1
Infringement Data and Noncompliance
For years, the European Commission has been complaining about a growing compliance deficit in the EU (Commission of the European Communities 1990, 2000, 2011, 2016). Some scholars have contended that the level of compliance with EU law compares well to the level of compliance with domestic law in democratic states (Keohane and Hoffmann 1990, 278; Zürn and Joerges 2005). However, many share the view of the Commission that noncompliance poses for the EU a serious problem that is “pathological” (Weiler 1988) and “systemic” (From and Stava 1993; cf. Krislov, Ehlermann, and Weiler 1986; Snyder 1993; Mendrinou 1996; Tallberg 2003; Cremona 2012). The contradicting assessments of member state compliance may at least partly be due to the absence of common assessment criteria and reliable data. Yet, this chapter shows that existing data provides no evidence that the EU has a compliance problem. EU scholars have to rely for their assessments on violations of EU law detected and reported by the European Commission. Some of them have compared noncompliance in the EU to an iceberg (Hartlapp and Falkner 2009). Only its tip is visible. However, how can we know how much of noncompliance remains under the surface?1 We have no means to measure the actual size of the iceberg. Moreover, like icebergs in the age of global warming, noncompliance in the EU has been diminishing over the past twenty-five years. Time series data on violations of EU law does not tell us how big the iceberg is; but it shows that the iceberg has been melting.
Compliance and Noncompliance
Compliance is defined as behavior that is consistent with (international) norms and rules (Young 1979, 104; Raustiala and Slaughter 2002). It is a broader concept than implementation, which refers to the putting into practice of a norm, rule, or policy (Mazmanian and Sabatier 1981; cf. Treib 2008). At the same time, compliance is narrower than effectiveness, which relates to the impact a policy or law has on the socioeconomic environment in solving the problem it was adopted to address (cf. Levy, Young, and Zürn 1995; Nollkämper 1992; Bernauer 1995).
Since international and EU law is addressed to states, compliance refers primarily to their rule-consistent behavior. As in any other international organization, member states have to incorporate EU law into their domestic legal orders as well as apply and enforce it (Raustiala and Slaughter 2002; Simmons 1998). If they fail to do so, they are in noncompliance.
Research focuses on noncompliance rather than compliance as its explanandum. This may or may not have to do with scholars expecting compliance to be the default behavior of states. I suspect that it is more related to the available data. International institutions report on violations of their member states, not on their law obedience. Reported noncompliance is a common, if not the most common, indicator for noncompliance.
The European Commission reports on four types of noncompliance with EU law (table 1.1). The different types of violation are largely defined by the form EU law can take. EU law is divided into primary and secondary law. Primary law refers to the articles in the treaties, which are made by the member states. Similar to national constitutions, these articles set the ground rules for all EU action. They include the fundamental values on which the EU is built, such as human rights, rule of law, and democracy (Article 2 TEU). Secondary law is made by EU institutions according to the rules and procedures set by EU primary law. The treaties provide for five different legal instruments, which differ in their legal bindingness and scope of addressees. Regulations are legally binding and directly applicable in all EU member states. Directives are equally biding for all member states as to the results to be achieved; they have to be transposed into the national legal framework. They leave the member states discretion in choosing the means of implementation. Decisions are directly applicable but binding only for those to whom they are addressed. Recommendations and opinions, finally, are nonbinding, declaratory instruments. Treaty articles, regulations, and directives form the essence of EU law—they constitute legal obligations for all member states.
TABLE 1.1 Types of EU law violations
The first three types of violations (tov) refer to directives, which are not directly applicable but require transposition into national law.
First, member states may fail to notify the European Commission of all national measures taken in order to legally implement directives in due time (delayed transposition into national law, referred to by the European Commission as noncommunication of national implementing measures, tov_1).
Second, member states may incorrectly transpose directives. Parts of the obligations of the directive are not enacted, or national regulations deviate from European obligations because they are not amended and repealed, respectively (nonconformity with directives, tov_2).
Third, even if the legal implementation of a directive has been timely, correct, and complete, member states might still not practically apply and enforce it. Noncompliance involves the active violation of member states taking conflicting national measures or the passive failure to invoke the obligations of the directive. The latter also includes failures to enforce effectively EU law—that is, to take positive action against violators, both by national administration and judicial organs, as well as to make adequate remedies available to the individual against infringements that impinge on her rights (bad application of directives, tov_3).
Fourth, member states may not or only incorrectly apply and enforce treaty provisions, regulations, and decisions, which are directly applicable and therefore do not have to be incorporated into national law.2 Noncompliance can also take the form of enacting, or not repealing, national measures that contradict EU law (violations against treaty provisions, regulations, and decisions, tov_4).3
The Methodological Challenge
Studies on compliance with international norms and rules face the methodological challenge of measuring their dependent variable. Many have developed their own assessment criteria and collected the empirical information in laborious case studies (e.g., Falkner et al. 2005; Zürn and Joerges 2005; Mitchell 2003a; Börzel 2003a; Finnemore and Sikkink 1998). This renders the comparison of empirical findings and theoretical claims difficult, particularly since some policy sectors (social affairs, environment) and some member states (UK, Germany, France, Italy, Greece) are more studied than others (Luetgert and Dannwolf 2009). Others, therefore, have drawn on statistical data provided by the monitoring bodies of international regimes and organizations (e.g., Reinhardt 2001; Mbaye 2001; Linos 2007; Perkins and Neumayer 2007; König and Luetgert 2009; Haverland, Steunenberg, and van Waarden 2011).
EU research has used different types of data published by the European Commission to measure noncompliance with EU law: statistics on transposition notification and on infringement proceedings (cf. Hartlapp and Falkner 2009).
Timely Transposition as a Proxy of Noncompliance
Since 1990, the Commission has reported the directives for which member states have notified transposition as percentage of the directives in force. In addition, the EU makes the notifications of national implementing measures and other information related to the transposition process publicly available in its EUR-Lex database (Sector 7 within the CELEX system).4 A number of studies have used this data on the notification of national implementing measures as a proxy of noncompliance (Mastenbroek 2003; Steunenberg 2006; Kaeding 2006; Berglund, Grange, and van Waarden 2006; Toshkov 2007a; Thomson 2007; Haverland, Steunenberg, and van Waarden 2008, 2011; Steunenberg and Kaeding 2009; König and Luetgert 2009; Luetgert and Dannwolf 2009). Delayed transposition, however, is only one of four types of violation of EU law, and arguably not the most relevant one. Noncommunication (tov_1) only refers to directives and their timely (and complete) incorporation into the domestic legal order of the member states. Regulations, as the other major form of EU secondary law, do not require transposition into national law because they take immediate effect. Their number by far exceeds that of directives. Directives make up only 13 percent of the legislation in force (see chapter 4).
Moreover, the failure to transpose directives is low and has been diminishing. According to the data published by the European Commission, member states have always had more than 90 percent of the directives in force on their books. Transposition rates have improved over the years, from an average of 91 percent in 1990 to an average of 99.1 percent in 2010, and 99.4 percent in 2013—exceeding the EU’s 2007 target of a transposition deficit of 1.0 percent of the total number of directives in force.5 The range between transposition laggards and leaders, accordingly, is narrow, ranging from 97.8 percent of Belgium to 99.9 percent of Sweden. The EU’s transposition scoreboards also testify to the exemplary performance of the new member states, which joined in 2004 and 2007 (cf. European Commission 2005; Dimitrova and Toshkov 2009). In other words, problems with timely transposition are limited and do not show much variation among member states. Arguably, transposition failures (noncommunication) cover only a very small and increasingly irrelevant part of noncompliance in the EU.
Studies that focus on the timeliness of transposition claim that transposition problems are more severe than suggested by the high percentage of directives transposed. Data collected at the national level shows that a significant number of important directives got transposed with serious delays, undermining the effective application of EU law and creating competitive disadvantages for industries in compliant member states (Haverland, Steunenberg, and van Waarden 2011; Haverland and Romeijn 2007; Steunenberg and Rhinard 2010; Borghetto, Franchino, and Giannetti 2006; Kaeding 2006; Mastenbroek 2003). The problem may be even more pronounced since notification data seems to underestimate the actual delay in transposition (Hartlapp and Falkner 2009). However, distortions between notified and actual transposition are greater for some member states, which points to a serious bias in the data (ibid.). Moreover, delays say little about whether transposition is correct (König and Mäder 2013; Hartlapp and Falkner 2009; Zhelyazkova and Yordanova 2015). Member states have substantial discretion when transposing EU directives. Even if they literally translate the letters of an EU directive into national law, they may fail to adhere to its spirit (Dimitrakopoulos 2001). Directives usually set framework legislation to allow accommodating country-specific context and conditions. Member states are left the choice as to the form and methods of implementation,6 which provides substantial opportunities for incorrect and incomplete transposition.
Research on international law has shown that ratification is a poor indicator for compliance (Simmons 1998; Risse, Ropp, and Sikkink 2013). In the EU, there is evidence that speedy transposition indicates noncompliance rather than compliance. Notification data relies on the member states’ self-reporting of the legal implementation measures they have taken to transpose a directive into national law. National governments may have an incentive to exaggerate their actual compliance, particularly if they anticipate problems. Thus, member states report pre-existing national measures to notify the Commission of the transposition of a directive (Zhelyazkova and Yordanova 2015). Unless a member state was able to fully upload its domestic policy to the EU level (Héritier 1996; Börzel 2002a), the lack of any measures to adjust domestic legislation to a new directive is likely to result in problems of incorrect or incomplete transposition. The study by Falkner et al. (2005) on the implementation of EU social policy finds that some member states notify the Commission before they correctly transpose a directive. Notifying legal compliance helps a member state buy time to muster necessary resources and overcome domestic opposition to make the necessary adjustments (Zhelyazkova and Yordanova 2015; Zhelyazkova, Kaya, and Schrama 2017). The Commission routinely opens infringement proceedings if a member state fails notifying national implementing measures within the deadline specified by the directive. Notifying early or timely transposition, by contrast, may hold off the Commission by signaling that a member state is already in compliance with a new directive or at least prepared to comply. Not surprisingly then, timely transposition is associated with more rather than less noncompliance. The more timely the transposition, the less complete and correct is its legal implementation (Zhelyazkova and Yordanova 2015; Zhelyazkova, Kaya, and Schrama 2017).
Infringement Proceedings as a Proxy of Noncompliance
To work around the narrow focus and the bias of the transposition data, students of noncompliance have taken recourse to the European Commission’s Annual Reports on Monitoring the Application of EU Law7 (e.g., Mbaye 2001; Tallberg 2002; Sverdrup 2004; Börzel et al. 2010; Thomson, Torenvield, and Arregui 2007). Since 1984, the Commission reports each year on the actions it has taken against member states for all four types of noncompliance with EU law. Regardless of the type of violation at stake, Article 258 TFEU entitles the European Commission as the “guardian of the treaties” (Article 17.1 TEU) to bring legal action against member states for failing to fulfill their obligations under EU law.
From the very inceptions of European integration, infringement proceedings were to ensure compliance with EU law.8 Many international organizations have dispute settlement procedures. They increasingly rely on courts rather than negotiated solutions (Alter 2014; Alter and Hooghe 2016). These developments notwithstanding, the EU still has the most elaborate compliance system at the international level. Infringement proceedings are an expression of the EU as a “community of law” (Hallstein 1972). They have been devised as an instrument t...
