key: cord-0057635-o0ql9oej authors: Mišík, Matúš; Prachárová, Veronika title: Perfect Compliance? Nuclear Power in Central and Eastern Europe and the EU Membership date: 2020-07-17 journal: From Economic to Energy Transition DOI: 10.1007/978-3-030-55085-1_15 sha: bcc41c1403dc95f568e9ed9bbb8c917a95936544 doc_id: 57635 cord_uid: o0ql9oej This chapter examines three Central and Eastern European EU members (Bulgaria, Lithuania, and Slovakia) that agreed to shut down their nuclear reactors not fulfilling the EU safety requirements during the accession process. Since this request was part of the membership requirements, conditionality seems to be able to explain their willingness to do such difficult step impacting their energy policies and energy security. However, while Bulgaria phased-out its reactors in Kozloduy already during the pre-accession process—in line with conditionality expectations—Lithuania and Slovakia did so only after they became full EU members in 2004. The puzzle this chapter is trying to shed light on is why these two countries fulfilled these (demanding and unpopular) obligations if the EU conditionality was not in place anymore. The chapter analyses factors that could have contributed to the post-accession compliance of Lithuania and Slovakia and thus contributes to the general discussion about the activities of Central and Eastern European countries in the EU. conditionality connected to EU membership negotiations replaced financial incentives as the crucial element in the strategy of the West. However, this chapter claims, this process was not that straightforward. The EU conditionality has been considered a crucial driving force of CEE countries on their way to EU membership (Moravcsik and Vachudova 2005; Epstein and Sedelmeier 2008) . During the accession process a system of "carrots and sticks" was developed to foster compliance with the EU acquis communautaire by the candidate countries (Schimmelfennig 2005; Schimmelfennig and Sedelmeier 2005) . The pivotal position of conditionality in this process has been repeatedly stressed (Gateva 2015) , but also criticized (Sedelmeier 2008) . The direct influence of conditionality expectedly decreased after CEE countries became members, when the main incentive-the membership itselfdisappeared (Böhmelt and Freyburg 2012; Zhelyazkova et al. 2017) . The dominant explanation of compliance-the external incentives model (Falkner and Treib 2008 ) that stood the test of time (Schimmelfennig and Sedelmeier 2020 )-argues that successful adoption of rules by the CEE countries is linked to membership promise by the EU as an external actor. Thus, it was assumed that the absence of incentives after accession would "significantly slow down or even halt the implementation process" (Schimmelfennig and Sedelmeier 2005, p. 226) . This chapter aims to contribute to the ongoing discussion about the post-accession compliance (Ademmer 2018; Börzel and Sedelmeier 2017 ; Buzogány forthcoming) by analysing nuclear energy policy in the CEE. During the pre-accession negotiations, three countries from this region were asked to shut down their Soviet-style nuclear reactors that were considered to be unsafe by the international community. This was a very sensitive topic as nuclear power played an important role in the energy mixes of these countries; moreover, the issue was challenging also at the political level (Mišík 2010) . Therefore, we could have expected the EU conditionality characterized by "highly effective means of influence" (Epstein and Sedelmeier 2008, p. 795) to ensure compliance in the preaccession period. It was a straightforward issue-the closure of nuclear power reactors, unlike most other requirements for EU membership, could have been easily defined and measured (Ivanov 2008 ). Yet, the problematic reactors at NPPs in two CEE states (Lithuania and Slovakia) were shut down only after several years after entering into the EU. Only Bulgaria that joined the EU in 2007 shut down its NPP before the accession date and thus acted in line with theoretical expectations prescribed by the external incentive model. The puzzle this chapter is trying to shed light on is why Lithuania and Slovakia fulfilled the obligations agreed upon during the pre-accession period, if the EU membership promise as powerful incentive was not present after the accession to the EU anymore. Moreover, a significant part of public and domestic elites in these states opposed such a step. In spite of the significance of this issue for both countries, existing analyses provide only very limited insights and are dated (Ivanov 2008; van Oudenaren 2001) . Analysis in this chapter indicates that changing post-accession incentive structure did not lead to deterioration of compliance with EU requirements and rules, thus supporting existing findings (Börzel and Sedelmeier 2017; Zhelyazkova et al. 2017) . Although membership conditionality lost its leverage after the accession, there have been other factors connected to indirect influence of conditionality that led to the compliance with EU rules. EU compliance studies and the Europeanization literature is based either directly on rational understanding of rule compliance (as that of external incentive model) or have been rationalized within the framework of a positivist research agenda (Brosig 2012) . The role of actors in translation of the effects of instruments (e.g. rule incorporation, application) and the motives of actions are, from a large share, overlooked (exemptions include Sasse 2008 or Thomann and Sager 2017) . However, externally initiated pressures have limited influence if actors refuse to react to them. Even though we agree that the decision to close NPPs was based on (positive or negative) motivations created by institutional arrangements, actors needed to interpret European institutions as well as the pressures that were supposed to emanate from them, in order to be able to act (Jacquot and Woll 2003) . What is, therefore, important is how exactly the very sensitive topic of NPPs was internalized by actors and how they understand institutional arrangements. The chapter argues that the direct conditionality-as presented by the external incentives model-has a limited explanatory power when it comes to the shutting down of NPPs. However, conditionality had an indirect effect-not based on rational consideration/calculationscausing the CEE countries to get used to follow "the rules of the game". To be more precise, indirect and long-term processes of social learning have led to norm-conforming behaviour (Jacoby 2005; Schimmelfennig and Trauner 2009) . In this sense, although the membership "carrot" lost its relevance, non-compliance was less likely to occur (Krizsan 2009 ). Lithuania and Slovakia considered the requirement to close down their NPPs as their obligation. We, therefore, argue that one of the most important factors influencing compliance is social learning. In comparison to Western countries, the CEE countries had adjusted their legislation prior to accession. They learned to behave in a particular way, so, it is not that surprising that based on data on notification records and infringement cases, the CEE member states even appear to outperform their Western counterparts, as they implement the EU directives before the specified deadlines and are less likely to be sanctioned for non-compliance by the European Commission (EC) (Zhelyazkova and Yordanova 2015) . The second section provides a theoretical background by discussing the role of conditionality in member states' compliance with the EU's rule. It examines the limits of post-accession compliance and argues that conditionality impacts it mostly in an indirect way, via social learning. The third section examines the EU's requirement to close down NPPs and examines all three countries that were affected by such a demand while looking especially close at the case of Ignalina NPP and Jaslovské Bohunice NPP located in Lithuania and Slovakia. The conclusion presents the main finding of the analysis arguing that these two countries engaged in social learning that supported compliance with EU rules including their own commitments from the pre-accession period. Compliance has been studied in connection to international regulatory agreements (Tallberg 2002; Treib 2014) , but especially to the EU. Compliance as incorporation and application of EU policies and rules by member states is essential not just for the practical functioning of specific policy areas (Sedelmeier 2008) , but also for the success of the whole European integration project (Zhelyazkova et al. 2017) . That is why the expectations of inadequate application and enforcement of EU rules by the CEE countries led the EU to impose a strict pre-accession conditionality (Sedelmeier 2005) . Contrary to the earlier enlargement rounds, candidate states for the 2004 and 2007 EU enlargements faced the challenge of adjusting their legislation prior to accession. Given the number of EU laws and regulations, as well as the accompanying administrative and institutional requirements, this obligation presented a mammoth task making CEE countries face much greater challenges than those encountered by previous applicants (Grabbe and Hughes 1998; Böhmelt and Freyburg 2012) . The EC even reviewed the behaviour of each applicant country according to explicit criteria within annual reports (Dinan 2004 ). If we take into account that the accession process can be described as a series of negotiations "over the extent to which the applicant, at any given stage of its preparation, satisfies the conditions set by the Union and the extent to which the EU is willing to continue supporting the applicant's candidacy" (Steunenberg and Dimitrova 2007, p. 6) , it generated enormous pressure on candidate countries and their representatives to conform to what the EU expected of them (Linden 2002) . The foremost approach so far in the field of enlargement studies concerning the CEE countries' adaptation performance-the external incentives model (Falkner and Treib 2008; Brosig 2012 )-explains the successful adoption of EU rules primarily by the indicated membership "carrot" promised to the candidates by the EU as an external actor. It holds that governments comply with EU law/requirements if the calculated benefits of membership exceed the expected political costs of compliance that are associated with the accession criteria. So, if, as Moravcsik and Vachudova (2005) claim, applicant states conceded to most of the demands from the EU "because the basic benefit offered to them-membership-is of such great value" (Moravcsik and Vachudova 2005, p. 49), such conditionality should be one of the most important foreign policy tools of the EU. Kelley (2004) even states that as early as in the 1990s membership presented a unique and valuable asset-i.e. positive incentive-for all post-Communist states, as for most of politicians "the membership served as proof of being modern and liberal and as a sign of their states' return to Europe as a democratizing, lawful country" (Kelley 2004, pp. 20-21) . The prospect of EU membership thus provided at that time an incentive for aspiring member states to persist with painful reforms that were needed anyway (Dinan 2004) . Conditionality in this sense presented a way how the EU was able to influence or change processes in domestic policy areas of non-member states-i.e. Europeanization. In fact, the application of the concept of Europeanization has thus no longer been restricted to political changes in member states of the EU (Bauer et al. 2007) . Accordingly, Schimmelfennig and Sedelmeier (2005) conclude that the EU's huge impact on the Europeanization (defined as "process by which domestic policy areas become increasingly subject to European policy-making"; Börzel 1999, p. 574) of CEE countries was caused by the size of the net benefits of EU accession to the candidate countries and of the credibility of EU conditionality. Their findings automatically triggered questions and concerns about the future development as/when the instruments will not be present anymore. They saw good reasons to be concerned as "in the absence of high conditional external benefits, domestic structures such as adoption costs, veto players […]that were superseded in the conditionality context-will again have a causal impact" (Schimmelfennig and Sedelmeier 2005, p. 226) . There was no more a strong domestic consensus on the goal of membership that could silence potential veto players (Sedelmeier 2008) or even overcome domestic opposition in highly sensitive issue areas (Kelley 2004 ) once the countries joined the Union. Sedelmeier (2011) hence notes that the EU's attempt to influence domestic adjustment prior to accession raises the question of how durable the adjustment patterns and the EU's preaccession influence were. Moreover, from his point of view, it is not only about "how effective Europeanisation through conditionality is, but also whether post-conditionality Europeanisation remains" (Sedelmeier 2011, p. 6) . We argue that although the EU membership conditionality lost its leverage after the accession of CEE countries to the EU, there have been other factors that led to the compliance with EU rules. Before addressing the factors determining post-accession compliance, the chapter discusses two perspectives of the existing research. First, from the methodological point of view, the chapter argues that a comprehensive review of literature can not only help us to obtain an in-depth understanding of complex matter of adaptation performance and role of actors therein but also to highlight theoretical discrepancies. Second, the chapter claims that the role of actors, when describing factors accounting for (non-)compliance of member states, is usually overlooked. In fact, by underestimating the importance of making the decision of countries' representatives whether to comply or not, scholars narrow and limit the explanation power of conditionality. The member states' transposition and implementation performance in the absence of conditionality with focus on domestic factors in explaining member states' behaviour is the main interest of compliance studies in the enlarged Europe (Börzel and Buzogány 2019). We identify two main traits in these studies: the dominance of quantitative approach and difficulties with difference identification between formal transposition and de facto implementation. This chapter sides with the research by Zhelyazkova et al. (2017) in criticizing this scholarship. First, quantitative studies (e.g. Böhmelt and Freyburg 2014) use official data on rule transposition to individual member states, which are often based only on notification by the member states. Such data can be problematic as countries can report less than honestly (Mastenbroek 2005) . Although the potential gap between transposition, i.e. formal incorporation, and de facto implementation is widely acknowledged (Falkner and Treib 2008; Sedelmeier 2011 ), 1 given the need for extensive data gathering at the micro level to study the actual implementation of policies, transposition studies still dominate compliance literature (Dimitrova and Steunenberg 2017) . These studies themselves recognize the limits of such an approach while qualitative studies are trying to fill this gap (Maniokas 2009 ). While we understand that small/medium-N case studies focusing only on specific countries and policy sectors can offer findings that are limited in their power both to explain and to generalize (Toshkov 2008) , they should not be underestimated as they can point out obvious theoretical discrepancies. Second, analyses (although not all; Schimmelfennig and Trauner 2009) indicated rather good transposition at the formal level (Sedelmeier 2008) and problems in (not only; Versluis 2007) CEE countries when it comes to practical implementation of EU rules (Falkner and Treib 2008) . Comparative designs of compliance studies have, however, tended to favour some member states and even policy sectors over others, especially environmental and social ones (Börzel and Buzogány 2019; Dimitrova and Steunenberg 2017) . Although scholars have usually clearly stated reasons for case selection, the questions about the generalizability of the findings are more than appropriate. Indeed, Zhelyazkova et al. (2017) argues that the indicated gap between formal and practical compliance can be much smaller than previously thought as the literature focused on social policy that is especially challenging for CEE countries. These new insights were obtained thanks to a combination of large-N quantitative approach and qualitative case studies (Ibid.). Considering these arguments this chapter claims that qualitative approach using case studies can help us to shed light on compliance of (not only CEE) EU member states. Conditionality had an impact, however, different from the one foreseen by the external incentives model. Using the cases of Lithuanian and Slovak nuclear reactors, which were not shut down in the preaccession period like the ones in Bulgaria, this research argues that there are three main theoretical deficiencies of the external incentives model in explaining compliance. First, following the rationale of the external incentives model, the membership conditionality should have been employed to achieve compliance with the most difficult EU requirements towards applicant countries. This was, however, not the case of the highly political sensitive (and therefore problematic) issues of NPPs shutting down in Lithuania and Slovakia. This most significant and controversial matter of their accession negotiations (Lofstedt 2008) had not been completed before these countries became full members. So, conditional incentive of membership was not that important a tool in helping the EU to ensure these countries' compliance as would the external incentives model suggest. Nor was it helping to overcome domestic opposition in highly sensitive issue areas (Kelley 2004 ) as the EU membership incentive ("carrot") was not on the table anymore. Second, incentive-based approaches expect the influence of international institutions to be weakest in areas in which the EU never applied conditionality (i.e. they are neither part of the acquis, nor explicitly included in its accession conditionality), like the closure of NPPs (Epstein and Sedelmeier 2008) . Therefore, it seems that the best evidence that other instruments effectively replaced the membership as the main incentive is the fact that even though the NPPs in question were crucial for the energy security of those countries, they were, as promised to the EC, eventually shut down. Third, if the incentive-based conditionality hypothesis is correct, with the circumstances under which non-member states comply-a credible membership incentive and non-prohibitive domestic adjustment costs (Kelley 2004; Schimmelfennig 2005 )-the decreasing influence of international institutions should be particularly salient right after accession (Epstein and Sedelmeier 2008) . Furthermore, an incentive-based perspective would also expect new member countries to use weaker sanctioning capacity of the EU institutions (it takes a long time for Court of Justice of the EU to impose financial penalties) and start with good formal compliance that is not followed by proper application and enforcement (Sedelmeier 2011) . None of these was confirmed in the case of CEE countries (Zhelyazkova et al. 2017 ). The literature focusing on factors that account for (non-)compliance of member states has been so far inconclusive. For example, Sedelmeier (2008) claims that compliance of CEE member states is only incremental if these are threatened with sanctions or the (temporal) safeguard clause in the accession treaty that could had been used to suspend certain membership rights during the first three years of membership. He highlights the importance of legislative capacities built during the pre-accession period, as well as socialization. Many empirical studies, however, point out that compliance is connected primarily to the functional qualities like administrative capacity or co-ordination within national ministries (Toshkov 2010) . This suggested that the new CEE member states would lag behind their Western counterparts, as they have a relatively lower administrative capacity. Zhelyazkova et al. (2017) argue that the opposite was actually true in most cases. However, before any conditions or factors can influence the ability of a member country to comply with EU rules, 2 there is actors' decision whether to conform to what is expected of them. Since the compliance in form of incorporation and implementation of EU policies and rules by member states is essential for the success of the whole European integration project (Zhelyazkova and Yordanova 2015) , it is quite understandable that there have been instruments (e.g. external incentives, pressures) developed to influence member states decisionmakers (actors) in order to achieve this goal as smoothly as possible. Scholars have analysed causal conditions (such as these instruments) and their relevance in explanation of variations in member states' compliance performance. The role of actors in translation of the effects of these instruments and the motives of actions (i.e. constructivist approach) is underdeveloped as there has been a tendency to put emphasis "on structural elements and institutional pressures with less attention to the mechanisms through which these induce change" (Jacquot and Woll 2003, p. 2) . These instruments are, however, useless if actors refuse to react to them. Even though we agree that the decision to close NPPs was based on motivations created by institutional arrangements, it should not be forgotten that actors, in order to be able to act, need to interpret European institutions as well as the pressures that are supposed to emanate from them (Ibid.). While this chapter does not rule out the influence of incentives, it nonetheless focuses on social context, socialization, and social learning (Epstein 2008; Checkel 2001) . In fact, even the main proponents of the external incentives model (Schimmelfennig and Sedelmeier 2005; Kelley 2004 ) acknowledge the importance of socialization and social learning in the process of (domestic) change. In the accession process they, however, found it hard to empirically pinpoint socialization and social learning as it was difficult to separate them from rationalist calculations (Sasse 2008) . However, in the post-accession period the conditional incentive of membership-as presented by the external incentives model-has no explanatory power in matters of compliance with EU rules. Thus, the overlap of social learning and direct effects of the EU membership conditionality would not be problematic. Moreover, this chapter focuses on indirect effects of conditionality, i.e. how the CEE countries got used to following "the rules". Even though the EU membership conditionality loses its leverage shortly before or as soon as the accession treaties are signed, it does not mean that it does not have any consequential effects on involved actors and thus also an indirect effect on member countrys' compliance. For example, it has been argued that the CEE member states have higher efficiency in meeting the respective transposition deadlines than Western countries because they, thanks to conditionality, "acquired experience in adopting vast amounts of legislation within a limited amount of time" (Zhelyazkova et al. 2017, p. 231) . Three out of the EU's four enlargement rounds resulted in a better compliance of the new members with EU law compared to the original members (Börzel and Sedelmeier 2017) . The only exception was the Southern enlargement as it led to a substantial increase in non-compliance. The application of the preaccession conditionality explains why the new members from the CEE performed so well. Indeed, as the explanation based on social learning shows, the CEE countries considered good compliance as appropriate behaviour that made them more susceptible to shaming (Sedelmeier 2008) . The experience of regular monitoring and assessments of progress with alignment (following a strict conditionality) during the pre-accession period has created for decision-makers in post-Communist countries a clear link between compliance and being a deserving and acceptable member state (Börzel and Sedelmeier 2017) . Consequently, CEE countries have been more eager to demonstrate their status as legitimate EU members by notifying national measures to the EC on time (Sedelmeier 2008) . This point of view is especially important because, as Sedelmeier (2011) claims, a crucial challenge for studies that want to examine the distinctiveness of post-conditionality compliance is to "establish whether explanatory factors more generally explain variation in compliance across old and new members in the enlarged EU" (p. 27). It is necessary, he argues, to find out if explanatory factors are specific to new members from the CEE, or if their impact on these countries differs from their impact on the old members. Thus, since candidate states for the 2004 and 2007 EU enlargements faced the challenge of adjusting legislation prior to accession, there is an important distinction in the way how Europeanization proceeded before and even after the accession. Indeed, one of the ways how the EU effect can manifest itself is through a steady continuity of deepening structural or behavioural trends from the pre-accession period (Sasse 2008) . These trends were in the CEE member states, contrary to the Western countries, primarily influenced by conditionality. As a part of the pre-accession negotiation three CEE countries promised to shut down their NPPs. Lithuania as well as Slovakia did this after their accession, in an era when membership as the main incentive was not relevant anymore. Only Bulgaria shut down the problematic reactors at Kozloduy NPP before the actual accession. Politicians and even the broad public considered this requirement as an essential one in relation to membership, although the possibilities that the EU (or the EC) had at its disposal in the case of non-compliance were limited and rather difficult to achieve (the above-mentioned judicial review; Sedelmeier 2011). This chapter argues that it was the indirect effect of conditionality in the form of social learning that prepared the actors (decision-makers) to conform to behaviour that was expected of them (i.e. compliance). Therefore, the chapter claims that the most important factors influencing compliance with the requirement to shut down nuclear reactors in Slovakia and Lithuania was social learning (besides the EU's financial and political support 3 ). The research presented in this chapter is based on secondary literature and primary data obtained from interviews with different stakeholders in Lithuania, Slovakia, and at the EU level. Utilized interviews are part of a bigger database that has been created by both authors for energy policy related research. While not conducted especially for this particular analysis, the interviews provide detailed information on the process surrounding shutting down of examined nuclear units. Moreover, since there has not been a significant time difference between the discussed events and the period when the interviews were conducted, they provide more reliable and precise account of the events than interviews that would be conducted nowadays. Since the Chernobyl NPP accident, the safety and operation of NPPs has not been, especially in Europe, considered a national matter but rather a regional issue. From the early 1990s, the international community provided financial aid to countries operating certain types of Soviet-built nuclear reactors on the condition that these countries close down these reactors. This, however, did not happen. On contrary, many nuclear facilities in CEE countries undergo significant (not only safety) upgrades with the aim of prolonging their life span. Then the EU and its membership conditionality entered the discussion. The vision of EU membership changed from being a distant prospect to a medium-term policy objective that replaced financial incentives as the crucial element in Western strategy (van Oudenaren 2001) . While there were still loans and support funds in place helping candidates' countries to fulfil the EU's requirements connected to nuclear energy, they were of secondary importance (Ibid.). The EU repeatedly highlighted the importance of a high level of nuclear safety. In June 2001, the Council of the European Union even in its Report on Nuclear Safety in the Context of Enlargement recommended candidates countries to continue their national safety improvement programmes, including the safe management of spent fuel and radioactive waste, to proceed with implementation of the proposed revision of the nuclear law, and to take needed steps to ensure the safety of their research reactors (Council of the EU 2001) . All candidate countries responded to these recommendations and as a Status Report from June 2002 concludes all of them were clearly committed to fulfil the recommendations. However, the EU insisted on the early closure of certain types of nuclear power units (Council of the EU 2002) . While problematic reactors at Kozloduy were closed before Bulgaria entered the EU on 1 January 2007, Lithuanian Ignalina (Units 1 and 2) and Slovak Jaslovské Bohunice V1 (Units 1 and 2) were closed only after the latter two countries entered the EU in 2004 (see Table 1 ). From the perspective of this chapter the closure of four units at Bulgarian NPP Kozloduy followed the conditionality logic. The power plant experienced "prolonged inspection" resulting in the International Atomic Energy Agency's demand for "substantial improvement of the safety situation" (Tchalakov and Mitev 2019, p. 53 Energy chapter was the most difficult issue during Bulgarian negotiations with the EC; however, it was temporary closed already in November 2002, five years before the entrance of the country to the EU. As part of the agreement the EC dropped visa requirements for Bulgarian citizens that sent a strong political signal (Tchalakov and Hristov 2019) . The country managed to cope with the decrease of domestic electricity generation by replacing nuclear capacity by coal-fired power plants (Tchalakov and Mitev 2019) . However, Bulgaria made an official request in 2007 to restart the then recently shut down reactors based on the accession treaty that in Article 36 allowed the country to take protective measurements to prevent serious deterioration of the economy. This request was, comparably to a similar one during the 2009 gas crisis, declined by the EC despite support from the neighbouring countries importing Bulgarian electricity (BBC 2007). While we have seen the pre-accession conditionality resulting in the closure of four problematic reactors at Kozloduy NPP, the situation in Lithuania was different with Ignalina "making it" into the EU. The European Agreement between Lithuania and the EU was signed on 12 June 1995. The pressure exercised by conditionality was rather strong and the shutting down of Ignalina NPP was one of the most important pre-accession issues (Maniokas 2009 ). Even before the opening of negotiations it was clear that the EC would consider 2005 and 2009 as possible dates of the closure of two Ignalina reactors. The parliament adopted a resolution in 1999 that the first unit of the power plant would be shut down before 2005 and that the decision about the deadline for the second one would be made until 2004 (Seimas 1999) . Lithuania had several motives not to fulfil the requirement to shut down its NPP: the step meant a risk of the country's energy security deterioration as the power plant generated a significant share of total electricity production (up to 87.1%; Vilemas 2010), the country was able to export surplus of (cheap) electricity to Latvia and Kaliningrad, and it was clear that the electricity mix would change making Lithuania more dependent on Russian natural gas (Mišík and Prachárová 2016) . Public debate in Lithuania articulated the question whether it was worth "sacrificing" the power plant in exchange for EU membership (Baltic Times 2002), however, the political discourse primarily focused on the financial support from the EU (Vilpišauskas 2003) . Even though Lithuania eventually agreed to the EU's principal position to close the plant, it managed to obtain considerable compensations and link these to the modernization of the whole energy sector (European Commission 2003). There were three attempts to postpone the closure of the power plant, including one immediately after the country joined the Union in 2004, the second one during the period of 2007-2008 when the Parliament requested the Government to start negotiations with the EC (that replied "no"), and a third one in 2007 when a life prolongation of the NPP was suggested (Maniokas 2009 ). An unsuccessful referendum (a 50% quota was not met) connected to the last attempt to prolong a lifespan of the NPP in 2008 was accompanied by a firm EC negative position on the issue. It was not "a mix of credible threats of sanctions and potential political damage", both negative tools, that "replaced the role of preaccession conditionality" (Maniokas 2009, p. 9 ). This chapter argues that more positive incentives in the form of social learning-i.e. doing things the "right way" prevailed. It was understood that there are some costs connected to EU membership and that the shutting down of Ignalina NPP was "a bigger cost of membership" that would, however, be compensated by financial and political support from the EU. 4 This was "part of the deal to become the member" of the EU. 5 After gaining independence, there was an agreement that Lithuania shall join the EU. When the country's representatives signed the accession agreement, they knew "that in 2009 we should close our nuclear power plant". 6 The government concluded negotiations with the EC on the closure of Ignalina, in spite of the fact that "a big part of the population was against" 7 and there was a lot of investment into Ignalina's security standards. The only alternative to this development was "non-membership", but since this was a very sensitive issue, "no party was going to take that risk. No one resisted, [membership] was a unanimous decision". 8 Moreover, it was questionable if keeping the NPP running would prevent the country from being dependent on Russia as with "Ignalina we were completely dependent on Russia, without Russia we couldn't run the plant". 9 Due to the central position of energy security in the domestic political discourse, this issue had been strongly politicized. It was argued during our interviews that even a brand-new NPP would not necessarily mean increased independence since the plant would mean a different type of dependency-on foreign lenders as the country does not have enough funds to finance a new NPP itself. Besides, electricity imports are cheap. Contrary to a negative perception of Russia, EU membership is seen as providing shelter from external threats, especially Russia as it "addresses the key challenge" of energy security. 10 The interests of Lithuania and the EU overlap in the energy area, because the EU is also interested in "bigger choice, secure supply, and the lowest possible price for the consumer". 11 Although the closure of the Ignalina NPP has been considered by our interviewees as a necessary pragmatic trade-off with the EU, they have also noted a lot of experience gained during the accession process (also afterwards) that helped the country to deal with its energy policy issues. On the one hand, Lithuania "lost influence, we lost money, on the other hand we have common EU energy policy [and we understand] that for all EU member states this kind of policy is needed". 12 The EU is viewed as a "very good tutor, its supervision is very helpful"; during the accession period the country got a lot of experience and knowledge-especially connected to privatization of the energy sector. 13 According to one interviewee, Lithuanian politicians even tend to change their opinion after their visits to Brussels. 14 Unbundling presents another visible example of social learning in the Lithuanian experience. The "European style of management in the energy sector" that it introduced was considered to be a very positive EU influence that helped the country to deal with its energy security issues. 15 The country followed EU rules in this area very closely and adapted the strictest unbundling opinion that enabled the country to limit the engagement of Russian Gazprom in the domestic gas system. Slovakia has been one of the main proponents of nuclear energy within the EU stressing the technology's climate friendliness. Its support for nuclear energy is underlined by the fact that it is one of only three countries within the EU that currently build new reactors (Mochovce Units 3 and 4; the other two are Finnish Olkiluoto Unit 3 and Hungarian Paks 2). The newest reactor in Slovakia was commenced quite recently in 2000 (Mochovce Unit 2) and nuclear energy is responsible for more than half of domestic electricity production (Eurostat 2020) . One of the main reasons why Slovakia is so supportive of nuclear power is that it is considered to be a domestic source of energy. The country "is so dependent on import of all types of energy sources so we want to be independent there where it is at least a bit possible and that's electricity production from nuclear". 16 The European Nuclear Energy Forum that was established and is currently hosted by the Czech Republic and Slovakia under auspices of the EC means for Slovaks recognition of nuclear energy at the EU level. 17 Especially before the 2011 Fukushima accident, Slovakia had been trying to make nuclear energy at the EU level an energy source equal to others so it would qualify for a similar financial support as had, for example, renewable energy sources (RES). 18 The situation in the Slovak nuclear sector before joining the EU had been similar to the Lithuanian one. Two of the country's nuclear units (1 and 2 at Jaslovské Bohunice V1 NPP) were considered to be unsafe by the international community and therefore the government was asked by the EC during the pre-accession negotiations to shut them down. The Slovak government adopted a resolution in 1999 on closure of the two reactors at the end of 2006 and 2008 and thus fulfilled the EC's requirements that belonged among the main ones (Mišík 2019 ). This has provoked significant domestic expert and political criticism (Rousek 2010 ) not only right after the decision, with the opposition criticizing what it believe was a premature closure of safe and efficient reactors (Rybář 2011 ), but also later on when the right-wing government was replaced after two consecutive terms by a left-wing one in 2006 (Kratochvíl and Mišík 2020) . However, our respondents noted that the decision to shut down the two reactors was voluntary and served for the Slovak government as a bargaining chip in the pre-accession process. 19 This strategy was part of a broader learning process that started even before the membership negotiations. Indeed, the importance of social learning facilitated by personal contact was highlighted already by the Slovak representatives' experience from negotiation within the International Atomic Energy Agency. The experience was later utilized at the EU level as it helped them to "learn how to negotiate" within the international arena. 20 Slovak representatives managed to learn how to influence the EU's activities in nuclear policy that turned this issue into "an area where we are quite strong and are able to push through our preferences". 21 This has helped them to negotiate with Austria, that was identified by several of our respondents as the most problematic actor in connection to Slovak nuclear energy policy. Especially during the pre-accession process Austria pushed hard for shutting down the two Slovak units in question as soon as possible. 22 Further application of social learning could be found in the handling of the 2009 gas crisis. The crisis started only a couple of days after the formal closure of the second unit at Jaslovské Bohunice V1 power plant so the restart of the reactor was one of the first ideas on how to remedy at least the electricity-related issues that emerged during the gas crisis. However, the EC's reaction was, similarly to the Bulgarian case, very negative when it spoke about a "clear violation" of the accession treaty if the reactor was restarted. The government did not materialize these plans and the media discourse reflected its critical stance towards the EC for such a harsh reaction (Kratochvíl and Mišík 2020) . However, the most significant consequences for Slovakia would not come from the EC; according to our interviewee, the re-start of Unit 2 at Jaslovské Bohunice V1 would mean a "breach of important political deals that could have been misused by many member states". 23 So, it was not the EU institutions and their threats, but expectations of shaming by other EU members that was of the highest importance in this process. Moreover, given the age of existing fuel in the power plant and the non-existent supplies any fresh ones, the possible very limited amount of electricity that could have been produced under these circumstances was just not worth the trouble. 24 Social learning has been utilized as a strategy also during the negotiations about the future of decommissioning support for Jaslovské Bohunice V1 Units 1 and 2 in the post-2020 period that was originally supposed to be without EU funds (provided via European Bank for Reconstruction and Development). The main Slovak strategy was to get the support of the large member states, because when you get their backing, "you will get also other countries on board". 25 This strategy indicates that the Slovak representatives have learned the "rules of the game" and adjust their negotiations strategies at the EU level accordingly. However, the goal of receiving further decommissioning support has been achieved only partially as Slovakia ended up in a group with Bulgaria, while Ignalina received a separate project (European Commission 2020). The main puzzle this chapter tried to shed light on is why Lithuania and Slovakia fulfilled their commitments in the nuclear energy area, agreed upon during the accession negotiations, once they were already full members of the EU and thus the pre-accession conditionality was gone. For both countries would the non-compliance bring significant profits, as the nuclear units they promised to shut down were crucial (Lithuania) or very important (Slovakia) for their energy policy and energy security. Non-compliance is well-documented within the EU and "original" as well as "new" (i.e. CEE) member states have been having issues connected to transposition of the EU rules (not transposing properly or not at all). Therefore, if the main EU's tool (pre-accession conditionality) securing that countries follow the EU's requirement had already "expired", why did Lithuania and Slovakia decided to fulfil their commitments and closed down their "pillars" of energy policy? This chapter argued that these countries shut down their problematic reactors during the post-accession period because both countries (and their representatives) engaged in social learning during the pre-accession negotiations that supported their compliance with EU rules (i.e. shutting down the reactors). Both countries have been thinking about "reversing" their compliance and restarting or postponing the closure of their NPPs. However, they never did that and we argue that the positive effects of "doing the things the right way" outweighed the possible negative consequences in the form of the EC's low credibility threats. This chapter's main findings thus are that, first, we should not underestimate the power of social learning and the role of actors in adaptation performance. Second, conditionality should not be linked solely to rational approaches such as the external incentive model because it narrows its explanation flexibility. Indeed, conditionality can be-through social learningeffective even after accession. The examined case studies indicate that conditionality had an indirect influence as Lithuania and Slovakia got used to complying with "the rules" and thus, consequently, they took the requirement to shut down their nuclear reactors as a duty. The main criticism of the argument presented in this chapter can stem from the legal framework between member states and the EU that is formed foremost by the accession agreement. Indeed, also the EC mentioned this document in its criticism of examined member states' efforts to restart or delay closure of the nuclear reactors in question. However, the expectations that the member states cannot break rules they agreed upon during the accession negotiations has been challenged repeatedly by several CEE member states. The rule of law and a democratic regime, the main preconditions for even opening membership negotiations with a candidate country and a crucial membership condition, have been confronted by autocratic tendencies in Hungary and Poland for several years now (Ágh 2016; Buzogány 2017; Szent-Iványi and Kugiel forthcoming) . While the EU, its institutions (the EC, the European Parliament) and even other member states have been trying to influence these countries by negative incentives (including the latest German idea to link post-pandemic (Covid-19) support to democratic values and thus using it as a tool against illiberal tendencies in Hungary and Poland; Reuters 2020) these have not proven to be very effective. What can be however more problematic concerning studies focused on compliance is the fact that different scholars refer to different aspects when analysing implementation-that could potentially lead to misunderstandings and confusion The EU offered target countries decommissioning support and funds to modernize their energy sectors. Moreover, these countries benefited also from the EU's political support in the energy security area Interview 1, Energy policy expert Energy policy expert Ibid. 10. Ibid. 11. Interview 6, President's advisor Energy analyst Head of a department Senior official, Ministry of Economy Energy expert, Slovak permanent representation at the EU, Brussels Official, Nuclear Regulatory Authority Senior official, Ministry of Economy Senior official, Ministry of Foreign Affairs Senior official, Nuclear Regulatory Authority Capitalist Diversity and Compliance: Economic Reforms in Central and Eastern Europe After EU Accession The Decline of Democracy in East-Central Europe Life in a Post-Ignalina Town Differential Europeanization in Eastern Europe: The Impact of Diverse EU Regulatory Governance Patterns EU Rejects Balkans Nuclear Plea The Temporal Dimension of the Credibility of EU Conditionality and Candidate States' Compliance with the Acquis Communautaire Diffusion of Compliance in the "Race towards Brussels?" A Spatial Approach to EU Accession Conditionality Towards Convergence in Europe? Institutional Adaptation to Europeanization in Germany and Spain Compliance with EU Environmental Law: The Iceberg Is Melting Larger and More Law Abiding? The Impact of Enlargement on Compliance in the European Union No Space for Constructivism? A Critical Appraisal of European Compliance Illiberal Democracy in Hungary: Authoritarian Diffusion or Domestic Causation? Democratization Beyond Balkan Exceptionalism. Assessing Compliance with EU Law in Bulgaria and Romania Why Comply? Social Learning and European Identity Change Report on Nuclear Safety in the Context of Enlargement Peer Review Status Report, 5034/02 ATO 1 ELARG 1 The Power of Implementers: A Three-Level Game Model of Compliance with EU Policy and Its Application to Cultural Heritage Europe Recast: A History of European Union The Social Context in Conditionality: Internationalizing Finance in Postcommunist Europe Beyond Conditionality: International Institutions in Postcommunist Europe After Enlargement Protocol No. 4 on the Ignalina Nuclear Power Plant in Lithuania Decommissioning of Nuclear Facilities Energy Datasheets: EU28 Countries Three Worlds of Compliance or Four? The EU-15 Compared to New Member States European Union Enlargement Conditionality Enlarging the EU Eastwards. London: The Royal Institute of International Affairs Legitimate Conditionality? The European Union and Nuclear Power Safety in Central and Eastern Europe External Incentives and Lesson-Drawing in Regional Policy and Health Care Usage of European Integration-Europeanisation from a Sociological Perspective Ethnic Politics in Europe. The Power of Norms and Incentives Bad External Actors and Good Nuclear Energy: Media Discourse on Energy Supplies in the Czech Republic and Slovakia From Formal Adoption to Enforcement. Post-accession Shifts in EU Impact on Hungary in the Equality Policy Field Norms and Nannies: The Impact of International Organizations on the Central and East European States Are Renewables an Alternative to Nuclear Power? An Analysis of the Austria/Slovakia Discussions Conditionality and Compliance in Lithuania: The Case of the Best Performer EU Compliance: Still a "Black Hole Security First: Energy Policy in the New Member States of the European Union External Energy Security in the European Union: Small Member States' Perspective Before "Independence" Arrived: Interdependence in Energy Relations Between Lithuania and Russia Preferences, Power and Equilibrium: The Causes and Consequences of EU Enlargement Fact Sheets-Chernobyl Accident and Its Consequences Germany Hinges EU Recovery Funding on Democratic Values Risk Assessment in the Slovak Energy Sector National Determinants of International Preferences in Post-Communist Europe: The Case of Slovakia in the European Union The Politics of EU Conditionality: The Norm of Minority Protection During and Beyond EU Accession Constructing the Path to Eastern Enlargement: The Uneven Policy Impact of EU Identity After Conditionality: Post-Accession Compliance with EU Law in East Central Europe Europeanisation in New Member and Candidate States Lietuvos Respublikos Seimas: The Seimas and the European Union Strategic Calculation and International Socialization: Membership Incentives, Party Constellations, and Sustained Compliance in Central and Eastern Europe The Europeanization of Central and Eastern Europe The Europeanization of Eastern Europe: The External Incentives Model Revisited Post-Accession Compliance in the EU's New Member States Compliance in the EU Enlargement Process: The Limits of Conditionality The Challenge from Within: EU Development Cooperation and the Rise of Illiberalism in Hungary and Poland Paths to Compliance: Enforcement, Management, and the European Union The Bulgarian Nuclear Sector in Transition: Adopted Russian technology, State Sovereignty and Accession to the EU Energy Dependence Behind the Iron Curtain: The Bulgarian Experience Moving Beyond Legal Compliance: Innovative Approaches to EU Multilevel Implementation Embracing European Law Compliance with EU Directives in Central and Eastern Europe Taking Stock: A Review of Quantitative Studies of Transposition and Implementation of Implementing and Complying with EU Governance Outputs The Limits of Conditionality: Nuclear Reactors Safety in Central and Eastern Europe Even Rules, Uneven Practices: Opening the "Black Box" of EU Law in Action Lithuanian Energy Policy in 1990-2009 and the Prospects of Its Development Until 2020 The Fnal Stage of the EU-Accession Game: The Baltic States-The Likely Victims of Their Own Success Notified and Substantive Compliance with EU Law in Enlarged Europe: Evidence from Four Policy Areas Signalling "Compliance": The Link Between Notified EU Directive Implementation and Infringement Cases Acknowledgements This work was supported by the Slovak Research and Development Agency under the project "Priorities of the Central and Eastern European Countries in the context of Energy Union" (APVV-16-0062).