Scholarly article on topic 'Embracing European Law: Compliance with EU Directives in Central and Eastern Europe'

Embracing European Law: Compliance with EU Directives in Central and Eastern Europe Academic research paper on "Political Science"

0
0
Share paper
Academic journal
European Union Politics
OECD Field of science
Keywords
{""}

Academic research paper on topic "Embracing European Law: Compliance with EU Directives in Central and Eastern Europe"

European Union Politics

DOI: 10.1177/1465116508093490 Volume 9 (3): 379-402 Copyright© 2008 SAGE Publications Los Angeles, London, New Delhi and Singapore

Embracing European Law

Compliance with EU Directives in Central and Eastern Europe

^ Dimiter Toshkov

Leiden University, The Netherlands

ABSTRACT

Accession to the European Union (EU) demands the adoption of a vast body of legislation. This paper analyses compliance with EU directives in eight post-communist countries during the Eastern enlargement and tries to account for the puzzling embrace of EU law in Central and Eastern Europe. Drawing on a new data set tracking the transposition of a sample of 119 directives, the paper finds effects of both political preferences and government capacity on the likelihood of timely transposition. Furthermore, important sectoral differences are uncovered, with trade-related legislation having a better chance and environmental legislation having a significantly worse chance of being incorporated into national legal systems on time. Beyond the conditionality of the accession process, the paper unveils a complex causal structure behind the ups and downs in transposition performance.

KEY WORDS

■ Central and Eastern Europe

■ compliance

■ Enlargement

■ implementation

■ transposition

Introduction

The transformations of government and society in Central and Eastern Europe (CEE) since the beginning of the 1990s have proven a fertile ground for the study of institutional persistence and change (Elster et al., 1998; Grzymala-Busse, 2006; Stark and Bruszt, 1998). More recently, the confluence of these reforms with the European integration of the region presented a rare chance to witness how internal impulses and external pressure for change combine to produce patterns of political and policy reforms (Dimitrov et al., 2006; Vachudova, 2005). This paper deals with the impact of the European Union (EU) on law and policy-making in CEE. Analysing the adoption of EU law in the new member states, I explore and try to explain the variation in the pace of implementing EU directives during the last enlargement in eight post-communist countries: the Czech Republic, Estonia, Hungary, Lithuania, Latvia, Poland, Slovakia and Slovenia.

States aspiring to become members of the EU have to incorporate into their national legal systems more than 80,000 pages of legislation in the course of a few years.1 Given the Herculean proportions of this task, and the systematic troubles of the 'old' member states to transpose EU law on time, it is a genuine puzzle (and one that has remained largely unnoticed) that the post-communist countries have been rather successful in adapting to EU law. As Figure 1 shows, the mean transposition deficit in the new member states (EU-10) is lower than the average percentage of non-transposed directives in the EU-15. Furthermore, as of 2007 the best-performing countries are new entrants and the worst performer amongst the EU-10 still does better than the worst EU-15 country.

The contributions of this paper go beyond drawing attention to the unlikely embrace of EU law in CEE. The performance of the new members is not uniform and important differences across countries, policy sectors and types of directives exist. In order to address this variation, I perform a quantitative analysis of a new data set tracking the transposition performance of the eight countries from the start of the accession negotiations in the late 1990s until 2005 (a year and a half after the actual enlargement). Using logistic regression, I attempt to capture simultaneously the effect of EU-level factors, the national institutional context and the changing landscape of political preferences in CEE on the likelihood of timely transposition.2

National government preferences appear related to the rate of incorporating EU law. Both government positions on a socioeconomic left/right scale and attitudes towards EU accession have a significant effect on the likelihood of timely transposition. The capacity of governments to transpose is, however, constrained by a high number of parties in government and bureaucratic

EU15 (min) EU15 (mean) EU15 (max) -EU10 (min) -EU10 (mean) -EU10 (max)

Figure 1 Transposition deficit in the EU (2004-2007). Mean (14.34) and maximum (40.4) values for EU-10 in 2004 are truncated. Own data compilation based on the European Commission's Internal Market Scoreboards.

quality, as well as by the regulatory environment in a country. Directives related to trade barriers have higher chances of being implemented within the deadline, whereas implementing legislation, highly politically complex laws and costly directives are more likely to suffer delays.

The findings of the analysis have considerable societal relevance because the transposition of European legislation is the basis of the internal market. A detailed knowledge of the state of implementation and the reasons behind the ups and downs in the member states' performance is valuable for policy experts, as well as for social scientists. The comparative study of the adoption of EU law in CEE presents a new setting for applying and testing the growing number of implementation/compliance theories, but it also provides insight into how policy-making in transition settings works.

The article is structured as follows. The next section places the current text in the context of EU enlargement, transposition and implementation studies. The discussion of the contributions and shortcomings of existing work is followed by a section explicating the theoretical arguments investigated in the empirical analysis. Next, the operationalization, data collection and research methodology are presented in some detail. The subsequent section contains the results of the empirical analysis. Lastly, the conclusion sums up the findings.

Theorizing transposition in the context of enlargement negotiations

The problem of adapting to EU law in CEE is of major interest for two research agendas: understanding the unfolding and the impact of the last EU enlargement round, and understanding compliance in a multi-level system of governance. The intersection of these two research areas provides the immediate intellectual context of the current study.

Relatively few studies directly address the question of transposing/ implementing EU rules in the new member states. Studies that are built on a comparative approach or a systematic empirical analysis are even harder to find. One of the few exceptions is the study by Hille and Knill (2006), who construct a measurable indicator of implementation performance in CEE for the period from 1999 to 2003 and try to match the resulting patterns with the bureaucratic strength of the governments and the national political constraints. The authors conclude that 'the functioning and the quality of the domestic bureaucracy constitute crucial preconditions for effective alignment with EU policy requirements' (Hille and Knill, 2006: 549). The study, however, employs a rather indirect measure of implementation performance - an indicator measuring 'the frequency, direction and intensity of criticism or approval of the candidate countries' performances' as expressed in the European Commissions' reports on the progress towards accession (Hille and Knill, 2006: 541). A further shortcoming is the possible endogeneity of the dependent variable and the major explanatory factor - bureaucratic quality as measured by the expert ratings collected for the World Bank governance indicators. Although the use of expert scores to assess various aspects of governance is not problematic as such, the analysts' assessments are likely to stem at least to some degree from the criticisms or approvals contained in the Commission reports.

The relatively strong explanatory power of administrative efficiency vis-à-vis party political preferences and institutional factors is also uncovered in an analysis of the transposition and implementation of social policy directives (Toshkov, 2007). Measuring the legislative 'productivity' of the CEE applicants in each six-month period from 1998 to 2005, the study finds administrative efficiency to have a strong and positive effect, government EU integration support to have a smaller but still significant impact and left/right ideological positions to have no effect at all. Owing to the peculiarities of the social policy sector, however, it is an open question whether these results can be generalized to the entire spectrum of EU activities.

The adaptation in the social policy field has also been analysed in depth in the cases of Poland (Leiber, 2007) and Slovakia (Dimitrova and Rhinard,

2005). Leiber shows the good performance of Poland in terms of legal implementation and emphasizes that the conditionality of accession is a major driving force. Interpreting the Slovakian case, Dimitrova and Rhinard find evidence for the impact of norm compatibility.

Comparative case studies possess clear advantages in addressing the issue of cross-sector and cross-country differences in implementation performance. Andonova (2004) examines the impact of the EU on environmental policies in Bulgaria, the Czech Republic and Poland. Analysing the performance in several subsectors, the author reveals the crucial role of domestic industry interests in successful adaptation and the role of the political institutions in furthering or constraining the change.

Exploring variation in a single country over time is another way of addressing the causal structure behind transposition/implementation performance. Working with transposition data from Poland during the time of accession, Zubek (2005) finds that the crucial factors that speeded up the process of adopting the acquis were the centralization of authority in the executive, manifested in the exercise of strong political leadership as well as the centralization of coordination structures, and the improvement in administrative capacity. In addition, the dominant role of the executive vis-à-vis the legislature has enhanced performance in Poland.

The literature reviewed provides some analytical leverage in regard to the problem of explaining the adoption of EU law in CEE. The strong positive impact of bureaucratic quality and administrative efficiency and the focal role the executive plays seem to be corroborated by most studies. Political preferences and the institutional structure are rarely found to be important, at least at the aggregate level. The main shortcoming of the literature is the limited ability to provide generalizations. The failure to go beyond policy sectors, countries and rather short time periods can be traced to the lack of comparable, reliable and valid measures of transposition performance.

The main theoretical focus of the more general literature on Euro-peanization in CEE (for a review see Sedelmeier, 2006) has been the influence of accession conditionality: the power of the EU to demand reforms in exchange for the prospects of accession to the Union (Dimitrova, 2002; Grabbe, 2003; Schimmelfennig and Sedelmeier, 2005). The impact of con-ditionality, however, cannot explain the cross-national variation in the transposition performance of the CEE countries. Furthermore, it is an open question whether the finer details of policy changes in the region can be related to the conditionality applied by the EU. What we need in order to explain the patterns of transposition in the region is a framework that incorporates country- and policy-level influences, as well as EU-wide pressures.

Hypotheses

Implementation performance depends on the willingness and capability of the relevant domestic actors to 'download', process, interpret and adapt European rules and adopt the corresponding national legislation (Borzel and Risse, 2003; Haas, 1998; Jensen, 2007; Tallberg, 2002). Cast at such a general level, this thesis can accommodate all of the factors suggested by the existing literature (and probably all conceivable factors). Nevertheless, it is useful as a starting point in order to structure the theoretical discussion. Also, since the implicit causal model is quite complex,3 I will present the variables according to the level at which they operate (national, EU, policy sector).

National level: The willingness to comply

Who the relevant actors are, what their relevant preferences are and what affects their capabilities are the main questions the theoretical argument has to address. In the context of accession negotiations, the domestic actors that have the greatest potential to affect adaptation outcomes are the government in power and the bureaucracies they steer. Although the CEE countries inherited from their communist past weak executives without strategic policy-making capabilities, it is the governments that conduct the negotiations and are held responsible by their electorates for the results of the talks and by the EU authorities for honouring their commitments. Although the degree of centralization of EU-related power by the executive (and within the executive) varies, the governments in power are the main actors in regard to the process of legal implementation of the directives of the EU (Dimitrov et al., 2006; Goetz and Wollmann, 2001).

How do the ideological profile and positions of governments affect the likelihood of EU rules being transposed on time or not? Transposition takes time. The European law has to be identified, translated, consulted and adapted. Each step of the process demands time and resources, and this implies missed opportunity costs for the government to deal with other issues. Because directives allow a certain amount of discretion, even more time and resources can be spent on bringing the provisions of the directive closer to the policy preferences of governments. In short, the less the provisions of the law coincide with government preferences, the longer a government will take to transpose a piece of legislation and, by implication, the lower will be the likelihood that it will be transposed on time. Unfortunately, it is practically impossible to get information on the fit of each and every EU law with national government preferences. At a higher level of aggregation, however, it is reasonable to assume that the respective governments' positions towards European

integration and the prospects of accession should be positively related to their assessment of individual EU rules. Furthermore, especially in the context of accession, government support for the EU should have an additional positive impact on the speed of transposition, because the governments will be more willing to sacrifice their substantive policy positions in order to fulfil the requirements of the EU and secure accession.

The link between societal support for the EU and implementation outcomes has been proposed in the past (Lampinen and Uusikyla, 1998; Mbaye, 2001). The effect of membership benefits, a closely related theme, has also been explored (Perkins and Neumayer, 2007). The distinctive feature of the hypothesis tested in this article is that it looks specifically at the impact of government preferences.

It is more difficult to hypothesize the impact of socioeconomic left/right government positions (for a preference-based explanation of compliance, see Thomson et al., 2007). The body of EU legislation is quite heterogeneous and it is difficult to theoretically derive an expectation for their impact. Still, to the extent that a lot of the acquis implies trade liberalization and the retreat of the state from certain industries (e.g. telecommunications, transport, energy), left (socialist) governments can be expected to be less happy with the implications of these reforms and thus to postpone the adoption of implementing national legislation (Linos, 2004).

H1: Pro-European and economically liberal government preferences increase the

probability of timely transposition.

National level: The capability to comply

The willingness of governments to adopt and implement EU laws is constrained by their capability to do so. Several sources of limited capacity can be identified. Some of these are connected to internal government factors, others are conditioned on the wider public administration system, and yet others are related to the different types of EU legislation to be adopted in the course of enlargement.

Government capacity can be conceptualized and operationalized in several ways. First of all, when governing in a coalition, political parties have to reconcile different political preferences. The more distant government partners are, the more difficult it is for them to direct reforms and adopt legislation, including EU-related acts. Whereas single-party majority governments have fewer hurdles to overcome to enact their proposals, multi-party cabinets have to accommodate more actors. The constraining effect of coalition government can be expected to reveal itself in various ways. The number of parties in government and the preference distance between the most extreme

partners provide different possibilities to measure and test the proposed impact of government capacity.

H2: Higher government capacity increases the probability of timely transposition.

The number of parties in government and the ideological distance are closely related to the concept of veto players (e.g. Kaeding, 2006; Steunenberg and Rhinard, 2006). However, given the broad institutional similarities in the CEE region, the number of parties in government captures de facto the variation in veto points. Only the Czech Republic and Poland have a bicameral parliament and the involvement of presidents in the legislative process is broadly similar in all the countries concerned. Also, none of the countries has a federal structure. Other national institutional features such as established systems of relationships between government, labour and business do affect the capability of government to steer through, and later implement, institutional and legislative changes (see Kaeding, 2006). However, in the case of the CEE countries there is no accepted measure of the degree of corporatism at the national level - a shortcoming limiting the opportunities to explore the argument in this paper.

Governments do not operate in an institutional vacuum. What they can achieve is conditioned on the broader governance system and the capabilities of the public administration. Especially for the case of transposition in post-communist Europe, it seems quite likely that the bureaucracies will have a constraining effect on the success of EU-related reforms. At the start of accession negotiations, all the candidate countries from CEE were still experiencing the consequences of the political, economic and social changes resulting from the collapse of the communist regimes (Elster et al., 1998). Their bureaucracies, simultaneously agents and subjects of reform, had to adapt to functioning in a completely different environment (Dimitrova, 2002). An understanding of the implementation performance therefore has to take into account the political institutions and the public administrations of the applicant countries. All else being equal, in countries and time periods of higher administrative efficiency, the chance of EU legislation being adopted within the deadlines should be greater. In a different twist of the same argument, regulatory quality should also be positively related to successful approximation.4

H3: Higher administrative efficiency and better regulatory quality increase the

probability of timely transposition.

EU level: Characteristics of the laws and policy sectors

The hypotheses discussed so far concern (slowly changing) attributes of the applicant countries. The causal process behind transposition, however, is

more complex and combines the influence of country-, policy- and directive-specific variables. The policy sector and individual directive-specific aspects can be linked to the capacity of being able to implement EU laws in a timely and proper way.

Laws that have a wider field of application, affect politically salient issues and deal with complex matters demand more from governments and administrations (see Mastenbroek, 2003; Kaeding, 2006). On the other hand, implementing legislation that only further develops and specifies framework rules already applied should create fewer problems and require fewer resources to be transposed. Policy sectors regulated by complex and costly rules are likely to be subject of transposition delays. The implied costs of legislation should be a significant hurdle for the timely adoption of national transposing measures and the implementation of the European legislation.

H4: The political complexity and implied costs of legislation decrease the

probability of timely transposition.

Finally, trade relations between the applicant countries and the EU underpin the entire integration process and they should have an effect on the transposition of the acquis because a large part of the EU body of rules concerns product standards, barriers to trade and other measures that regulate the trade exchange with the Single Market. Countries that do not transpose and implement this type of legislation fast will suffer from restricted access to the markets of the Union and thus face significant incentives to approximate their legislation to the acquis. Further, the removal of barriers to trade is a process of negative integration and, as such, involves fewer adaptation costs for the countries.5

H5: Trade-related legislation is more likely to be transposed on time.

This set of hypotheses does not represent a fully integrated theory of transposition. Rather, it adapts propositions introduced in the literature on Europeanization and compliance to the case of transposition during accession negotiations and combines them to take account simultaneously of the will and the capability of governments and bureaucracies to implement EU law. In the remainder of the paper these hypotheses will be operational-ized and the results of their empirical test will be presented.

Data and operationalization

The empirical analysis is based on a new data set derived from CELEX and other sources. In this section the operationalization of the concepts and the hypotheses will be described briefly, as well as the measurement of the

variables, the sources of the collected data, the structure of the data set and other related issues.

Sample selection

A quantitative analysis of transposition performance requires more data than the readily available snapshots of transposition deficits published by the European Commission. The CELEX (now incorporated by EURLEX) database (König et al., 2006) offers the opportunity to gather information on the individual directive level: it is possible to follow the transposition of each directive in all the member states. At the time of EU enlargement there were approximately 1700 directives in force that were eligible for transposition. A random sample of 119 directives was drawn from this population. The sample was checked to make sure that it is sufficiently representative of the entire population of directives in terms of important characteristics such as the policy sectors covered and the period of adoption. The randomness of the selection of the directives maximizes the chances that the particular pieces of legislation chosen for further study are representative of the entire body of EU directives and that the sample is not biased towards specific types of directives. The sample was deliberately not filtered to pick up, for example, only highly important laws, because the focus of this study is transposition of EU legislation in general.

Figure 2 presents the temporal dynamics of the transposition of the selected sample of directives in the candidate countries (the 'Luxembourg' group started negotiations for membership in 1997; negotiations were opened with the 'Helsinki' group countries in 1999). It is clear that the overwhelming majority of the legislation was transposed in 2004, although the peak for Lithuania, Estonia and the Czech Republic was in 2003, whereas Latvia has a more even spread. The Polish efforts are the most highly concentrated (in 2004). The drop in 2005 is an artefact of the data (because very few observations had deadlines for transposition after 2004) and does not necessarily imply a decline in transposition rates. The sample does not allow us to draw any strong conclusions about the post-accession phase of compliance. There is no observable effect of the moment of signing the Treaty of Accession, in April 2003, which theoretically strips conditionality power from the EU. Since legislative preparation takes time, most of the acts adopted in 2004 were already in the legislative pipeline by the time of concluding the treaty.

Dependent variable

The phenomenon of interest in the current study is transposition performance. Transposition performance can be measured in a variety of ways. First

Luxembourg group

■a 20

1999 2000 2001 2002 2003

2004 2005

- Czech Rep Estonia Hungary ^^^ Poland Slovenia

Helsinki group

— Latvia - - Lithuania ................Slovakia

Figure 2 Transposition of a sample of 119 directives in CEE over time. Source: own compilation.

of all, a binary variable can be created that detects whether or not an individual transposition process has been completed on time. An alternative approach is to measure the duration of each transposition process, or the time between the transposition and the deadline of the directive. Aggregating the

data over certain time and/or spatial units provides yet another opportunity. In this way we measure the number of completed transpositions within a temporal and /or spatial unit (or the backlog - the non-completed transpositions) and create a count variable. Finally, the proportion of completed transpositions from the total outstanding cases can be computed as a cross-sectional snapshot of transposition performance. Each measurement approach is related with specific statistical methods of analysis and the associated opportunities and constraints for the exploration of the data. This paper operationalizes transposition performance as a binary variable, taking the value of '1' if a transposition process has been completed within the deadline in a certain country, and the value of '0' otherwise.6

For all directives with a transposition deadline before 1 May 2004, the effective deadline is taken to be the date of accession (1 May 2004). For the rest of the directives, the effective deadline is the normal transposition deadline as recorded in CELEX. CELEX reports data on the national implementing measures (NIM) notified by each member state in regard to every directive. A directive is considered to be non-transposed within the deadline if no NIMs were found in the database or if the latest NIM was adopted after the deadline had passed. Because member states sometimes report national legislation as a NIM even when it has only a superficial connection to the EU directive, I excluded such legislation whenever it was possible to detect this. The procedure resulted in a total of 952 observations (119 directives in 8 member states). Table 1 presents the distribution of the dependent variable.

Table 1 Distribution of the dependent variable (timely transposition)

Deadline May 2004 Deadline after May 2004

on time delayed on time delayed

Czech Rep. 77 32 5 5

Estonia 101 8 7 3

Hungary 71 38 8 2

Latvia 95 14 6 4

Lithuania 88 21 5 5

Poland 70 40 7 3

Slovakia 74 35 6 4

Slovenia 60 49 5 5

Independent variables

The hypotheses presented in the theory feature both properties of the individual directives and preference-based and institutional attributes of the political and administrative systems of the member states.

The data related to the traits of the EU directives were again taken from the CELEX database. The relation of a law to EU trade was considered to be direct and significant if a directive's subject matter included 'technical barriers'. The information was coded from the CELEX subject matter descriptors field.7 An alternative measure based on the CELEX directory classification was also constructed: legislation concerning the Internal Market, Competition and the Free Movement of Goods, Services and Capital (categories 2, 6, 8, 10 and 13) was grouped into one category, and all the rest into another. The results using this measure were largely the same as from using the 'technical barriers', so results based on only one indicator (technical barriers) are reported. The coding of the remaining policy sectors was also based on the CELEX database and follows the 'classification heading' field. The environment sector (category 15) is usually considered to involve the most costly legislation and it is therefore included in the model estimation.

The political complexity of legislation was measured by the number of recitals. Recitals Kaeding, 2006 precede the body of a directive and give the reasons for the contents of the enacting terms (i.e. the articles). A large number of recitals indicates that the directive has an extensive scope, as well as addressing a high number of important issues. The number of recitals is also positively related to the salience of a directive. Implementing legislation implies low political complexity because it usually specifies, adapts and updates a legislative framework laid down by general directives adopted by the Council, or by the Council and the European Parliament.

The measurement of implementing legislation is straightforward. Directives adopted by the Commission are implementing (delegated) legislation . As such, Commission directives tend to be of less importance and scope (salience) than general directives.

The next set of variables stem from features of the political and administrative systems of the member states. Four measures are preference based. Government preferences towards European integration as well as positions on a socioeconomic left/right scale are computed on the basis of the data provided by the 'Party Policy in Modern Democracies' project (Benoit and Laver, 2006). For coalition governments, the positions were calculated by weighting the score for each party participating in the government by its share

of votes in the parliament relative to the remaining parties. For example, for a two-party government with one partner holding 30% of the seats and the other 20% of the seats of the legislation, the position of the first party will be weighted by a factor of 3/5 and that of the second by 2/5, and the two scores will be added to form the total score for the government. Generally, the available data covered all relevant parties, with two exceptions - for one of the missing parties, a measure was imputed using the Chapel Hill expert survey on party positions (Marks et al., 2006).

The party position data provide the basis for the computation of distance measures to test the government capacity argument as well. The preference distance was again measured in relation to two dimensions - EU integration and general left/right position. The measure simply tracks the maximum possible distance between a pair of two coalition partners in government.

The data on the composition of governments in the CEE countries needed for the construction of the government preferences and ideological distances data stemmed primarily from the European Journal of Political Research annual country reports (1999-2006) and were checked and complemented by other sources. Once collected, the data allow for a straightforward measure of the number of parties in government as well.

The measures of administrative effectiveness and regulatory quality are based on the World Bank Governance Indicators (Kaufman et al., 2005). 'Government effectiveness' measures the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies. 'Regulatory quality' measures the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development. All the preference-based and country-specific variables are averaged for the three years prior to the transposition of each directive and are, thus, directive specific. For directives not implemented within the deadline, an average score covering the entire span of the study (1999-2004/5) was computed.

Analysis

Because the dependent variable is binary, a binomial logistic regression is employed as the method of analysis. The model estimates the probability of a successful (on-time) transposition given a combination of values of the independent variables. The probability of success given a combination of explanatory variables is transformed into the odds of success (how often success happens relative to failure) and then the log of the odds (logit) is taken

so that the function has a range from - <» to + <». The model is estimated using the Design and GLM packages for fitting generalized linear models in R. An estimation using the MLlibrary package was also performed, yielding practically the same results. The Huber-White method was used to adjust the variance-covariance matrix to correct for heteroscedasticity and for correlated responses from cluster samples.

Table 2 reports the results of the model estimation. Two models were used, with Model 2 incorporating two interaction effects in addition to the base Model 1. A positive sign of the coefficients indicates that higher values on the independent variable increase the probability of a timely transposition. Negative values show that the factors decrease the chances a directive will be transposed within the deadline.

Surprisingly, implementing legislation is less likely to be transposed on time than regular directives. Commission directives are usually considered to be of lesser importance, specifying or updating a regulatory framework already laid down elsewhere. The negative relationship can perhaps be explained by the candidate countries focusing on the more substantially important legislation and leaving the Commission directives to be dealt with later. Political complexity shows a slight negative effect but the coefficient is not significant, indicating that the relationship might be due to chance.

The measures of government preferences have the expected signs and are statistically significant. Legislation is more likely to be transposed within the deadline if in the previous three years governments with higher support for European integration have been in power. Similarly, governments leaning towards the right on a left/right socioeconomic dimension are more likely to complete the transposition of EU legislation on time.

The measures for administrative effectiveness and regulatory quality are highly correlated and cannot be used in the same model. The reported results are based on the 'regulatory quality' measure (the results using 'administrative effectiveness' are essentially the same). The effect of this variable is positive and significant. Legislation is more likely to be transposed in countries with higher regulatory quality (measured through an average over the three years preceding the event).

The impact of government capacity depends on the precise conceptualization used. When using the preference distance measures (over the left/right and the EU dimensions), the effects are not significantly different from zero (not reported). On the other hand, a model including the number of parties in government returns a strong negative and statistically significant relationship between the number of parties in government and the probability that EU legislation will be transposed on time.

Table 2 Explaining on-time transposition: a logistic regression model

Model 2

Model 7 (Interaction effects) Expected odds of

Estimate (st. error) Estimate (st. error) timely transposition

(Intercept) -10.26 (2.72)*** -4.89 (3.77)

Implementing legislation -0.35 (0.18)** -0.36 (0.17)** 0.70

Political complexity (N recitals) -0.02 (0.01) -0.55 (0.26)** 0.81 (10 more recitals)

Government support for EU 0.51 (0.15)*** 0.20 (0.22) 1.66 (1 point increase)

Socio-economic left-right position 0.28 (0.06)*** 0.25 (0.06)*** 1.32 (1 point increase)

Number of parties in government -0.52 (0.14)*** -0.53 (0.14)*** 0.59 (1 party increase)

Regulatory quality 1.38 (0.59)** 1.46 (0.60)** 1.15 (0.1 point increase)

Trade relatedness (Internal Market) 0.45 (0.22)** -1.30 (1.05) 1.63

Environment -0.57 (0.27)** -0.53 (0.25)** 0.59

Social Policy -0.14 (0.35) -0.10 (0.34)

Transport 0.17 (0.30) 0.21 (0.29)

Interaction between left/right position and Internal Market / 0.17 (0.10)*

Interaction pol. complexity and government EU support / 0.03 (0.02)**

Log-Likelihood: -525 Log-Likelihood: -520

Chi-square: 81.62 Chi-square: 90.09

Prob x > chisq: 0 Prob x > chisq: 0

Dependent variable: probability of on-time transposition. N = 952. * significant at 10%, **significant at 5%, *** significant at 1%, two-sided tests.

Turning to the sectoral differences in transposition performance, Internal Market legislation is significantly more likely to be transposed within the deadlines. Measured by a binary variable marking the policy sector and by the presence of 'technical barriers' in the directive's subject (the measure reported in the model), the effect is positive and substantial. On the other hand, in line with the hypothesis, environmental legislation, being the most costly, is less likely to have been transposed on time. Two other policy sectors included as controls - transport and social policy - are not significantly different from the total pool of directives.

In Model 2, some of the more subtle implications of the theoretical arguments are explored. The multi-level structure of the data allows different interaction effects to be estimated. The paper tests two interaction effects that are especially relevant for the major claims of this paper. Since socioeconomic left/right positions are expected to have an effect because governments positioned towards the right end of the spectrum should be more supportive of trade liberalization and the retreat of the state from the economy (relative to left parties), the effect of the variable should be greater for the subset of Internal Market legislation. The positive sign and statistical significance of the interaction term between the left/right positions and Internal Market legislation in Model 2 support this conjecture.

Government support for EU accession should have the strongest impact on legislation that is politically complex. The transposition of non-controversial directives dealing with trivial matters should be less affected by the pro-European attitudes of governments than is politically complex legislation with wide implications. Precisely in these cases the willingness of a government to forgo its own policy preferences owing to its overall support for EU accession will result in a decrease in the negative effect of political complexity on the likelihood of a timely transposition. The positive sign of the second interaction term in Model 2 yields support for this interpretation.

The direction and significance of most of the effects are in line with the hypotheses presented in the theory section. However, it is instructive to look also at the size of the effects and to examine how much of a difference they make to the probability of timely transposition. First of all, the range of the predicted probabilities on the basis of Model 1 ranges from .34 to .94, highlighting that, even in the case of a combination of explanatory variables most unfavourable for timely transposition, a directive still has a substantial chance of being transposed within the deadline. A regular directive with an average number of recitals during a two-party government and having mean scores on the EU and left/right dimension has a probability of 86% to be transposed on time if it is related to trade. The probability drops to 68% if instead the directive is part of the environment field and is 79% for the rest.

Figure 3 plots the predicted probabilities for Internal Market and environment directives being transposed on time for a different combination of parties in government and values on the socioeconomic left/right dimensions, ranging from the observed minimum to the observed maximum. The figure illustrates the joint effects of government type and preferences and policy sectors. For example, Internal Market legislation is in general more likely to be transposed on time than environmental legislation. However, under a four-party government, Internal Market laws are less likely to be incorporated within the deadlines in comparison with environmental laws in a country governed by a one-party cabinet. Further, directives have a better chance of being transposed on time by a coalition government that strongly supports EU integration than by a single-party cabinet that only moderately supports the EU. The plot provides a clear representation of the changing probabilities of timely transposition, but it should be handled with care

Internal market legislation (one-party gov) Environmental legislation (one-party gov) Internal market legislation (4-parties gov) ---Environmental legislation (4-parties gov)

Left/right government position

Figure 3 Socio-economic left/right government positions and the probability of timely transposition.

because some combinations of the variables might not have been observed in the actual data.

Another useful way to assess the magnitude of the effect of the explanatory variables is to calculate the odds ratios. The odds ratio can be interpreted as the change in the expected odds of success (on-time transposition) as the value of one variable changes while holding the other factors constant.8 The right-most column of Table 2 presents the change in the expected odds of timely transposition for the variables of main interest (the calculation is based on Model 1). Values between 0 and 1 indicate a decrease in the expected odds, and values above 1 show higher odds of a transposition within the deadline. For example, the odds of implementing legislation being transposed on time are 0.70 smaller than of regular directives. Ten more recitals provide odds that are 0.81 times smaller for successful transposition. The odds of directives related to technical barriers are 1.63 times larger, whereas environmental directives have 0.59 smaller odds of being transposed on time. A change of one point on the EU accession support scale (the original scale ranges from 1 to 20) brings 1.66 greater odds and a 1 point change to the right on the socioeconomic dimension brings 1.32 greater odds. Clearly, the relative effect of the European dimension is bigger. The impact of regulatory quality is small, with 0.1 change in the value of this indicator (the observed range in the data is 0.6) bringing 1.15 greater odds of success. Overall, the size of the effects is not exceptionally large, but a change in some of the explanatory variables still brings substantial increases/decreases in the probability of timely transposition. As expected, some of these variables are directive related, some of them capture policy-specific influences and yet others depict features of the different candidate countries.

Conclusion

The results of the empirical analysis presented in this paper paint a complex picture of transposition in the CEE during enlargement. Pro-European and right-wing governments have been beneficial for timely transposition. In some respects this finding is in line with intuitions and existing theories of compliance. On the other hand, it dismisses the picture of EU law adoption being a purely administrative and technical exercise. All countries might have been willing to transpose the acquis, but those ruled by governments supportive of the EU and liberalization have done better. Government preferences have been translated to some extent into the varying outcomes of transposition. Moreover, the party position effects are stronger exactly where expected: in the Internal Market sector for left/right positions, and in politically complex legislation for European integration positions.

The positive effect of pro-Europe government attitudes is quite interesting in light of the lack of a relationship (or rather the inverse relationship) between support for European integration and transposition performance in the 'old' member states (see Lampinen and Uusikyla, 1998; Mbaye, 2001). The effect could well be specific to the enlargement period: a pro-Europe stance is directly related to how susceptible governments are to the lure of accession and the pressure of conditionality. The less supportive of integration that CEE governments have been, the fewer efforts and resources they seem to have put into meeting the transposition requirements. It is rather likely that overall success in terms of transposition is a result of the generally very high support for integration expressed by the governing parties in CEE during enlargement. Even the least EU-supportive governments ruling in Hungary (until 2002) and Poland (until 2001) have shown enough commitment to European integration to proceed with the incorporation of EU law, although at a slower pace than their more EU-supportive counterparts.

The capacity and capability of governments also appear influential on the basis of the data analysis. In line with the existing literature, regulatory quality has a positive effect on transposition performance, although the magnitude of the effect is not great. A greater number of parties in government seems to limit the capability of countries to transpose faster. On the other hand, measures of preference distance proved to have no explanatory value. The influence of regulatory quality, however, might to some extent be spurious and reflect the effect of time. With a few exceptions, such as the Czech Republic in 2003 and Poland in 2001, administrative efficiency and regulatory quality in the region improve in all countries over the analysed time period. At the same time, transposition is more likely as the date of accession approaches. As a result, the measures of public administration quality are to some extent collinear with time, leading to a potentially inflated size of the effect. Nevertheless, this paper brings some tentative evidence that regulatory quality has had a positive effect on the speed of transposition - a finding in line with previous research (Hille and Knill, 2006; Toshkov, 2007). Further theoretical and empirical research would have to address the possible subtle interaction between preferences and capability.

Even minor delays in adopting the EU legislation in the new member states might have important consequences for the economies and societies in the EU. The sectoral differences in transposition rates are important because they might lead to asymmetric adaptation in CEE. As areas of negative integration, such as the Internal Market, adapt faster to EU rules, whereas areas of positive integration such as environment lag behind, the market-correcting efforts of the EU might fail to compensate for some of the negative side-effects of European integration. Clearly, delays in transposing and implementing the European nature protection legislation imply potential

threats to the environment in the new member states. More importantly, companies and business projects in a country incorporating the EU rules belatedly enjoy a comparative advantage, because they do not have to carry the burden of the impact assessments demanded by the EU nature protection directives.

A direct form of conditionality related to opening trade (working separately from the grand conditionality of access to full accession) might be responsible for the observed swifter transposition of trade-related measures. Interpretation of the effect of trade, however, should be treated with care, because it does not capture cross-country or temporal variation but differentiates essentially between policy sectors and types of legislation. Transposition of trade-related directives is 'negative' integration in the sense of removing existing barriers rather than applying a new regulatory regime. In this regard, it is interesting to note the growing acceptance of EU rules and standards even beyond the borders of the EU and the candidate countries.9

Despite the importance of explaining the complex variation in the data, the big picture of timely and relatively successful transposition of EU directives in CEE during the period of enlargement should also be kept in mind. As the data indicate, even directives subject to the most unfavourable conditions have had a substantial chance of being processed within the deadline. It is an interesting question whether the drives behind transposition performance have fundamentally changed now that the enlargement period is over. Further studies will have to assess how durable and how special the embrace of EU law by the former communist EU members really is.

Notes -

1 In contrast to previous enlargements, when some countries were given a period of adaptation to the acquis during which full transposition and implementation were in practice not expected and infringement procedures were started against the newly acceded members only after several years, in the case of the Eastern Enlargement there was almost no leeway for the application of the legislation. The first infringement procedures in regard to the new members came within six months of accession.

2 A note on terminology: transposition is the process of formal adoption of rules and regulations that adapt the national legislation to the norms of European directives; implementation is a broader process that includes the formal legal aspects (transposition) as well as the practical application of the rules; compliance, defined as acting in accordance with certain standards, is broadly synonymous with implementation but emphasizes the exogenous source of the rules to be complied with. The current study deals primarily with transposition, or the formal legal aspects of compliance.

3 Over recent years, a multitude of theories and causal mechanisms have been proposed (and tested) in studies of transposition, implementation and compliance in the EU. Nevertheless, most of the analyses focus on and explore the influence of specific variables. The studies by Steunenberg (2006, 2007) are examples of the very few advances that have been made towards revealing the micro-foundations of compliance in the EU.

4 For other ways to capture the effect of administrative efficiency and reform, see Borghetto et al. (2005). Coordination problems are analysed by Masten-broek (2003). The effect of parliamentary scrutiny is investigated by Berglund et al. (2006) and the effect of administrative oversight mechanisms by Jensen (2007). Administrative capability in EU affairs might increase with the length of membership (Haverland and Romeijn, 2007).

5 I am grateful to an anonymous reviewer for drawing attention to this point.

6 The reasons for preferring a dichotomous measure instead of a duration-dependent variable are related to the problems of establishing a valid starting date when the directives become 'at risk' of being transposed for all cases, the level of measurement precision, and the related fact that a large percentage of the directives are transposed in a single period of observation, which creates problems for duration statistical models. Nevertheless, a duration analysis of the data has been performed (using Cox proportional hazards), and the results are to a very large degree consistent with the results of the logistic regression analysis presented below.

7 An alphabetically structured list of over 200 keywords based on the subdivisions of the treaties and the areas of activity of the institutions providing a general overview of the content of the document.

8 For a change of S in the values of the variable x, the odds are expected to change by a factor of exp(Px X S), holding all other variables constant. Px is the estimated coefficient for the effect of x from the logistic regression model (Scott Long, 1997: 80).

9 See the speech by the EU Trade Commissioner Peter Mandelson, 'Openness, Trade and the European Union', Chambre de Commerce et de l'Industrie de Paris, Paris, 30 June 2007, as well as the article 'Brussels Rules OK', The Economist, 20 September 2007.

References

Andonova, Liliana (2004) Transnational Politics of the Environment. The European Union in Central and Eastern Europe. Cambridge, MA, and London: MIT Press.

Benoit, Kenneth and Michael Laver (2006) Party Policy in Modern Democracies. London: Routledge.

Berglund, Sara, leva Gange and Frans van Waarden (2006) 'Mass Production of Law. Routinization in Transposition of European Directives. A Sociological-Institutionalist Account', Journal of European Public Policy 13(5): 692-716.

Borghetto, Enrico, Fabio Franchino and Daniela Giannetti (2005) 'Complying with the Transposition Deadlines of EU Directives: Evidence from Italy', paper presented at the 3rd ECPR Conference, Budapest.

Borzel, Tanja A. and Thomas Risse (2003) 'Conceptualizing the Domestic Impact of Europe', in Kevin Featherstone and Claudio M. Radaelli (eds) The Politics of Europeanization. Oxford: Oxford University Press.

Dimitrov, Vesselin, Klaus H. Goetz and Hellmut Wollmann (2006) Governing after Communism: Institutions and Policymaking. Lanham, MD: Rowman & Littlefield.

Dimitrova, Antoaneta (2002) 'Enlargement, Institution-Building and the EU's Administrative Capacity Requirement', West European Politics 25(4): 171-90.

Dimitrova, Antoaneta and Mark Rhinard (2005) 'The Power of Norms in the Transposition of EU Directives', European Integration Online Papers (EloP) 16(9).

Elster, Jon, Claus Offe and Ulrich Preuss (1998) Institutional Design in Post-Communist Societies: Rebuilding the Ship at Sea. Cambridge: Cambridge University Press.

Goetz, Klaus H. and Hellmut Wollmann (2001) 'Governmentalizing Central Executives in Post-Communist Europe: A Four-Country Comparison', Journal of European Public Policy 8(6): 864-87.

Grabbe, Heather (2003) 'Europeanization Goes East: Power and Uncertainty in the EU Accession Process', in Kevin Featherstone and Claudio M. Radaelli (eds) The Politics of Europeanization, pp. 303-27. Oxford: Oxford University Press.

Grzymala-Busse, Anna (2006) 'The Discreet Charm of Formal Institutions. Post-communist Party Competition and State Oversight', Comparative Political Studies 39(3): 271-300.

Haas, Peter (1998) 'Compliance with EU Directives: Insights from International Relations and Comparative Politics', Journal of European Public Policy 5(1): 17-37.

Haverland, Markus and Marleen Romeijn (2007) 'Do Member States Make European Policies Work? Analysing the EU Transposition Deficit', Public Administration 85(3): 757-78.

Hille, Peter and Christoph Knill (2006) '"It's the Bureaucracy, Stupid": The Implementation of the Acquis Communautaire in EU Candidate Countries, 1999-2003', European Union Politics 7(4): 531-52.

Jensen, Christian B. (2007) 'Implementing Europe: A Question of Oversight', European Union Politics 8(4): 451-77.

Kaeding, Michael (2006) 'Determinants of Transposition Delay in the European Union', Journal of Public Policy 26(3): 229-53.

Kaufman, D., A. Kraai and M. Mastruzzi (2005) Governance Matters IV. Governance Indicators for 1996-2004. Washington, DC: World Bank.

König, Thomas, Brooke Luetgert and Tanja Dannwolf (2006) 'Quantifying European Legislative Research: Using Celex and Prelex in EU Legislative Studies', European Union Politics 7(4): 553-74.

Lampinen, Risto and Petri Uusikyla (1998) 'Implementation Deficit - Why Member States Do Not Comply with EU Directives?', Scandinavian Political Studies 21(3): 231-51.

Leiber, Simone (2005) 'Implementation of EU Social Policy in Poland: Is There a Different "Eastern World of Compliance"', Journal of European Social Policy 17(3): 349-60.

Leiber, Simone (2007) 'Transposition of EU Social Policy in Poland: Are There Different "Worlds of Compliance" in East and West?', Journal of European Social Policy 17(4): 349-60.

Linos, Katerina (2004) 'How Can International Organizations Shape National Welfare States? Evidence from Compliance with EU Directives', Center for European Studies Working Paper 107.

Marks, Gary, Liesbet Hooghe, Moira Nelson and Erica Edwards (2006) 'Party

Competition and European Integration in East and West. Different Structure, Same Causality', Comparative Political Studies 39(2): 155-75.

Mastenbroek, Ellen (2003) 'Surviving the Deadline. The Transposition of EU Directives in the Netherlands', European Union Politics 4(4): 371-95.

Mbaye, Heather (2001) 'Why National States Comply with Supranational Law: Explaining Implementation. Infringements in the European Union, 1972-1993', European Union Politics 2(3): 259-81.

Perkins, Richard and Eric Neumayer (2007) 'Do Membership Benefits Buy Regulatory Compliance? An Empirical Analysis of EU Directives 1978-99', European Union Politics 8(2): 180-206.

Schimmelfennig, Frank and Ulrich Sedelmeier (eds) (2005) Europeanization of Central and Eastern Europe. Ithaca, NY: Cornell University Press.

Scott Long, J. (1997) Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: SAGE.

Sedelmeier, Ulrich (2006) 'Europeanisation in New Member and Candidate States', Living Reviews in European Governance 1(3).

Stark, David and Laszlo Bruszt (1998) Postsocialist Pathways. Transforming Politics and Property in East Central Europe. Cambridge: Cambridge University Press.

Steunenberg, Bernard (2006) 'Turning Swift Policy-Making into Deadlock and Delay: National Policy Coordination and the Transposition of EU Directives', European Union Politics 7(3): 293-319.

Steunenberg, Bernard (2007) 'A Policy Solution to the European Union's Transposition Puzzle: Interaction of Interests in Different Domestic Arenas', West European Politics 30(1): 23-49.

Steunenberg, Bernard and Mark Rhinard (2006) 'Hare or Turtle? Member States Catching up with the Transposition of EU Directives', Mimeo, Leiden University.

Tallberg, Jonas (2002) 'Paths to Compliance: Enforcement, Management and the European Union', International Organization 56(3): 609-43.

Thomson, Robert, Rene Torenvlied and Javier Arregui (2007) 'The Paradox of Compliance: Infringements and Delays in Transposing European Union Directives', British Journal of Political Science 37(4): 685-709.

Toshkov, Dimiter (2007) 'Transposition of EU Social Policy in the New Member-States', Journal of European Social Policy 17(3): 335-48.

Vachudova, Milada Anna (2005) Europe Undivided. Democracy, Leverage, and Integration after Communism. Oxford: Oxford University Press.

Zubek, Radoslaw (2005) 'Complying with Transposition Commitments in Poland: Collective Dilemmas, Core Executive and Legislative Outcomes', West European Politics 28(3): 592-619.

About the author

Dimiter Toshkov is a PhD fellow at the Department of Public

Administration, Leiden University, Wassenaarseweg 52, PO Box 9555,

2300 RB, The Netherlands.

Fax: +31 71 527 3979

E-mail: DToshkov@fsw.LeidenUniv.nl