Changes in the political criteria of the enlargement process of the European Union

(1)

Changes in the political criteria of the

enlargement process of the European Union

A comparison between the accession process of Bulgaria

and Romania and the accession process of Croatia

MA Thesis Sanne Sterenborg

Parelmoerhorst 184, The Hague S1724215

0638742225 Supervisor

(2)

2

Appendix 3 Declaration by candidate

DECLARATION BY CANDIDATE

I hereby declare that this thesis, ―Changes in the political criteria of the enlargement process of the European Union. A comparison between the accession process of Bulgaria and

Romania and the accession process of Croatia’’, is my own work and my own effort and that it has not been accepted anywhere else for the award of any other degree or diploma. Where sources of information have been used, they have been acknowledged.

Name: S.M. Sterenborg Signature:

(3)

3

1. Introduction

After the Second World War the European Coal and Steel Community (ECSC) was founded to secure peace throughout Europe. Although the ECSC evolved over the years into the European Union (EU) the founding principles of democracy, human rights and the rule of law never lost in importance. Concerning these principles a new challenge arose when ten Central and Eastern European States (CEES) applied for membership in the 1990s. Most of the new applicants were a long way from meeting the standards of the EU. To guarantee that all the applicants would meet these standards the Union introduced the Copenhagen Criteria, a first step in the development of its enlargement strategy. The first of the Copenhagen criteria required that future member states had ‘’stable institutions guaranteeing democracy, the rule of law, human rights and respect for and protection of minorities’’, therewith emphasizing the importance of the Union’s founding principles (European Commission 2012).1

The Union’s enlargement strategy led to the accession of the first group of CEES in 2004. Although Bulgaria and Romania started the accession process at the same time as these countries they were not ready to join the Union at the same time.2 During the accession process both countries experienced problems with the implementation of the political criteria. Despite the troublesome accession process of Bulgaria and Romania the enlargement of the Union continued when the accession negotiations with Croatia started in 2005. Considering the troublesome accession process of Bulgaria and Romania, especially concerning the political criteria, the Union was faced with the challenge to prevent this from happening in the accession process of Croatia.

Research question

Faced with the accession of Croatia the Union thus had to rethink its enlargement strategy, especially concerning the political criteria. Was it necessary to adjust the strategy, and if so which elements of the strategy should be adjusted and how? In this thesis I look into the accession process and the changes herein from the side of the EU and therewith answer the following question: How has the EU changed the process of enlargement in the field of

democracy, human rights and the rule of law for the accession of Croatia in comparison to the 2007 enlargement round with Bulgaria and Romania?

(6)

6 In order to answer the research question I have formulated five sub questions, each of which is answered in a separate chapter. The first chapter focuses on the theoretical framework by answering the question ‘’how can the combination of the external incentives model and the social learning model explain the changes in the enlargement process?’’ The answer to this question is the basis for the analysis that follows in the next chapters. In the second chapter I define the most important values of the EU in the political criteria which form the common ground in the comparison between the accession process of Bulgaria and Romania and the accession process of Croatia. Herewith I answer the question ‘’what do the values of democracy, human rights and the rule of law entail for the Union and how has this changed between the accession of Bulgaria and Romania and the accession of Croatia?’’.

The subsequent chapters focus on the empirical analysis of the accession process. The third chapter elaborates upon the workings of the enlargement process and answers the question ‘’how does the enlargement process work?’’. The fourth chapter then focuses on the accession process of Bulgaria and Romania and answers the question ‘’what was the approach of the EU in the accession process of Bulgaria and Romania and how is this reflected throughout the process?’’ Bulgaria and Romania are grouped together here because the EU applied the same enlargement strategy towards both countries. The focus of this chapter will therefore be on the general aspects of the accession process both countries went through. Nevertheless, the countries differ from each other concerning their domestic politics. As a consequence it is in some cases important to discuss the domestic reactions of both countries separately. The fifth chapter touches upon the accession process of Croatia and answers the question ‘’what was the approach of the EU in the accession process of Croatia and how has this changed in comparison to the accession process of Bulgaria and Romania?’’. By combining the answers to these sub questions it is possible to answer the research question.

Social and scientific significance of the research question

(7)

7 change of the political accession criteria regarding democracy, human rights and the rule of law can contribute to the knowledge about the accession process of the EU. This can result in a more effective negotiation and accession policy and better safeguards for compliance in the areas so dire to the Union’s identity. On the basis of these results the EU can make significant changes in the accession criteria between enlargement rounds which can be of positive influence on the countries that are currently in the process of joining the EU. Also it can be relevant in their preparation on, and focus throughout, the negotiation process leading to more clarity and efficiency benefiting both the EU and the candidate states. Furthermore, more effective negotiations and better safeguards can contribute to diminishing the so called ‘enlargement fatigue’ the EU may suffer from.3

Existing literature in the field of International Relations (IR) mainly focuses on the period after accession in Bulgaria and Romania. Emphasis is put on the situation within Bulgaria and Romania (Papadimitriou and Phinnemore 2008), an evaluation of the Cooperation and Verification Mechanism (CVM) and the implementation of it by both countries (Noutcheva and Bechev 2008). Less attention is however paid to the process of accession that preceded the introduction of the CVM and the way in which the EU responded to the accession process of Bulgaria and Romania. This research will therefore focus on the perception of effectiveness of the accession process as seen from the side of the EU and examine how the EU is reacting to the results of the accession process of Bulgaria and Romania and adjusting its enlargement strategy. In doing this I use the external incentives model and the social learning model. Herein the former is based on conditionality as driving factor behind progress whereas the latter depends on the constructivist notion of a perceived and unwanted gap in the own behaviour as ideal compared to that of the other. By comparing the accession process of Bulgaria and Romania with the accession process of Croatia this research can provide empirical evidence about how the use of different factors of the external incentives model and the social learning model has changed through the years. This contribution can therewith enhance the scholarly understanding of the process of EU enlargement.

Research design and methodology

To conduct the research I make use of document analysis. Several kinds of documents will be used to study the topic. In the first place scholarly articles and books will be used in the

(8)

(9)

9

1. Theoretical framework

To explain the changes in the political accession criteria of the EU between the enlargement round of 2007 and the accession of Croatia it is necessary to use a model that is based on both the values and norms of the EU as well as the pressure exerted by the EU on the candidate countries. Schimmelfennig and Sedelmeier (2005) introduced a framework to analyse Europeanization in the CEES that consists of three alternative models that together combine the use of norms and values and the use of pressure exerted by the EU. Within these models they distinguish between adaptational pressures driven by the EU and driven by the CEES. In the research into the change in the accession criteria I focus on the accession process from the side of the EU, as this is the actor that can make significant changes in the accession process. Therefore I use only the two models of Schimmelfennig and Sedelmeier that focus on EU driven changes in the process of Europeanization. The first model, the external incentives model, focuses on the pressure exerted by the EU on the candidate countries to implement the norms and values of the EU through four external incentives that influence the cost benefit balance candidate states make during the accession process. Although a much heard point of critique on this model is that the domestic factors of the candidate country are not accounted for in the model (Hughes, Sasse and Gordon 2004) this is of less concern for my research as I focus on the EU side of the enlargement process. The second model, the social learning model, emphasizes the identification of the CEES with the EU and the persuasion of the CEES by the EU of legitimacy of its rules as key conditions for rule adoption (Schimmelfennig and Sedelmeier 2005, 9). The combination of these models serves as a framework for my further research wherein I thus study the enlargement process from an EU perspective. Without going into the post-Cold War IR debate between naturalist and conservative notions of the world I thus pay attention to both rational driven change (external incentives model) and sociological driven change (social learning model).

2.2 The external incentives model

(10)

10 The logic of consequences is based on the rationalist assumption that states enter social relations with predefined interests. According to March and Olsen (1998) actions of actors are driven by the calculations of the consequences and measured against prior preferences. This definition was used in the work of Schimmelfennig and Sedelmeier (2005) on the accession of the CEES in the 2004 enlargement round. However in his individual 2003 work Schimmelfennig (18) had already noted that with regard to the environment states act in, anarchy is the defining factor. Herein the preferences of the state are based on the assessment of its welfare (and security) and the state will always choose the option that maximizes its own welfare, security and therewith its chances of survival in the earlier described anarchical setting (Ibid, 18, 19).4

The main characteristic of the external incentives model is conditionality. Conditionality became the central policy in the enlargement strategy of the EU with the introduction of the Copenhagen criteria and has since been the subject of considerable research. The contributions to the debate on conditionality vary in their focus and geographical spread. While the main focus of research on conditionality has always been on the role of conditionality in the accession process, recently the emphasis has shifted more towards the role of conditionality after accession to the EU (O'Dwyer 2010, Sedelmeier 2011, Trauner 2009). Furthermore, whereas Kelley (2004) and Sasse (2008) take a narrow approach and focus on conditionality in the case of minority protection, Vachudova (2005) takes a broader approach in her research as she focuses on the behaviour of political actors and policy change in six CEES.

Schimmelfennig and Sedelmeier (2005, 11) define conditionality as reinforcement by reward. The EU pays the reward of membership only when the candidate country complies with the conditions set by the Union. When the candidate country fails to comply the Union will withhold the reward. They furthermore argue that the Union does not intervene or penalize the candidate country by offering extra benefits or introduce extra punishments to the candidate country (Ibid). The withholding of the reward of membership should be punishment enough for the candidate country to choose to comply with the conditions of the EU. Steunenberg and Dimitrova (2007, 3) add to this that this so-called carrot and stick approach is justified because the EU views itself as a club with specific rules. Only when a state is capable of implementing the adaptations of the key features of the EU, a stable democratic

(11)

11 regime and a well-functioning market economy, a state can become a member of the club (Ibid).

The workings of conditionality can be further explained by the asymmetrical relationship between the EU and the candidate countries (Schimmelfennig and Sedelmeier 2005, 10, Hughes, Sasse and Gordon 2004, Moravcsik and Vachudova 2003). The EU is capable of making these demands because the candidate countries have a lot at stake, more than the EU. For the candidate country not complying with the conditions of the EU means not being rewarded with the much desired membership. For the EU on the other hand, there is less at stake. Although a failure of the accession process could affect the credibility of the Union’s enlargement process or security in the EU-region, this is a considerable lower price to pay than the price the candidate countries pay. The power in the accession process thus lies in the hands of the EU, the superior bargaining power. According to Moravcsik and Vachudova (2003, 49) countries that enter into the asymmetrical relationship are thus willing to make concessions. The applicant countries have accepted the costs of adaptation, and consider these costs to be lower than the costs of exclusion from the EU club.

Adding to the argument of Moravcsik and Vachudova (Ibid), Schimmelfennig and Sedelmeier (2005, 12) argue that candidate countries are only willing to make these concessions when the rewards of adaptation are higher than the costs thereof. This cost-benefit balance forms the basis of the external incentives model that assumes that ‘’a government adopts EU rules if the benefits of the EU rewards exceed the domestic adaptation costs’’ (Ibid). The cost-benefit balance depends on four factors: 1) the determinacy of conditions; 2) the size and speed of the rewards; 3) the credibility of conditionality and; 4) the role of veto players and adoption costs (Ibid). According to the external incentives model each of these four factors influences the likelihood of rule adoption by the candidate country.

(12)

12 The first factor of the external incentives model, the determinacy of conditions, refers to the formality of the rule and its substantive detail and clarity (Ibid). The more legalized the rule the clearer it is about the type and extent of domestic change expected, the higher its determinacy value the more effective it will be (Magen 2005-2006, 414). Not only the determinacy of the concrete rule, but also the determinacy by which the EU generally treats a certain issue in the enlargement process will determine the impact the EU exerts on the issue in question (De Ridder 2011, 667). The determinacy affects the informational value, giving the target country a clear idea of what they have to do in order to receive the reward of membership. Furthermore, determinacy matters because it enhances the credibility of conditionality (Schimmelfennig and Sedelmeier 2005, 12). It reduces the chance of reinterpretation or misinterpretation of a rule by the target country and at the same time it binds the EU to awarding the candidate country with membership when the conditions are met. The external incentives model thus asserts that the likelihood of rule adoption increases, when clarity of the rules set as conditions for reward increases (see fig. 1) (Schimmelfennig and Sedelmeier 2005).

Fig. 1: The external incentives model and the influence of the different factors on the likelihood of rule adoption.

In line with the external incentives model, several studies focused on the influence of the conditions of the EU on the accession of new members to the Union. In her research into the influence exerted by the EU through the accession process Heather Grabbe (2006, 91-94)

External incentives model Veto players and adoption costs Decreases the likelihood of rule adoption Credibility of conditionality Increases the likelihood of rule adoption Asymmetries in information Decreases the credibility of conditionality Cross-conditionality Decreases the credibility of conditionality Consistency Increases the credibility of conditionality High capabilities and low costs Increases the credibility of conditionality

(13)

13 describes the determinacy of conditions as one of the six dimensions of uncertainty in the accession process. According to her the rules of the EU were not determinate at all in the accessions between 1989 and 2004 (Ibid). Often it was not clear what exactly counted as meeting the conditions of the Union and complying with its various demands. Furthermore, there was uncertain linkage between fulfilling particular criteria and receiving particular benefits, because of the very general and vague Copenhagen criteria (Ibid). Nevertheless, all the states that started the accession process between 1989 and 2004 eventually entered the EU, despite the lack of determinate conditions. Research by De Ridder (2011) on the Czech Republic and Slovakia also shows that despite the lack of determinate conditions both countries did adopt the EU-rules and joined the Union, albeit after a difficult accession process. As stated above, the determinacy of conditions is thus more of influence on the speed of the accession process than on the actual accession of a candidate country. In my further research it is therefore interesting to see confirmed whether change in the determinacy of conditions also leads to change in the speed of the accession process.

Adding to the conclusions of Schimmelfennig and Sedelmeier (2005) and Grabbe (2006) on the determinacy of conditions, Steunenberg and Dimitrova (2007, 12) find that not all conditions matter equally in the accession process. While some of the conditions are singled out in the yearly progress reports, other conditions are not mentioned at all. Moreover, they conclude that the pressure of the EU on candidate countries also differs between candidates (Ibid). Some candidates have been pressured more and on a wider range of issues than other candidate countries. Both the importance of certain conditions and the intensity of the pressure exerted by the EU with regard to these conditions thus varies per country (Ibid). For this research it is therefore important to take into account that the studied changes in the accession process can either be caused by a deliberate change of the EU in the accession process in general or because the EU uses a country specific approach, that differs from the general approach used in the process of other countries.

(14)

14 to make changes in the domestic situation when it is rewarded with membership than when it will be rewarded with merely economic support.

My critique towards Schimmelfennig and Sedelmeier’s vision on the size and speed of rewards is however that within the EU accession process there is little to no room to shift the moment of payment of rewards. I state that in sense of time the reward (membership) may sometimes be far away, however in procedural terms the moment of its distribution stays static; when all chapters are closed the candidate country is deemed to be fit for membership and, after agreement of the Council, the treaty can be signed. Also the attractiveness of the reward cannot be altered from the EU perspective. It might be that due to candidate state’s internal developments the perspective of the reward (membership) becomes more attractive; however that is not the starting point of Schimmelfenning and Sedelmeier’s perspective. Herein it however remains important that the conditions are determinate and the candidate country thus knows what is expected in order to receive the reward of membership.

That the reward of membership is nevertheless a decisive factor in the compliance of the candidate countries to the demands of the EU is illustrated by several researches into the European Neighbourhood Policy (ENP) (Magen 2005-2006, Lavenex 2011, Kelley 2006, Smith 2001). The ENP is not a membership policy, but specifically an instrument to reach out to those European countries that do not yet fall within the realm of prospective member states. The absence of this membership perspective is seriously weakening the external incentives of compliance. It is however important to note that countries that had the reward of membership in sight did not always feel compelled to comply with the conditions of the EU, as shows the case of Slovakia under the leadership of Mečiar in the mid-1990s (Kelley 2006, Schimmelfennig, Engert and Knobel 2005).5

A third set of factors has to do with the credibility of the conditionality applied by the Union. For the candidate country to comply with the conditions of the EU, it has to be certain that the reward will be paid at the end of the road (2005, 13-16). Furthermore, the EU has to be a credible superior bargaining power for threats to be credible. Schimmelfennig and Sedelmeier (Ibid) have divided this factor into four conditions that can either enhance or impair the credibility of conditionality. The first condition is the capabilities and costs for the EU in employing conditionality (see fig. 1). The EU must be able to withhold the reward if the candidate country does not meet the conditions (Ibid, 14). The asymmetrical relationship

(15)

15 between the EU and the CEES proves that the target countries are more dependent on the EU than the other way around. Furthermore, the credibility depends on the consistency of the allocation of the reward (Ibid, 15). If the EU rewards membership to a candidate country without it fulfilling the conditions or the other way around, it will lose its credibility and candidate countries will fail to adopt EU rules. Additionally, cross-conditionality must be absent or minor (Ibid). Conditionality would not be effective if there would be another organisation offering the same benefits as the EU, but at lower costs. The final factor concerns asymmetries of information. If the EU is not able to monitor the target state, or if the information by other monitoring agencies is inconsistent or if the target state is able to conceal its compliance record from the monitors, the credibility of conditionality is weakened (Ibid). Following these four factors that influence the credibility of conditionality, Schimmelfennig and Sedelmeier formulate the following hypothesis: ‘’the likelihood of rule adoption increases with the credibility of conditional threats and promises’’ (see fig. 1) (Ibid, 16).

In addition to the external incentives model of Schimmelfennig and Sedelmeier (Ibid), Böhmelt and Freyburg (2012) focus on the credibility of conditionality in the different phases of the accession process. Like Steunenberg and Dimitorova (2007), they consider the accession talks as a series of negotiations consisting of different phases, namely 1) the application, 2) recognition as a candidate state, 3) the opening of negotiations, 4) the official closure of negotiations, and 5) the signing of the accession treaty. Böhmelt and Freyburg (2012, 254) add to this that each phase is characterised by a different probability of membership. Whereas the credibility of both the threat of exclusion and the promise of membership of the EU are of pivotal importance in the beginning of the accession process, the likelihood of rejection by the EU decreases the further the accession process progresses. From this Böhmelt and Freyburg (Ibid) conclude that the credibility of the EU’s threat declines and the bargaining power of the candidate grows. It is however important to note that despite the fact that the bargaining power of the Union decreases throughout the process, the EU remains the superior bargaining power. With the accession date coming closer there is more at stake for the EU as they already invested a lot in the accession of the new member state. Nevertheless, for the candidate country the stakes are higher as they risk membership of the Union.

(16)

16 of EU rules than on the benefits. Rule adoption is assumed to always be costly; otherwise conditionality would not have been necessary (Ibid). If the political costs are too high this might prevent rule adoption, moderate costs may lead to compliance of the candidate country as a result of effective conditionality (Schwellnus, Balázs and Mikalayeva 2009). Additionally, veto players might block decisions that are in favour of meeting the conditions of the Union. Schimmelfennig and Sedelmeier (2005) assume that the number of veto players is small in most CEES. In sum, they formulate the following hypothesis: ‘’the likelihood of rule adoption decreases with the number of veto players incurring net adoption costs from compliance’’ (see fig. 1) (Ibid).

The studies of Steunenberg and Dimitrova (2007), Böhmelt and Freyburg (2012), Grabbe (2006) and the study of Moravcsik and Vachudova (2003) already showed that the external incentives model has been the foundation for several other studies. These studies also showed that, contrary to this research, the external incentives model has primarily been used to examine the effectiveness of enlargement conditionality. Nevertheless, one can also say that the conditionality based external incentives model is of great use to examine the changes in the accession process as the process is based on conditionality. However, when it comes to the political criteria several studies find that the conditionality based external incentives model alone is not capable of explaining the compliance of candidate countries to the criteria of the EU.

(17)

17 Despite these difficulties, Kelley (2004) agrees with Tomić (2013) on the use of a combination of conditionality and socialization based approaches. However, whereas Tomić (2013) argues that the external incentives model should not be used without combining it with the social learning model, Kelley (2004) approaches the issue the other way around. She argues that a socialization based approach alone is not capable of changing government behaviour and should thus always be used in combination with the external incentives model (Ibid). Socialization based approaches like the social learning model have been used for several years but the effects were minimal. Socialization based approaches were rarely capable of changing behaviour and often failed when the domestic opposition was too strong (Ibid). Only when the linkage to EU membership was made, countries were motivated enough to reform and meet the conditions of the EU (Ibid).

Besides the above given arguments to combine the external incentives model and the social learning model it is however also important to emphasize that both models approach the accession process from a different starting point and assume a different motive for norm implementation. In the case of the external incentives model conditionality is the driving force behind norm implementation. Candidate countries only implement change because of the pressure exerted by the EU and the reward they get in return. The social learning model on the other hand assumes that a norm is implemented because the candidate country itself beliefs that their country should be governed according to this norm. The norms of the EU are not per se EU-specific norms, but can be granted to the Western-European democratic society wherein non-EU members Norway and Switzerland also take their place.

(18)

18

2.3 The social learning model

The social learning model is based on the notions of the logic of appropriateness, which in turn is based on the core trades of social constructivism. Social constructivism assumes that material resources only acquire meaning for human action through the structure of shared knowledge (Wendt 1995, 73). Furthermore, social constructivism emphasizes the role of norms and institutions, the assumption on which the logic of appropriateness is build (Reus-Smit 2009, 220-221). The logic of appropriateness assumes that decision-making is based on what the actor believes are the right norms and values, instead of the cost-benefit analysis in rationalist institutionalism (Balsinger 2014). In social constructivism the variables constituting the world are thus not predefined or given as in rationalism, but developed through interaction with others; in other words: they are socially constructed. Actors thus follow certain rules because over time they have grown to perceive these rules as natural, valid and legitimate (Ibid). Actors who want to belong to certain institutions or organisations shall internalize the norms and belief systems promoted by these institutions and organisations they want to become a member of (March & Olsen 1998; Börzel & Risse 2000, 2009). The model assumes that actors who interact with each other on a regular basis will develop the same norms and belief systems (Cengiz and Hoffmann 2012). Although this might be the case in a situation in which both actors start on equal foot, this is not the case with the relation between the EU and the candidate countries. As the EU has already developed a set of rules, norms and values through the years, the candidate country is the only actor that has to adjust its norms and belief system. Nevertheless, it is the EU’s role to persuade the candidate country of its rules, norms and values.

(19)

19 looks into the role of domestic resonance in the accession process of Slovenia and Latvia and Noutcheva (2009) examines the role of legitimacy in the Western Balkans. Nevertheless, one has to conclude that compared to the literature on rationalism and European enlargement, specifically the role of conditionality, research on the role of constructivism and social learning is still in its infancy.

The social learning model offers the most prominent alternative to the rationalist explanations of conditionality (Schimmelfennig and Sedelmeier 2005, 18). Just as the external incentives model, the social learning model focuses on Europeanization from an EU perspective. However, contrary to the rational external incentives model this model is based on the logic of appropriateness. In this perspective the EU is the formal organisation of a European international community defined by a specific collective identity and a specific set of common values and norms (Ibid). Whether a candidate country adopts the norms and values of the EU depends on the degree in which the candidate country sees the EU rules and demands for rule adoption as appropriate in terms of collective identity, values and norms (Ibid). The proposition of the social learning model is therefore: ‘’a government adopts EU rules if it is persuaded of the appropriateness of EU rules’’ (Ibid).

There are three factors that influence the transfer of rules from the EU to a non-member state. The first factor refers to the quality of the rules and the process through which the rules were established and transferred to the target country (Ibid). In the external incentives model this factor is referred to as the determinacy of conditions, as the adoption of rules depends on the clarity of the rule. However, the social learning model assumes that if a rule is formulated ambiguously or used inconsequently by the EU member-states, the capacity of the rule to bind the non-member state suffers (Ibid). Furthermore, the legitimacy of the rule will be strengthened by the extent to which the rule is tied to the norms and values of the EU. The legitimacy of the rule is further dependent upon the way in which it is transferred to the non-member state (Ibid). If the rule is forced upon the non-member state the rule is likely to lose in legitimacy. However, when the EU engages the target states in a process and takes their concerns into account by the interpretation of the rule the idea of the imposition of a rule will be mitigated (Ibid).

(20)

20 sum Schimmelfennig and Sedelmeier formulate the following hypothesis: ‘’the likelihood of rule adoption increases as the legitimacy of the rule increases’’ (see fig. 2) (Ibid).

Fig. 2: The social learning model and the influence of the different factors on the likelihood of rule adoption.

The second factor of the social learning model is identification. Schimmelfennig and Sedelmeier (Ibid) assume that when a target state can identify with the organisation of which they have to adopt the rules, the likelihood of rule adoption will increase. When the target country and the EU share the same values and norms the target state will identify itself with the EU. Schimmelfennig and Sedelmeier therefore propose the following identity hypothesis: ‘’the likelihood of rule adoption increases with the identification of the target government and society with the community that has established the rules’’ (see fig. 2) (Ibid).

Whereas the social learning model thus assumes that, besides the use of conditionality as in the external incentives model, the adoption of the rules, norms and values of the EU can also follow when a candidate country can identify with the norms of the EU, Jordan (Jordan 2003, 665) beliefs that conditionality is responsible for the implementation of EU norms and rules. She states that only after the adaptation of EU norms and rules under EU pressure identification with these norms will follow through a spill-over effect. Adaptation of the norm through conditionality leads to a more structural change which then can spill over in a process of identity formation in the candidate country.

Although conditionality is an important factor in the implementation of EU norms by candidate countries, this is not the only explaining factor. The assumption that rule adoption will automatically follow when a candidate country identifies with the norms of the EU is however also not the defining explanation. For the implementation of the EU norms one can

(21)

21 say that the model responsible for the implementation differs per norm. However, for the implementation of some norms there is neither EU pressure nor identification with the EU necessary, as the candidate country already governs its country according to the norms of the EU. Implementation of other rules can follow because of identification with the EU. There will however also remain a category of norms which the candidate country cannot identify with and where the use of conditionality is necessary for the implementation thereof. As is argued by Jordan (Ibid), it is then possible that identification will follow through a spill-over effect. Nevertheless, there also remains the possibility that identification does not happen. In that case it is possible that the implemented rule will change after the actual accession of the candidate country.

The final factor of the social learning model concerns resonance. According to Schimmelfennig and Sedelmeier (2005, 20) ‘’the likelihood of rule adoption increases with domestic resonance’’ (see fig. 2). According to Engert (2010, 93) domestic resonance depends on the cultural background of the state, thus the shared values, collective beliefs and common understandings that are derived from historical, cultural and religious traditions. These elements influence the ideology of the ruling government (Ibid, 94). It is however possible that this ideology changes. There are two situations wherein the willingness of the target state to adopt the rules of the EU can increase. First if there not yet exists a rule on the issue on which the EU insists on introducing new rules or if these rules are no longer legitimate because of a crisis or policy failure (Schimmelfennig & Sedelmeier, 2005, 20). Second, in the case that there do exist rules on matters that are important to the EU, the likelihood of rule adoption will increase if the principles of the EU rules and the rules of the target country correspond (Ibid).

(22)

(23)

23

Analysis

3. Political criteria in the accession of the EU

Following the elaboration on the theoretical framework in the previous chapter, this chapter zooms in on the political criteria and therewith gives a better understanding of the meaning of these criteria and the precise criteria that the candidate countries have to fulfil. Herewith the following question is answered: ‘’what do the values of democracy, human rights and the rule of law entail for the Union and how has this changed between the accession of Bulgaria and Romania and the accession of Croatia?’’ As the Union discusses democracy and the rule of law in a combined section and human rights and minority protection in another section, I also answer the subquestion of this chapter according to this division.

The concepts of democracy, human rights and rule of law are considered of utmost importance in the field of international relations. This is emphasized by the fact that they can be found in multiple international treaties6 and that many states and organisations focus on the export of these concepts, seeing it as essential values to be implemented in states in transition. It therefore comes as no surprise that the EU also has a particular interest in these concepts, and the export of it to other future member states. The importance of these values for the EU is amongst others demonstrated by the fact that the EU mentions democracy, human rights and the rule of law as the values upon which the Union is founded (Council of the European Communities and Commission of the European Communities 1992). Furthermore, these concepts are determined in the first item of the Copenhagen criteria, the most fundamental prerequisites for a state to be even considered for membership. Herein it is stated that states need to have a ‘’stability of institutions guaranteeing democracy, the rule of law, human rights and respect for and protection of minorities’’ (European Commission 2012). The great importance that the EU attaches to these values makes it more legitimate for the Union to export them and demand from future members states that they attach the same importance to these values. From the perspective of the social learning model the legitimacy the EU attaches to these values increases the likelihood of rule adoption by the candidate states. However, what the EU precisely means with democracy, the rule of law and human rights is not further

6

(24)

24 elaborated upon in the Copenhagen criteria or in the treaties of the EU. In the next paragraphs I therefore elaborate upon the meaning of these concepts for the EU.

3.1 Democracy and the rule of law

The Union does not have a single definition of the concepts of democracy and rule of law. Neither in the Copenhagen criteria, nor in any treaty does the EU mention the meaning of these highly regarded concepts. Despite the lack of a clear definition, the EU has applied the concepts for quite some time; even before the introduction of the Copenhagen criteria democracy and the rule of law were important factors in the enlargement process.7 Therefore it is possible to determine the definition of these concepts based on how they were applied during the latest enlargement rounds. In 1997 the Union introduced progress reports to make an assessment of the progress made by candidate countries on a yearly basis (Commission of the European Communities 1997). These reports state amongst others whether each of the candidate countries meets the Copenhagen criteria and assesses the preparedness of the candidates for accession. As democracy and the rule of law are extensively elaborated upon in these reports, they can help to further define these concepts. Furthermore, it is possible to make a comparison between the definition of the concepts in the accession process of Bulgaria and Romania and in the accession process of Croatia. The reports on the Central and Eastern European applicants from the 1990s,8 further on mentioned as the 1990s applicants, serve as a baseline in this comparison.

Before going into detail on the EU’s definition of the values it is important to look at the structure of the progress reports, as this gives an idea of the most important elements of democracy and rule of law. The most striking is that the European Commission does not make a distinction between both concepts, but reduces the two fractions to a common and presents it as an organic combination (Hoffmeister 2002). Although the organic combination of democracy and the rule of law did not change between the enlargement rounds, the subjects that were discussed under this heading did change. In the accession process of Bulgaria and Romania the heading was divided into four sections, namely the parliament; the executive; the judiciary and anti-corruption measures. In the progress reports of Croatia the Commission replaced the section on the executive with a section on public administration. Furthermore, the Commission added the section government. The newly introduced sections in the progress

7_{Nevertheless one should be aware that the rule of law standard demanded from for example Greece in the} 1980s was of a different level than what was asked from those countries that joined the EU post 2000. 8

(25)

25 report of Croatia suggest that the four sections that were used in the progress reports on Bulgaria and Romania were not specific enough to observe all relevant problems.

Under the first heading of the reports on the 1990s applicants, the functioning of the parliament, the Commission discussed the functioning of the parliament in general, the position of the opposition and the elections that have taken place in the period before composing the progress reports. To discuss these topics the Commission used almost the same wording in all the progress reports: ‘’The […] Parliament continues to operate satisfactorily. Its powers are respected and the opposition plays a full part in its activities’’ (Commission of the European Communities 1998a, 8). Furthermore the Commission pointed out the importance of minority representation in the parliament. The Commission was thus very clear about the requirements that had to be fulfilled. This increased the determinacy of conditions, an important element in the cost-benefit balance of the external incentives model. On the other hand however, there was a lack of determinacy on how the candidate countries can best achieve a functioning parliament that meets the standards of the EU. However, with a Union consisting of 15 member states before the 2004 enlargement round, there were and still are enough examples and knowhow on how to achieve this. The lack of determinacy on how to achieve a functioning parliament could thus be diminished through cooperation with the member states. It is however the responsibility of the candidate country to consult the member states hereon, as they wanted to join the Union.

In the reports on Croatia the Commission attached importance to exactly the same elements in the section on the functioning of the parliament. The only difference was that this section was complemented by the section government (Commission of the European Communities 2006c, 6). Herewith the Commission drew more attention to both topics and therewith showed that a good functioning parliament and government was of great importance for joining the Union. Furthermore, the separation of the section parliament in to two separate sections created more determinate criteria, which would increase the likelihood of rule adoption according to the external incentives model. As both topics were discussed in more detail, future member states, amongst which Croatia, were better informed about what was expected by the Union.

(26)

26 Under the subheading of state administration the Commission discussed in most cases decentralisation; there had to be a certain level of self-government for local governments and the state administration should be transparent.9 Concerning the civil service, emphasis was put on a professional, impartial, and a political neutral civil service.10 An ideal executive would thus be effective, professional, accountable, well regulated and transparent in the eyes of the Commission. As was the case in discussing the functioning of the parliament, the Commission was very clear on the requirements for a functioning executive. Nevertheless, it again refrained to determine how the candidate countries could establish an executive that met the standards of the EU.

From 2005 onwards, when the Commission published the first report on the progress made by Croatia towards joining the EU, the Commission no longer spoke of the executive but introduced the section public administration (Commission of the European Communities 2006c, 6). Although the content under the new section remained the same as under the section executive, there was some striking change that came with the change of the title. Whereas the Commission usually rather shortly discussed the functioning of the executive, the functioning of the public administration was elaborated upon in more detail. Herewith Croatia was thus better informed about the requirements on the executive than Bulgaria and Romania were during their accession process. Despite the more extensive elaborated upon demands of the EU on the public administration, the Commission still refrained from giving a detailed description on how to achieve a public administration, or executive, that meets the demands of the EU. The determinacy on the way in which the candidate countries could achieve an executive up to the standards of the EU thus had not changed in the period between the first reports on Bulgaria and Romania and the reports published on Croatia.

The following element under the heading democracy and the rule of law in the progress reports was the judiciary, both in the case of Bulgaria and Romania as in the case of Croatia. According to the Commission the judiciary is the vital element in ensuring respect for the rule of law and for the effective enforcement of the Acquis (Commission of the European Communities 2001c). The importance of the judiciary for the Union is also evident from the fact that chapter 23 of the Acquis discussed its functioning in more detail. According to chapter 23 the judiciary needed to be independent and efficient. Furthermore, impartiality, integrity and a high standard of adjudication by the courts were essential for safeguarding the

(27)

27 rule of law (European Commission 2013). These characteristics were complemented by the discussion on the judiciary in the progress reports. In its strategy paper of 2002 the Commission summarized the most important characteristics of the judiciary: the adaptation of basic legislation, strengthening of human resources of the judiciary, improving working conditions, the introduction of due-enforcement of court decisions, the improvement of citizens’ access to justice and tackling the problem of backlogs (Commission of the European Communities 2002). Comparing the reports of Croatia with those of the 1990s applicants, the Commission insisted on the same elements to create a well-functioning judiciary. The determinacy of the judiciary conditions was thus in both of the accession processes high, leading to an increased likeliness of rule adoption according to the model of Schimmelfennig and Sedelmeier. Bulgaria and Romania as well as Croatia were well informed about what their judiciary should look like according to EU standards.

As shown above, it is thus possible to create a list of elements that the judiciary of a European-member state has to entail. However, as was the case with the parliament and the executive, the Commission was not very clear on how the candidate could achieve a judiciary that meets the standards of the EU. On this matter the candidates could consult with the member states of the Union, however, their judicial systems are all organised in different ways. The advice of the member states could thus differ depending on the member state giving advice (Vigenin 2012, 61). The confusion and disorder this could create is illustrated by the case of Bulgaria. After recommendations and consultations with the Commission the Bulgarian Parliament adopted changes in the Bulgarian constitution to increase the accountability of the judiciary (Ibid). The day after, the Commissioner of Enlargement issued a statement warning that the changes adopted were threatening the independence of the Bulgarian judicial system and that new changes to the Constitution had to be introduced (Ibid).

(28)

28 accession processes. The determinacy on how to meet the criteria of the EU is thus both in the case of Romania and Bulgaria and in the case of Croatia insufficient.

The final section in the discussion on democracy on the rule of law elaborated on anti-corruption measures. In this case the title of the subheading clearly indicated the goal of the EU: ruling out corruption. In the progress reports on the 1990s applicants and the progress reports on Croatia the Commission discussed measures that could be taken to prevent corruption, which came down to the introduction of new laws and joining of conventions. All ten CEES and Croatia became a member of the Group of States Against Corruption (GRECO) and adopted the OECD Anti-Bribery Convention. Complementary to these conventions each country introduced its own laws to prevent corruption. These laws emphasized the definition of Transparency International that describes corruption as ‘’the abuse of entrusted power for private gain’’ (Transparency International n.d.). Concerning the fight against corruption the conditions were thus determined as it was clear what was expected of the candidate countries, both in the accession process of Bulgaria and Romania and in the process of Croatia.

Whereas determinacy in the external incentives model is about the clarity and formality of the EU rules, the social learning model puts more emphasis on the legitimacy of the rules, which is amongst others influenced by the implementation of EU rules in the EU’s member states. Looking into the compliance of the member states with the rules of the EU one can conclude that some of the current member states do not meet the criteria that new members have to meet in order to become a member of the Union. Hungary, for example, met the EU criteria when they joined in 2004 but the situation has changed since then. The Orbán government in Hungary leaves little to no room for opposition and has increased its control over the justice system, contrary to the criteria of the EU (Sedelmeier 2014, Mueller 2014). The Union’s capacity to force the candidate countries to implement its rules suffers from the fact that the current member states do not follow its rules. Furthermore, it shows that the power of conditionality is the strongest during the accession process, when membership is still at risk. When the reward of membership is paid the EU has no more leverage to force its members to comply with the rules.

(29)

29 check the functioning of the government. Another vital element was the implementation of basic legislation and the training of judges in order to be able to apply the EU law. Finally, corruption was fought by the implementation of various laws and conventions that prevent the abuse of entrusted power for private gain. The criteria thus seemed to be very determinate for both Bulgaria and Romania as well as for Croatia. In the case of Croatia the criteria were even more determinate due to the more extensive elaboration on the subjects by the EU. Nevertheless, there was a lack of determinacy regarding the implementation of the criteria. The EU remained rather vague on how to achieve its standards and member state advice was not always sufficient. From the perspective of the external incentives model this could lead to a decreased likeliness of rule adoption, as it is not clear for the applicants how they could meet the EU requirements. The decrease of the likeliness of rule adoption is also visible from the perspective of the social learning model. The fact that current member states do not always meet the standards of the EU regarding the political criteria impairs the legitimacy of the Union’s criteria. Concluding one can thus state that the criteria of the EU decrease the likeliness of rule adoption, both from the perspective of the external incentives model as well as from the perspective of the social learning model.

3.2 Human rights and minority protection

The meaning of human rights was already laid down in 1950, in the European Convention on Human Rights (ECHR) by the European Council (Council of Europe 1950). In 59 articles divided into 3 sections the ECHR clarifies the rights and freedoms that the countries that ratified the ECHR need to protect. These are the right to life, freedom and security; freedom of expression; freedom of thought, conscience and religion and the right to vote and stand for election (Council of Europe 1950). Additionally the convention prohibits the death penalty, torture or inhumane or degrading treatment and discrimination of amongst others minorities (Ibid). Although the Union was not able to sign the ECHR the Union attaches great importance to the treaty.11 This is evident from the mentioning of the ECHR in several treaties of the Union and recent debates in the European Parliament about the EU joining the ECHR (European Parliament 2012).12 Moreover, the same rights were discussed in the progress

11

The Lisbon treaty of 2007 established a legal personality for the EU in art. 47 TEU (European Union 2007). The legal personality made it possible for the Union to conclude and negotiate international agreements, to become a member of international organisations and to join international conventions like the ECHR. 12

(30)

30 reports of the EU under the subheadings civil and political rights; economic and social rights and minority rights and minority protection. Therewith the Commission confirmed that applicant countries should meet the criteria laid down in the ECHR.

From the perspective of the external incentives model one can argue that the ECHR increased the likelihood of rule adoption. Because of the very determinate conditions of the ECHR the applicant countries were well informed about what was expected of them. The ECHR left no room for misunderstandings or misinterpretations which increased the likeliness that the applicants would implement the human rights of the EU. The determinacy increased even more with the introduction of the EU Charter of Fundamental Rights. The Charter was meant to make the importance and relevance of human rights more visible for the EU citizens (European Union 2000). The 54 articles of the Charter are divided into six categories: dignity, freedoms, equality, solidarity, citizens’ rights and justice, guaranteeing amongst others the right to life, non-discrimination, the prohibition of child labour and the right to vote (Ibid). As was the case in the ECHR, the Charter also discusses minority protection. Article 21 of the Charter states that ‘’any discrimination based on the grounds of […] membership of a national minority […] is prohibited’’ (Ibid). The treaty however only became legally binding after Bulgaria and Romania joined the Union. For them the determinacy of conditions did not increase. For Croatia on the other hand the conditions became more determinate halfway their accession process in 2009.

From the perspective of the social learning model the correspondence between the belief systems of the applicants and that of the EU are of pivotal importance for an increased likeliness of rule adoption. Initially it seemed like the belief systems of Bulgaria, Romania and Croatia corresponded with that of the EU as they ratified the ECHR respectively in 1992, 1994 and 1997, before they officially started the accession process. One should however also keep a critical stance towards this correspondence. Despite the early ratification of the ECHR all three countries did not always honour and respect the human rights that are laid down in the ECHR (Amnesty International 1995, Amnesty International 1994, Amnesty International 1999). One can however argue that ratifying the treaty is about creating an image that corresponds to the ideal candidate for EU membership, rather than signing the treaty because of the correspondence to the domestic belief systems in the candidate countries. The treaty should therefore be seen as a lever in the negotiations, a means to talk to the candidates about their behaviour concerning human rights. Besides the lack of compliance of the applicants, Protection of Human Rights and Fundamental Freedoms (European Communities, 1987; Council of the

(31)

31 there is however also a lack of compliance with the ECHR and the Charter on the side of the member states.13 This inconsistent compliance of the member states impairs the credibility of the EU and makes it harder for the EU to demand these standards from applicant countries. This in turn decreases the likeliness of rule adoption according to the social learning model.

Contrary to democracy and the rule of law the EU is very clear on the meaning of human rights. The ECHR comprises all the rights and freedoms that member states should respect. Furthermore, the ECHR amongst others prohibits the death penalty, torture or inhumane degrading treatment and discrimination. The ECHR is thus very determinate on the human rights conditions that have to be met, which leads to an increased likeliness of rule adoption according to the external incentives model. The determinacy increased, although very little, with the introduction of the EU Charter on Fundamental Rights. As the Charter became legally binding in 2009 the criteria were thus more determinate for Croatia.

Initially it seemed that there was domestic resonance in Bulgaria, Romania and Croatia, as all three countries ratified the ECHR even before the accession process began. However, taking a closer look it turned out that not all the rights laid down in the ECHR are honoured. Ratification could thus points towards creating the perfect image rather than correspondence to the belief systems of the three candidates. In addition to the likeliness of rule adoption decreasing because of a lack of domestic resonance, it further decreases because of the incoherent application of the ECHR by the member states of the EU. As a consequence the EU’s capacity to obligate the candidates to meet the ECHR criteria suffers.

(32)

32

4. The enlargement process

With a clear understanding of the way in which the European enlargement process can be analysed through the external incentives model and the social learning model and by having established the definitions of democracy, the rule of law and human rights, it is possible to move towards the factual analysis of the changes in the accession process. This chapter elaborates on the workings of the enlargement process and answers the question: ‘’how does the enlargement process of the Union work?’’ To understand the changes in the enlargement process it is important to first paint a picture of the steps in the enlargement process and the role that the European institutions play in this process.

The enlargement process of the EU is anchored in the basic provisions of the EU treaties and is based on the experience from previous enlargement rounds. Article 49 of the Treaty on European Union (TEU) states that any European state may apply for membership, providing that it respects the principles of liberty and democracy, respect for human rights and fundamental freedoms, as is set out in Article 6(1) TEU (European Union 2010). In this regard the Copenhagen European Council of 1993 laid down certain criteria that were confirmed at the Madrid Council meeting in 1995 (European Commission 2012). A new member state should meet three criteria. In the first place the political criteria which demand a stability of institutions guaranteeing democracy, the rule of law, human rights and respect for and protection of minorities (Ibid). The second set of criteria are economic and demand the existence of a functioning market economy and the capacity to cope with competitive pressure and market forces within the Union (Ibid). Finally the new member state has to accept the community Acquis and be able to take on the obligations of membership, including adherence to the aims of the political, economic and monetary union (Ibid).

(33)

33 state to the level of a de facto veto, as there is only one state necessary to withhold the application.

Although it is not a legal prerequisite, it is customary for the Council to decide on the application after the Commission has given its opinion. In its opinion the Commission identifies the main country specific issues that have to be dealt with during the accession process and that can have a major influence on the shaping of the negotiations (Baun 2000, 12). Besides the main issues all the applicant countries have to fulfil the political Copenhagen criteria before the accession negotiations can be opened. These criteria form the first baseline of political criteria that have to be met before further progress in possible. In accordance with the argument of Schimmelfennig and Sedelmeier (2005) in the external incentives model there are thus intermediate rewards in the accession process. In this case fulfilling the political criteria is rewarded with progress towards the next step in the accession process, the accession negotiations. Moreover, the intermediate reward stresses that the accession process is based on conditionality, the central element of the external incentives model, as the political criteria first have to be fulfilled in order to receive the intermediate reward.

After the Commission forms its opinion about the readiness of the applicant country, it gives its recommendation to the Council.14 When the Council decides in favour of the applicant and the European Parliament supports the application (Cini and Pérez-Solórzano Borrgán 2010, 423), the second stage in the process of joining the EU can start: the membership negotiations. These are an intergovernmental conference between the ministers and ambassadors of the EU governments and the applicant country (European Commission 2013b). Although this phase is described as negotiations between the EU and the applicant country, in reality there is not much to negotiate for the applicant country. As Schimmelfennig and Sedelmeier (2005) already stated in their external incentives model, the relationship between the EU and the candidate country is asymmetrical. The EU is the superior bargaining power that lays down the rules, in this case the implementation of the full

Acquis Communautaire. The candidate country on the other hand, has little to say in the

process. Besides the possibility of limited derogations or transition periods for particular aspects for EU legislation, the candidate country has to accept the requirements that the EU sets for joining the EU (Baun 2000, 14). Whereas long-time members like the UK and Denmark have in the past secured opt-outs in certain policy areas, for nowadays candidate

(34)

34 countries that are not willing to meet these requirements, membership is off the table.15 With the power over the process in the hands of the EU, the EU is thus able to manipulate the outcome of the process to their advantage as long as the willingness to become a member remains high in a candidate country.

At the heart of the second stage in the accession process lays the Acquis

Communautaire, the body of EU policies and legislation. Whereas the political Copenhagen

criteria covered the basis of the political criteria that had to be met by the applicants, the

Acquis chapters further define the political criteria. The negotiation phase starts with the

screening of the Acquis, which is the basis for the bilateral negotiations between the EU and the candidate country. The aim of the screening is to ensure that the candidate countries fully understand their obligations as future members and to identify potential issues for the negotiations (Avery 2004, 42). The central institution in the screening process is the Commission, who carries out a detailed examination of each policy field to see what progress has to be made (European Commission 2013b). The findings are presented to the member states in a screening report. In the conclusion of the report the Commission makes a recommendation to either open negotiations directly or to require that certain conditions should be met first (Ibid). The Council then decides, based on the recommendation of the Commission, whether the negotiations will be opened.

If the Council decides that the negotiations can be opened the candidate country submits its position and the EU adopts a common position, after which the negotiations for membership can start (Ibid). The pace and duration of the negotiations differ per country, depending on the speed of the reforms and the alignment with EU law in the candidate country (Ibid). During the negotiations the EU positions and statements are delivered to the applicant countries by the country holding the presidency over the Council, acting as the spokesperson of the EU (Avery 2004, 39). The role of the Commission in the negotiations is mostly behind the scenes. This role is however fundamental in the negotiation process as the Commission carries out the actual negotiations with the candidate country, which have a highly technical nature (Ibid). The Commission is therefore perceived as the ’honest broker’, seeking to find solutions in the interest of the enlarged EU instead of the interest of the

15_{The UK has an opt-out for the Schengen agreement because of which it does not participate on the border} control elements (European Union 2010, 291). Both Denmark and the UK have an opt-out concerning

Changes in the political criteria of the enlargement process of the European Union