European Union Influence on Institutional Lock-In in the Countries in Transition

This paper explores implications of the EU accession process for consolidation of institutions important for sustainable economic growth in the states aspiring to be EU members. Much of its attention concentrates on the 10 member states from Central and Eastern Europe that have already joined the Union, but the Western Balkans countries also receive attention as they have been on the path towards the membership for some time. The core of the paper consists of three case studies illustrating both the EU influence and its limits and reversals. The first one looks at state aid and competition policies – an area where the EU exercises substantial influence already prior to a country’s accession and where the member states have to respect a dominant role of the European Commission. The next case study contrasts this with the issue of central bank independence, where the accession and membership are crucial, but also that the EU can act to, at the same time, support consolidation of bank autonomy and provide an opportunity to undermine it. The third case looks at EU attempts to mandate civil service reforms in the acceding countries, particularly creation of independent regulators and creation of clear demarcation lines between political and administrative level. In the absence of anchoring in the community legal framework, initial success in the pre-accession periods has been followed by significant backsliding after the membership was achieved. Overall, the conclusion is that the EU influence on institutional design and consolidation for countries aspiring to membership has been powerful when the Union chose to make an issue a priority one, be it because of its own legal framework for its own members (competition, central banking) or as a specific priority vis-à-vis future members (administrative capacity). However, the paper also showed that when the priority is not anchored in firm legal framework and not applied consistently for all existing members, the sustainability is threatened and the membership can easily bring backsliding. Therefore, the value of the Union as an agent of institutional consolidation appears to be precisely in the indefinite and rules-bound aspect of the EU membership. Where the EU accession is a distant and uncertain prospect (e.g. Caucasus, Moldova, Ukraine), the EU’s attempt to use conditionality without explicit promise of membership have, so far, not been successful in achieving its objective.


Introduction
This paper explores implications of the EU accession process for consolidation of institutions important for sustainable economic growth in the states aspiring to be EU members. It is based on original research conducted for this study as well as previous work (particularly Beblavý, 2007 and Beblavý, forthcoming). Much of its attention concentrates on the 10 member states from Central and Eastern Europe that have already joined the Union, but the Western Balkans countries also receive attention as they have been on the path towards the membership for some time. For other postcommunist countries, the EU influence has been limited due to the lacking prospects for the ultimate prize -the membership itself. Nonetheless, the paper briefly examines whether and how the EU has made a difference.
To achieve its objectives, the paper contains, in addition to an introduction and a conclusion, six short sections. The first three sections present a brief background sketch on the pathway to the membership and some pointers from the literature on how the EU has generally influenced creation and consolidation of political and economic institutions. It explores concepts such as conditionality, credibility and legitimacy to explain the process by which the route towards the EU membership has locked in institutions and how the membership continues to act in this way. This is followed by three case studies illustrating both the EU influence and its limits and reversals. The first one looks at state aid and competition policies -an area where the EU exercises substantial influence already prior to a country's accession and where the member states have to respect a dominant role of the European Commission. The next case study contrasts this with the issue of central bank independence, where the accession and membership are crucial, but also that the EU can act to, at the same time, support consolidation of bank autonomy and provide an opportunity to undermine it. The third case looks at EU attempts to mandate civil service reforms in the acceding countries, particularly creation of independent regulators and creation of clear demarcation lines between political and administrative level. In the absence of anchoring in the community legal framework, initial success in the pre-accession periods has been followed by significant backsliding after the membership was achieved.
The final section draws some tentative conclusions both from the literature and the case studies.

The pathway to membership as the EU instrument for institutional consolidation
The influence of the European Union over institutional consolidation in countries undergoing transition has been wielded primarily through conditionality tied ultimately to the EU membership. The road to membership typically progresses through the following stages. First of all, the countries interested in closer partnership with the EU negotiate an association agreement (called Stabilization and Association Agreement for Western Balkans countries), which usually entails access of the associated countries to the EU market as well as financial assistance. Countries wishing to join the EU then have to apply for the membership and the EU member states have to agree that the country fulfils criteria for opening the negotiations (the so-called Copenhagen criteria 1 ). Once the negotiations are completed, the accession process can be finalized through ratification of the membership by all current EU members as well as the country concerned. To some extent, these stages can overlapi.e. a membership application may be lodged even before conclusion of the association agreement.
The Copenhagen criteria were set for the first time in 1993, but their practical implementation and the resulting scheme of requirements and rewards was quite hazy initially. However, by the end of the 1990s, the EU developed a formalized enlargement pathway. As a part of it, it not only conceptualized accession stages based on conditionality, but also hardened the conditions for joining. The membership path for the postcommunist states has thus been relatively onerous compared to the past. In enlargements of 1970s and 1980s, the EU was much more benevolent: "It allowed candidates to negotiate long transitional periods and it provided much more aid to them after accession than it has offered to the current applicants. It was partly the experience of relatively expensive previous enlargements and very slow adaptation by poorer countries like Greece that led to such tight conditions for the central and east European countries" (Grabbe, 2003). This made the EU accession a long and uncertain process (ibid.), clear in stages, but with individualized schedules reflecting the complexity and conditions covering most aspects of political and institutional framework and in most cases requiring rebuilding, reengineering or even resetting of the former arrangements. Table 1 shows the progress made by the postcommunist states of Central and Eastern Europe and the former Soviet Union in the EU accession process. The table also shows that the situation is quite different between three groups of countries: -postcommunist countries of Central and Eastern Europe outside the former Soviet Union + 3 former Baltic republics of the Soviet Union -former Yugoslavia and Albania -other former republics of the Soviet Union, currently falling under the European Neighborhood policy -stable institutions that guarantee democracy, the rule of law, human rights and respect for and protection of minorities; -a functioning market economy, as well as the ability to cope with the pressure of competition and the market forces at work inside the Union; -the ability to assume the obligations of membership, in particular adherence to the objectives  of political, economic and monetary union  Latvia  1998  1999  2004  Lithuania  1998  1999  2004  Poland  1994  1998  2004  Slovakia  1995  1999  2004  Slovenia  1999  1998  2004  Bulgaria  1995  1999  2007  Romania  1995  1999 (2007) The first group has now completed the EU accession process, while the former Yugoslav republics (except for Slovenia) as well as Albania are at various stages on the formal path to membership. The third group has not yet really started on the path towards membership and it is not certain it ever will.
The European Union established the Stabilization and Association Agreement partnership to help the countries of the Western Balkans region move towards satisfying the criteria for the EU membership. In June 2005, the European Council reaffirmed its commitment to the process, noting that each country's progress towards European integration depends on its efforts to comply with the Copenhagen criteria and the conditionality of the Stabilization and Association Process. According to the European Commission (European Commission, 2006) further strengthening of the SAA partnership policy have proved an effective policy framework for EU action on the way of the Balkan states to their future accession. The geopolitical logic behind southeastern accession is to progressively absorb the entire Balkan region into the integration and Europeanization process. In the meantime, EU has largely taken responsibility for the whole region already by providing considerable financial and technical aid, with focus on developing administrative capacity, building its potential to effectively utilize the provided aid and opportunities. So far, Croatia and Macedonia have been allowed to become official candidates, while the others have signed Stabilization and Association Agreements. However, even within these groups, the EU makes some careful distinctions. For example, while Croatia is already negotiating the accession, Macedonia's progress has been blocked by its row with Greece over the country's name. Even within the SAA group, Albania's and Montenegro's agreements have not become effective yet. This allows the EU to pursue a strategy of highly graduated pathway, which allows it to fully utilize its conditionality strategy (see the next section).
The EU remains selective in offering the perspective of membership officially. Looking further east, the Commission claimed that at present situation, Belarus is too authoritarian, Moldova too poor, Ukraine too large and Russia too scary for the EU to contemplate membership even in the distant future (Grabbe, 2003). The European Neighborhood Policy (ENP) was developed in 2004 to prevent the emergence of new dividing lines between the enlarged EU and its neighbors, offering them a privileged relationship, building upon a mutual commitment to common values (for reference see European Commission, 2004). The countries covered by the ENP include North African states along the Mediterranean coast, as well as former Soviet republics united in confederation of Commonwealth of Independent States (with the exception of Russia and Kazakhstan) and covering the region of Caucasus and Eastern Europe. The ENP is built upon existing agreements between the EU and the partner in question (Partnership and Cooperation Agreements (PCAs) 2 , or Association Agreements in the framework of the EuroMediterranean Partnership). As already noted (and further explored in the section on state aid and competition policy), PCAs are less specific and intrusive than SAAs negotiated with Western Balkans countries in terms of institutional development.

The EU influence prior to the membership
The candidate country does not negotiate with the EU on the substance of the legal framework -the acquis -which must be fully adopted. The only negotiation is essentially about interpretation of the acquis and its sequencing, as well as on the interpretation of the Copenhagen criteria that extend beyond the acquis. The accession of the postcommunist states to the EU was a much more demanding process than for the previous rounds of enlargement since the Copenhagen macro-guidelines were designed primarily to minimize the risk brought into EU by the new entrants becoming politically unstable and economically burdensome (Grabbe, 2003).
The pre-accession process of institutional and economic consolidation is predominantly conducted by the conditionality mode of rule transfer (see Schimmelfenning and Sedelmeier, 2004). The logic of conditionality is underpinned by the external incentives model viewed as the most effective method of rule transfer necessary to meet the requirements of the accession negotiations. (ibid., Börzel 2000).
The dominant logic of the EU conditionality is a bargaining strategy of reinforcement by reward, under which the EU provides external incentives for a target government to comply with its conditions. (Schimmelfenning and Sedelmeier, 2004) of conditionality depends on the credibility and legitimacy of EU intrusion in the aspirant state. The relation between credibility and conditionality is one of linear dependency. The higher the credibility of conditionality, the stronger the bargaining power of EU incentives. From the candidate state point of view, the more advanced a country is in liberalization, efficient privatization, restructuring and international competitiveness, the better is prepared to face challenges of adjusting to conditions for membership in the EU. (Inotai, 2000) The Europeanization process can result in divergence between formal and informal rule transfer and between short-term achievement and long-term sustainability. The formal channels stream the rules through bargainable attributes of credibility and conditionality, while informal channels stream the standards primarily through legitimacy, socialization and internalization effects of learning and lesson drawing modes. (Schimmelfenning and Sedelmeier, 2004) Combination of both qualitiesformal and informal -can thus be successful if external pressure and internal social potential are together sufficient to trigger a spiral of changes and if the changes in the political and economic spheres can reinforce each other in democratization and marketization. (Milanese, 2001) The somewhat vague conceptualization of Copenhagen accession criteria left considerable space for their interpretation by the Commission and the member states. Indeed, the EU has redefined, elaborated, widened and detailed the criteria of what constitutes meeting the accession conditions and added specific requirements for individual countries (such as the emphasis on upgrading administrative capacity increasingly, particularly for the second wave enlargement, and even more so for the Western Balkans countries). It has added new policy areas and further refinements for both the first wave, and additionally hardened the conditions for the second wave of new member states. The "soft-law" coercion will undoubtedly add more requirements to future enlargement, as is already the case of Croatia and other Balkan states (for example, the ICTY compliance and resolution of political status).
With regard to countries of the former Soviet Union, evidence of substantial effects on consolidation of economic institutions is scarce, despite billions of Euro provided in technical assistance during the transition. Trying to achieve institutional change without promising membership is proving extremely difficult. As Light noted: "The EU expects the new neighbors to adopt EU norms in the same way that the accession countries did, but it is not prepared to offer them the same motivation -that is, the prospect of accession, which is what the new neighbors really want." (Light, 2007, p. 15) Mogilevsky (2007) remarks that the difference between effectiveness of technical assistance to the Central and Eastern European countries on one hand and the Commonwealth of Independent States countries on the other can be, to a large extent, attributed to the missing incentives for the CIS states, which has been compounded by the fact that the legitimacy and the desirability of the Copenhagen criteria -building liberal democracy and market economy -have become questioned throughout much of the post-Soviet space.
At the same time, the intensifying desire for the EU membership in Ukraine and Georgia in particular (where it was inextricably linked to the so-called color revolutions), but also in Moldova and, to some extent, in Armenia, have led the EU to construct a conditional and graduated path of institutional consolidation similar to the accession path, but without the explicit promise of future membership. The efficacy of this path has not been tested yet. The path consists of a shift from the Partnership and Cooperation Agreements towards Action Plans under the European Neighborhood Policy. These have been agreed with Ukraine and Moldova in 2005 (for a 3-year period) and the Caucasus countries in 2006 (for a 6-year period) and each contains an extensive, but shallow set of benchmarks (Light, 2007, p. 14). Despite the relatively low level of conditionality (ibid.), only Ukraine has managed to pass on to the next stage so far. This stage should lead to a new agreement on mutual relations including the so-called "deep free trade area", which would contain "harmonisation of the regulatory environment in certain sectors with that of the EU". (Shapovalova, 2008, p. 6) This would mean that the effect of the EU on the Ukrainian institutional consolidation would go beyond that of the WTO membership conditions. However, the negotiations on the overall agreement started in early 2007 have not yet led to any conclusions (even though the FTA negotiation themselves started in February 2008).

The EU influence after the accession
Once some of the postcommunist countries became members, Europeanization as a process shifted to a new shape: circular rather than unidirectional, and cyclical rather than one-off (see Dyson and Goetz, 2003). The former pre-accession relationship that relied on coercive instruments and enhanced the superiority of the EU are being supplemented by new forms, and the one-way reception of EU laws and policies (Bulmer and Burch, 2001) is being replaced by softer forms of integration. (Dyson and Goetz, 2003) The post-accession balance, attained by incorporating formal accession requirement into original institutional structure of candidate states, is transitory and as such is provisional in nature.
The post-accession experience is providing some ground for the view that the EU conditionality in general provides a relatively blunt instrument that may create some effects, but does not necessarily lead to deep and lasting institutionalisation. Once full membership status is granted, the states begin to lack equally stimulating rewards for compliance. Consolidation in the new member states linked to coercive adaptation, has been followed by short-term tactical calculations. (Goetz, 2006) As a result, the initial institutionalization produced only shallow institutional consolidation. (Goetz, 2002) The post-accession situation re-opens the previously exported standards to modification that can either proceed to further consolidation of EU standards or recede to traditional institutional and economic domestic patterns.
Of course, the EU is not powerless in face of such backsliding. The standard procedure for coercive behavior vis-à-vis member state is the so-called infringement proceedings, where the European Commission can refer a lawsuit against a member state deemed in violation of its legal commitments to the European Court of Justice for redress. As the case study on the civil service will show, this instrument is essentially irrelevant for requirements imposed without a clear reference to the legally binding acquis.
To compensate for this weakness, the EU has been negotiating more extensive asymmetric powers vis-à-vis the new entrants for the second wave Bulgarian and Romanian accession in 2007 and will undoubtedly continue to do so for future enlargements. This included both enhanced powers of the Commission with regard to the European funding provided to the member states as well as the more extreme and temporary power to suspend recognition of judicial decisions by the new member states AFTER the accession if their judicial system is deemed to be unsatisfactory. The Commission in July 2008 to an unprecedented degree used the former power when it stopped the whole Bulgarian Structural Funding process because of fraud suspicion. According to OLAF -EU anticorruption organ established in 1999 -a criminal organization (which included 50 domestic and foreign companies, plus businessmen, with ties to prominent politicians) in Bulgaria abused about 32 millions Euro. (OLAF, 2008) Both Bulgaria's Paying Agencies were suspended and lost their licenses.

Direct, powerful and sustainable EU influence: State aid and competition policy
The first case concerns state aid and competition policy. In the EU policy and legislation, these issues are closely intertwined and linked to one of the key objectives of the Union -to ensure a smooth functioning of the single European market. According to this logic, private and public actions that might distort or hinder competition in the single market need to be vigorously opposed. Additionally, since many of those actions could be taken, assisted or approved by national governments, strong powers for the European level are needed to make sure that the rules are enforced even in cases where there is a national political interest not do to so.
The EU legislation, particularly in Articles 81, 82, 86 and 87 of the Treaty, closely regulates both state aid and competition. Specifically, it subjects all state aid to businesses that can impinge on trade between member states to scrutiny by the Commission unless it fits into one of the so-called block exemptions agreed to by the Commission and the member states. Additionally, both the Commission and the member states have an obligation to prosecute cases of anticompetitive behavior by businesses and the power to approve or forbid mergers depending on their impact on competition in local, national and European markets.
The competition and state aid powers are important both in obvious and subtler ways. For example, issues such as subsidies for public transport, ability of various types of organization to receive Structural Funds or the transparency of relationships between public authorities and real estate developers have all been impacted by these rules, generally acting to limit discretion of local or national authorities in such matters. These powers have enabled the EU, and particularly the Commission, to shape domestic policies of the member states in far-reaching ways. (Paroulková and Zemanovičová, 2009) From a legal point of view, some of these powers (particularly those related to the authority of national competition institutions) are implemented indirectly -through transposition of EU directives into the national law -which is the more frequent approach to the EU legislation. However, the European-level policy in this area is promulgated through direct legal instruments -regulations, which are directly enforceable and applicable in each member state.
Since the process of association with the EU and the subsequent accession usually involves an early and largely unfettered access to the single EU market, 3 the associated states gain, to some extent, de facto membership in this regard as soon as they gain market access approaching that of the official members (and comparable to the status of closely aligned non-members such as Norway or Switzerland). As a result, the Union insists on transposition of the institutions related to competition early on and they are usually embedded already in the association agreement. However, it would be legally and politically difficult to apply the EU legislation directly in the non-member state, be it in form of regulations or individual administrative decisions. Therefore, the solution chosen is for the EU to require the associated state to mimic these institutions.
A good example is the Association Agreement between the Union and Montenegro signed in 2007 (though not yet effective). In Article 73, it lists unacceptable practices (i.e. anticompetitive behavior and state aid) and binds Montenegro to use EU standards in their assessment. It however stipulates that the associated country shall establish "operationally independent public body" to do the actual assessment. (Montenegro, 2007) This practice was also followed for the postcommunist countries that had signed association agreements earlier -in the 1990s and early 2000s. The EU concept of operational independence means only that the decision of the authority is not subject to change by the government. However, personal independence, particularly as related to appointment and dismissal, is crucial for effective independence (see Beblavý, 2007), but this has not been well defined with regard to the association agreements.
It should be mentioned that the EU practice was in any case a natural role model for domestic legislation on competition policy even prior to the signature of association agreements. Therefore, the stylized progression in institutional design and consolidation in this area has been as follows (Paroulková and Zemanovičová, 2009): • Stage 1: in early transition, competition authorities are established, usually modeled on what is perceived as the EU best practice, with the EU involvement in an advisory capacity and largely based on bilateral links with national authorities in the EU states • Stage 2: the association agreement brings a more rigorous definition of anticompetitive practices based on the EU legislation and brings state aid into the picture. However, the legislation continues to be implemented by domestic institutions • Stage 3: the EU accession means demise of domestic state aid institutions, with all powers being transferred to Brussels. The domestic competition authority continues to deal with national and local issues, but the Commission can preempt cases of multinational nature or even cases of national nature where the national authority is unable to make headway.
As a result, this is an example of a policy area, where the EU institutional influence has been long-term, direct, sustainable and powerful. However, even in such a case, the progression towards institutional consolidation has been very gradual and depended, at all stages, at the continuing EU authority.

Where EU and Europeanization go hand in hand: Central bank independence
The second case study looks at central bank independence. Its development is a good example of a process where the initial influence on transition countries was exercised more through the overall Europeanization than through formal EU requirements, but in the later stages the formal EU requirements took over to complete the process of institutional consolidation. At the same time, the need to review legislation opened an opportunity for politicians to restrict the central bank independence in ways that did not run afoul of the EU legal requirements as embodied in the Maastricht Treaty.  -Hungary in 1987, Poland in 1989and Czechoslovakia in 1990. In 1990and 1991 to the democratization of the political regime, change to the central bank law was considered vital in all four countries. In Poland, several substantial amendments of the NBP law occurred in the early 1990s. In Hungary, a new law became effective in December 1991. In Czechoslovakia, the new law came into effect in February 1992. These laws were generally based on the Bundesbank law and the law on the Austrian National Bank. (Siklos, 1994) They were crucial in establishing a relatively high level of independence. Principal features of these laws conducive to independence were the following: • relatively long terms of office Under these laws, central banks of the region became powerful political and economic players that played crucial roles in economic reforms and the preservation of macroeconomic stability during the following decade. During the following period of the early and mid 1990s, the legal position of central banks in all four countries remained largely unchanged. New laws were passed in the Czech Republic and Slovakia following the dissolution of Czechoslovakia, but they largely reflected the Czechoslovak law. In Hungary, minor changes were passed that marginally increased central bank independence.
In the late 1990s and early 2000s, major changes in central bank law were passed in all four countries in order to achieve compliance with the Maastricht Treaty. The Maastricht Treaty requires all states that wish to join EMU to grant considerable central bank independence. This is because the national central banks together exercise dominant influence in the decision-making of the European Central Bank and so the mandate and position of national central bank leaders became an issue of European interest. Even though central banks in Central Europe had already been quite independent in terms of personal independence, the following changes were usually required to make them fully compatible with the Maastricht Treaty: However, the need to satisfy the Maastricht criteria and to amend the central bank acts opened a 'Pandora's box' -an opportunity to change the statutes in other ways. Political players in the Czech Republic, Poland and Slovakia used it to weaken the actual autonomy of the central bank and the autonomy of its career officials from the current governing majority or even attempted to use it to weaken overall central bank independence. (However, proposals included in the latter category were usually defeated eventually as they would contradict the EU accession criteria.) This can be seen as a backlash against central banks, which were perceived as too powerful and too independent. The weakening of central bank autonomy usually meant changes in the make-up of the bank board (see Beblavý, 2007): • in Poland, key decision-making powers were switched from a bank board made up of bank officials to Monetary Policy Council, where the bank president is only one of ten members and the remaining nine are outsiders appointed by three centres of political power in the country. • in the Czech Republic, the power to appoint the bank board was switched from president to several centres of power (both houses of parliament and the president) though this was later reversed • in Slovakia, three out of eight members were explicitly designated as external members and the ultimate veto power of the Cabinet over all appointments for the board was preserved • in Hungary, a similar veto power of the prime minister remained as well There were also ultimately unsuccessful attempts to circumscribe central bank independence, which reached various levels of approval (by the government, by one house of the parliament etc.). These included, for example, in the Czech Republic, a proposal that the operating budget of the bank should be approved by the parliament or, in Poland, a proposal that those nominating bank board members should have the power to recall them at their own discretion.
To summarize these developments in a temporal perspective, we present, in Chart 1 developments of legal independence assigned within the widely used index created by Cukierman (1992). During transition, the legal framework for central banking in Czech Republic, Hungary, Poland and Slovakia underwent repeated and significant changes. To provide a comprehensive picture, the evaluation is limited to major developments where the provision influencing the index changed. 4 The chart ends in 2001 and later changes in the central bank law are not analysed as they have not changed the scores in any significant and, by 2001, these countries have finished the process of legal approximation.
Chart 1: Central bank independence in Central Europe as measured by the Cukierman index -1993Cukierman index - -2001 Source: Beblavý (2007) As we can see, with the exception of Poland, the stylised fact about independence is the high initial level, which was slightly increased at the end of 1990s and beginning of 2000s to comply with the Maastricht criteria. Poland experienced a more dramatic increase in 1997, reflecting both lower base and the decision to change the institutional set-up radically.
To get a more international perspective, Table 2 presents comparison of legal central bank independence produced for the Central European countries with the developed countries. The comparison is between values in Central European countries in 1990s and early 2000s and values in developed countries during 1980s. The reason for temporal mismatch is that the data for industrial countries in 1990s are not always available and also reflect changes implemented in the run-up to EMU. Therefore, the current values for western European EU members would essentially match the current ones for the new member states. It is clear that the level of legal independence enjoyed by central banks in Central European countries was very high from the very start. In the Cukierman index, all the value for Central European countries are above any value of developed countries during 1980s, including West Germany and Switzerland.  Cukierman (1992) and Beblavý (2007) Overall, the case study reveals that the prospect of EU and EMU membership was important in consolidating central bank independence in Central Europe, but that a relatively high measure of independence preceded the EU accession negotiation. The accession itself was more of a double-edged sword for central bank independence than appears at the first glance. While the EU accession was undoubtedly the driver behind the total ban on central bank credit to government, it also opened the 'Pandora's Box', which, in the Czech Republic and Slovakia in particular, allowed a decrease in CBI measured by legal indices in other areas.

Limits of the EU power: the civil service reform
The last of the three case studies examines the EU influence on developments of the legal and policy framework for the civil service. The importance of the civil service and the overall administrative capacity for functioning of the economic institutions is obvious. Nonetheless, bulk of the accession process has traditionally been about transposition of legislation, with very little attention paid to the actual capacity for its implementation. This changed during the accession negotiations of the postcommunist states. As Sedelmeier (2006) notes: 'A novelty in the context of eastern enlargement is that the EU has undertaken considerable efforts to influence administrative practices in candidate states, including civil service reforms, or anticorruption… and there is evidence of considerable changes in formal rules induced by conditionality' (p. 16). While these requirements were not part of the formal acquis, they have gained in prominence in preparation of the 2004 accession of the ex-communist states of Central and Eastern Europe due to scepticism about their administrative capacity and also the Copenhagen criteria for membership, particularly as interpreted by the Madrid summit in 1996. (Grabbe, 2001, Laffan, 2005 Since the EU had no clear benchmarks for administrative capacity (Nicolaides, 1999), the pressure for changes in these areas, were embodied by the push by the Commission to create independent civil service regulators in the accession countries and to create formal safeguards for depoliticization.
The formal vehicle for the pressure has been the Commission annual reports on the readiness of candidates for the accession, linked also to SIGMA reports (Laffan, 2005). 5 This monitoring plays a key role in decisions on the conduct of negotiations. These negotiations are an open-ended process, the outcome of which cannot be guaranteed beforehand. Since there is no acquis on civil service, the EU conditionality on service reform has been subsumed under one of the general Copenhagen criteriademocracy and the rule of law. This meant that the Commission had substantial leeway in its interpretation of the criterion, but also that there was very little it could do if the countries backtracked after the accession (whereas for commitments embodied in the EU legislation, there is a judicial redress analyzed above).
In practical terms, during the pre accession period, significant attention was paid by the European Union to ensuring personal autonomy of civil servants in the acceding states (Collins, 2002, Verheijen, 2002, with focus on creating distinct civil service with formally clear division lines between political and administrative positions securing depoliticization of the administrative level of civil service, among other criteria (see Beblavý, forthcoming). The last years before the EU accession saw a move in most candidate states towards the formalization of politico-administrative relations and the introduction of constraints to politicians' freedom to appoint and dismiss senior officials at will, particularly by creation of independent regulatory authorities for civil service. (Verheijen, 2007). However, as Table 3 shows, the European Commission was not equally successful with this pressure in all the states, with Estonia and Hungary, for example, successfully resisting creation of such a regulator.  (2001 -present) under the authority of the Minister of State Administration of Council of Ministers; no civil service regulator in place politicization is high in the administration because civil servants depend on the discretion of managers, who are often political appointees; the distinction between political and administrative positions is unclear

Romania
The National Agency for Civil Service (2001 -present) The regulation of civil service remains underfinanced, underestimated, and understaffed failing to function as foreseen by the Law originally envisioning the Civil Service reform. Politicization remains strong on local level, but is still present through entire civil service structure.
Source: Author, Hacek (2006), Meyer-Sahling (2006), Verheijen (2007), OECD/SIGMA Public Service and Administrative Frameworks Assessment and Partner Country Surveys More striking is what happened following the accession. In just 3 years after the accession of the first eight postcommunist countries, there was a significant amount of reversal towards pre-accesssion practices. By 2007, there was not a strong and independent civil service regulator in any of the eight postcommunist countries that joined the EU in 2004. In three countries where it had been created or legislated (Czech Republic, Poland, Slovakia), it was either abolished (Slovakia), lost its independence (Poland) or never actually created (Czech Republic) after the accession removed the political pressure to do so. The same developments can be observed with regard to safeguards for protecting senior administrators from politicization. Even in Hungary, where no explicit legal change took place, the politicization was on the rise (Meyer-Sahling, 2006, Meyer-Sahling, 2008. In Slovenia, despite legal change towards depoliticization, this did not change the reality of a highly politicized civil service (Hacek, 2006). Therefore, while there is a differentiated performance in this respect among the new members, the formal pre-accession conditions had weak or no effect. On the contrary, where the changes were most strongly linked to the accession-Czech Republic, Poland, Slovakia -the reversals have been most dramatic.
The European Commission has accepted that the first wave states meet the requirements of the acquis and seems disinclined to apply pressure in this area following the accession. (Verheijen, 2007) Although there is some attempts to keep up the pressure with the second wave states (Bulgaria, Romania), in the absence of a binding acquis or other clear standards applicable across all EU members, any special conditions are likely to be only temporary in nature and subject to the same reversal later on (unless one assumes they will become so entrenched in another 1-2 years that this will change the domestic political economy, which is unlikely. For example, in Poland, the independent regulator was stripped of its autonomy after nearly 9 years of operation).  (2000 -present) is an independent supervisory body responsible directly to the parliament Civil service law lacks mechanisms effectively preventing the politicization of recruitment; the powers of Civil Service Agency are limited, which weakens its function as independent depoliticization safeguard;

Montenegro
Human Resources Management Authority currently under the Ministry of Interior, Administration and Selfgovernments (2004 -present), in position of advisory and implementation body The implementation of the civil service legislation has not been followed through consistently, due to cumbersome or unclear regulations and partly due to simple non-application resulting from the absence of sanctions, the HRM Authority lacks monitoring capacities in practice, the merit system remains not defined legally and unimplemenented At the same time, the Commission, together with SIGMA, continue to apply a similar methodology to countries of the Western Balkans preparing for the accession. Civil service reforms have been launched, with civil service authority formally present in all countries. However, according to recent Commission reports (European Commission, 2008) the legal basis for building a modern and professional civil service is still incomplete and those already in place are not implemented systematically. There has been some attempt to broaden the benchmarks for analysis from formal, legalistic ones to actual assessment of human resource management systems as can be seen from the assessment in Table 4, which has largely been derived from SIGMA reports. Even this shift cannot hide the fact that any pressure based on conditionality suffers and will continue to suffer, by definition from the problems that plagued attempts at externally driven civil service reform in the new member states.

Conclusion
The paper explores implications of the EU accession process for consolidation of institutions important for sustainable economic growth in the states aspiring to be EU members. The EU influence on consolidation in the countries in transition has been formally wielded primarily through conditionality tied ultimately to the EU membership. However the potential of membership to lock in EU standards has proven questionable in cases when it is not supported by continuing coercive ability of the Union. It appears that the stabilizing effect of conditionality has its limits and weaknesses, and the ability of the EU to intervene depends on the legal status of the commitments.
The paper looked at three specific areas -competition and state aid, central bank independence and civil service reform -where the EU accession process has involved substantial institutional change. In case of state aid and competition policies, the EU institutional influence has been long-term, direct, sustainable and powerful. However, even in such a case, the progression towards institutional consolidation has been very gradual and depended, at all stages, at the continuing EU authority. For central bank independence, the prospect of EU and EMU membership was important in consolidating central bank independence, but a relatively high measure of independence preceded the EU accession negotiation. The EU accession itself was undoubtedly the driver behind the consolidation of independence along some axes, but it also opened the 'Pandora's Box', which, in the Czech Republic and Slovakia in particular, allowed a decrease in the central bank autonomy. Regarding the EU attempts to mandate civil service reforms in the acceding countries, particularly creation of independent regulators and creation of clear demarcation lines between political and administrative level, the absence of anchoring in the EU legal framework meant that initial success in the pre-accession periods has been followed by significant backsliding after the membership was achieved.
Overall, the conclusion is that the EU influence on institutional design and consolidation for countries aspiring to membership has been powerful when the Union chose to make an issue a priority one, be it because of its own legal framework for its own members (competition, central banking) or as a specific priority vis-à-vis future members (administrative capacity). However, the paper also showed that when the priority is not anchored in firm legal framework and not applied consistently for all existing members, the sustainability is threatened and the membership can easily bring backsliding. Therefore, the value of the Union as an agent of institutional consolidation appears to be precisely in the indefinite and rules-bound aspect of the EU membership. Where the EU accession is a distant and uncertain prospect (e.g. Caucasus, Moldova, Ukraine), the EU's attempt to use conditionality without explicit promise of membership have, so far, not been successful in achieving its objective. The innovative idea of using the so-called deep free trade agreement as a substitute might be a partial solution, but remains practically untested in implementation.