Particularly Important State Interests as Constraint on the Decision-Making of the European Court of Justice The Example of "COM v. Member State" Litigation A. Theory The European Union constitutes a supranational legal order. This means that the Member States maintain their status as nation states, but are in the same time integrated in a legal and political order that has emancipated itself from the classical form of an international organization and the legal doctrines of international law. The Union’s acts have direct effect within the Member States and establish directly applicable rights for national citizens. The European legal order enjoys supremacy over the legal orders of the Member States. Furthermore, the Union has political bodies that exercise powers independently of the will of the Member States, most importantly the Commission, the European Parliament, and the European Court of Justice (CJEU). Yet simultaneously, the European Union enlists the Member States as important part of the European legal order and as agents of European integration. Most importantly, the Member States exercise the executive function of implementation and administration of European legislation ("executive federalism"), and the national court system functions as part of the European judiciary, too. The official motto of the European Union is "United in Diversity". The EU respects the Member States´ national identities and their essential State functions, including ensuring the territorial integrity of the State, maintaining law and order and safeguarding national security (Art. 4 TEU). Yet, the express aim of the European project is to create an ever closer union among the peoples of Europe (Art. 1 TEU). In practice, the Union is therefore characterized by a strong political tension between unity interests that call for the harmonization of the Member States´ legal orders, and diversity interests calling for the preservation of national particularities and national autonomy. This tension establishes the dynamics of European politics: The political process in which European legislation enacted is driven by the generally pro-European policy initiatives of the Commission. But usually, initiatives by the Commission face strong claims of national interests that are supposed to limit integration, i.e. the "Europeanization" of substantive policy fields. The most important political mechanism to deal with this tension and to balance between national and European interests is a differentiated system of decision-making rules. In some fields of policy, the enactment of legislative acts by the Council, which is constituted by a governmental representative of each Member State, requires only a (qualified) majority of votes. In other fields of legislation, there needs to be unanimity of Member States. The latter establishes a heightened threshold for enacting legislation that works as a procedural protection of national interests, since under unanimity is not possible to overrule any Member State’s interest. Yet, the tension between national and European interests is usually not resolved completely within the legislative process. National interests also come into play when Member States implement and administer European law. Oftentimes, national agencies interpret and apply European legislation restrictively and try to account for national interests, such as economic interests, national security, restriction of immigration, limiting access to the national system of social security, and so forth. Member States´ efforts to give effect to national interests and to limit the application and effects of European law often times undermine the effectivity of EU legislation to accomplish its purposes. The Commission, which has the function to promote the general interest of the Union, is supposed to ensure the application of the Treaties, and of measures adopted by the institutions pursuant to them (Art. 17(1) TEU). It does so by way of the infringement proceedings laid out in Art. 258 TFEU. The Commission monitors the compliance of Member States with their obligations under EU law. In cases where it finds non-compliance by a Member State, it initiates a quasi-administrative infringement procedure that starts with the Commission delivering a reasoned opinion on the matter after giving the State concerned the opportunity to submit its observations. If the State concerned does not comply with the opinion within the period laid down by the Commission, the latter may bring the matter before the CJEU. The Court will then decide whether the Member State breached an obligation under EU law. Our study deals with the relevance and influence of Member States political interests in this kind of infringement proceeding before the European Court of Justice, i.e. in litigation "COM v. Member State". We assume that the CJEU is a strategic actor. Its decisions will neither be guided purely by considerations of the law (which is often not very clear and leaves a lot of leeway for interpretation), nor by ideological preferences of the Justices (about which we know little since the individual vote is not published) alone. We assume that the Court has a number of reasons to be sensible to Member States interests. The first is that for the implementation of EU law the European Union heavily depends on the Member States. It is thus crucial for the integrity and effectiveness of the European legal order that Member States are willing to comply with and do indeed apply EU law in good faith. Therefore, the EU and the CJEU as its judicial agent need to maintain acceptance, compliance and a certain good will by the Member States, and avoid situations in which Member States are no longer willing to implement EU law. We assume that the CJEU will account for this and consider the interests of Member States in its decision-making. Furthermore, the Member States can, by means of amending the treaties that govern the European Union, overrule decisions by the CJEU. They can change the substantive law, limit European powers in general and are also able to strip the Court’s jurisdiction. We assume that to avoid this, the Court will act strategically and give weight and effect to Member States´ interests. B. Hypothesis In the infringement proceedings initiated by the Commission against a Member State ("COM v. Member State"), the probability that the Court will rule in favor of the states (and against the Commission) is higher in cases which involve “particularly important state interests” (hereinafter also PISI). C. Variables To test this hypothesis we needed to set our independent and dependent variables. The dependent variable is the outcome of the case – whether the Court ruled in favor of the Commission or the defending state. For the purpose of reaching a sustainable balance between national and European interests, the Court needs to make a judgment about the weight and importance of a certain national interest and needs to make an assumption about how vital a certain substantive policy is for the Member State. We assume that the Court relies on two kinds of "signals" that indicate a particularly important state interest, and those will form the independent variables. To operationalize the "particularly important state interest" as independent variable we will firstly use a proxy based on the distinction in the EU primary law (see above) between unanimity and (qualified) majority rule for the decision-making of the Council. We assume that in the infringement proceedings initiated by the Commission, the Court will be more deferential to the Member States in cases where it enforces an EU legal act adopted by “unanimity procedure”. This proxy is reasonably reliable, because each act of EU’s secondary legislation needs to have a legal basis in the treaties and the legal basis is quite easily identifiable. We are convinced that the proxy is also valid to measure the existence of "particularly important state interest", as the remaining instances of unanimity procedure in the treaties after the Lisbon Treaty are generally considered to reflect areas where states want to retain control, indicating that the policy is relevant for particularly important state interests. The areas where the Council of EU acts unanimously are defined in the EU treaties, and they include inter alia taxation, social security, national and common security or citizenship. Those policies thus bear on what we call "particularly important state interest". In order to make our results more valid and tied to “particularly important state interests”, we are including a second independent variable, which is whether the EU Member States used their right to intervene in the Court’s proceedings (Art. 40 of the Statute of the Court). We assume that (generally speaking) the more Member States intervene on the side of the defendant, the more likely is the Court to rule in the favor of the defending state. We may operationalize this variable in two ways. First, we may use a plain number of states intervening on the side of the defendant, and second, we may use a proxy that indicates the relative strength, or importance, of Member States as compared to each other. The proxy is based on the number of representatives of the Member State in the European Parliament (number of MEPs, see Appendix). We expect that the second operationalization of this variable will be more relevant, as it better reflects the political-strategic considerations of the Court we expect to find. We expect that an intervention of France and Germany will carry more weight than intervention of Malta and Estonia, for example. D. Dataset We would use data from the public database “curia.eu” which includes all the decisions issued by the Court; the database contains all the data relevant for our research with the exceptions mentioned in the „selection bias“ part (see below). We are able to pin down the decisions initiated by the Commission’s petition against the state, decisions from certain time period etc. and we can easily identify the disposition of the case and the (non)existence of state interventions. There are some minor practical problems with the dataset but all of them can be overcome. For example, the information about state interventions are not presented in a very user-friendly manner and we would need to go through the decisions (their statement of facts and previous procedure). In the actual research, we would use the infringement cases decided between 2010 and 2014, which would provide us with several hundreds of cases. We would exclude two types of cases from our dataset: 1. Old pending cases: cases in which the enforced legal act was initially adopted by unanimity procedure, but the unanimity requirement was removed by Lisbon treaty. This is motivated by the fact that unanimity procedure after Lisbon treaty reflects the state interests more accurately and we want to avoid distortion of results by inter-temporal issues. 2. Cases where the only question is whether the State missed the deadline for implementation of the directive. By excluding this group of cases (where acting strategically is practically impossible) our results would be clearer. E. Conducting the research 1. After collecting the dataset, we will divide the cases in two groups based on the first independent variable (unanimity cases vs. other cases). And we will measure the rate of success of the Commission and Member States by a very simple percentage score. We expect that the rate of success of Member States will be higher in the unanimity cases. 2. In a second step, we will use the second independent variable in both forms (number of states and their relative strength). We would use regression analysis to control for unanimity (hold the unanimity variable constant). We expect that the regression analysis will reveal correlation between states’ interventions and the outcome of the case and we assume that the correlation will be stronger if we use the “relative voting strength” proxy than if we use the plain number of states. F. Possible problems Our research proposal suffers from a possible selection bias problem. The Commission is the most frequent and possibly the most qualified litigant before the Court and can as such be compared to the U.S. Solicitor General. The Commission will often act strategically and it might be well aware of the “political” tendencies we suspect to find in the Court’s case-law. Consequently, the Commission might adapt its litigation strategy and choose not to litigate certain PISI cases it would litigate otherwise. The Commission will probably settle PISI cases in which it sees a low chance of success in the administrative proceedings and will not bring those cases before the Court. This means that we will likely not see many cases before the Court in which the facts and the law work more in favor of the Member States. If this suspicion or ours would be true, it would affect the results of our research in that we would probably see a relatively high overall success rate of the Commission even in cases in which our independent variable indicate a PISI. Yet, we are confident that we would still be able to measure a difference in the Commission’s success rate between cases that involve a PISI as compared to cases that do not do so. APPENDIX -- COMPOSITION OF THE EUROPEAN PARLIAMENT 2009 - 2014 Member State Number of MEPs BELGIUM 22 DENMARK 13 GERMANY 99 IRELAND 12 FRANCE 72 ITALIE 72 LUXEMBURG 6 NETHERLANDS 25 UNITED KINGDOM 72 GREECE 22 SPAIN 50 PORTUGAL 22 SWEDEN 18 AUSTRIA 17 FINLAND 13 CZECH REPUBLIC 22 ESTONIA 6 CYPRUS 6 LITHUANIA 12 LATVIA 8 HUNGARY 22 MALTA 5 POLAND 50 SLOVENIA 7 SLOVAKIA 13 BULGARIA 17 ROMANIA 33 Total EU 736