A constitutional court for the EU?

Transcription

1 ARTICLE A constitutional court for the EU? Bo Vesterdorf* The question forming the title of this paper may be taken two different ways. Do we have a court in the EU in charge of constitutional adjudication? Or, alternatively, ought we to have a court in the EU that would specialize in constitutional adjudication? Such a court might resemble, say, France s Conseil Constitutionnel or the Bundesverfassungsgericht in Germany. I take the liberty of opting for the second reading Should we have a specialized constitutional court? since, in my view, the answer to the first question is quite straightforward. As I will explain briefly in section 1, the European Court of Justice (ECJ) already performs the duties of a constitutional court. However, because the ECJ has broader duties than pure constitutional adjudication, I think it resembles more a supreme court than a specialized constitutional court. I will, therefore, examine in section 2 whether the specialized constitutional court model has inherent advantages over the supreme court model. I conclude that neither is inherently superior to the other; nonetheless, I argue in section 3 that the ECJ as a supreme court should be allowed to concentrate on its constitutional duties as well as on its role as a guardian of the unity and consistency of EU law. Finally, I summarize in section 4 my thoughts on the means that we can use to redirect the ECJ s functions. 1. The constitutional duties performed by the ECJ There is no single accepted definition of what a constitutional court is, but everyone would probably agree that, at a minimum, it requires, first, a constitution (or any other supreme law of the land or Magna Carta) and, second, a court defending and interpreting this basic charter. As for the existence of a constitution in the EU legal order, it is not necessary to wait for the ratification of the constitutional treaty, 1 or any * President of the Court of First Instance of the European Communities. All views expressed are personal. This contribution is based upon a speech I delivered at the conference The Future Of The European Judicial System The Constitutional Role Of European Courts, held in Berlin on Nov. 2 4, Papers from this conference are available at and are being published in a collection by Nomos Verlag. bo.vesterdorf@curia.europa.eu 1 Treaty establishing a Constitution for Europe, 2004 O.J. (C 310) 1. For recent comments, see George Tridimas, The Relevance of Confederate Structures in the Judicial Architecture of the Draft EU Constitution, in INTERNATIONAL CONFLICT RESOLUTION 281 (Stefan Voigt, Max Albert & Dieter ª The Author Oxford University Press and New York University School of Law. 607 All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org I CON, Volume 4, Number 4, 2006, pp doi: /icon/mol026

2 608 Int l J Con Law, Vol 4, No 4 (Oct 2006) B. Vesterdorf other formal constitution, before we may say that such a set of norms exists. The proposition that the treaties should be regarded as a constitution dates back to the beginning of the Communities history. 2 As Ole Due explains: The Treaties not only define the tasks of the Communities; they also set out the general principles to be respected by the Member States as well as by the Community Institutions; they establish the Institutions and provide them with the powers necessary for performing the tasks of the Communities; they install a complicated system of checks and balances between the Institutions; they define the relations between the Communities, their component States, third countries and international organizations; they regulate the effect of Community acts within the internal legal order of the component State and they authorize the Community Institutions to make rules and take decisions directly binding on the individual subjects of these States. In so doing, they go much further than normal treaties establishing international organizations and they perform the same functions as the Constitution of a federal State. 3 In classical terms, a constitution performs three main functions. First, it organizes and allocates powers among the different governing bodies within a given society. We could call these a constitution s horizontal provisions. Second, it provides for the protection of certain principles and individual rights, which, for the very reason that they derive from the constitution, are called fundamental. These we could term its vertical provisions. The third function is to provide safeguards to ensure that the governing bodies, in exercising the powers granted to them under the horizontal provisions, do not infringe the rights granted to individuals under the vertical provisions. Schmidtchen eds., Mohr Siebeck 2006); Richard Bellamy, Which Constitution for What Kind of Europe?: Three Models of European Constitutionalism, in The Federal Trust Constitutional Online Paper Series No. 03/03 (2003), available at Koenraad Lenaerts, The Constitution for Europe: Fiction or Reality?, 11 COLUMBIA J. EUR. L. 465 (2005); Jeremy Lever, The Treaty Establishing a Constitution for Europe, 28 FORDHAM INT L L.J (2005); J.H.H.WEILER, THE CONSTITUTION OF EUROPE. DO THE NEW CLOTHES HAVE AN EMPEROR? AND OTHER ESSAYS ON EUROPEAN INTEGRATION (Cambridge Univ. Press 1999). For a recent comparison between the U.S. and European systems, see George Nolte, European and U.S. Constitutionalism: Comparing Essential Elements, in EUROPEAN AND U.S CONSTITUTIONALISM 9 (George Nolte ed., Council of Europe Publishing 2005). 2 See, e.g., Donal Barrington, The Emergence of a Constitutional Court, in HUMAN RIGHTS AND CONSTITUTIONAL LAW: ESSAYS IN HONOUR OF BRIAN WALSH 251 (James O Reilly ed., Round Hall Press 1992); Vlad Constantinesco, La Constitutionalisation de l Union Européenne, in DE LACOMMUNAUTÉ DE DROIT À L UNION DE DROIT 133 (Joël Rideau ed., Librairie Générale de Droit et de Jurisprudence 2000). 3 Ole Due, A Constitutional Court for the European Communities, in CONSTITUTIONAL ADJUDICATION IN EUROPEAN COMMUNITY AND NATIONAL LAW: ESSAYS FOR THE HON. MR. JUSTICE T. F. O HIGGINS 4 (Butterworths 1992).

3 A constitutional court for the EU? 609 The question of how these functions are fulfilled by the treaties within the framework of the EU legal order is complex, and the contributions to this debate in the form of articles, textbooks, and speeches are legion. 4 But it is obvious that the existing treaties as interpreted by the ECJ contain many provisions aimed at defining, distributing, and limiting powers and competences among the various EU political entities (including the member sates) 5 as well as other provisions aimed at protecting certain fundamental rights and principles. 6 Several provisions of the treaties are of a constitutional nature. The treaties are our basic constitutional charter, as noted by the ECJ in a line of cases starting with Les Verts v. Parliament, where the Court held that the European Economic Community is a Community based on the rule of law, inasmuch as neither its Member States nor its Institutions can avoid a review of the question whether the measures adopted by them are in conformity with the basic constitutional charter, the treaty. 7 Once it is agreed that we have a comprehensive set of constitutional norms, it is a straightforward matter to conclude that we also have a judicial body vested with the powers usually wielded by constitutional courts. The ECJ s role under article 220 EC is to ensure that, in the interpretation and application of the EC treaty, the law is observed. On this basis, the ECJ is called upon regularly to settle questions regarding the allocation of powers among the various bodies within the EU legal order and to defend the 4 See, as a very first selection, António Goucha Soares, The Division of Competences in the European Constitution, 11 EUR. PUB. L. 603 (2005); Alan Dashwood, The Limits of European Community Powers,21EUR.L.REV. 128 (1996); Stephen Weatherill, Better Competence Monitoring,30EUR.L. REV. 25 (2005); Alan Dashwood, The Relationship Between the Member States and the European Union/European Community, 41 COMMON MKT. L. REV. 365 (2004); and Grainne de Búrca, The Constitutional Challenge of New Governance in the European Union, 28 EUR. L. REV. 828 (2003). 5 Art. 5 EC, in particular, is the fundamental provision enshrining the principles of attributed powers and of subsidiarity. 6 In particular, art. 6 TEU lists the fundamental principles in which the Union is rooted and commits the Union to respecting the European Convention for the Protection of Human Rights and Fundamental Freedoms, signed in Rome on Nov. 4, Case 294/83, Parti Écologiste Les Verts v. European Parliament [1986] E.C.R (point 23). The Court has made repeated references to the Community of law. See, e.g., Case 2/88, Imm., J. J. Zwartveld and others [1990] E.C.R. I-3365; Opinion 1/91, Opinion on the Draft Agreement Between the Community, on the One Hand, and the Countries of the European Free Trade Association, on the Other, Relating to the Creation of the European Economic Area [1991] E.C.R. I-6079; Case C-314/91, Beate Weber v. European Parliament [1993] E.C.R. I-1093, Joined Cases C-181/91 and C-248/91, European Parliament v. Council and Commission [1993] E.C.R. I Furthermore, already in his Opinion of June 12, 1956, in Case 8/55, Fédéchar [1956] E.C.R. 231, 263, Advocate General Maurice Lagrange affirmed that, although the treaty had been concluded in the form of an international treaty, it should, nevertheless, be regarded as the internal charter of the Community. See also DENYS SIMON, LE SYSTÈME JURIDIQUE COMMUNAUTAIRE 73, paras. 48 and 49 (2001).

4 610 Int l J Con Law, Vol 4, No 4 (Oct 2006) B. Vesterdorf fundamental rights and principles of good governance, as defined by the treaties. 8 For example, it is obvious that the ability of the ECJ to rule de facto on the compatibility of domestic law with EC law by means of proceedings spelled out in article 234 EC may have substantial constitutional implications. From a more academic point of view, it is also widely accepted that the ECJ frequently resorts to interpretative techniques and trade-offs that are quite typical of constitutional adjudication. 9 Thus, there can be no doubt that the ECJ already carries out constitutional tasks. But it does much more than that. The ECJ also decides cases that do not have a constitutional dimension: (a) actions for annulment of an EU administrative decision; (b) actions regarding an EU institution s failure to act; (c) actions concerning infringement (as when a member state is charged with failure to fulfill an obligation under EU law); (d) preliminary rulings (when member states national courts refer issues to the court for a preliminary ruling on the interpretation of a point of EU law); and (e) appeals (of rulings by the Court of First Instance [CFI]) on points of law. For example, issues of a constitutional nature will arise only rarely in the context of preliminary rulings that specifically request the ECJ to interpret the meaning of ancillary provisions of the directives prescribing the general principles to be incorporated into the environmental laws of the member states. For this reason, the ECJ may be said to have wider duties than a specialized constitutional court. Its role also extends to promoting the unity and the consistency of the law, whether constitutional or not, by advising national courts through preliminary rulings and by judging on appeal from the CFI. Accordingly, when seen in this light, the ECJ looks more like a 8 On the ECJ, see generally ANTONY ARNULL, THE EU AND ITS COURT OF JUSTICE (Oxford Univ. Press 1999). More specifically, for a comparison between the ECJ and the U.S. Supreme Court, see Federico G. Mancini, The United States Supreme Court and the European Court of Justice, in DEMOCRACY AND CONSTITUTIONALISM IN THE EUROPEAN UNION 163 (2000); Walter van Gerven, Constitutional Aspects of the European Court s Case-law on Articles 30 and 36 EC as Compared with the U.S. Dormant Commerce Clause, in MÉLANGE EN HOMMAGE À MICHEL WAELBROECK 1628 (Bruylant 1999). 9 For the role of the ECJ as a constitutional court, see Andreas M. Donner, The Constitutional Powers of the Court of Justice of the European Communities, 11 COMMON MKT. L. REV. 127 (1974); Rolv Ryssdal, On the Road to a European Constitutional Court, in COLLECTED COURSE OF THE ACADEMY OF EUROPEAN LAW: THE PROTECTION OF HUMAN RIGHTS IN EUROPE 7 (Frank Emmert ed., Martinus Nijhoff 1991); Ole Due, A Constitutional Court for the European Communities, in CONSTITUTIONAL ADJUDICATION IN EUROPEAN COMMUNITY AND NATIONAL LAW: ESSAYS FOR THE HON. MR. JUSTICE T. F. O HIGGINS (Butterworths 1992); Francis G. Jacobs, Is the Court of Justice of the European Communities a Constitutional Court?, in CONSTITUTIONAL ADJUDICATION IN EUROPEAN COMMUNITY AND NATIONAL LAW: ESSAYS FOR THE HON. MR. JUSTICE T. F. O HIGGINS (Butterworths 1992); Jens Rinze, The Role of the European Court of Justice as a Federal Constitutional Court, PUBLIC L. 426 (1993).

5 A constitutional court for the EU? 611 supreme court than a constitutional court, at least to the extent that a constitutional court is supposedly confined strictly to constitutional adjudication. 10 Provided this view is correct, it immediately invites the following question: Is it a problem that the ECJ performs all these functions, and, if so, should the ECJ be transformed into a constitutional court with more specialized powers? 2. Supreme courts versus specialized constitutional courts In my view, there is no obvious reason to prefer the model of a specialized constitutional court over that of a supreme court. The supreme-court model operates successfully in many countries, an obvious example being the United States Supreme Court, which sometimes rules on questions of federal law without any constitutional question at issue. The U.S. Supreme Court is endowed with many functions, including the power to prescribe rules of procedure that bind the lower courts of the United States. 11 In short, I am not aware of a compelling reason why the specialized constitutional court would be inherently superior to a supreme court. Nonetheless, I can identify one situation in which the multiplicity of functions performed by a supreme court can raise legitimate concerns, namely, when the court s caseload is so heavy it cannot dispose of its cases within a reasonable time. I seriously doubt that a supreme court faced with the Sisyphean task of adjudicating an ever-increasing docket would be able always to dedicate the time and attention required to assess important questions of constitutional law. Such a court, operating under time constraints due to its caseload, runs the risk of forgoing comprehensive consideration and precisely at the point where the court s rigor and depth of analysis are central to its legitimacy. Against this background, it is interesting to note that the caseloads of the ECJ and the U.S. Supreme Court are completely different. According to the Supreme Court s statistics, its caseload has increased steadily to a current total of more than 7,000 cases on the docket per term, whereas plenary review with oral arguments by attorneys is granted in fewer than 10 In the many member states where a constitutional court exists, it is usually vested only with constitutional functions, whereas the supreme court is the last appellate judicature of the system. That is the case, for example, in Germany (Bundesverfassungsgericht, on the one hand, and Bundesgerichtshof and Bundesverwaltungsgericht, on the other); in France (Conseil Constitutionnel, on the one hand, and Cour de Cassation and Conseil d État, on the other); in Italy (Corte Costituzionale, on the one hand, and Corte di Cassazione and Consiglio di Stato, on the other); or in Spain (Tribunal Constitucional, on the one hand, and Tribunal Supremo, on the other). 11 See 28 U.S.C. x

6 612 Int l J Con Law, Vol 4, No 4 (Oct 2006) B. Vesterdorf 100 cases. Of those, formal written judgments are typically delivered in 80 to 90 cases, although this figure has varied over time. 12 This is in stark contrast to the situation in the EU; the ECJ has mandatory jurisdiction, 13 and it must rule on all cases lodged with its registry. In 2004, the ECJ completed no less than 665 cases, and it is widely known that its workload has increased steadily for decades despite the transfer of wide competences to the CFI. This increase has had an adverse effect on the ECJ s ability to deliver its judgments within a reasonable length of time. In 1983, it took the ECJ 12 months, on average, to deal with preliminary rulings under article 234 EC. In 2004, the average time taken for the same type of case had increased to 23.5 months. Most would agree that an average allocation of almost two years per case is problematic, given that the case pending before the national court is suspended until the ECJ has ruled on the question referred to it. We are all keenly aware of what is at stake: if national judges turn their backs on preliminary rulings, there could be dire consequences for the uniform and direct application of EC law. In 2004 and 2005, we saw some positive developments, with the average length of proceedings before the ECJ reduced substantially in all types of actions. The duration of proceedings concerning references for a preliminary ruling decreased to 20.4 months, as compared with 23.5 months in 2004, and 25.5 months in The average length of time taken for a direct action or appeal in 2004 and 2005 was approximately 21 months, whereas in 2003 it had reached 24.7 months for direct actions, and 28.7 months for appeals. 14 These figures are convincing evidence that the arrival of ten new judges after the enlargement and amendment of the ECJ s working methods have borne fruit. But we should be cautious about giving full rein to our satisfaction, since we can foresee a substantial increase in the ECJ s workload. In three or four years the number of references and direct actions coming from the new member states will have increased significantly. In addition, the upward trend will accelerate if, as envisaged by the constitutional treaty, 15 the rules governing the ECJ s jurisdiction are extended. Lastly, the substantive scope of 12 Federico G. Mancini, The United States Supreme Court and the European Court of Justice, in DEMOCRACY AND CONSTITUTIONALISM IN THE EUROPEAN UNION 163, 166 (Valerie Staats ed., European Communities Studies Association 2000). 13 The ECJ does not have the power to decide whether or not to hear a case lodged with its registry. On the contrary, according to SUP. CT. R. 10, [r]eview on writ of certiorari is not a matter of right, but a judicial discretion. A petition for writ of certiorari will be granted only for compelling reasons. 14 See the Court s Annual Report for 2005, 10 11, available at presentationfr/index.htm (last visited Sept. 1, 2006). 15 Art. 28 EC.

7 A constitutional court for the EU? 613 European law continually increases. Practically all forms of economic activity are already covered by Community legislation, and this legislation is subject to continuous modification and expansion. I hope that the future will prove me too pessimistic or even wrong. Yet I have few doubts that the ECJ s overburdened docket will remain so, assuming no structural change is put in place. That being the case, what changes can we envision to remedy the situation? 3. Should the ECJ be transformed into a specialized constitutional court or remain a supreme court? I will focus on what is, in my view, the best way to provide the ECJ with some breathing space. 16 As suggested above, there is no urgent need to transform the ECJ into a specialized constitutional court. It is perfectly feasible for the ECJ to remain a supreme court in charge not only of ruling on constitutional issues but, moreover, of safeguarding the unity and coherence of EU law. In the long term, I think, the ECJ should be doing only that, which means that it should not be called upon to decide mundane cases or, to put it differently, cases that do not raise constitutional issues or jeopardize the unity and consistency of EU law. Limiting the ordinary jurisdiction of the ECJ to important questions would have at least two advantages. First, the Court would be spared having to consider relatively unproblematic cases, those not requiring the time and attention of a supreme court for their resolution. As for the more important questions, the judges would have more time in which to balance the decisional options available to them and to provide more detailed legal reasoning in their judgments and orders. Second, it would allow the ECJ to hear more cases in the grand or the plenary chamber, on a more regular basis than is the situation today. This would ensure, at least, that a majority of the legal systems in the EU were represented when important judgment calls were required, since judges would no longer have to spread themselves thinly throughout the smaller chambers, as they do now, in order to address swiftly as large a number of cases as possible. In 1990, approximately 45 percent of cases were judged by chambers and 55 percent by the ECJ sitting in plenary session, whereas in 2005 the plenary and the grand chamber ruled on only 12 percent of cases. 17 Five-judge chambers are thus becoming the usual setting for hearing cases brought before the ECJ, which is unsatisfactory if one remembers that all, or at least a much larger number, of 16 I will not analyze other proposals, such as the relaxation of the test under Case 283/81, CILFIT [1982] E.C.R. 3415, point 16, as advocated by Advocate General Francis Jacobs in his opinion of July 10, 1997 in Case C-338/95, Wiener [1997] E.C.R. I-6495, para See the Court s Annual Report for 2005, supra note 14, at 11.

8 614 Int l J Con Law, Vol 4, No 4 (Oct 2006) B. Vesterdorf the national judicial systems should in principle have a say when important questions are addressed. Accordingly, in the long term, the ECJ should focus only on the main constitutional issues and on safeguarding the unity and coherence of EU law. This shift need not happen immediately, but it is the goal toward which the judicial system should and, I think, will evolve. 4. Limiting the jurisdiction of the ECJ to the most important issues: Practical problems and practical remedies While the prospect depicted above seems appealing, the devil, as always, is in the details. Thanks to the Treaty of Nice, we already have at our disposal the tools needed if we wish to ensure the ECJ is seized of only the most important cases. 18 Article 225(3) allows for the transfer of preliminary rulings to the CFI in specific areas. Such transfers would obviously alleviate the ECJ s caseload. Several commentators have suggested that the specific areas envisioned by article 225(3) could embrace such diverse fields as competition law, intellectual property law, customs law, or judicial cooperation in civil matters. 19 Advocate General Dámaso Ruiz-Jarabo Colomer, however, is emphatic about the need to maintain the integrity of the preliminary-ruling procedure: The uniform interpretation of Community law must, without exception, remain subject to the jurisdiction of the Court of Justice for preliminary rulings...the Court of First Instance should not be asked to share the task. The key to the success of the preliminaryruling procedure has lain in the centralisation of the interpretative function, which promotes uniformity... The day that two different interpretations are given by the two Courts in respect of the same precept of Community law, the death knell will sound for the preliminary-ruling procedure. The risk of confusion is not avoided by the [reference to] specific matters, since...different matters share common categories, institutions and legal principles... The preliminary-ruling procedure seeks to protect the law, in the manner 18 Treaty of Nice amending the Treaty on European Union, the Treaties establishing the European Communities and Certain Related Acts, 2001 O.J. (C 80). 19 See, e.g., the contributions of Koen Lenaerts, Michel Petite, Ingolf Pernice and James Flynn, The Role of the Court of Justice, the Court of First Instance and Specialised Tribunals in the Long Run, in PAPERS FROM THE COLLOQUIUM ON THE JUDICIAL ARCHITECTURE OF THE EUROPEAN UNION, Nov. 15, 2004, available at (last visited Sept. 1, 2006).

9 A constitutional court for the EU? 615 of a court of cassation, and there must be only one court of cassation in each legal order. 20 Others, too, have voiced their concern that a transfer of jurisdiction in any fields could jeopardize the unity and consistency of EU law, because this could result in the CFI having the final judicial word with regard to questions that may have a constitutional dimension or are inextricably linked with other areas of law not covered by the transfer of jurisdiction. 21 I understand these concerns, although I do not think those risks will materialize very often. What is more, I believe that a close look at the Treaty of Nice and the Statute of the ECJ reveals a comprehensive toolbox that may be relied on to avoid any such risk. In particular, article 225(3) says: [w]here the Court of First Instance considers that the case requires a decision of principle likely to affect the unity or consistency of Community law, it may refer the case to the Court of Justice for a ruling. It should be recalled that the CFI itself is sensitive to the consistency and unity of EU law. Accordingly, the CFI s right to transfer a question to the ECJ is in my view a safety valve that guarantees that the ECJ will continue to decide the most important questions. In addition, even if one were to doubt the CFI s ability or willingness to make such referrals, an additional remedy is provided in article 225(3): [d]ecisions given by the Court of First Instance on questions referred for a preliminary ruling may exceptionally be subject to review by the Court of Justice, under the conditions and within the limits laid down by the Statute, where there is a serious risk of the unity or consistency of Community law being affected. Pursuant to article 62 of the Statute of the ECJ, such exceptional review by the court is proposed by the First Advocate General. There is little chance that this procedure will create further delay because the same provision requires: (a) that the proposal be made within one month of delivery of the decision by the CFI; (b) that the ECJ decide whether or not the decision should be reviewed within one month of receiving the proposal of the First Advocate General; and (c) that the ECJ must give a ruling by means of an expedited procedure. 22 The process will be demanding for the First 20 Case C-17/00, De Coster [2001] E.C.R. I-9445, para See, e.g., Antonio Tizzano, La Cour de Justice Après Nice: Le Transfert de Compétences au Tribunal de Première Instance, REVUE DU DROIT DE L UNION EUROPÉENNE 665, 685 (2002), who warns of the risks of breaking the toy of preliminary rulings; and Angus Johnston, Judicial Reform and the Treaty of Nice, 38COMMON MKT.L.REV. 499, 509 (2001), who fears that there is still the danger that such a procedure could undermine the authority of preliminary rulings made by the CFI, having a detrimental effect on the respect paid to them by national judges. 22 Protocol on the Statute of the Court of Justice, signed at Nice on Feb. 26, 2001, 2001 O.J. (C 80)1, as lately amended by Council Decision of Oct. 3, 2005, 2005 O.J. (L 266) 60.

10 616 Int l J Con Law, Vol 4, No 4 (Oct 2006) B. Vesterdorf Advocate General and his chambers, but there is no reason to expect that he or she will not be able to cope with these new duties. Before rejecting the possibility of transferring preliminary rulings to the CFI in specific areas, it should first be proven that the remedies laid down in the Treaty of Nice are ill-conceived and insufficient. I have read many articles on the future of the EU judicial architecture and have found no compelling reason to distrust these safety valves. Let us consider other types of actions, in particular, appeals on points of law. Here, the Treaty of Nice provides that the Council may create judicial panels to hear and determine at first instance certain classes of action or proceeding brought in specific areas (art. 225a). With the creation of such specialized tribunals, the CFI will then have jurisdiction to hear and determine cases brought against any decisions of these judicial panels (art. 225[2]), which, in turn, will lighten the workload of the ECJ. Here again, for those who may voice concerns about the unity and consistency of the law, the safety valves already mentioned for preliminary rulings will be available. Nevertheless, when contemplating these reforms we must ensure that we transfer competences and not problems. Therefore, were it decided to transfer new functions to the CFI, we must be certain that the latter has the resources to deal with the new cases more quickly than the ECJ currently is able to do. This would require the creation of several specialized courts. I believe, therefore, that in the long term the ECJ should be in charge mainly of (a) interinstitutional litigation; (b) review of the CFI s judgments, either in appeals on points of law or, on a more exceptional basis, on judgments that relate to a competence transferred to the specialized courts; and (c) cases brought before the CFI that raise new constitutional questions and thus demand automatic transfer to the ECJ. Finally, should we consider transferring infringement actions to the CFI as well? I concede that it may not be palatable to the member states to relinquish the privilege of having their cases heard directly by the supreme court of the Union, and that, all in all, such a transfer would not necessarily bring about a significant gain in efficiency. On the other hand, few of these actions raise questions of constitutional law. For instance, actions that concern the noncommunication of transposing measures (that is, a member state s failure to comply with reporting requirements regarding the steps taken to incorporate EU directives into domestic law) usually do not raise questions of the magnitude to be treated by the ECJ. On reflection, we can envision a system in which most infringement actions would be dealt with by the CFI, while the ECJ would review the CFI s judgments as a juge de cassation (appeals on points of law). It could also rule directly on cases brought before the CFI that had a (new) constitutional dimension and, therefore, should be referred automatically to the ECJ.

11 A constitutional court for the EU? Conclusion In my view the time is ripe to make full use of the tools provided by the Treaty of Nice. I say full use because one of these tools has already been employed: October 1, 2005, saw the swearing in of the first seven judges of the new Civil Service Tribunal, the first panel created by the Council on the basis of article 225a. 23 This will be a significant test by which to assess the new system envisioned by the treaty. Staff cases typically raise several issues requiring the application of general principles of administrative law, which, by definition, also apply in other fields of EU law. We should, therefore, soon be able to determine whether the unity and consistency of these principles have unraveled to any significant degree. I am confident this will not happen. But I stress the decisive character of this test, since structural shifts, such as those contemplated by the Treaty of Nice, generally have to be implemented on a piecemeal basis. Changes of this magnitude are usually put in place on a large scale only after a first trial has proved successful. From a historical point of view, the first competences transferred to the CFI were limited. It was only after the CFI proved that it was up to the job that additional functions were transferred to it. The same applies to the Nice system. Reality naturally outwits our imagination, but I am sure that the new tribunal, with the CFI as a juge de cassation, will soon fulfill its new task with flying colors, which will also reinforce our confidence in the Nice system and bolster the political will that will be required to extend it to other fields of law. 23 By virtue of Council Decision 2004/752/EC, Euratom, O.J. (L 333/7) (establishing the European Union Civil Service Tribunal). The first judgment was rendered on April 26, 2006 (Case F-16/05, Nicola Falcione v. Commission [not yet reported in the E.C.R.]).