereby the existing theories on the topic will be explored. Finally, drawing from the theories of the external incentives model and the social learning model, expectations about the cases will be presented. These expectations will be used to scrutinize the impact of EU conditionality on the democratic development of the countries of Serbia, North Macedonia, Albania, Georgia and Moldova.

3.1 Defining EU conditionality, democracy and democratic development

Conditionality as defined by Oxford Public International Law (2013, para 1) describes “the practice of international organizations and states of making aid and co-operation agreements with recipient states conditional upon the observance of various requirements, such as financial stability, good governance, respect for human rights, democracy, peace and security”. Aside from the EU, there are also other international organizations which apply the principle of conditionality, for example, the International Monetary Fund (IMF) and the World Bank. There is some difference between the definitions that the various institutions provide. For example, the IMF denotes conditionality when a state borrows from the fund and its government agrees to amend its fiscal policies to resolve the issues that led to seeking financial help. The amendments of these policies function as conditions for obtaining the IMF loans and ensure repayment from the side of the receiver (IMF, 2019). In its 2005 Enlargement strategy paper, the European Commission sets out three important principles for the EU’s approach towards enlargement, namely consolidation, conditionality and communication (European Commission, 2005). The Commission (2005, p. 3) notes that the application “fair and rigorous conditionality” is necessary and the EU shall be firm in its demands towards fulfilment of requirements and fair in the respective rewarding. Thereby, candidate countries can advance to the next stage of the process only if they satisfied the required criteria. Additionally, the report explains that if there a severe breach in the process and EU principles are impeded, then the Commission will act accordingly and will sanction the applicant (European Commission, 2005). If the two ways in which conditionality is described are compared, it can be seen that the IMF discusses economic conditionality, while

(13)

13 the Commission refers to conditionality which can be both economic and political. Hughes et al. (2004, pp. 3-4) conceptualize conditionality as the interaction of various actors, their interests and opinions, who receive certain rewards or sanctions when their government meets or disobeys the conditions set by the donor. Furthermore, Hughes et al. (2004, p. 14) explain that conditionality represents a sort of power asymmetry, i.e. there is an actor who dominates over other actors by setting the conditions, monitoring compliance and providing the rewards or sanctions, and there are subordinate actors who need to comply with the conditions to receive the end reward. Schimmelfennig (2007, p. 127) explains that in the case of European regional organizations and inability of the targeted party to meet the conditions, the donor organization usually withholds the reward, thus it does not force or support (non-) compliance. Considering the abovementioned conceptualizations, this thesis will refer to EU conditionality as the conditions which are set by the European Union for states aspiring to become EU members and need to be fulfilled by these states in order to receive the promised reward.

Democracy is a rather difficult concept for providing a definition to because it can mean different things in different situations and countries. There are also various explanations of the term by the different scholars. For example, Dahl (1989) presents democracy as a “process of making collective and binding decisions”. Furthermore, he identifies five features which describe the democratic process, namely effective political participation, voting equality, understanding of the citizens about the decision-making matters, control on the part of the society as to what gets to the political agenda and inclusion. Additionally, Schumpeter (1962) and Bohman (2008) refer to democracy in a similar way by describing it as an institutional arrangement or procedures which lead to individuals having the power to choose or change representatives and terms that are decisive for their life as a group. However, when considering the EU and especially the EU’s conditionality in the Enlargement and ENP policy, the EU clearly links democratic development to the role of political parties, electoral participation, rule of law, freedom of press, a powerful and active civil society and protection of minority rights (e.g. European Commission, 2014; Wider Europe, 2003). In its strategy papers, the European Commission focuses on specific democratic values such as the rule of law, democratic and well-functioning institutions, fight against corruption, freedom of expression, civic education and human rights. As a consequence, the thesis will look at the extent to which there is a transparent and competitive political process, the implementation and effectiveness of reforms related to the rule of law, control of corruption and freedom of

(14)

14 expression to assess the degree to which the EU has impacted the democratic development of the case countries. Details about how these concepts will be measured will be discussed in the following chapter.

3.2 Literature review

There is an extensive existing work on the factors which encourage or reduce compliance with EU conditionality and criteria. For example, presumably the most cited authors when it comes to Central and Eastern EU Enlargement, namely Schimmelfennig and Sedelmeier (2004, 2005), propose an external incentives model which attempts to explain the situations in which states are predisposed to adopt EU rules and the situations in which state actors are constrained or unwilling to comply with the EU requirements. Schimmelfenning and Sedelmeier (2004) argue that the external incentives model of governance best elucidates the adoption of rules set by the EU in the CEECs applicants. Moreover, they contend that the effectiveness of political conditionality depends on 1) the determinacy of conditions, 2) the credibility of EU conditionality, 3) the size, kind and speed of rewards, and 4) the domestic compliance costs for the countries’ governments (Schimmelfenning and Sedelmeier, 2004; Schimmelfenning, 2008). Furthermore, the external incentives model relies on the rational choice theory and is presented as a bargaining model, which assumes that individual actors are utility-maximizers concerned with the growth of their own power (Schimmelfennig and Sedelmeier, 2004). They are highly self-interested and will make cost-benefit calculations to determine the most beneficial course of action for themselves (Shepsle, 2008).

Vachudova (2005) argues that countries vary in their intention to comply with EU criteria and that the domestic costs of compliance are different for each applicant state. She explains this variance by the limited political competition, high levels of corruption in the economy, as well as ethnic nationalism to uphold and succeed political power in less democratic states. Vachudova (2005, p. 72) contends that while in liberal democracies governing elites are not contingent upon these tactics to gain electoral votes, in more fragile or “less” democratic countries, political leaders utilize all of the abovementioned strategies to a certain degree. This means that fulfilling the EU requirements might reduce the power of the governing party and strengthen oppositions, decrease the possibilities for rent-seeking and for winning over ethnic support in elections (Vachudova, 2005, p. 73). Unsurprisingly, in the eyes of the ruling parties this would bring higher domestic costs, as they fear losing their power and control.

(15)

15 However, Vachudova (2005) reports that in the case of the CEECs, political leaders had no choice but to follow the EU recommendations and criteria since the prospective EU membership was very popular amongst the electorate, offered high financial rewards and opportunities for economic growth. She (2005, p. 258) asserts that the EU leverage impacts domestic groups and that in the case of less democratic states, the EU influences the institutional environment by supporting the creation of a stronger opposition and more open and competitive political field.

Similarly, Noutcheva and Aydin-Düzgit (2012) examine the effect of the EU incentives on the rule of law in a comparative study of the Western Balkans and Turkey, arguing that the external incentives model might not be as effective in cases when domestic political leaders’ interests are only partially or not aligned at all with the EU criteria. They explain that in such situations the domestic costs become too high and reforms will be inconsistent or unreliable if there are any at all. A different approach in analyzing the effectiveness of EU conditionality was presented by Steunenberg and Dimitrova (2007) who argue that EU conditionality becomes less effective after the announcement of the accession date for the country. This is because candidates are not subject the negotiations anymore which leads the slowing down of new reforms. Additionally, Steunenberg and Dimitrova (2007) note that to reduce ineffectiveness, the EU should reveal the accession date just before the official accession.

Kelley (2004) studies the impact of European institutions on states’ national policies by comparing rational choice and socialization mechanisms, e.g. membership conditionality and socialization-based efforts, such as social influence and persuasion. Likewise Schimmelfennig and Sedelmeier (2004), she argues that the incentive-based approach is more effective for influencing domestic policies and bringing for change. On the other hand, she claims that socialization-based efforts can guide the changes. However, in order for socialization-based efforts to have an impact on policies, government oppositions need to be weak (Kelley, 2004). Furthermore, Kelley (2004, p. 434) adds that the readiness to compromise is contingent upon the extent to which political leaders want the reward. Hence, the necessity of the EU to utilize conditionality increases with stronger domestic opposition.

Scholarly debates linked to abovementioned notion have been distinguished between two main strands: on the one hand, there are authors arguing that the EU constitutes a normative

(16)

16 power (e.g. Manners, 2002) and on the other, there are authors arguing that the EU constitutes a strategic power (e.g. Youngs, 2002; Hyde-Price, 2006). For example, Noutcheva (2009) in her study on the Balkans states that the EU’s policy lacks a considerable normative power in the Western Balkans. This in turn influences the degree to which countries comply with the EU criteria on matters concerned with sovereignty and results in applicants responding with “fake, partial and imposed compliance” (Noutcheva, 2009, p. 1067). Nonetheless, Heather Grabbe (2006) who looked both at rational and constructivist mechanisms used by the EU to affect CEE candidates to change, concluded that precisely soft power brought for stronger compliance. Also, in an article about the lessons learned, Grabbe (2014) argues that the 2004 Enlargement was successful and presented long-lasting effects not only because of rational incentives and sanctions but also because of normative power, persuasion and social learning.

Comparing the ENP to the EU enlargement conditionality, scholars describe the ENP’s leverage to be relatively weaker mainly due to the lacking membership perspective (e.g. Sasse, 2009; Wolczuk, 2007). Sasse (2009, p. 371) notes that the EU conditionality in the ENP is “vague and in flux”. Sasse (2008, p. 296) defines conditionality in the ENP as “conditionality-lite” precisely because incentives and enforcement structures are more vague and unclear for both parties. In this sense, the ENP conditionality is seen as a main reference point which ruling elites in neighboring partner countries can refer to when it fits their political agenda. Secondly, conditionality functions as a loose form of socialization both for the EU and the ENP countries. In his study on the EU’s impact on democracy and human rights in ENP countries, Schimmelfenning (2005) argues that the ENP conditionality is ineffective predominantly because of weak EU incentives and high domestic costs for the governments of the semi-authoritarian or semi-democratic states. Sasse (2008, p. 303) criticizes Schimmelfennig and Sedelmeier’s external incentives model when applied to ENP countries, stating that incentives, credibility and adoption costs in these countries are better to be described as vague than weak or high. She (2008, p. 303) agrees with Vachudova (2005) that it is more important to see whether political opposition, certain elites or citizens will mobilize against national veto players as result of the EU influences. On this notion, Sasse (2009) states that mobilization and socialization in the ENP can also happen consecutively, i.e. socialization might lead to mobilization and vice versa.

(17)

17 The existing literature shows that the majority of scholars have been positive about the effectiveness of EU conditionality in the CEECs. Nonetheless, the same effect has not been observed in the case of the Western Balkan candidates. In relation to the EU conditionality in ENP countries, authors have been generally critical about its effectiveness, especially when compared to the enlargement policy (e.g. Börzel & Schimmelfennig, 2017; Schimmelfenning, 2005). For instance, Börzel and Schimmelfennig (2017, p. 284) argue that the EaP countries “have not made any sustainable progress towards democracy”.

Given the rather negative views on progress in democratic development both in the EaP and the Western Balkans, this thesis aims to compare the effects of EU conditionality on democracy in certain countries of the Balkans and the EaP. The thesis’s objective is to see whether the progress (if any) in the EaP case countries in democratic development can be comparable to this in the Western Balkans, in terms of the vaguer conditionality and socialization-related method for encouraging change. To assess whether the selected case countries support Börzel and Schimmelfennig’s (2017) assumption that states with weaker EU incentives do not undergo as quick and effective democratic changes, the work will utilize Schimmelfennig and Sedelmeier’s external incentive and social learning model. This theoretical framework seems most appropriate for comparing countries that are subject to the Enlargement and Neighborhood policy. Firstly, this is because the external incentive model includes the condition for size of rewards. Hence, the model represents a basis for illustrating whether the size of rewards is indeed a factor for compliance. Secondly, the incorporation of the social-learning model includes the possibility that socialization-based mechanisms account for the impact on governmental decisions towards further EU integration as suggested by Kelley (2004). The following sub-section will explain the two models and suggest expectations for analysis.

3.3 External incentives model

The external incentives model is a bargaining model which relies on rational choice theory and assumes that individual actors are highly self-interested utility maximizers. They will make cost-benefit calculations to determine the most beneficial course of action for themselves (Shepsle, 2008). The model explains external governance basing it on the principle of conditionality, which in this case means that the EU is the institution presented as a rule of the game which defines the set of players and the rules or conditions that each state

(18)

18 has to fulfil in order to receive the final reward. In other words, the EU acts as the donor and determines the conditions that the candidate and neighboring countries need to comply with to obtain the reward – for example, a membership or a trade or association agreement. If the target country is unable to meet the requirements, then the EU withholds the reward. Schimmelfennig and Sedelmeier (2004, p. 664) describe the bargaining process as a ‘domestic status quo’ or ‘domestic equilibrium’ which is disturbed by EU conditionality, by the provision of incentives for the compliance with EU rules. In this sense, conditionality influences the recipient’s government either through bargaining or through empowering national actors which favor the adoption of the rules. Since rule adoption ultimately lies in the hands of the government, it starts making cost-benefits calculations about the most beneficial action for itself given all external and domestic pressures. According to Schimmelfennig and Sedelmeier (2004), these calculations are contingent upon the clarity of conditions, the size and speed of rewards, as well as the credibility of rewards or sanctions and the size of the adoption costs. A more common assumption of the external incentives model is that “a state adopts EU rules if the benefits of EU rewards exceed the domestic adoption costs” (Schimmelfennig & Sedelmeier, 2004, p. 664).

Considering the above stated citation, the reward of membership to a candidate country shall be greater and more beneficial both in short and long-term, providing greater political and economic assistance compared to trade and association agreements with neighboring countries. Also, Schimmelfennig (2008) notes that non-financial incentives based on social learning and persuasion are too weak to exceed the domestic costs of a government. Such non-material incentives can for example be the offer of expertise from EU institution or certain technical assistance. Schimmelfenning et al. (2006) even argue that financial aid or association agreements that do not include a promise of EU membership are not powerful enough to overcome the costs for adoption. As a consequence, the first expectation is as following:

E1: The larger the EU reward for the candidate or neighboring country, the greater will be the progress in democratic development for the country.

The second expectation concerns the domestic costs a government may experience when having to adopt a EU rule. Depending on the size of the domestic costs, a state’s government decides whether to comply with the requirement or not. Domestic, or adoption costs relate to

(19)

19 the power costs of the states’ governments which either increase or decrease given the specific EU conditions for compliance (Schimmelfennig, 2008). For instance, if domestic costs in a certain way threaten the authority and power, sovereignty or integrity of the state, then these costs for adoption of EU rules will be very high. According to Schimmelfennig and Sedelmeier (2004, p. 666) domestic costs are always present in the adoption of EU rules; if not the rules would have been implemented without the need of a conditionality. They present two possible implications of domestic costs: for example, domestic costs can act as opportunity costs i.e. losing other alternative rewards by choosing the reward and rule of the EU or creating “welfare or power costs for private and public actors” (Schimmelfennig & Sedelmeier, 2004, p. 666). However, while domestic actors are making cost-benefits analyses, they are also taking into account the benefits that the EU reward may hold. Thus, if the costs are higher than the benefits offered with the reward, then the chance that the government adopts the rules is significantly lower. The second expectation is:

E2: The lower the domestic costs in a candidate or neighboring country for the adoption of EU rules, the greater will be the progress in democratic development for the country.

3.4 Social-learning model

Alternative explanation about the extent to which EU conditionality is effective is provided by the social influence model. This model is based on the logic of appropriateness and assumes that actors make decisions based on internalized rules, values and norms, i.e. they take into account the interest of the institution juxtaposed to their self-interest (March & Olsen, 1983). According to this logic, governments will then adopt EU laws if they are convinced of the legitimacy of the rules, values and norms of the EU (Schimmelfinning & Sedelmeier, 2004). In this standpoint, the EU acts as a formal institution for the European society which can be distinguished by a particular shared identity and a particular set of values and principles. States that aspire to the EU and identify themselves with EU are more likely to believe in the legitimacy of the EU values and see these values, rules and principles as fair, righteous and appropriate. As a consequence, such states are more likely to be committed and comply with EU rules. Hence, the third expectation is:

E3: The stronger the commitment of the government and society in the candidate or neighboring country to the EU, the greater will be the progress in democratic development for the country.

(20)

20 The last expectation regards the exchange of goods and services with the EU, or the economic interdependence of the aspiring state and the EU. Schimmelfenning et al. (2003, p. 501) suggest that the “opportunity costs of non-accession will mobilize societal actors most forcefully in favour of EU conditions”. Thus, it is assumed that countries with higher economic interdependence with the EU will experience higher societal mobilization. Additionally, some of the countries might also have high interdependence with Russia. Consequently, it is interesting to test the following expectation:

E4: The higher the economic interdependence between the candidate or neighboring country and the EU, the greater will be the progress in democratic development for the country.

3.5 Conclusion

The chapter provides a definition of the key concepts of EU conditionality, democracy and democratic development. Additionally, the existing literature on the topic under study is discussed. The thesis will borrow from Schimmelfennig and Sedelmeier (2004, and 2005)’s theory to assess whether the expectations are present in the case countries. The expectations make for a basis of comparison between the case countries and help to give an answer to the research question. They are based on four elements discussed by Schimmelfennig et al. (2003) and Schimmelfennig and Sedelmeier (2004, and 2005). These are the size of rewards, domestic costs in the adoption of EU rules, governmental and societal commitment in the candidate or neighboring country and economic interdependence. Since the thesis is comparing countries which are given a promise of EU membership with countries which are currently denied such possibility, the size of the reward – being a membership or an agreement – is an important characteristic. Additionally, based on the results of Schimmelfennig et al. (2003), the size of incentives and domestic adoption costs are likely to be the two leading conditions for the effectiveness of EU conditionality. For this reason, this research focuses exclusively on them. The other two elements which are derived from the social-learning model and also utilized by Schimmelfennig et al. (2003) are governmental and societal commitment and economic interdependence. Here, it will be interesting to see whether the external incentives model will hold in the case of the countries of the Neighborhood given the higher level of socialization in the ENP. The social-learning model is included in the analysis due to the likelihood that a combination of both models will better

(21)

21 represent the situation of the case countries both in the ENP and the Western Balkans. The following chapter will discuss the research design of the thesis, including both data collection and data analysis.

(22)

22

4. Methodology

This chapter will firstly specify the design of the study and justify the choices made behind the selection of cases. Then, it will provide the operationalization of concepts of the main explanatory factors, namely, EU conditionality and progress in democratic development. Afterwards, it will cover how the data will be collected and analyzed, as well as what are the main drawbacks that can be expected of the research design. The chapter will conclude with final remarks.

4.1 Research design

The research design of this study is a qualitative comparative case study. Ragin (1987, p. 3) explains that researchers who undertake qualitative studies look at their cases as wholes i.e. they compare configurations – a combination of elements between the cases. This distinguishes them from quantitative studies where the emphasis is much more on the variation in the variables, having the end objective to generalize results on the entire population. Reflecting on the similarities between the case countries that are investigated and the different outcomes in their progress towards democracy and the EU, a most similar systems design is utilized (MSSD). The MSSD, introduced by Mill in 1843, aims at finding the most important characteristics which are diverse among a number of similar states and maintains that these differences account for the final outcome (Landman, 2008). In simpler words, it is the presence or absence of a certain key characteristic that brings for the difference in the result. For this method of design to work, the similarities between the case countries have to be held constant – only then one can see the variance in the result and in components that account for the result. According to Landman (2008, p. 71), researchers who are investigating cases in a particular area, such as Europe, Asia, etc. are often relying on the MSSD because of the similarities that the cases share with each other – for example, culture, history, a form of government. This correspondents to the motivation behind this thesis to account for the progress in democratic development in two European regions which are similar to each other in several factors listed in the next sub-section.

4.2 Case selection

The case countries which have been selected for the study are five: the western Balkan countries of Serbia, North Macedonia and Albania, which are given a candidate status for EU

(23)

23 membership, and the two European Eastern Neighborhood of Georgia and Moldova which have signed Association Agreements with the EU and are part of the Deep and Comprehensive Free Trade Area (DCFTA). The major reason for the case selection is the interest for the effectiveness of EU conditionality and specifically whether softer conditionality can make for a similar impact in the targeted by the EU governments compared to hard conditionality. All of the case countries are aspiring to join the Union, and subsequently all of them are attempting to approximate or harmonize their laws to the EU. However, only the Western Balkan countries is given the promise of future possibility for EU membership. The countries of the Western Balkans with a EU membership perspective are Albania, North Macedonia, Montenegro, Serbia, Bosnia and Herzegovina and Kosovo (European Commission, 2019a). Since the countries that are already candidates in the process share greater political and economic similarities with the DCFTA countries, Bosnia and Herzegovina and Kosovo have been considered as irrelevant for the study. Additionally, Serbia has shown higher variation in key characteristics, and thus it was preferred for analysis over Montenegro. Lastly, from the countries of the DCFTA, namely Georgia, Moldova and Ukraine, the latter was omitted due to the difficulties to control for the country’s recent crisis. All of the case countries were selected on the independent variable, while aiming for variation in it. The features that are shared between the states are listed below.

- All states are aspiring to become EU members;

- All states have the same government type, i.e. parliamentary republic;

- All states have recently declared independence with the exception of Albania, which has a longer history of sovereignty compared to the rest of the countries; this will be taken into consideration in the analysis.

- According to the World Bank (2019), all of the case countries in the Western Balkans are described as upper-middle income (3,896-12,055) economies based on gross national income (GNI per capita US$), while all of the case countries from the EaP are considered as lower-middle income (996-3,895) economies.

Serbia North

Macedonia

Albania Georgia Moldova

Independence 1992, SFRY 2006 with Montenegro 1991 1912 1991 reestablishment of independence from 1918 1992

(24)

24 Candidate status or AA 2009 applied; 2012 2004 applied; 2005 2009 applied; 2014 Signed in 2014; 2016 AA Signed in 2014; 2016 AA Government type Parliamentary republic Parliamentary republic Parliamentary republic Parliamentary republic Parliamentary republic GDP, World Bank 15,431 15,290 12,943 10,674 5,710 GNI, World Bank, 2019 5,180 4,880 4,320 3,770 2,200 Unemployment IMF 2019; World Bank, 2018 13.4 13.5 18.9 21.6 13.7 13.9 No data 14.1 4.0 3.4 Table 1: Similarities between the case countries

4.3 Data collection and analysis

Democratic development will be measured by the utilization of data from three sources, namely, the World Bank, Freedom House and Bertelmann stiftung. The worldwide governance data from 2017 will be considered in the case of the World Bank. This are the most recent statistics on governance indicators presented by the World Bank. The democratic total score will be calculated by the scores derived from all six indicators that the World Bank uses – voice and accountability, political stability and absence of violence/terrorism, government effectiveness, regulatory quality, rule of law and control of corruption. The total score will be estimated by summing the scores of all aforementioned indicators and then dividing them by six for each case country. The period in which the calculation will be undertaken will be 2009 – 2017. The starting point has been chosen based on the application of the Western Balkan states for membership, as well as the launch of the Eastern Partnership in 2009. Only North Macedonia applied earlier compared to the rest of the countries, however due to the problem with the name, it is assumed that there has not been much progress between its application and 2009. The end point of analysis is 2017 because the World Bank’s project captures the period from 1996 only until 2017. The scores vary from 0 until 100, meaning that the higher the score, the higher will the country score on democracy.

The thesis will also rely upon democracy scores provided by Freedom House’s latest Nations in Transit project (2018). The data is more recent and will contribute to the validity of the results. The indicators here are divided into freedom, civil liberties and political rights

(25)

25 indicators. The scores range from 1 until 7, with 1 being the highest and 7 the lowest (Freedom House, 2018). In cases where there are larger discrepancies between the advancements of the countries between the Freedom House and the World Bank, the country reports provided by Bertelsmann Stiftung will be used as a control indicator. The foundation developed a transformation index (BTI) which assesses the democratic development of countries in transition according to five criteria – stateness, political participation, rule of law, stability of institutions, political and social integration (Bertelsmann Stiftung, methodology, 2019).

The key characteristics for the external incentives model are size of the reward and domestic costs. The size of the reward will simply be measured by looking at the ultimate promised reward by the EU for the countries’ compliance with conditionality. The reward can take different forms, for example, economic or financial assistance and agreements, official institutional association agreements or membership (Schimmelfenning, 2008). Hereby, membership is considered to be the largest reward which allows the country to fully benefit from joining the EU; second will come the association agreements since they can provide countries with certain advantages, such as free movement of people, access to market, visa liberalization, etc.; the smallest reward would be some kind of financial or other non-material assistance to the country.

Based on what domestic costs entail, they could have been measured by the case countries’ democracy level. Since scholars argue that domestic costs are higher for authoritarian or semi-authoritarian regimes, states that are more advanced in democracy shall have lower domestic costs (Schimmelfennig and Sedelmeier, 2005). However, since this research evaluates democratic progress as part of the dependent variable, two other measures are regarded as more suitable. Firstly, the domestic costs for a country’s government will be measured by looking published European Commission progress reports for each country for the years of 2009, 2014 and 2018. Hereby, the three years are chosen based on the start and end date of the period for analysis. Furthermore, 2014 is selected since this year marks the beginning of the new framework for the EU’s financial and technical assistance in both regions through the Instrument for Pre-accession assistance and the European Neighborhood Instrument. Thus, the overall countries’ results in this year are expected to be higher. When assessing each country’s report, the one which perform better in terms of political integration to the EU and has less problematic areas to tackle with will be regarded to have lower

(26)

26 domestic costs. Secondly, an additional measure will compare the level of corruption in each state in the timeframe between 2009 and 2018. Here, the thesis will rely on data made publicly available by Transparency International. The country with lowest scores in corruption will be considered as experiencing lower domestic costs for rule adoption. This is based on the notion presented by Schimmelfennig (2004, p. 5) that liberal democratic norms, which result from political conditionality, generally restrict the power of governments. Such norms forbid undemocratic measures which governments may undertake to control for their power and autonomy. Examples are restricting media freedom, rigging election votes, suppressing opposition and civil society organizations. As a consequence, higher levels of corruption will most presumably facilitate such practices. Then, in cases in which corruption is greatly prevalent, domestic costs are expected to be high as well.

The key features for the social-learning model are governmental and societal commitment and economic interdependence. Governmental commitment will be measured by examining two factors: the ruling party’s orientation (e.g. whether it is pro-European or not) and the country’s membership in other international organizations. These two measures are borrowed from Schimmelfennig et al. (2003). The ruling party’s orientation will be assessed based on the information found in its official website. Furthermore, the organizations under focus are the United Nations (UN), the North Atlantic Treaty Organization (NATO) and the World Trade Organization (WTO). The first two organizations have been chosen based on their missions for preserving peace and security, while the latter aims at contributing to fairer and more transparent conditions for trade. These missions correspond to the EU’s values and the case countries’ involvement in them shows to an extent the readiness of key political figures to adopt pro-western and European views. The official web sites of each organization will be used to account for the year of the country’s accession. The countries which do not participate or accessed the latest will be regarded as less committed to joining the EU. Societal commitment will be measured by the help of two external surveys. The surveys gather citizens’ opinion on the EU potential membership of the countries. A comparable measure was utilized by Schimmelfennig et al. (2003). For Serbia, North Macedonia and Albania, it will be considered the ‘Balkan Barometer’ published by the Regional Cooperation Council. For Georgia and Moldova, the ‘EU neighbours east’ will be used. Both projects are published in 2017 and funded by the EU. This measure is borrowed from the study of Schimmelfennig et el. (2003) on Latvia, Slovakia and Turkey. Nonetheless, the authors

(27)

27 utilize data from Central and Eastern Eurobarometer from 1998. However, since this is outdated evidence, the thesis will rely on two abovementioned external surveys.

Economic interdependence between the EU and the case countries will be estimated based on two measures: the share of the imports and exports with the EU compared to the total share, and the value of balance between imports and exports with EU. Such measures are often used in studies for the assessment of economic or trade dependency or transactional exchanges (e.g. Schimmelfennig and Scholtz, 2008; Schimmelfennig et al., 2003). Firstly, it will be compared what is the share of trade between the EU and the country at hand to the share of trade between the country and the world. Secondly, it will be looked at the balance of imports and exports between the EU and countries. Countries with lower or negative balance of imports and exports will be more dependent on trade with the EU. The data will be gathered from statistics on trade published on the web site of the European Commission.

4.4 Limitations

As already indicated this research is a small-N comparative study which is based on Mill’s method of difference or the MSSD. One of the main limitations when it comes to small-N comparative study is the external validity of the research. This means that because small-N research is concentrated on a limited number of cases, it is difficult to make inferences about the larger population (Toshkov, 2016). Generalizations can only be made for countries or other units of analysis which share the same key characteristics as the selected cases. Hence, inferences about components outside the context of analysis are invalid. However, this research’s objective is not to generalize to different and greater number of units but to study the relationship between the effectiveness of EU conditionality and democracy and to assess whether the case countries are illustrations of the theory offered by Schimmelfennig and Sedelmeier (2004, and 2005).

Another threat that arises when conducting small-N studies, while utilizing the MSSD is the problem of spuriousness. Spuriousness, also described as omitted variable bias, refers to the notion that an “unidentified factor is responsible for the outcome, while the identified factor is mistakenly attributed to having an effect on the outcome” (Landmand, 2008, p. 40). It may not only be that the outcome is dependable on this factor but also the other key explanations in the research. Lieberson (1991) also states that MSSD has difficulties in identifying

(28)

28 possible interaction effects between two or more explanatory factors i.e. it might be the combination of several explanatory factors that lead to the outcome. As a solution to the problem Landman (2008, p. 41) proposes the identification all possible relevant factors that might influence the outcome. However, in the comparison of only few countries, as in this case, it is very difficult to account for all potential factors that might have an impact on the result. According to Landman (2008, p. 41) the additional specification of factors that can possibly impact the outcome can lead to another issue, namely to have too many factors but too less countries to account for those factors. Additionally, the incorporation of more cases for investigation can be restricted by time or other material or non-material resources for conducting the research. Another answer to the problem would be to select cases which correspond to the conditions of the theory. Nonetheless, this would result in selection bias. As a consequence, Landman (2008, p. 41) states that “the comparativist is forced to recognize these various trade-offs while maximizing the types of inferences that can be made given the countries and the evidence in the study”. Presumably the greatest limitation of this study is exactly the threat of an omitted variable bias since not all potential factors have been accounted for. The provided timeframe for the thesis did not allow for the inclusion of additional factors for analysis. Furthermore, the results of the comparative study cannot be as accurate as for example an experimental study. This is because it is rarely possible to find two or more countries which are indistinguishable to each other (Lieberson, 1991). For instance, there are minor differences in the numbers corresponding to the GDP and GNI of the case countries, which have to be acknowledged.

Landman (2008) also identifies case selection as one of the main drawbacks of small-N studies. He asserts that if cases are selected randomly or intentionally without respecting established rules, then the research can lead to invalid inferences and wrong assumptions about the topic under investigation. For example, if cases are chosen randomly, it might be that all of the cases support the theory and cases which are in contrast are disregarded, thus the theory is viewed as reliable. The problem of selection bias also occurs if cases are selected on the dependent variable. This can result in the belief that some explanatory factors are more important than others and to ignore their significance (Landman, 2008). In the context of selection bias, the second sub-section of the methodology of this research aimed at justifying as precisely as possible the decisions behind case selection.

(29)

29 There are some additional pitfalls when considering the validity of measures used for the research. Firstly, regarding the measures of size of rewards, it does not consider the overall financial incentives offered to the target countries which might have influence on the cost-benefit calculations of domestic actors. Secondly, while the research will reflect on progress reports of the countries to depict any domestic costs, it merely does so from the perspective of the European Commission which might be inconsistent with the standpoint of governmental actors. Thirdly, the assessment of governmental commitment to the EU might prove to be difficult to evaluate, if for instance, ruling parties portray themselves as pro-European but their practices show otherwise. Fourthly, given that societal commitment is measured by two different external sources, there might be certain discrepancies between the results. Lastly, the two measures related to trade in the assessment of economic interdependence do not include other potential economic factors which might add to the level of dependency.

4.5 Conclusion

This chapter provided insights into the research design and methods of the study. Accordingly, the thesis uses a most similar systems design through which it compares the selected case countries, namely Serbia, North Macedonia, Albania, Georgia and Moldova. The case selection sub-section explains why these countries have been preferred over others and outlines the major similarities that have been found between them. The data collection and analysis sub-sections show how is the data collected and how are important key concepts operationalized. There are a number of data sources that will be regarded when measuring the concepts. These are: World Bank’s statistics on governance, the Bertelsmann Stiftung’s and Freedom House’s evaluations on the degree of democracy, the European Commission’s progress reports on the countries under investigation, Transparency International’s data on corruption, the official web sites of the states’ ruling parties, as well as the official web sites of several international organizations, the projects of ‘Balkan Barometer’ and ‘EU neighbours east’ and the European Commission’s records on trade. The next chapter will discuss the findings of the study.

(30)

30

5. Results

This chapter will present the necessary findings for the assessment of EU conditionality factors and the progress in democratic development in the case countries. Firstly, each country’s democratic development is described considering the period of investigation 2009-2018. Hereby, the thesis relies on academic literature and reports by the Freedom House and Bertelsmann Stiftung for the above stated period. Nevertheless, each description will start with short background information about the country’s beginning in the transition towards democracy. Hence, this first section reveals information about the process in democratic development. It constitutes the basis for understanding the outcome in democratic development since 2009 in each country. Secondly, all of the four expectations will be discussed, namely the size of rewards, domestic costs, political identification and societal commitment and economic interdependence with the EU. The reported data will be divided per expectation, specifying the country with highest scores given the particular observation. 5.1 The progress in democratic development

Serbia

The beginning of the transition period towards democracy in Serbia started in 2000. According to the Freedom House, Serbia’s democratic transition took place much later compared to many other post-Communist states (Freedom House, Nations in Transit, 2007). The period was marked with massive anti-Miloševič protests when he was finally ousted in the presidential elections in October the same year (Bieber, 2003; Zakoek, 2008). Consequently, a considerable number of the previously leading political elites were replaced and there was a re-establishment of the state institutions in the country (Ördögh, 2017). In 2006, the country introduced a new democratic constitution since the old was largely favoring the authoritarian power of the president (Ördögh, 2017). 2006 was a rather difficult year for the state not only because of the political struggle for new constitution but also because of the dissolution of the Union with Montenegro, international pressures for cooperation with the ICTY, the question with the status of Kosovo and the status of Serbia in regard to the Stabilization and Association Agreement with the EU (Freedom House, 2007). Negotiations between the EU and Serbia were even stalled due to lack of cooperation with the ICTY on the side of Serbia (Freedom House, 2007).

The effect of EU conditionality on the democratic development in the Western Balkans and the European Eastern Neighborhood: The cases of Serbia, North Macedonia, Albania, Georgia and Moldova