INTERINSTITUTIONAL CONVENTION AS CHECKS AND BALANCES IN EU LAW

KATHOLIEKE UNIVERSITEIT LEUVEN FACULTY OF LAW INTERINSTITUTIONAL CONVENTION AS CHECKS AND BALANCES IN EU LAW Thesis submitted in order to obtain the...
Author: Joel Richards
2 downloads 2 Views 7MB Size
KATHOLIEKE UNIVERSITEIT LEUVEN FACULTY OF LAW

INTERINSTITUTIONAL CONVENTION AS CHECKS AND BALANCES IN EU LAW

Thesis submitted in order to obtain the degree of Doctor in Laws

BART DRIESSEN

Supervisor: Prof. Dr KOEN LENAERTS

Academic year 2006-2007

Members of the Examination Commission Prof. Dr ANDRÉ ALEN (KUL) Prof. Dr DEIRDRE CURTIN (University of Utrecht) Prof. Dr BRUNO DE WITTE (European University Institute) Prof. Dr RENÉ FOQUÉ (KUL)

© 2006 Bart Driessen, Brussels, Belgium.

KATHOLIEKE UNIVERSITEIT LEUVEN FACULTY OF LAW

INTERINSTITUTIONAL CONVENTION AS CHECKS AND BALANCES IN EU LAW

Thesis submitted in order to obtain the degree of Doctor in Laws

BART DRIESSEN

Supervisor: Prof. Dr KOEN LENAERTS

Academic year 2006-2007

v

TABLE OF CONTENTS

Note to the reader.......................................................................................................................... xi Abbreviations ..............................................................................................................................xiii 1 1.1 1.2 1.3 1.4 1.5

The Research Question: checks and balances and interinstitutional convention..............3 Checks and balances.........................................................................................................................3 Interinstitutional convention ............................................................................................................5 Interinstitutional convention influences the checks and balances at different levels...................6 Research Question ............................................................................................................................7 Structure of this study ....................................................................................................................11 PART I: THEORETICAL OBSERVATIONS........................................................................13

2 2.1 2.2 2.2.1 2.2.2 2.3 2.3.1

2.3.2

2.3.3

2.3.4

2.4 2.4.1 2.4.2 2.4.3 2.4.4 2.4.5 2.4.6

Separation of powers and checks and balances .....................................................................15 Introduction to this Chapter ...........................................................................................................15 Separation of persons .....................................................................................................................17 Incompatibilities .............................................................................................................................17 Appointments of officers in other institutions ..............................................................................19 Separation of functions ..................................................................................................................19 Its general logic ..............................................................................................................................19 The alleged obsolescence of the separation of functions theory .................................................21 The ‘new powers’ ...........................................................................................................................22 Its implementation in national constitutions.................................................................................23 In presidential constitutions ...........................................................................................................23 In parliamentary systems: relationship between the executive and parliament..........................29 In parliamentary systems: positive or negative operation of the confidence rule; collective and individual responsibility ........................................................................................32 In parliamentary systems: parliament’s involvement in execution .............................................35 Separation of functions in the Member States and the United States .........................................36 Separation of functions in the European Union ...........................................................................37 An argument in favour . . . .............................................................................................................37 . . . And a critique ...........................................................................................................................39 Separations of powers as a doctrine in EU law ............................................................................43 A brief flirtation with separation of functions ..............................................................................43 ‘Institutional balance’ or ‘balance/separation of powers’............................................................44 Could separation of powers be a legal doctrine in the European Union? ...................................48 Separation of institutions ...............................................................................................................49 Main principle of the separation of institutions............................................................................49 The natural boundaries and different ‘esprit de corps’ in the different institutions ...................50 The division of state functions.......................................................................................................51 The organisational autonomy of the institutions ..........................................................................52 Autonomous and interinstitutional checks and balances .............................................................53 Interinstitutional checks and balances divided into those created by the European constitution and law, those created by interinstitutional convention and those formed by other institutions .............................................................................................................................54

TABLE OF CONTENTS

2.4.7 2.5 3 3.1 3.2 3.2.1 3.2.2 3.2.3 3.2.4 3.3 3.4 3.4.1 3.4.2 3.4.3

3.4.4 3.4.5 3.4.6 3.5

vi

Checks and balances on institutions should regulate the appointment or dismissal of their members ..........................................................................................................................................55 Conclusions of this Chapter ...........................................................................................................55 Interinstitutional convention......................................................................................................57 Introduction to this Chapter ...........................................................................................................57 Interinstitutional agreements in the European Union...................................................................57 Their relationship with law ............................................................................................................57 Specific legal bases for interinstitutional agreements ..................................................................63 General legal bases for interinstitutional agreements ..................................................................68 To what extent are interinstitutional agreements sufficiently concrete to be law? ....................75 Interinstitutional convention in the Member States .....................................................................80 Interinstitutional convention in the European Union ...................................................................82 The forms that it can take...............................................................................................................82 Its nature..........................................................................................................................................85 Its creation.......................................................................................................................................86 The generally more formalised character of conventions in the European Union .....................86 Formal requirements ......................................................................................................................87 The requirement of consent ...........................................................................................................88 Limitations to the capacity to conclude interinstitutional agreements........................................89 Can convention be abrogated?.......................................................................................................90 The Nice Declaration .....................................................................................................................90 Conclusions of this Chapter ...........................................................................................................92 PART II: SEPARATION OF PERSONS.................................................................................95

4 4.1 4.2 4.3 4.4 4.5 4.5.1 4.5.2 4.5.3 4.6 4.6.1 4.6.2 4.7

The appointment of the Commission .......................................................................................97 Introduction to this Part..................................................................................................................97 Development of the appointment procedure.................................................................................98 The Commission under the Constitution ....................................................................................103 Appointment of the president ......................................................................................................107 The Minister of Foreign Affairs ..................................................................................................108 The Secretary-General/High Representative ..............................................................................108 Appointment of the Minister for Foreign Affairs.......................................................................110 The MFA’s position in the Commission.....................................................................................110 Appointment of the College.........................................................................................................111 The vote of investiture; the possibility to refuse confidence .....................................................111 The allocation of portfolios; mid-term appointments; reshuffling............................................114 Conclusions of this Chapter .........................................................................................................116

5 5.1 5.2 5.2.1 5.2.2

The parliamentary system.........................................................................................................119 Introduction to this Chapter .........................................................................................................119 Political responsibility..................................................................................................................120 What is political responsibility? ..................................................................................................120 Extent of the Commission’s political collective responsibility .................................................123 Responsibility and censure ..........................................................................................................123 The Commission’s responsibility towards Parliament: ratione personae .................................124 The Commission’s responsibility towards Parliament: ratione materiae .................................126 The president’s role ......................................................................................................................127 Sanctioning responsibility: the motion of censure .....................................................................128

5.2.3 5.3

TABLE OF CONTENTS

vii

5.4 5.4.1 5.4.2 5.5 5.6 5.6.1 5.6.2 5.7

The Community’s parliamentary system ....................................................................................130 Rare use of the motion of censure ...............................................................................................130 From 1999.....................................................................................................................................133 The requested vote of confidence................................................................................................139 Parliamentary responsibility and the Council.............................................................................141 No relationship of responsibility between Council and Parliament in law...............................141 The relationship between Commission and Council..................................................................143 Conclusions of this Chapter .........................................................................................................144

6 6.1 6.2 6.3 6.4 6.4.1 6.4.2 6.4.3 6.4.4 6.5

Appointments of members of other institutions and bodies .............................................147 Introduction to this Chapter .........................................................................................................147 Executive offices ..........................................................................................................................147 Judicial offices ..............................................................................................................................151 Legislative and mixed offices......................................................................................................153 Members of the Court of Auditors ..............................................................................................153 The Executive Board of the European Central Bank .................................................................154 The Ombudsman ..........................................................................................................................156 The European Data Protection Supervisor..................................................................................157 Conclusions of this Chapter and of this Part ..............................................................................158 PART III: PARLIAMENTARY OVERSIGHT....................................................................159

7 7.1 7.2 7.2.1 7.2.2 7.2.3 7.2.4 7.3 7.3.1 7.3.2 7.3.3

7.3.4 7.3.5 7.4 7.4.1 7.4.2

7.4.3

7.4.4

Parliament’s information rights ..............................................................................................161 Introduction to this Part................................................................................................................161 Parliament’s general right to information ...................................................................................163 Development of Parliament’s general right to information .......................................................163 The Access to Documents Regulation ........................................................................................165 Sensitive information ...................................................................................................................167 A brief evaluation .........................................................................................................................169 Presence of the Commission and Council in Parliament and vice versa ..................................170 Organisation of this section .........................................................................................................170 The representation of the Commission before Parliament.........................................................170 The representation of the Council before Parliament ................................................................172 In plenary sessions........................................................................................................................173 In committees and similar meetings............................................................................................177 In the trialogue, the Conference of Presidents and other contexts ............................................178 The representation of Parliament and other institutions before the Council ............................179 A brief evaluation .........................................................................................................................180 Parliamentary questions ...............................................................................................................180 Different types of questions.........................................................................................................180 Written questions..........................................................................................................................181 Procedural aspects ........................................................................................................................182 Written questions posed to the Council ......................................................................................183 Oral questions ...............................................................................................................................184 Procedure ......................................................................................................................................184 In the CFSP and PJCC spheres....................................................................................................185 Question time................................................................................................................................186 General issues ...............................................................................................................................186 Procedure ......................................................................................................................................187

TABLE OF CONTENTS

viii

7.4.5 7.4.6 7.5 7.5.1 7.5.2 7.6

Other modes of Parliamentary questions ....................................................................................189 A brief evaluation .........................................................................................................................190 The Annual General Report and similar documents ..................................................................193 Green papers, white papers and reports ......................................................................................193 Reports ..........................................................................................................................................194 Conclusions of this Chapter .........................................................................................................195

8 8.1 8.2 8.3 8.3.1 8.3.2 8.3.3 8.4 8.4.1 8.4.2 8.5

Parliament’s erga omnes rights of oversight .........................................................................199 Introduction to this Chapter .........................................................................................................199 General right to adopt resolutions ...............................................................................................199 Hearings and committees of inquiry ...........................................................................................200 Before Maastricht .........................................................................................................................200 Committees of inquiry and temporary committees after Maastricht.........................................201 A brief evaluation .........................................................................................................................205 The right to petition and the Ombudsman ..................................................................................206 The right of petition......................................................................................................................206 The Ombudsman ..........................................................................................................................208 Conclusions of this Chapter and of this Part ..............................................................................211 PART IV: LEGISLATIVE PROCEDURES .........................................................................213

9 9.1 9.2 9.2.1 9.2.2

9.3.3 9.4 9.5

Legislation: general observations ............................................................................................215 Introduction to this Part................................................................................................................215 The Commission’s right of initiative ..........................................................................................216 Legislative and policy planning and the right of initiative ........................................................216 Legislative and policy planning...................................................................................................217 The Commission’s annual programme .......................................................................................217 Policy initiation by the Council and European Council.............................................................220 Exceptions to the Commission’s right of initiative ....................................................................225 A brief evaluation .........................................................................................................................231 Voting rules and conventions in the Council..............................................................................231 Qualified majority voting in law and practice ............................................................................231 The Luxembourg Compromise....................................................................................................232 Its background ..............................................................................................................................232 Its relationship with interinstitutional convention......................................................................235 Other voting conventions in the Council ....................................................................................238 Interinstitutional contacts during the legislative process ...........................................................240 Conclusions of this Chapter .........................................................................................................246

10 10.1 10.2 10.3 10.4 10.5 10.6 10.7

Parliament’s right of consultation ..........................................................................................249 Introduction to this Chapter .........................................................................................................249 The matters on which Parliament must be consulted .................................................................249 The consultation procedure in Parliament ..................................................................................250 Reconsultation ..............................................................................................................................253 The conciliation procedure contained in the 1975 interinstitutional agreement ......................256 Member State initiatives ..............................................................................................................261 Conclusions of this Chapter .........................................................................................................262

11 11.1

Co-decision ..................................................................................................................................265 Introduction to this Chapter .........................................................................................................265

9.2.3 9.2.4 9.3 9.3.1 9.3.2

TABLE OF CONTENTS

ix

11.2 11.2.1 11.2.2 11.3 11.3.1 11.3.2 11.4 11.4.1 11.4.2 11.4.3 11.4.4 11.4.5 11.5

Development of the co-decision procedure ................................................................................265 Interinstitutional convention and the co-operation procedure ...................................................265 Early co-decision ..........................................................................................................................268 Interinstitutional convention and co-decision.............................................................................271 The trialogue .................................................................................................................................271 Exchange of information..............................................................................................................273 The procedure in practice.............................................................................................................274 In first reading ..............................................................................................................................274 In second reading..........................................................................................................................275 May the Commission alter the common position?.....................................................................277 The Conciliation Committee........................................................................................................279 The third reading ..........................................................................................................................281 Conclusions of this Chapter .........................................................................................................282

12 12.1 12.2 12.3 12.3.1 12.3.2 12.3.3 12.4 12.5

Special legislative procedures...................................................................................................285 Introduction to this Chapter .........................................................................................................285 The assent procedure....................................................................................................................285 Open method of co-ordination.....................................................................................................287 Different ways of doing business ................................................................................................287 The role of Parliament and of the Commission ..........................................................................291 A brief evaluation .........................................................................................................................294 Sui generis acts .............................................................................................................................295 Conclusions of this Chapter .........................................................................................................300

13 13.1 13.2 13.2.1 13.2.2 13.2.3 13.2.4 13.2.5 13.3 13.3.1 13.3.2 13.3.3

Implementing policy ..................................................................................................................303 Introduction to this Chapter .........................................................................................................303 Interinstitutional convention before the Constitution ................................................................304 Comitology before the Single European Act..............................................................................304 Following the Single European Act ............................................................................................306 Improving the information flow to Parliament...........................................................................309 The ‘ultra vires’ action .................................................................................................................313 Parliament’s influence on executive decision: the right to ‘call-back’ .....................................315 Policy implementation under the Constitution ...........................................................................319 Hierarchy of norms.......................................................................................................................319 Delegated legislation vs implementing acts................................................................................321 Delegating legislation...................................................................................................................324 Current law: the regulatory committee procedure......................................................................324 The nature of delegation ..............................................................................................................325 What are ‘non-essential’ elements?.............................................................................................328 Article I-37: Implementing acts...................................................................................................329 Agencies........................................................................................................................................330 Two types of agencies..................................................................................................................330 Executive agencies .......................................................................................................................331 Statutory agencies.........................................................................................................................332 Conclusions of this Chapter and of this Part ..............................................................................334

13.3.4 13.4 13.4.1 13.4.2 13.4.3 13.5

PART V: BUDGET AND EXTERNAL RELATIONS ........................................................337 14 14.1

The budgetary procedure .........................................................................................................339 Introduction to this Chapter .........................................................................................................339

TABLE OF CONTENTS

14.2 14.3 14.4 14.5 14.6 14.7 14.7.1 14.7.2

x

14.7.3 14.8 14.9 14.10

Operation of the budget procedure under the EC Treaty ...........................................................343 The Gentlemen’s Agreement.......................................................................................................347 The classification of expenditure.................................................................................................348 The trialogue .................................................................................................................................351 The financial perspective .............................................................................................................352 The relationship between budget and legislation .......................................................................356 A problem with two sides ............................................................................................................356 The budget as the sole legal basis for expenditure; the legal status of remarks in the budget............................................................................................................................................357 Financial amounts mentioned in legislation ...............................................................................361 CFSP expenditure.........................................................................................................................362 The budgetary procedure under the Constitution .......................................................................364 Conclusions of this Chapter .........................................................................................................365

15 15.1 15.2 15.2.1 15.2.2 15.3 15.3.1 15.3.2 15.3.3 15.3.4 15.3.5 15.3.6 15.3.7 15.4 15.5

External relations.......................................................................................................................367 Introduction to this Chapter .........................................................................................................367 The presence of MEPs in EU delegations to international conferences ...................................368 The procedure ...............................................................................................................................368 A brief evaluation .........................................................................................................................371 International agreements ..............................................................................................................371 Introduction...................................................................................................................................371 Parliament’s right of consultation ...............................................................................................375 Its right of assent ..........................................................................................................................376 Trade agreements..........................................................................................................................379 Mixed agreements ........................................................................................................................381 Other agreements..........................................................................................................................385 Suspension of an agreement ........................................................................................................388 Interinstitutional convention in the context of CFSP .................................................................389 Conclusions of this Chapter .........................................................................................................390

16 16.1 16.2 16.3 16.3.1 16.3.2 16.3.3 16.4

Conclusions .................................................................................................................................393 Introduction to this Chapter .........................................................................................................393 Separation of institutions and checks and balances, revisited ...................................................397 Interinstitutional conventions, revisited ......................................................................................403 Their status in law ........................................................................................................................403 Interinstitutional conventions and the law ..................................................................................405 Interinstitutional conventions as a cause for constitutional change ..........................................407 Interinstitutional convention as checks and balances in EU law...............................................408 Cited literature, case law and documents...............................................................................413 Cited literature ..............................................................................................................................413 Case law ........................................................................................................................................430 European Parliament and Ombudsman documents....................................................................434 Council, Council presidency and European Council documents ..............................................436 Commission documents ...............................................................................................................441 Documents from other institutions and government authorities ...............................................443 Press reports..................................................................................................................................444

xi

NOTE TO THE READER

In this study I often refer to the European ‘constitution’ and to the ‘Constitution’. The first (with an undercast ‘c’) refers to the fundamental rules that govern the decisionmaking and controlling processes at Union level. The second (with a capital ‘C’) relates to the Constitutional Treaty that was negotiated in the IGC 2003 and which, at the time of writing, is in limbo, pending a resolution of the ratification process.1 With Von Bogdandy,2 I hope that the ‘Constitutional’ element of its title will ultimately be better remembered than the ‘Treaty’ bit. I therefore stuck to, what I hope, will turn out to become popular usage. As a further observation on terminology, the term ‘convention’ in this study does not refer to the ad hoc body that drafted the Constitutional Treaty, but is rather used analogously with the concept of ‘constitutional convention’. Furthermore, in order to improve readability, I refer to the European Parliament as to ‘Parliament’ wherever this could be done safely without causing confusion. I have endeavoured to describe events until the autumn of 2006. Soms it was possible to include later events. Lastly, the opinions in this study are solely attributable to me. They are not necessarily supported by the Council of the European Union, by its Legal Service or by any other institution. Brussels, 12 November 2006. B.D.

1 2

On the dichotomy between the two concepts Wade, Bradley a.o., Constitutional and administrative law (1993) at 4-5. ‘The prospect of a European republic: what European citizens are voting on’ (2005) at 916-917.

xiii

ABBREVIATIONS

CATS CE CFI CFSP Const CoR ECB ECHR ECJ EcoFin ECR ECSC EDF EESC EIB EPC ERDF EUA Euratom FAO GAERC GAG GATS IGC IIA IMO JHA JO MEP MFA MP MRI NCE OLAF OMC OOPEC PDB PJCC QMV SG/HR TEC TEEC TEU TRIPs

Article 36 Committee (Comité Article Trente-Six) Compulsory expenditure Court of First Instance Common Foreign and Security Policy Draft Treaty establishing a Constitution for Europe Committee of the Regions European Central Bank European Court of Human Rights European Court of Justice Council for Economic and Financial Affairs European Court Reports (Treaty establishing the) European Coal and Steel Community European Development Fund European Economic and Social Committee European Investment Bank European Political Co-operation European Regional Development Fund European Units of Account (Treaty establishing the) European Atomic Energy Community Food and Agriculture Organization of the United Nations General Affairs and External Relations Council General Affairs Working Party (Groupe Affaires Générales) General Agreement on Trade in Services Intergovernmental Conference Interinstitutional agreement International Maritime Organization Justice and Home Affairs Journal officiel (Official Journal French-language edition) Member of European Parliament Union Minister of Foreign Affairs Member of Parliament Maximum rate of increase Non-compulsory expenditure Office Européen de lutte anti-fraude (European Anti-Fraud Office) Open method of co-ordination Office for Official Publications of the European Communities Preliminary draft budget Police and Judicial Co-operation in Criminal Matters Qualified majority voting Secretary-General/High Representative Treaty establishing the European Community Treaty establishing the European Economic Community Treaty establishing the European Union Agreement on Trade-Related Aspects of Intellectual Property Rights

xiv

UN WTO

United Nations Organization World Trade Organization

1

‘Je voudrais rechercher, dans tous les gouvernements modérés que nous connaissons, quelle est la distribution des . . . pouvoirs, et calculer par là les degrés de liberté dont chacun d’eux peut jouir. Mais il ne faut pas toujours tellement épuiser un sujet, qu’on ne laisse rien à faire au lecteur. Il ne s’agit pas de faire lire, mais de faire penser.’1

‘But what is government itself, but the greatest of all reflections on human nature?’ 2

1 2

Montesquieu, De l’Esprit des Lois (1748), Livre XI, Chapitre XX. Madison, Federalist Paper 51 (1987).

1

THE RESEARCH QUESTION: CHECKS AND BALANCES AND INTERINSTITUTIONAL CONVENTION

1.1

Checks and balances

The question that this study seeks to answer is based on the following hypotheses. The idea of separation of powers is one of the fundamental principles underlying the constitution of the European Union.1 Many disagree, however, what it entails and to what extent it inspires the constitutional set-up of the Union. Separation of powers is, of course, not an end in itself. Questions relating to separation of powers theories are embedded in more general discussions as to the ability of the European Union to function effectively as a democratically legitimate constitutional system governed by the rule of law, in short, as a liberal democracy (Etat de droit). This disagreement on separation of powers is not limited to the European constitution. It is often observed that separation of powers theories have been only partially implemented in continental Europe and not at all in the United Kingdom (although both Montesquieu2 and the writers of the Federalist Papers got their inspiration from that country). Many lawyers contrast separation of powers with the parliamentary systems dominating most EU Member States and treat it as a synonym for ‘trias politica’. From a European perspective, separation of powers theories raise several fundamental questions. To begin with, one could wonder whether they are a model for analysis or a constitutional prescription. The distinction was not very clearly made by Montesquieu and, up to this date, some critics of separation of powers in the EU reason, perhaps not quite logically, that separation of powers is not relevant to the EU because (in their view) the EU’s constitutional system is not based on it. Then there is the question at what level separation of powers applies. At least in the EU context, there are several possible levels: On the individual level, primary law foresees certain incompatibilities of offices. For example, Article 6(1) of the Act concerning the election of the representatives of the Assembly by direct universal suffrage3 lays down that the office of MEP is incompatible with that of member of the government of a Member State, member of the Commission, judge, and a series of other offices listed in that provision. There is the functional level, based on a separation of ‘legislation’, ‘execution’ and ‘adjudication’. As Lenaerts has shown, 4 it is very well possible to analyse the Union in these terms (and indeed, the system for implementing policy proposed in the Constitution relies heavily on it). In this sense, separation of functions operates as a mechanism for the organisation of the constitution, balancing functions against each other. It also helps as an ‘organising device’ enabling citizens to come to grips with a constitutional system that differs radically from the well-known institutions of their 1 2 3 4

Lenaerts, ‘Some reflections on the separation of powers in the European Community’ (1991) at 14. Montesquieu, De l’Esprit des Lois (1995), Livre XI, Chapitre 6. OJ (1976) L278/5, amended by Council Decision of 25 June and 23 September 2002 amending the Act concerning the election of the representatives of the European Parliament by direct universal suffrage, annexed to Decision 76/787/ECSC, EEC, Euratom, OJ (2002) L283/1. Op. cit. footnote 1, ibid., ‘A new institutional equilibrium? In search of the “trias politica” in the European Community’ (1992) at 141 et seq.

CHAPTER 1: THE RESEARCH QUESTION

4

home state. However, separation of functions is not geared to act as a break on the powers of individual actors (persons, institutions) towards citizens. To the extent that it is conceived as a mechanism limiting the powers of public bodies, functional separation of powers theory takes on a different aspect: limitations on the powers of institutions (‘checks and balances’) presume a link between the ‘functions’ and, in the view of most authors, a triad of institutions. This trias politica variety of separation of powers owes much to James Madison. 5 It is this theory that is on most people’s mind when they argue that the constitutional system of the European Union, or of Member States, is not based on separation of powers because there is no correspondence between a triad of functions with a triad of key institutions. In this form, functional separation of powers would appear less suitable as a framework of analysis of the EU’s constitutional system. As one example—the question is discussed in more detail below—, one can point to the system of the current Treaties, which is based on an attribution of powers whereby no clear distinction is made between ‘legislation’ and ‘execution’ in the functional separation of powers sense. For instance, the Commission’s right under Article 85(1) TEC to investigate is (from a functional separation of powers point of view) executive, whilst its prerogatives under Article 81(3) TEC6 arguably are more legislative in character. The institutional level. The system of attribution of competences characterising the Treaties caused the ‘functions’ or ‘powers’ to be separated from the institutions. For example, whilst the American Constitution contains comparatively clear-cut rules assigning the executive and legislative functions to specific institutions, EU law comprises few corresponding guidelines.7 Thus, depending on the legal basis, the legislative ‘function’ is exercised by the Council; by the Commission, Council, and Parliament; by the Commission alone; by the Commission and Council; by the Council and Parliament; etc. At first sight, the Commission may seem to be the ‘executive’ in the Union, but even this is not a rule without important exceptions in, for example, the second and third pillars. The Constitution proposed important simplifications in the set-up of the Union, but more on the ‘pure’ functional separation of powers level than by providing checks and balances on a trias politica level. This study will argue that there is room for another level of analysis of the separation of powers system in the EU, namely at the institutional level. The unit of analysis—the institutions—lies in between the individual and the functional ‘power’. From this perspective, it is not so much the functions of authority that are (or should be) separated, but rather the institutions. On an institutional level, the separation of powers doctrine is then very much evident as the mechanism balancing different institutions (rather than individuals or ‘functions’). It is at this level that, for example, the debate is held whether comitology affects the powers of Parliament,8 or where Parliament’s influence over the open method of co-ordination is debated, or whether the European Council can be equated with the Council in its formation of heads of state or government. The appropriate framework of analysis here is not the Montesquivian trias politica, but rather the ‘checks and balances’ that ensure 5 6

7 8

This is explored in sub-section 2.3.1, below. ‘The provisions of [Article 81(1)] may . . . be declared inapplicable in the case of: — any agreement or category of agreements between undertakings, — any decision or category of decisions by associations of undertakings, — any concerted practice or category of concerted practices . . .’ Such general rules as exist are laid down in e.g. Articles 7(1), 189, 192, 202 and 211 TEC. In the following pages I will refer to the European Parliament simply as ‘Parliament’, where no confusion arises.

CHAPTER 1: THE RESEARCH QUESTION

5

equilibrium between the institutions. This brings the discussion into the nitty-gritty analysis of the ‘nuts and bolts’ that connect the institutions and keep the European governmental engine running. 1.2

Interinstitutional convention

The European Court of Justice emphasises9 that Parliament’s prerogatives are ‘one of the elements of the institutional balance created by the Treaties’. These prerogatives and other links between the institutions are partially forged in primary law, partially the result of secondary law, partially the result of interinstitutional convention, and partially of practice. An example of the first is formed by the majority of provisions in Part Five of the EC Treaty. The Financial Regulation10 is an illustration of the second category. Most lawyers tend to concentrate on these two categories. However, the legal ties binding the institutions are only a very limited part of the overall picture and the constitutional lawyer who limits himself to studying them will have studied the skeleton, but not the flesh, of the whole organism. Constitutional systems may develop a body of convention that is not quite law but is more than custom. As Dicey observed in the context of the British constitution, ‘a lawyer cannot master even the legal side of the . . . constitution without paying some attention to the nature of those constitutional understandings which necessarily engross the attention of historians or of statesmen’.11 This conclusion holds equally true for EU institutional law. In the EU this interinstitutional convention—formed through interinstitutional agreements, correspondence between institutions, statements by politicians, or simply practice—plays a crucial, but often overlooked, role. Such interinstitutional convention provides many of the checks and balances which (albeit imperfectly) balance the institutions, or vice versa, impair the spirit of the Treaties. It is not difficult to illustrate this with several random examples: For example, the right of inquiry is an example of interinstitutional convention that enhanced Parliament’s prerogatives. This right, used to great effect in the BSE case, started in the 1980s as a unilateral initiative of Parliament that later was sanctioned by interinstitutional practice and ultimately codified as Article 193 TEC.12 As a second example, in the budgetary procedure, all expenditure is divided into compulsory and non-compulsory. Parliament’s prerogatives are considerably more extensive where the latter of these categories is concerned and, consequently, the classification of expenditure is the Achilles’ heel of Parliament’s powers in this field. Throughout the last thirty years, the classification question has been resolved largely through interinstitutional convention.13 Thirdly, in the assent procedure, for acts other than international agreements, the Parliament has no right of amendment. In the past, this led to complaints that the prerogative amounts to little more than a negative veto, not permitting Parliament to participate seriously in the legal process. Even on this legally barren ground the seeds of Parliamentary practice germinated and Parliament’s Rules of Procedure now comprise a provision under which the parliamentary committee responsible can adopt

9 10 11 12 13

Case C-70/88, Parliament vs Council (Chernobyl), [1990] ECR I-2041, § 21. Council Regulation 1605/2002 of 25 June 2002 on the Financial Regulation applicable to the general budget of the European Communities, OJ (2002) L248/1. Dicey, Introduction to the study of the law of the Constitution (1982) at 277. The matter is discussed in more detail in section 8.3, below. The development of this is analysed in section 14.4, below.

CHAPTER 1: THE RESEARCH QUESTION

6

an interim report containing recommendations. 14 This can be the basis of interinstitutional discussions. It is well possible that, in due course, this practice will evolve into a more active Parliamentary participation in the decision-making process.15 1.3

Interinstitutional convention influences the checks and balances at different levels

The impact of interinstitutional convention operates at several levels. Firstly, there is the immediate level. Take, for example, the right of petition. This right was originally not regulated in the Treaties and started from the shallow basis of a unilateral Parliamentary instrument setting up the right to petition as well as a Committee on Petitions.16 Parliament subsequently concluded an interinstitutional agreement with the Commission and Council in which those institutions essentially recognised the petition procedure.17 This agreement arguably is not an instrument with the force of law of Article 249 TEC but, as will be evident from the discussion in sub-section 8.4.1, below, it certainly went beyond mere practice. Secondly, there is the ‘meta-constitutional’ impact of interinstitutional convention. The right to petition again provides a fine illustration of this: under the influence of the successful operation of the interinstitutional agreement, the drafters of the Maastricht Treaty included the current Article 194 TEC. This upgrades the petition procedure to constitutional status.18 The right to petition subsequently found its way into the Charter of Fundamental Rights of the European Union19 and into the Constitution.20 Another example is found in the ‘open method of co-ordination’ (OMC), a decision-making process that easily qualifies as the litmus test for any separation of powers theory. Apart from a few related procedures in the EC Treaty,21 this amorphous decision-making process has no explicit legal basis or structure foreseen in the Treaties. Under OMC procedures, Union-wide guidelines tend to be established which set specific targets, achievement of which is monitored under a common procedure for assessing results.22 OMC clearly has the potential to influence the distribution of powers foreseen by the Treaties, an issue Parliament has woken up to.23 It also potentially affects the Commission’s political responsibility to Parliament.

14 15 16 17 18

19 20 21 22 23

Rule 75(3). The assent procedure is discussed in more detail in section 12.2, below. OJ (1975) C179/31 and OJ (1977) C299/26. Exchange of letters between the European Parliament, the Council and the Commission of the European Communities, OJ (1989) C120/90. Somewhat superfluously, Article 21 TEC largely repeats Article 194, adding only that ‘[e]very citizen of the Union may write to any of the institutions or bodies referred to in this Article or in Article 7 in one of the languages mentioned in Article 314 and have an answer in the same language’. Article 44, OJ (2000) C364/1. Article I-10.2(d), II-104 and III-334 Const. See Articles 99 and 128 TEC. OMC is discussed in more detail in section 12.3, below. In 2001 it noted that it had ‘fewer reservations about the open method of coordination if it is confined to the exchange of procedures that have proved their worth and the comparison of progress achieved and does not apply in the sphere of preparation of European Council meetings from which Parliament is excluded’: Resolution on the Commission White Paper on European governance, OJ (2002) CE153/314, § 37.

CHAPTER 1: THE RESEARCH QUESTION

7

As such, it is an interesting example of interinstitutional convention affecting the checks and balances that govern the relationship between the institutions. These few examples suggest that the balance of powers between the institutions is affected by interinstitutional convention in multiple ways, whilst there is an interaction between law and interinstitutional convention. It is my contention that an analysis of the separation of powers system in the EU at institutional level cannot ignore interinstitutional convention, from the point of view of both its direct operation as well as its ‘meta-constitutional’ value. 1.4

Research Question

Before framing the Research Question that this study seeks to address, it is useful to place separation of powers in the broader context of the requirements of a (written or unwritten) democratic constitution. An extensive analysis of the concept of the requirements of the Rechtsstaat or, what in Anglo-Saxon legal cultures is termed, a liberal democracy would transgress the limits of this study, so I will limit myself to noting three parameters. Within these, many varieties of constitutions are possible, ranging from presidential systems to parliamentary democracies. Firstly, a good constitution should enable the polity to function; in Hamilton’s words, ‘[t]he public business must, in some way or other, go forward’.24 It should thus provide the institutions with the means to do what they have been conceived to do. Secondly, the constitution should ensure that public action is legitimate. This is not just a matter of having ‘more Parliament’. In the harsh words of two observers, ‘[i]n a constitutional order like that of the European Union, which is composed of sovereign States, it is futile to suppose that democratic legitimacy can be provided exclusively by way of the link of political accountability between individual citizens and a directed European Parliament’.25 Legitimacy implies more: a high level of transparency (to the extent that the first principle is not impaired), but also that the people must be able to provide policy making with a sense of direction through elections.26 Lastly, legitimacy implies guarantees of the rights of individuals and minorities.27 The legitimacy of its constitutional system is a problem not limited to the political system of the European Union, but it plays there to a greater extent than it does in the Member States. Notwithstanding the large number of odd entities that litter international law up to this day, in the post-Westphalian world most citizens are used to think of their public authorities in terms of states. Moreover, most Member 24 25 26

27

‘Federalist Paper 22’, in Madison, Hamilton and Jay (‘Publius’), The Federalist Papers (1987) at 180. Dashwood and Johnston, ‘The institutions of the enlarged EU under the regime of the Constitutional Treaty’ (2004) at 1482. I prefer putting it this way rather than using the difficult term ‘democracy’, which has very different connotations depending on culture and history. This theme fals outside the scope of this study and I will limit myself to the following comments. Certainly democracy is more than majority rule (at least since Thucydides’ time). Arguably, respect for minorities is part of the concept and, in a (semi-)federal structure such as the European Union, that justifies the overrepresentation of smaller states. This is my main criticism of Peters, ‘European democracy after the 2003 Convention’ (2004), who seems to regard the imperfect equality of individual citizens in European Parliament elections a blemish on the democratic character of the institution. Aguably, it is rather a necessary consequence. A matter not further discussed in this study.

CHAPTER 1: THE RESEARCH QUESTION

8

States have existed long enough to be part of the ‘mental furniture’ of their citizens and the legitimacy of the state as such is rarely contested. On the other hand, at least in some Member States there appears to be a disaffection with the (for want of a better word) political culture which is reflected in disappointing voter turnouts and a general disinterest in politics. The European Union suffers from the compounded and intrinsic problems that it is no state and that its political centre is far away for most voters. Moreover, its political system operates differently from that which is known by citizens (and therefore—rightly or wrongly—tends to be taken by them as the yardstick for democracy). As a result, the perception of citizens tends to differ. A scandal in domestic politics is a scandal, but the loyalty to the state is hardly affected. On the contrary, any political conflict in European politics is inflated to a crisis (at least in the eyes of the press) and part of the European electorate rejects the European Union or, at least, the image of the European Union that is presented to them. This study is hardly the place to address these issues and I will limit myself to the following comments. Firstly, in most national political systems, parliament is seen as the institution able to change the colour of the government. Indeed, this gives relevance to national elections. In the EU, there seems to be a public perception that the composition of the European Commission does not depend on European elections. Consequently, those elections are considered irrelevant by a large part of the electorate as a means to ‘vote the scoundrels out’ or even to change policy. When the electorate perceived a vote to have direct tangible results (such as the 2005 referenda in France and the Netherlands on the Constitution), turnout was considerably higher than in European Parliament elections28 This fact raises, among others, the interesting question whether the public assumption that European elections have no influence on the direction of European politics, is indeed correct. And if it is, can anything be done about it? These questions of—in today’s jargon—‘input legitimation’29 have relevance for the discussion below and, to the extent that they relate to the separation of powers system of the Union, will be discussed below. Thirdly, each institution must be checked through autonomous or interinstitutional checks and balances. In many constitutional systems parliament is given an important role as far as interinstitutional checks are concerned, but this role is far from exclusive. Indeed, as will be explored in the following Chapters, in many cases in the European Union, the intervention of other EU institutions is required to ensure the development, adoption and implementation of policy. It has been argued that ‘a more thorough study of the requirements flowing from [the trias politica principle] is needed so as to streamline a number of possible institutional adaptations in the Community government towards a more transparent system of “checks and balances”’. 30 If I am permitted to replace ‘trias politica’ with 28

29 30

In fact, very many factors contributed to the result of the referenda. It is impossible to list all of them but the following may convey a flavour: partly issues irrelevant to the question were at play (e.g., the accessions, the popularity of the national government, the euro); partly a perceived lack of information on the Constitution must be admitted. Additionally, the public perception in the Netherlands seems to have been that all of the Constitution concerned new rules which, of course, was factually equally incorrect. For a thorough analysis of voting behaviour in the Netherlands referendum see European times— Public opinion on Europe & Working hours, compared and explained (European Outlook 3), Netherlands Bureau for Economic Policy Analysis, 2005 at 13-42. Puder, ‘Constitutionalizing government in the European Union: Europe’s new institutional quartet under the Treaty establishing a Constitution for Europe’ (2004-05) at 88. Lenaerts, op. cit. footnote 1 at 14.

CHAPTER 1: THE RESEARCH QUESTION

9

‘separation of powers at institutional level’, then I believe that a review of the checks and balances that exist between the institutions will, in a number of cases, point to imbalances. Former ECJ Judge Koopmans observed nearly thirty years ago that the debate on Parliament’s competencies ‘suffers from the irritating peculiarity that no problem is seen separately from the others’;31 observations on the relationship between the Commission and the Council imply a statement on Parliament’s influence. Notwithstanding the drastic changes brought about by amendments to the Treaties, this remains a truism for the interaction between all political institutions. Take, for example, the second reading of the co-decision procedure. According to the letter of Article 251(3) TEC, the Commission is formally sovereign to decide whether it wishes to support Parliament’s amendments. It is true that Parliament can hardly resort to the motion of censure in order to enforce its amendments in a specific case. 32 In this sense the motion of censure is the nuclear option: the weapon is too strong to be really useful to use directly. I would argue, however, that Parliament’s real institutional position cannot be understood without reference to the intricate web of interinstitutional practice that gives colour to the provisions of the Treaty. The MEPs concluded a series of interinstitutional agreements with the Commission in which the latter accepted restrictions on the use of its rights. Thus in 2000, when it brought down the Santer Commission by successfully threatening to use its motion of censure,33 Parliament had the political clout to negotiate the Framework Agreement in which the Commission undertakes ‘to take the utmost account of amendments adopted at second reading; should it decide on important grounds, and following examination by the College, not to adopt or support such amendments, it shall explain its decision before the European Parliament . . .’ 34 In practice, rules such as these bind the Commission, perhaps not in law, but certainly politically (the reasons for that will be explored in Chapter 3). They reduce the policy discretion of the institution. The example illustrates two points. Firstly, interinstitutional convention may serve to channel the extreme but empty threat of the motion of censure into an obligation that forces the Commission to behave in a certain way. Secondly, as a direct consequence of this provision, the voting rules for the Council under Article 251(3) were affected and, unsurprisingly, that institution was miffed about it.35 This example leads to the following hypothesis: changes in the checks and balances regulating the relations between the institutions affect other checks and balances. Checks and balances are to a large extent created through interinstitutional 31 32 33 34 35

Koopmans, ‘Euro-verkiezingen: symbool of panacee’ (1978) at 2. Lenaerts, op. cit. footnote 1 at 25, writing in the context of the co-operation procedure. The matter was admittedly somewhat more complicated (as is explored below), but that does not affect the argument here. The 2000 Framework Agreement, OJ (2001) C121/122, § 5. It is currently § 31 of the 2005 Framework Agreement, OJ (2006) CE117/125 . The Council was ‘concerned at the fact that several clauses of the framework agreement provide for slippage in the institutional equilibrium created by the Treaties. It . . . reserves the right to preserve its rights and take any measure appropriate should the application of those clauses of the agreement impinge upon the Treaties’ allocation of powers to the institutions or upon the institutional equilibrium that they create’: 2282nd Council meeting (General Affairs), Brussels, 10 July 2000, Council doc. 10085/00 of 10 July 2000.

CHAPTER 1: THE RESEARCH QUESTION

10

convention. Hence changes in interinstitutional convention will often affect the constitutional position of institutions other than those party to the convention, either directly or indirectly. This brings me to the following fundamental question: To what extent is it possible to alter the balance between the EU institutions through the use of interinstitutional convention and what are the conditions for doing this? This is the Research Question that this study seeks to answer. Most analyses of the European Union’s separation of powers system (often relying on a paradigm different from the separation of institutions) are limited to providing anecdotal evidence. As may have followed from the previous argument, the Research Question requires a detailed analysis of current interinstitutional convention among the political institutions. This analysis is, however, subject to several limitations and caveats. Firstly, this study touches upon the other elements of the ‘Etat de droit’ referred to above. That means on the one hand, that the discussion must have regard for the requirement of efficiency. On the other hand, the analysis relates to what is commonly referred to as the ‘democratic deficit’ problem36 and the related question as to the legitimacy of the EU’s political system. Part of the latter issue concerns the disconnection between the citizen and the mechanisms for making and controlling policy at European level. As noted above, in the public perception there is no direct link between elections for Parliament and the composition of the Commission or the political direction of the Union. In this sense, there is a correlation between the legitimacy question on the one hand and the balance between the European institutions on the other. This implies that the Research Question cannot be regarded in abstracto but requires discussion against the background of the European political system as it currently functions. Secondly, in legal circles, part of the democratic deficit debate is due to a Babylonic confusion over the meaning of ‘democratic’, because most authors tend to measure the European Union system against their own Member State background.37 For example, the idea of an executive prerogative to legislate without recourse to parliament is abhorrent to British lawyers trained to consider Parliament as sovereign, but is not alien to French lawyers accustomed to Article 37 of their Constitution. What might be considered quite bizarre in the context of one Member State can be perfectly acceptable in another constitutional context. This is no different for the EU, and it may well be one of the intrinsic weaknesses in the European Union’s popular legitimacy. Europe’s constitutional practice contains many elements that were developed in one Member State and adopted, in a fashion, by the EU. A proper evaluation of these presumes a willingness to accept that a constitutional system may be different but not necessarily inferior to what is tried and tested. A recurring theme of this study is therefore that the EU system may have borrowed many elements from Member State constitutional practice; but although the offspring often resembles, it seldom is identical to the parent. Some comparison with 36 37

For a very thorough discussion of the literature on this issue, Verhoeven, The European Union in search of a democratic and constitutional theory (2002) at 57 et seq. Moreover, it raises the question whether ‘democracy’ implies ‘more parliament’ or more protection of the individual against the state—a ‘Rousseau vs Montesquieu’ distinction not dissimilar from the debates that plagued the drafting of the US Constitution: Kramnick’s ‘Editor’s introduction to the Federalist Papers’ (1987) at 43 et seq.

CHAPTER 1: THE RESEARCH QUESTION

11

Member State constitutional practice is useful since it may give an inkling how concepts in EU institutional law develop. However, great care must be taken not to judge the Union by reference to a particular Member State legal system. Thirdly, I do not argue that a good separation of powers system is the panacea for the democratic deficit problem. As noted above, the democratic issue also has to do with facets such as participation and inclusiveness of decision making, accountability and respect for fundamental rights. However, a proper system of checks and balances between the different institutions in the EU is a necessary element to solve the democratic deficit. Lastly, this study focuses on the ties that bind the Commission, Council, European Council and Parliament. The question whether (or to what extent) the European Council is an institution is discussed in sub-section 9.2.2, below. I will not discuss in detail the issue of standing of the institutions before the Courts. 1.5

Structure of this study

Before briefly sketching the structure of this study, two general observations are in order. Firstly, the writing of this study coincided with the drafting of the Constitutional Treaty (or, in the remainder of this study, simply the ‘Constitution’). Of course, this has not yet entered into force and, following the referenda in France and the Netherlands, it is at the time of writing not certain that it will do so in its current form.38 Still, it remains possible to analyse to some degree how it is likely to be implemented, for two reasons. To begin with—and this is no denial of some of the Copernican changes proposed in the Constitution with respect to the Treaties—, many of the Constitution’s provisions are very similar to those of the Treaties. This is especially so for many of the institutional provisions of the EC Treaty that have been reproduced almost literally in Title VI of Part III of the Constitution. Moreover, even where the Constitution would change the relationships between the institutions, Parliament will not easily accept interinstitutional conventions that provide it with less influence than what it currently enjoys. Current interinstitutional convention will therefore remain of relevance, even after the Constitution has entered into force. Additionally, even if the Constitution would not enter into force, it is well possible that, in the future, parts of the current text of the Constitution will find their way into primary law in the form of another amending treaty. Secondly, there is no correlation between the importance of a particular part of constitutional law and the quantity of interinstitutional convention it has produced. For example, the MEPs’ right to ask questions has produced considerable interinstitutional convention, even though it may not be the most important of their prerogatives. Consequently, the fact that certain conventions are discussed in great detail in this study does not imply a value judgment on their importance in the greater picture of the Union’s constitution. Vice versa, some areas of institutional law are discussed only marginally, since they failed to generate much convention. This study is divided in five Parts. The first Part discusses the two poles that make up the framework of analysis. Chapter 2 discusses different modes of separation of powers and provides a framework of analysis of checks and balances. The 38

Piris, The Constitution for Europe—A legal analysis (2006) at 30 et seq. argues cogently that, at the time of writing, the race has not been run yet. In any event, even if thisConstitution is ultimately not adopted, many parts of it would likely find there way into the Treaties through some other amending instrument.

CHAPTER 1: THE RESEARCH QUESTION

12

discussion will not overly focus on the history of separation of powers thinking, but is rather geared towards the specific problems posed by the EU’s constitutional system. Interinstitutional conventions are analysed in the following Chapter. Part II deals with checks and balances related to what will be called the ‘separation of persons’. In Chapter 4 the appointment procedure of the Commission is discussed, both as regards past practice and convention and as it would operate under the Constitution. Chapter 5 is devoted to the parliamentary system as it operates between the Commission and Parliament. This implies an analysis of the Commission’s responsibility to Parliament and of its sanction, the motion of censure. The following Chapter discusses the appointment procedures for a number of offices in the European Union and focuses on interinstitutional conventions that have come into being in this field. Part III discusses the checks and balances that constitute Parliament’s rights of oversight over policy making by the Commission and the Council. These are the tools with which Parliament can obtain political responsibility. They fall apart into rights that the MEPs exercise vis-à-vis Commission and or the Council, and into rights with an erga omnes remit. Chapter 7 deals with the first, concentrating on Parliament’s information rights. The next Chapter analyses the interinstitutional conventions that came into being as far as the latter are concerned. The most important prerogatives concerned are the committees of inquiry and the right of petition. The whole of Part IV is dedicated to the checks and balances concerning the influencing of policy and its result, legislation. Chapter 9 discusses horizontal aspects of legislative procedures and focuses on the interinstitutional conventions that rule the Commission’s right of initiative, legislative planning, and policy initiation by the Council and European Council. That Chapter also analyses the conventions that govern voting in the Council. This is followed in Chapter 10 by a discussion of the interinstitutional conventions developed in the context of the consultation procedure. The succeeding Chapter deals with interinstitutional conventions in the field of co-decision. In this context, the co-operation procedure, which in a way preceded codecision, is also briefly discussed. Chapter 12 discusses the remaining legislative procedures. For reasons explained in section 12.3, the open method of co-ordination is also discussed in this Chapter. Lastly, the assent procedure and the remaining sui generis procedures will be briefly reviewed. The last Chapter of this Part deals with implementing policy. Apart from a review of the interinstitutional conventions that have come into being in the field of comitology, this includes an analysis of the checks and balances in the system foreseen by the Constitution. Part V deals with two rather different aspects of EU policy that each engendered considerable interinstitutional conventions. Chapter 14 discusses the interinstitutional conventions that have come into being in the field of the budget. The following Chapter reviews the interinstitutional conventions in the field of external relations. Finally, in Chapter 16 the conclusions of the study are drawn, and the Research Question posed above will be answered.

PART I: THEORETICAL OBSERVATIONS

2

SEPARATION OF POWERS AND CHECKS AND BALANCES

2.1

Introduction to this Chapter

This Chapter analyses the main aspects of separation of powers theories and attempts to suggest a separation of powers theory that works for the European Union. Many different models have been developed that aimed at mitigating the power of the state. I have no desire to venture into any detailed historical analysis of these.1 Often the term separation of powers is associated with Montesquieu but this is a gross simplification.2 Similarly, separation of powers is not the same thing as the ‘trias politica’ and it is well possible to imagine models that are very different.3 Montesquieu and other thinkers further developed the theories of mixed government that had been in vogue especially in England. In brief, the French jurist stressed the need to separate ‘la puissance législative’, ‘la puissance de juger’ and ‘la puissance exécutrice’.4 Academically, there is some discussion whether the separation element was really the key of Montesquieu’s theory.5 Mabille6 notes that Montesquieu never used the expression ‘séparation des pouvoirs’. This is correct (he only uses the word ‘séparée’ in the context of the independence of the judiciary), although at least Montesquieu’s compatriots interpreted him in this manner. For example, Article 16 of the Déclaration des Droits de l’homme et du citoyen of 26 August 1789 pronounced that ‘[t]oute Société dans laquelle la garantie des Droits n’est pas assurée, ni la séparation des Pouvoirs déterminée, n’a point de Constitution’. The French Constitution of 1791 applied this with such dogmatic rigidity that the point of the theory got lost and the institutions of the French state became unworkable. Be this as it may, Montesquieu’s contribution was to put the descriptive theory of separation of powers on the map as a normative one.7 The idea exerted great influence

1

2 3

4 5 6 7

The idea of distinguishing different governmental powers possesses a pedigree well older than Montesquieu: see e.g. Aristoteles’ Politica at book IV, Chapter 14. For the history of the concept see Vile’s Constitutionalism and the separation of powers (1998), Crossland, The separation of powers and the European Communities (1983) and Georgopoulos, ‘The “checks and balances” doctrine in Member States as a rule of EC law: the cases of France and Germany’ (2003), 532 et seq. Sári’s A hatalommegosztás (1995) at 9-74 is similarly very instructive. For one example see Allio and Durand, From legislation to implementation: the future of EU decision-making (2003), 7 et seq. For example, some of the medieval republics in Italy devised a system of government under which a foreigner would be appointed magistrate for a predetermined period. The conditions and rewards of his labours would be agreed between the governor and the governed. History tells us that the administrator ‘was not permitted to purchase land, to contract an alliance, or even to accept an invitation in the house of a citizen; nor could he honorably depart till he had satisfied the complaints that might be urged against his government’: Gibbon, The decline and fall of the Roman Empire (1994), Chapter LXIX, Volume 6 at 541. Montesquieu, De l’Esprit des Lois (1995), Livre XI, Chapitre 6. E.g. Crossland, op. cit. footnote 1 at 49-54, Marc-Lipiansky, ‘Problématique de la séparation des pouvoirs’ (1993) at 7. Législatif, exécutif, judiciaire—Les rapports entre les pouvoirs (1998) at 20. Although Locke in his Two Treatises of Government (1965), Second Treatise §§ 143-148 at 409-412 could equally claim the credit—but his distinction of ‘powers’ into a legislative, executive and federative branch was different.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

16

in constitutional legal thinking.8 In fact, Montesquieu’s ‘trias politica’ theory became so popular that, as late as 2002, the Belgian Prime Minister pleaded for it to be introduced into the European constitution.9 The popularity of Montesquieu’s brand of separation of powers (or at least, of what he was interpreted to say) had one consequence that has remained a consistent feature of separation powers theories up to this day. Montesquieu disguised his comments in the form of a description of the idealised constitution of England. Much of the discussion that followed him has been caused or influenced by the failure to distinguish between the prescriptive and analytical virtues of separation of powers theories.10 The importance of the distinction becomes particularly pertinent in the context of EU constitutional law. As will be seen in sub-section 2.3.3 below, many of the criticisms of separation of powers in the European Union pivot on the argument that the EU lacks a trias politica structure. Following this logic, the EU’s constitutional structure cannot clearly be divided in three ‘powers’; hence separation of powers should not apply. I will return below to this issue. Montesquieu’s views were taken up but substantially modified by Madison in the Federalist Papers and, as such, found currency in American constitutional thinking.11 From there, they were re-exported to Europe and, as we shall see, influenced the principles underlying many European constitutions. In the meantime two simultaneous thought processes took place. The first was the restriction of separation of powers theories to the trias politica. The second is that as early as Kant’s writings,12 the trias politica theory in Continental Europe formalised into a dogma and this influenced much of the legal literature on the subject in the following centuries.13 As a result of this, much of the legal literature confounds separation of powers with trias politica and treats the whole matter as a religion that is either followed as received wisdom, or rejected as theoretical folly. In the following pages I intend to analyse the concept of separation of powers. Vile has observed that the word ‘power’ can have five different meanings: ‘the possession of the ability through force or persuasion to attain certain ends, the legal authority to do certain acts, the “function” of legislating, executing, or judging, the agencies or branches of government, or the persons who compose these agencies’.14 In my opinion, this causes the separation of powers doctrine to operate at three levels.15 I will discuss these under the following headings: separation of persons 8 9 10 11 12 13 14 15

For more historical background see Sári, op. cit. footnote 1 at 9-40. ‘België voor scheiding van machten in EU’, De Tijd of 19 November 2002 and ‘Du Montesquieu dans l’Europe’, Le Soir of the same date. For the sake of completeness I must note that it is also possible to consider the separation of powers as a legal doctrine. This is discussed in sub-section 2.3.4, below. I will not treat into the historical controversy to what extent Montesquieu’s influence was decisive. For a description of the dispute see Vile, op. cit. footnote 1 at 133-135. Kant, Metaphysische Anfangsgründe der Rechtslehre (1988), Zweiter Teil, § 45 at 127 et seq. For an example of this century, see Dreier, ‘Die drei Staatsgewalten im Zeichen von Europäisierung und Privatisierung’ (2002), 537-547. Op. cit. footnote 1 at 13. Stern, Das Staatsrecht der Bundesrepublik Deutschland—Band I (1984) at 795 distinguishes a ‘substantive’, ‘organisational’ and ‘personal’ separation of powers. See also Wade, Bradley a.o., Constitutional and administrative law (1993) at 58, Sári, op. cit. footnote 1 at 15 and Woodhouse, ‘The English judges, politics and the balance of power’ (2003) at 920.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

17

(section 2.2), functional separation of powers or, more logically, separation of functions (section 2.3); and separation of institutions (section 2.4). I should note that in the past not every author on the subject has taken care to make this distinction and the terminology used in much of the literature on the subject tends to be confused. Moreover, different legal traditions accord different interpretations to it. For example, since the words regterlijke macht (‘judicial power’) were introduced in Dutch constitutional law through the Constitution of 1814, they acquired the specific meaning of the aggregate of certain courts and legal offices. 16 Obviously, any meaningful analysis of separation of powers obliges the reader to transcend such national connotations. 2.2

Separation of persons

Separation of persons17 consists of two types of prescriptions. The first is known as incompatibilities (sub-section 2.2.1). The second concerns limitations on the appointment of officers in other institutions (sub-section 2.2.2). 2.2.1

Incompatibilities

Incompatibilities18 are the only level of separation of powers that is a direct implementation of the unchecked Montesquivian brand of separation of powers.19 In the words of the French baron, ‘[t]out serait perdu, si le même homme, ou le même corps des principaux, ou des nobles, ou du peuple, exerçaient ces trois pouvoirs: celui de faire des lois, celui d’exécuter les résolutions publiques, et celui de juger les crimes ou les différends des particuliers.’20 Generally, the Member States’ constitutional laws proscribe government ministers or members of parliament to be judges. In many Member States there is a prohibition between combining the membership of parliament with that of the government.21 In EU law, incompatibilities occur at two levels. Firstly, there is the Union level. Secondly, certain functions within the Member States cannot be combined with certain Union offices. As far as Member State ministers are concerned, the two levels merge since they are qualitate qua members of the Council of the EU. The main provisions are the following:  Article 6(1) of the Act concerning the election of the representatives of the Assembly by direct universal suffrage22 lays down that the office of MEP is incompatible with that of member of the Government of a Member State, 16 17 18 19 20 21 22

Van der Pot, Donner a.o., Handboek van het Nederlandse staatsrecht (1995) at 438. The term is Vile’s, op. cit. footnote 1 at 18. For an overview of the incompatibilities between European Parliament membership and national offices in Member State law see Crespo Allen a.o., National provisions concerning ineligibility and incompatibility with regard to the European Parliament (1997). Jorda, Le pouvoir exécutif de l’Union européenne (2001) at 51. Op. cit. at footnote 4. In a similar vein Madison in ‘Federalist Paper 47’, in Madison, Hamilton and Jay (‘Publius’), The Federalist Papers (1987) at 303. In Belgium the incompatibility between the membership of parliament and the function of minister was introduced only in 1995: Mabille, op. cit. footnote 6 at 10. OJ (1976) L278/5, amended by Council Decision of 25 June and 23 September 2002 amending the Act concerning the election of the representatives of the European Parliament by direct universal suffrage, annexed to Decision 76/787/ECSC, EEC, Euratom, OJ (2002) L283/1.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

18

member of the Commission, judge, and a series of other offices listed in that provision;23  Article 213(1) TEC24 requires that the ‘independence [of the Commissioners] is beyond doubt’, which implies inter alia an incompatibility with the membership of a Member State government.25 Article I-26.4 Const says essentially the same thing, apart from requiring Commissioners to show ‘European commitment’;  Article 223 TEC requires that ECJ Judges and Advocates-General be chosen from ‘persons whose independence is beyond doubt’. Article 224 TEC lays down a similar requirement for CFI Judges. A similar requirement is provided in Article I-29.2 Const. In addition to these provisions, Article 4 of the Statute of the Court of Justice provides inter alia that ‘[t]he Judges may not hold any political or administrative office’;  Article 195(3) TEC and I-49 Const, repeated in Article III-335.3 Const, require the Ombudsman to be ‘completely independent in the performance of his duties’;  Under Articles 263 TEC and III-386 Const, it is impossible to combine membership of the Committee of the Regions with that of the European Parliament;  Article I-22.3 Const provides that the president of the European Council ‘shall not hold a national office’. Apart from these, note Article 203(1) TEC: ‘The Council shall consist of a representative of each Member State at ministerial level, authorised to commit the government of that Member State’.26 This provision therefore defines the composition of that institution, rather than laying down an appointment procedure for it. Hence the nature of the Council implies its independence vis-à-vis other institutions. There is one other provision of interest in EU law. The former Article 5 of the Act concerning the election of the representatives of the Assembly by direct universal 23

24 25

26

In a consolidated text: ‘The office of representative in the Assembly shall be incompatible with that of:  member of the Government of a Member State,  member of the Commission of the European Communities,  Judge, Advocate-General or Registrar of the Court of Justice of the European Communities or of the Court of First Instance,  member of the Board of Directors of the European Central Bank,  member of the Court of Auditors of the European Communities,  Ombudsman of the European Communities,  member of the Economic and Social Committee of the European Economic Community and of the European Atomic Energy Community,  member of committees or other bodies set up pursuant to the Treaties establishing the European Economic Community and the European Atomic Energy Community for the purpose of managing the Communities’ funds or carrying out a permanent direct administrative task,  member of the Board of Directors, Management Committee or staff of the European Investment Bank,  active official or servant of the institutions of the European Communities or of the specialised bodies attached to them or of the European Central Bank.’ Paragraph 2 of the same provision lays down that ‘[i]n addition, each Member State may, in the circumstances provided for in Article 7(2), lay down rules at national level relating to incompatibility’. Formerly Article 10(1) of the Merger Treaty. Following Commissioner de Silguy’s short-lived plans to seek election in a local council in Brittany, Commission President Santer declared during the Parliamentary debate of 28 January 1998 that ‘the exercise of an elected mandate, even at local level, risks harming the independence and availability of our Commissioners’. Roughly corresponding to Article I-23.2 Const.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

19

suffrage27 provided explicitly that being an MEP was compatible with a national parliamentary mandate. This provision was a borderline exception to the separation of persons. It was repealed in 2002 as part of the overhaul of the Act,28 following the wishes of the MEPs.29,30 2.2.2

Appointments of officers in other institutions

The second kind of prescriptions flowing from the separation of persons rule concerns the appointment of officers in one institution by another. Theoretically, the point of incompatibilities would be lost if the officer who cannot be simultaneously a member of two ‘powers’ would have an unfettered right to appoint the officers of another ‘power’. To a certain extent, the basic concept lay dormant in Montesquieu’s work, but it is expressed best by Madison: ‘In order to lay a due foundation for that separate and distinct exercise of the different powers of government, which to a certain extent is admitted on all hands to be essential to the preservation of liberty, it is evident that each department should have a will of its own; and consequently should be so constituted that the members of each should have as little agency as possible in the appointment of the members of the others.’31 That prescription cannot, however, be taken to its logical extreme. For example, should one insist that the judiciary be fully independent of the other ‘powers’, the consequence should be a kind of co-optation system among judges—not a healthy prospect at all, and I know of no modern state where it was even contemplated. Madison arrived at a similar conclusion and this aspect of separation of persons is best discussed below, in the context of ‘checks and balances’. 2.3

Separation of functions

2.3.1

Its general logic

The second level at which separation of powers is relevant is the substantive32 or functional level. Which powers form ‘the legislature’, ‘the executive’ and ‘the judiciary’? Separation of functions focuses on the functions of government. Rather than separating the officials, the analysis focuses on the functions of government, which are listed as the triad of legislation, administration (execution) and adjudication. Montesquieu argued that ‘[i]l pourra arriver que la constitution sera libre, et que le citoyen ne le sera point. Le citoyen pourra être libre, et la constitution ne l’être 27 28 29 30

31 32

OJ (1976) L278/5. See footnote 22, above. Contrary to what is suggested by Crespo Allen a.o., op. cit. footnote 18 above, Parliament disfavours double mandates: Resolution on the incompatibility between the office of Member of the European Parliament and member of a national parliament, OJ (1988) C235/131. In addition to these incompatibilities of a constitutional nature, there are other legal incompatibilities, such as e.g. the incompatibility between the duties of authorising officer and accounting officer in the institutions: Article 58 of the Financial Regulation, OJ (2002) L248/1. These are not further dscussed in this study. Madison in ‘Federalist Paper 51’, op. cit. footnote 20 at 319. In the language of Stern, op. cit. footnote 15 at 795.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

20

pas’.33 Elsewhere34 he observes that ‘les trois pouvoirs peuvent être bien distribués par rapport à la liberté de la constitution, quoiqu’ils ne le soient pas si bien dans le rapport avec la liberté du citoyen’. Interestingly, the Court of Justice made much the same distinction in Vreugdenhil:35 ‘the aim of the system of the division of powers between the various Community institutions is to ensure that the balance between the institutions provided for in the Treaty is maintained, and not to protect individuals’. The doctrine covers essentially two approaches to separation of functions and these correspond to the two arms of this paradox. I refer to these in shorthand as I. The ‘pure’ form and II. The ‘Madison-style’ separation of functions. In its pure form, separation of functions holds that ‘each function can fulfil its tasks in an efficient way only when at least one power co-operates to that effect, thus controlling the use which the first power makes of its authority’.36 In this form, the theory is relevant for the European Union and indeed, influenced the separation between legislation and implementation that characterises Articles I-33 to I-37 Const.37 Separation of functions, in this interpretation, acts as an organisational principle laying down which institutions should not simultaneously do what. However, it does not further limit the actions of each institution separately. ‘Madison-style’ separation of functions. Often separation of functions is interpreted somewhat differently. The authors of the American Constitution faced the question, not only how to distribute the powers of the institutions, but also how to check each of them. Vile considers that the American constitutional system is characterised by the ‘twin doctrines of the separation of powers and checks and balances’.38 Interestingly, this was not the opinion of the authors of the Federalist Papers: Madison noted that ‘[o]n the slightest view of the British Constitution, we must perceive that the legislative, executive, and judiciary departments are by no means totally separate and distinct from each other.’39 Clearly, Montesquieu had not intended to imply that the legislative, executive and judicial branches of government should be wholly separate. On the contrary, the departments should have ‘partial agency’ or ‘control’ over the acts of each other: ‘where the whole power of one department is exercised by the same hands which possess the whole power of another department, the fundamental principles of a free constitution are subverted . . . [Separation of powers] goes no farther than to prohibit any one of the entire departments from exercising the powers of another department.’40

33 34 35 36 37 38 39 40

Montesquieu, op. cit. footnote 4, Livre XII, Chapitre 1. On this issue and its impact on Montesquieu’s separation of powers thinking, Péter, ‘Montesquieu’s paradox on freedom and Hungary’s Constitutions, 1790-1990’ (1991) at 3 et seq. Op. cit. footnote 2, Livre XI, Chapitre 18. Case C-282/90, Industrie- en Handelsonderneming Vreugdenhil BV vs Commission, [1992] ECR I-1937, § 20. Lenaerts, ‘Some reflections on the separation of powers in the European Community’ (1991) at 11. This distinction is discussed in more detail in Chapter 13, below. Vile, Politics in the USA (1999) at 6. Madison in ‘Federalist Paper 47’, op. cit. footnote 20 at 303. Ibid. at 304-305.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

21

Madison’s brand of separation of functions consisted of mapping each of the three ‘functions’ onto a triad of institutions, paired with a separation of persons. The concern here is less the distribution of power within the state, but rather the use of separation of powers to protect the rights of individuals. In this form, separation of functions became part and parcel of the American constitution. Whether or not Madison’s separation of powers and checks and balances are one or two ‘theories’ is, in my view, by now largely a semantic or academic matter that need not detain us.41 For example, as Sári42 argues, it is a matter of viewpoint whether the presidential veto in the US is considered as part of the legislative procedure, or as a check on the rights of the legislature. In the following, I will treat Madison’s views as one aggregate whole of ideas and refer to it as the ‘Madison-style’ variety of separation of functions. In this form, the theory received much criticism in Europe. Critics often confound separation of powers theory with an exaggerated interpretation of Montesquieu’s teachings, which they then reject.43 In other cases, more subtle objections have been raised against the separation of functions concept (although the object of the critic was not always termed or identified as such), that I need to address. In a way, the whole British constitution can be seen as a criticism of the separation of functions principle. This aspect is discussed in sub-section 2.3.2, below. The remaining criticisms concern its alleged insufficiency to deal with the interventionist character of modern governments, and the ‘new powers’. The alleged obsolescence of the separation of functions theory It is almost commonplace to observe that the function of ‘execution’ or ‘administration’ no longer describes the activities of modern government. In the liberal states of the 19th Century, government involvement was minimalist44 and this fitted with the notion that states are ruled by laws, not men. Since those days, governments discovered the subsidy tool to steer developments in society. Limitations on the funds states could spend, and the number of causes deserving them, necessitate choices to be made. This causes conflicts between needy goals. Governing became conflict management.45 It is thus that governments had to develop policy guidelines in order to obtain their objectives. The executive grew in tandem with this burgeoning government intervention. Governments nowadays comprise more than the ministers; in any modern state, large bureaucracies govern46 and develop policy.47 41 42 43 44 45 46

47

The two theories view is powerfully supported by Vile, op. cit. footnote 1 at 36 et seq., although he notes the close connection between them. For the opposite view see Witteveen a.o., De sociale rechtsstaat voorbij. Twee ontwerpen voor het huis van de rechtsstaat (2002) at 61-62 and at 119. Op. cit. footnote 1 at 47. For example Hood Phillips and Jackson, Constitutional and Administrative Law (1987) at 13 are not wholly innocent. Van der Pot and Donner, op. cit. footnote 16 at 123. The term and the gist of the argument are from Van der Pot and Donner, op. cit. footnote 16 at 469-470 and 536. For similar views see Dreier, op. cit. footnote 13. Koopmans, ‘De rol van de wetgever’ (1970) at 10 and Club de Florence, Europe: l’impossible statu quo (1996) at 99. Scheuner, Staatstheorie und Staatsrecht (1978) at 39 distinguishes ‘Gesetzgebung’, ‘Regierung’, ‘Verwaltung’ and ‘Rechtsprechung’. He notes that the ‘feine und bewegliche Spiel des Zusammenwirkens’ between government and parliament ‘kann weder von einem starren, als Funktionentrennung mißdeuteten Gewaltenteilungsprinzip noch von einem reinen Kompetenzdenken erfaßt werden’ (ibid. at 295). Crijns, ‘Geen wetgevende bevoegdheden voor het Europees Parlement!’ (1987) at 72.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

22

Critics acknowledge that the tools of the policy maker certainly continue to include legislation and in parliamentary democracies the government, at some point, needs the sanction of statute law to cover its policies. However, guidelines and targets have become more important than norms. Policy replaces the simple notion of execution. Modern parliaments possess neither the expertise nor the means48 to adopt all generally applicable rules themselves and, often, must leave the adoption of details to the administration. Thus, in many cases, delegation is a more apt term than the trias politica notion of execution.49 The consequence, a much closer relationship between the legislative process and its implementation, renders the separation between the two outdated as a product of the 19th Century liberal state.50 Nevertheless, these criticisms do not in themselves discredit the separation of functions. An argument can be made that the critics of the separation of functions ignore the intrinsic connection between the rule of law—which presumes that the administration rests on a firm foundation of statutes—and the controlling functions of parliament. The combination of both tenets creates a system in which the parliament exercises final authority:51 it can amend or ultimately veto legislation, which is the ultimate basis for both the administrative power and the secondary legislation promulgated by the administration. Parliament’s prerogatives of oversight should thus be arranged in a manner allowing the effective control of the administration. Hence it is unnecessary that the parliament is involved in the creation of every single norm. As long as implementing legislation is based on a statute passed by parliament, such implementing legislation can be considered ‘execution’ rather than ‘legislation’. In this perspective, even the creation of law through jurisprudence is compatible with the separation of functions doctrine. The ‘new powers’ As early as 1972, Ipsen noted that non-state ‘powers’ (the social partners, the media, etc.) intervened and participated in the decision-making process. This, in his view, made the ‘überlieferte staatsrechtliche Gewaltenteilung’ outdated even for regular states, let alone the EEC: ‘[e]s gibt kein absolut herrschendes Gewaltenteilungsprinzip . . . das für jedwede Verfassungsgestaltung als maßgebliche Rezeptur mit dem Anspruch einer a priori-Geltung bestimmend wäre’.52 The defendants of the separation of functions theory can easily point towards Ipsen’s confusion of function with person: the requirement that different state functions be balanced against each other remains as valid or invalid as it was, no matter how many actors participate in them or influence the process. It is often observed that separation of powers theories have been only partially implemented in continental Europe, and not at all in the United Kingdom.53 The ‘trias politica’ is often contrasted with the parliamentary systems that dominate most EU Member States. In this view, the relationship of confidence between government and

48 49 50 51 52 53

Ibid. Van der Pot and Donner, op cit. footnote 16 at 501. In this direction Fuß, ‘Zur Rechtsstaatlichkeit der Europäischen Gemeinschaften’ (1964) at 581 and, implicitly, the Wetenschappelijke Raad voor het Regeringsbeleid, De toekomst van de nationale rechtsstaat—bestuurlijke voorwaarden voor een mobiliserend beleid (2002) at 45. Maas, ‘Parlementaire democratie in de Europese Gemeenschappen’ (1965) at 896. Ipsen, Europäisches Gemeinschaftsrecht (1972) at 318-319. E.g. Lenaerts and Verhoeven, ‘Towards a legal framework for executive rule-making in the EU? The contribution of the new Comitology Decision’ (2000) at 649.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

23

parliament cannot be reconciled with the distance that separation of functions requires. Of course, the criticism simply is often not valid as far as the ‘pure’ separation of functions is concerned. As far as the Madison-style variety of separation of functions is concerned, this study is not the place to compare the various constitutional traditions in Europe. Nevertheless, an analysis of separation of functions in the EU is best made against the background of existing relevant constitutions. In the following sub-section I can do no more than discussing, in a somewhat generalised form, some of the tendencies that shaped the constitutions of European countries and the United States, and shed the light of practice over the above-mentioned objections. I will concentrate on the constitutional traditions that were the main influences of EU constitutional law, namely those of the original founder states (and then especially France, Germany and the Benelux); the United Kingdom; and the federal systems of the United States and Switzerland. The comments below54 are divided into those concerning parliamentary systems and those concerning presidential constitutions. Admittedly, this is an extremely blunt division. Academically, different criteria have been suggested for defining parliamentary systems. The topic of this study does not require a dogmatic exposé on that issue: parliamentary systems are those forms of government where the government normally depends for its existence on the confidence (or lack of censure) of parliament.55 2.3.2

Its implementation in national constitutions In presidential constitutions

There are probably as many different presidential systems as there are presidents, but it will suffice to study three extreme poles of the spectrum and devote some words to Switzerland, France56 and the United States. The Swiss might object to their constitution being discussed as a ‘presidential’ system of government since the office of president does not concentrate much power.57 However, in my view, the lack of a parliamentary system between the Conseil fédéral and the Swiss Parliament precisely puts it in the category of a presidential system. Switzerland attests to the fact that it is perfectly possible for a constitution to operate without a single-headed head of state. Its Conseil fédéral consists of seven 54 55

56 57

Sub-section 2.3.2 owes a lot to Koopmans’ Vergelijkend publiekrecht (1986). As an interesting different definition, I note the one advanced by Prélot and Boulouis (Institutions politiques et droit constitutionnel (1984) at 104-107). They maintain that parliamentary systems are characterised by three elements: the head of government is neither directly nor indirectly elected by the people but is appointed by a head of state not responsible to parliament; the (houses of) parliament must accept the government thus constituted; and thirdly, the government should equally enjoy popular support. I disagree with first and third elements of the definition, which seem to generalise for the world what once was practice in Britain. The definition assumes real power for the head of state, which may be a reality in France and Victorian Britain, but which certainly is absent in many parliamentary democracies of today. As far as the third criterion is concerned, Britain was no less a parliamentary system under the rule of Ms Thatcher who, through much of her tenure as Prime Minister, seems to have relied more on the voting system than on popular support. Technically speaking, France’s constitution is presidential with parliamentary responsibility. In my view the first seems to dominate the second element and for this reason I discuss it at this place. Kriesi, Le système politique suisse (1998) at 219.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

24

members who are not politically responsible to parliament and who combine the function of head of state (one of their number, alternating every year) with that of executive.58 This institution also shows that it is not necessary for the executive to be directly or indirectly elected by the electorate, although most presidential systems foresee some sort of public mandate that cloaks the executive with sufficient stature vis-à-vis the legislature. This being said, the Swiss Constitution is a very peculiar construct that works as a result of a very long constitutional history.59 Its precepts could not be easily transplanted to other political systems operating in dissimilar circumstances. Montesquieu was French and Rousseau Genevan, but the ideas of the latter perhaps influenced French constitutional law more profoundly than the French baron’s views did. French constitutional developments have been compared to a wheel spinning so fast that one cannot notice it moving. The simile is apt: since the times of the French Constitution, the country adopted sixteen constitutional charters60 (not counting the still-born ones), but the great features of its separation of powers system oscillated between two fixed poles. The first of these concerned a Roussovian view on democracy that tended to equate it with parliamentary representation.61 The last constitutions suffering from this were those of the Third and Fourth French Republics, both of which became infamous for their lack of direction: ‘Parliament, and especially the popularly elected Chamber of Deputies, was the government as well as the legislature. The cabinet was merely its executive arm, but had no power, its existence being at the mercy of whimsical everchanging parliamentary majorities.’62 The chaos and crumbling of both Republics led the founding fathers of the Fifth Republic to rely largely on the other extreme. Since Louis XIV’s reign, France has had considerable experience with government whereby all sovereignty was focused in one person. In the Fifth Constitution this was mixed with the notion that the head of state should be the repository of popular sovereignty. The result is a presidency well suited to the country’s conditions. It not only faces up to Parliament, but in constitutional prerogatives easily outshines it. De Gaulle’s introduction of direct presidential elections, intended to improve his ‘légitimité électorale’, was the keystone of the process.63 The French President is not responsible to Parliament. He can dismiss the Prime Minister. The other members of the government are similarly appointed and dismissed by the President, but on the Prime Minister’s nomination and with his countersignature.64 The President chairs the weekly meetings of the council of ministers and, seemingly, strongly influences its decisions. The Constitution grants him an almost free and almost unchecked hand over foreign policy.

58 59 60 61 62 63 64

Ibid. at 218-219. Scheuner, op. cit. footnote 46 at 318. Prélot and Boulouis, op. cit. footnote 55 at 295-296. Koopmans, op. cit. footnote 54 at 19. Shirer, The collapse of the Third Republic (1970) at 80. Cadart, Institutions politiques et droit constitutionnel (1990) at 921; Duhamel, ‘Les logiques cachées de la Constitution de la Cinquième République’ (1988) at 19. Kortmann, ‘The French Republic’ (2004) at 257.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

25

The 1958 Constitution paired the emboldened executive with a drastically curtailed Parliament. Use of the motion of censure was limited and, following the example of the German Constitution, now requires a cooling-off period of 48 hours between its tabling and vote.65 Moreover, at least a tenth of the MPs must sign the motion of censure and an absolute majority is required for it to pass. On top of this, no MP can sign more than three motions of censure during the same ordinary session of Parliament and more than one during any extraordinary session. In the light of these obstructions, the multiple conditions imposed on the European Parliament by Articles 201 TEC and III-340 Const66 seem almost frivolous. Compared with other parliaments, the French Parliament’s other powers as provided by the Constitution are similarly limited. Article 28 of the Constitution restricts the number of days during which it can convene to one hundred and twenty days in ordinary session. Additional session days require the approval of the Prime Minister or of the majority of the House of Parliament concerned. Extraordinary sessions can be approved by them as well, but are limited in time and function. This rule too, makes the restrictions on the European Parliament (such as the rules on its places of venue) almost look liberal. French MPs do not have the general right to adopt motions or resolutions that parliaments elsewhere, including in Strasbourg, consider their natural prerogative.67 No ‘classical’ separation of functions doctrine underpins the basic structure of the French Constitution. In the legislative field its Articles 34 and 37 provide for a system of explicit attribution of competencies to Parliament68 reminiscent of EU constitutional law. The MPs can adopt ‘lois’ on the policy domains enumerated there. All other policy fields have a ‘caractère réglementaire’, implying that the Constitution allows the executive to legislate in many policy domains by decree without Parliament’s involvement. De Gaulle and his governments felt little scruples in doing exactly that, although later practice moderated the letter of the Constitution.69 The first cohabitation between Socialist president Mitterand and a conservative government in the mid-1980s had an additional moderating effect on French constitutional practice and improved the standing of the French Parliament. As a result of this, as well as of the case law of the Conseil constitutionnel and government practice, the French Parliament has regained much of the lost territory.70 Contrast this with the American situation. In its first decade after independence, the United States experienced a situation somewhat comparable to that in the French Third and Fourth republics. The recent experience of oppressive British rule, itself a factor shaping the minds at the Convention of Philadelphia,71 had led most states to embrace constitutions that—like the Third and Fourth Republics in France—allotted a 65 66 67 68

69 70 71

Article 49 of the French Constitution, following Article 68(2) of the German Constitution. Discussed in section 5.3, below. Kortmann, op. cit. footnote 64 at 211. It is probably a chicken and egg question whether the drafters of the EEC Treaty got their inspiration from the French constitutional process, or whether De Gaulle’s men borrowed the concept of attribution of powers from the on-going drafting of the Treaty of Rome. In any event, the French have had fewer difficulties than others to accept that a constitution can encompass different legislative procedures, which may not always involve Parliament. Pezant, ‘Loi/règlement, la construction d’un nouvel équilibre’ (1988), 342-374; Classen, ‘Parlamentarismus in der V. Republik Frankreichs’ (2004) at 270-271 and Kortmann, ‘PostTchernobyl’ (1991) at 167. Kortmann, ‘Het normenstelsel in het gemeenschapsrecht’ (1996) at 117. Helms, ‘Präsident und Kongreß in der legislativen Arena. Wandlungstendenzen amerikanischer Gewaltenteilung am Ende des 20. Jahrhunderts’ (1999) at 842.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

26

preponderant role to their legislatures. Many Americans were convinced that popular representation was at best a surrogate of the ideal of free individuals. By extension, they frowned upon a not directly-elected executive and judiciary. This created a political environment in which the state legislatures occasionally were not beyond interfering in the states’ administration and even justice systems. Interstate relations were largely dealt with through the Congress of the Confederation but, as a rule, this body decided with unanimity. The ensuing constitutional lethargy strikes a familiar note with students of EU history: ‘The measures of the Union’, Hamilton lamented, ‘have not been executed; the delinquencies of the States have, step by step, matured themselves to an extreme, which has, at length, arrested all the wheels of the national government, and brought them to an awful stand’.72 The pivotal difference between the American situation before 1787 and the French crisis of 1958 was the existence of thirteen separate American sovereignties. A single American centralised government was not generally supported and the only possibility for creating an effective central executive consisted of sharing sovereignty between the states and the central entity. The drafters in Philadelphia had the genius of transplanting Montesquieu’s separation of powers theory, framed against the background of an analysis of social classes in English politics, to the situation in the United States. The doctrine was altered in the process, with stress laid on the checks and balances element. The Founding Fathers went at lengths precisely to avoid anything resembling absolutist royalty73 and this led to an executive with less formidable prerogatives than those of the Fifth Republic’s president. This different conceptualisation starts with the President’s appointment. The drafters of the Constitution explicitly refrained from proposing the direct election of the President, but instead preferred a system whereby electors would be chosen who—in Hamilton’s words—would be ‘much less apt to convulse the community with any extraordinary or violent movements’ than a directly-elected President: ‘every practicable obstacle should be opposed to cabal, intrigue, and corruption’.74 ‘And chadding voting machines’, he should have added: the Founding Fathers of the United States failed to foresee that the resources needed for presidential campaigns would polarise the political spectrum. This, in turn, brought about the enormous power that currently resides in the two main political parties. ‘The simple, obvious fact about the office of President’, notes Vile, ‘is that only one person can fill it. The Senate or the House of Representatives could dissolve into a multitude of factional groups, but only one person can occupy the President’s chair in the White House’.75 This, coupled with the now substantial powers of the president, make the presidential election the most important one that can be fought. The extent to which parties are willing to go in such elections was shown in the presidential contest of 2000 between Mr Gore and Mr Bush. As a result, unlike European politics, which are characterised by a multitude of political parties, two (albeit rather loose) political parties dominate the American political landscape. Bipartism is a direct consequence of the Madison-style separation of functions. The rise of bipartism in turn gave a whole new meaning to the letter of the Constitution, which foresees in its Article II(1) the possibility that the House of Representatives would choose the President from the five candidates with the highest number of votes. As a result of the bipartisan system, the political system in the last 72 73 74 75

Hamilton in ‘Federalist Paper No 15’, op. cit. footnote 20 at 151. Ibid. in ‘Federalist Paper’ 67 and 68 at 389-396. Ibid., at 393. Vile, op. cit. footnote 38 at 44.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

27

century has only once, in 2000, come close to a situation where Congress might have determined the outcome of the elections. The normal situation is that the House of Representatives is presented with a clear winner. The consequence of this development for the separation of powers system has been that the electors normally carry little political weight. In the public perception, elections revolve around the presidential candidates rather than around them. As a result, the presidency has gained some of the direct democratic legitimacy that makes its French counterpart so constitutionally strong through a constitutional practice that renders the letter of the American Constitution with a colouring quite different than that envisaged by its drafters. As far as the execution of the institutions’ powers is concerned, apart from some important exceptions, the Constitution maps the state functions to the key triad of institutions and provides for a separation of persons. In other words, not only does the American Constitution entrust legislation, administration and adjudication to different institutions, with the exception of the Vice President (who chairs the Senate), it also disallows any person to be a member of any two of these institutions. The executive power is exercised by the presidency, the legislative by Congress and the judiciary is headed by the Supreme Court. However, none of these institutions can work in isolation, and checks and balances between them ensure that the powers are mutually dependent.76 In this manner, institutions representing different elements of sovereignty are forced to act together. The American legislature lacks the wide-ranging powers of the British Parliament. Its powers are circumscribed by the Constitution. It consists of the two Houses of Congress, each of which channels a different legitimacy claim: on the one hand, although they are elected on a state-by-state basis, the Representatives represent the American people. On the other hand, in its original set-up, each state legislature appointed two senators to represent them. This rule embodies the equality of the different states. Since the adoption of the Seventeenth Amendment in the early 20th Century, the senators are directly elected and this further enhanced their democratic legitimacy. Historically, the system for appointing the Senate must be seen against the views on popular democracy popular at the time of the adoption of the Constitution. In many states, the legislature was seen as the true representative of the people and endowed with the legitimacy that the judiciary or government (if they were at all independent from the legislature) lacked. The separation between Congress and the executive branch is not absolute. In fact, their relationship has been termed as ‘separated institutions sharing power’.77 Thus, although the Constitution places the ‘legislative power’ firmly in the hands of Congress, the executive substantially influences the legislative process. To begin with, currently it drafts some two-thirds of all laws passed by Congress.78 Moreover, the president possesses the right of veto which, in the opinion of most commentators, he can exercise at will.79 This veto can in turn be overturned with a two-thirds majority in Congress,80 but that may not be easy in a bipolar Congress. Moreover, in practice, instruments such as the ‘pocket veto’81 developed, which enhance the 76 77 78 79 80 81

Koopmans, op. cit. footnote 54 at 151. Helms, op. cit. footnote 71 at 842. Marc-Lipiansky, op. cit. footnote 5 at 15. Helms, op. cit. footnote 71 at 847. Article I:7 of the US Constitution. Article I:7 of the Constitution provides inter alia that ‘[i]f any Bill shall not be returned by the President within ten Days (Sundays excepted) after it shall have been presented to him, the Same

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

28

presidential influence over the legislative process. 82 Lastly, as in all developed countries, the executive passes much implementing legislation. In the field of external relations, the Senate must approve international ‘treaties’ with a two-thirds majority.83 However, this does not stop Congress from permitting presidents to conclude ‘executive agreements’, which can bind the United States internationally. Since American case law accepts that Acts of Congress override ‘treaties’ in domestic law,84 a fortiori executive agreements are subject to Congress’ legislative powers85 and can be used to evade the two-thirds requirement in the Senate.86 In comparison, the European Commission’s attempts to have a similar power to conclude ‘international administrative agreements’ recognised foundered on the Court of Justice’ absolute disapproval.87 In the area of appointments of ministers and other members of the executive, the first obvious difference between France and the United States is the lack of a Prime Minister-type of person in the latter. In France, the office is necessary since the ministers are responsible to parliament, implying that the government must be an entity separate from the head of state. In the United States, where no ministerial responsibility exists, a Prime Minister or separately organised government would not have fitted the logic of the Constitution very well. The Cabinet meetings—which are chaired by the President—lack the formal status they have in France. Since the government does not rely on a majority in Congress, its members are more the President’s trustees than their French colleagues are. The US Constitution requires that high civil servants and judges are nominated by the President and appointed ‘with the Advice and Consent of the Senate’.88 Barring the heavy impeachment procedure, the American Congress has no formal power to censure any member of the government—let alone the President (and, in return, he cannot dissolve Congress). It is questionable, however, whether a Secretary could function in the face of sustained hostility from the majority of one of the Houses. The Congressmen could easily block legislative initiatives from or the budget for any member of the government they wish to see deposed. In this sense, Capitol Hill indubitably has more real powers towards the executive than the European Parliament has and, in the course of the last two centuries, refined interinstitutional negotiation mechanisms lubricating mutual contacts have developed.89

82 83 84 85

86 87 88 89

shall be a Law, in like Manner as if he had signed it, unless the Congress by their Adjournment prevent its Return, in which Case it shall not be a Law‘. From this rule an absolute veto right developed concurrently to the regular right of veto. For the history of this so-called ‘pocket veto’ see Fisher, The Pocket Veto: its current status (2001). A third type, the so-called ‘line item veto’, was declared unconstitutional by the Supreme Court: Nicola, ‘Line Item Veto Act unconstitutional: Clinton v. City of New York’ (1998). Line item vetoes targeted parts of legislation, rather than the whole text. Article II:2 of the Constitution. US Supreme Court, Head Money Cases (Edye vs Robertson), 112 (1884) at 599. Validity of Congressional-Executive agreements that substantially modify the United States’ obligations under an existing treaty—Memorandum for Alan J. Kreczko, Special Assistant to the President and Legal Adviser to the National Security Council of 25 November 1996, . Yoo, ‘Laws as treaties?: the constitutionality of Congressional-Executive agreements’ (2001) at 758. Case C-327/91, France vs Commission, [1994] ECR I-3641. Admittedly, there was one big difference: in the United States the authority to conclude executive agreements is granted by the legislature. The Commission was arguing the existence of an autonomous power. Article II:2, clause 2 of the US Constitution. Helms, op. cit. footnote 71 at 847.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

29

To the extent that the European Union can be equated to a republic, it combines elements of all the above political systems. Like Switzerland, it has no clear singleperson executive. From the United States it inherited the notion of circumscribed powers for the institutions although—as will be shown in more detail below—its constitution is hardly comparable to its American counter part. From the French experience it inherited the limitations on the powers of Parliament and obstacles to sacking the executive. On the previous pages I juxtaposed the French and American constitutions as far as the relationship between legislature and executive is concerned. The latter constitution is based on a separation of functions model, whereas the former is not, or only very partially. In either case, political developments changed the spirit of the Constitution. In France, the more moderate approach of De Gaulle’s successors, as well as the appearance of cohabitation situations, benefited the standing of the French Parliament. Conversely, the development of the bi-party system in the United States ultimately increased the democratic legitimacy of the president, which upgraded his office to an importance not foreseen by the authors of the Constitution. The lesson to be learnt is that constitutional conventions and practice can radically affect the separation of powers laid down in the Constitution. In parliamentary systems: relationship between the executive and parliament Dicey noted in the 19th Century that ‘It were curious to follow out the historical growth of the whole theory as to the “separation of powers”. It rests apparently upon Montesquieu’s Esprit des Lois, Book XI. C. 6, and is in some sort the offspring of a double misconception; Montesquieu misunderstood on this point the principles and practice of the English constitution, and his doctrine was in turn, if not misunderstood, exaggerated, and misapplied by the French statesmen of the Revolution.’90 There may be some truth in this, but to do justice to the theory, the learned British jurist perhaps also himself misunderstood it.91 Nevertheless, the British constitution, although it was the example of parliamentary democracy that guided much of Europe, differs from Continental parliamentary democracies in many respects, especially in its organisation of separation of powers. Even a superficial glance at the British constitution suggests that it challenges notions of separation of powers by its emphasis on the sovereignty of Parliament. A first point to make is that ‘Parliament’ in this context is shorthand for the sovereign, government, Lords and Commons (collectively referred to as the ‘Queen in Parliament’).92 From the 17th Century onwards, the English courts accepted that the sovereign too was bound by the common law.93 From this principle—combined with the 90 91 92 93

Dicey, Introduction to the study of the law of the Constitution (1982) at 220. Vile, op. cit. footnote 1 at 23-113 is interesting and detailed on the influence of the English constitution on separation of powers thinking. Limon a.o. (eds), Erskine May’s treatise on the law, privileges, proceedings and usage of Parliament (1997) at 9. Sári, op. cit. footnote 1 at 11.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

30

recognition of the sovereignty of Parliament—flowed the important rule that all government must rest on law. In many Continental Member States a comparable process occurred when it was realised that (in the words of an American author), ‘the greatest injustice of all is done when the umpire forgets that he too is bound by the rules’.94 In Continental Europe, this 19th Century notion became the principle of gesetzmäßiger Verwaltung, which is a cornerstone of the Etat de droit.95 That principle meant that all administration must be based on statute law or on the Constitution.96 Since on the Continent law was ultimately derived from the Constitution or from statutes, gesetzmäßiger Verwaltung caused parliaments to extend their reach to all fields of policy, as they became competent to adopt statutes in all fields of law, forcing the administration to bow to the legislature.97 In the United Kingdom the process evolved somewhat differently. Dicey construed common law as ‘subordinate legislation, carried on with the assent and subject to the supervision of Parliament’.98 This ‘assent’ is, of course, silent and more a theoretical construct than reality: one modern author somewhat irreverently speaks of the ‘nakedness of judicial law-making’.99 The ‘supervision’ consists in the possibility for the legislator to override the judges through statute law. Nevertheless, the influence of common law is so persuasive that the judges are certainly more than merely ‘la bouche de la loi’. Of course, it would take a certain Montesquivian fundamentalism to claim that this job description fits Continental judges. Note that Montesquieu did not consider the judiciary a ‘puissance’ of the same stature as the other two ‘puissances’—it was only upgraded under the influence of Blackstone.100 The point is that—in this respect—the UK and Continental parliamentary systems differ in shades rather than in colours. The UK has no written constitution, that is to say, it possesses no constitutional charter in the Continental sense. This fact is closely related to the sovereignty of Parliament doctrine, which was first expounded as an analysis of the constitution but gradually obtained a normative force.101 As early as Edward VI’s reign, the King lost the power to issue ‘proclamations’ and ‘ordinances’, leaving only the ‘Queen in Parliament’ as the legislative power.102 Since a very long time it is accepted that the Commons, if push comes to shove, wins in power struggles with the Lords.103 Parliament can vote out the government. It can change the law without effective opposition of the monarch. It can amend the common law. Ultimately, it can change 94 95

96 97 98 99 100 101 102 103

Morley, Freedom and Federalism (1981) at 13. Stern, op. cit. footnote 15 at 772, Bos, Demokratie en rechtsstaat (1987) at 49 and Van der Pot and Donner, op. cit. footnote 16 at 494. ‘Statute’ does not automatically include ‘Constitution’: according to Voermans, ‘Artikel 44 Grondwet als bevoegdheidsbasis voor ministeriële regelingen’ (2003) at 179, Dutch law goes even further by not allowing ministerial orders directly based on the Grondwet. By way of typical example this was decided in the Netherlands the late 19th Century: Hoge Raad (Supreme Court of the Netherlands), Meerenberg, Weekblad van het Recht (1879) 4330. Of course, there are differences between countries, depending on their constitutional development. The exceptional provisions of Articles 34 and 37 of the French Constitution were already noted in sub-section 2.3.2, above. Dicey, op. cit. footnote 90 at 18. Hartley, The foundations of European Community law (2003) at 133. Vile, op. cit. footnote 1 at 114. Koopmans, op. cit. footnote 54 at 16. Dicey, op. cit. footnote 90 at 11. Ibid. at 280; Hood Phillips and Jackson, op. cit. footnote 43 at 126. Together with the convention that the Queen must assent to any bill passed by the Houses of Parliament (the latter ibid. at 124), this ultimately gives the Commons the final word on the constitution.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

31

the constitution. In the classic formula (which medical technology has overtaken), it ‘can do everything but make a woman a man, and a man a woman’.104 Westminster is thus certainly not characterised by a ‘pure’ separation of functions theory.105 A fortiori, there is no congruence of three state functions with three institutions. This is not to say that Parliament is unchecked. Madison noted that the strength of Britain’s constitution does not lie as much in the separation of the governmental functions as in their mutual checks: the legislative, executive, and judiciary departments are not totally separate and distinct from each other.106 These checks and balances are found within Parliament (in the broad sense of the word). There are several noteworthy factors here. Firstly, the mere size of the Commons (or, for that matter, most parliaments in Europe) allows it to legislate, but renders it ill equipped to govern the country. The British electoral system normally ensures that one party will have an absolute majority in the Commons and mechanisms such as the ‘whip’ system are intended to enforce party unity. Nevertheless, partially as the effect of the district system under which it is elected, the Commons is not a unity. Secondly, although in form the monarch wields the chief executive power, in substance—more thoroughly than in many other European monarchies—this devolved onto the Cabinet (or better, the ministers) who are responsible to the Commons. In Britain, unlike some Continental parliamentary democracies, the ministers must be members of one of the Houses of Parliament.107 Incompatibilities exist for, inter alia, holders of judicial offices and peers (since they are already member of the Lords) to be elected to the Commons, although the House of Lords fulfils important judicial tasks.108 Thirdly, the government, led by the Prime Minister, has in the course of time acquired tools allowing it to become a power not wholly under control of the two Houses of Parliament. Ultimately, however, the British Prime Minister’s power is not unlimited. He cannot alienate his backbenchers without risking a revolt. These internal balances within the Commons have in practice been offset by the development of the party system. The Commons counts more than two parties, but the first-past-the post voting system heavily favours the Conservatives and Labour. These parties are centrally run. The combination of both factors leads to a situation whereby the party in power usually commands a large majority in the Commons as well as government. In practice ‘Parlement et gouvernement ressemblent à deux machines animées par un même moteur’.109 More pessimistically, Dicey considered party government ‘the very foundation of our constitutional system’.110 In this fashion, the 104 105 106 107

Dicey, op. cit. footnote 90 at 5. See sub-section 2.3.1, above. Madison in ‘Federalist Paper 47’, op. cit. footnote 20 at 303. The only restriction here appears to be that not more than 95 holders of ministerial offices are entitled to sit and vote in the Commons at any time. Some further limitations on the size of the government apply: Limon, op. cit. footnote 92 at 50. A separation of persons between the Commons and office holders was almost included into the British constitution when the Act of Settlement was contemplated: Vile, op. cit. footnote 1 at 81. 108 Limon, op. cit. footnote 92 at 41 and 60-63. There is no need for British law to require that MEPs cannot be government ministers since this already flows from Article 6(1) of the Act concerning the election of the representatives of the Assembly by direct universal suffrage, OJ (1976) L278/5. Membership of both the European Parliament and the Commons is possible but, depending on the party, not always encouraged: Crespo Allen a.o., op. cit. footnote 18 at 33. 109 Duverger, Les partis politiques (1981) at 516-517. 110 Dicey, op. cit. footnote 90 at cxviii.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

32

executive is part of the legislature and, even if Madison’s interpretation of the English Constitution111 were correct in his time, it certainly no longer is today.112 On the Continent, most countries have a tradition of written constitutions. As a direct consequence, the national parliament was subject to a supreme legal instrument and the government institutionally separated from it. This precluded any sovereignty of parliament doctrine. At most, some constitutions give parliament pride of place amongst the institutions. For example, the German Constitution follows the Montesquivian triangle imperfectly and does not seem to separate ‘functions’ clearly.113 In parliamentary systems: positive or negative operation of the confidence rule; collective and individual responsibility Parliamentary systems are characterised by a confidence relationship between the executive and parliament. The former depends on the latter for its existence and for its demise. Detractors of the view that a parliamentary democracy can be a separation of functions system point mostly to this confidence relationship.114 In this logic, dependence in office implies dependence in action. This, however, firstly presumes the Madison-style interpretation of separation of functions rather than its pure form. Secondly, it misconstrues the nature of the parliamentary system. It is interesting to look at the three ways in which parliamentary responsibility can operate; at its connection with the right to dissolution; and at collective versus individual responsibility. The three ways in which parliamentary responsibility can operate. In generalised terms, parliamentary responsibility can operate in three modes. To begin with, confidence can be expressed positively through a motion of confidence—‘vote of approval’ in the terms of Article 214(2) TEC—providing democratic legitimacy115 to a new executive. Such a procedure exists for example in Germany, where the Federal Chancellor is elected by the Bundestag, and in the EU, but not in, for example, the Netherlands. Secondly, confidence can be presumed until it is expressly withdrawn. For example, in the Netherlands ministers are presumed to enjoy the confidence of Parliament unless a lack of confidence becomes apparent. The two systems can co-exist, as prove the constitutions of France and the EU.116 In the system of the German Constitution, the Bundestag can only express its lack of confidence in the Federal Chancellor, but not in the ministers. Since the ministers in his government depend on his appointment, his dismissal implies the censure of the whole government. A further German peculiarity is the requirement that the motion of censure be constructive: the Bundestag must in its motion of censure suggest a 111 Madison in ‘Federalist Paper 51’ in op. cit. footnote 20 at 303. 112 The realisation that political mechanisms no longer suffice to balance the British constitution have in recent years led to constitutional changes to introduce more checks and balances. On these Irvine, ‘Sovereignty in comparative perspective: constitutionalism in Britain and America’ (2001) at 15 et seq. 113 Scheuner, op. cit. footnote 46 at 308-309. 114 In this direction Marc-Lipiansky, op. cit. footnote 5 at 13. Majone, ‘Delegation of regulatory powers in a mixed polity’ (2002) at 324 seems to consider the two as opposing poles of a spectrum. 115 The extent to which the Commission rests on such democratic legitimation is discussed in Chapter 4, below. 116 Articles 49 and 50 of the French Constitution; Articles 214 TEC, I-27.2 Const (appointment) and 201 TEC and I-26.8, III-340 Const (motion of censure), respectively.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

33

successor. Since the president has no margin of decision in this matter, effectively the Bundestag appoints the new Federal Chancellor.117 The third way in which the confidence issue can be posed is through systems whereby the government attaches its fate to a certain matter under debate. Again, there are differences between constitutional systems on the modalities: in the Netherlands for example, there are few formalities or rules for ‘posing the confidence question’. In Germany on the other hand, the Constitution regulates the matter. The Federal Chancellor can request the Bundestag for an express vote of confidence. If this is refused, he can propose to the President to dissolve the Bundestag within 21 days. However, within that period the Bundestag can elect another Federal Chancellor, after which the right to dissolution dissipates.118 In France, the Constitution foresees the possibility for the Prime Minister to engage ‘la responsabilité du Gouvernement devant l’Assemblée Nationale sur le vote d’un texte’.119 Under this procedure, a ‘texte’ is considered adopted unless a motion of censure is adopted within the following 24 hours, during which the debate in Parliament is suspended. Since the adoption of such motions has been made very difficult,120 this is a powerful weapon in the hands of the government. Its connection with the right to dissolution. The parliament’s right to censure is normally, but not always, coupled with the government’s right to dissolve parliament (although this normally entails the dissolution of the government as well). Some authors consider this linkage so natural, that they consider the right to dissolution as a defining characteristic of a parliamentary system. There are, however, exceptions to the rule and I am not convinced it is an intrinsic necessity for the existence of a parliamentary system. For example, in France (admittedly not a country with a traditional parliamentary system entailing a politically fully responsible executive), the right to dissolve parliament is not a prerogative of the government but of the President.121 He, however, is legally—and in times of cohabitation factually—a separate political institution with a separate political will. In Britain the right is largely the Prime Minister’s, and in the past, at several occasions, he failed to consult his colleagues before using it.122 Moreover, I believe the essence of the parliamentary system to lie in the fact that the government must enjoy the confidence of (at least one house of) parliament. The government’s right to dissolution, where it exists, is merely a check against the abuse of that principle by the MPs.123 If a parliament had an unchecked right to censure the government, it would be apt to use this in the manner characterising the pre-1958 French republics. In other countries, such as the Netherlands and Belgium, this risk of abuse has diminished somewhat when strong party organisations began to dominate politics, but it never faded totally. It is, however, perfectly possible to counterbalance the motion of censure in other ways. Notably, it can be checked by making its use more difficult. I already noted the

117 Article 67(1) of the German Constitution. 118 Article 68(1) of the German Constitution. 119 Article 49 of the Constitution. The same provision requires him to first discuss this in the Council of Ministers. 120 Although the restriction that only one motion of censure can be tabled per session does not apply, since the initiative is taken by the Prime Minister: Prélot and Boulouis, op. cit. footnote 55 at 848. 121 Article 12 of the French Constitution. 122 Formally, of course, the sovereign dissolves the Commons—but in practice the decision is largely taken by the Prime Minister: Hood Phillips and Jackson, op. cit. footnote 43 at 301. 123 In this sense Vis, ‘Demissionair en quasi-demissionair—een pleidooi voor vereenvoudiging’ (2003) at 13.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

34

conditions to which the motion of censure is subject in French constitutional law. Restrictions exist in Germany and the EU as well.124 Collective versus individual responsibility. In some constitutional systems, such as the United Kingdom,125 France and Germany, the government’s responsibility towards parliament is collective. There are several aspects to this. Firstly, collective responsibility implies that the ministers stand by the policies of their government even if privately they disagree with them, or resign. Moreover, at least in Germany126 (but not in Britain)127 the resignation of the Prime Minister automatically entails the demise of his government. The European Commission is collectively responsible to the European Parliament. In other countries, such as the Netherlands, responsibility is individual. There is a Council of Ministers in The Hague, but the main function of this body is to prevent and resolve conflicts between different ministers.128 As a further consequence of the lack of collective responsibility, the Dutch Prime Minister has neither the advantage of being leader of the sole party in power (like his British counterpart) nor the German Federal Chancellor’s Richtlinienkompetenz.129 This lack of collective responsibility is in practice largely compensated for by the fact that governments in the Netherlands are always composed of coalitions of parties. Unlike the British situation, the often-difficult formation of a Cabinet does not serve to create a homogenous government, but rather to ensure a majority in the House of Representatives (Second Chamber).130 As a further consequence, the sacking of one minister, which normally requires at least one government party to side with the opposition, will not often go unpunished and, in the past, tit-for-tat sackings131 or the total collapse of the government both occurred.132 Nevertheless, in practice the government is normally not delivered to the whims of the Second Chamber and the call for more ‘dualism’ (meaning a more independent stance by the government parties from the government) in Dutch politics suggests that a relationship of confidence certainly need not imply dependence. These brief comments show that there are different modalities for a parliamentary system to exist. Marc-Lipiansky,133 reasoning from a French perspective, draws the somewhat hasty conclusion that ‘[l]a séparation des pouvoirs revêt un caractère fictif dans les régimes neo-parlementaires’. This, however, is not quite correct. The 124 The motion of censure in EU practice is discussed in more detail in section 5.3, below. 125 In the United Kingdom ministerial responsibility is both individual and collective: Hood Phillips and Jackson, op. cit. footnote 43 at 309. 126 Article 69(2) of the German Constitution. 127 Although a successful motion of censure against the Prime Minister by convention brings down the whole government: Hood Phillips and Jackson, op. cit. footnote 43 at 125. 128 Van der Vlies, ‘De bevoegdheid van de ministerraad’ (1982) at 422. 129 Van der Pot and Donner, op cit. footnote 16 at 390. 130 Kraan, ‘Homogeniteit van het kabinet als staatsrechtelijk probleem’ (1977) at 1083. 131 For example, in 1988 Minister Van Eekelen (a Liberal) resigned over an obscure crisis known as ‘the passport affair’. The Liberal party subsequently retaliated by renouncing its confidence in a junior minister from the other government party. 132 Artikel 12(2) of the Rules of Procedure of the Dutch Council of Ministers provide that under no circumstance a minister or staatssecretaris shall act against a decision of that Council. This measure instills an element of coherence and common responsibility into it, although the coalition agreement concluded between government parties at the time of the formation of the government is probably as important. 133 Op. cit. footnote 5 at 13. She notes that such a system requires ‘bipartisme’ or at least ‘deux coalitions stables’. The Dutch system has been pretty stable throughout most of its history whilst lacking either.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

35

Madison-style separation of functions, which calls for a correspondence between three state functions and three state institutions, may not have been fully implemented in Continental parliamentary systems because ‘legislation’ is normally a task in which both government and parliament participate. However, the fact that the government relies on the confidence of parliament does not, in itself, make separation of functions irrelevant for parliamentary systems.134 In parliamentary systems: parliament’s involvement in execution Parliaments in parliamentary democracies exercise the right of legislation in cooperation with the government and the head of state.135 Can such a legislature adopt measures that by their nature are of an executive kind rather than legislation? In the framework of this study it is impossible to review all parliamentary constitutions of Europe, but the examples of Britain and the Netherlands may convey the outlines of the answer. As far as Britain is concerned, the answer is simple, at least in theory. Under the sovereignty of Parliament doctrine, the Queen in Parliament can expressly interfere in individual cases that in separation of functions terms would normally be termed execution or adjudication. Parliament ‘may adjudge an infant, or minor, of full age. To attaint a man of treason after his death. To naturalise a mere alien, and make him a subject born. It may bastard a child that by law is legitimate, viz. begotten by an adulterer, the husband being within the four seas.’136 Although the British constitution thus clearly is not based on a separation of functions (of either variety), in practice the brunt of the execution of legislation as well as the formulation of policy is left to the government. On the Continent, where the institutions are placed in a written constitutional straitjacket, the situation is not always clear. For example, the Dutch constitution does not explicitly prohibit Einzelfalljustiz through statute law. Nevertheless, indications can be found supporting a separation of functions analysis. The judges are instructed by law not to judge the ‘intrinsic value or reasonableness of laws’. 137 Moreover, judges are instructed not to adopt general regulations in cases subject to their adjudication.138 The restriction goes even further: Article 120 of the Dutch Constitution provides that ‘[t]he judge shall not judge the constitutionality of statutes and treaties’. This is a peculiar formula sanctioned more by time than by logic: the prohibition to declare statutes unconstitutional seems to have flowed from a fear that the courts would override the better judgment of parliament.139 However, international agreements ratified by the Netherlands apply directly and can set conflicting statutes

134 For example, in 2002 Dutch Justice Minister Korthals defended constitutional reform in the new judicial organisation firmly on that basis: ‘Het publieke vertrouwen in de rechtsstaat’ (2002) at 789790. 135 In some countries the head of state is not member of the government. 136 Coke, quoted in Dicey, op. cit. footnote 90 at 9. 137 Article 11 of the Wet van 15 mei 1829, houdende algemeene bepalingen der wetgeving van het Koningrijk. 138 Ibid., Article 12. 139 Kortmann, De Grondwetsherziening 1983 (1983) at 30.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

36

aside.140 Following these pointers towards a separation of functions system, some have argued that in the Netherlands the legislature (which comprises the two Houses of parliament and the government) is not empowered to legislate for individual cases, situations or objects. However, this is contested.141 In any case, in practice the Dutch legislature is not prone to legislate for individual cases. Generally, Continental constitutions do not formally bar legislation for individual cases. However, in practice parliaments tend to be reluctant to go down that road and, in this respect, separation of functions does influence constitutional practice. Separation of functions in the Member States and the United States This brief and, of course, not more than sketchy review of some constitutional elements in Member States and the United States clearly shows that none of the Member State constitutions corresponds to a Madison-style congruence of three functions with three institutional actors. Moreover, even in the United States, where the constitution very clearly is inspired by a pure separation of functions, the Madison-style variety of the theory is more complex than the authors of the Constitution foresaw. Evidently, it does not suffice for a constitution to be based on a separation of powers mechanism. Other principles (rule of law, democratic legitimisation of decision making, etc.) fulfil their role in shaping what in America is called a liberal democracy and on the Continent the Etat de droit. But conversely, few people would deny the epithet of ‘democratic’ to the constitutions of Western Europe merely because there is no strict separation of functions. Armed with this light framework of reference, an investigation of the separation of functions in the EU’s constitution142 can now be attempted. 140 A matter more of theory than of law: the courts are extremely reluctant to declare statutes void on this basis: Van der Pot, Donner a.o., op. cit. footnote 16 at 517. 141 Van der Pot and Donner, Handboek van het Nederlandse staatsrecht (1983) at 444 did not believe it. 142 The question could be (and has been) raised to what extent the current EU has a constitution in the traditional sense. A distinction must be made between ‘Constitution’ in the sense of constitutional charter and ‘constitution’ in the sense of a set of rules that are ultimately the supreme rules in the legal order forming the ultimately legal basis for public action. For example, Britain has the latter (substantive) constitution but lacks the former. The question whether the EU has a ‘constitution’ lies outside the scope of this work and the following comments will suffice. For an interesting overview of political theories on the matter see Dobson, ‘Citizenship, political authority and constitutionalism in the European Union: a normative theoretical approach’ (2001) at 337 et seq., Pache, ‘Eine Verfassung für Europa—Krönung oder Kollaps der europäischen Integration?’ (2002) at 770 et seq., Pernice, ‘Multilevel constitutionalism and the Treaty of Amsterdam: European constitution-making revisited?’ (1999) at 711 et seq., and Verhoeven, The European Union in search of a democratic and constitutional theory (2002), 75 et seq. Some authors simply accept the Treaties as the ‘Constitution’: Curtin, ‘The constitutional structure of the Union: a Europe of bits and pieces’ (1993) at 21, Ehlermann, ‘Sachkompetenzen und Organverfassung der Gemeinschaft’ (1971) at 164, Ipsen, ‘Zum Parlaments-Entwurf einer Europäischen Union’ (1985) at 325, Oeter, ‘Die Genialität der Verträge’ (2001), Pernice, ‘Multilevel constitutionalism in the European Union’ (2002) at 514 and Petersmann, ‘Proposals for a new constitution for the European Union: building-blocks for a constitutional theory and constitutional law’ (1995) at 1126. For a further overview of the literature in this sense see Haack, ‘Der Begriff der Verfassung’ (2004), 785 et seq. and Voermans, ‘Europese Grondwet in de stellingen’ (2005) at 197. I respectfully disagree with Piris, ‘Does the European Union have a constitution? Does it need one?’ (1999) and with Cromme, ‘Verfassungsvertrag der Europäischen Union’ (2002) at 594, who argue that the Treaties cannot serve as a constitution. The former argues that the Treaties are not comparable to the Constitutional charters (with a capital C’) of states, since it lacks some of the attributes of such fundamental laws. The latter believes that a true Constitution presumes statehood.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

2.3.3

37

Separation of functions in the European Union

There are different views on the appropriateness of analysing the EU in terms of separation of functions. One of the reasons for that is that different authors have a different understanding of what separation of functions means. I will first examine the strengths of the argument in favour. Following that, the position of the opposite school of thought will be explored. An argument in favour . . . To what extent can one analyse the EU in terms of separation of functions? To start with its ‘pure’ variety, an argument can be made that the ‘trias politica’ is one of the principles underlying the Union’s institutional set-up.143 ‘It can hardly be doubted’, concluded Lenaerts in 1991, ‘that the principle of separation of powers, as defined in this functional perspective, is a structural feature underlying the Community legal order’.144 Indeed, both the EC Treaty and the Constitution separate ‘adjudication’ from ‘legislation’ and ‘execution’. As far as the separation between the latter two functions is concerned, the distinction is more subtle in the EC Treaty145 and perhaps not always as clear where I would respectfully submit that the discussion is semantic in that it depends on the criterion one uses to define a ‘Constitution’. I incline towards Barents’ emphasis on the internal autonomy of EC law as the criterion for determining the nature of the Treaties: De communautaire rechtsorde (2000) at 357-358; similarly Pescatore, L’ordre juridique des Communautés européennes (1973) at 56-57. The US Constitution is a prime example of a constitutional charter that started its life as an international treaty. Arguably, the EU got a substantive constitution on the day the Court of Justice pronounced its judgment in case 26/62, Van Gend en Loos, [1963] ECR 3. Whatever the discussion, the Court of Justice simply referred to the EEC Treaty as the Community’s ‘basic constitutional charter’ without further ado: case 294/83, Parti écologiste ‘Les Verts’ vs Parliament, [1986] ECR 1339, § 23. The subject merits two more observations. Firstly, Lenaerts and Gerard, ‘The structure of the Union according to the Constitution for Europe: the emperor is getting dressed’ (2004) at 293 et seq. make the point that a ‘constitution’ does not presume a state. Perhaps the slight nuance could be made here that different languages cause different connotations: in English, both the ILO and a soccer club can have a ‘constitution’. In Dutch, neither can have a ‘Grondwet’, which is a term reserved for the states (and for the EU’s Constitution). Secondly, of course the term ‘constitution’ has a political connotation as well: Bernhardt, ‘Les sources du droit communautaire: la “Constitution” de la Communauté’ (1983) at 80. 143 Lenaerts, ‘A new institutional equilibrium? In search of the “trias politica” in the European Community’ (1992), 141 et seq. See also Crossland, op. cit. footnote 1, especially at 491 et seq., Lenaerts and Van Nuffel (Bray, ed.), Constitutional law of the European Union (2005) at 563-564, Schütz, ‘Die Legitimation der Europäischen Zentralbank zur Rechtsetzung’ (2001) at 299-300 and, implicitly, Capotorti, ‘La structure institutionnelle de l’Union Européenne’ (1985) at 512. Louis, ‘Indépendance des banques centrales, séparation des pouvoirs et démocratie’ (1999) at 473 et seq. is caught in Montesquivian categories but fails to address the place of the ECB in the separation of powers structure of the EU. 144 Op. cit. footnote 36 at 14. 145 Critical (but without the qualifying ‘functional’): the European Commission in its ‘Report on European Union’, Bull. EC Suppl. 75/5, § 91. Advocate General Jacobs argued in case C-316/91, Parliament vs Council (Lomé Convention) (Opinion), [1994] ECR I-625, § 85 that ‘[i]t will be easier to accept that a Community institution may undertake an executive function on the mandate of the Member States . . . than to accept that a Community institution, especially the Council, may undertake a legislative function on the mandate of the Member States’. This is a curious argument: it presumes a hard and fast rule for determining between legislation and execution, and precisely this is lacking in the EU’s institutional structure.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

38

implementing legislation is concerned. Nevertheless, the structure of the EC Treaty forces a separation between ‘legislation’ (a function performed by the legislator, which varies in function of the Treaty provision) and ‘execution’ where the Member States and the Commission share the primary responsibility, but where the Commission’s hands are bound by comitology.146 As far as the EU Treaty is concerned, most if not all of the measures foreseen under the second and third pillar fall within the province of ‘execution’ rather than of ‘legislation’. As is explored in more detail in Chapter 13 below, the Constitution would radically revise this system and is definitely based on a functional separation of ‘legislation’ and ‘execution’. The question is more difficult where the Madison-style variety of separation of functions is concerned.147 Whilst the American Constitution comprises comparatively clear-cut rules assigning the executive and legislative functions to specific and separate institutions, the EU system provides few corresponding guidelines.148 However, it is possible to attempt an analysis on these lines. To start with the easy part, the ‘adjudication’ of conflicts of law between two or more parties is entrusted to the Community Courts. The EU’s judges monopolise, or at least have the final say, on the adjudication of disputes. The Courts’ attitude to precedents is similarly in line with separation of powers-inspired thinking. They do not recognise any ‘binding precedent’ or stare decisis, 149 which fits in with the conception that their task is to rule on individual disputes rather than to set general rules. Not only is the adjudication of disputes (apart from the above-mentioned exceptions in competition and trade law) separate from the other ‘functions’, there is even an organic monopoly on it. An argument could be attempted that the Courts’ attitude150 towards the individuals’ ability to challenge instruments of general application should be interpreted in terms of separation of functions. The logic behind the ‘individual concern’ requirement in Article 230 TEC would then arguably be to distinguish norms of general application (substantive laws) from (disguised) decisions in individual cases (‘execution’). However, such an interpretation of the Article 230 case law is not convincing. The Courts have never reasoned this matter in separation of powers terms. 146 Contra Curtin, op. cit. footnote 142 at 35, Jacqué, Droit institutionnel communautaire (1997) at 50 and Petersmann, op. cit. footnote 142 at 1135. Bradley, writing in 1987, argued that the ‘legislative function is split into the legislative power—exercised by the Council of Ministers, or by delegation, by the Commission—and the deliberative function, which is vested in the European Parliament’: ‘Maintaining the balance: the role of the Court of Justice in defining the institutional position of the European Parliament’ (1987) at 49-50. See also Parliament’s Resolution on the democratic deficit in the European Community, OJ (1988) C187/229, § 4. 147 Critical about the identification of three powers was the Vedel report, Bull. EC Suppl. 4/72 at 40. 148 Such general rules as exist are laid down in e.g. Articles 189, 192, 202 and 211 TEC. 149 The (limited) exceptions to this rule concern Article 54 of the EC Statute of the Court (Article 55 of the Euratom Statute of the Court of Justice), or where the principle of res judicata applies: joined cases T-177/94 and T-377/94, Altmann and Casson vs Commission, [1996] ECR II-2041, § 80. One could argue that these cases—the fruit of common sense rather than legal dogmatism—are exceptions confirming the rule. 150 Or at least the Court of Justice’ attitude. The Court of First Instance, with the warm support of Advocate General Jacobs (case C-50/00 P, Unión de Pequeños Agricultores vs Council (Opinion), [2002] ECR I-6677, § 41 et seq.) proposed to considerably widen the scope for individuals to attack regulations or directives under Article 230 TEC: case T-177/01, Jégo-Quéré et Cie SA vs Commission, [2002] ECR II-2365, § 51. The Court of Justice disagreed: Unión de Pequeños Agricultores, §§ 44-45.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

39

The Constitution goes further on the separation of functions path by allowing legal actions from individuals against delegated legislation (if I may summarise Article III-365(4) Const151 in this manner). The situation is somewhat more complicated for the relationship between the functions of ‘legislation’ and ‘execution’. On the one hand, in the system of the Treaties, the Commission is—apart from the Member States152— the prime executive in the system153 (although important exceptions to the rule exist, notably where CFSP is concerned, where the Council fulfils this role). On the other hand, there is no single legislature.154 Its composition depends on the applicable Treaty provision. Although in some cases this causes the Commission to exercise both functions of legislation and execution (i.e., in the field of competition law), in most cases the legislature comprises the Council alone, or the Council and Parliament. The Commission’s right of initiative can in this view be explained as a check on the legislators.155 Besides making a clearer distinction between the different functions, the Constitution assigns them to different institutions156 although, as will be shown in Chapter 9, important exceptions to the rule remain. . . . And a critique A more critical view has been argued,157 although not always on solid grounds.158

151 ‘Any natural or legal person may, under the conditions laid down in paragraphs 1 and 2, institute proceedings against an act addressed to that person or which is of direct and individual concern to him or her, and against a regulatory act which is of direct concern to him or her and does not entail implementing measures.’ 152 For an analysis of the Member States’ role as an example of Vollzugsföderalismus see Lenaerts, op. cit. footnote 36 at 14 et seq. 153 Differently: Barents, ‘Een Grondwet voor Europa (IV): besluiten en besluitvorming’ (2004) at 353. Balanced is Mistò, ‘La collégialité de la Commission européenne’ (2003) at 193. 154 Dashwood, ‘The constitution of the European Union after Nice: law-making procedures’ (2001) at 218 argues that the pre-Constitution EU does not, as such, have a legislature. 155 The Club de Florence, op. cit. footnote 46 at 98-99 argued against the functional separation of powers theory (without identifying it as such) with the argument that ‘[l]a Commission, embryon d’exécutif européen, exerce une gamme de compétences qui relèvent tantôt de la fonction législative—grâce à son droit de initiative—, tantôt de la fonction judiciaire—en matière de concurrence, par exemple’. That, however, would not be the argument why the functional separation of powers analysis is problematic: these Commission competencies can still be explained in terms of Madison’s emphasis on ‘partial agency’: ‘Federalist paper 47’, op. cit. footnote 20 at 304. 156 See for legislation: Articles I-20.1, I-23.1; for execution: I-26.1; for adjudication: I-29.1 Const. 157 E.g., Achterberg, ‘Die Gewaltenteilung in den Europäischen Gemeinschaften’ (review, 1968) at 240-245, Bobbert, Interinstitutionelle Vereinbarungen im Europäischen Gemeinschaftsrecht (2001) at 2, Fuß, op. cit. footnote 50 at 581-583, ‘Justus Lipsius’, ‘The 1996 Intergovernmental Conference’ (1995) at 237 and Kortmann, op. cit. footnote 70 at 122. From a political science point of view, Majone, op. cit. footnote 114. Jacqué, op. cit. footnote 146 at 48-50 sees the organisational plan of the Community rather in a ‘representation of interests’ than in a ‘separation of powers’, but are these two principles mutually exclusive? 158 For example, Dreier rather bluntly notes that the European Parliament is not the legislative power, since it does not possess an unrestricted right of initiative (op. cit. footnote 13 at 540). That presumes an extreme interpretation of separation of powers. Moreover, his interpretation of the— formal and informal—powers of Parliament seems to be unduly restrictive. Dann, ‘European Parliament and executive federalism: approaching a parliament in a semi-parliamentary democracy’ (2003) at 553, believes that the system whereby national executives legislate in the Council is an ‘inappropriate’ but necessary consequence of executive federalism. See further Crijns, Het europese perspectief van het nederlandse staatsrecht (1989), 112 et seq, who emphasises the ‘quadripartite’ character of the EC.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

40

One line of criticism argues that separation of powers would not be relevant for the EU simply because it is not a state.159 It is a view based on a dogmatic distinction between states and other international actors exercising sovereignty that finds no basis in international law, history or logic.160 To the extent it possesses state-like features, the EU can be analysed in similar terms. This corresponds with the line taken by the Court of Justice, which pragmatically speaks of the EU as a ‘Communauté de droit’161 by analogy to the Etat de droit. Then there is the view of authors arguing that there is a certain correspondence between the three ‘functions’ and three institutions, but who consider that the Council, Parliament and Court of Justice correspond to the triad of executive, legislature and adjudicator. In that view, the Commission is a new element breaking the Montesquivian mould.162 This view fundamentally misreads both the institutional roles of the Council and Parliament, as well as Montesquieu’s (or Madison’s) theory. Where the Council acts in the field of domestic policy, it can—notwithstanding its composition—never be a part of the central executive power. Clearly, the members of the Council do not represent only their respective governments but their Member States. The Member States might execute much of EU legislation, but they are not merely executors. ‘The term “the State” . . . necessarily encompasses all the bodies which exercise legislative, executive and judicial powers. The same is true of the bodies which, in a federal state, exercise those powers at federal level.’163 Notwithstanding these false criticisms, some real objections can be formulated against the application of (especially the Madison-style) separation of functions to the EU system. Firstly, the system of the Treaties does not allow for the easy identification of the three ‘powers’.164 An obvious exception here is the adjudicating power which, without much difficulty, can be said to rest wholly with the Union’s courts. The legislative and executing powers, however, pose problems because the system of attribution of competencies in the Treaties caused the ‘functions’ or ‘powers’ to be separated from the institutions. The main problems can be summarised under the five headings hierarchy of norms; the attribution system of the Treaties; ‘execution’ by Parliament, the Member States and the Council; comitology; and agencies. To start with the first of these issues, the hierarchy of norms problem is typical to the European Union. This was not part of separation of powers theories in Montesquieu’s time. The gesetzmäßigen Verwaltung principle implies that the administration should act within rules that either have been adopted by the ‘legislative power’ (which should include parliament), or which are based on such rules. In this logic, a functional separation between legislation and execution necessitates a 159 I.e. Achterberg, op. cit. footnote 157, to some extent; and Klein, ‘Entwicklungsperspektiven für das Europäische Parlament’ (1987) at 100. 160 In a slightly different context Koopmans cogently argued that the debate about the statehood of the European Union is not very fruitful: ‘Federalism: the wrong debate’ (1992) at 1047 et seq. 161 Opinion 1/91, [1991] ECR I-6079, § 21. Similar Seeler, ‘Die Legitimation des hoheitlichen Handelns der Europäischen Gemeinschaft/Europäischen Union’ (1998) at 724. 162 Schrans, ‘La Communauté et ses institutions’ (1982) at 18. Majone, op. cit. footnote 114 at 324 considers the Commission’s monopoly on legislation to be ‘the most striking violation of separationof-powers’ (which he appears to understand as Madison-style). 163 Case C-323/96, Commission vs Belgium, [1998] ECR I-5063, § 27. See also case 14/83, Von Colson and Kamann vs Land Nordrhein-Westfalen, [1984] ECR 1891, § 26. 164 In this sense already the Rapport van de werkgroep ad hoc voor de behandeling van het vraagstuk van de uitbreiding van de bevoegdheden van het Europese Parlement (Vedel report), Bull. EC Suppl. 4/72 at 40.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

41

hierarchy of norms between statutes (legislative acts) and what in many Member State are called ‘(royal) orders’, that are of an executive nature. The current Treaties totally lack such a clear hierarchy.165 Both ‘statute’-like instruments and ‘orders’ can take the form of regulations, or even of Council regulations. The consequences to which this leads are further discussed in section 13.3.1. The Constitution would signify considerable improvements in this respect. These are discussed in detail in section 13.3. Secondly, the attribution system of the Treaties represents another method of executing functions than the classical triad of legislation, execution and adjudication.166 Namely, the goal of separation of powers theories (or what some critics understood to be the trias politica) was achieved by limited attribution of competencies. This system causes the legislative power to change, depending on the specific competence. This must be seen in connection with Article 7(1) TEC: ‘Each institution shall act within the limits of the powers conferred upon it by this Treaty’.167 As a result, depending on the legal basis, the power to adopt or approve rules of general application is exercised by the Commission alone;168 or by the Commission and Council;169 or by the Council and Parliament;170 or by the Commission, Council and Parliament;171 by that trio plus the Member States;172 etc. Some support for this line of thinking can be derived from case law. When the UK argued that in the EC Treaty ‘all original law-making power is vested in the Council, whilst the Commission has only powers of surveillance and implementation’, the Court of Justice disagreed: ‘the limits of the powers conferred on the Commission by a specific provision of the Treaty are to be inferred not from a general principle, but from an interpretation of the particular wording of the provision in question’.173 The third objection concerns the ‘execution’ of EC law by Parliament, the Member States and the Council. The Commission may at first sight seem to be the ‘executive’ in the Union,174 but this notion can be challenged from three corners. To begin with, Parliament does not wish to restrain itself to a purely legislative or controlling role. Every last day of its monthly session Parliament adopts a large number of resolutions relating to foreign policy, specific and individual human rights cases, or both.175 That activity qualifies neither as legislation nor as oversight over the 165 Bieber and Salomé, ‘Hierarchy of norms in European law’ (1996) at 913; Monjal, ‘La Conférence intergouvernementale de 1996 et la hiérarchie des normes communautaires’ (1996) at 686 et seq. 166 Ipsen, op. cit. footnote 52 at 319 et seq. In a similar vein Pescatore, ‘l’Executif Communautaire: Justification du quadripartisme institué par les traités de Paris et de Rome’ (1978) at 388 et seq. and the Vedel report of 25 March 1972, Bull CE Suppl. 4/1972, Part II at section II. Everson a.o., The Role of specialised agencies in decentralising EU governance—Report presented to the Commission (no date) at 26-27 argue that the failure of the founding fathers to draft the Communities on the scheme of a trias politica organisation was the consequence of a combination of advanced functionalist analysis and concerns about sovereignty. I am not sure whether either reason at the time was articulated in such terms; but this question seems more of historical than practical interest. 167 See also Article I-19.2 Const. 168 E.g. Articles 38, 86(3), 95(6), 134 TEC; see also Articles III-165.3, III-166.3 and III-168.4 Const. 169 E.g. Articles 14(3), 26, 45, 49, 57(2), 60(2), 64(2), 67(2), 87(3)(e), 96, 100(1), 119(3), 132(1), 133(2) and (4) and 301 TEC; see also Articles III-151.5, III-159 and III-169 Const. 170 Article 190(5) TEC; see also Article III-335.4 Const. 171 E.g. Articles 251 and 252 TEC; see also Article III-396 Const. 172 E.g. Article 22, 172, 269 TEC; see also Article I-54.3 Const. 173 Joined cases 188-190/80, France, Italy and United Kingdom vs Commission, [1982] ECR 2545, §§ 4 and 6. 174 In this vein Pescatore, op. cit. footnote 166 at 392. 175 It is neither necessary nor possible to be exhaustive. As one of many examples see the Resolution on

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

42

executive. It either interferes in the domain of foreign relations (which in Locke’s separation of powers theory was separate from the executive)176 or in that of the executive (where the application of EU law in individual cases is at stake). Moreover, the executive federalism (Vollzugsföderalismus) characterising the Union’s structure177 implies that the Member States can lay some claim to being at least partially part of the executive.178 Articles 10 TEC and I-5.2 Const require the Member States to assist the Union. Moreover, Article I-37.1 Const explicitly provides that ‘Member States shall adopt all measures of national law necessary to implement legally binding Union acts’. I am, however, not sure whether this in itself suffices to qualify the Member State authorities as part of the executive since much of the implementing work done by the Member States consists of legislation. Lastly, the Council can challenge the Commission’s claim as the executive,179 especially where it acts in the second and third pillars, but also (at least in the view of some)180 as far as it is involved in first-pillar policy co-ordination activities. The Council is itself formed of representatives from the executives in the Member States.181 However, I agree with Pescatore182 that such a construction of the Council requires much effort to squish reality into the mould of Montesquivian ‘tripartisme’. One would have to conclude that the Commission and the Council are both executive powers (the latter as a resultant of the combined Member State executives). That, however, again requires a serious adaptation of reality to theory: in practice the Commission and the Council are disposed to be quite independent from each other. Fourthly, the problem of comitology must be mentioned. This is the subject of detailed analysis in Chapter 13, so my observations here can be concise. The 1999 Nigeria: the case of Mrs Amina Lawal, OJ (2003) CE272/486. 176 Although he admitted that, for practical reasons, the two will normally be united in one hand. ‘Second Treatise of Government’, Chap. XII and especially § 148, op. cit. footnote 7 at 412. 177 On this in more detail Dubey, ‘Administration indirecte et fédéralisme d’exécution en Europe’ (2003), 90 et seq. Dann, Philipp, Parlamente im Exekutivföderalismus (2004) prefers the term ‘Exekutivföderalismus’ but seems to intend the same notion. 178 In this sense the anonymous author of ‘The report of the Committee of Independent Experts: an ill wind . . .’ (1999) at 270. Due, ‘Artikel 5 van het EEG-Verdrag—Een bepaling met een federaal karakter ?’ (1992) at 355 argues that there is no executive in the normal sense of the word since the Member States have to implement much of EU law. 179 In this sense Advocate General Tesauro in case C-327/91, France vs Commission (Opinion), [1994] ECR I-3641, § 33. Blumann, ‘Le pouvoir exécutif de la commission à la lumière de l’Acte unique européen’ (1988) at 25 notes the literature but (at 29-30) argues that the amendment of the preAmsterdam Article 155 TEC clarified that the Council was obliged to delegate the execution of legislation to the Commission. Hix considers the EU to be blessed with two executives—the Commission and the Council: ‘Executive selection in the European Union: does the Commission President investiture procedure reduce the democratic deficit?’ (1997). Kirchhof (‘The balance of powers between national and European institutions’, 1999 at 227) argues that ‘the European Community, as an association of states largely, makes law through institutions made up of representatives of the Member State governments, that is, the executive branch’. That, however, goes too far: neither the Commission nor Parliament is made up of Member State government officials and the statement suggests that the MEPs only have a consultative right. The German case law referred to by him may in this sense well be out of date, and his argument would have benefited from taking account of the Van Gend en Loos case law dating from the early 1960s. His conclusion that ‘the European Assembly that calls itself “Parliament” cannot, at present, be a parliament with legislative power and budgetary sovereignty’ is not hindered by developments from the early 1970s up to the Treaty of Nice. 180 In this sense e.g. Curtin, ‘The executive(s) of the European Union—Out of the shadow(s), under the spotlight(s)?’ (2005) at 84 et seq. 181 Articles 203(1) TEC and I-23.2 Const. 182 Pescatore, op. cit. footnote 166 at 389-390.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

43

Comitology Decision foresees for a possibility for the Council to exercise itself directly certain implementing powers (Article 1), although this is not common.183 Under the management, regulatory and safeguard procedures, the Council can adopt a decision differing from that proposed by the Commission. This, however, is the case, whether or not the matter concerns implementing rules or the application of rules in individual cases184 (albeit under slightly different conditions, depending on the procedure used).185 Fifthly, some of the newest generation agencies that have been founded seem to confound ‘legislative’ and ‘executive’ functions. For example, the European Aviation Safety Agency186 has the right to ‘assist the Commission by preparing measures’ (which are generally applicable) to implement the Regulation, as well as to investigate individual cases. Agencies are discussed in more detail in section 13.4. In conclusion, as far as separation of functions is intended as a means to organise the constitution, it is possible to analyse the European Union in terms of its ‘pure’ variety. However, as far as the Madison-style variety of the theory is concerned, there is no clear mapping of three functions onto a set of three institutions. Whilst it is not impossible to analyse the European Union in such terms, this theoretical framework therefore constitutes a less than ideal paradigm for an analysis of checks and balances. 2.3.4

Separations of powers as a doctrine in EU law

To what extent is separation of powers then a part of EU law as a legal doctrine? Case law alternatively refers to the ‘balance of powers’ or the ‘institutional balance’ doctrine. These two concepts in fact cover one set of ideas187 although the Courts tend to attach a different emphasis to each of these names. Of course, the analysis below is retroactive: it is unlikely that the Court of Justice consciously established its case law with a clear conceptual distinction between separation of functions and the separation of institutions in mind. Nevertheless, in the end it supported the latter, although in the beginning the separation of functions briefly had the upper hand. A brief flirtation with separation of functions In Meroni188 the Court of Justice concluded that ‘there can be seen in the balance of powers which is characteristic of the institutional structure of the Community a fundamental guarantee granted by the Treaty . . .’ The judges distinguished delegation of powers provisions into  ‘clearly defined executive powers the exercise of which can . . . be subject to strict review in the light of objective criteria determined by the delegating authority’ and 183 Council Decision of 28 June 1999 laying down the procedures for the exercise of implementing powers conferred on the Commission, OJ (1999) L184/23. 184 Case 16/88, Commission vs Council (Comitology), [1989] ECR 3457, § 11. 185 Articles 4 to 6 of the 1999 Comitology Decision. 186 Established by Regulation (EC) 1592/2002 of the European Parliament and of the Council of 15 July 2002 on common rules in the field of civil aviation and establishing a European Aviation Safety Agency, OJ (2002) L240/1. 187 In this sense Prechal, ‘Institutional balance: a fragile principle with uncertain contents’ (1998) at 275, Majone, op. cit. footnote 114 at 323 and implicitly, Fischer-Appelt, Agenturen der Europäischen Gemeinschaft (1999) at 169-170. Different Everson a.o., op. cit. footnote 166 at 26-28. 188 Cases 9/56, Meroni & Co., Industrie Metallurgiche, SpA vs High Authority, [1958] ECR 133 and 10/56, Meroni & Co. vs High Authority, [1958] ECR 157.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

44



‘discretionary power, implying a wide margin of discretion which may, according to the use which is made of it, make possible the execution of actual economic policy’. The second variety ‘replaces the choices of the delegator by the choices of the delegate’ and ‘brings about an actual transfer of responsibility’.189 The ghosts of Montesquieu and Madison hovered over Meroni: although the Court did not state it explicitly, the first type of delegation corresponds nicely with administration in the narrow sense of the word (‘clearly defined executive powers’), whereas the second type is now termed delegated legislation. The criterion used by the Court was not based on the competencies or powers of the institutions, but on the nature of the functions to be performed. In later years the Courts rarely referred to the delegation dicta in Meroni.190 In Ladbroke Racing191 the applicant unsuccessfully attempted an argument based on the Meroni dictum. In Van der Wal192 the Court ruled that ‘the applicant has not shown how the principle of the separation of powers . . . would not be respected if the replies provided by the Commission to national courts . . . were not made accessible . . .’ This seems to imply that the judges were amenable to the idea of separation of powers as a legal doctrine. However, they did not indicate whether they understood it as the separation of functions or the separation of institutions. Four years later they again referred to it, but only in passing and without clarifying the issue.193 The earlier Prakash judgment perhaps counts as the real end of this short-lived flirtation with separation of functions. The Court decided there that it could not ‘without breaking the rules concerning the separation of judicial and administrative powers’ put its value judgment in an administrative appreciation in the place of that of the Commission.194 ‘Institutional balance’ or ‘balance/separation of powers’ Meanwhile, a separate strand of case law developed that proved more enduring.195 This case law strikes in the direction of a separation of institutions rather than of functions. The Court has not been consistent in its use of terminology: before 1970 the Court had a slight preference for the term ‘separation of powers’, whilst from that year196 the Court more often referred to it as ‘institutional balance’.197 189 Ibid. The cases were ruled under the ECSC but, arguably, the same logic would apply under the (E)EC and Euratom Treaties. 190 See for one example joined cases T-369/94 and T-85/95, DIR International Film Srl a.o. vs Commission, [1998] 1998 II-357, § 52. However, this judgment no longer breathes the spirit of separation of functions thinking. 191 Case T-67/94, Ladbroke Racing Ltd vs Commission, [1998] ECR II-1, § 170. 192 Case T-83/96, Van der Wal vs Commission, [1998] ECR II-545, § 55. 193 Case C-491/01, The Queen and Secretary of State for Health ex parte: British American Tobacco (Investments) Ltd a.o., [2002] ECR I-11453, § 110. A rare recent repetition can be found in joined cases C-154/04 and C-155/04, Alliance for Natural Health and Nutri-Link Ltd vs Secretary of State for Health, [2005] ECR I-6451, § 90. 194 Joined cases 19/63 and 65/63, Prakash vs EAEC Commission, [1965] ECR 533. 195 It has been argued that there is continuity between the Meroni doctrine and the later ‘institutional balance’ case law: e.g. Verhoeven, op. cit. footnote 142 at 205 et seq. and at 251 et seq. For the reasons explained on these pages I would rather see the two as separate doctrines 196 Case 25/70, Einfuhr- und Vorratsstelle für Getreide und Futtermittel vs Köster et Berodt & Co. Kg., [1970] ECR 1161, §§ 4 and 9, and case 30/70, Scheer vs Einfuhr- und Vorratsstelle für Getreide und Futtermittel, [1970] ECR 1197, §§ 13 and 18. 197 For a critical note on this ‘phantom’ concept see Bieber, ‘The settlement of institutional conflicts on the basis of Article 4 of the EEC Treaty’ (1984) at 518-520. Jacqué, ‘The principle of institutional

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

45

It is useful to follow this strand of case law in order to perceive its limitations. Not long after Meroni, the judges in Fonderies de Pont-à-Mousson vs High Authority198 spoke of ‘the separation between the powers vested in the High Authority alone and those vested jointly in the High Authority and the Council’. The use of the words ‘separation of powers’ is deceptive: what are separated are neither functions nor institutions, but the powers ‘vested’ in them. That verb suggests that ‘powers’ should be read as ‘competencies’ so that, effectively, the Court spoke of the separation of competencies of the different institutions. One year later, the Court ruled that ‘[t]he ECSC Treaty is based on the principle of a strict separation of the powers of the Community institutions and those of the authorities of the Member States’.199 Subsequently, in the Isoglucose cases, the Court of Justice ruled that the ‘power’ of Parliamentary consultation ‘represents an essential factor in the institutional balance intended by the Treaty’.200 That mantra has been repeated in a series of judgments with some minor variations.201 ‘Power’ must again be read as ‘competence’ and this, combined with the reference to the Treaty, essentially forced the Court’s hand in interpreting ‘institutional balance’ as referring to the attribution of competencies.202 From the mid-1970s, the Court filled the contours of its institutional balance doctrine with more colour. In Rey Soda it gave, as a reason for a strict interpretation of a Treaty provision on agriculture, the ‘preservation of the balance between the powers of the Council and the Commission’.203 Note that it referred not to the competencies of Commission and Council (which would have been a static approach based on the Treaties as they then stood), but rather to the balance of their powers, which has a more dynamic ring to it. In 1985 the Court ruled that the ‘balance of powers between the institutions provided for by the Treaties’ disallowed Parliament to ‘deprive the other institutions balance’ (2004) offers a more positive analysis. 198 Case 14/59, Société des fonderies de Pont-à-Mousson vs High Authority, [1959] ECR 215. 199 Case 6/60, Humblet vs Belgium, [1960] ECR 559, § 2. 200 Cases 138/79, Roquette Frères vs Council (Isoglucose), [1980] ECR 3333, § 33 and 139/79, Maizena GmbH vs Council, [1980] ECR 3393, § 34. 201 Cases 1253/79, Battaglia vs Commission, [1982] ECR 297, § 17; 828/79, Adam vs Commission, [1982] ECR 269, § 17; 817/79, Buyl a.o. vs Commission, [1982] ECR 245, § 16; C-65/93, Parliament vs Council (GSP), [1995] ECR I-643, § 21; C-21/94, Parliament vs Council, [1995] ECR I-1827, §§ 17 and 26; C-180/97, Regione Toscana vs Commission (Order), [1997] ECR I-5245, § 6; C-95/97, Région wallonne vs Commission (Order), [1997] ECR I-1787, § 6; C-392/95, Parliament vs Council (Visas), [1997] ECR I-3213, §§ 14 and 22; C-408/95, Eurotunnel SA a.o. vs SeaFrance, [1997] ECR I-6315, § 45; and CFI cases T-89/96, British Steel plc vs Commission, [1999] ECR II-2089, § 165 and joined cases T-125 and 152/96, Boehringer Ingelheim Vetmedica GmbH and C.H. Boehringer Sohn, [1999] ECR II-3427, § 130. The Court of Justice at times has used terms which seem to connote the same concept. For example, in case C-327/91, France vs Commission, [1994] ECR I-3641, § 28 it interpreted Article 228 (the current 300) TEC as ‘establishing a balance’ between the institutions. It then went to describe the ‘powers’ (in fact, competencies) of the different institutions in this field. In joined cases T-68/89, T-77/89 and T-78/89, Società Italiana Vetro SpA, Fabbrica Pisana SpA and PPG Vernante Pennitalia SpA vs Commission, [1992] ECR II-1403, § 319 the Court of First Instance speaks of the interinstitutional balance, which seems to denote the same concept. Lastly—and this increases the confusion surrounding the concept—in case 204/86, Greece vs Council, [1988] ECR 5323, § 17 the Court used the term ‘separation of powers, as laid down in the Treaty, between the institutions’ where it meant ‘institutional balance’. 202 Another, more dynamic interpretation has been forwarded, in the sense that the institutional balance as not yet been reached from a democratic point of view. Thus e.g. Hilf, ‘Das Klagerecht des Europäischen Parlaments im Organstreit’ (1990) at 278. ‘Institutional balance’ in this sense is a political, not a legal, notion. 203 Case 23/75, Rey Soda vs Cassa Conguaglio Zucchero, [1975] ECR 1279, § 9.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

46

of a prerogative granted to them by the Treaties themselves’.204 Shortly afterwards it ruled that the balance of powers between the institutions precluded it from considering the objectives pursued by the Commission in pursuing infringement actions.205 In brief, in these judgments the Court accepted that the institutional balance doctrine limits an institution in the exercise of its competencies where these deprive another institution from using theirs. In 1988 the Court implicitly rejected Parliament’s suggestion that the ‘institutional balance’ implied a ‘principle of equality’.206 The most pertinent judgment came two years later, when the Judges in Chernobyl described in more detail what the ‘institutional balance’ implied. They repeated that ‘[o]bservance of the institutional balance means that each of the institutions must exercise its powers with due regard for the powers of the other institutions. It also requires that it should be possible to penalize any breach of that rule which may occur’. It followed that the Court ‘which under the Treaties has the task of ensuring that in the interpretation and application of the Treaties the law is observed, must therefore be able to maintain the institutional balance’. Since the Court drew to itself the task of being the guarantor of the ‘institutional balance’207 it shut that concept off from extra-legal checks and balances: ‘the Parliament’s prerogatives, like those of the other institutions, cannot be breached without it having available a legal remedy, among those laid down in the Treaties, which may be exercised in a certain and effective manner’. Lastly, ‘[t]he absence in the Treaties of any provision giving the Parliament the right to bring an action for annulment may constitute a procedural gap, but it cannot prevail over the fundamental interest in the maintenance and observance of the institutional balance laid down in the Treaties establishing the European Communities’.208 From this logic three elements may be distilled. 209 Firstly, each institution has a role assigned to it in the institutional structure created by the Treaties. That structure comprises a ‘system for distributing powers’. Each of the institutions must exercise its powers with due regard for the powers of the other institutions.210 Secondly, it should be possible to penalise any breach of that rule which may occur. The Court has the task of ensuring that in the interpretation and application of the Treaties the law is observed and must therefore be able to maintain the 204 Case 149/85, Wybot vs Faure a.o., [1986] ECR 2391, § 23. 205 Cases 415/85, Commission vs Ireland, [1988] ECR 3097, § 9 and 416/85, Commission vs United Kingdom, [1988] ECR 3127, § 9. In case T-243/94, British Steel plc vs Commission, [1997] ECR II-1887, § 182 the plaintiff argued that the ‘balance of powers’ had been disturbed by the Commission’s alleged failure to adhere to the terms of Article 95 ECSC. The Court of First Instance did not reply to that argument, but ruled that the Commission had in fact respected the terms of that provision. 206 Case 302/87, Parliament vs Council (comitology), [1988] ECR 5615, §§ 19 and 21. Parliament had pleaded to be put ‘on at least an equal footing’ for a long time: see e.g. its Resolution on European Union, OJ (1975) C179/28. 207 On a general approving note on the Court’s influence on the institutional balance, Lenaerts and Corthaut, ‘Judicial review as a contribution to the development of European constitutionalism’ (2004) at 2 et seq. 208 Case C-70/88, Parliament vs Council (Chernobyl), [1990] ECR I-2041, §§ 22-26. 209 Without using the words ‘institutional balance’ the logic of the Court was already described by Advocate-General Van Gerven in case 70/87, Fédération de l’industrie de l’huilerie de la CEE (Fediol) vs Commission (Opinion), [1989] ECR 1781, § 19. 210 See also case C-58/94, The Netherlands vs Council, [1996] ECR I-2169 at 41.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

47

institutional balance by means of a legal remedy.211 In a way the Court’s selfproclaimed monopoly of maintaining the institutional balance limits the scope of that concept to the strictly legal domain. In the logic of the Court ‘powers’ means ‘competencies’.212 Competencies, however, are tied to institutions, not to functions. Thirdly, the absence in the Treaties of any provision giving Parliament the right to bring an action for annulment may have constituted a procedural gap, but could not prevail over the fundamental interest in the maintenance and observance of the institutional balance. The Court makes a dimension leap here: the institutional balance is laid down in the Treaties, but where the Treaties seem to clash with it by not providing the relevant competence, the Court can nevertheless rely on the ‘institutional balance’.213,214 This certainly goes further than a ‘practical short-hand’215 of the attribution of competencies principle since the Court effectively proclaims that a certain element of ‘balance’ should exist between the institutions.216 It is respectfully submitted that the Court demonstrated a certain blindness to possible remedies to the EU’s institutional structure other than those based on law. As will be discussed in the following Chapter, interinstitutional convention and practice have a large role to play here, next to recourse to the Courts. In this sense, the Court’s monopoly on maintaining the institutional balance perhaps goes too far. Admittedly, the reasoning in Brasserie du Pêcheur and Factortame does not fit the logic of the previous case law. In those cases the German Government argued that the creation of a general right to reparation for individuals could be created only by legislation. The recognition of such a right by judicial decision would, in its view, be ‘incompatible with the allocation of powers as between the Community institutions and the Member States and with the institutional balance established by the Treaty’. The Court rejected this argument, inter alia because ‘in many national legal systems the essentials of the legal rules governing State liability have been developed by the

211 Similar in case C-316/91, Parliament vs Council (Lomé Convention), [1994] ECR I-625, § 12. 212 In case C-417/93, Parliament vs Council (TACIS), [1995] ECR I-1185, § 26 it was not entirely certain whether the Court meant ‘competencies’ or ‘influence’ when it observed that ‘the overall balance of the powers allocated to the Commission and the Council is not decisively affected by the choice between the two types of committee at issue’. 213 This is difficult to reconcile with the Court’s ruling in joined cases 188-190/80, France, Italy and United Kingdom vs Commission, [1982] ECR 2545, § 6, where it said that ‘the limits of the powers conferred on the Commission by a specific provision of the Treaty are to be inferred not from a general principle, but from an interpretation of the particular wording of the provision in question . . . analysed in the light of its purpose and its place in the scheme of the Treaty’. Unless, of course, a different standard is applied to measure the prerogatives of Parliament than is used for those of the Commission. 214 In case 109/75R, National Carbonising Company Ltd vs Commission (Order), [1975] ECR 1193, § 8, the President of the Court of Justice refused to adopt an interim measure on the basis that it would be contrary to the balance between the institutions for him to exercise the Commission’s prerogatives. 215 Prechal, op. cit. footnote 187 at 277. Hofmann, ‘Rechtsstaatsprinzip und Europäisches Gemeinschaftsrecht’ (1996) at 328 and Jacqué, op. cit. footnote 197 at 383, tend to equate the institutional balance with the ‘division of powers between the different institutions’. Kortmann rightfully recognises the Chernobyl ruling as a ‘creatio ex nihilo’: op. cit. footnote 69 at 170. Different, in the sense of denying any link with Article 7 TEC, Bieber, ‘Artikel 7 EG’ (2003) at 664. 216 A similar dynamic view underpins the judgment in case C-417/93, Parliament vs Council (TACIS), [1995] ECR I-1185, § 26: ‘the overall balance of the powers allocated to the Commission and the Council is not decisively affected by the choice between the two types of committee at issue so that the amendment to the Commission’s proposal is not substantial’.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

48

courts’.217 That, however, is a separation of powers argument rather than a position based on the attribution of competencies. In the late 1990s the Court of Justice added an additional element to the concept of institutional balance. The president of the Court disallowed an interpretation of ‘Member States’ that included governments of regions amongst the Member States, because this would ‘undermine the institutional balance provided for by the Treaties, which . . . govern the conditions under which the Member States . . . participate in the functioning of the Community institutions’.218 I would respectfully submit that, whilst it is obvious that such regions are not Member States, this has little to do with the ‘institutional balance’. The region of Wallonia did not attempt to act as an institution ultra vires the Treaty; rather it is neither an institution nor a member of one. By 1996, the Court firmly placed the doctrine in the centre of the EU’s institutional structure: ‘the institutional rules governing the allocation of powers between the various Community institutions occupy an essential place in the Community legal order’.219 Around the same time it promoted the ‘institutional balance’ concept to a ‘principle’.220,221 Could separation of powers be a legal doctrine in the European Union? Could the Court have taken a different view and expounded a separation of powers doctrine in the way the US Supreme Court has done? As shown above, it certainly started such a line of thought. In my view, the Court was condemned to link the balance of powers concept to the attribution of competencies and this for two reasons. Firstly, the Court, by the nature of its task, has to work from within the context of the EU’s constitution. The Court can apply the Treaties. It cannot shape them ad infinitum.222 Secondly, the very words ‘balance of powers’ or ‘institutional balance’ raise the difficulty of the measurement: when are institutions out of balance? ‘Power’ in an abstract sense (i.e., influence) cannot easily be measured in legal terms and is hopelessly inadequate as a legal tool. When judges look at powers, they see competencies. This, however, changes the ‘balance of power’ from a dynamic ‘balance of influence’ to a more static ‘balance of competencies’. From there, the step to measuring the balance to the competencies defined in the Treaties is logical. There is another element at play here. Influence is a fluid concept, changing to a large measure independently from competencies. Parliament managed to increase its 217 Joined cases C-46/93 and C-48/93, Brasserie du Pêcheur SA vs Germany and The Queen vs Secretary of State for Transport, ex parte: Factortame Ltd a.o., [1996] ECR I-1029, §§ 24 and 30, respectively. 218 Cases C-95/97, Région wallonne vs Commission (Order), [1997] ECR I-1787, § 6 and C-180/97, Regione Toscana vs Commission (Order), [1997] ECR I-5245, § 6. 219 Joined cases C-239/96R and C-240/96R, United Kingdom vs Commission (Order), [1996] ECR I-4475, § 69. 220 Prechal, op. cit. footnote 187 at 277 believes this implies an institutional upgrading of its legal value. Under the name of ‘separation of powers’ it was similarly declared a principle: case T-83/96, Van der Wal vs Commission, [1998] ECR II-545, § 55. Subsequently, the Member States referred to it in Article 2 of the Protocol on the application of the principles of subsidiarity and proportionality adopted in Amsterdam. 221 Schorkopf believes that the Council, when it fails to decide on a Commission proposal, would violate the institutional balance: ‘Die Untätigkeit des Rates der EU im Gesetzgebungsverfahren’ (2000) at 370. To my mind, this is more lex ferenda than current law. 222 I agree in this with Kortmann, op. cit. footnote 69 at 172.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

49

influence during the last forty years even in periods when its formal competencies were not amended. If it were possible to weigh influence, one would perceive the balance to have shifted gradually in favour of the MEPs. On the other hand, if competencies could be weighed, the scales would have moved towards Parliament only in quantum leaps corresponding to the successive amendments of the Treaties. Since the Court can neither change the Treaties nor measure influence, its measurement is logically limited to the ‘balance’ of competencies at the state of the law as it stands at a particular time. Separation of powers analysis, however, goes beyond that. For example, Parliament must be consulted on the appointment of the ECB Executive Board. This is part of the ‘institutional balance’ laid down in primary law. Separation of powers theories, however, also have a normative, meta-constitutional component. In the example, the question transforms into the following: should Parliament only have the right to be consulted? That, however, is a question that cannot be answered by a court. It is a question of political theory and, ultimately, a political choice. The problem facing the judges in Luxembourg thus differs from that faced by the US Supreme Court. The American Constitution is permeated with the separation of functions principle. The Supreme Court may have had great difficulties in the 20th Century in establishing precise distinctions between the different domains of government, but the general principle of separation of functions was recognised as a legal doctrine as early as in 1816.223 I believe that this was less because of the intrinsic value of the political theory and more because it patently underlies the organisation of the American Constitution. As noted above, the European Union can be analysed through the prism of different separation of functions models, but it cannot be said that the separation of functions permeates the Treaties through and through. In summary, although the concepts developed in case law are useful, we need to look beyond them for the theory of separation of powers that fits the European constitution. That is the object of the next section. 2.4

Separation of institutions

2.4.1

Main principle of the separation of institutions

All modern separation of powers theories are built on a presumption and on a precondition. The presumption is that any possibility for abuse of power ultimately leads to abuse of power. The precondition is that all state power must be bound by law and that all public action is subject to the separation of powers principle. In order to be effective, separation of powers implies gesetzmäßigen Verwaltung. Montesquieu framed his theory on the different social forces that worked together under the British constitution and deduced the elements he believed to have universal application. Madison formulated his views on separation of functions and checks and balances against the background of American federalism and its relationship with the states. The overview in sub-section 2.3.2 of the application of separation of powers in some Member States shows that separation of powers, whilst a normative theory, must take account of the political conditions of the constitutional system it analyses. Similarly, the model of separation of powers that should underpin the European 223 US Supreme Court, Martin vs Hunter’s lessee, 14 (1816) 304.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

50

constitution must, in turn, be moulded to fit European conditions. As analysed in the previous section, the limitations of the Courts’ case law on separation of powers are inherent to the Court’s role. Nonetheless, I believe that this case law comprises pointers towards a viable and logical separation of powers theory. In the following pages, I shall refer to that theory as the separation of institutions.224 As we have seen, the separation of persons inter alia prohibits the same person to be minister and Member of Parliament. The separation of functions inter alia requires the constitution to be organised so that the legislative and executive functions are not controlled by the same hand. The problem, however, is often not that abuse follows from one institution’s combined use of two or three ‘functions’ of government. It lies rather in the unimpeded use by a single institution of its powers. In practice, those powers do not let themselves divide neatly into three fields, nor need they be: if power corrupts, power must be checked, whether it is legislative, executive, or any combination of these. In short, separation of institutions can be phrased as follows: for a constitution to be democratic it is at least required that all government activity ultimately rests on law. Moreover, no institution should control policy preparation, policy adoption, policy implementation or adjudication in an unchecked manner. This calls for some explanation and observations which I will discuss in the following sub-sections: the natural boundaries and different ‘esprit de corps’ in the different institutions (2.4.2); the division of state functions (2.4.3); the organisational autonomy of the institutions (2.4.4); autonomous and interinstitutional checks and balances (2.4.5); interinstitutional checks and balances divided into those caused by law, by interinstitutional convention and by other institutions (2.4.6); and the checks and balances on institutions that should regulate the appointment or dismissal of their members (2.4.7). 2.4.2

The natural boundaries and different ‘esprit de corps’ in the different institutions

In a separation of institutions analysis, it is not so much the state functions that are (or should be) separated, but rather the institutions.225 This unit of analysis lies in between the functional ‘power’ and the individual. Rather than facing the vexing and dogmatic question of how to separate ‘execution’ from ‘legislation’ (or the ‘executive power’ from the ‘legislative’ one) the analysis shifts to the mechanisms that keep the different institutions in check. Rather than fearing—like the classical authors did— that the legislator would have any possibility to administer or adjudge, the emphasis turns to ensuring that no single institution can legislate, adjudge, etc., unchecked. Putting the issue in these terms has the advantage of building onto reality. Since the early days of the Communities, the different institutions have had different policies and interests.226 There is a natural distance between the political institutions 224 Stern, op. cit. footnote 15 at 795 distinguishes a substantive, organisational and individual separation of powers. Sári, op. cit. footnote 1 at 15, uses the term ‘organic’ (‘szervezeti’) to distinguish this level from the functional one. Perhaps the term would be better, if it did not cause confusion in the context of EU constitutional law. For this reason I prefer to stick with accepted EU legal terminology. 225 I lay no claim to having ‘invented’ the idea to analyse separation of powers in terms of institutions. As only one example, Sári (op. cit. footnote 1 at 47) notes the theory according to which the US Constitution is similarly best explained in terms of separation of institutions rather than separation of functions. 226 Fuß, op. cit. footnote 50 at 582. Sasse, ‘Kommission und Rat—Konstituelle Rivalen oder

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

51

caused by a different esprit de corps and colliding vested interests. This operates as a natural check on each of them, in much the same way as the different social classes separated the ‘powers’ that made up the English political spectrum in Montesquieu’s day. 2.4.3

The division of state functions

As Lenaerts has shown, it is possible to analyse the EU’s constitution in terms of a ‘pure’ separation of three state functions.227 However, as analysed above, it is more difficult to do the same in terms of the Madison-style separation of functions. Whilst possible, there is no strict need to organise government along the lines of the ‘mystical’228 trinity of ‘legislator’, ‘administration’ and ‘adjudicator’ or to define the functions of legislation, etc. (although in the coming Chapters I freely use the term ‘administration’ or ‘executive’ when the Commission is intended). Although all of these functions are performed by the institutions, lawyers and political theorists alike have struggled for centuries to distinguish them in practice. In EU law, this problem is compounded by the peculiar structure of the Union, as a brief review of issues such as comitology and OMC reveals. The underlying problem is that the functions of government evolved. What held true in political theory in a 17th Century unitary kingdom no longer necessarily applies in a 21st Century union of Member States. Academia grudgingly acknowledges the problem: ‘The simple fact is’, says Vile, ‘if one abandons the Montesquieu functions altogether, closely related as they are to the concept of the supremacy of law, one is left without criteria for the orderly conduct of government business. Day-to-day expediency becomes the only guide for action, and few people would be prepared to admit that expediency alone should determine the organization and powers of government.’229 This, I think, is too pessimistic a conclusion. Clearly, in a separation of institutions analysis, government functions can be divided, but along different lines. I believe that the modern concept of ‘policy’ (i.e., the authorities’ expressed will) can be of help here.230 Government functions can be divided into 1) policy preparation and adoption, 2) policy implementation and 3) the adjudication of disputes. For a constitution to be ‘free’ it is necessary that none of these functions be exercised by one institution unchecked. Perceived from this231 angle, Parliament’s participation in the legislative process is not its partaking in one of the government functions. It is one of the checks and balances ensuring that policy formulation and execution are bound within a legal framework and not left to the mercy of one single other institution. Adjudication,

funktionale Partner?’ (1972) at 349 even speaks of the ‘Rivalität’ between Commission and Council. See sub-section 2.3.3, above. Vile, op. cit. footnote 1 at 16. Crossland, op. cit. footnote 1 at 61, agrees. Vile, op. cit. footnote 1 at 260. Geelhoed, ‘De Europese Unie’ (1975) at 702-703 goes somewhat in this direction but on the basis of the concept of ‘planning’. That, however, fails to fully grasp all elements involved in government action. 231 Of course, besides the matter of separation of powers, there are other elements that render the European Union into a ‘union de droit’, such as legitimacy. Qualifying Parliament’s role from this point of view is, of course, another matter. 227 228 229 230

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

52

whilst mainly a function to regulate conflicts between third parties, doubles as one of the checks and balances on the institutions. It is not necessary that each of the policy functions be performed by a separate institution. For example, the Commission—like any executive—is involved in all bar adjudication. This is not much of a problem, as long as it is checked in each field. Thus, the European Council and Parliament have partial agency in the preparation of policy, the Council and Parliament in its adoption, and the Council in its implementation. In addition, the Court’s jurisdiction provides a check on most of these aspects. 2.4.4

The organisational autonomy of the institutions

One of the most important constitutional mechanisms to balance the institutions is the principle that they should be as free as possible to organise themselves as effectively as possible.232 ‘[W]here the whole power of one department is exercised by the same hands which possess the whole power of another department, the fundamental principles of a free constitution are subverted.’233 Moreover, ‘[i]n order to lay a due foundation for that separate and distinct exercise of the different powers of government . . . it is evident that each department should have a will of its own’.234 The general principle of the political235 institutions’ autonomy is evident from primary law.236 This includes autonomy vis-à-vis the Member State authorities.237 Articles 199, 207 and 218 TEC provide that the Parliament, Council and Commission respectively adopt their own Rules of Procedure.238 In comparison, bodies that are not institutions generally lack such autonomy in adopting their Rules of Procedure,239 although they do have a certain ‘administrative autonomy’.240 232 Jacqué, op. cit. footnote 146 at 53. Because of this important constitutional principle I cannot agree with Barents, ‘Het Verdrag van Amsterdam en het Europees gemeenschapsrecht’ (1997) at 359, who feels that the decision of the Council to call itself ‘Council of the European Union’ had no basis in law and hence was illegal. Lefevre, ‘Rules of procedure do matter: the legal status of the institutions’ power of self-organisation’ (2005) at 802 and espec. at 811 et seq., is perplexed that the adoption of rules of procedure is not more regulated, but this perhaps betrays a French legal background. In that country parliament has no organisational autonomy. 233 Madison in ‘Federalist Paper 47’, op. cit. footnote 20 at 304. 234 Ibid. in ‘Federalist Paper 51’ at 319. 235 Not so for the Courts and the Court of Auditors, whose Rules of Procedure must be approved by the Council: Articles 223, 224 and 248 TEC. See also Articles III-355, III-356, III-359 and III-384 Const. 236 Ganshof van der Meersch a.o., Les Novelles—Droit des Communautés européennes (1969) at 215. 237 Case 208/80, Lord Donington vs Aspden, [1981] ECR 2205, § 19 and case 93/85, Commission vs United Kingdom, [1986] ECR 4011, § 21. Vice versa, the Court recognised that the Member States enjoy an ‘institutional autonomy’ vis-à-vis the institutions: case T-39/90, NV Samenwerkende Elektriciteits-Produktiebedrijven vs Commission, [1991] ECR II-1497, § 59. VerLoren van Themaat, ‘De Europese Raad van Edinburg als goudmijn voor juristen’ (1993) at 424 asks whether Member States at the European Council of Edinburgh were permitted to limited the autonomy of Parliament by providing for several places of operation for that institution. He points to the text of Article 289 (then 216) TEC, which speaks of the ‘seat’ (in singular) of the institutions. 238 Articles III-339, III-344 and III-352 Const. In the case of the last institution, Article 218 TEC adds the peculiar and superfluous admonition that its Rules of Procedure should ensure that ‘both it and its departments operate in accordance with the provisions of this Treaty. It shall ensure that these rules are published’. 239 See e.g. Council Decision of 12 July 2002 adopting the rules of procedure of the European Development Fund Committee, OJ (2002) L212/8. The Rules of Procedure of Eurojust were adopted by its College and approved by the Council: OJ (2002) C286/1. Even some organs of such bodies can have their own Rules of Procedure: Rules of Procedure of the OLAF Supervisory

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

53

Of course, the institutions’ internal autonomy does not imply that in practice no requests or pressures from the side of other institutions occur.241 In this respect it is possible for the Commission’s parliamentary responsibility to play its part. It is critical, however, that no institution can decide the organisation of another one. What is the extent of the institutions’ autonomy? Does it include autonomy towards the Court? To the extent that measures purely concern the internal organisation of the institution, they fall within its autonomy and are off-limits, even for the Courts. For example, Rule 16 of Parliament’s Rules of Procedure lays down the term of office of certain officers of that institution. That Rule does not, in itself, bind anyone other than the institution and could arguably not be challenged even in court. However, once a measure is intended to produce legal effects towards third parties, it falls outside the institution’s protected autonomy and becomes actionable.242 From this situation of illegal decisions under its Rules of Procedure must be distinguished the case where an institution violates its own Rules of Procedure. In the latter case, the Court of Justice has not been willing to respect the opinion of the institution as to the respect for its own Rules of Procedure. This occurred for example when the Court declared void a directive because it had been adopted in breach of the Council’s Rules of Procedure.243 Bar the prerogatives of the Court, most institutions will normally respect or ignore, but not violate the Rules of Procedure of other institutions.244 2.4.5

Autonomous and interinstitutional checks and balances

The institutions’ autonomy is not unlimited. The existence and functioning of institutions depends either on balancing institutions with other institutions, or on some

Committee, OJ (2000) L41/12. 240 This, at least, is the term used in recital (4) of the Interinstitutional agreement of 25 May 1999 between the European Parliament, the Council of the European Union and the Commission of the European Communities concerning internal investigations by the European Anti-fraud Office (OLAF), (1999) OJ L136/15. See also recital (4) of the Model Decision attached to it. 241 See for one example the intervention of Ms Maij-Weggen during Parliament’s plenary debate of 14 March 2001 on EU election assistance and observation. Conversely, it has been longstanding practice for the Commission (and the Council) to be involved in the determination of Parliament’s agenda: Commission doc. COM (73) 999 final of 30 May 1973, Annex at paragraph 4. 242 Cases 294/83, Parti écologiste ‘Les Verts’ vs Parliament, [1986] ECR 1339, § 44, C-68/90, Blot and Front National vs Parliament (Order), [1990] ECR I-2101, §§ 9 and 12, C-314/91, Weber vs Parliament, [1993] ECR I-1093, §§ 9-10 and joined cases T-222/99, T-327/99 and T-329/99, Martinez a.o. vs Parliament, [2001] ECR II-2823, §§ 52-53. MEPs can be such third parties: Martinez, § 61. As early as 1983, Parliament argued on the basis of ‘the principle of the separation of powers’ that it possessed the right to adopt resolutions ‘based on the sovereign power of the Parliament to organize the way in which it performs its tasks’: case 230/81, Luxembourg vs Parliament, [1983] ECR 255, § 14. At the time, the Court of Justice sidelined the separation of powers link, concentrating instead on the duty of the Member States to co-operate sincerely whilst having regard to Parliament’s power to determine its internal organisation. The Court did admonish the Member States that their decisions (in casu concerning the working place of the MEPs) ‘do not impede the due functioning of the Parliament’: ibid., §§ 28 and 37. 243 Case 68/86, United Kingdom vs Council, [1988] ECR 855. 244 For a case of the Commission giving an interpretation of Parliament’s Rules of Procedure see written question E-600/01 by Ms Rühle and Ms Stauner to the Commission, OJ (2001) CE318/82. Even though the Council made a point of noting that it is not bound by Parliament’s Rules of Procedure, Council Members acting within Parliament normally play the game according to the MEPs’ rules.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

54

other force. I will call the first interinstitutional checks and balances, and the second category autonomous ones. The elections that must be held at regular intervals to elect parliaments are an example of the latter category. To the extent that this obligation is laid down in a constitutional charter that a parliament cannot by itself change, this rule provides an autonomous guarantee (in the sense that it is independent from the actions of other institutions) that checks the power of the MPs. An example of the opposite proves the point. In the United Kingdom, where the sovereignty of (the Queen in) Parliament is paired with the lack of a constitutional charter, on several occasions a sitting Parliament decided to extend its own lifetime. The last occurrences of this coincided with the special circumstances of the First and Second World Wars.245 However, when in 1715 Parliament decided to extend its own lifetime to seven years without such an emergency, there was a public outcry.246 Whether or not the matter was legal at the time is beside the point. What concerns us is that perhaps this set-up exhibits a constitutional defect from a separation of institutions point of view. The Parliament Act 1911 now provides that any extension cannot be passed without consent of the Lords,247 but cannot this statute itself ultimately be amended without their consent? The adoption of legislation in the European Union provides an example of interinstitutional checks and balances. Ultimately, all legislation must rest on primary law. Depending on the legislative procedure, this is an effort shared between the Commission, the Council and Parliament. The Commission and, in many cases, Parliament disposes of prerogatives to ensure that the Council’s legislative rights are not unchecked. Policy implementation falls largely within the province of the Commission, but that institution is not uncontrolled in this: both Parliament and Council dispose of levers to influence and check the process. Moreover, part of policy implementation falls to the Member States and they are subject to their own constitutional checks and balances. 2.4.6

Interinstitutional checks and balances divided into those created by the European constitution and law, those created by interinstitutional convention and those formed by other institutions

Interinstitutional checks and balances can be divided in two types. Firstly, the institutions are obviously bound to the limits drawn by the European constitution248 and some of these interfere with institutions’ autonomy. The first example that springs to mind is the Council’s approval needed for the Rules of Procedure of the Courts and the Court of Auditors.249 As another example, Article 190(5) TEC (III-330.2 Const) requires the approval of the Council for the regulations and general conditions governing the MEPs, thereby indirectly limiting the autonomy of Parliament250 (although, admittedly, this provision concerns not so much the operation of Parliament but more the status of its Members). Lastly, in a number of 245 246 247 248

Hood Phillips and Jackson, op. cit. footnote 43 at 139. Dicey, op. cit. footnote 90 at 8, Hood Phillips and Jackson at 61. Hood Phillips and Jackson at 139. Joined cases 7/56, 3/57 to 7/57, Algera a.o. vs Common Assembly of the ECSC, [1957] ECR 39. See also case 1/55, Kergall vs Common Assembly of the ECSC, [1955] ECR 151. 249 See footnote 235, above. 250 In a similar direction De Zwaan, ‘Het Verdrag van Nice—Een bescheiden stap in het proces van Europese Integratie’ (2001) at 48.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

55

cases, the Treaties prescribe some of the matters that should be provided for in the Rules of Procedure.251 Secondly, the institutions are bound by interinstitutional convention, which will be explored in the next Chapter of this study. For example, the series of interinstitutional agreements concluded between the Commission and Parliament that culminated in the 2005 Framework Agreement252 obviously affected those institutions’ way of doing business. Finally, no institution can bind another one against its will. At several occasions when Parliament amended its Rules of Procedure, the Council made it known that these ‘do not in any way’ bind it.253 This is in line with the Chernobyl case law.254 2.4.7

Checks and balances on institutions should regulate the appointment or dismissal of their members

No single institution should control the appointment or dismissal of the members of itself or another one without proper autonomous or interinstitutional checks and balances.255 The relative independence of each institution requires that no institution be delivered to the mercy of a single other one for its appointment and dismissal. For example, the composition of the Council results from the constitutional processes in the Member States. Parliament’s composition is subject to the autonomous check of elections. The Commission’s situation is more complicated. As will be analysed in more detail in Chapter 4 of this study, its appointment requires the efforts of several institutions. 2.5

Conclusions of this Chapter

Sári256 made the point that, although one can discuss the constitutional technique of separation of powers as an abstract model, it operates in the organic development of individual constitutional systems. Since Montesquieu’s day, constitutional systems developed and became infinitely more complex. The (semi-)absolutist prince common to 17th Century Europe gave way to a large bureaucracy requiring some form of democratic legitimisation. The relatively simple executive and administrative functions of the sovereign were replaced by more complex notions such as policy making, planning and governing. The sheer volume of modern-day rule making makes it impossible for parliaments to be consulted on every small detail and the

251 See for Parliament, Council and Commission Articles 255 TEC, I-50.4 and III-399.1 Const (rules on access to documents); for the Court of Justice, Article 13 of its Statute (Assistant Rapporteurs), Article 16 of the same (composition of Chambers); for the EESC Article 261 TEC (methods of composition and terms of reference of its specialised sections and subcommittees; for the ECB Article 11.5 of the Protocol on the Statute of the ESCB and of the ECB (voting arrangements); for the EIB Article 12 of the Statute of the EIB (delegation of votes and quorum). 252 Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125. 253 See e.g. the ‘I’ item note from the Working Party on General Affairs to COREPER (Part 2)— Follow up to the General Affairs Council of 22 July 2002—Draft letter to the President of the European Parliament concerning the amendments to the European Parliament’s Rules of Procedure, Council doc. 12315/02 of 24 September 2002. 254 Case C-70/88, Parliament vs Council (Chernobyl), [1990] ECR I-2041, § 22, discussed in section 2.3.4, above. 255 A flavour of the basic idea is conveyed by Madison in ‘Federalist Paper 51’, op. cit. footnote 20 at 319. 256 Op. cit. footnote 1 at 14.

CHAPTER 2: SEPARATION OF POWERS AND CHECKS AND BALANCES

56

delegated legislation that ensues turns the executive into a formidable power, even when wholly separated from the ‘legislature’. Separation of powers models followed this development. Montesquieu’s abstractions were made from his perception of the role of the different classes in English politics where the three estates of the realm were ‘wonderfully combined’.257 Madison worked to unite the popular sovereignty dominant in the thirteen states with an element of action sorely missed from American pre-Constitutional politics. Similarly, the separation of institutions standard outlined above is purposely framed against the background of the EU’s general development. One can study animals by looking at their skeleton, but it is a poor substitute to seeing creatures living in their habitat. It follows that separation of powers models, even though they are simultaneously descriptive and normative, have a sell-by date.258 They are like clothes: there is an ideal fit, but the fit depends on the built of the person. Without going too far in abstract theoretical reflections, this means in practice that the models that ‘work’ or ‘worked’ in older constitutions need not necessarily fit the peculiar constitutional situation of the European Union. The model that in my view best fits the European Union is the separation of institutions model described above. That model emphasises the necessity to balance institutions against each other. It develops the checks and balances part of Madisonian separation of functions to its next logical step. In the process, I have striven to use terminology that allows easier use in the EU context than older models did. Clearly, although many of the guarantees that provide such checks and balances are found in law, this is only the skeleton of interinstitutional relations. As noted above, interinstitutional checks and balances can consist in law, interinstitutional convention and practice. The first of these categories is reasonably straight-forward and, before turning to an in-depth analysis of the checks and balances that make up interinstitutional relations, I must first discuss the tendons that connect the bones and make the constitutional skeleton move.

257 Majone, op. cit. footnote 114 at 327. On the development of the social-political realities behind separation of powers also Scheuner, op. cit. footnote 46 at 480. On the degree of idealisation in Montesquieu’s description of the English constitution see Vile, op. cit. footnote 1 at 92-93 and 111. 258 Crossland makes the same point: op. cit. footnote 1 at 375.

3

INTERINSTITUTIONAL CONVENTION

3.1

Introduction to this Chapter

This Chapter explores the place and role of interinstitutional convention in the European constitution. The analysis comprises two halves. Firstly, the place of interinstitutional agreements in EU law and the legal issues raised by them will be analysed in section 3.2. The second half of this Chapter is devoted to outlining a theory on the function and role of interinstitutional convention. This is started by making some brief comments on the role of constitutional conventions in several Member States (section 3.3). This will be followed by an argument to recognise interinstitutional convention in the European Union for what it is (section 3.4). 3.2

Interinstitutional agreements in the European Union

3.2.1

Their relationship with law

In the past twenty-odd years interinstitutional agreements1 have become more and more part and parcel of EU practice.2 It is now also more common for secondary 1

2

As far as I am aware, the earliest official document using the term ‘interinstitutional agreement’ was Parliament’s Resolution on the relations between the European Parliament and the Commission of the Community with a view to the forthcoming appointment of a new Commission, OJ (1980) C117/53, § 9. To my knowledge, the first ‘interinstitutional agreement’ concluded under that name is the Interinstitutional Agreement on budgetary discipline and improvement of the budgetary procedure, OJ (1988) L185/33. The first reference in academic literature I am aware of can be found in Isaac, ‘La rénovation des institutions financières des communautés européennes depuis 1970’ (1975) at 727. I do not possess an exhaustive list of all interinstitutional agreements ever concluded among the institutions, if it is at all possible to collate a full listing. Not all of those agreements have been published (e.g., the administrative agreement on access to documents referred to in Council doc. 7957/03 at 11), or published formally (e.g. the Luns procedure of 1964). The Council General Secretariat used to publish an extremely useful Digest of texts on relations between the European Parliament and the Council, but the last edition of this publication—which only covered agreements to which both the Council and Parliament were parties—dates from 1988. The following chronological list provides the references to the most important interinstitutional agreements, without attempting any judgment as to their legal status:  Luns procedure: Rapport du COREPER sur les propositions des Gouvernements luxembourgeois et allemand tendant au renforcement du rôle de l’Assemblée, Council doc. S/861/63 of 18 December 1963 at 7;  Résolution sur les compétences et les pouvoirs du Parlement européen, JO (1963) 1916 (Furler report), accepted by the Commission: Septième rapport général sur l’activité de la Communauté (1963-1964) at 344;  Unnamed agreement between the Commission and the Council on the accreditation of ambassadors from third states mentioned in OJ (1966) 4049.  Décision du 16 janvier 1969 portant installation de l’Office des publications officielles des Communautés, JO (1969) L13/19, amended by OJ (1980) L107/44;  Gentlemen’s agreement between Council and Parliament relating to the section of the budget concerning Parliament and the Council: Treaties establishing the European Communities, Vol. I, OOPEC, Luxembourg 1987 at 1091 (for the Council’s commitments);

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

58

 Rey-Scelba agreement, Débats du Parlement européen, JO (1970) Annexe 121/133-134;  Nameless agreement concluded in 1971 between the Council and Parliament on the follow-up in the Council of Parliament’s opinions, referred to in Resolution on the conciliation procedure, OJ (1989) C69/151 at 153, footnote 1.  Collaboration entre l’Assemblée et le Conseil dans le cadre de la procedure budgétaire, JO (1971) C124/62 and Council doc. R/788/72 of 18 April 1972;  Collaboration entre l’Assemblée et le Conseil lors de l’examen des actes ayant une incidence financière, Council doc. R/2020/72 of 29 September 1972;  Practical measures to strengthen the powers of control of the Parliament and to improve relations between the Parliament and the Commission, Commission doc. COM (73) 999 final of 30 May 1973;  Westerterp procedure: Relevé des décisions prises par le Conseil lors de sa 257ème session tenue le 15 octobre 1973, Council doc. R/2641/73 of 5 November 1973;  Letter from the President of the Council on the introduction of question time, OJ (1973) C19/4;  Joint Declaration of the European Parliament, the Council and the Commission concerning the institution of a conciliation procedure between the European Parliament and the Council, OJ (1975) C89/1;  Projet—orientations qui se sont dégagées des réunions tenues par la Commission de concertation les 7 et 22 novembre 1977 à Bruxelles, Council doc. R/2911/77 of 29 November 1977;  Joint Declaration by the European Parliament, the Council and the Commission of 30 June 1982 on various measures to improve the budgetary procedure, OJ (1982) C194/1;  Joint Declaration of the European Parliament and the Commission, Parliament doc. 1-328/84/rev of 23 May 1984;  Letter of M. Delors to Lord Plumb, president of Parliament, Parliament doc. PE 114.634/BUR of 7 July 1987;  Letter of M. Delors to Lord Plumb, president of Parliament (Plumb-Delors agreement), Commission doc. SG (88) IV 03026 of 14 March 1988;  Interinstitutional Agreement on budgetary discipline and improvement of the budgetary procedure, OJ (1988) L185/33;  Exchange of letters between the European Parliament, the Council and the Commission of the European Communities (right of petition), OJ (1989) C120/90;  Contacts avec le Parlement européen, Council doc. CIRC 3624/90 of 17 May 1990;  ‘A’ item note from COREPER to the Council—Accession of the Community to the FAO— Arrangement between the Council and the Commission regarding preparation for FAO meetings, statements and voting, Council doc. 10478/91 of 18 December 1991;  Code of Conduct concluded between the Commission and the Parliament, OJ (1991) C280/165;  Klepsch-Millan agreement (Code of Conduct on the implementation of structural policies by the Commission), OJ (1993) C255/19.  Interinstitutional Declaration on democracy, transparency and subsidiarity of 25 October 1993, OJ (1993) C329/133;  Interinstitutional Agreement of 28 October 1993 between the European Parliament, the Council and the Commission on procedures for implementing the principle of subsidiarity, OJ (1993) C329/135;  Arrangements for the proceedings of the conciliation committee under Article 189b, OJ (1993) C329/141;  Interinstitutional Agreement of 29 October 1993 between the European Parliament, the Council and the Commission on budgetary discipline and improvement of the budgetary procedure, OJ (1993) C331/1;  Code of Conduct concerning public access to Council and Commission documents, OJ (1993) L340/41;  Nouvelle procédure pour les ‘suites données aux résolutions d’initiative du Parlement Européen, Commission doc. SP (94) 2760/2 of 25 November 1994 ;  Joint Declaration on the legislative programme for 1994 concluded by the ‘European Parliament and the Commission, in the presence of the Council’, OJ (1994) C60/1;  Decision on the adjustment of the financial perspective annexed to the Interinstitutional Agreement of 29 October 1993 on budgetary discipline and improvement of the budgetary procedure, OJ (1994) C395/1;  Code of conduct for negotiations within the WTO, Council doc. 7926/95 of 8 June 1995;

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

59

 Code of Conduct between the Commission and Parliament, OJ (1995) C89/69;  Decision of the European Parliament, the Council and the Commission of 6 March 1995 on the detailed provisions governing the exercise of the European Parliament’s right of inquiry, OJ (1995) L113/2;  Joint declaration on the legislative programme and other activities for 1995, OJ (1995) C225/1;  Modus Vivendi of 20 December 1994 between the European Parliament, the Council and the Commission concerning the implementing measures for acts adopted in accordance with the procedure laid down in Article 189b of the EC Treaty, OJ (1995) C293/1, corr. OJ (1996) C102/1;  Interinstitutional agreement of 20 December 1994—Accelerated working method for official codification of legislative texts, OJ (1995) C293/2, corr. OJ (1996) C102/2;  Declaration by the European Parliament, the Council and the Commission of 6 March 1995 on the incorporation of financial provisions into legislative acts, OJ (1995) C293/4, corr. OJ (1996) C102/4;  Exchange of letters between Mr Weber and Commissioner Gradin of 25 July 1996 and 4 October 1996 concerning the co-operation between the Court of Auditors and the Commission in cases of suspected fraud;  Joint declaration on the legislative programme and other activities for 1996, OJ (1996) C282/1;  1996 Samland-Williamson agreement on comitology, OJ (1996) C347/125 at paragraph 72;  Joint Statement of 12 December 1996 concerning the improvement of information to the budgetary authority on fisheries agreements, OJ (1997) C20/101;  Nameless agreement on the agricultural component of the budget, OJ (1997) C150/13;  Decision on the adjustment of the financial perspective to take account of the conditions of implementation pursuant to paragraph 10 of the Interinstitutional Agreement of 29 October 1993, OJ (1997) C150/22;  Interinstitutional Agreement between the European Parliament, the Council and the European Commission of 16 July 1997 on provisions regarding financing of the common foreign and security policy, OJ (1997) C286/80;  Potsdam Agreement: Outcome of proceedings note of the Working Party of Chief Veterinary Officers on 4 and 5 June 1998—Draft conclusions of the Chief Veterinary Officers’ meeting (4 and 5 June 1998), Council doc. 9359/98 of 19 June 1998;  European Parliament and Council Decision on the adjustment of the financial perspective to take account of the conditions of implementation (submitted by the Commission pursuant to paragraph 10 of the Interinstitutional Agreement of 29 October 1993), OJ (1998) C138/155;  Interinstitutional Agreement of 13 October 1998 between the European Parliament, the Council and the Commission on legal bases and implementation of the budget, OJ (1998) C344/1;  1999 Cooperation Agreement between the Economic and Social Committee and the Committee of the Regions (not published);  Exchange of letters between Messrs Gil-Robles (doc. 201360 of 13 April 1999) and Santer (doc. D(99)81923 of 28 April 1999) on the treatment of legislative files on the entry into force of the Amsterdam Treaty;  Interinstitutional Agreement of 22 December 1998 on common guidelines for the quality of drafting of Community legislation, OJ (1999) C73/1;  Interinstitutional agreement of 25 May 1999 between the European Parliament, the Council of the European Union and the Commission of the European Communities concerning internal investigations by the European Anti-fraud Office (OLAF), OJ (1999) L136/15;  Joint Declaration on practical arrangements for the new co-decision procedure (Article 251 of the Treaty establishing the European Community), OJ (1999) C148/1;  Tripartite agreement between the European Investment Bank, the Commission and the Court of Auditors (not published) of 19 March 1999;  Interinstitutional agreement of 6 May 1999 between the European Parliament, the Council and the Commission on budgetary discipline and improvement of the budgetary procedure, OJ (1999) C172/1;  Brittan/Gil-Robles Agreement, OJ (1999) C177/47;  Council and Commission statement on Council Decision 1999/468/EC of 28 June 1999 laying down the procedures for the exercise of implementing powers conferred on the Commission, OJ (1999) C 203/1;  Code of conduct on the implementation of structural policies by the Commission (Gil-RoblesSanter agreement), OJ (1999) C279/488;

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

60

 Decision of the European Parliament, the Council, the Commission, the Court of Justice, the Court of Auditors, the Economic and Social Committee and the Committee of the Regions of 20 July 2000 on the organisation and operation of the Office for Official Publications of the European Communities, OJ (2000) L183/12;  Agreement between Parliament and the Commission on procedures for implementing the Council decision of 28 June 1999 laying down the procedures for the exercise of implementing powers conferred on the Commission, OJ (2000) L256/19;  Joint declaration of the President of the European Commission and the President of the Committee of the Regions, CoR doc. DI CdR 81/2001 rev. 2 FR/OU/ss of 20 September 2001;  Joint declaration of the President of the European Commission and the President of the EESC, EESC doc. R/CES 1004/2001 rev. Item 9Ba) Appendix FR/ET/ht of 24 September 2001;  Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121/122;  Joint declaration relating to Regulation (EC) No 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, OJ (2001) L173/5;  Statement on Common Foreign and Security Policy expenditure charged to the general budget, OJ (2001) C197/339;  Statement on the provision of information to the budgetary authority Statement on Common Foreign and Security Policy expenditure charged to the general budget, OJ (2001) C197/339;  Pilot projects and preparatory actions adopted according to point 37(a) of the Interinstitutional Agreement of 6 May 1999 on budgetary discipline and improvement of the budgetary procedure, OJ (2001) C197/339;  Interinstitutional Agreement on financial statements, OJ (2001) C197/354;  Agreement between the European Police Office (EUROPOL) and the European Central Bank (ECB), OJ (2002) C23/9;  Decision of the European Parliament and of the Council of 13 December 2001 on the mobilisation of the flexibility instrument, OJ (2002) L53/28;  Interinstitutional agreement on the financing of the Convention on the Future of the European Union, OJ (2002) C54/1, extended by OJ (2002) C320/1;  Interinstitutional Agreement of 28 November 2001 on a more structured use of the recasting technique for legal acts, OJ (2002) C77/1;  Joint declaration by the European Parliament, the Council and the Commission (2003 Budget), OJ (2002) CE177/270;  Joint declaration concerning the implementation profile for the 2002 budget, OJ (2002) CE177/270;  Joint declaration concerning an action plan to put an end to abnormal outstanding commitments, OJ (2002) CE177/271;  Decision of the European Parliament, the Council, the Commission, the Court of Justice, the Court of Auditors, the Economic and Social Committee, the Committee of the Regions and the Ombudsman establishing a European Communities Personnel Selection Office, OJ (2002) L197/53;  Decision of the Secretaries-General of the European Parliament, the Council and the Commission, the Registrar of the Court of Justice, the Secretaries-General of the Court of Auditors, the Economic and Social Committee and the Committee of the Regions, and the Representative of the European Ombudsman of 25 July 2002 on the organisation and operation of the European Communities Personnel Selection Office, OJ (2002) L197/56;  Interinstitutional Agreement concerning access by the European Parliament to sensitive information of the Council in the field of security and defence policy, OJ (2002) C298/1;  Memorandum of understanding between the services of the European Parliament, the Council and the Commission on the application of Article 4(4) of Regulation (EC) 1049/2001 regarding public access to European Parliament, Council and Commission documents of 9 July 2002, not published;  Memorandum of Understanding concerning the signing of legislative texts adopted under the codecision procedure, Council doc. 15106/03 of 19 November 2003;  Decision of the European Parliament and of the Council of 19 May 2003 on the adjustment of the financial perspective for enlargement, OJ (2003) L147/25;  Decision of the European Parliament and of the Council of 19 May 2003 on the revision of the financial perspective, OJ (2003) L147/31;

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

61

 Statement by the European Parliament and the Council on the revision of the Agencies Financial Regulation, OJ (2003) L245/46;  Interinstitutional agreement on better law-making, OJ (2003) C321/1, corr. OJ (2004) C4/7;  Declaration of the European Parliament, Council and Commission on the financing of the common foreign and security policy in accordance with the Interinstitutional Agreement of 6 May 1999, OJ (2004) CE31/239;  Joint declaration of the European Parliament, Council and Commission on the Global Fund to fight HIV/AIDS, tuberculosis and malaria (European Parliament Proposal), OJ (2004) CE31/240;  Declaration on potentially abnormal RAL, OJ (2004) CE31/240;  Joint declaration of the European Parliament, Council and Commission concerning the implementation profile for the 2003 Budget, OJ (2004) CE31/243;  Joint declaration concerning financial consequences of enlargement, OJ (2004) CE31/246;  Declaration on budget procedure 2004, OJ (2004) CE82/468;  Declaration on the adaptation of reference amounts in co-decided legal bases following Enlargement, OJ (2004) CE82/469,  Declaration on the legal bases for subsidies previously under chapter A-30, OJ (2004) CE82/469,  Joint declaration on the EU 25 Budget, OJ (2004) CE91/615;  Declaration on the basic acts for grants (ex A-30), OJ (2004) CE91/618;  Joint declaration 3 annexed to the Resolution on the draft general budget of the European Union for the financial year 2005 as modified by the Council (all sections) and Letters of amendment 1/2005 and 3/2005 to the draft general budget of the European Union for the financial year 2005, OJ (2005) CE226/340;  Decision of the European Parliament and of the Council of 7 September 2005 on the revision of the financial perspective 2000 to 2006, OJ (2005) L269/24;  Better Regulation—Inter-Institutional Common Approach to Impact Assessment (IA)—Adoption, Council doc. 14901/05 of 24 November 2005;  Joint Declaration—Guidelines for legislative proposals related to the 2007-2013 multiannual financial framework, COBU News of 24 October 2005 at 6;  Joint Declaration on payment appropriations attached to the Resolution on the draft general budget of the European Union for the financial year 2006 as modified by the Council (all sections) and Amending Letters 1/2005 and 3/2005 to the draft general budget of the European Union for the financial year 2006, adopted on 15 December 2005;  Joint Declaration on CFSP, attached to the Resolution on the draft general budget of the European Union for the financial year 2006 as modified by the Council (all sections) and Amending Letters 1/2005 and 3/2005 to the draft general budget of the European Union for the financial year 2006, adopted on 15 December 2005;  Framework Co-operation Agreement between the European Parliament and the European Ombudsman, letter 10255 arrived at Parliament on 21 September 2005;  Decision of the European Parliament and of the Council of 17 November 2005 on the mobilisation of the European Union Solidarity Fund according to point 3 of the Interinstitutional Agreement of 7 November 2002 between the European Parliament, the Council and the Commission on the financing of the European Union Solidarity Fund supplementing the Interinstitutional Agreement of 6 May 1999 on budgetary discipline and improvement of the budgetary procedure, OJ (2006) L8/44;  Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125;  Interinstitutional agreement between the European Parliament, the Council and the Commission on budgetary discipline and sound financial management, OJ (2006) C139/1;  Decision of the European Parliament and of the Council of 27 April 2006 on the mobilisation of the EU Solidarity Fund according to point 3 of the Interinstitutional Agreement of 7 November 2002 between the European Parliament, the Council and the Commission on the financing of the European Union Solidarity Fund supplementing the Interinstitutional Agreement of 6 May 1999 on budgetary discipline and improvement of the budgetary procedure, OJ (2006) L154/20;  Statement by the Commission (re Article 7(3)), OJ (2006) C171/21;  Statement by the Commission (re Article 5 and Article 5a), OJ (2006) C171/21;  Statement by the European Parliament, the Council and the Commission concerning the Council Decision of 17 July 2006 amending Decision 1999/468/EC laying down the procedures for the exercise of implementing powers conferred on the Commission, OJ (2006) C255/1.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

62

legislation to refer to such agreements or to enact commitments made in them. This happens most often in the sphere of budget law.3 For example, in 1988 the Council noted in its Decision of 24 June 1988 concerning budgetary discipline the 1988 interinstitutional agreement on budgetary procedure and decided that the annual reference framework for the budget should take ‘account of the financial estimates of the Interinstitutional Agreement, for compulsory expenditure other than EAGGF Guarantee expenditure’. It furthermore decided that ‘[t]he budgetary discipline applicable to non-compulsory expenditure shall be ensured on the basis of the arrangements contained in the Interinstitutional Agreement’.4 In other areas, legislation has merely noted the existence of interinstitutional agreements as a fact to take into account.5

3

4 5

The Agreement between Eurojust and EUROPOL, Council doc. 15829/03 of 9 December 2003, the Co-operation agreement between Commission and Europol, press release in Commission doc. IP/03/241 of 18 February 2003 and the Charter of Fundamental Rights of the European Union, OJ (2000) C364/1 are border line cases. It is not clear to what extent the Commission agreed to the Collaboration entre le Conseil et la Commission sur la base de l´article 162 du Traité CEE, Bull. CE 1966/3 at 8 et seq. adopted during the special Council session of 28 and 29 January 1966; it seems that the Commission was at least not present during the meeting where the Council agreed on the text: Bull. CE 1966/3 at 5. At least in certain respects, the text led to interinstitutional agreements (see e.g. the 1966 interinstitutional convention listed above). In addition to the texts listed above, there are many interinstitutional texts of a political nature that influence substantive law. In this category fall, for example, the Joint Statement adopted at the end of the tripartite conference on employment and stability in the Community (Luxembourg, 24 June 1976), OJ (1976) C173/1; the Joint Declaration by the European Parliament, the Council and the Commission concerning the protection of fundamental rights and the European Convention for the protection of Human Rights and Fundamental Freedoms, OJ (1977) C103/1; the Joint statement by the Council, Commission and Parliament issued following the conciliation procedure on the reform of the ERDF, OJ (1985) C72/59; the Declaration by the European Parliament, the Council, the representatives of the Member States, meeting within the Council, and the Commission against racism and xenophobia, OJ (1986) C158/1; the Joint statement by the European Parliament, the Council and the Commission concerning Decision 819/95/EC establishing the Community action programme ‘Socrates’, OJ (1995) L132/18; the Declaration by the European Parliament, the Council and the Commission relating to lifts, OJ (1995) L213/32; the Joint declarations of the European Parliament, the Council and the Commission, OJ (1999) L155/6 and 12; the Interinstitutional agreement of 7 November 2002 between the European Parliament, the Council and the Commission on the financing of the European Union Solidarity Fund supplementing the Interinstitutional Agreement of 6 May 1999 on budgetary discipline and improvement of the budgetary procedure, OJ (2002) C283/1; the Joint statements entered in the Council minutes relating to the Merger Regulation, Council doc. 15449/03 ADD 1 of 12 December 2003 and the Joint statement by the Council and the representatives of the governments of the Member States meeting within the Council, the European Parliament and the Commission on European Union Development Policy: ‘The European Consensus’, OJ (2006) C46/1. These texts have little or no bearing on the separation of powers between the institutions and are therefore not discussed in this study. See for example the Final Adoption of amending and supplementary budget 1 of the European Communities for the financial year 1992, OJ (1992) L142/1. Other instances can be found in OJ (1992) L174/1, OJ (1992) L349/1, OJ (1993) L345/1, OJ (1994) L34/1, OJ (1994) L169/1, OJ (1994) L362/1, OJ (1995) C109/110, OJ (1995) C308/122, OJ (1996) L22/1 and OJ (1997) L44/1. Articles 14 and 15 of the Council Decision of 24 June 1988 concerning budgetary discipline, OJ (1988) L185/29. See for example Council Regulation 837/90 of 26 March 1990 concerning statistical information to be supplied by the Member States on cereals production, OJ (1990) L88/1. For similar formulas see, for example, OJ (1990) L117/28, OJ (1990) L185/1, OJ (1990) L347/1, OJ (1992) L364/32.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

63

This example—which can easily be multiplied with others—provokes the question as to the status of such interinstitutional agreements and their relationship with EU law.6 This matter comprises two big themes. Firstly, there is the question to what extent interinstitutional agreements are concluded on a legal basis. It is a fundamental principle of EU law that ‘[e]ach institution shall act within the limits of the powers conferred upon it’ by the Treaties.7 In practice, though, the situation is not always evident. For example, recital (7) of the Interinstitutional agreement of 22 December 1998 on common guidelines for the quality of drafting of Community legislation8 provides that it is to be regarded as an instrument ‘for internal use by the institutions’ and is ‘not legally binding’. For other texts baptised with the name of interinstitutional agreement, however, such a clear intention is absent and the obvious question arises on what legal basis the three institutions adopted it.9 Many of the interinstitutional agreements on budgetary issues fall within this category. Some interinstitutional agreements, such as the above-mentioned text on legal drafting, concern fairly prosaic subjects. However, other interinstitutional agreements, such as some of the budgetary texts, go far in regulating constitutional relations. The Court of Justice noted in France vs Commission that ‘the binding nature of any act intended to have legal effects must be derived from a provision of Community law which prescribes the legal form to be taken by that act and which must be expressly indicated therein as its legal basis.’10 This applies also to ‘negotiated acts’.11 Although this position has been criticised as ‘extreme’,12 it makes perfect sense in the context of a separation of institutions system. I will therefore in sub-sections 3.2.2 and 3.2.3 examine to what extent interinstitutional agreements have a legal basis. Secondly, the extent to which the content of interinstitutional agreements enables their application in law or, to be more precise, enables them to be invoked before the courts, is analysed in sub-section 3.2.4. 3.2.2

Specific legal bases for interinstitutional agreements

In some instances, primary or secondary law provides for specific legal bases for interinstitutional agreements.13 It is difficult to present an exhaustive list, but the 6 7

8 9 10 11 12 13

Debard, ‘Textes interinstitutionnels, constitutionnalisation et révision des Traités’ (2001) at 100 simply bases them on Article 10 TEC. This, however, is much less straight-forward than it would seem, as is discussed in sub-section 3.2.2; below. Articles 7(1) TEC and 19(2) Const. The consequences of a lacking legal basis for the validity of interinstitutional agreements seem to have been missed by Eiselt and Slominski, ‘Sub-constitutional engineering: negotiation, content, and legal value of interinstitutional agreements in the EU’ (2006), 209 et seq. OJ (1999) C73/1. Perhaps the question is not that obvious; not many authors posed it directly. Exceptions are Bobbert, Interinstitutionelle Vereinbarungen im Europäischen Gemeinschaftsrecht (2001) at 3 and Schwarze, ‘Möglichkeit und Grenzen interinstitutioneller Vereinbarungen nach Maastricht’ (1995) at 50 et seq. Case C-325/91, France vs Commission, [1993] ECR I-3283, § 26. Ibid., § 28. Evans, A textbook on EU law (1998) at 33. Surprisingly, Tournepiche, ‘La clarification du statut juridique des accords interinstitutionnels’ (2002) at 215 believes that the Declaration adopted in Nice on Article 10 of the Treaty establishing

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

64

following examples should give a reasonably comprehensive overview of the types of provisions giving rise to interinstitutional agreements, and of some cases that do not. The first provision of primary law to be mentioned is not an example. Article 161 TEC, added by the Treaty of Nice, explicitly refers to an interinstitutional agreement: ‘From 1 January 2007, the Council shall act by a qualified majority on a proposal from the Commission after obtaining the assent of the European Parliament . . . if, by that date, the multiannual financial perspective applicable from 1 January 2007 and the Interinstitutional Agreement relating thereto have been adopted . . .’ Since 1988, the multi-annual financial perspective has been part of the 1988, 1993, 1999 and 2006 budget interinstitutional agreements. Although, according to some, these interinstitutional agreements had legal force,14 this could certainly not rest on Article 161, which was only inserted with the Treaty in Nice and which speaks only of the situation from 2007. Moreover, that provision takes the multi-annual financial perspective as a fact, rather than providing a legal basis for it. The fifth paragraph of Article 272(9) TEC sets out a special type of interinstitutional agreement: the Council and the Parliament ‘may’ agree to a different maximum rate of increase of expenditure for the budget. This is a rare case of primary law attaching procedural requirements to the conclusion of an interinstitutional agreement.15 In 1986 the Court of Justice annulled the decision of the president of Parliament to declare the budget adopted, since such an agreement was absent.16 Another special type of interinstitutional agreement is foreseen in Article 248(3) TEC:17 ‘In respect of the European Investment Bank’s activity in managing Community expenditure and revenue, the Court’s rights of access to information held by the Bank shall be governed by an agreement between the Court [of Auditors], the Bank and the Commission . . .’ Although the EIB is no institution, the Commission and the Court of Auditors of course are. An Article 248 agreement was signed in 1992 and renewed in 1999.18

14 15

16 17 18

the European Community, OJ (2001) C80/77 constituted the first ‘prise en compte “officielle” de l’existence des accords interinstitutionnels’. As shown in the list of specific legal basis in this subsection, far from it! As early as 1989 this view was heard: Timmann, ‘Das Haushaltsverfahren 1989—Erste Erfahrungen mit der Interinstitutionellen Vereinbarung über die Haushaltsdisziplin’ (1989) at 28. The financial perspective is discussed in more detail in section 14.6, below. Article 272(9) TEC provides in its relevant part that: ‘[w]here the European Parliament, the Council or the Commission consider that the activities of the Communities require that the rate determined according to the procedure laid down in this paragraph should be exceeded, another rate may be fixed by agreement between the Council, acting by a qualified majority, and the European Parliament, acting by a majority of its Members and three fifths of the votes cast’. The procedure in Parliament is regulated in Article 2 of Annex IV to its Rules of Procedure. Case 34/86, Council vs Parliament (Budget), [1986] ECR 2155. The provision recurs with slight textual changes as Article III-384.3 Const. See currently the Tripartite agreement between the European Investment Bank, the Commission and the Court of Auditors (not published) of 19 March 1999.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

65

Besides these, some authors have proposed several other primary law legal bases. Some19 believe that Articles 193 and 195 TEC20 constitute such legal bases, but this requires some qualification. Article 195(4) TEC lays down that ‘[t]he European Parliament shall, after seeking an opinion from the Commission and with the approval of the Council acting by a qualified majority, lay down the regulations and general conditions governing the performance of the Ombudsman’s duties.’ The Treaties require here the approval of the Council, but the decision is Parliament’s. The outcome of the procedure21 is a Decision which can be equally viewed as an interinstitutional agreement or as a sui generis act. Article 193 TEC provides that ‘[t]he detailed provisions governing the exercise of the right of inquiry shall be determined by common accord of the European Parliament, the Council and the Commission’. The outcome of this provision was a Decision of the three political institutions on provisions governing the right of inquiry,22 which again has the form and content of a Decision. Whether or not this sui generis act is an interinstitutional agreement is in this case too a theoretical question. The bottom line is that it is legally binding.23 Article 252(g) TEC provides for the possibility for the Council and Parliament to extend some of the deadlines in the co-operation procedure with one month. This procedure is discussed in more detail in sub-section 11.2.1, below. Lastly on primary law, the Protocol on the application of the principles of subsidiarity and proportionality adopted in Amsterdam ‘tak[es] account’ of the ‘Interinstitutional Agreement of 25 October 1993 between the European Parliament, the Council and the Commission on procedures for implementing the principle of subsidiarity’. That does not provide the latter with a legal basis; it rather enacts at primary law level the commitments it contains. As far as secondary law is concerned, references to interinstitutional agreements occur regularly, as for example in the following cases:  Council Regulation 610/90 inserted a reference to the 1982 Joint Declaration into the 1977 Financial Regulation.24 Moreover, in 1990 a new Article 3(3) was

19

20 21 22 23

24

As regards the TEC predecessors of these provisions: Bobbert, op. cit. footnote 9 at 53-55, Monar, ‘Interinstitutional agreements: the phenomenon and its new dynamics after Maastricht’ (1994) at 697, Reich, ‘La mise en œuvre du Traité sur l’Union européenne par les accords interinstitutionnels’ (1994) at 81 and Rideau, Droit institutionnel de l’Union et des Communautés européennes (2002) at 171. Which correspond to Articles III-333 and III-335 Const. Decision of the European Parliament of 9 March 1994 on the regulations and general conditions governing the performance of the Ombudsman’s duties, OJ (1994) L113/15. Decision of the European Parliament, the Council and the Commission of 6 March 1995 on the detailed provisions governing the exercise of the European Parliament’s right of inquiry, OJ (1995) L113/1. Similar logic applies to the Decision 1247/2002/EC of the European Parliament, of the Council and of the Commission of 1 July 2002 on the regulations and general conditions governing the performance of the European Data-protection Supervisor’s duties, OJ (2002) L183/1, which is required by Article 43 of Regulation 45/2001 of the European Parliament and of the Council of 18 December 2000 on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data, OJ (2001) L8/1. Article 1(28) of Council Regulation 610/90 of 13 March 1990 amending the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities, OJ (1990) L70/1.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION







25 26

27 28 29 30 31

66

inserted,25 requiring the legislature in cases of multi-annual operations to take account of the ‘financial perspective’ annexed to the 1988 interinstitutional agreement on budgetary procedure. In addition, Article 127 of the 1977 Financial Regulation26 comprised an implicit reference to the 1975 Joint Declaration of the European Parliament, the Council and the Commission concerning the institution of a conciliation procedure between Parliament and Council.27 Furthermore, Article 39 of the same Financial Regulation provided for the possibility that the institutions agree to use shortened deadlines for the adoption of the budget. Such a ‘pragmatic timetable’ was indeed agreed; it is discussed in section 14.2, below. In the recitals of the Europe against AIDS programme the Council considered it necessary that the amount of the Community financial resources ‘must be included in the financial perspective defined in the interinstitutional Agreements’.28 Contrary to the reference in Article 161 EC discussed above, there is here an active requirement. However, this occurs in the recitals rather than the operative provisions of the act and therefore does not constitute a legal basis.29 Recital 9 of the 2001 Access to Documents Regulation30 specifically calls for an interinstitutional agreement and Article 9(7) of the same Regulation provides for a corresponding legal basis by requiring that ‘[t]he Commission and the Council shall inform the European Parliament regarding sensitive documents in accordance with arrangements agreed between the institutions’. Under Article 15(2) of the same Regulation, an interinstitutional committee must be established ‘to examine best practice, address possible conflicts and discuss future developments on public access to documents’. Presumably addressing conflicts implies the power to solve them and take the necessary decisions. Article 2 of the Staff Regulations31 provides that ‘one or more institutions may entrust to any one of them or to an inter-institutional body the exercise of some or all of the powers conferred on the Appointing Authority, other than decisions relating to appointments, promotions or transfers of officials’. On the legal basis of this provision Parliament, the Council, the Commission, the Court of Justice, the Court of Auditors, the Economic and Social Committee, the Committee of the Regions and the Ombudsman agreed to found an office for the recruitment of EU Article 6 of Council Regulation 610/90 of 13 March 1990 amending the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities, OJ (1990) L70/1. Inserted by Regulation 610/90, which provided that ‘[a]t three-year intervals, the European Parliament and the Council shall examine this Financial Regulation in the light of a proposal from the Commission. Any financial regulations amending this Financial Regulation shall be adopted by the Council after recourse to the conciliation procedure if the European Parliament so requests’. The current text is Article 184 of Council Regulation 1605/2002 of 25 June 2002 on the Financial Regulation applicable to the general budget of the European Communities, OJ (2002) L248/1. OJ (1975) C89/1. OJ (1991) L175/26. Similarly, the Council Decision of 24 June 1991 providing medium-term financial assistance for Bulgaria pre-empted the conclusion of a successor to the 1988 Budget agreement without providing a legal basis: OJ (1991) L174/36. Similar references are easily found. Regulation 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, OJ (2001) L145/43. Staff Regulations of officials of the European Communities, and to the Conditions of employment of other servants of the European Communities as laid down by Regulation 259/68, OJ (1968) L56/1, as last amended by OJ (2004) L124/1.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

67

staff.32 This agreement was itself the legal basis for an agreement between the secretaries-general of those institutions.33  A footnote to Article 11(2) of the Decision of the European Parliament of 9 March 1994 on the regulations and general conditions governing the performance of the Ombudsman’s duties34 provides for a ‘joint statement by the three institutions’ on the Ombudsman’s staff. This is a unique instance of a Parliament Decision providing the legal basis for an interinstitutional agreement. Parliament’s willingness to involve the Council is less spectacular if it is remembered that under Article 195(4) TEC, which is the legal basis for the Decision, Parliament needed the Council’s approval for it.  The Regulation establishing the European Maritime Safety Agency requires that body to accede to the OLAF interinstitutional agreement.35 A virtually similar text features in Article 50(2) of the Regulation establishing the European Aviation Safety Agency.36 Article I-26.1 Const would have provided inter alia that the European Commission ‘shall initiate the Union’s annual and multiannual programming with a view to achieving interinstitutional agreements’. This, however, does not in itself say anything about the legal validity of such agreements and cannot serve itself as a legal basis. Lastly, a declaration was attached to the Treaty of Amsterdam in which:37 ‘The Conference considers that the three institutions involved in the procedure for adopting Community legislation, the European Parliament, the Council and the Commission, should lay down guidelines on the quality of drafting of the said legislation . . . Therefore, the Conference declares that the European Parliament, the Council and the Commission ought to . . . establish by common accord guidelines for improving the quality of the drafting of Community legislation . . . ‘38 Declarations attached to the Treaties have, however, no separate legal binding force, and consequently can hardly be a legal basis. This motley assembly of instances covers many interinstitutional agreements. However, many more lack any such specific legal basis. The question thus arises to what extent such interinstitutional agreements find a general basis in EU law. 32 33

34 35 36 37 38

Decision of the European Parliament, the Council, the Commission, the Court of Justice, the Court of Auditors, the Economic and Social Committee, the Committee of the Regions and the Ombudsman establishing a European Communities Personnel Selection Office, OJ (2002) L197/53. Decision of the Secretaries-General of the European Parliament, the Council and the Commission, the Registrar of the Court of Justice, the Secretaries-General of the Court of Auditors, the Economic and Social Committee and the Committee of the Regions, and the Representative of the European Ombudsman of 25 July 2002 on the organisation and operation of the European Communities Personnel Selection Office, OJ (2002) L197/56. OJ (1994) L113/15. Article 20(2) of Regulation 1406/2002 of the European Parliament and of the Council of 27 June 2002 establishing a European Maritime Safety Agency, OJ (2002) L208/1. Regulation 1592/2002 of the European Parliament and of the Council of 15 July 2002 on common rules in the field of civil aviation and establishing a European Aviation Safety Agency, OJ (2002) L240/1. Declaration 39 on the quality of the drafting of Community legislation annexed to the Treaty of Amsterdam, OJ (1997) C340/139. See also the considerations of the Protocol on the application of the principles of subsidiarity and proportionality, OJ (1997) C340/105.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

3.2.3

68

General legal bases for interinstitutional agreements

Although one could argue that interinstitutional agreements would find their bases in logic39 and common sense, lawyers are left with the difficulty that each institution must act within the limits of the powers conferred upon it (Articles 7(1) TEC and I19.2 Const). Apart from the special legal bases noted in the previous sub-section, the Treaties and the Constitution contain several provisions that are of interest. I will review the possible general legal bases under the following headings: I. Case law; II. Article 218 TEC; III. ‘Self binding’ of institutions; IV. Article 6(4) TEU; V. Articles 7(1) TEC; VI. Parliament’s Rules of Procedure; and VII. General principles of law. I. The Court of Justice’ case law can be separated in three different strands.40 Firstly, in Luxembourg vs Parliament the Court commented that Parliament’s practice to hold half of its plenary sessions in Luxembourg ‘had been decided upon by the Parliament of its own motion and had never been approved either expressly or by implication by the Member States. On the contrary the French government several times denied that the practice was compatible with the Decisions of the Member States and requested that it be changed.’41 Thus far little news: the analysis seemed to fit with the background of the rule that the Member States decide the seat of Parliament. However, the Court continued that ‘[t]he Luxembourg government is therefore wrong in alleging that the practice had created a custom in its favour supplementing the Decisions of the Member States in the matter and requiring the Parliament to hold part of its plenary sittings in Luxembourg.’ This dictum is more difficult to understand. The Court does not deny that a ‘custom’ of Parliament approved by the Member States can supplant positive law. It merely states that in fact this did not happen. Had France not protested, then the Court might have found this ‘custom’ to bind Parliament and the Member States—even though the decision to determine Parliament’s seat was the Member States’ alone.42 The key word in the first quotation thus becomes ‘approved’, which suggests agreement. Under the influence of international law, scholars in the past sometimes argued that customary law has a place in the Community’s legal system, provided this does not contradict written primary and secondary law.43 However, in my view, (the current) Article 7 TEC militates against the recognition of ‘customary law’ within the European Union’s constitutional system. In any event, Luxembourg vs Parliament was ruled against the failure of the Member States to discharge their obligation to 39 40 41 42 43

And this is almost what Monar, op. cit. footnote 19 at 700 has to resort to, together with relying on the mutual co-operation obligations. In the view of Schwarze, op. cit. footnote 9 at 51, case law goes little beyond ‘eine Reihe von Hinweisen auf die prinzipielle Zulässigkeit interinstitutioneller Vereinbarungen’. As the analyses on these pages may show, I believe the judges went beyond that. Case 230/81, Luxembourg vs Parliament, [1983] ECR 255, § 44. At the time Articles 77 ECSC, 216 TEEC and 189 Euratom. Läufer, ‘Das Europäische Parlament nach der Direktwahl: Positionsstärkung durch intrakonstitutionellen Wandel?’ (1979) at 271 and the literature mentioned there.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

69

adopt a final arrangement for the seat of Parliament and the importance of the judgment for our topic should not be exaggerated. The second line of case law started with a problem in the budgetary sphere. In Council vs Parliament (Budget 1986) the validity of interinstitutional agreements was not directly addressed. After the Court had struck down the decision by the president of Parliament declaring that the budget for 1986 had been adopted, the judges noted that ‘the problems regarding the delimitation of non-compulsory expenditure in relation to compulsory expenditure are the subject of an interinstitutional conciliation procedure set up by the Joint Declaration of the European Parliament, the Council and the Commission of 30 June 1982 . . . and . . . they are capable of being resolved in that context.’44 That did not imply that the Joint Declaration45 had legal value,46 although it was a legal recognition of the existence of that text. Two years later this was followed by Greece vs Council. The Court’s reasoning in that judgment is better explained from a historical than from a logical point of view. The question arose whether the budgetary authority was empowered to transfer a financial aid amount of ECU 10 million from one budgetary heading to another. According to Article 21(2) of the Financial Regulation at the time, the precise procedure for doing this depended on the question whether the amount constituted ‘compulsory’ or ‘non-compulsory’ expenditure. The Council relied on a provision in the 1982 Joint Declaration that classifying the aid as compulsory. This brought up the question whether the Declaration was legally determinative for the classification. The Court started by pointing out that the scope of those two classes of expenditure had given rise to budgetary crises between Parliament and Council. These had been resolved by the 1982 Joint Declaration. The judges recalled the dictum in the Budget 1986 case quoted above and noted that ‘the operation of the budgetary procedure, as it is laid down in the financial provisions of the Treaty, is based essentially on inter-institutional dialogue. That dialogue is subject to the same mutual duties of sincere cooperation which, as the Court has held, govern relations between the Member States and the Community institutions (see the judgment of 10 February 1983 in Case 230/81 Luxembourg v European Parliament . . .)’47 This dictum forces to pause and reflect. Luxembourg vs Parliament—discussed above in another context—concerned a resolution of the Assembly concerning the place where it performed its activities. The Court upheld that resolution with the logic that Article 10 TEC (then Article 5 TEEC) lays down a ‘rule imposing on Member States and the Community institutions mutual duties of sincere co-operation’.48 However, the letter of the provision merely provides for a one-way obligation: 44 45 46 47 48

Case 34/86, Council vs Parliament (Budget 1986), [1986] ECR 2155, § 50. Joint Declaration by the European Parliament, the Council and the Commission of 30 June 1982 on various measures to improve the budgetary procedure, OJ (1982) C194/1. Similarly cautious Glaesner, ‘Haushaltsverfahren der Europäischen Gemeinschaften, Befugnis des Europäischen Parlaments zur Erhöhung der nichtobligatorischen Ausgaben’ (1987) at 161. Case 204/86, Greece vs Council, [1988] ECR 5323, § 16. Luxembourg vs Parliament, § 37.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

70

‘Member States shall take all appropriate measures, whether general or particular, to ensure fulfilment of the obligations arising out of this Treaty or resulting from action taken by the institutions of the Community. They shall facilitate the achievement of the Community’s tasks. They shall abstain from any measure which could jeopardise the attainment of the objectives of this Treaty.’ In Luxembourg vs Parliament the Court proceeded to turn the obligation around in what effectively constitutes a creation of constitutional law ex nihilo: ‘in accordance with the above-mentioned mutual duties of sincere cooperation, the Decisions of the Parliament in turn must have regard to the power of the governments of the Member States’.49 Back to Greece vs Council. The Court pointed out that the institutions possess a discretionary power as far as the classification of expenditure is concerned. This discretion is limited by ‘the separation of powers, as laid down in the Treaty, between the institutions’.50 The Court implicitly noted that this discretion should also not be exercised in an arbitrary manner. It then noted that the institutions had classified the expenditure concerned as compulsory. And, by accepting that classification, it implicitly recognised the legal validity of the 1982 Joint Declaration which laid it down. Hence, through a singular revolution, the Court read out of the obligation of Member States to co-operate with the Community the right of institutions to conclude agreements with each other.51 This creative reasoning was perhaps necessary but leaves the uncomfortable feeling that the judges were forced to do the constitutional legislator’s work in creating the underlying principle of loyal co-operation. In my humble opinion, this slender legal foundation hardly supports the weight of interinstitutional agreements,52 some of which are the substantive equivalent of national (organic) laws. Consider, for example, the interinstitutional agreement on the European Union Solidarity Fund.53 Whilst the Fund itself is based on a legislative act,54 the agreement comprises a budgetary ceiling of one billion euro. That, however, is a matter 49 50 51

52

53

54

Ibid., § 38. Greece vs Council, § 17. In the words of Fletcher, ‘Extending “indirect effect” to the third pillar: the significance of Pupino?’ (2005) at 871 et seq., the Court has been similarly ‘rather inventive’ in extending the duty of Member States to loyal co-operation to the EU Treaty: case C-105/03, Pupino, [2005] I-5285. Maurer, Kietz and Völkel, ‘Interinstitutional agreements in the CFSP: parliamentarization through the back door?’ (2005) at 177 somewhat optimistically conclude that ‘Article 10 TEU [sic], which obliges national institutions and European institutions to cooperate loyally, is commonly interpreted as an implicit legal basis for the conclusion of IIAs’. Kapteyn and VerLoren van Themaat (ed. Gormley), Introduction to the law of the European Communities (1998) at 158 argue that the current Article 10 TEC is the expression of such a broader fundamental principle of EU law. However, this benevolent reading, however useful to the institutions, seems hard to combine with the rigour of Article 7(1) TEC. Interinstitutional agreement of 7 November 2002 between the European Parliament, the Council and the Commission on the financing of the European Union Solidarity Fund supplementing the Interinstitutional Agreement of 6 May 1999 on budgetary discipline and improvement of the budgetary procedure, OJ (2002) C283/1. Council Regulation (EC) 2012/2002 of 11 November 2002 establishing the European Union Solidarity Fund, OJ (2002) L311/3.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

71

of policy affecting third parties. Whilst understanding the need for the Fund, it is difficult to accept that a ceiling to protect the Member States’ interests can be based on a provision requiring them to support the work of the European Community. Lastly, there are academic views suggesting that the Court’s case law cannot be generalised to all interinstitutional agreements55 and indeed, in the following Chapters many interinstitutional agreements will be encountered which clearly cannot have been intended to be legally binding. II. The third strand of case law must be discussed together with Article 218(1) TEC.56 In the early 1990s, the FAO accepted the Community as a Member. On 19 December 1991 the Council and the Commission concluded an ‘Arrangement regarding preparation for FAO meetings, statements and voting’.57 Under this Arrangement, the Commission and the Member States were to co-ordinate their actions in the FAO. The Arrangement gave rise to a judgment in which the Court of Justice declared that ‘[i]t is clear . . . from the terms of the Arrangement, that the two institutions intended to enter into a binding commitment towards each other . . . Consequently, by concluding that the draft Agreement concerned an issue whose thrust did not lie in an area within the exclusive competence of the Community and accordingly giving the Member States the right to vote for the adoption of that draft, the Council acted in breach of section 2.3 of the Arrangement which it was required to observe.’58 And the Court ruled the matter accordingly. Casual readers could be mistaken to assume that the legal basis of the Arrangement was based on the intention of the two institutions to commit themselves in a binding manner. Remarkably, neither the Court, nor Advocate-General Jacobs even mentioned (the current) Article 218(1) TEC. Nevertheless, this Article, which states that ‘[t]he Council and the Commission shall consult each other and shall settle by common accord their methods of co-operation’, provides for a general basis for interinstitutional agreements between these institutions.59 Does Article 218(1) also allow for agreements including Council, Commission and Parliament? Or including one of the two former and the Parliament? Van Craeyenest defends the view that this provision (then Article 15 of the Merger Treaty) ‘peut d’ailleurs être regardé comme l’expression spécifique d’un

55 56

57

58 59

Bobbert, op. cit. footnote 9 at 4. This provision started its life as 162 TEEC, after which it became Article 15 of the Merger Treaty. It corresponds to III-397 Const. Its TEU pendant is Article 3 of that Treaty (Article III-292.3 Const), which lays down that ‘[t]he Council and the Commission shall be responsible for ensuring . . . consistency and shall cooperate to this end. They shall ensure the implementation of these policies, each in accordance with its respective powers.’ See also Article III-423 Const (co-operation in the field of enhanced cooperation). That this provision can serve as a legal basis was determined in case 22/70, Commission vs Council (ERTA), [1971] ECR 263, § 87. ‘A’ item note from COREPER to the Council—Accession of the Community to the FAO— Arrangement between the Council and the Commission regarding preparation for FAO meetings, statements and voting, Council doc. 10478/91 of 18 December 1991. It is discussed in more detail in sub-section 15.3.5, below. Case C-25/94, Commission vs Council (FAO fisheries agreement), [1996] ECR I-1469, §§ 49-50. Indeed, this was also the opinion of Advocate General Mancini in case 204/86, Greece vs Council, [1988] ECR 5323. See also the literature mentioned there.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

72

principe plus général de collaboration loyale entre les institutions de la Communauté’.60 This view is problematic because the Courts never openly embraced this ‘general principle’ theory. Moreover, it raises questions as to the democratic legitimacy of law created outside the procedures specifically foreseen by the Treaties. Additionally, it offends the clear wording of Article 7 TEC. Lastly, the question remains what the status is of interinstitutional agreements concluded with the EESC and the Committee of the Regions,61 or with the Member States.62 In summary, I believe that Article 218(1) TEC remains an imperfect legal basis for interinstitutional agreements concluded with institutions other than the Council and the Commission. In any event, the FAO Arrangement fell exactly within its terms and, notwithstanding the Court’s silence in this respect, one could well argue that it was based on that provision. Therefore, the scope of the Court’s judgment should not be extrapolated to interinstitutional agreements between other institutions than the Commission and the Council. Perhaps this is the reason why the Declaration attached to the Treaty of Nice63 solidly follows the Greece vs Council case law, whilst completely ignoring the intention criterion of FAO. Lastly, Article III-397 Const would rephrase Article 218(1) by providing that ‘[t]he European Parliament, the Council and the Commission shall consult each other and by common agreement make arrangements for their cooperation. To that end, they may, in compliance with the Constitution, conclude interinstitutional agreements which may be of a binding nature.’ The addition of Parliament would resolve one of the problems noted above. The ‘may be’ in the second sentence indicates that interinstitutional agreements can be binding. The text leaves it open whether this ‘binding nature’ implies legally binding force (considering its mention in the Constitution, presumably it does). Moreover, the provision explicitly recognises that interinstitutional agreements may not have (legally) binding value. In summary, the addition adds little to the existing law on interinstitutional agreements since it does not clarify when an interinstitutional agreement has legally binding force. A fortiori, the provision remains silent on agreements that include other institutions than the three mentioned in it, and does not address the hierarchy between interinstitutional agreements that are purportedly legally binding with other acts of secondary law. A reasonable interpretation of Article III-397 would be to restrict the ‘binding nature’ category to interinstitutional agreements that are foreseen by the Constitution or by secondary law. 60 61

62

63

‘Article 211’ (1987) at 10. Very similar De Witte, ‘Zaak C65/93, Europees Parlement t. Raad’ (1997) at 239. E.g. the 1999 Cooperation Agreement between the Economic and Social Committee and the Committee of the Regions (not published), the Joint declaration of the President of the European Commission and the President of the Committee of the Regions, CoR doc. DI CdR 81/2001 rev. 2 FR/OU/ss of 20 September 2001 and the Joint declaration of the President of the European Commission and the President of the EESC, EESC doc. R/CES 1004/2001 rev. Item 9Ba) Appendix FR/ET/ht of 24 September 2001. For example, the Code of Conduct for negotiations within the WTO which was negotiated between the Council, the Member States and the Commission and which explicitly involves the Member States: ‘A’ Item note from COREPER to the Council on the WTO—code of conduct, Council doc. 7926/95 of 8 June 1995. OJ (2001) C80/77. The Declaration is analysed in more detail in sub-section 3.4.6.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

73

III. An argument could be attempted that the Courts’ case law provides for the possibility of ‘self binding’ of an institution on the basis of the principle of patere legem quam ipse fecisti. This would imply the possibility for third parties to exact the respect of commitments taken upon the institution by its own volition. In such logic, an interinstitutional agreement is a combination of the unilateral commitments of two or more institutions. However, on closer inspection, this logic misreads the case law. One the one hand, in Carvel the Court of First Instance found that the Council had violated its own unilateral Council Decision. However, that Decision rested on the solid basis of the Council’s Rules of Procedure64 and was, therefore, indeed of an unmistakably legal nature. Academia65 points to WWF,66 in which the Court of First Instance at least gave the impression that the binding nature of Commission Decision 94/90 on access to documents67 rested on the ‘self binding’ of that institution. It is, however, respectfully submitted that both the judges and academia overlooked the fact that that Decision in its recitals clearly lists the former Article 162 (currently 218) TEC. In other words, the Decision was directly based on the EC Treaty and the deus ex machina of a ‘principle of self binding’ seems hardly necessary. IV. Article 6(4) TEU provides that ‘[t]he Union shall provide itself with the means necessary to attain its objectives and carry through its policies’. This rather vague provision, occurring among the ‘Common provisions’ of the Treaty on European Union, must probably be interpreted in conjunction with the principle of Article 7 TEC. It is, however, doubtful whether the words ‘shall provide itself’ allow a reading as ‘is empowered’. In any event, its equivalent in Article I-54.1 Const is placed in the provision on the Union’s own resources, limiting the word ‘means’ to the EU’s financing and thereby barring it from being a potential legal basis for interinstitutional agreements. V. I already mentioned Article 7(1) TEC above. That provision lays down the important principle that ‘[e]ach institution shall act within the limits of the powers conferred upon it by this Treaty.’ Its intended successor Article I-19.2 Const would continue by admonishing the institutions to ‘practice full mutual co-operation’.68 Would Article I-19.2 convey any interinstitutional co-operation with the stamp of law? I believe that such a conclusion cannot be drawn from it. The second phrase merely seeks to codify the case law discussed above. Note that, in contrast to Article III-397 Const (discussed above under II), Article I-19.2 does not mention anything concerning the binding nature of interinstitutional co-operation. VI. Parliament is the only institution69 that provided for interinstitutional agreements in its Rules of Procedure. Rule 120(1) clarifies that such agreements must be concluded with ‘other institutions in the context of the application of the Treaties or in order to improve or clarify procedures’. The Rule makes it clear that such

64 65 66 67 68 69

Case T-194/94, Carvel and Guardian vs Council, [1995] ECR II-2765, § 65 et seq. The Council pointed out that the Decision rested on its Ruled of Procedure: ibid., § 47. Lenaerts and Desomer, ‘Bricks for a Constitutional Treaty of the European Union: values, objectives and means’ (2002) at 399-400. Case T-105/95, WWF UK vs Commission, [1997] ECR II-313, §§ 53-55. Commisison Decision 94/90 of 8 February 1994 on public access to Commission documents, OJ (1994) L46/58. Since the Treaty of Amsterdam, Article 3 TEU provides for a similar duty for Commission and Council to co-operate. The Committee of the Regions, which is not an institution, adopted a similar legal basis in Rule 66 of its Rules of Procedure, OJ (2004) L175/1.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

74

agreements may take the form of joint declarations, exchanges of letters or codes of conduct or other appropriate instruments. As far as Parliament is concerned, this somewhat drains the legal morass. Since institutions are under a duty to comply with procedural rules laid down in their Rules of Procedure,70 Parliament arguably is bound to interinstitutional agreements concluded on the basis of Rule 120. The obvious question remains, however, how the Rules of Procedure of one institution could bind another one. In addition, the provision raises other issues. Parliamentary rapporteur Mr Corbett noted that if interinstitutional agreements ‘deal with matters which, in the absence of the agreement, would have been dealt with in the Rules of procedure themselves, they have to be transposed into the Rules in a way which satisfies the requirements of Article 199 of the EC-Treaty (adoption by a majority of its Members) . . .’71 The criterion for determining whether a matter ‘would have been dealt with in the Rules of procedure themselves’ presumably is whether Parliament can deal with it on its own; Parliament cannot bind other institutions through its Rules of Procedure. A problem with Corbett’s interpretation remains that it presumes that interinstitutional agreements are formalised. Although this is often the case, it need not be, as the Luns and Westerterp procedures and the Gentlemen’s Agreement on the institutions’ budgets72 clearly show. VII. General principles of law. Schwarze assumes the existence of a ‘right of self-organisation’ that would allow institutions to conclude interinstitutional agreements. In his view, the provisions allowing the institutions to adopt their rules of procedure are only declaratory.73 I argued in sub-section 2.4.4 that indeed, as a general principle and a meta-constitutional requirement, the institutions should be autonomous. However, this autonomy is not unlimited; especially not when a measure affects third parties.74 Interinstitutional agreements may often have such effects for third institutions75 or citizens.76 I therefore have my doubts whether this ‘general right of self-organisation’ can be a basis for interinstitutional agreements, certainly when these go beyond mere administrative co-operation. An argument has been attempted77 to construe interinstitutional agreements as resting on the principle of legitimate confidence that the Court of Justice enunciated in Commission vs Council (Salaries).78 In my view, however, this reasoning hits the same snag as the ‘self-organisation’ logic: legitimate confidence cannot go further than primary law allows, or impair the rights of other institutions. Moreover, it 70 71 72 73 74 75 76 77 78

Case 68/86, United Kingdom vs Council, [1988] ECR 855, § 48. Draft report on the General Revision of the Rules of Procedure, Parliament doc. PE 294.770 of 8 June 2001 at 14. Discussed in sections 15.3.2, 15.3.4 and 14.3, respectively. Schwarze, op. cit. footnote 9 at 53. The self-organisation view is shared by Rideau, op. cit. footnote 21 at 169 and by Läufer, op. cit. footnote 43 at 273. In this sense also Hilf, ‘Die rechtliche Bedeutung des Verfassungsprinzips der parlamentarischen Demokratie für den europäischen Integrationsprozeß’ (1984) at 20-21. The Council’s vehement protest against the 2000 Framework Agreement concluded between Commission and Parliament springs to mind: 2282nd Council meeting (General Affairs), Brussels, 10 July 2000, Council doc. 10085/00 of 10 July 2000 at I. For example the Code of Conduct concerning public access to Council and Commission documents, OJ (1993) L340/41. Schwarze, op. cit. footnote 9 at 54. Case 81/72, Commission vs Council (Salaries), [1973] ECR 575, § 13.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

75

presumes a willingness on the part of the institution to legally bind itself, which often is absent. The overview presented in this sub-section indicates that the legal basis issue is not at all straight-forward. There is no rule in EU law establishing in a generic way that all interinstitutional agreements have a legal value. Many of such agreements clearly do have a legal basis, but at all times this depends on the agreement and on the institutions concluding it. I will now proceed to investigate a different dimension of the issue: the problems with ascribing intrinsic legal value to interinstitutional agreements. 3.2.4

To what extent are interinstitutional agreements sufficiently concrete to be law?

In the past, the view has been argued in general terms that secondary law can be annulled if it violates interinstitutional agreements.79 However, in my view, the situation is more complex. Many such agreements do not intend to lay down enforceable obligations. For example, in 2002 Parliament and Council adopted a ‘joint statement’ in which they agreed ‘to provide Community institutions with adequate appropriations and incentives to prepare for enlargement in 2003’.80 Such a commitment clearly is too vague to be legally enforceable en soi. Or look at the language of paragraph 21 of the 1999 Budget IIA: ‘Except for expenditure under heading 2, the institutions will examine the scope for reallocating expenditure between the programmes covered by the heading concerned by the revision, with particular reference to any expected under-utilisation of appropriations. The objective should be that a significant amount, in absolute terms and as a percentage of the new expenditure planned, should be within the existing ceiling for the heading. The institutions will also examine the scope for offsetting raising the ceiling for one heading by lowering the ceiling for another.’81 The language used in these instances is distinctly loose: the institutions ‘examine’, the objective ‘should’ be. Clearly, even if it were accepted that these texts have a legal basis, their content would preclude the courts to acknowledge them as creating legal obligations enforceable in court. They are rather framed as political commitments. On the other hand, there are some interinstitutional agreements that appear intended to have the force of ‘hard’ law. To what extent is it possible to make generalised statements about the intrinsic value of these texts? Different views have been argued,82 with parts of academia arguing that interinstitutional agreements are to be construed as tacit authorisations to formalise by 79 80 81 82

In this sense Waelbroeck and Waelbroeck, ‘Les “Déclarations communes” en tant qu’instruments d’un accroissement des compétences du Parlement européen’ (1988) at 84. Similarly Advocate General Mancini in case 204/86, Greece vs Council (Opinion), [1988] ECR 5323. ‘A’ item note from the General Secretariat to the Council—Establishment of the draft general budget of the European Communities for the financial year 2003, Council doc. 11211/02 of 19 July 2002. Interinstitutional agreement of 6 May 1999 between the European Parliament, the Council and the Commission on budgetary discipline and improvement of the budgetary procedure, OJ (1999) C172/1. Emphasis added. Blanchard, La constitutionnalisation de l’Union Européenne (2001) at 373 doubts whether any generalised statements can be made for interinstitutional agreements not having a specific legal base.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

76

agreement undecided or lacking administrative rules. In this view, interinstitutional agreements are legal instruments and can be relied upon.83 A somewhat different view considers IIAs as binding where the institutions have the intention to be bound.84 Yet again others have argued that such agreements are ‘para-constitutional’ political texts.85 An even more moderate view considers them as soft law,86 with some noting the necessity of agreement between its parties.87 Lastly, in the past some authors tended to view such texts as mere political declarations.88 In the 1970s and early 1980s, IIAs would sometimes go by the name of ‘joint declarations’.89 Originally, the Council was of the opinion that such ‘joint declarations’ were little more than political in nature.90 There may have been some 83

84

85 86

87 88

89 90

At least, one cannot assume legal validity in such cases. In this sense Reich, op. cit. footnote 19 at 81. Bieber, ‘Artikel 7 EG’ (2003) at 663 lets it depend on the content of the respective agreement and on the intention to be bound. Monar, op. cit. footnote 19 at 703 takes this view for at least some IIAs (‘hard-law’ IIAs). The remainder (‘soft-law’ IIAs) might in his opinion have full legal effect. In my view, this confuses the status of soft law: soft law without a legal basis is no law, hence no soft law. In this sense Advocate General Tesauro in case C-106/96, United Kingdom vs Commission (Opinion), [1998] ECR I-2729 at footnote 11. He notes that the failure to observe such IIAs is actionable in proceedings before the Court. The problem with that view is that many obligations, even in ‘hard’ IIAs, cannot be readily applied in law. Further opinions in this direction may be found in Rideau, op. cit. footnote 19 at 169, Lenaerts and Van Nuffel (Bray, ed.), Constitutional law of the European Union (2005) at 791 and RégnierHeldmaier a.o., Commentaire J. Mégret—Les finances de l’Union européenne (1999) at 31. Bobbert, op. cit. footnote 9 at 116 believes in the ‘Vertragsfähigkeit der Institutionen im interinstitutionellen Bereich’ but has to base this on the ‘Interesse der Funktionsfähigkeit der Gemeinschaft’. A hardly satisfying detour from positive law into logic and common sense. Bernhardt, ‘Les sources du droit communautaire: la “Constitution” de la Communauté’ (1982) at 84. Cammarata, ‘Guerre et paix dans les finances publiques européennes: les divergences d’interprétation des accords interinstitutionnels entre les institutions signataires’ (1995) at 49-50; Godet, ‘Le nouveau “code procédure budgétaire” de l’Union européenne’ (2000) at 296; Wellens and Borchardt, ‘Soft law in European Community law’ (1989) at 300; Snyder, Interinstitutional agreements: forms and constitutional limitations (1995) at 2. In relation to the 1993 Code of conduct concerning public access to Council and Commission documents (OJ (1993) L340/41) this is also argued by Kranenborg, ‘De Eurowob in de hand, de EU transparant?’ (2002) at 448. Critical on the soft law theory are Bobbert, op. cit. footnote 9 at 118 and Läufer, op. cit. footnote 43 at 272. Compare (in an international law context) Tammes, ‘Soft law’ (1983) at 187 et seq. Zangl, ‘The interinstitutional agreement on budgetary discipline and improvement of the budgetary procedure’ (1989) at 677, as far as the 1988 Budget IIA is concerned. However, one page further he refers to it as a ‘gentlemen’s agreement’ with ‘apparent legal force’, which is a contradictio in terminis. Griese, ‘Die Finanzierung der Europäischen Union—Bestandsaufnahme und Ausblick’ (1998) at 468 simply denies any legal value on the basis that interinstitutional agreements have no legal basis. An early and more subtle view is expressed by Bieber, ‘The settlement of institutional conflicts on the basis of Article 4 of the EEC Treaty’ (1984) at 520-522. He argues that one also has to take account of the content of the text concerned. For example the 1975 Joint Declaration of the European Parliament, the Council and the Commission concerning the institution of a conciliation procedure between the European Parliament and the Council, OJ (1975) C89/1. ‘The [joint declarations such as that of 4 March 1975 by the Council, the Commission and Parliament on the conciliation procedure] are political undertakings and in the final instance it would be for the Court of Justice to assess their legal implications. As regards more particularly the Joint Declaration instituting the conciliation procedure, the Council considers that by this Declaration the three signatory institutions have undertaken to apply the procedure as soon as all the conditions for its implementation have been met, and in accordance with the arrangements indicated therein’: reply to written question 169/77 of Mr Maigaard to the Council, OJ (1977) C259/4. This view was supported by Advocate-General Reischl in case 64/80, Giuffrida and Campogrande vs Council (Opinion), [1981] ECR 693. Since the agreement was enacted into the

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

77

fears amongst the ministers that Parliament would consider the 1975 Joint Declaration on conciliation to be binding in law. For many texts bearing the title ‘(joint) declaration by the Commission, Council and Parliament’ undoubtedly there is neither a legal basis nor any intention to conclude a legal obligation.91 The Council’s view seems to have changed somewhat following the budgetary wars between Council and Parliament in the early 1980s. These conflicts were (at least partially) settled by the 1982 Joint Declaration.92 Clearly, this text was intended to be more than a mere political declaration. As was discussed above, the Court of Justice has implicitly considered it to be legally binding when it refused to deal with a request from the Council, noting that such problems were ‘subject of an interinstitutional conciliation procedure set up by the joint declaration’.93 In the view of the Commission, ‘[i]t is not possible to give a single answer which would apply to all joint declarations’. As far as the 1975 Joint Declaration was concerned, the Commission considered it a ‘political and moral commitment’ with ‘no binding legal effect’.94 Parliament has been at the forefront of attributing value to interinstitutional agreements but has no generalised view on the matter. In 1993 it saw great potential in the instrument: ‘the practice of concluding interinstitutional agreements . . . has transferred legislative responsibilities to Parliament and has improved the legislative procedure and relations between the three institutions’.95 Nine years later it toned down its enthusiasm somewhat by noting that interinstitutional agreements ‘have an important role to play in coordinating and organising working relations among the institutions’ and argued for a possibility ‘to refer such acts to the Court for an opinion’.96 This, however, does not quite square with other Parliamentary resolutions, in which it at times considered specific interinstitutional agreements as political agreements.97 Moreover, the MEPs’ enthusiasm for legally binding interinstitutional agreements contrasts somewhat with their plea for an ‘ending of the frequent use of legal instruments not provided for in the Treaty on European Union (resolutions, recommendations and conventions), so as to allow democratic control to be exercised’.98 There has been less case law on the intrinsic legal value of interinstitutional agreements than one would imagine and much of it is less than straight-forward on the

91 92 93 94 95 96 97 98

Financial Regulation (see footnote 26 above), this position is in my view overtaken, at least as far as that Regulation is concerned. At least at the time of its conclusion the Council considered even the 1982 Joint Declaration to be ‘un accord politique’: Addendum à la note de transmission du Secrétariat Général—Trilogue interinstitutionnel en matière budgétaire, Council doc. 8361/82 add.1 of 22 July 1982. See for example the Joint declaration on a progress report to be submitted by the Commission by 30 June 2001 mentioned in the Resolution on the draft general budget of the European Union for the financial year 2001 . . ., OJ (2001) C232/281, § 5. Joint Declaration by the European Parliament, the Council and the Commission of 30 June 1982 on various measures to improve the budgetary procedure, OJ (1982) C194/1. Case 34/86, Council vs Parliament (Budget 1986), [1986] ECR 2155, § 50. Reply to question 170/77 by Mr Maigaard to the Commission, OJ (1977) C180/18. Resolution on the conclusion and adaptation of interinstitutional agreements, OJ (1993) C 115/253. Resolution on the typology of acts and the hierarchy of legislation in the European Union, OJ (2004) CE31/126, § 23. E.g. its Resolution on the guidelines for the 1999 budget procedure—Section III—Commission, OJ (1998) C138/149, § 6. Resolution embodying (i) Parliament’s opinion on the convening of the Intergovernmental Conference, and (ii) an evaluation of the work of the Reflection Group and a definition of the political priorities of the European Parliament with a view to the Intergovernmental Conference, OJ (1996) C96/77.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

78

issue.99 In Hauer100 the judges mentioned the Joint Declaration on Human Rights.101 In my view, Monar102 reads more out of this than is warranted: the judges merely referred to the text as a political fact rather than as a rule or a ‘source of inspiration’. In Advernier103 the Court concluded that the conditions of the 1975 Joint Declaration concerning the conciliation procedure were not fulfilled without, however, explicitly pronouncing on the legal value of that text. In Council vs Parliament (Budget 1995) the Court went out of its way to avoid an issue raised by the Council. That institution had argued inter alia that ‘the Parliament has acted in breach of the duty of cooperation in good faith between institutions by unilaterally and arbitrarily proceeding to reclassify compulsory and non-compulsory expenditure in a way which contravenes the undertakings given in the context of the 1982 declaration and the 1993 agreement.’104 What was more, Parliament had defended itself on the same basis, among others by arguing that the 1982 agreement was largely ‘superseded’.105 The question before the Court, whether Parliament had indeed wrongly classified the expenditure, was preceded by another issue: assuming the institutions can agree on the classification of expenditure, could the interinstitutional agreement be invoked in court? It goes perhaps too far to read into the judgment an implicit affirmative answer. The Court focused on the adoption procedure during the Parliamentary session and noted that the budget had been adopted by that institution, even though Parliament and Council had failed to agree the total amount of expenditure to be classified as non-compulsory. Consequently, they had failed to agree the new rate of increase of non-compulsory expenditure. The Court concluded that the budgetary procedure had not been finished, and the president of Parliament thus not been entitled to declare the final adoption of the budget. The judges thus missed an important opportunity to pronounce on the legal status of interinstitutional agreements. Around the same time the Court of First Instance went one step further by arguing in an obiter dictum that its interpretation of Decision 93/731 on access to documents was ‘consistent with the provisions of the Code of Conduct . . . which [the Decision] was intended to implement’.106 However, one year later, the Court of Justice analysed the provisions of the Code and noted that it ‘is an act which is the 99

100 101 102 103 104 105 106

Advocate General Léger implicitly considered interinstitutional agreements binding, but I humbly submit that his reasoning is not unambiguous: ‘In the interinstitutional declaration on democracy, transparency and subsidiarity, the Community authorities concluded that each institution must show that [the principle of subsidiarity] has been observed’. He added that this being the case, ‘it does not seem excessive to expect those institutions, in the future, systematically to state reasons for their decisions in view of the principle of subsidiarity’: case C-233/94, Germany vs Parliament and Council, [1997] ECR I-2405, §§ 88-89; footnotes omitted and emphasis added. Case 44/79, Hauer vs Land Rheinland-Pfalz, [1979] ECR 3727, § 15. See also case 222/84, Johnston vs Chief Constable of the Royal Ulster Constabulary, [1986] ECR 1651, §§ 18-19. Joint Declaration by the European Parliament, the Council and the Commission concerning the protection of fundamental rights and the European Convention for the Protection of Human Rights and Fundamental Freedoms, OJ (1977) C103/1. Op. cit. footnote 19 at 703. Of course, the Joint Declaration concerns mostly substantive policy rather than checks and balances on the institutions. Case 211/80, Advernier a.o. vs Commission, [1984] ECR 131, §§ 22-23. Case C-41/95, Council vs Parliament (1995 Budget), [1995] ECR I-4411, § 13. Ibid., § 15. Case T-194/94, Carvel and Guardian Newspapers Ltd vs Council, [1995] ECR II-2765, § 66.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

79

expression of purely voluntary co-ordination and is therefore not intended in itself to have legal effects’.107 We already encountered the FAO case108 in which the Court of Justice ruled that the defining criterion for a legally-binding interinstitutional agreement is the intention that the instrument have legal effects. Apart from the legal basis question, the emphasis on intention is problematic because it presumes that all parties share similar expectations. That may not necessarily always occur.109 Another problem arises that not all clauses in an interinstitutional agreement may be equally concrete. Although the Court of Justice seems reluctant to pronounce in a clear-cut manner on the direct legal value of one of these texts, it is more prone to accept them as evidence in Court,110 an aid in interpretation111 or, as—to borrow the words of Advocate General Mancini—’droit de complément’.112 In summary, case law affords neither a general basis for the view that all interinstitutional agreements have legal value, nor a criterion for distinguishing which ones do. The ‘intention’ view of FAO is a rare case where a clear criterion has been given. Sometimes the legal situation is straight-forward,113 but more often it is not. The problem thus occurs that, while some texts are political declarations, others may be ‘hard’ enough to be invoked in court.114 Moreover, different clauses within one text may have a different level of legal ‘concreteness’, with some parts being clearly intended as political statements. Qualifying these as law or ‘soft law’ does not satisfy and raises more questions than it resolves:115 if interinstitutional agreements can be law, what is their relation to other acts? To put it differently, if a conflict were to arise between an interinstitutional agreement and an act of secondary legislation, which one should then prevail? This question may be less straight-forward than it seems. What happens if the Council and Parliament were to conclude an interinstitutional agreement that would override the Comitology Decision? Could the Council still rely on that Decision in violation of the interinstitutional agreement? In addition, what is the situation if not the institutions but the secretariats of the institutions conclude an agreement? I believe that attempts to ‘legalise’ interinstitutional agreements for which no specific legal basis exists start from the wrong premise. Should such agreements be part of the legal domain? In the following sections I will argue that it is possible to 107 Case C-58/94, The Netherlands vs Council, [1996] ECR I-2169, § 27. 108 Case C-25/94, Commission vs Council (FAO fisheries agreement), [1996] ECR I-1469, §§ 49-50. 109 For the 1999 Budget IIA see Chevalier, ‘L’accord interinstitutionnel du 6 mai 1999 et les perspectives financières 2000-2006: de nouvelles ambitions pour l’Union européenne?’ (2000) at 443. 110 See for an example of the Court accepting the evidential value of interinstitutional agreements case C-106/96, United Kingdom vs Commission, [1998] ECR I-2729, §§ 25-26. 111 See for an instance where the institutions were at loggerheads over the use as interpretation aid of a ‘declaration’ signed between them case C-360/93, Parliament vs Council (Procurement) (Opinion), [1996] ECR I-1195, § 13. 112 Case 204/86, Greece vs Council (Opinion), [1988] ECR 5323 at section 7. 113 For example recital (7) of the Interinstitutional agreement of 22 December 1998 on common guidelines for the quality of drafting of Community legislation, OJ (1999) C73/1, which was discussed earlier in this Chapter. 114 For example the Décision du 16 janvier 1969 portant installation de l’Office des publications officielles des Communautés, JO (1969) L13/19. This ‘Decision’, which is loosely based on the Treaties (without indication of any specific Treaty provision as a legal base), was concluded by Parliament, the Council, the Commission, the Court of Justice and the EESC. 115 Which does not imply that soft law does not exist: for example, the recommandations of Article 249 TEC would typically fall within this category. But such recommendations have a legal basis (see e.g. Articles 21 and 39 TEU and 37(2), 53, 77, 110, 149(4), 151(5) and 152(4) TEC.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

80

take a different view on the matter by considering such texts as ‘interinstitutional convention’. In order to show that this is not as strange as it seems, first a brief discussion of some aspects of Member States’ constitutional practice is proffered. 3.3

Interinstitutional convention in the Member States

To my knowledge, constitutional conventions were first recognised as such in the context of British constitutional law; but they exist in other constitutional systems as well. The extreme case remains the British constitution, which to a large extent revolves around unwritten rules that cannot easily be challenged in court, but which are no less binding. The other extreme is the American Constitution, which proves that even in a system based on a written Constitution in the form of a Treaty between its constituent states there is ample space for convention to play its part. For example (which I copy from Dicey),116 until the XXIIth Amendment was adopted, it was a convention that no president served more than two terms (the Amendment was adopted following president F.D. Roosevelt’s violation of the convention). Similarly, convention binds electors in their constitutional freedom to choose the president. In this section I will make some general comments on constitutional conventions in some EU Member States. I will concentrate on the United Kingdom and the Netherlands, since between them they cover most of the territory that I want to cover. I will, however, use the French system as a measure to gauge the universality of the notion. As regards the nature of conventions, in sub-section 2.3.2 the argument was made that in Britain checks and balances operate within Parliament rather than between different separated institutions. This is so because the government is part of Parliament. The immediate consequence of this development was that those checks and balances have taken the form of political agreements or conventions rather than of rules that can be invoked in court. Conventions in the British context are—in Dicey’s classic definition—‘customs, practices, maxims, or precepts which are not enforced or recognised by the Courts’.117 Elsewhere the same author uses the term ‘constitutional morality’118 to describe the same concept but (at least when seen through the glasses of EU institutional law) perhaps this phrase is less helpful to describe the nature of the animal. Morality connotes higher rules felt to be constraining. Dicey argued that breaches of convention ultimately would lead to breaches of the law: ‘the sanction which constrains the boldest political adventurer to obey the fundamental principles of the constitution and the conventions in which these principles are expressed, is in fact that the breach of these principles and of these conventions will almost immediately bring the offender into conflict with the Courts and the law of the land.’119 This view has been criticised by later authors: there are many conventions in Westminster that are regarded as binding, even though their infringement would not lead to a breach of the law. Different explanations for the binding character of 116 117 118 119

Dicey, Introduction to the study of the law of the Constitution (1982) at cxliv. Ibid. at 277. Ibid. at cxli. Ibid. at 296-297.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

81

conventions, such as political or psychological factors have therefore been suggested.120 In the view of more modern authors, ‘conventions are observed because of the political difficulties which arise if they are not’.121 On the Continent the true character of British convention has often been misunderstood. For example, in both France and the Netherlands authors attempted to supply the basis of these rules by construing them as unwritten constitutional law.122 Prélot and Boulouis qualify the convention in the United States before President F.D. Roosevelt that no president should seek a third term as ‘coutume’, a concept close to unwritten law.123 This, however, interprets the American Constitution through French glasses. There clearly was a rule in that sense. However, it was of no more than a political nature. No court would have maintained that Roosevelt’s third and fourth terms as president violated the law. If the Constitution had not been amended, the rule would not have become law. To qualify the rule then as unwritten law that cannot be upheld before a court, is essentially to say that it is little more than a political precept that lasted as long as it was perceived to be binding. In other words, ‘unwritten law’ that cannot be invoked before the courts does not differ from convention in the Diceyan sense of the word. The French-style ‘coutume constitutionnelle’ is a much more formalised animal: if one accepts its existence as a source of law, it requires, apart from the repeated124 act, opinio necessitatis. 125 Friends and foes of ‘coutumes constitutionnelles’ seem to agree that these at least pretend to have a legal value.126 There have been even views in France arguing that this would suffice to override the letter of the Constitution.127 Whatever the legal situation in France, it would be incorrect to qualify British conventions as law and not quite right to consider that they could ‘override’ the constitution. The parliamentary systems of Continental Europe too, rely on a larger or lesser degree on unwritten rules that give effect to the letter of the Constitution. In some countries, such as France or Germany, the constitutional upheavals of the 20th Century led to the detailed enactment of many of these rules in the Constitution. In other countries, such as the Netherlands, more room could be left to practical implementation to determine the details of the rules of the game. Conventions in Westminster and elsewhere can be created as the result of an act, a series of acts, or a refusal to act. In Britain, the Netherlands or other parliamentary democracies, convention arises—as a matter of speaking—as the ‘accident’ of constitutional life. A single fact or incident may suffice. Obviously, a long history of uninterrupted constitutional practice is conducive to the development of a body of convention. 120 Hood Phillips and Jackson., Constitutional and Administrative Law (1987) at 122-123. 121 Wade, Bradley a.o., Constitutional and administrative law (1993) at 25. 122 Prélot and Boulouis, Institutions politiques et droit constitutionnel (1984) at 205. Van der Pot, Donner a.o., Handboek van het Nederlandse staatsrecht (1995) at 145-146 provides the history of the development of this assimilation in the Netherlands, but do not make the same mistake. Scheuner sees a role for ‘Verfassungskonvention’ in Germany: Staatstheorie und Staatsrecht (1978) at 322. 123 Ibid. at 206-207. 124 At least in France some discussion exists on the necessity of the act to be repeated. See the discussion by Maestre, ‘A propos des coutumes et des pratiques constitutionnelles: l’utilité des constitutions’ (1973) at 1280-1281. 125 Van der Pot and Donner, op. cit. footnote 122 at 148. 126 See the discussion by Maestre, op. cit. footnote 126 at 1275 et seq. 127 Ibid. at 1277-1278.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

82

As concerns their form, in Britain conventions are unwritten unless they are subsequently enacted. In the latter case, the convention receives by Act of Parliament the force of law128 and stops being a convention. After these general comments, it is time to analyse the matter in the European Union context with a fresh look and note the differences between Member State and EU practice. 3.4

Interinstitutional convention in the European Union

3.4.1

The forms that it can take

The relations between the Parliament, Commission, Council and European Council are governed by law, practice and interinstitutional convention. The legal precepts that govern interinstitutional relations consist of primary and secondary law, as interpreted by case law. They can be relatively easily ascertained, since they are published. Most of the constitutional legal EC literature focuses on these rules, abandoning the study of their implementation to political scientists. Practice consists of the acts, letters, speeches, more generally: behaviour that the different institutions display towards each other, i.e., as the usual and regular method of proceeding not (so far as is known) instituted by express authority.129 Practice does not bind in any way as long as no legitimate expectations are created and certainly cannot override the provisions of the Treaty,130 although convenience and a tendency to rely on precedent often cause the institutions to perpetuate ways of handling certain factual or legal matters. This may give rise to an expectation that certain matters are handled in certain ways without, however, giving rise to an obligation of any sort. Interinstitutional contacts are as old as the Communities131 but the practice of interinstitutional meetings, not foreseen in the Treaties, has recently become more popular. One example of such contacts are the interinstitutional meetings which are nowadays held regularly, especially in the context of the co-decision procedure. Another example is formed by so-called Conventions (with a capital ‘C’)—a conference format used for the preparation of the Charter on Fundamental Rights132 and for the drafting of the Constitution.133 I will not further discuss this type of Conventions. 128 Dicey, op. cit. footnote 116 at lxix. 129 I borrow some elements of the definition from Limon a.o. (eds), Erskine May’s treatise on the law, privileges, proceedings and usage of Parliament (1997) at 4. 130 Case C-84/94, United Kingdom vs Council, [1996] ECR I-5755, § 19. See also case C-327/91, France vs Commission, [1994] ECR I-3641, § 36. 131 See for an early example Parliament’s Resolution on relations between the European Parliament and the Council, OJ (1981) C234/52. For an even earlier example of the Council requesting an agreement, the Note verbale du Président de l’Assemblée au Président du Conseil, Council doc. R/1358/69 of 17 July 1969 at Annex at 4; and the agreement between Council and Parliament based on Council resolution 3 attached to the Treaty of 22 April 1970 on the modalities of collaboration between the two institutions in the budgetary procedure. The Council resolution required the Parliament to agree to those modalities, which it did on 18 November 1971. The Council then formally approved the modalities: Collaboration entre le Conseil et l’Assemblée dans le cadre de la procédure budgétaire, Council doc. R/788/72 of 18 April 1972. 132 Charter of fundamental rights of the European Union, OJ (2000) C364/1. 133 A curious interinstitutional forum is the Monitoring Group set up between the Commission, Parliament and Council to monitor the implementation of the Lamfalussy package (legislation on financial services). The only (hardly legal) basis for the group seems to be the Final report of the Committee of Wise Men on the regulation of European securities markets produced by the Lamfalussy Committee (2001) at 41. See also Council doc. 10503/02 of 28 June 2002.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

83

The most popular form of systemised contact between Parliament, Commission and Council are interinstitutional agreements, but there are many more arrangements between the institutions that bind them in practice, even though they are often missed by the lawyer’s eye. Adapting the terminology used in some Member States, I will use the generic term interinstitutional convention for arrangements between two or more institutions.134 Parliament, the Council and the Commission regularly agree to such interinstitutional conventions, occasionally involving other institutions and bodies.135 Often such conventions are formalised in the form of an interinstitutional agreement. They may go by names such as ‘joint declaration’, ‘modus vivendi’, ‘joint statement’ or formally unilateral or bilateral ‘statement’.136 Unlike Tournepiche,137 I doubt whether different titles imply any distinction in their value. Practice seems to be confused: at least some texts adopted under the name ‘joint declaration’ seem to be interinstitutional agreements, whereas other ‘joint declarations’ are no more than substantive138 or interpretative139 statements. 134 There is not much literature on interinstitutional—or to use another term—constitutional conventions in the European Union and, strangely, much of it seems to be in German. To my knowledge Ehlermann, ‘Sachkompetenzen und Organverfassung der Gemeinschaft’ (1971) at 171 was the first to discuss the application of ‘constitutional conventions’ in the European Community, although he seems to read more law into conventions than they really contain. Zuleeg, ‘Der Verfassungsgrundsatz der Demokratie und die Europäischen Gemeinschaften’ (1978) at 46-47 and Jacqué, ‘Parlement européen (janvier 1981-juillet 1982)’ (1982) at 672 mention the possibility without really elaborating on the binding nature of conventions. Läufer, op. cit. footnote 43 at 274 is critical, but I fear that he too makes more of British-style constitutional conventions than what they really are. Similarly Hilf, op. cit. footnote 74 at 19-20 considers conventions ‘as alien to the system’ of European constitutional law. However, he too seems to misunderstand the political nature of conventions. Bradley, ‘Institutional design in the Treaty of Nice’ (2001) at 1108 and Hartley, The foundations of European Community law (2003) at 23-25, consider the ‘Luxembourg Compromise’ to be a convention but, as I will argue in sub-section 9.3.2 below, this is a simplification. Evans, footnote 74 at 67 considers a convention equating a refusal to provide discharge to the Commission with censure. Finally, Corbett, Jacobs and Shackleton, The European Parliament (2000) at 4 refer to the 1975 interinstitutional agreement as ‘a sort of constitutional convention’. Beukers, ‘Enhancing Parliamentary control over the European Commission and the Member States’ (2006) at 21 et seq. places the individual political responsibility of Commissioners in the context of constitutional conventions (see Chapter 4, below). 135 Examples of the latter are the Joint declaration of the President of the European Commission and the President of the Committee of the Regions, CoR doc. DI CdR 81/2001 rev. 2 FR/OU/ss of 20 September 2001 and a similar text concluded between the Commission and the Economic and Social Committee, EESC doc. R/CES 1004/2001 rev. Item 9Ba) Appendix FR/ET/ht of 24 September 2001. In 2002 the EESC, bolstered by its success in concluding a co-operation agreement with the Commission, wrote to the Council with a similar proposal: Lettre de M. Frerichs, Président du Comité économique et social européen en date du 14 mai 2002 à M. Piqué i Camps, Ministre des Affaires Etrangères du Royaume d’Espagne, Président du Conseil ‘Affaires Générales’ de l’UE, Council doc. 9061/02 of 23 May 2002. The latter turned down the offer politely but firmly: ‘I’ item note from General Secretariat of the Council to COREPER, Council doc. 15036/02 of 2 December 2002. 136 E.g. the Statement on the provision of information to the budgetary authority, OJ (2001) C197/345, which was an agreed text: Parliament Resolution on the draft general budget of the European Union for the financial year 2001, OJ (2001) C197/339, § 34. 137 Tournepiche, op. cit. footnote 13 at 215-216. Similarly, Klabbers, ‘Informal agreements before the European Court of Justice’ (1994) at 1011. Lenaerts and Van Nuffel (Bray, ed.), op. cit. footnote 84 at 791 consider the title of the agreement relevant. 138 See the examples mentioned at the end of footnote 2, above. 139 E.g. the joint declarations attached to the interinstitutional agreement on recasting: OJ (2002) C77/3.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

84

In other cases, the convention remains unwritten. This does not mean that the convention is never recorded on paper, but rather that it has not been formalised qualitate qua. The ‘Gentlemen’s agreement’ under which Parliament and Council agreed not to amend the estimates of each other’s expenditure is an example of such unwritten interinstitutional convention. Technically, the agreement takes the form of separate resolutions from Council and Parliament not referring to each other.140 There has never been any suggestion that the agreement creates a legal obligation for each institution not to interfere in the budget of the other. The point to retain is that, up to this day, the convention is adhered to and affects the effective influence that institutions have over each other. Similarly, the declaration made by Mr Prodi in Parliament on 5 February 2002 in which he committed the Commission in the field of comitology, was followed by a resolution in which Parliament took note of ‘the statement detailing the Commission’s undertakings to Parliament’.141 The result is a set of obligations that bind at a political level. Interinstitutional conventions are thus distinct from legal obligations. For example, agreements based on Article 272(9) have a legal basis and are sufficiently concrete to be enforced before the Courts. They are therefore not interinstitutional conventions. In the case of many other interinstitutional arrangements, the situation is less clear-cut. Some interinstitutional conventions concern prosaic administrative matters such as administrative co-ordination, recruitment or security matters.142 Many other interinstitutional conventions are clearly not intended to be enforced by the courts. One example of this is formed by the 2005 Commission-Parliament Framework Agreement,143 which provides detailed commitments for those two institutions in the field of their political behaviour. Yet this agreement is intended to be binding and in practice it is, at a political level. It is no exaggeration to say that interinstitutional convention covers the full length and width of the areas in which the European Union is active. Some of it covers very broad and important fields of constitutional law (such as, for example, the 1999 interinstitutional agreement on the budgetary procedure).144 Other interinstitutional conventions deal with prosaic small details, such as for example the codification technique.145 A notable difference between European and Member State practice is the use of convention as a means to decide on substantive policy. This happens especially in the field of budget policy where the three political institutions since 1988 have agreed ‘financial perspectives’. In this study I focus on the role of interinstitutional conventions as checks and balances. I will therefore not further discuss such substantive political agreements in any detail. In summary, between mere practice and Treaty-based law, there is a category of rules that are concluded by agreement between the institutions and, although they are not actionable in court, bind them in practice. A review of the checks and balances governing the European constitution cannot be complete without a review of these. 140 The Gentlemen’s Agreement is discussed in section 14.3, below. 141 For the background of the matter see sub-section 13.2.5, below. 142 For example, OJ (1993) C207/46 (on the recruitment of the disabled people in the Community institutions). 143 Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125. 144 OJ (1999) C172/1. 145 Interinstitutional agreement of 20 December 1994—Accelerated working method for official codification of legislative texts, OJ (1996) C102/2;

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

3.4.2

85

Its nature

Interinstitutional convention in the European Union may be described as rules that, whilst not being binding in law, are recognised by two or more institutions146 as constraining their behaviour. It differs from mere practice in that there is an element of perceived obligation. On the other hand, it differs from droit coutumier—if such a notion exists in European law at all147—in that there is no legal, but rather a political obligation. The key to understanding the nature of interinstitutional convention lies in the distinction between legal rules (i.e. rules that can be enforced by the courts) and rules that constrain the institutions solely at a political level. As in Britain, conventions may effectively influence the behaviour of the institutions to a high degree. For example, the Commission instructed its staff that the 2000 Framework Agreement ‘must be applied fully, coherently and consistently at all levels and in all dealings with the European Parliament’.148 Whatever one’s view of the so-called Luxembourg Compromise that once dictated the proceedings within the Council, none can argue that the unanimity rule read into it was not a living reality. I certainly do not claim that every rule of interinstitutional convention (be it laid down in an interinstitutional agreement or otherwise) is always respected by the institutions. Rather, one could say that such conventions guide the behaviour of the institutions to a large extent. It depends on the convention how well it is followed and probably no hard and fast rule can be formulated in this respect. As noted in section 3.3, Dicey explained the binding nature of conventions by arguing that their breach would ultimately result in a break of the law. Whether or not this applies for every single convention in Westminster; it certainly does not for interinstitutional conventions in Brussels and Strasbourg. It seems more correct to say that conventions are respected as binding because there is a mutual interest to do so. It follows from the discussion thus far that the extent to which a text adopted by two or more institutions is intended to be binding is a matter of interpretation. Consequently, the denomination of a certain text as ‘interinstitutional agreement’, ‘joint declaration’, something else, or nothing at all, is of minor importance. This being said, the name ‘interinstitutional agreement’ may indicate the intention of the parties to it. Interinstitutional conventions fulfil two different constitutional functions. Firstly, such conventions may have an implementation function where they execute a Treaty or legislative rule. For example, the Joint Declaration on practical arrangements for the new co-decision procedure149 set out detailed rules implementing the co-decision procedure of Article 251 TEC. The OLAF Interinstitutional agreement150 was the cornerstone of a legal package intended to improve anti-fraud action in the Union. 146 Or Member States. In section 9.3 the voting conventions within the Council are discussed. 147 Kranz, ‘Le vote dans la pratique du Conseil des ministres des Communautés européennes’ (1982) at 412 doubts it, with some support in case law: case 7/71, Commission vs France (Euratom Supply agency), [1971] ECR 1003, § 6. 148 Practical guide to the Framework Agreement, Commission doc. SP (2000) 3785-5 of 20 December 2000 at 2. 149 OJ (1999) C148/1. 150 Interinstitutional agreement of 25 May 1999 between the European Parliament, the Council of the European Union and the Commission of the European Communities concerning internal investigations by the European Anti-fraud Office (OLAF), OJ (1999) L136/15.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

86

Secondly, interinstitutional conventions may fulfil a planning function if they establish a framework of future, agreed policy. In this sense, interinstitutional agreements have proved both successful and popular in the domain of budget law. As noted earlier, this can even go so far that the legislation concerned refers to the interinstitutional agreement.151 3.4.3

Its creation The generally more formalised character of conventions in the European Union

The creation of conventions often is a more formal process in the EU than it is in the Member States, and tends to be more based on explicit agreement than on accepted unilateral custom. This more formal nature might cause especially British lawyers to object to the use of the term ‘convention’. Indeed, in Westminster conventions ‘vary from generation to generation, almost from year to year’.152 In the Member States, conventions frequently are the result of political accident. In the European Union this is well possible (as discussed below), although interinstitutional conventions tend to be explicitly sought as such by the institutions. For example, in a 1981 resolution the Parliament regarded ‘joint declarations by Council, Commission, and Parliament as both necessary and feasible, especially in the fields of exchanges of information, consultation and conciliation on internal and foreign issues and in the budgetary procedure’.153 By way of another example, around the time of the entry into force of the Maastricht Treaty the MEPs called ‘on the Council and the Commission finally to acknowledge the need for a political agreement on the implementation of the co-decision procedure’.154 Lastly, the downfall of the Santer Commission led the emboldened Parliament to demand a series of concessions from the Commission covering the whole spectrum of their mutual relations.155 These demands led to the 2000 Framework Agreement.156 The initiative for interinstitutional conventions may emanate from the Member States,157 the Commission,158 the European Council159 or Council160 and even from 151 By way of examples, see OJ (1988) L185/24, OJ (1988) L374/1, OJ (1989) L388/18, OJ (1989) L388/1, OJ (1994) L130/1, OJ (1995) L86/10, OJ (1996) C65/37. 152 Dicey, op. cit. footnote 116 at cxlv. 153 Resolution on relations between Parliament and Council of 9 July 1981, OJ (1981) C234/52, § 2. The Commission was willing to take up the offer: Commission communication The institutional system of the Community—restoring the balance, Bull. EC Suppl. 3/82 at 12. 154 Resolution on the interinstitutional agreements of 16 September 1993, OJ (1993) C268/165, § 13. Emphasis added. 155 Resolution on the ‘Prodi Commission’, OJ (2000) C54/49. 156 Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121/122. 157 E.g. Declaration 39 on the quality of the drafting of Community legislation attached to the Treaty of Amsterdam, OJ (1997) C340/139. 158 E.g. in reply to written question E-1368/93 by Mr Lafuente López, OJ (1994) C140/49. 159 E.g. § 2.3.6 of the Solemn Declaration on European Union, Bull. EC 6-1983 at 24: ‘The Council will enter into talks with the European Parliament and the Commission with the aim, within the framework of a new agreement, of improving and extending the scope of the conciliation procedure

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

87

the Court of Auditors,161 although most often Parliament seems to be demandeur.162 This reflects the fundamental interest that institution has in developing constitutional law. In the budgetary field the negotiations have often been initiated or conducted through the ‘trialogue’ of Parliament, Council and Commission, which itself resulted from the 1982 Joint Declaration.163 Formal convention often is drafted by a working party composed of legal experts from the relevant institutions and voted with a normal majority in the Parliament. In 1993 the three political institutions even negotiated a package of interinstitutional agreements following on the Maastricht Treaty during an interinstitutional ‘conference’.164 Formal requirements Article 253 TEC requires that ‘[r]egulations, directives and decisions’ be reasoned. In addition, Article 254 TEC lays down publication requirements.165 However, formal interinstitutional conventions do not take the form of Article 249 TEC (I-33 Const) instruments. Consequently, the formal requirements for such instruments (publication) do not apply. For example, in 1970 the Council consented through letters from Mr Harmel and Mr Scheel dated 20 March and 22 July 1970 to inform Parliament of its reasons for not complying with any of the latter’s opinions, whenever Parliament thought it useful to request this.166 This interinstitutional convention was not published. The Access to Documents Regulation167 sets out in its Article 13 a more detailed list of acts that are to be published in the Official Journal but, pointedly, the EU legislator failed to include interinstitutional agreements. Notwithstanding this

160 161 162

163 164 165 166 167

provided for in the Joint Declaration of 4 March 1975’. Parliament and the Commission followed this up with their Joint Declaration of the European Parliament and the Commission, Parliament doc. 1-328/84/rev of 23 May 1984, § 17: ‘Parliament and the Commission call on the Council to decide on further advances to be made in institutional matters in the context of the Treaties by means of tripartite declarations, pursuant to the Solemn Declaration on European Union’. See also Annex 1 to Part A of the Conclusions of the Edinburgh European Council of 11 and 12 December 1992 at item I.4 and the Presidency Conclusions of the Berlin European Council 24 and 25 March 1999, §§ 17-18. Which initiated the negotiations leading to the 1982 Joint Declaration: Communication from the Commission to the budgetary authority concerning legal basis and maximum amounts, Commission doc. SEC (94) 1106 final of 6 July 1994 at 5. See, for example, OJ (1996) C287/1, § 91. Examples may be found in resolutions at the following places: OJ (1980) C117/53, § 9, OJ (1995) C89/60, OJ (1995) C126/75, OJ (1995) C339/28, OJ (1996) C96/288, OJ (1997) C132/31, OJ (1997) C304/131, OJ (1998) C80/36, OJ (1998) C138/177, OJ (1998) C379/92, OJ (1998) C379/88;, OJ (1999) C104/135. Lastly, see the Resolution on the interinstitutional agreements, OJ (1993) C268/165. Sometimes Parliament will even lay down its negotiation position in its resolution, as e.g. in its Resolution on the conclusion and adaptation of interinstitutional agreements of 12 March 1993, OJ (1993) C115/253. Timmann, ‘Die Interinstitutionelle Vereinbarung über die Haushaltsdisziplin vom 29. Juni 1988’ (1988) at 274. The package of agreements can be found in OJ (1993) C329/132 et seq. With some mutatis mutandis changes, Article I-38.2 Const. Note from COREPER to the Council of the European Communities, Council doc. 5749/1/82 of 2 April 1982. See also the reply to written question 1901/82 by Mr Moreland to the Council, OJ (1983) C118/16. Regulation 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, OJ (2001) L145/43.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

88

omission, many interinstitutional agreements are published in the Official Journal or in the Bulletin of the EU. In the Council formal interinstitutional agreements of a conventional nature will be adopted by consensus. The requirement of consent The essential difference between conventions in the Diceyan sense and interinstitutional conventions is that the latter are conventions within the literal meaning of the French convenir. Unlike its British corollary, in Europe its only essential ingredient is the consent of the institutions party to it. This consent must cover both the substance of the convention, as well as the recognition of its binding nature upon the institutions party to it. Consent must be evident, but need not be expressly given. Often it is: for example, many interinstitutional agreements are put to paper and signed by the institutions that are party to them. In other cases consent takes the form of acquiescence. For example, the Rome European Council of 1 and 2 December 1975 adopted a procedure for the representation of the Community at the Conference on International Economic Co-operation. This procedure imposed obligations on both Council and Commission and, at first sight, the impression could arise that the Commission had made itself subordinate to the European Council. A closer reflection, however, shows that this is not the case. The European Council was, at the time, not yet codified in Community law and hence was itself purely a product of interinstitutional convention.168 Its membership included the heads of state or government and the Commission president. Since, at the time, the European Council (notwithstanding its very real importance in practice) did not exist in EU law, the pre-SEA European Council is best construed as a conference of the heads of state or government with the Commission president. In that interpretation, the European Council’s guidelines for the Community’s representation at the Conference on International Economic Co-operation essentially are an agreement between the Member State governments and the Commission. One can go even further: uncontested practice can create convention, provided the institutions party to the convention recognise the binding quality of such practice. For example, in 1973 the Commission proposed measures to improve the involvement of Parliament.169 By relying on these, Parliament implicitly accepted them. Consent may not always be a once-off matter. For example, the role of the Council presidency—which originally was largely symbolic170—largely grew from convention over a long period. Obviously, consent is only required between the parties to the interinstitutional convention. Parliament was not party to the unanimity convention flowing from the Luxembourg Compromise, but clearly its position was influenced by the less flexible decision-making within the Council. Vice versa, the Council’s position has been affected by the 1990, 1995 and 2000 interinstitutional conventions between Parliament and Commission that required the latter inter alia to report to the MEPs what went on inside the Council. 168 The conventions surrounding the European Council are analysed in sub-section 9.2.2. 169 Practical measures to strengthen the powers of control of the Parliament and to improve relations between the Parliament and the Commission, Commission doc. COM (73) 999 final of 30 May 1973. 170 Constantinesco a.o., Traité instituant la CEE—Commentaire article par article (1992) at 861.

89

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

A convention, then, cannot bind parties that explicitly repudiate it and continue not to honour it. For example, the Council regularly signifies to Parliament when the MEPs changed their Rules of Procedure, that these ‘do not in any way bind the Council’.171 Similarly, the Council protested against the Framework Agreement of 2000 and 2005 negotiated between Parliament and the Commission, arguing in each instance that ‘the undertakings entered into by those institutions cannot be enforced against it in any circumstances’.172 Although, from a legal point of view, this seems obvious, it is a clear sign that the Code of conduct and the changes in the Rules of Procedure did not create any convention between Parliament and Council. That position could change, however, if the Council’s subsequent behaviour would imply that it considered itself bound by the Framework Agreement. Sometimes formal interinstitutional convention is accompanied by declarations on the part of one or several of the parties to the interinstitutional agreement.173 To the extent that not all parties to the interinstitutional agreement agreed to such declarations, these can of course at most bind their signatory on the political level, or as a result of the doctrine of legitimate expectations. 3.4.4

Limitations agreements

to

the

capacity

to

conclude

interinstitutional

Is there a limit to the power of the institutions to conclude interinstitutional agreements? Tentatively, this question must be answered positively. Article 7(1) TEC lays down the rule that each institution must act within the limits of the powers conferred upon it.174 It follows that it is not possible for an interinstitutional agreement to change the Treaties by, for example, replacing a unanimity requirement with qualified majority. What such agreements can do is to build upon primary law. For example, in my view it would be perfectly legal for the institutions to agree that in a matter where Parliament has no right of consultation such consultation will be undertaken (and in fact, the Council has agreed to such a commitment).175 It would, however, not be legally binding in the sense that the commitment could be enforced through the Courts. Conversely, in 1985 the Court of Justice ruled that the ‘balance of powers between the institutions provided for by the Treaties’ disallowed Parliament to ‘deprive the other institutions of a prerogative granted to them by the Treaties themselves’.176 It clearly follows that interinstitutional convention may not directly contravene the text of the Treaties or secondary law based on it in a way that impairs the prerogatives of an institution or the Member States. Although convention may not violate the letter of primary law, it can nullify its effects. Just as the sovereignty of the British monarch in the course of centuries was mitigated not by reducing his powers but by forcing him to exercise them in 171 See e.g. the Draft letter to the President of the European Parliament concerning the amendments to the European Parliament’s Rules of Procedure, Council doc. 12315/02 of 24 September 2002. 172 2282nd Council meeting (General Affairs), Brussels, 10 July 2000, Council doc. 10085/00 of 10 July 2000 at I and Council statement concerning the framework agreement on relations between the European Parliament and the Commission, OJ (2005) C161/1. It appears that the Council similarly protested against the 1995 Code of conduct that preceded the Framework Agreement. 173 E.g. the declarations attached to the 1999 Budget IIA declarations: OJ (1999) C172/21. 174 I-19.2 Const . A comparable obligation may be read from Article 5 TEU. 175 Note from COREPER to the Council of the European Communities, Council doc. 5749/1/82 of 2 April 1982. 176 Case 149/85, Wybot vs Faure a.o., [1986] ECR 2391, § 23.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

90

accordance with Parliament’s will, convention in the European Union at times bludgeoned the sharpness of the law in unexpected ways. In 1995, Parliament’s Budget Committee complained about the 1993 Budget Interinstitutional Agreement that it ‘a eu pour effet, dans le cadre des relations entre les deux branches de l’autorité budgétaire, de réduire à une simple formalité certaines phases de la procédure budgétaire qui se trouvent ainsi limitées à un exercice de forme plutôt que de fond.’177 3.4.5

Can convention be abrogated?

Interinstitutional convention can be abrogated as easily as it is concluded. In the case of formal convention, this happens most often through the replacement of interinstitutional agreements with newer ones.178 As another possibility, the convention may be limited in time (as happened with the 1988, 1993, 1999 and 2006 Budget IIAs). Additionally, it is possible that conflicting law abrogates the convention. For example, Annex XV of the Accession Treaty for the ten new Member States comprises a financial perspective that effectively replaced the financial perspective agreed in the 1999 Budget Interinstitutional Agreement. This move irked Parliament and, following threats from the MEPs to withdraw from the 1999 Budget Interinstitutional Agreement; a compromise was found through a new interinstitutional arrangement.179 3.4.6

The Nice Declaration

In response to the conclusion of the 2000 Framework Agreement between Commission and Parliament, a declaration on interinstitutional agreements was attached to the Treaty of Nice. This text clarifies that ‘. . . the duty of sincere cooperation which derives from Article 10 of the Treaty establishing the European Community and governs relations between the Member States and the Community institutions also governs relations between the Community institutions themselves. In relations between those institutions, when it proves necessary, in the context of that duty of sincere cooperation, to facilitate the application of the provisions of the Treaty establishing the European Community, the European Parliament, the Council and the Commission may conclude interinstitutional agreements. Such agreements may not amend or supplement the provisions of the Treaty and may be concluded only with the agreement of these three institutions.’180 The declaration is somewhat difficult to interpret. It seems obvious that interinstitutional agreements cannot ‘amend’ the Treaties. The institutions cannot conclude interinstitutional agreements whilst ignoring rules of law.181 Still, it is less evident why interinstitutional agreements could not ‘supplement’ such rules of law. 177 Parliament doc. PE 212.450/def./Partie II of 4 May 1995 at 33. 178 E.g., the Declaration by the European Parliament, the Council and the Commission of 6 March 1995 on the incorporation of financial provisions into legislative acts, OJ (1995) C 293/4, corr. OJ (1996) C102/4 was explicitly replaced by the 1999 Budget IIA: point 7 of the Interinstitutional Agreement of 6 May 1999 between the European Parliament, the Council and the Commission on budgetary discipline and improvement of the budgetary procedure, OJ (1999) C172/1. 179 The matter is discussed in section 14.1, below. 180 OJ (2001) C80/77. 181 As the Court of Justice made clear in case 204/86, Greece vs Council, [1988] ECR 5323, § 17.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

91

For example, the 1999 Budget IIA or the Recasting Agreement182 can hardly be considered illegal for providing supplementing rules. It would seem that, as far as interinstitutional agreements set additional detailed rules interpreting and implementing the Treaties, they ipso facto do ‘supplement’ it. Since, in any event, interinstitutional convention is bound by the margins of the law, it can at best provide an interpretation of the law. Can the Nice Declaration prohibit the institutions to agree on an interpretation of the Treaties? The declaration limits interinstitutional agreements to texts agreed between the Parliament, Council and the Commission. This limitation reflects the Member States’ fears that agreements between the executive and Parliament would impair their prerogatives. A possible example of such a case is formed by paragraph 3.5 of the 1995 Code of Conduct between Commission and Parliament: ‘The Commission shall keep the competent parliamentary committee regularly informed of the principal positions emerging from discussions within Council bodies, particularly where they depart from the original proposal, and shall also forward any Commission amendments to the original proposal on the basis of which the Council would continue its discussions. The Commission shall inform Parliament as soon as possible if it adopts a favourable opinion on amendments by the Council to its proposals.’183 To some extent, the Member States’ fears are understandable. Nevertheless, the restriction leaves open the question how agreements between only two institutions ought to be qualified. For example, what to make of the agreement between the Court of Justice and the Commission on the oath to be taken by the latter upon its investiture?184 Similarly, one would hardly consider the FAO voting arrangement between Commission and Council185 now to be illegal. Tournepiche186 appears to find the Nice Declaration consistent with pre-Nice practice since she believes that no texts under the name ‘interinstitutional agreement’ have been adopted by only two institutions. However, that interpretation lays undue stress on the title of the interinstitutional agreement: is the ‘declaration’ ‘agreed’ between the Parliament and the Council on the question ‘des conditions et de la flexibilité des budgets en fonction du résultat global des négociations et du plafond des futures perspectives financières’187 a political agreement or is it to be ignored? Interestingly, those two institutions did call the text an ‘accord’ in the French text. More examples of interinstitutional agreements between the Council and only one institution readily present themselves: the ‘gentlemen’s agreement’ between Parliament and Council188 and the European Parliament and Council Decision on the adjustment of the financial perspective to take account of the conditions of implementation (submitted by the Commission pursuant to paragraph 10 of the Interinstitutional Agreement of 29 October 1993).189 Lastly, the interinstitutional agreement between Parliament and Council concerning Parliament access to sensitive 182 Interinstitutional Agreement of 28 November 2001 on a more structured use of the recasting technique for legal acts, OJ (2002) C77/1. 183 OJ (1995) C89/69. 184 Bieber, ‘Legal developments in the European Parliament’ (1985) at 350. 185 See footnote 57, above. 186 Op. cit. footnote 13 at 220. 187 2150th Council meeting (Budget), Council doc. 13917/98 of 8 December 1998. 188 See section 14.3, below. 189 OJ (1998) C138/155.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

92

information in the field of security and defence policy190 was negotiated following the Nice Declaration. Logically, the Nice declaration cannot have been intended to be law. On the other hand, the question arises whether its value is merely political.191 I belief that, by itself, the text has no more than a political value but it proves an important point: the Member States accepted in abstracto that interinstitutional convention binds at a political level. This, coupled with the knowledge that the declaration—as a mere political instrument—cannot override a legally binding text, leaves only two possibilities. Either one believes the declaration to be a fluke since it tries to impose a constitutional rule through an instrument without legal value. Alternatively—and this is the only interpretation in which the Declaration makes sense—one accepts the existence of interinstitutional convention as binding at a political level (meaning it cannot be enforced in court). However, in such an interpretation, the Declaration tries to state something that it cannot: as a practical matter, interinstitutional conventions come into being, whether or not the Declaration allows it. 3.5

Conclusions of this Chapter

In 1993 Parliament somewhat enthusiastically noted that ‘the practice of concluding interinstitutional agreements between the Council, the Commission and Parliament has transferred legislative responsibilities to Parliament and has improved the legislative procedure and relations between the three institutions’.192 Parliament’s enthusiasm hardly surprises: it sees in interinstitutional agreements the possibility to extend its rights without Treaty change, to steadily improve its institutional position193 and to fence off assaults on its prerogatives. Thus, its general relations with the Commission have been formalised in political commitments which have been systematised at regular intervals. When, following the Santer Commission’s discomfiture, Parliament gained the spotlight, it solicited concessions from Mr Prodi. These were formalised in the 2000 Framework Agreement. Parliament subsequently capitalised on its initial refusal to approve Mr Barroso’s Commission by extracting concessions that were formalised in the 2005 Framework Agreement. On the other hand, some view interinstitutional convention more as a stabilising or, to use a more negative term, petrifying factor.194 It probably is both: not all institutions accept the institutional balance as ideal. Over the decades, especially Parliament has sought to improve its influence. Political agreements between the

190 Interinstitutional agreement of 20 November 2002 between the European Parliament and the Council concerning access by the European Parliament to sensitive information of the Council in the field of security and defence policy, OJ (2002) C298/1. 191 In its otherwise fiery reaction to the 2005 Framework Agreement concluded between Parliament and the Commission, the Council timidly refers to the ‘spirit’ of the Nice Declaration: Council statement concerning the framework agreement on relations between the European Parliament and the Commission, OJ (2005) C161/1. 192 Resolution on the conclusion and adaptation of interinstitutional agreements, OJ (1993) C115/253, recital D. 193 Schwarze, op. cit. footnote 9 at 66. 194 In this sense Magiera, ‘Diskussion auf dem Kolloquium für Herrn Professor Dr. Hans-Joachim Glaesner’ (1995) at 85.

CHAPTER 3: INTERINSTITUTIONAL CONVENTION

93

institutions merely reflect this struggle between the more ‘conservative’ forces in the Union and the more ‘integrative’ ones and help to smooth relations between them. The variety of arrangements between the institutions is such that the simple term ‘interinstitutional agreement’ does not do it justice. Moreover, that term suggests a legal status that such arrangements often do not and should not have. Even when an ‘interinstitutional agreement’ has a legal basis, not all of its provisions may have been intended to be invoked in court. In Britain the institution of convention took off because the country lacks a written constitution. Since most power relationships take place within Parliament (in the broad sense of the word), these relationships are almost by definition political rather than legal. In the European Union, conventional arrangements are popular for a wholly different reason. Because the legislative power of the institutions is strictly circumscribed by Article 7 TEC, there is little possibility to regulate interinstitutional relationships in a Regulation. Such relationships must be either set out in primary law (which is cumbersome and petrifies them) or in the Rules of Procedure of the different institutions (which are unilateral instruments). Interinstitutional convention has been the way in which the gaps in-between are filled. The concept of constitutional convention better describes the nature of these arrangements and clarifies their relationship with law in a more satisfying manner than any of the legal base theories put forward in literature. I must recall, however, that the constitutional system of the Union differs radically from that of Britain. Consequently, conventions in the Union are a somewhat different animal than they are in Westminster. The analysis above showed that conventions in the Union are often written and more formal than they are in Britain. Undoubtedly the Union’s origins as a quasiinternational organisation have instilled it with a diplomatic culture that still leaves its mark on the spirit in which institutions do business. The further development of the European Union into a constitutional unit may diminish this ‘diplomatic’ character somewhat. In the following Chapters of this study I will review in more detail the checks and balances that govern the relationships between the political institutions with special emphasis on interinstitutional convention. Since there is so much of it (and a lot of it imperfectly recorded at that), it is impossible to be complete. Still, I believe it must be possible to convey at least a more colourful and detailed picture of interinstitutional relations than a mere reading of the legal texts allows.

95

PART II: SEPARATION OF PERSONS

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

4

THE APPOINTMENT OF THE COMMISSION

4.1

Introduction to this Part

97

In the first Chapter of this study I formulated the Research Question to what extent the EU institutions can be balanced through the use of interinstitutional convention and what the conditions are for doing this. Balancing institutions presumes a measure of separation between them. In Chapter 2 different aspects of separation of powers theories were investigated and a perspective on separation of powers was formulated. The argument was made that this is not the only perspective from which the European Union can be interpreted. Instead, the separation of institutions perspective chosen affects the conclusions on the checks and balances that keep the institutions in check. The analysis in Chapter 3 showed that checks and balances are not only of a legal nature but also consist of silent or express agreements that constrain the institutions at a political level. I defined interinstitutional convention. The argument was made that a mere reading of the Treaties not always conveys the complete picture of the power relations between the institutions. In this and the following Parts of this study, interinstitutional convention will be looked at in closer detail. The focus will not be on testing abstract theories, but rather on assessing concretely to what extent interinstitutional convention adds flesh to the bones of the constitution in practice and to what extent it modifies it. In Chapter 2 I noted that checks and balances between the institutions operate at several levels. One of them is the separation of persons level (incompatibilities and checks on appointment and dismissal procedures). The argument was made that it also follows that no single institution should control the appointment or dismissal of the members of itself or another one without proper autonomous or interinstitutional checks and balances. For example, there would be little point in providing for interinstitutional checks and balances on the legislative process if the Member State governments would control the appointment of all the institutions involved (which they do not). This Part analyses the checks and balances relating to the appointment of the different offices in the Union. I will dispense with a discussion of the status of MEPs and Council members, because both categories of offices are controlled by autonomous checks and balances (elections and Member State constitutional requirements, respectively). This Chapter deals with the Commission’s appointment. It is a field of institutional law that has seen a lot of change in the last twenty years and which was contested during the Constitutional Convention and the ensuing IGC. I will briefly discuss the interinstitutional conventions that played a part in it before turning to the law and expected practice under the Constitution’s procedures. I will discuss these as they concern the appointment of the Commission (section 4.3), its president (4.4), the Minister of Foreign Affairs proposed in the Constitution (4.5) and the other members of the College (4.6). Chapter 5 discusses the Union’s parliamentary system and Chapter 6 reviews the appointment of several other categories of officers.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

4.2

98

Development of the appointment procedure

In most constitutional systems the executive is by far the most powerful ‘branch’ of government. That makes it necessary for any democratic constitution to provide checks and balances on its operation, but also to ensure that its appointment and dismissal are checked, democratic and transparent. This is not just a matter of separation of institutions. There is an obvious link between the democratic legitimacy of the European Union’s political system and the appointment of its (primary)1 executive. The low and decreasing turnout in European Parliament elections2 (at least, in those Member States were voting is not compulsory) attests to the fact that citizens do not perceive Parliament to be very influential in setting the ‘colour’ of the European executive. One theme of this Chapter is the evolution which has taken place in this respect. A review of the changes in this field in the last decades, and especially of the interinstitutional convention involved, will give more relief to the achievement of M. Giscard d’Estaing’s Convention and the ensuing IGC. The obvious question to be answered is, whether the voters’ perception and reality still correspond. It is true that, originally, there was no role foreseen for Parliament in the appointment of Commissioners. On the other hand, the matter was not solely in the hands of the Member States either. The notion that the pères fondateurs of the original Communities had wished for each Member State to be represented in the Commissions is a tenacious myth3 that negatively influenced the debates in the Constitutional Convention. Although the procedures for nominating the first EEC Commissions were purely intergovernmental, this was not the case for the ECSC’s High Authority. The six Member States appointed eight of the members of the very first High Authority by common accord. These eight members co-opted a ninth, who was elected if he got the votes of at least five of his colleagues.4 The High Authority thus selected remained in power for six years. The following High Authorities were to be appointed using the same procedure, although it was now possible for the Member States to appoint the High Authority with a five-sixth-majority vote, with a limited right of veto. The ninth member remained co-opted. Lastly, following the initial six years, one third of the High Authority was to be replaced every two years. This system was abandoned in the EEC and Euratom, whose Commissions were appointed by the Member States.5 In the early Communities there was no linkage between the second Commissioner and large Member States. The drafters of the Treaties considered a High Authority and an EEC Commission of six members too small and opted for nine members instead. The Euratom Commission consisted of five members, which made it impossible for each Member State to have a national on it. In the early executives it was even possible for a small state to have two Commissioners.6 1 2 3 4 5 6

As discussed in Chapter 2, it is not quite correct to refer to the Commission as ‘the’ executive, especially in view of the Council’s role in, inter alia, the second and third pillar. That, however, is a matter falling outside the scope of this Chapter. A comparison of turnout results can be found on Parliament’s website: . E.g. Ponzano, ‘La réforme des institutions de l’Union européenne dans le cadre de la Constitution’ (2004) at 33 believes it. Article 10 of the ECSC Treaty in its 1951 text. Article 213 TEC and 127 Euratom. Bradley, ‘The Institutional Law of the European Union in 2000’ (2002) at 289 and same author,

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

99

Following the Merger Treaty, the Treaties set the number of Commissioners at the total number of Member States plus the number of large Member States. Moreover, no Member State was allowed to provide more than two Commissioners. The letter of the Treaties thus allowed small Member States to provide two Commissioners. Nevertheless, the important unwritten convention developed that the large Member States proposed two, and the smaller Member States one candidate each. After each Member State had selected its candidate(s), a conference of Member State governments was called (the practice grew for this to coincide with a European Council meeting). Since the Commission is independent from the Council, it was this conference rather than the Council that appointed the Commissioners,7 even though there was almost personal identity between (European) Council and conference.8 It has been pointed out that the difference in practice between a unanimous Council decision and the pre-Maastricht system was that the former allowed for abstentions whereas the latter did not.9 Of course, in practice abstentions could hardly have come into play anyway, since each Member State de facto appointed its own Commissioners. Additionally, the Commission president is part of the European Council but not of the Council, making the personal identity incomplete. Under this system the Commission had little democratic legitimacy. This was no great problem for the High Authority, which perhaps was comparable more to a large agency than to an executive.10 However, it became an issue where the traité-cadre EEC Treaty was concerned. For this reason, the centre of legislative gravity in the EEC was laid in the Council.11 Soon the custom developed that, in principle, the Member States accepted the candidates of other Member States unless there were strong reasons not to.12 The reverse occurred occasionally: when the French government decided not to extend the candidature of incumbent Euratom Commission president Mr Hirsch, it raised comments from the other Member States. Ultimately, the irritated French government threatened to block the appointment procedure unless the other Member States accepted its position. This tactic worked.13 None of this was law or convention, but rather a practice propelled by the expectation that vetoing a candidate would be returned in kind. Moreover, there may have been a realisation that the appointment of Commissioners was, to some extent, a matter decided by the domestic political dynamics of the appointing Member State. The custom was not always followed. Some smaller Member States would occasionally voice objections to a particular candidate. In 1957 the Netherlands objected to a German candidate who, during the Second World War, had signed the order expropriating the House of Orange. In 1970 some Member States successfully opposed the appointment of Mr Achenbach, another candidate with a doubtful war

7 8 9 10 11 12 13

‘Institutional design in the Treaty of Nice’ (2001) at 1100-1101. Van Miert, ‘The appointment of the president and the members of the European Commission’ (1973) at 258. Grabitz a.o., Direktwahl und Demokratisierung—Eine Funktionenbilanz des Europäischen Parlaments nach der ersten Wahlperiode (1988) at 502. Van Miert, op. cit. footnote 7 at 258. Majone, ‘Delegation of regulatory powers in a mixed polity’ (2002) at 330 goes so far as to consider even the current Commission as an agency—a view denying its political aspects in a very extreme fashion. A point cogently made by Lenaerts and Van Nuffel (Bray, ed.), Constitutional law of the European Union (1999) at 8. Van Miert, op. cit. footnote 7 at 259. Ibid. at 264.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

100

past.14 There have also been instances of countries co-ordinating the political colour of their respective candidates. However, these exceptions confirmed the practice that Member States tended to accept each other’s candidates. From early on this system was heavily criticised15 as depriving the Commission of any political colour, making it unwieldy,16 unduly clipping the duration of the president’s appointment and sometimes17 turning it into a ‘pensioner’s home’ for unwanted politicians. As far as size is concerned, Ehlermann’s complaint in this direction, voiced in a time when the college consisted of fourteen persons,18 might mildly amuse those who followed the debates of the last decade on the reform of the Treaties. Continuous enlargements multiplied Commission portfolios, and, since all of these Commissioners needed to be entertained with sufficient policy, fragmented responsibilities.19 As early as 1980, Parliament noted the ensuing co-ordination problems.20 Certain Member States preferred to restrict the duration of the Commission Presidency so as to weaken that office. Under the pre-Maastricht rules, the president and vice-presidents were appointed for two years. In the 1970s a gentlemen’s agreement existed among the Member States under which the president’s two-year mandate could be renewed only once and then only by exception. Moreover, each country would have its turn in proposing a candidate for the Presidency, whereby Belgium and the Netherlands counted as one, as did Denmark, Luxembourg and Ireland.21 A more insidious consequence of the old appointment procedure was the entanglement of interests. 22 Several incidents occurred which suggested that the guarantees for the Commission’s independence were not to be found in the mode of its appointment. For example, as early as 1971, the complaint was heard that Commissioners had consulted ‘their’ respective governments on draft agricultural measures.23 In 1973, French minister Chirac accused Commissioner Cheysson of having betrayed ‘the French interests’.24 In 1988 the British government refused to renominate Lord Cockfield, apparently as a result his vigorous support for the Internal Market project.25 If in 2002 Commissioner De Palacio blocked the reform of the 14 15

16 17 18 19 20 21 22 23 24 25

Ibid. at 259. See for example the Faure report of 7 November 1960 (Parliament doc. 1960-1961, nº 64); Parliament’s Résolution sur les compétences et les pouvoirs du Parlement européen, JO (1963) 1916; Resolution on the relations between the European Parliament and the Commission of the Community with a view to the forthcoming appointment of a new Commission, OJ (1980) C117/53, § 7; Article 25 of the 1984 Draft Treaty establishing the European Union (Spinelli report). Constantinesco, ‘Die Institutionen der Gemeinschaft an der Schwelle der 80er Jahren’ (1981), espec. at 237. As Ms Thatcher experienced with both Lord Cockfield and Mr Brittan, this could sometimes backfire. ‘Das schwierige Geschäft der Kommission’ (1981) at 339. On the ensuing co-ordination problems, Mandelkern Group on Better Regulation, Final Report of 13 November 2001 at 64. Resolution on the relations between the European Parliament and the Commission of the Community with a view to the forthcoming appointment of a new Commission, OJ (1980) C117/53, § 5. Van Miert, op. cit. footnote 7 at 266. Van Miert, ‘Benoeming van voorzitter en leden van de Europese Commissie’ (1972) at 1687-1688 noted the risk as early as in 1972. Written question 10/71 by Mr Vredeling to the Commission, OJ (1971) C61/1. Kapteyn and VerLoren van Themaat (ed. Gormley), Introduction to the law of the European Communities (1998) at 196. ‘The Economist looks at the power of governments to appoint their choice of commissioners’,

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

101

fisheries policy due to a direct order from the Spanish government, this would be a questionable interpretation of Article 213(2) TEC.26 Lastly, Parliament’s influence on the appointment was minimal and its freely offered opinion was freely ignored.27 Both the Treaties of Maastricht and Amsterdam amended the later Article 214. Subsequent to the latter Treaty, the ‘governments of the Member States’ nominated the Commission’s president-elect ‘by common accord’, which nomination needed the approval of the MEPs. Following this, the other members of the Commission had to be nominated ‘by common accord’ with the president-elect. At least in theory, this strengthened his position.28 The whole team then had to obtain Parliament’s approval, following which the Member States appointed them ‘by common accord’. Importantly, and following Parliament’s wish,29 the term of office of the Commission was aligned to that of Parliament. The procedure now went somewhat in the direction of Parliament’s wishes30 but was still flawed, for essentially two reasons. In 1999, Mr Prodi, whose Commission was formally nominated under the pre-Amsterdam procedures, discovered that the president-elect’s influence on the composition of his team was still not overwhelming. In practice most governments continued to present one candidate for each post.31 Moreover, even under the post-Amsterdam procedures, Parliament had only the possibility to approve or disprove the whole team when it found individual Commissioners objectionable. Following the déconfiture of the Santer Commission, Parliament used its muscle to unilaterally extend its rights of control from the choice of Commissioners to the organisation and programming of the institution.32 The 2000 Framework Agreement concluded between Parliament and Commission mirrored this by providing for the incoming Commission to present ‘as soon as possible, its political programme, containing all its proposed guidelines for its term of office’ and to establish a dialogue with Parliament.33 Moreover, the MEPs warned the Member States not to agree to a

26

27 28 29 30

31 32 33

Economist of 30 July 1988 at 16. For the accusation and the details of the latter affair see ‘De Palacio facing “quit” calls over role in fisheries battle’, European Voice of 2 May 2002 at 1, ‘Polemics and suspicions behind the reshuffling of some Directors General in the European Commission’, Agence Europe of 1 May 2002 and Parliament’s plenary debate of 29 May 2002. As far as the appointment of the second Delors Commission iis concerned see Bradley, ‘Legal developments in the European Parliament’ (1993) at 525. Barents, ‘Het Verdrag van Amsterdam en het Europees gemeenschapsrecht’ (1997) at 358. Institut d’études européennes de l’Université Libre de Bruxelles, Les conférences intergouvernementales avant le Conseil européen de Maastricht (1991) at 26. Parliament set its hopes on strongly influencing the composition and policies of the Commission: Resolution on the investiture of the Commission, OJ (1994) C128/358. Mr Santer indicated in 1994 that he would withdraw his candidacy as Commission president if he failed to garner the support of a majority of MEPs: Dehousse, ‘European institutional architecture after Amsterdam: parliamentary system or regulatory structure?’ (1998) at 610-611. ‘Greep Prodi op Commissie niet gehonoreerd’, NRC Handelsblad of 19 June 1999. Resolution on the institutional implications of the approval by the European Parliament of the President of the Commission and the independence of the members of the Commission, OJ (1999) C104/59. That resolution also called for a stronger position for the Commission President. Framework Agreement on relations between the European Parliament and the Commission, OJ (2001), C121/122, § 7. See currently § 28 of the Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125. The European Council supported the political programme idea, but made sure it would be proposed following an exchange of view with the heads of state or government: Part II.II of the Berlin European Council 24 and 25 March 1999 Presidency Conclusions.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

102

candidate president before the elections34—with little success, since Mr Prodi was proposed as candidate-president before the political colour of the new Parliament was known. Furthermore, Parliament foresaw ‘negotiations’ with the president-elect inter alia on ‘the undertakings given by the latter with regard to the political guidelines he intends to follow in office, the quality of interinstitutional relations [and] the criteria he will follow when cooperating with the governments over the nomination of members of the Commission’.35 Lastly, the MEPs requested the president-elect to make a ‘statement of intent’ in Parliament, which would be followed by a debate.36 The Nice Treaty again brought some important changes and formalised the president’s improved status.37 Firstly, rather than the ‘governments of the Member States’ it is now the Council meeting in the composition of heads of state or government who nominates the president-elect.38 It has been suggested that from this point, the Commission lost its claim not to be subordinate to the Council.39 I cannot agree. The Commission does not become subordinate to an institution, merely because that institution has the formal power to appoint. By comparison, the British government is not subordinate to the Queen, merely because she formally appoints the ministers. The main relevant factor is not who makes the final appointment, but which institutions can influence it. Moreover, the Commission can only be sacked by Parliament,40 but not by the Council. Secondly, the Council has to decide with qualified majority. Since in practice the choice is made during European Councils,41 which operate with consensus, this change is not extremely important in practice. Thirdly, although the president-elect and the Member States would still together draw up the list of the other Commissioners, this list has to be ‘in accordance with the proposals made by each Member State’. At first sight this seems to leave little room of manoeuvre to the president-elect. However—as will be demonstrated below—, reality may be more complex, because Parliament is required to give its vote of approval to the new Commission. Even following the vote in Parliament, the executive is not appointed. This happens only by the Council, ‘acting by a qualified majority’. That very last step makes the Nice procedure ambiguous. Does the Council retain any freedom of decision? If the Parliamentary vote is decisive, 42 there is arguably little need for a Council decision. Alternatively, if such a decision has no discretion to it, it obviates the need for qualified majority. It is unlikely that the ministers would veto a Commission they themselves proposed, but the voting requirement suggests that a negative decision remains possible even after Parliament’s vote of investiture. 34 35 36 37 38 39 40 41 42

OJ (1999) C104/59, §§ I.3-4. Ibid., § I.5. Ibid., § IV.3. Calls for this were made by inter alia Von Weizsäcker, Dehaene and Lord Highbury, The institutional implications of enlargement—Report to the European Commission of 18 October 1999. Bradley, op. cit. footnote 6 (first) at 289-290, referring to case 297/86, CIDA vs Council, [1988] ECR 3531, notes that the direct consequence of this is that the Courts can review the decisions to nominate the president-elect and appoint the Commission. Bradley, op. cit. footnote 6 (second) at 1105. Apart from the special case of Articles 216 TEC and III-349 Const. See for the nomination of Mr Barroso the Press release following the 2595th Council meeting (at Heads of State or Government level), held at Brussels, Council doc. 10995/04 of 29 June 2004 at 2. The MEPs were quick to cling on to their right of ‘assent’: Resolution on the institutional implications of the approval by the European Parliament of the President of the Commission and the independence of the members of the Commission, OJ (1999) C104/59 at D.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

103

Construing the Council’s vote as an indication that Parliament’s ‘vote of approval’ is merely indicative would negate the meaning of the words ‘after approval by the European Parliament’ in Article 214(2) TEC. The only possible conclusion is thus that both Parliament and Council have the possibility to block the appointment of the Commission. Moreover, even though the qualified majority allows for appointment with a minority of Member States opposing, the whole body of Commissioners, including the ones coming from opposing states, is appointed. The next step in the development of Parliament’s prerogatives in this field was taken when Parliament initially refused the Barroso Commission the right of investiture. This matter is discussed in some detail in sub-section 4.6.1. Lastly, in 2005 Parliament developed guidelines for the approval of the Commission which, somewhat optimistically, were based on the assumption that the Constitution was to enter into force in 2006. The guidelines spell out the criteria for assessment, the organisation of the hearings and their follow-up.43 This brief overview shows that the procedures to appoint the Commission underwent considerable changes over the past fifty years. The letter of the Treaties provided Parliament with some influence on the composition of the executive. However, this influence is mostly negative, in the sense that Parliament can block the appointment of the whole College. In practice, though, it has started to use this power to refine its influence. Moreover, the Constitution would further amend the primary law procedure. 4.3

The Commission under the Constitution

Since (part of) the appointment procedures laid down by the Constitution would enter into force (if at all) only in 2009, there is little practice or new convention to guide their application. This loss is somewhat compensated by the fact that the text of the Constitution spells the procedures out in much greater detail than the Treaties did. The regular appointment procedure under these rules is regulated in Articles I-26, I-27 and—as far as the Minister for Foreign Affairs is concerned—I-28 Const. Article I-26.5 Const provides that the first Commission appointed under the terms of the Constitution (including its president and the Minister of Foreign Affairs) would consist of one member from each Member State, leading to a assembly of possibly 27 or 28 people (depending on the number of countries that accede to the Union). The following Commissions will consist of two thirds of the number of Member States, unless the European Council decides with unanimity to alter this number.44 At the time of writing, it is uncertain whether the Constitution will enter into force. However, it does seem likely that, even if the provisions on the Commission’s appointment do not find their way into law with the Constitution, a future amendment of primary law will be inspired by them. This justifies a discussion in the following sections of the appointment of the Commission president (section 4.4); of the 43 44

For the details see Report on guidelines for the approval of the European Commission, Parliament doc. PE 355.359v03-00 of 7 June 2005. M. Giscard d’Estaing’s Convention had proposed the creation of ‘non-voting Commissioners’. The idea was criticised by the Commission (who dubbed them ‘castrati’) and in academia: A Constitution for the Union—Opinion of the Commission, pursuant to Article 48 of the Treaty on European Union, on the Conference of representatives of the Member States’ governments convened to revise the Treaties, Commission doc. COM (2003) 548 of 17 September 2003 at 5, Geelhoed, ‘Een Europawijde Europese Unie: een grondwet zonder staat?’ (2003) at 294, and ‘“Elk EU-land een commissaris!”Euro-commissaris Neil Kinnock mordicus tegen “castraten-commissarissen”‘, NRC-Handelsblad of 16 September 2003. It was dropped by the IGC.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

104

proposed MFA (4.5) and of the other Commissioners (4.6). This section will make some general observations on the Commission’s size; nationality requirements; and its political colour. The Commission’s size became a sensitive issue among the drafters of the Constitution. Strangely enough, as early as 1996, Parliament pleaded for maintaining at least one Commissioner per Member State.45 This was the Commission’s stance as well,46 with the argument that the legitimacy of its decisions would suffer if not all Member States are equally represented. M. Giscard d’Estaing proposed a team counting fewer members than there are Member States. Especially the smaller Member States insisted on keeping ‘their’ Commissioner, partly for fear that a small Commission would become a more administrative and less political body, partly for reasons of legitimacy of the EU,47 and partly for the reason of having a channel of influence in the institution. To my mind, all of these points are fallacious and a wrong approach to the problem. Firstly, as noted in section 4.2 above, in the original Treaties there was no automatic link between the number of Commissioners and the Member States and certainly no ‘right’ to a (second) Commissioner. Secondly, the fears that the institution will suffer from the (perceived) lack of representation of all Member States are, in my view, misplaced. Of course, it were somewhat naive merely to refer to the Commissioners’ duty not to seek or take instructions from any party,48 although clearly this injunction has an important role to play by giving the Commissioners a weapon to fence off overly intrusive national governments. Indeed, often the Member States overtly or secretly use their direct ‘line of communication’ with the College. At times this can be useful for all parties. The Commissioner can provide first-hand input about the situation in his Member State. On the other hand, the Commission can use his political connections at home. This argument should, however, not be pushed too far: having one Commissioner in a College of nine gave the Member States in the 1960s indeed a powerful communications channel. However, it remains to be seen whether the voice of a lone Dutch, Lithuanian or Greek Commissioner is loud enough to be heard in an assembly of twenty-seven or more Commissioners. The Executive Board of the ECB provides a precedent for an institution not counting nationals from all Member States49 and the experience with that body indicates, if anything, that its small size reinforces an esprit de corps that rises above the national interests. In a College ‘representing’ all nationalities a Commissioner can morally ‘justify’ an intervention on behalf of his Member State with the argument that his action will be, or at least could be, balanced by that of his colleagues. In a College not including all nationalities, ‘nationally-minded’ actions assume a more odious character since they cannot be balanced by ‘un-represented’ nationalities. Actions on behalf of a particular Member State could then have repercussions for the State involved the next time the Commission has to be re-appointed. This would force 45 46 47 48 49

Resolution on (i) Parliament’s opinion on the convening of the Intergovernmental Conference; and (ii) evaluation of the work of the Reflection Group and definition of the political priorities of the European Parliament with a view to the Intergovernmental Conference, OJ (1996) C96/77. A democratic constitution for the European Union, Speech by Mr Prodi, Commission doc. Speech/03/381 of 3 September 2003 and Commission doc. COM (2003) 548 at 5. Best, ‘The Treaty of Nice: not beautiful but it’ll do’ (2001) at 3, Mattera, ‘Les zones d’ombre du projet de Constitution dans l’architecture institutionnelle de l’Union: la composition de la Commission. La Convention touche au cœur du “projet européen”‘ (2003) at 11. Article 213(2) TEC, copied in this respect by Article I-26.7 Const. In this direction the Club de Florence, Europe: l’impossible statu quo (1996) at 128.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

105

Commissioners to be much more careful in playing the ‘national card’ and actually help to ‘de-nationalise’ the institution. Ultimately, Article I-26.6 Const proposes a system whereby the Commissions following the first Commission appointed under the Constitution (including the president and the MFA) would consist of a number of members corresponding to twothirds of the Member States. After the accessions of Bulgaria, Rumania, Croatia and possibly two or three other Balkan countries, this would still bring the Commission’s membership to about twenty. This solution—a heavily fought political compromise—still starts from the wrong premiss. Ideally, the reform debate should have started with the question ‘how many identifiable functions are there?’ rather than with the question ‘how many people do we have to fit functions to?’ One glance at the functions performed by the institution suggests that a team of fifteen is probably not a bad number. The reduction of the number of Commissioners hopefully will lead to a reduction of the number of directorates-general. The resources thus freed from fighting inter-DG turf wars, mutual controls and duplication of work are massive and could be used to good effect in understaffed policy areas. This in turn, together with the strengthened internal coherence of the institution, would translate in a much more able and less bureaucratic political actor than the current giant is. If anything, a smaller College will enhance the institution’s coherence. The successive expansions of the College led to the splitting up of portfolios in an attempt to provide every Commissioner with some tangible policy area. Reducing the number of Commissioners makes it possible to reduce the trend. The example of the Swiss Conseil fédéral, which operates in a federation of 23 entities, clearly shows that it is possible for an executive to be small, effective and above all, neutral as regards local interests. Nationality requirements. As far as their nationality is concerned, Article 213(1) TEC merely provides that Commissioners have to be nationals of Member States, that the executive must number at least one, but no more than two, nationals from each Member State. The Constitution’s rules necessarily are more complicated because the Commission eventually will have fewer members than there are Member States. Essentially, the viable options for the selection of Commissioners fall between two extremes. One the one hand, one could leave the president-elect a free hand (possible with a restriction on the maximum number of candidates from any single Member State). Such a system would imply a vote of confidence in the European system and would have emphasised the supranational character of the Commission. It would have fitted the principle that the Commission should be autonomous in its organisation.50 The other extreme, preferred by the Convention and following the solution adopted in Nice,51 is to have the Member States select the order in which Member States have a ‘right’ to a Commissioner. That presumes a decision setting out this order, which the European Council would adopt, and which is based on two principles (Article I-26.6 Const): (a) Member States must be treated on a strictly equal footing as regards determination of the sequence of, and ‘the time spent by’, their nationals as Members of the College. The ‘time spent by’ clause may create difficulties: if a Commissioner is sacked, his or her Member State will automatically be entitled to fill the post. That 50 51

See sub-section 2.4.4, above. Article 4(2) of the Protocol on the enlargement of the European Union adopted in Nice, OJ (2001) C80/49.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

106

may not always be easy for the very small Member States. For them, finding one qualified candidate may be difficult, finding two would be worse. Article I-26.6 also provides that ‘the difference between the total number of terms of office held by nationals of any given pair of Member States may never be more than one’. This system ensures equality between Member States, but not between Commissioners. It certainly does not guarantee that the appropriate man gets the job. (b) Subject to point (a), each successive College must be so composed as to reflect satisfactorily the demographic and geographical range of all Member States. In other words, there should be a spread between the different corners of the Union as well as between large and small Member States. One may wonder whether it was necessary to carve this in the Constitution. It is hardly likely that a Commission with no North Europeans or with only one Central European would seriously be contemplated. The Commission’s political colour. As early as the 1970s complaints were heard that the appointment system ensured that the Commission reflected all main streams of the European political spectrum.52 On the other hand, some argued53 that the Commission should be representative for the major political points of view and social economic interest groups. For this reason both the United Kingdom and Germany tended to nominate one of their Commissioners from the ranks of the (largest) government party and one from the largest opposition party.54 As late as 1999, president-elect Prodi made a point in composing a college that should rely on broad support in Parliament. Part of academia opposes the idea of a more politicised Commission.55 An argument could indeed be made that the European constitutional structure does not have the same deep historical roots as the political systems in most Member States and that an executive representing all major political views would gain somewhat in legitimacy what is lost in distance to the electorate. Indeed, at first sight, it would seem that a politically-coloured Commission would find it harder to deal with Member State governments of a different colour. Moreover, as the Swiss Conseil fédéral shows,56 it is not impossible to conceive a homogenous executive of several political parties and cultural backgrounds. On the other hand, however, very strong objections can be fielded against such a view. Firstly, the EU is an entity that lacks the long historical legitimacy, cultural coherence and relative communality of interests that characterise Switzerland. Rambow57 noted in the early 1980s that the very mixed composition of the Commission led to a lack of homogeneity, which in turn impaired its capability to conduct coherent policies. In addition, Commissioners tended to only get acquainted 52 53

54

55 56 57

Patijn, Het Europees Parlement—de strijd om zijn bevoegdheden (1973) at 96. Van Miert, op. cit. footnote 22 at 1686. See also Frowein, ‘Die rechtliche Bedeutung des Verfassungsprinzips der parlamentarischen Demokratie für den europäischen Integrationsprozess’ (1983) at 304 and Touscoz, ‘Brèves remarques juridiques sur les institutions de l’Union Européenne après le Conseil Européen de Thessalonique’ (2003) at 423. At the occasion of the nomination of the Prodi Commission, Chancellor Schröder’s government broke with tradition by nominating Ms Schreyer from the Greens, the junior coalition party, even though the Conservative CDU/CSU had gained in the European Parliament elections: ‘Greep Prodi op Commissie niet gehonoreerd’, NRC-Handelsblad of 19 June 1999. Dashwood and Johnston, ‘The institutions of the enlarged EU under the regime of the Constitutional Treaty’ (2004) at 1486. Frowein, op. cit. footnote 53 at 304 pleads for ‘Konkordanzdemokratie’, in which all major political colours remain represented in the executive. Kriesi, Le système politique suisse (1998) at 229-230. ‘Möglichkeiten und Grenzen der Verbraucherpolitik im Gemeinsamen Markt’ (1981) at 249.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

107

after they have been appointed; which again did not enhance coherence.58 Moreover, a system whereby the executive lacks inherent political consistency is difficult to match with the institution’s parliamentary responsibility, unless some sort of mechanism to provide coherence is foreseen. In the Commission this mechanism is collective responsibility, but in the past this did not always work as well as the Commission’s position called for. Secondly, the legitimacy of the European political system is not necessarily improved by a politically all-inclusive executive without opposition. Most citizens are used to governments relying on parliamentary majorities. They see political debates in their national parliaments that have relevance because the decision makers and the opposition put forward different opinions. A European executive constituted from all major political colours will, almost by definition, make it impossible for Parliament to provide genuine opposition. The consequence of this is that the accountability of the Commission to Parliament suffers (at least in the public’s eye). Additionally, debates in Parliament rarely demonstrate the liveliness that characterises most of its national counterparts. Most importantly, it causes citizens to confound all institutions with an invisible and amorphous ‘Brussels’ that takes decisions far removed from their sight and influence. More political colour in the European Commission will spice its relationship with Parliament and enhance the parliamentary system in Europe. As discussed below, the Constitution would break away from the tradition of consensus politics, albeit in an incomplete manner. 4.4

Appointment of the president

Unlike what some59 have concluded, the Constitution in some respects radically breaks with the current procedure for the appointment of the Commission president. Article I-27 Const reinforces Parliament’s influence over his election largely along the lines the MEPs wanted it. 60 Rather than approving the nomination of the person which the Council ‘intends to appoint’ as Commission president (Article 214(2) TEC), the MEPs would elect the candidate proposed by the European Council.61 The innovations start with his selection by the European Council. The Constitution provides for the European Council to select its candidate with qualified majority. Given the history of working by consensus in this institution, it remains to be seen whether it will actually do so in practice. Perhaps its expansion after enlargement will render consensus unpractical. The practical implication of the provision lies most likely in the possibility to override the resistance of one or two members. Article I-27.1 Const exhorts the European Council to select its candidate ‘[t]aking into account the elections to the European Parliament’. These few words may embody a much more radical change in the position of the Commission than the whole rotation system. 58 59 60

61

Compare the Spierenburg Report of 24 September 1979, § 14. I.e. Ponzano, ‘La réforme des institutions de l’Union européenne dans le cadre de la Constitution’ (2004) at 30. Declaration on the election of the President of the Commission by the European Parliament, OJ (2004) CE38/332. See also ‘“European Constitution” Intergroup calls for Commission president to be elected by Parliament—Germany not in favour of European Council “with executive power”, stresses Jo Leinen—Carlos Carnero Gonzalez on separation of powers’, Agence Europe of 17 October 2002 at item 9. Blumann, ‘Les institutions de l’Union dans le cadre du Traité établissant une Constitution pour l’Europe’ (2005) at 366 rightfully considers this a major change.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

108

As noted above, in 1999 the European Council did not await the European elections before suggesting a candidate president. Similarly, the debate on the selection of his successor started already in 2003. More than half a year before the elections there was already a consensus that the successor should be Christian Democrat.62 In other words, under the current rules, the electorate’s opinion is virtually irrelevant for the political colour of the Commission president.63 Under the system as foreseen in the Constitution, the European Council could not propose a candidate until after the elections have taken place. Moreover, the candidate’s political colour should roughly reflect the tendency in which the elections went. This requirement is tricky: if it were interpreted as implying that the presidentelect should come from the largest group in Parliament, it would automatically reduce the choice to either a Christian Democrat/Conservative or a Socialist. Presumably, the Convention would not have wanted to limit choice in this sense. However, one may expect politicians from the largest group to claim the right to provide the presidentelect. The runner-up group then will have a natural interest in agreeing a common platform with smaller groups (although this factor probably will not lead to fullyfledged coalitions as they exist in some of the Member States). One may speculate that the increased importance of the president may thus become a factor polarising Parliament into two blocks. Article I-27.1 Const provides that the European Council, if its candidate were not approved by Parliament, must propose a new one within one month. That might prove to be a logistical nightmare if it means that the European Council has to reconvene on short notice. More likely, the heads of state or government will make sure that their candidate is sufficiently moderate to garner the support of at least the PPE-DE and PSE groups so that the situation will not occur. If he would really prove unpalatable to a majority of MEPs, then the new nomination may have to be made by means of a written procedure within the European Council. 4.5

The Minister of Foreign Affairs

4.5.1

The Secretary-General/High Representative

The creation of the Union Minister of Foreign Affairs (MFA) is one of the most radical changes proposed by the Constitution in respect of the old Treaties. His office replaces that of the Secretary-General/High Representative for the common foreign and security policy (SG/HR), who resorts under the Council.64 The combination of the (until then) somewhat hidden Secretary-General of the Council with the new title of High Representative resulted from the Treaty of Amsterdam. It was the outcome of the far from satisfying experiences with the CSFP as it had operated since its inception. Governments realised that the purely intergovernmental system agreed in Maastricht precluded decisive action.65 In this 62 63 64 65

‘Postenkarussell beginnt sich zu drehen’, Die Presse of 4 November 2003. Beukers, ‘Enhancing Parliamentary control over the European Commission and the Member States’ (2006) at 23 argues that the appointment of Mr Barroso probably caused a rule pre-empting Article I-27(1) Const. This, however, seems optimistic to me in view of the facts. As Piris, The Constitution for Europe—A legal analysis (2006) at 147-148 points out, he is ‘triplehatted’, since he would also take over some of the CFSP functions currently excercised by the Council presidency. Some of the key texts on the origin of the SG/HR are found in 1996 Intergovernmental Conference (IGC’96)—The Reflection Group Report and other references for documentary purposes (1995) at 26-27, Commission opinion—Reinforcing political union and preparing for enlargement of 28

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

109

respect, the main question before the IGC was, whether the new office should be of a political or of a civil servant nature. The Dutch draft for the Treaty of Amsterdam limited the appointment of the Secretary-General (by the Council with qualified majority) to five-year terms.66 This was deleted in the final text of Article 26 TEU but became practice anyway.67 Such a time restriction would have rendered the function with a more political than civil servant character. The solution chosen in Amsterdam was a compromise. Notwithstanding the great ability of the person filling the post, from a separation of institutions perspective it was not ideal. The SG/HR is somewhere in the middle of a civil servant and a political figure.68 On the one hand, in terms of appointment, he inclines to the former. He is not politically responsible towards any parliament. On the other hand, the nature of the Secretary-General’s work can hardly be described as apolitical. Article 207(2) TEC explicitly devolves the running of the Council Secretariat to the Deputy Secretary-General. The Secretary-General is intended to function in global politics with a high profile. Parliament clearly considers the office to be political in character, as is attested by Rule 85(2) of its Rules of Procedure. The Reflection Group for the 1996 IGC saw no major role for Parliament,69 which institution had declared itself squarely against the institution of a High Representative outside the Commission.70 Since, as a matter of principle, the institutions are autonomous in appointing their secretaries-general, the MEPs got little say over the appointment of the supreme foreign policy officer. Parliament did not peaceably agree to its exclusion from the adoption procedure and adopted (the current) Rule 85 in its Rules of Procedure, foreseeing a procedure whereby it would invite the SG/HR before he would take up his duties, to make a statement and answer questions to the responsible committee. This could be followed by a recommendation at the initiative of the responsible committee. This procedurewhich does not bind the Councilis hardly impressive and the effects of any recommendation (which follows the appointment) are limited, to say the least.

66 67

68 69 70

February 1996, § 28, Section III of the ‘Presidency report to the European Council on the Progress of the Conference, presented by the Italian government to the Florence European Council of 21 and 22 June 1996’, in Intergovernmental Conference on the revision of the Treaties—Italian Presidency—Collected texts (1997) at 105-106); the ‘General Outline for a draft revision of the Treaties submitted by the Irish Presidency of 5 December 1996’, in Intergovernmental Conference on the revision of the Treaties—Irish Presidency—Collected texts (1997) at 66; and the Presidency Conclusions on the Intergovernmental Conference of the Dublin European Council of 13 and 14 December 1996 (ibid. at 139). Article J.14 of the General outline for a draft revision of the Treaties, submitted by the Dutch Presidency in early 1997, reproduced in Intergovernmental Conference on the revision of the Treaties—Dutch Presidency—Collected texts (1998) at 58. Article 1 of Council Decision of 13 September 1999 appointing the Secretary-General, High Representative for the Common Foreign and Security Policy, of the Council of the European Union, OJ (1999) L248/33. The succeeding five-year appointment decision is published in OJ (2004) L236/16. Once the office was created, Parliament urged ‘that the High Representative should be a politician, rather than an administrator’: Resolution on the decision-making process in the Council in an enlarged Europe, OJ (1999) C150/353, § 26. Reflection Group Report at 27. Resolution embodying (i) Parliament’s opinion on the convening of the Intergovernmental Conference, and (ii) an evaluation of the work of the Reflection Group and a definition of the political priorities of the European Parliament with a view to the Intergovernmental Conference, OJ (1996) C96/77, § 18.9.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

4.5.2

110

Appointment of the Minister for Foreign Affairs

M. Giscard d’Estaing’s Convention proposed that the ‘High Representative’ half of the SG/HR’s office be transferred to a Commission vice-president. Thus the MFA was created. Article I-28 Const provides that the European Council, acting by qualified majority, appoints the MFA ‘with the agreement of the President of the Commission’. That implies that the MFA’s appointment can only follow that of the president. As with the other members of the College, the president is not presented with lists of candidates. In practice this would be the Achilles’ heel of the future Commission: the MFA, as an semi-appointee of the European Council with a special Constitutional role as regards external policy, would display a natural inclination to interpret his portfolio as largely as possible. The Constitution comprised two guarantees intended to keep the cuckoo in check. Firstly, the president has to approve his appointment. In this however, the pressures from twenty-five leaders of government anxious to retain their role in foreign policy may leave him with little choice. It is even more likely that, in the future, the European Council will in practice combine the choice of MFA with the selection of the president-candidate in one political agreement.71 The latter will then have to accept the candidate MFA as part of his nomination by the European Council, putting him on a weaker footing. If this were to become practice, the approval by the president perversely would weaken the latter’s status. The fact that the European Council can dismiss the MFA72 will, if anything, increase the tensions between him and the president. A stronger guarantee for the Commission’s coherence would be provided by the Parliament. The MFA would be a member of the College and, as such, part of Parliament’s vote of investiture. The MEPs would subject him to the same hearing procedure as the Commissioners. On his own, the president(-elect) may be unable to obtain a commitment from the MFA to resign when asked by the president. However, as Parliament has shown at the time of the appointment of Mr Barroso’s Commission, the MEPs may force the Commission’s (and the European Council’s) hand in this respect. 4.5.3

The MFA’s position in the Commission

The MFA would be an ex officio vice president of the Commission.73 This follows proposals made by the Commission.74 He would combine the SG/HR’s clout with the External Relations Commissioner’s budget.75 The MFA would be a peculiar member of the Commission. He would preside over the Foreign Affairs Council. Academically, the view has been defended that the MFA would be forced to follow the Member States’ line more than that of the Commission.76 This would, however, depend more on the issue involved: as regards 71 72 73 74 75 76

As far as the first MFA is concerned, already on 29 June 2004 the European Council decided that Mr Solana would get the job: ‘EU-regeringsleiders bezorgd over grondwet’, Standaard of 30 June 2004. Article I-28.1 Const. Article I-28.4 Const. Communication of the Commission on the institutional architecture, Commission doc. COM (2002) 728 final of 4 December 2002 at 12-13. Maddalon, ‘L’action extérieure de l’Union européenne’ (2005) at 515. Barents, ‘Naar een Europese constitutie? (VI)’ (2003) at 221. Dashwood, ‘The draft EU

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

111

the common foreign and security policy, the MFA would ‘conduct’ external relations, although he would work under the mandate of the Foreign Affairs Council77 and the European Council. As regards other external relations matters, the Commission’s collective responsibility would come into play.78 This explains the curious construction that the MFA would often ‘jointly’ act with the Commission, of which he is a member.79 It probably goes too far to conclude that the MFA would require a ‘schizophrenic personality’.80 The construction chosen by the drafters of the Constitution would probably improve the co-ordination of foreign policy between the Commission and the Council, but not without a certain loss of ‘constitutional transparency’.81 In separation of institutions terms, it is unfortunate that the autonomy of the Commission would be affected. Moreover, the influence of the Member States and the (European) Council on the appointment of the MFA would create two poles of power within the Commission that are likely to negatively affect the coherence of that institution.82 4.6

Appointment of the College

4.6.1

The vote of investiture; the possibility to refuse confidence

The idea of a vote of investiture has a long history.83 Following the first direct elections in 1979, Parliament unilaterally decided to hold a debate and a vote of confidence on the Thorn Commission. Although this vote was perhaps more symbolic than real,84 starting with the Delors I Commission four years later, the Commission members waited for that vote before taking the oath before the Court of Justice.85 The 1975 Tindemans report took a step back by proposing only a vote of investiture for the Commission president,86 but the practice did develop that the European Council consulted the president-elect on his candidate-colleagues. As Commission president Jenkins experienced, the obvious flaw of this system was that

77 78 79 80

81 82 83

84 85 86

Constitution—First impressions’ (2004) at 414 sees the problem, but does not venture a guess which institution would win. Article I-28.2 Const. Lenaerts, Binon and Van Nuffel, ‘L’Union européenne en quête d’une Constitution: bilan des travaux de la Convention sur l’avenir de l’Europe’ (2003) at 295. Articles III-293.2, III-322.1 and III-329.2 Const. In this sense Bulterman, ‘De externe betrekkingen van de Europese Unie in de Europese grondwet’ (2005) at 127. Similarly critical Barents, ‘Een grondwet voor Europa (VII): de externe betrekkingen’ (2005) at 86. I agree with Maddalon, op. cit. 75 at 522, who argues that the MFA’s constitutional position will to a large extent depend on the person filling the post. Geelhoed, ‘Een Europawijde Europese Unie: een grondwet zonder staat?’ (2003) at 309. In a similar vein Cremona, ‘The draft Constitutional Treaty: external relations and external action’ (2003) at 1355. Barents, ‘Naar een Europese constitutie? (V)’ (2003) at 170. Although Parliament did not ask for a right of investiture in its Resolution sur les compétences et les pouvoirs du Parlement européen, JO (1967) 1916, the Council debated the idea of such a vote as early as 1964: Council doc. I/4/64 of 20 March 1964 at 23. Parliament asked to be ‘involved’ in the appointment of the Commission through a ‘right of co-decision’ as early as 1972: Résolution à l’intention de la prochaine conférence au sommet des chefs d’Etat ou de gouvernement des Etats membres des Communautés européennes, JO (1972) C82/26. Constantinesco a.o., Traité instituant la CEE—Commentaire article par article (1992) at 846-847. Institut d’études européennes de l’Université Libre de Bruxelles, op. cit. footnote 29 at 25, Corbett, The European Parliament’s role in closer EU integration (1998) at 136-137 and 259. Bull. EC Suppl. 1/75, ad V.D.2.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

112

the president incurred the risk to have to work with colleagues whom he himself had declined.87 Following the 1981 Genscher-Colombo plan,88 which foresaw consultations between the Council president with Parliament’s Bureau before the nomination of the Commission president as well as an investiture debate without right of censure, the 1983 Stuttgart European Council promised to consult the enlarged Bureau of the Parliament before nominating the members of the Commission. This should be followed by a debate, to be concluded by a vote. Parliament adapted its Rules of Procedure accordingly. Less than two months passed before media leaks forced the Member States to break their pledge when the Delors Commission was nominated89 and the Parliament’s Bureau indicated its dissatisfaction with being briefed, rather than consulted, on the appointment.90 This episode suggested the limitations of how far the Member States could be expected to go. It led the MEPs to negotiate with the Commission an interinstitutional convention under which any new president would discuss with Parliament’s Bureau the Commission’s programme and composition. Following the Commission’s appointment, Parliament would vote on that programme.91 Although the matter was considered in the run-up to the Single European Act92 that instrument did not change matters. In 1988 the re-appointment of M. Delors was handled following the Stuttgart procedure. The Maastricht Treaty introduced the vote of investiture (‘vote of approval’) and made it a condition sine qua non for the appointment of the executive. The requirement that Parliament concedes its confidence to the new Commission of course implies the possibility that such confidence is refused. In 1999, following American practice,93 it held detailed hearings in Parliament of the candidate Commissioners for the Prodi Commission. The debating format of these hearings made it difficult for the MEPs to grill the Commission but, even though these hearings still showed less a barking than a yapping Parliament,94 they somewhat improved the MEPs’ influence on the appointments. The prospective Commissioners consented to the procedure and it has become a standard feature of Commission appointment procedures. This

87 88 89 90 91 92 93

94

Ehlermann, op. cit. footnote 18 at 341. The Project of European Act’, proposed by the governments of Germany and Italy to the European Council of 6-7 November 1981, ad II.3.4. Grabitz, op. cit. footnote 8 at 504. Council Secretariat General note of 26 July 1984, quoted in Digest of texts on relations between the European Parliament and the Council (1988) at 103-104. Joint Declaration of the European Parliament and the Commission, Parliament doc. 1-328/84/rev of 23 May 1984. The 1985 Report of the ad hoc-Committee for institutional questions chaired by Mr Dooge (Spaak II Committee) supported the idea of a vote of investiture. Magnette, ‘Appointing and censuring the European Commission: the adaptation of parliamentary institutions to the Community context’ (2001) at 297, argues that this copying of American procedures suggests that ‘they are conscious of the imbalance in the power of nomination’ between themselves and the Member States. Not necessarily: in the different circumstances of the European constitution, this solution seems to suggest itself as a way to give meaning to the right of investiture. Such hearings were held in 1994 as well, but resisted to the extent possible by the Commission: Beukers, op. cit. footnote 63 at 24-25. More cynical minds qualified them as ‘about as exciting as North Korean politburo elections’: ‘Parliamentary game show yields no clear winner but plenty to laugh at’, European Voice of 29 November 2001 at 11.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

113

interinstitutional convention was only formalised as such in Article 7 of the 2005 Framework Agreement.95 In 2004, president-elect Barroso still faced a refusal from at least most Member States to provide him with a choice of candidates. During the hearings in Parliament96 the Italian, Dutch, Hungarian and Latvian candidates incurred considerable opposition. Mr Barroso publicly stood by his proposed team,97 which hardened that opposition. He then proposed to withdraw some of the files from the responsibility of the Italian candidate, but this did not satisfy Parliament. Ultimately, when it became evident that a majority of MEPs would reject the investiture, Mr Barroso asked Parliament not to vote on the investiture yet.98 In the following weeks, he asked the Italian and Latvian governments to replace their candidates and switched the Hungarian to another position.99 These changes made, Parliament gave its vote of approval on 18 November 2004 and the Council appointed the team shortly thereafter.100 Parliament was quick to claim a victory101 and indeed, the MEPs gained a major concession. However, in my view, the Commission is not the losing party but is strengthened by the affair. As both Mr Prodi and Mr Barroso experienced, the Member States continue to have the tendency to present the president-elect with a single candidate who, more often than not, is selected in function of national politics. For example, in 1999 the British Government favoured fielding Mr Patten, so that he would not succeed Mr Hague as leader of the Conservatives. Parliament’s preparedness to refuse the investiture to the Commission will enable the next president-elect to refuse candidates who, in his estimation, will not garner sufficient support in Parliament. This will provide him with some means to negotiate with the Member State governments. In the long term, the affair may have much a deeper impact. As a multi-national body, the Commission does not have the natural cohesion that even coalition governments in the Member States have. In the past, the Member States deliberately ensured that all major political views were represented in it, to the point where the Commission had no distinct political colour. The politisation of the choice of president-elect caused by Article I-27.1 Const, which requires the European Council to take account of the elections of the MEPs, might be a step in the direction of a situation whereby a majority coalition in Parliament will force the president-elect to appoint a team sharing predominantly a similar political outlook. If this were to happen, it would be the beginning of a division of Parliament into a government bench and an opposition, thereby radically changing the nature of politics in that institution. 95

Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125, § 7. 96 For a very detailed description of the procedure followed see the Report on guidelines for the approval of the European Commission, Parliament doc. PE 355.359v03-00 of 7 June 2005 prepaed by Mr Duff MEP. 97 ‘Parliament leaders order Barroso to shift Buttiglione’, European Voice 21-27 October 2004. 98 Parliamentary debates of 27 October 2004; ‘Barroso defends delay in EU vote’, CNN.com of 27 October 2004; ‘Vertrek Buttiglione volstaat niet voor Europarlement’, Standaard of 30 October 2004; ‘Barroso haalt bakzeil’, Volkskrant of 27 October 2004. 99 The revised list of nominees was published in OJ (2004) L333/12. 100 Council Decision of 19 November 2004 appointing the President and Members of the Commission of the European Communities, OJ (20004) L344/33. 101 ‘M. Barroso retire sa Commission, l’UE confrontée à une crise inédite’, Le Monde of 27 October 2004.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

4.6.2

114

The allocation of portfolios; mid-term appointments; reshuffling

I will analyse several matters concerning the appointment of the Commissioners, focusing on the distribution of portfolios, mid-term appointments, and the reshuffling of portfolios. The distribution of portfolios amongst Commissioners is a matter legally within the sovereign domain of the president102 (in the system of the Constitution, the MFA would be an obvious exception to the rule).103 This apparent legal freedom is, however, in practice bound to some extent. Firstly, there are some traditions which, in practice, are respected. For example, traditionally the Agriculture portfolio is entrusted to a Commissioner from a small country.104 This is not a rule of convention, but mere practice: the stakes of the large Member States in agriculture are so high, that none of their nationals is entrusted with it. Both Mr Prodi and Mr Barroso gave the portfolio to a national from a small Member State to avoid political difficulties, but probably not because they felt obliged to do so. Secondly, notwithstanding the president-elect’s legal prerogatives, Member States may still attempt to push for a coveted portfolio. It depends largely on the power of the president-elect whether he is able to resist such pressures; apparently Mr Barroso had some success in this field.105 Mid-term appointments. There may be several reasons why an office in the Commission may fall open. A Commissioner can resign (voluntarily or at the request of the Commission president), die, or be compulsory retired by the Court of Justice.106 In the first case he remains member of the institution until replaced.107 Article 215 TEC lays down that deceased or (compulsorily) resigned Commissioners are to be replaced for the remainder of their term by a new member, unless the Council decided unanimously not to fill the vacancy. 108 Peculiarly, the Council can appoint the new Commissioner with a qualified majority without even being required to consult Parliament,109 an anomaly repaired in Article III-348.2 Const. The EC Treaty provides for Parliamentary influence in the mid-term appointment of the Commission president by referring to the normal appointment procedure of 214(2) TEC. Note that in such cases, the whole Commission is again subjected to a 102 This follows from Article 217 TEC. A similar principle is laid down in Article I-27.3(a) Const. 103 In Declaration (No 32) on the organisation and functioning of the Commission, attached to the Treaty of Amsterdam, OJ (1997) C340/137, the IGC considered that ‘the President of the Commission must enjoy broad discretion in the allocation of tasks within the College, as well as in any reshuffling of those tasks during a Commission’s term of office’. 104 ‘Chirac polst Veerman voor EU-post’, Volkskrant of 11 June 2004. 105 Mr Barroso apparently resisted a lot of pressure from the large Member States with some success: ‘Germany, France seek top economic posts after compromise on Barroso’, Financial Times of 29 June 2004; ‘New Commission president vows not to yield to pressure from big states’, Financial Times of 30 June 2004; ‘Bold as Barroso’, Financial Times of 15 July 2004; ‘A summer of heated haggling for Barroso’, Financial Times of 29 July 2004. 106 Articles 215 TEC, last paragraph and III-348 Const. 107 Joined cases T-227/99 and T-134/00, Kvaerner Warnow Werft GmbH vs Commission, [2002] ECR II-1205, § 52. Note that in this case the Court was—rightfully—more concerned with the Commission’s independence than with the de facto smaller number of Commissioners. 108 For an application of this provision see Council Decision of 9 July 1999 on the composition of the Commission, OJ (1999) L192/53. 109 For an example see Council Decision of 13 September 2004 appointing a new member of the Commission of the European Communities, OJ (2004) L298/24.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

115

vote of investiture. This is not illogical in view of the Commission’s collective responsibility (discussed in more detail in Chapter 5). The Constitution essentially kept the same identity of procedures for his initial appointment and mid-term appointment.110 At the time of the appointment of the Barroso Commission, Parliament called for a commitment from the Commission that, ‘in the event of a resignation, the replacement Commissioner shall not appear before Parliament or the Council in an official capacity until his or her nomination has been validated by the normal parliamentary procedure (hearing and vote in plenary)’.111 Mr Barroso agreed to these points, save where appearances before the Council were concerned. However, rather than stating that the Commissioner’s appointment should be ‘validated’ by Parliament, he merely agreed, somewhat tautologously, that ‘new Commission members should not appear officially before this Parliament until they have met with the relevant parliamentary committee’, which does not quite go as far.112 Indeed, the subsequent 2005 Framework Agreement concluded between the Commission and Parliament seems to go even less far by merely requiring the presidents of the two institutions to agree on ‘the manner in which the president of the Commission intends to ensure the presentation of the future Member before Parliament without delay and in full compliance with the prerogatives of the Institutions’. From its side, Parliament committed to ‘ensure that its procedures are conducted with the utmost dispatch’ so that the Commission president can be informed of Parliament’s position ‘in due time before the Member is called upon to exercise duties as the Commission’s representative’.113 This suggests that the Commissioner should not act (at least, not in Parliament) until confirmed; but it is a long cry from Parliament’s original wish that such Commissioners be equally barred from acting before the Council, which unrealistic demand was impossible to enforce towards that institution. Reshuffling portfolios. Article 217 TEC enables the Commission president to reshuffle the Commissioners’ portfolios.114 The pre-Nice procedures effectively left the assignment of portfolios in the hands of the Member States. The post-Nice Treaty text seems to empower the Commission president but, as could be expected, Parliament attempts to bind that power. Since the 2000 Framework Agreement, the Commission president is obliged to ‘immediately notify Parliament of any decision concerning the allocation of responsibilities to a Member of the Commission. Where the responsibilities of a Member of the Commission are changed substantially, that Member shall appear before the relevant parliamentary committee at Parliament’s request.’115 There has been only one case of a reshuffling until now (and even that concerned candidate Commissioners rather than the appointed body) when Mr Barroso switched the portfolios of candidate Commissioners Kovács and Piebalgs in order to respond to 110 111 112 113

Idem in Article III-348.3 Const. Resolution on the election of the new Commission, OJ (2005) CE201/113, § 5. Debates of 18 November 2004. Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125, § 4. 114 Article I-27(3) Const is more succinct, but probably not different in contentt. 115 Paragraph 5 of the 2005 Framework Agreement, which succeeds § 11 of its 2000 predecessor.

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

116

pressures from Parliament. This—at first sight innocuous—decision may prove to be a precedent for greater Parliamentary involvement in the allocation of portfolios. 4.7

Conclusions of this Chapter

Hix raises the question whether the post-Maastricht ‘quasi-parliamentary model of EU executive selection really allowed Europe’s voters to chose the EU executive’.116 As far as the president is concerned, the voters have some influence, although less than they have in some of the Member States. An argument could be attempted that the discrepancy between EU and national level is relative. For example, in the Netherlands voters arguably have only limited influence on the composition of the government. In Britain the Prime Minister is only voted into the Commons by a relative majority within his election district. This, however, is not the yardstick by which the problem should be measured. Firstly, as far as the democratic legitimacy is concerned, the Member State systems generally have the advantage that they are known and accepted by the population. Concerns about the influence of the population on the executive may be warranted, but are generally limited for this reason. Secondly, as far as the separation of institutions is concerned, the appointment procedure is an incomplete check on the institutions. It is true that the Commission operates between the Council and the Member States. Even if it had a political colour, its room for manoeuvre would be limited. As a check on the institutions, the reflection of the electorate’s majority preference in the Commission is not as important as its unity and relative independence from the Member States. The greater involvement of Parliament in its appointment procedure serves a purpose as a check on the Member States. The review of the development of the appointment procedure leads to two important conclusions. To begin with, although Parliament’s influence on the composition of the Commission is still imperfect, interinstitutional convention has improved it. It is likely that this trend will continue. That, possibly combined with the entry into force of the Constitution, may well lead to a system whereby Parliament will obtain a very tangible say in the composition of the executive. Additionally, even though the trend has been favourable towards a party-based Commission, this too is as yet an incomplete process. In this author’s view, this is to be regretted: appointing a Commission based on the support of all major political groups in the European spectrum does not enhance its accountability towards Parliament. Moreover, and this is a much greater problem, it reduced the institutional transparency of the European political system. To put it simply, ordinary citizens— who refer to their national political systems as their framework of reference—have difficulties understanding the European politics of an executive without opposition and a Parliament that seems to act as a unified institutional actor rather than as a forum for political debate. Seen from afar, ordinary citizens may easily be forgiven to confound all institutions into one amorphous ‘Brussels’. If casting a vote in the European elections does not seem to influence the composition of the ‘European government’, voting loses much of its relevance in the eyes of the citizen. A comparison of the election turnout in the last European elections with the turnout in 116 ‘Executive selection in the European Union: does the Commission President investiture procedure reduce the democratic deficit?’ (1997).

CHAPTER 4: THE APPOINTMENT OF THE COMMISSION

117

the Dutch and French referenda117 makes this point painfully evident. The consequence is that the standing and influence of Parliament must necessarily suffer from a lack of popular support. From this point of view, the appointment procedure is much more important than the immediate question of which Member States gets influence on what portfolio. It is a matter of vital importance to the legitimacy of the Union’s democratic system. Following these conclusions, it is now time to address the Research Question posed in Chapter 1 as to what extent it is possible to balance the EU institutions through the use of interinstitutional convention and what the conditions are for doing this. The appointment procedures for the Commission have given rise to interinstitutional conventions. As the appointment of the Barroso Commission shows, the Treaties offer considerable scope for further developing interinstitutional conventions that provide Parliament with a firmer grip on the composition of the executive. If anything, the Constitution would only enhance this tendency and would itself give rise to interinstitutional conventions. The key element here is Parliament’s ultimate veto over the new Commission. It is true that this is not lightly used, but the fact that Parliament can credibly threaten with it already puts other institutions in a position where they need to neutralise the threat. This provides the political basis for them to agree to interinstitutional conventions. As was explored in this Chapter, Parliament has successfully used interinstitutional conventions to obtain a firmer grip on the appointment of the Commission. It is likely that more interinstitutional conventions will develop that will enable the MEPs to block the appointment of undesired candidate Commissioners. Parliament’s Treaty-based right of investiture was ultimately the leverage that brought about these interinstitutional conventions.

117 The French turnout in the 2004 European elections was 52.7%, whilst in the Netherlands it was 35.7% (). This compared with a turnout in the 2005 referenda of 70% in France and 62.8% in the Netherlands (‘Referenda on EU Constitution—state of play in the member states’, of 13 February 2006 and ‘Dutch reject EU constitution’, CNN.com of 1 June 2005).

CHAPTER 5: THE PARLIAMENTARY SYSTEM

5

THE PARLIAMENTARY SYSTEM

5.1

Introduction to this Chapter

119

The relationship between the Commission and Parliament is characterised by a parliamentary system.1 That implies, firstly, that the Commission is responsible towards Parliament. Secondly, it means that the MEPs can sanction that responsibility through a motion of censure. Construing the relationship as a parliamentary system may not be obvious, since the political system in the European Union differs from the constitutional set-ups of the Member States in several ways. Firstly, a glaring difference with practice in the Member States is the absence of the right to dissolve parliament and issue new elections, which in many Western European parliamentary systems balances the government’s duty to resign when confidence is absent.2 This absence is linked with the fixed term of Parliament and is less of an exception than it appears at first sight. In many Member States a rule in law or convention connects the duration of office of the government to the term of Parliament. At first sight, both the Treaties and the Constitution suggests the two are not dependent on each other. Articles 214(1) TEC and I-26.3 Const autonomously set the Commission’s term of office at five years. This, however, corresponds to a similar five-year term for Parliament.3 A second difference with Member State practice that might cast doubt on the parliamentary system is the fact that relationships of responsibility are not limited to the Commission and Parliament. For example, Rule 106 of the Parliament’s Rules of Procedure foresees the possibility for the ECB president to present the Bank’s Annual Report, which is followed by a general debate. He is ‘invited’ to attend Parliamentary committee meetings ‘at least four times a year to make a statement and answer questions’. Rule 87 provides for a roughly similar procedure for the SG/HR. One difference between these situations and the Commission-Parliament relationship, however, is that the former cannot be sanctioned by censure.4 Thirdly, in the Member States with the oldest parliamentarian traditions, political responsibility originally referred to the responsibility of the government for the actions of the head of state. For example, in the succinct language of the Dutch

1

2 3 4

Zuleeg, ‘Der Verfassungsgrundsatz der Demokratie und die Europäischen Gemeinschaften’ (1978) at 32. Dehousse, ‘European institutional architecture after Amsterdam: parliamentary system or regulatory structure?’ (1998), 595-627 sees the EU rather as a mix between a parliamentary system and a regulatory one. I would agree with him that it is possible to some extent to consider the EU in terms of a regulatory entity, but I am not sure whether such an analysis is useful in terms of democratic development. Pech, ‘La solution au “deficit démocratique”: une nouvelle gouvernance pour l’Union Européenne?’ (2003) at 134 is not convinced that there is a ‘logique parlementaire’ in the European political system, but postulates rather than demonstrates. Patijn, Het Europees Parlement—de strijd om zijn bevoegdheden (1973) at 40, Cadart, Institutions politiques et droit constitutionnel (1990) at 665. Articles 190(3) TEC and I-20.3 Const. Until the Maastricht Treaty the respective terms of office differed. They were harmonised due to Parliamentary pressure: Corbett, ‘The Intergovernmental Conference on Political Union’ (1992) at 293-294. The Minister of Foreign Affairs under the Constitution, who can be dismissed by the European Council, is a half exception: Article 28.1 Const.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

120

constitution, ‘the King is inviolate; the ministers are responsible’.5 That responsibility was exacted by parliament which, for this reason, obtained rights of oversight over the King’s servants. Such rights included e.g. the right to ask questions from the government, to debate with it, to inquire into its actions, and ultimately, to sack it. Since the European Union has no sovereign, the (European) Council has assumed some of the prerogatives that in states normally befall to the head of state (e.g. the accreditation of foreign ambassadors, appointment of the executive)6 and almost acts a head of state by proxy. In this Chapter, I will first indicate what is meant by collective political responsibility (section 5.2). In section 5.3 its sanctionthe motion of censureis discussed, followed by an analysis of the parliamentary system in practice (5.4). The subsequent section analyses the requested vote of confidence. Finally, section 5.6 examines the Council’s involvement in the parliamentary system. 5.2

Political responsibility

5.2.1

What is political responsibility?

Political responsibility means that the Commission is obliged to justify to Parliament its past, current and planned actions and policies. The instruments by which Parliament exercises this responsibility are manifold: debates, Parliamentary questions, committees of inquiry, the Ombudsman, etc. They are reviewed in the following Chapters of this study. The existence of political responsibility is a question of fact rather than of law. In this sense, the Commission’s responsibility towards Parliament is more evident from the obligation to reply to Parliamentary questions (Articles 197 TEC and III-337.2 Const) than from the declaratory formula of Article I-26.8 Const stating that ‘[t]he Commission, as a body, shall be responsible to the European Parliament’. On the one hand, political responsibility does not entail that the executive has to bow to Parliament’s every whim. On the other, it does imply that ignoring Parliament’s view can ultimately be sanctioned with censure. The Commission then, should not be the slave of Parliament in the fashion of the governments of the French Fourth Republic, which tended to be dominated by their MPs. For this reason, the Commission has been unwilling to adopt automatically proposals promoted by Parliament. ‘I tell you quite frankly’ said president Thorn, ‘that we are not prepared to accept an automatic arrangement, for that would be failing to meet our own responsibilities’.7 It is in the (political) nature of things that the matter does not provoke much case law. The closest the Court of Justice has come to the matter was when it had to determine whether Parliament was entitled to bring its own actions under Article 173 TEEC (now, revised, Article 230 TEC). At the time, the Court initially found that Parliament

5 6

7

Article 42(2) of the Dutch Constitution of 1983. The matter is the subject of an interinstitutional agreement based on Article 162 TEEC (the current Article 218 TEC). It was agreed in 1966 and is described in the response to question 78 to the Commission, OJ (1966) 4049. Under the agreed procedure, the accreditation is automatic and following the transfer of the accreditation letter to the Council and the Commission. Commission President Thorn, Debates of Parliament of 14 October 1981, OJ Annex (1981) 1-275/132.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

121

‘is empowered, on the one hand, to exercise political control over the Commission which, pursuant to Article 155 [currently 211 TEC], is required to “ensure that the provisions of this Treaty and the measures taken by the institutions pursuant thereto are applied”, and on the other, to censure the Commission, where necessary, if the latter should fail properly to discharge that task’.8 However, soon after this ruling, the Judges realised that this ‘political control’ does not imply slavish dependence: ‘while the Commission is required to ensure that the Parliament’s prerogatives are respected, that duty cannot go so far as to oblige it to adopt the Parliament’s position’.9 The Commission makes its own political assessment. Responsibility implies an addressee: responsible to whom? This not-so-straightforward question is tied up with the collective nature of the Commission’s responsibility. A distinction must be made at this point between the Commission’s collective responsibility in law and its political collective responsibility. The Courts noticed collective10 responsibility in law from 1986 onwards11 although rarely in the context of its relationship with Parliament. The Judges deduced from the principle laid down in (the current) Article 219 TEC12 that ‘decisions should be the subject of a collective deliberation and that all the members of the college of Commissioners bear collective responsibility on the political level for all decisions adopted’.13 The Commission’s Rules of Procedure elaborated the principle, by requiring in Article 1 that the Commission ‘shall act collectively’ andsince 1999‘in compliance with the political guidelines laid down by the president’.14 The reference to ‘collegiality’ in Article 217(1) TEC in its post-Nice drafting made explicit what was law anyway.15 The question before the Court in AKZO was whether collective responsibility allowed delegation. The Court allowed such delegation ‘within certain limits and subject to certain conditions’, whilst noting that the Commission remained ‘fully responsible’ for any decisions taken on the basis of it.16 The Court ruled in BASF that individuals ‘affected by the legal consequences of such decisions’ have a right to know that these were taken by the College17 and correspond to its intention.18 From 8 9 10 11 12

13 14 15 16 17

Case 302/87, Parliament vs Council (Comitology), [1988] ECR 5615, § 12. Case C-70/88, Parliament vs Council (Chernobyl), [1990] ECR I-2041, § 19. It has been the guiding principle of the institution from the beginning: Article 1 of the Commission’s Rules of Procedure of 1963, JO (1963) 181/63. Case 5/85, AKZO Chemie BV and AKZO Chemie UK Ltd vs Commission, [1986] ECR 2585, § 30, where the Judges referred to it as ‘principle of collegiate responsibility’. ‘The Commission shall act by a majority of the number of Members provided for in Article 213. A meeting of the Commission shall be valid only if the number of Members laid down in its Rules of Procedure is present.’ A similar rule (minus the quorum requirement) is provided in Article III-351 Const. I thus do not quite agree with Bradley, ‘The Institutional Law of the European Union in 2000’ (2002) at 290 who argue that the Commission’s collective responsibility entered expressly into primary law only at Nice. It has been a very evident priniciple since the beginning. AKZO, § 30. Article 1 of the Rules of Procedure of the Commission, OJ (2005) L347/83. In practice, voting is relatively rare in the College. Most decisions are taken under delegation procedures or by consensus: ‘Meeting of College with Valéry Giscard d’Estaing mainly allows discussion on way Commission works’, Agence Europe of 7 November 2002 at item 3. AKZO, §§ 35-36. To what extent the Court of Instance is willing to go in assessing whether a decision has been

CHAPTER 5: THE PARLIAMENTARY SYSTEM

122

this, it deduced a distinction between ‘straight-forward measures of management’ (which could be delegated) and decisions finding infringements,19 which could not be delegated since they are decisions of principle.20 The BASF judgment caused the Commission to amend its Rules of Procedure,21 but third parties in competition proceedings continued challenging the procedure for the adoption of fines, although not always with success.22 In later case law, the Court of Justice has gone further. In Commission vs Germany, an infringement proceeding, it concluded from the collective (or, in the terminology of the judgment, collegiate) responsibility that the decision to issue a reasoned opinion as well as the decision to apply to the Court for a declaration of failure to fulfil obligations cannot be described as a measure of administration or management and may not be delegated.23 Strangely, the Court referred to the requirement that members of the College bear ‘collective responsibility at political level’24 (something which ipso facto lies beyond the power of the Court to control). Notwithstanding this, it effectively develops a collective responsibility in law requiring ‘that decisions should be the subject of collective deliberation’ by the Commissioners. Referring to BASF, the Court then concluded that affected individuals are entitled to be assured that the operative part of decisions and the statement of reasons were actually adopted by the College.25 Without going too deep into the distinction then made by the Court between ‘pre-litigation stage’ and other acts, the arresting point is the formal requirements for the cuisine interne within the Commission which are deduced from the collective responsibility in law. These requirements can be checked in Court. In other words, the Commission cannot simply assume the responsibility for the decision whatever the exact decision-making procedure used. This ruling is not a priori logical. The Court could have assumed a more deferential attitude towards the autonomy of the Commission to organise itself. If the decision at issue is endorsed by the Commission, it is a Commission decision; if there was a problem within the Commission, one would assume that the institution would either endorse the decision anyway, or decide on its own behalf to withdraw it. In any event, it is these procedural requirements which set the collective responsibility in law apart from the Commission’s political responsibility.

18 19 20 21 22

23 24 25

discussed by the whole College is evident from case T-151/94, British Steel plc vs Commission, [1999] ECR II-629, §§ 126 et seq. Case C-137/92P, Commission vs BASF AG a.o., [1994] ECR I-2555, § 64. See also case T-442/93, Association des Amidonneries de Céréales de la CEE a.o. vs Commission, [1995] ECR II-1329, § 97. BASF, § 71. Case T-254/99, Maja Srl vs Commission, [2003] ECR II-757, § 41. Commission Decision of 8 March 1995 amending its Rules of Procedure, OJ (1995) L97/82. E.g. cases T-275/94, Groupement des Cartes Bancaires ‘CB’ vs Commission, [1995] ECR II-2169, T-218/95, Azienda Agricola ‘Le Canne’ Srl vs Commission, [1997] ECR II-2055, § 35 et seq, joined cases T-305/94, T-306/94, T-307/94, T-313/94 to T-316/94, T-318/94, T-325/94, T-328/94, T-329/94 and T-335/94, Limburgse Vinyl Maatschappij NV vs Commission, [1999] ECR II-931, § 308 et seq. Cases C-191/95, Commission vs Germany, [1998] ECR I-5449, §§ 36 and 37 and C-198/97, Commission vs Germany, [1999] ECR I-3257, § 20. Case C-191/95, Commission vs Germany, § 39. Ibid., § 40 et seq.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

123

Lastly, although the Commission has attempted to deduce from the principle of collective responsibility a justification for the secrecy of the Commission’s deliberations,26 the Court of First Instance did not follow this. The Commission’s political collective responsibility towards Parliament is quite a different animal, with a different scope and serving a different purpose. It is in the nature of things that the Court only rarely had a chance to pronounce on this— political rather than legal—form of collective responsibility. The Court’s statement in BASF that ‘all the members of the college of Commissioners should bear collective responsibility at political level for all decisions adopted’27 (note the ‘should’) could be at best a factual observation. 5.2.2

Extent of the Commission’s political collective responsibility Responsibility and censure

There is no precise formula encompassing the whole extent of the Commission’s political collective responsibility towards Parliament.28 Certainly the matter is not one of positive law but rather of convention and politics. It comprises a personal and a substantive element, both of which are analysed in the sub-sections below. Referral to the text of Article 201 TEC (III-340 Const) does not shed much light on the matter. The sanction of responsibility is the motion of censure (discussed in section 5.3, below) but responsibility and censure are separate matters. It is not even at all clear that parliamentary responsibility and censure cover the same scope. To begin with, a Commission functioning in a caretaker capacity29 following a successful motion of censure remains responsible to Parliament, even though that responsibility can no longer be sanctioned. Compulsory retirement, the EU equivalent of Americanstyle impeachment, is a procedure that can only be initiated by the Commission or the Council,30 leaving Parliament largely unarmed until a new Commission has been appointed. In real life this problem is less dramatic than it appears, since in the last fifty years it has been rare for the Commission to operate in a caretaker capacity for long. In addition, Parliament and the Commission concluded interinstitutional conventions (discussed in the following Chapters) which have a bearing on the matter. Lastly, the Commission has unilaterally laid down rules of conduct by which Parliament can take it to task.

26 27

28 29 30

Joined cases T-134/94, T-136/94, T-137/94, T-138/94, T-141/94, T-145/94, T-147/94, T-148/94, T-151/94, T-156/94 and T-157/94, NMH Stahlwerke GmbH a.o. vs Commission (Order), [1997] ECR II-2293, § 18. BASF, § 63, repeated in Association des Amidonneries de Céréales de la CEE a.o. vs Commission, § 83, case T-435/93, Association of Sorbitol Producers within the EC (ASPEC) a.o. vs Commission, [1995] ECR II-1281, § 102, case C-191/95, Commission vs Germany, § 39, and case C-1/00, Commission vs France (British beef), [2001] ECR I-9989, § 79. Sinnaeve, ‘De positie van het Europees Parlement na het Verdrag van Maastricht’ (1994) at 149 seems to believe that the political responsibility of the Commission towards Parliament was only introduced in Maastricht—a view at least not shared by the Court of Justice. For an—incomplete—attempt see Communication from the Commission to the Intergovernmental Conference on the reform of the institutions, Commission doc. COM (2000) 771 final of 22 November 2000 at 3 et seq. Article 215 TEC (III-348.5 Const). Article 216 TEC (III-349 Const).

CHAPTER 5: THE PARLIAMENTARY SYSTEM

124

The Commission’s responsibility towards Parliament: ratione personae There are three aspects to the matter, namely the responsibility of the Commission for its members and staff; the responsibility of Commissioners for the Commission and its staff; and the responsibility of each Commissioner for his own actions and portfolio. Responsibility of the Commission for its members and staff. Politically, the investiture granted to the whole College implies that the College is responsible for its actions and for those of its members. The Barroso Commission laid down in a code of conduct that ‘the Commission acts collectively on the basis of the political guidelines set by the President. The chain of responsibility runs from the level of the College to the level of the Commissioner, who is answerable to the College for his/her activities and those of his/her departments. The Commissioner lays down, in the light of the guidelines and work programme of the Commission and the principle of collegiality, the policy guidelines and priorities relating to his/her portfolio. He/she shall ensure that it is implemented by his departments, supervising the management of his/her Director-General or Head of Service . . . and giving him/her, where necessary, the guidelines or general instructions to put these policies and priorities into practice. The chain of responsibility continues down to department level in the person of the Director-General, who is answerable to the Commissioner and the College for the proper implementation of the guidelines set by the College and the Commissioner . . .’31 In this system it is irrelevant whether a Commissioner (or the Commission) actually is unaware of the goings-on in the Directorate-General.32 The difference with the collective responsibility in law, discussed above, is evident. The Commission is politically responsible for the running of the administration. The same document gives the cabinets a special role in ensuring that ‘the principle of collective responsibility operates correctly by keeping the Commissioner informed about matters outside his or her own area of competence’. Responsibility of Commissioners for the Commission and its staff. Each Commissioner is responsible to Parliament for any Commission act,33 even if he did not join the decision making process, or voted against. In consequence, it is perfectly

31

32

33

Code of conduct for Commissioners, Commission doc. SEC(2004) 1487/2 (undated), adopted on 24 November 2004. Its predecessor, the Code of Conduct ‘les Commissaires et leurs services’, communication from President Prodi, Commission doc. SEC (1999) 1481 of 16 September 1999, comprised a somewhat more succinct definition. Vandersanden and Levi, ‘La réforme administrative de la Commission—quelques considérations générales’ (2005) at 292 point to the necessity of the Commissioner being properly informed, but fail to draw the logical conclusion that the Commissioner is also responsible for ensuring that he is properly informed. Merely stating that one did not know may dissolve the responsibility for the matter, but not for being ignorant about it. Adapting the institutions to make a success of enlargement—Commission Opinion in accordance with Article 48 of the Treaty on European Union on the calling of a Conference of Representatives of the Governments of the Member States to amend the Treaties, Commission doc. COM (2000) 34 of 26 January 2000 at 11.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

125

possible that the Commission president is asked to explain in Parliament comments made by individual Commissioners. Within the Commission, a Commissioner has to support the outcome of the decision-making process. If he disagrees with the Commission’s course, he has the choice between shutting up or resigning.34 Commissioners should not have dissenting opinions. Lastly, whilst in theory a Commissioner can threaten to resign if the College is unwilling to follow him,35 I am sceptical whether the threat has been posed or executed often. The responsibility of each Commissioner for his own actions and portfolio. The 2005 Framework Agreement concluded between Parliament and the Commission provides that ‘[e]ach Member of the Commission shall take political responsibility for action in the field of which he or she is in charge, without prejudice to the principle of Commission collegiality’.36 This individual form of political responsibility does not rest on law. 37 Although it is a product of interinstitutional convention rather than of primary law, it is no less real. It is additional to the Commission’s collegial responsibility38 and to the individual responsibility of Commissioners towards the Commission president. The latter, in return, is responsible for identifying conflicts of interest and for the action subsequently taken. 39 It follows that each Commissioner is politically responsible for its own actions40 and for those of the staff hierarchically subordinate to him. This is the logical corollary of collective responsibility, fitting with the role of the president-designate in the appointment of the College. It is also a necessary addition to avoid collective responsibility becoming what one MEP called the ‘legal shield of collegiality’.41 The possibility of sanctioning the individual responsibility of Commissioners is discussed in sub-section 5.4.2, below.

34

35 36 37 38 39 40 41

The 1999 Code de conduite des Commissaires, which was adopted as part of the reform programme by the Prodi Administration, draws the logical conclusion: ‘Out of respect for the principle of collective responsibility, Commissioners shall not make any comment which would call into question a decision taken by the Commission. They shall also refrain from disclosing what is said at meetings of the Commission’: Commission doc. SEC (1999) 1479 of 16 September 1999 at 3. The same text features in § 1.2.2 of the 2004 Code of conduct for Commissioners. Van der Burg, ‘De portefeuillekwestie’ (1970) at 6-7. Paragraph 2. This text is essentially similar to § 9 of the 2000 Framework Agreement. This is also the Commission’s view: Practical guide to the Framework Agreement, Commission doc. SP (2000) 3785/5 of 20 December 2000 at 9. This is my main criticism of Coutron, ‘Le principe de la collegialité au sein de la Commission européenne après le Traité de Nice’ (2003) at 249, who seems to ignore interinstitutional convention altogether. 2005 Framework Agreement, § 2. Including, where these are relevant, those of his spouse: § 1.1.3 of the 2004 Code of conduct for Commissioners. Mr Cox, during the debate of 11 January 1999. Following the Commission’s resignation in that year, Parliament asked ‘the Council that a procedure be laid down as soon as possible which enables individual Commissioners to be called to account’: Resolution on the resignation of the Commission and the appointment of a new Commission, OJ (1999) C177/19, §§ 6-7. It is hard to see how the Council could possibly satisfy a request which lies far outside its field of competence.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

126

The Commission’s responsibility towards Parliament: ratione materiae The Commission is responsible to Parliament for its actions in respect of Parliament, other institutions,42 or third parties. It accepted as long ago as 1973 that its responsibility was not confined to past events but also included current activities.43 Clearly, the Commission cannot be held responsible for matters to the extent that these fall outside its control. Logically, there can be no political responsibility without competence.44 For example, whilst OLAF is part of the Commission, it is in principle independent from the institutions. Normally, this precludes the Commission being held politically responsible for acts of OLAF or of its staff. Of course, where OLAF drafts anti-fraud legislation under Article 2(4) of the Commission Decision,45 the Commission is not responsible for the draft, but does have a responsibility as regards its attitude towards it. Similarly, the Commission is responsible for its duties towards OLAF (e.g., the creation of administrative conditions for OLAF to function). This last point became an issue when Parliament voiced criticisms as to the Commission’s personnel policy, which it feared compromised the anti-fraud office. Commissioner Schreyer, instead of declining to assume responsibility for the affair, explained the Commission’s position in substance.46 Parliament finished the debate by rapping the Commission over the knuckles, considering ‘it unacceptable . . . that applicants for managerial posts in OLAF are to be scrutinised by a Commission-nominated Advisory Committee on Appointments’.47 Obviously, the Commission is not responsible for the acts of other institutions or independent offices to the extent it cannot influence them. The Commission cannot be responsible for its own nomination48 since this precedes the appointment of the College. This causes no democratic void, since Parliament has the right of investiture. A more problematic matter is the responsibility for the policy domain of the Minister for Foreign Affairs created by the Constitution. He ‘shall contribute’ to the development of the common foreign, security and defence policies. However, he is bound to carry that policy out ‘as mandated by the Council’ (Article I-28.2 Const). Clearly, the MFA’s (and by extension, the Commission’s) responsibility would be limited to the extent that it has freedom of action. Article I-28.4 adds that, in handling external relations and co-ordinating ‘other aspects of the Union’s external action’ within the Commission, he is ‘bound by Commission procedures’. This raises the question whether the Commission’s collective responsibility includes the MFA. The argument has been made that the ‘Commission procedures’ formula should be 42 43 44

45 46 47 48

As far as the Commission’s actions before the Council are concerned, see the Résolution sur la position du Parlement européen à l’égard de l’évolution institutionnelle récente des Communautés européennes, JO (1966) 3465, § 10. Bieber, Organe der erweiterten Europäischen Gemeinschaften: Das Parlament (1974) at 31. In this vein also Blanchard, La constitutionnalisation de l’Union européenne (2001) at 375. Consequently, Parliament had more control over the High Authority than it had over the EEC or Euratom Commissions: Maas, ‘Parlementaire democratie in de Europese Gemeenschappen’ (1965) at 888. For the legislation creating OLAF see OJ (1999) L136. See the intervention of Mr Bösch during Parliament’s debates on 13 March 2001. Resolution on the Commission’s 1999 Annual Report ‘Protecting the Communities’ financial interests—The fight against fraud’, OJ (2001) C343/187, § 5. Unlike the practice in for example the Netherlands, where this is perfectly possible: Oud and Bosmans, Honderd jaren (1982) at 288.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

127

interpreted as a restriction on the Commission’s political responsibility.49 The general principle would apply that the Commission would be responsible for the MFA’s actions to the extent it can influence them. Vice versa, the MFA would be a full member of the College and therefore politically responsible for that institution. However, it remains to be seen whether the MEPs will easily acquiesce in a simple explanation by the MFA that he was ordered by the Council to act in a certain fashion. 5.2.3

The president’s role

There has been a trend for the Commission to become more ‘presidential’.50 The term is perhaps not perfect, since it connotes the French or American presidencies. Instead, the Commission president has started to resemble more the figure of the German Federal Chancellor or British Prime Minister. The growth of the president from a primus inter pares to leader of the executive was largely the work of the Commission itself51 even though, twenty years before, president Jenkins still had opposed any such move.52 The Treaty of Amsterdam53 introduced the formula in Article 219 TEC that the Commission works under the ‘political guidance of its president’.54 It is a development probably caused by the growth of the College due to recent accessions.55 Mr Prodi’s initiative to require a commitment from his Commissioners that they would resign at his request was a laudable next step. It was enacted in the Treaty of Nice, with one important limitation: Article 217(4) TEC now requires the president to ask the approval for such a step from the College. The same Article brought other changes. Henceforth, the Commission president can decide on the internal organisation of the institution, with the right to ‘reshuffle’ the responsibilities of the Commissioners. The ‘Members of the Commission’ are obliged to ‘carry out the duties devolved upon them by the president under his authority’. These are elements reminiscent of the German Kanzlerprinzip or the privileges of the British Prime Minister. As a further step, the Commission’s Rules of Procedure of 1999 further concentrated power in the hands of the president,56 for example through his new right to propose the discussion of questions not on the agenda. 49 50

51 52 53 54 55 56

Lenaerts, Binon, and Van Nuffel, ‘L’Union européenne en quête d’une Constitution: bilan des travaux de la Convention sur l’avenir de l’Europe’ (2003) at 294. A very detailed description of the changes within the Commission is given by Mistò, ‘La collégialité de la Commission européenne’ (2003), 189 et seq. Wouters, ‘Institutional and constitutional challenges for the European Union—some reflections in the light of the Treaty of Nice’ (2001) at 348 expresses a concern that the presidentialisation of the Commission would require additional checks and balances. His analysis is limited, however, to the legal situation. The interinstitutional convention that has developed is discussed in section 5.4.2, below. Commission doc. COM (2000) 34 of 26 January 2000 at 39 and Communication from the Commission to the Intergovernmental Conference on the reform of the institutions, Commission doc. COM (2000) 771 final of 22 November 2000 at 6. Jacqué, ‘Le Parlement européen (juillet 1979-décembre 1980)’ (1981) at 525. See Parliament’s Resolution on the functioning of the Treaty on European Union with a view to the 1996 Intergovernmental Conference—Implementation and development of the Union, OJ (1995) C151/56, § 21. Article 219, first paragraph as added by the Amsterdam Treaty and Declaration 32 attached to it. The Nice Treaty removed the first paragraph but rewrote Article 217 TEC. The French text of the pre-Nice Article 219 was a mite stricter: ‘dans le respect des orientations définies par son président’. Best, ‘The Treaty of Nice: not beautiful but it’ll do’ (2001) at 3. See now Article 6 of the Rules of Procedure, OJ (2005) L347/83.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

128

The (until now) final step was taken when in 2004 Parliament declared to hold the Commission president ‘fully accountable for identifying a conflict of interest which renders a Commissioner unable to perform his or her duties’ and for any subsequent action taken on it.57 Some have argued that the Kanzlerprinzip contradicts the principle of ‘collegialité’.58 That is true to the extent that this latter principle requires the Commission to decide by majority and each of its members is equal (internal operation of the legal principle of collective responsibility). It does not, however, contradict the political responsibility of the Commission towards Parliament, even though this has become more focused in the president. In practice, the president’s capacity to provide political guidance depends on the person. Mr Prodi found himself publicly criticised by his fellow Commissioners, something hardly comparable with the spirit of Article 217(1) TEC.59 5.3

Sanctioning responsibility: the motion of censure

In most Member States a rule exists which forces the government, or a single minister, to resign if it, or he, is censured by parliament. Some differences exist in the modalities or term used. For example, in the Netherlands the matter is not called censure but an issue of confidence. The classic definition of this rule is that a cabinet or minister, to whom more or less expressly has been denied confidence, has either to resign, or to obtain the dissolution of the Second Chamber of Parliament followed by elections.60 It is a convention that was purposely not enacted when the Constitution was rewritten in 1983.61 Since by convention the government resigns on the day of the elections, a lack of confidence in the government will in any event bring it down. The Dutch Parliament scarcely resorts to the vote of no confidence; but that is rather a mark of its effectiveness. If push comes to shove, ministers will normally go far not to loose its confidence. In this manner, the confidence rule is the linchpin in the system of parliamentary responsibility. In the European Union, the system operates differently. The Treaties provide for two modes in which Parliament can express its censure. The first is the positive vote of confidence (discussed in section 5.5, below), whereas the second is the motion of censure of Article 201 TEC.62 Two comments are in order. Firstly, the Court of Justice has stressed that ‘legal concepts do not necessarily have the same meaning in Community law and in the law of the various Member States’.63 Secondly—and this is different in many Member States—, these provisions do not require the Commission to enjoy Parliament’s confidence. Rather than exposing 57 58

59

60 61 62 63

Resolution on the election of the new Commission, OJ (2005) CE201/113, § 5(d). E.g. Nemitz, ‘Europäische Kommission: Vom Kollegialprinzip zum Präsidialregime?’ (1999) at 678. Mistò, op. cit. footnote 50 at 231, argues that the President’s power to give guidelines is incompatible with the principle of collegiality. I would rather argue that it has changed that principle. See e.g. for Mr Bolkestein’s criticisms of Mr Prodi’s comments on the Stability Pact: ‘Kritiek Bolkestein op EC-voorzitter Prodi’, Radio Nederland Wereldomroep Nederlands Nieuws of 17 November 2002. This arguably violated the 1999 code de conduite mentioned in footnote 34 of this Chapter. Van der Pot, Donner a.o., Handboek van het Nederlandse staatsrecht (1995) at 471. Kortmann, De Grondwetsherziening 1983 (1983) at 162. Articles I-26.8 and III-340 Const. Case 283/81, Srl CILFIT and Lanificio di Gavardo SpA vs Ministry of Health, [1982] ECR 3415, § 19.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

129

the political relationship underlying the motion of censure, primary law provisions focus on the cumulative conditions for such a motion to be carried. I will review those in the order as they occur in Articles 201 TEC. To begin with, the motion of censure must concern the ‘activities of the Commission’ and this requirement poses a somewhat perplexing problem in both the Treaties and the Constitution. The different language versions of Article 201 TEC suffer from gross differences in translation.64 The Court of Justice noted in CILFIT that ‘it must be borne in mind that Community legislation is drafted in several languages and that the different language versions are all equally authentic’.65 The Court added that ‘[a]n interpretation of a provision of Community law thus involves a comparison of the different language versions’.66 This, now, is not easy in the case of Article 201. The English text of the provision speaks of the ‘activities of the Commission’, and the wording of the Danish (‘virksomhed’), German (‘Tätigkeit’), Greek (‘δραστηριότητας’), Finnish (‘toiminta’), Irish (‘ghníomhaíocht’), Italian (‘operato’), Portuguese (‘actividades’) and Swedish (‘verksamhet’) texts seems to render this literally.67 Contrast this with the French and Spanish ‘gestion’ of the Commission. There may well be ‘activities’ of the Commission (or of its officials) which do not qualify as ‘management’. Mme Cresson’s hiring of her dentist as a consultant was perhaps an act of management, but his trips to the town where she was mayor68 would seem to have constituted ‘activities’ rather than ‘management’. The problem is worse. The Dutch text of Article 201 limits the scope of the motion of censure to ‘het beleid van de Commissie’, which is best rendered as ‘the Commission’s policies’. To what extent is the Commission under this standard responsible for the secret Eurostat accounts that surfaced in 2003? The question thus arises how these language versions should be reconciled, and different views are possible. For the more restrictive view one could argue that in the ECSC Treaty a linkage existed between the motion of censure and the Annual Report. It could be argued that this suggests that the scope of the motion of censure was not intended to be wider than that of the Annual ECSC Report. However, this position does not appear convincing. To begin with, it negates the possibility of motions of censure for procedural reasons. Moreover, it would seem reasonable to assume that, even though the parliamentary system and the motion of censure are separate notions, it remains up to the MEPs to determine what the reasons for a motion of censure are. Lastly, as early as in 1958, the EEC Commission accepted that the EEC Treaty constituted a break with the ECSC’s linkage between the Annual Report and the motion of censure.69 64 65 66

67 68 69

Bieber, op. cit. footnote 43 at 38 relies on the German text but it is not at all certain that this is the definitive one. CILFIT, § 18. Emphasis added. This mode of interpretation is of long standing. See e.g. case 29/69, Stauder vs City of UlmSozialamt, [1969] ECR 419, § 3, case 9/79, Wörsdorfer, née Koschniske, vs Raad van Arbeid, [1979] ECR 2717, § 6, case 55/87, Alexander Moksel Import und Export GmbH & Co. Handels-KG vs Bundesanstalt für landwirtschaftliche Marktordnung, [1988] ECR 3845, § 15, and case C-449/93, Rockfon A/S vs Specialarbejderforbundet i Danmark, [1995] ECR I-4291, § 28. Of the new official languages, I can only vouchsafe that the Hungarian ‘tevékenység’ corresponds to ‘activity’ or ‘activities’. Committee of Independent Experts, First report on allegations regarding fraud, mismanagement and nepotism in the European Commission of 15 March 1999, § 8.1.26. Premier rapport général (1958) at 26.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

130

This position is to some extent supported by Parliamentary practice. Motions of censure are rarely tabled in Parliament and none was ever adopted. Nevertheless, the events surrounding the downfall of the Santer Commission (discussed in sub-section 5.4.2, below) indicate that the net must be cast wide and includes, in addition to the Commission’s ‘policy’, also its ‘management’ and ‘activities’. Theoretically, Parliament could adopt a motion of censure for any reason, no matter what language version of Article 201 is used. How can this be reconciled with the principle of the Commission’s autonomy discussed in sub-section 2.4.4, above? The answer lies in the realisation that the checks and balances here are, ultimately, political rather than legal. The point is not that, theoretically, each time the Commission disobeys Parliament’s wishes, the MEPs could sack it. Rather, the truth is that, in practice, they will not do that lightly and certainly not as a regular course of action. The remaining conditions for the motion of censure need less discussion. It must be tabled by one-tenth of the MEPs. 70 Until 1973 individual MEPs had the right as well.71 The motion must be called ‘motion of censure’. It follows that it must be presented in writing. It must supported by reasons.72 Articles 201 TEC and III-340 Const require at least three days between the tabling of the motion and the vote thereon. Parliament has divided this period in the 24 hours between the announcing of the motion and the beginning of the debate, and the 48 hours between the beginning of the debate and the vote.73 Note that these deadlines start running from different bases. Since inevitably some time will lapse between the tabling of a motion and the announcing of it, effectively the three-days’ rule will always be met under Parliament’s Rules. Articles 67.2 of the German Constitution, 49 of the French Constitution and 46 of the Belgian Grondwet also lay down a compulsory resting pause. However, such a period is not ubiquitous in parliamentary systems. It is, for example, alien to Dutch law and practice. Lastly, the vote must be open74 and the adoption of a motion of censure requires a two-thirds majority of the votes cast, representing a majority of the MEPs.75 By comparison, French constitutional law only recognises the absolute majority requirement. The absolute majority requirement currently implies a quorum of (732 ÷ 2 + 1 =) 367 MEPs, which derogates from the normal quorum rule.76 Since censure must be positively established, logically only positive votes count for the two-thirds and absolute majority requirements. That implies that abstentions must be counted as negative votes as far as the absolute majority requirement is concerned. 5.4

The Community’s parliamentary system

5.4.1

Rare use of the motion of censure

The Commission’s responsibility towards Parliament can be seen in action almost every day the MEPs meet. Its ultimate sanction, however, is a rare event. The first 70 71 72 73 74 75 76

Rule 100(1) of the Parliament’s Rules of Procedure. Jacqué, ‘Le Parlement européen (1er janvier 1973-31 mai 1974)’ (1974) at 734. Rule 100(2). Rules 100(4) and (5). Articles 201 TEC and III-340 Const. Ibid. and Rule 100(7). One-third of Parliament: Rule 149 of the Rules of Procedure.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

131

motion of censure was proposed only in 1972,77 whilst the first vote on such a motion took place four years later.78 Until the crisis of January 1999, the motion of censure had been tabled only eight times. At two occasions it was withdrawn whilst in the other six it was rejected. 79 There are several reasons for this lacklustre use of the instrument, which I will discuss under the following headings: the link with the Commission’s (re-)appointment; the influence of the (European) Council; the lack of a clear government/opposition divide in Parliament; procedures within Parliament; and the requirements for a motion of censure. The link with the Commission’s (re-)appointment. Until not so very long ago, the Parliament exerted little influence on the composition of the new Commission. Although, after the first direct elections, some predicted that it would be politically more difficult for the Member States to ignore a motion of censure,80 it was unclear what would stop the Member States from re-appointing a Commission evicted from office.81 In theory, Parliament could have forced the Member States to take account of its feelings by consequently voting down any unpalatable Commission but, under the political constellation of the day, such a possibility was hypothetical.82 In my view, the question is often wrongly posed. Of course, the letter of the preMaastricht Treaties could not stop the Member States from re-appointing a censured Commission. Nor could it force them actually to replace a censured Commission. However, political practice might not have made this prudent and it would depend on the reasons for the censure whether the Member States could have ignored the will of Parliament. Since no motion of censure has ever been adopted, it is hard to tell with certainty what would have happened in such a case. The closest precedent was the resignation of the Santer Commission. It appears that at least one Commissioner implicated in the

77 78 79 80 81

82

Motion of censure tabled by Mr Spénale, Parliament doc. 204/72 of 16 November 1972, reproduced in Rideau a.o., Droit institutionnel des Communautés Européennes (Recueil des textes) (1974) at 216. Evans, A textbook on EU law (1998) at 66. Jorda, Le pouvoir exécutif de l’Union européenne (2001) at 370. Other motions of censure were rejected on 4 May 2004 and on 8 June 2005. Cohen, ‘Het Europees Parlement en de democratie’ (1982) at 165. The theoretical question has divided academia: Bieber, op. cit. footnote 43 at 39, Zuleeg, ‘Die Anwendbarkeit des parlamentarischen Systems auf die Europäischen Gemeinschaften’ (1972) at 3-4 and, writing just after the Maastricht Treaty, Sinnaeve, op. cit. footnote 27 at 149 believed that such re-appointment would be permissible. Legally that was correct, but whether it would be politically is quite a different matter. Schmutzer, ‘Some legal aspects of the work of the European Parliament’ (1967-68) at 87 seemed to argue on the opposite that the Member States could not reappoint the same Commission. Interestingly, Evans, op. cit. footnote 78 at 67 argues that such a re-appointment would violate (the current) Article 213 (ex 157) TEC, which requires the Commissioners to be chosen on the grounds of their ‘general competence’. He also points to the obligation in Article 10 (ex 5) TEC for Member States to loyally co-operate. I am not convinced that these obligations in themselves can be stretched so far as to preclude the reappointment of a sacked Commissioner. It appears that in 1973, after M. Spénale tabled the first motion of censure, but before it was scheduled to be voted upon, some Member States proposed ‘their’ sitting Commissioners for the next Commission: Bieber, op. cit. at 41. The matter did not come to the crunch since the French MEP withdrew his motion. Kapteyn and VerLoren van Themaat, Inleiding tot het recht van de Europese Gemeenschappen— deel 1A (1987) at 100. The 1998 English edition of the same work (Introduction to the law of the European Communities, ed. Gormley) at 223, written following the Maastricht Treaty, argued that it would be difficult to imagine serious politicians agreeing even to stand for office in such circumstances.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

132

fraud scandals surrounding that affair was proposed to Mr Prodi, who apparently managed to refuse the incumbent.83 Parliament’s successful pressure to block certain candidate Commissioners at the time of the appointment of the Barroso Commission suggests that nowadays Parliament is not only legally, but also politically capable of blocking the (re-) appointment of candidates it finds undesirable. The influence of the (European) Council. In many cases, in the eyes of the MEP the real culprit will not be the Commission, but rather the Council, the European Council or the Member States.84 As one Commission president warned the MEPs, ‘make no mistake in identifying your partner, and above all make no mistake in identifying your enemy.’85 As far as legislation is concerned, the Single European Act and the Maastricht, Amsterdam and Nice Treaties made Parliament co-responsible for much of the EU’s legislation. In the executive domain, the Council is firmly in control of most second and third pillar policies as well as over some first-pillar policy coordination domains. Many problems in the execution of legislation occur at Member State level and are both too obscure to be noticed by the national parliament and too far away from Strasbourg. Once they are noticed by other bodies, such as the Court of Auditors, the Commission can devolve its responsibility on the Member State and Member State politicians on ‘Brussels’. For Parliament (and for the citizen) it then becomes difficult to pin-point those responsible. The lack of a clear government/opposition divide in Parliament. As noted above, Parliament has no very clear government and opposition segments. No MEP’s future specifically depends on the successes of the Commission,86 whose composition covers the broad political spectrum. In the current state of public awareness, Parliament gains or suffers as such from the Commission’s achievements. This does not remain without consequence for the checks and balances between the institutions. Moreover, it strongly affects the political legitimacy (input legitimation) of the European political system—as any student of the turnout of European elections is painfully aware.87 The lack of a clear government/opposition divide in Parliament tends to depoliticise the relationship between that institution and the executive, resulting in a more legal relationship. In the words of one MEP, ‘We are not a House of confrontation, as a national parliament would be, because we do not have a government here to support or oppose. The Commission has not been elected by us, but it is our duty to help it work better.’ 88 Similarly, as noted in section 4.3, in the past the appointment procedure caused the Commission to a large extent to be politically colourless89 even though there is no 83 84 85 86 87 88

89

‘Greep Prodi op Commissie niet gehonoreerd’, NRC-Handelsblad of 19 June 1999. In a similar vein Constantinesco a.o., Traité instituant la CEE—Commentaire article par article (1992) at 845 and Teitgen, Cours de droit institutionnel communautaire (1972) at 330-331. Commission President Thorn during the sitting of Parliament of 14 October 1981, OJ Annex (1981) 1-275/131. Similarly Nemitz, op. cit. footnote 58 at 686. See footnote 117 of Chapter 4, above. Mr Garriga Polledo during the sitting of Tuesday, 15 December 1998. The press observed caustically that ‘it is something of a tribute to Mr Santer’s clumsiness that they should have rubbed along so badly in recent years’: ‘Little respect, less love, but growing power’, Economist of 10 June 1999. For some interesting data on the Parliamentary support for motions of censure see Magnette,

CHAPTER 5: THE PARLIAMENTARY SYSTEM

133

Treaty obligation for the Commission to be so.90 I argued in sub-section 4.6.1, above that this is likely to change somewhat following the appointment of the Barroso Commission. The entry into force of the Constitution would have reinforced such a development. A politically more coloured Commission may be somewhat more prone to attract motions of censure, although the required two-thirds majority will probably still not make this a regular occurrence. The procedures within Parliament are such that MEPs need each other and cannot afford a too confrontational stance vis-à-vis their colleagues. Moreover, in addition to their political (party) interests, MEPs have strong systemic interests that tend to work against a split into opposition and government blocs.91 For example, Article 251(2) TEC requires an absolute majority of votes for the rejection or amendment of common positions. In practice, this often assures that the largest two parties are condemned to each other’s support. This factor disfavours overly sharp political differences between the major groups. Of course, the spirit of consensus is not unlimited. During the debate on the motion of censure against the Santer Commission, the Socialist group at some point opposed a motion of censure directed at only Socialist Commissioners Cresson and Marín.92 Similarly, during the 2003 Eurostat scandal, the Socialists attempted to draw other Commissioners into the scandal in order to avoid seeing only Socialists implicated. The requirements for a motion of censure. The requirements for the spontaneous French motion de censure are so severe that rarely a government of the Fifth Republic stumbled over one. It is similarly difficult to trip up the Commission via Articles 201 TEC or III-340 Const. A motion of censure has been filed at several occasions, only once leading to the downfall of the Commission and even that happened before the vote. In 1997 a bitter MEP complained that ‘the European Parliament has denounced [the Commission], the European Parliament has condemned it, and at the end of the day the European Parliament has backed down’.93 Hence the academic view that the motion is not a weapon for which the Commission need to be too afraid.94 In 1999 a motion of censure was almost passed. Five years later, the appointment of the Barroso Commission caused more convention to develop in this field. It is to these events that I must now turn. 5.4.2

From 1999

The events that led to the collapse of the Santer Commission provide a good illustration of the operation of the parliamentary system in the European Union, whilst they simultaneously influenced that system. Since they were widely publicised, I will only recall them in detail where they are of interest to constitutional practice.

90

91 92 93 94

‘Appointing and censuring the European Commission: the adaptation of parliamentary institutions to the Community context’ (2001) at 304. Different (for no good reason): Anonymous, ‘Who’s afraid of two European presidents?’ (2003) at 2. It is true that Mr Prodi sought his team to be ‘colourless’: not to function along party lines: Address delivered to Parliament by Romano Prodi, President-designate of the Commission, on 21 July, Bulletin EU 7/8-1999 at item 2.2.1. For an analysis of some of these Hix, Kreppel and Noury, ‘The party system in the European Parliament: collusive or competitive?’ (2003) at 318 et seq. ‘Parliament versus commission’, Economist of 14 January 1999. M. Fabre-Aubrespy during the debate of 20 February 1997 on the motion of censure. Kapteyn and VerLoren van Themaat (ed. Gormley), Introduction to the law of the European Communities (1998) at 224.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

134

In 1998 Parliament postponed the discharge for 1996 to enable the Commission to respond to certain complaints concerning mismanagement of funds. Parliament’s suspicions were dwarfed by the Van Buitenen report. On 9 December 1998 that Commission official sent a file with strong allegations of mismanagement to a Belgian Green MEP. Subsequently the Commission suspended him. This itself sparked the MEPs’ ire.95 More momentous to the rest of Europe, however, were the consequences for the Commission. Mr Van Buitenen’s allegations had been detailed and substantiated and made the position of especially Socialist Commissioners Cresson and Marín very difficult. The debate on the discharge for 1996 followed a few days later. Most political groups disfavoured granting discharge, much to the anger of the Socialists, who spoke of ‘an unholy alliance of Conservatives, Liberals and extremists’.96 Before the vote, three days later, the Commission attempted to sway the MEPs’ opinion by issuing a statement suggesting that, if the budget discharge was not given, Parliament should clarify the situation by voting on a motion of censure. This ‘blackmail’97 led MEPs roundly to reject the discharge98 and the Socialists, by now a lone voice, announced a motion of censure in the hope that its rejection would be construed as a vote of confidence in the Commission.99 Other groups levelled a resolution of censure specifically aimed at ousting Commissioners Marín and Cresson. During the plenary debate of 11 January 1999 on both motions of censure, Socialists group leader Ms Green refused to support those attempts to bring down only these two Commissioners. Legally this was correct, since a motion of censure against individual Commissioners is not foreseen in Article 201 TEC (although, of course, nothing stops Parliament from indicating its displeasure with individual Commissioners). The impasse between the PSE on the one hand and most other parties on the other caused no single motion of censure to be adopted at that occasion. However, the resolution adopted three days later100 was a compromise calling inter alia for a committee of independent experts ‘to be convened under the auspices of the Parliament and the Commission’ which would ‘examine the way in which the Commission detects and deals with fraud, mismanagement and nepotism’.101 The Commission could do little more than agree to this last lifeline. As one newspaper put it, ‘something important has happened’.102 Parliament had shown its teeth. The committee of independent experts’ first report accused several Commissioners, but most of all French Commissioner Cresson, whilst concluding in respect of the College that ‘[i]t is becoming difficult to find anyone who has even the slightest sense of responsibility’.103 This was the iceberg onto which the Santer 95 96 97 98 99 100 101 102 103

‘Europarlement woedend om schorsing ambtenaar’, NRC Handelsblad of 6 January 1999. Mr Blak during the sitting of 15 December 1998. In the verdict of Ms Aelvoet, Mr Cox and Mr Miranda during Parliament’s sitting of 17 December 1998. Minutes of 17 December 1998, OJ (1999) C98/194-195. See the interventions of Ms Green during the debate of 17 December 1998. ‘Parliament versus commission’, Economist of 14 January 1999. Resolution on improving the financial management of the Commission, OJ (1999) C104/106, § 1. ‘Shaken in Strasbourg’, Economist of 14 January 1999. Committee of Independent Experts, First report on allegations regarding fraud, mismanagement and nepotism in the European Commission of 15 March 1999, at § 9.4.25. Commissioner Van Miert suspected the hand of a Member State in drafting this phrase in order to dilute Mme Cresson’s abuses: Mijn jaren in Europa—Herinneringen van een Eurocommissaris (2000) at 309.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

135

Commission foundered. The institution’s president could do little more than protest that the conclusions of the Independent Experts were ‘distorted’ and ‘unjustified’104 and resign with his colleagues in order to avoid a now certain censure.105 On 4 May 1999 Parliament implicitly refused discharge.106 A first, obvious, conclusion from the affair is that the fear107 that governments would simply re-appoint their dismissed Commissioners108 is in the current political and constitutional constellation no longer realistic. Of course, the appointment procedure laid down in the Constitution in any event would no longer permit this. Several other conclusions can be drawn from the episode. I will discuss the censure of individual Commissioners; discharge and censure; and the Commission’s caretaker capacity. Censure of individual Commissioners. The first issue to analyse are Parliament’s attempts to censure individual Commissioners. Whereas Parliament before 1999 was a staunch supporter of the collective responsibility of Parliament,109 since 1999 calls are heard for the right to censure individual Commissioners.110 Of course, the wording of Article 201 TEC, combined with the principle of collective responsibility, does not foresee the possibility of a motion of censure targeting individual Commissioners. In national constitutional law, the corollary of collective responsibility normally is the right of the Prime Minister to dismiss ministers, or to have them dismissed by the head of state. Mr Santer was, however, virtually powerless towards Ms Cresson and recourse to Article 216 TEC was difficult in practice. Mr Prodi learnt the lesson and obliged his new colleagues to sign an undertaking promising to step down if he asked them to.111 Not long thereafter, Parliament negotiated an obligation for him to ‘consider seriously’ to fire a Commissioner if Parliament had expressed its ‘lack of confidence’ in that particular person, ‘subject to political support for such a view, in terms both of substance and of form’.112 The undertaking required by Mr Prodi, which was of a conventional nature, was enacted into the Treaty of Nice, but the dismissal was made subject to the approval of the College (Article 217(4) TEC in its post-Nice version). Article I-27.3 Const would 104 Statement by President Jacques Santer the day after the resignation of the members of the Commission, Commission press release IP/99/179 of 16 March 1999. 105 Anonymous, ‘The report of the Committee of Independent Experts: an ill wind . . .’ (1999) at 270-271. It appears that a number of MEPs attempted to table a second motion of censure during the April session that followed the resignation, in an attempt to obtain some sort of limitation on the Commission’s activities whilst it was in caretaker status. That proposal, however, was ruled inadmissible since its rejection could have been interpreted as a vote of confidence: Bradley, ‘The Institutional Law of the European Union in 1999’ (2000) at 585. 106 OJ (1999) C 279/114. 107 As expressed by Mr Cox during the debate of 22 March 1999 on the Council statement following the resignation of the Commission. 108 Discussed in sub-section 5.4.1 above. 109 At least until the Treaty of Maastricht: Corbett, op. cit. footnote 3 at 294. 110 For an earlier call in this direction see Parliament’s Report on the European Commission’s follow-up of the recommendations made by the Committee of Inquiry into BSE drawn up by the Temporary Committee to follow up the recommendations on BSE, Parliament doc. PE 223.656/fin.2 of 14 November 1997 at 22. See also its Resolution on improvements in the functioning of the Institutions without modification of the Treaties, OJ (1999) C219/427, §§ 7–10 and its Resolution on the Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions ‘Second Bi-annual BSE Follow-up Report’, OJ (1999) C219/433, § 43. 111 ‘The new team for Europe’, Economist of 15 July 1999. 112 Framework Agreement on relations between the European Parliament and the Commission, OJ (2000) C 121/122, paragraph 10.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

136

remove that latter hindrance. Either way, the system renders the responsibility for individual Commissioners enforceable because of the notion of ‘lack of confidence’. To the extent that they are relevant, the events surrounding the appointment of the Barroso Commission are described in sub-section 4.6.1, above. At the time of the vote of investiture, Parliament adopted a resolution in which it demanded that a vote of censure of an individual Commissioner would require the Commission president to ‘either require the resignation of that Member or justify his refusal to do so before Parliament’.113 Mr Barroso subsequently undertook such a commitment,114 which further reduces his room for manoeuvre towards the MEPs. These new engagements were subsequently formalised in the 2005 Framework Agreement.115 This convention does not oblige the president to fire the Commissioner, but defying the majority of the MEPs may put the responsibility of the whole Commission on the line. Parliament’s initial reluctance to approve the Barroso Commission shows that this is no idle threat. Discharge and censure. The affair shed some light on the relationship between censure and Parliament’s right to grant discharge. That latter prerogative originally belonged to the Council. 116 Although foreseen in the financial provisions of the EC Treaty and the Constitution,117 properly speaking it concerns a right of oversight. Indeed, Articles 276(2) and (3) TEC, added by the Maastricht Treaty,118 make this very clear: Parliament may ‘ask to hear the Commission give evidence with regard to the execution of expenditure or the operation of financial control systems’ before giving discharge, or even ‘for any other purpose in connection with the exercise of its powers over the implementation of the budget’. The Commission ‘shall submit any necessary information’ to the MEPs at their request. In addition, the Commission is obliged to ‘take all appropriate steps to act on the observations in the decisions giving discharge and on other observations’ by Parliament relating to the execution of expenditure, as well as on comments accompanying the recommendations on discharge adopted by the Council. Parliament resolved in 2001 to follow up discharge recommendations by letting its Budgetary Control Committee automatically putting the matter onto its agenda six months later.119 These obligations put the Commission’s entire financial management—and with that, its policies—squarely under the oversight of the MEPs (and, to a somewhat lesser extent, of the Council). One can readily see why, in practice, the discharge has never been a simple accounting affair. The basis for the discharge are the accounts of the preceding financial year, submitted annually by the Commission to the Council and Parliament,120 the Court of Auditors’ annual report together with the replies of the institutions under audit, the statement of assurance referred to in Article 248(1), second subparagraph, and any relevant special reports by the Court of Auditors.121 The Commission obliged itself to

113 Resolution on the election of the new Commission, OJ (2005) CE201/113, § 5(a). 114 Commission president Barroso during the debate of 18 November 2004. 115 Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125, § 3. 116 Jorda, op. cit. footnote 79 at 393. 117 Article 276 TEC and III-409 Const, respectively. See also Articles 139-147 of the Financial Regulation, OJ (2002) L248/1, corr. OJ (2003) L25/43. 118 Before the Maastricht Treaty, similar rules were already used in practised on a conventional basis: Strasser, ‘Les dispositions financières du Traité de Maastricht’ (1994-95) at 201. 119 Resolution on reform of budgetary control procedures and institutions, OJ (2001) C232/194, § 29. 120 Article 275 TEC. 121 Article 276 (1) TEC.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

137

‘forward all information necessary for supervising the implementation of the budget for the year in question, which the chairperson of the parliamentary committee responsible for the discharge procedure pursuant to Annex VI of the Parliament’s Rules of Procedure requests from it for that purpose. If new aspects come to light concerning previous years for which discharge has already been given, the Commission shall forward all the necessary information on the matter with a view to arriving at a solution which is acceptable to both sides.’122 The procedure laid down in Article 276 TEC is fairly simple. The Council submits a recommendation to Parliament (the only occasion where the Council adopts recommendations forming the basis for an act of Parliament). The Council votes by qualified majority. Before giving discharge, Parliament may ask the institution concerned for more information.123 The MEPs consider that the remarks they attach to the discharge decision should be ‘constraining’ on the institution involved.124 The information requested must be related to the accounts: in 2003 Parliament’s Budgetary Control Committee sent a questionnaire to the Council, which the latter thought not to be directly related to the accounts of the Council or to the Court of Auditors’ report. A reply was subsequently refused.125 Parliament can delay discharge, which can be useful if it wishes to obtain more information from the Court of Auditors or from the Commission (or from other institutions). In Rule 105 of its Rules of Procedure, Parliament provided a possibility for the president of the Court of Auditors to take the floor ‘in order to present the comments contained in the Annual Report, special reports or opinions of the Court, or in order to explain the Court’s work programme’. The second paragraph of the same Rule adds that ‘Parliament may decide to hold a separate debate on any questions raised in such statements with the participation of the Commission and Council, in particular where irregularities have been reported in financial management’. The first time Parliament delayed the discharge concerned the execution of the Budget for 1980.126 It now happens regularly (for instance for the discharges of 1996, 1998 and 2000), with Parliament tying conditions to the discharge.127 In its Rules of Procedure128 Parliament claimed the right to refuse discharge. It first exercised this in 1984, in connection with the discharge over the Budget for 1982. The question arises whether such a refusal is to be equated with censure.129 122 2005 Framework Agreement, §26, which succeeds § 17 of the 2000 Framework Agreement, OJ (2001) C121/122. 123 Article 276(2) TEC. 124 Resolution embodying (i) Parliament’s opinion on the convening of the Intergovernmental Conference, and (ii) an evaluation of the work of the Reflection Group and a definition of the political priorities of the European Parliament with a view to the Intergovernmental Conference, OJ (1996) C96/77, § 2.VII. 125 Council doc. 15806/03 of 9 December 2003. 126 Notenboom, Het Europees Parlement en de financiën (1988) at 296. 127 See the Resolution of 13 April 2000 on postponement of the decision concerning discharge to the Commission in respect of the financial management of the sixth, seventh and eighth European Development Funds for the 1998 financial year, OJ (2000) L191/14. See for other examples OJ (2001) C40/381 and 383 and for the postponement of discharge to other bodies OJ (2003) L148/16. For completeness’ sake, I should note that Parliament’s Rules of Procedure contain a spearate monitoring procedure over the implementation of the current year’s budget in Rule 72. This monitoring (the Rules use the gallicism ‘control’) is entrusted to the Budgetary Control Committee. 128 Article 3(1)(a) of Annex V of Parliament’s Rules of Procedure. Under its pre-1999 Rules of

CHAPTER 5: THE PARLIAMENTARY SYSTEM

138

There is no legal link between the two, in the sense the way the motion of censure was linked to the Annual Report by Article 24 ECSC. Under the terms of Article 274 TEC, the Commission executes the budget ‘on its own responsibility’. In 1979 Commissioner for Budgetary Affairs Tugendhat threatened that the Commission would interpret a refusal of discharge as a motion of censure and resign.130 However, when in 1984 discharge was refused, the Commission failed to resign, apparently because it felt that Parliament had abused the procedure.131 This suggests that, even then, the door was indeed open for a lack of discharge to be seen as censure. However, the end of the Commission’s term was in any event nigh and this made the question of the consequences moot.132 It appears that the 1999 crisis was the only other occasion in which the Commission was refused discharge,133 but that happened after the Santer Commission resigned. However, there is here more than meets the eye. Although the direct reason for Mr Santer’s resignation was the outcome of the Committee of Independent Experts procedure, with which the Commission had agreed, this procedure was itself the consequence of the discharge procedure. In 1998 Mr Santer had written to Parliament asking that, in the event of a negative vote on discharge, the MEPs should ‘clarify’ the situation by proceeding to a vote on a censure motion. In other words, on that occasion the Commission did not go as far as automatically to link its fate to the discharge vote. Similarly, a majority in Parliament tended to see a possible refusal of discharge as perhaps a good reason for a motion of censure, but not as an alternative to it.134 It is not evident that future Commissions are bound to resign if discharge is not given—even if the demise of the Santer Commission may have reinforced the political135 weight of the discharge procedure. This practice fits with the scheme of the Treaties and the Constitution. Articles 201 TEC and III-340 Const lay down very strong requirements for bringing down the Commission. It is at least not likely that the drafters would have wished Parliament to be able to bring the Commission down through the discharge procedure with a normal majority vote without période d’apaisement.

129 130 131 132 133 134

135

Procedure, Parliament required an absolute majority of its members to refuse discharge, a strange limitation not required by the Treaties: Article 5(2) of Annex VI of the Rules of Procedure of the Parliament, OJ (1993) L293/62; see Rüping, ‘Die parlamentarische Haushaltskontrolle in den Europäischen Gemeinschaften und deren Mitgliedstaaten’ (1982) at 213. In 1999 the limitation was removed. A sensible improvement, since the old Rule arguably violated Article 198 TEC (‘Save as otherwise provided in this Treaty, the European Parliament shall act by an absolute majority of the votes cast’). Müller, The Powers of the European Parliament in the European Union (1999) at 25 considers a refusal ‘akin to a vote of no confidence’. Jorda, op. cit. footnote 79 at 393 tends towards a similar view. Freestone and Davidson, The institutional framework of the European Communities (1988) at 84. Evans, op. cit. footnote 78 at 67. Corbett, The European Parliament’s role in closer EU integration (1998) at 125. See rapporteur Morgan during the debate of 9 April 2002 on the discharge for 2000 (sections IV, V, VI (+ discharges 1996 to 1999), VII, VIII of the General Budget/postponement of the discharge 2000 for section II). See the interventions of Rapporteur Elles during the debate on 15 December 1998 and of Mr Cox, Ms Aelvoet, Mr Miranda, Mr Pasty and Ms Lalumière during the debate on 17 December 1998. This is consistent with the line taken in the Resolution on reform of budgetary control procedures and institutions, OJ (2001) C232/194, § 19: ‘Urges that a decision to refuse discharge be followed by a debate on the political position of the Commission, following a statement by the latter pursuant to Rule [103] of the Rules of Procedure’. Mr Crowley came closest during the debate on 17 December 1998 by arguing that ‘[i]t is open to the individual Commissioners to resign if they feel they do not have the support of this House’.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

139

A further argument against a linkage between refusal to discharge and censure lies in the discharge of the other institutions and bodies. Parliament is similarly involved in this exercise136 and a refusal to give discharge to institutions other than the Commission obviously cannot be construed as censure of them. In summary, the better view is that a refusal of discharge cannot be equated to a motion of censure, although of course it can provoke one. That being said, there is nothing in the law that would stop the Commission and Parliament from agreeing that a lack of discharge should be taken as a lack of confidence. Even though this term features neither in Article 201 TEC or III-340 Const, in 1999 Mr Santer as well as several MEPs framed the political question in terms of confidence.137 The decision to grant or withhold discharge is a political one138 in which Parliament is sovereign. The MEPs are bound by neither the Court of Auditors nor the Council. Some consider that Parliament, by using the discharge procedure as a political exercise, ‘détourne l’esprit des procédures’,139 but I disagree. One can hardly put the discharge decision in the hands of 732 politicians (most of whom are not lawyers) and expect the result not to be a political assessment. The Commission’s caretaker capacity. Following censure or resignation, the Commission continues to deal with current business in a caretaker capacity until it is replaced.140 In the past it has been suggested that this would require a Commission in such a position to determine which matters are ‘controversial’ and which ones are not.141 I doubt whether such a copy of Dutch constitutional practice can be read into the phrase ‘current business’. Rather, the word ‘current’ intimates to the executive not to start up any big policy initiatives. I believe, however, that the wording is so broad that it would be hard to make such a precise distinction, which—as Dutch practice shows—in any event is a matter of political judgment rather than of law.142 5.5

The requested vote of confidence

In some Member States the government does not have to wait until Parliament passes a motion of censure. Apart from a vote of investiture, the government can request parliament to express its confidence. The modalities vary per Member State.143 For example, in the United Kingdom there are few formalities. The Prime Minister can indicate before an important vote in the Commons whether he considers its outcome as a matter of confidence.144 In France the matter is dealt with in detail by Article 49 of the Constitution. The Prime Minister can engage ‘la responsabilité du Gouvernement devant l’Assemblée Nationale’ on either his political programme or on 136 However, under the terms of the 1970s Gentlemen’s Agreement with the Council (discussed in section 14.3, below), the two institutions agreed not to question each other’s budget. 137 E.g. President Santer during the debate of 11 January 1999. 138 In so many words: Resolution embodying comments forming part of the decision granting a discharge to the Commission in respect of the implementation of the budget of the European Communities for the 1983 financial year, OJ (1985) L132/24. 139 Blanchard, op. cit. footnote 44 at 379-380. 140 Articles 201, 215 TEC and III-340, III-348.5 Const. 141 Schmutzer, op. cit. footnote 81 at 90. 142 The Court of First Instance’s judgment in case T-219/99, British Airways vs Commission, [2003] ECR II-5917, § 56 provides some support for this. 143 Constantinesco a.o., op. cit. footnote 84 at 845 read from the lack of a vote of confidence (other than the investiture) that no parliamentary system exists in the Union. Their view is coloured by French glasses: a positive vote of confidence is hardly a defining feature of parliamentary systems in general. 144 Hood Phillips and Jackson, Constitutional and Administrative Law (1987) at 150.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

140

the vote of a text. In the latter case, the text is adopted unless a motion of censure is voted within the following twenty-four hours. Since the conditions for such a motion are very strong, this provides the government with a crow bar, the threat of which may suffice to apply pressure. Note that the French Council of Ministers must have ‘délibéré’ before the Prime Minister can put the question of confidence. In Germany, Article 68 of the Constitution allows the Federal Chancellor to request the confidence of the Bundestag, a procedure used rarely. If confidence is refused, he can request that institution to be dissolved within 21 days, unless the parliamentarians elect a new Federal Chancellor. Similar to the motion of confidence, a 48 hours ‘cooling off’ period is required between the request for the confidence motion and the vote. In addition, the German Constitution foresees the possibility to link the motion of confidence to a particular legislative proposal, but this has never been used in practice.145 European primary law foresees a positive vote of confidence only at the time of the investiture of the Commission, discussed in sub-section 4.6.1, above. This does not mean, however, that the Commission could not attempt to force a vote of confidence by asking Parliament to vote on a motion of censure. As discussed in the previous sub-section, Mr Santer tried this tactic during the 1998-1999 crisis. However, MEPs questioned both the constitutionality and wisdom of this move146 and it backfired. Mr Santer’s error indeed was caused by the specific conditions pertaining to the motion of censure. In abstract terms, one can imagine the parliamentary system to revolve around the ‘unwritten’ concept of confidence (as in, for example, the United Kingdom and the Netherlands) or around a ‘formalised’ system of censure (as in, for example, Germany and France). In the former case, the majority needed for a noconfidence vote is automatically set at the level of the number of MPs that can make the work of the government impossible. In the—very hypothetical—case that a Dutch or British government would ignore a motion of no confidence, parliament would have the requisite majority to obstruct the government into submission. Similarly, the same mechanism that is used to express a lack of confidence can be used by the government to obtain a positive statement of confidence. In the latter case however, the motion of censure is tied to specific majorities and conditions. Not meeting these thresholds does not mean that parliament has expressed its confidence in the government. It merely signals that it failed to meet the conditions for sending the executive home. This does not necessarily mean that the majority in parliament likes the current government. By extension, it is not possible for such a motion of censure to function as a positive motion of confidence. Indeed, it is therefore logical that both the German and French constitutions lay down the modalities for a positive motion of confidence. The motion of censure of Article 201 TEC is much closer to the German/French vote of confidence than to the British/Dutch system of confidence. Parliament’s failure to adopt a motion of censure was not more than exactly that. Under the circumstances, a motion of censure had been surely defeated if the Socialist Group still supported the Commission. That is an entirely different majority and message than a vote of confidence. This is not to say that a vote of confidence is impossible in the EU’s parliamentary system—on the contrary. However, any such vote is more likely to be 145 Article 81 German Constitution; Feldkamp, ‘Chronik der Vertrauensfrage von Bundeskanzler Gerhard Schröder im November 2001’ (2002) at 5. 146 Messrs Pasty, Martens and Cox during the Sitting of 11 January 1999.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

141

the fruit of interinstitutional convention based on the right of discharge, rather than the outgrowth of the motion of censure. As a final issue, I should note the (in the current scheme of things, theoretical) possibility of a Commissioner or Commission threatening to resign in order to pressure the Council. It appears that in the 1960s such an event almost occurred when Commissioner Mansholt stated the Commission would no longer be able to bear the responsibility for the food surpluses situation if the Council did not change its position. When the French and German press interpreted this as a threat to resign, the Commission hastily clarified that this was not the intention.147 5.6

Parliamentary responsibility and the Council

5.6.1

No relationship of responsibility between Council and Parliament in law

The letter of primary law suggests that the Council is not politically responsible to Parliament. There is a subtle difference in wording in the provisions that comprise the tools for such responsibility:  Article 197 TEC (III-337 Const, oral and written questions),  Article 200 TEC (III-337.3 Const, discussion of the Annual General Report),  Article 214 TEC (I-27.2 Const, approval of the appointment of the Commission),  Article I-26.8 Const (responsibility of the Commission to Parliament, with no corresponding provision concerning the Council). For example, whereas the Commission ‘shall reply’ to questions put to it by the MEPs without further limitation, the Council is heard ‘in accordance with the conditions laid down’ in its Rules of Procedure. Occasionally the argument is made that the Council should be politically responsible to Parliament,148 a claim presumably fuelled by the executive competencies that the Council has accumulated in the second and third, but also in the first pillar. This claim is not quite correct and, in addition, it obfuscates the distinction between parliamentary responsibility and its possible sanction of censure. Differently from the Commission or national governments, ministers in the Council are not members à titre personnel but rather representatives of their respective governments.149 When this distinction is lost, confusing situations occur. When, during the plenary debate on 14 February 2001, an MEP expressed his concern that new EU rules on transparency would in effect imply a limitation of Swedish legislation on the issue, Secretary of State Danielsson assured the Honourable Member that ‘the Swedish government will not accept any change whatsoever to Sweden’s Freedom of the Press Act’—a pledge peculiar not because of its content, but because of the place where it was made.150 This matter, as the Secretary of State later in the debate realised, was perhaps best ‘continued in a more national context’. Nevertheless—and this will be discussed at some lengths in Chapters 7 and 8, below—the Council has accepted to a large extent to explain its policies and actions 147 Van der Burg, op. cit. footnote 35 at 7. 148 Bracke, ‘Het democratisch tekort na Maastricht’ (1993) at 154 and Evans, op. cit. footnote 78 at 67-69. 149 From a separation of functions point of view this organisation may well be criticised, but that question, however interesting, directly concerns the logic of the basic plan of the EU’s constitution. I will not further discuss it since it lies beyond the horizon of both this study and the EU’s legislative powers. 150 Parliamentary debates of 14 February 2001.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

142

in Parliament, and to respond to Parliamentary questions. In this sense, it has to a certain extent become politically responsible to the MEPs. Of course—and this needs to be emphasised—, any relationship of responsibility cannot be sanctioned since, in the nature of things, Parliament cannot censure the Council.151 To start with, the Council’s character differs fundamentally from that of the Commission. 152 The Council is formed by ministers who are appointed and dismissed in function of their respective national constitutions. In addition, the Council is composed of national representatives who do not necessarily agree. In this sense, as far as its composition is concerned, the Council resembles a deliberative assembly when acting in (most of) the first, and an imperfect executive in the second and third pillars.153 There is a parallel here with the Court of Justice’s judgment in Eurocoton vs Council,154 in which the Court rebuked the Council for not having reasoned its lack of a decision in an implicitly terminated anti-dumping proceeding. The measure at issue in Eurocoton concerned a proposal for definitive anti-dumping duties. Such measures are by their very nature, hybrid from a separation of functions point of view: the Commission proposal is drafted following an (ostensibly) administrative procedure. The Council is then called upon to adopt the definitive measures by Regulation, which applies to any imports of the product concerned. In the field of Article 230 TEC this led to a rich case law155 on which companies may contest the resulting act. The Court of Justice leaned heavily on this case law to (over)emphasise the administrative over the political nature of the procedure in question. I respectfully submit that the judgment overlooks that the problem was precisely that the Council in the case at hand could not agree on what was an extremely political156 matter. Hence it was patently unable to reason its lack of a decision in any meaningful manner. How could the Council have reasoned its decision if there was no, and could be no,157 Commission proposal to terminate the proceeding? Ultimately, the Court confused the situation of a decision terminating the case with no decision and one is reminded of the old saying that whilst a doctor buries his errors, those of the judge become the law of the land.

151 Evans, op. cit. footnote 78 at 68 argues that this is different for the Council’s legal responsibility, which is expressed by Parliament’s standing in Article 232 TEC proceedings. I do not find it helpful to consider the standing of Parliament in Court actions against the Council in terms of responsibility of the latter. In this manner, one could dilute the concept of responsibility by claiming the institutions are ‘responsible’ towards any individual who is directly and individually affected. 152 Rapport van de werkgroep ad hoc voor de behandeling van het vraagstuk van de uitbreiding van de bevoegdheden van het Europese Parlement (Vedel report), Bull. EC Suppl. 4/72 at 60. 153 Blumann, ‘Le Conseil européen’ (1976) at 12 cherished the idle hope that the Council could evolve into a ‘véritable gouvernement européen’. The nature of the institution diametrically opposes such a a development. 154 Case C-76/01P, Eurocoton vs Council, [2003] ECR I-10091. 155 For a discussion of this case law in the area of anti-dumping—which would exceed the limits of this study—see Vermulst and Waer, E.C. anti-dumping law and practice (1996) at 152 et seq. 156 On the matter see Driessen and Bordalba, ‘Commercial defence actions and other international trade developments in the European Communities XIV: 1 July 1997-30 June 1998’ (1999). 157 Article 9(2) of the Basic anti-dumping Regulation (OJ (1996) L56/1) provides that termination decisions are adopted by the Commission rather than the Council. What therefore should have happened is that the Commission, after noticing the political impossibility to impose definitive antidumping duties, ought to have adopted a Decision terminating the proceeding. This act would have been actionable in court. In other words, I humbly submit that Eurocoton and the Court of Justice wrongly reprimanded the Council when the real problem was that the Commission failed to adopt a termination decision.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

143

To the extent that the Council legislates, from a separation of institutions point of view, some checks and balances exist in the form of the Commission’s right of initiative, Parliament’s influence on the legislative process (at least to the extent that the applicable legislative procedure provides the MEPs with tangible prerogatives), Parliament’s rights of oversight (discussed in Part III, below) and—at least in theory—the control of national parliaments over their respective ministers. Notwithstanding the greatly improved transparency of the Council in recent years,158 these efforts have been considered insufficient by some and been ignored by others.159 Be that as it may, Article I-50 Const, which would have obliged the Council to legislate in public, would have improved Parliament’s oversight. As far as the Council acts in the second and third pillars and in the policy coordination areas of the first pillar, the check of the Commission’s right of initiative is largely absent, as is Parliament’s formal influence on decision making. To this extent, Parliamentary oversight of the Council is perhaps more of an issue in these areas, even though it is least developed in these parts of primary law. Here too, the weave of primary law does not allow the responsibility of the Council to Parliament to be sanctioned.160 5.6.2

The relationship between Commission and Council

The Commission as an institution is not responsible to the Council. This is evident from comparing Parliament’s controlling rights towards the Commission with the Council’s rights towards the executive. Moreover, whereas Parliament can dismiss the Commission, the Council has only the possibility to initiate the compulsory retirement of Commissioners pursuant to Article 216 TEC (III-349 Const).161 The idea of political responsibility towards the Council is not as alien as it might seem.162 It would give the Council a more recognisable place in the Union’s constitutional architecture and enhance the Commission’s democratic legitimacy. A —surmountable—disadvantage would be that the classical counterpart of the parliamentary system, the right of the executive to dissolve parliament, of course could not operate towards the Council. The solution for this would lie in an enforced

158 For example, the much improved access to documents, which currently implies that one half of documents is immediately published on the Internet, whilst many others are made publicly available later: Brunmayr, ‘Transparency’ (2005) at 136; the publication of voting records and explanations of votes as per Article 9 of the Council’s Rules of Procedure. On the Access to Documents Regulation see sub-section 7.2.2, below. 159 On 1 February 2006, years after this was made possible in the great majority of cases, Mr De Rossa MEP asked the Council whether ‘it will be possible to ascertain the voting position of each and every delegation in Council in these votes and that a public record of these votes and the positions taken by each and every delegation will be later available on the internet?’: question H-59/06 and Article 7(5) of the Council’s Rules of Procedure of 6 December 1993, OJ (1993) L304/1, respectively. 160 Separation of functions theorists would analyse the problem by arguing that the Council fulfils executive functions but is unchecked by the legislative ‘power’. 161 The original text of the EEC Treaty foresaw a power for the Council to temporarily suspend and replace individual Commissioners. This power, however, was abolished with the Merger Treaty: Article 160 TEEC in its original text. 162 And at the time of M. Giscard d’Estaing’s Convention, the Commission toyed with it, with the distinction that it was the European Council rather than the Council that would have exercised the sanction: Article 52 of the Feasibility study—Contribution to a preliminary draft Constitution of the European Union—Working document (the ‘Penelope’ draft) of 4 December 2002.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

144

majority, along the lines of the requirements for a motion of censure from Parliament.163 The Minister of Foreign Affairs proposed in the Constitution would constitute an exception to the rule. He contributes to the Union’s foreign and defence policies, which ‘he or she shall carry out as mandated by the Council’ (Article I-28.2 Const). As far as the sanction on responsibility is concerned, Article I-28.1 Const provides that the European Council could end the MFA’s tenure with qualified majority with the agreement of the Commission president. This solution of partial responsibility, which appears to have originated within the Commission,164 raises the question whether the agreement of the president falls under the Commission’s collective responsibility. As far as the other members of the College are concerned, Article I-27.3 Const is more charitable to the president than the post-Nice Article 217(4) TEC, which required the approval of the College for the dismissal of a Commissioner. The MFA would not be an ordinary Commissioner, but there is no good reason why he should be any more an exception to the Kanzlerprinzip than the Constitution requires. It would thus seem that, just like the prerogatives of Article I-27.3 Const, the president’s rights under Article I-28.1 vis-à-vis the MFA are unfettered by a decision from the College. 5.7

Conclusions of this Chapter

In Chapter 1 of this study the Research Question was posed to what extent it is possible to balance the EU institutions through the use of interinstitutional convention and what the conditions are for doing this. As far as the parliamentary system within the EU is concerned, the answer must be as follows. The parliamentary nature165 of the relationship between the Commission and Parliament flows from the rights of oversight that the Treaties and the Constitution bestow on the MEPs. The Commission’s responsibility to Parliament does not require it to give in to every demand of the majority of MEPs. Responsibility implies the power to make one’s own decision and to be accountable for it. The executive is not, and should not be, the hostage of Parliament. In the context of the rules on judicial protection this led the Court to reason that ‘while the Commission is required to ensure that the Parliament’s prerogatives are respected, that duty cannot go so far as to oblige it to adopt the Parliament’s position and bring an action . . . which the Commission itself considers unfounded’. The Court added that ‘each of the institutions must exercise its powers with due regard for the powers of the other institutions’.166 This reflects what in the constitutional practice of some Member States is known as ‘dualism’: a certain balance between government and parliament. In most Member States this is easier to maintain than in the European Parliament since the governments in many Member States wield the power of dissolution. Whether it is wise to use this depends on the political situation of the day; but when the system works properly, MPs will often be mitigated in their control over the government by the knowledge that extreme parliamentary pressure can backfire. 163 As Dutch constitutional practice shows, it is perfectly possible for an executive to be responsible to two houses of parliament simultaneously and to require the confidence of each of them. 164 ‘Positions and opposing positions on key points of institutional reform just as the debate begins’, Agence Europe of 13 December 2002 at item 1. 165 Dann uses the term ‘semi-parliamentary’, but this is based on his logic that Parliament has only very partial control over the Commission’s appointment: ‘European Parliament and executive federalism: approaching a parliament in a semi-parliamentary democracy’ (2003) at 551. 166 Case C-70/88, Parliament vs Council (Chernobyl), [1990] ECR I-2041, §§ 19 and 22.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

145

As in other parliamentary systems, parliamentary responsibility can be sanctioned through censure. Some have argued that ‘the threat of censure . . . is a horribly blunt weapon, too crude and powerful to be used on anything like a regular basis’.167 This is true, but it misses the point. In most parliamentary systems successful motions of censure are rare.168 However, it is not the sacking but the threat of sacking that puts urgency to the executive’s parliamentary responsibility. In this sense, the Santer crisis was a successful demonstration of the motion of censure,169 even though none was adopted. The EU’s parliamentary system does not provide for a right to dissolve Parliament. This is perhaps the reason (maybe less so historically than logically) why the motion of censure is dressed in a legal straitjacket: a reflection period of three days is required between the tabling of the motion and the vote, which must be open. In addition, a double majority is required of two-thirds of the present MEPs, who must constitute more than one-half of the total number. These guarantees are intended to ensure that Parliament’s competence to censure is offset with an appropriate check. These strict criteria moderate the threat of immediate dismissal, but do not change the principle of the parliamentary system. In addition to the legal concept of censure of the collective College, a conventional relationship of confidence of individual Commissioners is gaining hold. Parliament exacted from Mr Prodi the commitment that individual Commissioners could be sacked by him, especially if Parliament had expressed its ‘lack of confidence’.170 This, together with the Barroso commitments, led to a system whereby Parliament is getting very close to a system where individual Commissioners can be censured if they no longer commend the confidence of a majority of the House. Clearly, interinstitutional convention has changed the parliamentary system in place between the executive and the MEPs. Notwithstanding this possibility of individual responsibility and no-confidence, the ‘proud doctrine’171 of the collective responsibility of the Commission remains a vital tenet of the European constitution, needed to provide the executive with a coherence that it does not have by virtue of its appointment and which enhances its democratic legitimacy.172 The levers that enabled Parliament effectively to change the operation of the parliamentary system are twofold. Firstly, Parliament effectively exploited the possibilities of the discharge procedure. In 1999 it was essentially the impending refusal to grant discharge that caused the Santer Commission to founder. Even if a refused discharge does not equate censure, it weakens the executive politically to an extent that it becomes vulnerable to censure. Secondly, as far as the individual responsibility of Commissioners is concerned, Parliament successfully used its right of investiture at the time of the Barroso Commission to exact more concessions from the executive. 167 ‘Europe has to scratch its head’, Economist of 18 March 1999. 168 Raworth, ‘A timid step forwards: Maastricht and the democratisation of the European Community’ (1994) at 31. 169 Dann, op. cit. footnote 165 at 557. 170 Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121/122, § 10. 171 ‘Europe has to scratch its head’, Economist of 18 March 1999. 172 On this last point see Adapting the institutions to make a success of enlargement—Commission Opinion in accordance with Article 48 of the Treaty on European Union on the calling of a Conference of Representatives of the Governments of the Member States to amend the Treaties, Commission doc. COM (2000) 34 of 26 January 2000 at 11.

CHAPTER 5: THE PARLIAMENTARY SYSTEM

146

Whilst these developments must be applauded, a word of caution is needed. The European political system does not operate on the lines familiar in virtually all European countries of a ‘government’ opposed by an ‘opposition’. Whilst the example of Switzerland demonstrates that is it is not impossible for a political system to function on the basis of institutionalised ‘Konkordanzdemokratie’, this is alien to most European citizens. The consequence is that, for many citizens, the institutional transparency of the European political system is low. This, in consequence, affects Parliament (which is elected by fairly low turnouts)173 and, besides the incidence on the ‘input legitimation’ of the system, affects Parliament’s standing vis-à-vis the other institutions. In summary, the reply to the Research Question is that the development of the parliamentary system in the European Union demonstrates Parliament’s ability to alter the relationship it has with the executive. It managed to do so by agreeing interinstitutional conventions which, themselves, influenced primary law. It was able to do so because, ultimately it could rely on the ‘hard’ rights of the motion of censure (in 1999) and the right to refuse investiture (in 2004).

173 Although, if compared with e.g. federal elections in the United States, this should not be exaggerated. In years when no presidential elections take place, the turnout for House of Representative elections typically dips well under 40%: .

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

147

6

APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

6.1

Introduction to this Chapter

‘In order to lay a due foundation for that separate and distinct exercise of the different powers of government, which to a certain extent is admitted on all hands to be essential to the preservation of liberty’, argued Madison, ‘it is evident that each department should have a will of its own’. He concluded that ‘the members of each [department] should have as little agency as possible in the appointment of the members of the others’.1 In sub-section 2.4.4, above, the argument was made that the same principle applies for the separation of institutions. The original simplicity that adorned the Communities’ institutional set-up in the 1950s has long since been lost in successive Treaty amendments. Many offices and functions have been added to the original quadrangle Commission-ParliamentCouncil-Court. The appointment procedure for these offices differs for each of them. This Chapter analyses the interinstitutional conventions surrounding the appointment of the most important ones with the exception of the Commission, which was the subject of the preceding Chapters. Firstly the appointment procedures for executive offices (section 6.2), then those for legislative and mixed offices (6.4) and judicial offices (6.3) are reviewed. I dispense with discussing the appointments in the purely consultative EESC and Committee of the Regions. 6.2

Executive offices

Both the Treaties and the Constitution comprise certain categories of officers in the field of external and justice policy whose appointment procedures have caught Parliament’s attention. These are mainly CFSP Special Representatives, heads of Commission external delegations, EUROPOL appointments, and the president of the European Council. Special Representatives, who sometimes are styled Special Envoys, find their legal basis in Articles 18(5) TEU and III-302 Const. Since the Nice Treaty, the Council appoints them with a qualified majority.2 Special Representatives have been appointed for diverse regions and problems, such as Afghanistan, the African Great Lakes Region and Rwanda, Belarus, Bosnia and Herzegovina, Cambodia, FYROM, Ethiopia/Eritrea, Kosovo, Macedonia, the Middle East peace process and Palestine, the Stability Pact for South Eastern Europe, the Southern Caucasus and Yugoslavia.3

1 2 3

Madison in ‘Federalist Paper 51’, in Madison, Hamilton and Jay (‘Publius’), The Federalist Papers (1987) at 319. Article 23(2) TEU. The Council adopted standard guidelines on the appointing procedure: ‘I/A’ item note from the Secretariat to COREPER/Council—EU Special Representatives: Guidelines on appointing procedure and administrative arrangements, Council doc. 7089/00 of 20 March 2000. See for example Council Joint Action of 10 December 2001 concerning the appointment of the Special Representative of the European Union, OJ (2001) L326/1. A somewhat special case was the Administrator for Mostar. For a list of special representatives see Piris, The Constitution for Europe—A legal analysis (2006) at 148-149.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

148

The appointment decision sometimes includes brief pointers delimiting the Special Representative’s mandate.4 The Treaties make the appointment of Special Representatives solely a Council affair. Once the Council agrees on the advisability of a Special Representative, the Member States and the institutions propose candidates. The Council Secretariat prepares a draft Joint Action which sets out the mandate, duration, terms of reference, criteria for the Representative’s team, criteria and modalities for reporting, financing arrangements including co-operation with the Commission, and provisions on immunities and privileges. The financial arrangements are co-ordinated with the Commission. The Member States’ Foreign Relations Counsellors discuss the draft Joint Action and related financial statements, which are then approved by the hierarchy of preparatory bodies and, ultimately, the Council. The appointed Special Representative selects his own team in consultation with the Presidency, ‘assisted by the SG/HR and in full association with the Commission’.5 Neither the EU Treaty nor the Constitution6 foresees a role for the MEPs in the appointment of Special Representatives. Nevertheless Parliament, following the practice of the US Congress,7 occasionally insists on the appointment of a Special Representative.8 Moreover, it adopted provisions in its Rules of Procedure for cases where the Council proposes to appoint a Special Representative. Rule 86(1) foresees that: ‘[w]here the Council intends to appoint a Special Representative . . . the President [of Parliament], at the request of the committee responsible, shall invite the Council to make a statement and answer questions concerning the mandate, the objectives and other relevant matters relating to the tasks and role to be performed by the Special Representative.’ The second paragraph adds that ‘[o]nce the Special Representative has been appointed, but prior to taking up the position, the appointee may be invited to appear before the committee responsible to make a statement and answer questions’.

4

5 6 7 8

See e.g. Joint Action of 8 June 1998 adopted by the Council on the basis of Article J.3 of the Treaty on European Union concerning the nomination of an EU Special Representative for the Federal Republic of Yugoslavia (FRY), OJ (1998) L165/2, corr OJ (1998) L 187/56 and Council Joint Action of 14 December 2000 appointing the European Union Special Representative for the Middle East peace process and repealing Joint Action 96/676/CFSP, OJ (2000) L318/5. Forwarding note from the Secretariat to DelegationsSuggestions for procedures for coherent, comprehensive EU crisis management, Council doc. 5633/01 of 24 January 2001 at 34. Article III-302 Const would change the procedure in one important respect by giving the Minister of Foreign Affairs the initiative for proposing special representatives. See for example ‘Dalai Lama meets Powell, China grumbles’, CNN.net of 23 May 2001. See e.g. its Resolution on the appointment of an EU special representative for Tibet, OJ (1998) C34/169, soon followed by another Resolution on Tibet, OJ (1998) C167/223. A few months later the Council stated in reply to a written question that ‘it has not decided to appoint a special representative for Tibet’: reply to written question 1842/98 by Mr Moorhouse to the Council, OJ (1999) C13/98. Other instances of requests for Special Representatives include the Resolution on the ASEM process (Europe-Asia relations), OJ (1998) C104/217, which led to the appointment of a Special Representative; a Special Representative for Albania (Resolution on the situation in Albania, OJ (1997) C115/158); an envoy for East Timor (Resolution on the 2001 budget in view of the conciliation procedure before the Council’s first reading, OJ (2001) C121/146); and an envoy for children’s rights (Resolution on the EU positions in the Special Session on Children of the UN General Assembly, OJ (2002) CE65/366, § 7).

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

149

Within three months of the hearing, the committee may propose a Recommendation relating directly to the statement made and answers provided. The Special Representative is then ‘invited to keep Parliament fully and regularly informed as to the practical implementation of his mandate’. Article 21 TEU is a shallow legal basis for these latter rules, since this provision strictly speaking obliges only the Presidency and the Commission to inform Parliament. The Special Representative, however, is an official of neither. To the extent that Rule 86 is respected, it is at most the reflection of interinstitutional convention but perhaps not more than comity. For completeness’ sake, the special Representatives of the Presidency must be mentioned. These are a somewhat different type of animal, although the outside world will hardly know the difference. They find their legal basis in Article 18(2) TEU and fall under the Presidency’s responsibility.9 They would appear to fall within the remit of Article 21. Under the regime of the Constitution, the Special Representatives would fall under the political responsibility of the MFA and this would give Parliament a better basis to influence their appointment and operation. Heads of Commission external delegations. Presumably with an eye to the scrutiny of candidate ambassadors by the American Congress, Parliament adopted rules foreseeing the heads of Commission delegations to be invited to make a statement and answer questions. 10 Rule 88 of its Rules of Procedure foresees that ‘[w]ithin three months of the hearing . . . the committee responsible may adopt a resolution or make a recommendation, as appropriate, relating directly to the statement made and answers provided’. The logic of interviewing heads of delegations is less evident11 than it is in the United States, where many ambassadors tend to be political appointees. In the European Union on the other hand, most heads of delegations tend to be career officials. These rules do not seem to be used in practice. EUROPOL. In 2002 Parliament added the current Rule 93, third paragraph to its Rules of Procedure. This states that the procedure in Parliament for the appointment of the members of the Court of Auditors, discussed in sub-section 6.4.1 below, applies mutatis mutandis to the appointment of EUROPOL’s Director and Board members. Parliament’s rapporteur Corbett justified the amendment by noting that this procedure ‘should apply likewise to appointments based on special legal instruments’.12 That suggests that other special officers, or new offices to be created in the future, may be submitted to a similar procedure. Thus, the amendment may have a potential impact reaching considerably beyond the appointment of EUROPOL officers. The president of the European Council. The office of president of the European Council is proposed in the Constitution. It is the result of a double compromise. On 9 10 11 12

‘I/A’ item note from the Secretariat to COREPER/Council—EU Special Representatives: Guidelines on appointing procedure and administrative arrangements, Council doc. 7089/00 of 20 March 2000. This follows from its Resolution on consultation of the European Parliament on the appointment of senior officials by the Commission and on the Community’s diplomatic representation, OJ (1989) C120/340. Gosalbo Bono, ‘The international powers of the European Parliament, the democratic deficit and the Treaty on European Union’ (1993) at 128 considers it ‘excessive’. Draft report on the General Revision of the Rules of Procedure, Parliament doc. PE 294.770 of 8 June 2001 at 21.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

150

the one hand, some Member States favoured a strong character able to make a powerful European voice heard in international politics. On the other hand, the Commission displayed some concerns as regards the Commission president’s prerogatives.13 Additionally, many politicians from smaller states feared that the post would automatically be occupied by the representative of a large Member State and they favoured more a ‘chairman’ than a ‘president’. Although the connotations behind these two notions are very different, in French they are represented by the same word and in the end, the term ‘president’ won the day. However, more substantively, the president’s powers were clipped considerably, making him more of a chairman of the quarterly European Council meetings or a ‘German-style’14 president than a full-time equivalent to his American homonym. On the one hand, Piris argues that the president of the European Council might be inclined to distance himself from his Member State in order to obtain the trust from the others.15 If the person was a ‘heavy-weight’ politician, he could successfully work towards a more efficient and visible Union. It is, on the other hand, not certain that the office would develop in this manner. Large Member States fear each other more than they fear other states. Moreover, no large Member State leader is prone to take second place to a European ‘superior’:16 neither when that ‘superior’ originates from a small Member State (as a matter of power), nor when he comes from a large one (as a matter of jealousy). The IGC agreed formally to involve the European Council president in the preparation of European Council meetings. Contrary to some views,17 I doubt whether this really added much that was not implied already in the office as foreseen by M. Giscard d’Estaing’s Convention: it is hard to imagine that the European Council president would remain inactive during the 250 working days a year when there is no European Council. Since Article III-295 Const would provide him with the power to call for a European Council if international developments so require, his involvement in the preparation of these meetings would seem implied in any event. The office cannot be seen separately from those of the Commission president and the MFA. First of all, the Constitution left ample room for turf wars between these three officers.18 The European Council president would have a role in the external representation of the Union in the CFSP arena, ‘without prejudice to the powers of the Union Minister for Foreign Affairs’.19 This role is presumably intended to make him a counterweight to the US president. However, where the latter’s role in external relations flows naturally from the Madison-style separation of powers structure of the American Constitution, the European Council president’s involvement can only hinder the other actors in this field without any such separation of powers logic.20 13 14 15 16 17 18 19 20

Smulders, ‘Kritische kanttekeningen bij de gevolgen van het “ontwerp-verdrag tot vaststelling van een grondwet voor Europa” voor het institutionele evenwicht’ (2003) at 248. Gordos and Ódor, ‘Az EU alkotmányozási folyamata magyar szemszögből a Kormányközi Konferencia alapszaka után’ (2004) at 21. Piris, op. cit. footnote 3 at 185. Jacobs, ‘De geboorte van de Europese Grondwet nadert’ (2003) at 1433. Jacobs, ‘De Europese Grondwet weer een stap dichterbij’ (2004) at 1534. Ponzano, ‘La réforme des institutions de l’Union européenne dans le cadre de la Constitution’ (2004) at 29-30. Article I-22.2 Const. Maddalon, ‘L’action extérieure de l’Union européenne’ (2005) at 525 attempts to separate the respective roles of the president and of the MFA into a ‘constitutional’ one for the former and a more ‘executive’ one for the latter. It remains to be seen whether practice would neatly align to such a distinction.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

151

Secondly—and this is of some more importance for this study—, his appointment would be politically linked to that of the other two officers. The IGC adopted a Declaration on Articles I-22, I-27 and I-28 Const, stating that ‘[i]n choosing the persons called upon to hold the offices of President of the European Council, President of the Commission and Union Minister for Foreign Affairs, due account is to be taken of the need to respect the geographical and demographic diversity of the Union and its Member States’.21 The Declaration can hardly be more than a political understanding among the Member States that the persons nominated as Commission president and MFA and the one appointed as European Council president should not originate from the same state. If anything, these limitations are likely to limit the choice of candidate Commission presidents, now that the European Council could only select the latter taking into account the elections of Parliament. It is hard to see how this would contribute to strengthening the link between the European elections and the appointment of the Commission. If anything, the linkage must therefore be regretted. 6.3

Judicial offices

The judges of the European Court of Justice and of the Court of First Instance are appointed by common accord of the governments of the Member States for a term of six years.22 Each Court comprises one national from each Member State. The Judges select their own president as well as their Registrar.23 The current appointment procedure in practice amounts to every government nominating ‘its’ judges, because Member State governments will not object to the choices of other Member States. If they were to do so, it would invite reciprocity and, ultimately, unhelpful tit-for-tat behaviour. The consequence of the appointment system under the Treaties is that every Member State government has a certain measure of control over the appointment of its judges, but no government controls the appointment of the bench as such. Firstly, it obviously takes more than one judge to render a judgment.24 Secondly, generally the quality of European judges is high.25 Thirdly, an appointed judge has little interest to be profiled too politically. The chance that the government appointing him is still in office by the time his reappointment comes up is slim. From a separation of institutions point of view, the European system of appointments leads to a much less politically inclined Court than the American system, where Congress and president heavily influence the political line of the Supreme Court.

21 22

23 24 25

Declaration on Articles I-22, I-27 and I-28, OJ (2004) C310/420. Article 223 and 224 TEC and I-29.2 Const, respectively. In early 2000 the idea was floated to appoint the judges with qualified majority: Presidency note— IGC 2000: Possible extension of qualified majority voting—Articles which could move to qualified majority voting as they stand, CONFER 4706/1/00 of 11 February 2000. It made it neither into the Nice Treaty nor into the Constitution. I will not further discuss the three-year renewal of the bench. The exception is the unus iudex of Article 11 of the Court of First Instance’s Rules of Procedure, OJ (2003) C193/41. For a different opinion see Wouters, ‘Institutional and constitutional challenges for the European Union—some reflections in the light of the Treaty of Nice’ (2001) at 345.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

152

Since the early 1960s, Parliament has attempted to be involved in the appointment procedure,26 apparently something resisted by the Judges.27 The outcome of this pressure is that Article III-357 Const would foresee the participation of one Parliament-appointed member in a consultative panel of seven jurists. With this solution, which does not give the panel any more than advisory powers on the appointment, the overall political neutrality of the Court is maintained. A Parliamentary right of approval of the appointments would lead to a politicisation of the Courts.28 Parliament’s Rules of Procedure, which comprise procedures for the appointment of the Court of Auditors and the ECB Executive Board, provide no rules for the Judges. This difference is not illogical. The Treaties grant the MEPs at least a right of consultation on the appointment of the Executive Board and the Auditors. In the case of the Judges there is no such fertile ground on which interinstitutional practice could grow. Lastly, the appointment of judges for the Civil Service Tribunal should be mentioned. That Tribunal, which finds its Treaty basis in the Articles 225a TEC and 140b Euratom inserted with the Treaty of Nice, consists of seven judges and is consequently the first judicial body consisting of less judges than there are Member States. Article 225a provides that the judges are to be appointed by the Council. The Decision establishing the Tribunal provides for a committee of seven persons chosen from former members of the two Courts and lawyers of recognised experience that must give an opinion on candidates’ suitability to perform the duties of a judge at the Tribunal. The committee must also draw up a list of most-suitable candidates. That list must comprise at least twice the number of names as there are candidates.29 The committee’s composition is to be decided by Council Decision, acting by a qualified majority on a recommendation by the president of the Court of Justice. As to the substantive criteria for the judges, Article 225a lays down little more than that they ‘shall be chosen from persons whose independence is beyond doubt and who possess the ability required for appointment to judicial office’. The Decision adds that the Council must ensure ‘a balanced composition of the Tribunal on as broad a geographical basis as possible from among nationals of the Member States and with respect to the national legal systems represented’.30 It seems that the

26

27 28 29 30

See e.g. its Résolution sur les compétences et les pouvoirs du Parlement européen, JO (1963) 1916 and its Resolution embodying (i) Parliament’s opinion on the convening of the Intergovernmental Conference, and (ii) an evaluation of the work of the Reflection Group and a definition of the political priorities of the European Parliament with a view to the Intergovernmental Conference, OJ (1996) C96/77. Five years later, in its Resolution on the Treaty of Nice and the future of the European Union, OJ (2002) CE47/108, it deplored ‘the fact that the members of the courts will continue to be appointed by common accord of the Member States and that their case therefore constitutes the only exception to the general rule established by the Treaty of Nice whereby appointments are made by decision of the Council on the basis of a qualified-majority vote’. Corbett, Jacobs and Shackleton, ‘The European Parliament at fifty: a view from the inside’ (2003) at 368. Piris, ‘After Maastricht, are the Community institutions more efficacious, more democratic and more transparent?’ (1994) at 482. Article 3 of the Annex to the Statute of the Court, as added by Council Decision of 2 November 2004 establishing the European Union Civil Service Tribunal, OJ (2004) L333/7. A subsequent Council Decision of 18 January 2005 concerning the conditions and arrangements governing the submission and processing of applications for appointment as a judge of the European Union Civil Service Tribunal, OJ (2005) L50/7 provides for some other minor conditions in point 5 of its Annex.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

153

geographical criterion would render it illegal to recruit two judges from the same Member State.31 This appointment procedure leaves the ultimate appointment solely in the hands of the Council but that is not an important matter for three reasons. Firstly, the ‘technical’ remit of the Tribunal, whilst important for the European Union’s civil service, will rarely allow it to deal with momentous questions of policy. Secondly, the Tribunal’s judgments are subject to appeal to the Court of First Instance. Thirdly, the advisory committee procedure will discourage the Council from making appointments radically different from those proposed to it. 6.4

Legislative and mixed offices

6.4.1

Members of the Court of Auditors

Article 247(3) TEC (III-385.2 Const) grants the MEPs only a consultative right on the appointment of the members of the Court of Auditors, even though Parliament has requested a right of assent for a long time.32 The MEPs were disappointed in both Amsterdam and Nice. Parliament has tried to achieve through resolutions at least part of what it failed to acquire during the IGCs33 and, even though this unilateral path has not led to Treaty revision, it was able to put procedures in place that are respected by the Council in practice. These interinstitutional conventions are both procedural and substantive34 in nature. I will focus on the procedural aspects. The nominations for candidates must be presented with their CVs and any other relevant information to Parliament at least ten weeks before the appointment date.35 The candidates are invited to make a statement before the Parliament’s Committee on Budgetary Control and answer questions put by its members.36 In its 1992 resolution on the matter, Parliament requested to be given ‘[a] reasonable amount of time for consideration’ so as to enable the Committee to organise hearings and vote. The votes in committee and in the plenary are secret. The discussion at the hearing is reduced to a minimum ‘in order not to subject the candidates to personal criticism in public’37—a peculiar rule, which is not applied to candidateCommissioners. The Committee then makes a recommendation to Parliament’s plenary.38 In principle, Parliament votes within two months of receipt of the nomination unless Parliament, at the request of the committee responsible, a political group or at least thirty-seven Members, decides otherwise. The vote is secret, per candidate and a 31 32

33 34 35 36 37 38

Osztovits, ‘Az Európai Unió Közszolgálati Törvényszéke—megjegyzések az Európai Közösségek Bírósága szervezeti reformjának legújabb eredményéhez’ (2005) at 13 sees no legal problem in such a situation, but considers it highly unlikely in practice. E.g. its Resolution on procedures to follow when Parliament is consulted in connection with appointment of Members of the Court of Auditors, OJ (1995) C43/75, and Resolution embodying (i) Parliament’s opinion on the convening of the Intergovernmental Conference, and (ii) an evaluation of the work of the Reflection Group and a definition of the political priorities of the European Parliament with a view to the Intergovernmental Conference, OJ (1996) C96/77. OJ (1995) C43/75. For the substantive requirements see § I.4 of the Resolution on the procedure for consulting the European Parliament on the appointment of Members of the Court of Auditors, OJ (1992) C337/51. OJ (1995) C43/75, § 2. Rule 101(1) of the Parliament’s Rules of Procedure. OJ (1992) C337/51. Rule 101(2).

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

154

majority of votes cast is required.39 Since Parliament has no right of assent, a negative vote can only induce the president of Parliament to ‘request the Council to withdraw its nomination and submit a new nomination to Parliament’.40 Parliament votes with absolute majority in the plenary session.41 The Council never agreed to a formal legal commitment to Parliament. It did, however, issue a political statement to the effect that ‘[t]he Council confirms that when appointing members of the Court of Auditors, it will give full consideration to the Opinion on this subject delivered by the Parliament’.42 Practice bears out that, at least to some extent, the ministers feel themselves bound by Parliament’slegally not bindingopinion. For example, in November 1989 the MEPs failed to approve the French and Greek candidates. The French candidate was indeed withdrawn, whilst the Greek government was unable to find a better candidate. However, in 1993 a rejected candidate was not withdrawn.43 Academically, Parliament’s quest for the right of assent on the appointment of Court of Auditors has found some support, especially now that the MEPs have the right of investiture as regards the appointment of the Commission.44 Practice suggests, however, that interinstitutional convention may well develop in this respect, filling to a large extent the void caused by the Treaties.45 From a separation of institutions point of view, this development may be applauded. However, from that perspective, little would be gained if the MEPs were to obtain full control over the appointment of the Court of Auditors. 6.4.2

The Executive Board of the European Central Bank

The president, vice-president and other members of the Executive Board46 of the European Central Bank are appointed by the European Council on the basis of a Council recommendation. For this purpose only the Member States that have adopted the euro have voting rights47 although, for the remainder, the Council functions as normal. Thus, in 1998, the Council recommendation was signed by British Chancellor of the Exchequer Brown, who happened to be Council president at the time,48 even though the United Kingdom has not adopted the euro. 39 40 41 42 43 44 45 46

47 48

Rule 101(3). Rule 101(4). OJ (1992) C337/51, § I.5. That text uses the somewhat Stalinist-sounding wording ‘Parliament will ensure that candidates are approved by an absolute majority’, which probably must not be read too literally. Annex IV of Renforcement des pouvoirs budgétaires de l’Assemblée, Council doc. R/227/75 of 23 January 1975. Corbett, The European Parliament’s role in closer EU integration (1998) at 128-129. In this sense Dashwood, ‘The constitution of the European Union after Nice: law-making procedures’ (2001) at 230. Article III-385.2 Const does not grant Parliament more than a right of consultation. Notwithstanding the clear language of Article 112(2)(a) TEC the president is styled in full the ‘president of the European Central Bank’ as opposed to the ‘president of the Executive Board of the ECB’. In 2002 the Council recommended Mr Papademos for the function of vice-president of the Executive Board of the ECB (OJ (2002) L101/17), which provoked a rather sharp rebuke from the Bank’s Governing Council: OJ (2002) C100/8, § 4. For future practice, the two institutions agreed informally to skip the ‘Executive Board’ bit in the title. The matter is practically only relevant for the size of business cards. Articles 122 TEC and III-197 Const. Council Recommendation of 3 May 1998 on the appointments of the President, the Vice-President and the other members of the Executive Board of the European Central Bank, OJ (1998) L139/36.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

155

Under current practice, the candidate is first agreed during an informal ECOFIN Council meeting. The recommendation is then formally adopted by the next available Council. The ECB’s Governing Council and Parliament are consulted on this recommendation, following which the heads of state or government of the eurozone make the appointment. This order of events was used when the first Executive Board was appointed in 1998 and again when a new vice-president was appointed in 2002. Apparently, in 1998 some in Parliament murmured that it should have been given a choice between different candidates.49 The text of the provision concerned indeed provides some elements which could support this view. Under Article 112(2)(b) TEC, ‘[t]he President . . . shall be appointed . . . by common accord of the governments of the Member States at the level of Heads of State or Government, on a recommendation from the Council, after it has consulted the European Parliament and the Governing Council of the ECB.’50 The intriguing word here is ‘it’. In the English text it can only refer to ‘the Council’. The words ‘after . . . ECB’ clarify that the appointment by the heads of state or government takes place following the consultation of Parliament and the ECB’s Governing Council. The English and French texts leave open the possibility that the Parliament and the ECB would be consulted before the Council adopts its recommendation. The French text is even ambiguous who consults the MEPs and the bankers.51 The situation is rather different in the German and Dutch versions. The latter uses the completed past tense (‘de Raad, die . . . heeft geraadpleegd’), indicating that the Council recommendation follows the consultation. In German, the appointment occurs after ‘Empfehlung des Rates, der hierzu das Europäische Parlament und den EZB-Rat anhört’. The word ‘hierzu’ (‘for this purpose’) again allows only for the interpretation that the consultations take place before the Council recommendation is adopted.52 Based on the Court’s case law an argument could be attempted that the correct interpretation would call for the application of the Dutch and German language versions, since all other language versions can be reconciled with this procedure.53 However, Parliament accepted a different interpretation, apparently because it was unsure on what it otherwise would be consulted. Rule 102(3) of its Rules of Procedure54 provides that the plenary vote must take place ‘within two months of the 49

50 51

52 53 54

Strangely enough, Parliament has not maintained this stance. In 2002 it merely asked for greater involvement in nominating ECB executive board members, including the capability to confirm their appointments: Resolution on the 2000 Annual Report of the European Central Bank, OJ (2002) CE65/159, § 16. See also the Resolution on democratic accountability in the third phase of EMU, OJ (1998) C138/177, § 16. Emphasis added. ‘Le président, le vice-président et les autres membres du directoire sont nommés d’un commun accord par les gouvernements des États membres au niveau des chefs d’État ou de gouvernement, sur recommandation du Conseil et après consultation du Parlement européen et du conseil des gouverneurs de la BCE . . .’ The Italian appears to be similarly vague: ‘Il presidente . . . sono nominati . . . di comune accordo dai governi degli Stati membri a livello di capi di Stato o di governo, su raccomandazione del Consiglio e previa consultazione del Parlamento europeo e del consiglio direttivo della BCE’. Article 112(2)(b) in its Hungarian version similarly uses this sequence in a formula that leaves no ambiguity: ‘az Európai Parlamenttel . . . folytatott konzultációt követő ajánlása alapján’ (literally ‘on the basis of the recommendation that follows the consultation of the European Parliament . . .’). See the case law cited in footnotes 63 and 63 of Chapter 5 of this study. Rule 102 is quite similar to Rule 101, which regulates the nomination of the members of the Court

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

156

receipt of the nomination’. The last word, in singular, indicates that the MEPs do not expect to be presented with more than one candidate for each function. Deprived of the element of choice, the consultation procedure becomes a rather toothless affair within Parliament. The sole candidate is asked to fill out a questionnaire55 and will normally make a statement to Parliament’s Economic and Monetary Affairs Committee. Contrary to practice elsewhere in Parliament, the ensuing debate tends to be relatively spontaneous although the outcome can rarely be in doubt. The Committee adopts a recommendation, which is voted upon in plenary. If Parliament’s opinion is negative, its president must request the Council to withdraw its nomination and submit a new nomination to it.56 Following the consultations, the heads of State or government of the countries that have adopted the euro appoint the candidate. Consequently, the consultation process is so formalised that it provides Parliament with virtually no real influence, but the MEPs must largely blame their own interpretation of the Treaty and the ensuing interinstitutional convention that they subsequently incorporated in their Rules of Procedure. The drafters of the Constitution chose to enact current practice into the text of Article III-382 Const. The Dutch and German versions were adapted accordingly. Note that the European Council assumes the role previously played by the ‘governments of the Member States at the level of Heads of State or Government’. Additionally, the European Council makes the appointment by qualified majority. As a final comment, the procedure is top-heavy. Originally devised to appoint the whole top of the Executive Board in a grandiose formal manner, it requires an enormous effort simply to replace its individual members. Whether the institutional transparency of the Union gains from such a procedure is to be doubted. 6.4.3

The Ombudsman

The appointment of the Ombudsman is Parliament’s prerogative. Article 195(2) TEC (III-335.2 Const) provides that it takes place following each election of Parliament, for the duration of its term of office. The Ombudsman is eligible for reappointment. Candidates must have the support of at least 37 MEPs and are heard by the Committee on Petitions.57 Article 195(2) provides for the possibility to dismiss the Ombudsman.58 The dismissal may be pronounced ‘by the Court of Justice at the request of the European Parliament if he no longer fulfils the conditions required for the performance of his duties or if he is guilty of serious misconduct’. It follows from this wording and from the nature of the office thatbar death59 and resignation60these are the only grounds for which an Ombudsman may lose his office.

55 56 57 58

59

of Auditors. Council doc. 8299/98 of 6 May 1998. Rule 102(4) of the Parliament’s Rules of Procedure. The remainder of the procedural rules can be found in Rule 194 of Parliament’s Rules of Procedure. Essentially repeated in Article 8 of the Decision of the European Parliament of 9 March 1994 on the regulations and general conditions governing the performance of the Ombudsman’s duties, OJ (1994) L113/15. The Decision was amended in 2002 to adapt it to changes in the budget: Decision of the European Parliament of 14 March 2002 amending Decision 94/262/ECSC, EC, Euratom on the regulations and general conditions governing the performance of the Ombudsman’s duties, OJ (2002) L92/13. An eventuality overlooked in the Treaties, but not by Parliament: Rule 194(8) of its Rules of Procedure.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

157

Parliament has regulated the dismissal procedure in its Rules of Procedure. In short, at least one tenth of Parliament has to initiate the procedure. The request is forwarded to the Ombudsman and to the Committee on Petitions. This Committee will present a report to the plenary with the approval of a majority of its members. The Ombudsman may be heard, after which a decision is taken for which a quorum of at least one half of Parliament is required. The Ombudsman is thus requested to vacate his office. If he would refuse to do so, the president of Parliament must request the Court of Justice to provide a ruling.61 The Council criticised this procedure, arguing that the request to the Ombudsman to resign is not provided for in the Treaty.62 However, in my opinion the situation seems more complicated. Parliament (or for that matter, anybody) is free to request the Ombudsman to step down. In principle, the Ombudsman is equally free to refuse. To build in such a procedure does not impair the Ombudsman’s prerogatives. Rather, it provides for a more gracious way out than a dismissal by the Court would be. No Ombudsman has ever been dismissed, giving little opportunity for practice or convention to develop in this area. 6.4.4

The European Data Protection Supervisor

The European Data Protection Supervisor (EDPS) is the product of the Data Protection Regulation,63 which applies to the institutions what the Data Protection Directive64 imposes onto the rest of Europe. He is an independent supervisory authority overseeing rules and practice in the field of data protection within the European institutions.65 Article 42 of the Regulation provides that Parliament and the Council appoint the EDPS and the Assistant Supervisor by common accord on the basis of a list drawn up by the Commission following a public call for candidates. Like the Ombudsman, on whose office his is loosely modelled, the duration of his term is five years. The first EDPS is a Dutchman.66 His deputy is Spanish, partially to ensure a balance between North and Southern Europe.67 It is probable that a similar practice will be followed in the future, although this does not seem to be convention. Lastly, Article 43 of the Regulation provides a specific legal basis for an interinstitutional agreement between the three political institutions for determining his duties, salary, allowances and any other benefits.68 60 61 62 63 64 65 66 67 68

A logical possibility allowed by Article 7 of Parliament’s Decision of 9 March 1994. Rule 196 of Parliament’s Rules of Procedure. Letter from the President of the Council to the President of the Parliament, Council doc. 4605/94 of 2 February 1994. Regulation (EC) 45/2001 of the European Parliament and of the Council of 18 December 2000 on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data, OJ (2001) L8/1. Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data, OJ (1995) L281/31. Ibid., Article 41. Decision of the European Parliament and of the Council of 22 December 2003 appointing the independent supervisory body provided for in Article 286 of the EC Treaty (European Data Protection Supervisor), OJ (2004) L12/47. ‘Nederlander aan hoofd nieuw Europees bureau’, NRC-Handelsblad of 20 December 2003. This took the form a Decision No 1247/2002/EC of the European Parliament, of the Council and of the Commission of 1 July 2002 on the regulations and general conditions governing the performance of the European Data-protection Supervisor’s duties, OJ (2002) L183/1.

CHAPTER 6: APPOINTMENTS OF MEMBERS OF OTHER INSTITUTIONS AND BODIES

6.5

158

Conclusions of this Chapter and of this Part

The analysis of the appointment and dismissal procedures for different officers in the EU’s institutions and bodies leads to several conclusions. Firstly, although some of these procedures have spawned interinstitutional practice and convention, some of this benefited Parliament, but some of it did not. As regards the appointment of the Commission, the practice prevalent until a few years ago that each Member State selects ‘its’ Commissioners nationalised the Commission. Partially as a result of pressures from Parliament, existing interinstitutional conventions were enacted in Maastricht into a veritable right of investiture. Subsequently, Parliament could use this prerogative to expand its influence on the Commission. Whereas, up to the 1990s, the appointment of the Commission was more a matter of practice among the Member States only, the appointment of the ECB’s Executive Board does require the consultation of Parliament. Peculiarly, Parliament essentially agreed to a convention that provides it with very little influence over the procedure. On the other hand, Parliament has been somewhat more successful as far as the appointment of the Court of Auditors is concerned. Secondly, unlike the British situation, many of the conventions in the European Union are, so to speak, ‘pinned’ to a rule of constitutional law. For example, the convention resulting from Parliament’s initial reluctance to invest the Barroso Commission were only possible because the MEPs had the legal right to refuse investiture. Thirdly, the individual responsibility of Commissioners towards Parliament, built purely on political obligations agreed by the Commission, is but one example of an interinstitutional convention that shifts the balance of influence between the institutions. Fourthly, in Chapter 1 the possible ‘meta-constitutional’ impact of interinstitutional convention was noted. Indeed, some of the appointment procedures discussed in this Part illustrate the possibility of interinstitutional convention to lead to changes in primary law. Again the appointment of the ECB’s Executive Board can serve as an example. After the institutions agreed that Parliament was to be consulted only following the Council’s recommendation, the authors of the Constitution adapted the text of the provision to fit the existing convention. In this way, interinstitutional convention influenced the drafting of primary law. Of perhaps more moment is Mr Prodi’s commitment—which was required and accepted by the MEPs—to require from each of his Commissioners to step down if so requested. This requirement, an essential step in the creation of individual political responsibility of Commissioners, was subsequently anchored in the Treaty of Nice and found its way into the Constitution. These examples show that, as far as the appointment and dismissal of the Union’s officers is concerned, interinstitutional convention plays both a direct and a metaconstitutional role.

159

PART III: PARLIAMENTARY OVERSIGHT

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

7

PARLIAMENT’S INFORMATION RIGHTS

7.1

Introduction to this Part

161

The previous Part discussed the checks and balances between the institutions at the level of the appointment of their respective officers. In this and the following Parts, I investigate the checks and balances that keep the institutions separate yet connected at the level of their operation. This Part concentrates on Parliament’s rights of oversight over policy making by the Commission, Council and European Council. The next Part discusses the checks and balances concerning the influencing of policy and the resulting legislation. Parliament’s rights of oversight fall apart in its right to information towards the other political institutions (the subject of this Chapter) and its prerogatives of oversight that are not specifically aimed at the Council or Commission, but can be used towards them (Chapter 8). Most textbooks on European constitutional law treat the MEPs’ information rights as a more mundane matter than it is. Whilst it is true that Parliament’s right to ask questions is not the heaviest artillery it can wield against other institutions,1 it is the most used weapon. For any parliament, information is a sine qua non for effective oversight. The information rights are the nuts and bolts of parliamentary responsibility. This area is rich in interinstitutional convention. Article 197 TEC2 provides as follows: ‘The European Parliament shall elect its President and its officers from among its Members. Members of the Commission may attend all meetings and shall, at their request, be heard on behalf of the Commission. The Commission shall reply orally or in writing to questions put to it by the European Parliament or by its Members. The Council shall be heard by the European Parliament in accordance with the conditions laid down by the Council in its Rules of Procedure.’ As far as the Commission is concerned, this provision implies a duty to reply. Hence it is responsible to Parliament. For the Commission there is only a freedom of means. It is within its province to decide whether the reply is to be oral or in writing. The wording of Article 197 testifies to the fundamentally different legal position of the Council and European Council. In the formative years of the Communities some considered the Council as a ‘House of Representatives of Government’.3 Such a view, however, does little justice to neither the hybrid character nor the central role of the institution. Moreover, since the 1980s both the Council and the European Council acquired considerable executive functions in addition to their legislative/policy tasks.

1 2 3

See in this sense the research by Van Schendelen, Het Europees Parlement (1984) at 80. Roughly corresponding to Article III-337 Const. E.g. Forging ahead. Thirty years of the European Parliament, 1952-1982, unnumbered doc. (1983) at 166.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

162

As argued in Chapter 5, the parliamentary responsibility of either of those institutions towards Parliament cannot be sanctioned. In 1958 the Council decided to reply to questions from Parliament, under the condition that the Council must have taken a decision. Moreover, the Council specified that it must have acted as an institution.4 The first condition is logical. The Council is naturally loath to pronounce on the reasons why it has not (yet) taken a decision and cannot be expected to comment on the position of one or more of its members.5 Nevertheless, often it will be interesting to know why a decision was not taken6 and what the voting patterns within the Council are. This latter problem is considerably alleviated with the improved transparency of Council sessions (at least in its legislative capacity),7 although the work in its subordinate bodies8 is less accessible the general public. Moreover, in 1960 it accepted to reply to questions concerning matters that have already been the subject of a ‘certaine exploration ou élaboration’ in that institution.9 Article 197 refers to the Council’s Rules of Procedure (the European Council has none). The Council’s Rules offer little guidance other than providing for the Council’s representation in Parliament (a matter discussed below) as well as implementing rules for the Access to Documents Regulation. They do provide that the Council may ‘present its views to the European Parliament by means of a written statement’ (Article 26). The 1958 decision was not just a matter of practice: although the letter of Article 197 TEC (and even of III-337 Const) allow the Council to revoke it,10 this would be politically impossible.11 The Council’s duty to reply to Parliamentary questions, then, may not exist in law but is a product of interinstitutional convention. The direct consequence of the Council’s decision was that its relationship to Parliament changed, compared with what the authors of the EEC Treaty had envisaged. Although—as was noted in Chapter 5—it is not enforceable through censure, and perhaps notwithstanding the Council’s intentions, it was the beginning of the acceptance by that institution of a certain political responsibility towards the MEPs. In the course of the years, Parliament obtained more concessions that required the Council to provide information to the MEPs. These rights to information—and those that govern

4 5 6 7

8 9 10 11

Patijn, Het Europees Parlement—de strijd om zijn bevoegdheden (1973) at 77-78. It confirmed these general guidelines in its Letter from Council president Mr De Koster to Mr Scelba, President of the Assembly, Council doc. 10891 of 11 November 1969. Which, in cases where a deadline is set for the decision, may imply a negative decision that needs to be reasoned: case C-76/01P, Eurocoton vs Council, [2003] ECR I-10091, discussed in sub-section 5.6.1, above. The Council legislates in public in co-decision cases: Article 8 of its Rules of Procedure, as interpreted by the Council conclusions on improving openness and transparency in the Council, Council doc. 15834/05 of 15 December 2005. It is Council practice to release many preparatory documents before the legislative process is finished, if need be with deletion of the names of the delegations (but not of their positions). For a statistical analysis suggesting the importance of these bodies for the Council’s decisionmaking process see Van Schendelen, ‘“De Raad beslist”. Beslist de Raad?’ (1995) at 706 et seq. Council doc. 168/60 rév of 25 March 1960 at 13. Save where the common foreign and security policy is concerned: see Articles 21 TEU and III-304 Const. An argument could be attempted that a (rather hypothetical) roll-back would injure the duty of loyal co-operation between the institutions. However—as explored in sub-section 3.2.3 ad I—this latter concept was only recognised or invented by the Courts in the 1980s, when the Council had replied to Parliamentary questions for decades.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

163

the relationship between Parliament and the Commission—are the subject of this Chapter. The topic falls apart in three great themes. First Parliament’s general information rights are analysed (section 7.2). Then I will discuss the representation of each institution in the other (7.3). Subsequently, the MEPs’ right to ask questions to Commission and Council will be analysed (7.4). Lastly, I will discuss the different reporting obligations for Commission and Council (7.5). 7.2

Parliament’s general right to information

7.2.1

Development of Parliament’s general right to information

The Treaties originally foresaw only an obligation for the Commission to produce an annual report and a right for Parliament to question the Commission. On the face of it, the Council was not even under any general obligation to provide information. Certainly, in 1958 the notion of loyal co-operation between the institutions did not yet exist. To some extent, the MEPs’ rights to receive information are subsumed in the institutions’ more general obligations as regards the release of documents to the general public (discussed in the next sub-section). However, this development diminishes neither the value of Parliament’s prerogatives for achieving that general change towards transparency, nor their value as safeguards for the MEPs. The different Treaty obligations in this field for the Commission on the one hand, and for the (European) Council on the other, caused different interinstitutional conventions to develop between Parliament and the Council on the one hand, and between Parliament and the Commission on the other. In 1957 Parliament proposed to the Council to organise an annual ‘colloquium’ during which certain questions pertaining to integration could be discussed. The Council accepted the idea, provided it did not exceed a simple exchange of views and it did not create the impression that the Council was being questioned by the Assembly.12 This colloquium fulfilled an important role in the Communities’ early years but, with the development of Parliament’s other information rights, gradually withered away. After 1966 only the Council president participated on behalf of his institution. This required the ministers to co-ordinate their position before colloquia. Since, however, at the time it became more difficult for the Council to agree in advance to any detailed positions, the Council’s participation lost its meaning and the colloquium its relevance. From 1973 this instrument of interinstitutional convention was no longer organised13 and its place was taken by other information rights. From 1960 the Council presidency provided the Assembly in an annual speech with an overview of the Councils’ activities.14 In 1974 the Council declared itself prepared to make every effort to furnish information requested from it, in particular on the outcome of Council meetings. 15 This commitment is not without limitations: the MEPs have at times sought a commitment from the Council to report regularly on

12 13 14 15

Procès-verbal de la vingtième session du Conseil de la Communauté Economique Européenne tenue à Bruxelles, les 16 et 17 mars 1959, Council doc. 146/59 rév. of 14 April 1959 at 10. Bieber, Organe der erweiterten Europäischen Gemeinschaften: Das Parlament (1974) at 49. Council doc. 168/60 rév of 25 March 1960 at 19. Reply to written question 198/74 from Lord O’Hagan to the Council, OJ (1974) C113/16.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

164

the actions taken on resolutions adopted by Parliament at its own initiative,16 but this has been steadfastly refused by the ministers. Currently, there are three information flows from the Council to Parliament. Firstly, Parliament can ask for information or documents as an institution under Article 197 TEC, under the discharge procedure of Article 276(2) TEC, or under other arrangements. It is standard Council practice that it will only recognise a request to emanate from Parliament if it is sent from, or routed through, the office of the president of Parliament. For example, a request for information sent by a chairman of a Parliamentary committee not endorsed by the president of Parliament is considered to be a request from an individual citizen. Of course, even if the request emanates from Parliament, it still depends on the applicable interinstitutional arrangement whether the document or information will be transmitted. Secondly, individual members of Parliament have a privileged route to information by using the questions procedures (section 7.4, below). Lastly, MEPs—like all citizens and residents—are entitled to request documents under the Access to Documents Regulation and information under the Council’s Code of Good Administrative Behaviour (discussed in the next sub-section). Notwithstanding the great development of Parliament’s relationship with the Council, the MEPs had more success with the Commission. The pressure of the budget discharge procedure led the two institutions to agree to a detailed body of interinstitutional convention in this field. Although as early as the 1960s, the Commission tended to send communications and policy documents to Parliament, this practice was at the time not systematic.17 In 1984 Parliament’s Enlarged Bureau and the Commission concluded an interinstitutional agreement in which the Commission promised inter alia to refer texts of general importance (memoranda, general studies and programmes) directly to Parliament at the same time as they were forwarded to the Council.18 This was followed in 1987 by an interinstitutional agreement in which the Commission promised the MEPs access to the documents on which Commission decisions are based and which Parliament needs for the exercise of its budgetary control function.19 The agreement foresaw that the two institutions would decide ‘sur une base pragmatique’ the equilibrium between the good functioning of the Commission services and the necessity of the Budgetary Control Commission to have complete and actual information. The agreement’s third article provided that ‘[s]i la Commission prend une décision qui s’écarte de l’avis de son contrôle financier, la commission du contrôle budgétaire en est informée, tous les documents ayant été à la base de cette décision lui étant transmis’. In return, the president of the Budgetary Control Committee (and presumably, although that was not explicitly stated, the full Committee) was bound to treat the Commission’s information on a confidential basis.

16 17 18

19

See e.g. the Resolution on relations between the European Parliament and the Council, OJ (1986) C283/36. Ganshof van der Meersch a.o., Les Novelles—Droit des Communautés européennes (1969) at 217-218. Joint Declaration of the European Parliament and the Commission, Parliament doc. 1-328/84/rev of 23 May 1984 at point 6. In 1986 Parliament reminded the Commission of the obligation: Annex to Resolution on relations between the European Parliament and the Commission in the institutional context of the Treaties, OJ (1986) C283/39, § 3. Letter of M. Delors to Lord Plumb, president of Parliament, Parliament doc. PE 114.634/BUR of 7 June 1987. Parliament’s Bureau approved it on 7 July 1987.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

165

The 1995 Code of conduct negotiated between Parliament and Commission20 extended the 1984 agreement by committing the Commission in the context of the legislative and budgetary procedures to ‘ensure that Parliament is kept informed on an absolutely equal footing with the Council, due regard being had to the Treaty’. In addition, outside the legislative sphere the Commission undertook to forward to the Parliament and the Council ‘on an absolutely equal footing all its strategic initiatives, such as Green or White Papers, in areas in which the Community could be required to act’. This broadened existing convention. This exchange of information was further broadened as a result of an interinstitutional agreement implementing Parliament’s discharge powers to obtain more information in the field of CFSP.21 The 1995 Code of Conduct was succeeded by the 2000 Framework Agreement22 which, in turn, was followed by the 2005 Framework Agreement.23 The latter two texts detailed the Commission’s diverse commitments concerning the information of Parliament in much greater detail. I will discuss the 2005 Framework Agreement in more detail in the sections below. Annex I of the 2005 Framework Agreement provides rules on confidentiality (discussed in sub-section 7.2.3). Lastly, the Interinstitutional agreement on better law-making24 provides for a general obligation between the Commission, Council and Parliament to keep each other informed about their work throughout the legislative process. 7.2.2

The Access to Documents Regulation

A separate factor assisted Parliament in obtaining more information. Public pressure caused the institutions to be more transparent in the distribution of documents, culminating in the Access to Documents Regulation.25 This legislation found its legal basis in Article 255 TEC. The direct predecessors of this text are Council and Commission Decisions.26 These were based on a Code of Conduct concluded between Commission and Council,27 which was a pure product of interinstitutional convention, inspired by a declaration adopted by the European Council of Birmingham of 16 October 1992.

20 21 22 23 24 25

26 27

OJ (1995) C89/69. Interinstitutional Agreement between the European Parliament, the Council and the European Commission on provisions regarding financing of the Common Foreign and Security Policy, OJ (1997) C286/80 at L, discussed in section 15.4, below. Framework agreement on relations between Parliament and the Commission, OJ (2001) C121/122. Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125. OJ (2003) C321/1, corr. OJ (2004) C4/7. Regulation (EC) 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, OJ (2001) L145/43. In addition, some other measures have been taken to make the decision making process more transparent. Notably, the Council now holds a number of public debates, on the basis of Article 8 of the Council’s Rules of Procedure, OJ (2006) L285/47. The meetings of COREPER and the other preparatory bodies in the Council are not open to the public. Council Decision 93/731 of 20 December 1993 on public access to Council documents, OJ (1993) L340/43 and Commission Decision 94/90 of 8 February 1994 on public access to Commission documents, OJ (1994) L46/58. Code of Conduct concerning public access to Council and Commission documents, OJ (1993) L340/41.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

166

A detailed analysis of the Access to Documents Regulation falls beyond the scope of this study, 28 which will limit itself to sketching some of the main elements. In the words of the Court of First Instance, ‘the principle of transparency is intended to secure a more significant role for citizens in the decision-making process and to ensure that the administration acts with greater propriety, efficiency and responsibility vis-à-vis the citizens in a democratic system. It helps to strengthen the principle of democracy and respect for fundamental rights’.29 Although applicants need not provide reasons for their requests,30 the institutions must provide reasons for any refusal fully to release a document.31 The Regulation provides a right to the general public to request documents but also lays down all possible exceptions to this right.32 For example, in principle, the Commission and Council must deny the (full) release of documents when a decision on the issue at hand has not yet been taken by the institution and the decision-making process is seriously undermined. The exceptions to granting access must be construed and applied restrictively.33 However, the Regulation concerns the access to documents, not information. Hence it obliges the institutions to provide existing documents, but not to produce them.34 In 2001 Mr Bonde MEP attempted to use the possibilities of the Transparency Regulation in a letter to the Belgian Council presidency.35 Since that time, more applications under the Access to Documents Regulation followed emanating from MEPs, their assistants or from officials in Parliament. The Council separates the privileged right to information that MEPs have under the questions procedures (discussed in section 7.4, below) from their rights as citizens. Thus, if an MEP fails to request information or documents using the questions procedures, the matter is dealt with under the Access to Documents Regulation or, where relevant, the Council’s Code of Good Administrative Behaviour.36 This makes sense: for example, once the 28

29 30 31 32

33 34 35 36

For a detailed discussion Bauer, ‘Transparency in the Council’ (2004) at 367 et seq. For a discussion of the Council’s practice under the Regulation see Driessen, ‘The Council, Member States, and access to documents’ (2005) at 113 et seq. and ibid., ‘The Council of the European Union and access to documents’ (2005) at 675 et seq. Cases T-211/00, Kuijer vs Council II, [2002] ECR II-485, § 52. and T-309/97, Bavarian Lager vs Commission, [1999] ECR II-3217, § 36. Cases T-174/95, Svenska Journalistförbundet vs Council, [1998] ECR II-2289, § 109 and T-111/00, BAT International (Investments) vs Commission, [2001] ECR II-2997, § 42. Cases T-188/98, Kuijer vs Council I, [2000] ECR II-1959, § 36 and T-123/99, JT’s Corporation Ltd vs Commission, [2000] ECR II-3269, § 63. Article 4 of the Regulation. Article 6 of the common position for a Regulation of the European Parliament and of the Council on the application of the provisions of the Aarhus Convention on Access to Information, Public Participation in Decision-making and Access to Justice in Environmental Matters to EC institutions and bodies, OJ (2005) CE264/18, which is currently in codecision, will add to these exceptions. Case T-211/00, Kuijer vs Council II, [2002] ECR II-485, § 55. Decision of the European Ombudsman on complaint 1184/2002/PB against the European Commission. See also Decision of the European Ombudsman on complaint 1015/2002/(PB) IJH against the Council of 17 September 2003. Council doc. 6188/02 of 13 February 2002. The Council’s reply can be found in Council doc. 6190/02 of 25 March 2002. Decision of the Secretary-General of the Council/High Representative for Common Foreign and Security Policy of 25 June 2001 on a code of good administrative behaviour for the General Secretariat of the Council of the European Union and its staff in their professional relations with the

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

167

Council releases a document, this is released erga omnes and published on the institution’s website.37 By maintaining this distinction between the different procedures, MEPs retain the possibility to receive information on a privileged basis, without requiring the documents concerned to be released into the public domain. 7.2.3

Sensitive information

Although, as early as in 1989, Parliament adopted guidelines for the handling of confidential documents communicated to it,38 these were not such that they always enabled the Commission and the Council to entrust the MEPs with such information. This led to interinstitutional friction (and on 10 December 1976 even to an unsuccessful motion of censure).39 This conflict was resolved a large extent through interinstitutional agreements based on Article 9(7) of the Access to Documents Regulation. Parliament and the Commission agreed to conventional rules in Annex III to the 2000 Framework Agreement. These are repeated in Annex I to its 2005 successor. Under these rules, the Commission ensures that Parliament is given access to confidential information whenever it receives from the president of Parliament, the chairperson of the parliamentary committees concerned, the Bureau or the Conference of presidents a request to that effect.40 The Annex excludes information on infringement procedures and competition procedures, which makes it clear that the Annex is not restricted to the ‘sensitive’ documents of Article 9 of the Access to Documents Regulation.41 This still leaves much to be desired: there are many more similar proceedings where the need for confidentiality may clash with the ‘right’ of access granted by the Annex I rules. Of course, where the law requires confidentiality the Commission will be obliged to refuse requests under Annex I. For example, the information that an anti-dumping proceeding will be initiated may not be disclosed.42 As a general rule, the responsible Commissioner and the chairman of the parliamentary body concerned decide on the mode in which confidential information is to be made available. The Framework Agreement foresees a procedure in case disagreements arise. These rules were challenged by a number of MEPs, who felt that their rights to ask questions and receive information were infringed. A series of challenges was rejected by the president of the CFI, who held that the Annex does not restrict the rights of individual MEPs to request confidential information and even less so the right of the Commission to refuse it.43 In addition, the rules do not affect the right of 37 38 39 40 41 42

43

public, OJ (2001) C189/1. Article 10 of Annex II to the Council’s Rules of Procedure. See currently Annex VII of Parliament’s Rules of Procedure. Parliament doc. PE 480/76 of that date. For details see Annex I to the 2005 Framework Agreement. These are the document with classification codes ‘Très Secret/Top Secret’, ‘Secret’ and ‘Confidentiel’. It is hard to imagine that these would be used for any document relating to infringement or competition procedures. Article 5(5) of Council Regulation (EC) 384/96 on protection against dumped imports from countries not members of the European Community, OJ (1996) L56/1. Similarly, Articles 5.5, 6.5 and 12.2.2 of the WTO Agreement on implementation of Article VI of the General Agreement on Tariffs and Trade 1994 (the ‘Anti-Dumping Agreement’), OJ (1994) L336/103 provide obligations that override the Access to Documents Regulation. Of course, there are many more similar international obligations. Case T-236/00R, Stauner a.o. vs Parliament and Commission I (Order), [2001] ECR II-15. The

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

168

individual MEPs to ask questions.44 This concords with the Commission’s view that it did not consider the transmission of confidential information to only certain selected bodies or officers of Parliament a violation of Annex VII of Parliament’s Rules of Procedure.45 In its view, the Agreement simply does not apply to questions asked by individual MEPs.46 Simultaneously, Parliament adopted a resolution47 regretting that, ‘even after the conclusion of the Framework Agreement, the Commission has not forwarded certain information and documents requested by the discharge authority’. It noted that from its part, it had respected the Framework Agreement ‘despite the fact that the rules . . . are open to very broad interpretation, in particular as to whether a document is deemed confidential’ and this ‘may cause undue delays’. The MEPs’ request for a revision of the text has to date remained unanswered and in 2005 did not lead to different rules. Article 9(7) of the Access to Documents Regulation lays down that the Commission and the Council must inform Parliament regarding sensitive documents ‘in accordance with arrangements agreed between the institutions’. This also followed from the requirements of Article 21 TEU and settled an interinstitutional conflict as to the legality of the Council’s Security Regulations.48 By way of implementation, in 2002 the Council concluded an interinstitutional agreement with Parliament on sensitive information in the field of security and defence policy.49 The agreement covers both ‘sensitive’ documents and information. The former are those as defined in Article 9 of the Access to Documents Regulation. The MEPs that are to have access to such sensitive documents must be appropriately security cleared. In view of its special legal basis, the agreement is not of a conventional nature but a legally binding text. Under paragraph 3.1 of the agreement, the president of Parliament or the chairman of its Foreign Affairs Committee may request that the Presidency or the SG/HR inform the Foreign Affairs Committee of developments in European security policy. To this end, the Conference of Presidents designates a special committee chaired by the chairman of the Foreign Affairs Committee and composed of four MEPs. The president of Parliament and the special committee will be informed by the Presidency or the SG/HR of the content of the classified information where it is required for the exercise of the Parliament’s powers. The agreement provides for

44 45 46 47 48

49

MEPs unsuccessfully attacked this Order at two successive occasions: cases T-236/00R, Stauner a.o. vs Parliament and Commission II (Order), [2001] ECR II-2943 and T-236/00R, Stauner a.o. vs Parliament and Commission III (Order), [2002] ECR II-135. Stauner a.o. vs Parliament and Commission III, § 61. Reply to written question E-0600/01 by Ms Rühle and Ms Stauner to the Commission, OJ (2001) CE318/82. Reply given to written question P-0203/01 by Ms Stauner to the Commission, OJ (2001) CE235/155 and reply to written question E-0539/01 by Ms Maes and Ms Stauner to the Commission, OJ (2001) CE261/139. Resolution containing the observations which form an integral part of the decision concerning discharge in respect of the implementation of the general budget of the European Union for the 1999 financial year (Commission), OJ (2002) CE21/199, § 1. The Security Regulations can be found in OJ (2001) L101/1. Parliament challenged them in court: case C-260/01, Parliament vs Council, OJ (2001) C245/13. The challenge was withdrawn as part of the package comprising the Interinstitutional Agreement concerning access to sensitive information in the field of security and defence policy. Interinstitutional Agreement of 20 November 2002 between the European Parliament and the Council concerning access by the European Parliament to sensitive information of the Council in the field of security and defence policy, OJ (2002) C298/1.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

169

different modalities for the president of Parliament and the special committee to be acquainted with sensitive documents. Different possible modes of informing Parliament are foreseen, depending on the classification level and circumstances. 50 The Council must agree with the choice, giving it leverage over the decision how the information will be treated in Parliament. Paragraph 4.2 of the interinstitutional agreement provides that the two institutions are willing to discuss ‘comparable Interinstitutional Agreements covering classified information in other areas of the Council’s activities, on the understanding that the provisions of this interinstitutional Agreement do not constitute a precedent for the Union’s or the Community’s other areas of activity and shall not affect the substance of any other Interinstitutional Agreements’. At the time of writing, no such agreements have been concluded, although negotiations over a text on classified information in the third pillar are continuing. In the meantime, Parliament is granted access to such documents on the basis of ad hoc procedures. Lastly, these agreements do not derogate from other interinstitutional arrangements such as those covering the right of inquiry51 or the 1999 Interinstitutional Budget Agreement.52 7.2.4

A brief evaluation

In summary, in the past forty years the combined effects of interinstitutional convention and the general tendency towards greater access for citizens to documents led to a considerable increase in the MEPs’ ability to exercise their rights of oversight. These days, the Commission is much more exposed than it was in the 1960s. If the change is remarkable in the Commission, it has been Copernican in the Council. Here the agent of change has been mostly the law on access to documents. This developed from the 1993 Code of Conduct onwards, which Code was an interinstitutional convention not even involving Parliament. This increase in information beyond doubt helps Parliament, although sometimes it is difficult to suppress the impression that it does not know very well how to digest all of this information. Generally, though, the increased exposure of the Commission and Council generally helps Parliament to increase its grip onto policy making. In this sense, it has changed the checks and balances between the institutions to the benefit of the MEPs.

50

51 52

See the interinstitutional agreement for the details of the information procedures. Parliament adopted a Decision with implementing provisions: Decision of the European Parliament of 23 October 2002 on the implementation of the Interinstitutional Agreement governing European Parliament access to sensitive Council information in the sphere of security and defence policy, OJ (2002) C298/4. Discussed in sub-section 8.3.2, below. Interinstitutional agreement of 6 May 1999 between the European Parliament, the Council and the Commission on budgetary discipline and improvement of the budgetary procedure, OJ (1999) C172/1.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

170

7.3

Presence of the Commission and Council in Parliament and vice versa

7.3.1

Organisation of this section

Legal rules governing the presence of institutions in each other’s meetings are succinct and interinstitutional convention plays a major role. These apparently innocuous arrangements reflect rather sensitive institutional issues. They also betray, in the case of the Council-Parliament relationship, a shift from the Council’s constitutional status as foreseen by the Treaties’ drafters to a very different one in practice. Sub-section 7.3.2 analyses the representation of the Commission before Parliament, whilst the following sub-section does the same for the Council. Subsection 7.3.4 discusses the representation of Parliament and other institutions in the Council. 7.3.2

The representation of the Commission before Parliament

The representation of the Commission before Parliament is a component of the executive’s parliamentary responsibility to the MEPs. Article 197 TEC clarifies that ‘Members of the Commission may attend all meetings and shall, at their request, be heard on behalf of the Commission.’ This clear statement does, however, raise some questions. Notably, who exactly is included among ‘members of the Commission’? And what is the scope of ‘all meetings’? The meetings of the Parliament can be usefully divided in plenary sessions, committee and delegation meetings, and trialogue and other (organisational) meetings. I discuss each category separately. In plenary sessions. The word ‘meetings’ in Article 197 obviously includes the most important ones (although the MEPs’ attendance sometimes suggests otherwise): the plenary sessions. The law gives the Commission a right to attend: ‘may attend all meetings’; however, convention turned the ‘may’ into ‘shall’. A series of interinstitutional agreements, of which the 1990 and 1995 Codes of Conduct and the 2000 and 2005 Framework Agreements are the highlights, regulate the matter. The last of these agreements formalised a number of commitments made by Mr Prodi’s and Mr Barroso’s Commissions upon entering in office.53 The gist of these agreements is that the Commission ‘shall ensure’ that ‘the Commissioner responsible is present whenever requested by Parliament at plenary sittings for agenda items falling under his or her responsibility’ (§ 37 of the 2005 Framework Agreement).54 The 1995 Code of Conduct stipulated that this also included Question Time. In view of the wide-ranging scope of that item, this was an unfortunate addition, since its literal implementation would imply the monthly presence of almost the complete Commission. A solution was found in an obligation for Parliament to re-organise the conduct of Question Time, which it did.55 As a general principle, Parliament will attempt to organise its agenda in such a way that questions falling within the remit of any single member of the Commission are 53 54 55

See also Parliament’s Resolution on the election of the new Commission, OJ (2005) CE201/113. The text on the presence of Commissioners in plenary was accepted by Mr Barroso during the debates of 18 November 2004. Interinstitutional agreement on better law-making, OJ (2003) C321/1, corr. OJ (2004) C4/7, § 8 essentially repeats the obligation, but without the request from Parliament and with the qualifier ‘as a general rule’. See section 7.4.4, below.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

171

grouped.56 Curiously, the 2000 Framework Agreement no longer comprises this provision. It is, however, still parliamentary practice. The 1995 text also provided for the presence of one of the external relations Commissioners on Friday,57 but this became irrelevant when the MEPs weighed their Friday evening against Strasbourg’s travel connections and abolished the Friday sittings. In committees and similar meetings. Parliament established twenty standing committees58 as well as temporary committees and delegations. Delegations, Joint Parliamentary Committees and similar structures group representatives from Parliament with representatives of other parliamentary bodies. The proliferation of international agreements concluded by the European Union lead to a concomitant increase in the number of such joint bodies. Since 1973,59 the Commission is always represented at any parliamentary committee, but the level of representation differs. Paragraph 40 of the 2005 Framework provides rather optimistically that ‘[a]s a general rule, the Member of the Commission responsible for an item under consideration in a parliamentary committee shall be present at the relevant committee meeting, when invited’.60 In practice this remains rather the exception. Commissioners tend to attend the relevant committee in person at least a number of times every year. Indeed, the last sentence of paragraph 40 is of more practical relevance: ‘[w]here the presence of a Member of the Commission is not explicitly required at a parliamentary committee meeting, the Commission shall ensure that it is represented by a competent official at a senior level.’ This, in fact, is an upgrading from the ‘best efforts’ obligation from 1973. It remains to be seen whether the Commission will be able to meet it in practice. Paragraph 40 also provides that in committee meetings Commissioners have the right to be heard, but in practice this is not denied to Commission civil servants either: contrary to their counterparts from the Council, Commission civil servants do tend to give statements to Parliamentary committees. The Commission will keep the relevant committee regularly informed of the principal positions emerging from discussions within Council bodies, particularly where they depart from the original proposal and committed to forwarding all Commission amendments to the original proposal. This provision has been slightly broadened throughout the years. In 199061 it did not go beyond discussions within the Council, while five years later all Council bodies were included. In view of the importance of these bodies in preparing the Council’s work, this is a hardly surprising but very substantial addition that changed the checks and balances between the institutions. The text from the 1995 Code of conduct found its way into the 2000 Framework Agreement, but it would seem that this was restricted to co-decision. 62 56 57 58 59 60 61 62

1995 Code of Conduct between Commission and Parliament, OJ (1995) C89/69,.§ 8 Ibid. The list can be found in Annex VI to Parliament’s Rules of Procedure. Practical measures to strengthen the powers of control of the Parliament and to improve relations between the Parliament and the Commission, Commission doc. COM (73) 999 final of 30 May 1973 at 7-8. This succeeds the virtually identical rule in § 24 of the 2000 Framework Agreement. The 1990 Code of Conduct concluded between the Commission and the Parliament, OJ (1991) C280/165. This follows from a careful reading of §§ 5 to 7 of Annex I to the 2000 Framework Agreement. The Interinstitutional agreement on better law-making, OJ (2003) C321/1, corr. OJ (2004) C4/7, § 8 essentially repeats the obligation, but without the request from Parliament and with the qualifier ‘as a general rule’.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

172

Somewhat surprisingly, it was dropped in the 2005 Framework Agreement, although this is unlikely to signify a significant change in practice. In 2004, Mr Barroso accepted the MEPs’ demand63 that Commissioners make their presence in committees a ‘priority’. It remains to be seen whether this will have any practical impact on the Commission’s behaviour. In the trialogue, the Conference of Presidents and other contexts. The trialogue, which is discussed in more detail in the context of co-decision and the budgetary procedure,64 brings together the presidents (or their representatives) of Council, Parliament and Commission and is chaired by the latter. The gathering became a success and its use spread from the budgetary field to the legislative. The Commission is equally presented at the meetings of the Conference of Presidents where the draft agenda for the following part sessions is adopted.65 In 2004 the Barroso Commission committed its president and vice-president responsible for interinstitutional relations to remain in regular contact with the Conference of Presidents.66 Lastly, there are many contacts between the two institutions at civil servant level.67 7.3.3

The representation of the Council before Parliament

Although Article 197 TEC creates the impression that representatives of the Council do not appear before Parliament at all, they have done so since the early days of the Communities.68 A description of the interinstitutional conventions concerned exhibits both the sensitivity of the matter and the level of detail in which the checks and balances between the institutions are arranged. As far as the early days of the Communities are concerned, in 1960 the Council president started the practice of presenting in Parliament an overview of the Councils’ activities.69 Apart from this, Council appearances before Parliament became only systematic in 1969, when the ministers promised to be represented in Parliament in principle by their president or another Council member.70 It appears that the first debriefing of a Council meeting by the Council president in the plenary took place in 1974.71 As is attested by the numerous complaints from MEPs about the Council’s lack of presence in their meetings,72 this is a much more contentious issue than the 63 64 65 66 67 68 69 70 71 72

Resolution on the election of the new Commission, OJ (2005) CE201/113, § 5(f). Sub-section 11.3.1 and section 14.5, below. Rule 130(1) of Parliament’s Rules of Procedure, second sentence: ‘The Commission and Council may attend the deliberations of the Conference of Presidents on the draft agenda at the invitation of the President.’ Mr Barroso during the debate of 18 November 2004 in response to the Resolution on the election of the new Commission, OJ (2005) CE201/113, § 5(g). Lastly, the interinstitutional committee foreseen by Article 15(2) of the Access to Documents Regulation must be mentioned. As attested by the Commission’s Deuxième rapport général (1959) at 18. Troisième rapport général (1960) at 48. See Note verbale du Président de l’Assemblée au Président du Conseil, Council doc. R/1358/69 of 17 July 1969. Reply to written question 198/74 by Lord O’Hagan to the Council, OJ (1974) C113/16. See for example the interventions of Messrs Pirker, Colom i Naval and Maaten during the sitting of 12 June 2001 and of Mr Staes during the plenary debate of 3 April 2001. In 1986 Parliament bluntly demanded that the Council ‘be represented at all meetings of all parliamentary bodies [sic!], at the appropriate political level, and should speak in debates to make its position clear at every juncture’:

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

173

Commission’s presence, because the Council’s relationship with Parliament is hardly straight-forward. From the Madison-style separation of functions point of view, Parliament’s criticisms would appear hardly justified to the extent that the Council acts as a (co-)legislator. Perceived from this angle, it should appear strange that Parliament insists on the presence of the Council in its meetings. The British House of Lords is not called upon to explain itself in the Commons, nor is the German Bundesrat responsible to the Bundestag. I am not aware of any bicameral parliamentary system where this is different. Of course, in many instances, the Council acts rather like an executive instead. This is mainly so when it acts under the second and third pillars. In addition, one could think of the cases when it exercises directly certain implementing powers itself under Article 202 TEC (although such specific instances of comitology rarely attract Parliament’s attention). Lastly, an argument could perhaps be made that the Council acts as an executive in its co-ordination role in matters such as economic governance and employment. Still thinking from a separation of functions perspective, it would be more logical for the Council to be represented in Parliament. From a separation of institutions perspective, the MEPs’ insistence on the Council’s presence is more understandable, even when it acts as legislator. In the following I analyse the conventions between the Council and Parliament by first examining the situation in plenary; then the situation in parliamentary committees and last, in other bodies. In plenary sessions A distinction must be made between active and passive representation. Passively, the Council is represented at all times by representatives of its General Secretariat. However, these do not speak in Parliament. Legally, it is up to the country having the presidency to decide who will represent it in Parliament. In practice however, the institution is normally actively represented in plenary by the presidency73 at ministerial or state secretarial level. The Council’s Rules of Procedure allow for the following presidency74 to represent it in Parliament,75 but in practice few presidency politicians willingly concede the spotlight. In 2002, the Council’s Rules of Procedure belatedly allowed the SG/HR to represent it in plenary,76 even though Mr Solana spoke there as early as on 17 November 1999. This caters to the MEPs’ appetite for more Council presence.77 Currently, the Council is present at ministerial level in principle one day during each Strasbourg plenary session and one afternoon of the Brussels session. However, depending on presidency’s preferences, the Council may decide to appear an additional one half day per session in Strasbourg. This, however, is not a matter of

73 74 75 76 77

Resolution on relations between the European Parliament and the Council, OJ (1986) C283/36, § 8. Article 26 of the Council’s Rules of Procedure. Until 2002 it was still possible for the Council to be represented by ‘any other of its members’ instead of the Presidency: Article 26 of the 2000 Rules of Procedure, OJ (2000) L149/21. From 2004, the agreement of the presidency is required for it to be replaced. Before that it was presumably a Council decision, taken with ordinary majority. The 2000 version of Article 26 of the Rules of Procedure foresaw the presence of the SecretaryGeneral only in the committees of Parliament. See Parliament’s Resolution on the reform of the Council, OJ (2002) CE112/317 in which it called for the presence of the Council Secretary-General in its plenary meetings.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

174

convention and even less of law. Any further presence must be specifically agreed between the institutions. The organisation and conduct of the Council’s presence is largely the product of convention. This covers the following issues: the planning of the draft agenda, time, the Presidency’s speech, the contents of the Council’s interventions, the organisation of the debate, emergency debates, and some sundry rules. Co-operation on the planning of the draft agenda of the plenary session goes back a long time. As long ago as 1959, the Council decided to examine the Assembly’s agenda before each Parliamentary session.78 In 1970 the Council agreed to take account of Parliament’s calendar in fixing its own. This was not only prompted by the wish that Council members be present at Parliament, but also by the desire that Commissioners should not be faced with overlapping Council and Parliament meetings.79 The following year this was given more practical value, when the Council agreed that one of its officials should participate in the meetings of the Bureau (from 1993, the Conference of Presidents)80 in which the draft agendas are agreed. The time spent by the ministers in Strasbourg may seem a mundane issue, but it has always been a political problem. In order to simplify planning for the Presidency, the items on Parliament’s agenda requiring its presence are in principle booked for the Wednesday of the session.81 Additionally, the Council may be asked to attend other important debates although, in such cases, sufficient prior notice is necessary.82 The Council has never gone beyond promising the presence of a minister during one day, whilst leaving it to the presidency concerned whether they wished to be present for a further half day per session. Any such decision does not bind future presidencies.83 For a long time, Parliament pressed for more.84 Additionally, the MEPs demanded the Council president to report three times per half-year presidency to Parliament (as the Swedish presidency had done)85 and considered ‘it essential, in the context of the legislative procedure . . . that the Council be present when legislative texts are adopted in Parliament’.86 The Council is reluctant to commit more of the ministers’ time than necessary, especially in view of ‘the sometimes relatively low level of presence of the Members of the European Parliament’.87,88 78 79 80 81 82

83 84

85 86 87

Procès-verbal de la vingtième session du Conseil de la Communauté Economique Européenne tenue à Bruxelles, les 16 et 17 mars 1959, Council doc. 146/59 rév. of 14 April 1959 at 10-11. Reply to written question 96/70 by Mr Vredeling, JO (1970) C97/5. Corbett, Jacobs and Shackleton, The European Parliament (2000) at 102. See also the second sentence of Rule 130(1) of Parliament’s Rules of Procedure. Tuesday when France holds the presidency. Letter from the President of the Council to the president of the Assembly, Council doc. R/259/77 of 3 February 1977. In 1982, the Council agreed that the Council president would be available for emergency debates provided they occurred on the day of its presence in Strasbourg: Council doc. 10079/82 of 13 October 1982. As is clear from e.g. the Letter of Belgian foreign minister Michel to president of Parliament Mme Fontaine of 29 November 2001, Council doc. 15528/01 of 19 December 2001. See e.g. the Letter of president of Parliament Mme Fontaine to Council president Michel, Council doc. 14980/01 of 5 December 2001 and the Resolution on the reform of the Council, OJ (2002) CE112/317, § 18. The latter was an unsuccessful attempt to codify the practice under the Belgian and Swedish presidencies into an obligation. Resolution on reform of the Council and transparency, OJ (2003) CE180/491, § 8. See also Council doc. 14980/01. OJ (2003) CE180/491, § 9. Note from the Working Group on General Affairs to COREPER—Relations between the European

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

175

The gradually increased importance of the Council presidency for Parliament is reflected by the programme speech and debate at the beginning of each half-year presidency and an oral report at the end of it on progress achieved. This practice rests on commitments made or confirmed by the European Council in its 1983 Stuttgart Declaration.89 On such occasions, the presidency typically is represented by the Prime Minister or president. In 1982 the Council agreed to provide Parliament sufficiently in advance with the documents forming the basis for the debate.90 Even though presidencies tend to hype up the event, the value of such programme speeches is limited. In practice, much of the presidency’s programme is dictated by work in progress and this can be instigated by wholly different actors, such as the Commission or previous presidencies, or even Parliament. As to the contents of the Council representative’s interventions in Parliament, the latter has not adopted many substantive rules.91 As noted above, the Council focuses its information on decisions that have been taken and the diffuse nature of the Council leaves that institution with little other choice. The rigour of the main rule is somewhat mitigated in practice. In 1983, the Council acknowledged the right of the presidency to express ‘his own views or those of his Government on matters on which the Council has not yet adopted a position’.92 Parliament’s Rules of Procedure lay down some guidelines for the organisation of the debate during the plenary session affecting the other institutions. Rule 142(5) codifies the custom that Commission and Council representatives shall be heard in the debate on a report as a rule immediately after its presentation by the rapporteur. After this first round, the Rule provides that the Commission and Council may be heard again, in particular to respond to statements of MEPs. In practice it is assumed that these institutions will wish to react to questions or comments directed at them. Another important guideline is laid down in Rule 103, which allows ‘Members of the Commission, Council and European Council’ (in practice, the SG/HR is included as well) to ask for permission to make a statement ‘at any time’. Such statements may be followed by either thirty minutes of questions or a full debate. In practice, often Parliament asks the Council to make a statement.93 Either way, the procedure is attractive for MEPs who want to profile themselves through debates.

88

89 90 91 92 93

Parliament and the Council—Letters from the Council to the European Parliament, Council doc. 8431/01 of 4 May 2001. The matter was raised again during the negotiations leading to the Interinstitutional Agreement on better law-making, OJ (2003) C321/1, corr. OJ (2004) C4/7. Its § 8 provides in its relevant part that ‘[t]he Council will continue the practice of maintaining intensive contact with the European Parliament by means of regular participation in plenary debates, as far as possible by the Ministers concerned . . .’ (emphasis added). Solemn Declaration on European Union, § 2.3.4. Much of this text was inspired by Parliament’s Resolution of 18 December 1981 on the role of the European Parliament in its relations with the European Council, OJ (1982) C11/192. Note from COREPER to the Council, Council doc. 5749/1/82 of 1 April 1982. For the request from Parliament see its Resolution on relations between the European Parliament and the Council, OJ (1981) C234/52. One such provision is Rule 107 of Parliament’s Rules of Procedure, under which the Council is invited to inform Parliament of the position taken by the European Council as regards the broad guidelines of the economic policies of the Member States and the Community. Reply to written question 376/83 from Mr Pannella, OJ (1983) C246/7. This applies equally for committee meetings. Such requests then emanate from its Conference of Presidents: Council Guide—Part I: The Presidency Handbook (2001) at 40.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

176

In 1973 the Council had already agreed to be represented ‘in principle’ during ‘certains débats particulièrement important sur des problèmes d’actualité’.94 In 1982, the Council slightly strengthened this by offering to attend emergency debates ‘whenever possible’.95 This is now practice, provided the debate takes place on the day a Council member attends the session. The commitment covered both EC affairs and European Political Co-operation (now replaced by the CFSP). In line with the Council’s general position, it limited its participation to ‘matters on which [it has] already adopted a joint position’. Secondly, ‘if Parliament wants a debate on such an occasion, the topics to be discussed and the details of the president’s participation will be decided upon in advance by common agreement’.96 Since the 2002 revision of Parliament’s Rules of Procedure, urgency debates are called ‘debates on cases of breaches of human rights, democracy and the rule of law’.97 They tend to follow the same formalised pattern so characteristic of the plenary sessions in Parliament. Some sundry rules deserve to be noted. Firstly, Article 21 TEU requires the presidency98 to consult Parliament on the main aspects and the basic choices of the CFSP and to ensure that its views are ‘duly taken into consideration’. Additionally, the presidency must regularly inform Parliament of the development of CFSP.99 Lastly, the MEPs are directed to hold an annual debate on progress in implementing it. Of course, it falls within the organisational autonomy of Parliament to debate any issue it wants. The value of this last provision therefore rather lies in clarifying that the duty of mutual co-operation extends also to the Council’s participation in this debate. Since 2002, Rule 89(4) of Parliament’s Rules of Procedure clarifies that ‘[t]he Council and/or the High Representative and the Commission shall be invited to each plenary debate that involves either foreign, security or defence policy’. Beyond the annual debate, this implies no obligation for the Council to appear in Parliament and in practice Mr Solana appears in Parliament on average three times a year. Secondly, Article 39(2)-(3) TEU lays down for the ‘third pillar’ somewhat more succinct provisions to almost the same effect (without, naturally, the involvement of the SG/HR). Parliament’s Rules of Procedure foresee the possibility of relegating the debate largely to the relevant committee, which may sit in camera.100 Lastly, the Council resolved in 1970 that the Council president or another Council member should be present during the deliberations of Parliament on the draft budget.101

94 95 96 97 98

Annex to Council doc. R/2641/73 of 5 November 1973 at 8. Letter of the Council to the president of Parliament, Council doc. 10079/82 of 13 October 1982. Ibid. Currently Article 115 and Annex III of the Rules of Procedure. In Article III-304.1 Const the task is devolved to the Minister for Foreign Affairs, who may involve his special representatives in it. 99 This provision succeeds the Decision of 28 February 1986 on the practical application of the EPC adopted by the Ministers of Foreign Affairs at the time of the adoption of the Single European Act. This followed itself older convention: the Letter from Council president Mr Fitzgerald to the president of Parliament of 18 February 1975, reprinted in the Relevé des textes relatifs aux relations entre le Parlement Européen et le Conseil (1984) at 87. 100 Rule 92 of Parliament’s Rules of Procedure. 101 Council Resolution of 22 April 1970 on cooperation between the Council and the European Parliament in matters of budgetary procedure, Treaties establishing the European Communities Vol. I (1987) at 1091-1092. Interinstitutional conventions in the budgetary field are analysed in Chapter 14.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

177

In committees and similar meetings Primary law mentions the possibility of the Council appearing before Parliamentary committees only once, in the context of the procedure concerning the follow-up of economic guidelines by Member States.102 Originally there was some reticence amongst the ministers to appear in Parliamentary committees. As early as 1959 the idea was debated within the Council. It decided to allow such presence only in ‘exceptional cases’.103 However, already in 1960 the Council participated in the work of the Assembly’s Budget Commission, whilst leaving the door open to extending that practice.104 Indeed, a few years later the Council was fully prepared to participate in Parliamentary committees.105 Parliament’s Rules of Procedure now provide as a basic rule that ‘[t]he Commission and Council may take part in committee meetings if invited to do so on behalf of a committee by its chairman’.106 As far as resources go, the Council is passively represented in most Parliamentary committees at a civil servant level. As far as political representation is concerned, the Council can be represented by the presidency or, with his agreement, by the following presidency, the Secretary-General, Deputy Secretary-General or (rarely) senior officials of the General Secretariat, acting on the presidency’s instructions.107 Since 2005, this includes—logically—also the Counter-Terrorism Co-ordinator appointed in the Secretariat-General of the Council.108 Representation by the presidency is the normal situation. Since decades, most ministers from the country fulfilling the presidency appear at least once in the respective Parliamentary committees.109 Parliamentary committee chairmen can invite the Council president once during his six-month stint without going through the president of Parliament or the Bureau. All other invitations for him to attend committee meetings must be routed through the president of Parliament.110 By 2001, the average committee could reasonably expect to be visited by the relevant minister at the beginning of his presidency and once at the end of it. These visits partly fulfil a protocol function. Partly they serve to acquaint the relevant parties with each other. Notwithstanding the considerable logistical burden for small countries, Parliament presses continually for more ministers to spend more time in its committees. The other side of the coin is that some committees with wide-ranging 102 Article 99(4) TEC (III-179.5 Const): ‘[t]he President of the Council and the Commission shall report to the European Parliament on the results of multilateral surveillance. The President of the Council may be invited to appear before the competent committee of the European Parliament if the Council has made its recommendations public’. 103 Procès-verbal de la vingtième session du Conseil de la Communauté Economique Européenne tenue à Bruxelles, les 16 et 17 mars 1959, Council doc. 146/59 rév. of 14 April 1959 at 12-13. Bieber, op. cit. footnote 13 at 46 discusses whether the Council has the right to participate in Parliamentary committees. However, the question if it was obliged to do so is more germane. 104 Council doc. 168/60 rév of 25 March 1960 at 12. 105 Council doc. I/4/64 of 20 March 1964 at 22. 106 Rule 183(2) of Parliament’s Rules of Procedure. 107 Article 26 of the Council’s Rules of Procedure. 108 Note from the General Secretariat of the Council to the Members of the General Affairs Working Party—Meeting of the Delegation for Relations with the United States of the European Parliament with Mr De Vries, EU Counter-Terrorism Coordinator, Council doc. 5735/05 of 26 January 2005. 109 Council docs 9827/1/81 at 5 and 5749/1/82 of 1 April 1982. 110 Decision of the enlarged Bureau of 25 September and 28 October 1980, Parliament doc. PE 69.767 BUR, quoted in Digest of texts on relations between the European Parliament and the Council (1988) at 8. Originally, a limited exception was made for the Foreign Affairs Committee.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

178

competencies can count on a large number of visiting politicians, which has started to impede the committee’s work. Following complaints from Parliament, the Council suggested to the president of Parliament to focus the presence of ministers in committees (‘quality, rather than quantity’).111,112 Remarkably, Parliament—while pushing for more presence in plenary—failed to press the point in its 2001 Resolution on the reform of the Council. It did, on the other hand, request somewhat optimistically that ‘the High Representative for the CFSP . . . at any time report to . . . the committee responsible, at Parliament’s request, if necessary together with the Commissioner for External Relations’.113 In view of the SG/HR’s time schedule, this seems a victory of hope over realism. In the trialogue, the Conference of Presidents and other contexts Contacts between the Council presidency and Parliament have been a regular feature of interinstitutional relations for a long time. I already noted the trialogue and the presence of the Council Secretariat in the Conference of Presidents meeting where Parliament establishes its draft agenda. There are, however, more fora where the institutions meet. There is an on-going dialogue between the two institutions by regular appearances of the Minister for Foreign or European Affairs of the country holding the presidency in the Conference of Presidents, sometimes in lieu of a presence before the plenary when Parliament’s dates of session do not permit otherwise.114 Another procedure that needs to be briefly mentioned here is the ‘procédure de contacts’ approved by the Council on 7 May 1990. This foresees that, following each quarterly meeting of the ministers dealing with immigration issues, the Council presidency meets with the chairmen of the relevant Parliamentary committees.115 This procedure, agreed before immigration was brought under the roof of the European Union, seems to have been subsumed in Parliament’s general rights of oversight. Lastly, in 2002 the Danish presidency tried a new structure, the ‘European Summit’. This combines the European Council president and the president and group leaders of Parliament. Such meetings were organised by the Danish presidency in order to discuss the presidency’s priorities.116 In addition, the Danish presidency of

111 Note from the Working Group on General Affairs to COREPER, Council doc. 8431/01 of 4 May 2001. 112 The Interinstitutional Agreement on better law-making, OJ (2003) C321/1, corr. OJ (2004) C4/7, § 8 does not go beyond a meek obligation for the Council to ‘endeavour’ to participate ‘regularly’ in the work of committees, ‘preferably at ministerial level or at some other appropriate level’. 113 Resolution on the reform of the Council, OJ (2002) CE112/317, § 17. 114 See e.g. Cover note from Mr Borrel Fontelles, President of the European Parliament, to Mr Balkenende, President of the Council of the European Union—Invitation to an extraordinary meeting of the Conference of Presidents on 21 December 2004, Council doc. 15566/04 of 2 December 2004. 115 Contacts avec le Parlement européen, Council doc. CIRC 3624/90 of 17 May 1990. 116 ‘Presidency prepares to implement several Council reforms agreed in Seville24 and 25 October maintained for European Council of BrusselsPat Cox pleased with meeting between Rasmussen, Möller and Haarder’, Agence Europe of 26 June 2002 at item 2. See also the Letter from president of Parliament Mr Cox to Council president Mr Aznar López, Council doc. 7271/02 of 18 March 2002 and the Letter from Mr Aznar to Mr Cox—Relations between the Parliament, the Council and the Commission, Council doc. 9577/02 of 3 June 2002.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

179

2002 started with ‘Summit level meetings’ of the trialogue before each European Council meeting.117 This experiment seems to have been discontinued. 7.3.4

The representation of Parliament and other institutions before the Council

The Commission is the only institution enjoying a widespread right to attend most meetings of the Council and its preparatory bodies, although the Council may always decide to deliberate without it. In a limited number of cases representatives from other institutions participate in the Council’s work. For example, the ECB participates in meetings when it exercises its right of initiative.118 The Court of Auditors may attend the Council’s Budget Committee sessions when the Court’s Annual Report is discussed.119 Other institutions are sometimes invited to attend meetings of the Council or its preparatory bodies dealing with matters of special concern to them.120 MEPs meet with the Council in the context of the conciliation procedure or within the framework of the co-decision procedure. Moreover, MEPs are occasionally present in the margin of Council meetings, notably in the context of the budgetary procedure. The MEPs’ presence during such meetings on budget matters goes back to before 1975.121 Parliament has for a long time sought more access to Council meetings,122 but this has been firmly denied by the latter institution with the logic that its Rules of Procedure foresee only the Commission’s presence.123 Article 5, which regulates the matter, also provides that in principle Council meetings are not public. As has been argued academically, the presence of Parliament and the confidentiality question are linked. In the long run, allowing MEPs into the meetings of its preparatory bodies would force the Council to abandon the confidentiality of these meetings,124 which would completely change the way in which the institution operates. Parliament pushes for a presence, including the right to speak, when the Council takes final legislative decisions.125 In 2002 the MEPs even adapted their Rules of Procedure, adding a new Rule foreseeing the possibility.126 This remains wishful thinking, since the Council is extremely unwilling to go along this road. The only

117 ‘Rasmussen calls for “historical opportunity” to re-unite Europe to be seized—Summits with Parliament and Commission before European Councils in Brussels and Copenhagen’, Agence Europe of 4 July 2002 at item 16. 118 Article 5(2) of the Council’s Rules of Procedure. 119 Note from the Budget Committee to Permanent Representatives Committee/Council—Discharge to be given to the Commission in respect of the implementation of the budget for the financial year 1999, Council doc. 6307/01 of 20 February 2001. 120 Council Guide, Part II—Comments on the Council’s Rules of Procedure (2000) at 11-12. I will leave the special case of the revision of the Staff Regulations aside. 121 Isaac, ‘La rénovation des institutions financières des communautés européennes depuis 1970’ (1975) at 724. 122 Such suggestions were already made in the early 1970s: Fuß, ‘Positionsstärkung des Europäischen Parlaments—Zum Vedel-Bericht und zu anderen Vorschlägen für die institutionelle Reform der Gemeinschaften’ (1972) at 370. See also the footnote to the Joint Declaration on the Legislative Programme for 1994, OJ (1994) C60/1. 123 Reply to written question E-1254/02 by Ms Kinnock to the Council, OJ (2002) CE309/102. 124 Bieber, op. cit. footnote 13 at 53 expresses strong doubts that MEPs could be restrained by obligations of secrecy in the Council. 125 Resolution on the reform of the Council, OJ (2002) CE112/317, § 20 and Resolution on reform of the Council and transparency, OJ (2003) CE180/491, § 9. 126 Rule 62a of Parliament’s Rules of Procedure, OJ (2002) L61/1. This became the current Rule 38.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

180

measurable impact these efforts have had is that since the late 1980s the president of Parliament is invited to be present during part of European Council meetings.127 7.3.5

A brief evaluation

In the past fifty years, Parliament successfully used Article 197 TEC to ensure the involvement of Commissioners in its work. This caused the relationship between the two institutions to become much more political than foreseen by the letter of the original EEC Treaty. The Commission got used to appearing in Parliamentary committees and this considerably facilitates the dialogue between the institutions. This, in turn, greatly improved Parliament’s ability to hold the Commission politically responsible. The policy fields covered by the TEU apart, the Council’s presence in Parliamentary meetings is a product of interinstitutional convention: whilst possible according to the letter of the Treaties, in practice the Council could not easily decide to cease such presence. Its conditions are very much negotiated between the two institutions. The increased interinstitutional contacts that have resulted from these arrangement improved Parliament’s tools of oversight over the Commission, whilst establishing a measure of oversight over the Council where the Treaties foresaw virtually none. 7.4

Parliamentary questions

7.4.1

Different types of questions

Article 197 TEC obliges the Commission to reply ‘orally or in writing’ whilst the Council128 ‘shall be heard’ in accordance with its Rules of Procedure. This provision suggests on the one hand that, as far as the letter of the law is concerned, the conditions for the Commission’s answers are to be found in Parliament’s Rules of Procedure. On the other hand,—apart from the exceptions of questions concerning CFSP and police and judicial co-operation in criminal matters, policy domains in which Parliament is entitled in law to pose questions to the Council—the Council may merely be ‘heard’ when the Council wants. It follows that, second and third pillars excepted, the MEPs’ right to answers from the Council does not rest on Article 197. Rather, it is a product of interinstitutional convention,129 starting with the Council’s 1959 commitment of principle to reply to questions. Article 197 and the other relevant primary law provisions speak of ‘Parliament’ rather than ‘MEPs’. Parliament has never really known what to do with its right as an institution to ask questions. In 1962 it deleted from its Rules of Procedure the provision summarily regulating the matter.130 In practice, only a right for individuals or groups of MEPs to ask questions remains. This fall apart in written questions (in interinstitutional jargon referred to as ‘E-questions’, or ‘P-questions’ for priority ones, discussed in sub-section 7.4.2), oral questions (‘O-questions’, sub-section 7.4.3), 127 I will not further discuss the involvement of Parliament representatives in the preparatory work on the Staff Regulations. 128 And in Article III-337 Const the European Council. 129 See in this sense Council president Solbes Mira during the Parliamentary debate of 11 April 1989, OJ (1989) 2-377/62. In 1973 the Council ‘confirmed’ its willingness to reply to written questions and accepted to reply to oral questions: Annex to Council doc. R/2641/73 of 5 November 1973. 130 Patijn, op. cit. footnote 4 at 48.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

181

question time (‘H-questions’, 7.4.4) and some other modes of Parliamentary questions (7.4.5). Before proceeding to discuss the different types of parliamentary questions, some general principles can be deduced. Firstly, oral questions are, other than the name suggests, submitted in writing. They are officially and more properly called questions for oral answer—but following common usage I will stick to the shorter ‘oral questions’. Secondly, neither the Commission131 nor the Council132 will reply in substance on questions falling beyond their remit. This is a logical corollary of Article 7(1) TEC and makes sense from a separation of institutions point of view. Thus, the Council does not reply to comments made by Member State politicians133 or comment on the internal affairs of other institutions.134 Additionally, the Commission refuses to answer questions on matters sub judice. 135 It understandably refuses to answer questions that concern decisions of the intergovernmental conference (which is not a Council event),136 opinions or positions of one of its members,137 opinions of (members of) other institutions,138 or statements made to the press.139 Thirdly, although replies to oral questions and question time are prepared in writing, nothing can stop a minister to deviate from a prepared text and to reply solely on behalf of the Presidency.140 In that case, he should make it evident that he does not represent the Council’s point of view. 7.4.2

Written questions

Written questions are the most common vehicle of parliamentary responsibility. In comparison with (at least some) Member State parliaments, Parliament uses the written question procedure enthusiastically.141 The probable cause of this is the somewhat complicated arrangement governing Parliament’s seat.142 Most parliaments 131 Reply to written question E-546/02 by Ms Schröder to the Council, OJ (2002) CE229/99. 132 ‘Les Conseils . . . ne pourront répondre, en principe, qu’à des questions concernant des matières dans lesquelles ils ont pris des décisions en tant qu’Institutions des Communautés’: Council doc. 146/59 rév. of 14 April 1959 at 14. See also the reply to written question E-1741/00 by Mr Duff to the Council, OJ (2001) CE72/107. 133 Reply to written question P-2920/01 by Mr Tajani to the Commission, OJ (2002) CE40/249. 134 Reply to written question E-531/02 by Mr Tannock to the Council, OJ (2002) CE277/59. 135 Written question E-2778/01 by Messrs Rod, Lipietz and Jonckheer to the Commission, OJ (2002) CE147/52. 136 Reply to written question E-3659/00 by Mr Duff to the Council, OJ (2001) CE163/187. 137 Replies to written question 55 by Mr Vredeling to the Council, JO (1967) 178/1, to written question 1573/84 by Mrs Cassanmagnago Cerretti to the Council, OJ (1985) C83/32 and joint reply to written questions P-748/02 by Mr Rutelli, P-756/02 by Mr Di Pietro, P-757/02 by Mr Di Lello Finuoli and P-779/02 by Mr Cossutta to the Council, OJ (2002) CE229/131. 138 Reply to written question E-262/02 by Lord Inglewood to the Council, OJ (2002) CE205/102. 139 Reply to written question E-3060/02 by Mr Vanhecke to the Council, OJ (2003) CE155/83. 140 Westlake, The Council of the European Union (1999) at 335. 141 By way of (somewhat anecdotal) evidence: in 1981 the total number of written questions amounted to 1991 (1744 to the Commission, 210 to the Council, and 37 to the Ministers in the context of the EPC). The total amount of oral questions was 103 (67, 29 and 7 respectively) and 850 questions were posed for question time (510, 238 and 102, respectively): Forging ahead. Thirty years of the European Parliament, 1952-1982, unnumbered Parliament doc. (1983) at 199. During the first half of 2000 the number of written questions posed to the Council alone had risen to 202, with 38 oral questions and 190 questions for question time, all to the same institution (of these 190, 63 concerned the CSFP). One may expect that the Commission must have felt a similar increase. The increase is, of course, mostly due to Parliament’s enlargement between 1981 and 1991. 142 Mathijsen, A Guide to European Community Law (1985) at 18.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

182

in Europe tend to hold their committee and other meetings in the morning and their plenary sessions in the afternoon and evening. Such a time schedule is beyond the possibilities of the MEPs as long as most of their plenary sessions take place in Strasbourg. This considerably limits the total number of days on which MEPs can hope to encounter the other institutions in plenary, making the possibility to ask questions between plenary sessions more important. Written questions may serve multiple purposes. They are a vehicle of parliamentary oversight of the executive, force the Commission to take a position, stimulate new Commission initiatives, or are intended to obtain information.143 All of these are different aspects of exercising parliamentary responsibility on the preparation, adoption and execution of policy. In addition, many questions give the impression to be asked to profile the MEP concerned or simply to make a disguised statement for domestic consumption. Written questions do not produce legal effects144 and constitute at best a very limited—if any—factor for establishing legitimate expectations.145 Procedural aspects Article 197 TEC uses the peremptory ‘shall reply’ as far as the Commission is concerned.146 The Council’s much more detached position is mirrored in the applicable arrangements. Written questions can be asked by individual MEPs. They must be submitted in writing to the president of Parliament who forwards them to the Commission or the Council. Parliament unilaterally imposes a deadline of six weeks, and three weeks for questions ‘which require an immediate answer but not detailed research (priority questions)’.147 The deadlines are counted from the date the question is forwarded to the institution concerned. The Commission considers itself bound by these unilateral deadlines.148 Each MEP is entitled to one priority question monthly. The president of Parliament decides whether a question fulfils the criteria of a priority question. If a question is not answered within the time limit set by Parliament its author may request it to be placed on the agenda of the next committee following the ‘question time’ procedure149 (although this last remedy seems to be used rarely, if ever). Until October 1993, the written question had to ‘be brief and relate to specific points falling within the sphere of activities of the Communities’.150 From the eighth 143 Patijn, op. cit. footnote 4 at 49. 144 Case T-229/94, Deutsche Bahn AG vs Commission, [1997] ECR II-1689, § 114. Bieber in Commentaire J. Mégret 9 (2000) at 113 argues that the legal value of a reply depends on the content of the individual reply. However, he must acknowledge that there is little case law in this respect. The clear statement in Deutsche Bahn would rather point towards the other direction. 145 Ibid. and joined cases 424/85 and 425/85, Coöperatieve Melkproducentenbedrijven NoordNederland BA (‘Frico’) a.o. vs Voedselvoorzienings In- en Verkoopbureau, [1987] ECR 2755, §§ 32-33. 146 In the early 1960s the obligation was not always accepted by each of the Commissions: Feuer, ‘L’activité du Parlement européen du 1er janvier au 30 juin 1965’ (1965) at 625. 147 Rule 110(4) of the Rules of Procedure. In the early days of the EEC, Parliament would publish the question in the Official Journal if the institution failed to reply within two months: Ganshof van der Meersch a.o., op. cit. footnote 17 at 281. 148 Replies to written questions P-2640/02 by Mr Hannan to the Commission, OJ (2003) CE137/101 and E-2808/01 by Mr Meijer to the Commission, OJ (2002) CE115/164. 149 For all of these rules see Rule 110 of Parliament’s Rules of Procedure. 150 Article 62(1) of Parliament’s Rules of Procedure, seventh edition.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

183

edition of Parliament’s Rules of Procedure onwards this laudable rule was deleted. Currently, Parliament does not limit the subject of the question151 but—as noted above—this does not force the other institutions to provide a meaningful reply.152 Nevertheless, by decision of the Enlarged Bureau, a written question may not exceed 25 lines,153 so as to avoid that the question becomes a statement in disguise. In 1959 Commission and Council agreed to consult each other where written questions to one institution could interest the other.154 Notwithstanding such consultation, the responsibility for the reply rested with the institution to which the question was asked.155 The MEPs disliked this156 and in 1970 the Commission formally denied that it submitted as a matter of course its draft replies to COREPER.157 Whatever the extent of this co-operation, it seems to have softly expired, even though the Commission is represented in the Council’s General Affairs Working Party (GAG) which prepares the draft replies to questions addressed to the Council. Written questions posed to the Council In practice, the procedure for posing questions to the Council is, on Parliament’s side, almost identical to that used for questions posed to the Commission. The main difference concerns the deadline for replying, which is set unilaterally by Parliament. Although the Council agreed to a speedy reply,158 it refuses to be bound by Parliament’s Rules of Procedure.159 This notwithstanding, internal guidelines in the Council call for a fast-paced procedure, which at least aims to have the question answered within two months.160 This is no mean feat considering the number of Member States and officials that deal with the draft reply. The written question is sent in all official languages by Parliament’s services to the Council’s General Secretariat. In the Council’s view, failure to provide all language versions implies that the question has not been received at all. The question is then forwarded to all Member State delegations and to the relevant desk officer, who will prepare a draft reply.161 This is fine-tuned by the Legal Service, after which it is distributed to the GAG, who should in principle receive it within ten working 151 Since its 2002 revision, the Rules of Procedure makes this explicit (currently Rule 110(1)). 152 In at least one case the Commission declined to reply because the question was posed too bluntly: reply to written question E-2537/97 by Ms Breyer to the Commission, OJ (1998) C76/122 and 124. 153 Decision of the Enlarged Bureau, Parliament doc. PE 104.739/BUR of 25 and 26 February 1986. 154 Council Minutes of 15-17 March 1959, Council doc. 146/59 rév. at 15. Apparently, such cooperation was in practice limited to the consideration of the facts: Ganshof van der Meersch a.o., op. cit. footnote 17 at 219. 155 See the Council’s reply to written question 130/68 by Mr Bading to the Council, OJ (1968) C83/21, and to written question 186/70 by Mr Vredeling to the Commission, OJ (1970) C135/3. 156 Patijn, op. cit. footnote 4 at 53-54. 157 Written question 109/70 by Mr Vredeling to the Commission, OJ (1970) C102/7. 158 In 1972/73 the Council still needed on average three and a half month to reply to a question: Bieber, op. cit. footnote 13 at 50. In 1973 the Council adopted internal measures enabling it to reply to written questions within a period of less than two months: Annex to Council doc. R/2641/73 of 5 November 1973. 159 Joint answer to written questions E-1299/00, E-1300/00 and P-1541/00 by Messrs Tannock and Ebner to the Council, OJ (2001) CE46/167 and reply to written question E-2272/02 by Mr Esteve to the Council, OJ (2003) CE92/147. 160 Report from the Working Party on General Affairs on 22 November 1979 to COREPER—Procedure for replying to parliamentary questions, Council doc. 10427/1/79 of 26 November 1979. 161 For this and the remainder of the procedure within the Council, ibid.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

184

days from receipt of the question. The Member States (and, at least in theory, the Commission) have a few weeks to submit comments. On a small part of the draft replies comments are indeed received, which are then discussed in this forum. Subsequently, the (eventually amended) reply is approved by COREPER as an ‘I/A’ point (meaning that no discussion takes place on it in COREPER or Council).162 It is then forwarded in all languages to Parliament. Rule 110(3) of Parliament’s Rules of Procedure (transformation of a written question into question time before committee)163 is not widely used and it seems somewhat unrealistic to expect Council politicians to participate in such exercises. Moreover, it is possible that the deadline set by paragraph 4 (priority questions) would cause the matter to be brought before the relevant committee before the Council would have had a chance to approve an answer. 7.4.3

Oral questions Procedure

Oral questions with debate164 constitute a considerably more heavy-handed instrument than the written question, since oral questions require scarce plenary time as well as the presence of a Commission or Council politician. This form of questioning was introduced in 1962. Although the Council originally feared that the spirit and letter of the Treaties objected to the possibility of a resolution concluding the debate,165 currently such resolutions are optional. In any event, the Council accepted to reply to all oral questions and to participate in debates following such questions.166 In 1974, the European Council agreed that questions put to the Ministers of Foreign Affairs in the EPC context would also be answered;167 consequently, oral questions in this field became possible as well. A big difference with written questions and question time questions is that Parliament restricted the MEPs’ right to ask oral questions. Rule 108(1) of Parliament’s Rules of Procedure requires that the question be tabled by committee, a political group or at least thirty-seven members.168 The Conference of Presidents decides whether, and in what order, questions should be placed on the agenda. This filter is necessary since the reply to such questions requires the presence of either 162 Under the Council’s Rules of Procedure the replies to written questions may be adopted with a simple majority but, as follows from the procedure described, this hardly happens in practice. 163 ‘If a question cannot be answered within the time limit set it shall, at the request of the author, be placed on the agenda of the next meeting of the committee responsible. Rule 109 shall apply mutatis mutandis.’ Before 1993 Parliament simply published unanswered questions. 164 The clarification ‘with debate’ was added to Parliament’s Rules of Procedure in 2002. See now Rule 108 for more details on the procedure within Parliament. Until 1993, Parliament’s Rules of Procedure provided for both oral questions with debate and oral questions without debate. The latter never became popular and disappeared from the Rules. 165 Patijn, op. cit. footnote 4 at 80. 166 Relevé des décisions prises par le Conseil lors de sa 257ème session tenue le 15 octobre 1973, Council doc. R/2641/73 of 5 November 1973. The decision was confirmed by the 1983 European Council of Stuttgart: Solemn Declaration on European Union, § 2.3.3. 167 Communiqué of the Conference of Heads of Government in Paris on 9 and 10 December 1974, Bull. EC 1974-12 at 7. This commitment was formalised by the Decision of 28 February 1986 on the practical application of the EPC adopted by the Ministers of Foreign Affairs at the time of the adoption of the Single European Act. 168 In addition, Parliament’s Enlarged Bureau decided that a question may not exceed twenty-five lines: Parliament doc. PE 104.739/BUR of 25 and 26 February 1986.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

185

Commission or Council politicians and time is limited. Moreover, the debate needs to be scheduled into the agenda. Under Rule 108(2), questions to the Commission must be referred to the addressee at least one week, and questions to the Council at least three weeks before the sitting on whose agenda it is to appear. As far as the Council is concerned, this Rule implies that its General Secretariat and the GAG are forced to work within very brief deadlines. As a result, the procedure within the Council differs somewhat from that used for written questions. In practice, if time permits, draft replies are circulated at the Council, leading the Council representative to speak on behalf of that institution without necessarily possessing an approved brief. Parliament’s Rules of Procedure foresee that ‘[o]ne member of the institution concerned shall answer’ the oral question.169 This is normally the presidency at ministerial level. Although the Council’s Rules of Procedure technically allow it, in practice the Secretary-General does not reply to oral questions. An additional problem arises because the Council can only meaningfully reply to the extent it has an opinion. Thus, although the initial reply to the oral question is prepared through the GAG procedure, during the follow-up debate the presidency will be forced to ad lib. As a result, it is during the initial reply not customary for the presidency to put forward its own view. When he has to respond to the MEP comments that follow this response, he has more opportunity to give his own (country’s) view. For the Commission it is typically the relevant Commissioner who replies to the oral question. The different rules concerning the presence in Parliament differentiate the Commission’s situation somewhat from that of the Council. In the 1995 Code of Conduct, the Commission and Parliament agreed that Commissioners would be present ‘as far as possible (in particular, the appropriate Commission members when oral questions are discussed)’. In principle they agree on a timetable for such attendance.170 Paradoxically, Parliament’s increased powers in terms of questions thus limited its autonomy to set its agenda. Parliament’s Rules of Procedure no longer foresee the automatic transformation into written questions of oral questions not included in the agenda. Such questions simply lapse.171 In the CFSP and PJCC spheres Articles 21 and 39(3) TEU give Parliament the right to ‘ask questions of the Council or make recommendations to it’. If these provisions are to have any effet utile, the right to ask questions implies the duty of the Council to respond within a reasonable time. However, the special rule of Articles 21 and 39(3) TEU does not imply that the Council is bound by Parliament’s Rules of Procedure. Article 197 TEC applies to both pillars.172 It follows that the restriction in Article 197 applies in full and that the Council is heard in accordance with the conditions laid down in its own Rules of Procedure. Parliament interprets these provisions as exceptions to, or at least as situations separate from, the main rule of Article 197 TEC. Rule 108(3) provides that in cases of oral questions concerning CFSP or PJCC matters, the three-week deadline for 169 170 171 172

Rule 108(4) of Parliament’s Rules of Procedure. Code of Conduct, OJ (1995) C89/69, § 8. In 2002 this was laid down explicitly in (the current) Rule 108(1). This is evident from Articles 28 and 41 TEU, respectively.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

186

submitting questions to the Council does not apply. Rather than providing for another deadline for Parliament to submit the question, Rule 108(3) lays down that the Council ‘must reply with sufficient promptness to keep Parliament properly informed’. Whilst it is clear that—contrary to first pillar matters—there is a legal obligation for the Council to reply, it is submitted that the reference to Article 197 leaves the modalities to be agreed by the Council. 7.4.4

Question time General issues

Question time arrived with the first British MEPs at Parliament in 1973, although a proposal in that direction had been studied by Parliament since 1971.173 At least as far as the Council is concerned, it is the result of pure interinstitutional convention, being instituted through an exchange of letters between the two institutions.174 Like oral or written questions, such ‘H-questions’ may be asked to the Commission or the Council. Like the other types of questions, they can cover the second175 and third pillars. In the United Kingdom the institution exists since 1832. Question time in the Commons forces the government to debate current affairs without much preparation. The notified questions ‘often serve as pegs on which to hand a more insidious “supplementary”’.176 In Strasbourg too, question time is an opportunity for MEPs to obtain more information or to profile themselves. In Westminster, the practice is to have 45 to 50 minutes of question time on every sitting day bar Fridays. Obviously, the more limited number of session days in Strasbourg—itself a consequence of the imbroglio surrounding Parliament’s seat—necessitates a somewhat longer period in the European Parliament. This period is now set at one and a half hour (but only on one day) for the Commission (typically on Strasbourg Tuesdays) and a similar period for the Council (typically on the Wednesday). Whereas in the UK individual ministers are questioned, the MEPs question institutions. The obvious consequence is that the replying Commissioner or minister will need considerable more (written) preparation of the answer (which more often than not will lie out of his professional purview), whereas a UK minister may be expected to reply more impromptu to questions concerning his own portfolio. This turns question time into a rather more formal affair in Strasbourg than it is in Westminster. Question Time’s formatreply by the Commissioner or minister followed by follow-up questions and replies, its low thresholdindividual MEPs can table questionsand its emphasis on topical issues necessitated some procedural rules 173 Jacqué, ‘Le Parlement européen (1er janvier 1973-31 mai 1974)’ (1974) at 730, who explains the original procedure in detail. See also Parliament’s Resolution on the addition of a Rule 47a to the Rules of Procedure introducing a question time in the European Parliament followed by a debate if requested, OJ (1973) C4/17. 174 The Council’s letter is published in OJ (1973) C19/4. 175 The Ministers of Foreign Affairs agreed in 1976 to include the EPC in the scope of Question Time: Letter of 10 May 1976 from M. Thorn to the president of Parliament, quoted in Relevé des textes relatifs aux relations entre le Parlement Européen et le Conseil (1984) at 92. This was followed by the Decision of 28 February 1986 on the practical application of the EPC adopted by the Ministers of Foreign Affairs at the time of the adoption of the Single European Act. This was the predecessor of Article 21 TEU. 176 Hood Phillips and Jackson, Constitutional and Administrative Law (1987) at 229.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

187

especially to protect the Council. This explains why, other than with the written and oral questions, Parliament devotes the whole Annex II to its Rules of Procedure to the matter. Procedure Rule 109(1) suggests more autonomy for Parliament in organising question time than it really enjoys: ‘Question Time to the Council and Commission shall be held at each partsession at such times as may be decided by Parliament on a proposal from the Conference of Presidents . . .’ Although there are more limitations on substance than for oral questions, the procedural thresholds are low. MEPs need the agreement of neither their group nor colleagues. In order to avoid overly enthusiastic use of the instrument, the Rules of Procedure limit the right to ask such questions for each MEP to one question to the Council and the Commission at a given part-session.177 The time limits for question time questions are short. The Council insists on having the questions in all languages at the latest on the Friday of the week before the week preceding the session, 178 that is, ten days (seven working days) before the question time session. 179 For the MEPs this effectively implies the deadline is the Thursday preceding that.180 Even then, it requires great efforts to prepare the replies within the deadline and occasional clashes occur with Parliament over the Council’s refusal to accept late questions.181 Unlike other types of questions, the rules on substantive admissibility and restrictions are quite developed. The Council requested Parliament in 1982 to ensure that especially question time questions be ‘brief and specific’.182 This particular concern with question time flows from the Council representative’s exposed position under this procedure. Apart from the less thoroughly prepared reply, the possibility of supplementary questions provides an opportunity for surprise difficult to combine with the often compromise responses that the representative of the Council has to present. This difficulty led Parliament to adopt detailed rules on admissibility and substance. Questions must be submitted to the president of Parliament, who rules on their admissibility.183 Annex II.A of Parliament’s Rules of Procedure contains guidelines in 177 Rule 109(2) of Parliament’s Rules of Procedure. 178 The original five working days before the sitting concerned were in 1979 in mutual agreement extended to seven: Parliament doc. PE 62.676, quoted in Digest of texts on relations between the European Parliament and the Council published by the Secretariat General of the Council (1988) at 139. Since question time for the Council normally takes place on the Wednesday of the Strasbourg session, this puts the deadline at the Monday of the week preceding the part-session: Note from COREPER to the Council, Council doc. 5749/1/82 of 1 April 1982. 179 The request was made, and accepted by Parliament’s Enlarged Bureau, in 1979: Problèmes d’organisation des travaux du Parlement (Enlarged Bureau minutes), Parliament doc. PE 265/79 of 22-23 November 1979. 180 Minutes of the meeting of the Bureau of Parliament, Parliament doc. PE 163.014/BUR of 19 November 1992. 181 See for example, the Letter of Mr Cox to Mr Pique i Camps, President of the Council, Council doc. 8048/02 of 16 April 2002. 182 Council doc. 5749/1/82 of 1 April 1982. 183 Rule 109(3) of Parliament’s Rules of Procedure.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

188

order to avoid that question time turns into a cross-examination of the other institutions, or is abused by MEPs to make speeches.184 Of course, this does not stop the Council from applying its own admissibility test, especially where deadlines are concerned. Another check protecting the other institutions is the rule that an MEP may follow up the reply with at most two supplementary questions.185 Until 2002, the Rules allowed only one supplementary question to the Council, a distinction caused by the different constitutional roles of Commission and Council. Of course, these rules can be played around with by having different MEPs ask the questions. In order to avoid this to happen, the Rules provide Parliament’s president with some powers to stop such circumvention.186 The Council has accepted the principle of follow-up questions, but has never been enthusiastic about its presidency being led into an unprepared debate. A last restriction requires that the questioner or his substitute be present before question time begins.187 In 1973, Parliament still favoured a system where questions where this condition was not met were answered in writing. The Council refuses to go along that road. Interestingly, Parliament has itself recognised a problem here. In 2002 it added a new rule laying down that ‘[i]f a Member tables a question, but neither he nor his substitute are present at Question Time, the President shall write to the Member reminding him of his responsibility to be present or substituted. If the President has to send such a letter three times in the space of any twelve-month period, the Member concerned shall lose his right to table questions at Question Time for a sixmonth period.’188 Rapporteur Corbett’s laconic justification was that ‘[u]nfortunately it happens frequently that authors of questions are not present’.189 This rule makes sense. On the one hand, the MEPs’ right to ask questions is an important democratic asset but, on the other hand, answering them requires considerable work and resources. For example, in the Council a reply passes the hands of dozens of administrators, translators, Legal Service and Member State officials before it is given. Even a slight increase in the number of questions involves an additional heavy administrative burden. Under such circumstances, it is important that especially MEPs take the instrument seriously, if they wish other institutions to do the same. Occasionally MEPs try to play around the limitation by requesting the Council document containing the text of the draft reply under the Access to Documents Regulation. The Council refuses such access under Article 4(3) of that Regulation.190 184 Annex II.A of the Rules of Procedure, which ultimately is the outcome of the realisation of Parliament in 1979 that MEPs had a habit of crowding each question: Échange de vues au sujet des travaux de la présente période de session, Parliament doc. PE 60.637/BUR/extr of 23 October 1979. See also the comments made by acting President of Parliament Mr Puerta, during the debate of 14 February 2001. 185 Rule 4 of Annex II.A of the Parliament’s Rules of Procedure. 186 Ibid, Rule 6 of Annex II.A and Annex II.B for the details. 187 Articles 9-10 of Annex II.A. 188 Article 11 of Annex II.A to the Rules of Procedure. 189 Draft report on the General Revision of the Rules of Procedure, Parliament doc. PE 294.770 of 8 June 2001 at 42-43, and Second draft report on the General Revision of the Rules of Procedure, Parliament doc. PE 304.283 of 29 August 2001 at 160. 190 ‘I/A’ item note from the Working party on Information to COREPER (2nd part)/Council—Public

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

189

Indeed, an argument can well be made that, as a matter of loyal co-operation between the institutions, (Members of) Parliament should not benefit from their own lack of attendance.191 The status of replies is a matter of interest. One major consequence of the extreme deadlines is that there is no time for the GAG—let alone COREPER or the Council—to discuss the draft reply. Therefore, the responsibility for the content of the reply lies more with the Council presidency than with the Council. As a result, the impromptu observations of the Council or Commission presidents do not carry as much weight as reflections of Council policy as the carefully calibrated replies to written questions do. The led to a problem as regards the status of unanswered questions. The limited amount of time available for question time causes, on average, roughly half of the questions to remain unanswered. Parliament’s Rules of Procedure provide that these questions are converted into written priority questions unless their authors request them to be dealt with in committee.192 In the past, a Council civil servant passed on those prepared written replies to Parliament’s services immediately following question time. This situation resulted in question time becoming something of a ‘shortcut for a more rapid written answer’.193 In 2005 the Council and Parliament agreed that such draft replies should be accompanied with a statement to the effect that the sole responsibility for the draft reply lies with the Council presidency.194 One year later the GAG agreed further guidelines. These lay down that questions not answered during question time are answered in writing under the responsibility of the presidency (provided the MEP was present during the session). Member States have a chance to provide comments to the draft reply. The final reply makes clear that it does not commit the Council.195 In 2006, the Austrian presidency and Parliament agreed to focus question time by having a different minister attend at each session. He would reply only to question pertaining to his field of competence.196 At the time of writing, it is not yet known how this system functions in practice. Lastly, I must mention the possibility for a parliamentary committee to hold question time.197 Each committee decides its own rules for the conduct of it.198 It is certainly not practice in each committee. 7.4.5

Other modes of Parliamentary questions

Besides the question types discussed above, there are a few more modes in which MEPs can question Commissioners or the presidency. Obviously, in the context of

191 192 193 194 195

196 197 198

access to documents—Confirmatory application 05/c/01/05, Council doc. 5316/05 of 28 January 2005, § 8. In analogy with case C-65/93, Parliament vs Council (GSP), [1995] ECR I-643, espec. § 27. Articles 12 and 13 of Annex II.A. Draft report on the General Revision of the Rules of Procedure, Parliament doc. PE 294.770 of 8 June 2001 at 43. Note de transmission de: M. Borrell Fontelles, Président du Parlement européen à M. Alexander, Président du Conseil—Heure des questions au Conseil, Council doc. 15261/05 of 1 december 2005. Note from the General Secretariat to the Working Party on General Affairs—Guidelines applicable during the transitional period for the procedural handling of written replies to ‘H’ questions not answered orally during ‘Question Time’—Agreed text by the Working Party on General Affairs, Council doc. 7561/06 of 15 March 2006. Notice to Members of the European Parliament—New system for Council Question Time from Mr Harald Rømer (no number) of 6 February 2006. Corbett, The European Parliament’s role in closer EU integration (1998) at 129. Rule 187 of Parliament’s Rules of Procedure.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

190

any debate questions are posed and, usually, answered. In addition, there are two issues deserving notice. Firstly, Parliament’s Rules of Procedure foresee liberal other opportunities for (European) Council and Commission members to make statements. They may ask ‘at any time’ the president of Parliament for permission to do so. The latter may decide when the statement is to be made, and whether it is to be followed by a full debate or by thirty minutes of ‘brief and concise’ questions.199 In practice, it is normally Parliament that takes the initiative to invite the Commission or Council to make statements. The president of Parliament, after approval from the Conference of Presidents, can invite the Commission to make a statement in Parliament following each meeting of the Commission, explaining the main decisions taken.200 The statement is again followed by a debate ‘of at least thirty minutes in which Members may put brief and concise questions’. Secondly, the ACP-EC Council of Ministers does not answer questions put to it by the ACP-EC Assembly.201 This appears to have set a precedent for other similar inter-parliamentary bodies (Joint Parliamentary Committees, Joint Co-operation Committees etc.) that involve the European Parliament and other parliaments in the world. 7.4.6

A brief evaluation

Several observations can be made, concerning the increased use of written questions; problems with replies given to them; the need to filter written questions; the development of the Council’s commitment to reply; and the reasons for the differences in popularity amongst MEPs between oral questions and question time. The increased use of written questions. The Members of Parliament use the questions procedures enthusiastically. The twofold reason for this is on the one hand the arrangement of Parliament’s seat, which leaves the MEPs with limited opportunities to engage the other institutions in debates. Additionally, Parliament’s size has inflated dramatically in the last twenty years as a result of both accessions and Treaty revisions. Nonetheless, between 1963 and 1996, the average number of written questions per MEP per year increased almost six-fold from 1.13 to 6.60.202 To some extent, this is the result of the vastly increased scope of European Communities/Union work. Still, even when this is factored in, the increase remains surprising.203 This is put a bit in context when the corresponding ratio of question time questions is taken into account. The data for question time show a similar rising tendency, with an increase from 1.27 per MEP in 1975 to 1.68 in 1996.204 Of course,

199 Ibid., Rule 103(1). 200 Rule 104. 201 Cover note from the ACP Working Party to COREPER—Relations with the ACP States—Oral question put by a member of the ACP-EEC Consultative Assembly on the migration of black workers to South Africa, Council doc. 11500/82 of 2 December 1982. 202 Calculated on the basis of raw data from Parlement, op. cit. footnote 3 at 199 and from Commission document SP (97) 179 of 20 January 1997. 203 And, of course, one has to account for annual idiosyncrasies. Ipsen reports that in the parliamentary year 1969-1970 out of 508 questions 222—i.e., some 44 per cent—were posed by Dutch social democrat Vredeling: Ipsen, Europäisches Gemeinschaftsrecht (1972) at 332. 204 Raw data from Parlement, op. cit. footnote 3 at 199, from Commission doc. SP (97) 179 of 20 January 1997, and from my own calculations.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

191

the time limitations to which question time is subjected constitute an upper limit to its growth. While the use of written questions has experienced a continuous increase, the number of oral questions remained roughly stable, which is a result of the limitation on Parliament’s overall debating time. Problems with replies given to them. There are some differences of opinion between the institutions, especially concerning the replies received from the Council. Occasionally the complaint has been heard that ‘the grace of the Council’s reply is more eye-catching than the relevance of its content’205 and consists of the lowest common denominator of agreement within that institution.206 However, this perceived lack of information is intrinsic to the institutional set-up in the Treaties. The Council is often not able to provide a substantive reply because it has no position yet, or because different Member States have sensitivities in the area. Of course, often the MEPs will be precisely interested in the different points of view within the Council,207 but the Council’s refusal to provide such information puts this beyond the scope of the written question. In my opinion, these limitations on the Council’s replies are defensible in the light of the nature of the institution. The dual nature of the Council as a legislative body and as an executive limits its capabilities to provide a meaningful reply if no decision has been taken. There is no hierarchy between the Member States, or any intrinsic reason why they should easily agree on any particular matter. The Commission, however, was subject to a similar criticism of providing uninformative replies208 although it is bound by the principle of collegiate responsibility which should allow it to provide Parliament with a united view. Does this perceived lack of information imply that the right to ask questions ‘has lost its attractiveness’ to MEPs?209 At first sight, such a conclusion is not borne out by statistical analysis. In the final analysis, the amount of questions per MEP has on average grown since 1979, although the growth is decreasing.210 On the other hand, the inflation in the number of questions seems to have led to a certain debasement of their value. While in the early 1970s the legal literature still followed the questions and their answers,211 the perception of the value of written questions was much lower by the 1980s.212 There is little reason to assume this is any better now, especially not now that Parliament has acquired ‘hard’ prerogatives in other fields. The need to filter written questions. In fairness, the decreased value of the written question as an instrument of control is at least partly the consequence of the MEPs’ sometimes excessive use of it. This problem was identified as early as 1986, when Parliament ruled that written questions had to fall within the sphere of activities of the Communities.213 Regrettably, this limitation was abolished. Notwithstanding their 205 Sassen, ‘Politiek, parlement en democratie in de Europese Gemeenschap’ (1975) at 203. See for Parliament’s criticisms in this direction inter alia its Resolution on relations between the European Parliament and the Council, OJ (1981) C234/52. 206 Bieber, op. cit. footnote 13 at 49. 207 Patijn, op. cit. footnote 4 at 77. 208 Ibid. at 51. 209 In this sense Maurer, (Co-)governing after Maastricht: the European Parliament’s institutional performance 1994-1999 (1999) at 49. 210 Ibid. 211 For example Jacqué’s regular reviews of the activities of Parliament in the Revue trimestrielle de droit européen in the first half of the 1970s. 212 Van Schendelen, Het Europees Parlement (1984) at 80. 213 Bieber, ‘Legal developments in the European Parliament’ (1987) at 372.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

192

importance, one sometimes wonders whether time and money should be spent on issues such as the use of English in the Eurovision song contest,214 the harmonisation of Mothers’ Day in the Member States,215 sex equality in chess tournaments,216 or the fate of children raised by stray dogs in Chile.217 Additionally, the 2004 enlargement of Parliament to up to 732 members increased the already formidable workload with almost 17 per cent in one blow.218 Unlike many national parliaments, the European Parliament imposes no filter on the right to ask written questions. In 1973, Parliament introduced a filter to avoid motions of censure tabled by individual MEPs.219 A similar discouragement, for instance by requiring approval of questions by the groups, would reduce the number of written questions, raise their status and, if anything, induce the other institutions to reply with even more care than resources currently admit. However, no matter how laudable, it is unlikely that such a filter will be introduced soon. The MEPs are unlikely to relinquish one of the few possibilities for backbenchers to profile themselves or to bring their constituents’ problems to public attention. The development of the Council’s commitment to reply. Would it be possible to conceive of a legal duty for the Council to reply to questions? In my opinion, the drafters of the Treaties (and of the Constitution) rightfully decided in the negative. Parliament cannot logically insist on the Council to develop into a house of parliament representing the governments while simultaneously insisting on a duty for it to reply to questions. This is only different to the extent that the Council acts as an executive. In most cases this concerns the second and third pillar, for which Articles 21 and 39(3) TEU provide a specific rule. For the remainder, the Council acts as an executive in the first pillar to the extent that its different co-ordination activities (economic policy, employment, etc.) can be so qualified. Separation of functions purists will point to the Council’s ability to adopt implementing measures under Article 202 TEC, but such Council measures—which, as discussed in Chapter 13 below, are infrequent—will rarely be of sufficient political import to attract the MEPs’ attention. This leaves the Council’s duty to reply to questions in the first pillar largely in the domain of interinstitutional convention. The reasons for the differences in popularity amongst MEPs between oral questions and question time. Oral questions never became very popular with MEPs, partially because the deadlines and their planning into Parliament’s agenda make them less appropriate for actual affairs. This inevitably decreases their spontaneity and information level.220 Moreover, question time provides MEPs with better opportunities to profile themselves since it allows for follow-up questions. Furthermore, the threshold for asking a question in question time is much lower than what is required for oral questions. Another factor may be at play as well. Due to the deadline problem, ministers are necessarily less thoroughly prepared for question time than they are for oral questions 214 Written question E-1514/01 by Mr Xarchakos to the Commission, OJ (2001) CE350/176. 215 Written question 582/81 by Mr Remilly to the Commission, OJ (1981) C222/26. 216 Written question E-2511/99 by Ms Sornosa Martínez and Ms Valenciano Martínez-Orozco to the Commission, OJ (2000) CE225/154. 217 Written question P-1964/01 by Mr Podestà to the Commission, OJ (2002) CE81/82. 218 732 ÷ 626 = 1.17. 219 Jacqué, op. cit. footnote 173 at 734. 220 Grabitz a.o., Direktwahl und Demokratisierung—Eine Funktionenbilanz des Europäischen Parlaments nach der ersten Wahlperiode (1988) at 449.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

193

and are therefore forced to operate more à l’improviste than they would be during an oral question procedure. This provides question time with somewhat more of a potential ‘spectacle level’ than oral questions can. For the remainder, the oral question shares some of the criticisms that have been levelled against the written question. Parliament has complained of receiving ‘uninformative’ replies.221 The reasons for this are similar as those for the replies to written questions discussed above. Summarising these observations, although the oral question procedure is the most formalised procedure, even question time has become a somewhat more petrified affair than some222 hoped at the time of its introduction, at least where questions to the Council are concerned. In Westminster, question hour has a certain spontaneity because ministers normally can answer the question without first having to discuss the matter with their colleagues. In Strasbourg, however, the surprise element that is the strength of the system collides with the nature of the Council, where a logical tendency persists to attempt to ensure the coherence of the institution’s output. This does not render the questions procedures irrelevant as checks and balances between the institutions. They remain useful as ‘low level’ instruments to obtain information. Their use increases if MEPs embed them in a general approach encompassing other parliamentary tools (legislation, budgetary rules, investigations, questions, resolutions, hearings, etc.). 7.5

The Annual General Report and similar documents

7.5.1

Green papers, white papers and reports

The Council, European Council and especially the Commission produce a great number of documents and reports which typically are sent to Parliament and the Council. Even though there is no rule in the Treaties or Constitution obliging it to do so, it is Commission practice to announce major policy initiatives by publishing them. The Commission sends such communications to Parliament and Council as a matter of course.223 Some are published in the Official Journal. As precursors to legislation, such papers have considerable value and Parliament often expresses an opinion on them so as to influence policy making and adoption at the earliest possible stage.

221 See e.g. Parliament’s Resolution on relations between the European Parliament and the Council, OJ (1981) C234/52. 222 Such as Bieber, op. cit. footnote 13 at 35. 223 Commission communication The institutional system of the Community—restoring the balance, Bull. EC Suppl. 3/82 at 9.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

7.5.2

194

Reports

The Commission provides the Annual General Report as well as specialised reports. Every year in February (a month before the opening of the session) the Commission publishes the Annual General Report.224 This instrument has gradually declined in political importance. Article 24 ECSC coupled it to the motion of censure. In the later (E)EC and Euratom Treaties this linkage disappeared (although, theoretically, the report can of course provoke such a motion). In 1967 Parliament adopted a complex procedure for dealing with it225 but in 1975 it ceased assigning a special rapporteur for the matter.226 Article 200 TEC obliges Parliament to discuss it in open session. In practice it is dealt with by means of a report submitted to the plenary.227 Besides the Annual General Report, the Commission, Council and European Council produce a plethora of annual, multi-annual or incidental reports on specialised subjects.228 These reports can have different origins. As far as the Commission is concerned, firstly, both primary law229 and secondary legislation230 require reports in a number of instances. Secondly, the Member States can agree on annual reporting requirements in political texts. Thirdly, the Commission may undertake an annual reporting obligation as the result of a request from Parliament itself.231 Fourthly, annual reporting requirements may rest on little else than an interinstitutional convention.232 The increase in the number of MEPs in the last thirty years has led backbenchers to upgrade the political importance of the most minute policy fields, leading to a considerable amount of requests for annual or one-off Commission reports on the most tiny policy fields.233 The result of all this activity is that the MEPs are inundated with reports on a wide variety of subjects. It also causes the institutions to adopt a microscopic view on policy areas, a tendency reinforced by the division of the Commission’s policy fields over too many Commissioners. It is doubtful whether the coherence of the Commission’s policy benefits from this tendency.

224 Articles 200 and 212 TEC and 113 and 125 Euratom. Article III-337.3 Const repeats Article 200 TEC, which is perhaps a little bit silly in view of the irrelevance that the matter nowadays assumes. 225 Résolution concernant la procédure d’examen des rapports généraux sur l’activité des Communautés européennes, JO (1967) 973, on which Feuer, ‘Parlement européen—1er janvier-30 juin 1967’ (1968) at 112. 226 Constantinesco a.o., Traité instituant la CEE—Commentaire article par article (1992) at 919. 227 Rule 112 of Parliament’s Rules of Procedure. 228 I will not discuss the reports of other institutions or bodies. 229 E.g. in Articles 22, 143 and 145, 159, 173 and 280(5) TEC as well as in Articles III-129, III-216, III-218, III-221, III-255 and III-415(5) Const. See also Article 9 of the Protocol on the application of the principles of subsidiarity and proportionality attached to the EC Treaty and to the Constitution. 230 E.g. Article 11(8) of the Regulation 1073/1999 of the European Parliament and of the Council of 25 May 1999 concerning investigations conducted by the European Anti-Fraud Office (OLAF), OJ (1999) L136/1 or Article 4(14) of Council Regulation 2667/2000 of 5 December 2000 on the European Agency for Reconstruction, OJ (2000) L306/7. 231 See e.g. Parliament’s Resolution of 16 December 1981 on the Community’s anti-dumping activities, OJ (1982) C11/37. 232 E.g. the Interinstitutional Agreement of 28 October 1993 between the European Parliament, the Council and the Commission on procedures for implementing the principle of subsidiarity, OJ (1993) C329/135. 233 E.g. Parliament’s Resolution on development of relations between the European Union and the southern Caucasus, OJ (2002) CE87/261, §§ 9-10 and the Resolution on cultural cooperation in the European Union, OJ (2002) CE72/142.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

195

Pursuant to a request from Parliament,234 the Council too produces an annual report which, like the Commission’s, discusses only past events. In recent years it has dwindled to a paper of minor importance. More importantly, in the CFSP field, Parliament successfully used its budgetary rights as a crowbar to force the Council Presidency to consult the MEPs annually ‘on a document established by the Council on the main aspects and basic choices of the CFSP, including the financial implications for the Communities budget’.235 In addition, various provisions of law require the Council to prepare reports on specialised subjects.236 Occasionally, the Council reports on substantive issues such as, for example, the European Union Code of Conduct on Arms Exports.237 The European Council promised in the Solemn Declaration on European Union of 1983 to ‘address a report to the European Parliament after each of its meetings’.238 It also undertook to address a written annual report to Parliament on progress towards European Union. Both commitments became legal obligations in Article 4 TEU, but these annual reports do not tangibly enhance Parliament’s powers of oversight. Most reports discussed in this section have a considerable informational value, although the debate following the State of the Union speech by the president of the Commission in the Parliament has more political importance.239 Since the reports concern past events, Parliament tends to be more interested in the ‘green papers’ and ‘white papers’, in which the Commission discusses its future policies. 7.6

Conclusions of this Chapter

In sub-section 2.4.3 of this study I argued that a separation of institutions requires checks and balances on policy preparation, policy adoption, policy implementation and the adjudication of disputes. The information rights discussed in this Chapter are the staple of Parliament’s controlling rights concerning policy preparation, policy adoption and policy implementation. Apart from the special rules in the second and third pillars, the Treaty provisions in this respect have remained largely unchanged since 1958. In fact, the main difference of substance between Article III-337 Const and the original Article 140(2)-(4) TEEC is the addition of the European Council. Much of the structure of commitments that regulate the flow of information between Commission and Council on the one hand, and Parliament on the other consists of interinstitutional convention. The MEPs were largely successful in using their rights vis-à-vis the Commission to establish a general flow of information from that institution. Especially after the 234 Bieber, op. cit. footnote 13 at 47. 235 Resolution on the proposal for an Interinstitutional Agreement between the European Parliament, the Council and the European Commission on provisions regarding financing of the Common Foreign and Security Policy, OJ (1997) C286/80. This text was subsumed in the Interinstitutional Agreement between the European Parliament, the Council and the Commission of 6 May 1999 on budgetary discipline and improvement of the budgetary procedure, OJ (1999) C172/9, § 40. 236 E.g. Article 34(1) of the Convention based on Article K.3 of the Treaty on European Union, on the establishment of a European Police Office (EUROPOL Convention), OJ (1995) C316/2 and Article 17 of the Access to Documents Regulation. 237 See e.g. Parliament’s Resolution on the Council’s Second Annual Report according to Operative Provision 8 of the European Union Code of Conduct on Arms Exports, OJ (2002) CE87/136. 238 Point 2.1.4 of the ‘Solemn Declaration on European Union’ adopted by the European Council of Stuttgart (19 June 1983). The address codified a practice originating from 1981 when Ms Thatcher became the first European Council chair to address Parliament on the results of a European Council meeting: Jacqué, ‘Parlement européen (janvier 1981-juillet 1982)’ (1982) at 685. 239 Mathijsen, op. cit. footnote 142 at 22.

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

196

collapse of the Santer Commission, Parliament managed to strengthen its hand by obtaining a very active presence of the Commission during its plenary debates. These achievements were formalised with the 2000 and 2005 Framework Agreements. With the Council the MEPs never enjoyed quite the same leverage. Notwithstanding this handicap, the general thrust of Parliament’s efforts has been to increase the involvement of the Council in Parliament’s work, although Parliament still tends to be unsatisfied about the information it receives from the Council. Here too, the MEPs registered some successes. There is, however, a limit as to how much interaction Parliament can reasonably expect from the Council. In my view, ‘more Council presence’ does not imply ‘more transparency’. The Council’s structure makes it difficult for the presidency to make any meaningful statement to Parliament before a Council position has been adopted. More Council involvement in Parliament’s debates is therefore likely to disappoint those expecting genuine spontaneity or interesting information. In the future, Parliament will continue to face the perplexing problem of its relationship with the Council, occurring because of Council’s dual nature. On the one hand, the MEPs try to treat the Council as a government with reporting obligations similar to those of the Commission. The MEPs ask questions to the Council and they insist on increased Council presence during the debates. On the other hand, there is the view that the Council is—or rather should develop into—a chamber of Member State representatives, somewhat on the lines of the German Bundesrat. From a ‘Madison-style’ separation of functions perspective, there is a fundamental contradiction between these two positions. From a ‘pure’ separation of functions point of view, the two distinct functions of the Council may be much less of a problem, but the relationship between the institutions will differ depending on the function fulfilled by the Council.240 To some extent, the nucleus for this exists already in Article 7 of the Council’s Rules of Procedure, which clarifies when the institution acts in its ‘legislative capacity’ and when it does not. The consequences of the distinction are felt mainly241 in the field of transparency.242 Looking at the matter from a separation of institutions point of view, the first point to note is that the Council’s role as executive in the second and third pillars is catered for by the specific Treaty provisions of Articles 21 and 39(3) TEU. As far as the first pillar is concerned, the Council’s activities are overwhelmingly of a policymaking or legislative kind. That, however, does not imply that they should be free of checks and balances, certainly since the Council is de facto at least as influential as the Commission in setting the policy agenda. Against this background, the MEPs’ question rights—even though the nature of the Council necessarily subjects these to some practical limitations—make perfect sense. Although it technically is a process separate from the development of Parliament’s prerogatives, the Copernican changes in the general law and practice on transparency must be emphasised. Whilst the changes were very considerable in the Commission, they have been revolutionary in the Council. In the latter institution, at 240 An argument could be made that the Council’s first-pillar activities as an executive correspond to those provided in Article 51 of the Court of Justice’ Statute, as amended (OJ (2004) L132/5). However, that definition concerns rather the division of responsibilities between the Court of Justice and the CFI. In any event, in the framework of analysis preferred in this study, the matter is of little relevance. 241 But not solely. The distinction has also some relevance for the drawing up of the Council’s agenda (Article 3(5) of the Council’s Rules of Procedure) and the presidency’s powers (Article 21). 242 Ibid., Article 9(1) (publicity of the results and explanations of votes, etc.) as well as Annex II, Article 11 (documents directly accessible to the public).

CHAPTER 7: PARLIAMENT’S INFORMATION RIGHTS

197

the time of writing roughly two-thirds of all produced documents are directly available to the pubic, whilst a large share of the remainder is provided in whole or in part upon request. This development benefits MEPs; but this is dwarfed by the obvious impact on the general legitimacy of the decision-making process In Chapter 1 of this study the Research Question was posed to what extent it is possible to balance the EU institutions through the use of interinstitutional convention and what the conditions are for doing this. In the field of information rights, interinstitutional convention certainly adjusted the relationship between Commission and Parliament. Parliament’s successful negotiation of the series of interinstitutional agreements that culminated in the 2005 Framework Agreement tangibly reduced the Commission’s scope of action and provided the MEPs with the means to hold the Commission political responsible both at the College level and at the individual Commissioner level. As far as its relationship with the Council is concerned, Parliament managed to increase its influence on this institution through dual ways. On the one hand, the Council has declared itself more and more willing to provide information to the MEPs and is bound to do so by a web of interinstitutional conventions. On the other hand, Parliament had some success in obtaining from the Commission commitments that force it to be the MEPs’ ‘eyes and ears’ in the Council. In the field of information rights, interinstitutional convention thus played a major role in changing the balance between the institutions.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

8

PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

8.1

Introduction to this Chapter

199

Some of Parliament’s prerogatives, such as the motion of censure or written questions, are exercised in respect of a particular institution but others, such as the right of inquiry, have no specific addressee. For the purposes of analysis, the checks and balances that regulate Parliament’s powers of oversight may thus be differentiated into bilateral and erga omnes prerogatives. The former consist of, inter alia, the information rights discussed in the previous Chapter. In this Chapter, Parliament’s erga omnes rights of oversight are discussed. These are the privileges that it can exercise vis-à-vis any natural or legal person, but which may equally be employed towards the Commission or the Council. In section 8.2 I discuss Parliament’s right to adopt resolutions. The following section deals with hearings and committees of inquiry. Lastly, section 8.4 discusses the right to petition and the Ombudsman. 8.2

General right to adopt resolutions

Unlike the French Assemblée,1 the European Parliament can adopt resolutions on any topics that catch its fancy. The ‘right to discuss any question concerning the Communities, to adopt resolutions on such questions and to invite the governments to act’ is—in the words of the Court of Justice—’inherent in the Parliament’.2 Resolutions are Parliament’s prime method of expressing its position on a political issue. Many such resolutions are adopted each year. They may be simply an expression of Parliament’s feelings, or call upon other institutions or third parties to act. The question is what effect such resolutions have as interinstitutional checks and balances. The Court held that ‘a resolution of the Parliament is not binding and cannot give rise to the legitimate expectation that the institutions will comply with it.’3 The basic rule is thus that, contrary to the legislative resolutions adopted in the context of some of the legislative procedures, they have no legal effect. This rule is not, however, without exception. To begin with, resolutions may affect the working of the institutions. Where a resolution affects the internal operation of Parliament itself, the question arises whether it can prejudice other instruments. The matter came to the fore in the controversy surrounding Parliament’s seat. In Luxembourg vs Parliament the Court of Justice was faced with the dilemma whether a resolution of Parliament violated the 1965 Decision on the provisional location of the institutions.4 Interestingly, Parliament played down the effects of its own resolution. The Court disagreed: ‘a determination of the legal effect of the contested resolution is inseparably associated 1 2 3 4

The Conseil constitutionnel laid down several restrictions in its Decision 59-2 DC of 17 June 1959 re the Rules of Procedure of the Assemblée nationale. Case 230/81, Luxembourg vs Parliament, [1983] ECR 255, § 39. Joined cases 87, 130/77, 22/83, 9 and 10/84, Salerno a.o. vs Commission and Council, [1985] ECR 2523, § 59. OJ (1967) L152/18.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

200

with consideration of its content’. It found that the resolution concerned indeed was of a ‘decision-making nature’.5 Finally, the extent to which legislative initiatives embodied in resolutions pursuant to Articles 192(2) TEC (III-332 Const) have legal effects is discussed in subsection 9.2.3, below. Both the Commission and Council accepted commitments relating to Parliamentary resolutions. These are discussed in section 9.4, below. 8.3

Hearings and committees of inquiry

8.3.1

Before Maastricht

Parliaments have a natural right to investigate and obtain information.6 The right to organise hearings and committees of inquiry thus concerns less the right to organise the hearing than the obligation of witnesses, officials and citizens to attend and Parliament’s means to enforce that they do.7 Hearings organised by Parliament’s standing committees must be distinguished from committees of inquiry and temporary committees. Parliament’s standing committees organise hearings on a regular basis. In 1973, the Commission declared itself in principle prepared to co-operate with them.8 Hearings normally do not specifically target the Council or the Member States but they can of course investigate either. Among the Member States, there has been a tendency to co-ordinate the approach towards such hearings. As early as 1982, Member States recognised that it was up to Parliament to decide whom it would invite to hearings and for the Member States to decide whom to send. It is agreed practice for the Member States to inform the Council General Secretariat of any invitation to participate in a Parliament hearing. Within the framework of the GAG, the Member States can exchange information on how they intend to respond to the invitation.9 Such hearings tend to be fairly sedate affairs. They may be useful as informationgathering exercises, but are hardly the epitome of aggressive controlling. Committees of inquiry are quite a different animal. Proposals for such committees were made decades ago.10 The instrument took off only in the 1980s, 11 based on a provision in Parliament’s Rules of Procedure of 1981.12 In 1986, Parliament amended its Rules to limit the potential subject matter of committees of inquiry to ‘alleged contraventions of Community law or instances of maladministration with respect to Community responsibilities’.13 Up to the entry into force of the Treaty of Maastricht,

5 6 7 8 9 10 11 12 13

Joined cases C-213/88 and C-39/89, Luxembourg vs Parliament, [1991] ECR I-5643, §§ 24 and 27. The parliaments in all of the first fifteen EU Member States have a right of investigation. For a description see Beckedorf, Das Untersuchungsrecht des Europäischen Parlaments (1995) at 33-109. Dölle, Het recht van parlementaire enquête (1985) at 11-12. Practical measures to strengthen the powers of control of the Parliament and to improve relations between the Parliament and the Commission, Commission doc. COM (73) 999 final of 30 May 1973, paragraph 14. Note from the General Secretariat to the Working Party on General AffairsParticipation by the Member States in European Parliament hearings, Council doc. 10839/82 of 23 November 1982. For example by Cohen, ‘Het Europees Parlement en de democratie’ (1982) at 162 and by Morgan, ‘Nieuwe taken voor het Europese Parlement’ (1979) at 679-681. For the establishment of the first such committee see OJ (1983) C242/18. Beckedorf, op. cit. footnote 6 at 188. Ibid. at 214.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

201

some eight committees of inquiry did their work,14 although not all of them really dealt with contraventions or maladministration.15 In this period, co-operation of the other institutions and of the Member States was a matter of comity rather than of interinstitutional convention or law. In 1988 the Council decided that, under EC law, there was no obligation for members of Member State governments or national civil servants to be present at meetings of Parliament’s committees of inquiry. Nevertheless, there was nothing stopping such participation either, although Member States could prohibit their national civil servants to cooperate. National civil servants would appear on instruction of their respective Member State governments. Requests from Parliament for the appearance of members of Member State governments or national civil servants to be present at meetings of committees of inquiry were to be addressed to the members of the General Affairs Council. National civil servants were bound to professional secrecy and confidentiality and should at all times be able to refuse questions. The policy towards invitations to appear before committees of inquiry was to be co-ordinated by the GAG or COREPER.16 Parliament accepted that national civil servants had to have the agreement of their authorities to appear before a committee of inquiry. It also agreed to inform the Foreign Ministers of any invitation to national civil servants to that extent.17 I this fashion, interinstitutional convention brought about a degree of co-operation from the side of the Member State authorities. However, the powers of the committees of inquiry were clipped in practice and the MEPs remained discontented with the situation. 8.3.2

Committees of inquiry and temporary committees after Maastricht

The Maastricht Treaty inserted the current Article 193 into the EC Treaty.18 An interinstitutional agreement19 and Rule 176 of Parliament’s Rules of Procedure provide some additional guidelines. In addition to the committees of inquiry20 that find their legal basis under these provisions, Rule 175 foresees temporary committees. In practice, these perform similar functions, although they lack the cover of Article 193 TEC and the interinstitutional agreement. As far as their scope is concerned, the setting up of the committee of inquiry must be ‘in the course of [Parliament’s] duties’. The remainder of the text of Article 193 14 15 16 17

18 19 20

For a list of these see Corbett, Jacobs and Shackleton, The European Parliament (2000) at 261-262. Beckedorf, ‘Das Untersuchungsrecht des Europaischen Parlaments’ (1997) at 238. Council doc. SN 1087/88 of 1988, quoted in Digest of texts on relations between the European Parliament and the Council (1988) at 11-12. Letter of the president of Parliament, Council doc. 5455/88 of 23 March 1988, quoted in Digest of texts on relations between the European Parliament and the Council (1988) at 13. Technically, informing the Ministers of Foreign Affairs is not the same thing as informing the General Affairs Council. Article III-333 Const would replace the interinstitutional agreement with a law of Parliament, acting with the consent of Commission and Council. Decision of the European Parliament, the Council and the Commission of 6 March 1995 on the detailed provisions governing the exercise of the European Parliament’s right of inquiry, OJ (1995) L78/1, corr. OJ (1995) L113/1. Confusingly, both Treaty and Constitution call committees of inquiry ‘temporary committees of inquiry’, although they differ from Rule 175 ‘temporary committees’. An example of a Rule 175 (then: 150(2)) committee was the Temporary Committee on Foot-andMouth Disease: OJ (2002) CE271/51.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

202

defines the scope as ‘alleged contraventions or maladministration in the implementation of Community law, except where the alleged facts are being examined before a court and while the case is still subject to legal proceedings’.21 In addition, the investigation by the committee of inquiry is ‘without prejudice to the powers conferred by this Treaty on other institutions or bodies’. Three such bodies spring to mind, namely OLAF, the Court of Auditors and the Ombudsman. The ‘alleged contraventions or maladministration in the implementation of Community law’ exclude matters falling under the second and third pillars.22 This lacuna would be repaired under Article III-333 Const. The limitations on the scope of committees of inquiry were the main reason for Parliament to retain the temporary committees of Rule 175 of its Rules of Procedure.23 These latter committees can deal with any topic Parliament considers useful, even when it does not fall within the scope of Article 193, but do not benefit from the same procedural guarantees as committees of inquiry. Notwithstanding the wide definition of Article 193, it is not sure that in practice Parliament can always fully enjoy its theoretical rights. The American noncooperation with the Echelon committee (a temporary committee rather than a committee of inquiry) suggests that practice draws boundaries where law fails to. In at least one Member State the experience is similar.24 The main procedural requirement for setting up a committee of inquiry is found in Article 193, which provides that it ‘may’ be established ‘at the request of a quarter of [Parliament’s] Members’.25 For the establishment of temporary committees a majority vote in Parliament is needed, this time without any condition as to a request from MEPs. The ‘may’ in Article 193 allows Parliament to refuse the request, even when it is made by a quarter or more of the MEPs. This happened, for example, in the Echelon case. Instead, a temporary committee was installed. Indeed, since a committee of inquiry under the interinstitutional convention and law at the time was unable to obtain the co-operation of the other institutions in the field of justice and home affairs, this decision made good sense; a committee of inquiry would have seen its wings clipped.26 Parliament’s Rules of Procedure lay down that the request for a committee of inquiry must ‘specify precisely the subject of the inquiry and include a detailed statement of the grounds for it’. Parliament decides on a proposal by the Conference of presidents whether to set up a committee.27 Parliament’s restrictive interpretation of Article 193 is not a priori obvious. Whereas, for example, in the Netherlands the majority of the Second Chamber of Parliament must agree to install a committee of inquiry, this is not the case in, for

21 22 23 24 25 26 27

This rather formal broad scope of the right to inquiry compares with the practice in e.g. the Netherlands, where the right to inquiry is limited only to matters within the purview of parliament: Dölle, op. cit. footnote 7 at 17. Note that neither Article 28(1) nor 41(1) TEU list Article 193 TEC. Wiersma, ‘De parlementaire enquête op z’n Brussels’ (2004) at 225. For the Netherlands see Dölle, op. cit. footnote 7 at 27-29. The requirement is somewhat superfluously repeated in Rule 176(1) of Parliament’s Rules of Procedure. OJ (2001) C121/131 and Report on the existence of a global system for the interception of private and commercial communications (Echelon interception system), Parliament doc. PE 305.391 of 11 July 2001 at 22. Rule 176(3) of the Rules of Procedure.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

203

example, Germany, where a minority of the Bundestag can force an inquiry.28 Article 193 TEC merely requires a request by a quarter of the MEPs, without requiring the approval of the majority of the plenary. The word ‘may’ in those provisions allows for a separate decision but does not necessitate it. An argument could be made that Parliament’s controlling function would be enhanced if a minority were allowed to pursue an inquiry. However, in the context of European politics, the chosen solution is, at least for the time being, justified. German politics is characterised by anessentially bipartisan system and a voting system that often guarantees the government coalition a strong majority in the Bundestag. Requiring a majority of German parliamentarians to agree with an inquiry would effectively kill the instrument. On the other hand, Dutch politics is characterised by a multitude of parties, none of which cherish hopes to acquire more than one-third of the seats in the Second Chamber. Under these circumstances, a majority vote is more logical. The European Parliament similarly consists of many parties which, moreover, are not divided into coalition and opposition blocks. In this setting, a decision by a majority of Parliament makes sense, although this might change if in the future the Commission were to develop into a genuine government-type body and Parliament were characterised by more bipolarism. Under the interinstitutional agreement Parliament, the Council and the Commission agreed that the decision to set up a committee of inquiry, specifying in particular its purpose and the time limit for submission of its report, must be published in the Official Journal.29 It is within Parliament’s province to determine the rules of procedure of the committee of inquiry but of course, only to the extent that the interinstitutional agreement or the Treaties do not provide them. Lastly on procedure, it is practice for the chairman and the rapporteur of the same committee of inquiry to belong to different political groups.30 An interesting question concerns the categories of persons that may be invited to committees of inquiry. Article 3(2) of the interinstitutional agreement provides that the Committee may ‘invite an institution or a body of the European Communities or the Government of a Member State to designate one of its members to take part in its proceedings’. The ‘members’ of the Commission are the Commissioners31 and those of the governments ministers;32 in either case, civil servants are excluded from Article 3(2). The provision must therefore be read separately from Article 3(3), which concerns the appearance of European or national civil servants. Under this latter provision, the committee of inquiry has to make a ‘reasoned request’, after which the institution or body ‘shall designate’ the civil servant authorised to appear, ‘unless grounds of secrecy or public or national security dictate otherwise by virtue of national or Community legislation’. Article 3(3) also obliges civil servants to speak on behalf of their government or institution and thus disallows them to provide their own opinions to the committee.33 In return, civil servants are shielded by ‘obligations 28 29 30 31 32 33

Van der Pot, Donner a.o., Handboek van het Nederlandse staatsrecht (1995) at 549, SchmidtBleibtreu and Klein, Kommentar zum Grundgesetz für die Bundesrepublik Deutschland (1983) at 619. Article 2 of the Decision of 6 March 1995 on the detailed provisions governing the exercise of the European Parliament’s right of inquiry, OJ (1995) L113/2. Wiersma, op. cit. footnote 23 at 227. Article 213 TEC and I-26 Const. Article 203 TEC and I-23 Const. In practice, as far as the committee poses written questions to the Council, roughly the same

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

204

arising from the rules to which they are subject’. These restrictions do not cover Article 3(2) and hence ministers and Commissioners cannot rely on them. However, even ministers and Commissioners benefit from Rule 176(7) of Parliament’s Rules of Procedure: ‘Any person called to give evidence before a temporary committee of inquiry may claim the rights they would enjoy if acting as a witness before a tribunal in their country of origin. They must be informed of these rights before they make a statement to the committee’. There have been few problems in this regard as far as committees of inquiry are concerned, but an issue did arise once in the context of a temporary committee. This occurred when former UK Minister for Agriculture Mr Hogg refused to appear before the Temporary Committee responsible for following up the recommendations on BSE. His government considered that the interinstitutional agreement did not provide for a legal obligation for him to appear. Indeed, Mr Hogg was no longer minister and thus fell outside the scope of both Article 3(2) and (3(3). Parliament disagreed34 but could do little more than request the Commission to institute infringement proceedings against the United Kingdom for infringement of Article 3(2). The Commission refused to do so since it shared the United Kingdom’s interpretation.35 Indeed, the temporary committee was not a committee of inquiry and the interinstitutional agreement did not apply to it. In my opinion, the interinstitutional agreement could bind at most the institutions that signed it but not third parties. In the end, the episode ended with Parliament not quite knowing whom it should criticise, so it took the easy way out by reprimanding the Commission ‘that there continues to be fundamental disagreement between Parliament and the Commission on [the issue of] proceedings against the British government for the failure of Mr Hogg, the UK Minister of Agriculture, to appear . . .’36 Article 3(4) of the interinstitutional agreement provides for a duty for Member State authorities, institutions and EC bodies to co-operate. Parliament’s Rules on confidentiality apply,37 as do confidentiality provisions such as Article 287 TEC (III-430 Const). Limitations on the investigative powers of committees of inquiry: Academia distinguishes the right of committees of inquiry to ask written and oral questions, the right to invite people to hearings, the right to investigation of documents and the right to visit the place of the acts under investigation.38

34 35 36 37 38

procedure is followed as that for written questions. Resolution on the report of the Temporary Committee instructed to follow up the recommendations on BSE, OJ (1997) C371/81, § 9. Report on the European Commission’s follow-up of the recommendations made by the Committee of Inquiry into BSE drawn up by the Temporary Committee to follow up the recommendations on BSE, Parliament doc. PE 223.656/fin.2 of 14 November 1997 at 46. Resolution on the Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions ‘Second Bi-annual BSE Follow-up Report’, OJ (1999) C219/433, § 43. Rules 176(8) of the Rules of Procedure. Beckedorf, op. cit. footnote 6 at 204.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

205

Article 2(2) of the interinstitutional agreement provides that hearings and testimony shall in principle take place in public, but that in-camera testimony is possible. Although it was a temporary committee rather than a committee of inquiry, this procedure was for instance used at a number of occasions by the Echelon Committee. Article 2(3) delineates in more detail the competencies of the committee of inquiry from those of the courts and from the Commission’s function as guardian of the Treaties. Termination of such committees: Article 2(4) of the interinstitutional agreement lays down that the committee of inquiry ceases to exist on submission of its report, or after twelve months after being set up, or at the close of the parliamentary term. Parliament may extend the twelve-month period twice. This is more restrictive than Article 193 TEC, which merely lays down that ‘[t]he temporary Committee of Inquiry shall cease to exist on the submission of its report’. Paragraph 5 provides for an antiharassment clause: a new committee of inquiry may not be set up concerning matters into which an inquiry has already been held by a committee of inquiry in the preceding year unless any new facts have emerged. In the BSE case this barrier was easily evaded by instituting the second committee as a temporary committee, rather than as a committee of inquiry.39 8.3.3

A brief evaluation

Since the Maastricht Treaty, three committees of inquiry have been instituted,40 whereas a handful of temporary committees have done their work.41 In the context of Dutch practice, it has been observed that the main importance of the right of inquiry is its sheer existence.42 Committees of inquiry there are seldom used, but are a formidable threat if the executive is not forthcoming with information. In view of the fairly rare use of committees of inquiry by the European Parliament,43 it is perhaps still too early to say whether the right of inquiry will develop into a similar role. Interestingly, although the committees of inquiry performed very well, it seems that some of the temporary committees, such as the second BSE committee and the Echelon committee, have been at least as successful. Apart from the Hogg affair described above, Koopmans’ view that an obligation to appear is not necessary for committees of inquiry and temporary committees to operate satisfactorily, has been proven right.44 Temporary committees are not covered by Article 193 TEC. The interinstitutional agreement and any acceptance of the other institutions to co-operate with such committees rests at most on convention rather than on law. However, in practice, Commissioners and ministers will normally not refuse an invitation to appear and give evidence to them.45 The biggest case of refusal 39 40

41 42 43 44 45

Decision on the setting up of a temporary committee instructed to monitor the action taken on the recommendations made concerning BSE, OJ (1997) C150/9. The first one concerned the Community transit system, OJ (1996) C7/1. Note the call for evidence published in OJ (1996) C60/1. The second one concerned BSE (bovine spongiform encephalopathy), OJ (1996) C239/1 and OJ (1996) C261/132. The third, set up in 2006, investigates the collapse of the Equitable Life Assurance Society. The most notable ones of these have been the Temporary Committee on the Echelon interception system, OJ (2001) C121/131 and the Temporary committee on foot-and-mouth disease, OJ (2002) CE271/51. Van der Pot and Donner, op. cit. footnote 28 at 550. The logic seems to be that Parliament wants to maximise the publicity factor by only rarely using the instrument: Haag, ‘Artikel 193’ (2004) at 105. ‘Euro-verkiezingen: symbool of panacee’ (1978) at 7. See e.g. the Activity report of the Temporary Committee on Foot and Mouth Disease, Parliament

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

206

of co-operating occurred in the Echelon case, from the side of the United States government.46 However, no European constitutional rule could have changed this. Its co-operation with, and the outcome of committees of inquiry and temporary committees is part of the subject matter for which the Commission is politically responsible. In the past, the MEPs threatened on at least one occasion with a motion of censure if the committee of inquiry’s recommendations were not executed. However, the matter was solved peacefully when the ensuing temporary BSE committee ascertained that the Commission had indeed executed the recommendations.47 Temporary committees and committees of inquiry could perhaps gain more importance as instruments of parliamentary oversight if they were used more creatively. Surely, an inflation of such investigations would probably take away the sense of gravity that surrounds them. But Parliament could perhaps enhance its own toolbox of oversight instruments if it made it possible to combine investigations together with other relevant assemblies, such as Member State parliaments. In the past the suggestion has even been made to conduct joint investigations with e.g. the American Congress.48 Of course, this would require appropriate changes to the Parliament’s Rules of Procedure and adequate checks on the protection of confidential information. 8.4

The right to petition and the Ombudsman

8.4.1

The right of petition

Although the ECSC Assembly foresaw a right to petition in its Rules of Procedure as early as 1953,49 Parliament set up a Committee on Petitions only in 1975.50 The Commission soon agreed that Parliament should act as a first filter for petitions and should possibly forward them with the opinion of its Petitions Committee.51 The right of petition became so successful52 that the Milan European Council of 28-29 June 1985 suggested an interinstitutional agreement on it. 53 In 1989 the presidents of the Parliament, Council and Commission agreed in an exchange of letters that ‘in appropriate cases where the questions involved related to matters of Community competence, the European Parliament should continue to send the Commission requests for assistance, in its capacity as guardian of the Treaties,

46 47 48 49 50 51 52

53

doc. PE 319.300 of 19 December 2002. ‘VS willen niet over Echelon praten’, NRC-Handelsblad of 11 May 2001. Wiersma, op. cit. footnote 23 at 226-227. Morgan, op. cit. footnote 10 at 681. Marias, ‘The right to petition the European Parliament after Maastricht’ (1994) at 169. OJ (1975) C179/31. Reply to written question 979/79 by Mr Van Miert to the Commission, OJ (1980) C86/16. From a very modest beginning of 25 petitions up to the beginning of 1974: Bieber, Organe der erweiterten Europäischen Gemeinschaften: das Parlament (1974) at 103. Currently the number of such petitions lies around 500-1000 annually, with apparently some fluctuation between years. I do not intend to discuss the right of petition beyond its implications for the separation of powers between the institutions. Following a request from Parliament in this direction: Resolution on strengthening the citizen’s right to petition the European Parliament, OJ (1985) C175/273. See also the Resolution on initiatives to strengthen co-operation between the institutions in the examination of petitions submitted to the European Parliament, OJ (1986) C283/86.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

207

or ask it to forward them after Examination [sic] to the Member States concerned’. They hoped that Parliament ‘would be able to obtain as clear and swift replies as possible to those questions which the Commission might decide, after due examination, to forward to the Member States concerned. They pointed to the principle . . . requiring the Member States and the Commission institutions to cooperate wholeheartedly in applying the Treaties’.54 Already one year later, Parliament expressed its satisfaction over the operation of this interinstitutional convention, whilst pressing for the time to reply to petitioners to be reduced.55 The right of petition is a case where the fruits of interinstitutional convention were enacted into primary law. In Maastricht it became one of the components of EU citizenship. The current Article 194 TEC provides that: ‘[a]ny citizen of the Union, and any natural or legal person residing or having its registered office in a Member State, shall have the right to address, individually or in association with other citizens or persons, a petition to the European Parliament on a matter which comes within the Community’s fields of activity and which affects him, her or it directly.’ Somewhat superfluously, Article 21 TEC essentially repeats this, adding only that citizens may use any authentic treaty language and expect a response in the same language. Although Articles 28(1) and 41(1) TEU do not refer to Article 194 and thus seem to exclude the right of petition from the CFSP and PJCC areas, 56 the sweeping wording of Article 44 of the Charter of Fundamental Rights of the European Union57 contains no such restriction and arguably, petitions concerning these two pillars could be based on this provision.58 The right to petition subsequently found its way into Article III-334 Const. In this last provision ‘Community’ is replaced with ‘Union’, thus removing the awkward exclusion of the second and third pillars.59 It has been argued that the right to petition implies the right for a petition to be accepted, checked and replied to.60 At least since the enactment of the right in 54 55 56

57 58 59 60

Exchange of letters between the European Parliament, the Council and the Commission of the European Communities (right of petition), OJ (1989) C120/90. Resolution on the deliberations of the Committee on Petitions during the parliamentary year 19891990, with indications as regards future procedure for handling petitions, OJ (1990) C175/214. Writing before the Charter in this sense Sinnaeve, ‘De positie van het Europees Parlement na het Verdrag van Maastricht’ (1994) at 155. Pliakos, ‘Les conditions d’exercice du droit de pétition’ (1993) at 326-327 defended the exclusion shortly after Maastricht with the argument that citizens have no right to touch these areas of intergovernmental organisation. Surely, by now that position seems outdated. OJ (2000) C364/1. Hölscheidt, ‘Die Ausgestaltung des Petitionsrechts in der EU-Grundrechtecharta’ (2002) at 441. See also Articles I-10(2)(d) and II-104 Const. In this sense inter alia Haag, ‘Artikel 194’ (2004) at 109. Pliakos, op. cit. footnote 59 at 335-336, assumes a duty for Parliament to act. This, in my view, is stated in too broad terms. There may be petitions that, without being indecent, go radically against EU policies.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

208

Maastricht, this seems correct: any other interpretation would deprive Article 194 of effet utile. Rule 192 of Parliament’s Rules of Procedure provide guidelines ensuring that opinions are forwarded to the Commission or the Council. Alternatively, the Petitions Committee can advise petitioners to approach the ombudsmen or petitions committees in their Member State if the petition is better dealt with at that level.61 The Rules do not foresee many active interventions from Commission or Council in Parliament on the subject of petitions, perhaps because petitions do not command as high a priority in Council and Commission as they do in Parliament. Some authors have argued that the principle of Community loyalty implies a duty for the Commission and Council to co-operate with Parliament’s Committee on Petitions.62 This duty, however, is obviously not limitless: for example, the right of petition does not create a shortcut to circumvent limitations on citizens’ rights under law. Notwithstanding Parliament’s success in establishing the right of petition, it was not quite satisfied with the way in which the procedure operated especially since, in its view, the Member States take too much time dealing with petitions.63 The Parliament felt that ‘mandatory’ deadlines were in order. Parliament seeks a firmer legal basis for the co-operation of Council and Commission in the petition procedure. In 2001 it called for a revised interinstitutional agreement,64 something at least the Commission was prepared to consider65 but which, to date, did not materialise. Lastly, the number of cases where the Petitions Committee could correct violations of European law is not overwhelming.66 This is partly because its role, as far as this aspect is concerned, has largely been taken over by the Ombudsman. Indeed, Parliament strives to improve the co-ordination between its Committee on Petitions and the latter officer.67 There is a rough division of powers under which the Committee on Petitions focuses on more political matters or political aspects of maladministration.68 8.4.2

The Ombudsman

Contrary to the right of petition, the Ombudsman is not the product of interinstitutional convention. Before the Treaty of Maastricht, Parliament was not very enthusiastic about the idea of a European ombudsman.69 During the IGC leading 61 62 63

64 65 66 67 68 69

Marias, op. cit. footnote 49 at 180-181. E.g. Haag, op. cit. footnote 60 at 113. See for example the interventions of Messrs Chanterie during the debate of Parliament of 18 July 1996, Schmidbauer during the debates of 9 June 1997, Perry during the debate of 16 July 1998, and Ms Lambert (debate of 6 July 2000), and the Resolution on the deliberations of the Committee on Petitions (2004-2005) adopted on 13 June 2006, not yet reported, § 11. Resolution on the right of petition at the dawn of the 21st century, OJ (2002) CE34/89, § 5 and Resolution on the deliberations of the Committee on Petitions during the parliamentary year 20012001, OJ (2002) CE72/337, § 14. See the intervention of Commissioner De Palacio during the Parliament debate of 6 July 2000. Maniatis, ‘Le règlement des pétitions au Parlement européen’ (2002) at 140. Resolution on the deliberations of the Committee on Petitions during the parliamentary year 20012002, OJ (2003) CE273/251, § 9 and Resolution on the annual report on the activities of the European Ombudsman for the year 2001, OJ (2003) CE273/247, § 21. Marias, op. cit. footnote 49 at 182. Haag, ‘Artikel 195’ (2004) at 115 and Bieber, ‘Legal developments in the European Parliament’ (1986) at 352; although in 1979 the MEPs favoured the idea: Resolution on the appointment of a

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

209

to that Treaty, Denmark suggested the office70 and it is perhaps for that reason that it has been modelled on that of his Danish counterpart.71 In line with the Research Question formulated in Chapter 1, the discussion in this sub-section is limited to his position in the institutional system; his use by MEPs as a means to reinforce their information rights; deadlines for replying to complaints; and confidential information. His position in the institutional system. The Ombudsman is not an institution in the sense of Article 7 TEC.72 The CFI emphasised the Ombudsman’s de facto independence by disallowing an action against Parliament concerning acts committed by him.73 I respectfully submit that the connection between Parliament and Ombudsman is such that he would be better considered as a body of Parliament, albeit with a very large measure of independence. This follows from the place of Article 195 TEC (III-335 Const) in the Section called ‘The European Parliament’. Moreover, the close organic relationship between the Ombudsman and Parliament is also clear from Article 3(4) of the Ombudsman’s Statute: ‘If the assistance which he requests is not forthcoming, the Ombudsman shall inform the European Parliament, which shall make appropriate representations’.74 Under Article 3(7) of the Statute, the follow-up action on maladministration consists of a report to Parliament. The fact that the office effectively acts as an independent body does not necessarily change this. For example, any decision of OLAFanother body with fargoing independencealso needs to be attacked in court through an action against an institution (the Commission) that has little effective control over it. Moreover, the Ombudsman is appointed by Parliament, which also adopts the Decision setting out the conditions of his office. Unlike the Treaties, the Ombudsman’s Statute provides for personal separation of powers as far as he is concerned. During his term of office, the Ombudsman may not engage in any other political or administrative duties, or any other occupation.75 His use by MEPs as a means to reinforce their information rights. The ambit of the Ombudsman’s powers is to check whether European institutions or organs committed ‘maladministration’.76 This term has been borrowed from the British Parliamentary Commissioner Act 1967, where it includes bias, neglect, inattention, delay, incompetence, ineptitude, perversity, turpitude and arbitrariness.77 In his Annual Report over 1995, the Ombudsman clarified that

70 71 72 73 74

75 76 77

Community Ombudsman by the European Parliament, OJ (1979) C140/153. Jacqué, Commentaire J. Mégret 9 (2000) at 66-67. Bonnor, ‘The European Ombudsman: a novel source of soft law in the European Union’ (2000) at 40. Or, for that matter, within the sense of Article 232 TEC: case T-103/99, Associazione delle cantine sociali venete vs Ombudsman and Parliament (Order), [2000] ECR II-4165, § 46. Case T-209/00, Lamberts vs Ombudsman and Parliament (Order), [2001] ECR II-765. The Court of Justice implicitly agreed with this position: case C-234/02P, Ombudsman vs Lamberts, [2004] ECR I-2803. Previously Haag, op. cit. footnote 69 at 117 argued on the same lines. The Ombudsman’s Statute is laid down in Decision of the European Parliament of 9 March 1994 on the regulations and general conditions governing the performance of the Ombudsman’s duties, OJ (1994) L113/15. The Decision was amended in 2002 to adapt it to changes in the budget: Decision of the European Parliament of 14 March 2002 amending Decision 94/262/ECSC, EC, Euratom on the regulations and general conditions governing the performance of the Ombudsman’s duties, OJ (2002) L92/13. Article 10(1) of the Ombudsman’s Statute. Article 195(1) TEC deliminates this mandate, especially as regards the Courts in their judicial role. Schipper, ‘Enkele kanttekeningen bij de Europese ombudsman’ (1995) at 492.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

210

‘[a]ll complaints against decisions of a political rather than an administrative nature are regarded as inadmissible . . . Nor, for example, is it the task of the Ombudsman to examine the merits of legislative acts of the Communities such as regulations and directives’.78 Notwithstanding these limitations, Parliament (or better, MEPs) occasionally tries to use the Ombudsman in order to obtain more information from the Council and the Commission. A precedent arose in 1996, following a complaint by a Danish MEP, about the failure of the Council to make certain documents public. The Council considered, inter alia, that the request was of a ‘political’ nature rather than a matter of maladministration. The Ombudsman, however, disregarded this argument, after noting that ‘[t]he complainant’s status as a Member of the European Parliament had no bearing on the matter’.79 Since that time, the Council treats requests from MEPs that are not made through the questions procedures similar to requests originating from other citizens. In 2004, an MEP complained to the Ombudsman that the Council’s Rules of Procedure did not foresee public deliberations every time the institution acts in its legislative capacity.80 The Council’s reply was based on the premise that it had correctly applied Article 207 TEC as well as its Rules of Procedure.81 Notwithstanding the fact that no one had argued that the Council misapplied its Rules of Procedure, the Ombudsman found that these were themselves an instance of maladministration since, in his view, they violated Article 1(2) TEU.82 It is respectfully submitted that, from a separation of institutions point of view, this finding is flawed. The adoption of the Council’s Rules of Procedure is a matter falling within the autonomy the institution concerned, subject of course to the existence of higher rules. Irrespective of what one’s views are on the usefulness of the Council legislating in public, it takes a certain creativity pour le besoin de la cause to read a ‘general principle’ into the programmatic wording of Article 1(2) TEU. This provision proclaims among others that the EU Treaty marks a ‘new stage’ in the ‘process’ of ‘creating’ an ‘ever closer’ union in which decisions are taken as openly as possible and as closely as possible to the citizen. The Ombudsman’s interpretation sacrifices the effet utile of all of these guarded clauses to teleology. Moreover, unlike some of the provisions in the Rules of Procedure dealing with access to documents, the criticised rules on public deliberations are not intended to provide rights to third parties. Arguably, the adoption of an act of such constitutional

78 79 80

81

82

Ombudsman’s Annual Report over 1995 (1996), § 1.3.2. Ombudsman’s Annual Report over 1996 (1997) at 43. Cover note from Mr P. Nikiforos Diamandouros, European Ombudsman to Mr Javier Solana, Secretary General/High Representative of the Council of the European Union—Complaint made by Mr Brok, MEP and by Mr Piesbergen to the European Ombudsman (2395/2003/GG), Council doc. 5366/04 of 21 January 2004. ‘I/A’ item note from the Working party on Information to COREPER (2nd part)/Council—Follow-up to the complaint by Mr Brok, MEP and by Mr Piesbergen to the European Ombudsman (2395/2003/GG, 375/2004/GG and 478/2004/GG), Council doc. 11450/04 of 19 July 2004. Article 8 of the Rules of Procedure does foresee public debates in certain cases (notably in the codecision field). In addition, each presidency proposes a list of topics on which open debates could be held. Cover note from Mr P. Nikiforos Diamandouros, European Ombudsman to Mr Javier Solana, Secretary General/High Representative of the Council of the European Union—Complaint 2395/2003/GG to the European Ombudsman—Special report to the European Parliament, Council doc. 13507/05 of 19 October 2005.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

211

importance as the Rules of Procedure hardly qualifies as ‘administration’ or, for that matter, ‘maladministration’. Deadlines for replying to complaints. Although Article 195 TEC provides for only one occasion for the accused institution to defend itself, in practice the procedure consists of two steps. The institution is notified of the complaint and given three months83 to reply. Following this, the Ombudsman will issue a draft recommendation, which is referred to the institution for its comments in accordance with the Treaty provision. Subsequently, the Ombudsman adopts his final recommendation.84 In 2004, the Ombudsman suggested to the Council to shorten the three months’ deadline in the first step of the procedure to two months in cases concerning the access to documents. The Council politely refused the idea of such a convention.85 The Ombudsman’s access to confidential information caused a minor controversy between his office and the Commission. The latter noted in 2002 that it wished to maintain the practice whereby the Ombudsman can take note of certain confidential documents without further divulging them.86 In response, the Ombudsman simply denied that such a practice existed and refused to accept that he should decide on the basis of documents not seen by both parties to the dispute. In such cases, the document is not taken into account and returned to the Commission.87 8.5

Conclusions of this Chapter and of this Part

Interinstitutional conventions altered Parliament’s rights of oversight to a large extent. Special examples are the committees of inquiry/temporary committees and the right of petition. The conventions that shaped these mechanisms give substance to especially the Commission’s responsibility towards Parliament. As such, they amended the system of checks and balances between the institutions. In Chapter 1 the Research Question was posed, to what extent it is possible to balance the EU institutions through the use of interinstitutional convention and what the conditions are for doing this. The discussion in this Part clearly shows that interinstitutional conventions do affect the ways in which Parliament monitors the other institutions. Interinstitutional conventions gave substance to the political responsibility towards the MEPs. The development of the questions procedures are, certainly where the Council is concerned, mostly the product of interinstitutional convention. The development of Parliament’s rights of oversight thus suggests a positive answer to the Research Question.

83 84

85 86 87

The three months’ deadline is laid down in Article 4.3 of the Decision of the European Ombudsman adopting implementing provisions. Articles 4 and 7 of the Ombudsman’s Implementing Provisions (which are only published on his website) provide for a possibility to close the matter with a maladministration finding without providing the institution with a chance to comment on the draft finding. It is somewhat difficult to combine this ‘fast track’ procedure with the letter of Article 195 TEC. ‘I/A’ item note from the Working Party on General Affairs to COREPER (Part 2)/Council—Draft reply to the European Ombudsman concerning the time limit for an opinion on a complaint concerning access to documents, Council doc. 10894/04 of 28 June 2004. Commission Opinion on amendments to the regulations and general conditions governing the performance of the Ombudsman’s duties, Council doc. 7180/02 of 15 March 2002, § 10. Copy of letter from Mr Söderman, European Ombudsman to Mr Haarder, President of the CouncilRegulations and general conditions governing the performance of the Ombudsman’s duties (‘the Statute’), Council doc. 9543/02 of 19 September 2002.

CHAPTER 8: PARLIAMENT’S ERGA OMNES RIGHTS OF OVERSIGHT

212

Moreover, several of the interinstitutional conventions discussed in this Chapter have been enacted into primary law. The right of petition springs to mind, as well as the question rights as far as external relations are concerned.

213

PART IV: LEGISLATIVE PROCEDURES

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

9

LEGISLATION: GENERAL OBSERVATIONS

9.1

Introduction to this Part

215

In Chapter 2 of this study I argued that it is possible to divide state functions into 1) policy preparation and adoption, 2) policy implementation and 3) the adjudication of disputes. Viewed from a separation of institutions angle, the right of institutions to participate in the legislative process is then best interpreted as a check on the other institutions’ ability to adopt and implement policies. The Commission’s right of initiative guarantees an input in policy. Similarly, Parliament and the Council have guarantees to influence policies through their right to amend and adopt legislation. The Treaties foresee no general right of legislation. The main principle of attribution of competence implied that the drafters allocated legislative powers on a provision-by-provision basis. Moreover, the variety of procedures foreseen in the Treaties is bewildering: regulations, directives and other instruments are adopted by the Council, by the Commission, with or without Parliament, or by any combination of these three, with or without the assistance of other institutions or bodies. The addition of the co-decision procedure during the Treaty revisions of the 1990s did little to simplify EU law and added to the many—often quirky—procedures that litter primary law. The Constitution would have improved on this state of affairs by establishing the ‘ordinary legislative procedure’ of Article III-396 Const, which is based on the codecision provision of Article 251 TEC. However, a complete harmonisation of legislative procedures proved not feasible, leading to a multiple-step system. Firstly, the Constitution distinguishes between (framework) laws1 and other rules of general application (for example, regulations) based on it. The distinction between legislative and regulatory matters evokes the separation of the ‘pouvoir réglementaire’ and the ‘pouvoir législatif’ that characterises French constitutional law, but there is a difference. Unlike Articles 34 and 37 of the French Constitution, Part I of the draft European Constitution does not clearly spell out which matters are dealt with through which procedure and it is necessary to resort to the relevant provision of Part III to ascertain the applicable legislative procedure. To a large extent, legislative acts are reserved for those instruments that lay down the essential elements and fundamental political choices. 2 However, this distinction has not always been straight-forward since, in one of the rare cases where the Constitution proposes to empower European institutions to adopt penalties for civilians, they must do so by regulation rather than by law (Article III-190.3).3 1

2 3

In this study the somewhat useless adjective ‘European’ added to each of the instruments of Article I-33 Const is dispensed with. In addition, in general terms, the legislative procedures for framework laws do not differ for those of laws. Therefore I will in the remainder of this study only refer to ‘laws’. Lenaerts, Binon, and Van Nuffel, ‘L’Union européenne en quête d’une Constitution: bilan des travaux de la Convention sur l’avenir de l’Europe’ (2003) at 293, § 36. Dougan, ‘The Convention’s draft Constitutional Treaty: bringing Europe closer to its lawyers?’ (2003) at 783-784, Geelhoed, ‘Een Europawijde Europese Unie: een grondwet zonder staat?’ (2003) at 297 and Lenaerts, ‘Structuurelementen van de Unie volgens de Grondwet voor Europa’ (2004) at 404 forward similar comments.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

216

Secondly, the distinction between (framework) laws and regulations is to a large extent one of form, rather than indicative of the authorship of the instrument.4 Some regulations are adopted by the Commission (for example those of Articles III-133.3(d) and III-165.3 Const) whilst others are the result of a Commission proposal adopted by the Council (for example Articles III-130.3 and III-163) and again others are solely the domain of the Council (Article III-187.4) or even the ECB (Article III-190.1(a)). Whether intentionally or not, by labelling certain instruments ‘regulation’, the Constitution’s authors put them outside the realm of the small constitutional review procedure of Article IV-444.2 Const. Thirdly, although all (framework) laws are in principle of equal legal value, even they would not all be adopted according to the same procedure. Whilst most5 would follow the ordinary legislative procedure, others follow special procedures as provided for in the Constitutional provision on which they are based. It is difficult to perceive general rules for the adoption of such special legislative procedures, other than that the authors of the Constitution were at times forced to bow to political reality. Lastly, there are implementing instruments that can be decisions of an individual nature, implementing rules, or a mixture of both. In the system currently in force, this complex of rules and procedures is referred to in shorthand as ‘comitology’ and, for easiness’ sake, I will do so as well. The above divisions usefully divide the subject of this Part even under current primary law and they largely guide the organisation of this and the following Chapters. The next Chapter deals with the consultation procedure, which is the most venerable of the legislative procedures. The interinstitutional conventions that surround co-decision are discussed in Chapter 11. The subsequent Chapter deals with other legislative procedures, whilst Chapter 13 analyses implementing instruments. This Chapter deals with general observations on policy-making and legislative procedures. The next section discusses interinstitutional convention in the field of legislative and policy planning. This is followed by an analysis of conventions concerning voting in the Council (9.3). Finally, the conventions concerning interinstitutional contacts during the legislative procedure are discussed. 9.2

The Commission’s right of initiative

9.2.1

Legislative and policy planning and the right of initiative

Unlike Article I-26.2 Const, the Treaties do not explicitly lay down that, in principle, the right of legislative initiative is the Commission’s. Some authors regard the quasi-exclusivity of the Commission’s right of initiative as an infringement of Parliament’s natural rights. In this logic, inspired by functional separation of powers thinking, Parliament’s inability to adopt a general norm without the intervention of other institutions is an infringement of democratic principles. However—contrary to what some scholars seem to imply6—, even in the Member States, legislatures are not fully autonomous in norm setting. For example, the Dutch Second Chamber has the right of initiative, but any legislative proposal passed by it 4 5 6

Barents, ‘Naar een Europese constitutie? (IV)’ (2003) at 140. 95 per cent, in Ponzano’s estimation: ‘La réforme des institutions de l’Union européenne dans le cadre de la Constitution’ (2004) at 27. E.g. Martens, ‘Die rechtsstaatliche Struktur der Europäischen Wirtschaftsgemeinschaft’ (1970) at 211.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

217

must receive the assent of the First Chamber (Senate) and of the government. Moreover, as follows from Chapter 2 of this study, the purported ‘right’ of legislatures to legislate or to take the initiative for legislation, whilst defendable from a separation of functions point of view, is not a priori logical when viewed from the angle of separation of institutions. As argued above, from such a point of view, the right of legislation is better seen as one of the checks that parliaments have to influence the implementation of policy than as a state ‘function’. This is not to deny the importance as to the question of the legitimacy of the political system. The point is rather that, in modern political systems, policy is more relevant than legislation. In view of the amount of policy that is required in any modern polity, parliaments are hopelessly out-competed by national administrations. Viewed from that angle, construing parliaments’ legislative prerogatives as checks and balances has at least the advantage that it starts from realistic expectations. In the older literature the implicit assumption was made that exclusivity of legislative initiative implies exclusivity of policy initiative.7 Even if that would have been true in the past, the situation is much more complicated nowadays. In practice, the policies embodied by legislation often are put on the rails or influenced by other institutions. In this section I first discuss interinstitutional conventions concerning legislative planning, and subsequently the exceptions to the Commission’s right of initiative. 9.2.2

Legislative and policy planning The Commission’s annual programme

Parliament has no right of legislative initiative, but managed to carve out a role in policy formulation through interinstitutional convention. In addition, the Commission had to allow the Council and the European Council beside it in this field. The combined effect of the pressures from these institutions causes some 90 per cent of legislation to be initiated at the instigation of actors other than the Commission.8 According to recent research, Commission proposals have the following origin:  20-25 per cent are a follow-up to Council or Parliament resolutions or to requests on the parts of the social partners or economic operators;  About 30 per cent arise from the EU’s international obligations;  10-15 per cent relate to obligations under the Treaty or secondary legislation; and  About 20 per cent update existing European regulation.9 In practice, there is long tradition of the Commission announcing and explaining its policies in Parliament.10 Since the early days of the Communities, the Commission agreed to present its annual work programme during Parliament’s February session.11 7 8 9 10 11

In this sense e.g. Freestone and Davidson, The institutional framework of the European Communities (1988) at 93. ‘Short account of the Convention on the Future of Europe—The institutional debate begins and is significantly revealing but essential points still remain’, Agence Europe of 19 September 2002 at item 1. Mandelkern Group on Better Regulation, Final Report of 13 November 2001 at 55. Ponzano, ‘Le processus de décision dans l’Union européenne’ (2002) at 39 makes a similar point. This started with the speech by the Commission President at the first session of the Parliamentary year: Commission doc. COM (73) 999 final of 30 May 1973 at 3 and at the Annex. This happens on a systematic basis since the Rey Commission: Débats du Parlement européen, JO (1970) Annexe 121/133-134. The commitment was repeated in Annex 1 to Commission doc. COM (73) 999 final and in Parliament doc. 1-328/84/rev of 23 May 1984.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

218

This is subject of a debate and vote in Parliament. In addition, the president of a new Commission would make a declaration in Parliament. Since the Malfatti Commission (1970) this is systematically followed by a debate.12 In 1980 the Commission, Council and Parliament agreed to collaborate on an ‘annual programme of routine or periodic work’,13 enabling Parliamentary committees and the Council presidencies to plan their meetings. In 1986, Parliament amended its Rules of Procedure, requiring the Commission to provide an annual legislative programme and timetable for submission of proposals.14 Following the adoption of the Maastricht Treaty, the three political institutions agreed new interinstitutional convention in the field of legislative planning. The procedure laid down in the Interinstitutional declaration on democracy, transparency and subsidiarity15 foresees that the Commission ‘proposes’ an annual legislative programme, on which Parliament adopts a resolution. Subsequently the Council would ‘state its position’ in a declaration and ‘undertake to implement as soon as possible the provisions to which it attaches priority’ on the basis of ‘formal’ Commission proposals. The spirit of this declaration was that the Commission’s right of initiative was a somewhat formal matter. Rather, the policies behind the legislative proposals mattered. It is on these that the other two institutions sought to get a grip. In this respect, the interinstitutional agreement evidenced realism instead of a formalistic (or legalistic) approach. Indeed, in 1994 the Commission duly proposed its legislative programme, which was followed by two Parliamentary resolutions,16 a Council declaration17 and even a Joint Declaration on the Legislative Programme for 199418 in which Parliament and the Commission, ‘in the presence of the Council’ ‘agreed’ their joint priorities for the year. That latter Declaration was not foreseen by the 1993 interinstitutional agreement, but was not illogical. It implemented the 1993 convention inter alia by foreseeing an interinstitutional calendar for the adoption of common positions and second readings, activities that involve all three institutions. This procedure has, largely,19 been followed in the following years.20 In 1995 and 1996 Parliament and the Commission adopted a ‘joint declaration’ in which they agreed on the work programme.21 The 1995 joint declaration included commitments from the Commission to submit the work programme by October of the year n-1, so that Parliament could debate it. In 1995, Mr Santer committed the Commission to 12 13 14 15 16 17 18 19 20 21

Patijn, Het Europees Parlement—de strijd om zijn bevoegdheden (1973) at 36-37. Currently, this latter convention is reflected in paragraphs 28 and 29 and Annex 2 of the Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125. Parliament doc. PE 69.767 BUR, quoted in Digest of texts on relations between the European Parliament and the Council (1988) at 17. Bieber, ‘Legal developments in the European Parliament’ (1987) at 368. OJ (1993) C329/133. Resolution on the proposal for the legislative programme for 1994, OJ (1994) C44/98 and Resolution on the statement by the Commission on the Legislative Programme for 1994, OJ (1994) C60/26. Council Declaration (legislative programme), OJ (1994) C60/25. OJ (1994) C60/1. In 2001 a minor scandal arose when Parliament refused to accept the proposed work programme as such: Resolution on the presentation of the Commission’s programme for 2002, OJ (2002) CE177/287. For a description of the procedures followed to establish the annual programme see Communication from the Commission to the Council and the European Parliament—Stocktaking of the follow-up to the Annual Policy Strategy for 2003, Commission doc. COM (2002) 427 final of 28 August 2002. For 1995 see OJ (1995) C225/1, for 1996 OJ (1996) C282/1.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

219

providing detailed information during the course of the year on cases where the work programme had not been followed. A year later, the two institutions agreed to exchange their respective legislative calendars on a monthly basis. In each year the Council adopted a separate statement of its priorities. With the 2000 Framework Agreement, the Commissioners further caved in to Parliament by agreeing to ‘take the utmost account of the guidelines’ suggested by the latter. The Commission promised to present its programme in sufficient time to permit a wide public debate on it.22 Even though, since 2002, the Commissioners appear in Parliamentary committees to discuss policy planning and the Commission sends the other two political institutions regular updates of its operational planning,23 the MEPs still complained that their institution should be more closely involved in the development of the work programme through preliminary presentations by the respective Commissioners of ‘an outline programming document’.24 In addition to the legislative planning, the Commission and Parliament agreed that the latter should receive ‘all its strategic initiatives, such as Green or White Papers, in areas in which the Community could be required to act’.25 The 2003 Interinstitutional agreement on better law-making26 attempted to further co-ordinate these efforts by adding yet another layer to them. The agreement foresees in a general obligation of the three institutions concerned to co-ordinate their legislative timetables. Under the agreement, Parliament and Council (but not the Commission) are to agree for each co-decision proposal an indicative timetable for the various stages of its adoption. Although these commitments are, on the face of it, purely procedural, they do take the Commission’s initiative out of its hands as soon as it has been forwarded to the other institutions. Lastly, paragraph 27 of the 2005 Framework Agreement requires the Commission to present proposals for the EU’s multi-annual programming, ‘with a view to achieving consensus on inter-institutional programming between the Institutions concerned’. Annex 2 to the agreement spells out a timetable requiring the Commission to present an Annual Policy Strategy to Parliament each February. This is followed with a debate during the February/March part-session, after which the Parliamentary committees continue the dialogue with the Commission. Overall overview is retained by Parliament’s Conference of Committee Chairs, which reports to the Conference of Presidents in September. During the following November session, the Commission president then presents the legislative and work programme for the following year.27 Tellingly, the Framework Agreement keeps the door open for an arrangement involving the Council, by clarifying in Annex 2 that it ‘shall not prejudice any future agreement on inter-institutional programming’.

22

23 24 25 26 27

See Annex 1 to the Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121/122 for the details of the commitments. That convention contained other detailed arrangements concerning the organisation of policy planning by Parliament, which I will not further discuss. Mr Prodi during the Parliamentary debate of 20 November 2002. In this sense e.g. Parliament’s Resolution on the Commission’s legislative and work programme for 2003, OJ (2004) CE27/146. Paragraph 3.1 of the 1995 Code of conduct, OJ C89/69 of 10 April 1995. This commitment follows up on the Joint Declaration of the European Parliament and the Commission, Parliament doc. 1-328/84/rev of 23 May 1984. OJ (2003) C321/1, corr. OJ (2004) C4/7. The details of the procedure can be found in Annex 2 to the Agreement.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

220

In addition to these annual and shorter-term planning documents, the Commission also adopts ‘strategic objectives’ covering a longer (five-year) period.28 As a direct consequence of all of these developments the Commission’s right of initiative, unrestrained in law, has been bound with the silk ties of political obligation. Parliament beyond doubt gained some influence over policy initiation and making in the Union but other factors worked to the MEPs’ disadvantage. It is to these that we turn now. Policy initiation by the Council and European Council The Council presidency, which gradually developed from merely chairing meetings to a policy centre in its own right, briefs Parliament on its programme. This, however, is to a large extent dominated by on-going business, inherited dossiers and the political events of the day. As far as large policy lines are concerned, the European Council29 is more important in practice. Up to the entry into force of the Single European Act, the European Council was purely the child of convention. Although in time it acquired a solid legal basis and some30 formal competencies, its operations remain largely outside the field of law. The European Council arguably is a much bigger factor than the presidency is in influencing the Commission’s freedom to propose policies. For this reason it is necessary to expand somewhat on its policy-initiating role, and notably on its tasks and institutional position; its distinction from the Council; its preparation and its relationship with the Commission. Its tasks and institutional position. Overcoming Dutch resistance,31 the heads of state or government decided at the Summit Conference of Copenhagen of 14-15 December 1973 to meet ‘whenever justified by the circumstances’ and ‘when it appears necessary to provide a stimulus or to lay down further guidelines for the construction of a united Europe’. Following a speech by French president M. Giscard d’Estaing on 27 August 1974,32 the European Council was born. The Stuttgart European Council of 17 and 19 June 1983 adopted the first detailed description of this virtual institution’s tasks, which included providing ‘a general political impetus to the construction of Europe’, defining ‘approaches to further the construction of Europe’, issuing ‘general political guidelines’ for the Communities and EPC and initiating ‘co-operation in new areas of activity’.33 VerLoren van Themaat34 argues that ‘orientations’ of the European Council cannot be legally binding to the Council, since this would impair Parliament’s co-decision rights as well as the Council’s political and legal responsibility. That misses the point somewhat: 28 29 30

31 32 33 34

Strategic objectives 2005-2009: Europe 2010: A Partnership for European Renewal. Prosperity, Solidarity and Security—Communication from the President in agreement with Vice-President Wallström, Commission doc. COM (2005) 12 final of 26 January 2005. I do not know who coined the name ‘European Council’. The name was not even mentioned in the Communiqué of the 1974 Paris Summit that announced its birth: Everling, ‘Die Rolle des Europäischen Rates gegenüber den Gemeinschaften’ (1995) at 41. Articles 4, 13, 17(1), 23(2), 40a(2) TEU, 11(2), 99(2), 128(1) and 128(2) TEC. In the draft Condtitution the European Council’s competencies are strongly expanded: Articles I-20.2, I-21, I24.4, I-26.6, I-27.1, I-28.1, I-40.2, I-40.3, I-40.7, I-41.2, I-55.4, I-59.2, I-60.2, I-60.3, III-136.2, III179.2, III-198.2, III-206.1, III-258, III-270.3, III-271.3, III-274.4, III-293.1, III-295.1, III-300.2, III302.3, III-329.3, III-337.1, III-382.2, IV-440.7, IV-443.2, IV-444.1, IV-444.2, IV-445.2 Const. Van Oosten, ‘Supranationaliteit—principes, paradoxen en competenties’ (1976) at 613. Lauwaars, ‘The European Council’ (1977) at 25. Paragraph 2.1.2 of the Solemn Declaration attached to the Stuttgart European Council conclusions. ‘De constitutionele problematiek van een Europese Politieke Unie’ (1991) at 442.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

221

the Council can be legally responsible, but it will be difficult politically for any national member of parliament or MEP to trace the political responsibility for any decision taken in execution of such general political guidelines. The Single European Act enacted the European Council’s existence in its Article 2.35 That provision, which itself was largely copied from the 1983 Stuttgart Solemn Declaration,36 became Article 4 TEU with some minor drafting changes. The latter Treaty also formalised the European Council’s tasks. Articles 13(1) and (2) TEU comprise a special provision with regard to the CFSP, whilst Article 99 TEC gives the European Council a role in the formulation of the broad guidelines of the economic policies of the Member States and of the Community. Some argue that the European Council has absolute decision-making power.37 This perhaps betrays a somewhat formalistic way of thinking about the institutions. Moreover, it presumes a unity of the European Council that often is not prevalent. A different and, probably, more correct observation would be that listing the European Council in the Constitution as one of the institutions may open the door to a European Council secretariat and bureaucracy.38 Its distinction from the Council. At least since the Single European Act,39 the European Council is an entity different from the Council and, for that matter, from the ‘Council, meeting in the composition of the Heads of State or Government’40 and from the ‘governments of the Member States at the level of Heads of State or Government’.41 The reasons for this are twofold. Firstly, the Council consists of ‘a representative of each Member State at ministerial level, authorised to commit the government of that Member State’.42 On the contrary, Article 4 TEU makes it clear that the president of the Commission is a member of the European Council.43 It follows that, certainly since the Single European Act44 or better, since the 1983 Stuttgart Declaration, the European Council can no longer be considered as a specialised formation of the Council. For this reason I cannot agree with Glaesner,45 who seems to argue that the distinction between the European Council and the Council meeting in the composition of heads of state or government is one of ‘spitzfinder Formalismus’.

35 36 37 38 39 40 41 42 43

44

45

Which, contrary to the view of VerLoren van Themaat, ‘De Europese Akte’ (1986) at 464, implies more than a mere ‘codification’ of existing practice. Paragraph 2.1.1 of the Solemn Declaration on European Union. Einem, ‘Eine Verfassung für Europa—Anmerkungen zu ausgewählten Aspekten des Verfassungsentwurfs’ (2004) at 210. Ponzano, op. cit. footnote 5 at 29. Until the Single European Act, the Council’s view was that the European Council, when deliberating matters falling within the scope of the Treaties, was a Council ‘meeting at the highest level’: reply to written question 326/79 by Lord O’Hagan to the Council, OJ (1979) C310/4. This features in Articles 7 TEU, 121(2), (3) and (4), 122(2) and 214(2) TEC. Articles 112(2)(b) and 117(1) TEC. Currently Article 203 TEC. Martenczuk, ‘Der Europäische Rat und die Wirtschafts- und Währungsunion’ (1998) at 153 tries to square the circle by trying to establish that the European Council, when deciding on EU matters, does not include the President of the Commission. Interesting reasoning, but unfortunately hard to reconcile with the letter of either the Treaties or the Presidency Conclusions, which make no such distinction. Blumann, ‘Aspects institutionnels’ (1997) at 742 sees the difference between the European Council and the ‘Council, meeting in the composition of the Heads of State or Government’ mainly in the voting modalities. However, in practice the two get confounded during European Council meetings and decisions taken by the latter are adopted by consensus. ‘Der Europäische Rat’ (1994) at 26.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

222

Moreover, the Protocol on the location of the seats of the institutions and of certain bodies and departments of the European Communities and of EUROPOL adopted in Amsterdam lays down that ‘the Council shall have its seat in Brussels. During the months of April, June and October, the Council shall hold its meetings in Luxembourg’. Nevertheless, the European Council never followed that schedule of meetings. In fact, the Nice IGC decided that from the moment the Union would count eighteen Member States all European Council meetings will be held in Brussels. 46 Viewing the European Council as a specialised Council thus would imply a violation of primary law and—as Parliament has found out—the law on the seats of the institutions is no matter to be ignored lightly.47 Its preparation. M. Monnet had conceived the European Council as a gathering of the heads of state and government in a small, informal setting. Already during the Copenhagen meeting in 1974 this concept failed: ‘les chefs d’État et de gouvernement perdent le contrôle de leur réunion qui redevient une conférence internationale classique envahie par les experts’.48 In the words of Article 4 TEU, the original role of the European Council consists of providing ‘the Union with the necessary impetus for its development’ and to ‘define the general political guidelines thereof’. Since most Member States were weary of a new body taking the place of the Community’s institutions, there was to be no formal secretariat, no rules of procedure and no agenda. The function of the latter was taken over by the president’s letter.49 The London European Council of 29 and 30 June 1977 put further flesh on the bones of the European Council by deciding that it should have three types of discussions. These were, in short, informal exchanges ‘held in the greatest privacy and not designed to lead to formal decisions or public statements’, discussions designed to produce policy effects, and discussions aimed to settle issues outstanding from discussions ‘at a lower level’. A statement (to be prepared in advance) would be issued at the end of each meeting. The discussions aimed at reaching decisions or statements would be reflected in the Presidency Conclusions (issued under its authority) but no record would be kept of the informal discussions. 50 The Stockholm European Council decided that its annual spring meeting should be ‘the focal point for an annual review of economic and social questions’.51 In its following meeting the European Council added sustainable development to these themes.52 In the early days of the European Council, the Council presidency was charged with its preparation.53 However, soon this role was taken over by the General Affairs Council (now styled ‘General Affairs and External Relations Council’ or GAERC). Of course, this Council must itself be prepared by someone. The emphasis moved here from the Commission to the Council presidency. Of course, it would be a gross 46 47 48 49 50 51 52 53

Declaration 22 on the venue for European Councils, attached to the Treaty of Nice, OJ (2001) C80/85. It is for this reason that I disagree with Blumann, ‘Les institutions de l’Union dans le cadre du Traité établissant une Constitution pour l’Europe’ (2005) at 350, who argues that the European Council can still meet as a Council formation. Monnet, Mémoires (1976) at 600. Council Guide—Part I: The Presidency Handbook (2001) at 38. European Council of London of 29 and 30 June 1977 at section II. Presidency Conclusions of the Stockholm European Council of 23 and 24 March 2001, § 3. Presidency Conclusions of the Göteborg European Council of 15 and 16 June 2001, §§ 24-25. Hummer and Obwexer, ‘Die “EU-Präsidentschaft”—Entwicklung, Rechtsgrundlagen, Funktionen und Aufgaben sowie künftige Ausgestaltung der “Präsidentschaft” des Rates der Europäischen Union’ (1999) at 417.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

223

simplification to conclude that presidencies are fully in control of the agenda. Rather, each presidency is confronted with the legacy of its predecessors and the wishes of its peers as well as the Commission’s initiatives. On top of this, the events of the day may totally re-focus even the best-prepared presidency. In 1975 the Villa Marlia procedure was adopted to enhance the General Affairs Council’s co-ordinating role, but in practice this did not lead to much.54 Later efforts in this direction were similarly difficult.55 This lacklustre result is not the consequence of ‘laziness’56 or incompetence of the General Affairs Council, but rather something that must necessarily happen under the pre-Seville set-up of the institution. One of the reasons for this was the de facto role that the European Council before its Seville meeting exercised in the legislative process (which role was criticised by Parliament).57 Legally, the European Council can occasionally itself act as legislator, but more often it will ‘guide’ the legislative process.58 The European Council was originally not intended to take formal decisions and indeed, it generally left these to the Council.59 There have been a few exceptions: for example, the European Council decided to renegotiate the United Kingdom’s accession conditions, to found a passport union and to elect Parliament directly.60 Although not formally obliged, the Council and Commission are under strong political pressure to execute the decisions of the European Council. The Seville European Council intended to enhance the co-ordinating role of the GAERC in several ways.61 The GAERC now channels the work of the other Councils and prepares the European Council’s agenda. This draft agenda, drawn up at the latest four weeks before the European Council meeting, classifies items into four categories:  items to be approved or endorsed without debate;  items for discussion with a view to the definition of general political guidelines;  items for discussion with a view to the adoption of decisions in the context of enlargement and in exceptional cases. In the case of the previous and this bullet point, the Presidency must prepare an outline paper setting out possible policy options;  items for discussion but not intended to be the subject of conclusions.62 The GAERC’s monopoly on preparing the draft agenda further reduces the Commission’s influence on the process, since the Council does not have to act on the basis of a Commission proposal. 54 55 56 57 58

59 60 61 62

Note introductive du Secrétariat général au COREPER/Conseil—Méthodes de travail du Conseil, Council doc. 10297/98 of 14 July 1998 at 1. Westlake and Galloway, The Council of the European Union (2004) at 53. In 1976 Blumann, ‘Le Conseil européen’ (1976) at 10 criticised a ‘certaine paresse’ of the Council in this respect. See e.g. Parliament’s Resolution on the reform of the Council, OJ (2002) CE112/317 at J and at 6. As one example, the Madrid European Council adopted the name ‘euro’ for the common currency in the disguise of an ‘agreed and definitive interpretation of the relevant Treaty provisions’: Section A.I.A.1 of the Presidency Conclusions of the Madrid European Council of 15 and 16 December 1995. Martenczuk, op. cit. footnote 43 at 168 has some difficulties reconciling this decision with the role of the Court of Justice in the Union’s institutional system, especially since the European Council is not the author of the Treaties. Writing well before the Single European Act, Ipsen was of the view that the right to decide included the practice whereby ‘Rechtsverformung’ of decisions was left to the Council: ‘Das ist seine Praxis der “Umsetzung”‘: ‘Marginalien zum Europäischen Rat’ (1981) at 329. Lauwaars, op. cit. footnote 32 at 33 et seq. These changes were subsequently reflected in Article 2 of the Council’s Rules of Procedure: OJ (2006) L285/47. Presidency Conclusions of the European Council of Seville of 21 and 22 June 2002, Annex I.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

224

The Seville Conclusions foresaw the adoption by the European Council of a three-year ‘multi-annual strategic programme’, leading to an ‘annual operating programme’ submitted by the two presidencies concerned. In 2006 the Council decided to replace this somewhat heavy-handed system with planning for 18 month periods, to be adopted based on a draft submitted by the three relevant presidencies.63 In practice, the Commission’s planning has taken second place to the European Council’s work. In 2002 the Commission presented its 2003 legislative programme as ‘an important contribution to the Council’s first annual operating programme as decided at the Seville European Council’,64 thereby bowing to the heads of state and government. The relationship between the European Council and the Commission has never been straight-forward. The presence of the latter at the ‘summits’ in the period before 1973 was accepted only gradually and hesitatingly.65 Although the Commission’s president is a member of the European Council, he does not wield overwhelming influence in it. Nowadays—-apart from those cases where it has powers under the Treaties66— the European Council generally will ‘invite’ the Commission to submit proposals. Precisely because of the Commission president’s membership of the European Council, in practice the invitation almost equals an order and, even though the Commission’s right of initiative is untouched in law, its hands are bound politically. In the 1970s the view was expressed that the creation of the European Council would strengthen the Commission, leading it ‘to find new ways to play its traditional role’.67 Neither theory nor past practice bears out this conclusion. At the time, the Commission had lost the initiative in any event,68 up to the point were one observer called it caustically ‘a research office of the European Council’.69 In the 1980s it regained its importance as the policy motor of the Communities but, since the early 1990s, even though its legal prerogatives are largely still in tact, it has been unable to maintain its aura of the Delors period. Consequently, the political initiative has shifted back to the European Council and—by extension—to the Council. There are two problems with the current arrangements from a point of view of checks and balances. To begin with, the concurrent policy planning processes in Commission, Parliament and (European) Council lead to a less than transparent situation: who has the final say over legislative planning? 63 64

65 66 67

68 69

‘I’ item note from the Antici Group to COREPER—Amendment of the Council’s Rules of Procedure—18-month programme, Council doc. 7365/06 of 14 March 2006. Mr Prodi during the Parliamentary debate of 20 November 2002. Parliament criticised him for it: Resolution on the Commission’s legislative and work programme for 2003, OJ (2004) CE27/146, § 10. Blumann, op. cit. footnote 56 at 11. One such important exception is Article 128(2) TEC, which requires the Council to adopt employment policy guidelines on a proposal from the Commission ‘[o]n the basis of the conclusions of the European Council’. Report of the Committee of Wise Men (Messrs B. Biesheuvel, E. Dell and R. Marjolin) of October 1979 at 21. Similarly, Mr Tindemans was assured that the presence of the Commission would guarantee that the European Council, when deciding on EC matters, would act in accordance with the ‘form and procedures’ of the Treaties: Mr Tindemans in his report of 29 December 1975, Bull. EC Suppl. 1/76 at V.B. Blumann, op. cit. footnote 56 at 11. The Commission admitted as much: report on European Union, Bull. EC Suppl. 5/75, § 122. Mr Tindemans arrived at a similar conclusion in his report: Bull. EC Suppl. 1/76 at V.B. Constantinesco, ‘Die Institutionen der Gemeinschaft an der Schwelle der 80er Jahren’ (1981) at 238.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

225

Additionally, as far as the Commission’s legislative planning is concerned, its political70 responsibility towards Parliament might become an issue, since the Commission can easily disclaim responsibility by hiding behind the European Council. In the Interinstitutional agreement on better law-making71 the Commission, Parliament and Council attempted to co-ordinate the different policy planning processes by adding yet another structure. At the time of writing, the precise format remains to be decided. 9.2.3

Exceptions to the Commission’s right of initiative

Primary law comprises several real as well as quasi-exceptions to the Commission’s right of initiative. The real exceptions consist firstly of those procedures where either another institution or a Member State has an exclusive or concurrent right of initiative. For example, in the procedures of Articles 190(4) and 195(4) TEC72 the right of initiative is Parliament’s. Article 107 TEC,73 which concerns the ESCB Statutes, allows the ECB a concurrent right of initiative. Article 67(1) TEC allowed, during a transition period,74 Member States to submit initiatives. A similar right continues to exist in Article 34 TEU, whilst Article III-264 Const (legislation in the framework of the area of freedom, security and justice) allows a right of initiative to a quarter of the Member States.75 A second, real (but at the time of writing, in volume limited), challenge to the Commission’s right of initiative is caused by the prerogatives of certain statutory agencies (especially the European Aviation Safety Agency). These agencies are discussed in section 13.4, below. A third exception, proposed in Article I-47.4 Const, would have provided for the possibility that one million citizens who are nationals of ‘a significant number’ of Member States could invite the Commission to submit a proposal. Blumann76 considers it an ‘[h]eureuse innovation’ which may lead to the creation of a ‘conscience collective’ in Europe. This may be overoptimistic. The provision begs the question, whether such an initiative would bind the Commission. On the one hand, one can hardly expect the Commission to take the political responsibility for a proposal that it may not support. On the other hand, the word ‘appropriate’ and the one-million threshold do suggest some sort of legal consequence. Any citizen can ‘invite’ the Commission to take an initiative and, indeed, such requests are frequently made in the form of petitions. Does it then matter whether 999,999 citizens or 1,000,001 sign the petition?77 Perhaps the provision’s primary value does not lie so

70 71 72 73 74 75 76 77

As well as legal, since European Council are not assailable under Article 230 TEC: case C-253/94P, Roujansky vs Council, [1995] ECR I-7, § 11. The wording of Article III-365.1 seems to change at least this latter issue. OJ (2003) C321/1, corr. OJ (2004) C4/7. III-330.2 and III-335.4 Const. III-187 Const. I will not further discuss the Member States’ right of initiative under Article 67 TEC, which right expired in 2004. Article I-26.2 Const somewhat superfluously recognises these exceptions. Op. cit. footnote 47 at 368-369. For similar criticisms, Dougan, op. cit. footnote 3 at 773-774 and Schwarze, ‘Ein pragmatischer Verfassungsentwurf—Analyse und Bewertung des vom Europäischen Verfassungskonvent vorgelegten Entwurfs ines Vertags über eine Verfassung für Europa’ (2003) at 557.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

226

much in providing any tangible influence to citizens, but rather in helping them to overcome a ‘feeling of powerlessness’ vis-à-vis the European institutions.78 The fourth exception is formed by the procedure of Article 138 TEC79 discussed in section 12.4, below. The fifth type of alleged exceptions concern general provisions in which the Commission can be requested to submit a proposal. These concern Article 208 TEC and Article 192(2) TEC.80 Article 208 TEC 81 provides that ‘[t]he Council may request the Commission to undertake any studies the Council considers desirable for the attainment of the common objectives, and to submit to it any appropriate proposals.’ Opinion and the institutions have been divided on the precise scope and meaning of this provision. The main question here is whether the request includes a right to an answer. More specifically, can the Council request a specific initiative? And is the Commission entitled not to present an initiative? In 1962 the Council requested a proposal from the Commission concerning the applicability of competition rules to transport policy, a matter on which the institutions disagreed. Apparently, partially under influence of (then) Article 152 TEEC, the Commission conceded to the Council’s pressure.82 In later instances, the Commission considered Article 152 to be binding (without indicating whether it was legally or politically binding).83 However, it appears that it now takes the opposite view and seems to instruct its staff that, whilst there is no legal obligation, such requests create a certain legal pressure. Commission policy tends to be not to enter into commitments or accept language that could be interpreted as an obligation. The policy furthermore requires Commission departments to tone down the scope of any specific request to the Commission included in a legislative instrument, for example by including wording such as ‘where appropriate’.84 If the request is contained in Council conclusions or resolutions, the Commission must include a statement to the effect that it takes note of the request and reserves the right to respond in accordance with the rules of the Treaty.85 78 79 80

81 82 83 84

85

In this sense Peters, ‘European democracy after the 2003 Convention’ (2004) at 45. More positive on the idea Laurent, ‘Le droit d’initiative citoyenne en attendant l’entrée en vigueur de la Constitution européenne . . .’ (2006), 221 et seq. Article III-211 Const. Very occasionally, secondary legislation requires the Commission to propose legislation. See e.g. Article 4 of Directive 1999/96/EC of the European Parliament and of the Council of 13 December 1999 on the approximation of the laws of the Member States relating to measures to be taken against the emission of gaseous and particulate pollutants from compression ignition engines for use in vehicles, and the emission of gaseous pollutants from positive ignition engines fuelled with natural gas or liquified petroleum gas for use in vehicles and amending Council Directive 88/77/EEC, OJ (2000) L44/1. I will not further discuss this cas de figure. Article III-345 Const uses a slightly different wording. Kapteyn and VerLoren van Themaat (ed. Gormley), Introduction to the law of the European Communities (1998) at 410. See the reply to written question 865/82 by Mr Radoux, OJ (1982) C298/9. For example in Article 17(2) of Regulation (EC) 1049/2001 of the European Parliament and of the Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, OJ (2001) L145/43: ‘the Commission . . . shall make recommendations, including, if appropriate, proposals for the revision of this Regulation and an action programme of measures to be taken by the institutions’. Interviews with Commission staff.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

227

The Council (or at least, some presidencies) still seem to interpret Article 208 TEC as a rather direct possibility to order an initiative and to indicate the lines this should follow.86 One view argues that this provision shows there is no ‘institutional equality’ between Commission and Council.87 Alternatively, the argument is made that Article 208 is a corrective exception to the general principle of the Commission’s right of initiative aimed to deal with inertia on the side of the executive.88 However, the best view seems to be that the provision must be interpreted narrowly. A compulsive interpretation of the provision would demote the right of initiative to a mere formality once the Council can reach agreement on the line to be followed.89 Consequently, the Council’s right to request does not imply a duty for the Commission to provide a proposal, let alone provide a specific one.90 Any different view would impair not only the Commission’s rights, but also the MEPs’ prerogatives. On the other hand it does not seem to be necessary (at least in the Commission’s view) for the Council to explicitly invoke Article 208.91 Article III-345 Const would add to current law the rule that the Council votes by simple majority and that—similar to Article 192(2) TEC—the Commission must provide reasons for not submitting a proposal. That addition would make it very evident that there is no obligation to submit a proposal. Can the Commission simply ignore the request under Article 208 TEC? Apart from the question whether such a course is politically wise, a cogent argument has been made it would also be difficult to reconcile with the duty of co-operation between Council and Commission laid down in Article 218(1) TEC92 and—in the logic of the Court of Justice—Article 10 TEC. In this view, to which I subscribe, the Commission is under a duty to examine the request in good faith.93 In practice, Article 208 does not pose many problems.94 The Council only rarely explicitly refers to it,95 largely because of the preponderant role of the European Council in policy initiation.96 Nevertheless, the political reality is such that the Commission will rarely refuse a proposal and often will be forced to conform to the political constellation in the Council.97

86

87 88 89 90 91 92 93 94 95 96 97

As Lenaerts notes in Constantinesco a.o., Traité instituant la CEE—Commentaire article par article (1992) at 902, the German text of the provision provides some support for this: the term ‘auffordern’ would seem to be more imperative than the English ‘may request’ or the French ‘demander’. I agree, however, that the majority of the language versions use rather neutral language. Oppermann and Kilian, ‘Vergangenheit und Zukunft des Europäischen Parlaments’ (1981) at 390. In this sense e.g. Constantinesco, Compétences et pouvoirs dans les Communautés Européennes (1974) at 316. Kapteyn and VerLoren van Themaat (ed. Gormley), op. cit. footnote 82 at 410. In this sense Lenaerts, ‘Some reflections on the separation of powers in the European Community’ (1991) at 22, Noel, ‘The Commission’s power of initiative’ (1973) at 126 and Verhoeven, The European Union in search of a democratic and constitutional theory (2002) at 232. Reply to written question 865/82 by Mr Radoux to the Commission, OJ (1982) C298/9. Lenaerts and Van Nuffel (Bray, ed.), Constitutional law of the European Union (2005) at 582. Jacqué, Droit institutionnel communautaire (1997) at 136. Ibid. at 137. Constantinesco a.o., op. cit. footnote 86 at 904. Westlake and Galloway, op. cit. footnote 55 at 34. The issues relating to the exclusivity of the Commission’s right of initiative in the context of implementing legislation (comitology) are discussed in Chapter 13.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

228

At several other places, the Treaties provide the Council with a right to request the Commission to submit a proposal. In some of these cases also the Member States are given the right to request the Commission.98 Article 192(2) TEC.99 Parliament begged for a long time100 for a prerogative equal to that of Article 208 TEC,101 either through interinstitutional agreement102 (which option for a long time was disfavoured by the Commission) or through Treaty amendment.103 In practice, the MEPs could of course always request the Commission to come forth with a proposal but the latter was not prepared to go beyond recognising ‘the utmost political importance to the initiatives put forward by Parliament’, whilst keeping its freedom of action unfettered: ‘The Commission is prepared to use these initiatives as a basis for its work provided that it has no substantial objections to them. Otherwise, it shall submit detailed reasons for its attitude to Parliament within a reasonable period’.104 That latter commitment was taken up and found its way into paragraph 3.3 of the 1995 Code of conduct, which added that ‘Commission decisions on such requests shall be duly reasoned on a case-by-case basis, if necessary even in a sitting of Parliament’.105 The 2000 Framework Agreement repeats the obligation in different words and specifies in addition that the Commission may have to provide a ‘sufficiently detailed reply to any such request within the relevant parliamentary committee’.106 The 2003 Interinstitutional agreement on better law-making107 added that the Commission will reply ‘rapidly and appropriately’. Paragraph 14 of the 2005 Framework Agreement in essence repeats the existing commitments.

98 E.g. Articles 14(4) TEU, 11(1), 67(2), 95(8) and 185 TEC. 99 Article III-332 Const uses a somewhat different wording. 100 See e.g. its Resolution on the right of legislative initiative and on the role of the Parliament in the legislative process of the Community, OJ (1981) C234/64. The Commission contemplated it as early as 1975: Report on European Union, Bull. EC Suppl. 5/75, § 115. In later years, the Commission has been less willing to concede on this point. 101 It was promised a full right of initiative in the draft Treaty of 10 March 1953 establishing the European Political Community (Article 23(1)), but this text sunk with the European Defence Community. The Vedel Committee brought it back from oblivion only to reject it: Vedel report of 25 March 1972, Bull CE Suppl. 4/1972, Part I at section II § 5. Contrariwise, Mr Tindemans in his report of 29 December 1975 (Bull. EC Suppl. 1/76 at V.A.1) thought such a step desirable. It has been on Parliament’s agenda ever since. 102 Resolution on the right of legislative initiative and on the role of the European Parliament in the legislative process of the Community, OJ (1981) C234/64. In response, the Commission indicated its preparedness ‘to be inspired’ by such Parliamentary initiatives: Les relations entre les institutions de la Communauté, Commission doc. COM (81) 581 final of 7 October 1981 at 11. This, of course, implies that the Commission retained its own decision-making capacity in this regard. It had already made that last point in its reply to written question 247/79 by Mr Caillavet, OJ (1979) C253/8. 103 See e.g. Article 37(2) of the Draft Treaty establishing the European Union adopted by Parliament on 14 February 1984. 104 Joint Declaration by the Enlarged Bureau of the European Parliament and the Commission on relations between the two institutions, Parliament doc. 1-328/84/rev of 21 May 1984 at § 5. 105 Resolution on the Commission’s annual programme of work, OJ (1995) C89/69. 106 Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121-122, § 4. 107 OJ (2003) C321/1, corr. OJ (2004) C4/7.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

229

Meanwhile, the Maastricht Treaty inserted the current Article 192, second paragraph: ‘The European Parliament may, acting by a majority of its Members, request the Commission to submit any appropriate proposal on matters on which it considers that a Community act is required for the purpose of implementing this Treaty.’ This provision required an absolute majority and, in that sense, was a retrograde step for Parliament.108 Moreover, the right to request studies is excluded (although in practice, in the context of the parliamentary system governing relations between Commission and Parliament this is unlikely to be a big problem in practice). Furthermore, proposals are limited to those ‘required for the purpose of implementing this Treaty’ (as opposed to those ‘desirable for the attainment of the common objectives’ as per Article 208 TEC). In practice, however, the difference in wording between the two provisions does not reflect a difference in scope. In either case, the requesting institution judges how the request should be framed, and the Commission whether it can be accepted. Parliament provided in its Rules of Procedure for a possibility to fix a deadline for a request pursuant to Article 192. It furthermore laid down that inter alia such a request ‘shall indicate the appropriate legal basis and be accompanied by detailed recommendations as to the content of the required proposals, which shall respect fundamental rights and the principle of subsidiarity’ (Rule 39). Of course, these rules could by their nature only bind Parliament internally, but not the Commission and even less the Council.109 The question arises to what extent Article 192 obliges the Commission in law to submit a proposal. On the one hand, a case can be made that such an obligation indeed exists. It could be argued that a mere right to request without a right to be heard would lose much of its value. Moreover, Article 192 requires an absolute majority of the MEPs to vote for the request, suggesting that the request cannot be simply compared with an ordinary resolution. Third, the scope of the request is defined in substance by the limitation that the request must concern ‘matters on which it considers that a Community act is required for the purpose of implementing the Treaty’, which presumes some consequence that sets it apart from requests outside the framework of Article 192. On the other hand, there are arguments pointing against an overrated importance of the provision and these seem more persuasive. An interpretation requiring the Commission to submit a proposal pursuant to an Article 192 request would amount to a Parliamentary right of initiative. Any such initiative provokes the question as to who will defend it in Council. Presumably this would be the Commission, which, however, may not necessarily support the proposal. Would a request be void if it concerned a matter not required for implementing the Treaty? It is unlikely that this would become a legal issue, since Article 192 108 In this sense Boest, ‘Ein langer Weg zur Demokratie in Europa—Die Beteiligungsrechte des Europäischen Parliaments bei der Rechtsetzung nach dem Vertrag über die Europäische Union’ (1992) at 184 and Sinnaeve, ‘De positie van het Europees Parlement na het Verdrag van Maastricht’ (1994) at 148. 109 This seems to be accepted by Parliament, which uses its newly obtained right sparingly: Maurer, (Co-)governing after Maastricht: the European Parliament’s institutional performance 1994-1999 (1999) at 30.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

230

clearly makes Parliament the judge of that. In practical terms, this is in any event unlikely to become a big problem. Clearly, Parliament could request any initiative without the Treaty specifically empowering it to do so. In practice, it has at times made quite elaborate requests. 110 The Commission is legally not obliged to respond with a proposal if it disagrees with the need for it. The interinstitutional convention discussed above considerably sharpens the Commission’s obligations. 111 The requirement to explain a refusal in Parliament may make it politically difficult in practice for the Commission to ignore a request. The Commission presumably will have more freedom where the request is very specific.112 Verhoeven113 concludes because of ‘the general obligation of inter-institutional loyalty’ that the Commission’s undertaking applies similarly to requests emanating from the Council. I am not sure whether one can read into the principle of loyal cooperation any obligation for the Commission to apply ‘equal treatment’ to Parliament and the Council. Current convention lays down in some detail how the Commission should respond to Parliament.114 It would seem to stretch the general obligation of inter-institutional loyalty to read such procedural requirements into it. On the other hand—and apart from the common sense for the Commission to do so—, if abstracted from interinstitutional convention, the principle of loyal co-operation may support a duty to provide some sort of reply to Council requests. However, the matter is better analysed in terms of effet utile. In order to render meaning to the terms of Article 208, the Commission could not callously ignore Article 208 requests from the Council, but will have to respond meaningfully. In summary, the Commission considers itself politically bound to requests made under Articles 192 and 208 TEC.115 In practice, Article 192 seems to be less of a threat to the Commission than it would appear to be on paper. The first legislation adopted pursuant to it was the Fourth Motor insurance Directive,116 for which the proposal was submitted some four years after the entry into force of the provision.117 Lastly, the drafters of Article III-332 Const widened the scope of the provision to all Union matters. Moreover, they proposed to enact the convention that the Commission, when it disfavours submitting a proposal, must explain the reasons to Parliament. Article III-345 extends this obligation to requests from the Council.

110 See e.g. its Resolution on the Commission communication on safe operation of mining activities: a follow-up to recent mining accidents, OJ (2002) CE65/382 and the Resolution on the Commission report on the application of the Directive on the establishment of a European works council or a procedure in Community-scale undertakings and Community-scale groups of undertakings for the purposes of informing and consulting employees (Council Directive 94/45/EC of 22 September 1994), OJ (2002) CE72/68. 111 In this sense White, ‘The European Parliament-Commission Framework Agreement’ (2000) at 333. 112 Parliament has at times made very detailed requests to the Commission, e.g. in its Resolution with recommendations to the Commission on a fifth European Parliament and Council directive on motor vehicle liability insurance to improve the legal protection of accident victims, OJ (2002) CE65/46. 113 Op. cit. footnote 90 at 232. 114 2005 Framework Agreement, § 14. 115 Ponzano, op. cit. footnote 9 at 40. 116 Activity report (1 May 1999 to 31 July 2000) of the delegations to the Conciliation Committee, Parliament doc. PE 418.584 (no date) at 5. 117 OJ (1997) C343/11.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

9.2.4

231

A brief evaluation

Although the Commission’s right of initiative is intact in law (save for the few real exceptions discussed above), the different layers of policy planning have to a large extent eroded it in practice. The right of initiative may still serve as a check that the Commission has on the other institutions and the value of this should not be underestimated. However, this prerogative no longer goes hand in hand with the considerable influence on policy setting that the Commission enjoyed in the 1980s. The discussion of the Commission’s right of initiative demonstrates three points. To begin with, it is very much possible for a Treaty prerogative to be bound by convention. In other words, whilst none of the conventions on policy planning or the right of initiative contravene the Commission’s prerogative, they steer its use into a certain direction. In that sense, interinstitutional convention has provided a very different flavour to the text of the Treaties. The second point concerns the impact on the separation between the institutions. This silken web of obligation is woven by, on the one hand, Parliament and, on the other, the European Council and Council. Especially where policy planning is concerned, this leaves the Commission with very limited room to develop its policy planning. One could argue that, perhaps, the point where the Commission’s policy setting is checked by the other institutions has long been passed. Rather, the Commission can at best hope to use (or rather, not use) its right of initiative to check the policy setting by Parliament and the Member State representatives. The concurrent ambitions of the other institutions in a way balance them against each other. As a final point, it is to be regretted that all of this concurrent planning activity does not improve the institutional transparency. The Interinstitutional agreement on better law-making118 was partly intended to improve on this state of affairs, but it is not yet clear whether it will in practice. 9.3

Voting rules and conventions in the Council

9.3.1

Qualified majority voting in law and practice

Article 205 TEC lays down the basic voting rules for the Council. The default rule is a decision with a simple majority. This majority is used for example for procedural matters. However, in practice, most Treaty provisions prescribe qualified majority voting (QMV), and a vote with simple majority for non-procedural issues is quite rare.119 In the original scheme of the EEC Treaty, the Commission’s quasimonopolistic right of initiative was coupled with its right to amend its proposals.120 This would have been combined with QMV so as to give the executive a negotiating edge, especially since the Council’s right to amend Commission proposals is not unlimited. 118 OJ (2003) C321/1, corr. OJ (2004) C4/7. 119 One such case is the imposition or suspension of anti-dumping duties under Articles 9(4), 12(3), 13(3) and 14(4) of Council Regulation 384/96 on protection against dumped imports from countries not members of the European Community, OJ (1996) L56/1. 120 In 1990 the executive instructed its officials to ‘veiller à ce que . . . les groupes du Conseil travaillent dans la perspective du vote [et] la Commission soit saisie lorsqu’une modification de sa proposition initiale permet d’obtenir une majorité. En cas de problème les services concernés alerteront le secrétariat général’. The Commission checks systematically whether the Council Secretariat General has correctly indicated whether a proposal can be subject of a majority vote. For these rules and further details see the Commission’s Vade-mecum institutionnel (1990) at 53-54.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

232

Under QMV as it operates in the EC Treaty following the 2004 accessions, each Member State has a number of votes, totalling 321. A decision is taken if at least 232 votes are cast in favour by a majority of states when it is based on a Commission proposal. If the decision is not taken on the basis of a Commission proposal, 232 votes plus at least two-thirds of the Member States must be in favour. In addition, since the entry into force of the Nice Treaty, any Member State may request verification that the qualified majority represents at least 62 per cent of the EU’s total population. In the field of CFSP, Article 34(2) TEU uses the term ‘initiative’ rather than ‘proposal’ of the Commission. It follows that, for such initiatives, the two-thirds majority rule applies. Article 34(3) TEU makes this clear. The practice of QMV, especially before the Nice Treaty entered into force, differed from the letter of the Treaties for essentially two reasons. On the one hand, the organisation of the Council promotes a drive for consensus. The agenda of the Council is divided into a list of ‘A’ items for which approval by the Council is possible without discussion, and a part ‘B’ covering the remainder.121 This practice serves the Council’s efficiency well and it is hard to see how the institution in its current form could be organised differently, but it propels working groups to seek unanimity wherever possible. 122 Arguably, this natural tendency to acquire as broad a basis as possible for a decision is not, in itself, a bad thing. Indeed, from a perspective of output legitimacy it is quite defendable. Notwithstanding this, the MEPs have accused the Commission of tending ‘not to submit a Commission proposal to the Council before it has been the subject of unanimous agreement among the members of the competent Council working party, regardless of how long this takes’.123 This seems a somewhat exaggerated view of reality. On the other hand, conventional rules within the Council play a role. The system in the original Treaties was in its original form largely frustrated by what became popularly known as the ‘Luxembourg Compromise’. Although this text no longer applies, it had a pervasive effect on the voting culture within that institution and that justifies an analysis below (sub-section 9.3.2). The remaining relevant conventions within the Council are discussed in subsection 9.3.3. 9.3.2

The Luxembourg Compromise Its background

The so-called Luxembourg compromise itself may be largely history but some discussion of it is necessary for two reasons. Firstly, it led to convention that is still applied. Secondly, the peculiar nature of the ‘Compromise’ makes it interesting from the point of view of interinstitutional convention.

121 Article 3(6) of the Council’s Rules of Procedure. 122 For an interesting quantitative analyse of the work of the Agricultural Council see Van Schendelen, ‘”De Raad beslist”. Beslist de Raad?’ (1995) at 711-715. 123 Resolution on the decision-making process in the Council in an enlarged Europe, OJ (1999) C150/353, § 28.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

233

The background of the matter can be described briefly.124 Already before the entry into force of the EEC Treaty, the High Authority often sought to obtain unanimity in the Council when a qualified majority would have sufficed,125 presumably in order to prop up its tender political legitimacy by consensual support from the Member States. This practice was copied by the early EEC Commission:126 in the period 1958-1965 the Council took about a hundred decisions based on provisions foreseeing majority votes. However, in only some ten of them a vote was actually taken.127 The practice of unanimity that arose from the constitutional crisis of 1965 had quite a different flavour to it. Even before that crisis, many decisions had to be taken by QMV.128 Matters came to a head when the time approached when, according to the EEC Treaty, the scope of QMV would be increased. Although the direct cause for the crisis was a package of proposals tabled by the EEC Commission in March 1965 aimed at, inter alia, introducing the direct financing of the Communities and granting some budgetary powers to Parliament,129 on the background French resistance to QMV played a major part.130 The French government stood not entirely alone in its opposition on this point: in 1964 the German government, under electoral pressures, had requested a decision on wheat prices to be delayed until 1967. Germany obtained as a concession from France that the cereal price would be determined with unanimity131—a ‘concession’ De Gaulle’s government was probably most willing to grant. In the view of the Council, this political agreement had no legal value.132 The French resistance led to the ‘policy of the empty chair’, forcing the Council to operate through the written procedure, which requires the unanimous approval of all Council members. After seven months of deadlock and stagnation, a compromise was reached in January 1966133 under which the Member States agreed to attempt to reach unanimity. France continued to disagree with the remaining five what should happen if such unanimity remained elusive after ‘un délai raisonnable’. In the following years, this ‘délai raisonnable’ was interpreted as longer and longer. No Member State—not even those opposing the Luxembourg Compromise— chose to force a vote in the face of the opposition of one of the Council Members. The unanimity practice extended throughout Community policy making134 and soon the 124 There is an ocean of literature on the matter. In addition to the works mentioned below, I usefully consulted Jacqué, op. cit. footnote 93 at 150-153; Teasdale, ‘The Luxembourg Compromise’ (1999) at 104-111. On the historical background see Pryce, ‘Historical development of the EC’ (1980) at 52. 125 Kranz, ‘Le vote dans la pratique du Conseil des ministres des Communautés européennes’ (1982) at 405. 126 Lagrange, ‘Le pouvoir de decision dans les Communautés Européennes: théorie et réalité’ (1967) at 25. 127 Kranz, op. cit. footnote 125 at 407. 128 Ophüls, ‘Die Mehrheitsbeschlüsse der Räte in den Europäischen Gemeinschaften’ (1966) at 193. 129 Bull CE 5/65. 130 See Gen. De Gaulle’s comments during his press conference of 9 September 1965, reported in Dalloz, Textes de politique étrangère de la France (1989) at 107. 131 Kranz, op. cit. footnote 125 at 408-409. 132 Reply to written questions 47/65 and 115/66 by Mr Vredeling to the Council, OJ (1966) 75/1121. 133 Conclusions of the extraordinary session of the Council in Luxembourg on 17, 18, 27 and 28 January 1966, Bull CE 3/66 at 5 et seq. 134 The budget was always an exception where majority voting was used. The reason was that the French government (at least initially) did not consider it very important: Kranz, op. cit. footnote 125 at 427.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

234

Communities experienced the truth of Hamilton’s warning that the operation of unanimity is to ‘embarrass the administration’ and ‘to destroy the energy of the government’.135 There was, however, an additional pernicious effect. The Commission could do little more than sadly confirm that the cancer had spread to all Member States ‘concerning almost all subjects at all levels of decision making’136 and that it had eroded the rule that amendments to Commission proposals required unanimity.137 Of the first ten EC Members only Italy and the Benelux countries never invoked their ‘right’ of veto.138 When Denmark ratified the Single European Act, it laid down that the ‘Luxembourg Compromise’ would remain in force.139 This statement may have affected Danish constitutional law, but had no legal consequences for EC law. As far as this author is aware, Germany invoked it formally once, when vetoing a Commission proposal to decrease wheat prices in 1985.140 The ‘Compromise’ did most of its pernicious work in preventing majority voting. Ehlermann141 observed in the early 1980s that its consequences were more of a psychological nature than of direct mechanical impact. It may be added that every enlargement of the EC directly compounded to the problem of reaching unanimity in the larger group of Member States. This in turn had a direct effect on the possibility of the Commission to promote its own policy line. In the 1980s, two factors contributed to a rollback.142 Firstly, in 1982 the United Kingdom attempted to block the adopting of agricultural prices in an effort to obtain satisfaction on another issue (financial compensation). The other Member States refused to accept this, effectively establishing a rule that the invoked vital interest must be connected to the item under discussion. 143 One year later, when the Stuttgart European Council adopted the (on this point rather bland)144 Solemn Declaration on European Union, some Member States indicated that the national interest needed to be ‘directly related to the subject under discussion’ and confirmed ‘in writing’.145 This indeed became practice. 135 Hamilton in ‘Federalist Paper 22’, in Madison, Hamilton and Jay (‘Publius’), The Federalist Papers (1987) at 180. For a very comprehensive analysis of the effects of the ‘Luxembourg Compromise’ see Constantinesco, op. cit. footnote 69, espec. at 217 et seq. 136 Les relations entre les institutions de la Communauté, Commission doc. COM (81) 581 final of 7 October 1981, § 6. The European Council of Paris of 9 and 10 December 1974 had implicitly already come to the same conclusion: Communiqué, Bull. EC (1974) 12, 8, § 6. 137 Commission communication The institutional system of the Community—restoring the balance, Bull. EC Suppl. 3/82 at 7-8. 138 The other Member States invoking it did so on the basis of the principle of reciprocity: Rapport van de werkgroep ad hoc voor de behandeling van het vraagstuk van de uitbreiding van de bevoegdheden van het Europese Parlement (Vedel report), Bull. EC Suppl. 4/72 at 27. 139 Vogelaar, ‘De Europese Akte’ (1986) at 79. 140 Vasey, ‘Decision-making in the Agriculture Council and the “Luxembourg Compromise”‘ (1988) at 727. 141 Ehlermann, ‘Das schwierige Geschäft der Kommission’ (1981) at 346. 142 Commission’s Vade-mecum institutionnel (1990) at 53. 143 Nugent, The government and politics of the European Union (2003) at 172 and Vasey, op. cit. footnote 140 at 731-732. 144 Paragraph 2.2.2 of the Solemn Declaration on European Union: ‘The application of the decisionmaking procedures laid down in the Treaties of Paris and Rome is of vital importance in order to improve the European Communities’ capacity to act. Within the Council every possible means of facilitating the decision-making process will be used, including, in cases where unanimity is required, the possibility of abstaining from voting.’ 145 In this sense the declarations of France and Ireland. Britain and Denmark maintained the other view that ‘when a Member State considers its very important interests to be at stake, discussion should be continued until unanimous agreement is reached’ without qualifying those interests. Greece similarly did not mention any linkage between the national interest and the item under discussion

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

235

Secondly, even though a number of governments at the time stated otherwise,146 the Single European Act proved to be the second turning point for the veto practice, for several reasons. To begin with, the co-operation procedure introduced by it contains a number of explicit ‘safety valves’, making it politically difficult for a Member State to insist on an additional right of veto. In addition, the Single European Act foresaw the possibility of national derogations, which made a veto less necessary for Member States. 147 Lastly, the internal market program embodied in the Single European Act required more efficiency than the unanimous voting machine could offer and the need for more efficient decision making procedures became stronger felt. These factors led the Council on 20 July 1987 to change its Rules of Procedure.148 Under the new system, the president is obliged to open voting proceedings ‘on the invitation of a member of the Council or of the Commission, provided that a majority of the Council’s members so decides’.149 The direct result of this change was that no single state was capable anymore to hold the Council hostage over what it claimed to be its vital interests. Most authors would agree that the ‘Luxembourg Compromise’ is (almost) dead and buried.150 Available information on voting patterns, whilst not conclusively proving this, at least suggests so.151 However, this view is not universally accepted152 and the Council itself has been ambivalent about it. The ministers noted in 2001 that ‘its status will remain the same after the entry into force of the Nice Treaty’ without clarifying what that status is.153 In any event, when Poland tried to resuscitate it in 2006, the attempt met with very little support and was quickly dropped. Its relationship with interinstitutional convention The ‘Luxembourg Compromise’ was neither law nor convention but, instead, led to convention. The ‘Compromise’ was a Council text154 and hence not an international treaty.155 Indeed, it lacked even the most basic aspect of an international treaty, since agreement and the Benelux, Germany and Italy favoured voting when the Treaties so provide. 146 Vasey, op. cit. footnote 140 at 725. 147 Jacqué in Constantinesco a.o., op. cit. footnote 86 at 873. 148 According to Glaesner, ‘Das Verfahren der Zusammenarbeit (Art. 149 Abs. 2 EWGV)’ (1988) at 123, the change in the Rules of Procedure merely reflected changed practice. 149 Article 5(1) of the Amendment of the Council’s Rules of procedure adopted by the Council on 20 July 1987, on the basis of Article 5 of the Treaty of 8 April 1965 establishing a Single Council and a Single Commission of the European Communities, OJ (1987) L291/27. 150 Bradley, ‘Institutional design in the Treaty of Nice’ (2001) at 1108; Lenaerts and Van Nuffel (Bray, ed.), op. cit. footnote 92 at 420 and Westlake, The Council of the European Union (1999) at 97. Jacqué, op. cit. footnote 93 at 152-153 and Nugent, op. cit. footnote 143 at 171 argue that it has almost disappeared. 151 Reply to written question E-2893/01 by Mr Bonde to the Council, OJ (2002) CE134/136. 152 Implicitly (writing in 1990) Temple Lang, ‘Community constitutional law: Article 5 EEC Treaty’ (1990) at 662-663 and in the last six years (!) Schorkopf, ‘Die Untätigkeit des Rates der EU im Gesetzgebungsverfahren’ (2000) at 368 and Huber, ‘Das institutionelle Gleichgewicht zwischen Rat und Europäischem Parlament in der künftigen Verfassung für Europa’ (2003) at 579. 153 Written question E-1109/01 by Mr Bonde to the Council, OJ (2001) CE364/48. In a similar vein the reply to written question P-2887/01 by Mr Bonde to the Council, OJ (2002) CE134/134. 154 The text given to the press was an excerpt from the procès-verbal of the Council’s meeting of 17-18 and 28-29 January 1966: reply to written question 116 by Mr Vredeling to the Council, OJ (1966) 1124. 155 In favour of the international treaty thesis Klabbers, ‘Informal agreements before the European

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

236

went not much further than as to the existence of disagreement. A fortiori, it was not a Decision (‘Beschluß’) either, since at no point the EEC Treaty empowered the Council to bypass the Treaty voting rules. In the Court of Justice’ view, ‘the rules regarding the manner in which the Community institutions arrive at their Decisions are laid down in the Treaty and are not at the disposal of the Member States or of the institutions themselves’.156 It follows that the ‘Luxembourg Compromise’ could not have been legally binding.157 Some authors have argued that ‘Luxembourg Compromise’ was a convention158 but this view is problematic, too. As noted in sub-section 3.4.3 above, the prime and only requirement for convention to exist, is consent among the parties to it that there is a binding rule. In the case of the Luxembourg Compromise—at least for the longest period of its existence—this requirement was not fulfilled. The Member States agreed that when ‘des intérêts très importants d’un ou de plusieurs partenaires sont en jeu’, they should strive to reach unanimity. Still, this agreement went only as far as ‘un délai raisonnable’. Not all Member States shared the (originally) French position that the discussion should be continued until a unanimous agreement was reachedwhich effectively allowed each Member State a veto. Hence the prime condition for the coming into force of conventionconsentwas lacking even within the Council. A fortiori, the ‘Compromise’ did not constitute interinstitutional convention: both Parliament and the Commission vehemently opposed it.159 That said, it did lead to an unwritten convention among the Member States, requiring them to strive to consensus for principally all but administrative/budgetary decisions:160 ‘The fact that that Treaty provides in many cases for adoption by majority does not prevent Members of the Council from endeavouring, as a general rule, to narrow their differences before the Council votes’.161 Indeed, even though, since the 1987 amendment of the Council’s Rules of Procedure, Member States are less willing to accommodate individual countries up to the bitter end162 there remains a reluctance to isolate them.163 This element was both the

156 157 158 159 160

161 162 163

Court of Justice’ (1994) at 1000. In my view, this ignores the fact that the Treaties constitute a new legal order of law (in this sense case 26/62, Van Gend & Loos vs Netherlands Inland Revenue Administration, [1963] ECR 1). Kranz, op. cit. footnote 125 at 410 similarly opposes the construction as an international agreement, simply because it was not an agreement. Case 68/86, United Kingdom vs Council, [1988] ECR 855, § 38. A similar conclusion is reached by Jacqué, ‘L’Acte unique européen’ (1986) at 588 and by Vasey, op. cit. footnote 140 at 725-726. Bradley, op. cit. footnote 150 at 1108 and Vasey, op. cit. footnote 146 at 731. It is in this sense that I understand Edward’s comment that ‘the so-called Luxembourg Compromise is, in a sense, no more than a reflection of political reality’: ‘The impact of the Single Act on the institutions’ (1987) at 25. Kranz, op. cit. footnote 125 at 417 and Vasey, op. cit. footnote 140 at 726. Parliament strove to get the ‘Compromise’ revoked or at least, to have recourse to it made more difficult: e.g. in its Resolution on relations between Parliament and the Council, OJ (1981) C234/52. In this sense (as far as the consequential behaviour is concerned) Constantinesco, op. cit footnote 69 at 217-218. This, of course, does not exclude the case where Member States prefer consensus as a trade off in package deals: Röhl, ‘Die Beteiligung der Bundesrepublik Deutschland an der Rechtsetzung im Ministerrat der Europäischen Union’ (1994) at 414-415. Reply to written question 317/96 by Mr Moorhouse to the Council, OJ (1996) C217/22. Corbett, ‘The Intergovernmental Conference on Political Union’ (1992) at 289. Wallace, The Council of the European Union—An observer’s view (1998) at 27 and Westlake and Galloway, op. cit. footnote 55 at 12. For some anecdotal figures see the reply to written question

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

237

contributing factor to, and the consequence of, the longevity of the ‘Luxembourg Compromise’. The Luxembourg Compromise is not quite extinct. It was enacted in the context of Articles 23(2) (CFSP) and 40(2) TEU (closer co-operation in the field of the third pillar) as well as 11(2) TEC (closer co-operation) in their Treaty of Amsterdam versions. The latter two provisions were revised a few years later in Nice and were never invoked.164 Article 23(2) provides that ‘[i]f a member of the Council declares that, for important and stated reasons of national policy, it intends to oppose the granting of an authorisation by qualified majority, a vote shall not be taken. The Council may, acting by a qualified majority, request that the matter be referred to the European Council, for decision by unanimity.’ This ‘Luxembourg Compromise-ish’165 formula differs from the original Compromise in several respects. Firstly, the point has been made that ‘important and stated reasons of national policy’ implies a difference with the ‘intérêts très importants’ from the 1966 text. Arguably, the national interest is by definition for a Member State to determine, whilst the degree to which national policy is affected can be a matter of debate between Member States.166 Secondly, Article 23(2) is primary law, whereas the Luxembourg Compromise clearly was not. As a practical effect, this enabled—within the limits of Article 35 TEU—some controls by the Court as to the decision-making process. Thirdly, the decision to refer to the European Council must be taken with a qualified majority. If a blocking minority wishes to compromise, the decision cannot be taken. Fourthly, the president of the Commission is a full member of the European Council under Article 4 TEU. Hence (although this may be a theoretical point), under Article 23 he wields a similar right to veto as any of the heads of state or government. There is little need to discuss this clause—which, to date, has never been invoked—in more detail. 167

164 165 166 167

E-3938/00 by Mr Huhne to the Council, OJ (2001) CE364/4. A quantitative analysis of Council voting is presented by Hayes-Renshaw, Van Aken and Wallace, ‘When and why the EU Council of Ministers votes explicitly’ (2006), 161 et seq. The Commission, of course, had an interest in majority voting and instructed its staff where majority voting was permitted to make every effort to avoid talks dragging on until unanimity was achieved: Institutional vade-mecum (provisional edition) (1986), § 53. Thus, it seems that Dashwood’s fears that such clauses would start to infest the EU’s legal order have not materialised: ‘External relations provisions of the Amsterdam Treaty’ (1998) at 1035. This is the qualification of Posthumus Meyjes, ‘Nieuwe kleren voor het buitenlands en veiligheidsbeleid’ (1997) at 367. Ibid. Article III-300.2 would succeed Article 23(2), with the difference that ‘vital and stated reasons’ replace the ‘important and stated reasons’. Additionally, it introduces a mediating role for the MFA. In addition, Articles III-270 and III 271, which deal with the sensitive issue of judicial co-operation in criminal matters, provide for a special referral procedure. However, this procedure is a far cry from the ‘Luxembourg Compromise’ and would not allow any Member State to block decision making.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

9.3.3

238

Other voting conventions in the Council

There are three more developments that must be briefly mentioned, namely, the Ioannina Compromise; the Declaration attached to the Final Act of Maastricht on voting in the field of the common foreign and security policy, and the 2003 reform of the Council’s voting rules. The Ioannina Compromise. During the accession negotiations of Finland, Austria, Norway and Sweden some Member States felt that the new accessions unbalanced the QMV arrangements in the Treaties since these countries are all relatively small. This would cause the minimum population represented by a qualified majority to fall considerably. In this logic, the democratic legitimacy of Council decision-making would require lowering (in percentage terms) the threshold for a blocking minority.168 Finally a compromise text was adopted by the Council meeting in Ioannina on 29 March 1994. Under this Ioannina Compromise, any minority representing 23 to 25 votes (i.e., just less than a blocking minority) could request the Council to reach a solution that could be supported by at least 65 votes.169 Apart from this ‘“soft” suspensive veto’,170 the Ioannina Compromise comprised several other elements not published in the Official Journal. Firstly, it was agreed that the threshold in the Treaties for a decision with QMV would be raised upon the next accessions.171 Moreover, the issue of QMV and the weighing of votes would be examined in the 1996 IGC. Lastly, the various elements of the Declaration were to apply ‘until the entry into force of an amendment to the Treaties, following the 1996 Conference’. This compromise is not applied unless specifically invoked. Apparently, this happens rarely172 and with little lasting success.173 The United Kingdom invoked it during the Council meeting on 24 October 1995. This led to further negotiations and a solution that carried the support of all Member States bar Italy and the abstaining UK.174 Note that these two states had a combined total of 20 votes, which technically should not even have been sufficient for invoking the Ioannina procedure. The Ioannina Compromise is an entirely different animal than the Luxembourg text is.175 To begin with, it took the form of a Council Decision.176 Moreover, it did

168 Club de Florence, Europe: l’impossible statu quo (1996) at 104. 169 Council Decision of 29 March 1994 concerning the taking of Decision by qualified majority by the Council, OJ (1994) C105/1, amended by OJ (1995) C1/1. 170 Moberg, ‘The Nice Treaty and voting rules in the Council’ (2002) at 267. 171 Press release following the discussions of the Ministers for Foreign Affairs at their informal meeting in Ioannina on 26 and 27 March 1994, Council doc. 6004/94 of 11 April 1994. The Compromise was reached on 64 votes, but this was based on the assumption that Norway would accede. After this failed to materialise, the threshold was set at 62. 172 Ponzano, op. cit. footnote 115 at 43. 173 According to Piris, The Constitution for Europe—A legal analysis (2006) at 105, the Compromise never delayed the adoption of the act concerned with more than one day. 174 Reply to written question 316/96 by Mr Moorhouse to the Council, OJ (1996) C280/19. 175 Similar Jacqué, op. cit. footnote 93 at 154. 176 I cannot subscribe to the conclusion of Koenig und Pechstein, ‘Die EU-Vertragsänderung’ (1998) at 134, that ‘[s]owohl die in ihrer Rechtsqualität zweifelhafte “Luxemburger Vereinbarung” als auch der “Kompromiß von Ioannina” illustrieren, daß die Vertragsstaaten bei heiklen Fragen durchaus Kompromisse schließen, welche materielle Änderungen jenseits des in Art. N Abs. 1 EUV/Maastricht (Art. 48 EUV/Amsterdam) vorgeschriebenen Verfahrens einvernehmlich herbeiführen, ohne dabei Rat, Kommission und Europäisches Parlament am Verfahren zu beteiligen’. The Ioannina text was a Council compromise approved by the Corfu European Council and subsequently formalised in a Council Decision. The ‘Luxembourg Compromise’ was not agreed, but at least it was not agreed in the Council.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

239

not override the voting rules laid down in the Council’s Rules of Procedure,177 which entitle any minister or the Commission to request the voting procedure to be opened, subject to the will of the majority in the Council.178 In my view, the Ioannina Compromise—notwithstanding its form as a Council Decision—was an agreement of a political nature (i.e., convention) amongst the Member States and their representatives in Council. To begin with, the Decision mentions no legal basis. Secondly, the decision is not even addressed to anyone in particular, which means it is not an Article 249 TEC instrument. Lastly, its validity was extended through a ‘Declaration’ attached to the final act of Amsterdam.179 ‘Declarations’ from the Member States cannot amend or extend secondary law. Hence the Ioannina Compromise must have had no more than a politically binding value.180 The Council took the view that the agreement should apply until 1 January 2005,181 meaning it is now presumably history. However, the Declaration on Article I-25 attached to the Constitution comprises a similar mechanism182 which should have remained in force ‘at least until 2014. Thereafter the Council may adopt a European Decision repealing it’. The Declaration attached to the Final Act of Maastricht on voting in the field of the common foreign and security policy183 provides that ‘[t]he Conference agrees that, with regard to Council decisions requiring unanimity, Member States will, to the extent possible, avoid preventing a unanimous decision where a qualified majority exists in favour of that decision’. I am not aware of this declaration having had any great impact on decision-making. The 2003 reform of the Council’s voting rules. In the 1960s the six members of the Council may have deliberated almost with the intimacy of a small committee. Since that time, the increase of ministers and Council formations necessitated the institution to operate somewhat more like a parliament-type body. In March 2003 the Council adopted a ‘code of conduct’ with rules of behaviour that are a further small step to changing the nature of the institution. These rules, intended to use meeting time optimally, will in practice enhance the role of the presidency.184 The restrictions on speaking time and the suggestion that ‘like-minded delegations are encouraged to hold consultations with a view to the presentation by a single spokesperson of a common position on a specific point’ suggest a very timid first step towards a grouping-based organisation of the Council.185 These organisational changes agreed amongst the Member States potentially may have long-term effects on the Council’s relations with the Commission. One can hypothesise that, following future accessions, blocks of states will become more 177 Reply to written question 788/96 by Mr Lange to the Council, OJ (1996) C 280/81. 178 Article 11(1) of the Council’s Rules of Procedure. 179 Declaration (No 50) annexed to the final act of Amsterdam relating to the Protocol on the institutions with the prospect of enlargement of the European Union, OJ (1997) C340/142. 180 At the time of its origin, apparently one Member State argued it was legally binding: Westlake and Galloway, op. cit. footnote 55 at 245. 181 Reply to written question E-1109/01 by Mr Bonde to the Council, OJ (2001) CE364/48. 182 OJ (2004) C310/421. 183 Declaration on voting in the field of the common foreign and security policy, OJ (1992) C191/104. 184 There is no need to discuss them in detail here. For details see the ‘I/A’ item note from the Antici Group to COREPER/Council—Working methods for an enlarged Council—Code of Conduct, Council doc. 7105/03 of 11 March 2003, approved by the Council on 18 March 2003: 2495th Council meeting (External Relations), Council doc. 6941/03 of 18 March 2003. 185 The suggestion was soon taken up by at least the Special Committee on Agriculture: Outcome of proceedings from the Special Committee on Agriculture—Working method for the Special Committee on Agriculture, Council doc. 9468/03 of 19 May 2003.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

240

natural. Depending on the strength of such blocks, this could have some impact on the possible creation of a blocking minority and hence on the Commission’s negotiation flexibility. In 2004 the Council enacted the code of conduct by incorporating it into its Rules of Procedure.186 9.4

Interinstitutional contacts during the legislative process

During the legislative procedure there are many interinstitutional contacts. Much of this is practice, but some interinstitutional conventions have developed that provide a normative framework implementing the Treaties. The general information rights of Parliament were discussed in Chapter 7 and much of what was said there, applies in the field of legislative procedures as well. As far as such rules relate to specific legislative procedures, they are discussed in the following Chapters. In this section I will concentrate on the general rules that apply to all legislative procedures. I discuss these conventions under the following headings: the ‘suites données’ procedure; the forwarding of proposals to Parliament; the Commission’s reaction to Parliament’s amendments; informing Parliament of the discussions within the Council; the Commission’s role in ensuring that the Council does not adopt legislation before Parliament’s opinion has been received; the Commission’s commitments concerning the withdrawal of legislation; and conventions concerning legal bases. The ‘suites données’ procedure. In 1994 the Commission adopted the currently applicable ‘suites données’ procedure,187 which rests on conventions agreed with Parliament. To begin with, the Commissioners committed themselves to inform the respective Parliamentary committees, along modalities to be agreed between the respective cabinets and committees. In addition, following each plenary, the Commission selects some resolutions adopted by Parliament. Those resolutions are selected on the basis that they make reference to a Commission action foreseen in the legislative programme or require a Commission proposal or have an important impact on a current problem. Resolutions adopted on the Commission’s annual report on the application of Community law, Parliamentary initiatives under Article 192 TEC and resolutions adopted in the framework of the follow-up of Commission proposals are always included. The Commission’s Directorates-General prepare replies, which have to be communicated to Parliament within three months (in practice this happens on a half-yearly basis). Parliament publishes tables indicating the follow-up action taken in its Bulletin (European Parliament Activities). As one of the commitments of Mr Barroso given at the time of his Commission’s investiture, he promised in 2004 to extend the mechanism for replying to Parliament to cases other than legislation.188,189 186 Article 20 and Annex IV of the Council’s Rules of Procedure. 187 Nouvelle procédure pour les ‘suites données aux résolutions d’initiative du Parlement Européen, Commission doc. SP (94) 2760/2 of 25 November 1994. This followed older convention. See point 10 of the Practical measures to strengthen the powers of control of the Parliament and to improve relations between the Parliament and the Commission, Commission doc. COM (73) 999 final of 30 May 1973, point 10 of the Joint Declaration of the European Parliament and the Commission, Parliament doc. 1-328/84/rev of 23 May 1984; the Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121/122, § 3 and the Framework agreement on relations between Parliament and the Commission, OJ (2006) CE117/125, § 14. The latter two texts expanded the Commission’s obligations towards information to a very broad field. 188 Commission president Barroso during the debate of 18 November 2004.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

241

The Council agreed, as early as 1959, to put Parliament’s resolutions on its agenda for information.190 In 1969 it agreed to inform Parliament orally or in writing on the reasons why it deviated ‘sensiblement’191 in an adopted act from Parliament’s opinion, provided the request was made by the competent Parliamentary committee and channelled through the office of that institution’s president.192 Any follow-up questions should take the form of written or oral questions.193 As always, the Council refused to discuss the positions taken by its members. In 1973 the Council agreed that the differences between the draft report to COREPER/Council and Parliament’s opinion should be pointed out in a document and agreed to meet the president of Parliament regularly in order to examine ‘the action taken by the Council on the Parliament’s Opinions’.194 A decade later, the Stuttgart European Council solemnly promised Parliament that the Council would respond to ‘resolutions concerning matters of major importance and general concern, on which Parliament seeks their comments’.195 Two years afterwards, the Council expressed a preference for such information to be given orally.196 It appears that the Council in 1988 decided that the chairmen of its working parties must ensure that Parliament’s opinions are given due weight in the discussions. Notwithstanding Parliament’s wishes,197 the Council has not been able to go much further than that. In practice, all texts adopted by Parliament are transmitted to the Member States, followed by a summary document prepared by the Council’s Secretariat-General summarising all requests to the Council.198 The presidency will briefly draw the 189 For the sake of completeness, I mention the procedure of Article 276(3) TEC (III-409.3-4 Const), which foresees a ‘suites données’ procedure following the discharge decision. For an example of a Commission communication under this procedure see Commission doc. COM (2002) 696 final of 10 December 2002. 190 Council doc. 146/59 rév of 14 April 1959 at 13, confirmed in Council doc. 168/60 rév of 25 March 1960 at 13. The latter text restricted it to resolutions addressed to the Council. 191 This criterion features in the Letter of Council president Harmel to president of of the Assembly Mr Scelba, Council doc. 3050 of 20 March 1970. 192 Following complaints from Parliament (Resolution on relations between the European Parliament and the Council, OJ (1982) C234/52), the Council confirmed these commitments: Council doc. 5749/1/82 of 2 April 1982. 193 Letter from Council president Mr De Koster to Mr Scelba, President of the Assembly, Council doc. 10891 of 11 November 1969. This followed a request from Parliament: Letter from SecretaryGeneral of Parliament Mr Nord to the Secretary-General of the Council Mr Calmes, Council doc. 109442 of 25 April 1969. The Council did not agree to any deadlines and wished to retain the right to judge on a case by case basis whether it could reply: Letter of Council president Harmel to president of Parliament Mr Scelba, Council doc. 3050 of 20 March 1970. 194 Relevé des décisions prises par le Conseil lors de sa 257ème session tenue le 15 octobre 1973, Council doc. R/2641/73 of 5 November 1973. 195 Solemn Declaration on European Union, § 2.3.3. 196 Report from the Working Party on General Affairs to COREPER—Consideration of European Parliament Resolutions and informing that body on the action taken, Council doc. 7546/85 of 17 June 1985, as changed by the Draft summary record of the 1213th meeting of COREPER held in Brussels on Monday 24 and Wednesday 26 June 1985, Council doc. 7963/85 of 11 September 1985. 197 Resolution on the conclusion and adaptation of interinstitutional agreements, OJ (1993) C115/253, § 12. 198 Council doc. 7546/85 of 17 June 1985, as changed by Council doc. 7963/85 of 11 September 1985. For an example see Resolutions, decisions and opinions adopted by the European Parliament at its part-session in Strasbourg from 17 to 20 November 2003, Council doc. 15402/03 of 27 November 2003. Annex III to the Presidency Report submitted to the European Council of Brussels of 29 October 1993, § 1 promised a parallel obligation for the Presidency to bring the resolutions etc. of Parliament to the attention of the Council in the domain of CFSP but, in practice, this is subsumed in the general procedure.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

242

Council’s attention to those resolutions that it deems most important. Non-legislative resolutions are taken ‘into account insofar they are relevant’.199 Realistically speaking, a very detailed discussion in the Council is unlikely to take place, since the ministers are under no obligation to act upon non-legislative opinions expressed by Parliament. Moreover, it would be hard to fit any such discussion into the overloaded agendas of the Council and its working parties.200 The forwarding of proposals to Parliament. The Commission agreed to forward to Parliament all legislative proposals which it submits to the Council,201 as well as ‘any other Commission document relating to such proposals, which are forwarded to the Council at any time during the decision-making procedures’.202 In addition, in 2000 the executive agreed not to make public any legislative or significant policy initiative without notifying Parliament in writing or orally.203 From a separation of institutions point of view, this is an important commitment since it enables Parliament to influence policy making in the Commission from early on, rather than at its implementation stage. The Commission’s reaction to Parliament’s amendments. In the early 1960s, Parliament attempted to have the Commission consult it on draft proposals for regulations and to indicate its position on Parliamentary amendments before forwarding the proposal to Council.204 However, neither the Commission nor the Council accepted these suggestions. Ultimately, the MEPs managed to obtain three political commitments that gave them partial satisfaction. Firstly—as noted above—, after some difficulties and pressure from the MEPs in the early 1960s the Commission agreed systematically to notify Parliament of its amended proposals. Moreover, it consented to amend its proposal,205 either in writing or orally,206 once it accepts Parliamentary amendments. Thirdly, the Commission agreed to take a reasoned position, in principle orally, with respect to each of Parliament’s amendments. The suggestion was made in the Furler report and agreed in the Rey-Scelba agreement.207 It was repeated in 1973 and 1984.208 The commitment returned in a slightly reworded form in paragraph 2 of the 1990 Code of conduct and in paragraph 3.7 of its 1995 successor. 199 Note from COREPER to the Council, Council doc. 5749/1/82 of 1 April 1982. 200 Reply to written question 1901/82 by Mr Moreland to the Council, OJ (1983) C118/16. 201 Practical measures, Commission doc. COM (73) 999 final of 30 May 1973, Annex 1, § 3, confirmed in the Joint Declaration of the European Parliament and the Commission, Parliament doc. 1-328/84/rev of 23 May 1984, § 6. 202 1995 Code of conduct, OJ (1995) C89/69. See also the Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121/122, § 12. As far as amended proposals are concerned, the commitment dated from the 1960s: Schmutzer, ‘Some legal aspects of the work of the European Parliament’ (1967-68) at 82-84. 203 2000 Framework Agreement, § 13. That commitment contained several additional requirements that need not be further discussed. For the procedures in place to implement the commitment see the Practical guide to the Framework Agreement, Commission doc. SP (2000) 3785-5 of 20 December 2000 at 11. 204 The Furler report: Résolution sur les compétences et les pouvoirs du Parlement européen, JO (1963) 1916, and Louis, Les règlements de la Communauté Économique Européenne (1969) at 9. 205 Commission doc. COM (73) 999 final of 30 May 1973, Annex 1, § 6 and the Commission’s Vademecum institutionnel (1990) at 48. 206 Oral amendment is allowed by the Court: Case C-280/93, Germany vs Council, [1994] ECR I-4973, § 36. 207 Débats du Parlement européen, JO (1970) Annexe 121/133-134. 208 Commission doc. COM (73) 999 final of 30 May 1973 at § 10 and Annex 1 at § 5, and the 1984 Joint Declaration of the European Parliament and the Commission, § 8.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

243

If the Commission cannot accept an amendment, it informs Parliament of the reasons.209 The logic of this convention finds its basis in (the predecessor of) Article 250(2) TEC, which allows the Commission to amend its proposals ‘at any time’ during the legislative procedure.210 This, coupled with the rule of Article 250(1), according to which the Council needs unanimity to amend Commission proposals, would in theory have given the Commission considerable leeway in determining the content of legislation. The Commission has, however, always insisted that this convention should not prejudice its right of initiative.211 After the introduction of the co-operation and co-decision procedures, the convention was adapted. Since the 2000 Framework Agreement, the Commission can now only reject a Parliamentary amendment whilst taking ‘utmost account’ of amendments adopted at second reading. It must do so ‘on important grounds and following examination by the College’ and explain its decision before either the full Parliament or (in co-decision) in its opinion on Parliament’s amendments by virtue of Article 251(2)(c).212 Lastly, in 1984 Parliament and the Commission agreed that, before submitting its proposals to the Council or before making substantial changes to them, the executive would ‘where appropriate, hold talks with the competent committee’.213 Informing Parliament of the discussions within the Council. As early as 1970, the Council committed itself to inform Parliament of the reasons for which the Council failed to act upon Parliament’s opinion, whenever this is the case. This commitment was not fulfilled to the MEPs’ satisfaction and in 1981 they reminded the Council of its existence.214 The Council subsequently confirmed it.215 In 1990 the Commission undertook regularly to inform the relevant Parliamentary committee of the main positions arising from discussions in the Council, notably where they are at odds with the original proposal.216 This followed a long-standing request from Parliament in this direction.217 In 1995 the commitment was, importantly, extended to discussions in Council 209 In the 1995 Code of conduct the Commission promised to explain its reasons for not adopting Parliamentary amendments ‘before Parliament or at the next meeting of the competent committee’. 210 On the application of Article 250 in the co-decision procedure, see sub-section 11.4.3, below. 211 Läufer, ‘Das Europäische Parlament nach der Direktwahl: Positionsstärkung durch intrakonstitutionellen Wandel?’ (1979) at 264. 212 See currently § 31 of the 2005 Framework Agreement. In the Interinstitutional Agreement of 28 November 2001 on a more structured use of the recasting technique for legal acts, OJ (2002) C77/1, Commission, Parliament and the Council set up the recasting procedure, which entails the substantial amendment of an act whilst simultaneously codifying it. This necessitated a procedure for determining which parts of recasting proposals are pure codification and which parts are intended as substantial amendment, since the Council’s and Parliament’s right of amendment extends only to the ‘object’ of the Commission proposal rather than to the whole act. An interinstitutional committee set up for this purpose reviews the proposal and, in practice, advises on the extent to which the scope of amendment goes. 213 1984 Joint Declaration of the European Parliament and the Commission, § 6. In 1963 the two institutions agreed that the executive would inform the MEPs of its own position during the negotiations in Council: Furler report at II. 214 Resolution on relations between the European Parliament and the Council, OJ (1981) C234/52, § 15. 215 Briefing of the European Parliament on the action taken by the Council regarding the Resolutions on relations between the European Parliament and the Council and the right of legislative initiative and the role of the European Parliament in the legislative process of the Community, Council doc. 5749/1/82 of 2 April 1982. 216 1990 Code of conduct, OJ (1991) C280/165, § 3. 217 Résolution sur la position du Parlement européen à l’égard de l’évolution institutionnelle récente des Communautés européennes, JO (1966) 3465.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

244

bodies.218 In 2000 the logical addition was made that the executive must also inform the MEPs ‘as soon as possible’ when it agrees with Council amendments. In practice this is often done through briefings of the rapporteur.219 The Council has since long held the view that the Commission is precluded from divulging the positions of individual Member States.220 The Commission tends to abide by this rule. The Commission’s role in ensuring that the Council does not adopt legislation before Parliament’s opinion has been received. Following the wishes of the MEPs,221 the Commission undertook in 1990 to remind the Council not to adopt a political agreement on Commission proposals until Parliament had delivered its opinion. The executive furthermore undertook to request the ministers to conclude their discussion after they have had a reasonable time to consider Parliament’s opinion.222 In the 2000 Framework Agreement the two institutions drew the logical conclusion that this commitment is mainly of importance outside the context of the co-decision procedure;223 during much of co-decision, the Council evidently cannot act but upon Parliament’s amendments. The Commission’s commitments concerning the withdrawal of legislation. The Commission undertook to withdraw ‘where appropriate’ legislative proposals rejected by Parliament. If the Commission decides to maintain its proposal, it shall explain the reasons for that decision in a statement to Parliament.224 The Commission undertook to give Parliament and the Council prior notification before withdrawing its proposals.225 The words ‘if appropriate’ take the sting out of what otherwise would be approaching a ‘right of veto’ for Parliament over all (including non-codecided) legislation. Conventions concerning legal bases. A full review of the legal issues surrounding the legal bases of Union legislation would exceed the scope of this work and the discussion is limited to the interinstitutional convention in this field. The legislative procedure to be followed depends on the legal basis of the proposed legislation. Consequently, the legal basis chosen directly affects the prerogatives of the institutions and the checks and balances between them. This caused legal basis disputes to become one of the battlegrounds between especially Parliament and the Council from the moment the Single European Act introduced the co-operation procedure,226 and even more so after the introduction of co-decision with the Maastricht Treaty. Without going into any detail on the case law in this field, the main rule to be noted is that ‘where an institution’s power is based on two provisions of the Treaty, it is bound to adopt the relevant measures on the basis of the two relevant provisions’.227 If, however, the procedural requirements in those two 218 1995 Code of conduct, § 3.5, respectively. These commitments were repeated in the 2000 Framework Agreement, Annex 1, § 6. 219 Practical guide to the Framework Agreement, Commission doc. SP (2000) 3785-5 of 20 December 2000 at 21. 220 Reply to written question 55 by Mr Vredeling, OJ (1967) 178/1. 221 See e.g. Parliament’s Resolution on relations between the European Parliament and the Council, OJ (1990) C284/58. 222 1990 Code of conduct, § 1 and 1995 Code of conduct, § 3.4. 223 Annex 1, § 7(i). 224 1995 Code of Conduct, § 3.8. 2000 Framework Agreement, Annex 1, § 9 did not contain the explanation formula, but § 33(iii) of its 2005 successor restored it. 225 1995 Code of Conduct, § 3.9 repeated in the 2000 Framework Agreement, Annex 1, § 9. 226 On the history of the problem Cullen and Charlesworth, ‘Diplomacy by other means’ (1999) at 1246 et seq. 227 Case 165/87, Commission vs Council, [1988] ECR 5545, § 11.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

245

provisions are incompatible, then the single appropriate legal basis must be determined.228 In Rotterdam Convention the Court clarified that the criterion is not so much whether the two legal bases are incompatible, but also whether ‘the use of two legal bases is liable to undermine the rights of the Parliament’.229 The legal basis should be clearly stated in the proposal itself.230 Although the Constitution simplifies legislative procedures to a certain extent, it does so incompletely. Although perhaps somewhat less in number, it remains likely that interinstitutional conflicts on legal base issues would continue. The Court of Justice’ case law would thus remain relevant. The Court of Justice ruled in 1987 that ‘the choice of the legal basis for a measure may not depend simply on an institution’s conviction as to the objective pursued but must be based on objective factors which are amenable to judicial review’.231 This practically implies that the Court must be able to deduct the legal basis from the instrument and that the ‘objective factors’ are to be found in the measure itself. In the light of this ruling, paragraph 6 of the 1990 Code of conduct foresaw that ‘[t]he choice of the legal basis is determined in the light of the contents of the planned proposal’. The two institutions then agreed to a procedure to deal with differences on the legal basis: ‘The Commission is willing to step up informal contacts at the appropriate level with Parliament and the Council and, notably between their legal services by far-reaching exchanges of views whenever the Commission is preparing to take initiatives in new areas.’ Five years later, point 2 of the 1995 Code of Conduct brought some important changes to this convention. Firstly, the Council was apparently excluded from this dialogue à trois. Moreover, the two institutions almost literally copied the Court case law by noting that ‘in accordance with the Community’s structure of powers . . ., the choice of legal basis must be founded on objective factors which admit of judicial review, including, in particular, the objective and content of the act’. More importantly, the Commission committed itself to ‘take the utmost account of all changes to the legal bases of its proposals contained in Parliament amendments. The Commission undertakes to set out in detail the reasons for its position’. Furthermore, in cases where the Council considered amending the legal basis of a proposal, the Commission committed itself to reminding the ministers ‘of the need to adhere to the rules evolved by the Court of Justice requiring Parliament to be reconsulted. The Commission shall inform Parliament that it has so reminded the Council’.232 Lastly, 228 Case C-300/89, Commission vs Council (titanium dioxide), [1991] ECR I-2867, §§ 17–21. In order to avoid such complications, the Commission normally attempts to base a proposal where possible on a single legal basis: Vade-mecum institutionnel (1990) at 59. 229 Case C-94/03, Commission vs Council (Rotterdam Convention), Judgment of the Court of 10 January 2006, § 52; see also case C-178/03, Commission vs Parliament and Council, Judgment of the Court of 10 January 2006, not yet reported, § 57. 230 This flows from Article 253 TEC (I-38.2 Const). By way of exception, the legal basis need not be explicitly mentioned when it may be determined from other parts of the measure: case 45/86, Commission vs Council (GSP), [1987] ECR 1493, § 9. 231 Case 45/86, Commission vs Council (GSP), [1987] ECR 1493, § 11. See also cases 68/86, United Kingdom vs Council, [1988] ECR 855, § 24 and 131/86, United Kingdom vs Council, [1988] ECR 905, § 29. 232 This happens in practice by letter from the Commissioner to the chair of the Parliamentary committee concerned: Practical guide to the Framework Agreement, Commission doc. SP (2000) 3785-5 of 20 December 2000 at 20.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

246

the Commission undertook to submit to the MEPs at the same time as to the Council, any proposals amending legal bases. This, of course, hardly surprises in the light of the Commission’s general commitments on informing Parliament of revised proposals. Paragraphs 3 and 4 of Annex 1 to the 2000 Framework Agreement essentially repeat these commitments but, peculiarly, they were omitted from its 2005 successor. This, however, is unlikely to imply any great change in practice. Parliament implemented these conventions in its Rules of Procedure. Rule 35 provides that the responsible committee of Parliament must first verify the legal basis of Commission proposals and other legislative documents. If the committee concerned disputes the legal basis chosen by the Commission, it must request the opinion of the Committee on Legal Affairs, which can also act ex officio. If this body rules that the legal basis is incorrect, the full Parliament must first vote on this question before it can vote on substance. If the Commission does not agree to amend the legal basis, ‘the rapporteur or the chairman of the committee responsible for legal affairs or of the committee responsible may propose that the vote on the substance of the proposal be postponed to a subsequent sitting’. There is no restriction of this process repeating itself—but it is doubtful whether, legally speaking, this could lead to obstruction or deferral ad infinitum.233 9.5

Conclusions of this Chapter

What conclusions can be drawn as regards the role of interinstitutional convention as a factor influencing the checks and balances? The discussion above reveals several noteworthy trends. Firstly, although the Commission’s right of initiative is quasi-exclusive in law, it has been limited by interinstitutional convention and practice. On the one hand, the Commission is not always free to propose or frame policies underlying legislation, but in many cases reacts to policies adopted by the European Council, which has become the ‘engine’ of the process of integration.234 On the other hand, interinstitutional conventions have made it more difficult for the Commission to ignore Parliament. Whereas the European Council has made itself felt since the 1970s, this latter aspect is a relatively recent development. Although the 1990 Code of conduct and the omnibus interinstitutional agreements preceding it still witnessed a relative ‘equality of arms’ between Commission and Parliament,235 the 1995 Code of conduct and the two Framework Agreements succeeding it marked a distinct shift. From this period on the Commission—with some exaggeration—started to become ‘Parliament’s agent in the Council’. In any event, the Commission no longer masters the European agenda236 and its right of initiative has turned from a power to set the agenda to a check on the policy aspirations of other institutions.

233 Not where the Treaties do not grant Parliament a right of veto. Any course of action which de facto would amount to vetoing draft legislation arguably violates the duty to sincerely co-operate. See in this sense case C-65/93, Parliament vs Council (GSP), [1995] ECR I-643, § 27. 234 Barents, ‘Een Grondwet voor Europa (I): algemeen overzicht’ (2004) at 240. 235 Rideau, Droit institutionnel de l’Union et des Communautés européennes (2002) at 550 attributes the Commission’s status in the early 1990s to personal factors (M. Delors’ handling of the institution) and to the internal market programme that put the executive in the driving seat. 236 Lassalle and Levrat, ‘Un triangle à quatre côtés: l’équilibre institutionnel et le Conseil européen’ (2004) at 435.

CHAPTER 9: LEGISLATION: GENERAL OBSERVATIONS

247

Secondly, with slight exaggeration, ‘the process of the Council remains consensual in its ethos’237 and the more this is so, the more the final product acquires the characteristics of a compromise. A tendency towards consensus—whether by law or by practice—is no fertile ground for creative use of the Commission’s right of initiative. This does not necessarily imply that ‘je effizienter der Rat, desto erfolgreicher die Kommission’.238 On the contrary, the Commission has lost prestige and efficiency. Although the times of the Luxembourg Compromise are well behind us, the enduring pressure amongst Member States to seek consensus continues to make it difficult for the Commission fully to enjoy the margin of manoeuvre that the authors of the Treaties foresaw for it. This issue touches on a logical consequence of separation of institutions thinking: no power should be able unilaterally to change the balance of powers. The voting practice of the Member States throughout much of the seventies and eighties is a good example where exactly that happened. The Commission’s decreased influence over decision-making reflected on Parliament. In short, a convention between the Member States unbalanced the Community’s decision-making machinery. In Chapter 1 of this study I asked to what extent it is possible to balance the EU institutions through the use of interinstitutional convention and what the conditions are for doing this. As far as the Commission’s right of initiative is concerned, the reply must be that interinstitutional convention scaled the balances towards both (European) Council and Parliament. As far as the other general principles of the legislative procedures are concerned, the picture similarly suggests that the checks and balances foreseen by the Treaties can be influenced, if not neutralised, by conventions.

237 Dashwood, ‘The constitution of the European Union after Nice: law-making procedures’ (2001) at 220. 238 Ehlermann, op. cit. footnote 141 at 352.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

10

PARLIAMENT’S RIGHT OF CONSULTATION

10.1

Introduction to this Chapter

249

The consultation of Parliament by the Council is the oldest legislative procedure. According to the Court of Justice ‘[d]ue consultation of the Parliament in the cases provided for by the Treaty . . . constitutes an essential formality disregard of which means that the measure concerned is void’.1 Since the consultation procedure looks back at a long history, it unsurprisingly gave rise to many interinstitutional conventions. Some of these were already discussed in the previous Chapter. Here I will discuss the matters on which Parliament is consulted (sub-section 10.2), the procedure in Parliament (10.3), reconsultation (10.4) and the conciliation procedure (10.5). Section 10.6 deals with consultations on Member State initiatives. 10.2

The matters on which Parliament must be consulted

The matters on which Parliament should be consulted as a legal obligation are listed in the Treaties.2 Besides these instances, I note optional consultations, sui generis instruments, and Commission communications. Optional consultation. In many instances the Treaties did not even require consultation, but what the MEPs failed to obtain in law they got in the form of interinstitutional convention. As early as 1960, the Council promised to consult Parliament on important problems even where there was no legal obligation to do so.3 Optional consultation is applied to all important matters. The Council retains the right to decide—after consultation of the Commission—to decide what qualifies as an important matter. In practice the Council has tended to be quite liberal in this respect, and if criticisms have been voiced about optional consultation, they concerned the weight attributed to Parliamentary opinions rather than a failure to request them.4

1

2

3

4

Case 138/79, Roquette Frères vs Council (Isoglucose), [1980] ECR 3333, § 33. See also cases 817/79, Buyl a.o. vs Commission, [1982] ECR 245, § 16, 1253/79, Battaglia vs Commission, [1982] ECR 297, § 17, C-21/94, Parliament vs Council, [1995] ECR I-1827, § 17 and C-392/95, Parliament vs Council (Visas), [1997] ECR I-3213, § 14. As Bradley, ‘Maintaining the balance: the role of the Court of Justice in defining the institutional position of the European Parliament’ (1987) at 51 points out, Roquette was not really surprising in the light of the Court’s treatment of the consultation claim in case 2/54, Italy vs High Authority, [1954] ECR 37. The Court of Justice has ruled that implementing regulations and implementing directives adopted without consultation of Parliament, must respect the basic elements laid down in the basic legislation: cases 46/86, Romkes vs Officier van Justitie for the District of Zwolle, [1987] ECR 2671, § 16 and C-303/94, Parliament vs Council (plant protection products), [1996] ECR I-2943, § 23. Patijn, Het Europees Parlement—de strijd om zijn bevoegdheden (1973) at 110, Troisième Rapport général (1960) at 48 and Procès-verbal de la 31ème session du Conseil de la Communauté Economique Européenne tenue à Bruxelles, les 9 et 10 mars 1960, Council doc. 168/60 rév. of 25 March 1960 at 16. Patijn at 110-111.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

250

This consultation is a favour rather than a right. Both Article 7 TEC and the Council’s prerogative to decide whether it will consult Parliament object to it becoming a right of customary law.5 In 1982 the Council confirmed its intention, while noting that this was done on a ‘flexible’ basis, especially if the matter was urgent.6 Since 1969 the convention developed to note Parliament’s opinion in the preambular language of legislation, even if the consultation is not compulsory. The Council may decide by unanimity not to do this. Notwithstanding these conventional improvements, optional consultation is not the same as compulsory consultation. If for no other reason, compulsory consultation creates a legal obligation for Parliament to provide an opinion within the limits of the possible.7 There is no such legal obligation in the case of optional consultation. Sui generis instruments. Unless obliged by law, the Council does not consult Parliament on sui generis instruments such as resolutions. It does not resort to resolutions to evade the Parliamentary consultations, but rather to lay down a work programme that later on leads to the consultation of Parliament.8 This manner of handling things is imperative in view of the diverse character of the Council and the necessity to develop one policy line. An unfortunate side effect is that it somewhat undercuts the practical value of the right of consultation to the extent that the MEPs are asked to give their opinion on what is already adopted policy. Commission communications. Parliament is normally consulted on Commission communications, which have no legal value but which indicate (potential) policies. Of course, this does not make them less interesting to Parliament. Indeed, for an assembly aspiring to change the great lines of policy, examining such papers may well prove more rewarding than scrutinising the legislative outcome. 10.3

The consultation procedure in Parliament

Contrary to the first reading in co-decision, it is the Council9 and not the Commission that consults Parliament. This is evident10 from the formula used: ‘the Council shall act . . . after consulting the European Parliament’. There are three aspects to the matter: Parliamentary delaying tactics; the treatment of Parliament’s opinion in the Council; and interinstitutional contacts during the legislative process. Parliamentary delaying tactics. In Parliament, matters for compulsory consultation are normally discussed in the committee(s) concerned. This is followed by a report adopted in plenary approving the texts with or without proposed amendments, or rejecting the proposal. 5 6 7 8 9 10

Bieber, Organe der erweiterten Europäischen Gemeinschaften: Das Parlament (1974) at 63, cherished some hopes about a development in the direction of customary law. I have grave doubts whether customary law exists at all within the EU’s legal order (see sub-section 3.2.3 ad I, above). Council doc. 5749/1/82 of 2 April 1982. In this sense Advocate General Jacobs in case C-316/91, Parliament vs Council (Lomé Convention) (Opinion), [1994] ECR I-625, §§ 25-28 and the judgment in that case, § 16. Council doc. 5749/1/82 of 2 April 1982. The consultation is preceded by a written procedure in which the Member States are asked to agree to a particular consultation (even when it is compulsory). Rideau, Droit institutionnel de l’Union et des Communautés européennes (2002) at 554, who argues that ‘[l]es traités ne précisent pas l’institution appelée à consulter le Parlement. En pratique, c’est au Conseil que cette tâche revient’. Indeed, in the French text of the EC Treaty, the formula ‘après consultation du Parlement européen’ leaves some doubts. The German and Dutch versions are similarly ambiguous. The English ‘consulting’ is clear.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

251

The Treaties prescribe no deadline for Parliament to deliver its opinion.11 At both the committee and the plenary stage Parliament built delaying possibilities into its Rules of Procedure intended to force the Commission to amend its proposal.12 These delaying tactics are only possible because of the Roquette case law quoted above.13 Of course, such a delaying tactic cannot be overly stretched, or it will snap: Parliament cannot block legislation by failing to provide an opinion.14 As noted in section 9.4 above, the Commission is bound by convention to express its view on each amendment adopted by Parliament. The logic of this convention becomes clear at this point: if the Commission fails to do so or declares that it cannot accept all of the amendments, the committee may postpone the vote.15 This delaying process can recur at the plenary stage: if the Commission proposal as a whole is approved, but on the basis of amendments which have also been adopted, the vote on the draft legislative resolution is postponed until the Commission has stated its position on each of Parliament’s amendments in plenary. If the Commission announces that it is unable to adopt all amendments the delaying manoeuvres start, Parliament may decide to suspend consideration of the item. If the vote is postponed, the item is automatically referred back to the responsible committee, which has to report back in two months (this term can be extended). The idea behind the proceeding is betrayed by Rule 53(2) of Parliament’s Rules of Procedure, which provides inter alia that only amendments seeking to reach a compromise with the Commission are admissible at this stage. Several objections have been levelled against this Roquette procedure. Firstly, the Council complained that the system would lead to delays in the legislative procedure.16 This objection seems not to have borne out in practice. On average, the MEPs do not seem to take more time than the Council does. Secondly, the Commission was initially wary of the procedure, warning Parliament not to identify the wrong enemy.17 The executive’s fear was indeed not unfounded. Indeed, often Parliament’s disagreement is not so much with the Commission as it is with the Council. Forcing the Commission to adopt Parliament’s amendments puts it in the uncomfortable position of acting as Parliament’s agent. This renders it difficult to maintain the independence presupposed by its political responsibility. Thirdly, some legal observers have noted that the Rule 53 procedure ‘sails very close to the borders of legality’18 for two reasons. On the one hand, it walks the thin line between delay and obstruction. The first is permitted and indeed, in a healthy political system, may be an acceptable means of pressure. But if the consultation 11 12 13 14

15 16 17 18

Article 39 TEU provides a limited exception, entitling the Council to set a deadline of at least three months. For an example of Parliament using this procedure, see procedure CNS/2002/0298 (review of the second Comitology Decision) in Parliament’s OEIL database (), at 13 May 2003. Footnote 1 of this Chapter. In this sense also Jacobs, ‘Isoglucose resurgent: two powers of the European Parliament upheld by the Court’ (1981) at 222 and De Witte, ‘Zaak C65/93, Europees Parlement t. Raad’ (1997) at 238-239. In my view this was already clear from the reasoning in Roquette. In any event, the ruling in case C-65/93, Parliament vs Council (GSP), [1995] ECR I-643 leaves no doubt. Rule 50. Council doc. 5749/1/82 of 2 April 1982. Commission President Thorn during the Parliamentary debate of 14 October 1981, OJ (1981) Annex 1-275/131. Freestone and Davidson, The institutional framework of the European Communities (1988) at 99. Similar: Jacqué, ‘Parlement européen (janvier 1981-juillet 1982)’ (1982) at 664.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

252

procedure were to be delayed ad infinitum without a real search for a compromise with Commission or Council, delay would turn into obstruction.19 An example how not to do things was the first time the procedure was used (in connection with a Commission proposal concerning the harmonisation of tobacco tax).20 The matter was referred three times back and forth from plenary to committee, after which the Commission decided to withdraw it. Of course, these delaying practices cannot be used when the proposal is urgent. In such cases, the Council will indicate the urgency in the request for consultation and will not consider itself bound to wait for Parliament once this deadline has expired.21 The Council is, however, obliged to use all means to its disposal, such as requesting the urgency procedure22 or if necessary, by requesting Parliament’s president to convene the institution in special session. In the past, Parliament has complained of— in its view—excessive use of the urgency procedure.23 Parliament’s Roquette rules have always remained unilateral. As long as the MEPs do not transgress the legal limits of their prerogatives, the other institutions cannot object, but the procedure is practice rather than interinstitutional convention. Treatment of Parliament’s opinion in the Council. The conventions concerning the follow-up of Parliament’s resolutions in the Council were discussed above.24 In cases of compulsory consultation, the Council agreed in 1973 that ‘any differences between the draft report to the Permanent Representatives Committee or the Council and the Opinion delivered by the European Parliament should be pointed out either in the text of the draft report itself or, if necessary, in a separate paper.’25 Two years later, the Council agreed that the footnote in the recitals of the legislation concerning the citation referring to Parliament’s opinion must indicate the date on which Parliament delivered it, as well as the Official Journal reference.26 In cases of optional consultation the Council agreed in 1969 that the Parliament’s opinion should be referred to in the citations of the legislation, save where a decision to the contrary had been adopted unanimously. If the Council had sought no opinion, but Parliament nevertheless adopted one, this rule explicitly does not apply.27 Interinstitutional contacts during the legislative process. Lastly, as is the case with the ordinary legislative procedure, in practice there are numerous contacts between the Council presidency and the key players in Parliament. The former has an 19 20 21 22 23 24 25 26

27

Then-MEP Wijsenbeek spoke of a ‘suspensive veto’, which sounds perhaps a bit too optimistic to describe the extent to which Parliament can effectively hold up legislation: ‘Het vernieuwde Reglement van het Europees Parlement en de inter-institutionele verhoudingen’ (1982) at 465. Freestone and Davidson, op. cit. footnote 18 at 99. Dashwood, ‘Community legislative procedures in the era of the Treaty on European Union’ (1994) at 344. Roquette, § 36. The urgency procedure can be found in Rule 134 of Parliament’s Rules of Procedure. Bradley, ‘Legal developments in the European Parliament’ (1997) at 439. Section 9.4, above. Council doc. R/2641/73 of 5 November 1973. Working document—Reference to the Opinions of the European Parliament, Council doc. R/1351/75 of 26 May 1975 and Réponses du Conseil aux lettres du Président de l’Assemblée relatives aux laissez-passer des membres de l’Assemblée, à la procédure de concertation, à la mention des avis de l’Assemblée, Council doc. R/1535/75 of 11 June 1975. Note verbale du Président de l’Assemblée au Président du Conseil, Council doc. R/1358/69 of 17 July 1969 at 5.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

253

important role in facilitating the legislative process in such a way that the result is acceptable to the Council.28 10.4

Reconsultation

The issues related to reconsultation cover three closely related situations. Firstly, there is the situation that the Commission, following the work of the Council’s preparatory bodies, amends its proposal. Secondly, circumstances in the outside world may have changed in the time passed since the adoption of Parliament’s opinion. Thirdly, it is possible that the Council preparatory bodies substantially amend the Commission proposal, before they have received Parliament’s opinion. As a consequence, the text to be adopted by the Council may be very different from the proposal on which Parliament rendered its opinion and the question arises whether the Commission should adapt its proposal.29 To start with the first situation, since the 1960s the Commission routinely communicates its amended proposals to Parliament.30 The Council’s position is that, if the modifications do not go beyond the ‘matière traitée’ in the initial proposal, a new consultation is not required but is possible.31 The second situation arises if some time elapses between the moment when Parliament adopts its opinion and the adoption of the text in question by the Council. If in the meantime new circumstances or legal developments have occurred which change the situation in which Parliament gave its opinion, different possibilities arise. The Commission may amend its proposal, which may give rise to a reconsultation requirement. It is also possible that Parliament itself decides to adopt a revised opinion. The Council has agreed that it ‘will not fail to take [this] into account when examining the relevant dossier’.32 The third case is more difficult. In 1971 the Council and Parliament agreed that the presidents of both institutions should meet whenever the COREPER report to the Council differed significantly from Parliament’s opinion on a given proposal. This convention, however, seems to have fallen in desuetude.33 The Court of Justice has ruled that Parliament need not be reconsulted in a case where the Regulation finally adopted broadly conforms to the proposal submitted to it (and it was in a position to re-examine the proposal had it wanted to do so).34 In 1973, the Council committed itself to not examining Commission proposals until Parliament’s opinion was received. The ministers did make an exception for cases of urgency.35 In practice, however, this applies only at the level of the ministers. 28 29 30 31 32 33 34 35

The Group of Three Wise Men in its Report of October 1979 to the European Council (Annex 3) made valuable suggestions for presidencies to ‘lubricate’ the contacts between the ministers and the MEPs. In this case, Parliament is technically not ‘reconsulted’ but consulted on an amended proposal: Advocate General Léger in case C-417/93, Parliament vs Council (TACIS) (Opinion), [1995] ECR I-1185, § 55. Ganshof van der Meersch a.o., Les Novelles—Droit des Communautés européennes (1969) at 216. Demande de reconsultation de l’Assemblée, Council doc. R/2673/73 of 31 October 1973. Council doc. 5749/1/82 of 2 April 1982. Bradley, ‘Legal developments in the European Parliament’ (1990) at 247. Nevertheless, Parliament still referred to it in its Resolution on the conciliation procedure, OJ (1989) C69/151 at 153, footnote 1. Cases 41/69, ACF Chemiefarma NV vs Commission, [1970] ECR 661, §§ 68-70; 817/79, Buyl a.o. vs Commission, [1982] ECR 245, §§ 23-24 and 1253/79, Battaglia vs Commission, [1982] ECR 297, §§ 24-25. Mésures pratiques visant à améliorer les relations entre le Conseil et l’Assemblée, attached to

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

254

The working parties and COREPER do not wait for Parliament’s opinion but start processing proposals generally soon after they have been received. The Council may even arrive at a ‘political agreement’ (meaning that the text has not yet been finalised by the jurists-linguists)36 or ‘general orientations’, meaning that the conditions for adoption have not yet been fulfilled. In response to the above-mentioned Roquette case law, the Council put a mechanism in place under which it cannot vote on items on its agenda unless all the compulsory opinions have been given by Parliament, the Committee of the Regions, the EESC, the ECB and/or the Court of Auditors. Parliament protested against the Council’s practice to adopt political agreements before having considered the MEPs’ opinions.37 It felt that this would force the MEPs to pronounce on a text that may be outdated by the time their opinion is rendered, turning consultation into ‘a mere sham or fiction’.38 Nevertheless, the Court of Justice saw no such problem, as long as the Council does not formally adopt the proposal.39 Even the fact that in some cases Parliament’s opinion was received only the day before the formal adoption (making it hardly possible in practice for the Council seriously to consider it) does not change this.40 Parliament’s prerogatives are reinforced somewhat by the commitment from the Commission to remind the Council bodies in good time not to reach a political agreement before Parliament has given its opinion.41 However, this hardly compensates for the loss of influence relating to the late stage at which Parliament’s opinion is filtered into the process. Parliament tried in several ways to improve the relevance of its opinions in the Council. Firstly, in at least one instance, a rapporteur of Parliament got around of the problem by tabling the Council’s amendments as his own, in some cases with the advice to vote against them. This solution enabled Parliament at least to express an opinion on them.42 More structurally, the reconsultation issue led to both case law and interinstitutional convention. The case law is based on the principle that Parliament ‘must be consulted again whenever the text finally adopted, taken as a whole, differs in essence from the text on which the Parliament has already been consulted, except in cases in which the amendments substantially correspond to the wishes of the Parliament itself’.43,44

36

37 38 39 40 41 42 43

Council doc. R/2641/73 of 5 November 1973. The Council can work on the proposal at working group or COREPER level but, under Article 3 of the Protocol on the role of national parliaments in the European Union that was adopted in Amsterdam, is precluded from placing it on its agenda unless a six-week period has elapsed from the moment it was made available in all languages. E.g. in its Resolution on relations between the European Parliament and the Council, OJ (1990) C284/58 and in its Resolution on the obligation for the Council to await Parliament’s opinion, OJ (1990) C324/125. Case C-417/93, Parliament vs Council (TACIS), [1995] ECR I-1185, § 8. Ibid., § 10, following the very convincing reasoning of Advocate General Léger in his Opinion: [1995] ECR I-1185, §§ 45-49. Case 114/81, Tunnel Refineries Ltd vs Council, [1982] ECR 3189, § 18. Discussed in section 9.4, above. Draft report on the proposal for a Council decision adopting an action programme for administrative co-operation in the fields of external borders, visas, asylum and immigration (ARGO), Parliament doc. PE 311.004 of 25 January 2002 at 11/17. Case C-408/95, Eurotunnel SA a.o. vs SeaFrance, [1997] ECR I-6315, § 46. See also (inter alia) Chemiefarma, §§ 178-179, Buyl, § 23, C-388/92, Parliament vs Council (transportation services), [1994] ECR I-2067, § 15 and C-392/95, Parliament vs Council (Visas), [1997] ECR I-3213, § 15.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

255

It is difficult to present a general rule as to when a modification is so substantial that reconsultation is necessary.45 In the past, the Council set the standard at ‘questions of fundamental importance not yet submitted to the Parliament’.46 In TACIS the Court went into great detail to determine what the intention of the legislature was, analysing the text on an amendment-by-amendment basis and occasionally made almost political assessments.47 The Courts’ thorough examination of the text in these cases contrasts with the marginal character of the test applied in others: in some cases the Court of Justice ruled that the Commission’s amendments to the proposal did not ‘affect the very essence of the Regulation taken as a whole’, obviating the need for reconsultation.48 A change of legal base is by definition substantial. A change in comitology procedure can be, but not necessarily so.49 If the Council argues that the adopted text broadly corresponds to the MEPs’ wishes, it is up to that institution to prove this is so.50 The Council may not rely on the opinions of parliamentary committees before the adoption of a legislative resolution ending the consultation procedure.51 Interinstitutional convention and practice have given hands and feet to this case law. In Rules 54 and 55 of its Rules of Procedure, Parliament provided a procedure to monitor the follow-up of opinions to ensure that the other institutions keep their promises concerning the adoption of Parliamentary amendment and, if necessary, to request reconsultation. Since the MEPs are not themselves represented in the Council and its preparatory bodies, they extracted commitments from the Commission to ensure that Parliament’s prerogatives are safeguarded. In 1990, the executive committed to seize the Court of Justice if the Council failed to reconsult Parliament when it was legally obliged to do so.52 The question remained who was to judge whether an action should be brought. As Parliament argued in Chernobyl, it could not always rely on the Commission to plead its case.53 After the Court of Justice recognised the validity of this argument and allowed Parliament to defend its own prerogatives in Court,54 the interinstitutional convention was amended. The 1995 Code of conduct still provided that the Commission should ‘ensure’ that the Council reconsults, but no longer required the executive to take the Council to Court over it.55

44 45 46 47 48 49 50 51 52 53 54 55

The Council cannot refuse reconsultation on the basis that it is aware of Parliament’s wishes on the essential points of the proposal: Parliament vs Council (Visas), §§ 21–22. Jacobs, ‘Isoglucose resurgent: two powers of the European Parliament upheld by the Court’ (1981) at 225. Reply to written question 409/79 by Lord O’Hagan to the Council, OJ (1980) C27/3. For an example see case T-154/96, Chvatal a.o. vs Court of Justice, [1998] ECR SC IA-527, II-1579, § 160. Case C-280/93, Germany vs Council, [1994] ECR I-4973, § 42. TACIS, § 24-26. Case C-388/92, Parliament vs Council (transportation services), [1994] ECR I-2067, § 15. Ibid., § 17. Code of Conduct concluded between the Commission and the Parliament, OJ (1991) C280/165. Case C-70/88 Parliament vs Council (Chernobyl), [1990] ECR I-2041, § 6. Ibid., § 27. 1995 Code of Conduct, § 3.6. This provision subsequently found its way into § 7(ii) of Annex 1 to the Framework Agreement on relations between the European Parliament and the Commission, OJ (2001) C121/122.

256

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

10.5

The conciliation procedure interinstitutional agreement

contained

in

the

1975

The increase in Parliament’s powers following the Second Budgetary Treaty56 led to a realisation that the MEPs should have a bigger say on legislation, at least as far as it had budgetary consequences.57 Indeed, there had been a precedent for this. Following the first Budgetary Treaty, Parliament and Council agreed a procedure for the follow-up in the Council of Parliament’s opinion on acts having a financial impact.58 This procedure did not go much further than requiring the Council president to meet the president of Parliament and the rapporteur. As such, it involved no dialogue. The Council also confirmed its preparedness to explain to Parliament the reasons that had led it to disregard the opinion of the latter.59 In 1975 the Commission, Council and Parliament concluded a ‘Joint Declaration’ instituting the conciliation procedure, replacing the previous convention. The English ‘conciliation’ does not quite have the same flavour as the French ‘concertation’ and suggests a closer convergence of views.60 The aim of the conciliation procedure is ‘to seek an agreement between the European Parliament and the Council’.61 The following discussion will focus on its status; the conditions in which it can be invoked; its operation; the conciliation committee; and its potential to provide Parliament with more influence on legislation. Its status. The conciliation procedure is the pure product of interinstitutional convention. Parliament attempted the view that the interinstitutional agreement created legal obligations62 but both the Commission and Council63 held the opposite view, the latter describing joint declarations explicitly as ‘political undertakings’. There are two exceptions. Firstly, in 1981, Parliament made it clear it did not consider itself to be bound by a Financial Regulation on which it had not been consulted. Notwithstanding a political declaration inscribed into the Council’s minutes,64 this was a dubious legal position at best. The MEPs requested the Commission to indicate that the Financial Regulation can form the subject of a conciliation procedure.65 Ultimately, the Parliament managed to get text included in 56 57 58 59 60 61 62 63 64

65

1975 Treaty amending certain budgetary provisions of the Treaties establishing the European Communities and of the Treaty establishing a single Council and a single Commission of the European Communities, OJ (1977) L359/1. Vergès, ‘De quelques méthodes de développement institutionnel des Communautés européennes’ (1981) at 507. Council doc. R/2020/72 of 29 September 1972. On this Isaac, ‘La rénovation des institutions financières des communautés européennes depuis 1970’ (1975) at 723. For the interinstitutional conventions that preceded the conciliation procedure further Isaac at 728-729. Forman, ‘The conciliation procedure’ (1979) at 77. Joint declaration of the European Parliament, the Council and the Commission concerning the institution of a conciliation procedure between the European Parliament and the Council, OJ (1975) C89/1, § 6. Oosterman-Meulenbeld, ‘De concertatie-procedure en het Europese Parlement’ (1986) at 743. See the reply to question 170/77 by Mr Maigaard to the Commission, OJ (1977) C180/18 and the reply to question 169/77 by Mr Maigaard to the Council, OJ (1977) C259/4, respectively. ‘Le Conseil convient que la procédure de concertation prévue dans la déclaration commune de l’Assemblée, du Conseil et de la Commission, en date de 4.3.1975, sera appliquée pour l’adoption des règlements financiers’: Relevé des décisions des Conseil, 329ième session des 10/11.2.1975 (no doc. nr known). OJ (1981) C287/77. See also Resolution embodying the opinion of the European Parliament on the proposal from the Commission of the European Communities to the Council for a Regulation

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

257

the Financial Regulation, thereby enacting the conciliation procedure as far as amendments to that Regulation are concerned.66 Secondly, Article 13 of the Act concerning the election of the members of the European Parliament by direct universal suffrage67 comprises a procedure modelled on the Joint Declaration, again with the crucial difference that it grants a legal right to Parliament to invoke it. A Declaration attached to the Act specifies that especially the provisions of paragraphs 5, 6 and 7 of the procedure laid down in the joint declaration of the European Parliament, the Council and the Commission of 4 March 1975 are foreseen. The—open—legal question remains, which procedure applies if the interinstitutional agreement were to be revised. A procedure akin to the conciliation procedure found its way into the budgetary procedure.68 I will not further discuss these exceptions. The Joint Declaration can be invoked for ‘Community acts of general application’ which have appreciable financial implications, and the adoption of which is not required by virtue of acts already in existence (paragraph 2). At the time of its negotiation, the most important reasoning behind this formula was the exclusion of agricultural price developments.69 This wording does not necessarily encompass only regulations and directives, but can also include decisions establishing programmes of general application. Parliament’s Rules of Procedure do not mirror the requirement that the draft legislation should have appreciable financial implications, nor that its adoption must be required by virtue of acts already in existence. Rule 56(1) merely refers somewhat vaguely to ‘certain important Community decisions’. Under the Treaties, the conditions that must be fulfilled impede the full application of the procedure in policy areas with a lot of compulsory legislation (typically agriculture).70 In many other fields, Parliament acquired more extensive legislative rights under the Single European Act and the Treaties of Maastricht, Amsterdam and Nice. The conciliation procedure concomitantly decreased in importance. The ‘appreciable financial implications’ criterion has given rise to differences between the institutions and has not always been applied consistently. On the one hand, the Council refuses to open the procedure when, in its opinion, this condition is not met.71 Peculiarly, on the other hand, the very first application of the procedure

66

67 68

69 70 71

amending the Financial Regulation of 21 December 1977 applicable to the budget of the European Communities, OJ (1983) C277/64. Actually, the Council accepted this as long ago as in 1975: Annex V to Council doc. R/227/75 of 23 January 1975. The provision was inserted by Council Regulation 610/90 of 13 March 1990 amending the Financial Regulation of 21 December 1977 applicable to the general budget of the European Communities, OJ (1990) L70/1. See currently Article 184 of the Financial Regulation, OJ (2002) L248/1, corr. OJ (2003) L25/43. OJ (1976) L278/5. Interinstitutional Agreement between the European Parliament, the Council and the European Commission on provisions regarding financing of the Common Foreign and Security Policy— General Provisions, OJ (1997) C286/80. Interinstitutional convention in the budgetary sphere is further discussed in Chapter 14. Forman, op. cit. footnote 60 at 86. Grabitz a.o., Direktwahl und Demokratisierung—Eine Funktionenbilanz des Europäischen Parlaments nach der ersten Wahlperiode (1988) at 289. The Court of Justice’s judgment in case 260/80, Andersen a.o. vs Council, [1984] ECR 177, § 27 leaves at least the possibility open that the Council’s determination that no appreciable financial implications are at play, is challengeable in court.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

258

concerned the 1977 Financial Regulation, which did not itself have any budgetary implications.72 The Council has held that the procedure does not apply simultaneously with the co-operation procedure (discussed in section 11.2.1).73 Using the same logic, and notwithstanding an attempt by Parliament to this effect,74 the conciliation procedure is similarly not applied to matters covered by co-decision. Under Article 3 of the Joint Declaration, the Commission must indicate whether its proposal (the Joint Declaration curiously speaks of ‘the act’) is capable of being subjected to conciliation. In practice, this indication has little relevance since it is consistent practice that neither Parliament nor the Council is bound by the Commission’s opinion.75 This seems only logical, since the need to start the procedure will normally be felt by Parliament, and the need to refuse it by the Council. Although the interinstitutional agreement entitles Parliament, Commission76 and the Council each to initiate the procedure, it are normally the MEPs who do so by requesting in a legislative resolution that ‘the conciliation procedure [be] initiated should the Council intend to depart from the text approved by Parliament’. If the Council accepts that the procedure should be used, exploratory talks between the respective presidencies will be held. If these do not lead to a solution, the Council will not adopt the legislation, but merely a common position, which is forwarded to the Parliament. The conciliation procedure may then be formally opened at the request of any of the three institutions involved. Its operation. Following agreement between the institutions that the procedure did not really operate satisfactorily,77 on 17 December 1981 the Commission submitted proposals to improve it.78 The debate on its reform went on for several years and focused on its scope and on the chairmanship of the conciliation meetings. Notwithstanding the intention of the European Council that a new interinstitutional agreement be concluded,79 by 1984 this still proved elusive.80 Parliament then agreed with the Council some ‘pragmatic’ improvements, which implied the introduction of exploratory contacts between the presidents of Parliament, Commission and Council to ensure a harmonious course of the procedure. Moreover, a permanent interinstitutional working group was set up, tasked to serve as the secretariat of the conciliation committee. Parliament and the Council agreed on that occasion that, if 72 73

74 75 76

77 78

79 80

Läufer, ‘Das Europäische Parlament nach der Direktwahl: Positionsstärkung durch intrakonstitutionellen Wandel?’ (1979) at 263. Council conclusions of 13 November 1987, quoted in Digest of texts on relations between the European Parliament and the Council (1988) at 60. From the side of Parliament hopes had been cherished in the other direction. In this sense Bieber, Pantalis and Schoo, ‘Implications of the Single Act for the European Parliament’ (1986) at 786. Resolution on the conciliation procedure, OJ (1993) C21/138. Kapteyn and VerLoren van Themaat, Inleiding tot het recht van de Europese Gemeenschappen— deel 1A (1987) at 185. The Jonit Declaration does not in so many words provide such a right to the Commission. However, the Council accepted that the executive can initiate conciliation: Annex II to Draft joint declaration by the European Parliament and the Commission on the setting up of a conciliation procedure, Council doc. R/3556/74 of 17 December 1974. Jacqué, op. cit. footnote 18 at 679. Council doc. 5749/1/82 of 2 April 1982, Commission communication The institutional system of the Community—restoring the balance, Bull. EC Suppl. 3/82 at 18 and Commission communication to the Council and the European Parliament on the conciliation procedure, Commission doc. COM (81) 816. Solemn Declaration on European Union, § 2.3.6. Bieber, ‘Legal developments in the European Parliament’ (1985) at 345.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

259

differences of opinion between the institutions are likely, exploratory contacts should take place between the presidencies of the three institutions. At the time of the adoption of Parliament’s opinion the Council would adopt an ‘orientation’ which it would send to Parliament. The conciliation procedure can be opened at the request of either the Council or Parliament. As part of the 1984 improvements, Parliament proposed81 that the three presidents of the institutions concerned would establish contacts and an interinstitutional high-level group to ensure the harmonious conduct of the procedure. These arrangements had little direct impact in the context of the conciliation procedure but were important predecessors of the trialogue procedure that now characterises co-decision. Following the 1984 changes, the procedure now operates as follows. The first official step in the procedure is the transmission by the Council of its ‘position’ to Parliament. Following the subsequent adoption of Parliament’s opinion on this text, the conciliation committee meets. This consists of the Council and representatives of Parliament. The Commission participates in its work. If the differences can be sufficiently narrowed, Parliament adopts a new opinion. The Council subsequently adopts the measure. In principle, the whole conciliation process should not take more than three months and two meetings.82 Subsequently, Parliament is given a chance to adopt a revised opinion. If, however, no agreement is reached, the Council can adopt the legislation, taking account ‘as far as possible’ of Parliament’s views. The Council agreed to inform Parliament of the reasons for departing from the MEPs’ opinion.83 The idea of a conciliation committee is largely borrowed, on the one hand, from the French institution of the navette, a conciliation committee tasked to bridge the differences between the French Assemblée nationale and the Sénat,84 and on the other hand from the Vermittlungsausschuß that works out compromise texts in cases where the German Bundestag and Bundesrat disagree. The latter committee consist of one member for each Land and an equal number for the Bundestag. Whilst in Germany the conciliation committee procedure is mainly invoked by the representatives of the Länder,85 this is precisely the reverse in Europe, reflecting the reverse political power relationship between the directly elected representatives and the governments of the states. The composition of the conciliation committee is essentially similar to that of the Conciliation Committee in the co-decision procedure (discussed in sub-section 11.4.4, below).86 The Commission is also represented. Parliament’s delegation must reflect the composition of the institution in political groups. The parliamentary delegation to the conciliation committee decides by majority of its members. All members of the Council may take part in the discussions of the Conciliation Committee,87 although typically, the discussion from the side of that institution is led by the presidency. During the meetings of the conciliation committee the Secretariat81 82 83 84 85 86 87

Parliament’s position on the reform is published in OJ (1984) C10/32. Section 4.01.012 of the Digest of texts on relations between the European Parliament and the Council (1988) at 48-51. Forman, op. cit. footnote 60 at 79. According to him, this commitment is stronger if the act concerned has financial consequences. Prélot and Boulouis, Institutions politiques et droit constitutionnel (1984) at 819 et seq. Burkens and Van Schagen, ‘The Federal Republic of Germany’ (2004) at 333. As to this latter precedent Foster, ‘The new Conciliation Committee under Article 189b EC’ (1994) at 191–193. Rule 64 of Parliament’s Rules of Procedure. Council doc. R/3556/74 of 17 December 1974.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

260

General of the Council makes notes which, before formal approval by the Council, are sent to the Parliament for its agreement. For access to documents purposes, such notes are treated similarly to minutes of Council meetings.88 Its potential to provide Parliament with more influence on legislation. Conciliation has at times made a difference. For example, during the conciliation for the 1977 Financial Regulation Parliament managed to obtain agreement on the principle that the (current) Article 272 TEC be applied to commitments and to appropriations separately.89 However, although expectations were initially high on the Parliamentary side, as a vehicle to influence legislation, the conciliation procedure has disappointed in other cases. 90 Only a limited number of procedures were completed at all: until June 1984 in total 46 procedures were started but the conciliation committee met only 17 times.91 Moreover, in the period up to 1985, Parliament only achieved limited results. Apart from the Financial Regulation, the only tangible success seems to have been the reform of the European Social Fund in 1983. Indications are that, since 1984, Parliament has not fared better. This altogether less than excellent success rate has different causes. Firstly, requests to open it were honoured only in a minority of cases. In the period 1979-1982 this occurred with merely twice out of twelve requests.92 The main reason for this lies in a difference of interpretation between the MEPs and the Council concerning the conditions for invoking it.93 The main difference centred on the financial implications criterion.94 Secondly, the Council’s common position was often itself a compromise leaving that institution with very little margin for concessions, 95 presented by a carefully mandated Council president.96 Consequently, Parliament rarely was engaged in a genuine dialogue with the Council.97 Where the matter is covered by unanimity decision-making, the Council has to adopt its common position in the conciliation procedure also unanimously. Fast and efficient negotiations require that the institution speak with one voice. Where the Parliament can achieve that through voting and the Commission through collective responsibility, the Council has an intrinsic difficulty here since it is not ruled by hierarchy or voting in such cases. This effectively forces Parliament to negotiate with a large number of parties, reducing the compromise opportunities even further. The shadow of the ‘Luxembourg Compromise’, which was very much a factor in the years after 1975, compounded to the Council’s difficult position.

88 89 90 91 92 93

94 95 96 97

Reply to written question 1158/92 by Lord O’Hagan to the Council, OJ (1992) C285/29. Notenboom, Het Europees Parlement en de financiën (1988) at 81. Glaesner, ‘Willensbildung und Beschlußverfahren in der Europäischen Union’ (1994) at 29, Notenboom, op. cit. footnote 89 at 27. Parliament shares this assessment: Resolution on the conciliation procedure, OJ (1989) C69/151, § A. Grabitz a.o., op. cit. footnote 70 at 290. Oosterman-Meulenbeld, op. cit. footnote 62 at 742-743. The first elected Parliament asked for participation in all Community legislation: Resolution on the draft second Joint Declaration of the European Parliament, the Council and the Commission on the conciliation procedure, OJ (1984) C10/34. As noted above, Rule 56(1) of Parliament’s Rules of Procedure draws the scope of the procedure much larger than the Joint Declaration. Jacqué, op. cit. footnote 18 at 680. Oosterman-Meulenbeld, op. cit. footnote 62 at 743. Dankert, ‘“Niets dan het Verdrag”’ (1981) at 541. Commission communication The institutional system of the Community—restoring the balance, Bull. EC Suppl. 3/82 at 18.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

261

The third cause for the limited success rate—and this was a structural defect in the procedure— is the lack of incentive for the ministers to engage in the procedure. Even if Parliament wishes to interpret the application conditions broadly, it remains the Council’s call whether the procedure is used. The idea that the conciliation procedure could evolve to some kind of colegislative right98 was therefore flawed from the outset. Separation of institutions implies the impossibility of unilateral amendment of the parameters of the political system. No single ‘power’ ought to be able to amend the constitution. Such an issue is here at play: the Council retains the possibility of stopping the conciliation process, or not even entering into the procedure at all. In other words: notwithstanding the interinstitutional convention, conciliation remains a favour granted by the Council rather than a right acquired by Parliament.99 And this considerably limits the direct constitutional value of the procedure. 10.6

Member State initiatives

Title VI TEU, comprising police and judicial co-operation in criminal matters, foresees a procedure under which Member States can initiate legislation.100 A few interesting interinstitutional issues have cropped up, concerning the consultation procedure, and the subject matter of the consultation. The consultation procedure. It is custom that Member States intending to propose an initiative first contact other Member States and the Commission to ascertain what other initiatives are being prepared. Subsequently, normally a discussion paper is submitted to the Council outlining the proposed legislation. This does not give rise to an in-depth discussion, since premature political agreement could impair Parliament’s rights. In principle, the normal consultation procedure applies under Article 34 TEU, with the exception of ‘common positions’ adopted under Article 34(2)(a). Contrary to consultations under the EC Treaty,101 the Council may set a deadline for the consultation which, however, must be at least three months.102 It appears that Parliament would like the Council to express its opinion on the Parliament’s amendments before the vote in Parliament. This would be more or less in analogy to the procedure followed when it is consulted on a Commission proposal, which allows Parliament to refer the matter back to the relevant committee, if necessary. The Council does not favour such an approach. The subject matter of the consultation. A question has arisen exactly on what Parliament is consulted. In 2001, Mr Corbett MEP complained that in one case the Council had not consulted Parliament on a French initiative, but rather on the Council’s position on it.103 Indeed, an argument could be attempted that consultation 98 99

100 101 102 103

A view which may be found in Bull. EG Suppl. 3/82 at 11. This seems to be also the line of the Court of Justice, which pronounced that ‘the sole fact that the Parliament had requested the initiation of the conciliation procedure [in vain] is not capable of affecting the legality’ of the legislation concerned: case 262/80, Andersen a.o. vs Parliament, [1984] ECR 195, § 25. Title IV TEC similarly comprised a possibility for Member State initiatives, but this expired on 1 May 2004. The only exception in the EC Treaty is the Council’s right to lay down the deadline in the context of international agreements. Note that Article 300(3) TEC does not mention the three-month restriction. Article 39(1) TEU. Mr Corbett’s point was made during Parliament’s sitting of 12 March 2001. This concerned the

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

262

on a text already politically agreed by the Council reduces the real value of the consultation exercise.104 Contrariwise, it can be argued that Article 39(1) TEU leaves it undecided on what Parliament is consulted. Consultation on the Council’s position has the benefit of giving the MEPs a somewhat more accountable counterpart. The Council has committed itself to appear in Parliament. The Member States have not, although in practice Member State representatives do appear in committees to explain their initiatives. The problem appears to have been restricted to at least no more than a minority of Member States initiatives. Under current practice, the Member State initiative is translated and finalised by the Council’s legal linguists and is subsequently sent to Parliament. The Council’s Article 36 Committee (in its French acronym CATS) agreed that ‘Council should ensure that the European Parliament is given sufficient time for rendering opinions and that it is consulted as early as possible on new initiatives’.105 The Constitution’s authors suggested to continue Member State initiatives, but with two major changes. Firstly, they proposed that a quarter of the Member States be needed for such an initiative. Secondly, for the remainder, the legislative procedure to be followed is the ordinary procedure (co-decision).106 Since a quarter of the Member States implies at least seven, it is unlikely that the Member State initiative will find widespread use in the future. It is therefore unlikely to generate much interinstitutional convention. 10.7

Conclusions of this Chapter

Consultation can—and often does—have practical effects. For example, in December 2000 Parliament was consulted on a French initiative to set up a European judicial training network.107 Following the negative opinion from the MEPs (and, admittedly, some resistance by some Member States) the proposal was withdrawn and an alternative text submitted.108 However, notwithstanding this example, the value of the consultation right has traditionally not been estimated as very high, with some MEPs complaining that the Council tends to ignore their carefully drafted opinions.109 Empirical research bears out that the right of consultation is not an important means of influence. 110 There are several reasons for this. Firstly, in the view of Van Schendelen this is to a large extent inherent in the working methods of the Council that often most political agreements are concluded at the working party level. In many cases this happens at least in part well before

104 105 106 107 108 109 110

Initiative of the French Republic with a view to adopting a Council directive on mutual recognition of decisions on the expulsion of third country nationals. The logic of Advocate General Léger in case C-417/93, Parliament vs Council (TACIS) (Opinion), [1995] ECR I-1185,, § 49, seems to support this position (albeit it in a first pillar context). Annex to the Outcome of proceedings from the Article 36 Committee to COREPER concerning the working methods in the JHA area, Council doc. 10336/01 of 26 June 2001 at item 5. See Articles I-34 and III-264 Const, respectively. OJ (2001) C18/9. Working document on the Initiative of the French Republic with a view to the adoption of a Council Decision setting up a European Judicial Training Network Committee on Citizens’ Freedoms and Rights, Justice and Home Affairs, Parliament doc PE 311.008 of 1 February 2002. Dankert, op. cit. footnote 96 at 541. Grabitz a.o., op. cit. footnote 70 at 432. For data on perception by the MEPs covering the first elected Parliament see Van Schendelen, Het Europees Parlement (1984) at 80.

CHAPTER 10: PARLIAMENT’S RIGHT OF CONSULTATION

263

Parliament’s opinion has been received.111 To the extent that this is true, it is a systemic matter inherent in the set-up of the institutions. Secondly, sometimes Parliament’s opinion may come at a point in the negotiations between the Member States when it is politically very difficult for them seriously to take account of it. For example, during the negotiations of the 1994 industrial GSP Regulation,112 Parliament duly submitted a (non-compulsory) opinion,113 which, however, was adopted when the Member State representatives and the Commission were close to finalising the new Regulation. Even if the Member States had wanted to take account of Parliament’s views where these differed from their own, this would have been difficult since that opinion was rendered at a very late stage in the legislative procedure. In the end, Parliament’s opinion was largely a wasted effort and a missed opportunity. Thirdly, sometimes Parliament rejects a proposal outright. This puts an obligation on the Commission to consider withdrawing it. However, this may not always be a practical option and often it will be unable to do so. The Council has in such cases dispensed with its obligations towards Parliament and is free to adopt the legislation without considering any amendment from the MEPs. The conciliation procedure has not been able to increase directly the value of Parliament’s prerogatives. The Council is often not bound or simply not interested in conciliation; moreover, the procedure allows Parliament little means to force the ministers to negotiate a text which they are legally capable of adopting without adopting Parliament’s amendments. Indeed, both as a check on the other institutions and as a possibility to engender input legitimacy, the consultation procedure does not score very high. However, this is not to say that the conciliation procedure was useless. Rather, its main importance lay in its mere existence. It prepared the minds for the co-operation procedure and, later, co-decision. The idea of the conciliation committee, first developed with moderate success in the context of the conciliation procedure, found its way into co-decision. To sum up, in Chapter 1 of this study the question was asked to what extent it is possible to balance the EU institutions through the use of interinstitutional convention, and what the conditions are for doing this. Whilst there is a considerable body of interinstitutional convention in the field of the right consultation, its direct effect on the balance of influence between the institutions appears to be limited. On the other hand, its indirect consequences (influence on the Treaty) are more notable.

111 Van Schendelen, ‘”De Raad beslist”. Beslist de Raad?’ (1995) at 713. 112 Council Regulation (EC) 3281/94 of 19 December 1994, applying a four-year scheme of generalised tariff preferences (1995-1998) in respect of certain industrial products originating in developing countries, OJ (1994) L348/1. 113 Resolution on the Commission Communication to the Council and the European Parliament on the integration of developing countries in the international trading system—role of the GSP 1995-2004, OJ (1994) C341/243.

CHAPTER 11: CO-DECISION

11

CO-DECISION

11.1

Introduction to this Chapter

265

Article 251 TEC lays down what—notwithstanding successful British pressure to keep the term out of the Treaties1—is commonly known as the co-decision procedure. The procedure was established with the Maastricht Treaty and amended by the Amsterdam Treaty. The ‘ordinary legislative procedure’ proposed in Article III-396 Const is a slightly amended version of it. This Chapter investigates interinstitutional convention in the field of co-decision. Since some of the conventions that govern that procedure originally came into being in the context of the (now almost defunct) co-operation procedure, the following section commences with some brief comments on that subject. This is followed by an analysis of the growth of conventions in the field of co-decision (section 11.3). Then the practice of co-decision is reviewed in section 11.4. 11.2

Development of the co-decision procedure

11.2.1

Interinstitutional convention and the co-operation procedure

The co-operation procedure, inserted into the EC Treaty by the Single European Act, was largely replaced by the co-decision procedure through the Maastricht, Amsterdam and Nice Treaties. Currently, the procedure, which is laid down in Article 252 TEC, is only called for by Articles 99(5), 102(2), 103(2) and 106(2) TEC, all of which concern economic and monetary policy. During the 1996 IGC there was support for the view that the procedures for the EMU should not be changed,2 resulting in this remaining the sole field were co-operation still is used. In view of its currently limited use, the discussion of the interinstitutional convention engendered by it can be brief. The procedure is built upon the consultation procedure discussed in Chapter 10. It starts with the Council’s consultation of Parliament, essentially copying the consultation procedure.3 The procedures developed pursuant to the Roquette case law discussed in section 10.3 can play their full part.4 A difference with the consultation procedure is that the Council adopts a ‘common position’, which is communicated to Parliament. Council and Commission inform the MEPs fully of the reasons leading the Council to adopt it and of the Commission’s position on it. Parliament then has three months to act, which raises the question from which point in time this deadline should be calculated. On the one hand, some in Parliament suggested that the proper starting point should be the first meeting of the 1 2 3 4

Shackleton, ‘The politics of codecision’ (2000) at 339-340. Report from the presidency to the European Council—Progress report on the Intergovernmental Conference, IGC doc. CONF/3860/96 of 12 June 1996 at 21. Dashwood, ‘Community legislative procedures in the era of the Treaty on European Union’ (1994) at 347. Lenaerts, ‘Some reflections on the separation of powers in the European Community’ (1991) at 16. This includes, for example, sending back the file to committee in cases where the Commission does not compomise sufficiently: Pennera and Schoo, ‘La codécision—dix ans d’application’ (2004) at 537.

CHAPTER 11: CO-DECISION

266

Parliamentary committee following communication of the common position by the Council. The Commission takes a similar approach.5 On the other hand, at a meeting on 29 March 1987 the three presidents of Parliament, Commission and Council agreed that the deadline would start running on the date when Parliament receives the common position. In addition, the Council agreed that common positions would be sent to Parliament with due regard to the dates of its part-sessions—not a minor matter, since a common position received just before the Summer holidays would hardly enable the MEPs to take a position. Half a year later COREPER concluded that ‘[t]he three-month time-limit . . . starts to run as soon as a common position has been lodged, and cannot be suspended’.6 Presumably, in cases where a hiatus would occur between sending the common position and receiving it, ‘lodged’ must be interpreted as ‘logged by the Parliament’s registry’, in other words, ‘received’. If Parliament approves the common position or fails to act, the Council definitively has to adopt the act in question in accordance with the common position. Alternatively, Parliament can adopt amendments or reject the common position with an absolute majority, in which case the Council needs unanimity in second reading. Parliament amended its Rules of Procedure so that the Commission had to express its opinion on the amendments before the vote on the rejection would take place. In this manner, pressure on the Commission was maximised to adopt Parliament’s amendments.7 Rejections have always been rare,8 partly because it is not always easy to muster the required absolute majority,9 partly because Parliament is often reluctant to be made responsible for the échec, and partly because often no legislation is worse than bad legislation. A rejection can be overturned by the Council with unanimity. Following Parliament’s second reading, the Commission has one month to reexamine ‘the proposal on the basis of which the Council adopted its common position’. Article 252 does not clarify what the consequences are if this deadline is missed. The re-examination is limited to ‘taking into account the amendments proposed by the European Parliament’. It is implicit in the text of Article 252(d) that the ‘re-examination’ involves an amendment of its proposal by incorporation of the amendments it can live with. Subsequently, this becomes the basis for the Council’s discussions.10 Notwithstanding the fact that Article 250(2) allows the Commission to amend its proposals,11 this provision does not apply at this stage in view of the lex specialis of Article 252(d).

5 6

7 8 9 10 11

Procédure de coopération—procédure à suivre pour approuver la position de la Commission sur les amendements parlementaires et transmettre des déclarations sur les positions communes, Commission doc. SEC (89) 1364 of 20 July 1989 at 3, § 14. Exchange of views in the Permanent Representatives Committee on the co-operation procedure: Council conclusions (Brussels, 13 November 1987), Council doc. SN 3743/87, quoted in Secretariat General of the Council, Digest of texts on relations between the European Parliament and the Council (1988) at 60-61. Corbett, The European Parliament’s role in closer EU integration (1998) at 264. This occurred seven times. In three cases the Council overturned it and adopted the legislation nevertheless: joint answer to written questions E-1860/96 and E-1861/96 by Ms Schleicher to the Commission, OJ (1996) C356/106. Raworth, ‘A timid step forwards: Maastricht and the democratisation of the European Community’ (1994) at 20. Jacqué, ‘L’Acte unique européen’ (1986) at 592 Before the Treaty of Maastricht, Article 149(3) TEEC spelt this out explicitly. In Maastricht the Treaty was reorganised but—to my knowledge—the new text was not intended as a restriction of the current Article 250(2) in the co-operation procedure.

CHAPTER 11: CO-DECISION

267

The Commission forwards to the Council, at the same time as its re-examined proposal, Parliament’s amendments which it did not accept and expresses its opinion on them. In the 1990 Code of conduct12 the Commission committed itself to giving ‘special consideration’ to Parliament’s amendments adopted in second reading ‘which it does not intend to incorporate in its reviewed proposal’ and to explain its decision before the next Parliamentary committee.13 The Commission was unwilling to accept any general commitment to withdraw its proposals if Parliament rejected them or if the Council adopted a proposal without taking on board all of Parliament’s amendments.14 Similarly, it refused to adopt automatically Parliament’s amendments. From a separation of institutions point of view, this stance is sensible. The Commission is autonomous as an institution. Its responsibility to Parliament does not equal slavish dependence. The Council may adopt the revised proposal with a qualified majority or may amend it with unanimity. The value of the procedure lies in the possibilities it affords when Parliament can coalesce with one or more Member States to avoid unanimity and force the Council to act with qualified majority. At the time of the inception of the co-operation procedure, some presumed that unanimity would be seldom reached in the Council,15 giving Parliament a considerable advantage in the procedure. Often this proved to be wishful thinking: even when the Luxembourg Compromise no longer poisoned decision making, the convention that the Council should try to reach as large a consensus as possible still has its effects. In practice, this somewhat impaired the full operation of the qualified majority option. The Council must act in second reading within three months or see the legislation fail. This period can be extended with one month by common accord between the Council and Parliament. Such a ‘common accord’ clearly has a legal basis and is sufficiently concrete to be legal. It is therefore no interinstitutional convention. The Single European Act spelled out the co-operation procedure in great detail, but this did not suffice to establish a true interinstitutional dialogue. The gap was soon filled with convention. As early as 1987, COREPER decided that, throughout the cooperation procedure, the Council presidency should keep in touch with the various relevant officers in Parliament (president, committee chairmen and rapporteurs). The presidency was also charged with ensuring that Parliament was fully informed of the Council’s reasons for the common position including, ‘[a]s far as possible’, ‘the grounds for rejecting/modifying amendments proposed by the European Parliament and accepted by the Commission’. However, ‘[w]hen appearing before Parliamentary Committees and when replying to written or oral questions, the Presidency will not refuse to explain reasons but will stress that those given in writing are full and final’.16

12 13

14 15 16

Code of conduct on improving interinstitutional relations, OJ (1991) C280/165. The Commission concluded that it was under ‘un engagement politique’ to defend Parliamentary amendments adopted by it with conviction. Any decision to abandon such amendments needed to be vetted by the College: Vade-mecum institutionnel (1990) at 51. See also the 1990 Code of Conduct, § 2. In half of the cases,, the Commission actually accepted second reading amendments: Commission doc. SEC (91) 1097 final of 5 June 1991 at 5. Debate of the European Parliament of 13 February 1990, OJ (1990) 3-386/30. In this sense Bieber, Pantalis and Schoo, ‘Implications of the Single Act for the European Parliament’ (1986) at 782. Exchange of views in the Permanent Representatives Committee on the co-operation procedure: Council conclusions (Brussels, 13 November 1987), Council doc. SN 3743/87, quoted in Secretariat General of the Council, Digest of texts on relations between the European Parliament and the Council (1988) at 60-61.

CHAPTER 11: CO-DECISION

268

At least in sensitive cases, the three institutions developed a culture of interinstitutional dialogue. In practice this procedure, which applied to about one-third of all legislation considered by Parliament between the entry into force of the SEA and the Maastricht Treaty,17 worked well, although it fell short of at least Parliament’s expectations for several reasons. Firstly, the co-operation procedure was not geared towards pushing legislation against the will of the Council, although it strengthened Parliament’s hand to stop or amend legislation that the Council did want. This worked especially well when the Council was divided and Parliament could force the choice between legislation supported by the MEPs and no legislation at all. For example, Par