Regulation through Agencies in the EU: A New Paradigm of European Governance 1845422678, 9781845422677

The past decade has witnessed a proliferation of regulatory agencies at both the national and the EU level. This coheren

253 82 16MB

English Pages 296 [304] Year 2006

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Regulation through Agencies in the EU: A New Paradigm of European Governance
 1845422678, 9781845422677

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Regulation through — gencies in theE A New Paradigm of European Governance

Digitized by the Internet Archive in 2022 with funding from

Kahle/Austin Foundation

https://archive.org/details/regulationthrougO000unse_p9d0

wim

«=|GLY NTAFF

REFERENCE ONLY

Regulation through Agencies in the EU

Regulation through Agencies in the EU A New Paradigm of European Governance

Edited by Damien Geradin

Professor of Law, University of Liége and College of Europe, Belgium

Rodolphe Munoz Research Fellow, University of Liége, Belgium

Nicolas Petit Research Fellow, University of Liége, Belgium

Edward Elgar Cheltenham, UK + Northampton, MA, USA

© Damien Geradin, Rodolphe Mufioz and Nicolas Petit 2005 All right reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of publisher. Published by Edward Elgar Publishing Limited Glensanda House Montpellier Parade Cheltenham Glos GL50 1UA UK

aid

Edward Elgar Publishing, Inc.

Z*

136 West Street



Suite 202

Northampton Massachusetts 01060

USA

soiee

(ett ft

Or 4S

_ &an? ee ae

nie

12998850

A catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication Data Regulation through agencies in the EU : a new paradigm of European governance / edited by Damien Geradin, Rodolphe Mufioz, Nicolas Petit.

p. cm. ISBN 1-84542-267-8 1. Administrative agencies—European Union countries. 2. European Union. I. Title: Regulation through agencies in the European Union. II. Geradin, Damien. III. Mufioz, Rodolphe, 1972- IV. Petit, Nicolas, 1978-

JN32.R44 2005 351.4—de22 2005049463

ISBN 1 84542 267 8 Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall

Contents

List of Figures and Tables List of Contributors Editors’ Preface PART I: RATIONALES FOR SETTING UP AGENCIES

A

The Politics of Regulation in the European Union Paul Magnette Regulation of Liberalised Markets: A New Role for the State? (or How to Induce Competition Among Regulators) Phedon Nicolaides

2

PART II: THE DESIGN OF AGENCIES IN THE EU

SF

Independent Regulatory Agencies and Elected Politicians in Europe Mark Thatcher Agencies for European Regulatory Governance: A Regimes Approach Colin Scott Delegation to EU Non-majoritarian Agencies and Emerging Practices of Public Accountability Deirdre Curtin Independence, Accountability and Transparency of European Regulatory Agencies Ellen Vos

47

67

88

120

PART III: CHALLENGES RAISED BY AGENCIES IN THE EU 7.

Good Governance and European Agencies: The Balance Michelle Everson

141

vi

Contents

8.

Coordination of European and Member State Regulatory Policy: Horizontal, Vertical and Transversal Aspects Pierre Larouche 9. The Proliferation of National Regulatory Authorities alongside Competition Authorities: A Source of Jurisdictional Confusion? Nicolas Petit

164

180

PART IV: AGENCIES IN A COMPARATIVE AND INTERNATIONAL PERSPECTIVE 10.

11.

Index

The Development of European Regulatory Agencies: Lessons from the American Experience Damien Geradin Regulation and Globalization: Interactions between International Standard-setting Agencies and the European Union Jan Wouters and Sten Verhoeven

pAie

246

274

Figures and Tables FIGURES 2:1 a2

Convergent Regulation Divergent Regulation

38 39

TABLES 1.1 pe pm: 3.1

= Me 3.4 oe 3.6 5.1 ME

The Causes and Possible Options which make the Politics of Delegation Shifting Regulatory Concerns Regulatory Centralisation vs Decentralisation Independent Regulatory Agencies for Market Competition in Britain, France, Germany and Italy in Selected Domains Party Activism and Public Affiliations of IRA Members 1990-2001 Resignations of IRA Members before End of Term 1990-2001 Party Activism and Public Affiliations of Communications and General Competition of IRA Members 1990-2001 Resources of IRAs 2000 Average Tenure of Senior Members of General Competition Authorities 1990-2001 Delegation to Non-EU Majoritarian Agencies Overview of Potential Divergences between NRAs and Competition Authorities Approaches

Vil

106 187

Contributors

Deirdre Curtin, Professor of Law at the University of Utrecht Michelle Everson, Lecturer in European Law at Birkbeck College, University of London

Damien Geradin, Professor of Law at the University of Liege and at the College of Europe, Bruges Pierre Larouche, Professor of Law at the University of Tilburg, Director of AMOUKE

Paul Magnette, Professor of Political Science at the Free University of Brussels, Director of the Institute of European Studies Phedon Nicolaides, Professor of Economics at the European Institute of Public Administration, Maastricht

Nicolas Petit, Doctoral Candidate and Assistant at the University of Liége, Institute for European Legal Studies Colin Scott, Reader in Law at the London School of Economics Mark Thatcher, Senior Lecturer in Public Administration and Policy at the London School of Economics

Sten Verhoeven, Assistant at the Institute for International Law, Katholieke Universiteit Leuven (KUL)

Ellen Vos, Professor of Law at Maastricht University Jan Wouters, Professor of International and European Law at the Katholieke Universiteit Leuven (KUL), Director of the Institute for International Law Vill

Editors’ Preface

The past decade has witnessed an unprecedented proliferation of regulatory agencies in the EU. This proliferation can be observed at the national level where a large number of regulatory authorities have been created in response to the liberalisation directives adopted in the field of network industries (telecommunications, energy, rail etc.) as well as in response to domestic policies in a variety of economic and social areas. It is also witnessed at the EU level where the number of agencies has increased from four in 1993 to sixteen in 2004. These agencies cover a heterogeneous range of subject matters, from vocational training to environmental, maritime and food safety regulation. Only in the past two years, the Commission has proposed the setting up of three new agencies. There are reasons to believe that this trend is gathering pace. Indeed, the Commission tends to see agencies as a positive evolution in the EU institutional machinery. As explained in the White paper on European Governance adopted in 2001: ‘The creation of further autonomous EU regulatory agencies in clearly defined areas will improve the way rules are applied and enforced across the Union’. The setting up of agencies offers a number of advantages over the regulatory methods employed previously in the EU: increased expertise on highly technical issues, enhanced visibility for the regulated sectors and substantial cost savings for the regulated entities. In addition, the creation of agencies may help lighten the workload of other EU institutions, which can, in turn, focus on their core strategic functions.

However, the establishment of national regulatory authorities and European agencies through the impetus of EU law leaves a large number of legal, political and economic questions unanswered. In light of the above developments, the objective of the book is to take a closer look at the complex issues raised by the proliferation of regulatory agencies in the EU through an interdisciplinary approach. It gathers papers written by highly esteemed experts in the field of economics, law and political science. The book is organised in four parts. The first part analyses the rationales 1s

Xx

Editors’ Preface

underlying the setting up of agencies at both national and European levels. The second part concentrates on the design of regulatory agencies in the EU. The third part identifies a number of challenges raised by the setting up and functioning of agencies. Finally, the fourth part compares the current EU developments with the US experience as well as with other agencies set up at international level. This book constitutes the final outcome of a conference organised on 18-19 March 2004 by the Institut d’Etudes Juridiques Européennes (IEJE) of the University of Liége with the financial support of the following institutions: Interuniversitary Poles of Attraction Programme P4/04, Belgian State, Prime Minister’s Office, Federal Office for Scientific, Technical and Cultural Affairs; Fonds National de la Recherche Scientifique (FNRS); French Community of Belgium.

We would like to thank here all the speakers and panels’ chairs for their active participation. In particular, we would like to mention here: Giandomenico Majone, Bart Kiewiet, Ludwig Kramer and Leigh Hancher. Finally, we want to express our warm gratitude to Delphine Vannerum, Stéphanie Hanson, Vassiliki Mitrias and David Henry for their involvement in the organisation of the conference as well as the editing process. Damien Geradin, Rodolphe Mufioz, Nicolas Petit

Liege March 2005

PART ONE

Rationales for Setting up Agencies

Ow

:

Wer me Sh

red, yee ta

aie

mi ae

ie)

wal: ss

2a

vy

a

eer

oe nia

(eee

se er

.

or

hae eat

we wae

May

aa

hante

| ?

nO

Ea

me

7

whad a, bw or), optibug

Ho wii, thy Bett (po) Seavert ;

her

SET

wT

ee

ae)

i

mt

asics

im Weed

ew

a et

(ieee

uri’ ae

ae

7

Re

Pra th! “ee

TWitae Y

zISemel)

Fi a Oi che

§

7

1.

The Politics of Regulation in the European Union Paul Magnette

INTRODUCTION Why should a public authority regulate a given policy area, and how should it do this? This question, which has been a key concern of democratic politics since its inception, has given rise to a new flow of academic literature since the 1980s, for at least two major reasons: first, most Western states have since then

been redefining the scope of their interventions on the markets; second, the emergence of international regimes at the global and regional levels has given rise to a new generation of public authorities whose status and articulation to national authorities need to be clarified. Using analytical tools drawn from contract theory, political scientists have endeavoured, in the last decade, to explain why and how public authorities choose to regulate a given field, and to examine the consequences of these political decisions in terms of public control and accountability. The issue of the ‘legitimacy’ of new forms of regulation has become, in the recent years, one of the main concerns of this research. The idea that new forms of regulation call for alternative modes of legitimation is widespread in these discussions. Yet, most authors acknowledge that the principles and concrete forms of this alternative remain to be defined. This chapter aims at showing how these questions emerged on the European agenda, and how the ‘dilemma of accountable independence’ can be addressed. Section one recalls the reasons why governments decide to regulate a given field, and how they choose between national or supranational agencies, and between different types of agents. Section two shows. that when they set up agents, the governments also have to deal with the consequences of this decision in terms of public control and accountability: these are the reverse side of the ‘politics of regulation’. Section three concludes that, given its broader impact

3

4

Regulation through Agencies in the

EU

on EU politics, the discussion on the delegation to regulatory agencies cannot be divorced from broader reflections on the EU’s institutional evolution.

WHY AND HOW REGULATE ? The question of the state’s intervention on the market has been the major cleavage of democratic politics in the twentieth century, leading to a quite stable opposition between a non-interventionist right and a statist left.’ In the first half of the century, the debate was cast in different terms in Europe and the US, and the two conceptions only converged in the 1980s. Initially, in Europe, the advocates

of public intervention did not reason in terms of

‘regulation’: as it opposed a public sphere governed by equal rights, to a private market ruled by unequal wealth, the worker movement aimed at transferring some of the market’s functions to the public realm, and to replace private rules by public laws.’ In the US, on the contrary, in the absence of a strong worker movement, the ‘left’ was dominated by a more liberal vision,

less hostile to the capitalist economy. While supporting stronger public interventions on the market, the New Dealers saw the state as a corrective, not

a substitute to the market. Such a reasoning became widespread after the 1929 crisis, and legitimised the government’s wide programmes and the progressive abandonment by the Supreme Court of its sanctified doctrine of private property.

From Public Property to ‘Regulation’: the Ideological Context In Europe, this liberal and pragmatic vision only gained ground in the 1980s. The ‘New Right’ denounced the failures of the post-war Keynesian state and dominated the electoral cycle for most of the decade. The left underwent a deep ideological crisis and progressively renounced its collectivist credo. The collapse of the Soviet Union strengthened the belief that the market was the best engine of growth. In conformity with the new dominant ‘paradigm’ of public action, large segments of the public sector were privatised or liberalised: it is in this context that the state had to be redefined as a ‘regulator’. In the same time, the emergence of new social movements denouncing the negative externalities of the market (in terms of public goods such as health, consumer and environment protection...) gave public authorities a new raison d’étre. The ‘paradigm shift’, from the Keynesian to the regulatory state, generated a new stream of economic and political research dedicated to the foundations of public action. Researchers aimed at identifying the criteria which make a public action both necessary and efficient. In so doing, they merely argued that public authorities emerge when the negative externalities of the market are perceived

The Politics ofRegulation in the European Union

5

as a public problem, and when the regulation of this problem by the market itself does not seem possible. An ever-growing literature has also tried to explain how leaders choose between different forms of public regulation. Why do they sometimes preserve public monopolies; why do they, in other circumstances, opt for external regulation of a service transferred to a private actor; or why do they decide to let private parties define their own rules through self-regulation? In trying to address these questions, most European scholars drew on the literature developed in the US in similar circumstances. Influenced by the dominant functionalist approach, the European debate was merely cast in terms of ‘delegation’. The reason why scholars largely focused on the issue of delegation is obvious. The major reforms of public authorities in the 1980s all followed the same direction: functions formerly performed by state agencies were either privatised or transferred to new public authorities which were granted broader margins of autonomy than classic state departments. Where new forms of regulation emerged — such as in food safety, environment... -- the choice between coregulation, legislation and regulation by an ‘agency’ more often than not ended up privileging the last option. The Assumptions of the Functionalist Approach In this context, political scientists found in the ‘principal-agent’ (P-A) approach forged in the US a useful analytical tool. Initially designed to explain delegation of legislative authority to Congress committees, this schema had also been used to analyse delegation of executive functions to federal agencies. Given the rationalist assumptions on which it is based (actors rationally try to maximise their benefit), and its functionalist logic (the nature of an agent is defined by the function it is supposed to perform), such a language seemed abstract enough to be extended to new objects and different contexts. The principal-agent approach is indeed based on a rationalist assumption which is widespread among academics nowadays: any public authority is said to be moved primarily by cost-benefit calculation. An authority will therefore regulate a given field on its own as long as the benefits overweigh the costs. In some circumstances, public officials realise that the cost of their regulation is higher than the benefits they draw from it, and search for alternatives. Political

scientists have identified four main reasons why an authority may believe that it has an interest in delegating one of its functions to an ‘agent’? a) delegating a function to an agent can first help reduce the problem of credible commitment due to time inconsistency or non-compliance. When, for example, a government believes that its commitment to avoid inflation will not be trusted by economic actors — who fear that it will be tempted to relax its

6

Regulation through Agencies in the

EU

efforts in periods of electoral competition — it may choose to delegate the monetary policy to an independent central bank. b) anon-governmental agent can also provide policy expertise needed by governments at low cost, and reduce the workload of the administration. c) recourse to non-governmental experts can also be useful to enhance the efficiency of decision making, particularly in fields characterised by a high level of technicality. d) delegating decisions to a third party is also a means to shift blame for unpopular decisions from governments to other actors. These are usually presented as ‘universal’ explanations that can be used to analyse situations as diverse as the US Congress, the French or British administration, or the European Union. In practice, these motivations are often combined: the ECB was created first and foremost to address a potential problem of commitment, but it was also supposed to make efficient choices informed by sound expertise, and governments hoped that the Bank would bear the cost of taking unpopular decisions. General Trends and National Variations

Although the P-A approach pretended to provide general explanations, scholars soon noted that governments faced with similar problems often opted for different solutions, and they endeavoured to understand these variations through comparative analysis. Explaining why governments opt for independent agencies, and which powers they give them, is a task which largely bypasses the limits of this chapter — and the competence of the author. I will simply recall that, drawing on comparative research, Mark Thatcher has identified four sets of factors which help explain the pace and nature of institutional choices made by European governments in the recent past:* a) policy learning and institutional mimetism in large part explain why the ‘agency’ model spread among Europe: lessons drawn in one sector often served as examples to reform other sectors; experiences made in other countries often influenced domestic choices; in some instances, the exchange of practices

across countries were encouraged — sometimes coercively — by supranational bodies; b) the pre-existence of a national tradition of regulatory commissions, or

the compatibility of this model with the legal culture of a given country also helps explain why it spread more or less rapidly across countries; c) the nature of the government’ political majority determines both the rhythm and the shape of these reforms: a single-party majority’s strong leadership makes reforms easier and favours single-persons agencies, while coalition governments face more difficulties and usually opt for multi-party models;

The Politics of Regulation in the European Union

i

d) the broader context of state reforms, that can be summarised as the trinity of privatisation, liberalisation and new public management, facilitates decisions to set up regulatory agencies. In turn, these domestic factors can explain the facility or difficulties of institutional choices made at the European level. On the one hand, if a decision

to delegate powers to a supranational body is primarily moved by concerns for credible commitment and cost-benefit calculation, the presence of a model with

which governments and public opinion are familiar may facilitate supranational choices. But on the other hand, when governments have set up national regulatory agencies in a given field, they may also be less willing to create a European agency that would replace their own organs. This uncertainty explains that so much attention has been devoted to the problems inherent in the logic of supranational regulation. Supranational Agencies: Addressing Problems of Collective Action

Given the rise of European debates on regulation, political scientists have been encouraged to explain why governments sometimes decide to create supranational ‘agents’. Their main explanation can be summarised in these terms: in an international regime such as the European Union, specific problems of public action arise, which are usually not encountered in a nation state. Institutionalised cooperation between states faces at least two specific problems, which can be addressed by the technique of delegation. Upstream, the rivalry between governments may undermine their coordination and prevent the formation of agreements. Downstream, they may be tempted to elude the obligations they have subscribed to, so as to reduce the cost of their commitment. The EU institutional setting may be understood, in this prospective, as a set of contracts designed to cope with problems of collective action. Andrew Moravesik argues that, although they always seek to maximise their interests, Member states have sometimes chosen to pool or delegate sovereignty. These decisions are, in his book, primarily instrumental: governments ‘define a series of underlying objectives or preferences, bargain to substantive agreements concerning cooperation, and finally select appropriate international institutions in which to embed them’.° In other words, ‘the decision to delegate or pool sovereignty in international regimes is analytically separate from (and subordinate to) bargaining over substantive cooperation’.® The motivation of these institutional choices is stable: pooling or delegating sovereignty are best understood as ‘efforts by governments to constrain and control one another — in game-theoretical language, by their effort to enhance the credibility of commitments’.’ Opting for qualified majority voting or granting a monopoly of

initiative to the Commission may reduce the risks of non-cooperation; setting

8

Regulation through Agencies in the

EU

up agents endorsed with the task of guarding the treaties or of regulating a given field may prevent free-riding. In the same vein, Mark Pollack has identified four types of agents in the EU context, characterised by the function they perform:* a) agents which benefit from a power of formal agenda setting (Commission) to counterbalance tensions and prevent non-decisions; b) agents designed to monitor the principal’ compliance with common rules (Commission, Court); c) agents designed to adjudicate disputes among principals (Court); d) independent regulators in defined ‘technical’ fields (Commission on competition law, ECB, other ‘regulatory’ agencies). These approaches give us three useful insights. First, they help reduce the complex nebula formed by EU ‘agencies’ by offering a typology of ‘agents’ based on their raison d’étre and on the powers they are granted. Second, they draw our attention to the fact that a single agent can be endowed simultaneously with different tasks. The Commission is given a power to set the agenda and to monitor the governments’ attitude in a broad range of matters; in addition to that, in some specific fields such as competition policy, it acts as a regulatory body — setting the agenda, regulating, implementing decisions, adjudicating disputes and monitoring the principals. Third, this approach indicates that a similar problem can be addressed through different means: when they face a problem of asymmetric information, the governments can decide to set up advisory groups, or to delegate the decision to an agent; when they seek to reduce non-cooperation, governments can decide to give a monopoly of initiative to an agent, but they can also, more modestly, set up agencies endowed with the task to collect and disseminate information, or to encourage coordination through exchange of information, benchmarking, mutual learning...; when the governments face problems of commitment, they can delegate the decision to an agent, or simply give it the power to implement their decision or to monitor their implementation. The principals thus have a broad choice between different options, and their decision is the outcome of political negotiations (among governments and/or, depending on the decision-making rule, between the governments gathered in the Council and the Commission and European Parliament). Although the P-A approach is based on ‘abstract’ assumptions and aims at discovering universal laws of delegation, the discovery of different empirical solutions to identical ‘problems’ reminds us that politics matter here: facing similar difficulties, principals make different institutional choices. The following table summarises the causes and possible options which make the politics of delegation. This typology differs from the one proposed by the Commission? and from those put forward by lawyers." First, it does not include those ‘agencies’ that simply operate as subcontractors to the European institutions — such as the

The Politics of Regulation in the European Union Table 1.1

9

|The Causes and Possible Options which make the Politics of Delegation

Nature of the problem faced by the principal

Function granted to the agent

Example in the EU context

Lack of technical knowledge (asymmetry of information)

Advisory

European Food Safety Agency

Regulation

Office for Harmonisation in the Internal Market!

Information and awareness-rising

European Monitoring Centre on Racism and Xenophobia!”

Coordination

European Centre for the Development of Vocational Training!3

Agenda-setting

Commission

Monitoring

Commission

Lack of cooperation

Credibility of commitment

European Aviation Safety Agency'4

Disputes among the principals

Implementation

Commission (competition)

Regulation

ECB

Adjudication

Court

European Training Foundation and the Translation Centre for the Bodies of the EU — because these are simple executive bodies which do not address problems of regulation and collective action examined here. Second, this typology, includes institutions such as the Commission, ECB and the Court, which are more than

‘agencies’ from a legal point of view, but which do act as ‘agents’ of the principals —read the governments — from a political point of view. The purpose of this classification is to remind that there is no ‘objective’ solution to a given problem. The decision to create an agent, and the choice between different types of agents — more or less independent, more or less supranational — is the outcome of

10

Regulation through Agencies in the

EU

political conflicts; or to put it in more fashionable terms, ‘the perception that delegation to non-majoritarian institutions is the best option for dealing with certain problems is socially constructed’.'° It remains difficult to draw general laws explaining the principals’ choices. Each decision is the outcome of a sui generis context, so that the general dynamic can only be described in negative and relative terms. Looking at recent European experiences, it is reasonable to believe that: a) the governments will generally prefer to keep a given field under their control than to delegate it to an independent agency; b) when they find it difficult to cope with a given task by themselves (because of their lack of expertise and/or expected gain of resources), they will accept to delegate it, while privileging a weakly independent agency over a truly regulatory body;

c) when they think they have to set up an independent regulator, because of the problems of credible commitment revealed by the pressures of the stakeholders and/or public opinion, they will prefer a national to a supranational agency; d) they will only create a supranational agency if they fear the consequences of non-coordination; e) EU institutions will only support an independent and supranational regulator if they think this widens EU competences and if the new regulator will not undermine their own existing or potential powers. The combination of these conditions explains why so few regulatory agencies have been created at the European level so far: the weight of one or several of these constraints explains why the governments have opted, in most cases, for agencies endowed with information, coordination or strictly limited . implementation tasks. As the word agency is used in a large variety of different cases in the EU, and as this “conceptual stretching’ may arise confusion, Majone suggests to coin a new concept to describe truly independent regulatory agencies: he distinguishes ‘agents’ which are only given a very restricted competence and margin of independence, and ‘trustees’ which are given a much larger mandate and stronger guarantees of independence.'® This distinction is important not only as an analytical tool, but also, as we will see, because it helps understand the dilemmas faced by politicians when they try to address the consequences of their decision

to set up a regulatory agency.

THE CONSEQUENCE OF DELEGATION When the governments agree to set up an independent regulator, despite their reluctance, they will usually try to reduce the potential costs of this decision. In

The Politics of Regulation in the European Union

1]

other words, the politics of regulation do not only involve the decision about delegation, but also about its consequences: ‘Every decision to delegate essentially involves two choices — what powers to delegate and what institutional control mechanisms to craft’.'” Guarding the Guardians

Experience indeed shows that an agent may develop an interest of its own, and use its institutional assets to promote it at the expense of the principals’ objectives. The asymmetry of information between the principal and the agent, the conflicts of preferences among the principals, the gap between the principal’s electoral cycle and the agent’s mandate, are resources that can be exploited by the agent to pursue its own policy. If it does not manage to control it and prevent these deviations, the principal will face ‘agency losses’. This is the reason why, anticipating this risk, the principal is induced to set up mechanisms to keep its agent under its control: a) the principal will define the agent’s mandate as narrowly as possible; b) it will set up oversight procedures so as to monitor it— or have it monitored by the stakeholders or by a third party; c) it will usually preserve a right to sanction the agent and/or to alter its mandate.'* Yet, these mechanisms may sometimes conflict with the raison d’étre of the agency. When the main task of the agent is to collect information or expertise to facilitate agreements among governments, its powers and autonomy can be restricted. When, on he contrary, an agent is set up by governments to make their commitments credible, it will be granted large powers of compliance and/ or adjudication and its autonomy will be protected. If these two conditions were not met, the ‘trustee’ would be unable to fulfil its functions because it would

lack the necessary institutional resources and/or would not be seen as a neutral arbiter. The governments are thus constrained to design oversight procedures which are ‘soft’ enough to be compatible with the trustee’s autonomy. This brings us to the second consequence of the politics of regulation: while respecting the trustee’s independence, the governments must also ensure that their creature will not be understood as a breach of the democratic principles underlying the regime. Independent regulators raise issues of legitimacy because their status seems to break the ‘chain of delegation’ of the parliamentary state. If a classic majoritarian model, which gives the elected authority the power to sanction its agent, is not compatible with the agent’s independence, another mechanism must be forged to hold the regulator accountable. True, the legitimacy of a trustee in part lies in the efficiency of its action. Or to put it in Sharpf’s terms, a trustee must rely on ‘output legitimacy’ rather than on ‘input legitimacy’. Since it was created to solve problems inherent in classic

i

Regulation through Agencies in the

EU

governmental actions — time inconsistency, lack of coordination, capture of the authority by powerful interests... — it is merely by demonstrating that it can indeed solve these problems that it will become legitimate in the eyes of the public. Moreover, Scharpf argues that this is consistent with the public’s fundamental expectations: ‘the weakening of political legitimacy in Western Europe is a consequence of the loss of problem-solving capacities of political systems which has been brought about by the dual and interrelated processes of economic globalization and European integration’.'!” This theory seems selfevident. Comprehensive transnational surveys have shown that the citizens’ declining support for the state was linked to the feeling they had that public authorities could no longer respond to global changes. In their view, the state was not able any more to guarantee their welfare, which is its raison d’étre.*° If such a prognosis is true, the legitimacy of European agents is thus mainly — and even exclusively — dependent on their capacity to respond to the desires of the represented. Output legitimacy cannot, however, replace input legitimacy; at best, it can supplement it. A political regime whose legitimacy is only based on its acts is condemned to precariousness. Historical examples are numerous — the worst of which being the Republic of Weimar — that prove that a political regime or an institution which is not legitimate in itself, but only in function of its acts, may be globally rejected if it is no longer able to come up to the citizens’ expectations. 7! The Narrow Path of ‘Accountable Independence’ It is thus widely accepted nowadays that, to be seen as legitimate, trustees must not only deliver the outcomes for which they were created, but that they must also be held accountable. The question becomes: how can an independent regulator be held accountable; can we identify procedures guaranteeing that the agent will account for its action, without jeopardising its independence? The conventional answer to this question is that ‘procedural legitimacy is a fair and democratic substitute for electoral accountability, given the broader purposes of delegation’. Most scholars argue that the transparency of an agency, its duty to consult the stakeholders, weight their arguments and give the reasons for its decision, amount to a form of ‘soft accountability’ which is an acceptable substitute for classic electoral mechanisms where these do not fit the institution. Yet, in practice, it remains difficult to agree on a set of ‘procedural conditions’. While trying to identify objective standards of ‘procedural legitimacy’, Majone acknowledges that this is not an uncontroversial choice. His case for a procedural approach is, he concedes, ‘only viable if all the actors understand the fiduciary principle and the duties that flow from it’.”’ In other words, ‘soft accountability’ is a valuable alternative to classic methods of control based on constraints

The Politics of Regulation in the European Union

13

and threats of sanctions if and only if the principals are convinced that the ‘fiduciary principles’ actually work. Majone implicitly recognises the limits of his functionalist reasoning: it is perfect if one assumes that the governments share the rationalist assumptions of the scholar; but if some governments or other public actors are not convinced that soft accountability is efficient and/or legitimate, the legitimacy of a strongly independent agent will be called into question. In other words, the ideas shaping the governments’ attitudes may be less ‘rational’ than those of the functionalist scholar, and this can alter the practice of oversight. In Thatcher and Stone-Sweet’s words, the procedural model

‘typically assumes ideological consensus on the role, function, and benefits to be accrued from delegation, whereas these points are contested by parties and groups’.** This is the most sensitive aspect of the politics of regulation. It is difficult to assess the need to regulate a given field, and the opportunity to do it at the European level or not, because this will always arouse conflicts of interests; it is even more difficult to decide how a trustee should be held accountable, because

this will generate conflicts of standards. Although ‘democratic principles’ are common foundations of European politics, concrete doctrines of democratic legitimacy — defining the way these principles should be implemented — remain deeply rooted in national and ideological patterns of thought which are often contradictory.” Those who support this point of view do not pretend to replace a democratic logic with another non-democratic (technocratic) logic. On the contrary: they argue that these procedures give a new form to the classic democratic principles. Nowadays, most political theorists agree that the major ambition of a democratic polity is to produce public deliberation about collective issues: *° a democratic constitution should organise a sound process of confrontation of different views, so as to produce the best possible policies, well informed and responding to the citizen’s expectations.”’ The ‘fiduciary principles’ which constrain the agents to consult the stakeholders, to weight their views, to give access to the documents

they use to prepare their decisions, and to give the reasons why those decisions were taken, pursue the same deliberative ambitions. In other words, “these canons are best understood not as barriers but as catalysts, promoting democracy’.”® First, they constrain the regulators to take a large array of viewpoints into consideration and to assess the different options carefully: this is supposed to prevent the risks of captation by a given interest and the danger of grouppolarisation (when they deliberate in closed circles, and are not exposed to contesting views, like-minded persons may be induced to strengthen their initial assumptions, rather than look for a balanced solution). Second, these canons offer the stakeholders — and those segments of the general public who benefit from enough resources to scrutinise the regulator — means to contest their decisions: anticipating the risk to be discredited in the eyes of

14

Regulation through Agencies in the

EU

those who are concerned by its decisions, or even the possibility to be overruled by a judicial adjudicator, the regulator is supposed to be responsive — very much like an elected politician who anticipates an electoral sanction. Designing an accountable agency amounts to adopting a set of rules likely to force it to balance different views, while avoiding constraints that would undermine its independence. This can be done either negatively, or positively. From a negative point of view, an agent can be made responsive by being threatened with political interference in case it would deviate from it mandate. The independence of the agent can be limited — actually or potentially, by playing on three factors: a) the rules of appointment defining the regulator’s personal independence can favour or reduce political interference: long and non-renewable mandates; multi-institutional appointment procedures; participation of the regulator in the appointment; rules of rotation reducing the scope of new appointments; conditions of professional expertise; incompatibilities and ‘cooling-off periods’; absence of sanctions; those are the most often used criteria to reduce political interference a priori and to put the regulator in such a situation that he does not have to anticipate the principal’s reaction. In the case of the ECB, the conditions most favourable to the Board’s independence were chosen. b) the definition of the regulator 's mandate may also prevent or encourage political interference: a narrow and clearly defined mandate, neatly separate from the principal’s own responsibility, reduces the risks of interference and conflict. Again, the ECB was given a narrow mandate, which does not overlap with the government’s own tasks — contrary to the Fed, whose responsibility is defined in much broader terms overlapping with the government’s mission. c) the /egal nature of theses conditions also favours or discourage political interference: the easiest it is for the political branch to alter the agent’s mandate or independence, the more the agent will be tempted to anticipate its reaction. Here too, the ECB benefits from the strongest guarantee: included in the treaty, its mandate and status can only be modified by an IGC, acting by unanimity. Beside these negative conditions, the accountability of an agent may be defined in positive terms: a) the agent can be obliged to consult the stakeholders or the principals themselves. In some cases, they may even be invited to participate in the deliberation of the regulator, so as to have an opportunity to have their voice heard; b) the agent can be imposed a duty to explain its decisions: giving access to its documents (a condition that can be limited to some of its documents, and

that can be suspended before a given time limit has expired), giving the reasons underlying its decisions, reporting on its actions, are classic mechanisms designed to create the conditions for a debate on the agent’s action; these mechanisms pursue a deliberative objective in that they set the conditions for the development of counter-arguments;”?

The Politics of Regulation in the European Union c)

15

additionally, its decisions, or some of them, can be submitted to Judicial

review andfinancial accountability; in the EU, extra-judicial reviews such as those of the Ombudsman, complete these classic mechanisms. The Conflict of Standards: the Example of the ECB

Given their potential effect, these institutional choices are the object of deep political divisions. In the case of the ECB, for example, the Council’s President and one member of the Commission were given the opportunity to participate, without being allowed to vote, in the meetings of the Bank’s working Council. In addition, the consultative tasks granted to the Economic and Financial Committee — composed of two members of the ECB, the Commission and each

Member state — were also supposed to ensure that the principals would have an opportunity to have their voice heard. Strict guarantees of independence were thus counter-balanced by mechanisms designed to favour a dialogue between the regulator and the political authorities. Despite these precautions, the ECB remains contested. Without calling into question its existence or its status, many voices denounce its ‘lack of accountability’. Rejecting the arguments made by the central bankers themselves and by their supporters, several academics argue that ‘severing ties to democratic representatives and relying on technocratic expertise does not apoliticise monetary policy. Rather, delegation to independent central banks produces partisanal policies, with significant distributional effects that raise important questions of democratic accountability’ .*°This critic is widespread in academia, and echoed in a large array of political parties and trade unions:*' the ECB should not be insulated from political pressures, because contrary to what is often stated, its decisions cannot be reduced to ‘technical’ choices between different means; they are political choices between different values. Similar

reasoning could be made, and is often made, vis-a-vis other trustees: any agent who benefits from a real decision-making power makes choices which entail distributional consequences. It should therefore account for its action before those who represent the citizens. This argument does not necessarily amount to defending the status quo ante. First, most observers know that classic administrative departments were rarely transparent, and that the political representatives who were supposed to account for their action rarely did.** Transforming an administrative committee into an agency often implies a gain of clarity and transparency.*? Second, most critics focus on the agent’s attitude rather than on its status: it is reasonable to believe that, if the ECB followed a less orthodox line — thereby reducing the gap between its own policies and the dominant preferences of the governments — most of these critics would cease. The more an agency departs from the dominant social demand, the more its legitimacy is called into question.

16

Regulation through Agencies in the

EU

One should not, however, underestimate the conflicts on the ECB’s lack of

accountability; they cannot be reduced to pretexts used by the opponent’s of the Bank’s policy, or to temporary problems of cultural adaptation. Since the very first days, and well before its alleged orthodoxy rendered it ever less popular, the ECB’s status has given rise to principled discussions.** Three aspects of the process designed to hold it accountable remain open to political discussions.** First, its relation to its principal. A large part of the governments, and a stable majority in the EP, argue that the ‘principal’ should be strengthened: a more structured Eurogroup would act as the EU’s ‘economic government’ and force the ECB to take the governments’ view more seriously into consideration; some even argue that the power to define the inflation target should be transferred from the ECB to this political branch. Second, the idea that the ECB should give the reasons for its decisions, not

only in terms of price stability but also with regard to the other general objectives assigned to it by the treaty, remains popular not only in academia but also in the political spheres. The article which defines the ECB’s mandate in the treaty clearly gives priority to the first objective, but some argue that this does not mean that secondary objectives should be neglected.** Within the Convention on the future of Europe, a significant minority pleaded for a reform of the ECB’s mandate, so as to force it to justify its action in broader terms, like its US counterpart.

Third, the ECB’s transparency also gives rise to discussions. On the one hand, some argue that the ECB should be fully transparent. It should publish its individual voting records, the minutes of its deliberations, its inflation forecasts

and the other instruments used to take its decisions. This point of view, inspired by British and Scandinavian traditions of openness,*’ lies on the cognitive belief that a transparent society can reach harmony because each actor is able to get the information she needs to adapt its actions to the context. On the other hand, accountability means, for other actors, that ‘we do what we are supposed to do’.** In other words, an independent organ is accountable if it respects its mandate. This conception, closer to the continental tradition, is based on another cognitive assumption: as a member of the ECB clearly put it, it reflects a ‘recognition of the limits of knowledge’.*? Those who share this point of view do not believe in the self-regulatory capacity of markets and fear that full openness may disturb the actors’ strategies, and the public’s perception. According to this point of view, information should be channelled by authorities. Where transparency is seen as a condition of clarity and harmony by some, it is described as leading to cacophony and opaqueness by others. These conflicts in large part reflect the differences between national civic patterns and ideological preferences. To a certain extent, the EU can cope with this problem. Many conflicts of standards about the role and status of agents can indeed be addressed by EU agents themselves. The role played by the Court

The Politics of Regulation in the European Union

17

since its inception, and more recently by the Court of First Instance (CFI) and the ombudsman,” contributes to the emergence of a ‘common constitutional language’ which could reconcile national and partisan views. The recent Opposition on the meaning of ‘transparency’ illustrates this process. After transparency had been presented as one of the principles of the EU in the Maastricht treaty, a strong conflict emerged between the governments who argued that it should be given a restricted meaning in order to preserve the efficiency of confidential negotiations within the Council (France, United

Kingdom) and those who required a broad and rigorous definition of transparency and access to documents, perceived as key conditions of the EU’s accountability. Partly addressed by the governments themselves who negotiated an agreement and later on a treaty amendment to clarify the meaning of the concept, it was also submitted to the CFI and the ombudsman. Individual litigants (journalists, academics and MEPs) complained about the lack of transparency of the Council before the CFI and the ombudsman,

and this gave the two institutions the

opportunity to forge a strong definition of the concept and to contribute to convey it in the EU institutional setting. This saga illustrates a classic functionalist mechanism, where agents themselves complete the contract adopted by the principal. It also underlines the ever-growing role of ‘accountability agencies’ in the EU: in an ever more complex Union, it may be necessary to rely on transversal principles which consists in subjecting ‘all of the Union’s institutions to standard sets of rules and procedures, or scrutiny by agents who are dedicated to a single task but responsible for applying it across the entire EU institutional system’.*!

THE LIMITS OF PROCEDURAL LEGITIMACY The reflection on the accountability of regulatory agencies in the EU tends to focus on their legal status and on the procedural conditions imposed upon them to hold them accountable. When approaching this debate, one should not lose sight, however, of the broader institutional context. Even if a consensus could

be reached on the conditions that would make an agency accountable, without jeopardising its independence, it would remain necessary to reflect upon the consequences of the politics of regulation on the EU as a whole. At present, at least two broader problems arise. First, the hybrid status of some EU institutions, mixing regulatory and other functions, causes confusion. Acting as a regulatory organ in some fields, as an executive body in others, and being a key element of the legislative process, the Commission’s profile is often misunderstood. Moreover, given its essential hybridity, the Commission is subject to a very large array of oversight procedures: its members are appointed by the governments and approved by the EP; it must

18

Regulation through Agencies in the

EU

respect a large set of ‘fiduciary principles’; it is monitored by committees of national experts and EP committees and can, in certain circumstances, be imposed judicial sanctions. In practice, however, this may lead to a pernicious ‘mélange des genres’: the Commission’s role as a guardian of the treaty can be and often is politicised, while as a policy-initiator it can be judged on technocratic rather than political criteria.” Some would argue that this limit is due to an incomplete application of the logic of delegation, rather than to the logic itself: since the Commission is ever more politicised, Everson and Majone argue, and given its role in the legislative process, its regulatory tasks — in the field of competition policy — should be transferred to an independent agent.** Others could suggest that the Commission’s task as a guardian of the treaty could be transferred to a US-like General Prosecutor. Long supported by lawyers within the Commission itself, this proposal is, however, contested by those who argue that those are the key powers of the Commission, and that the management of the competition policy, or the guardianship of the treaties, are highly political tasks, that should remain under the control of accountable officials.” A second, separate problem, is linked to the complexity of an institutional system relying on delegation. The Court’s reluctance to accept delegations which threaten the EU’s ‘institutional balance’* can be understood in these terms: like its US counterpart,” the Court’s scepticism seems to be motivated by its fears that a complex system would cause accountability gaps. This fear is not unfounded. The ESB crisis has illustrated this problem. The report on the handling of this crisis by the Commission — submitted by the Committee of Inquiry set up by the European Parliament — gave an apocalyptic description of how decisions were made in the EU. First, it argued that the allocation of tasks

between different organs and different levels of power does not follow any clear logic. The Committee also denounced the consequences of interference between politicians, officials and experts. Less than two years after, a Committee of Independent Experts in charge of examining the way the Commission dealt with internal cases of fraud, mismanagement and nepotism reached similar conclusions. As an echo to the observations made by the parliamentary Inquiry Committee on the handling of the mad cow disease crisis, the second report showed that the complex distribution of management tasks between national administrations, the Commission, the Executive Committees and private subcontracting companies, had often led to ‘depriving the accountable body of its responsibility’.*’ This crisis illustrated a strong warning made by Hamilton more than two centuries ago: ‘One of the weightiest objections to a plurality in the executive [...] is that it tends to conceal faults, and destroy responsibility. [...] It often becomes impossible, amidst mutual accusations, to determine on whom the

blame or the punishment of a pernicious measure, or series of pernicious

The Politics of Regulation in the European Union

19

measures, ought really to fall. It is shifted from one to another with so much dexterity, and under such plausible appearances, that the public opinion is left in suspense about the real author’.** This is not simply a question of delineation of competence. A clarification of the respective tasks of the agencies might prevent this sort of conflicts, but it would not correct the widespread impression that, in the EU, crucial decisions are taken for which nobody can be praised and blamed. It is in this spirit that John Stuart Mill wrote, one century and a half ago, that ‘responsibility is null when nobody knows who is responsible. Nor, even when real, can it be divided without being weakened. To maintain it at its highest, there must be one person who receives the whole praise of what is well done, the whole blame of what is ill’.” This is a major problem in terms of democratic theory, which had also been identified by Dewey at the dawn of the emergence of the regulatory state in the US. Even if each agency were submitted to perfect procedures of accountability, and even if the stakeholders were actively mobilised to monitor these agencies, the system’s legitimacy would still remain imperfect: ‘The ramification of the issues before the public is so wide and intricate, the technical matters involved are so specialised, the details are so many and so shifting, that the public cannot for any length of time identify and hold itself. It is not that there is no public, no large body of persons having a common interest in the consequences of social transactions. There is too much public, a public too diffuse and scattered, and

too intricate in composition. And there are too many publics, for conjoint actions which have indirect, serious and enduring consequences are multitudinous beyond comparison, and each one of them crosses the others and generates its own group of persons especially affected with little to hold these different publics together in an integrated whole’.*° Robert Dahl, one of the most respected theoreticians of democracy in the last decades, has followed the same reasoning in his examination of the EU: ‘in practice, delegation might be so extensive as to move a political system beyond the democratic threshold’,*! he argued, adding that this implies that ‘an international organization is not and probably cannot be a democracy’.* In Dahl’s book, democracy requires political institutions that ‘provide citizens with opportunities for political participation, influence and control’ and this in turn implies that citizens need to be ‘concerned and informed about the policy decisions’ and that ‘political and communication elites [...] need to engage in public debate and discussion of the alternative in ways that [...] engage the attention and emotions of the public’.°* Viewed through this lens, the EU is not democratic because accountability remains confined to institutional actors and elitist stakeholders, and it will not be as long as it will be deprived of an ‘international equivalent to national political competition by parties and

individuals seeking office.*4 This does not mean that the politics of delegation should be condemned sans phrase, but that in examining the evolution of the

20

Regulation through Agencies in the

EU

EU’s regulatory structure, scholars should not lose sight of its parliamentary architecture.*

NOTES 1 K. Polanyi (1944), The Great Transformation, New York: Reinhart. 2 M. Teld (1988), Le New Deal européen, La pensée et la politique sociale-démocrates face a la crise des années trente, Brussels: Editions de |’Université de Bruxelles.

3 M. Pollack (1997), ‘Delegation, agency and agenda setting in the European Community’, International Organisation, 51 (1), 99-134; M. Thatcher and A. Stone Sweet (2002), “Theory and Practice of Delegation to Non-Majoritarian Institutions’, West European Politics, 25 (1), 1-22.

4 M. Thatcher (2002), ‘Delegation to Independent Regulatory Agencies : Pressures, Functions and Contextual Mediation’, West European Politics, 25 (1), 125-147.

5 A. Moravesik (1998), The Choice for Europe, New York/Ithaca: Cornell University Press, p. 5. Ibid., p. 21. Ibid., p. 9. M. Pollack, art. op. cit. supra note 3. an oon See Communication from the Commission on an operating Framework for European Regulatory Agencies of 11 December 2002, COM (2002) 428 final. 10 See D. Géradin and N. Petit (2004), ‘The development of Agencies at EU and National Levels : Conceptual Analysis and Proposals for Reform’, Jean Monnet Working Paper, 01/04. ‘1 The Community Plant Variety Office and the European Agency for the Evaluation of Medicinal Products belong to the same category. 12 The European Monitoring Centre on Racism and Xenophobia belongs to the same category. 13 The European Foundation for the Improvement of Living and Working Conditions and the European Agency for Safety and Health at Work belong to the same category. 14 As well as the European Maritime Safety Agency. 15 M. Thatcher and A. Stone Sweet (2002), ‘Delegation to Independent Regulatory Agencies: Pressures, Functions and Contextual Mediation’, 25, West European Politics, 1.

16 See G. Majone (2001), ‘Two Logics of Delegation’, European Union Politics, 2 (1), 103-122 and G. Majone (2001), ‘Nonmajoritarian Institutions and the Limits of Democratic Governance: A Political Transaction-Cost Approach’, Journal of Institutional and Theoretical Economics, 157 (1), 57-78. Without adopting the same terms, Géradin and Petit, art. op. cit. supra note 14, also end up privileging a summa divisio based on the independence and powers of the agency. 17 J. Tallberg (2002), ‘Delegation to Supranational Institutions: Why, How, and with What Consequences?’, West European Politics, 25 (1), 23-46, p. 28.

'8 For an excellent synthesis of these mechanisms, and their practice in the EU context, see M. Pollack, art. op. cit. supra note 3. '9 F. Scharpf (1999), Governing Europe, Efficient and Democratic, Oxford: Oxford University Press, p. 2.

20 H.-D. Klingemann and D. Fuchs (eds) (1995), Citizens and the State, Beliefs in Government, Oxford: Oxford University Press. 21 Jn the seventies, Jiirgen Habermas noted in his systemic analyses of the legitimacy crises that a state that based its legitimacy on its acts, suffered from the consequences of global economic cycles it could not really control, so that it was systematically condemned to undergo frequent crisis of legitimacy. See J. Habermas (1975), Legitimation Crisis, Boston; Beacon Press. 22M. Thatcher and A. Stone-Sweet (2002), ‘Delegation to Independent Regulatory Agencies: Pressures, Functions and Contextual Mediation’, 25, West European Politics, 1.

23 G. Majone, ‘Nonmajoritarian Institutions and the Limits of Democratic Governance: A Political Transaction-Cost Approach’, art. op. cit. see supra note 16, p. 120. 24 M. Thatcher and A. Stone-Sweet, art. op. cit. supra note 3, p. 19. 25 See V. Schmidt (1997), ‘European Integration and Democracy: the Differences among Member States’, Journal of European Public Policy, 4 (1), 128-145 and P. Craig (1999), ‘The Nature ofthe

The Politics of Regulation in the European Union

Zl

Community: Integration, Democracy and Legitimacy’ in P. Craig and G. De Burca (eds), The Evolution of EU Law, Oxford: Oxford University Press. 26 In a very wide literature, see S. Benhabib (ed.) (1996), Democracy and Difference, Contesting the Boundaries of the Political, Princeton: Princeton University Press; J. Bohman and W. Rehg (eds) (1997), Deliberative Democracy, Essays on Reason and Politics, Cambridge (Mass): MIT Press; J. Elster (ed.) (1998), Deliberative Democracy, Cambridge: Cambridge University Press; C. Sunstein (2001), Designing Democracy, What Constitutions Do, Oxford: Oxford University Press. 27 Additionally, deliberative politics are said to enhance the citizens’ civic competence and their acceptance of the system and of the policies, since they have had their voice heard and they have been enlightened by the discussion.

28 C. Sunstein, op. cit. supra, note 26, p. 151. 29 See M. Shapiro (2002), ‘The Giving Reasons Requirement’, in M. Shapiro and A. Stone Sweet, On Law, Politics and Judicialization, Oxford: Oxford University Press, pp. 228-257. 30 K. McNamara (2002), ‘Rational fictions: Central Bank Independence and the Social Logic of

Delegation’, West European Politics, 25 (1), 47-76, p. 47-48. In the same vein, see J. Stiglitz, ‘Une banque centrale indépendante ou démocratique?’, Le Monde, 27 February 2004. 3! See J.-V. Louis and A. Komminos (eds) (2003), The Euro, Law, Politics, Economics, London:

British Institute for International and Comparative Law. Studies on the media coverage of EU activities show that the institution against which ‘claims’ are most often raised is the ECB. See Le Torrec, Philippe Blanchard, Guillaume Garcia and Charles Patou, ‘Framing Europe: News, coverage and legitimacy of the European Union in five countries’, (2001) Seventh Biennial International Conference, European Community Studies Association, Madison Wisconsin, USA. 32 See R. Elgie (1998), ‘Democratic Accountability and Central Bank Independence: Historical and Contemporary, National and European Perspectives’, West European Politics, 21 (3), 53-76. 33 R. Dehousse (1997), ‘Regulation by Networks in the European Community: the role of European agencies’, Journal of European Public Policy, 4 (2), 246-261. 34 See L. Gormley and J. de Haan (1996), ‘The democratic deficit of the European Central Bank’, European Law Review, 21 (1), 95-112.

35 P. Magnette (2000), “Towards accountable independence ? The European central Bank and the Rise of Parliamentary controls’, European Law Journal, 6 (3), 326-340. 36 This argument is frequently made by left wing parties and trade-unions, but cannot be reduced to a left-right cleavage. In some countries (Belgium, France, Britain, Sweden...) this point of view is also supported by liberal and conservative leaders who argue that the ECB should take the general objectives into account and explain how its decisions pursue these objectives. 37 See C. Gronbech-Jensen (1998), ‘The Scandinavian tradition of open government and the European Union: problems of compatibility ?’, Journal of European Public Policy, 5 (1), 185-199 and W. Buiter (1999), ‘Alice in Euroland’, Journal ofCommon Market Studies, 37 (2), 181-209. 38 Q. Issing (1999), ‘The Eurosystem: Transparent and Accountable or “Willem in Euroland”’, Journal of Common Market Studies, 37 (3), 503-519.

39 Tbid., p. 507. 40 See P. Magnette (2003), ‘Between parliamentary control and the rule of law: the political role of the Ombudsman in the European Union’, Journal of European Public Policy, 10 (5), 677-694. 41 J. Peterson and M. Shackleton (eds) (2001), The Institutions of the European Union, Oxford:

Oxford University Press, p. 366. On the necessity of ‘accountability agencies’ see A. Przeworski, S. Stokes and B. Manin (eds) (1999), Democracy, representation and accountability, Cambridge: Cambridge University Press. 42 See P. Magnette (2001), ‘Appointing and Censuring the Commission: the Adaptation of Parliamentary Institutions to the Community Context’, European Law Journal, 7 (3), 289-307. 43 See M. Everson and G. Majone (2001), ‘Réforme institutionnelle: agences indépendantes, surveillance, coordination et contréle procédural’, in O. De Schutter, N. Lebessis and J. Paterson

(eds), La gouvernance dans |’Union européenne, Luxembourg: Office des publications officielles. 4 Interviews with Karel Van Miert, Romano Prodi and Philippe Busquin. 45 See K. Lenaerts and A. Verhoeven (2002), ‘Institutional Balance as a Guarantee for Democracy

in EU Governance’, in C. Joerges and R. Dehousse (eds), Good governance in Europe's Integrated Market, Oxford: Oxford University Press.

22

Regulation through Agencies in the

EU

46 See C. Sunstein, op. cit., supra note 26. 47 Committee of Independent Experts, Second Report on the Reform of the Commission, Analysis of current practice and proposals for tackling mismanagement, irregularities and fraud, 10 September

1999, p. 29; 48 A, Hamilton, J. Madison and J. Jay (1982), The Federalist Papers (1787-88), New York: Bantam

Books, No. 70, pp. 358-359. 49 J. S. Mill (1991), Considerations on Representative Government (1861), in On Liberty and Other Essays, Oxford: Oxford University Press, p. 394. 50 J. Dewey (1927), The Public and its Problem, New York: H. Holt — reprint: Ohio University Press, p. 187. 5! R. Dahl (1999), ‘Can international organizations be democratic? A skeptic’s view’, in I. Shapiro and C. Hacker-Cordon (eds), Democracy s Edges, Cambridge: Cambridge University Press.

52 Tbid., p. 19. 53 Thid.,p. 31. 54 Tbid. 55 See R. Dehousse (1998), ‘European Institutional Architecture after the Amsterdam Treaty: Parliamentary System or Regulatory Structure?’, Common Market Law Review, 35 (3), 595-627 and P. Magnette (2003), ‘European governance and civic participation, Beyond elitist citizenship’, Political Studies, 51 (1), 139-156.

2.

Regulation of Liberalised Markets: A New Role for the State? (or How to Induce Competition Among Regulators) Phedon Nicolaides

INTRODUCTION: CHANGING REGULATORY CONCERNS During the past decade the member states of the European Union have liberalised significant sectors of their economies. Statutory monopolies have been removed and healthy competition has appeared in all of those sectors. Liberalisation has also been accompanied by considerable harmonisation of technical aspects concerning the management of networks, the relationship between incumbent firms and new entrants and the regulatory supervision of these sectors by national authorities. As the EU rules expand and competition intensifies, there is now discussion on how long sector-specific regulation should be retained and whether and when competition law should replace sector-specific regulation. Indeed, it would appear, that after liberalisation and the establishment of regulatory frameworks, there is little for the state to do. This is even more so in the context of the EU. Overall policy objectives and many technical rules are set by the EU, while the Commission not only monitors the transposition of those rules by the member states, it also reviews important decisions of the national regulatory authorities and manages the various networks of regulators. So, a quick answer to the question posed in the title of this chapter is that the state has certainly much less to do than before liberalisation and the adoption of common EU rules. However, I will argue in this chapter that the state still has a non-trivial role to play. I do not mean the elaboration of general EU policy principles. Since liberalisation and regulatory harmonisation has been achieved by and large 23

24

Regulation through Agencies in the

EU

through directives, the state always has to carry out an important task: it has to define with sufficient precision and apply the provisions of the directives. But beyond this task, the state has to ensure that regulatory systems themselves function efficiently. I will show that the EU system of regulation has considerable gaps. These gaps are not the result of omission. They denote the limits of EU competence. I will identify these limits in relation to the institutional aspects of sectoral regulation. The scheme of the chapter is as follows. The next section reviews briefly the standard arguments for government intervention and regulation. This review is necessary in order to understand why liberalisation does not remove all reasons for intervention. The section after that, asks what should governments regulate. Not only should they lay down substantive rules on the behaviour of incumbents and new entrants, but they should also define open and transparent procedures and ensure that the regulators are accountable. The regulatory framework contains prescriptions for both those who have to comply with the rules (i.e. the regulated companies) and those that have to enforce the rules (i.e. the regulators). The rest of the chapter focuses on issues of regulatory governance and the efficient functioning of regulatory institutions. It is in this context that I will ask whether the regulatory system of the European Union defines in a sufficient manner those institutional aspects of regulation, which are necessary for it to function effectively and efficiently. By using the recent telecommunications framework as a benchmark, I conclude that important components of the institutional side of regulation are missing. Only the state can fill those gaps. The chapter concludes by deriving the most significant institutional tasks for the state after liberalisation. Before asking why government should intervene to regulate, it is instructive to sketch the changing landscape of regulation. After World War II there was a wave of nationalisations across Europe. It was thought that some industries were too important to be left to private ownership and control. In the 1960s and 1970s the focus changed slightly. Public policy shifted from controlling national industries to promoting national champions. Within the then EEC these policies were difficult to reconcile with the avowed Treaty objectives of free trade and competition. One outcome was the Commission-sanctioned crisis cartels that in today’s policy environment would never be allowed. In the 1980s, there was another but this time radical policy shift. As a result of dissatisfaction with the performance of nationalised industries and in an attempt to rid them of continuous political meddling in their financial and operational decisions and also to earn badly needed revenue for public coffers, privatisation became the buzz word.' Privatisation brought about liberalisation and then the definition of more precise social policy objectives and the establishment of regulatory authorities. Today’s concerns are about further deregulation, reliance on general competition principles and regulatory convergence across Europe.

Regulation of Liberalised Markets:

A New Role for the State?

Zo

These shifting regulatory targets are presented in Table 2.1. Although this chapter concentrates on regulatory governance, I nonetheless begin with a brief review of the arguments for regulation in the next section. Table 2.1

Shifting Regulatory Concerns

Role of the state

Functioning of markets

Social policy objectives

From state ownership

From entry: statutory monopoly, licensing,

From definition of universal service

certification, etc.

To regulatory governance;

To conduct: price caps,

To funding of universal

powers of NRA

rates of return, inter-

service

connection, liquidity, etc. Currently: relations between NRAs/NCAs; functioning of judicial system and appeals process

Currently: mixture of service and network competition; technological convergence

Currently: state aid to firms with public service obligations; public—private partnerships

WHY REGULATE ? The two fundamental theorems of welfare economics state that every pareto optimum outcome maximises welfare and every outcome that maximises welfare is pareto optimum. These theorems hold under conditions of perfect competition. In these conditions, there is no need for the government to intervene or to regulate business conduct. Therefore, the standard arguments for regulation presuppose some kind of market failure or market imperfection. And, the underlying assumption is that government intervention aims to improve welfare. Typical market failures are the following:*

Externalities: Gains and costs are not completely captured by the direct participants in the economic exchange. For example a production process may lead to appreciable pollution (an ‘external cost’) that impacts adversely on society; while an inoculation campaign against a major disease may benefit not just those who pay to be inoculated but others too by stemming the spread of disease (an ‘external benefit’). Public goods: These are goods and services where non-payers cannot be excluded from the benefits of the provision of a good or service; at the same

26

Regulation through Agencies in the

EU

time supplying one person with the good or service does not prevent supply to someone else. This means that the economic (opportunity) costs of extending supply to additional consumers is zero or negligible, so that there is no economic case for restricting the supply. In other words, public goods are associated with the two conditions of non-excludability and non-rivalry. Where a good or service is non-excludable, there will be a tendency for consumers to ‘free ride’, refusing

to pay for its provision but nonetheless benefiting from its consumption. Incomplete information: Markets will tend to allocate resources inefficiently where there are important information imperfections. For example, depositors may not know the true liquidity of the bank in which they put their money or of the company whose shares they buy. They may also be unaware of healthy and safety risks posed by products they buy. In addition, people may underestimate the risks of ill-health or having inadequate income in old age and therefore under-provide through private insurance. Markets work best where consumers and producers are well informed (ideally, perfectly or completely informed). Where information is incomplete, adverse selection and moral hazard can lead to ‘bads’ driving out ‘goods’ in market exchanges. Adverse selection occurs where one party ex ante to the contract exploits an information asymmetry to negotiate an especially favourable contract (e.g. selling to an unsuspecting party a defective second-hand car). Moral hazard arises where, ex post to the contract, one of the parties has an incentive to exploit the terms of the contract to the disadvantage of the other party (e.g. reducing care when driving because motor insurance has been obtained). Both adverse selection and moral hazard occur due to asymmetric information in market contracting. Both involve what has been termed in the literature as ‘opportunism’ or ‘self seeking with guile’. Incomplete markets: The market may have frictions so that price signals do not produce a socially optimal allocation of resources; for instance where there is factor immobility or other kind of transaction costs. Another example of this form of market failure is where markets are ‘missing’ or under-developed due to absence of appropriate legal framework, as for instance, to protect property rights. Monopoly/cartels: Markets may not be competitive. Competition or antitrust laws and economic regulation of natural monopolies exist to protect consumers from monopoly prices, poor quality services and cartel behaviour. This can be summarised, as it is in EU competition law, in terms of preventing firms from ‘abusing a dominant position’ in markets and preventing ‘restrictive and concerted practices’ between firms that reduce competition. Economic regulation is particularly evident where firms have a natural monopoly. Natural monopolies exist where there are sufficient economies of scale or scope in production so

Regulation of Liberalised Markets:

A New Role for the State?

P54

that competition raises supply costs. This is most likely where there are important sunk costs in the form of networks, pipelines and similar high fixed-cost infrastructure. Examples of industries with important network effects, and therefore where competition is restricted by the technology and economics of the industry, include electricity and gas transmission and distribution, rail infrastructure, fixed line telecommunications (though less so today because new technology is reducing network costs) and water and sewerage services. Social/political reasons: Governments intervene for other reasons as well, such as to reduce income inequality. Society may decide that free market outcomes are unacceptable because of the resulting distribution of income and wealth. Redistribution has been an important reason for state intervention in both developing and developed economies. Redistribution involves interference with private property rights (wealth redistribution) or interfering with the outcomes in terms of revenues received from these property rights (income redistribution). Merit and demerit goods: There may be some types of goods and services where society considers supply should not be restricted to those willing or able to purchase. Oft-cited examples of merit goods are education, housing and health care. Equally, there may be demerit goods, such as certain drugs, tobacco and alcohol, where society considers that supply should be prohibited or reduced by regulation and taxation. The importance of merit goods is controversial, however. The argument leads to allegations of a ‘nanny state’ because the state overrides the normal market signals and augments the supply through direct provision, regulation of private markets (e.g. price limits) or by providing subsidies.

Although the causes of market failure cited above provide a necessary condition for government intervention they do not constitute a sufficient case for intervention. The act of regulation itself may be too costly or prone to failure. Together with market failure there is also the possibility for government failure. The main causes of this kind of failure are the following: Imperfect and/or asymmetric information: Even well-meaning public measures may not be well designed. Those that have to comply with the regulations may exploit them or escape from their scope. Regulations may also create moral hazard, inducing agents to carry out more of the activity that is supposed to be avoided. Asymmetric information is also.a source of principalagent problems, whereby governments (the principals) have difficulty monitoring and assessing the performance of those they appoint as regulators (their agents). I will return to this theme later on.

28

Regulation through Agencies in the

EU

Rent-seeking: Economic agents try to secure and exploit public measures for their own benefit. Politicians, acting selfishly but rationally, respond to lobbying and other forms of political support by delivering public measures that protect the interests of their supporters. In summary, the fact that markets are imperfect does not mean that governments have the knowledge or the instruments to correct them, nor does it mean that whenever governments intervene they do so with altruistic motives or with the general good at heart. With these caveats in mind, the answer to the question posed at the beginning of this section is that governments may still have to intervene in liberalised markets because liberalisation removes only one of the causes of market imperfection — that of statutory monopoly. Other causes of imperfection remain and, therefore, regulation may in principle be required. Moreover, removal of statutory monopolies does not necessarily result in removal of structural monopoly, as the incumbent in an otherwise open market may still have significant market power. Nor, can the break up of a dominant incumbent guarantee that dominance will not be acquired again in the future. If liberalisation does not eliminate the need to regulate, the next question is whether it changes anything about the method or target of regulation?

WHAT TO REGULATE? MARKET BEHAVIOUR AND REGULATORY INSTITUTIONS Regulation, like any other policy, consists of at least two components: the substantive rules and the procedures for enforcing those rules. The substantive rules concern such things as price caps, provision of information to consumers, minimum qualification requirements for practitioners, etc. The procedural rules refer to things such as the processes for investigation, decisions, appeals, etc. In terms of substantive rules, there is a voluminous literature on this issue

and a consensus has emerged that when there are well-defined structural problems sector-specific regulation has the advantage over general competition rules in that it imposes specific and ex ante obligations of preventive nature (e.g. price caps, accounts separation, interconnection agreements, etc.) and that for this reason it has a wider range of instruments at its disposal. This advantage disappears as market structure changes. Therefore, during the opening stages of liberalisation, regulation should take the form of sector-specific rules. In later stages and as competition intensifies, sector-specific rules are gradually replaced by general competition rules. For example, in network industries regulators face the following dilemma in setting substantive rules during the early stages of liberalisation. On the one

Regulation of Liberalised Markets: A New Role for the State?

29

hand, they have to ensure that incumbent operators do not exploit consumers by charging excessively high prices or squeeze out competitors who seek interconnection to the network. On the other hand, setting prices too low, reduces profits and discourages further investments by the incumbent or the laying of alternative networks by competitors. Low retail prices and low interconnection charges favour competition in services. Higher prices and charges stimulate competition of networks. In essence, regulators have to perform a balancing act: deliver short-term benefits to consumers through low prices without discouraging long-term investments by extracting all profits from the operators. The intensity of this trade-off probably lessens as markets become more competitive in terms of both the number of players and the number of networks. On the institutional side, when governments regulate, they normally do so by assigning regulatory tasks to certain authorities. These authorities may be within existing ministries or departments or they may be independent agencies. It is for this reason that some authors have distinguished issues of substantive rules from issues of regulatory governance and they have used the term regulatory ‘framework’ to cover both the rules and the institutions responsible for regulation. In this context, governance is broader than regulatory procedures. Governance concerns also the relations between the government, the regulatory authority, the courts and the regulated industry and, in particular, the supervisory powers of the government, the powers assigned to the regulator, the extent of the review that may be carried out by the courts and the rights and obligations of the regulated industry. While there is a voluminous literature on the substantive rules of regulation, analysis of governance is relatively less extensive and more recent. This is especially true in the European Union where independent regulators have been established during the past decade in conjunction with the opening of markets which used to be statutory monopolies. Regarding regulatory governance, the primary concern of the literature has been expressed in two questions: why should regulatory tasks be assigned to a separate authority and why should that authority be independent. The answer to the first question is for reasons of efficiency. The answer to the second question is for reasons of consistency.’ Assignment of regulatory tasks to a separate authority enables that authority to develop technical, economic and legal expertise and to benefit from economies of scale in dealing with many similar cases. If the authority also has a narrow mandate it can also ignore extraneous or conflicting policy concerns that can dilute its aims and inject ambiguity and trade-offs in its decisions. The question of independence stems from the need to arrive at consistent decisions over time. Some authors have emphasised the need for long-term commitment so that the policy pronouncements of politicians become credible.

30

Regulation through Agencies in the

EU

Since no regulation can be completely defined in advance, regulation is a form of ‘incomplete contract’ that can be exploited by those that set the rules; i.e. the government. An independent authority that has a narrow mandate and does not take into account extraneous factors such as re-election or lobbying demands, would be less prone to change the rules. Similarly, if companies are to invest in network industries they need to be assured that there will be no policy reversal after they make their investments. This is because this type of investment is characterised by significant fixed costs that become sunk after investment. A politician would be tempted to try to expropriate their profits by regulating their prices so that they cover only their marginal costs. In effect, shareholders forfeit their profits, including the return on their capital. Since this ‘time inconsistent’ behaviour raises the costs of investment, there is less investment in regulated sectors under these conditions. The creation of independent regulators who apply stable and predictable rules is a way of encouraging investment. Another explanation has been recently advanced in the context of international negotiations such as those that take place in the European Union. Independent regulators are needed not because the rules cannot be made more precise but because the rules are intentionally left vague so as to achieve agreement among partner countries.* Agreement on some common rules is beneficial to all partner countries because their companies avoid the costs of complying with multiple rules. On the other hand, however, each partner country has to deviate from its national optimum. This is costly. Common rules can be adopted only when the negotiated outcome improves on the pre-negotiation or pre-integration status quo. The model developed by Nicolaides et al. (2003) and Nicolaides (2004) shows that even if the post-integration outcome is superior to the pre-integration situation, partner countries still have incentives to deviate from the agreed rules.° This is because of two reasons. First, the agreed rules are not optimum for each country even if they make each country better off. Second, each country has the possibility to deviate because the rules are not precise enough. The imprecision was necessary during the negotiations in order to make the outcome acceptable to all parties. This means that the ambiguity that makes the conclusion of negotiations possible creates a compliance problem later on. The solution to this compliance problem is to establish independent regulatory authorities with a narrow mandate. When markets are not competitive, governments should regulate for the purpose of simulating competition. If effective regulation means optimum simulation of competition, it follows that well-functioning regulatory governance is the institutional counterpart of well-functioning markets. Independent regulators should provide the institutional stability that eventually should stimulate investment and intensify competition.

Regulation of Liberalised Markets:

A New Role for the State?

54

Once independent regulators are created, the role of the government is to lay down the overall policy objectives and empower the regulators to adopt instruments to achieve those objectives. Within the EU, the overall policy objectives in liberalised sectors are determined to a large extent by the EU; but not completely. The next section elaborates on this incompleteness. So the answer to the question posed at the beginning of this section is that as long as significant imperfections remain in the market, the government may have to create a regulatory framework that comprises both substantive rules and institutions. The focus on sector-specific rules changes with the intensity of competition and structural evolution of the market. There is also a declining need for the government to define precise substantive rules as the institutional side and especially the regulators become more experienced and more credible. So in the presence of a well-designed regulatory framework, the government need only lay down the overall policy aims. But now we have arrived at two conundrums. The first is that liberalisation which is intended to inject competition requires independent authorities to bring about institutional and policy stability that stimulates investment. But if regulators are independent, how can the government ensure that they carry out their tasks correctly? The second is that the existence of regulators takes away the need for the government to define extensive ex ante rules and the existence of the EU takes away the need for the government to define overall policy objectives. But do these mean that there is nothing for the government to do after it establishes the regulatory framework? Is the task of the government one-off or continuous? The next section provides some answers to these questions by identifying some of the gaps in EU policy frameworks.

GAPS IN EU POLICY

FRAMEWORKS

Although EU directives such as those on telecommunications, electricity and postal services are very detailed, they still leave considerable leeway to member states to decide how to implement them. After all, this is the meaning of

directives. To demonstrate the extent of the discretion of the member states I will confine myself to institutional issues in the field of telecommunications.° Consider below the governance relationships defined in the main telecommunications Directive. The main governance provisions of Directive 2002/21 are as follows: Art. 3 (2):

‘NRAs must be legally distinct from and functionally independent of all organisations providing electronic communications networks, equipment or services.’

Regulation through Agencies in the

32 Art Art

3 (3): 3 (4):

Art 7

(Sy

Art

.4(1):

Art

5 (2):

Art

.7 (3):

Art

7 (4):

EU

‘NRAs exercise their powers impartially and transparently.’ ‘Member States shall ensure consultation and cooperation between NRAs and NCAs and other national authorities.’ ‘NRAs and NCAs shall provide each other with the necessary information.’ ‘Effective mechanisms must exist at national level under which any user or undertaking who is affected by a decision of an NRA has the right of appeal to an independent appeal body ... Pending the outcome of any such appeal, the decision of the NRA shall stand, unless the appeal body decides otherwise.’ ‘NRAs shall provide the Commission with the necessary information ... the Commission shall make that information available to another NRA in another Member State ... NRAs make information available to another NRA after a substantiated request.’ “Where an NRA intends to take a measure that affects trade or relates to market definition or market analysis, it shall make its draft measure accessible to the Commission and the other NRAs, together with its reasoning ... NRAs and the Commission may make comments.’ “Where the Commission considers that the draft measure would create a barrier or has serious doubts as to its compatibility, the draft measure shall not be adopted for two months ..., on

the basis of detailed and objective analysis, the Commission may take a decision requiring the NRA to withdraw the draft measure.’ Art eid boa

“NRAs shall take the utmost account of comments from other NRAs and the Commission and, except in cases under Art. 7

Art. Art lS (3):

Art

16 (1,2,4):

(4), may adopt the resulting draft measures and shall communicate them to the Commission.’ ‘NRAs shall promote competition ... the development of the internal market and ... the interests of citizens.’ ‘NRAs, taking the utmost account of the Commission recommendation and guidelines, define relevant markets appropriate to national circumstances, in accordance with the principles of competition law. NRAs shall follow procedures in Arts. 6 & 7 before defining markets that differ from those defined in the recommendation.’ ‘NRAs shall carry out an analysis of the relevant markets ... in collaboration with NCAs ... to determine whether a relevant market is effectively competitive.’

Regulation of Liberalised Markets:

Art. 19 (1):

Art. 21: (2):

A New Role for the State?

33

‘The Commission may issue recommendations on harmonised application of the directives ... Where an NRA chooses not to follow a recommendation, it shall inform the Commission giving its reasoning.’ *“NRAs shall coordinate their efforts in order to bring about resolution of disputes.’

The independence required by the Directive is different than the independence defined in the regulatory literature. The Directive in essence seeks to prevent conflicts of interest. The functional independence defined in the literature means protection from political interference. In the literature the role of independence is often examined in conjunction with accountability mechanisms. In this context, I define accountability to mean the obligation of a public authority to explain and justify its decisions. The Directive mentions appeals and judicial review, exchange of information with the Commission and other NRAs. Although these provisions impose a degree of accountability, there is neither direct mention of it, nor do these provisions cover all aspects of NRA tasks. For example, the Commission itself has found that judicial review is mostly confined to procedural issues, whereas most of the literature refers to substantive accountability which is accountability concerning the contents and effects of regulatory decisions. The literature defines several different aspects of accountability.’ The typical features of an accountable regulator concern obligations to consult widely, reach decisions in an open and transparent manner, respond to feedback and explain and publish decisions. Afterwards, judicial review is also thought to be indispensable. Some authors, in addition, favour reviews by specialised panels of experts and periodic peer reviews and performance assessments. None of these is required by the telecoms Directive. In fact, as far as I know, with two exceptions, EU regulations and directives do not oblige member states to evaluate the performance of their national authorities or agencies. The only two exceptions Iam aware of are the accreditation of national agencies managing EU agricultural funds and national authorities certifying the conformity of products with technical and safety standards. To conclude so far, it is largely up to member states to monitor and evaluate the performance of their regulators. If, as was concluded in the previous section, effective regulatory systems are founded on independent, autonomous or empowered regulators, then the most important task of the government is to ensure that regulators are accountable so that they perform as they are expected.

34

Regulation through Agencies in the

EU

MONITORING AND MEASURING REGULATORY PERFORMANCE IN THE EU The literature on the principal-agent theory has come up with many suggestions on how agents may be forced to perform according to the preferences of the principal. The solutions proposed usually fall into two categories: ex ante monitoring and ex post reviews. Most measures intended to improve the performance of regulators impose procedural obligations such as consultation, adequate justification and publication of decisions. Some measures also require the regulators to carry out ex ante and ex post impact assessments of the regulations in question. Ex post performance reviews may look into the internal administrative arrangements of the regulator, or the quality of its decisions or the actual effects that those decisions may have had on services, market structure, competitors and consumers. These ex ante and ex post instruments are not panaceas. The root of the problem in principal-agent relationships is that important aspects of the behaviour of the agent may be neither observable, nor measurable. For example, in terms of ex ante monitoring it is very difficult to know whether someone is trying hard enough to come up with innovative ideas to solve regulatory problems. In terms of ex post assessment, it is impossible to know what would have happened if a different measure were adopted. Since EU directives do not require that NRAs perform efficiently, a natural role for the government is to carry out period reviews of the efficiency of the internal administrative and other procedures of the regulator. In addition, the government should look into what happens after the regulator decides. This means two things: the effect of the regulations in general (their economic impact) and the outcome of the decisions in particular (the effectiveness of enforcement). The former requires an ex post impact assessment of the effects of regulation on the economy and consumers. The latter concerns how quickly individual decisions translate into action and what their specific effect may be. With respect to enforcement, the telecoms Directive 2002/21 states that the decisions of the regulators are subject to appeal. This Directive and, as far as | know, directives in other sectors do not specify how the appeals process should be organised. In some member states, such as Denmark and the Netherlands,

there are specialised tribunals or specific courts that rule on appeals concerning regulatory decisions. In other member states, the relevant appeals bodies are administrative courts or higher courts. In the UK it is the competition authority which is responsible to hear appeals. The Commission is currently concerned that incumbents may clog up the system by systematically appealing every decision of the regulator. There are two issues here: the subject matter of the appeals and whether appeals have

Regulation of Liberalised Markets: A New Role for the State?

a6

suspensive effect. The first issue relates to the ability of the appeals court to examine both procedural and substantive aspects of regulatory decisions. The government should ensure that the tribunal that hears appeal cases is indeed equipped in terms of the requisite expertise for that purpose. Moreover, national judges should not render judgements which are divorced from Community law. At minimum, they should be trained in EC law and be provided with up to date information on relevant case law and regulatory practices in other member states. Naturally, national judges can always ask the ECJ for preliminary rulings. With respect to the second issue, appeals should not result in suspension of regulatory decisions. Otherwise, the incumbents will abuse the opportunity to appeal. How about the quality of the decisions of the regulators? Oversight of national regulatory decisions is the task of the Commission. The question that arises is whether the Commission’s monitoring covers all aspects of the decisions of the regulators? The answer is that it does not. More precisely, the Commission has three primary tasks. First, as guardian of the Treaties, it initiates proceedings against the member state concerned whenever EU rules are not applied or misapplied. This can be the case when the member state fails to transpose them properly or when the regulator fails to enforce them properly. In either case the Commission can initiate proceedings only when there is a legal infraction. It cannot take legal action to force, for example, a national regulator to set prices at EUR 0.5/minute instead of EUR 0.6/minute even if it believes that the first price is more efficient. The second primary task of the Commission is to review those decisions of the regulators that according to the directives have to be notified to itself and other NRAs. Not all regulatory decisions are formally notifiable. The third task of the Commission is to manage the networks of European regulators. In the case of telecoms, the Commission has established the European

Regulators Group (ERG). Although the ERG has no formal decision-making power, in principle it can discuss any regulatory issue. The lack of power allows it to have a very wide agenda. Indeed, a cursory examination of the items that have been placed on the agenda of ERG for 2004 reveals that a very broad range of issues are due to be discussed.* The ERG has been established not only to advise the Commission but also to provide a forum for members to exchange views on issues of common interest and learn from each other through peer reviews. Although its work programme is comprehensive and involves much comparative analysis, it remains to be seen whether there is indeed tangible convergence towards best practices.° I have written elsewhere that if ‘competition’ among different national approaches and methods will lead to convergence towards best practices, the

36

Regulation through Agencies in the

EU

dominant role of the Commission should be weakened.'° The Commission should be under obligation to promote regulatory rules and methods for which there is a strong support among NRAs. In addition, if NRAs are to adopt such rules and methods, they should be obliged to adopt best practices or duly justify deviation from such practices. Even though regulators within the ERG are free to express opinions, they are not bound to explain their own regulatory practices or change them for as long as they are legally correct. For best practices to spread, there should not only be systematic identification of such practices but also systematic assessment of the practices and performance of NRAs. So far this has not happened. That is regrettable, because the EU offers a golden opportunity. It is the only jurisdiction in the world where distinct authorities have to apply similar rules. It is as if it has set up a laboratory on a huge scale that will allow different approaches to be tested. It should now go a step further and initiate the process of gathering of data and comparing the effectiveness of the various national approaches. This should be the new role of the state after markets are liberalised. It should oversee the regulators, introduce a sort of ‘competition’ among them and adjust the national regulatory framework in light of the findings on regulatory best practices. In the next section I show formally why this process needs to be systematised and why there can be no presumption that just because national authorities have similar tasks, somehow there will be convergence of regulatory practices.

A SIMPLE MODEL OF REGULATORY CONVERGENCE AND DIVERGENCE In the previous section I argued that if the networks of European regulators are to bring about improvement in the performance of regulatory authorities, peer reviews should be managed more closely. Simply because regulators talk to each other, it does not necessarily follow that there will be a natural tendency to identify best practices and adopt them. In this section I want to show that, first, there can be both convergent and divergent tendencies and, second,

explain why identifying a ‘European’ solution or practice is not necessarily good for the Community. So the interaction between regulators needs to be managed. In other chapters I have examined in more detail the reasons why countries agree to harmonise their regulations and what the likely outcomes may be from this kind of harmonisation." As indicated above, the purpose of this section is to explore two related themes: under which conditions there is regulatory convergence or divergence of regulatory practice and whether there can be excessive harmonisation.

Regulation of Liberalised Markets: A New Role for the State?

ay

I begin by assuming that when countries determine their regulations independently or in isolation they set their rules with respect only to their own specific economic and social conditions. This immediately raises the question why countries would voluntarily agree to harmonise their regulations. Harmonisation means that they have to shift away from their national optima. The answer is that integration of regulatory frameworks allows companies from all partner countries to operate under a single set of regulations. This saves them from the extra costs inflicted on them by regulatory duplication and fragmentation. They do not have to obtain authorisation more than once and they can reap whatever economies of scale are possible. Table 2.2 summarises the main theoretical arguments in favour and against regulatory centralisation and decentralisation. Table 2.2

Regulatory Centralisation vs Decentralisation

Reasons for centralisation of regulation

Internalisation of externalities: Single regulator takes into account all effects of the rules Economies of scale and avoidance of regulatory duplication: Uniform regulation avoids market segmentation and multiple compliance costs Prevention of race to the bottom: Avoidance of competition among regulators Expertise: Single regulator gains by dealing with higher number and variety of cases Avoidance of capture: Single regulator not likely to be as vulnerable to lobbying pressure as smaller, regionally based regulators Reasons for decentralisation of regulation

Diversity of preferences: Locally set rules reflect more closely local preferences Opportunities for experimentation: Comparison of diverse local regulations reveals their strengths and weaknesses Source:

R. Hahn, A. Layne-Farrar and P. Passell, ‘Federalism and Regulation’, Regulation (2003-2004), Winter, pp. 46-50.

38

Regulation through Agencies in the

EU

Therefore, regulatory integration, like most other forms of integration, has both positive and negative effects. The positive effect here is encapsulated in the benefits from the ‘commonality’ of the rules. The negative effect is the cost from deviating from the national optimum. For the purposes of this chapter, we can model these effects by measuring their impact on national welfare in terms of deviations from the pre-integration national positions. Assume for simplicity that there are only two partner countries, H and F, and that regulation can be expressed in cardinal values. Before integration the prevailing regulatory value is Rh in H and Rf in F, respectively. The gains from regulatory integration are measured by functions H (Rh — Re) and F (Rf— Re), respectively. Re is the value of the common post-integration rule that is to decided through negotiations between the partner countries. These are shown in Figure 2.1.

EU max

Rh Figure 2.1

Rec*

Rf

Convergent Regulation

To simplify the exposition, I further assume that the benefits for each country are maximised if the other would adopt its own regulations. That is, the best option for each country is Rh = Re and Rf = Re. This is shown in Figure 2.1

Regulation of Liberalised Markets:

A New Role for the State?

39

where each country achieves maximum gains by having its own rule become the common rule. Naturally, this cannot hold for both of them at the same time,

unless they happen to be identical. This is why they have to negotiate to determine the value of Re. In my other chapters I mentioned above, I explore in more depth the problems caused by this kind of bargaining. I conclude that one of the reasons for which regulatory authorities are needed is because negotiations leave too much ambiguity in the agreed rules. In this chapter I do not examine how countries bargain. Rather my intention is to look at possible outcomes, irrespective of how they are determined. However, I assume that whatever value of Re is chosen, it must be such that it

makes both countries better off in relation to the pre-integration level of regulation. That is why Figures 2.1 and 2.2 show also the sum of H and F for those portions that are non-zero.

Rh Figure 2.2

Re*

Rf

Divergent Regulation

Under these conditions, the best outcome for the EU as a whole is rather easy to identify, at least in principle. Expressed mathematically, the objective of the EU as a whole is to find a value of Re that maximises joint benefits. That is,

the EU maximises:

40

Regulation through Agencies in the

EU

max E (Re) = H (Rh — Rc) + F (Rf— Re)

(2.1)

The first-order condition for maximisation requires that the Re is set at such value so that: (dE/dRc) = 0 which implies that (dH/dRc) = — (dF/Rc)

(2:2)

The derived value of Rc, Rc*, is the optimum value or level of regulation for the EU as a whole. To ensure that Rc* is the global optimum, we also need to have: (dE/dRc) > 0 for Re < Rc* and (dE/dRc) < 0 for Re > Re*

(2.3)

Naturally, each country will be willing to agree on harmonisation of its regulations only if: Rc* > Rfand Rc* > Rh, where, Rf and RA are the pre-accession equilibrium points for each country.

(2.4)

Unstable Outcome

It is important to note two special features of this negotiated outcome. Even if the post-integration level of regulation improves on the pre-integration level of regulation and each country becomes better off as a consequence, the post-integration state of affairs is not stable. The reason is that even though (dE/dRc) = 0, (dH/dRc) # 0 and (dF/Rce) # 0, each country has an incentive to deviate from the agreed level of regulation. Under these conditions, EU rules have to be narrowly defined so as to prevent member states from deviating from the EU optimum of Re* through creative interpretation of the rules. It also makes sense to confer sufficient powers to the Commission to control national regulatory authorities. Here there should be convergence of national practices. This situation is depicted in Figure 2.1. We may call this the convergent or homogeneous case. The dotted lines indicate the boundaries of the common rule. Sub-optimum Outcome

It is possible, however, to have cost configurations that lead to different outcomes. For example, if the change in the derivative of E (Re) is as follows:

(dE/dRc) < 0 for Re < Re* and (dE/dRc) > 0 for Re > Rc*

(2.5)

Regulation of Liberalised Markets:

A New Role for the State?

4]

the derived value of Re* is minimum instead of maximum. Under these conditions a corner solution is superior. In other words, the agreed rule should be wide and flexible enough to allow member states to deviate from the average or middle value and opt for the best national practice. It may even be collectively advantageous for national practices to differ without causing too much harm due to differences in the national implementing rules of the common rule Re. Under these conditions divergence rather than convergence should be promoted. This situation is depicted in Figure 2.2. We may call this the divergent or heterogeneous case. Mathematically, it is easy to describe the conditions that lead to this divergent outcome. Take, for example, country H. Its regulatory function is described by first and second-order derivatives which are: (dH/dRc) < 0 in both diagrams, and

(2.6)

(d?H/dRc’) < 0 in Figure 2.1 but (d?H/dRc’) > 0 in Figure 2.2

Ga)

In plain words these conditions mean that for sufficiently small deviations from the pre-integration level of regulation, the benefits from common rules outweigh shifts from the national optimum in the convergent case (Figure 2.1) while the cost of shifting away from the national optimum overwhelms any benefits from common rules in the divergent case (Figure 2.2). Policy Implications

If the results derived above are any representative of real world processes, there are at least three significant policy implications. First, it cannot be presumed that narrow and inflexible EU rules that do not allow much discretion to national authorities are always good for the EU as a whole. Convergence of national practices is good only when the benefits from commonality are strong enough so as to compensate for shifts away from national preferences. Second, the process for identifying best practices differs in each case. In the convergent case, the emphasis should be in selecting practices which can be replicated across member states. In the divergent case, the emphasis should be in finding out practices that can address the particular problems and that can be adapted to suit the conditions of each member state. Third, if, and this is a big if, the European regulator networks under the management of the Commission have a natural tendency towards convergence, it cannot be presumed that ‘Europeanisation’ of this kind is always good for the member states. This means that member states need to maintain some supervisory responsibility over the conduct of their regulators. They cannot rely on the Commission’s oversight.

42

Regulation through Agencies in the

EU

CONCLUSIONS This chapter has asked a number of questions on the role of the state in liberalised markets. The first question is on the role of the state after markets are liberalised. The opening up of markets does not remedy all market imperfections and therefore does not eliminate all reasons for government intervention. This is even more so for as long as the operators of essential networks such as those for telecoms are dominant firms. The second question that has been raised in this chapter is whether the establishment of independent regulators and the setting of overall policy objectives by the EU remove the need for governments to regulate. Again the answer is not completely. Effective regulation requires sufficiently independent and accountable regulatory institutions. The government acts as a principal who regularly monitors and evaluates the performance of regulators. With respect to setting policy objectives at EU level, EU rules are not detailed enough, nor do they cover important facets of regulatory governance such as the efficiency of the appeals process. So again, governments have an important role in overseeing the functioning of the national regulatory systems. Lastly, the chapter has suggested that this oversight can be made more systematic by exploiting the opportunity offered by the EU processes for comparative performance assessment, peer reviews and identification of best practices. In some cases, convergence of national practices is desirable, but in other cases variability across member states may be a more efficient solution. Since, member state governments are not members of the regulators’ networks, it may be necessary to introduce new EU rules on peer reviews. At any rate, the member states themselves should maintain sufficient surveillance over the performance of their regulators and should periodically examine whether the practices adopted by those regulators address the problems in their economies. They should not rely on the Commission to blow the whistle on under-performing regulators. In conclusion, after the introduction of competition in markets, it may be expedient to introduce competition in regulatory systems as well.

NOTES ' See G. Yarrow (1986), ‘Privatization in Theory and Practice’, Economic Policy, 2, April; J. Vickers and G. Yarrow (1986), Privatisation, Cambridge, MA: MIT Press and M. Armstrong, S. Cowan and J. Vickers (1994), Regulatory Reform: Economic Analysis and British Experience, Cambridge, MA: MIT Press. ? For a more detailed review see A. Kahn (1988), The Economics of Regulation: Principles and

Institutions, Cambridge: MA: MIT Press; R. Baldwin and M. Cave (1999), Understanding Regulation: Theory, Strategy and Practice, Oxford: Oxford University Press; R. Hahn, Policy Watch (1998), “Analysis of the Benefits and Costs of Regulation’, Journal of Economic Perspectives, 12, 250.

Regulation of Liberalised Markets:

A New Role for the State?

43

3 See the analysis in J. Stern (1997), ‘What Makes an Independent Regulator Independent?’ Business Strategy Review, 8 , 67; J. Stern and S. Holder (1999), ‘Regulatory Governance’, Utilities Policy, 8, 33; J. Stern and F. Trillas (2003), ‘Independence and Discretion in Telecommunications Regulation’, Utilities Policy, 11, 191. 4 See G. Majone (1996), Regulating Europe, London: Routledge; G. Majone (1999), ‘The Regulating State and its Legitimacy Problems’, West European Politics, 22, 1; M. Thatcher (2002), ‘Delegation to Independent Regulatory Agencies’, West European Politics, 25, 125. > P. Nicolaides, A. Geveke and A.-M. den Teuling (2003), Jmproving Policy Implementation in an Enlarged European Union, Maastricht: EIPA; P. Nicolaides (forthcoming 2004), ‘The Political Economy of Multi-tiered Regulation in Europe’, Journal of Common Market Studies. © For a review of the EU system of regulatory authorities in all policy fields see the relevant section in European Commission, Green Paper on Services of General Interest, COM (2003) 580 final. 7 See T. Bergman et al. (eds), Delegation and Accountability in West European Parliamentary Democracies, Oxford: Oxford University Press, 2002 and Stern and Trillas (2003) op. cit., see supra note 3.

8 The full work programme of the ERG can be accessed at www.erg.eu.int/doc/work_progr_2004/ erg0403.doc.

9 Peer reviews do not solve all problems. The OECD has examined in detail the design, organisation and limits of peer reviews. Its main documents on peer reviews can be accessed at its websité (www.oecd.org) under the topic of regulatory reform. See in particular Peer Review: Merits and Approaches in Trade and Competition Context (Paris: OECD, 2002), COM/TD/DAFFE/ COMP(2002)4/FINAL. It describes the main weaknesses that have been encountered in reviews of

trade and competition policies. Weaknesses of the WTO TPRM: Limited resources; uneven participation in meetings; lack of recommendations or prescriptive elements; becoming increasingly adversarial. Weaknesses of the OECD competition policy reviews: Lack of follow-up. 10 See Nicolaides et al. (2003), see supra note 5 and Nicolaides (2004), see supra note 5. 11 See Nicolaides et al. (2003) and Nicolaides (2004), supra note 5.

7

| 7

-s ‘y 46

NG

7

ey: . ae y .aa, : a

Vere wr iol wiles mined it ‘oe > 4 yo) atee ,

palmer

=

Cy

ade a

te |

Os

=

:

.

ewan rs

pide one

a

|

PART TWO

The Design of Agencies in the EU

JH seit a

3.

Independent Regulatory Agencies and Elected Politicians in Europe Mark Thatcher

INTRODUCTION The 1980s and 1990s saw the creation and/or strengthening of independent regulatory agencies (hereafter, the ‘IRAs’) in Europe. The most significant changes concerned IRAs for market competition — general competition authorities, sectoral utility regulators and financial regulators. Most analyses, both theoretical and empirical, have focused on formal institutional arrangements — how and why elected politicians delegated to IRAs, the institutional form and powers of agencies and the powers remaining in the hands of elected politicians. This chapter seeks to examine the stage after formal delegation — how IRAs have behaved in practice and in particular, their relationship with their principals, elected politicians. Empirically, relatively little comparative work across both European nations and policy domains has been done on this.! Theoretically, the dominant principal-agent models treat the relationship between elected politicians and IRAs as one of control by elected principals over their agents, a problem that is dealt with through the formal institutional design of delegation. This chapter shows how formal institutional arrangements leave much room for choice. Moreover, the formal framework of delegation is overlaid with informal relationships and resources. Hence it looks both at how elected politicians have used their formal powers over IRAs, and other aspects of the relationship between the two sets of actors. It argues that elected politicians have chosen, in terms of membership, to constitute IRAs as a separate set of actors, distinct from elected politicians, and often not to exercise other formal

controls over IRAs such as reversing their decisions. It then explores the informal relationships and networks that have developed between IRAs and elected politicians. 47

48

Regulation through Agencies in the

EU

The chapter begins with a brief theoretical and methodological discussion of IRAs and their relationship with elected politicians, using pnineipal-agent models as its starting point; it then sets out the spread of IRAs in Europe in key market domains, their formal institutional autonomy from elected politicians, the controls that remained in the hands of politicians and expectations at the time of their creation in the 1980s and 1990s.It shows that elected politicians had great discretion over the use of controls and that expectations of IRA independence were low. Thereafter, it explores the relationships between IRAs and elected politicians through analysis of the use of the latter's formal powers, before discussing informal relationships and wider networks.

PRINCIPAL-AGENT MODELS: IRA INDEPENDENCE FROM ELECTED POLITICIANS Principal-agent (hereafter, the ~PA*) analyses are the dominant model of delegation to non-majoritarian institutions such as IRAs, albert that they are open to challenge and criticism.* They analyse changes in the formal institatonal framework, namely principals (in our case, elected politicians) creating and

giving the agent formal powers, but also limiting its discretion through controls, including those retained by principals. Those principals’ aim is to give the agent sufficient autonomy to fulfil the functional purposes for which it was set up (for instance, to enhance credible commitment or shift blame), but to limit ‘agency losses’, namely behaviour by IRAs contrary to the preferences of their principals. Agency losses can arise from two sources: “shirking’, because the agent follows its own preferences which diverge from those of its principal(s ‘slippage’ due to institutional incentives causing the agent to behave contrary te the wishes of its principal(s). To reduce ‘agency losses’, PA analyses look to formal institutional design. In particular, principals can establish mechanisms to attempt to reduce their losses, notably administrative procedures that apply before the agent acts (such as screening and selection mechanisms for the choice of agents), and ongoing devices that apply after delegation* Thus principals create formal controls such as powers over appointment, dismissal, budget setting and review or reversal of agents” decisions, to attempt to ensure that agents follow their preferences. Although PA frameworks have increasingly concentrated on formal institutional design,* they also provide a useful starting point for analysis of post-delegation relationships, if only in structuring discussion and offering assumptions and terms. First, they underline that whatever legal terms are used, IRA ‘independence* from elected politicians is not total (or zero), but varies. Second, they draw our attention to the expectations and calculations of principals at the time of delegation: principals know that IRAs will have a

Independent Regulatory Agencies and Elected Politicians in Europe

49

degree of discretion and hence a zone within which they can take decisions that may run counter to the preferences of their principals. Third, PA frameworks point to the importance of formal controls and institutional frameworks in the behaviour of IRAs and their relationship with elected politicians. These structure the nature of the relationship between the two: even if agencies are labelled ‘independent’, they are created by elected politicians and operate in a framework created by the latter and in which the latter continue to have powers over them. However, PA frameworks are also open to criticisms in the study of postdelegation behaviour. First, the concept of agent ‘independence’ is difficult to define and even more tricky to identify empirically. It faces the problems that principals may have incentives to hide whether IRAs are following their own preferences, and/or vice versa, not to admit agency loss. If the use of sanctions is utilised as an indicator, for instance, agents may appear to be autonomous if principals do not apply sanctions, but this may be due to controls being so. effective and agents rationally anticipating principals’ responses that sanctions do not need to be used, thereby producing ‘observational equivalence’.° Second, the focus on agency independence and agency losses may lead to a misleading focus on conflict between principals and IRAs, ignoring their common interests and cooperation. Third, as Moe points out, the ability of the principal to influence agency performance depends on both control mechanisms and its use of those mechanisms. The latter cannot be taken for granted.° Principals must have the desire and energy to use the controls; many different applications of controls are possible (for instance, appointment powers can be used to select political cronies or independent-minded individuals).’ A fourth issue is that the attention to formal instruments of control may also miss the many informal aspects of relationships between elected officials and IRAs that may go beyond and/or subvert formal institutional frameworks. Finally, the PA framework may be too static. Within the same formal institutional framework, relationships may change due to factors such as learning or the development of cooperation or norms as part of ‘iterated games’. The approach taken here is therefore to use PA analyses as a starting point by analysing the formal controls that they suggest are important in post-delegation relationships. Thereafter, however, it also examines informal relationships between IRAs and elected politicians. Since examining IRA independence from elected politicians faces theoretical and empirical difficulties, the chapter looks at conditions that structure the relationship between IRA and elected politicians — both formal controls and informal relations. These conditions may not predict IRA behaviour in particular cases, but do provide the context in which IRAs and elected politicians relate to each other on specific policies and decisions.

50

Regulation through Agencies in the

EU

IRAS IN EUROPE: FORMAL INSTITUTIONAL FRAMEWORKS AND INITIAL EXPECTATIONS Traditionally, very few IRAs existed in Europe — they were very much an American institution. Regulation — better known as industrial policy — was largely a closed game between government and suppliers (public or private), into which occasionally other actors such as trade unions intruded.* This position ended from the late 1970s onwards, as IRAs were created in all major European nations.? Each country has its own legal doctrines and definitions: Italy and France have ‘independent administrative authorities’ ;'° in Germany, there are several types of agency — public and private law agencies, federal and Land agencies, most within or subordinate to ministries,

with even ‘independent agencies’ such as the Federal Cartel Office and the telecommunications regulator, the RegTP, being (legally) subject to supervision and instructions from ministries; Britain does not have a legal doctrine of independent agencies. Moreover, the term ‘agency’ can be used for bodies that are in fact legally subordinate to ministers. Therefore minimum requirements for inclusion as an IRA refer to the formal institutional position and comprise: the agency has its own powers and responsibilities given under public law; it is organisationally separated from ministries; it be neither directly elected nor managed by elected officials. Table 3.1 offers an overview of IRAs in key market domains in the four largest European countries, Britain, France, Germany and Italy with dates of

creation. Table 3.1

Independent Regulatory Agencies for Market Competition in Britain, France, Germany and Italy in Selected Domains

Domain

Britain

France

Germany

General competition

Competition Commission 1998(1948) and OFT— Office of Fair Trading 1973

Conseildela Concurrence 1986 (1977)

Telecoms

Ofcom 2003 (Oftel — Office of Tele-

ART —Autorité de Régulation

RegTP— Regulierungs

AGCOM — Autorita per le

communications, 1984)

des Télécommunications 1996

behdrde fiir Telekommunikation und Post 1996

Garanzie nelle Communicazioni 1996

| Bundeskartel lamt (Federal Cartel Office) 1957

Italy

AGCM —Autorita Garante della Concurrenza e del Mercato 1990

Independent Regulatory Agencies and Elected Politicians in Europe Energy

Ofgem (Office of |Commission de Gas and Electricity Régulation de Markets) 2000 Energie 2003 (1986 for Ofgas (2000) and 1989 for Ofgem)

Water

Ofwat (Office of

51

AEEG — Autorita per l’energia elettrica ed il gas 1995

Water Services)

1989 Railways

Office of Rail Regulator 1993 and Strategic Rail Authority 1999 (1993)

Postal services

Postal Services Commission 1999

Media

Ofcom 2003 (Independent Television

Conseil Supérieur de l’Audiovisuel

No national body — Landesmedie

Commission 1990,

1989 (1982)

nanstalt for

and predecessors — 1954 Stock Exchange/ Shares

FSA — Financial Services Authority 1997 (1986)

RegTP — see telecoms AGCOM (see telecommunications)

each Land

AMF (Autorité Bafin — des Marchés Bundesantalt Financiers) fiir 2003 (COB— __—‘Finanzdiens! Commission eistungen — des opérations 2002 (1995) de bourse 1967 and Conseil des Marchés de Valeurs 1996)

Consob — La Commissione per le societa e la _borsa 1974

Notes: a. Dates refer to legislation creating of the IRA; in brackets are the dates on which an IRA was first created in the domain.

b. Empty boxes — no sectoral IRA; for a discussion of the German model, see Coen, Héritier and Bollhoff 2002, Bollhoff 2002, Eberlein 2000 and Lodge 2000.

c. There are often several financial regulators — the most important regulator for stock exchanges is taken; for Germany, analysis relates to Bundesaufsichtsamt fur Wertpapierhandel, which was absorbed into Bafin in 2002.

52

Regulation through Agencies in the

EU

Expectations of the independence of most IRAs from elected politicians in Europe were low at the moment of their birth. Previous experience had seen a very high level of politicisation of decision making. Many states had no tradition of IRAs and judicial doctrines of state power that ran counter to strong IRAs — notably in continental Europe. Moreover, at times, IRAs were established almost as an afterthought to accompany privatisations (for instance, in telecommunications in Britain in 1984) and with little political debate controversy (Thatcher 1999). Such expectations appeared to be well founded in the institutional framework of IRAs. Ministers kept vital powers, often over licensing and final decisions over mergers and takeovers. Moreover, elected politicians enjoyed strong controls over IRAs, with few restrictions over their use. In particular, they controlled: — nomination of the heads of IRAs — either by ministers (for example, in Britain), or by a combination of heads of the executive and the legislature (for instance, heads of lower and upper houses or election by the legislature), methods common in France and Italy. There are almost no limits as to how elected politicians can exercise those powers nor criteria over qualities that appointees should have; — the power to reappoint IRA members, who were often appointed for limited terms; — determination of IRAs’ budgets and staffing levels — budgets came from public sector funds within annual budgeting and even if raised from industry levies, were set by government ministries; — powers to overrule IRA decisions — for instance, in Germany, the Economics minister can overrule a Bundeskartellampt refusal of a merger/takeover, whilst in Britain, the Secretary of State for Industry had the power to accept or reject the Director General Fair Trading’s ‘recommendations’ as to whether a merger should be referred to the Competition Commission (formerly the Monopolies and Mergers Commission); he/she also had the power to reject the Competition Commission’s advice. The formal institutional framework also offered few limits on the activities and strategies of IRAs. Thus for instance, limits on renewal of terms are rare, whilst conditions and procedures for dismissal of a regulator are mostly unspecified. The objectives set for IRAs are often very broad.'! Similarly, the legislation offered little detail on matter such as IRAs determining their strategies or over activities such as publicising themselves, establishing procedures for consultation or creating networks with other actors. Similarly, very few provisions are made over the internal organisation of IRAs. Thus the formal institutional framework left much room for IRAs to determine their strategies and choose how to develop and use their resources.

Independent Regulatory Agencies and Elected Politicians in Europe

53

THE RELATIONSHIP BETWEEN ELECTED POLITICIANS AND IRAs: RESOURCES AND USE OF DISCRETION IRAs and elected politicians have many resources. However, in part following PA analyses, I focus first on the use of key formals: appointing and dismissing IRA members, and then setting IRA resources, notably their budgets, and reversing their decisions. Thereafter I examine informal relationships and wider networks of IRAs and elected politicians. Membership of IRAs

Nomination of IRA members is claimed to be the most visible and effective formal control over IRAs available to elected politicians available.'? Given the lack of formal controls over choice of nominees, one hypothesis is that elected politicians seek to control IRAs by appointing party politicians — either past politicians who ‘owe’ debts and may wish to re-enter politics — or aspirants whd use IRA membership as a route into politics. Table 3.2 takes two measures of the party politicisation of appointments to IRAs: holding national government office or standing for legislative or local elections; publicly known party affiliation. Party politicisation does not exclude expertise — some individuals may be both linked to parties and be experts in a domain (for instance, several members of the RegTP in Germany or Professor Giuliano Amato as head of the Italian competition authority 1994—7) but at the very least, appointing individuals with clear party links reduces the public distance between IRAs and partisan politics. The results show that in Britain, France and Germany, elected politicians have not used their appointment powers to choose party activists; even the broader category of publicly known party affiliations covers a minority of IRA members and generally arise in communications regulators. Britain is an extreme case: no regulator between 1990 and 2001 had been recently politically active or linked to a party (although a recent exception is the new head of Ofcom, Lord Currie, a Labour peer).'* This does not mean that regulators lack political views but does show that individuals with public ties to parties have not been appointed). Italy is an exception in that almost all members of AGCOM, the communications regulator, have clear party political affiliations and a high proportion have stood or held public office (della Cananea 2002). Premature departure from office provides another indication of the relationship between IRAs and elected politicians. No IRA member has been formally dismissed in the sample. Even resignations have been relatively rare, and have often been for personal or professional reasons, including taking other attractive posts, rather than pressure from elected politicians. There is little evidence of IRA members leaving office when the party in government changes hands.

54 Table 3.2

Regulation through Agencies in the

EU

Party Activism and Public Affiliations of IRA Members 1990-2001

% holding or standing for public office (local, national or European) before or after term on

Britain

France

Germany

Italy

3% (1 of 33)

9% (4 of 46)

15% (2 of 13)

23% (6 of 26)

0 (0 of 33)

46% (21 o0f46)

36% (Sof 13)

717% (20 of 26)

IRA % publicly affiliated with party

Notes: a. Coverage: Britain: members of all sectoral IRAs; only heads of OFT, Competition Commission and predecessor Monopolies and Mergers Commission, SIB/FSA and ITC; excludes temporary interim regulators. France: All members of sectoral IRAs; President of Conseil de la Concurrence and COB Germany: Presidents and Vice-Presidents of Bundeskartellampt, Bundesaufsichtsamt fiir den Wertpapierhandel and RegTP; 1 vice-President of Bundeskartellampt excluded due to lack of information. Italy: All members of AGCOM, AEEG and AGCM. b. Information derived from biographies, newspaper reports and Whos Who.

Table 3.3 shows both total numbers of IRA members resigning and those who left from those IRAs that existed throughout over the period 1990-2001 (the latter to correct for biases due to different lifespans of IRAs). Table 3.3

Resignations of IRA Members before End of Term 1990-2001 Britain

France

Germany

Italy

% resigning before end of term (or retirement if permanent post)

15% (Sof33)

13% (6 of 46)

17% (2 of 12)

19% (5 of 26)

% resigning of IRAs existing 1990-2000

29% (Sof l7)

18% Gof23)

0 (0 of 5)

23% (3 of 13)

Independent Regulatory Agencies and Elected Politicians in Europe

55

It may be thought that politicisation of appointments and early departures may be greatly linked to the characteristics of sectors — especially whether they are highly salient in party politics. One good comparison is between communications IRAs (telecommunications and broadcasting), which lie in highly contentious fields, and general competition authorities, which are generally less salient for party politics. Table 3.4 analyses party politicisation and resignations across the two groups of IRA. Table 3.4

Party Activism and Public Affiliations of Communications and General Competition of IRA Members 1990-2001 Communications IRAs _ General competition (telecommunications authorities and broadcasting)

% holding or standing for public office (local, national or European) before or after term on IRA

16% (9 of 56)

18% (5 of 28)

% publicly affiliated with party

48% (27 of 56)

43% (12 of 28)

% resigning beforeend of term (or retirement if permanent post)

18% (10 of 57)

22% (6 of 28)

Note: Coverage — no broadcasting IRA in Germany; AGCOM in Italy; CSA and ART in France; Oftel and ITC in Britain.

A comparison between the two sets of IRAs suggests that, perhaps surprisingly, the extent of politicisation is only slightly higher in communications than in general competition IRAs, despite the political sensitivity of the former. A similar picture of lack of great cross-domain differences emerges from resignations of IRA members. Cross-domain differences in party politicisation and early departures appear much smaller than cross-national variations. These data show that IRA members are not just a subgroup drawn from elected politicians or altering as governments change, but a group separated from their

56

Regulation through Agencies in the

EU

principals. However, this does not mean that elected politicians refuse to use their powers over nomination and dismissal to influence IRAs. Instead a detailed analysis suggests a more subtle and balanced relationships between IRA members and elected politicians than a simple model of control based on nomination and dismissal. On the one hand, IRA members frequently combine expertise, links to the dominant administrative elite and backgrounds that offer policy outlooks and sensitivity to the government’s policy aims. Thus for example, most senior IRA members in France are members of a grands corps, many of whom have served in ministerial cabinets and form part of the administrative-politico elite (48% — 22 of 46 of the sample in Table 3.1). Even in Britain, the most depoliticised of the four countries, heads of the competition authorities have mostly been economists, businessmen and lawyers likely to believe in the virtues of markets and law (as opposed, perhaps to trade unionists who might attach greater significance to jobs and industrial restructuring), whilst the three heads of Oftel, the telecommunications

authority between

1984 and 2002, were an

accountant and two businessmen, all known for their belief in competition.'* On the other hand, IRA nominees are rarely party hacks — they have their own professions and a degree of distance from party politics. They can and do criticise government policy. This has occurred even in Italy, the most politicised of the four countries — thus heads of the general competition authority have repeatedly attacked slow progress in privatisation and ending state monopolies, whilst the head of AGCOM the communications regulator, Professor Enzo Chelli,

publicly called into question the 2003 Communications Bill (which failed to tackle the conflict of interest between Berlusconi being both Prime Minister and owner of three major television stations). The difficulty and rarity of dismissal means that IRA members can choose whether to resign, especially if they have a profession to return to or the prospect of lucrative employment. Moreover, resignations may inflict embarrassment on the government (for example, resignations from the RegTP in Germany or Sir Bryan Carsberg as head of the Office of Fair Trading in Britain in 1994), as attention is drawn to government failings. Implementation — Staff, Information, Expertise and Reversal of IRA Decisions Regulation of markets is frequently highly technical and legalistic, requiring certain kinds of information and expertise. Thus for instance, decisions on takeovers and mergers can depend on matters such as definition of relevant markets, entry barriers and cost structures; in utility industries such as telecommunications, regulation of prices can be based on whether a supplier has ‘significant market power’ and long-run marginal costs. At the same time,

Independent Regulatory Agencies and Elected Politicians in Europe

57

as noted, the formal institutional framework generally offers IRAs only very broad objectives, leaving them much discretion over setting strategies and using their powers. Expertise, control of detailed implementation and setting strategies are vital in the relationship between IRAs and elected politicians. Many PAs have emphasised the importance of elected politicians’ powers to set agents’ budgets and staffing and to overrule agents’ decisions. Table 3.5 shows the budgets and staffing levels of general competition and communications IRAs in 2000.

Table 3.5

Resources of IRAs 2000 Britain

France

Germany

Italy

General competition authorities — spending

68.6m euros (OFT and Competition Commission/MMC)

1.9m euros

17m euros

28.6m euros

General competition authorities — staff numbers

547 (OFT and Competition Commission/MMC)

110

262

137

Telecommunications regulators — spending’

22:4m euros

14.1m euros

63.8m euros

26.8m euros

Telecommunications

190

145

4200

178

regulators — staff

numbers’ Note: * Germany — RegTP covers both postal and telecommunications services; Italy -AGCOM covers telecommunications and broadcasting.

The figures indicate that IRAs are relatively small in terms of numbers and spending. Nevertheless, although budgets and staffing may be important, the activities of IRAs, especially rule-making, may not require many staff. Indeed, specialist expertise may be equally important. Whilst IRAs may be disadvantaged vis-a-vis large, well-resourced powerful firms and hence prone to capture or reliance by such firms, our interest lies in the power of IRAs relative to elected politicians. A comparison of the two in terms of specialised expertise offers a more complex picture than that of figures for total staffing and budgets. Many heads of IRAs have existing high levels of knowledge of their domain — and/or

58

Regulation through Agencies in the

EU

relevant general skills, such as law, economics or accountancy.'* In contrast,

ministers are rarely regulation or domain specialists. Finally, the average length of tenure of regulators is high — well above that of ministers or even governments: longer tenure can not only offer greater expertise, but also allows IRA members to ‘wait out’ a minister. Table 3.6 looks at the tenure of senior members of general competition authorities to allow cross-national comparability, since such authorities have existed throughout the 1990s in all four countries. It takes those senior members who finished their term or whose appointment was renewed. Table 3.6

Average Tenure of Senior Members of General Competition Authorities 1990-2001

Average tenure

Britain

France

Germany

Italy

6.4 years

(5)

7.5 years

7.7 years

5.3 years

(17)

(3)

(7)

Notes: a. Average tenure: only those members who left during period 1990-2000, except if appointment renewed; where term of office began before 1990, time is included; excludes those deceased in office; for Germany, only Presidents of the Bundeskartelhampt are included. b. For France, all members of the Conseil de la Concurrence included; if only the President is included, average tenure would be 5.5 years. For Italy, all members of AGCM included; for Britain, DGFT (Director General of Fair Trading) and Chairman of Competition Commission/MMC (Monopolies and Mergers Commission).

Naturally IRA members and elected politicians rely heavily on their staff. Here both the number and composition of staff may be extremely important — notably having domain specialists and also lawyers and economists. Legislatures are generally very underresourced in terms of specialists — for example, the 2003 House of Lords Constitution Committee which conducted a review of independent regulators had one specialist advisor. Even governments frequently have fewer specialised staff than IRAs. Thus for instance, the German government abolished its Posts and Telecommunications ministry and transferred all the staff to the regulator, the RegTP. In Britain, most senior officials at the Department of Trade and Industry are policy generalists, not regulation specialists or economists or lawyers. Staffing and budgetary constraints have not prevented IRAs from deciding their own strategies. On the contrary, many IRAs have developed innovative approaches, at times contrary to sectoral or national traditions. Thus for instance, the Italian general competition authority has led repeated attacks on

Independent Regulatory Agencies and Elected Politicians in Europe

59

long-standing monopolies and oligopolies, including state ones in fields such as telecommunications and broadcasting.'° In Britain, Oftel made promotion of competition its main objective after 1984,'’ although this was only one of its secondary objectives under the 1984 Telecommunications Act. In some instances, IRAs were following directions already set by elected politicians — for instance, the ART in France has largely followed French government policy of balancing competition with a degree of protection for France Télécom. On some rare occasions, IRAs have pursued strategies that conflicted with the objectives of governments — for instance, the pro-liberalisation strategies of the Italian competition and energy regulators have been in sharp contrast to the policies of the Berlusconi government. However, in many cases, IRAs have acted in cooperation with governments or have accelerated and developed government policies. Thus for instance, the pro-competition stances of the telecommunications and energy IRAs in Britain accelerated and developed previously rather timid government policies, playing a significant role in the liberalisation of the entire telecommunications, gas and electricity markets. Another significant example concerns 3G mobile licensing, in which telecommunications IRAs in all four countries considerably influenced government decisions. If IRAs take decisions that elected politicians dislike, the latter sometimes have formal powers to overturn those decisions. Although there are few publicly-available data, the figures are very striking. Thus for instance, in Germany, under the 1973 amendment

of the competition law, the Federal

Economics Minister can overturn a refusal by the Bundeskartellampt to allow a merger.'* However, only six decisions have been overturned between 1973 and 2000, with the last dating back to 1989.'° In Britain, the Secretary of State

for Industry has the power to accept or reject the Office of Fair Trading’s ‘recommendations’ as to whether a merger should be referred to the Competition Commission (formerly the Monopolies and Mergers Commission); he/she also has the power to reject the Competition Commission’s advice.” Use of the two powers is extremely rare: the OFT’s recommendations were rejected in only 14 cases between 1990 and 1997, and only four cases have been found for the MMC’’s report (source: MMC annual reports). Informal Relationships and Wider Networks

Whilst PA theory focuses on the formal institutional structure governing IRAs, qualitative studies suggest that informal relationships and wider networks are equally if not more important for the relationship between IRAs and elected politicians.”! Although IRAs are legally and organisationally separated from governments, the two generally enjoy close informal relationships. At the senior levels, IRA

60

Regulation through Agencies in the

EU

members are drawn from national elites — for example, the grands corps and ministerial cabinets in France and the professori in Italy (48% of the sample in France and for Italy, 62% of the sample were professori — 16 of 26). Most have had lengthy experience of public life, but outside elected office. Thus for instance, they may have held senior posts in or close to government or served on public committees and commissions. Thus for instance, all three heads of German

competition authority since 1976 had been heads of department in government ministries, whilst the three heads of the Italian competition agency had been President of the Constitutional court (Saja), senior politician and academic (Amato) and professor and Advocate General to the ECJ (Tesauro). At lower levels, many IRA staff have been drawn from government departments Even in well-established IRAs, ‘second tier’ personnel are often civil servants — e.g. Anna Walker, deputy director general at Oftel in the 1990s, was a vital link with the government,” and indeed returned to a central executive department (energy) from her Oftel post. The result of these recruitment patterns are a web of personal ties linking IRAs with policy makers, both at senior levels between the appointed heads of IRAs and elected officials, and at more junior levels with government civil servants. These ties often result in frequent consultation and two-way communication between IRAs and governments, so that each influences the other.”? IRAs and elected politicians may form alliances and cooperate in order to deal with other actors — notably regulatees, other countries and international organisations. Sometimes they have done so in surprising directions — for example, governments have supported the extension of competition in utilities even though this eroded the domestic base of national champion firms, including state-owned ones. At times, IRAs can engage in battles within governments; thus for instance, the French telecommunications regulator, the ART, allied itself

with the Industry ministry against the Finance Ministry in debates about the system to allocate 3G licences and the amounts of money to be raised (the Finance ministry sought maximum revenues and an auctions system as opposed to a lower-revenue beauty contest). IRAs have also built up their own networks at both national and supranational levels. Thus for example, the utility regulators in Britain meet regularly and have agreed common policies on matters such as capital investment and risk.74 There are also Europe-wide ‘fora’ for telecommunications, electricity and gas regulators, with annual meetings, supplemented by continuing informal contacts. IRAs also have close links to the Commission, sometimes formalised (e.g. the European Regulators Group for Electronic Communications, previously the High Level Regulators’ Group). In EC sectoral committees of officials, countries are often represented by their IRAs rather than ministries. These networks allow not only policy learning across countries, but also influence on EC legislation at an early stage. Given the growth of the role of the EC in

Independent Regulatory Agencies and Elected Politicians in Europe

61

regulated markets, such access is important for IRAs. It offers a counterbalance to national governments, who play a crucial role in the EC’s legislative process. The EC level of decision making also allows a complex game to be played. On some issues, IRAs have allied themselves with their domestic governments, notably in seeking to ‘export’ national models of regulation to the EU level. Thus for example, Oftel and other British IRAs have cooperated with the British government and indeed incumbent suppliers such as BT to press for EC liberalisation. The EC level of decision making also allows a complex game to be played. On some issues, IRAs have allied themselves with their domestic governments, notably in seeking to ‘export’ national models of regulation to the EU level; thus for example, the Italian competition authority linked with the Commission’s Competition Directorate General (DGIV) in the 1990s to press the Italian government to liberalise telecommunications and to end discrimination against new entrants such as Olivetti.” Finally, IRAs can develop their own resources and networks with regulatees and the public. IRAs have created new procedures, such as producing consultation papers and draft decisions and inviting comments. They have published a much greater volume of information than governments did, even on sensitive commercial matters such as costs, profits and market shares. They have also made available the models on which their decisions are based,

including economic and pricing models. Such developments have occurred not only in Britain,*° but also in traditionally more closed countries such as Italy where IRAs have adopted procedures based on principles of openness and contestability.*”? The information produced by IRAs has been widely disseminated. Press coverage of IRAs has often been extensive, especially in the utilities and some regulators have become major public figures (for instance, in Britain, Don Cruickshank at Oftel or Clare Spottiswood at Ofgas, or in Italy, the head of the Communications Authority, Enzo Chelli). At times,

IRAs may form alliances with powerful regulatees or users, creating ‘consensus’ in favour policies, pre-empting or opposing certain parts of the government. One clear example came in French debates over 3G mobile licensing, in which the ART lobbied alongside suppliers for a beauty contest and limits on the cost of licences against the Finance ministry.

CONCLUSION The formal institutional framework governing the relationship between IRAs and elected politicians is broad and allows much discretion to both sets of actors. Moreover, analysis of formal structures misses important informal resources and relationships. To understand post-delegation behaviour, we have analysed

62

Regulation through Agencies in the

EU

how formal powers have been used and then the informal relationships and networks of IRAs. The data suggest that elected politicians have not used their powers of appointment to select party loyalists (except in Italy). Nor have they forced out IRA members before the end of their terms, even when governments have changed. The result has been IRAs with longer-lasting members than ministers or most governments. Equally, when available, elected politicians have not used their powers to overturn IRA decisions. At the same time, informal

relationships between elected politicians and IRAs have developed. Moreover, IRAs have built up their own resources in terms of expertise, informal linkages and wider networks and legitimacy. The evidence points to the fact that whereas previously regulation in Europe was largely a matter between governments and suppliers, today, contrary to expectations at the time of their creation, IRAs in key economic

domains in Britain, France, Germany

and

Italy constitute a distinct group of actors in regulation. Explaining why elected politicians have not used their formal powers of control or assessing the relationship between formal and informal controls and relationships would require a large-scale investigation. However, two broad interpretations can be put forward for further investigation. The first is that elected politicians have found means other than the formal controls examined to limit IRA autonomy and ensure that agency losses were small. These other means might be institutional design or ex ante controls, administrative procedures and controls exercised by third parties such as interest groups and the courts”* or non-statutory controls, notably the informal relationships examined. An alternative view is that elected politicians have chosen to leave IRAs considerable independence and hence not utilise their formal controls. This might be because for IRAs to perform their functions for elected politicians such as increasing credible commitment, they must be independent in practice.” It might also be that the costs for elected politicians of utilising controls are very high — for instance, in terms of the informational resources to effectively control IRAs.

Indeed, elected politicians may not want to control substantive policy, preferring to avoid the blame for difficult decisions and concentrate their energies in more rewarding domains.

NOTES ' Although there are studies of the most long-established IRAS, notably general competition authorities (Gerber 1998); Wilks (1999), In the Public Interest, Manchester: Manchester University Press. See Thatcher, M. and Stone Sweet, A. (2002) ‘Theory and Practice of Delegation to Non-

Majoritarian Institutions’, West European Politics, 25 (1), 1-22; for recent major PA work on agencies, see Epstein and O’Halloran (1999), Delegating Powers. A Transaction Cost Politics Approach to

Independent Regulatory Agencies and Elected Politicians in Europe

63

Policy Making under Separate Powers, Cambridge: Cambridge University Press and Huber and Shipan (2002), Deliberate Discretion? (Cambridge: Cambridge University Press). 3 See Kiewet, R.D. and McCubbins, M.D. (1991), The Logic of Delegation, Chicago: University of Chicago Press; McCubbins and Schwartz (1984), ‘Congressional Oversight Overlooked: Police Patrols versus Fire Alarms’, American Journal of Political Science, 28, 1, 165-179. * Pollack, M. (2002), ‘Learning from the Americanists (Again): Theory and Method in the Study of Delegation’, West European Politics, 25, 1. > See Weingast, B. and Moran, M. (1983), ‘Bureaucratic Discretion or Congressional Control:

Regulatory Policymaking by the FTC’, Journal of Political Economy, 91, 765-800. ® Moe, T. (1985), ‘Control and Feedback in Economic Regulation: The case of the NLRB’, American Political Science Review, 79 (4), 1101. 7 See Moe, T. (1985) supra note 6, cf. Moe, T. (1982), ‘Regulatory Performance and Presidential Administration’, American Political Science Review, 26 (2), 197-224.

8 See Hayward, J.E.S. (1986), The State and the Market, Brighton: Harvester Wheatsheaf; Hayward, J.E.S. (1995), Industrial Enterprise and European Integration, Oxford: Oxford University Press; Schmidt, V.A. (1996), From State to Market? The Transformation of French Business and Government, Cambridge: Cambridge University Press; Machin, H. and Wright, V. (1985), Economic Policy and Policy-Making under the Mitterand Presidency 1981-1984, London: Pinter; Cawson, A., Holmes, P., Webber, D., Morgan, K. and Steven, A. (1990), Hostile Brothers, Oxford: Clarendon Press; Dyson K. and Wilks S. (eds) (1983), Industrial Crisis, Oxford: Martin Robertson; Grant, W.

(1989), Government and Industry. A Comparative Analysis of the US, Canada and the UK,Aldershot: Edward Elgar; Muller, P. (1989), Airbus, l’ambition européenne: logique d état, logique de marché, Paris: |‘Harmattan. 9 See Thatcher, M. (2002), ‘Delegation to Independent Regulatory Agencies: Pressures, Functions and Contextual Mediation’, West European Politics, 25 (1), 125-147; Gilardi, F. (2002), ‘Policy Credibility and Delegation to Independent Regulatory Agencies: A Comparative Empirical Analysis’, Journal ofEuropean Public Policy, 9 (5); Levy, B. and Spiller, P. (1996), Regulation, Institutions

and Commitment, Cambridge: Cambridge University Press; Doern, B. and Wilks, S. (eds) (1996),

Comparative Competition Policy: National Institutions in a Global Market, Oxford: Oxford University Press. !0 See Cassese, S. and Franchini, C. (eds) (1996); Perez, P. (1996), ‘Autorita Indipendenti e Tutela dei Diritti’, Revista trimetrale di diritto publico, 1, 115-147; Gobbo, F. and Zanetti, G. (eds) (2000), ‘Istitutiozioni e mercato: il ruolo delle Autorita nell’economia italiana, special issue of L industria, XX1I/4 (2000); Martinelli, E. (2000), ‘Autorita indipendenti e politica’, Amministrare XXX (1-2), 127-146; Giraudi, G. and Righettini, M. (2001), Le autorita ammistrative indipendenti, Rome: Laterza. For France, see Conseil d’Etat 2001, Teitgen-Colly, C. (1990), “Les instances de régulation et la constitution’, Revue de Droit Public, 36, (1), 53-262; Colliard, C.-A. and Timsit, G. (eds) (1988), Les autorités administratives indépendantes, Paris: Presses Universitaires de France; Guédon, M.-J. (1991), Les autorités administratives indépendantes, Paris: Librairie Générale de Droit et de

Jurisprudence. !! Thus for instance, the Director General of Oftel had two primary duties under the 1984 Telecommunications Act: to ensure that ‘all reasonable demands for telecommunications were satisfied throughout the UK, except if not reasonable practicable’; to ensure that suppliers could finance the provision of such services. 12 See Majone, G. (1996), Regulating Europe, London: Routledge, 38; Wood, B. and Waterman, R. (1991), ‘The Dynamics of Political Control of the Bureaucracy’, American Political Science

Review, 85 (3), 801-828.

13 Sir Gordon Borrie, DGFT 1976-1992, stood unsuccessfully for Parliament in the 1950s but by the 1970s was not publicly linked to a political party. 14 The heads of the OFT were Sir Gordan Borrie (Barrrister and legal academic) 1976-1992, Sir

Bryan Carsberg 1992-1994 (accountant), John Bridgeman (1995-2000) (senior business executive) and John Vickers 2000-present (academic economist); for the MMC, the most recent have been Sir Sidney Lipworth 1988-1993 (senior business executive), Graeme Odgers 1993-1997 (senior business executive) and Derek Morris 1997-present (academic economist); the heads of Oftel have

been: Sir Bryan Carsberg 1984-1992 (accountant), Don Cruickshank 1988-1993 (accountant and business executive) and David Edmonds (civil service and senior business executive).

64

Regulation through Agencies in the

EU

1S Thus for example, for competition IRAS: in Italy, all three heads (Saja 1990-1994, Amato 19941997 and Tesauro 1998-present) were lawyers; for Germany two Presidents of the Bundeskartellamt were lawyers (Kartte 1976-1992 and Wolf 1992-1999) and one an economist (Boge 2000-present), for Britain, see above; in France, two of the three presidents of the Conseil de la Concurrence were senior members of the Conseil d’Etat, with expertise in administrative law (Laurent 1987-1993 and

Hagelsteen 1998-present). 16 See Perez, R. (2002), Telecomunicazioni e concorrenza, Milan: Giuffre. 17 See Carsberg, B. (1989), ‘Injecting competition into telecommunications’, in Veljanovski, C. (ed.), Privatisation and Competition, London: IEA. See also Carsberg, B. (1991), ‘Office of Telecommunications: Competition and the Duopoly Review’, in Veljanovski, C. (ed.), Regulators and the Market, London: IEA.

18 See Baake, P. and Perschau, O. (1996), ‘The law and politics of competition policy in Germany’, in G. Majone (ed.) (1996), Regulating Europe, London: Routledge. 19 The Daimler Benz-MBB decision, which although much discussed, represents the exception rather than the rule. 20 See Wilks, S. (1999), In the Public Interest, Manchester: Manchester University Press.

21 See Hancher, L. and Moran, M. (1989), Capitalism, Culture and Regulation, Oxford: Oxford University Press; Scott, C. (2001), ‘Analysing Regulatory Space : Fragmented Resources and Institutional Design’, Public Law (Summer), 329-353. 22 Hall, C., Scott, C. and Hood, C. (2000), Telecommunications Regulation, London: Routledge. 23 See Hall, C., Scott, C. and Hood, C., supra note 22. 24 See Moran, M. (2003), The British Regulatory State, Oxford: Oxford University Press.

25 See Cassese, S. (2001). “L’Arena Pubblica: Nuovi Paradigmi per lo Stato’, Rivista trimestrale di diritto pubblico, 3, 601-650; see Perez, R. (2002), Telecomunicazioni e concorrenza, Milan: Giuffre. 26 See Prosser, T. (1997), Law and the Regulators, Oxford Clarendon Press.

27 See Cananea, G. della (2002), ‘The Regulation of Public Services in Italy’, International Review of Administrative Sciences, 68, 73-93. 28 See McCubbins, M.D. and Schwartz, T. (1984), “Congressional Oversight Overlooked: Police

Patrols versus Fire Alarms’, American Journal of Political Science, 28 (1), 165-179. See also McCubbins, Mathew D., Noll, Roger G. and Weingast, Barry R. (1987), ‘Administrative Procedures as Instruments of Political Control’, Journal of Law, Economics and Organization, 3 (2), p. 243. 29 See Majone, G. (2001), ‘Two Logics of Delegation: Agency and Fiduciary Relations in EU Governance’, European Union Politics, 2 (1), 103-122.

REFERENCES Baake, P. and Perschau O. (1996), ‘The law and politics of competition policy in Germany’, in G. Majone (ed.), Regulating Europe, London: Routledge. Bollhoff, D. (2002), Developments in Regulatory Regimes — An Anglo-German Comparison on Telecommunications, Energy and Rail, Pre-prints Max Planck Institute: Bonn.

Cananea, G. della (2002), ‘The regulation of public services in Italy’, International Review of Administrative Sciences, 68, 73-93.

Carsberg, B. (1989), ‘Injecting competition into telecommunications’, in Veljanovski, C. (ed.), Privatisation and Competition, London: IEA. Carsberg, B. (1991), ‘Office of telecommunications: competition and the duopoly review’, in Veljanovski, C. (ed.), Regulators and the Market, London: IEA.

Cassese, S. (2001), ‘L’Arena Pubblica: Nuovi Paradigmi per lo Stato’, Rivista trimestrale di diritto pubblico, 3, 601-650.

Cassese, S. and Franchini, C. (eds) (1996), J garanti delle regole, Bologna: Il Mulino. Cawson, A., Holmes, P., Webber, D., Morgan, K. and Stevens, A. (1990), Hostile Brothers, Oxford:

Clarendon Press. Coen, D., Héritier, A. and Béllhoff, D. (2002), Regulating the Utilities: Business and Regulator Perspectives in the UK and Germany, Berlin: Anglo-German Foundation.

Independent Regulatory Agencies and Elected Politicians in Europe

65

Colliard, C-A. and Timsit, G. (eds) (1988), Les autorités administratives indépendantes, Paris:

Presses Universitaires de France. Conseil d’Etat, Rapport Public 2001, Paris: La documentation Francaise 2001. Doern, B. and Wilks, S. (eds) (1996), Comparative Competition Policy: National Institutions in a Global Market, Oxford: Oxford University Press. Dyson, K. and Wilks, S. (eds), Jndustrial Crisis, Oxford: Martin Robertson.

Eberlein, B. (2002), ‘Institutional Change and Continuity in German Infrastructure Management: the Case of Electricity Reform’, German Politics, 9 (3), 81-104. Epstein, D. and O’Halloran, S. (1999), Delegating Powers: A Transaction Cost Politics Approach to Policy Making under Separate Powers, Cambridge: Cambridge University Press. Gerber, D.J. (1998), Law and Competition in Twentieth Century Europe: Protecting Prometheus, Oxford: Clarendon Press. Gerber, B.J. and Teske, P. (2000), ‘Regulatory Policymaking in the American States: A Review of Theories and Evidence’, Political Research Quarterly, 53 (4), 849-886. Gilardi, F. (2002), ‘Policy Credibility and Delegation to Independent Regulatory Agencies: A Comparative Empirical Analysis’, Journal of European Public Policy, 9 (5), 873. Giraudi, G. and Righettini, M. (2001), Le autorita amministrative independenti, Rome: Laterza. Gobbo, F. and Zanetti, G. (eds) (2000), ‘Istitutiozioni e mercato: il ruolo delle Autorita nell’ economia italiana’, special issue of L '/ndustria, XX1/4 (2000). Grant, W. (1989), Government and Industry: A Comparative Analaysis of the US, Canada and the UK, Aldershot: Edward Elgar. Guédon, M.-J. (1991), Les autorités administratives indépendentes, Paris: Librairie Générale de Droit et de Jurisprudence. Hall, C., Scott, C. and Hood, C. (2000), Telecommunications Regulation, London: Routledge. Hancher, L. and Moran, M. (1989), Capitalism, Culture and Regulation, Oxford: Oxford University Press. Hayward, J.E.S. (1986), The State and the Market, Brighton: Harvester Wheatsheaf. Hayward, J.E.S. (ed.) (1995), Industrial Enterprise and European Integration, Oxford: Oxford

University Press. Huber, J. and Shipan, C. (2002), Deliberate Discretion?, Cambridge: Cambridge University Press. Kiewiet, R.D. and McCubbins, M.D. (1991), The Logic of Delegation, Chicago: University of

Chicago Press. Levy, B. and Spiller, P. (1996), Regulation, Institutions and Commitment, Cambridge: Cambridge

University Press. Lodge, M. (2002), On Different Tracks: Institutions and Railway Regulation in Britain and Germany, London: Praeger. McCubbins, M.D., Noll, R.G. and Weingast, B.R. (1987), ‘Administrative Procedures as Instruments

of Political Control’, Journal of Law, Economics and Organization, 3 (2), 243. McCubbins, M.D. and Schwartz, T. (1984), ‘Congressional Oversight Overlooked: Police Patrols versus Fire Alarms’, American Journal of Political Science, 28 (1), 165-179.

Machin, H. and Wright, V. (eds) (1985), Economic Policy and Policy-Making under the Mitterrand Presidency 1981-1984, London: Pinter. Majone, G. (ed.) (1996), Regulating Europe, London: Routledge. Majone, G. (2001), ‘Two Logics of Delegation: Agency and Fiduciary Relations in EU Governance’, European Union Politics, 2 (1), 103-122. Martinelli, E. (2000), ‘Autorita independenti e politica’, Amministrare XXX(1-2), 127-146. Moe, T. (1982), ‘Regulatory Performance and Presidential Administration’, American Journal of Political Science, 26 (2), 197-224.

Moe, T. (1985), ‘Control and Feedback in Economic Regulation: The Case of the NLRB’, American Political Science Review, 79 (4), 1094-1116.

Moran, M. (2003), The British Regulatory State, Oxford: OUP. _ Muller, P. (1989), Airbus, l'Ambition Européenne: Logique d’Etat, Logique de Marché, Paris: 1Harmattan. Perez, R. (1996), ‘Autorita Indipendenti e Tutela dei Diritti’, Revista trimestrale di diritto pubblico, 1, 115-147. Perez, R. (2002), Telecomunicazioni e concorrenza, Milan: Giuffre.

66

Regulation through Agencies in the

EU

Pollack, M. (2002), ‘Learning from the Americanists (Again): Theory and Method in the Study of Delegation’, West European Politics, 25 (1), 200.

Prosser, T. (1997), Law and the Regulators, Oxford: Clarendon Press. Schmidt, V.A. (1996), From State to Market? The Transformation of French Business and Government, Cambridge: Cambridge University Press. Scott, C. (2001), ‘Analysing Regulatory Space: Fragmented Resources and Institutional Design’, Public Law (Summer), 329-353. Teitgen-Colly, C. (1990), ‘Les instance de régulation et la constitution’, Revue de Droit Public, 36 (1), 53-262. Thatcher, M. (1999), The Politics of Telecommunications, Oxford: Oxford University Press.

Thatcher, M. (2002), ‘Delegation to Independent Regulatory Agencies: Pressures, Functions and Contextual Mediation’, West European Politics, 25 (1), 125-147. Thatcher, M. and Stone Sweet, A. (2002), ‘Theory and Practice of Delegation to Non-Majoritarian Institutions’, West European Politics, 25 (1), 1-22. Weingast, B. and Moran, M. (1983), ‘Bureaucratic Discretion or Congressional Control: Regulatory Policymaking by the FTC’, Journal of Political Economy, 91, 765-800. Wilks, S. (1999), In the Public Interest, Manchester: Manchester University Press. Wood, B. and Waterman, R. (1991), ‘The Dynamics of Political Control of the Bureaucracy’, American Political Science Review, 85 (3), 801-828.

4.

Agencies for European Regulatory Governance: A Regimes Approach Colin Scott

INTRODUCTION EU institutions and norms form a key part of many aspects of contemporary regulation in the member states. Equally, a large part of the EU’s activity is regulatory in character. The interdependence between member state and EU levels makes it difficult to apply models of regulatory governance developed to describe policy processes at national level. Nevertheless recent debate about EU regulatory processes, both official and scholarly, has included discussion of ideas about developing agencies closer in character to the American independent agency model. The debate about agencies, prominent in the mid1990s,' has been given a new lease of life first by the publication of the Commission’s White Paper on European Governance, second by the creation of a third wave of new agencies and thirdly by the report to the President of the European Commission of the independent high-level study group chaired by André Sapir. The Governance White Paper was part of the Commission’s response to the crisis which brought down the Santer Commission in 2000. It represented a taking stock of the institutions and mechanisms of EU governance.’ The Sapir report was concerned with giving guidance on how to make economic governance within the EU more effective and efficient.’ The primary objective of this chapter is to offer a mapping of the variety of forms of regulatory governance available at the level of the EU. The analysis is based on a regimes approach which starts from the observation that the different components of the regulatory system are widely dispersed among different organisations, at different levels, and of both governmental and nongovernmental character. I suggest that there are at least ten models of regulatory governance currently in play at the level of the EU. Amongst these ten regime types there are a number of different forms of agency. I suggest that 67

68

Regulation through Agencies in the

EU

of the available models for EU regulation first, that the independent agency model is one of the least well developed, and second, that there are some pretty

significant obstacles in the way of the development of such a model. A central methodological assumption of the chapter is that institutional development is likely to be incremental, building on existing successful models, rather than

discontinuous with existing patterns: Equipped with a map it is then possible to provide some hint of the normative issues surrounding the development of independent agencies in the concluding section.

A REGIMES APPROACH TO REGULATION Regulation is an attractive policy instrument for the EU because it is relatively inexpensive (when compared with expenditure based instruments such as subsidy or direct provision of services), and because its basis in legal rules enables a degree of separation between rule making and implementation processes, an important characteristic of the way that EU governance has developed. Within the EU, as with many governance systems, there is no single organisational focus for the observer seeking to understand how regulation works in any particular domain. This observation makes it appropriate to think in terms of regulatory regimes. A regulatory regime may be conceived of as ‘a historically specific configuration of policies and institutions which structures the relationship between social interests, the state and economic actors in multiple sectors of the economy’.* More specifically a complete regime consists of each of the elements necessary to comprise a complete system of control over some social or economic domain. These components include goals, standards, rules or norms which the system is seeking to achieve, a mechanism for monitoring or feeding back information about the state of the system, together with a mechanism for bringing back into alignment behaviour which deviates from the goals, standards, rules or norms.* It is common

to find

overlapping capacities for standard setting or rule making, monitoring and enforcement diffused among different levels of government, sub-national, national and supranational and between governmental, non-governmental and private organisations. Within a classical regulatory model the elements of a regulatory regime comprise legal rules, agency monitoring, and the application of formal sanctions for breach of the rules. The widely discussed American model of the independent regulatory agency often combines these components within a single agency. Agencies such as the Federal Trade Commission (established 1914) and the Federal Communications Commission (established 1934) provide the core cases from which theories of independent agencies are developed.° The establishment of such a complete regime within a single

Agencies for European Regulatory Governance

69

agency is much less common outside the United States. It is more usual to find that much of the regime is focused on ministerial government departments, or that there is wide diffusion of the components, perhaps with rule making reserved to legislative bodies, monitoring assigned to an agency, and the application of formal sanctions reserved to a court, perhaps at the instigation of an agency or of third parties.

INSTITUTIONAL MODELS FOR EU REGULATION The development of regulatory governance at the level of the EU is characterised by considerable variety. Application of a regimes analysis suggest there are at least ten different models currently in use. The particular forms taken are shaped in part by constraints in the constitutional structures of the EU itself. Thus any EU action must first be underpinned by demonstration that there is competence to act. The EU institutions are creatures of Treaty and may only do that which they are authorized to do. Thus there are some fields which, however attractive supranational regulation might appear, lack a legitimate basis for action. Where competence is shared between EU and member state levels a second constraint comes in the form of the subsidiarity doctrine with its central norm that the Community should act ‘only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States and can therefore, by reason of the scale or effects of the proposed action, be better achieved by the Community’.’ A third constraint on the possible forms of regulation is provided by doctrines on the ‘inter-institutional balance’ within the Community. Article 7(1) assigns the tasks set out in the Treaties to the five community institutions. Under the Meroni doctrine it has been held that the EC legislative institutions may not empower others to adopt legislation and that the Commission may not delegate a decision-making power conferred on it by the Treaty or by legislation except for the limited purposes of implementation.* A strict interpretation of the Meroni doctrine appears to limit the scope for delegation to agencies. Commission as Regulator The EU regulatory regime which appears closest in character to that of the classical regulatory model is that over competition policy. The main rules creating prohibitions on abuse of dominant position and restrictive agreements are set down in the EC Treaty.’ These provisions of the Treaty are unusual because they create binding obligations on undertakings within the members states. The Treaty itself has little to say about monitoring and enforcement of the rules, but as early as 1962 the Council legislated for the Commission to

70

Regulation through Agencies in the

EU

have extensive powers of monitoring and enforcement, including the capacity to fine undertakings found to be in breach of the rules.'® The EU competition regime is most like a classical regulatory system in those aspects where the Commission combines rule making powers with its monitoring and enforcement functions. Though the Commission has no control over the Treaty rules which make up the core of the normative structure, it has at least three different ways in which it can make norms which elaborate on that structure. First it has power to issue its own directives for the purposes of clarifying the nature of the Treaty competition obligations as they apply to the state."' In practice directives made under this power were central to the liberalisation of the European telecommunications sector,'* but have been not much used otherwise because of political sensitivities.’ Second, the power to issue Block Exemption Regulations under Article 81(3) of the EC Treaty was delegated to the Commission by the Council.'* Third, the Commission makes extensive use of soft law instruments providing guidance of various sorts as to the application of the competition rules.'° In other respects the competition regime diverges from the classical model, in particular because the Commission is a multi-purpose executive authority, not just a regulator. While the Commission has developed considerable expertise in respect of competition matters, final decision making is for the non-expert College of Commissioners and is liable to be diluted by other strategic considerations which would be alien to an independent regulatory agency. Reflecting the substantial move towards adoption of national competition laws similar to those of the EU, reforms to the procedural rules which took effect in May 2004 promote shared competence between national competition authorities (NCAs) and the Commission.'® The new Regulation abolishes requirements of prior clearance under Article 81(3), thus reducing the Commission’s rule making power. The net effects of the new Regulation are to move the EU competition regime further from a classical regulatory model, making it more like the kind of diffused, multi-level regime found in other EU policy domains. Metaregulation and the community method The ‘Community method’ of policy making and implementation, as the Commission labelled it in the 2001 White Paper on European Governance,!”

involves the Commission in initiating legislative policy which is then adopted by the legislative institutions of the Council and the Parliament. Responsibility for implementation, which includes transposition of the rules into national law, and assignment of monitoring and enforcement responsibilities falls to the member state governments, which may then assign some of these functions to other state agencies. The Commission retains a key role in monitoring and enforcing implementation processes. Thus, as far as oversight of the member

Agencies for European Regulatory Governance

vA

states is concerned, the role of the Commission is that of ‘metaregulator’. Metaregulators operate through seeking to steer the regulatory processes of others, influencing behaviour in the regime indirectly.'* In the case of the Commission emphasis has traditionally been given to its role in monitoring the conduct of member state governments in transposing and implementing directives. The Commission has a form of enforcement pyramid which it can deploy with states which do not comply with the legislative requirements, first opening a case with informal discussion and, where such negotiation is not successful seeking condemnation and ultimately financial penalties in the European Court of Justice.'° It is clear that the Commission also uses other metaregulatory strategies, for example convening regular meetings of national regulators so as to educate them in the requirements of the EU regime, or issuing guidance addressed directly to actors within the regulatory domain.” A good example of the Commission as metaregulator is provided by the communications sector. An extensive revised legislative framework was put in place by the EU legislative institutions in 2002.7! The legal framework makes provision for member state governments to put in place national regulatory authorities (NRAs) which carry the main burden of monitoring the conduct of the key actors in the sector for compliance with the rules and enforcement to realign behaviour. The Commission has been very active in encouraging NRAs to use their powers in the cause of creating a single market in communications, and speaks to firms directly through soft law instruments.” EU agencies and the community method In some policy domains the regulatory function at the EU level is not restricted to metaregulation, and involves direct regulatory activity concerned with such matters as the elaboration of the requirements of legal rules, issuing authorisations, and the taking of emergency regulatory action affecting health and safety. In some domains the Commission has initiated legislation creating independent agencies. These agencies are not full regulators, but generally operate to assist the Commission in fulfilment of its functions. The European Commission’s count lists between 15 and 20 such agencies, established in a number of ‘waves’ in the 1970s, the 1990s and most recently since 2000.” Further agencies are proposed.” For the European Commission an agency has a number of characteristics: it has its own legal personality and is distinct from the institutions established in the Treaties; it is established by legislation for the performance of technical, scientific or managerial tasks.*> The European Commission persists in describing agencies as a mechanism

of decentralised governance,” and this claim has

been adopted by most commentators.’ It must be clear that EU agencies only represent decentralization in the very narrow sense relating to their locations

72

Regulation through Agencies in the

EU

and relationship with the Commission itself. In all other respects, and in particular vis-a-vis the member states, EU agencies are instruments of centralization.* Not all the agencies are involved in what we might think of as regulatory domains, and those that are possess few regulatory powers. Thus, the European Medicines Evaluation Agency (EMEA) was established to advise the Commission on the granting of authorisations for medicinal products derived from biotechnology and on the Commission’s role in resolving disputes concerning the application of the principle of mutual recognition in respect of other medicinal products authorised by member state authorities.” The powers to grant authorisations and to determine disputes remain with the Commission. The European Food Safety Authority (EFSA) is largely advisory to the Commission on matters relating to food standards and food emergencies. The Regulation establishing EFSA creates a set of general principles for EU food law and provides the normative basis for a European network for the regulation of food safety involving member

state authorities,

the EFSA

and the

Commission.*° In effect EFSA’s creation partially displaces the role of comitology in policy processes which continue to be dominated by the Commission. The Community Plant Varieties Office and the Office for Harmonization in the Internal Market (OHIM) are unusual in that they do possess autonomous power to grant registrations of intellectual property rights with legal effect. Thus their actions do bind third parties. However, the two agencies barely comply with the definition of regulators since they do not engage in monitoring. Rather they are part of a wider and more complex regulatory regime in which norms are created by the EU legislative institutions and enforcement falls to holders of intellectual property rights (see the discussion of private enforcement on pages 75-6). It is striking that the development of EU agencies merits no discussion in the preparatory work for the White Paper on improving regulatory processes. This may be indicative of the ‘agencies strategy’ being somewhat detached from the mainstream of Commission thinking on better regulation, or perhaps just being irrelevant to it. If the latter is the case it is because it is relatively clear that as presently constituted the EU agencies have little regulatory power and are destined to be minor players, at best, within key regulatory regimes. It is difficult to see how the Commission could shuffle off its Treaty responsibilities sufficiently to create independent and potent regulatory agencies at the EU level without Treaty amendment. Commission agencies An alternative to the legally separate EU agency is the establishment of somewhat distinct units within the Commission itself. The creation of an independent unit within the Commission has been achieved in the case of

Agencies for European Regulatory Governance

73

Eurostat, which is described as a service of the European Commission, but

under the terms of its regulation (the ‘statistical law’) is an independent European statistical authority.*' The EU Fraud Office (OLAF) is said to be a department of the Commission,” but independent in its decision making both from other parts of the Commission and from anyone else. While Eurostat may be an independent unit within the Commission,

it does not, of course, exercise

regulatory functions. It has no powers to make or enforce rules against anyone. OLAF is a more interesting case. It does have law enforcement powers against anyone who spend EU funds. Thus it provides the monitoring and enforcement element of the wider regime of fraud control. An interesting case is provided by the Food and Veterinary Office, which is a unit of DG Consumer Protection and Health. The FVO is charged with monitoring the compliance by member state and third nation governments with food safety and animal health legislation. In pursuit of these functions the FVO is directly involved in inspection of private facilities in the member states, not as a direct regulator, but rather as a monitoring check on national regulatory systems. It has no enforcement capacity of its own, but its reports may be used as the basis for Commission decision making on enforcement action against member state governments or import controls affecting non-EU states on health and safety grounds. The plan to turn the FVO into an agency was rejected because the Commission felt its independence as a regulator could be better guaranteed within the hierarchy of the Commission than in an independent agency where it would be subject to greater oversight from member state governments.** In other words, independent agencies do not provide an appropriate model for regulation of member state governments.

Independent (Treaty) agencies The EU organisations which correspond most closely with the independent agency model are those established by the Treaties which are, nevertheless, not ‘institutions’. The leading examples are the European Central Bank and the European Ombudsman. The ECB’s mission is to control the money supply within the Eurozone and, linked to this, to maintain low inflation (below 2%). The strict independence given to the ECB in the Treaty was granted in order to insulate it from the macro-economic policy making of governments and of the EU institutions.** The decision to insulate the ECB in this way follows similar decisions made for national central banks in the EU by member state governments. The ECB demonstrates that it is possible to have an independent and powerful agency operating at EU level and to assign to it functions such as the making of regulations.** The independence of the ECB is, of course, open to criticisms, firstly that it is insufficiently transparent and accountable for its actions, and secondly that it is too insulated from vital macro-economic considerations such as levels of unemployment.

74

Regulation through Agencies in the

EU

The European Ombudsman was established under the Maastricht Treaty to investigate grievances against other EU institutions and is said to be independent, though under the general scrutiny of the European Parliament.”° The Ombudsman is chiefly a grievance handler, rather than a regulator, though his work may have regulatory effects to the extent that it sets standards for future conduct which are then monitored in further investigations. To this extent the Ombudsman may be conceived as a regulator of EU institutions. Transnational regulatory networks An alternative mechanism to the Community method for exploiting and coordinating regulatory capacities in the member states is to build EU regulation as a process of coordination through transnational regulatory networks.*’ Such networks bring together national regulatory authorities and use them as the engine for coordinated policy development. The European Securities Regulators Committee (RESCO), established in response to the recommendations of the Lamfalussy report in 2000, provides a key example of this mode of regulation.*? RESCO is advisory to the Commission on matters of securities policy, but also acts as an instrument of informal coordination among national regulators. Such transnational regulatory networks do not, of course, have to be the product of stimulation by the Commission or other supranational authorities. A network of competition regulators appears to have emerged more or less spontaneously as a means to coordinate international policy enforcement well beyond the EU,” though it has subsequently been given

statutory status.’ Similar observations might be made for telecommunications.*! The open method of coordination The open method of coordination (OMC) is the term coined at the Lisbon meeting of the Council of Ministers in March 2000 to encompass a range of mechanisms through which the Council could coordinate policy developments in different domains without recourse to traditional legislative mechanisms of the Community. An early analysis noted three distinctive elements to the mechanism: common assessment of the situation (through sharing of information); agreement on appropriate policy response; mutual adjustment by member state governments of their policy structures with peer pressure from other governments.” The leading case of deployment is in the field of Economic and Monetary Union where there is perhaps rather more effort on attempts to coordinate the fiscal policies of the members of the Eurozone than there is on the openness dimensions, implying as it does experimentation and learning from the experiences of others. In other policy fields there is greater openness, for example in respect of social policy. An interesting feature of the OMC is the extent to which it reduces the Commission’s responsibility and capacity for executive action.

Agencies for European Regulatory Governance

te)

Negative integration and mutual recognition An earlier and alternative method for addressing the tensions between diversity of national regulatory policies, on the one hand, and the need for harmonization on the other is found in processes of negative integration and the doctrine of mutual recognition. The main institutional focus for this form of regulatory regime is the European Court of Justice and on national legislative authorities. The ECJ held in the celebrated Cassis de Dijon case that the provisions of the EC Treaty directed towards completion of a single market in goods required mutual recognition of regulatory rules of one member state by another — permitting products capable of being lawfully sold in one member state to be similarly lawfully sold in another.*? The processes of ‘negative integration’ generated by the judgment effectively cut out the legislative institutions of the EC, though it does not preclude the adoption of positive harmonizing or approximating measures. Better regulation, coregulation and enforced self-regulation The eight models of regulatory regimes discussed hitherto focus very much on governmental capacities and organisations. An approach which focuses exclusively on such state and super-state activity neglects the observation that many non-governmental and private organisations have substantial resources relevant to the exercise of regulatory power.** Contemporary thinking about ‘better regulation’ recognises the benefits of conceiving of a wider range of actors as involved in regulatory regimes. Coregulation is generally thought of as a model of regulation under which an industry association engages in self-regulation “with some oversight or ratification by government’.*° In the proposals on ‘Better Regulation’ the Governance White Paper highlights the use which has already been made of processes of coregulation — unconventionally defined as the setting down of essential requirements in legislation with detailed regulatory rules to be worked out by non-governmental actors such as standard setting organizations or trade associations — and proposes its further development as a governance

strategy.*° A key example of coregulation is provided by the ‘new approach’ to technical regulation developed in the 1980s as a refinement of the Community method.*’ Within the ‘new approach’ the standard setting function is split between the legislative institutions, which set minimum

standards, and non-

governmental, but mostly supranational standard setting bodies which fill in most of the detailed standards, compliance with which is necessary to comply with the law. Another example is provided by the social dialogue, introduced in the Maastricht Treaty, under which the Commission is empowered to initiate policy discussions with the so-called social partners (employers and trade

76

Regulation through Agencies in the

EU

unions) with a view to them reaching agreement on norms which are then implemented by means of a Council and Parliament Directive. Private enforcement A final form important to EU regulatory regimes is that of private enforcement. Consumers may be involved as private enforcers of regulatory norms, though in most sectors the value of the products involved makes it unlikely that consumers will engage in formal enforcement. A number of EC Directives empower consumer groups to enforce Community law,” exemplifying the model of ‘tripartite regulation’.°° Perhaps more significant are the activities of firms in enforcing rules. Often such enforcement will be directed against competitors. Thus we are discussing regimes in which the standards are set elsewhere, for example in respect of anticompetitive behaviour (largely set down in the EC Treaty), or intellectual property rights (EU and national legislation). Where a firm is harmed by breach of the rules and this harm is significant they may consider they have an incentive to pursue formal legal enforcement. Such private enforcement may also be directed against the state for breach of the rules of a Community regime, for example in respect of procurement, or for a more general failure to comply with Community requirements such that EC directives may have a direct effect against state bodies or failure to implement properly may give rise to a right to damages against the state.

THE NORMATIVE CASE FOR AGENCIES This chapter offers an analysis of the variety of forms currently deployed for regulatory governance in the EU. The main argument of the chapter is that a regimes approach highlights the extent to which regulatory agency models are marginal within contemporary EU-level governance. Given the tendency of institutional development to occur incrementally the prospects for a radical shift towards independent European regulators appear limited. It might be more fruitful to think about how the objectives of the proposed radical institutional reform might be met through adaptation of the existing institutional arrangements.

While I may argue that other forms than the independent agency have greater prospects for successful adoption, I recognise that the arguments in favour of independent agencies are unlikely to go away, in particular because they are espoused by influential actors both in academic and within policy circles. Some assessment of the normative case for independent agenices is required. Questions about the appropriateness of developing an independent agency model for EU governance have a horizontal and vertical dimension. In the horizontal dimension there is first the question whether some operational or regulatory

Agencies for European Regulatory Governance

yi|

functions of the Commission would be better carried out by independent agencies. To the extent that a transfer of functions included the allocation of powers to make rules or apply sanctions it would require that constitutional limits on delegation of powers within the EU be addressed. A second aspect of the horizontal issue is whether some of the Commission’s meta-regulatory functions in monitoring compliance by national authorities with EU requirements would better be carried out by independent agencies, for example in the telecommunications sector. The vertical dimension to the question is whether regulatory functions carried out by member state governmental bodies would be better carried out by EU independent agencies. This is amuch more complex matter, involving as it does not only the prospect of a change in the balance of power between member state and EU level governance, but also a re-working if not an unpicking of the conventional methods of indirect governance of the EU, discussed above.

The central case for independent regulatory agencies, as set out by their’ most influential proponent, Giandomenico Majone, is that they increase the credibility of policy processes which might otherwise be tainted with political influence. Majone and his co-authors refer to a general trend towards the use of “non-majoritarian institutions’ as a means to insulate policy processes from politics.*' To the extent that the creation of EU agencies shifts powers from the Commission to agencies it is questionable whether the development of the EU agencies actually represents a shift from majoritarian to non-majoritarian institutions. The agencies are fulfilling tasks which, were they not established, would be carried out by the European Commission. Indeed, one might go so far as to suggest that the EU already has a ‘generalist independent agency’ in the form of the Commission itself.°* While there is no question that there is a complex political life in the Commission, and particularly among the College of Commissioners, the latter are appointed by national governments to be nonpartisan. The Commission itself is a non-majoritarian institution,*’ and the modest increase in influence of the European Parliament* shows little sign of fundamentally changing the position. If it is correct that the Commission is becoming politicised then it is a non-majoritarian form of politicisation. If there was a likelihood of transferring powers to agencies which are currently held by the EU legislative institutions, the Council and the Parliament, then the case for

agencies at EU level, might be closer to the non-majoritarian case in the governance of nation states. A variation on the case independent agencies is presented in the recent Sapir report which argues that institutional arrangements under which the Commission exercises regulatory functions combined with policy formation functions are too complex and the former should be hived off to independent agencies.*> Sapir suggests that regulatory functions over member states (including aspects of the competition regime dealing with state aids and

78

Regulation through Agencies in the

EU

services of general interest) should remain with the commission while regulatory functions over businesses should be hived off to independent agencies. The only example that Sapir mentions of the latter case is competition policy.” This is because it is the only example of the Commission having a direct regulatory role over businesses. Thus what appears to be a radical proposal involving the development of independent regulatory agencies in truth simply involves splitting the competition enforcement functions of the Commission in half, with the risk, no doubt, that the independent agency might develop different interpretations of the common rules to be applied from the residue of DG Competition in the Commission. The other main policy area mentioned by Sapir requiring independent agencies is in respect of safety of medicines and consumer products* where, they suggest, risk assessment is currently carried out at EU level while risk management is carried out at national level and/or by the Commission.*’ I do not think this analysis is correct. Most of the evaluation of medicinal products is carried out at national level and the same is true for consumer products. The principle of mutual recognition is deployed to permit the free movement of such goods around the EU.® There are some general norms relating to consumer and food safety at EU level , but under the new approach much of the detail is filled in either by non-governmental standards institutions (in the case of consumer products) and through national or subnational enforcement in the case of food products. Regulatory enforcement falls almost exclusively to national and subnational

authorities,

save for the limited role of the

Commission in addressing consumer product and food safety emergencies. Thus the proposal for independent EU agencies in these domains amounts to displacing a metaregulatory approach with a command and control approach, not unfamiliar within some national jurisdictions, but rather alien to governance at the EU level. There is implicit in Sapir’s approach a preference for EU-level regulation over the various metaregulatory forms which have been the hallmark of the distinctive arrangements for EU governance. The case for EU-level regulation is more directly made by Geradin and Petit who express surprise that the solution of EU-level independent agencies has not been taken up more vigorously by the Commission as a means of addressing the externalities associated with diverse national regulatory regimes.®! These externalities include both diversity in regulatory approaches, but also the consequences of weaknesses in enforcement. Such weaknesses, it is said, are demonstrated by

crises in the food safety area, such as that relating to BSE. My judgement is that the argument that EU agencies are to be preferred whenever national regulation causes externalities, pays insufficient attention to questions of proportionality — does the degree of harm from externalities justify the costs associated with the supranational remedy? The argument also comes up

Agencies for European Regulatory Governance

79

against subsidiarity arguments. It is far from clear how arguments based on efficiency of regulation should be weighed up against arguments that regulatory decision making should be made at the lowest level possible. While Majone and Everson argue that EU-independent agencies will be additional to national agencies, engaged in coordinating national efforts to implement EU regimes,” the proposed agencies must also take some autonomy from such national institutions. This involves decisions not just about regulatory efficiency, but also about the relative weight of national democratic governance arrangements as compared with the supranational institutions and processes of the EU. A key aspect of proposals for powerful EU agencies is that they should be both independent and accountable. Independence of agencies, as it is understood within the doctrines which may be derived from EU legislation, focuses largely on the relationship between agencies and businesses whose activities they oversee.” The regulatory and business activities are to be. organisationally separate; agencies are to show no discrimination in favour of particular firms. But independence in this dimension is more complex than the EU doctrine might suggest. Rigorous appointments processes, organisational separation, and controls over ‘revolving doors’ cannot remove the reality that most agencies are likely to be dependent on the businesses they regulate not just for the provision of information, but also for knowledge about how the world they regulate should be seen. Put another way, organisational independence is no guarantee of operational independence, and the latter is extremely difficult to achieve.” Recognition of this dependence is reflected in formal provisions which empower agencies to demand information for firms. It may be accentuated by provisions requiring not just consultation with businesses over changes in regulatory policy, but even the consent of the regulatee. This dependence is found not only in formal policy making, but it also typically characterises regulatory relations over matters of enforcement, where the absence of “epistemic independence’ forces regulators to accept interpretations of regulatory rules developed by and for the regulatees’ purposes.” With the current EU agencies it is clear that few are properly independent of the Commission. Not only is the Commission responsible for appointments to and financing of most of the agencies, but it also possesses key decision making powers within the policy domains.” The position of the Commission vis-a-vis the agencies has analogies to the relationship between ministers and agencies in the member states, and particularly the problem of independence.” It is common for ministries to combine functions concerned with regulatory policy and the promotion of the industry and these functions may conflict where the protection of national champions (whether or not still publicly owned) is inimical to the development of transparent and fair regulation. In many parliamentary systems of government it is actually very difficult for ministers

80

Regulation through Agencies in the

EU

to give away executive power. In many member states this factor is recognised in provisions which provide for the sharing of regulatory power between ministers and agencies. Formal provisions for ministerial oversight of agencies may justify nearly wholly opaque, but regular intervention in agency decision making by ministry officials. But greater independence may not be desirable. It is suggested that among the effects of establishing independent competition agenciesin Europe are the unintended consequences of promoting detached and over-zealous advocacy of competitive markets, together with a pursuit of expertise which has placed economists and lawyers at the heart of regulatory decision making.” As regards accountability, a focus on agencies alone might encourage us to focus on traditional mechanisms of political and judicial accountability. Certainly we can envisage a role for the European Parliament and for the Court of Justice in holding agencies accountable for meeting the objectives set down in legislation and for acting within their powers. The European Ombudsman would play a role in handling grievances about maladministration and the agencies would fall within the purview of the European Court of Auditors in respect of the regularity and efficiency of their expenditure. But, within a regimes approach such traditional mechanisms of accountability may not be as important as more pressing and systematic controls which arise out of relationships of interdependence between agencies and other key actors within the regime.”' The particular interdependencies are likely to vary from regime to regime. Each of the current EU agencies is in a relationship of interdependence with the Commission. In many cases the agencies depend on cooperation both from member state governments and from undertakings within member states, particularly for gathering information and for knowledge about the operation of activities within the domain. Agencies are liable to be caught within networks of accountability.” For those who support the assignment of more expansive powers to agencies a solution to somewhat greater problems of accountability is to import the mechanisms developed in the United States which accompany widespread delegation of powers. These mechanisms would include an administrative procedure act, creating general principles and procedures for transparent and fair decision making, and the creation of an organisation akin to the Office of Management and Budget to monitor the activities of agencies against a set of criteria of sound management, including the development of the kind of regulatory impact assessment which is widely used in the member states to assess regulatory proposals.” My own view is that such proposals are based on a false premise that the burgeoning of independent agencies at the EU level is politically possible. Such a development would appear to require Treaty revisions similar to those which established the ECB. The ECB provides an exceptional case and it appears likely that the member states and the Commission would

Agencies for European Regulatory Governance

81

Oppose 2 more extensive delegation of regulatory power. If | am wrong on this, and agency proliferation proves possible, we would then need to have a serious debate about the normative issues only hinted atinthis chapter.

CONCLUSIONS: PROSPECTS FOR INDEPENDENT AGENCIES We have already noted the structural constraints on mechanisms for regulatory govemance in the EU and may, atleast, hypothesise that they have played a role inshaping the variety of models discussed inthe previous section. My own methodological bias would suggest that future changes in the models for regulatory governance are likely to occur along the current, observable trajectories. Itisimportant to be clear that the making of a strong normative

case for the adoption of the independent agency model does not, by itself,. enhance the prospects that it will be adopted in the face of structural and political obstacles. Central to my doubts about whether an independent agency model might be widely adopted is the ambivalence of the European Commission. The Commission formula for the EU regulatory agencies appears to represent, simmultancously, an embracing of the agency model, and a rejection of its development along the lines ofthe independent regulatory agency, In its limited attention to institutional reforms at the level of operations or imaplementation in the Governance White Paper the Commission proposes the creation of new autonomous regulatory agencies to improve the application and enforcement of rules across the EU.” It isstated that such agencies can draw on greater expertise, achieve greater visibility, and reduce costs for businesses.” The Commission’s vision of this new breed of regulatory agencies is 2 restrictone. ed While they will have power to make decisions on particular cases, along the lines of the existing ‘regulatory’ agencies — Office for Harmonization in the Internal Market; Community Plant Varieties Office; European Medicines Evaluation Agency ~—they will not have powerto make regulatory rules.” Additionally they will not be permitted to exercise powers assi to the Commission in the Treaties (thus ruling gned out anEU competition agency), nor will they be assigned responsibilities which require them “to arbitrate between conflicting public interests, exercise political discretion or carry out complex economic assessments’.”’ Under the Commission’s classification ‘executive agencies’ are said to be responsible for ‘purely managerial tasks’ while ‘regulatory agencies are required to be actively involved in the executive function by enacting instruments which help to regulate 2 specific sector’.” The Commission proposes a different an inter-institutional agreement on an distinction 2s the basis for developing

82

Regulation through Agencies in the

EU

operating framework for the agencies. This distinguishes decision-making agencies from executive agencies, the former having the capacity to make decisions which legally bind third parties, the latter having lesser, managerial powers.” The Commission’s position could demonstrate the desire of the Commission to guard its own powers and protect them from diffusion or it might simply be a recognition of the cultural and political obstacles in the way of a more expansive regulatory agency model. The Commission has explicitly stated that the use of regulatory agencies does not dilute the Commission’s policy responsibility in the applicable domain, but rather represents a transfer of limited executive functions.*° The Commission seeks to delegate to agencies rather routine operational tasks, but seeks to retain control of higher profile and more important policy making, monitoring and enforcement processes.*' This preference may be explained both by reference to the limited resources available to the Commission and to a wish to engage in a form of bureaushaping, under which utility maximising officials are seen to increase the proportion of their work which relates to high policy making.** Whatever the explanation it is difficult to conclude in any other way other than to say it seems most unlikely that the independent agency model for the EU will rise sufficiently up the policy agenda to have a serious chance of adoption.

NOTES ' A Kreher (1997), ‘Agencies in the European Community: A Step Towards Administrative Integration in Europe’, Journal of European Public Policy, 4, 246. 2 The White Paper largely focuses on policy and legislative processes — and constitutes, in particular, a rearguard action to defend and deepen the “Community method’ in which the Commission plays a key role in initiating and brokering legislation, against incursions from new policy models such as the Open Method of Coordination (by which the Commission role is substantially undermined). 3 André Sapir (2003), An Agenda for a Growing Europe: Making the EU Economic System Deliver: Report of an Independent High-Level Study Group Established on the Initiative ofthe President of the European Commission, European Commission.

4 Marc Allen Eisner (2000), Regulatory Politics in Transition, 2nd edn, Johns Hopkins University Press, Baltimore. Eisner’s usage appears to be derived from scholarship in international relations where an influential definition of regimes is ‘sets of implicit or explicit principles, norms, rules, and decision-making procedures around which actors’ expectations converge in a given area of international relations’. Krasner (1983), /nternational Regimes (Cornell University Press, Ithaca). > Christopher Hood, Henry Rothstein and Robert Baldwin (2001), The Government ofRisk (Oxford University Press, Oxford). ® Stephen Wilks and Ian Bartle (2002), ‘The Unanticipated Consequences of Creating Independent Competition Agencies’, West European Politics, 25, 148.

7 EC Treaty A 5. Protocol (30) to the EC Treaty, on the application of the principles of subsidiarity and proportionality, adopted in 1997, introduces also a requirement of proportionality — Grainne de Burca (1999), “Reappraising Subsidiarity’s Significance After Amsterdam’, Harvard Law School,

1/99. 8 Case 9/56 Meroni v High Authority [1957-8] ECR 133, 151. 9 EC Treaty Arts 81, 82, 86, 87.

Agencies for European Regulatory Governance

83

'0 EEC Council: Regulation No. 17: First Regulation implementing Arts 85 and 86 of the Treaty Official Journal P 01, 21/02/1962 P. 0204 —021.The Provision for the application of fines, both for the failure to comply with the Commission investigatory powers and the Treaty rules is contained in Art 15. 'l_ EC Treaty Art 86(3).

'2 Commission Directive 83/301/EEC of 16 May 1988 on competition in the markets in telecommunications terminal equipment O./. L 131, 27.05.1988,

73-77; Commission

Directive

90/388/EEC of 28 June 1990 on competition in the markets for telecommunications services O.J. L 192, 24.07.1990, 10-16; Commission Directive 96/19/EC of 13 March 1996 amending Directive

90/388/EEC with regard to the implementation of full competition in telecommunications markets. O.J. 74, 22.03. 1996, 13-24. 'S These sensitivities were reflected in litigation initiated by member state governments to challenge the extensive use made of the Commission’s autonomous legislative power in the Terminals and Services Directives: French Republic vyCommission of the EC (Case C-202/88),

1991, ECR-I 2223; Joined Cases C-271, C-281, C-289/90 Spain v Commission of the EC [1992] ECR-I 5833. '4 Regulation No. 19/65/EEC of 2 March of the Council on application of Art 85 (3) of the Treaty to certain categories of agreements and concerted practices, O.J. P 036, 06/03/1965 P. 0533

- 0535. 'S_ Francis Snyder (1995), ‘The Effectiveness of European Community Law: Institutions, Processes,

Tools and Techniques’, in T Daintith (ed.), Jmplementing EC Law in the United Kingdom (Wiley, London). '6 Council Regulation (EC) No. 1/2003 on the implementation of the rules on competition laid down in Arts 81 and 82 of the Treaty. O.J. L1/1 4.1.2003. '7 European Commission, European Governance: A White Paper, COM (2001) 428 (2001).

'8 Christine Parker (2002), The Open Corporation: Self-Regulation and Democracy (Cambridge University Press, Melbourne), Bronwen Morgan (1999), “Regulating the Regulators: MetaRegulation as a Strategy for Reinventing Government in Australia’, Public Management, 1, 49.

19 Christopher Hood, et al. (1999), Regulation Inside Government: Waste-Watchers, Quality Police, and Sleaze-Busters (Oxford University Press, Oxford). 20 For the energy sectors see the discussion of the Florence and Madrid processes in Leigh Hancher’s contribution to this volume.

21 Directive 2002/21/EC of the European Parliament and of the Council of 7 March 2002 on a common regulatory framework for electronic communications networks and services (Framework Directive); Commission Directive 2002/77/EC of 16 September 2002 on competition in the markets for electronic communications networks and services (Text with EEA relevance) Official

Journal L 249, 17/09/2002 P. 0021 — 0026; Directive 2002/19/EC of the European Parliament and of the Council of 7 March 2002 on access to, and interconnection of, electronic communications networks and associated facilities (Access Directive) Official Journal L 108, 24/04/2002 P. 0007 —

0020. 22 e.g. Guidelines on the application of EEC competition rules in the telecommunications sector OJ C 233, 6.9.1991; Commission guidelines on market analysis and the assessment of significant

market power under the Community regulatory framework for electronic communications networks and services Official Journal No. C 165, 2002, Item 3 (2002/C 165/03). 23 Fifteen of the agencies were established under the EC Treaty, one under the Euratom Treaty and four under the second and third pillars of the European Union: Commission Communication “The Operating Framework for the European Regulatory Agencies’ COM (2002) 718 Final 11.12.202, at

p. 3. The EC Treaty Agencies are: European Centre for the Development of Vocational Training (Regulation (EEC) No. 337/73 of 10.02.75); European Foundation for the Improvement of Living and Working Conditions (Regulation (EEC) No. 1365/75 of 26.05.75); European Environment

Agency (Regulation (EEC) No. 1210/90 of 07.05.90); European Training Foundation (Regulation (EEC) No. 1360/90 of 07.05.90); European Monitoring Centre for Drugs and Drug Addiction (Regulation (EEC) No. 302/93 of 08.02.93); European Agency for the Evaluation of Medicinal Products (Regulation (EEC) No. 2309/93 of 22.07.93); Office for Harmonisation in the Internal

Market (Regulation (EC) No. 40/94 of 20.12.93); European Agency for Safety and Health at Work (Regulation (EC) No. 2062/94 of 18.07.94); Community Plant Variety Office.(Regulation (EC)

84

Regulation through Agencies in the

EU

No. 2100/94 of 27.07.94); Translation Centre for Bodies of the European Union (Regulation (EC) No. 2965/94 of 28.11.94); European Monitoring Centre on Racism and Xenophobia (Regulation

(EC) No. 1035/97 of 02.06.97); European Agency for Reconstruction (Regulation (EC) No. 2454/ 1999 of 15.11.99); European Food Safety Authority (Regulation (EC) No. 178/2002 of 28.01.02); European Maritime Safety Agency (Regulation (EC) No. 1406/2002 of 27.06.02); European Aviation Safety Agency (Regulation (EC) No. 1592/2002 of 15.07.02). The other agencies are: Euratom Supply Agency, Article 52ff of the Euratom Treaty (see also the Statutes of the Agency, published in OJEC No. 534, 06.12.58); European Union Institute for Security Studies (Joint Action of 20.07.2001, OJEC No. L 200, 25.07.01); European Union

Satellite Centre (Joint Action of

20.07.2001, OJEC No. L 200, 25.07.01). European Police Office-Europol (Convention of 26.07.95, OJEC No. C 316, 27.11.95); Eurojust (Decision of 28.02.02, OJEC No. L 63, 06.03.02). 24 The most developed is the proposal for a Regulation establishing a European Railway Agency, COM (2002) 23, 23.01.2002. 25 Commission Communication ‘The Operating Framework for the European Regulatory Agenicies’ COM (2002) 718 Final 11.12.202, at p. 3. It is reported that the Legal Service of the Commission has a fuller definition ‘according to which the concept of a Community agency relates to decentralised bodies have the following characteristics: creation by regulation, legal personality, autonomous management bodies, financial independence, staff covered by the Staff Regulations of Officials and the Conditions of Employment of Other Servants of the European Communities, defined missions and tasks’. White Paper on Governance, Work Area 3, Improving the Exercise of Executive Responsibilities, Report by the Working Group ‘Establishing a Framework of Decision-Making Regulatory Agencies’ (Group 3a) SG/8597/01-EN. 26 e.g. Commission Communication ‘The Operating Framework for the European Regulatory Agenicies’ COM (2002) 718 Final 11.12.202, p. 2. 27 e.g. Carl Fredrik Bergstr6m and Matilda Rotkirch, Decentralized Agencies and the IGC: A Question of Accountability, Swedish Institute for European Policy Studies, 14 (2003). Geradin and

Petit appear to recognise the contradiction between their acceptance of the argument that the development of the agencies represents a decentralisation, on the one hand, and their normative agenda which is very much orientated towards the centralisation of regulatory decision making within EU level agencies, at the expense of National Regulatory Authorities: Jean Monnet Working Paper No 01/04, New York University Law School.

28 A point effectively made in the case study of EMEA written by Les Metcalfe in M. Everson, G. Majone, L. Metcalfe and A. Schout (1999), The Role of Specialised Agencies in Decentralising EU Governance Report presented to the Commission, 178.

29 Regulation (EEC) 2309/93 of 22.07.93. 30 Regulation (EC) No. 178/2002 of 28.01.02. 31 M. Everson, G. Majone, L. Metcalfe and A. Schout (1999), The Role of Specialised Agencies in

Decentratising EU Governance Report presented to the Commission, chapter XII. 32 Case C-11/00, Commission vy ECB ECR [2003], not yet reported para 41. 33 R. Daniel Keleman (2002), ‘The Politics of “Eurocratic” Structure and the New European Agencies’, West European Politics, 25, 93. 34 Kathleen R. McNamara, ‘Rational Fictions: Central Bank Independence and the Social Logic of Delegation’. ibid. 47.

35 Art 110 EC. 36 Decision of the European Parliament on the regulations and general conditions governing the performance of the Ombudsman’s duties Official Journal L 113, 4.5.1994, p. 15 (European Ombudsman Statute), Art 9. 37 Giandomenico Majone (2000), ‘The Credibility Crisis of Community Regulation’, Journal of Common Market Studies, 38, 273. 38 Commission decision of 6June 2001 (2001/1501/EC). 39 Imelda Maher (2002), ‘Competition Law in the International Domain: Networks as a New Form

of Governance’, Journal of Law and Society, 29, 111. 40 Council Regulation (EC) No. 1/2003 on the implementation of the rules on competition laid down in Arts 81 and 82 ofthe Treaty. O.J. L1/1 4.1.2003.

41 Commission Decision No. 2002/627/EC establishing the European Regulators Group for Electronic Communications Networks and Services, O.J. L 200/38, 30.07.2002.

Agencies for European Regulatory Governance

85

#2 Dermot Hodson and Imelda Maher (2001), ‘The Open Method as a New Mode of Governance’, Journal of Common Market Studies, 39(4), 719.

43 Rewe-Zentral AG v Bundesmonopolverwaltung fiir Branntwein [1979] ECR 649; Kenneth Armstrong (2002), ‘Mutual Recognition’, in C. Barnard and J. Scott (eds), The Legal Foundations of the Single European Market (Hart, Oxford). # Colin Scott, ‘Analysing Regulatory Space: Fragmented Resources and Institutional Design’ (2001) Public Law, 329. 45 Tan Ayres and John Braithwaite (1992), Responsive Regulation: Transcending the Deregulation Debate (Oxford University Press, Oxford). 46 White Paper, p. 21. For a fuller discussion see Report of the Working Group ‘Better Regulation’ White Paper on European Governance Work Area no. 2, Handling the Process of Producing and Implementing Community Rules (2001). 47 Council Resolution of 7 May 1985 on a new approach to technical harmonization and standards Official Journal C 136, 04/06/1985 pp. 0001-0009; Council Resolution of 18 June 1992 on the role of European standardisation in the European economy Official Journal C 173, 09/07/1992 pp. 0001— 0002; Councii Resolution of 10 November 2003 on the Communication of the European Commission “Enhancing the Implementation of the New Approach Directives’ Official Journal C 282, 25/11/

2003 pp. 0003-0004. 48 Arts 138 and 139 EC. 4# e.g. Council Directive 93/13/EEC of 5 April 1993 on Unfair Terms in Consumer Contracts O.J. L 095, 21.04.1993, 23-34. 50 Tan Ayres and John Braithwaite, (1992), Responsive Regulation: Transcending the Deregulation Debate (Oxford University Press, Oxford). 5! M. Everson, G. Majone, L. Metcalfe and A. Schout (1999), The Role of Specialised Agencies in Decentralising EU Governance Report presented to the Commission, 9. 52 _R. Daniel Keleman (2002), ‘The Politics of “Eurocratic” Structure and the New European Agencies’, West European Politics, 25, 93. 53 Jonas Talberg (2002), ‘Delegation to Supranational Institutions: Why, How, and with What

Consequences?’ Ibid. 23; Mark Thatcher and Alec Stone Sweet, ‘Theory and Practice of Delegation to Non-Majoritarain

Institutions’,

West European Politics, 25, 1, define non-

majoritarian institutions as ‘those governmental entities that (a) possess and exercise some grant of specialised authority, separate from that of other institutions, but (b) are neither directly elected by the people, nor directly managed by elected officials’. They include the European Commission within the set. But compare Stephen Wilks and Ian Bartle (2002), ‘The Unanticipated Consequences of Creating Independent Competition Agencies’, West European Politics, 25, 148, who argue that as appointees of national governments commissioners are not immune from domestic political considerations. This is perhaps particularly true in the competition policy

field. 54 Cf. M. Everson, G. Majone, L. Metcalfe and A. Schout (1999), The Role ofSpecialised Agencies

in Decentralising EU Governance Report presented to the Commission, 19. 55 André Sapir (2003), An Agenda for a Growing Europe: Making the EU Economic System Deliver:

Report of an Independent High-Level Study Group Established on the Initiative of the President of the European Commission, European Commission. See Chapter 12. 56 Tbid. 57 [bid. does also mention sectoral regulation as a candidate for creating independent agencies. On their own analysis, which suggests that telecoms regulation in the EU is largely within the legal jurisdiction of the member states (‘national regulator acting under national laws’, 154), it is not clear why they think the Commission’s role as coordinator would best be fulfilled by an independent agency. Ona correct analysis the role of Community law within the network sectors is much greater than Sapir suggests (national regulators largely acting under national laws implementing Community laws), and accordingly the Commission has a major role as metaregulator, steering the activities of member state governments and agencies. Accordingly, on Sapir’s own analysis, this function should not be a candidate for an independent agency because it involves the regulation largely of member state governments and agencies. 58 Tbid. 59 Tbid.

86

Regulation through Agencies in the

EU

60 Kenneth Armstrong (2002), ‘Mutual Recognition’, in C. Barnard and J. Scott (eds), The Legal Foundations of the Single European Market (Hart, Oxford).

61 [D), Geradin and N. Petit, supra note 27. 62 Giandomenico Majone and Michelle Everson (2001), ‘Institutional Reform: Independent Agencies, Oversight, Coordination and Procedural Control’, in O. De Schutter, N. Lebessis and J. Paterson (eds), Governance in the European Union (European Commission, Brussels).

63 Tbid. A detailed account of the role of the BSE crisis in stimulating the creation of the European Food Safety Agency can be found in R. Daniel Keleman (2002), ‘The Politics of “Eurocratic” Structure and the New European Agencies’, West European Politics, 25, 93. 64 Giandomenico Majone and Michelle Everson (2001), ‘Institutional Reform: Independent Agencies, Oversight, Coordination and Procedural Control’, in O. De Schutter, N. Lebessis and

J. Paterson (eds), Governance in the European Union (European Commission, Brussels) . 65 A similar bias towards independence from regulated businesses is found in WTO documents. The Reference Paper on Telecommunications Services, which forms the basis for trade in telecommunications at the WTO level specifies that ‘The regulatory body is separate from, and not accountable to, any supplier of basic telecommunications services. The decisions of and the procedures used by regulators shall be impartial with respect to all market participants’, World Trade Organisation, ‘Telecommunications Services: Reference Paper’ (1997), 36 /nternational Legal Materials 367, para 5. 66 Michael Power (1997), The Audit Society: Rituals of Verification (Oxford University Press, Oxford). 67 Nancy Reichman (1992), ‘Moving Backstage: Uncovering the Role of Compliance Practices in Shaping Regulatory Policy’, in K. Schlegel and D. Weisburd (eds), White-Collar Crime Reconsidered (Northeastern University Press, Boston); Clare Hall, Colin Scott and Christopher Hood (2000),

Telecommunications Regulation: Culture, Chaos and Interdependence Inside the Regulatory Process (Routledge, London).

68 M. Everson, G. Majone, L. Metcalfe and A. Schout (1999), The Role of Specialised Agencies in Decentralising EU Governance Report presented to the Commission, 178. 69 Murray Horn (1995), The Political Economy of Public Administration (Cambridge University Press, Cambridge). 70 Stephen Wilks and Ian Bartle (2002), “The Unanticipated Consequences of Creating Independent Competition Agencies’, West European Politics, 25, 148. 71 Colin Scott (2000), ‘Accountability in the Regulatory State’, Journal of Law and Society, 27,

38. 72 Cf. M. Everson, G. Majone, L. Metcalfe and A. Schout (1999), The Role of Specialised Agencies in Decentralising EU Governance Report presented to the Commission, 14-15. 73 M. Everson, G. Majone, L. Metcalfe and A. Schout (1999), The Role of Specialised Agencies in Decentralising EU Governance Report presented to the Commission, 12-13. 74 European Commission, European Governance: A White Paper, COM (2001) 428 (2001). 7> Ibid. These virtues appear to be outlined by way of comparison with how things would be if the Commission were to fulfil the tasks. But arguments about reduced costs to business would also apply to EU agencies when compared with national regulatory authorities, especially for firms operating in more than one state and thus subject to multiple regimes. 76 This position set out in the Governance White Paper, p. 24, appears to be at variance with the later communication: Commission Communication ‘The Operating Framework for the European

Regulatory Agencies’ COM (2002) 718 Final 11.12.202, p. 4, in which the Commission states that “The Concept of European Regulatory Agency designates agencies required to be actively

involved in exercising the executive function be enacting instruments which contribute to regulating a specific sector’. The reference to enactment suggests that agencies will have general rule making powers, but in fact must refer only to the power to make legally binding decisions in individual

cases. 77 Tbid. 78 Commission Communication ‘The Operating Framework for the European Regulatory Agencies’ COM (2002) 718 Final 11.12.202, 4. 7 Commission Communication ‘The Operating Framework for the European Regulatory Agencies’

COM (2002) 718 Final 11.12.202, 7-8.

Agencies for European Regulatory Governance

87

80 Commission Communication ‘European Governance: Better Lawmaking’ COM (2002) 275 Final Deo.

81 R. Daniel Keleman (2002), ‘The Politics of “Eurocratic” Structure and the New European Agencies’, West European Politics, 25, 93. 82 Patrick Dunleavy (1991), Democracy, Bureaucracy and Public Choice (Harvester Wheatsheaf, Hemel Hempstead).

5.

Delegation to EU Non-majoritarian Agencies and Emerging Practices of Public Accountability Deirdre Curtin

INTRODUCTION Calls for more ‘accountability’ are increasingly heard at virtually every conceivable level of governance in the twenty-first century, from the global to the local, including all the intermediary levels in between. According to political theorists, accountability is not of necessity confined to democratic forms of governance, although it is in democracies that greater demands for accountability will generally be heard.'! Accountability is also increasingly expected with regard to those international organizations that have been delegated ‘powers’ by democratic Member States. According to some we even live in the ‘age of accountability’.” Be that as it may, the concept of accountability often appears to be used rather vaguely, as ‘a garbage can filled with good intentions, loosely defined concepts and vague images of good governance’.’ In the modern state, accountability can however be said to have two major meanings, which overlap.’ First there is the standard meaning that those who exercise power (governments, elected representatives, appointed officials) are in a sense stewards and must be able to show that they have exercised their powers and discharged their duties properly (ultimately to the voters at election time). This is often referred to as political accountability. Second, there is a somewhat more technical use of accountability as referring to those arrangements made for securing conformity between the values of a delegating body and those to whom powers are delegated (delegatee). This tends to be referred to as administrative accountability. In an effort to use the term accountability more rigorously, Bovens refers to ‘public’ accountability as the open process of explaining and justifying actions or omissions by

8&8

Delegation to

EU Non-majoritarian Agencies

89

public servants or those exercising public authority to an accountability forum.’ The latter can be institutionally specific (parliament, Ombudsman, auditors, courts of law) but it can also be more general such as the general public (media, Internet). It includes mechanisms of political accountability, such as that of the Minister (or Commissioner) to parliament but is not limited to this and can include transparency mechanisms, contro) by courts and other accountability institutions. Part of the mechanism of public accountability can consist in making information freely available, through the medium of the Internet or otherwise, but in addition the aspect of account giving (explaining, justifying) being compulsorily directed to a specific (institutional) forum that will debate and judge the same, is considered as crucial. Those authors who focus on (the lack of) accountability in the evolving political system of the European Union* (more than a classic international organization, less than a state) tend to be concerned with the (broader) issue of making decision making, at that level more democratic, as well as with the (narrower) issue of controlling delegated powers. The latter seems to be dominated by the former. Thus, in the words of Fisher, ‘while accountability within the EU context is often characterized as the principle for an era of innovative governance unshackled from conventional understandings of the constitutional state, it is doubtful whether accountability and accountability processes can operate outside the context of liberal democratic government’.’ Others, like Moravesik, defend the EU’s ‘democratic deficit’ on the basis that it pretty much parallels what isfound in the ‘real world’ of advanced industrial democracies; in any event, the EU is, on his findings, not ‘an unaccountable technocracy’ but is the subject of both direct and indirect mechanisms of democratic axcoumtability.” From the early days of the European Community political scientists, in particular, defended a rather dominant theory, ofunderstanding the process of European integration as 2 series of delegations of power from the foundational Member States to (most specifically) the Commission as a supranational actor.” In much more recent times a legal perspective has been that the Commission’s policy making and executive functions are exercised in a delegated relationship of agency for the politically legitimated Member States.” This is a rather isolated approach among lawyers; their focus tends to be on the legal (constitutional) parameters of the formal delegation of powers rather than the issue of agency after delegation. Given, however, the fact that the political system of the EU has evolved quite substantially in practice in recent years, to the point where the EU unquestionably possesses rather full scale legislative power and increasingly expansive executive power, it seems less accurate and less appropriate to view the exercise of power at the EU level itself as purely instances of delegated power to the EU institutional ‘agents’, to be controlled by Member State principals. Rather, the issue of

90

Regulation through Agencies in the

EU

(democratic) accountability can, it seems, be separately raised as part of the EU political process itself. Part of the problem is that it is not, however, easy to transpose to the EU the customary clear distinction between legislative and executive authority, i.e. between some institutions empowered only to pass laws and others merely implementing legislation or issuing regulations. On the contrary, the EU (Community) was founded upon the notion of the balance of powers,'' which can be interpreted as a commitment to the rationale underlying the trias politicus, that of the control of power.'* The problem in practice is, however, that there is a ‘growing confusion of roles between the various bodies, leading for example to conferring Council of Ministers’ responsibilities on the European Council, Commission tasks on the Council of Ministers and tasks which should be dealt with by COREPER on the specialist committees’.'° In recent years both the Commission and the Council have been establishing and de facto delegating (executive and administrative) powers and tasks to what are formally termed the ‘decentralised agencies’ of the European Union.“ Such bodies are non-majoritarian in nature, and are established in order to perform defined tasks in a relatively autonomous fashion at the European level. How does this affect the underlying principles of balance of powers among the institutions according to the original Treaty design? And how do the Commission and the Council as principals ensure that their agents do not drift away from the principals? To what extent can one speak of mechanisms of accountability (political, administrative and public) in the institutional design and subsequent practices of non-majoritarian decentralised EU agencies? What I propose to do in this chapter is to examine the ‘practices’ with regard to a number of fairly exemplary, and/or new, agencies with a view to shedding some light on the type of measures that are adopted in practice, in order to enable principals to retain some degree of control over the exercise of discretion by the agency in question.

CHAINS OF DELEGATION TO AND FROM EU NONMAJORITARIAN AGENCIES Conceptual and Terminological Issues In contemporary national democracies the conditional designation of others to make decisions, which the delegator (the party delegating) was originally authorized to do in their name and place, tends to be viewed as one single chain of delegation. The single chain of delegation runs all the way from the voters to the civil servants that ultimately implement public policy albeit with several chains.'° It is a single chain of delegation with multiple links. In the

Delegation to

EU Non-majoritarian Agencies

91

EU context it is very difficult to establish a single chain of delegation in this fashion. Moreover it is difficult to conceive of each link in terms of a single principal delegating to a single or multiple non-competing agents. Even if we view the national legislature (connected to the voters) as delegating power upwards via the national governments to the EU supranational actors (in particular the Commission, much less so the Council of Ministers), it is difficult to link that same authorisation by the national legislature with the output of the EU actors. Moreover the national legislatures may be connected to the national voters but the elections are on purely national matters not on the output of the supranational actors. It is thus virtually impossible to conceive of a chain of delegation in singular terms. At the European level itself it is clear that, although the European Parliament may indeed be directly connected to national voters through direct elections to the European Parliament, these elections, apart from the low turnouts, are not focused on European issues, nor do they give a mandate at the European level to a ‘government’. Moreover it is not the Eurepean Parliament as such that transfers power in a single chain to the other supranational actors at the European level, although it may, in certain circumstances, be a co-principal. It seems rather that, at the European level, we are forced to consider multiple chains of delegation with multiple links, largely because it is not the legislature that is both directly connected to the voters and, at the same time, delegating its powers. When we consider the delegation of powers at the horizontal EU level the chain of delegation fragments even further, as de facto there are two institutions performing tasks of an executive nature in the EU and both of them separately delegate discrete executive tasks to a variety of independent agencies for a variety of reasons. The issue whether they were originally delegated these executive tasks by the national parliaments (individually) is contested, but in my view difficult to maintain in practice as a single unbroken chain of delegation, given the manner in which the contemporary political system of the EU has evolved. It is, in any event, clear that the European Parliament as co-legislator has not delegated these executive powers to either the Commission or the Council in the first place. Delegation for the purposes of this chapter is defined as an authoritative decision, formalised as a matter of public law, that (a) transfers policy making

authority away from the established, representative organs (those that are directly or indirectly elected, or are appointed by elected politicians) to (b), a public non-majoritarian institution. The definition given by Thatcher and Sweet Stone'® has been adapted in four respects to make it ‘fit’ the specific institutional configuration of the EU. Thus it includes representative organs that are not only directly elected (the European Parliament), but also those that are indirectly elected (the Council of Ministers, the European Council).

92

Regulation through Agencies in the

EU

According to this definition, therefore, the European Commission is included as a putative principal in the relevant chain of delegation at the EU level itself, as it is appointed by those who are directly or indirectly elected. A public nonmajoritarian institution is a wide term for all those organisations that spend public money and fulfil a public function, but exist with some degree of independence from elected politicians. Principals can be further defined as those institutions (the Council of Ministers, the Commission and — potentially in the future — the European Council) that use their authority to establish non-majoritarian institutions through a public act of delegation. Agents, on the other hand, are those who govern by exercising delegated powers. This is the case of most of the original 15 ‘Community Agencies’ and the more recently established Union bodies as well as those in the pipeline, most of which were established by secondary legislation. The non-majoritarian institutions at the EU level are defined, moreover, for the purposes of this chapter as only those ‘technocratic’ agencies that are exercising public authority and are public in nature and are to some degree autonomous of the principal. It does not include therefore offices that enjoy no legal autonomy of their own and are allied as such to services of the Commission (such as, for example, the Humanitarian Aid Office, the Cooperation Office or the Development Office). Here one cannot regard policy-making authority as having been ‘transferred’ from the principal. Nor does it include those private companies that the Commission — for many years — was in the habit of ‘contracting-out’ various (largely management) tasks. This latter phenomenon is not insignificant, amounting currently to approximately one-sixth of the Community budget, in the order of 14 billion Euros per year,’’ but as it involves private contractors and offices (technical assistance offices as they are known) falls outside the frame of reference of this chapter. In the non-EU literature on delegation to non-majoritarian agencies it is commonly assumed that it is de facto the legislature that delegates powers to agencies or non-majoritarian institutions.'* This is premised on the assumption that in political systems, which can be somewhat loosely defined as parliamentary democracies, it will virtually always be the parliaments that are the ‘principals’ in the chain of delegation having been directly elected by the voters.'” In the EU system a quite different sort of logic (or logics) prevails (literally a non-majoritarian one), as it is not the parliaments, as such, that are the principals but, even in the closest analogy possible, always together with its legislative partner, the Council of Ministers (which is indirectly elected in the national arena). Moreover the Council of Ministers can act on its own in

establishing non-majoritarian institutions at the EU level without formally involving the European Parliament (nor indeed the national parliaments) in its decision to delegate powers. Finally, the Commission as a non-legislative

Delegation to

EU Non-majoritarian Agencies

93

power can delegate its powers and tasks to non-majoritarian institutions (with varying degrees of involvement of one or both arms of the legislative authority in some instances). What this description already reveals is that, at the EU level, there is quite often multiple (legislative and/or executive) principals involved in the establishment of non-majoritarian institutions, something that will have implications for the accountability framework to be studied later in the chapter. The piecemeal logic appears to dominate at the European level, making it extremely difficult to define a clear typology for the classification of each non-majoritarian agency in a homogeneous category’.”” As one report put it: ‘for each of them, the decision to create them is motivated by the need to

respond to the particular circumstances of the moment. In certain cases the occurrence of a crisis that has aroused public sensitivity is the basis of the decision to create the agency’. They include those Agencies that perform purely management tasks, those who gather and collate information and those whose tasks include more regulatory-like tasks including the power to adopt decisions that are legally binding on third parties.?! Not only can (and do) very different sorts of rationale apply for the delegation of powers to take place in the first place (enhancing efficiency of rule making, resolving commitment problems, overcoming information asymmetries in technical areas of governance, avoiding blame etc.) but it is virtually inherent in the process that some degree of discretion will be conferred on the agent which may be reined in or otherwise by a variety of control mechanisms. The ‘trick’ for political science analysis is whether the power delegated is just enough to enable agents achieve (the principals) desired outcomes without straying beyond that by developing their own interests. Lawyers, on the other hand, stumble at the notion of discretion being

delegated but have few tools other than a textual analysis of the constituting texts to enable them to establish this. This issue has been particularly important in the EU context where as far back as 1958 the Court of Justice ruled that no discretionary (regulatory) powers could be delegated to agencies.” Any delegation of power to Agencies must be limited to implementing powers that are precisely defined and can be rigorously monitored. The current institutional system does not allow delegation either of the power to adopt secondary legislation nor the power to formulate rules for implementing legislation. Thus some European agencies (for example, the European Agency for the Evaluation of Medicinal Products) do not themselves adopt decisions, but prepare opinions (on product authorizations for example) that are ‘subsequently formally adopted by the Commission itself. Other Agencies have been given the power to adopt decisions themselves, applying the rules of secondary legislation to specific cases. The thinking seems to be that the latter is permissible, once it involves a margin of technical appraisal related to

94

Regulation through Agencies in the

EU

precise objectives and not a ‘broad margin of appraisal in a complex economic field or a decision with political implications’. What is not permissible, it has been assumed until now, is, for example, the full-scale delegation by the

Commission of its decision-making power with regard to competition policy to a European Competition Agency.” Underlying both the legal debate and the political science literature are differing views on the nature of discretion. A helpful way of viewing it, following Thatcher and Sweet Stone,” is in terms of a theoretical ‘zone of discretion’. This zone is constituted by (a) the sum of delegated powers (policy discretion) granted by the principal to the agent, minus (b) the sum of control instruments, available for use by the principals to shape (constrain) or annul (reverse) policy outcomes that emerge as a result of the agent’s performance of set tasks. Needless to say explicit discussion rarely takes place on the existence or otherwise of such a zone of discretion with regard to different agencies. One rather gets the distinct impression that many of the principals at the European level are quite happy to leave the issue of discretion very much in the air. The Commission as Principal Most existing studies that apply the principal—agent framework to the EU view Member States as principals who delegate powers to supranational agents, typically the Commission.”* With the manner in which the political system of the EU has evolved over the course of the last decade or so, this analysis

reveals too limited an understanding of the nature of the integration process. It

overlooks, for example, the fact that the Commission acts itself as a principal in the case of delegation to (some) European agencies. Over time the Commission, for a variety of reasons, has sought to delegate its own powers and tasks to others. First, it did so in an informal fashion to private third parties and did not institute a system of control of the powers thus ‘delegated’.*° More recently, as a result of criticisms of this informal system, it instituted a novel system of so-called ‘executive agencies’ with clearly defined (management) tasks. The main objective underlying the creation of such Executive Agencies is to empower structures with their own legal personality, distinct from a Commission department with the execution of management tasks (inter alia implementation of a Community programme). These tasks are not directly linked to ‘tasks requiring discretionary powers in translating political choices into actions’’’ that the Commission wants to keep under its direct control (for example, its role as guardian of the Treaties or its right of initiative). Executive agencies are termed by the Commission as ‘Community bodies’. In fact this is a misnomer, the term ‘Commission bodies’ is more accurate, given the fact that they are exercising specific tasks of implementation and administration,

Delegation to

EU Non-majoritarian Agencies

95

directly delegated by the Commission itself to them without intervention of any outside body.” At the same time the Commission retains ‘control’ over its operational performance of its tasks (see further infra.). Obviously such executive agencies are designed to have no real zone of discretion and thus will not fall foul of the criteria laid down by the Court of Justice in the Meroni case. In its White Paper on Governance the Commission flagged the fact that, in its view, there was (in addition) a need for more independent and autonomous

structures, in the form of European Regulatory Agencies that would be delegated with certain specific (discretionary) regulatory functions.*” The putative establishment of such Agencies is to be seen as a further part of the Commission’s attempt to refocus on what it terms its ‘core functions’ of policy making and policy enforcement.*! They will have — this is the intention — a significant and autonomous role in performing essentially what can only be regarded as part of the Commission’s own executive duties. In other words, in specific policy sectors, the intention is that Regulatory Agencies will enact regulatory instruments subject to a general framework of conditions governing their creation, operation and supervision. The specific intention is that certain of these regulatory agencies will be empowered to adopt individual decisions, binding on third parties in the specified policy area. Unlike the new executive agencies such new-style regulatory agencies will be organisationally and functionally autonomous of the Commission. Indeed, there are currently avant /a lettre at least three agencies in existence

which satisfy the Commission’s criteria of regulatory decision-making power. At the end of April 2004 the Commission proposed a new Community Fisheries Control Agency that, in its own terms, will constitute a ‘regulatory agency’.* The intention is that it will be tasked inter alia in the field of ‘operational coordination’ of control and inspection by Member States in line with an ‘appropriate’ Community strategy. Operational coordination will include ‘forming multinational teams for inspection at sea and ashore in the interest of the Community’ as well as ‘issuing instructions concerning the geographical areas, the stocks, fisheries and the fleets to be surveyed and inspected in a given period’. The idea is that these tasks will not interfere with the functions of the Commission, as such, under the Treaty and the Common

Fisheries Policy, since control and enforcement of the rules of the CFP is primarily a matter for public authorities of the Member States. In other words it is effectively not the Commission’s own executive powers that have been delegated to it, but those of the Member States! Presumably for this reason, although it is not explicitly discussed in these terms, the Commission does not envisage any conflict with the Court’s ruling in the Meroni case of 1958. The issue with regard to the newer style regulatory agencies is how precisely is one to distinguish between what can be delegated in constitutional

96

Regulation through Agencies in the

EU

terms and what not? With regard to the proposed new ‘operational style’ Community Fisheries Control Agency, can it not be said that the fact that certain decisions will be addressed to (public authorities) of Member States will be ipso facto politically sensitive? Is the Commission not beginning the process of moving the institutional goal posts ever so incrementally beyond the parameters of what the Court laid down all those decades ago in the Meroni case? If this is so, has not the time come to give such regulatory agencies a clear Treaty base?** On the other hand, some will argue that what is involved in the construction of such control agencies is the ‘Europeanisation’ of the functions of the administrations of the Member States, rather than the

delegation of the powers of the Commission authorities as such.*° In that sense, such agencies go considerably further than the mandate in the new Financial Regulation*® given to the Commission to entrust tasks, when it shares administration to national public-sector bodies, to bodies governed by private law with a public service mission guaranteed by the State,*’as well as to purely private-sector entities with tasks involving technical expertise and administrative, preparatory or ancillary tasks involving neither the exercise of public authority nor the use of discretionary judgment.** The Council of Ministers as Principal In practice the Commission is not the single locus of executive power at the EU level that it clearly would like to be, but has seen its executive power diluted and constrained. At the same time what can be termed a parallel executive power developed in the shadows of the basic constituent texts. This stemmed, basically, from the fact that the Treaty of Maastricht (and its

successors), in setting up and giving substance to the EU as a novel international organisation, omitted to provide the possibility that the new executive power acquired by the Council (especially by the Presidency and in practice by the Secretariat General and its new Secretary General CFSP) could be delegated to the Commission and subjected to the straightjacket of ‘comitology’ in due course. As a result, what has happened over the course of the past ten years is that the Council, at precisely the same time as its legislative power developed (both as a result of its interaction with the European Parliament in co-decision and its independent legislative and policymaking power in the new policy areas and in particular with regard to justice and home affairs), developed its own original executive power in quite a substantial manner in the EU policy areas. In other words, the Council as an institution of the EU acquired and exercised a series of (executive) tasks that would normally have been entrusted to the Commission under the Community system. Thus the Presidency of the Council (aided considerably by the Secretariat General of the Council) is responsible for representing the Union

Delegation to

EU Non-majoritarian Agencies

97

in CFSP matters, implementing common measures and expressing the EU’s position in international organisations and at international conferences. The Treaty of Amsterdam strengthened this role by enabling the Council to instruct the Presidency to negotiate an international agreement on behalf of the EU and by entrusting the Presidency with the same tasks in JHA policy areas as in CESP. The Presidency likes to be perceived as the Council’s political ‘motor’: it drives things forward and has a policy planning role in setting the direction of the Council’s work and ensuring that it is completed within reasonable deadlines (often the end of a particular Presidency).*? On the other hand what is often underestimated in this explanation is the driving role played by the bureaucracy (permanent officials) and by the national civil servants who sit permanently in the Council’s underbelly of working parties, coordinating committees (article 36 EU), SCIFA and so on, in driving forward specific measures (within the broad parameters of the policy set by the Member State holding the Presidency at a given moment in time). The General Secretariat of the Council has evolved significantly during this time frame from having a passive notary/registrar role to a much more active role in assisting the Presidency, ‘not only in the application of procedures, but also in preparing for substantive negotiations; at the same time, the role of the Legal Service has become established and has developed to encompass intergovernmental conferences; in general, the six-monthly rotating Presidency with its increased role has made it more and more necessary to call upon the General Secretariat’s assistance in ensuring continuity and efficiency of work, by giving successive Presidencies the benefit of the experience it has accumulated over the years’.*° Political scientists are beginning to recognise that the General Secretariat of the Council has indeed a very pivotal role to play in behind-the-scenes outcomes and even has shifted outcomes closer to its own vision of the manner in which the EU has evolved on a number of occasions.*! The General Secretariat of the Council has incrementally assumed a role that parallels that played by the Commission administration in other (EC) policy areas. Since the Treaty of Maastricht, it has acquired specific executive tasks in practice that would normally have been delegated to the Commission in the Community system. For example, the development of significant aspects of CJHA policy making has been based on action plans, such as the Action Plan on Organized Crime*® or the Tampere Action Plan — this is closely linked to the emergence of a more autonomous role for the Council Secretariat. Thus the Secretariat has been charged with the task of implementing various recommendations contained in the High Level Report. In that context it has taken initiatives independently of the Member States themselves. More recently a number of Council instruments have conferred the Council (in practice its

98

Regulation through Agencies in the

EU

Secretariat General) with tasks concerning the evaluation of implementation measures, very similar to what is carried out by the Commission with regard to the classic EC areas. Moreover, the appointment of a CFSP High Representative (at the same time a Secretary General of the Council), provided for in the Treaty of Amsterdam, made this point particularly clear with the tasks including the framing, preparation and implementation of the EU’s foreign policy decisions. The Final Report of the European Convention’s Working Party on Justice and Home Affairs underscores the need for a clearer distinction between the Council acting in its legislative capacity and the Council exercising specific executive functions that specifically seems to turn on what is termed ‘operational collaboration’. This brought with it the creation of a more efficient structure for the coordination of operational cooperation at high technical level within the Council, namely the merging of various existing working groups and redefining the mission of the so-called “Article 36 Committee’, so that it would, in the future, focus on coordinating operational cooperation rather than becoming involved in the Council’s legislative work. The idea that this reformed structure would focus on the coordination and oversight of the entire spectrum of operational activity in police and security matters (inter alia police cooperation, fact-finding missions, facilitation of cooperation between Europol and Eurojust, peer review, civil protection) took root in the Draft Constitution as proposed to the IGC. As the CIE put it in their second report: ‘the distinction between formulating and implementing policy, namely between policy and operational matters is a falsely rigid one, difficult to define in principle and even more difficult to apply in practice’.* Europol is probably the best known example of an independent organ set up by the Member

States with links to the Council, whose tasks have been

incrementally expanded to cover those of a more operational nature such as its participation in so-called ‘Joint investigation teams’. It was established by mulltilateral Convention under the terms of the Maastricht Treaty. Europol has in recent years quite extensively exercised a power to negotiate and conclude international agreements with third countries and other international organisations. Eurojust was set up more recently and was established by Council decision, not requiring Member State ratification, precisely so its powers and tasks could be amended more easily with the passage of time. In the past months and years a number of new Agencies have been created by the Council and powers delegated to them. Unlike the Commission’s recent attempt to structure and rationalize its Agencies the Council seems to work on a purely ad hoc basis inventing a new or modified formula every time thus contributing considerably to the overall structural opacity. Thus, the European Agency for the Management of Operational Cooperation at the External Borders has been established by Council regulation.** In this case it can be

Delegation to

EU Non-majoritarian Agencies

99

argued that the Council did not delegate its own existing executive powers, but rather that the tasks in question had been largely exercised by the Member States and not the Council (or the Commission). At the same time, the Agency replaces in a sense much of the fragmentary and opaque structures that were attached to the Council in one form or another, for example the Common Unit for external border practitioners (SCIFA) that was a Council Working Party. Perhaps it is the latter fact that explains why, despite being proposed by the Commission, the institutional design of the Agency is very much leaning towards the Council as principal. In any event, it is rather clear that the Agency has been given (delegated) quite far-reaching powers, including operational powers and in the future ‘tasking the Agency with the carrying out of inspections at external borders and the facilitation of operational cooperation with relevant third countries and international organizations’ is explicitly envisaged (recital 19).

MOVING BEYOND AGENCY OVERSIGHT TO PRACTICES OF PUBLIC ACCOUNTABILITY? Conceptualising Public Accountability According to the classic delegation literature, principals engage in various forms of oversight of their agents. The literature on delegation identifies four major measures by which principals can contain agency losses,* namely contract design, screening and selection mechanisms, monitoring and reporting requirements, institutional checks (and balances). The former two are mechanisms by which principals seek to contain agency loss ex ante; the remaining mechanisms operate ex post. What is crucial here are the multifarious provisions of the secondary legislation establishing such agencies that exist alongside the formal principles of political accountability of ministers and commissioners to parliament (national or European), which is regulated at Treaty level. To what extent then can certain mechanisms be qualified as practices of public accountability on the ground as it were giving sustenance to a de facto increase in democratic control of EU governance? Moreover the manner in which the ‘practices’ of the Agency in question have evolved in their internal and external decision making is also highly relevant to this analysis. In an attempt to structure the discussion logically I stick to the four levels of analysis used in the classic literature on delegation, namely contract design, screening and selection mechanisms, monitoring and reporting requirements and institutional checks and balances.

100

Regulation through Agencies in the

EU

Contract design

In the case of a so-called ‘Community’ agency (the term used by the Commission and the EU) — in principle — the royal way is followed when a new agency is established. Thus the Commission makes a proposal and a decision is taken by the legislator, either the Council or in some cases in recent years the Council and the European Parliament jointly. It follows that the discussion, on the institutional design features of the proposed agency, on the whole takes place in any event on the basis of a publicly available proposal. For many agencies in the past the Council adopted the final text after extensive — secret — discussions in working parties with the European Parliament being given the opportunity to deliver an advisory opinion on the proposal as formulated initially by the Commission. In more recent years, however, the position of the European Parliament as colegislator has ensured much more openness in the discussions on institutional design. The dominant issue at the level of contract design is that of the legal basis in the Treaties. Depending on which legal basis is used, the process will be more or less open, or more or less closed. Thus, in the case of the European Agency for External Border Operational Cooperation, the choice of Article 66 TEC as the legal basis empowered the Commission to submit a proposal, the European Parliament to be consulted and the Council to take the decision. Thanks to the Register of Council Documents on the Internet, it is possible for an informed reader to keep abreast of (some of)* the discussions and changes made by the Council at Working Party and COREPER levels in the lead up to final decision making. In addition, due to the public nature of the Commission proposal and of the European Parliament’s committee deliberations, it is even possible for civil society, generally speaking, to follow and have some informal (political) input in the process if desired.*’ The contrast could not be greater with, for example, the establishment of Europol some eight years earlier at a time when no Internet Council register was available and when neither the European Parliament nor the national parliaments were informed of the latest state of discussions in advance of actual decision making. Of course, at that time, no Community legal base was available and the Member States had, under what was at that time the provisions of the ‘third pillar’ of the EU, to adopt an international public law convention that had to be ratified by all the Member States separately, in order to enter into force. Nonetheless the provisions and institutional design of Europol were discussed in secret in Council Working Parties etc. and agreed in a Council ministerial meeting with no role for any parliament, European or national, even in an advisory capacity. The institutional design of the agency and the inclusion or otherwise of ‘public’ moments of accountability for the exercise of future discretion will, to a significant extent, depend on who is taking the decision and according to what procedure. In practice what emerges is that (much) more ‘public’

Delegation to

EU Non-majoritarian Agencies

101

elements are built into the accountability mechanisms governing the agency and its operation when the European Parliament has a substantial role in the process.*® As the European Parliament’s legislative power increased it asserted a more powerful role in the oversight of existing agencies and in the design of new ones. For example, with regard to the new ‘executive’ agencies set up to assist the Commission in its management and other executive tasks, a Council regulation was adopted laying down the statute in advance for executive agencies to be (subsequently) entrusted with certain tasks in the management of Community programmes. This meant that a public debate could take place involving the European Parliament as to the overall contract design of such future executive agencies, as well as the state of the art public accountability provisions included. The idea of having a general framework instrument that in advance lays down the institutional design parameters in advance of the actual creation of more specific agencies seems to be taking root. At the end of 2003 the Commission produced a communication on the next stage, narhely the ‘operating framework for the European regulatory agencies’, with the intention being that it will produce a proposal for a statute for such regulatory agencies. This new trend is obviously preferable from the point of view of public deliberation and debate to the older practice of concocting, in a purely ad hoc fashion, the contract design of individual agencies, with no clear conceptual typology in mind. Another issue of contract design is whether either the principals or the legislature (either European or national) have the right to control changes made to the original contract design. If the original legal instrument is either a regulation of the Council or jointly of the Council and the European Parliament then, as a matter of European law, the procedure for amendments will follow the same procedure as that followed for the original legal instrument. With regard to EU agencies the situation is a little more complex. Take the example of Europol, originally established by means of an international Convention ratified by all the Member States as distinct from secondary EC or EU measures. Amendments to the Convention have to be ratified by the Member States in the same way as the original Convention, but the Council also has various quite far-reaching powers, acting unanimously, to extend Europol’s competences further and to modify the definitions of crime in the Annex to the Convention.” The European Parliament was not consulted at all on the initial Europol Convention or on any subsequent changes made to that Convention (by means of Protocols) until the entry into force of the Treaty of Amsterdam. The latter conferred the right on the European Parliament to be consulted on (future) Conventions and presumably all amendments to existing Conventions and all measures implementing Conventions. This meant, in practice, that the various

references in the Europol Convention to the Council’s power to adopt

102

Regulation through Agencies in the

EU

implementing measures henceforth entailed an obligation to consult the EP before those measures were adopted. This was considered as a significant broadening of the European Parliament’s right to, at the very least, express an opinion in public on the basis of public drafts before an implementing measure was formally adopted by the Council. In practice, however, it appears that the European Parliament has not been systematically consulted on all measures: on one calculation this happened in only 12 of the 17 relevant implementing measures.*? Moreover the European Parliament is still not consulted on financial decisions and also not on Europol Treaties with EU or non-EU bodies or States. The latter, in particular, is a significant omission, as Europol has, over the past five years, established an extensive network of Treaties exchanging information inter alia with third countries. Europol has exercised its power to negotiate and conclude international agreements with third countries and other international organisations quite extensively, most recently and controversially with the United States on the exchange of personal information. The ‘political control’ in such case is provided by the Council of Ministers, which must authorise the Director of Europol to conduct negotiations with the country or organisation in question; the results of such negotiations are then reported back to the Council and the Director of Europol is ‘authorised’ by the Council — behind closed doors and with no public information available nor provided to parliaments, either European or national — to proceed to sign the agreement as negotiated.*’ Screening and selection mechanisms As Table 5.1 shows virtually all the agencies selected make provision in the original contract design for the selection of the members of the Management Board as well as the Executive Director of the Agency in question. Thus, in rather typical fashion for those Agencies with the Commission as effective principal the European Environment Agency, provides* that the Executive Director be appointed by the Management Board ‘on a proposal by the Commission’ for a period of five years. It is clear therefore that it is the Commission itself that will undertake its own screening and selection processes in that regard and no further details or requirements are laid down. It recalls the selection provisions that apply with regard to the members of the Commission itself with the European Parliament being given the formal power of appointment in the Constitution for the EU, but on the basis of a name forwarded to it by the

European Council. Here too it is clear that the European Council exercises the real power (legal and political) with the European Parliament consigned to a rubber-stamping role (or not in the event it rejects the name put forward). In the early days of Agency design it was, in fact, the other way round: the Management Board had to submit a ‘list of candidates’ to the Commission who would then select and appoint the Executive Director.™

Delegation to

EU Non-majoritarian Agencies

103

A more modern example is provided in the terms of the European Aviation Safety Agency, which one suspects might become a model for future agency design. Here it is provided that the Executive Director of the agency ‘shall be appointed on grounds of merit and documented competence and experience relevant for civil aviation’ by the Management Board on the proposal of the Commission. The Executive Director may be similarly dismissed by the Management Board on the proposal of the Commission.*° Some criteria of competence and experience were thus laid down, on the basis of which a name should be proposed to the Management Board that then acts as the formal appointing authority. This was not, however, the first time, since, with regard

to the criteria for appointment to the Management Board of the European Monitoring Centre on Racism and Xenophobia (EUMC), it was (in 1997) provided very explicitly that they shall be ‘persons with appropriate experience in the field of human rights and analysis of racist, xenophobic and anti-Semitic phenomena’.*® With regard to the European Food Safety Authority, provision was made for a ‘list of candidates’ to be forwarded to the Management Board, which then presumably would be able to choose and appoint a specific candidate. The list of candidates was to be drawn up by the Commission on the basis of an open competition ‘following publication in the Official Journal and elsewhere of a call for expressions of interest’.*” Moreover before the Management Board could appoint a specific candidate it is noteworthy that it is to ‘be invited without delay to make a statement before the European Parliament and answer questions put by members of that institution’. With regard to the very recent European Agency for the Management of Operational Cooperation at the External Borders, the European Parliament went further and requested, on analogy with the practice that has developed regarding Commissioner designates, to hear the designate Executive Director prior to his appointment and to issue an opinion.** This involvement of the European Parliament in the selection and screening of candidates in a sense indicates a fairly recent trend to take account of the European Parliaments, increased status as co-legislature in the EU. The hypothesis is, moreover, that mention and involvement of the European Parliament in the selection process is directly related to the intensity of its role in the contract design (as co-legislator and thus co-principal, or only in an advisory capacity to the Commission and the Council). Thus, with regard to the European Food Safety Authority, with the European Parliament as colegislator we find rather advanced provisions (compared to earlier Agency designs)? involving the European Parliament in selection and screening issues. But even the earlier European Environment Agency, which was not adopted (in 1990) pursuant to co-decision by the European Parliament, included a provision enabling the European Parliament ‘to designate, as members of the management board, two scientific personalities particularly qualified in the

104

Regulation through Agencies in the

EU

field of environmental protection’. In later agency designs the European Parliament has often fought for this right to designate members of the Management Boards, alongside the Member States and the other EU institutions (Council and or Commission), sometimes successfully,®' other times not.” Very often it is moreover specified with regard to representatives of the European Parliament what specific expertise they must possess and the responsibilities they will exercise, whereas for the other representatives (from the Member States and from the Commission) no such specification is necessarily made.™ Most recently there is some indication that the European Parliament may well be changing tactic in this regard. Where it is involved in co-decision and the Agency is clearly being established according to the Community method, with the Commission as effective principal recent practice, indicates that it is not insisting on designating itself members to the Management Board. Where this is not the case, and where the agency is constructed in what it feels is a more inter-governmental fashion (with Council more in steering position, albeit falling under Community competence as such), the European Parliament attempts to shift the balance of power more towards the Commission as principal. Thus, with regard to the European Agency External Border Operational Cooperation the European Parliament has, for example, very deliberately suggested that it designate no members of the Management Board, in favour of achieving a more central role for the Commission as opposed to the Council (as principal) in the overall agency design. This is quite subtle in terms of approach, as the more overall involvement of the Commission, the more the European Parliament in any event exercises its general powers of political control over the actions and failure to act of the Commission. Be that as it may, this newer approach is in line with the sentiments recently expressed by the Commission to the effect that the European Parliament should not as colegislature be ‘represented’ by members on the Management Board, as this amounted to mixing up the legislative power with the executive power in an undesirable fashion; rather the European Parliament should see its powers of political control over Agencies increased.® This newer approach is confirmed in the statute for executive agencies, as well as the suggested operating framework for European regulatory agencies. However the quid pro quo in the Commission’s view at any rate is that the Commission should see its representation on the Boards considerably increased, to the point of parity with those of the Member States.” When it comes to those EU agencies established by the Council as principal, with little to no involvement of either the Commission or the European Parliament, a quite different picture emerges. Thus in the original agency design of Europol for example, not only was the European Parliament not formally consulted but no role was envisaged for it (nor indeed the

Delegation to

EU Non-majoritarian Agencies

105

Commission) in the designation of members of the Management Board nor with regard to the appointment of the Executive Director. The Director is appointed by the Council, acting unanimously and ‘after obtaining the opinion of the Management Board’.*’ Some discussion has taken place that representatives of the European Parliament should be included, at the very least as observers, in order to strengthen the political supervision of the activities of Europol, but this has not been effectuated in practice. A similar pattern emerges with regard to the later Eurojust,°* the EU Satellite Centre,” and the EU Institute for Security Studies” as well as the very recent European Defence Agency.”' With regard to the latter it is, for example, specifically stated that the EDA will act ‘under the Council’s authority’ and no provision whatsoever is made for any involvement by the European Parliament in the selection and screening mechanisms leading to the appointment of Agency members and Director. Monitoring and reporting requirements From the table of selected agencies (Table 5.1) it can be seen that provision is made for an Annual Report to be drawn up and sent to, at the very least, the Commission and/or Council; in a few more recent cases the European Parliament is also included explicitly in this list, in recognition, undoubtedly, of its expanded role, both in decision making and as the natural locus of political supervision of the Commission at any rate. This rule is included with regard to executive agencies in their general statute and is also envisaged with regard to the operating framework for regulatory agencies. This general rule seems to apply equally to the EU agencies (second and third pillar legal bases) although in the case of Europol in practice only an abrogated version is made public and sent to the European Parliament. In addition explicit provision may be made in certain Agencies for supplementary (scientific) reports to be made public, and the practice is to make them available in .pdf file on the respective web-sites.” Up until now it has not been the general practice for the European Parliament to debate the annual reports of the agencies nor to call in the Directors for questioning. It can, however, be observed that in the contract design with respect to some of the more recent agencies, the European Parliament is mandated to invite the Executive Director to report how the tasks of the Agency have been carried out. This is the case for the European Aviation Safety Agency” — since it only adopted its first Annual Report in 2004 it is not yet clear if and how the European Parliament will deal with its new-found powers. In the newly adopted European Agency External Borders Operational Cooperation” the European Parliament sought to push out the frontiers even further: thus it may request ‘at any time’ a hearing with the Executive Director on ‘any subject’ related to the Agency’s activities.”

106

AIOSIAPY uotuidg

Saye} QUON

lounog

:SULIOJIUOJA [eusjuy pue [eutoyxy

a

dANIOX| JOWOIIG BJGuyUNODOR ~—O} a

pue

dAIINIOX| | 109011 9]QeJUNODDe 0)

(Z)da

yy ga 4st] jo

qyy Aq

UOTSSTUIUIOD

uoneynsuos yoro Sj Woy Aq oy)

UOTSSTUOD

snd

pajutodde = ‘gy ouo lounog =u doquiow Aq

uonelnsay

jesodoig Aq UOTSsTUTUIOD JO AYNIOXY IOJOIIG [lounog (Q]) pue worsstuMW0D (Z)

jeuLojuy uoneynsuos Aq UOTSSTUUOD)

Arosiapy uoruldg

[Iounodg

uvedoing Sa [VUusIxyJIplog [euonessdg uoyneisdo00y

AIOAGoA sieak juopuodoput [BUIO}XxXO ‘UOTBN]RAD Sue}OUT yuNOSOR SMITA JO SIOP[OYOYLIS pue ope orjqnd

ylounog Are}0109¢ [erouay

JOJISISI[-ODQuON

jouonrnssy[lounod lounod = julof uonoy

vO0c

uvsdoing auajaq AouIs

vO0C

dATINIOX AIDA xs saead AloAqdaly sqedh ON IONIC] juopuodopur yuopuodopur DTQuJUNODOR 0} [BUI9}xd [euia}x9o aw uonenyead uonenyead

dAIINIOXY dAINIDXA {10}99s1q AW {4039911 AI saaryeyuasoidos soanejuasoida1 jo sj pue Jo sj pure UOTsstuTOD (Z)= uorsstumut0g (Z)

UOISSIWUIO7Z) UOISSTUIUOZ sainpad0i1gsosodoid asodoid

——-IOJLSIB9]-0D

jouonensey lounog gq pue

C007

uvodoing WOnRIAV Ayayes VW Aguas

‘uodo, A1ostapy pseog ON juoredsuen payynsuoo Jowid orqnd 0} SUTyeUI UOTJe]JNsuUOD =, UIRLIDD SUOISIOOp

uonusauog jo Joquiayy gq pue

CO0C

uvodoing uvodoing) pooy Ajayeg SUIDTIOg AyLIoyny (uonesIUes1Q Vy

jodoiny

9661

ylounog uOTR[NB2yyY 9u}

€661

uvodoing AQuasy doy uoneneagq SIUIDIIPITA]

uDLDLIOM Salauasy

uone}[Nsu0, uoNeyNsuoDd Aq JeIsadg dInpodoid sUON +=/SAIpP[OYIyeIS UOISSIWWOD poidope 4q AYQ190G WAWa juensind 0} sit sony Jo oINpsd01g

AIOSIAPY uoruldg

JUSUIDA[OAUT JO uvodoing qUIMILIL

0661

JO

ee

WMOTIAUY uO = AduIs

uvodoing

pounog uonensay

Po}BI1T)

Vy

eeeS

uOYvsajaq_ 0] ()J-UON

}0v.1U0DusIseg JUIUIN.ASUT

uons9]9¢ pue 3ulj0A

YIM JIAID

J AG

1BIX

arqvy, [6

UT

ssooon) 0}

Tpsune OH JO MOD jo s1OJIpHy uonoIpsiin jo e JNOD jo ausne

UTE

suonndo Jo YUeIOs oy

uoONwoON JO

ONTO

107

SOA

Aer

ON

ANIqeI

jenyoe.QUuoS-uoU yemoeyuos-uoUu

yenmornuos pure

ATUJO}

ON

Jeoddy Sa ON jenjoequos suorsioap JO pue -uou prvog Jo jeoddy jenjow.yuospuew LO} QN]Te} Ager 0} “OR OTIC] sjeodde Aq SW pus OF suonniysul

AUCIO}

0}

ON

ON

yuouedsuey,dA) $]809 Sounpaoold JO} WO [es9uos Suryey [eNplAIpulJo8pnq Jo a 9101) sounpoooid * 105 (da yeouar ssoooe UONRULIOJUL suolsAoid

YSTYy

SJUOWISSOSsE ponnugns ©} jrounos pur uorssrmwo0,) dq pue [luneAvw, AU DANNIONY 10,9911] O} WOddL FurAues JO ino uo sysey

IEpNFoy suodoau oO} jlounos

————

yuoWouINboY uonPoygng JO Jlounos OUUTITOIY JO YIOM orqnd suoruido sonssi IO] TUNUOD Jeo uoHeUNoOyU possouppe O} OY) souTapins =— PopwwAaAUoy 0} “UOISSTUNUO,) ysiy, 0} y “Aout = ‘lounoy JUSUaIINbOY Uo -UON uolsstwwo) pur EINO} YSTIQBISO Aouaiedsuny -easturwpe dq

OANNOONY LO} 0} proday uo WurAuwo [RIOUdH *sysu} JNO wodarsyd jenuue ayes WodaJO}

penuuy woday q4y pue jlouno, AeuuL, SAUL

SoA

yenjoenuos pue

AUCJO}

eUOFYON (SuoIstAoad

vONBOTQnNg JO suotuido pue ssagoR 0} uONRUNOyUI suoistaoid

ssaooy 0) uoneUoyut jo suotstAoid pue& [PA9] JO

penuuy jesouod penuuy Jesouod Woda ‘odo poevouge UOISIOAOPEL

SOK

yenuuy jesouod

wodar

[RUOHIPPY SyIYDQn,

Sunaodoy suOHEsY

ayqnd

pur

GO

oryqnd

108

Regulation through Agencies in the

EU

Article 34 of the Europol Convention makes provision for an annual report to the European Parliament on its activities, but this public non-confidential version has a more detailed confidential counterpart that goes to the Council of Ministers and remains secret. No formal right to request and exchange views on the Annual Report exists with either the Council Presidency or the Director of Europol. Moreover, there is no practice (or possibility according to the original Europol Convention) to request the Europol Director to appear before the competent committee of the European Parliament, although the European Parliament has in recent years put questions to the Presidency of the Council concerning the activities of Europol. In a subsequent protocol (which has not yet entered into force),”° provision has been made to enable the Director of Europol to appear before the European Parliament to discuss ‘general’ questions and only in order to assist the Council Presidency.” The European Parliament itself was highly critical of the proposed amendments pointing out that ‘the European Parliament will keep its current role with regard to Europol: discussing without anyone listening and making reports without anyone implementing’. In an analysis carried out under the auspices of the Council itself, the conclusion was reached that if the requisite political will was present, then the Council could involve the European Parliament much more in the political control of Europol’s activities, since formal Treaty change would be required with regard to a very few issues. Thus there was, in fact, nothing legally preventing the Council to voluntarily submit the reports it receives from the Joint Supervisory Board, to consult the EP on the budget of Europol and to consult the EP on Europol starting negotiations with third parties. In the Constitution Treaty (Article III-177) provision is made for the Europol Convention to be effectively replaced by legislation adopted by co-decision with the European Parliament. Moreover it is explicitly provided that ‘the procedure for scrutiny of Europol’s activities by the European Parliament, together with Member States” national parliaments’ will be provided for in co-decision with the European Parliament itself. If and when this provision enters into force and is implemented via legislative means (on a proposal from the Commission), a considerable step forward may well be taken. In terms of ongoing monitoring of the activities of Agencies quite far-reaching provisions have been agreed upon with regard to Executive Agencies. Thus a steering committee is established that consists of five members appointed by the Commission that has quite far-reaching tasks in terms of the operation of the Agency, and close structural links with the Commission itself and the Commission's programming. The idea is that the Commission will develop an infrastructure giving itself the capacity to play much more of a steering and monitoring role, both with regard to Steering Committees of Executive Agencies and also with regard to the supervisory boards of agencies in the future when Commission participation is increased.”

Delegation to EU Non-majoritarian Agencies

109

Finally it must be noted that, in many of the newer Agencies, provisions are included that mandate the holding of an independent external evaluation every three to five years depending on the Agency involved. This is an important improvement with regard to the earlier situation, and in some instances it is explicitly provided that the views of stakeholders and other interested parties will be taken into account and made public ((European Agency External Border Operational Cooperation). There is no practice with regard to the holding of such external evaluations yet, but the European Food Safety Agency is preparing to launch its first evaluation in 2005.” Institutional checks and balances The most significant additional checks and balances that are part of the surrounding membrane of Agency design are those relating to independent judicial control and independent financial control. In addition, provisions enabling theOmbudsman to play a role, applying the general EU provisions on access to information in this context, as well as enabling OLAF to play its role to the full in this context, are of importance. The idea behind providing such additional checks and balances into the system is to ensure that the Agencies, decentralised and autonomous as they might be, are tied into the overall Community (EU) system of legality and accountability. It is beyond the scope of this chapter to carry out a detailed study of these various additional mechanisms. In a }pacianinacy fashion the following general points can be noted. First, with regard to the issue of judicial control, the role of the Court of Justice is primordial, but until recently not well established. Thus, in the early Agencies the Court was only given an express role with regard to the contractual and non-contractual liability of the Agencies as such, thus obliging these Agencies to assume legal responsibility for acts attributable to them.” The lisnitation of judicial control in this manner probably had a lot to do with the fact that these carly agencies were perceived as purely information agencies without specific decision-making powers. Nevertheless, the European Monitoring Centre for Drugs and Drug Addiction (1993) also fell in this category of information provision, without decision-making power, but included a novel provision enabling actions to be brought against the Centre before the Court of Justice ‘under the conditions provided for in Article 173 of the Treaty’. This meant that both Community institutions (Commission, Council, European

Parliament) could challenge measures of the Centre as well as individuals who could fulfil the highly specific criteria of ‘direct and individual concern’.” Such provision extended the role of the Court of Justice to monitoring the legality of acts adopted by agencies that are intended to have binding legal effects.” Some of the later Agencies also included 2 wider access to the Court of Justice in order to challenge measures adopted by respective Agencies. Thus,

110

Regulation through Agencies in the EU

for example, the Office for the Harmonisation of the Internal Market (Trade Marks and Designs), set up in 1994, included even more far-reaching provisions enabling judicial control.** Two points were especially noteworthy in this regard. First this Agency (‘Office’) was the first Agency to be given explicitly regulatory tasks, albeit arguably not possessing decision-making powers as such.** Second, this was the first Agency to establish an internal ‘Board of Appeal’ system which enabled individuals to challenge decisions made by the Agency that adversely affected its interests and that could in the final analysis be appealed to the Court of Justice. This practice of a two-stage judicial system of appeals (one internal, although independent and the final appeal external) has been taken over with regard to a number of subsequent Agencies, especially where the tasks conferred on the Agency in question are regulatory in nature or involve decision-making, as such, by the Agency.*° With regard to the statute for Executive Agencies the Commission has been given rather extensive power to review any acts of an executive agency and to take measures itself of a judicial nature: Namely to suspend implementation of the measure of the Executive Agency in question, to prescribe interim measures in that regard or to prescribe amendment of the Agency measure in whole or in part, with the agency being placed under an obligation to comply with the Commission’s decision in that regard. In the final analysis an action for annulment can be brought to the Court of Justice to annul ‘the Commission’s explicit or implicit decision to reject the administrative appeal’.*’ With regard to the proposed operating framework for regulatory agencies, the Commission does not envisage a similar central role for itself in line with its thinking as to the more genuinely autonomous nature of such Agencies.** Rather, in the case of decision-making agencies, it is envisaged that interested third parties will be able to challenge the legality of decisions taken by the agency — possibly with an internal system of boards of appeal — before the Court of First Instance as the competent court in addition to providing recourse to Member States and the institutions to challenge any breaches to the general principles of Community law. Finally, with regard to agencies established under EU provisions as such (as opposed to the original Community system), it is noteworthy that the provisions regarding independent judicial control are in some respects weaker. Thus, the Europol Convention in its Article 40(1) provides that disputes between Member States on the ‘interpretation or application’ of the Convention first have to be discussed in the Council ‘with the aim of finding a settlement’. After six months, the Member States that are party to such a dispute, if they have not been able to end the dispute, are to agree among themselves how to settle it. The Convention itself does not include a provision guaranteeing that disputes about the Europol Convention will in the final analysis be sent to the Court of Justice for resolution, thereby clearly creating the possibility of deadlock.

Delegation to

EU Non-majoritarian Agencies

1%

Fourteen of the 15 Member States (at that time) however subsequently signed a declaration asserting that they would systematically send such disputes to the Court of Justice.*? In addition a large number of Member States were keen to ensure that national courts would be able to send references for a (so-called) preliminary ruling on the Europol Convention (but not on the validity of measures implementing the Convention) to the Court of Justice. A Protocol to the Europol Convention was ultimately adopted permitting Member States to decide for themselves whether or not to permit their own national courts to make such references to the Court of Justice.* In other words they could ‘optin’ to the system of judicial control at their discretion and they could further decide whether to restrict the power to refer to their final national courts or to enable it more widely. Despite this wide jurisdiction over references, no national court has yet sent any questions concerning the Convention to the Court. It should be noted that this Protocol to the Europol Convention, while less than what the situation is with regard to the rights and obligations of national courts under the general provisions of the EC Treaty, is stili much more than has ever been expressly included in the constitutive texts of other Agencies under the EC Treaties. With regard to the issue of independent financial control, the position is a bit varied. For those Community Agencies financed from the Community budget then many of the constituting acts provided for the normal financial procedures to apply. Thus, traditionally, in the early Agencies, the Commission would be sent an estimate of revenue and expenditure which would be forwarded to the Council with the preliminary draft budget of the EC.”' By the time the European Environment Agency was established in 1990, the Court of Auditors was given its normal role to examine the Agency’s revenue and expenditure.” This practice was continued in all the later Community Agencies. In addition that the normal financial rules were followed meant that the European Parliament very quickly obtained an instrument giving it some power over the Agencies via their budgets, since it retained the ultimate power of discharge over noncompulsory expenditure. This has enabled the Parliament to, in certain instances, reorient (or threaten to do so preventively) the budget of Agencies.” In addition the European Parliament used its role as co-legislator of the new Financial Regulation as a means of introducing much further reaching transparency in budgetary procedures and financial management in the case of all the so-called ‘decentralised’ Community Agencies. As from | January 2003, the constituent acts of all the existing Agencies were amended so as to provide the Financial Controller of the Commission

with a clear role. Moreover,

obligations were imposed on the respective Directors to provide at the request of the European Parliament any information required for the control of the implementation of the budget for the year in question. Only with regard to the

12

Regulation through Agencies in the

EU

statute for the ‘Executive Agencies’ was provision made, in certain circumstances,

for appropriations to the executive agency to be entered in the general budget of the EU ‘and shall be implemented by direct charging to that budget under the responsibility of the Commission’. This is explained by the Executive Agencies being in the unique position of, inter alia, being delegated tasks relating to the budget implementation of ‘operational appropriations for Community

programmes’.”* With regard to the non-Community (Union) Agencies, the position is a lot more varied. Europol, for example, is today still not financed through the Community budget but through Member State contributions. The financial controller is appointed by the Management Board of Europol itself and is supervised by the Management Board. The draft budget is forwarded to the Council and the European Parliament has no financial powers in the form of a discharge or even any scrutiny. An amending protocol to the Europol Convention in 2002® in addition to extending the Europol’s competence required that the Management Board’s five-year financing plan be forwarded to the European Parliament. With regard to the European Defence Agency different arrangements are made since its budget is financed exclusively from contributions of the participating Member States and it will receive no funding from the budget of the EC: In addition to judicial and financial control, a number of other institutional checks and balances are in existence. First, and appointed by the European Parliament, there is an independent Ombudsman who is empowered (and has certainly in practice interpreted his power in that fashion) to receive complaints of maladministration in the activities of the Community and Union Agencies. In the meantime, and irrespective of the formal treaty provisions, virtually all the decentralised bodies and agencies of the Union have adopted their own access of information provisions, including recently Europol,” the European Aviation Safety Agency”’ and the European Medicines Authorisation Agency, as well as the Ombudsman’s Code of Good Administrative behaviour, did adopt a Code of Administrative Behaviour.** Many of these Agencies make very extensive and exemplary use of the Internet. For example, the European Medicines Authorisation Agency makes publicly available, in an accessible fashion, not only its annual activity reports, its Management Board decisions, rules of procedure etc. but also the agenda and documents for the latter’s meetings as well as detailed European public assessment reports on various medicines. Finally it is striking how many of the decentralised Agencies have taken over the terms of an inter-institutional agreement between the Commission, the Council and the European Parliament on the European Anti-Fraud Office, OLAF, and specifically agree to cooperate with this office in the fight against fraud and corruption.”

Delegation to

EU Non-majoritarian Agencies

113

CONCLUSIONS The manner in which powers and tasks are being delegated in practice to a whole series of non-majoritarian

agencies, established and active at the

European level of governance, is a striking illustration of the development in practice of institutional structures outside the formal Treaty framework. Still today there is no legal basis in the constituent Treaties (nor the Constitution) enabling the structural establishment of such European-level Agencies. Yet in practice it is not just the Commission delegating functions to executive and regulatory agencies, as part of its ongoing re-focusing on its core (political) tasks, it is also (and increasingly) the Council delegating, or giving executive and operational-type tasks to European level agencies, that previously belonged to the Member States. At the same time there is a rich seam of legal and institutional practices surrounding the creation and functioning of such European level agencies. In particular, it seems the agencies may be breeding grounds and learning sites for state-of-the-art participatory practices that ultimately may be generalised across the wider spectrum of EU administration. For example, the Management Boards of two relatively new Agencies have themselves been proactive in developing participatory procedures for stakeholders prior to decision making (essentially a system of notice and comment), namely the European Aviation Safety Agency and the European Medicines Agency. This is streets ahead of the Commission’s own thinking and practice, despite the rhetoric surrounding the implementation of its White Paper on Governance and the inclusion of a formal provision in the European Constitutional Treaty on the place of more participatory forms of democracy in the EU. At the time when the European Constitutional Treaty adopts a horizontal approach to the EU, doing away with the divisive ‘pillars’ and introducing a single system of legal instruments and of decision-making procedures, the time has come to lift the veil on the Agencies and view in a horizontal fashion their activities, tasks and accountability provisions. In this view it can no longer make sense to ‘constitutionalise’ the Community administration — the same needs to happen with regard to the EU-wide Union administration. Moreover, in that context, it may be especially relevant to develop coherent thinking on an understanding of public accountability for action and inaction that exists alongside more vertical forms of accountability, via the mechanisms of representative democracy. This is all the more the case in a political system that is not grounded (and probably never will be) within a parliamentary system of governance, and where the distance between the evolving public administration (and its outposts) at the European level and its ‘subjects’, the citizens of the Member States of the EU, is particularly wide, and the structures of access and

voice both underdeveloped and mightily complex.

114

Regulation through Agencies in the

EU

NOTES | D. Robertson (2002), The Routledge Dictionary of Politics, London, Routledge, p. 3. 2 M. Strathern (2000), ‘Introduction: New Accountabilities’, in M. Strathern (ed.), Audit Cultures

Anthropological Studies in Accountability, Ethics and the Academy, London: Routledge, p. 1. 3 M. Bovens (2004), ‘Public Accountability’, in E. Ferlie, L. Lynne and C. Pollitt (eds), The

Oxford Handbook of Public Management, Oxford: Oxford University Press. 4 The account below follows D. Robertson (2002), The Routledge Dictionary of Politics, London: Routledge, p. 3. 5 M. Bovens (2004), supra note 3. 6 C. Harlow (2002), Accountability in the European Union, Oxford: OUP, and A. Arnull and D. Wincott (2002), (eds), Accountability and Legitimacy in the European Union, Oxford: OUP. 7 E. Fisher (2004), ‘The European Union in the Age of Accountability’, Oxford Journal of Legal Studies, vol. 24, 3, pp. 495-515, p. 496.

8 A. Moravesik, ‘In Defence of the Democratic Deficit: Reassessing Legitimacy in the European Union’, Journal of Common Market Studies, 40, pp. 603-624. 9 See further, M. Pollack (2003), The Engines of European Integration. Delegation, Agency and Agenda-Setting in the EU, Oxford. 10 P. Lindseth (1999), ‘Democratic Legitimacy and the Administrative Character of Supranationalism: the Example of the European Community’ Columbia Law Review, 99, p. 628. 11 See K. Lenaerts (1991), ‘Some Reflections on the Separation of Powers in the European Community’, Common Market Law Review, 28, p. 11.

12 See M. Everson (1995), ‘Independent Agencies: Hierarchy Beaters ?’, European Law Journal,

pp. 180-204, p. 181. 13 See Report of 28 January 1999 drawn up on behalf of Committee on Institutional Affairs on the decision-making process in the Council in an enlarged Europe, EP Doc A4-0049/99. 14 See the web site: http://www.europa.eu.int/index_en.htm. 15 See, K. Strom (2000), ‘Delegation and accountability in parliamentary democracies’, European Journal of Political Research, 37, p. 261. 16 M. Thatcher and A. Stone Sweet (2002), ‘Theory and Practice of Delegation to Non-Majoritarian Institutions’, West European Politics, 25, pp. 1-22, p. 2. 17 P. Craig (2003), ‘The Constitutionalization of the Community Administration’, Jean Monnet Working Paper 3/03, New York: NYU School of Law, p. 6, available at http:// www.jeanmonnetprogram.org/papers/03/030301 .pdf. 18 Thatcher and Sweet Stone (2002), supra note 16. 19 Strom (2000), supra note 15. 20 See Report by the Working Group “Establishing a Framework for Decision-Making Regulatory Agencies’ (Group 3a). Preparation of Commission White Paper on Governance, para. 8, available at http:/Avww.europa.eu.int/comm/governance/areas/group6/report_en.pdf. 21 Tbid. 22 Case 9/56, Meroni & Co., Industrie Metallurgiche, S.p.A. v. High Authority ofthe ECSC, E.C.R., 1957-58, p. 133 and Case 10/56, Meroni & Co., Industrie Metallurgiche, S.p.A. v. High Authority of the ECSC, E.C.R., 1957-58, p. 157. 3 See, for example, Ehlermann (1995), ‘Reflections on a European Cartel Office’, Common Market Law Review, pp. 471-486. 4 Thatcher and Sweet Stone (2002), supra note 16.

25 See, recently and very comprehensively, M. Pollack (2003), The Engines of European Integration. Delegation, Agency and Agenda-Setting in the EU, Oxford. 26 See Committee of Independent Experts (1999), Second Report on ‘Reform of the Commission — Analysis of current practice and proposals for tackling mismanagement, irregularities and fraud’, Volume I, para 2.1.37., at http://www.europarl.eu.int/experts/pdf/rep2-1en.pdf. 27 Council Regulation (EC) No. 58/2003 of 19 December 2002 laying down the statute for executive agencies to be entrusted with certain tasks in the management of Community programmes, O.J. L 11/1 of 16 January 2003, Art. 6 (1).

8 Council Regulation (EC) No. 58/2003, supra note 27, Preamble (7).

Delegation to

EU Non-majoritarian Agencies

143

29 In legal terms a framework regulation based on Art. 308 of the Treaty was adopted by the Council in 2002 Council Regulation (EC) No. 58/2003 supra note 27, The Commission resorts to it on a

case-by-case basis and establishes specific Executive Agency’s by Commission Decision. Commission Decision of 23 December 2003 setting up an executive agency, the ‘Intelligent Energy Executive Agency’, to manage Community action in the field of energy in application of Council Regulation (EC) No. 58/2003, O.J. L 5/85 of 9January 2004. 30 The White Paper on European Governance of 25 July 2001, COM (2001) 428 final, p. 24. 3! Communication from the Commission of 11 December 2002, The Operating framework for the European Regulatory Agencies, Com (2002) 718 final. Web-site: http://europa.eu.int/comm/ governance/docs/comm_agence_en.pdf. 32 These are the Office for the Harmonization in the Internal Market, the Community Plant Variety Office and the European Aviation Safety Agency. These were designed and set up in advance of the Commission’s more structured approach to regulatory agencies in the Community framework and each represent a more ad-hoc approach to institutional design than what the Commission has in mind for the future. 33 Proposal for a Council Regulation establishing a Community Fisheries Control Agency and amending Regulation (EC) No. 2847/93 establishing a control system applicable to the Common Fisheries Policy, (COM(04) 28 final). Prior to that (in 2002) the Commission proposed setting up a European Railway Agency which it also indicated would constitute a new-style regulatory agency but this time without the power to take individual decisions or discretionary regulatory powers. Communication from the Commission of 23 January 2002, Proposal for a Regulation of the European Parliament and of the Council establishing a European Railway Agency, COM (2002) 23 final. The

relevant committee of the European Parliament qualified it as ‘a technical tool in the hands of the Commission’. See this quotation in Report of 5 December 2002 on behalf of Committee on Regional Policy, Transport and Tourism on the proposal for a Regulation of the European Parliament and of the Council establishing a European Railway Agency (COM (2002) 23-C-5-0046/2002-202/0024 (COD), EP Doc A5-0441/2002, at p. 34. This was adopted in Regulation (EC) No. 881/2004 ofthe European Parliament and of the Council of 29 April 2004 establishing a European Railway Agency. 34 See too the analysis by D. Geradin and N. Petit, ‘The development of agencies at EU and national levels: conceptual analysis and proposals for reform’, Jean Monnet Working Paper 01/04. 35 See, for example, G. Majone and M. Everson (2001), ‘Institutional reform: independent agencies,

oversight, coordination and procedural control’, in O. de Schutter, N. Lebessis and J. Paterson, Governance in the European Union, Luxembourg Office for Official Publications of the European Communities (also published in French). 36 Council Regulation (EC, Euratom) No. 1605/2002 of 25 June 2002 on the Financial Regulation

applicable to the general budget of the European Communities, O.J. L 248 pp. 0001-0048 of 16 September 2002, Vol. 45. See further: P. Craig (2003), supra note 17.

37 Council Regulation 1605/2002, supra note 36, Art. 54 (2c). 38 Council Regulation 1605/2002, supra note 36, Art. 57 (2). 39 See Trumpf/Piris Report (1999), Operation of the Council with an enlarged Union in prospect — Report by the Working Party set up by the Secretary-General of the Council, CFSP Presidency

Statement Brussels. 40 Trumpf/Piris Report (1999), ibid. 41 See, for example, D. Beach (2004), ‘The Unseen Hand in the Treaty Reform Negotiations: the

Role and Influence of the Council Secretariat’, Journal of European Public Policy, pp. 408-439, and T. Christiansen (2003), ‘Out of the Shadows: The General Secretariat of the Council of Ministers’, R.M. van Schendelen and R. Scully (eds), The Unseen Hand: Unelected Legislators in the EU, London: Frank Cass, at p. 80.

42 Action Plan to combat organised crime adopted by the Council on 28 April 1997, O.J. C 251/1 of

15 August 1997 pp. 0001-0016. 43 See Committee of Independent Experts (1999), Second Report on ‘Reform of the Commission — Analysis of current practice and proposals for tackling mismanagement, irregularities and fraud’, Volume II, para. 7.9.1., available at http://www.europarl.eu.int/experts/pdf/rep2-2en.pdf. 44 Council Regulation (EC) No. 2007/2004 of 26 October 2004 establishing a European Agency for the Management of Operational Cooperation at the External Borders of the Member States of the European Union, of 2004 L 349/1, 25 November 2004.

116

Regulation through Agencies in the

EU

45 Kiewiet and Mc Cubbins and Strom (2000), supra note 15.

46 Some of the documents are not available or only partially available in .pdf file. 41 See, for example, Statewatch website: http://www.statewatch.org.

48 See, too, the analysis by D. Keleman (2005), ‘The Politics of “Eurocratic” Structure and the New European Agencies’, West European Politics, 25, pp. 93-118, who shows how inter-institutional politics have influenced Agency design in the EU context. 49 See further, for chapter and verse, S. Peers (2004), ‘Governance

and the Third Pillar: The

Accountability of Europol’, in D.M. Curtin and R.A. Wessel (eds), Good Governance and the European Union: Reflections on Concepts, Institutions and Substance, Antwerpen: Intersentia. 50 See Peers (2005), supra note 49.

5! Council of the European Union (2002). 52 Art. 9 of Council Regulation (EEC) 1210/90 of 7 May 1990 on the establishment of the European Environment Agency and the European environment information and observation network, O.J. L

120 of 11 April 1990. 53 See Art. 26 ‘The President of the European Commission’ of the Draft Treaty establishing a Constitution for Europe, Title IV: The Union’s Institutions, Chapter 1: The Institutional Framework, O.J. C 169 of 18 July 2003. 54 See, for example, Art. 6 of Regulation (EEC) No. 337/75 of the Council of 10 February 1975 establishing a European Centre for the Development of Vocational Training, O.J. L039 of 13 February OWS: 55 Tn virtually all the earlier Agencies no provision was made for dismissal. 56 See Art. 8 (1) of Council Regulation (EC) No. 1035/97 of 2 June 1997 establishing a European Monitoring Centre on Racism and Xenophobia, O.J. L 151 of 10 June 1997. 57 Fora similar approach see Art. 31 of Proposal for a Council Regulation establishing a Community Fisheries Control Agency and amending Regulation (EC) No. 2847/93 establishing a control system applicable to the Common Fisheries Policy, 28 April 2004, COM (2004) 289 final.

58 See Report of 24 February 2004 drawn up on behalf of Committee on Citizen’s Freedom and Rights, Justice and Home Affairs on the Proposal of Council regulation establishing a European Agency for the Management of Operational Co-operation at the External Borders (COM (2003)

687-C5-0613/2003-2003/0273(CNS)), EP Doc- A5-0093/2004 FINAL, Amendment 37. 59 See, for example, the agency design of the European Foundation for the Improvement of Living and Working Conditions, Regulation (EEC) No. 1365/75 of the Council of 26 May 1975, O.J. L 139 of 30 May 1975 with the European Parliament only being mentioned in its budgetary capacity. 60 For a similar provision with regard to the European Monitoring Centre for Drugs and Dug Addiction (EMCDDA), see Art. 8 of Council Regulation (EEC) No. 302/93 of 8 February 1993, O.J. L 36 of 12 February 2003. 61 See, for example, with regard to the European Agency for the Evaluation of Medicinal Products (EMEA), Art. 56 of Council Regulation (EEC) No. 2309/93 of22 July 1993, O.J.L214 of24 August 2003. See, too, the European monitoring Centre on Racism and Xenophobia (EUMC), Art. 8 of Council Regulation (EC) No. 1053/97 of 2 June 1997, O.J. L 151 of 10 June 1997. 62 For example, the Office for Harmonisation in the Internal Market (Trade Marks and Designs),

Council Regulation (EC) No. 40/94 of 20 December 1993, O.J. L 11 of 14 January 2004. See, too, the European Agency for Safety and Health at Work, Council Regulation (EC) No. 2062/94 of 18 July 1994, O.J. L 216 of 20 August 1994, Art. 4 of the European Agency for Reconstruction, Council Regulation (EC) No. 2667/2000 of 5December 2000, O.J. L 306 of 7December, Art. 11 of Regulation (EC) No. 1406/2002 of the European Parliament and of the Council of 27 June establishing a European Maritime Safety Agency, O.J. L 208 of 5 August 2002 and Art. 25 of Regulation (EC) No. 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, O.J. L 240 of 7 September 2002. 63 For example the case of the EMEA, supra note 61. 64 See Report of 24 February 2004 drawn on behalf of Committee on Citizens’ Freedoms and Rights, Justice and Home Affairs on the proposal for a Council regulation establishing a European Agency for the Management of Operational Co-operation at the External Borders, supra note 58. 6° See Communication from the Commission, ‘The operating framework for the European Regulatory Agencies’, COM (2002) 718 final.

Delegation to

EU Non-majoritarian Agencies

HZ

6° See Communication, ibid. at 9. The European Parliament in its report on the communication sees this as one option but in fact comes down in favour more of the model already used in the European Food Safety Authority, namely ‘to entrust the Commission with the task of drawing up a comprehensive list of candidates on which Parliament may express its opinion and on the basis of which the Council makes the appointments except for one member directly appointed by the Commission’. See Report of 4 December 2003 drawn up on behalf of the Committee of Constitutional Affairs on the communication from the Commission ‘The operating framework for the European regulatory

agencies’ (COM (2002) 718-2003/2089(INI)), EP Doc AS-0471/2003 FINAL, at para. 17 (d). 67 Art. 29 (1) of Council Act of 26 July 1995 drawing up the Convention based on Art. K.3 of the Treaty on European Union, on the establishment of a European Police Office (Europol Convention), O.J. C 316 of 27 November 1995 pp. 0002-0032.

68 Council Decision 187/JHA of 28 February 2002 setting up Eurojust with a view to reinforcing the fight against serous crime, O.J. L 63 of 6 March 2002. 6 Council Joint Action 555/CFSP of 20 July 2001 on the Establishment of a European Union Satellite Centre, O.J. L 200 of25 July 2001. 7 Council Joint Action 554/CFSP of 20 July 2001 on the Establishment of a European Union Institute for Security Studies, O.J. L 200 of 25 July 2001. 7 Council Joint Action 551/CFSP of 12 July 2004 on the Establishment of the European Defence Agency, O.J. L 245 of 17 July 2004. 72 See, for example, the annual safety reports made available to the public by the European Aviation Safety Agency and the publication by the European Environment Agency of the scientific committee reports: See, for example, the opinion of scientific committee on Global Monitoring on Environment and Security, available in .pdf at http://org.eea.eu.int/organisation/sciecomm.html/SC-files/ SC_opinion_on_GMES-final_adopted_6 Oct_2004.pdf. 73 See Art. 29 of Regulation (EC) No. 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, O.J. 240/1 of 7 September 2002. 7 Jn Draft Council Regulation establishing a European Agency for the Management of Operational Cooperation at the External Borders of the Member States of the European Union issued on 30 April

2004, Council Doc. 90158/04, Art. 25 reads ‘Functions and powers of the Executive Director’, para 2 states: ‘The European Parliament or the Council may invite the Executive Director of the Agency to report on the carrying out of his/her tasks’. 75 See Report of 24 February 2004 drawn up on behalf of Committee on Citizen’s Freedom and Rights, Justice and Home Affairs on the Proposal of Council regulation establishing a European Agency for the Management of Operational Co-operation at the External Borders, supra note 58, Amendment 40.

76 Initiative of the Kingdom of Denmark with a view to adopting a Council Act drawing up, on the basis of Art. 43(1) of the Convention on the Establishment of a European Police Office (Europol Convention), a Protocol amending that Convention, O.J. 2002 C 172/15 of 15 July 2002. 77 See further, Peers, supra note 49. 78 See Report of 26 March 2003 drawn up on behalf of Committee on Citizen’s Freedom and Rights, Justice and Home Affairs: 1. on the initiative by the Kingdom of Denmark with a view to

adopting a Council Act amending the Council Act of 3 November 1998 adopting Rules on the confidentiality of Europol information (13875/2002 — CS-0553/2002 — 2002/0823(CNS)); 2. on the initiative of the Kingdom of Denmark with a view to adopting a Council Act amending the Staff Regulations applicable to Europol employees (13873/2002 — C5-0555/2002 — 2002/0822(CNS)); 3. on the initiative of the Hellenic Republic with a view to the adoption of a Council Decision adjusting the basic salaries and allowances applicable to Europol staff (63 14/2003 — CS-0066/2003 — 2003/0806(CNS)), EP Doc. A5-0107/2003, at p. 11. 79 See, too, Report by the Working Group, ‘Establishing a Framework for Decision-Making Regulatory Agencies’, White Paper on Governance Preparation, June 2001, at 4.

80 See the 16th Management Board meeting of 14 September, available at http://www.efsa.eu.int/ mboard/mb_meetings/606/mb16_doc0S_evaluation_en1.pdf. 81 Thus, among others, Art. 21, European Foundation for the Improvement of Living and Working Conditions; Art. 17 of the European Centre for the Development of Vocational Training and Art. 18 of the European Environment Agency.

118

Regulation through Agencies in the

EU

82 See, too, Art. 15 of Council Regulation (EC) No. 1035/97, supra note 56. 83 See, too, Art. 22 of Council Regulation (EC) No. 58/2003, supra note 27.

84 Art. 63 of Council Regulation (EC) No. 40/94 of 20 December 1993 on the Community trade mark, O.J. L 11 of 14 January 1994.

85 See the analysis made in Report by the Working Group, ‘Establishing a Framework for DecisionMaking Regulatory Agencies’, June 2001, at 7. 86 For example, Art. 73 of Council Regulation (EC) No. 2506/95 of 25 October 1995 amending Regulation (EC) No. 2100/94 on Community plant variety rights, O.J. L 258 of 28 October 1995 Community Plant Variety Office (regulatory tasks and decision-making power) and Art. 41 of Council Regulation (EC) No. 1592/2002, supra note 62 (regulatory tasks and decision-making power). 87 Art. 22 (5) of Council Regulation (EC) No. 58/2003, supra note 27. 88 As opposed to the European Parliament which does envisage a similar role for the Commission regarding administrative complaints in the first instance against acts of regulatory agencies, similar

to that played with regard to executive agencies, see Report of 4 December 2003 drawn up on behalf of the Committee of Constitutional Affairs on the communication from the Commission ‘The operating framework for the European regulatory agencies’, supra note 66, at para. 18. 89 Convention based on Art. K.3 of the Treaty on European Union, on the establishment of a European Police Office (Europol Convention), O.J. 1995 C 316/32 of 27 November 1995. The exception being the United Kingdom which fought vigorously to prevent the Court of Justice being given any role with regard to the Europol Convention or measures subsequently adopted. 90 Council Act of 23 July 1996 drawing up, on the basis of Art. K.3 of the Treaty on European Union, the Protocol on the interpretation, by way of preliminary rulings, by the Court of Justice of the European Communities of the Convention on the establishment of a European Police Office, O.J. 1996 C 299/1 of 9 October 1996. The Protocol entered into force on 29 December 1998 and 14 of the 15 Member States at that time gave the Court of Justice jurisdiction and 11 of these 14 decide to permit all their respective national courts and tribunals to send questions to the Court of Justice. See further Peers (2004), supra note 49. 91 Art. 11 of Council Regulation 337/75, supra note 54. 92 Art. 13 of Council Regulation (EEC) No. 1210/90, supra note 52. °3 See Brinkhorst Laurens Jan (1996), ‘The future of European agencies: a budgetary perspective from the European Parliament’, in The New European Agencies, EUI Working Paper RSC, No. 96/ 49, edited by A. Kreher, Florence: EUI, pp. 75-81. 94 See Art. 16 (1) of Council Regulation (EC) No. 58/2003, supra note 27. 9 Jnitiative of the Kingdom of Denmark with a view to adopting a Council Act drawing up, on the basis of Art. 43(1) of the Convention on the Establishment of a European Police Office (Europol Convention), of a Protocol amending that Convention, O.J. 2002 C 172/15 of 15 July 2002 Convention on the establishment of a European Police Office (Europol Convention) and the Protocol on the privileges and immunities of Europol. °6 See Public Access to Europol Documents, whereas the Council Decision of 20 December 1993 on public access to Council documents is applying to Europol, available at http://www.europol.eu.int/ legal/other/files/PublicAccessToEuropolDocuments. pdf. °7 See Decision No. 1/2004 of the Management Board concerning the arrangements to be applied by the Agency for the public access to documents, EASA MB1/04 of 3 February 2004 available at http://www.easa.eu.int/doc/About_EASA/Manag_Board/2004/mb_decision_0104.pdf. °8 See, for example, the European Safety Food Authority, i.e. the Decision of the Management Board on EFSA Code of Good Administrative Behaviour of 16 September 2006 available at http://www.efsa.eu.int/mboard/statutory_texts/internal_rules/409/mboard_meeting 010 doc13 adopted_en11.pdf; the European Environmental Agency, i.e. the Decision of the EEA Management Board on Code of Administrative Behaviour of 22 June 2004, available at http://org.eea.eu.int/ documents/administrativedocuments/mbdecisionCodeofgoodbehavior.html; the European Foundation for the improvement of living and working conditions, i.e. the Decision on Code of Administrative Behaviour of 11 February 2000, available at http://www.eurofound.eu.int/about/ code.htm; the European Medicines Agency, i.e. the Decision of EMEA Administrative Unit on EMEA Code of Administrative Behaviour, Doc. Ref. EMEA/D/37674/99

of 3 December

available at http://www.emea.eu.int/pdfs/general/admin/Conduct/3767499EN pdf.

1999,

Delegation to

EU Non-majoritarian Agencies

119

9 See, for example, Europol, i.e. Administrative Arrangement between EUROPOL and OLAF of 8 April 2004, available at http://europa.eu.int/comm/anti_fraud/press_room/pr/2004/09.pdf; Eurojust, Memorandum of Understanding between Eurojust and OLAF of 14 April 2003, available at http:// europa.eu.int/comm/anti_fraud/press_room/pr/2003/memo_en.pdf; European Maritime Safety Agency. The Administrative Board of 9 December 2003 has approved the Draft Interinstitutional Collaboration Agreement between EMSA and OLAF; European Agency for the Evaluation of Medicinal Products. Decision No. 15007/99 of the Executive Director of | June 1999 concerning the terms and conditions for internal investigation in relation tot the prevention of fraud, corruption and any illegal activity detrimental to the Communities’ interest, available at http://www.emea.eu.int/ pdfs/general/direct/decision/1500799EN.pdf; European Food Safety Authority. The Administrative Board of 11 December 2002 adopted the Decision concerning the terms and conditions for internal investigation in relation to the prevention of fraud, corruption and any illegal activity detrimental to the Communities’ interest, available at http://www.efsa.eu.int/mboard/statutory_texts/internal_rules/ 409/decision_olaf mb 06 en1l2.pdf; Office for the Harmonization of Internal Market. Decision No. ADM-99-24 of the President of OHIM of 31 May 1999 concerning the terms and conditions for internal investigations in relation to the prevention of fraud, corruption and any illegal activity liable to give rise to administrative or criminal proceedings, available at http://oami.eu.int/EN/ office/aspects/decisions/24-99.htm.

6.

Independence, Accountability and Transparency of European Regulatory Agencies Ellen Vos

INTRODUCTION The creation of independent agencies within the organisational structure of government is attractive for a number of reasons. One of these reasons is that agencies may contribute to greater independence, transparency and accountability.' In the case of the European Union, it has been argued that the carrying out of clearly defined tasks by clearly identifiable agencies, instead of some obscure divisions of the European Commission or equally opaque committees, could, in principle, create greater transparency at EU level. Importantly, agencies would also provide greater consistency in implementing policies, because they operate at arm’s length from the political institutions.’ Faced with the findings of the Committee of Independent Experts, that inter alia revealed that it was difficult to find in the Commission persons who had ‘even the slightest sense of responsibility’,?the Commission highlighted in its reform plans of 2000 the importance of delegation and decentralisation of dayto-day executive tasks. At the same time, it advocated the need of an open government and accountability, to be built on new forms of partnerships between the different levels of European governance.* These views were largely in line with the ideas expressed by Dehousse and Majone. In view of the increasing politicisation of the European integration process and the lack of both competences and resources of the Commission to adopt measures to ensure uniform implementation of EU policies, they pleaded for a greater role of autonomous agencies.° These agencies could carry out specific administrative tasks, which would leave the Commission greater room to concentrate on its core tasks and policy priorities.® 120

Independence, Accountability and Transparency

121

Functional decentralisation to clearly identifiable agencies could furthermore be advantageous in several respects. Agencies could for example encourage uniform interpretation and implementation of Community law, where they form the nucleus of networks of national authorities.’ They could advance cooperation between national officials in charge of the implementation of Community policies, requiring that actors agree on the definition of a common problem and the range of possible responses, which is determined by their access to comparable data and expert opinions.‘ They could also further administrative integration,’ while playing an important role in the international arena. The creation of agency networks involving all interested parties could in principle also contribute to creating a better understanding and public confidence in EU action. Yet, resorting to independent agencies has not remained without criticism. Concerns have particularly been expressed about delegating decision-making powers to agencies.'° In the EU context, this has been accompanied by the worry about the institutional balance of powers and the appropriate legal basis for such delegation.'' In addition, delegation of greater decisional powers to agencies gives rise to fears that they are easily influenced and subverted to the ends of those whom they are supposed to regulate.'? Closely linked with this are the legitimacy concerns relating to the place of agencies within the system of separation of powers, and to the carrying out of governmental tasks by agencies that do not have a constitutional basis and are not subject to constitutional guarantees. Equally, the independence of agencies would not be sufficiently enough ensured under the system of public scrutiny.'? For instance, resorting to agencies could arguably lead to more fragmentation and in turn to erosion of accountability as the institutional complexity would lead to confusion as to which institution is accountable for which issue.'* Moreover, agencies’ transparencies and political accountabilities are perceived as rather doubtful. Critics point to the problematic disentanglement of expert findings from political strategies.'° Resorting to agencies may also be problematic in terms of judicial review and public involvement. While in the Netherlands currently some of the above-mentioned arguments are ventilated against independent authorities,'® the interest for agencies in the European context has only augmented.'’ The 2001 White Paper on European Governance pointed out several advantages of agencies.’* In its Communication on Better Lawmaking, the Commission placed the use of regulatory agencies in the broader context of the exercise of the executive function and definition of the responsibilities of the institutions.'’ Very recently, reflecting upon the policy challenges and budgetary means of the enlarged EU for the years 2007-2013, the Commission explicitly considered both regulatory and executive agencies of importance within the context of what it calls the principles for administrative governance: less direct management, better control of delivery and greater costeffectiveness.” Moreover, contrary to the current treaties, the Constitutional

122

Regulation through Agencies in the

EU

Treaty even explicitly refers to agencies*' Several proposals submitted by the Commission to create agencies are now awaiting approval from the Council and European Parliament.” Hence, seen within the context of the need for reform of the Commission and the general strive of the Community institutions for better lawmaking based on principles of good governance, it is not surpnsing that in the new millennium, the resort to, and creation of, European agencies is more popular than ever before. Having briefly touched upon the terminology and concept of European regulatory agency, this chapter examines how the European institutions have shaped independence, accountability and transparency of European regulatory agencies.

CONCEPT OF EUROPEAN REGULATORY AGENCY Currently there is no official definition of agency adopted in EU legislanon. The EU website defines agencies as bodies governed by European public law that are distinct from the Community institutions and have their own legal personality.** In its Communication on the Operating Framework for the European Regulatory Agencies (hereafter, “Communication on the Operating Framework’), the Commission highlights the formal characteristics: they are created by a regulation, they have clearly specified tasks, they have legal personality and have a certain degree of administrative and financial autonomy In this manner, an agency is ‘required to be actively involved in exercising the executive function by enacting instruments which contmbute to regulating a specific sector’.> Accordingly, other independent bodies of the Community, that are created by the Treaty itself, such as the European Central Bank and the Investment Bank, fall outside the scope of the concept of agency. Although the Treaty on the European Union refers in the second pillar to the creation of

Europol and Eurojust.*” the Commission is of the opinion that both bodies™ as well as the European Union Institute for Security Studies> and the European Union Satellite Centre® fall under the general concept of agency" On the basis of a functional classification, the existing European agencies can roughly be divided into: i) agencies that collect information and coordinate: ii) agencies that manage and administer programmes; iii) agencies that in some way are involved in regulation and decision making The latter agencies are in some way involved in regulating economic and social policies, such as the preparation and adoption of the regulatory framework: preparation and adoption of legislative acts for implementing the regulatory framework: and monitoring implementation of the regulatory and legislative framework. In its Communication on the Operating Framework, the Commission adheres to a two-fold classification: executive and regulatory agencies. Executive agencies

Independence, Accountability and Transparency

123

are entrusted, according to the Commission, purely managerial tasks and hence assist the Commission in implementing the Community’s financial support programmes. They are strictly supervised by the Commission. Regulatory agencies are actively involved in ‘the executive function by enacting instruments, which help to regulate a specific sector’.*? The latter agencies can in turn be subdivided into agencies with decision-making powers and agencies without such powers. In order to promote a single terminology, we will here stick to the two-fold typology as adhered to by the Commission.

INDEPENDENT

FROM WHOM?

Independence has been viewed by the Commission in its White Paper on Administrative Reform of 2000 as one of the key principles of a European public administration.*° This principle was further defined in the Code of Good Administrative Behaviour. This Code was adopted by the Commission in 2000, in order to ensure that the right to good administration, now laid down in the Constitutional Treaty,*® would be respected.*’ The Code thus stipulates that the staff of the Commission ‘shall act independently within the framework of the policy fixed by the Commission and their conduct shall never be guided by personal or national interest or political pressure’.** Independence is generally considered to be free of both political and industry interests.*’ In the Community context this also refers to national interest. Not surprisingly in its White Paper on European Governance the Commission views that agencies too should operate with ‘a degree of independence’.*’ In its Communication on the Operating Framework the Commission adds: [i]t is particularly important that they should have genuine autonomy in their internal organisation and functioning if their contribution is to be effective and credible. The independence of their technical and/or scientific assessments is, in fact, their real

raison d’étre. The main advantage of using the agencies is that their decisions are based on purely technical evaluations of very high quality and are not influenced by political or contingent considerations.*!

How far do agencies really have ‘genuine’ autonomy in their international organisation and functioning? As regards their organisational structure, it can be observed that Member States, the Commission and the Parliament can to a certain extent, through their representatives on the Management or

Administrative boards, keep an eye on the agencies’ activities. Management or Administrative boards are in fact the steering bodies of the agencies and generally are responsible for budgetary matters, the appointment of the executive directors and the monitoring of the agencies’ performance. The reverse side of this possibility of control is certainly that the independence of the agencies may be

124

Regulation through Agencies in the

EU

endangered.*? In its Communication on the Operating Framework the Commission attempts therefore both to lessen Member States’ control over agencies and to reduce the size of the Management boards. In this context it envisages an Administrative board of 15 members, including six representatives appointed by the Commission and six representatives of the national executives by the Council and three members, with no voting rights, representing the interested parties.“ It seems very unlikely that this proposal will be ever accepted by the Council and Member States. The removal of the role of Member State representatives from the Administrative board has been successful only with respect to one agency: the EFSA (the European Food Safety Authority). The EFSA Management Board thus includes one Commission representative plus 14 members appointed by the Council in consultation with the European Parliament from a list drawn up by the Commission. Four of the latter members have a background in organisations representing consumers and other interests in the food chain.** The readiness to have a more independent Management Board of the EFSA should be understood against the backdrop of the crisis of confidence in Community regulation, its scientific advice, and the food industry in general, calling for dramatically enhancing credibility and removing any inch of influence whatsoever. In all other cases the Council has refused to accept the Commission’s proposal on this point and has insisted on including one representative per country. The Commission is however right in arguing that in order for agencies to function effectively, smaller Administrative boards should be created which

‘reflect the executive at Community level while taking account of the expertise of the Member State executives’.*° This is the more so with regard to the enlargement. However, just as in the case of comitology,*’ the Member States should be recognised as partners, and participation of national representatives to the Community executive could be considered as transnational governance or a kind of ‘shared administration’.** The problem of large Administrative boards could be addressed in various ways, varying from division into working groups and annual plenary meetings together with the use of multimedia,’ to the solution that has been adopted with regard to the Commission itself in the new Constitutional Treaty: a reduced number of members (two-thirds of the Member States), selected on the basis of a system of equal rotation between the Member States while the composition of the boards needs to satisfactorily reflect the demographic and geographical range of all the Member States.©° As regards their financial situation, agencies are largely dependent on subsidies by the Community. The new Council Financial Regulation expressly refers to agencies. It introduces increased control of the agencies, determining that decisions imposing executive tasks on, inter alia, agencies must include all appropriate arrangements for ensuring the transparency of their activities. This means that agencies must make use of transparent, non-discriminatory

Independence, Accountability and Transparency

125

procurement and grant-award procedures; have an effective internal control system for management operations; and correct accounting arrangements.*! Importantly, the Regulation submits agencies to the control of the Commission’s internal auditor.** Discharge for the implementation of agency budgets is given by the Parliament on the recommendation of the Council. The Financial Regulation moreover stipulates that (future) executive agencies must conduct regular checks to ensure that activities to be financed from the budget have been implemented correctly. They must prevent irregularities and fraud.*4 Article 185 of the new Financial Regulation also expressly authorises the Commission to adopt a framework financial regulation for agencies receiving grants charged to the budget.*°

ACCOUNTABILITY The more use is made of agencies, the more important become the design of mechanisms for keeping agencies under control and making them accountable. Agencies must therefore be subject to an effective system of supervision and control,*° whereby the regulatory mandates may constantly be reviewed.*’ Can regulatory agencies be autonomous and yet be under control? At first sight this may seem contradictory but regulatory designs demonstrate that this is possible and necessary.** In its White Paper on European Governance, the Commission explains that the principle of accountability requires that: [rJoles in the legislative and executive processes need to be clearer. Each of the EU Institutions must explain and take responsibility for what it does in Europe. But there is also a need for greater clarity and responsibility from Member States and all those involved in developing and implementing EU policy at whatever level.°?

Above we already noted that agencies are from an organisational and financial perspective not completely independent. The Commission views that a specific provision must be created in order to ensure that the activities of regulatory agencies are consistent with the Community’s executive function. The Commission does not stop in underlining that it has the ultimate responsibility for the execution of Community law. This would imply that the Commission would have to exercise its role as guardian of Community law and be ensured that its overall responsibility for implementation of the Community budget is not infringed upon by acts of agencies.” In general, the agencies’ independent administrative structures facilitates their control. Because of their visibility, agencies may attract greater attention and fuel policy debate in their field. Their actions are likely to be identified with Community action and as such they will be held responsible.*' Yet to whom are European agencies accountable and for what?” In our discussion, particular

126

Regulation through Agencies in the

EU

focus will be put on the political or democratic accountability. This issue is currently being hotly debated within the Dutch context.

POLITICAL ACCOUNTABILITY In EU context the principle of political accountability to the European Parliament is generally accepted.* Does this also apply to regulatory agencies? Can regulatory agencies be independent and at the same time under control?° A recent study of the role of independent agencies in several national systems reveals that these agencies seem to reduce parliamentary control and eliminate certain fields from parliamentary supervision.®’ At the national level, direct parliamentary control over agencies has only just begun to be a topic of debate. In this respect, the EU is more progressive. The European Parliament has been given various instruments of control by means of the membership on the Management Board and, importantly, by means of its budgetary power. As regards the general practice of agencies having representatives of the European Parliament as members on the Administrative or supervisory board of agencies, it may be questioned whether this is a true means of control. In the practice of the EMEA

Management Board, for example, there seemed to be little or no

feedback among the experts appointed by the Parliament (generally university professors) and the Parliament itself. The Agency Working Group preparing the Governance White Paper proposed in 2000 to exclude representation by the European Parliament as its participation would arguably raise conflicts with the Parliament’s constitutional role as controller of the executive.” This line of thinking was copied by the Commission who proposed to limit membership of supervisory bodies. Participation of the Parliament in supervisory boards, which are responsible for approving the agency’s budget, so it was argued, could create an odd situation as the Parliament has also general budgetary powers.” The Parliament itself does not consider direct participation in the board indispensable for the exercise of political scrutiny as long as there is a kind of structure in which small administrative boards, such as the one of EFSA, are accompanied

by an advisory committee composed of all the Member States.”! The importance of the Parliament’s budgetary power should not be underestimated as many agencies depend partly or entirely for their revenues on Community subsidies. Being non-compulsory expenditure, these subsidies are finally determined by the Parliament, which can use its powers, in a certain sense, to adjust the budget.” Parliament’s readiness to invoke this power ‘in the name of transparency and accountability’ has been underlined several times.” The new Council Financial Regulation expressly confirms increased control of the agencies, conferring upon the Parliament the power of discharge to agencies for the implementation of their budgets.”

Independence, Accountability and Transparency

127

In addition to these two existing means of control, the Commission proposes an innovative structure of political supervision by the Parliament and the Council over the regulatory agencies.’° Here the Commission thinks of enabling these institutions to hear the agency’s executive director, being the agency’s representative, at any time, in particular upon publication of the agency’s annual report. These hearings could serve to verify whether the agency adheres to its mandate and whether its action is in line with Community policies and policy guidelines. Another ‘soft’ instrument of parliamentary control could be the power to appoint and dismiss the executive director of the current agencies.’° In view of the lack of a meaningful notion of responsibility and accountability within the Commission,”’ it could be considered essential that agencies are governed by directors who would have the support of the Parliament.’”* Today this power is for the decision-making agencies (the Office for Harmonisation in the Internal Market; the Community Plant Variety Office; the European Aviation Safety Agency) entrusted to the Council; for the non-decision-making agencies this power lays in the hands of the Management boards of the agencies. It should be carefully considered whether the Parliament too should have the right of taking appropriate measures in case of malfunctioning or mismanagement, with ultimately the possibility to dismiss the director. In its Communication on the Operating Framework, the Commission denies this and views that, holding the executive function, it should have itself the power to appoint and dismiss the executive director of decision-making agencies on the basis of a list of candidates prepared by the Administrative board. In the case of the executive agencies, appointment should be made by the Administrative board on the basis ofa list of candidates put forward by the Commission.” The Commission views that in any case, formal appointment of candidates for the post of the executive director would be made dependent on a hearing before the Parliament. The Commission seems herewith to link to concepts of ministerial responsibility. Yet, it can be maintained that, in the absence of a fully-fledged ministerial

responsibility as developed in various countries, it seems indispensable to establish true and political accountability of agencies. Under these circumstances, there would arguably be room for the Parliament to exercise also this kind of control, independent of the control exerted by the Commission.* In view of the fact that most agencies, both at national and European level, are created in order to operate at an arm’s length of the political arena, such a power of the Parliament may nevertheless not seem to be appropriate. In addition, whereas technical or scientific expertise is a primary criterion for agencies and their directors, there is a risk that Parliament would put more importance to other criteria.*' Empirical work on the relationship between elected politicians and independent regulatory authorities reveals that party politicisation does not exclude expertise, but ‘at the very least, appointing individuals with clear party

128

Regulation through Agencies in the

EU

links reduces the public distance between the [agencies] and partisan politics’. An important finding of this study is that there is no evidence that politicians used their powers of appointment to select their own party members (except for in Italy). Two explanations for this are largely put forward: either politicians have found means of control other than the formal controls to limit the agencies’ independence or politicians have chosen to leave the agencies considerable independence.* Clearly this highlights that more research on this topic is necessary. It may hence be wondered whether it is appropriate to develop a concept of ministerial responsibility** for agencies in relation to EU commissioners.** The Dutch discussion on regulatory authorities teaches us that even in systems where longstanding traditions of ministerial responsibility exits, such a responsibility is limited to the nomination of members

of the board, to the issuance of

instructions or to the approval of the budget, while it is restricted to the powers that the minister has in the steering of the authority. No responsibility exists for the decisions that the authorities adopt. Such a ministerial responsibility may easily turn out to be a hollow one, or at least a very unclear responsibility in the everyday practice.** For example, the Dutch minister of finances, G. Zalm, protested recently that he was held responsible by the Dutch Parliament for independent authorities for which he considered himself not to be responsible.*’ In its attempt to strengthen its control over the agencies, the Commission sees a role for itself in the approval of the annual reports, the budget and the financial control (as laid down in the financial regulation, to be performed by the internal auditor of the Commission). It nevertheless denies for itself a role of legal supervision, in which it would be empowered to give agencies instructions or require them to withdraw certain decisions.** Taken together this would arguably support the plea to look for new forms of accountability of agencies.

OTHER FORMS OF ACCOUNTABILITY: TOWARDS A SYSTEM OF ‘HORIZONTAL RESPONSIBILITY’ OR ‘PUBLIC ACCOUNTABILITY’? Over the years, several mechanisms to keep agencies under control have been developed. In the Dutch context, for example, we may witness in addition to the system of ‘vertical’ ministerial responsibility, the rise of several forms of, what is called in the circles of public administration, a ‘horizontal responsibility’, which attaches high importance to the relations with other organisations and the public or clients.*° In 2002 for example, several independent authorities adopted a so-called ‘charter on public responsibility’,°° while more recently a code of good administrative behaviour was adopted.*' Also in the EU we may

Independence, Accountability and Transparency

129

observe a growing interest in arrangements of what has also been called ‘public accountability’.°? On request of the European Ombudsman, all existing agencies, for example, have, just as the Commission,” adopted a code of conduct.” This practice is now consolidated in the new Constitutional Treaty that submits agencies to the administrative supervision of the European

Ombudsman.” Also arrangements as regards participation of interested parties and the public in general are forcefully promoted by the institutions. In its White Paper on European Governance the Commission views that ‘improved participation is likely to create more confidence in the end result and in the institutions which deliver policies’.** It thus has adopted a general strategy ona stronger dialogue with civil society.”’ In addition, all existing agencies have created networks by means of which they structure their relationship with the ‘external world’, with their ‘stakeholders’. Through these networks, the agencies build on the work of the existing institutions and collaborate with them. The creation of agency networks involving all interested parties could hence in principle contribute to ‘a Europe closer to the citizen’ and foster better understanding and public confidence in EU action. Several steps have already been taken to involve interested parties in the activities of the agencies.** The networks often promote interaction with both governmental actors and all other actors concerned in a specific policy area, although the practical operation of these networks and their legitimacy as European governance structures still needs scrutiny. This kind of accountability comes close to what Colin Scott has coined ‘redundancy accountability’. With this model he refers to the tendency to extend traditional mechanisms of accountability by the horizontal mechanisms of the market, in which overlapping (and superfluous) accountability mechanisms reduce the centrality of any one of them.” It may be wondered however whether this model could actually provide accountability in a system of transnational governance. '

TRANSPARENCY AND ACCESS TO DOCUMENTS The Court of First Instance has repeatedly recalled that transparency is essential for the citizens to monitor the exercise of the powers by the Community institutions!”! and to ‘strengthen the democratic character of the institutions and the trust of the public in the administration’.'°* Transparency is therefore considered as a prerequisite for accountability or sometimes even as part of accountability. For, if activities of public authorities are obscure or largely unknown to the public, the public cannot hold these authorities accountable for their actions.'® Transparency has therefore, for many years, ranked highly on

the political agenda. It was introduced by the Treaty of Amsterdam’ and has

130

Regulation through Agencies in the

EU

been proclaimed by the Commission in its White Paper on European Governance as one of the principles of good governance.'® Unsurprisingly, the European Convention has also put great weight on transparency and has further developed it. Article I-50 of the new Constitution now affirms the ‘telos’ of the concept of transparency: the function of transparency is ‘to promote good governance and ensure the participation of civil society’ .'°° To this end, ‘the Union Institutions, bodies and agencies shall conduct their work as openly as possible’. Curtin’s plea to extend the general obligation to provide extensive access to their documents to agencies, as well as the other institutions and bodies not referred to in Article 255 of the EC Treaty,'”’ has been accepted by the new Constitutional Treaty, which stipulates that the right of access to document also applies to agencies’ documents.'” Article III-399 of the new Constitution requires that agencies too will recognise the importance of transparency in their work and that they will lay down in their rules of procedure the specific provisions for public access to documents.'° The new Constitution also requires agencies, like the Community institutions, to adhere to principles of sound administration, such as the duty to state the reasons for decisions taken, the interested parties’ right to be heard and to receive an answer to reasoned questions and provide for time limits.'° The principle of transparency and access to documents are also here explicitly included. The procedure followed by agencies must guarantee transparency and access for all parties concerned with deliberations and the decision-making process. The operating framework as proposed by the Commission emphasises these requirements. As regards the access to documents, it should be noted, that although the right of access to documents laid down in the current Treaty does not cover documents by the agencies, in practice these rules have been extended by the Community to also cover agency documents. Encouraged by the Code of Conduct on Access to Documents adopted by the Council and the Commission and the Recommendation of the European Ombudsman,'"! most agencies have adopted decisions on access to their documents.'!” These decisions are all based on the Commission’s rules on access to documents. In practice, several agencies seem very open and accessible, yet the degree of transparency largely depends on the agency itself. The regulations governing the setting up of third-generation agencies already now include a provision on access to documents. Both EMSA and EASA are subjected to the general Access to Documents Regulation applicable to the Community institutions.''’ EFSA must “ensure wide access to the documents which it possesses’, leaving it to its Management Board to adopt provisions governing access to documents ‘taking full account of the general principles and conditions on access to documents governing the Community institutions’.''4 Nearly all agencies (in particular, the European Medicines Agency; the European Environment Agency, the Office

Independence, Accountability and Transparency

131

for Harmonisation in the Internal Market and the European Monitoring Centre for Drugs and Drug Addiction) allow access to many other documents via their websites. The Medicines Agency seems the most progressive in this and its website is impressive because of the amount of detailed information, including inter alia, its annual activity reports, its Management Board (press releases), its committees (calendar of meetings, press releases), and the addresses of the national authorities competent in the area of pharmaceuticals, standard operational procedures, guidelines and, amid some worry of the pharmaceutical industry, detailed European public assessment reports.''> EFSA too is committed to the principle of transparency in its founding regulation and must make public, without delay: agendas and minutes of the Scientific Committee and the Scientific Panels; the opinions of the Scientific Committee and the Scientific Panels immediately after adoption, minority opinions always being included; the information on which its opinions are based; the annual declarations of interest made by members of the Management Board, the Executive Director,

members of the Advisory Forum and members of the Scientific Committee and Scientific Panels, as well as the declarations of interest made in relation to items on the agendas of meetings; the results of its scientific studies; annual reports of its activities; and requests from the European Parliament, the Commission or

a Member State for scientific opinions.''® Access to documents seems generally still to be problematic in view of the many exceptions to the right to access.''’ Although the Constitutional Treaty reinforces the right of access to documents by improving the wording of the provision laying down this right, several issues, such as the precise legal nature of the right of access to documents, remain unclear. Access to agency meetings is more problematic. Whilst Article I-50 of the Constitutional Treaty explicitly stipulates that also the Council meets in public when considering and voting on a draft legislative act,''* this does not apply to the Commission, organs or agencies. They have though, as indicated, to work as openly as possible. Hence, contrary to the practice of the American ‘sunshine committees’,'!? agencies and their organs or committees do generally not meet in public. Only the Management Board of the EFSA holds, as a general rule,'”° its meetings in public and may authorise consumer representatives or other interested parties to observe the proceedings of some of its activities.'*! This illustrates that transparency in the EU context is limited. Opening up all meetings of agencies and their board and committees may however raise new dilemmas as it may lead to endless discussions of viewpoints and, in case of the agencies dealing with scientific matters, to further politicisation of science.'” In practice, the institutions attempt to address this dilemma by allowing some sort of participation of the ‘stakeholders’ and/or the general public. In the new ‘General Food Law’ for example, the principle of transparency sees actually to public consultation and information in the framework of the new farm to fork

132

Regulation through Agencies in the EU

approach. It stipulates thus that during the preparation, evaluation and revision of food law, ‘there shall be open and transparent public consultation, directly or through representative bodies’, ‘except where the urgency of the matter does not allow it’.!”?

CONCLUSIONS Agencies are mushrooming in the new millennium as they have an important role to play within the reform and modernisation process of the Commission. They are the expression of the desire to allow the Commission to become a more political administration and let agencies carry out technical tasks at operate at arm’s length from the political institutions. This underlines the need for independence, accountability and transparency of European regulatory agencies. In order to ensure more uniformity of the agencies the Commission proposed in 2002 to establish a general framework with rules and principles that are applicable to regulatory agencies, dealing explicitly with their independence, accountability and transparency. It confirms the restricted role that agencies can play within the process of modernisation by giving them strictly circumscribed tasks. We observed that both from a financial and organisational perspective, agencies are not completely made independent by the institutions. Financial ‘dependence’ on the Community and the Community’s new financial regulation ensures that agencies include all appropriate arrangements for ensuring the transparency of their activities. Organisational ‘dependence’ sees to the influence that Member States may still exercise by means of their membership of the Administrative boards of agencies. The attempt of the Commission to remove the national representatives from these boards has, with the exception of the EFSA, been unsuccessful. Here a choice must be

made between the lessening of Member State influence on the board and the recognition of the transnational character of governance and agencies. Not surprisingly, the Commission emphasises that it has the final responsibility for the executive function and as such it has also a responsibility towards agencies. While regulatory agencies generally respond to the Community’s demand for greater efficiency and expertise, political accountability will nevertheless remain of high importance. Therefore, the proposal to improve parliamentary control by means of inter alia hearings, annual reports, is to be applauded. In view of the experiences at the national level, however, it seems appropriate to explore alternative forms of accountability, such as horizontal responsibility or public accountability. In this context the agencies’ transparency becomes of even more importance.

Independence, Accountability and Transparency

133

NOTES ' See e.g. R. Baldwin and C. McCrudden (1987), Regulation and Public Law, London: Weidenfeld and Nicolson, pp. 4-7; R. Baldwin (1995), Rules and Government, Oxford: Clarendon Press and P. Craig (1999), ‘The Nature of the Community: Integration, Democracy and Legitimacy’, in: P. Craig and G. De Burca (eds), The Evolution of EU Law, Oxford: Oxford University, pp. 45-50. 2 R. Dehousse and G. Majone (1999), Reforming European Governance: Options for the New Commission, Centre Européen, Porte d’Europe, Paris, on file with the author. 3 See the Committee of Independent Experts in its First Report on ‘Allegations regarding Fraud, Mismanagement and Nepotism in the European Commission’ of 15 March 1999, para. 9.4.25, available at: http://www.europarl.eu.int/experts/report!_en.htm. + See Communication from the Commission, to the European Parliament, the Council, the Economic

and Social Committee and the Committee of Regions, Strategic Objectives 2000-2005, Shaping the New Europe, COM (2000) 154 final.

5 R. Dehousse and G. Majone, supra note 2. & G. Majone (1996), Regulating Europe, London: Routledge. 7 R. Dehousse (1997), ‘Regulation by Networks in the European Community: the Role of European Agencies’, 4 (2) Journal of European Public Policy, 246-261.

8 Tbid. ° A. Kreher (1997), ‘Agencies in the European Community — a step towards administrative integration in Europe’, Journal of European Public Policy, 4 (2), 238. 10 A. Heringa and L. Verhey (2003), “Independent Agencies and Politicai Control’, in L. Verhey and T. Zwart (eds), Agencies in European and Comparative Law, Antwerp: Intersentia Publishing, pp. 155-169. tL See K. Lenaerts (1993), ‘Regulating the regulatory process: “delegation of powers” in the European Community’, European Law Review, 18, 42. !2 See e.g. R. Baldwin and C. McCrudden (1987), Regulation and Public Law, London: Weidenfeld

and Nicolson, pp. 9-10. 13 See M. Everson (1995), ‘Independent Agencies: Hierarchy Beaters?’, European Law Journal, 1

(2), 180-204. '4 On 3 October 2000, during a plenary session of the European Parliament, Romano Prodi warned that a serious risk of fragmentation would be incurred if the creation of agencies would be facilitated on which the Council can then confer executive powers. There would according to Prodi be a real danger that this will create conflicting centres of power. Speech by R. Prodi before the European Parliament, 3 October 2002, SPEECH/00/352, see http://europa.eu.int/comm/commissioners/prodi/ speeches/index_en.htm. 1S M. Shapiro (1997), ‘The problems of independent agencies in the United States and the European Union’, Journal of European Public Policy, 4 (2), 276-291. 16 On the basis of the conclusion of a report entitled: Rapport Kohnstamm, Een herkenbare staat: investeren in de overheid, see: http://www.andereoverheid.nl/.

See inter alia Nrc Handelsblad,

‘Een quango is als een kind dat op kamers gaat’, 12 July 2004. 17 See in general E. Chiti (2000), ‘The Emergence of a Community Administration: The Case of European Agencies’, Common Market Law Review, 37 (2), 309-343; M. Everson, supra note 13; S.

Frank (2003), in the EU and Commission: 18 See White

A new model for European Medical Device Regulation, A comparative legal analysis the USA, Groningen: Europa Law Publishing; E. Vos (2000), ‘Reforming the European What role to play for EU Agencies?’, Common Market Law Review, 37 (5), 1113-1134. Paper on European Governance, COM (2001) 428 final, 24; see already White Paper

‘Reforming the Commission’, COM (2000) 200 final. 19 See Communication from the Commission, ‘European Governance: Better Lawmaking’, COM

(2002) 275 final. 20 See Communication from the Commission, ‘Building our Common Future: Policy challenges and Budgetary means of an Enlarged Union 2007-2013’, COM (2004) 101 final, annex 1.

21 See e.g. Articles I-49 to I-51, II-102 to I-103, I-111, I-112, I1-335, II-365, I1-367 to III-369 and III-395 to III-401 of the final text of the Constitutional Treaty, 29 October 2004, CIG 87/2/04,

REV 2.

134

Regulation through Agencies in the

EU

22 See e.g. European Centre for Disease Prevention and Control (ECDPC), COM (2003) 441 final; European Chemicals Agency (ECA), COM (2003) 664 final and Community Fisheries Control Agency (CFCA), COM (2004) 289 final. 23 See http://europa.eu.int/agencies/index_en.htm. This links to the conceptualisation of agency by the Legal Service of the European Commission, see SEC (2001) 340. This document is referred to by the report drafted by the Working Group ‘establishing a framework for decision-making regulatory agencies’ (group 3a) in preparation of the White Paper on European Governance, work area 3: Improving the exercise of executive responsibilities (June 2001). 24 See Communication from the Commission, ‘The operating framework for the European Regulatory Agencies’, COM (2002) 718 final, 3.

25 Tbid., 4. 26 These organs will therefore fall beyond the scope of this chapter. It should be considered though, whether these organs should not be considered as agencies too, See Resolution of the European Parliament on the Commission’s Communication on the Operating Framework, T5-0015/2004, and the draft-resolution, AS5-0471/2003.

27 28 L 29

Namely Article 29 of the EU Treaty. Convention of 26.07.95, O.J. C 316 of27November 1995 and resp. Decision of 28.02.02, O.J. 63 of 6March 2002. Council Joint Action of 20 July 2001 on the establishment of a European Union Institute for

Security Studies O.J. L 200, p. 1.

30 Council Joint Action of 20 July 2001 on the establishment of a European Union Satellite Centre, O.J. L 200, p. 5. 31 See Communication from the Commission, ‘The operating framework for the European Regulatory Agencies’, supra note 24, 3. 32 See E. Vos, ‘Agencies and the European Union’, in L. Verhey and T. Zwart (eds), supra note 10, 119-121. X.A. Yataganas (2001), ‘Delegation of Regulatory Authority in the European Union. The Relevance of the American Model’, Jean Monnet Working Paper, 3/01.

33 See Communication from the Commission, ‘The operating framework for the European Regulatory Agencies’, supra note 24, 4. 34 Tn its Communication on the operating framework on Regulatory Agencies, supra note 24, the Commission confusingly subdivides regulatory agencies into executive agencies and agencies with

decision-making powers. During the Conference ‘Which Regulatory Authorities in Europe?’ Brussels, 18-19 March 2004,

a Commission official declared that this was a mistake. See very

critical D. Geradin and N. Petit (2004), ‘The Development of Agencies at EU and National Levels: Conceptual Analysis and Proposals for Reform’, Jean Monnet

Working Papers,

1/04, 47; and

D. Geradin in this volume. 35 See White Paper ‘Reforming the Commission’, supra note 18. 36 Article II-101 of the Constitutional Treaty, supra note 21. See for an extensive discussion of the charter, the special issue of the Maastricht Journal, Maastricht Journal, 8 (1), 2001. 37 On 1 November 2000, the code came into force. See: http://europa.eu.int/comm/secretariat general/code/_docs/code_en.pdf. m 38 This is presented as a guideline for good behaviour in the code. 39 D. Geradin and N. Petit, supra note 34, 50. 40 See White Paper on European Governance, supra note 18, 24. 41 See Communication from the Commission, ‘The operating framework for the European

Regulatory Agencies’, supra note 24, 5. 42 See below, para. 4.

43 See M. Everson, supra note 13, 183; see also D. Geradin in this volume. 44 The commission fails to indicate who should appoint the latter members. 4S Article 25 of European Parliament and Council Regulation No. 178/2002 laying down the general principles and requirements of food law, establishing Safety Authority and laying down procedures in matters of food safety, O.J. 2002, 1. See Council Decision of 15 July 2002 appointing the members of the of the European Food Safety Authority, O.J.

of 28 January 2002 the European Food L 31 of 1 February Management Board

C 179 of 27 July 2002, 9.

4° Communication on the operating framework on Regulatory Agencies, supra note 34, 9.

47 See E. Vos (2004), ‘The Fall of Committees?’, in Jaap W. de Zwaan, Jan H. Jans and Frans A.

Independence, Accountability and Transparency

135

Nelissen (eds), European Union — An Ongoing Process of Integration, Cambridge University Press,

pp. 111-129. 48 See P. Craig (2003), ‘The constitutionalisation of Community administration’, European Law

Review, 28 (6), 25. 4° As proposed for comitology see E. Vos (2005), ‘The Role of Comitology in European Governance’, in D. Curtin and R. Wessel (eds), Good Governance in the European Union: Lessons

from National and International Law, Antwerp: Intersentia Publishing, pp. 107-124. 50 Article I-26 (6), supra note 21.

5! Article 48 of Council Regulation No. 1605/2002 of 25 June 2002 on the Financial Regulation applicable to the general budget of the European Communities, O.J. L 248 of 16 September 2002, 1.

52 53 54 55

Tbid., Article 185 (3). Tbid., Article 185 (2). Tbid., Article 54 (3). See Commission Regulation (EC, Euratom) No. 2343/2002 of 23 December 2002 on the

framework Financial Regulation for the bodies referred to in Article 185 of Council Regulation (EC, Euratom) No. 1605/2002 on the Financial Regulation applicable to the general budget of the European Communities, O.J. L 357 of31 December 2002, 72.

56 See White Paper on European Governance, supra note 18, 24.

57 See M. Everson in this volume. 58 See D. Geradin in this volume. ‘ 59 See White Paper on European Governance, supra note 18, 10. Harlow criticised this description of accountability, as it seemingly perceives accountability more as ‘responsiveness’: C. Harlow, Accountability in the European Union, Oxford: Oxford University Press 2002, p. 185. 60 See Communication from the Commission, Regulatory Agencies’, supra note 24, 14.

‘The operating framework for the European

61 See R. Dehousse, supra note 7, 258. 62 Colin Scott has identified that the questions who, to whom and for what are central in the debates

on accountability, see C. Scott (2000), ‘Accountability in the Regulatory State’, Journal of Law and Society, 27 (1), 38-60. 63 See for a discussion of the various meanings of accountability, C. Harlow, supra note 56, 23. See

for a discussion of various types of accountability of agencies, D. Geradin and N. Petit, supra note 34 and D. Geradin in this volume. 64 See Rapport Kohnstamm, supra note 16.

65 C. Harlow, supra note 56, 182; D. Geradin and N. Petit, supra note 34, 55. 66 M. Everson, supra note 13, 183. See also M. Everson in this volume. 67 A. Heringa and L. Verhey, supra note 10, 164. 68 See as regards France, e.g. A. Heringa, ‘Agencies in France. Autorités Administratives Centrales

Indépendantes/ National Independent Administrative Authorities’, in L. Verhey and T. Zwart (eds), supra note 10, 56; S. van Thiel (2003), Sturen op afstand: over de aansturing van verzelfstandigde organisaties door kerndepartementen. Management in Overheidsorganisaties, available at: http:// ep.eur.nl/retrieve/1507/BSK073.pdf. 69 See Agency Working Group, supra note 23. 70 See Communication from the Commission, ‘The operating framework for the European Regulatory Agencies’, supra note 24, 9. This is in line with the proposal submitted by the Agency Working Group, supra note 23. 71 Resolution of the European Parliament, supra note 26, para 17 (e-g). 72 See L. Brinkhorst, ‘The future of European Agencies: A budgetary perspective from the European Parliament’, in Kreher (ed.), The New Agencies. Conference Report, [EUI Working Paper RSC 96/ 49] (Florence 1996), 77.

73 Tbid. 74 Article 48 of Council Regulation No 1605/2002, supra note 51. 75 See Communication from the Commission, ‘The operating framework for the European Regulatory Agencies’, supra note 24, 13; see already Agency Working Group, supra note 23, 20. 76 M. Everson, supra note 13, 199. 77 See V. Mehde (2003), ‘Responsibility and accountability in the European Commission’, Common Market Law Review, 40 (2), 442.

136

Regulation through Agencies in the

EU

78 A. Heringa and L. Verhey, supra note 10, 165. 79 See Communication from the Commission, ‘The operating framework for the European Regulatory Agencies’, supra note 24, 13; see already White Paper on European Governance, supra note 18, 13; this is in line with the proposal made by the report drafted by the Agency Working Group, supra note 23.

80 81 82 83 84 85

A. Heringa, supra note 68. A. Heringa and L. Verhey, supra note 10, 165. M. Tatcher in this volume. Tbid. See in general V. Mehde, supra note 77. At this moment, only the President of the Commission can request an individual commissioner

to resign on the basis of Article 217 (4) of the EC Treaty.

86 Rapport Kohnstamm, supra note 16. 87 NRC Handelsblad, supra note 16.

88 See Communication from the Commission, Regulatory Agencies’, supra note 24, 14.

‘The operating framework for the European

89 §. van Thiel, supra note 68, 22. See also, for instance, C. Grave, M.V. Flinders and S. van Thiel

(1999), ‘Quangos: what’s in a name? Defining quangos from a comparative perspective’, Governance, an International Journal of Policy and Administration, 12 (1), 129-146.

90 http://www.publiekverantwoorden.nl. 91 http://www.publiekverantwoorden.nl/profiel/code_goed_bestuur.pdf. 92 See also, D. Curtin in this volume.

93 See above, para. 3. 94 Special report from the European Ombudsman to the European Parliament following the own initiative inquiry into public access to documents, O.J. C 44 of 10 February 1998, 9. See for example, the code of conduct of EMEA: http://www.emea.eu.int/pdfs/general/admin/Conduct/ 3767499EN. pdf. 95 Article II-103, supra note 21. 96 See White Paper on European Governance, supra note 18, 7. 97 See Communication from the Commission, ‘Towards a reinforced culture of consultation and dialogue — General principles and minimum standards for consultation of interested parties by the Commission’, COM (2002) 704 final. 98 See E. Vos (2004), Overcoming the Crisis of Confidence: Risk Regulation in an Enlarged European Union, Inaugural lecture, Maastricht: Unigraphic, 19-20. 99 C. Scott, supra note 62, 52. 100 C. Harlow, supra note 56, 184.

101 See inter alia, Interporc Im — und Export GmbH v. Commission (‘Interporc IT’) Case T-92/98, E.C.R., 1999, II-3521, para 39. 102 See inter alia Svenska Journalistforbundet v. Council, Case T-174/95, E.C.R. 1998 II-02289, para. 66 and Heidi Hautala v. Council, Case T-14/98, E.C.R.

1999 TI-02489, para. 83. See also

K. Lenaerts (2004), ““In the Union we trust”: Trust-enhancing principles of Community law’, 41 (2), Common Market Law Review, 317-343.

103 P. Dyrberg (2002), ‘Accountability and Legitimacy: What is the Contribution of Transparency?’, in D. Wincott and A.M. Arnull (eds), Accountability and Legitimacy in the European Union, Oxford: Oxford University Press, p. 83. 104 Article 255 of the EC Treaty and Article 1 of the EU Treaty. 105 See White Paper on European Governance, supra note 18. 106 See A. Peters (2004), ‘European democracy after the 2003 Convention’, CMLR, 41 (1), 65. 107 See D. Curtin (2000), ‘Citizens’ Fundamental Right of Access to EU Information: an Evolving Digital Passepartout?’, Common Market Law Review, 37 (1), 27-28. 108 Article I-50 (3), supra note 21.

109 Access to documents nevertheless raises still the question as to the specific language in which these documents can be accessed. From the perspective of legitimacy, the language regimes adopted by agencies remain a thorny issue. See for a discussion of this issue, E. Vos, supra note 32. 110 Article II-101, supra note 21. See already Agency Working Group, supra note 23, 18. ‘1 Special Report from the European Ombudsman, supra note 94.

Independence, Accountability and Transparency

137

'2' See, for example, the transparency rules of the EMEA of 31 October 2003: http://www.emea.eu. int/pdfs/general/manage/mbar/Transparancy%20pol/005203en.pdf. "3 Article 4 of European Parliament and Council Regulation No. 1406/2002 of 27 June 2002 establishing a European Maritime Safety Agency, O.J. L 208 of 5 August 2002, p.1 and Article 47 of European Parliament and Council Regulation No. 1592/2002 of 15 July 2002 on common rules in the field of civil aviation and establishing a European Aviation Safety Agency, O.J. L 240 of 7 September 2002, p.1.

4 US 6 ‘17 U8

Article 41 of Regulation No. 178/2002, supra note 45. See http://www.eudra.org (home page). Article 38 of Regulation No. 178/2002, supra note 45. See D. Curtin, supra note 107. See already D. Curtin (1999), “Civil Society” and the European Union: Opening Spaces for

Deliberative Democracy?’, in Collected Courses of theAcademy ofEuropean Law, Florence: Kluwer, pp. 185-279. 119 See the US Freedom of Information Act (‘FOIA’), 5 U.S.C. §552; the Federal Advisory Committee Act (FACA), 5 U.S.C. app. II and the Government in the Sunshine Act, codified in 5 U.S.C. section 552b. 120 Unless, acting on a proposal from the Executive Director, it decides otherwise for specific administrative points of its agenda: Article 38 (2) of Regulation No. 178/2002, supra note 45.

121 Tbid. y 22 R. Bal, W. Bijker en R. Hendriks (2002), Paradox van Wetenschappelijk Gezag: Over de Maatschappelijke Invloed van Adviezen van de Gezondheidsraad,

Gezondheidsraad, p. 315. 123 Article 9 of Regulation No. 178/2002, supra note 45.

1985-2001, Den Haag:

-_ (Seer

ae

ae

Anerae

a)

re

WAPed wD-

7

-

*> ln

| > Soe

?

—= aes

hobs

Gay



.ereeee Garey ae a 4 —— 'Sewelin

ems

a

f= —



7)

a

PART THREE

Challenges raised by Agencies in the EU

T4 ie:

-

B thi

26) ‘

ree 6 naa , ita

s #

;

&

¢.

Pe

e

| é



7 «

fia

iin

2

.

7.

Good Governance and European Agencies: The Balance Michelle Everson

INTRODUCTION: GOOD GOVERNANCE IN CONTESTED POLITIES The focus placed by the European Commission, not only upon governance, but more directly on ‘good’ governance, in its White Paper of 2001,’ is indicative of the fundamental theoretical and practical conundrum facing all efforts to create stable, effective and legitimate governing structures at post-national level. Between facts and norms, post-national governance is not only a (vitally important) matter of establishing institutions that derive a factual legitimacy from their ability to furnish a more effective, efficient and economic form of governance than is possible at national level. Instead, ‘good’ governance, at post-national, as well as at national level, is also a matter of the establishment

of ‘normative’ legitimacy, or, a scheme of governing that pays equal attention to the need to shape institutions of governance in a manner that guarantees the sustained acceptance and support of the governed, no matter which individual advantages or disadvantages post-national decision-making entails. The endeavour to furnish ‘good’ governance thus underscores the Commission’s dual attempt to fashion institutions of post-national governance in line, both with its commitment to the three ‘Es’ of efficiency, effectiveness and economy,” and with an eye to the maintenance of a ‘close relationship’ between governors and governed, or the ‘inclusion’ of the European citizen within the decisionmaking activities of European institutions. The establishment of institutions of a civil society within Europe, the inclusion of, say, consumer protection interests within institutions of European governance, as well as the heavy emphasis placed upon the quality and independence of the expert advise upon which European decision making rests; all are indicators of the Commission’s aim to bridge the gap between facts and norms, and to imbue European governance institutions with both technical and social legitimacy.

14]

142

Regulation through Agencies in the

EU

Efficiency, effectiveness, economy and, vitally, ‘accountability’, would thus appear to be the norms against which new institutions of European governance must be assessed for their legitimacy. And yet, valuable as any such scheme is for assessing the legitimacy of new

and proposed regulatory authorities at European Union (EU) level, recent events within Europe, more particularly constitutional controversy, would seem to indicate that the effort to bridge facts and norms and to supply European supranational governance with sustained legitimacy is still beset by difficulties. Above all, Europe remains a ‘contested polity’ in which the determinative norms and rationales of European governance, as well as the formal normative framework in which such governance is exercised, remain subject to doubts,

conflicting interests and contrasting proposals for its future development. Seen in this light, Commission yardsticks for ‘good’ governance accordingly cannot but be seen as being provisional in nature; they are important points of reference in relation to the mechanics of institutional design, but are provisional nonetheless. Concentrating the argument upon the creation of independent or semiindependent regulatory authorities at European level, the contested nature of the European polity accordingly determines that any analysis of the quality of European governance in this field must be conducted at a number of levels. Certainly, the Commission’s efforts to ensure good governance must be measured against its own legitimacy criteria. Nonetheless, these value yardsticks must also be tested for their resilience and governing sustainability within a contested European polity that may and does doubt any endeavour to concretise the norms and values in accordance with which it is governed. “Accountable to whom?’ ‘Effective and efficient in which manner?’ These are the questions that can be and are posed, not simply in relation to the day-today regulatory activities of European agencies,’ but also with regard to the underlying visions of the legitimate nature of the European polity that each and every act of agency design embodies. Are agencies, at least in their technocratic scientific character, reflective of a vision of Europe governance as an apolitical and independent ‘fourth branch of government’?‘ Are they by contrast, and in view of principles of non-delegation, or their continuing subordination to the Commission, merely a mirror of a wholly conventional approach to administration, which locates the nexus of the executive’s legitimacy in its ability to give unimpeded voice to the transmitted will of the traditional polity?® Alternatively, do they give voice to new governance notions which envisage an immediate relationship between ‘administrator’ and ‘administered’? The new and not so new European agencies are certainly an expression of a desire to give functional shape to an integrated European market; they also, however, lie at the heart of a storm of controversy on the meaning of legitimate governance within Europe.

Good Governance and European Agencies: The Balance

143

Between facts and norms, the following analysis thus seeks to give a current balance of ‘good governance’ within European agencies, not through the simple elaboration and review of the various devices and desires of contemporary agency design, but rather through: (1) an examination of the normative consequences for our traditional perceptions of the executive (agency design), and the legal control of the executive (administrative law), of the lack of conventional state-

based constitutional settlement within Europe; (2) an elaboration of reference concepts of ‘political administration’, ‘executive deliberation’ and ‘public law’ which might allow for the evolution ofa ‘neutral’ set of normative and practical criteria against which the quality of current European agency governance might be measured; and (3) a closer look at and examination of contemporary and proposed European agencies.’ Between facts and norms, European agencies can only truly contribute to good governance within the EU, where an equal eye is cast both to technical and social, or practical and normative, demands for legitimacy, and a cautious institutional and legal path is woven through a mass of conflicting visions of the proper nature of the European polity. ;

EUROPE’S CONTESTED PUBLIC ‘INTERESTS’ Transmission Rejected

Article 37(1) of Regulation 178/2002 of the European Parliament and of the Council,* establishing the general principles and requirements of food law and a European Food Safety Authority (EFSA), declares confidently that: The members of the Management Board, the Advisory Forum and the Executive Director [of the EFSA] shall undertake to act independently in the public interest.

The second paragraph of Article 37(1), making it clear that members ‘shall make a declaration of [...] interests indicating [...] the absence of any interests which might be considered prejudicial to their independence’, indicates that this provision was included within the Regulation in an effort to reproduce, at European level, the traditional ethos of civil service and administrative impartiality historically present at national level. Nonetheless, the partnership within this Article of notions of ‘independence’ and ‘service to a public interest’ might also be argued to be indicative of the normative and factual peculiarities that mark a nascent European ‘regulatory’ executive; peculiarities that become evermore apparent if the slightly contrasting provisions of Article 37(2), applying to members of the Scientific Committee, are also considered: the members of

the Scientific Committee and the Scientific Panels [of the EFSA] shall undertake to act independently of any external influence.

144

Regulation through Agencies in the

EU

This provision, shielding scientific experts from any ‘undue’ influence, and even from the influence of a public interest that might express itself as “public opinion’, immediately confirms the technocratic character of a tier within a European regulatory executive, which owes a duty only to the pursuit of ‘scientific truth’. By contrast, Article 37(1), applying to the governing bodies of EFSA, and demanding independence (presumably also from the named treaty institutions of the EU) in service of a public interest, would seem to suggest some form of ‘direct’ (whether normatively, politically or socially flavoured) interaction between a further governing tier within an emergent European regulatory executive and a wider European public. It is not only the tiered nature of the administration within Europe, but rather this apparent ‘direct’ connection, or establishment of ‘duty’ as between the European regulatory executive and a wider European ‘polity’, that seems to make it distinct from traditional administrative structures to be found at the national level. Certainly, this differentiation in duties owed, either to the public or to scientific truth, can be explained by the differences in the tasks given to Management Boards, Executive Directors, Advisory Councils and Scientific Committees. Thus, as a preamble to the Regulation explains, ‘the scientific and technical issues in relation to food and food safety are becoming increasingly important and complex’ (preamble (33)). Food law aimed at combating risks to health is accordingly broken down into a technically-refined process of ‘interconnected’ risk analysis, risk management and risk communication (preamble (17)), wherein: (a) risk analysis is a wholly apolitical task to be assigned to independent scientific committees; (b) risk management may be affected by non-technical issues such as ‘societal, economic, traditional, ethical and environmental factors’, as well as, ‘the feasibility of controls’ (preamble

(19)) and as such is devolved to ‘politicised’ Community institutions and national authorities; whilst (c), risk communication, the transparency of which is vital to ensure “consumer confidence and the confidence of trading partners’ within the internal market (preamble (22)), becomes a task for the administering institutions of the EFSA. Seen in the light of this division of tasks, although the non-scientific management bodies of the EFSA cannot be described as possessing political powers to intervene in regulation for, say ethical or social reasons, the ‘interconnected’ nature of risk assessment, management and communication, determines that they do have a vital agenda-setting role within food law, both through their structuring and oversight of independent scientific experts and through their ability to control the dissemination of information on risks. Accordingly, it appears appropriate for the Regulation to establish firmly that such a potentially agenda-setting role must be exercised in accordance with the public interest. However, by the same token, the recognition of the interconnected, possibly even indivisible, nature of risk analysis, management

and communication, not only raises the character of EFSA managing bodies

Good Governance and European Agencies: The Balance

145

above the status of simple technocrats, but also undoubtedly heralds a final departure from the notion that legitimate government entails the ex priori setting of a political agenda, the transmission of a political will to a neutral executive, and the final ex ante execution of a political mandate by the administration.’ In other words, constant monitoring of scientific risks, post-market control and the brief of the EFSA to supply independent scientific information on demand to the institutions of the Community and the authorities of the Member States, entail a recognition that, beyond general commitments to the maintenance of a ‘high level of protection for health and safety’, food law, and its administration, is not a proactive matter of the formulation of detailed policy and its application within the market. Instead, food law and food policy is a reactive matter of responding to the sometimes foreseeable and sometimes unforeseeable risks that may arise within modern production processes. Seen in this light, the institutionalisation of a monitoring authority as an agency embodies a simple practical acceptance that executive functions are not merely about implementation, but are also, however indirectly, about the day-to-day formulation of policy and regulation in the light of prevailing market conditions. Structuring of research and future research demands, shaping of the modes of the communication of risks, as well as the stimulation of public discussion on particular issues; all can contribute to the manner in which we as a polity respond to individual issues arising in food markets, and all can consequently have a major impact on everyday policy-making. Accordingly, Article 37(1), or what may be initially regarded as a simple measure to secure the impartiality of the administration, may also be argued to be a normative guarantee, given to the European polity as a result of a fundamental upheaval in the manner in which administrations are conceived of, and as a consequence of the recognition that a clear distinction can no longer be made between legislative and executive governance functions. The European Public Interest in ‘Polity Restraint’?

The weakening of the transmission belt model of administration and the consequent direct dedication of the EFSA to the independent service of the public interest, however, also raises further fundamental questions about the place of European agencies within European governance. The concept of ‘public interest’ is thus itself far from being a neutral normative term. Rather, lying at the core of the liberal constitutional settlement, the notion of public interest may be understood as an expression of the differentiated underlying rationales of entire schemes of governance, as well as a reflection of the substantive or institutional aims to which such governance schemes are dedicated. In its simplest form, an appeal to the public interest may thus embody no more and no less than the substantive conviction that the will of the polity,

146

Regulation through Agencies in the

EU

democratically expressed, should be executed without any deviation or room for executive manoeuvre.'® Alternatively, the public interest may also lie in measures designed to ‘restrain’ the powers of the polity, whether for liberal reasons of the protection of individual autonomy and minority rights,'' or for more functional reasons, such as the constituting and protection ofan autonomous and functioning market sphere." Finally, in a more complex institutional and procedural reading, the public interest can also be located within the maintenance of the principles and norms, which seemingly underpin the liberal constitution; principles and norms, such as deliberation, solidarity and universality.'* What then is the public interest, which the EFSA in particular, and European agencies in general,'* is required ‘independently’ to serve? At one level, the particular public interests that individual agencies are required to serve are easily identified within founding regulations. The European Environmental Agency (EEA), for example, is dedicated to the restraint of a European polity with regard to ensuring that Community legislation does not unnecessarily impact upon the maintenance of an environmental quality, which has been given quasi-constitutional status within the European Treaties. The Agency is thus required to furnish Community institutions and the member states with: [O]bjective and comparable information at European level enabling them to take the requisite measures to protect the environment, to assess the results of such measures and to ensure that the public is properly informed about the state of the environment.'>

The apolitical provision of information, it is thus hoped, will have a restraining effect on general legislative programmes, ensuring that such programmes are also governed by the polity’s normative commitment to the preservation of the environment. By the same token, the public interest to which EFSA would seem to be dedicated entails further restraints on a European polity, particularly with regard to the functioning of the internal market:'° The free movement of safe and wholesome food is an essential aspect of the internal market and contributes significantly to the health and well-being of citizens and to their social and economic interests (preamble (1)). Further:

There are important differences in relation to concepts, principles and procedures between the food laws of the member states. Where member states adopt measures governing food, these differences create unequal conditions of competition, and may thereby directly affect the functioning of the internal market (preamble (4)).

Certainly, the quality of food and the health and welfare of individual citizens of the EU are aims in themselves within the overall canon of European legislation.

Good Governance and European Agencies: The Balance

147

Nonetheless, the pursuit of health and welfare is also a matter of the restraint of fragments of the European polity (member state authorities) who might inadvertently do damage to the smooth functioning of the market through their uncoordinated pursuit of welfare aims. Likewise, a diverse consumer interest, to the protection of which EFSA is also dedicated, is inexorably linked to the smooth functioning of the internal market: ‘It is necessary to ensure that consumer confidence and the confidence of trading partners is secured through the open and transparent development of food law’ (preamble (22)). The maintenance of consumer confidence in the autonomous working of the market, rather than the securing of political sovereignty over the market, would seem to be the aim of the institutionalisation of the oversight of EU food law within a semi-autonomous agency. All this would appear to support a characterisation of European agencies as the institutional expression of the desire of the European polity to limit its political sovereignty with regard to particular goals and aims that have been given a

higher normative status and recognition; to whit, the smooth and autonomous functioning of the internal market and the protection of the European environment. Seen in this light then, a ‘European public interest’ might initially be thought to liein the concept of polity restraint; polity restraint with regard to individual rights (the right to 2 good environment or, indeed, the right toeconomic autonomy), or restraint with regard to particular functional needs (the need to ensure the smooth functioning of the internal market). Non-Delegation, the Institutional Balance of Powers and Renewed Adherence to Transmission

Europe’s public interest in polity restraint, however, is still subject to a large degree of doubt. EU agencies it should thus be remembered are not fully Rather, even in the case of the new supposedly more independent Food Standards Authority, the legal notion of non-delegation determines that while member state’s control over the agency has diminished,”’ the agencies retain an advisory function while the Commission is apportioned a final decisionmaking function, or responsibility for riskmanagement. Such continued ‘political institutional’ control over the activities of European agencies is a result of a reading of Article 202 [145] of the European Treaty that draws upon the doctrine of institutional balance \aid down by the European Court of Justice in the Case of Meroni v. High Authority.” Laying down a fundamental principle of European normative organisation (Article 7), which declares that only the named institutions of the Community/Union may exercise European competences, and then only within the limits laid down by the European treaties, the Meroni doctrine accordingly amounts to a ban on the delegation of Commission functions to newly established and proposed European agencies.

148

Regulation through Agencies in the

EU

At one level, the non-delegation principle is a simple expression of a European institutional desire to prevent the surreptitious increase in the powers of any one named treaty institution through the creation of ad-hoc satellite bodies in which the powers of various other institutions may be pooled. At another level, however, ‘non-delegation principles’ also seem to bring us back to transmission belt models of administration and a notion of European public interest that resides in polity facilitation, rather than polity restraint. Non-delegation, reflecting the notion that the powers of institutions are inalienable, furnishes a vision of a final apportionment of powers within Europe; an apportionment of powers that in turn must be protected absolutely by the law in order to preserve the primacy of an existing polity, or settled mode of government (Commission control) over all executive action. Normative Incoherence within the Contested Polity

As such, non-delegation is ultimately reminiscent; indeed ‘a child of’ an ancient constitutional/legal principle that derives, not from the form of fluid and contested polity that Europe currently represents, but instead from a principle of a ‘separation of powers’, that prizes transmission belt models of administration and locates a public interest, not in the ‘restraint’, but in the

ensuring of the political sovereignty of the polity. In this sense, non-delegation and the Meroni doctrine thus undoubtedly represent a crucial normative incoherence between a reality of European governance that is accepting of the blurring of lines between political sovereignty and executive action,'? and a deep normative structure of European governance that identifies the existence of a single European ‘legislative’ will; or, a ‘decision-making will’, which is further deemed to possess full directive powers over a subordinate executive body. The Meroni vision of institutional balance is not an uncontested one.” Nonetheless, the unfortunate correlation between the EU institutional balance

and principles of non-delegation not only opens up a gulf between facts and norms within the governance of the EU, but also furnishes further incoherence as its inherent and inevitable echo of traditional notions of the separation of powers both raises expectations that Europe subscribes to traditional notions of political sovereignty and dashes those expectations as European law cannot but fail properly to police transmission. It should thus be recalled that under a traditional separation of powers model transmission administration serves the political sovereignty of the polity. The rigid distinction between legislative policy making, executive implementation, as well as a judicial application of the rule of law to executive action that concentrates on the rigid executive adherence to an existing political mandate, all add up to a non-delegation principle that derives from a guiding normative

Good Governance and European Agencies: The Balance

149

governance principle that the democratically expressed will of the polity should be paramount. Clearly, such notions of traditional democratic primacy are wholly inappropriate within the EU. Equally, however, an additional and crucial technical-legal issue is one of the institutional incongruence of a transmission belt model of European administration in view of the necessarily underdeveloped rule of law policing functions of European law. The non-delegation principle and the transmission belt model of administration are tried and trusted mechanisms of control over executive decision making within systems of law founded within the western liberal compromise. Within the liberal state, executive decision making is a necessity. Nonetheless, the liberal state is also founded within the principle of political accountability. By this token, the transmission belt model allows for the maintenance of an axiom of political accountability through a functional differentiation between principle-based constitutional frameworks and an attendant administrative law founded in a policing-oriented rule of law. Within an administrative legal canon, the political will of a liberai polity is protected through notions (in traditional common law terms) of ultra vires or ‘unreasonableness’. Meanwhile, its very liberality is assured through notions of /ocus standi that derive from the protection of individual rights and individual interests. In other words, the technical terms of administrative law

serve the constitutional principles of unitary policy making within a polity of autonomous, rights-bearing citizens. Law thus retains its normative steering function and a functional capacity to act. It is nonetheless this, the founding differential core, of non-delegation that

is lacking within European law. To be sure, the establishment of a Court of First Instance is a step in the direction of functional differentiations within European law. Nonetheless, the exact nature of that Court’s jurisdiction is still a matter of evolution, while, most crucially, both the substantive grounds of

review and the form of standing that attach to the Article 230 ‘administrative’ review mechanism of the European Treaty are, perversely, conditioned by concepts of institutional balance that have very little to do with underlying historical rationale of the principle of non-delegation. Instead, ECJ application of Article 230 is conditioned by a formalist procedural reasoning that seeks to protect, not the political accountability of the wider European polity, but rather the demarcated powers of European institutions, thus negating any administrative oversight rational within Europe that is founded in transmission. Rejecting tests of reasonableness (substantive review) and restricting standing to exclude even those liberal attributes such as property rights protection found within the most restrictive of national review systems, European law fails to establish the functional demarcation within which a policing-oriented rule of law might lay claim to be preserving the founding normative contours of a

European polity.”!

150

Regulation through Agencies in the

EU

In short then, divergence between the readings afforded the institutional balance by the Commission and Court simply confirms Advocate General Mancini’s assertion that the balance of institutional powers remains ;” an object of fluid and, on occasions, contradictory constitutional ‘contingent’

development. A concept of agency design and oversight that is institutionally, if not practically, founded in one temporal definition of the institutional balance and a consequently uprooted and legally constitutionally isolated concept of non-delegation represents a striking normative incoherence within the framework of European law.

PUBLIC INTEREST(S), THE POLITICAL ADMINISTRATION AND THE NEW PUBLIC LAW The lack of clarity in the governing normative principles that underlie European agencies; the contradiction between ‘deep’ institutional frameworks for agencies, which sees them as being subordinate to a legislative will, and practical individual frameworks which directly connect European agencies to and root them within a European polity;*? as well as, an apparent contradiction between a ‘polityrestraining’ public interest that would require European agencies to delimit political activity within the Community/Union, and a ‘polity-facilitating’ public interest that would, by contrast, require agencies to be a conduit of a European political will: all are doubts and contradictions that derive from continued uncertainty about the legitimate nature of the European polity and character of ‘good’ governance within the EU. The simple reality of agencies within the EU would seem to be one of a new form of administration with: (1) functional distinctions between individual administrators dedicated to the pursuit of scientific truth and administrators called upon to directly interact with a European polity; (2) a creative attempt to include both ‘polity-restraint’ and ‘polity-facilitation’ elements; and (3), a degree of side-stepping of established structures for the exercise of policymaking functions and the embedding of the administration within the market (normative dedication of EFSA to the smooth functioning of the market) and within civil society (inclusion of consumer and environmental interests within the EEA and EFSA). Undoubtedly, such practical constructions are a result of the Commission’s effort to gain legitimacy for supranational governance by practically ensuring efficiency, effectiveness and economy in European regulation, as well as, and with an eye to demands for ‘good governance’, a degree of accountability for the European public. Nonetheless, at the level of the establishment of clear normative guiding principles for the good governance of Europe, such pragmatic efforts are far in advance of the deep institutional structures of the EU, giving rise to a striking degree of normative incoherence.

Good Governance and European Agencies: The Balance

151

Between facts and norms, the constant effort to find normative coherence in a contested policy may seem like a work of Sisyphus. Accordingly, the natural conclusion to such an analysis might thus initially be one of an augmented expression of regret for the failure to constitute Europe such that its political will might be properly safeguarded within a coherent legal system of differentiated constitutional principles and a law of administrative application. Nonetheless, and once again from a largely legal point of view (though bolstered by various political and economic observations) this chapter does not immediately grasp at what would in all likelihood amount to traditional transmission belt straws. Instead, through a brief historical review, it now moves

on to demonstrate the continuing incongruence of a model of transmission belt administration within Europe and begins to develop concepts of ‘political administration’ and ‘public law’. Transmission Rejected: the Historical Failure of Administration

In short, transmission belt models of administration, though supportive of functional legal differentiation and protective of the axiomatic liberal constitutional comprise, have always remained compromised by their inability clearly to distinguish legislative from executive action. In the bulky canons of national administrative law, this failure is readily apparent in doctrinal conflicts that have seen notions of transmission undermined by increased degrees of substantive review,~* or by the extension of the breadth of standing to a degree that wholly contradicts the supposedly inviolate nature of a single legislative will.” From a legal history point of view, the constant blurring of the administrative constitutional distinction that such doctrinal conflict embodies might be traced to the submersion of ancient monarchical legal principles of the preservation of the Sovereign’s power within the liberal constitutional compromise at a time when inevitable and irrevocable conflicts were being formed as to the ‘public interest(s)’ that constitutional settlement should serve. In short, a very simple legal mechanism designed to ensure the ‘no-one might cloak themselves with the Sovereign’s powers to do violence to that power’*® became intermeshed within settlements that served far more complex and occasionally competing aims of political accountability, social stability, competent decision making and bourgeois economic development. An instrument of autocratic self-preservation was the empty vassal into which more modern ideas of ‘limited’ self-governance, and the projection of ‘limited’ self-governance, were poured. Where the law had once been charged with the fairly simple task of ensuring that no servant of the Crown might take more power than his due, now it was faced with an infinitely more intricate task of balancing various public interests in personal autonomy, collective political action and stable economic development.

152

Regulation through Agencies in the

EU

To be sure, given this complexity, a transmission belt model of administration proved a useful and self-justificatory legal axiom, its functional differentiation furnishing a mechanism whereby ‘technical’ law might act on a day-to-day basis, secure in the knowledge that such action derived from settled constitutional principles. Nonetheless, educed from a notion of the separation of powers, which located the raison d’étre of constitutional settlement in the primacy of collective political action, the transmission belt model of administration was never sufficiently differentiated to capture the nuances of western liberal compromises that served multiple public interests. Most strikingly, since it is often cited in justification of ‘daring’ administrative legal deviance from the transmission model, the rigidly construed belt seemed particularly insensitive to the mixture found within most western constitutions (most notable the US and the UK) of separation of powers principles and balance of powers principles. In other words, while settlements may, under pressures of widened emancipation, have had as their goal political self-governance and collective political action, they also had as their aims, in their ‘limited’ liberal manifestation, the preservation of personal autonomy, social stability or competent decision making though the limitation of collective politics by personal rights, the creation of ‘estates’ (social classes with balanced powers) or the apportioning of power between diffuse institutions; the latter given power by virtue of their various imputed or proven functional abilities. Seen in this light, recurrent legal administrative preparedness, to lessen the rigidity of nondelegation and to readmit political interests within the exercise of administrative power, is not simply a response to complex modern governance conditions but is also a reflection of the original confusion between public interests within the liberal constitutional settlement. By that token, however, a further area of ‘original’ obfuscating potential that is of far greater interest to this chapter is undoubtedly that of an existent tension within the transmission belt model between notions of political action and economic autonomy. It is thus perhaps noteworthy that more recent administrative law approaches to the executive”’ are often characterised by their preoccupation with Weberian theories of the evolution of the administration and administrative law. Weber, it is often noted, was among the first to notice

an essential ‘paradox’ in the construction of the modern rule of law and state and administration that it purported to establish. In Weber’s analysis, economic autonomy, founded in stable rights of property and exchange, formed the primary impetus for the evolution of the modern formal rule of law. At the same time, economic development and the political emancipation which it drew in its wake, augmented desires already inherent within society and the law for the ‘materialisation’ of law and the imposition of a welfare-oriented collective political upon the bourgeois economy. The modern state was thus necessarily born on the horns of a dilemma, whereby intimately interconnected public

Good Governance and European Agencies: The Balance

153

interests in economic development and political self-governance were locked into an eternal waltz between the poles of comprehensive political selfdetermination and economic nihilism.”® In modern terms, there is a tendency to represent demands for the creation of politically independent regulation as a ‘new’ and blind surrender to Weberian notions of efficiency and a triumph of technocracy over democracy. Equally, there is an assumption that law, both in terms of institutional design and in terms of the oversight mechanisms and governing philosophy that it applies to independent institutions, is irredeemably anti-political. Such views nonetheless crucially overlook the historical nature of the process of the management of economic-political tensions, the far more differentiated role that law has played in the management of such tensions, as well as the vital nature of an underlying and fundamental

political—philosophical debate on the nature of the public interest. Current debates on executive decision-making and regulation are thus not new: they are not about the triumph oftechnocracy over politics, but are instead a continuing expression of an underlying and necessarily infinite political-philosophical debate on the pursuit of complementary but conflicting notions of the public interest. Certainly, at EU level, the debate on ‘public interests’ has reached a new level of intensity: (1) since concepts of economic autonomy are no longer solely served by application of simple complexes of property rights and exchange rules; and (2) since the social solidifying concept of a state-based constitutional settlement cannot be easily applied to the governing institutions of the EU. Nonetheless, for all evolution from simple schemes of contractual freedom to refined regulatory notions of economic accounting,” and for all, in all probability, continuing estrangement from socially solidifying state axioms, the issue is not one of a radical departure from politics, but is instead a traditional and intensely political one: the constant definition and redefinition of which public interest is appropriately followed where, when and how. Seen in this light, the pertinent question would seem not to be one of whether mantra-like adherence to notions of transmission and non-delegation is specifically ridiculous within the institutional setting of the EU, but is instead one of whether transmission and non-delegation have ever been appropriate mechanisms of normative guidance within traditional state polities. The Political Administration and the Evolution of a New Public Law

This final departure from the comfort blanket of transmission axioms, and final admission of the enduring fragmentation of public interests and polities is, however, not without serious consequences of its own, not least for law:

(1) A consequential lessening of the rules governing the apportionment of power beyond the traditional polity is not immediately congruent with an

154

Regulation through Agencies in the

EU

acceptance of the apolitical primacy of a regulatory state or fourth branch of government. Rather, where the ongoing and conditional character of the discourse on the appropriate nature of the public interest to be pursued is recognised, regulatory authorities located outside traditional institutional frameworks of politics are placed under a far greater onus, politically, to justify their operations beyond accepted standards of economic autonomy and rationality. In short, and in recognition of fragmented polities and public interests, resurrection of the largely discredited term, Politische Verwaltung, may thus be justified. Not since it reproduces the welfarist aim to subordinate economic organisation to material standards

of welfare, but since a modern political administration

must

continually justify and found its pursuit of economic autonomy in the face of equally valid ‘public interest’ claims (especially within schemes of ‘good governance’) for corrective political action and substantive justice. (2) Stripped of its axiom of transmission, the law necessarily faces a collapse in its historical functional differentiation between constitutional principle and administrative law. Vitally, the new public law thus created is also alienated from its historical normative justifications for action. Traditional formal administrative can thus no longer simply be applied in the confident knowledge that any exercise of legal authority is justified by its inherent, if merely axiomatic, link to the constitutional will of the polity. Accordingly, any action that European law might take with regard to the design and oversight of independent centres of European regulation must take care not only to justify external operations, but also take care to justify itse/f*°

A ‘POLITICAL ADMINISTRATION’: EXECUTIVE ‘DELIBERATION’ WITHIN EU REGULATORY AGENCIES At one level, the justifications given above for the abandonment of nondelegation principles within the EU, together with the evolution of a ‘public law’ of agency design and oversight, might be viewed as a mere exercise in utopian theory: after all, all experimentation with plural administrative law notwithstanding, no liberal state has fully liberated itself from transmission belt regulation. Nonetheless, where specific conditions of institutional contingency and constitutional non-settlement within the EU are taken into account, it might be argued that piecemeal design of European agencies, as well as tentative judicial efforts to increase the bases for agency oversight, of themselves constitute an incremental effort to establish a form of pluralist political administration that is suited to EU conditions. In the following, this chapter accordingly seeks both briefly to sketch out a model of agency design that may be suited to modern European conditions and to assess current modes of agency design and oversight within the EU for their

Good Governance and European Agencies: The Balance

155

sustainability and legitimacy within a fragmented polity that is governed by a sui generis law lacking recourse to principles of conventional constitutional settlement. Between facts and norms, the effort to create piecemeal legitimacy through the constant attuning and re-attuning of the guiding normative principles of ‘good’ governance must replace our traditional faith in the constitutional

axioms. Political Administration

Independence within the institutional balance To date, two particular readings of the character of the institutional balance have mitigated against the full institutional independence of agencies within Europe. Clearly, the Commission’s continued belief in principles of nondelegation have led to its restriction of regulatory mandates, its retention of the final power of decision making in the case of regulatory agencies and the relegation of most European agencies to a status of information gathering bodies. Equally constraining of independence, however, has been the recognition that the powers of regulatory authorities do not merely derive from the Commission, but also from the Member States and other EU institutions. Accordingly, management boards, with their various national, European and interest group representatives have become an institutional expression of the notion, wellknown in the context of ‘comitology’,*' that all exercise of European powers must pay attention to all national, European and private interests. Now, where the talk is of a ‘political’ administration, a ‘close’ relationship

between political actors and regulatory authorities is not necessarily a delegitimising factor and may even be a vital aid in the maintenance of a discourse dedicated to constant review of the appropriate pursuit of variable public interests. Nonetheless, given the potential raison d’étre of regulatory authorities and the fourth branch of government as guardians of a public interest in economic autonomy (or, indeed, at another level, in environmental integrity), the lack of formal institutional guarantees for regulatory autonomy appears incongruous. By this same token, it is possible to conceive of the ‘contingent’ institutional balance in terms which support greater regulatory autonomy, either since regulation is recognised as a legitimate focus for the exercise of a portion of a fragmented polity’s power, or since various functional characteristics of plural centres of power are recognised as being worthy of protection from other potentially disruptive social and political interests (balance of powers arguments). In this regard then, the changes in the management board structure of the European Food Safety Agency, and the concentration of national interests within a coordinatory ‘advisory’ board, together with the increased regulatory autonomy which Article 37(1) would seem to imply — both moves undoubtedly a crisis-led product of the BSE experience — appear to indicate that the principle

156

Regulation through Agencies in the

EU

of the institutional balance is flexible enough to allow it to be construed, where appropriate, in terms of a traditional constitutional balance of powers, holding that power should be divided amongst as many centres as possible. Political ‘Accountability’

BSE is one crisis. It is highly likely, as an EU regulatory complex grows evermore intricate and intense, that more crises will follow, in particular, say, in the areas

of pharmaceutical or environmental control. Accordingly, the political conditions for the establishment of increasingly independent regulatory authorities with ever less fettered powers of regulatory decision making are likely to grow more favourable. However, independence within this post transmission context can never be commensurate with a lack of political accountability. Instead, and within the context of intense debates on ‘good’ European governance, a ‘political’ administration must maintain its links with a social and political sphere within which valid demands for social justice and political action are formulated. In practice, such a stipulation surely extends far beyond a demand that accountability be secured through the application of the tried and trusted mechanism of control over a fourth branch of government established, for example, within the US regulatory sphere (for example, mechanisms such as parliamentary review of agency budgets). Rather, where final acceptance is made of the conditional nature of public interests, this issue is no longer a simple one of whether independent regulatory authorities are ‘efficiently and effectively’ performing their allotted roles. Instead, it also entails, the continuing political justification for such roles. This is perhaps the most demanding of challenges facing independent regulatory agencies. At core, the challenge would thus appear to be whoily counter-intuitive: if independent regulation is in large part justified by its apolitical pursuit of economic autonomy and development, how can a political character be legitimately reintroduced to such regulation? As noted, this is not a matter for the authoritative involvement of politically oriented management boards within agency operations; such involvement can only reduce the efficiency and effectiveness of agencies and fatally conflate and obscure the pursuit of conflicting public interests. Instead, the issue is far more one of establishing systems whereby the breadth of regulatory mandates may constantly be reviewed, and unexpected issues of social regulation be tackled with an eye to continual readjustments in public interests and still valid demands for social and political justice. The reintroduction of politics into an autonomous apolitical sphere of economic regulation is accordingly a multi-level problem, involving the creation of political and social links between a diffuse polity and the ‘political’ administration; links that neither politically compromise regulation, nor

Good Governance and European Agencies: The Balance

157

artificially re-establish the myth of a single, normatively determinative, public interest. At core, the issue is one of the redefinition of European politics to

ensure that while axiomatic constructions of social ‘solidarity’ (a potential cloak for the pursuit of particularist political interest) are absent from the regulatory sphere, a measure of ‘informed’ political contention between conflicting public interests is a feature of regulatory decision making: a) Limited Governance: Though the following comments remain necessarily tentative, an initial key to the establishment of successful political administration might be found in a partner notion to balanced or mixed constitutions; that is, the notion of regulative/legislative self-restraint and limited governance. Alternatively, where current European agency design is led by a preoccupation with the identification of ‘narrow’ regulatory mandates, an obsession potentially undermining of efficient and effective regulation, the issue should instead be redefined as one of the establishment of the normative conditions for the exercise of governance restraint within appropriate regulatory competences. Where possible and appropriate, regulatory authorities should possess a pre-emptive regulatory competence. Nonetheless, pre-emption should likewise be governed by wise self-restraint; a secession of decision making to social and political centres of power and influence where this seems to be more appropriate (for example, where issues of seemingly technical regulation are deemed to entail redistributive competences in the manner, say, of essential facilities doctrines). Seen in this light, recent European efforts to reformulate the Management Board of the EFSA, in order to establish a more non-directive collection of

political and social interests within the fourth branch of government, would seem to offer a blueprint for future agency design; a forum within which public interests can be weighed and the parameters for social, political or economic action be determined. b) ‘Deliberative’ Decision Making: Following on from this, however, a further question arises as to the possible character of the governing principles which might effectively regulate discourse between social, political and economic values, where each value is seen as being equally valid. Clearly, within this construction, the search is on for a mode of regulating interaction within a plural fragmented polity, within which the governing normative basis for action is no longer the primacy of an imputed (common) legislative will, but the context specific ‘rationality’ of any form of decision making. Which values should prevail, when and why? In this regard then, normatively founded politics (democracy) can and have been redefined as a deliberative process,” whereby the functional act of effective problem solving becomes the leading criterion for decision-making, while parties to the decision are subject to a duty to justify their arguments that is founded in a proven eschewal of all personal (non-rational) interests. Within a context of

158

Regulation through Agencies in the

EU

‘arguing’ rather than ‘bargaining’, a political administration might thus identify the appropriate basis for regulatory self-restraint; the context specific primacy of competing public interests. In short, ‘effective problem solving’ is a criterion that matches the Commission’s desire to ensure the factual legitimacy of European regulatory bodies through the three ‘E’s’ of efficiency, effectiveness and economy, while deliberation augments the normative legitimacy functions of ‘accountability’, especially as regards the adequate representation of all civil society interests.** c) Executive Deliberation: The notion of deliberation, however, is also highly

relevant beyond the drawing of agency mandates. As noted, the lessons of the BSE crisis have led to an increased degree of independence for EFSA. At one level, a response to a regulatory failure deriving from political interference within technical—scientific decision making, such a move can and should be praised. Nonetheless, care should also be taken not to overlook further lessons of the

BSE crisis, primarily the inability to provide comprehensive systems of risk analysis and management within ever more complex systems of modern protection. Just as the origins of BSE in part remain obscure to us, and although regulatory failure was also due to a confusion of political and regulatory interests (as well as a scientific failure in the transmission of relevant information), European regulatory decision making is still, on a day-to-day basis, haunted by the constant spectre of ‘scientific indeterminacy’, or an ability to clearly assess the risks that modern production processes pose. Alternatively, day-to-day regulatory activities can and might be characterised as being a substantive social—political decision to define risk and redistribute its consequences among producers, consumers or the general populace (through the economic costs of regulation). The precautionary principle, when seen in this light, is not simply, as the EFSA regulation would seem to suggest, a neutral technical or legal term, but instead entails value judgments on risk definition and redistribution. Given such a conclusion, ‘deliberation’ or the reconnection of regulatory authorities to the social and political sphere through the constant, though ‘rational’ non-interest led, evaluation of complementary social and political values is also an imperative. No independent authority can preclude the danger that some of its decisions will be ‘politically flavoured’, or even, ‘wrong’ in fact. Nonetheless, political authority, or the legitimacy of decision making, must always be maintained through vigilant and constant oversight of the value consequences of individual decisions. In short, the presence within independent regulatory authorities of explicitly social or political interests (be these present within formal advisory boards, or encouraged through more informal channels of information exchange with civil society, as may be argued to be the case in relation to EIONET) is a necessary legitimating mechanism for what should again be seen as a political administration.

Good Governance and European Agencies: The Balance

159

Legal Oversight

Finally, some thought should also be given to the role that European law does or could play in the oversight of an independent fourth branch of European government, or a European political administration. Public law A primary point to remember, here, however, is the fact that, post transmission,

the functional differentiation between constitutional principle and administrative law no longer has any application, leaving courts of review in the uncomfortable position, not only of reviewing administration, but at the same time, justifying their own normative basis for action. In other words, a European public law is denied recourse to the simple technical mechanisms of normative derivation to be found in notions such as ‘non-delegation’ and as a consequence must constantly draw upon its own internal legal values to justify its oversight functions and actions. j Systemic closure and a ‘modern’ proceduralism Now, in this regard, the interesting point for this chapter is not the manner in which such values might be evolved with regard to any future European ‘political’ administration, but is instead, the degree to which a sui generis European law has already evolved its own form of self-justificatory systemic closure and mode of procedural review that is supportive of political and executive deliberation, rather than merely reflective of axiomatic adherence to potentially inappropriate imputed models of the legitimate polity. Certainly, European law does display weaknesses in relation to the oversight (and design) of European regulatory agencies. Most particularly, Meroni seems to continue to stand in the path of truly effective delegation, while restricted bases for judicial review under Article 230 seem currently to foreclose one of the most important basis for legal oversight over the political administration. Nonetheless, the current genius of European law lies in the fact that, in view of the lack of conventional constitutional settlement, it has never been able simply to refer to normative shorthand axioms of legitimation and has instead always been required to respond to the political complexities of the EU drawing on legal internal values for its own legitimation. At this level of analysis then, attention is thus immediately drawn to the fact that while Meroni may currently appear to impose a non-delegation principle, the Court is capable of variable readings of the institutional balance and may, if the issue ever arises in its correct context of regulatory efficiency and effectiveness, vary its jurisprudence. Equally, notions of the real context specific meaning of the institutional balance are already being varied, at least within the

CFI, especially with regard to individual standing in cases of executive decision

160

Regulation through Agencies in the

EU

making.*4 Beyond such pragmatically attuned judicial flexibility, however, the most important facet within current and future European legal efforts to oversee delegated regulation is the evolution of a series of procedural doctrines within European law that both reflect legal internal values and seem to support ongoing processes of adjustment between equally valid public interests through political deliberation: (1) Proportionality: By far the most important doctrine in this respect, deriving from the legal internal value, or need, to maintain judicial independence while facilitating societal interaction and decision making, is that of proportionality. The demand that all decision making be fully justified both in relation to its necessity and with regard to the appropriate means used to solve problems currently plays a crucial role in setting a workable balance between the national and European jurisdictions. In the future, a mechanism that calls on all bodies to review the grounds for their decision making within an independent forum, could surely also play a primary role in policing ‘limited government’, or decisions on how, when and where to use or restrict the pre-emptive regulatory powers of the political administration. (2) State of the Art Decision Making: A further deliberation ensuring doctrine, informed by ancient legal values of ultra vires decision making, has been the insistence of the European Court of Justice upon the quality of the information used to reach decisions. Again, where all decisions are to be reviewed in the light of independent procedural criteria on the character of the information that feeds those decisions, the danger that bargaining replaces arguing as personal interests are cloaked as values is lessened. (3) Relevant Interests: Importantly, the ECJ has also recently sought to develop subtle means to maintain ancient legal notions of ‘equity’ within a modern European legal system.** In particular, the notion of equity dictates that no decision should be taken in a manner that leaves any parties affected by that decision without a means to register their views or interests. Seen in this light, such a line of European law could both allow for the extension of individual standing under Article 230 and go some way to answering one of the primary conundrums that decision making within a fragmented plural polity faces; that is, the identification of instruments that compensate for consequent restrictions in political representation.

In other words, one of the most immutable barriers to increased delegation of powers must be the claim that delegation would leave policy and decision making open to domination by a minority of public interests. In this regard, the presence of social interests within the management or advisory boards of agencies is to be prized. Nonetheless, any claim that these interests represent an adequate cross-section of civil society is surely illusory. However, where, and to the

Good Governance and European Agencies: The Balance

161

degree that, the law of review is tailored to ensure that all relevant interests might participate in decision making, either through a widened basis for /ocus standi or through the ‘deliberative’ stipulation that all relevant interests are reviewed during decision making, lack of representation within the plural polity presents a lesser problem.

CONCLUSION: CONSTITUTIONALISM INCREMENTALISM Between facts and norms, the primary conclusion reached by this chapter is that under current conditions of continuing contestation over the guiding normative principles of the EU, a seemingly simple question as to whether the Commission has achieved ‘good’ governance in relation to the design of European agencies, can only be answered through complex examination of the interplay between practical (functional) legitimacy and continuing efforts to readjust the minutiae of European institutional design and law in the effort to capture fleeting moments of normative legitimacy. Equally, however, this primary conclusion does not of itself result in a demand for the immediate concretisation of the contours of the European polity. On the contrary, the ‘real-world’ manifestations of a lack of constitutive concordance within Europe may be a modern phenomenon, occasioned by efforts to create a novel form of post-national organisation. They are not, however, a sui generis occurrence. Rather, behind the crumbling facades of national constitutional settlements, a continuing battle has always been fought between contrasting, even contradictory, visions of the public interest, and the true meaning of good governance. Seen in this light, although the manifold contradictions and incoherencies within current schemes of European agency design might, on the one hand, be considered to be a danger to good governance; on the other hand, they can also be regarded as an opportunity, a final impetus to admit of the shortcomings in constitutional settlement and to further refine the liberal compromise through an incremental constitutionalism that remains true to all liberal values and principles, but also admits to a simple reality of pluralist social organisation and political aspirations. Between facts and norms, an inherent danger in such an approach is of course one of the simple unreflective translation of ‘fact’ into ‘legitimacy’; fait accompli governance as good governance. And yet, as a review of EU design demonstrates, ‘fact’ can force a ‘quest’ for good governance. The Commission’s legitimacy value yardsticks of efficiency, effectiveness and economy may be derived from a simple necessity to justify supranational above national government and may not be the undisputed primary rationale for EU governance. And yet, they do force the Commission to enunciate its aims, and so open up

162

Regulation through Agencies in the

EU

room for disputation and demands, from other segments of European society, such that further governance rationales (in our case, the interests that cluster around the notion of accountability) are able to demand their place within European agencies. Equally, in the experimental mix and match process of agency design that follows, the existence of a law (European law), which has long been required to justify its own legitimate existence through sensitive incremental judgments and adherence to its own (inherently liberal) values, does in fact promote ‘political’ disputation on the appropriate values that European administration must follow and when. The resulting cacophony of values that centre around the modern task of responding to vital economic and social concerns, rather than seeking to preempt social and economic development (and which are coordinated by a wholly neutral law dedicated to its own values), is characterised here as a ‘political administration’; a term that takes on a normative character of its own where it is

argued to be founded in simple recognition of the irreconcilable nature of social, political and economic values and the impossibility of pre-empting social, economic and political development. Contestation, in this model, must both be preserved and disciplined by the set values of law. Applying this model to current agency design within the EU, the results accordingly lie on a positive side. To be sure, normative incoherence and inconsistency in agency design is disturbing. Yet, agency design in the EU is marked by contestation, experimentalism and, above all, is overseen by a law that retains its potential to submit governance within Europe to neutral (legal internal) values that both promote problem solving in fact and give normative support to disputation and incrementalism. Between facts and norms, the final balance on whether EU agencies furnish us with ‘good governance’ must be drawn in line with the following statement: “we are not sure what good governance is, but long may it remain so’.

NOTES '! See White Paper on European Governance, COM (2001) 428. 2 C. Harlow (2002), Accountability in the European Union, Oxford: Hart Publishing. 3 See Harlow, supra note 2. 4 See G. Majone (1994), ‘The European Community, an Independent Fourth Branch of Government?’, in Gert Briiggemeier (ed.), Verfassungen fiir ein ziviles Europa, Nomos: BadenBaden, pp. 23-44.

> See Richard B. Stewart (2003), ‘Administrative Law in the Twenty-First Century’, New York University Law Review, 78, 437.

6 See infra. 7 More particularly, the European Food Standards Agency and the European Environment Agency.

8 See European Parliament and Council Regulation No. 178/2002 establishing the general principles and requirements of food law and a European Food Safety Authority , O.J. L 31 of 7 September 2002, p. 16.

9 See Stewart, supra note 5. 10 See Harlow, supra note 2.

Good Governance and European Agencies: The Balance

163

'l See S. Holmes (1995), Passions and Constraints, Chicago: University of Chicago Press. '2 See E.-J. Mestmicker (1994), ‘On the Legitimacy of European Law’, Rabels Zeitschrift fiir ausldndisches und internationales Privatrecht, 58, 615. '3 Jurgen Habermas

(2001), ‘Der Demokratische

Rechtsstaat — eine paradoxe Verbindung

widerspriichlicher Prinzipien’, in: ibidem, Zeit der Ubergdnge, Frankfurt A.M., pp. 133-154. '4 Interestingly enough, direct reference to service to a public interest is a new feature within European agency design.

'S See Article 2 of Council Regulation No. 1219/90 on the establishment of the European Environment Agency and the European Environment Information and Observation Network, O.J. L 120, of 11 May 1990, pp. 1-6.

'6 See supra note 8. '7 The Management Board is no longer made up simply of national representatives and representatives of the

Commission, but now instead includes independent consumer interests; the

national interest within the internal structures of the EFSA is also greatly weakened by the stipulation in Article 25 of the EFSA regulation that not all nations will be represented within the Board. Instead, a mix of members should be chosen to ensure the ‘highest standards of competence, a broad range of relevant expertise and, consistent with these, the broadest possible geographic distribution within the Union’ (Article 25(1)). '8 See Meroni v. High Authority (Case 9/56), E.C.R., 1957/1958, p. 113. '9 In other words, the forging of direct responsibility between and executive and a European public.

20 See, for exampie, Mancini’s assertion that the balance of powers is ‘contingent’, Parliament V. Council (Case 302/87), E.C.R., 1988, 5615. 21 Although recent attempts by the Court of First Instance in cases such as Pfizer (Pfizer Animal Health SA v. Council of the European Union (Case T-13/99), E.C.R., 2002) to increase the breadth of individual standing within the EU, such that the impact of executive action upon individual economic interests might be reviewed by the Court, do give us an indication of the ambition of European Law to establish some form of functional constitutional-administrative demarcation. 22 See supra note 20. 23 Not least through the inclusion within the Management of the EFSA of consumer interest representatives, and the exhortation to the European Environment Agency to network with various environmental interests. 24 Tn the UK, for example, the move from concepts of unreasonableness and ultra vires to more diffuse concepts of ‘illegality’. 25 In the US, class actions and a definition of personal economic interest extending far beyond constitutionally guaranteed rights of property and personal autonomy. 26 The original formulation of transmission found within Blackstone. See, Blackstone/Katz (1979), Commentaries on the Laws of England, Chicago: Chicago University Press. 27 C. Harlow (1997), Law and Administration, London: Butterworths. 28 Paraphrasing greatly, Max Weber (1967), Economy and Sociology, Volume 2, G. Roth and C. Wittig (eds), Berkeley: University of California Press, Chapter 7.

29 See supra note 4. 30 Interestingly, such a conclusion may call for a degree of systemic ‘re-closure’ of the law. In other words, absent settlement and a consolidated polity, law may be best advised, in its pursuit of political justice (including social justice), to draw upon its own traditional justificatory values rather than those of any imputed polity or public interest. In practice, this would imply the adoptions of a modern legal proceduralism founded in neutrality, efficiency and societal facilitation. 31 See Ch. Joerges (2002), ‘The Law’s Problerns with the Governance of the European Market’, in Christian Joerges/Renaud Dehousse (eds), Good Governance in Europe’ ‘Integrated’ Market, Oxford: Oxford University Press, pp. 3-34. 32 See Cohen and Sabel (1997), ‘Directly-Deliberative Polyarchy’, European Law Journal, 3, 4. 33 See infra under ‘Legal oversight’, ‘Systemic Closure and a “Modern” Proceduralism, point (3).

34 See supra note 21. 35 See, in particular, the Social Insurance Cases, above all Albany, With, Drijvende, Bokken and

Bretjens, (Joined Cases C-115/97, C-116/97, C-117/97, C-219/97), E.C.R., 1999, I-6025 and E.C.R., 1999, I-6121.

8.

Coordination of European and Member State Regulatory Policy: Horizontal, Vertical and Transversal Aspects Pierre Larouche

INTRODUCTION Economic regulation has experienced spectacular growth in the EC since the 1990s. Riding on the wave of liberalization efforts, whole new regulatory frameworks were put in place in areas such as telecommunications (now extended to electronic communications), broadcasting, posts, air transport, rail transport and energy. Each of these frameworks entailed — more or less explicitly — the creation of a national regulatory authority (hereafter “NRA’) to administer that specific framework. On the competition law side, in addition to the growth of merger control, recent reforms have opened the door to a significant increase in activity in the enforcement of Articles 81 and 82 EC as well, through the

empowerment of national competition authorities (hereafter NCAs) and national courts as fully-fledged enforcers. Indeed the current institutional picture is as follows. At the European level, the Commission is active both as an enforcer of EC competition law and as a major player in the development of the various sector-specific regulatory frameworks. The Court of Justice plays its usual roles. In each Member State: — —



AnNCA is in charge of applying EC and national competition law; For each sector-specific regulatory framework (usually), an NRA is competent to apply national law (a large part of which often simply implements EC legislation); Finally, national courts handle numerous cases either under competition law (as a first instance alternative to the NCA or the Commission) or under 164

Coordination of European and Member State Regulatory Policy

165

sector-specific regulation (usually by way of judicial review). For the sake of keeping this chapter within bounds, national courts will not be dealt with in much detail. Bringing them into the analysis adds a further layer of complexity. Taking the EU enlargement into account, this means that hundreds of authorities are involved in one way or another in the operation of economic regulation in Europe. This chapter deals with the most obvious issue that can be expected to arise in view of such a substantive and institutional inflation, namely coordination. For the sake of limiting complexity, it focuses on only three frameworks, namely general competition law and sector-specific regulation in the telecommunications and broadcasting sectors. Furthermore, it does not go into a discussion of which of the abovementioned authorities qualify as agencies or any other category. It simply assumes that all of the above are relevant players in the operation of economic regulation. Finally, it is concerned with EC law. ; In order to organize the discussion, a distinction will be made — to the extent necessary — between three dimensions: —

— —

Horizontal coordination refers to the coordination between authorities from the various Member States active under the same framework (e.g. amongst all NCAs or amongst all NRAs for electronic communications, etc.); Vertical coordination refers to the coordination between the Member State and European levels, here as well as in a given framework; Transversal coordination refers to the coordination between authorities active under different frameworks (e.g. between the NCA and the NRA for electronic communications).

After a brief recall of why we came to this state of affairs (1), this contribution looks at whether coordination is needed (II) and if so, how it is taking place (III).

FACTORS LEADING TO THE PROLIFERATION OF RELEVANT AUTHORITIES It is interesting to note that, while following different paths altogether, competition law and sector-specific regulation came to the same result, namely that national authorities should play the leading role in the enforcement of the law. For competition law, the path was somewhat longer. The original enforcement system was fairly centralized, with the Commission holding a monopoly on

166

Regulation through Agencies in the

EU

exemptions pursuant to Article 81(3) EC and the NCAs not being empowered by EC law itself to apply Articles 81 and 82 EC. The new enforcement system resulting from Regulation 1/2003, coming into force on 1 May 2004, marks a sea change in that Article 81(3) EC is made directly applicable (thereby terminating the whole notification and exemption system), and the NCAs are empowered by Regulation 1/2003 itself to apply Articles 81 (in its entirely) and 82 EC. The main reasons put forward for this change were that: —

After 40 years, EC competition law had been fairly well mapped out, so that the monitoring and control function of the notification system were no longer necessary to ensure a consistent development and application of the



The Commission services were overburdened with notifications which were for the most part relatively uncontroversial and of limited interest. Given that the resources made available at EC level for the enforcement of EC competition law were not going to be increased, it was necessary to reduce that burden and free up resources for truly significant cases.

law;

As a consequence, the enforcement burden was shifted in great part onto NCAs and national courts, and beyond that (since notifications cease to be available) onto the shoulders of the private parties themselves. The latter indeed now have to figure out and decide for themselves whether their course of conduct breaches competition law and risks incurring the wrath of the competent authorities. For sector-specific regulation, the situation is simpler. In line with a basic tenet of EC law, the enforcement of the new regulatory framework was left to the Member States. At least as regards electronic communications, the possibility of EC-level enforcement, through a regulatory agency (e.g. a European communications authority), was discussed a number of times, but there was never any political will on the part of Member States to depart from that basic tenet and move part of the enforcement at the EC level. Accordingly, the NRAs ended up being the first-line enforcers of EC sector-specific regulation. In both areas, a common picture emerges, namely that of national authorities (be they NCAs or NRAs) as the centre of gravity for the ‘day-to-day’ enforcement of EC law, together with national courts. The Commission remains somewhat

removed, dealing with major policy issues as well as (in competition law) the most significant cases.

THE NEED FOR COORDINATION At first sight, one might be tempted to conclude immediately that coordination between the various competent authorities is direly needed. Yet it is worth

Coordination of European and Member State Regulatory Policy

167

dedicating some attention, as a preliminary matter, to whether coordination is needed at all. Horizontal and Vertical Coordination

As a starting point, it is trite to say that the most sectors of the economy now have a global dimension, and that firms, as soon as they reach a certain size,

plan and operate at a regional if not global level. The logical consequence is that regulation should correspondingly be coordinated — at the very least — at the level at which firms operate. For a pan-European or global firm, contradictory or even incompatible decisions arising from the respective authorities of various Member States engender significant costs, since they complicate the entire range of activities, all the way from business planning down to operations. They also give rise to distortions of competition, when competing firms conduct a large part of their respective operations under differing national legal and regulatory frameworks, some of which are more favourable than others. The situation is typical of failures in the internal market, and the solution involves greater coordination at EC level and beyond. At the same time, an examination of the literature on regulatory competition illustrates that there are some reasons for refraining from coordinating the actions of national authorities more than necessary: —

Learning-by-doing: There is widespread agreement on the general competition law and regulatory principles (e.g. access on a nondiscriminatory, transparent, cost-oriented basis). However, more specific decisions are notoriously difficult to take, given policy, technical and economic considerations (e.g. What type of access must be given? Which conditions are not discriminatory? What price level qualifies as costoriented?). A number of options are open, none of which can be dismissed out of hand or picked as the obvious winner. In such cases, authorities which





have to take decisions on the same or similar issues can benefit from the experience of those which already had to commit themselves to a decision. With time, one or more best practice(s) should emerge. Closeness to the playing field: National authorities are usually in a better position to know the situation in their respective jurisdictions, if only because their resources are devoted solely to that jurisdiction. Responsiveness and flexibility: Furthermore, the various decision-making procedures available under EC law are known to take some time: coordination at the EC level is thus likely to lead to longer decision-making processes. As a result, regulatory intervention will then be less responsive and flexible, since both its onset and its termination will not follow as closely the development of the market. Even under EC competition law, where the

168

Regulation through Agencies in the

EU

Commission holds decision-making powers in individual cases, it is likely that national authorities will be in a better position to respond more quickly to problems arising within their jurisdiction. This literature has begun to find its way into the case law of the ECJ, as the Court has started to emphasize that the mere fact that national laws differ is not as such sufficient to justify harmonizing them, in the absence of evidence that the difference leads to significant barriers to intra-Community trade or appreciable distortions of competition.' In other words, there are some benefits linked with leaving matters to be dealt with at national level, and a clear case must be made that the costs of doing so offset those benefits before harmonization at EC level can be envisaged. In the end, no simple solution can be put forward to ensure the appropriate level of coordination. Nonetheless, some guiding principles can be put forward: —



The benefits of a more decentralized approach, as described above, will not

arise automatically from the mere fact that decisions are taken by national authorities. For instance, there is little to be gained if a national authority takes a decision in complete isolation, without examining what other authorities have done or are contemplating doing. Similarly, taking a different decision for political reasons (power struggle, etc.) unrelated to the situation of the market is counterproductive. It would follow that, at the very least, national authorities should be obliged to take into account the work of other authorities, and that they should give reasons, especially if they choose a different route than their counterparts. The benefits of coordination can equally be lost through overly long or complicated procedures. The coordination procedure must therefore aim to keep costs and delays to a minimum.

The amount of coordination can vary depending on the level of the decision, which ranges from general policy decisions to individual cases. At the level of general policy decisions, i.e. legislation, a strong case can be made for coordination. At the opposite end of the range, i.e. decisions in individual cases, the benefits of a decentralized approach are overwhelming. Yet one of the main characteristics of the newest regulatory framework (electronic communications’) is that a number of significant decisions are taken at an intermediate level, somewhere between legislation and individual decisions. For instance, decisions concerning the main aspects of interconnection with the incumbent (tariffs, conditions of colocation, etc.) are derived from higher legislative principles (cost-orientation, non-discrimination); at the same time, they go beyond the

individual case, since they influence the operations of the whole industry. Similarly, under competition law, a number of significant decisions will be made

Coordination of European and Member State Regulatory Policy

169

in leading cases (often processed by the Commission), whose impact is felt much beyond the parties to the case. That intermediate level definitely benefits from coordination, yet at that level coordination must be more carefully balanced against the benefits of a more decentralized approach. Transversal Coordination

While a measure of ‘regulatory competition’ within a single framework might be healthy and beneficial, the situation is different when it comes to transversal coordination. First of all, transversal coordination issues only arise to the extent that the various authorities can render decisions which diverge and cannot both be complied with. For instance, it is unlikely that transversal coordination problems would arise between the NRAs in charge of post and energy (which does not prevent their decisions from contradicting each other at a more theoretical level, e.g. in the choice of accounting standards). Transversal coordination is most likely to be an issue between the authorities in charge of applying competition law (Commission, NCAs, national courts), on the one hand, and each of the sector-specific authorities (NRAs), on the other hand. Indeed, in the current

state of EC law, competition law has taken an expansive course whereby a large number of regulatory problems (access, discrimination, pricing, etc.) can also be dealt with under competition law, thereby creating a substantive overlap between competition law and each of the main areas of SSR under study here. Moreover, transversal coordination can also be an issue between NRAs dealing with neighbouring sectors, first and foremost electronic communications and broadcasting. Secondly, the various authorities have different remits, however much their

respective regulatory frameworks might converge. Accordingly, they do not truly ‘compete’ with each other. It is conceivable that their respective decisions might diverge because each of them sees itself as having a specific mandate, with no added value coming out of this divergence in the form of learning-bydoing, closeness, responsibility or flexibility. Thirdly, unlike horizontal coordination where the national authorities exert their jurisdiction over different geographical areas, transversal coordination can involve authorities whose competence extends to the same territory. In this case, the consequences of a lack of coordination are far more severe than simply having to incur extra costs to comply with contrasting decisions in different territories. From a practical perspective, the only redeeming feature of multiple authorities exercising overlapping jurisdiction over the same territory is that claimants enjoy a choice of fora in which to pursue their grievances; in each case, they will presumably pick the one which they perceive as the most favourable to their

170

Regulation through Agencies in the

EU

claim. Obviously, this may contribute to keeping the pressure on prospective defendants to comply with applicable law and regulation. For prospective defendants, however, it can become difficult to manage legal and regulatory matters when claims can arise in a number of fora. Most significant firms are likely to fall on the claimant side in some Member States and on the defendant side in others, so that they experience both advantages and disadvantages. The risk of contradictory decisions from these various instances affects claimants and defendants alike, however. Claimants who would have erred in their choice of forum might see their competitors obtain better results in another forum. More significantly, defendants might be put in a position where they have to comply with contradictory or perhaps even incompatible rulings. In fact, uncoordinated action by the various authorities undermines the very basis for regulatory intervention: because of excessive costs arising from contradictions and incompatibilities, regulatory intervention could very well fail to deliver any overall benefit. Moreover, even in the absence of contradictions and incompatibilities, a lack

of transversal coordination can seriously undermine the efficiency of enforcement. When two or more authorities working under different regulatory frameworks are involved in settling a given case, procedures can drag on for years. While it is important to safeguard the rights of defending parties, this should not come at the expense of effective enforcement of regulatory policy and of the rights of claimants. In particular, under EC law, it must not be forgotten that, as a matter of principle, EC law must be given its full effectiveness (effet utile) by Member States, which would imply a certain level of transversal coordination. Accordingly, in contrast with the more nuanced situation as regards horizontal or vertical coordination, it would be tempting to issue a strong plea for transversal coordination.

HOW COORDINATION IS ENSURED On the assumption that coordination (horizontal, vertical and transversal) would be required, the next issue is then how it is taking place. In order to fully understand how EC law seeks to ensure coordination between regulatory authorities, it is necessary to examine both substantive and procedural avenues. Substantive Means to Ensure Coordination

Usually, one would think that the coordination of regulatory policy as between the various authorities would be conducted through institutional or procedural measures. However, both EC competition law and electronic communications

Coordination of European and Member State Regulatory Policy

171

law show that it is possible to achieve some coordination via substantive law. These measures typically ‘harmonize’ the law to be applied by the various authorities or the substantive tasks of these authorities. Under EC competition law, it is interesting to look at Article 3 (1) of Regulation 1/2003, which provides that national authorities (NCAs and courts) must apply EC competition law as soon as it is applicable in a given case (i.e. if there is an effect on trade between Member States), irrespective of whether they also apply their own national competition law as well. Even though the final version of that provision differs from the original Commission proposal,’ in practice the effect is not much different: given the primacy of EC law in case of conflict, there is a fair chance that in the long run national authorities will focus most of their attention on the application of EC law, at the expense of national law. The Commission retains a significant role in the enforcement of EC competition law. It intends to continue dealing with major cases and issue communications and notices, so that it will still be the leading authority on EC competition law. Accordingly, the Commission will be able to influence the work of national authorities by shaping the evolution of the EC competition law which all authorities are bound to apply. This can already be seen in the Commission’s Guidelines on the application of Article 81(3) EC,* where it is attempting to reshape the case law on that point in such a way as to orient the work of the NCAs and national courts. As regards sector-specific regulation, the trend is to align it in substance with competition law. Indeed, Commissioner Monti has been advocating such an alignment in recent pronouncements. This trend is most visible in the regulation of electronic communications; other regulatory frameworks find themselves at a prior stage, where the regulation is infused with economic analysis but without explicitly using competition law concepts. The mainstay of electronic communications regulation, namely the heavier regulatory framework applicable to operators with Significant Market Power, has now been aligned with competition law. It runs along the three stages of market definition, market analysis and remedies, and at each stage reference is made to the corresponding competition law concepts (relevant market, dominance, etc.). While it is not empowered to take any leading decisions in individual cases, here as well the Commission is actively shaping the work of NRAs through its Guidelines on Market Definition and Analysis® and its Recommendation on Relevant Markets.° A common point between the two areas surveyed above is therefore the use of EC competition law to minimize the influence of national competition law or to streamline the substance of sector-specific regulation. Given that the Commission remains the key player in the development of EC competition law,’ it is then in a position to effect a certain coordination of regulatory policy (horizontal, vertical and transversal) without using any institutional or procedural means.

172

Regulation through Agencies in the

EU

Some reservations should nevertheless be made. First of all, the influence of

the Commission as the coordinating instance is felt mostly through soft-law instruments. In EC competition law, soft-law instruments (notices, guidelines) have traditionally been issued on the basis of decision practice in individual cases, where the Commission views were confronted with those of interested

parties and ultimately subject to the control of the ECJ. In more recent soft-law instruments, the Commission tends to set out its views as regards newer issues which have not yet been dealt with in individual cases. These views being persuasive if not outright authoritative, they are often not challenged subsequently in individual cases.* In the absence of ‘battle-testing’, there is a risk that the quality and accuracy of soft-law instruments would decrease. Secondly, the coherency of competition law is likely to suffer if it is used as an instrument to coordinate regulatory policy across different frameworks. In the longer term, it could be that either competition law concepts become formless or that ‘sector-specific competition law’ islands begin to emerge. Recommendation 2003/311 on relevant product and service markets for the purposes of electronic communications regulation,’ for instance, pays lip service to the concept of relevant market under competition law. In fact, however, it is an exercise in identifying areas of the telecommunications sector where regulation is likely to be needed because of high and persistent barriers to entry, behind which effective competition is unlikely to arise and where the application of competition law does not appear to provide a solution. This is a solid economic approach, but it does not entirely square in with competition law. Institutional and Procedural Means to Ensure Coordination

Beyond substantive measures, coordination of regulatory policy is mostly conducted through institutional and procedural measures. Here it is convenient for the purposes of analysis to revert to the distinction between horizontal and vertical coordination, on the one hand, and transversal coordination, on the

other hand. Horizontal and vertical coordination Coordination amongst national authorities and between the authorities and the Commission was a key point in the discussions leading up to Regulation 1/ 2003 and to the new regulatory framework for electronic communications. As concerns Regulation 1/2003, the whole decentralization endeavour received frequent and vocal criticism for creating a risk that the uniform application of EC competition law would be unwound through the independent actions of the various authorities. A cacophony of diverging or even conflicting interpretations — all of them valid on their face — would arise out of the work of the NCAs and national courts. In order to counter that risk, Regulation 1/2003

Coordination of European and Member State Regulatory Policy

173

requires close cooperation between all authorities involved.'° As a last resort, should it strongly disagree with the course of action envisaged by an NCA, the Commission holds the express power to relieve an NCA from a case and take it over.'' Leaving aside these extreme cases, Regulation 1/2003 rather rests on a premise of constant and institutionalized cooperation. In particular, it provides for the creation of a network regrouping the Commission and the NCAs, the European Competition Network,'? whose task is to ensure consistency in the work of the various authorities. As regards sector-specific regulation, the new EC regulatory framework for electronic communications strengthens the position of the NRA as the main authority in charge of the application and enforcement of the law. At the same time, in order to alleviate concerns not unlike those expressed as regards the decentralization process under competition law, the work of the NRAs is subject to a number of control mechanisms in order to ensure consistency across the EU and prevent over- or underenforcement. In particular, NRAs are bound to consult the other NRAs and the Commission before taking decisions-in application under the new SMP procedure (whether concerning market definition, market analysis or remedies).'* In some cases, the Commission holds something of a veto right over draft measures from NRAs.'° Outside of these exceptional cases, the new electronic communications framework also assumes constant

and institutionalized cooperation between the authorities. It makes room for the creation of a network of regulators comprising the Commission and the NRAs, called the European Regulators Group (ERG)."° Despite differences in institutional direction’’ and substance,'* the reform of EC competition law and the new electronic communications framework evidence a number of similarities when it comes to the institutional structure: —

— — —



National authorities are at the forefront, and EC law devotes considerable

attention to their setup and powers; The Commission stays in the background and deals with the most important cases (competition law) or with the broad lines (electronic communications); The Commission uses soft-law instruments to set out its position; The Commission monitors the work of national authorities and can if necessary impose its will against a national authority if it believes that the proposed course of action of that authority would impede the consistency of Community law; The Commission and the national authories cooperate within a network that is meant to ensure consistency without the Commission having to use the hard powers mentioned just above."

Under both competition law and electronic communications regulation, there is some tension between the horizontal and vertical dimensions of coordination.

174

Regulation through Agencies in the

EU

Vertical coordination is ensured with relatively classical means — rules of conflict in favour of the Community level?’ and ultimately a power to prevent national authorities from taking decisions that would endanger consistency in the eyes of the Commission. The Commission is thus clearly placed in a superior position with respect to the NCAs and NRAs.”! Horizontal coordination, in contrast, is

entrusted to a fairly innovative institution, namely the network of national authorities. It is in the essence of such a setting that the authorities cooperate on an equal footing, as peers. Yet the Commission is also part of the network. In view of this tension, it appears too early to predict how the system of coordination will work. On the one hand, it could be that the horizontal cooperation mechanism — the network — would become prevalent (leaving the harder vertical instruments unused), so that the leadership for the evolution of the law would truly pass to the authorities acting collectively. This setup would also allow some measure of regulatory competition (which can also be construed as a flaw in coordination), since network members are not obliged to agree. On the other hand, it could be that the vertical coordination mechanisms would turn out to be used more often than originally envisaged, ensuring perhaps greater consistency but turning the national authorities into some kind of branch office. Transversal coordination It was seen above that the case for transversal coordination might be stronger than for horizontal or vertical coordination. Yet here EC law is weaker. Of course, EC law seeks to avoid conflicts between the NRA and NCA. The Framework Directive on electronic communications, for instance, provides that Member States must ensure that the NRA consults the NCA and cooperates

with it.” Moreover, the NRA and NCA must exchange the information required for the application of the regulatory framework.’ In any event, even without specific provisions in secondary law, primary EC law would oblige the NRA to work together with the NCA, since the NRA must avoid contradicting or undermining EC competition law. Most Member States have implemented the above through a cooperation agreement between the NRA and NCA, which typically assigns common cases to one of the authorities — usually the NRA, with rules to refer cases from one authority to the other accordingly. When dealing with such a case, the leading authority must consult with the other. It is interesting to look closer into such a rule to try to assess whether it is a mere priority rule or a jurisdictional rule of more fundamental significance. A few cases can help in this assessment. Firstly, the NRA can simply fail in its mission by not launching proceedings or dragging its feet. In such cases, the NCA or the Commission should be able to intervene.”

Coordination of European and Member State Regulatory Policy

— 175

Secondly, the NRA can discharge its task but reach a result with which the NCA cannot agree (or even reach an agreeable result for reasons which the NCA cannot subscribe to). From a very formalistic perspective, it could be held that the NCA should be able to intervene to restore the primacy of competition law. However, competition law can often receive many interpretations, especially on difficult issues such as arise before NRAs. It could be that the NRA’s decision is consistent with another interpretation of competition law than the NCA’s, and that both interpretations are equally reasonable. In such a case, the two authorities should settle their divergences before the court on review from the NRA decision (where the NCA could intervene in the proceedings), instead of the NCA opening its own proceedings to stake its claim to a different interpretation of competition law. Thirdly, the NRA could discharge its tasks in a manner agreeable to the NCA, but see its decision subsequently quashed by the courts. If the NRA decision is invalidated on the merits, it is difficult to see how the NCA could carry on in the same case and reinstate the NRA decision in practice, on the basis of competition

law.* If, on the other hand, the NRA decision is invalidated on formal grounds (including lack of jurisdiction on the part of the NRA), the NCA should be able to take up the case, even if the parties are then being dragged through yet another proceeding.*° In the light of the foregoing, it would seem that cooperation agreements between NCAs and NRAs cannot do more than establish a priority rule, leaving the NCA with many possibilities to intervene after, or in the place of, the NRA. From the perspective of judicial and administrative efficiency, this is disappointing. While it might be important to offer adequate avenues for the enforcement of economic regulation, sequential interventions on the part of the NRA and the NCA in the same case, each with the possibility of judicial review, brings with it considerable delays, costs and inefficiencies. Compliance with EC law could actually require more than just a cooperation agreement between the NRA and the NCA. Indeed, it is not just a matter of implementing secondary legislation (the Framework Directive), but also to ensure the effectiveness (effet utile) of EC law, as discussed earlier.*” Lengthy and complex procedures do not meet that last standard. Among alternative options, it could be possible to integrate the NRA and NCA. For instance, the UK has given its NRA the power to apply competition law in addition to regulation. This solution does not eliminate the transversal coordination problem unless the NCA sees its jurisdiction curtailed accordingly. It would be counterproductive to do so (given that the NCA is otherwise competent across the whole economy), and further it would almost certainly lead to litigation on jurisdictional grounds. The Netherlands considered a more radical option, namely to bring the NRA within the NCA (as a special chamber within the NCA). This option has the great advantage of internalizing the problem

176

Regulation through Agencies in the EU

of transversal coordination. At the same time, the NRA is also entrusted with

tasks that are less immediately related to competition law.”* A unified authority would conceivably neglect these tasks; perhaps they could be left with a ‘rump NRA’, but then the coherency of regulatory policy would also suffer. Another option could be to improve the division of labour between NRAs and NCAs on the substance. In essence, the NCA would have the last say over market definition and analysis, while the NRA would focus on remedies. Such a division of labour can be observed in certain instances. Under the electronic communications framework, the influence of competition law is greatest at the level of market definition and analysis. These two areas are mapped out for the NRAs through the Guidelines on Market Definition and SMP assessment” as well as the Recommendation on Relevant Markets*’ issued by the Commission.*' In these two areas as well, the NRAs are subject to greater control by the Commission.” Presumably, the interplay with the NCAs will also be most intense for market definition and analysis. In contrast, the NRAs enjoy more leeway as regards the imposition of remedies. The regulatory framework does not provide for the issuance of any notices on that point,** and the Commission has fewer means of control.** This state of affairs matches the evolution of competition law. It seems that the competition authorities, first and foremost the Commission, are content to

use their superior investigative powers to gather market data and conduct the substantive assessment, leaving then the NRAs with the ‘dirty work’ of working out and implementing the remedies. This division of tasks was first evidenced in the inquiries opened in 1997, concerning international telephone prices and fixed and mobile telephony prices. In both cases, the Commission gathered substantial amounts of data on the strength of its investigative powers (essentially through questionnaires), analysed that data and came to the conclusion that there were serious competition law issues. It then proceeded to urge NRAs to act on these findings and adopt adequate remedies, which they did in most cases. The Commission was then able to close these inquiries without itself having to undertake enforcement action.** More recent inquiries** concerning leased lines or mobile roaming proceeded in the same fashion.*’ Furthermore, in Newscorp/Telepiu, a merger decision at the beginning of 2003, the Commission called upon the Italian Communications Authority to supervise the proper implementation of the commitments entered into by the parties as a condition for approval of their merger.**

CONCLUSION The coordination of European and Member State regulatory activities is a complex issue, which has only begun to be addressed. This contribution

Coordination of European and Member State Regulatory Policy

— 177

attempted to structure the analysis by distinguishing between horizontal (e.g. between NCAs or between NRAs), vertical (e.g. between NCAs and the Commission) and transversal (e.g. between the NCAs and NRAs) aspects of coordination. This distinction has proven useful, since on a number of points,

the conclusion reached with respect to transversal aspects differed from that reached with respect to horizontal and vertical aspects. The latter two also tended to differ in some respects. In the academic and policy discussion so far, a number of issues are in risk of being forgotten. First of all, the need for coordination is often simply assumed, without any question being asked. However, it is not established as a general proposition that coordination between authorities is always desirable. In the horizontal and vertical dimension, in any event, the case for regulatory competition should be studied more closely. In the transversal dimension, on the other hand, the case for coordination is much clearer. Secondly, primary EC law should not be discounted. Sometimes, the detailed provisions of secondary EC law hide the general principles of primary EC law, which always remain valid. For instance, even if secondary EC law instruments tend to contain provisions designed to ensure that the rights of parties (especially the defendants) are respected, primary EC law also required that the substantive rights given by EC law to complainants are effectively implemented (effet utile) in national law. While the requirements of secondary EC law might tend to produce lengthy or complex procedures, primary EC law puts a brake on this trend. Thirdly, using the substance of EC competition law as a tool for coordination, via soft-law instruments issued by the Commission, is a risky strategy. Such instruments will often not be ‘battle-tested’, and accordingly the legitimacy of competition law — as a relatively neutral policy framework evolving through major cases (as opposed to bureaucratic design) — could be endangered. Leaving these issues aside, the procedural and institutional aspects of the ‘new’ EC competition law and the recent regulatory framework for electronic communications evidence some progress in finding appropriate solutions to coordination issues. The introduction of networks, such as the European Competition Network or the ERG, could provide the institutional vehicle to reach the equilibrium between coordination and regulatory competition amongst national authorities. In both cases, however, the creation of the network

(horizontal dimension) is offset by the ‘hard’ powers left with the Commission to enforce its position as against national authorities (vertical dimension). Accordingly, only after a few years of practical experience will one be in a position to reach a conclusion on this point. In comparison, the mechanisms put in place to conduct transversal coordination — which in theory is more desirable than horizontal or vertical coordination — still leave a lot to be desired.

178

Regulation through Agencies in the

EU

NOTES 1 See Germany v. Parliament (Tobacco Advertising) (Case C-376/98), E.C.R., 2000, I-8419 and subsequent case law. 2 The same can be said of other regulated sectors, such as utilities (energy, post), transport of financial services. 3 Which provided that once there was an effect on trade between Member States, ‘Community competition law shall apply to the exclusion of national competition laws’. 4 Guidelines on the application of Art. 81 (3) of the EC Treaty, O.J. C 101 of 27 April 2004, p. 97. 5 See Commission guidelines on market analysis and the assessment of significant market power under the Community regulatory framework for electronic communications networks and services, O.J.C 165 of 11 July 2002, p. 6. 6 See Recommendation 2003/311 of 11 February 2003 on relevant product and service markets within the electronic communications sector susceptible to ex ante regulation in accordance with Directive 2002/21, O.J. L 114 of 8May 2003, p. 45. 7 It is too early to tell whether the new enforcement system of Regulation 1/2003 will jeopardize this position, see infra.

8 Often, Commission notices or guidelines will precisely lead parties to avoid litigation and to adopt the solution set out therein instead. 9 See supra, note 7. 10 See Regulation 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, O.J. L 1 of 4January 2003, p. 1, Art. 11, 15. 1! Note that, for constitutional reasons, no similar power exists with respect to national courts.

12 Mentioned in Regulation 1/2003, supra note 10 and created in October 2002. 13 Tt can be noted that national courts are left out of this network, here as well for obvious

constitutional reasons. 14 European Parliament and Council Directive 2002/21 of 7 March 2002 on a common regulatory framework for electronic communications networks and services (Framework Directive), O.J. L

108 of 24 April 2002, p. 33, Art. 7. 15 Directive 2002/21, ibid., Art. 7(4). This veto right concerns draft NRA measures that would

stray from the Commission’s recommendation concerning relevant markets or that would designate SMP operators (or abstain from doing so). See also Recommendation 2003/561 of 23 July 2003 on notifications, time limits and consultations provided for in Article 7 of Directive 2002/21/EC of

the European Parliament and of the Council on a common regulatory framework for electronic communications networks and services, O.J. L 190 of 30 July 2003, p. 13. The Commission recently used this right for the first time to require the Finnish communications authority (FICORA) to withdraw a draft measure. '6 Directive 2002/21, ibid., Rec. 36. The ERG was formally created through Decision 2002/627 of 29 July 2002 establishing the European Regulators Group for Electronic Communications Networks and Services, O.J. L 200, p. 38.

'7 The reform of EC competition law is meant to decentralize enforcement, while the new electronic communications framework rather amounts to a re-balancing, with the Commission getting more power over the actions of NRAs. '8 The core of EC competition law are directly effective EC Treaty provisions, while electronic communications regulation is secondary EC law, which must be implemented in national law. '9 This network can also have some noteworthy positive effects on the independence of its members as against both national administrations and market players (capture). By acting together, the network members see their positions on their respective national scenes strengthened. 20 Regulation 1/2003, see note 10, Art. 3 (2) for competition law, and for electronic communications regulation, the obligation to heed EC competition law, which given the substantive alignment with

competition law can also amount to a rule of precedence in favour of decisions taken at EC level. 21 The situation is different for national courts, as discussed above. 22 Directive 2002/21, see supra note 14, Art. 3(4). 23 Tbid., Art. 3 (5). This provision is bizarrely worded, given that the NCA is not entrusted with the application of electronic communications, so that it is difficult to see why the NRA would disclose

Coordination of European and Member State Regulatory Policy

— 179

information to the NCA for that purpose. 4 Indeed this is the point of view of the Commission in the Notice of 22 August 1998 on the application of competition rules to access agreements in the telecommunications sector — framework, relevant markets and principles, O.J. C 265, p. 2, para. 11 and ff. 5 Unless sector-specific regulation so differs from competition law on the issue at stake, in that case that the NCA could base its decision on reasons that would not contradict the judgment that invalidated the NRA decision. In view of the trend towards the alignment of regulation and competition law, this appears unlikely. 26 The case of mobile termination tariffs in the Netherlands offers a perfect example of this latter situation. 27 This principle is central to the ECJ case law, and it flows directly from primary EC law (amongst others, Art. 10 of the EC Treaty).

28 Such as the administration of the universal service regime and the tasks relating to scarce resources.

29 Supra, note 6. 30 Supra, note 7. 3! Within the Commission, it is also significant to note that these two documents were elaborated by, or under the close supervision of, DG COMP. 32 See Directive 2002/21, supra note 14, Art. 7 (4), which gives the Commission a right to veto draft NRA measures which would stray from the Recommendation on relevant markets or which designate (or fail to designate) SMP operators. Here as well, it should be noted that the treatment of notifications concerning these two issues is primarily entrusted to DG COMP and not DG INFSO, within the joint task force set up by the two DGs. 33 See the ERG Common position on the approach to appropriate remedies in the new regulatory framework, available at http://erg.eu.int, which attempts to map out the exercise of that discretion. 34 Draft NRA decisions concerning remedies fall outside of Art. 7 (4) of Directive 2002/21, supra note 14 and are therefore subject to the consultative procedure of Arts. 7 (3) and (5) only.

35 See ‘Commission sees substantial progress in its investigation into international telephone prices’, Press Release IP/99/279 (29 April 1999) and ‘Commission successfully closes investigation into mobile and fixed telephony prices following significant reductions throughout the EU’ Press Release IP/99/298 (4 May 1999). 36 These were sector inquiries conducted pursuant to the Commission’s competition law powers, formerly Art. 12 of Regulation 17, O.J. of21 February 1962, Art. 17 of Regulation 1/2003, supra note 10. 37 See the statements contained on the DG COMP’s website at http://europa.eu.int/comm/ competition/antitrust/others/sector_inquiries/local_loop/ and http://europa.eu.int/comm/competition/ antitrust/others/sector_inquiries/roaming/ respectively. 38 Decision of 2 April 2003, Case COMP/M.2876, Newscorp/Telepiu, O.J. L110 of 16 April 2004, pp. 73-125.

9.

The Proliferation of National Regulatory Authorities alongside Competition Authorities: A Source of Jurisdictional Confusion? Nicolas Petit

INTRODUCTION Within the framework of the debate on good governance in the European Union (hereafter, the ‘EU’), the promotion of swift, efficient and predictable regulatory mechanisms should unequivocally occupy a prominent position. In the field of network industries, the advent of National Regulatory Authorities

(hereafter, the ‘NRAs’) in the 1990s and their subsequent improvements has, to a large extent, contributed to the achievement of these regulatory standards. Yet, a closer look at the current situation tends to suggest that the ‘proliferation’ of NRAs alongside competition authorities (i.e. the European Commission and the National Competition Authorities) could paradoxically compromise the attainment of this goal. Indeed, as will be seen below, the proliferation of

NRAs with increased powers may generate a risk of jurisdictional conflicts with existing competition authorities as well as regulatory inconsistencies and therefore potentially run against the tide of “better governance’ pursued by the EU. A snapshot approach would lead to the bold conclusion that, for a variety of reasons, the risks of overlap between the missions of NRAs and competition authorities are limited. First, it could be argued that NRAs and competition authorities are formally entrusted with the implementation of two distinct sets of rules that apply in distinct situations (the former’s mandate is the application of sector-specific legislation while the latter’s mandate is the application of competition rules). Second, it could also be considered that overlaps between NRAs and competition authorities are limited because the

180

The Proliferation of National Regulatory Authorities

181

former generally intervene on an ex ante basis through the imposition of detailed and adjustable remedies (or the adoption of guidelines), while the latter essentially carry out ex post enforcement duties to eliminate actual abuses of market power through the adoption of prohibition decisions, as well as fines.” Finally, the argument could be advanced that competition authorities enjoy an exclusive jurisdiction over certain matters which are hence left out of the reach of NRAs.* These distinctions are, however, oversimplistic because they do not reflect the true face of reality. On the contrary, the differences between NRAs and competition authorities become increasingly blurred due to several factors.’ First, the growing convergence between competition law and sector-specific regulation leads to substantial overlaps between the NRAs’ and the competition authorities’ respective mandates. On the one hand, the progressive opening of markets and the elimination of bottlenecks accelerates the transition of network industries towards competitive markets. To complete this evolution, EC legislative reforms are bringing sector-specific regulation closer to competition law standards (e.g. in the telecommunications sector, the central concept of significant market power is identical to the notion of dominance under competition law). On the other hand, the extensive interpretation of the competition rules of the EC Treaty (e.g. the emphasis put on ‘access-based’ competition) and the development of ‘quasi regulatory’ enforcement policies by competition authorities extends the reach of classic antitrust rules to situations that are already subject to sector-specific regulation.* Second, in parallel to this substantive evolution, NRAs are increasingly entrusted with powers that resemble the powers enjoyed by competition authorities. The growing importance of the NRAs’ dispute settlement missions creates, for instance, a new procedural avenue for complainants. The overlaps that have just been exposed may become a critical issue in the coming years, especially when conjugated with a third factor. The proliferation of NRAs across all network industries sectors through the impetus of EC liberalisation directives and the increased decentralisation of competition rules towards national competition authorities (hereafter, the ‘NCAs’) and national courts could have a multiplier effect in terms of jurisdictional conflicts.° The discussion below is organised as follows. Following this introduction, Part Il examines the increasing overlaps between the NRAs and the competition authorities’ mandates as well as the risks they entail. Part II] deals with the mechanisms that have been adopted for limiting the risks of regulatory inconsistencies and conflicts of jurisdiction. Part IV identifies the remaining hypothesis where NRAs and competition authorities can both claim jurisdiction. Part V suggests alternative approaches to deal with the risks of cumulative jurisdiction. Part VI proposes some conclusions.

182

Regulation through Agencies in the

EU

THE INCREASING OVERLAP BETWEEN THE NRAS AND THE COMPETITION AUTHORITIES’ MANDATE The evolving nature of sector-specific regulation and competition rules increases the number of situations that can potentially fall within the scope of both sets of rules and trigger the concurrent jurisdiction of NRAs and competition authorities. In addition to this, the growing similarities of the missions entrusted to NRAs and competition authorities leads to overlaps in terms of jurisdiction. This creates a risk of regulatory inconsistency as well as a risk of jurisdictional confusion. Increasing Substantial Overlaps between Sector-specific Regulation and Competition Rules

A substantial number of issues that are at the core of sector-specific regulation can be translated under competition law terms and dealt with on the basis of competition rules.’ The best illustration of this can be found in the field of ‘access regulation’. Most sector-specific frameworks contain provisions requiring incumbent operators to grant effective access to their networks and set out conditions for access. In the field of telecommunications, for instance, Articles 9, 10 and 13 of the Directive 2002/19 (hereafter, the ‘Access

Directive’) impose a variety of transparency, non-discrimination and pricing requirements on operators enjoying a Significant Market Power (hereafter, ‘SMP’).8 In the field of energy, similar network access conditions are laid down under the so-called Third Party Access regime (hereafter, the “TPA’), which allows third parties (e.g. gas producers and suppliers that do not control the grid) to use the existing gas pipelines so as to reach gas consumers.” In parallel to this, the emergence of the ‘essential facilities’ doctrine in the field of competition law has been a powerful instrument for competition authorities to equate the control of an infrastructure with the concept of dominance and to impose on incumbents a general obligation to satisfy access demands pursuant to Article 82 EC.'° In addition, Article 82 EC has occasionally been used to define the conditions under which (timeliness, quality etc.) an undertaking controlling an essential facility should give access to its infrastructure.'' For instance, infrastructure owners giving access can be required under Article 82 EC to disclose technical information on their network." Finally, access pricing could potentially be regulated through Article 82 EC, especially in situations where excessive prices would be charged for

access to the network.'’ Access matters may hence be analysed through the spectrum of both sets of rules. Besides the issue of access regulation, a larger range of classic regulatory issues can be examined through the lenses of competition rules. This is, for instance, the case of the sector-specific

provisions that seek to ensure

The Proliferation of National Regulatory Authorities

183

accounting separation and structural separation with a view to avoiding the financing of competitive activities through monopoly rents.'* Similar requirements are being imposed under Article 82 EC (and Article 81 EC) in cases involving cross-subsidisation practices by dominant undertakings.'> Another example is the requirements, under sector-specific regulation, that vertically integrated undertakings do not discriminate between their subsidiaries and third parties.'® These situations can equally be analysed under competition rules. In the postal sector, for instance, the SVEPLD decision illustrates the possible application of Article 82 EC to discriminatory practices implemented by a postal incumbent when granting access to services of the postal monopoly to third parties.'’ Furthermore, in several Member States (hereafter ‘MS’), the question of access to subscribers’ lists is subject to regulation under national sector-specific frameworks.'* In parallel, the Commission’s /TT Promedia/ Belgacom decision made clear that the refusal to provide, on reasonable terms, data about customers of telephone services to a new entrant could amount to an

abuse of a dominant position within the meaning of Article 82 EC.'° Yet, new areas of overlap between the two bodies of norms have recently been brought to light. In addition to classic market regulation, sector-specific frameworks tend to increasingly take into account additional considerations such as, for instance, the protection of the environment or public health. To meet these standards, the telecommunications regulatory framework contains provisions on facilities and property sharing between operators.” Under these provisions, NRAs may encourage or require operators to share network elements (e.g. masts, antennae, support cabinets, power supplies etc.).”! In parallel to this, infrastructure-sharing strategies are often implemented through agreements between operators which thus fall within the scope of Article 81 of the EC Treaty. Often, the purpose of these strategies is to reduce the initial fixed costs of rolling out the network while ensuring maximal geographical coverage. Recently, two network infrastructure-sharing agreements were notified to the Commission. The network-sharing provisions were deemed not to fall within the scope of Article 81(1) EC because it concerned only a limited part of the infrastructure.*? However, the two cases provide very detailed guidelines on the conditions for implementing colocation and site-sharing arrangements.” Overlaps between the two bodies of norms may thus, in the near future, continue to increase.” Increasing Jurisdictional Overlaps between NRAs and Competition Authorities

Besides the large and growing area of overlap between sector-specific regulation and competition rules, NRAs

and competition

authorities tend to offer

increasingly substitutable remedial avenues for operators. Traditionally,

184

Regulation through Agencies in the

EU

competition authorities provide a natural venue for settling disputes arising between several operators or for acting on consumers’ requests. Because a substantial number of sector-specific regulation issues can potentially fall within the scope of Article 81 (provided that an agreement was concluded) or Article 82 (in case of refusal of access or of abusive conditions), competition authorities may deal with disputes between parties or of consumers complaints.” An example of this can be found in field of energy. In the Marathon case, Thyssengas GmbH, the German gas company had refused to grant TPA to its gas pipelines to Marathon, a Norwegian gas producer. Marathon lodged a complaint before the Commission on the basis of Article 82 EC. The latter brought the two companies to a commercial settlement.”° In parallel to this, the growing importance of the NRA’s dispute settlement functions creates a new jurisdictional route for complainants.”’ In the field of telecommunications, NRAs were already entrusted with a dispute settlement function by the first generation directive.** These functions were, however, reinforced by the new regulatory framework which provides that: In the event of a dispute arising in connection with obligations arising under this Directive or the Specific Directives between undertakings providing electronic communications networks or services in a Member State, the national regulatory authority concerned shall, at the request of either party, and without prejudice to the provisions of paragraph 2, issue a binding decision to resolve the dispute in the shortest possible time frame and in any case within four months except in exceptional circumstances. The Member State concerned shall require that all parties cooperate fully.?9

Some elements suggest, however, that the scope of the dispute settlement functions may be larger than disputes between operators on interconnection issues. The preamble of the Framework Directive implies that interconnection and access issues are only an example of possible disputes falling under the dispute settlement procedure.*° This is confirmed at the national level, where some MS have included, within the scope of the dispute settlement procedure, a wide range of issues that go beyond access matters. In France, for instance, the ART can act as a dispute settlement body on interconnection issues, but also with regards to site-sharing agreements, financial and technical conditions for transfers of subscribers list, cable related issues, etc.*!

Similar dispute settlement procedures were established in other sectors.*? In the field of energy, for instance, Article 20(3) of Directive 96/62 required that MS designate a competent authority (not necessarily the NRA) to settle disputes relating to the contracts and negotiations over access issues (TPA) and refusals to purchase.** The dispute settlement procedure was strengthened in Directive 2003/54 which provides that these missions should be entrusted to NRAs and imposes time limits for the settlement procedures.™ In a fashion similar to that

The Proliferation of National Regulatory Authorities

185

of the telecommunications sector, the concept of conditions of access was given a very broad meaning in some MS. In France, for instance, it includes the

interpretation and implementation of the conventions, as well as contractual liability claims.*> A further element that brings the NRAs closer to competition authorities is that several MS have entrusted the former with sanctioning powers (e.g. the imposition of fines or the revocation of authorisations). NRAs can thus impose sanctions on operators that fail to comply with the requirements of the regulatory framework or that do not implement its decisions. In Germany, France and Spain, the telecommunications regulators (and, for the latter, the minister) enjoy a sanctioning power.*° This constitutes an additional judicial remedy for victims of an infringement of the sector-specific legislation. As would be the case before a competition authority, complainants can lodge a complaint before the NRAs in order to deter the operator from infringing the rules through the imposition of a sanction. The Potential for Regulatory Inconsistencies and Jurisdictional Confusion between NRAs and Competition Authorities The existence of several jurisdictional routes for complainants creates two series of problems. First, on material grounds, although the similarities between sectorspecific regulation and competition rules limit the risk of conflicts (both sets of rules pursue related and complementary objectives), there is nonetheless a possibility that the decisions adopted under each body of rules differ in the medium run and lead to regulatory inconsistencies.*’ Indeed, the approach under sector-specific regulation may differ from the approach under competition rules.*® In the field of access regulation, for instance, NRAs and competition authorities may reach different decisions on whether to require an incumbent to grant access to a new entrant. On the one hand, provided that the requirements for applying Article 12 of the Access Directive are met (i.e. the relevant market is not effectively competitive and the operator enjoys a significant market power), the NRA can require the incumbent to satisfy all reasonable requests for access, ‘where the denia! of access would hinder the emergence of a sustainable competitive market at the retail level or would not be in the end-user’s interest’. On the other hand, competition authorities will apply Article 82 EC in line with the twofold test set out by the ECJ in Bronner.*? Competition authorities will only require the incumbent to grant access if (i) the refusal to grant access is likely to ‘eliminate all competition in the downstream market’ and cannot be objectively justified and (ii) the good, service of infrastructure in question is ‘absolutely indispensable’ for the carrying out of the economic activity. The conditions to be met for obtaining access are more stringent under competition law.*° Thus, there is a possibility that NRAs and competition authorities reach

186

Regulation through Agencies in the

EU

different outcomes in similar cases (i.e. most likely the NRA requiring the incumbent to grant access while the competition authority would refuse to condemn the refusal to grant access by the incumbent).*! Also, with regard to the regulation of access conditions, sector-specific frameworks often take into account the investments incurred by the owner of the infrastructure, the existence of intellectual property rights or the necessity to preserve competition in the long run.” In contrast, competition rules are less concerned with these objectives.** The ‘essential facilities’ doctrine does not mention the conditions under which access should be given. Competition authorities are primarily concerned with the elimination of actual restrictions of competition. Hence, they generally give little importance to the necessity that access conditions ensure sufficient rates of return on investments and do not undercut incentives for innovation etc. Therefore, NRAs may legitimise access conditions that a competition authority would on the contrary consider as incompatible with Article 82 EC. Conversely, while infrastructure sharing is encouraged under sector-specific frameworks for environmental or public health reasons, these considerations

may be less relevant for an assessment under competition rules. Site-sharing arrangements might thus, in the medium run, receive a less positive assessment under competition law. The issuance of diverging decisions under both sets of rules may lead to regulatory inconsistencies which can limit the effectiveness of regulatory reforms.* In addition, confused regulatory signals to the industry could prove potentially harmful for investments.* Finally, the existence of regulatory divergences between NRAs and competition authorities creates a risk of forum shopping for complainants.*” Second, on the jurisdictional ground, the existence of a variety of remedial routes for complainants is not coordinated by a rule of case allocation. On the contrary, the first generation Directives (i.e. the first sets of Directives that were adopted at the beginning of the liberalisation process) provide that the regulatory frameworks are ‘without prejudice of the application of competition rules’.“* A possibility thus exists that a similar case be simultaneously or successively analysed by NRAs and competition authorities. In addition, EC-sector-specific frameworks also provide that the initiation of a dispute settlement procedure before NRAs does not preclude parties: (i) from bringing an action before the courts; and (ii) from their rights of appeal under national and EC law.*? The reference to appeal under EC law is all the more curious. The Treaty of Rome does not provide any avenue for appealing decisions adopted at the national level. This provision is arguably a reference to the possibility to lodge a complaint before the Commission. Thus, a complainant is entitled to bring a similar action before several institutions, i.e. the NRAs, the NCAs, the Commission and national courts.°°

Also, there is a possibility that distinct complainants each bring closely related

The Proliferation of National Regulatory Authorities Table 9.1

187

Overview of Potential Divergences between NRAs and Competition Authorities Approaches Competition Authorities (Commission and NCAs)

Conflict of Rules/ Decisions?

Not Favourable to

More Favourable

No — Granting Access

Incumbent — Large Access for New Entrants — Article 12 of Access Directive

to Incumbent — More difficult for

Potentially Favourable to Incumbent — Return on Investments etc.

Less Favourable Potential — Excessive to Incumbents— — Access Pricing can be Not Concerned by an Infringement of Returns on Article 82 Investments etc.

NRAs

Grant of Access

by Incumbent

Conditions of Access

Infrastructure Sharing

Favourable — Environment, Public Health

New Entrants — Strict Conditions in Bronner

cannot be an

Infringement of Article 82 — Article 82 does not apply

Less Favourable — Less Concerned

Potential — Agreements containing Restrictions of Competition

cases (e.g. several interconnection conventions with a same incumbent operator) before distinct authorities or courts. The absence of rules of exclusivity, /itis pendentia and joining of claims creates a risk of duplication of proceedings. Besides potential abuses of predatory litigation,*' this situation is unsatisfactory because it may in turn generate delays, higher transaction costs and increased legal uncertainty for operators. From a public policy standpoint, it is also costly because it creates a risk of duplication of investigative resources.

MECHANISMS FOR LIMITING THE RISKS OF CONCURRENT JURISDICTION OF NRAS AND COMPETITION AUTHORITIES To address the risk of regulatory inconsistencies and ofjurisdictional conflicts, a variety of rules and principles have been established. First, a

188

Regulation through Agencies in the

EU

general duty of observance of EC competition rules has been placed on the NRAs through the case law of the European Court of Justice (hereafter, the ‘ECJ’) and sector-specific frameworks. Second, several specific mechanisms have been adopted at the EC and MS levels in order to prevent the risks of conflicts. The NRAs’ General Duty to Observe Competition Rules EC legislation in the field of network industries merely specifies that the mandate of the NRAs is the application of the regulatory tasks assigned in the specific directives through the application of the national rules of transposition.* As a result of this, NRAs have generally no jurisdiction to apply competition law.* However, this is not to say that NRAs are under no obligation to pay attention to the competition rules of the EC Treaty. On the contrary, the case law of the ECJ has made it clear that NRAs fall under a general duty of observance of EC competition rules when dealing with matters within their jurisdiction. In the landmark GB-Jnno-BM/ATAB case the ECJ held that pursuant to Article 10 EC: “While it is true that Article 86 (now 82) is directed at undertakings, nonetheless it is also true that the Treaty imposes a duty on Member States not to adopt or maintain in force any measure which would deprive that provision of its effectiveness’.** In addition to this, the ECJ considered that Article 86 EC implied that: In the case of public undertakings and undertakings to which Member States grant special or exclusive rights, Member States shall neither enact nor maintain in force any measure contrary inter alia to the rules provided for in Articles 85 (now 81) to 94 (now 90). Likewise Member States may not enact measures enabling private undertakings to escape from the constraints imposed by Articles 85 (now 81) to 94 (now 90).>5

These principles were confirmed and clarified by subsequent case law. In Ahmed Saeed, the ECJ evoked the interface between sector-specific legislation on the one hand, and the competition rules of the Treaty, on the other hand. The ruling provides interesting indications on the duties of sector-specific authorities. The ECJ held that where EC secondary legislation exists in a sector and leaves MS authorities free to encourage mutual consultations between operators: ‘the Treaty nevertheless strictly prohibits them from giving encouragement, in any form whatsoever, to the adoption of agreements or concerted practices with regard to tariffs contrary to Article 85(1) or Article 86, as the case may be’.*® A more recent case made it clear that such obligations rested on all sorts of public administrative authorities and entailed the duty to leave unapplied national legislation running counter to EC law. In the Italian Matches case the court held:

The Proliferation of National Regulatory Authorities

189

[...] In accordance with settled case-law the primacy of Community law requires any provision of national law which contravenes a Community rule to be disapplied, regardless of whether it was adopted before or after that rule. The duty to disapply national legislation which contravenes Community law applies not only to national courts but also to all organs of the State, including administrative authorities. [...], which entails, if the circumstances so require, the obligation to take all appropriate measures to enable Community law to be fully applied [...].57

The EC legislator discretely transposed these principles into sector-specific frameworks through the insertion of subtle references requiring NRAs to ensure the promotion of competition in general, as well as the protection against restrictions of competition in their respective sectors.** In the field of telecommunications, for instance, the Framework Directive provides that NRAs shall: promote competition in the provision of electronic communications networks, electronic communications services and associated facilities and services, by [inter

alia], ensuring that there is no distortion or restriction of competition in the electronic communications sector.*?

Similarly, in the electricity and gas sectors, NRAs shall monitor the level of competition and are responsible for ensuring ‘effective competition’.© The ECJ case law, combined with these references, clearly requires that NRAs shall not, for matters falling within their jurisdiction, frustrate the ‘effet utile’ of EC competition rules. This does not, however, entail an obligation for NRAs to

actively apply competition rules.*' Specific Mechanisms of Conflict Avoidance

The duty for NRAs to observe the provisions of the EC Treaty certainly limits the risks of regulatory inconsistencies in terms of decision making. However, it does not eradicate the possibility that a complainant brings a similar case before a NRA on the one hand, and a competition authority on the other hand (the Commission or a NCA).” To avoid this, the Commission clarified its position with the adoption of the Access Notice in the field of telecommunications. This document contains a number of principles for avoiding duplication of proceedings that can, given the similarities of issues, be transposed to other sectors such as, for instance, the energy sector.® In parallel to this, EC second generation directives (i.e. the sets of Directives replacing the first generation Directives with a view to push market liberalisation a step further) and MS have tried to strengthen cooperation between NRAs and competition authorities at the national level.

190

Regulation through Agencies in the

EU

The Commission’s approach for avoiding conflicts with NRAs The Commission’s doctrine regarding its relationship with NRAs was clearly set out in the Access Notice.“ One of the main purposes underlying the adoption of this document was to avoid the risk of duplication of proceedings or of investigative efforts by the Commission and by NRAs.® To this end, the Commission considers that a ‘/ex specialis’ principle should prevail. It will therefore not act on a complaint if parallel procedures are running before the NRA (or a national court or the NCA) unless: (i) the matter is not solved within a reasonable period of time (6 months); or (ii) adequate interim relief is not available to the complainant in national proceedings.® Furthermore, it is of note that the Commission considers that it retains the possibility to take action, in exceptional circumstances where the case involves substantial Community interests.° Arguably, a similar /ex specialis approach has been implemented in other network industries. In the field of energy, for instance, the HFC Bank plc/British Gas Trading Ltd case provides an illustration of this principle.®* In 1997, HFC Bank and British Gas Trading entered into a joint venture agreement for the launch of the so-called Goldfish card in the United Kingdom (hereafter, the ‘UK’). The Commission identified a risk of tying and wrote to Ofgas (the UK gas regulator), asking for its view on the agreement. Ofgas replied that it was already investigating the case. The Commission then decided to stay proceedings and to await Ofgas’ final decision. Following the finding by Ofgas that the risk of tying was unlikely, the Commission decided to close the case. This case provides a clear illustration that the Commission gives priority to the specialised authority and stays proceedings when an NRA is already investigating a case. It also shows that the Access Notice’s doctrine can be transposed in the field of energy. A further illustration of the Commission’s tendency to promote a /ex specialis principle can be found in the Synergen case. In the course of the year 2000, several joint venture agreements between ESB (the Irish dominant electricity producer) and Statoil (a Norwegian gas company) were notified to the Commission. To address the competition concerns raised by the Commission under Article 81(1), the parties offered various commitments. The interest of the case lies in the fact that the Commission relied on the Irish electricity regulator (the Commission for Electricity Regulation) for carrying out the settlement negotiations and for ensuring the monitoring of the remedies.” The development of a /ex specialis doctrine between the Commission and NRAs across different network industries is to be welcomed for a variety of reasons. First, the priority given to the specialised authority in case of parallel procedures lessens the risks of duplication of proceedings. Second, regulatory matters and in particular access regulation often involve pricing issues (and more generally access conditions issues) which NRAs are better placed to deal

The Proliferation of National Regulatory Authorities

19]

with than competition authorities.”? Third, access regulation may require regular adjustments that NRAs are better placed to ensure through their continuing oversight of a specific industry. Fourth, NRAs are generally required to take decisions within tight timeframes (e.g. four months in the field of telecommunications in a case of dispute, two months in the field of energy, ten days in the field of rail) whereas a competition authority is not obliged to respect time limits when investigating complaints.”! In spite of the development of a Jex specialis relationship between the Commission

and NRAs, three points remain, however, unclear. First, the

Commission’s practice offers no guidance on the question of the relations between NCAs and NRAs (whether a similar principle should apply in case of parallel proceedings before national authorities).”” Second, the Commission gives no clarification about the concept of ‘exceptional circumstances’. Third, the Commission does not explain how it will act if the NRAs’ intervention leaves competition concerns. Mechanisms to avoid the duplication of proceedings at the national level Because it is not clear whether a similar /ex specialis principle applies at the MS level, the risk of multiple proceedings before NRAs and NCAs remains high. To address this issue, most MS have, however, spontaneously designed cooperation procedures at the national level. These cooperation mechanisms either take the form of law, and/or take the form of ‘agreement’, ‘protocols’ or ‘memorandum of understandings’ between NCAs and NRAs.” A first mechanism that exists in most MS requires NRA to inform and/or seize the competition authority when the former has knowledge of anticompetitive practices in the sector they regulate.” For instance, in 2000, the ART (the French telecommunications regulator) examined the tariffs charged by France Telecom for a number of fixed line services. The ART informed the Competition Council (the French NCA) that the commercial offer was likely to constitute an abuse of a dominant position and decided to defer the case to the latter.” Ultimately, the Competition Council came to the conclusion that France

Telecom’s behaviour could be qualified as an abuse of a dominant position and imposed provisional measures on France Telecom. Reciprocally, a second mechanism requires NCAs to communicate with the NRA when the former are called to rule on a dispute that falls within the jurisdiction of the latter. The intensity of this collaboration is not the same in all cases. In a majority of cases, this leads to the consultation of the NRA and/or to the possibility of producing a report.’ In other cases, the NCA will rely on the NRA for some aspects of its decisions.” Finally, in some circumstances, this

leads to the referral of the case by the NCA to the NRA.” Third, even MS that opted for the option of entrusting the regulator with the enforcement of the competition rules have designed cooperation mechanisms.

192

Regulation through Agencies in the

EU

In the UK, a rule of priority determines whether the NRA or the NCA must deal with the case and reciprocal consultation requirements are set up. In case of conflict between the two authorities, the minister determines which authority shall have jurisdiction.” A majority of observers consider that these procedures have worked efficiently and that NRAs and NCAs do, in practice, effectively collaborate.*° At the EC level, awareness of the usefulness of such procedures is more recent. The second generation Directives in the field of telecommunications provide that: Member States shall ensure, where appropriate, consultation and cooperation [...] between those authorities [i.e. the NRAs] and national authorities entrusted with the implementation of competition law [...] on matters of common interest. Where more than one authority has competence to address such matters, Member States shall ensure that the respective tasks of each authority are published in an easily accessible form.®!

The Access Directive also contains interesting provisions. The recital’s formulation implicitly acknowledges the existence of a problem where cumulative action by NRAs and NCAs would lead to the imposition of overlapping and burdensome obligations. NRAs and NCAs are called on to cooperate on issues of access regulation and shall coordinate their action to ensure that the most appropriate remedy is applied.” The EC’s clutch on cooperation between NRAs and NCAs remains nonetheless limited to the telecommunications sector as well as it being constrained by the ‘institutional autonomy’ principle.® In sum, the development of cooperation mechanisms at both EC and national levels certainly limits most of the risks of multiple proceedings before NRAs and competition authorities (Commission and NCAs). A recent case suggests, however, that in spite of these mechanisms, NRAs action could be subsequently superseded by a competition authority’s action on the ground that it would leave competition concerns.

REMAINING HYPOTHESIS OF CUMULATIVE JURISDICTION OF NRAS AND COMPETITION AUTHORITIES The recent Commission decision in the Deutsche Telekom case provides a good illustration of the possibility of duplication of proceedings before an NRA and a competition authority.“ The approach followed by the Commission in this case suffers a number of drawbacks.

The Proliferation of National Regulatory Authorities

193

The Deutsche Telekom Case

The Deutsche Telecom decision is a typical example of a situation where a NRA applies a sector-specific remedy but leaves a number of competition concerns. This case concerned the prices charged by Deutsche Telekom (hereafter, ‘DT’) to its competitors and consumers for access to the local loop between the end of 1998 and 2002. The local loop is the physical circuit between the consumer’s premises and the telecommunications operator’s local switch. This infrastructure is generally controlled by the incumbent operator and new entrants need access on fair and non-discriminatory terms to the loop in order to offer retail services to consumers and compete with the incumbent. In March 1999, several of DT’s competitors had lodged complaints before the Commission arguing that DT’s prices for access to the local loop were incompatible with Article 82 EC. The prices charged by DT for competitors’ access on the wholesale local loop market were higher than the prices it charged to its own subscribers on the retail market. DT thus prevented new entrants from competing on the retail local loop market and deterred entrance on the market.*° DT argued that its conduct could not be held as an infringement of Article 82 EC in that its tariffs had previously been approved by the German telecommunications regulator, the RegTP.*’ DT considered that the Commission was not entitled to proceed against charges that have been the subject of regulatory decisions at the national level, and that if there was a violation of EC law, the appropriate course of action was the initiation of Article 226 EC infringement proceedings against Germany in spite of a direct procedure against the undertakings whose charge had been approved by the regulator. These arguments were rebutted by the Commission: the competition rules may apply where the sector specific legislation does not preclude the undertakings it governs from engaging in autonomous conduct that prevents, restricts or distorts competition.**

In fact, the Commission observed that the charges for wholesale and retail access to the loop were regulated but that DT was still left with a commercial discretion which allowed it to restructure its tariffs in order to put an end to the margin squeeze.*® DT could have avoided the price squeeze by increasing the retail charges, which DT actually did but to an insufficient extent. In addition, even when it enjoyed a limited commercial discretion, DT could have raised the prices of other non-regulated charges. Thus, the Commission considered that DT’s behaviour was the main source of the restriction of competition and that Article 82 EC was the appropriate course of action. DT was condemned to a €12.6 million fine. The prior application of a regulatory remedy was not admitted as a cause of justification. It was merely taken into account as a mitigating factor for the calculation of the fine.”°

194

Regulation through Agencies in the

EU

Drawbacks of the Commission’s Approach in Deutsche Telekom

The DT case provides a good example of the application of a sector-specific remedy by a NRA which nonetheless leaves concerns with regards to competition rules. The approach taken by the Commission is in line with the provisions of the Access Notice because the Commission let the RegTP deal with the case. However, the decision to subsequently open new proceedings against the regulated undertaking can be criticized in that it entails a series of negative consequences for regulated entities as well as for regulators.”' As far as the operators are concerned, the approval of tariffs, access conditions or the settlement imposed by a NRA does not mean that they cannot be prosecuted twice on similar issues, before a competition authority. In addition to high legal uncertainty, this could raise transaction costs through a multiplication of expensive litigation, increase further delays and have a disincentive effect on investments.” Also, the possibility that two remedies (potentially different) be imposed on a same operator with regard to a single situation could prove disproportionate, a situation that the Commission explicitly wishes to avoid in the Access Directive.” As far as complainants are concerned (e.g. competitors in case of a settlement on access conditions which they consider advantageous to the incumbent), they are entitled to bring a new action against the regulated operator before a competition authority, provided the sector-specific remedy left sufficient commercial freedom to the latter.** The risk of procedural abuse cannot be excluded because competitors will legitimately use all available means to obtain a more advantageous decision. Furthermore, it is questionable whether the concept of commercial freedom is appropriate because: (i) it is unclear how much freedom the sector-specific remedy must leave to the regulated entity; and (11) a substantial number of regulatory remedies do leave some margin of manoeuvre to regulated entities.”° With regard to NRAs, the decision can also be criticised. In Deutsche Telekom, the Commission refused to challenge the NRAs’ decision directly, although a parallel complaint had been introduced on the basis of Article 86 EC.*® However, one could consider that in the present case, the RegTP bears

some responsibility for the existence of the price squeeze.”’ The refusal, by the Commission, to target NRAs when they bear some responsibility in the infringement of the EC competition rules is not an adequate solution. Indeed, a status of immunity is not likely to give incentives to NRAs to take due account of the competition provisions of the EC Treaty. This could result in the development, side by side, of inconsistent decision-making practices that could in turn prove discouraging to investment.’® Finally, on public policy grounds, the Commission’s position is debatable. The successive intervention of different authorities on a similar matter

The Proliferation of National Regulatory Authorities

195

duplicates investigative resources. In addition, it could slow down the pace of implementation of regulatory reforms and inhibit the transition to increased competition.» Furthermore, there is some uncertainty as to whether the Commission’s line of action can equally be followed by NCAs. Unfortunately, there does not seem to be any theoretical limit to the transposition of these principles in the MS, since NCAs are equally entrusted with the duty to apply Article 82 EC.

ALTERNATIVES APPROACHES TO CUMULATIVE JURISDICTION For the reasons that have been outlined, the principle of cumulative application of sector-specific regulation and competition rules by two sets of institutions intervening successively is not satisfactory. A two-tier approach is suggested to neutralise the shortcomings of this principle. The first tier of the solution consists in refraining from introducing competition proceedings against operators when a sector-specific remedy has been applied to them. The second tier is to prefer to target the NRAs instead of the operators where the sector-specific regulation remedy has been applied but leaves competition concerns. First Tier — Non-Enforcement of Competition Rules where a SectorSpecific Remedy has been Applied As has been said, the Deutsche Telekom decision places a heavy burden on regulated entities (i.e. risks of cumulative remedies, legal uncertainty and duplication of proceedings).'” To avoid these risks, a suggested solution is that where a sector-specific regulation remedy is available, competition rules should no longer be enforced against the operator. A solution of this kind was recently adopted by the US Supreme Court in the Verizon Inc. v. Curtis V. Trinko LLP case (hereafter, ‘Trinko’).'°! In short, Verizon, one of the US local incumbents, is required pursuant to sector-specific legislation, to provide access to its operations support systems (OSS). OSS are interfaces which allow competitors to fill in customer’s orders. Competitors send orders for service through an electronic interface with Verizon’s ordering system, and Verizon completes certain steps in filling the order and subsequently confirms back through the same interface. Access to Verizon’s OSS is thus necessary for new entrants to provide services to customers. In late 1999, competitors complained to regulators that a number of customers’ orders were going unfilled, in violation of Verizon’s obligations to provide access to OSS. A settlement was reached pursuant to which Verizon accepted to implement

a series of regulatory remedies for ensuring access to OSS functions. Shortly

196

Regulation through Agencies in the

EU

after the settlement, a complainant (Curtis v. Trinko LLP) filed a class action lawsuit before a federal court claiming that Verizon’s conduct was a violation of US antitrust law (i.e. Section 2 of the Sherman Act) and the case was eventually brought before the US Supreme Court. Besides the interesting findings of the Supreme Court with regard to the essential facilities doctrine, the ruling states that, even where the sector-specific regulation explicitly rejects immunity from antitrust law (as is the case in the EC), it nonetheless must be assessed whether an expansion of the scope of antitrust provisions should be admitted in situations where sector-specific remedies can be enforced. The US Supreme Court puts emphasis on the costs of such an expansion through the existence of ‘false positives’ (i.e. false condemnations where purely legitimate actions by an undertaking are held to be an infringement of the antitrust rules). The US Supreme Court also points out the fact that courts of law are ill equipped to deal with the day-to-day enforcement of the remedies they could impose on the basis of antitrust rules. Hence, the US Supreme Court comes to the conclusion that in fields where sector-specific remedies can be applied, it is preferable to refrain from enforcing antitrust rules. The Jrinko ruling is an outstanding recognition of a ‘pre-emption’ or ‘exhaustion’ principle in the field of antitrust. Private claims on the basis of antitrust rules should be ‘exhausted’ where a sector-specific remedy exists. The rationale behind this is that where a regulatory remedy is available, the benefits from antitrust action are likely to be limited and the costs are likely to be significant. In addition, the specificities of US procedural rules led to the fear that an expansion of antitrust liability to regulated sectors would trigger a flood of litigation through the initiation of numerous class action procedures combined with substantial treble damages claims.'” The approach followed in the US is appropriate in that it solves the risks of duplication of procedures and of conflicts in terms of decision making. The ruling of the US Supreme Court implies that antitrust law should only have a subsidiary role in regulated sectors, in situations where no sector-specific remedies are available. At this stage, three remarks shall be made. First, a number of elements suggest that the Commission’s approach, while apparently at odds with Trinko, can be partly reconciled with the Supreme Court’s approach. There are some recent indications that the Commission is unlikely to initiate proceedings where sector-specific remedies can be applied.'® In the O2/T-Mobile decision, a notified agreement on infrastructure sharing entailed a risk of foreclosure on sites.'°* However, the Commission voiced no concern in light of the fact, inter alia, that a sector-specific remedy was provided for by Article 12 of the Framework Directive and could be used by NRAs ifa restriction of competition was observed. !°°

The Proliferation of National Regulatory Authorities

197

Second, it is not sure whether the US Supreme Court approach should be roughly transposed in the EC. In some situations, a regulatory remedy may exist but the NRA fails to enforce it. In these cases, the possibility to act on the basis of competition rules could prove extremely useful. The Commission would start the procedure (by, for instance, launching a sector inquiry) in order to bring the failing NRAs’ attention to a particular problem. The Commission would subsequently defer the case to the regulator. Eventually, the problem would be handled by the NRA through the application of sector-specific remedies. The competition rules are thus only enforced as an ignition device. This approach has already been followed in the telecommunications sector with regard to the pricing of leased lines inquiry and the mobile termination charges inquiry.'” In the latter case, the Commission launched an inquiry on the pricing of interconnection charges for the termination of calls. The inquiry led to the finding that several PSTN operators charged mobile operators more than fixed operators for call termination. Given that the sector-specific framework provided a remedy for tackling this issue the Commission decided to stay proceedings and shift the investigation to the relevant NRAs.'°’ The usefulness of this approach pleads for a partial transposition of Trinko in the EC legal order. Third, the fact that the US Supreme Court approach can be explained by the specific risks attached to antitrust liabilities before US courts of law (class actions and treble damages) does not rule out the possibility, in the EC, to derive several jurisdictional principles from the Trinko case. This is all the more true in the current context, where the Commission tends to increasingly promote private enforcement of competition rules before national courts.'° At any rate, the signals given by the Commission in recent times are contradictory. It is suggested that the Commission should strive towards the recognition of the two following principles: (i)

when a sector-specific remedy has been applied to an operator and leads to an infringement of competition law, competition authorities should no longer enforce competition rules against operators (Exhaustion approach);

(ii)

when a sector-specific remedy is available and where there is/could be a potential infringement of competition law, competition authorities shall not enforce competition rules (O2/T-Mobile approach), unless the remedy is not applied. In that case, competition authorities should initiate the proceedings and subsequently shift the matter to the regulators (Sector Inquiry approach).

While the first principle avoids duplication of proceedings and decision-making inconsistencies, the second principle presents the advantage that potential market

198

Regulation through Agencies in the

EU

failures are not left unchecked as well as the advantage that the sector-specific remedy (arguably the best suited) is applied.’ One could, however, question what shall happen if a sector-specific remedy has been applied but leads to or leaves competition restrictions. The application of the first principle eradicates the possibility of enforcing competition rules against the operator. However, it does not exclude the possibility of acting, on the basis of competition rules, against the NRA or of asking the NRA to revise its position and to extend the remedies coverage so as to tackle remaining anticompetitive concerns. Furthermore, it could be observed against the exhaustion principle that the enforcement of competition rules besides the enforcement of sector-specific regulation may be a quite efficient tool for promoting effective regulation. This is a valid point but it seems that it leads to a degree of legal uncertainty and of regulatory costs that is incompatible with the achievement of regulatory efficiency. Second Tier — Use of EC Law Remedies directed at NRAs Misconceived Decisions

In parallel to the above, several remedies should be used to ensure compliance of the NRAs with competition rules without targeting the operator. First, the NRAs’ decision can be directly challenged through appeal channels and eventually lead to an Article 234 EC reference to the ECJ. Second, a more straightforward solution is action by the Commission against NRAs’ decisions on the basis of Articles 226 or 86 EC (2.2). Appealing NRAs’ decisions Under national law, NRAs generally fit within the overall judicial framework. Depending on the judicial structure of each MS, NRAs’ decisions can be challenged before ordinary courts of law or special administrative tribunals. National courts will normally ensure that the decisions of NRAs comply with the requirements of national as well as EC law and, in particular, the competition rules of the EC Treaty.''? However, this may not always be the case. Thus, EC directives contain some limited requirements regarding the judicial review of the NRAs.

In the telecommunications

sector, for instance, the Framework

Directive requires that MS put in place an ‘effective’ procedure of appeal which allows for a party affected by the decision adopted by a NRA to have this decision reviewed.''' In other sectors, EC legislation leaves more discretion to MS and in some cases, nothing is provided for.'” A positive feature of setting up appeal procedures is that ultimately a judgment on the compatibility of the NRA decision with EC competition rules can be reached through an Article 234 EC reference to the ECJ.' Indeed, NRAs do not generally possess the attributes necessary for directly bringing a

The Proliferation of National Regulatory Authorities

199

reference to the ECJ. The requirement to set up appeal procedures reintroduces the possibility of having some light cast on the compatibility of an NRA decision with the competition rules. In addition, the interpretation by the ECJ is a guarantee of lawfulness, which is not necessarily the case with a Commission decision. Nevertheless, the long delays that may result are not fully compatible with the speedy evolution of certain regulated sectors (e.g. the telecommunications sector) as well as with the larger goal of accelerating transition to further liberalisation. Also, it can be subject to question whether courts of law are well placed to rule on NRAs’ decisions in highly technical fields. This could prove even more difficult when the assessment requires a combined analysis of both sector-specific regulation and competition rules.''4 Action by the commission against the NRA decision Two routes are potentially available to the Commission in order to bring an NRAs’ decision in line with the EC Treaty competition rules. A first route consists in the Commission issuing reasoned opinions pursuant to Article 226 of the EC Treaty (ex-Article 169 EC) if it considers that an NRA’s decision violates EC Law. The lack of compliance by the MS in question may serve as a basis for the

initiation of legal proceedings by the Commission before the ECJ.'!° The doctrine of the combined application of Article 3(g), 10, 81 and 82 EC requires MS to abstain from adopting or maintaining in force measures susceptible of eliminating the ‘effet utile’ of these provisions. A variety of reasons are, however, classically put forward to disqualify this possibility. A first reason is that this doctrine has rarely been used by the Commission. A large majority of the cases alleging a combined infringement of Article 3(g), 10 and 81 or 82 EC have been brought to the ECJ through the preliminary reference channel, in the context of proceedings pending before national courts.''® This is so because the Commission is unlikely to welcome Article 226 EC complaints as it would act as an appeal body against NRAs’ decisions and that Article 234 EC remains available for such purposes.''’ A second reason is that infringement proceedings might prove lengthy and thus poorly suited to remedy a breach of competition law.''* A second route for action against an NRA decision is Article 86 EC. In Ahmeed Saeed and Bodson, the ECJ held that MS measures that contribute to, or impose on undertakings enjoying special rights, a conduct incompatible with Article 81 or 82 EC fall within the scope of Article 86(1) EC. In the

Italian GSM fees and the Spanish GSM fees decision, Ministries’ decisions (at the time, in quality of NRAs) that had the clear effect of favouring the incumbent operator by strengthening its dominant position and extending it to new markets was challenged on the basis of Article 86(1) and 82 EC.'!° Thus, the Commission is entitled to proceed on the basis of Article 86(3) EC and adopt a decision against a MS and consequently, against an NRA’s action that

200

Regulation through Agencies in the

EU

favours an operator or a class of operators and lead to a violation of the EC Treaty competition rules.!”° The adoption of a decision on the basis of Article 86(3) has a number of advantages over the adoption of an Article 226 EC reasoned opinion. First, the procedure is quicker because Article 86(3) decisions are immediately binding in their entirety upon the MS to whom they are addressed, and thereby to the MS organs.!?! In the framework of Article 226 EC, a similar effect can only be obtained after a judgment of the ECJ, which takes generally a much longer time. Second, this decision can be invoked by individuals before national courts and can serve as the basis for a claim of damages against the MS under the rule in Francovich.'” Third, the procedure under Article 86 EC is much more advantageous for private undertakings than that under Article 226 EC. In Max.Mobil, the CFI held that while in the case of Article 226 EC the large discretion of the Commission does not allow the complainants to challenge the Commission’s refusal to proceed, Article 86(3) EC leaves less discretion to the Commission, and its refusal can be contested through the initiation of annulment proceedings pursuant to Article 230 EC.'* Thus, the Commission margin of manoeuvre is more limited (although the opportunity of taking a decision is only subject to marginal control by the ECJ and CFI). Article 86 EC complaints could prove a more efficient route for third parties than Article 226 EC. Whatever the pros and cons of Article 86 and 226 EC, a number of public policy considerations may better explain why the Commission has, so far, been reluctant to act against NRAs. Given that the infringement procedure is not directed to the NRA but to the MS in which it operates, the former seems to enjoy a certain degree of immunity. Of course, the MS will generally ask the NRA to bring its actions in line with EC law. But this situation creates a tension with the requirement of independence between the government and the NRA as it implies that the government ends up giving instructions to the NRA. Another limitation of this approach is that, from a policy-making perspective, Article 226 or 86 EC proceedings are unlikely to be initiated by the Commission. In sectors where the ‘networking’ pattern has been followed (such as, for instance, the telecommunications sector), the NRAs are the arm of the Commission for

bringing about market liberalisation.'** Thus, the development of conflictual relationships between the Commission and the NRAs may in the long run compromise vital collaboration between both levels. In addition, the initiation of formal legal proceedings may affect the credibility of the NRA with regard to the regulated entities. This is arguably illustrated in the Deutsche Telekom case, where the Commission had also been seized on the basis of Article 86 EC but subsequently decided to leave these claims and to proceed solely on the basis of Article 82 EC. A third limitation of this is that, from a purely legalistic standpoint, the initiation of proceedings for an alleged infringement of the competition rules

The Proliferation ofNational Regulatory Authorities

201

could give rise to a difficult conflict between equally ranking norms. It could be considered that the NRA applied the national sector-specific rules based on directives adopted pursuant to Article 95 EC and that it acted in full compliance with this provision. Thus, if the legitimate enforcement of sector-specific legislation led inevitably to an infringement of the competition rules, it would be difficult to decide which set of rules should prevail. However, a situation of frontal conflict of this kind is extremely unlikely to occur.!*5 For the reasons that have just been given, the Commission’s reluctance to act against NRAs is understandable. Nevertheless, it is submitted that in light of the drawbacks of the Deutsche Telekom approach, the Commission should not hesitate to commence proceedings directed at NRAs. An encouraging illustration of this has been recently reported in the telecommunications sector The Commission has evoked the possibility of using Article 226 EC to act against NRAs enforcing national sector-specific legislation that is contrary to the EC framework.'”° Pushing this a step further, the Commission could act against NRAs that do not comply with competition rules. It shall be recalled, in that respect, that during the year 2000, the Commission had already evoked the possibility of introducing actions against NRAs’ decisions that would be contrary to EC competition rules.!’

CONCLUSION The best approach for limiting conflicts and regulatory inconsistencies between competition authorities and NRAs is to propose the implementation of institutional reforms. A first option that was implemented in the UK is to entrust NRAs with the power to actively enforce the competition rules. This option was also explicitly considered in the postal sector where Directive 97/67 provides that NRAs may apply EC competition law. The problem of this approach, however, is that in the long run two diverging approaches to competition law (i.e. the NRA’s approach, on the one hand and the NCA’s approach, on the other hand) could progressively emerge.'** Also, it does not eradicate the risk of multiple proceedings, because two entities remain competent. For these reasons this option has been criticized.'”° A second option is to set up the NRA as a division/department of the NCA. The main advantage of this approach is that it eliminates the problem of duplication of proceedings (one stop shop) and consequently limits transaction costs for operators. Also, it limits the risks of regulatory inconsistencies. Does that amount to say that this is the best solution? This is not sure. In a system of this kind, specific issues of sector-specific regulation might be neglected (universal service, investment incentives etc.). In addition, this approach only internalises problems. Difficulties may thus arise in terms of decision-making

202

Regulation through Agencies in the

EU

transparency and difficulties for the authority to reach consensual rulings on

controversial issues.'*° It seems, thus, that no institutional approach can guarantee the elimination of potential regulatory inconsistencies and of multiple proceedings. The increasing proliferation of NRAs alongside competition authorities should be accompanied by the adoption of rules clarifying their relationships. Ideally, these rules would apply to all network industries sectors where NRAs have been set up. In terms of substance, the recourse to a /ex specialis approach and the admission that competition authorities should not target operators when a sector-specific remedy has been applied would probably constitute an improvement. The eventual restrictions of competition that could remain should then be treated on a cooperative/integrated mode between competition authorities and NRAs (through for instance, the competition authority asking the NRA to reopen the case). Ultimately, action against NRAs on the basis of the remedies provided for by the EC Treaty could be envisioned. In any event, the risks described above should rapidly disappear once full liberalisation has been achieved. When this time comes, NRA action will be nugatory. The risks linked to regulatory proliferation will vanish in the dawn of regulatory extinction.

NOTES 1 This chapter places an emphasis on the telecommunications sector, because it provides numerous illustrations of these problems. See Aurore Laget-Annamayer (2002), La régulation des services publics en réseaux, LGDJ-Bruylant, Bruxelles, p. 302. Note, also that potential situations of conflicts could additionally arise with national courts which are also competent to apply competition rules. 2 See Alexandre de Streel, Robert Queck and Philippe Vernet (2002), ‘Le nouveau cadre réglementaire européen des réseaux et services de communications électroniques’, 3-4, Cahiers de Droit Européen, 243, p. 294. For instance, under sector-specific regulation, the fact that an operator enjoys a dominant position on a market already triggers the imposition of remedies by an NRA on the basis of the anticipation of potential future anticompetitive behaviour by the undertaking. In contrast, under competition law, the existence of a dominant position cannot lead to a remedial action by a competition authority unless an actual abuse of this dominant position can be proven. 3 Such as, for instance, the control of concentrations between undertakings or the control of State aids by the European Commission. 4 See, on this, Damien Geradin and Greg J. Sidak, ‘European and American Approaches to Antitrust Remedies and the Institutional Design of Regulation in Telecommunications’, forthcoming in the Handbook of Telecommunications Economics, 2nd Vol. 5 Competition authorities tend to increasingly rely on their exclusive competences in the field of merger to achieve regulatory objectives through the imposition of commitments to operators. See, in the energy sector, Michele Piergiovanni (2003), ‘EC Merger Control Regulation and the Energy Sector: An Analysis of the European Commission’s Decisional Practice on Remedies’, 3, Journal of Network Industries, 227. 6 First, in most network industries sectors (telecommunications, electricity, gas, rail, and post),

NRAs have been or are being created and have been entrusted with substantial market regulation powers. In the telecommunications sector, most MS have set up NRAs and these authorities have

The Proliferation of National Regulatory Authorities

203

been active for a rather long time. The Framework Directive requires all MS to have an NRA. See Directive 2002/21 of the European Parliament and of the Council of 7 March 2002 on a common regulatory framework for electronic communications networks and services (hereafter, the ‘Framework Directive’), O.J. L 108 of 24 April 2002, pp. 33-50. In the field of energy, most MS have created sector-specific agencies or have entrusted the regulatory duties deriving from EC directives to economy-wide competition authorities pursuant to the single obligation imposed by the first generation directives. As far as the postal sector is concerned, Directive 97/67 required MS to set up NRAs. See Directive 97/67 of the European Parliament and the Council of 15 December 1997 on common rules for the development of the internal market of Community postal services and the improvement of quality of service, O.J. L 15 of 21 January 1998, pp. 14-25. Currently, most MS have done so and where MS have not, proceedings have been commenced before the European Court of Justice. Finally, as far as the rail transport sector is concerned, requirements of setting up regulatory institutions can be identified in Directive 2001/14. First, this Directive requires that the capacity allocation processes (the allocation of train paths) shall be performed by an allocation body that is independent in its legal form, organisation and decision making from any railway undertaking. Second, the Directive requires that MS establish a regulatory body. See Directive 2001/14 of 26 February 2001 on the Allocation of Railway Infrastructure Capacity and the Levying of Charges for the Use of Railway Infrastructure and Safety Certification, O.J. L 75/29 of 15 March 2001. In parallel to this, the modernisation of EC competition law under new Regulation 1/2003 leads to substantial transfers of powers from the European Commission to National Competition Authorities and national courts. See Council Regulation 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, O.J. L 1 of4 January 2003, pp. 1-25. There have been numerous comments on the decentralisation of the implementation of competition rules and the creation of a network of competition authorities: See Damien Geradin (2002), ‘Competition Between Rules and Rules of Competition: a Legal and Economic Analysis of the Proposed Modernization of the Enforcement of EC Competition Law’, Columbia Journal of European Law, 1, 1; Hans Gilliams (2003), ‘Modernisation: from policy to practice’, European Law Review, 28, 451; Jacques Bourgeois and Christophe Humpe (2002), ‘The Commission’s Draft “New Regulation 17”, European Competition Law Review, 2, 43; Claus-Dieter

Ehlermann and Isabela Atanasiu (2002), ‘The modernisation of E.C. Antitrust Law: Consequences for the Future Role and Function of the EC Courts’, European Competition Law Review, 2, 72; James S. Venit (2003), ‘Brave new world: The modernization and decentralization of enforcement under Articles 81 and 82 of the EC Treaty’, Common Market Law Review, 40, 545; Laurence Idot (2003), ‘Le nouveau systéme communautaire de mise en ceuvre des articles 81 et 82 CE (réglement 1/2003 et projets de textes d’application’, Cahiers de Droit Européen, 3-4, 283-371. 7 See on this Pierre

Larouche

(2000),

Competition

Law

and Regulation

in European

Telecommunications, Hart Publishing, pp. 296-298. 8 See Directive 2002/19 of the European Parliament and the Council on access to, and interconnection of, electronic communications

networks and associated facilities of 7 March

2002 O.J. L 108 of 24 April 2002, pp. 7-20. These obligations apply to operators enjoying a Significant Market Power. This concept can be equated to the concept of dominance under Article

82 EC. 9 See Article 20(3) of the Directive 2003/54 of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92, O.J. L 176 of 15 July 2003, pp. 37-56. 10 See on this, Antonio F. Bavasso (2002), ‘Essential facilities in EC law: the rise of an “epithet” and the consolidation of a doctrine in the communications sector’, Yearbook of European Law, 21,

63; John Temple Lang (2000), ‘The Principle of Essential Facilities in European Community Competition Law — The Position Since Bronner’, Journal of Network Industries, 1, 375; Sylvain Martin and Fabienne Levilain (2002), ‘L’application de la théorie des “facilités” essentielles dans le secteur des télécoms, de |’électricité et du gaz’, Gazette du Palais, Rec. Bimestriel, 5, 1512. The

case law has helped clarify the conditions under which an undertaking controlling an essential facility is under a duty to give access to its infrastructure. In Bronner, the European Court of Justice held that, for a refusal to give access to be an abuse of a dominant position, it must be shown that: (i) the refusal to grant access is likely to eliminate all competition in the downstream market and cannot be objectively justified; and (ii) the good, service of infrastructure in question is absolutely

204

Regulation through Agencies in the

EU

indispensable for carrying out the economic activity (i.e. there is no alternative to this). See ECJ, 26 November

1998, Oscar Brénner v. Mediaprint, C-7/97, [1998] ECR I-7791. The Commission

recently made use of the essential facilities doctrine in the rail sector. See Commission Decision of 27 August 2003, GVG/FS (especially §120) O.J. L 11 of 16 January 2004, pp. 17-40 and Oliver Stehmann (July 2004), ‘Applying Essential Facility Reasoning to Passenger Rail Services in the EU — the Commission Decision in the Case GVG’, European Competition Law Review, 25, 7, 290-

394. \! See Leigh Hancher (1993), ‘Competition, State Monopolies and the Internal Energy Market’, in Jules Stuyck and Ellen Vossestein (eds), State Entrepreneurship, National Monopolies and European Community Law, Kluwer, p. 31.

12 See European Commission, XVIth Report on Competition Policy (1984), §§94-95, Undertakings offered by IBM. 13 See Competition and Regulation in Telecommunications,

OECD,

DAFFE/COMP(2002)6,

p. 8. This hypothesis is, however, subject to discussion given the difficult practicability of the test for excessive pricing and the Commission’s reluctance to embark upon difficult pricing analysis. Nevertheless, according to some observers, it is probable that a competition authority would be entitled to act on the basis of Article 82 provided the test laid down by the ECJ in United Brands is met and prices are clearly excessive. See, on this, Jonathan Faull and Ali Nikpay (eds) (1999), The

EC Law of Competition, Oxford University Press, p. 189. 14 See, for instance, Article 13 of the Framework Directive, supra note 6. See also Article 14 of

Directive 97/67, supra note 6 in the postal sector. See Article 19 of Directive 2003/54, supra note 9 in the electricity sector. 15S See Notice on the application of the competition rules to the postal sector, §8. See, with regard to Article 81, Atlas and Phoenix [1996] O.J. L239/57 and Unisource [1997] O.J. L 381/1. 16 Tn the energy sector, this is guaranteed through the unbundling of the transmission and distribution functions from the other functions of a vertically integrated undertaking. See Articles 10 and 15 and 20(1) of Directive 2003/54, supra note 9.

17 See Commission Decision of 23 October 2001, on the lack of exhaustive and independent scrutiny of the scales of charges and technical conditions applied by La Poste to mail preparation firms for access to its reserved services, O.J. L 120 of 7 May 2002, pp. 19-37. 18 See, for instance, in France the Ordonnance of 25 July 2001 portant adaptation au droit communautaire du Code de la Propriété Intellectuelle et du Code des Postes et Télécommunications. 19 See Commission Press Release IP/97/292 of 11 April 1997, ‘Settlement reached with Belgacom on the publication of telephone directories — ITT withdraws complaint’. 20 See Article 12(2) of the Framework Directive, supra note 6 which provides that: ‘where undertakings are deprived of access to viable alternatives because of the need to protect the environment, public health, public security or to meet town and country planning objectives, Member States may impose the sharing of facilities or property (including physical co-location) on an undertaking operating an electronic communications network or take measures to facilitate the coordination of public works only after an appropriate period [...]’. 21 Several MS have adopted sector-specific rules to deal with this issue. In the UK, Oftel published a position in May 2001, ‘3G Mobile Infrastructure Sharing. Note for information’, available at http://www.oftel.gov.uk/publications/mobile/infrashare0501.htm. In Germany, the RegTP issued its ina document entitled ‘Interpretation of the UMTS Award Conditions in the light of more recent technological advance’, RegTP (6 June 2001), available at www.regtp.de. In France the ART adopted guidelines on 10 December 2001. See ‘Position de I’Autorité de régulation des télécommunications sur le partage d’infrastructures dans les réseaux mobiles de troisiéme génération’ available at http://www.art-telecom.fr. 22 See Nicolas Petit (July 2004), ‘The Commission’s Contribution to the Emergence of 3G Mobile Telephony: An Analysis of some Decisions in the Field of Competition Law’, European Competition Law Review, 25, 7, 429-438.

23 See Commission Decision, COMP/38.370, O2 UK Ltd./T-Mobile UK Ltd., of 30 April 2003,

O.J. L200 of 7 August 2003, pp. 59-84 and Commission Decision, COMP/38.369, T-Mobile Deutschland/VIAG Interkom of 16 July 2003. See also Commission Press Release IP/03/1026 of

16 July 2003. 4 It seems that, in the near future, a further variety of issues could potentially be brought under the

The Proliferation ofNational Regulatory Authorities

205

two umbrellas of sector-specific regulation and competition rules. This is, for instance, the case of

agreements between mobile communications operators (hereafter, the ‘MOs’) and Mobile Virtual Network Operators (hereafter, the ‘MVNOs’). MVNOs are operators that have their own mobile network code and their own range of mobile numbers but that do not have a licence to operate wireless frequencies nor own any substantial network infrastructure (e.g. only own some limited network elements). The form of MVNO is interesting for operators that cannot afford to incur the high fixed costs and sunk investments of acquiring licences and rolling out networks. Their cost base and risk are by definition much lower than those of network operators as they do not have significant investments in network equipment. MVNOs enter into agreements with MOs in order to offer competing services. Depending on the business model followed, a MVNO can be considered as a service provider or, in case it owns some network elements, as an operator willing to conclude

national roaming arrangements and thus asking for interconnection. In that respect, MVNOs could potentially fall within the telecommunications

framework, and benefit, for instance, from the

provisions on access and interconnection. These agreements could potentially fall within the scope of Article 81 and/or 82 EC. Given that the business model of these operators is, however, unclear it

is too early to assess whether the risk of overlap will raise problems. 25 It shall be noted that competition authorities may also initiate proceedings on their own motion. 26 See Commission Press Release IP/01/1641 of 23 November 2001, ‘Commission settles Marathon case with Thyssengas’. Also, national courts may be called to rule on contractual disputes arising from access conventions on the basis of competition rules. This is especially true in light of the modernisation of EC competition law, which will presumably increase the volume of litigation before courts on the basis of Articles 81 and 82. Let us, however, recall that the competence of the

national courts to apply the EC competition provisions is not a new thing. However, the Commission seeks to increase the involvement of national courts in the implementation of competition law and the decentralisation of Article 81(3) might contribute to this. 27 Since 2002, the ART (the French telecommunications regulator) and the CRE (the French energy regulator) have adopted respectively 16 and 14 decisions in the context of disputes between operators. There is a clear increase in the number of decisions adopted between 2002 and 2003. In addition,

this was already true in the field of classic NRAs’ regulatory functions, such as the prior approval of tariffs or the adoption of guidelines. The finding of the NRAs could subsequently be dealt with by a competition authority. Also, complainants should be understood in a broad meaning. See Article 9(4)b of the Framework Directive, supra note 6: “The national regulatory authorities shall promote the interests of the citizens of the European Union by inter alia: [...] ensuring a high level of protection for consumers in their dealings with suppliers, in particular by ensuring the availability of simple and inexpensive dispute resolution procedures carried out by a body that is independent of the parties involved’. In the energy sector, procedures must be open to any customer that wishes to introduce a complaint. See Article 20(5) and Article 3(5) of Directive 2003/54, supra note 9. This

option was followed in some MS which have set up additional conciliation mechanisms. A large number of customers can benefit from a conciliation procedure before the NRA. See Christophe Lemaire (2003), Energie et concurrence: Recherches sur les mutations juridiques induites par

la libéralisation des secteurs de I’électricité et du gaz, Presse Universitaires d’ Aix-Marseille, 95D: 2sSee Article 9(5) of Directive 97/33 of the European Parliament and of the Council of 30 June 1997 on interconnection in telecommunications with regard to ensuring universal service and interoperability through application of the principles of Open Network Provision (ONP), O.J. L 199 of 26 July 1997, pp. 32-52. 29 See Article 20 of the Framework Directive, supra note 6. 30 See Recital 32 of the Framework Directive, supra note 6: ‘In the event of a dispute between undertakings in the same Member State in an area covered by this Directive or the Specific Directives, for example relating to obligations for access and interconnection or to the means of transferring subscriber lists, an aggrieved party that has negotiated in good faith but failed to reach agreement should be able to call on the national regulatory authority to resolve the dispute’. 31 See A. Laget-Annamayer, supra note 1, p. 361. It is of note that in Spain, the dispute settlement function of the NRA is considered as the main function of the regulator. 32 See, for instance, Article 21(6) of Directive 2001/14, supra note 6: ‘In case of disputes relating

to the allocation of infrastructure capacity, a dispute resolution system shall be made available in

206

Regulation through Agencies in the

EU

order to resolve such disputes promptly. If this system is applied, a decision shall be reached within a time limit of 10 working days’. 33 See Article 20(3) of Directive 96/92 of the European Parliament and of the Council of 19 December 1996 concerning common rules for the internal market in electricity, O.J.

L027 of 30 January 1997,

pp. 20-29. 34 See Article 20(5) of Directive 2003/54, supra note 9: ‘Any party having a complaint against a transmission or distribution system operator with respect to the issues mentioned in paragraphs 1, 2 and 4 may refer the complaint to the regulatory authority which, acting as dispute settlement authority, shall issue a decision within two months after receipt of the complaint. This period may be extended by two months where additional information is sought by the regulatory authority. This period may be further extended with the agreement of the complainant. Such a decision shall have binding effect unless and until overruled on appeal. Where a complaint concerns connection tariffs for major new generation facilities, the two-month period may be extended by the regulatory authority’. MS such as the UK and Italy will thus have to entrust their NRAs with a dispute settlement competence.

35 See C. Lemaire, supra note 27, p. 356.

36 In France for instance, the ART can impose fines of up to 3% of the total sales of the offender. In the energy sector, NRA (UK, Spanish, Dutch etc.) have also been entrusted with administrative and financial sanction powers. 37 Several observers consider that the rules laid down in sector-specific frameworks go further than competition rules and lead to the imposition of more stringent conditions on incumbents than those required under competition law. See A. de Streel, R. Queck and P. Vernet, supra note 2, p. 294. For this reason, the risk of conflict is likely to be limited. While this is generally true, NRAs and competition authorities follow different approaches which could, in the long run, lead to substantial divergences. 38 Even the authors who see competition law and sector-specific regulation as a common set of rules do not rule out the risk of divergences. See, for instance, Paul Nihoul and Peter Rodford (2004), EU Electronic Communications Law, Oxford University Press, London, §4.370, who argue

that while sector-specific regulation should be seen as a ‘part of general competition law’ [...], “It cannot be excluded that discrepancies may appear in specific circumstances’. 39 See ECJ, Oscar Brénner v. Mediaprint, supra note 10.

40 See Damien Geradin and Gregory Sidak, supra note 4. 41 This should not, however, be too problematic because there would be no conflict between norms.

In that case, Article 82 EC would be held to be inapplicable. Sector-specific apply solely. 42 See Article 12(2) of the Access Directive, supra note 8. 43 This is illustrated by the controversial debates surrounding the adoption of the case or of the Microsoft decision. Claims by Microsoft that a generous access counter to its incentives to innovate have been rebutted by the European Commission Decision of 24 March 2004, Case COMP/C-3/37.792

legislation would

recent JMS Health policy might run Commission. See

Microsoft, C(2004)900 final,

especially §709. In addition, in the IMS case, the Court made clear that access to an input protected by an intellectual property right should only be granted provided it would prevent the marketing of a new product on the downstream market. See ECJ C-418/01, 29 April 2004, IMS Health GmbH & Co. OHG v. NDC Health GmbH & Co. KG, not yet published. There is, however, a considerable room for debate with respect to the degree of novelty that has to be proven. 44 A topical illustration of this can be found in the application of the ‘essential facilities’ doctrine to intellectual property rights. The JMS Health decision, for instance, shows that the Commission’s main concerns are very short-sighted and that arguments linked with the protection of innovation and investments incentives are not taken into account under Article 82 EC. 45 Tt shall be mentioned that the risk of inconsistencies between NRAs and NCAs is further aggravated by the fact that the approaches may vary from one MS to another.

46 An empirical study showed a clear relationship between regulatory effectiveness and industrial investment. Regulatory confusion might lead to inefficiency and lack of effectiveness and indirectly impact investment. See Regulatory Scorecard Report, Jones Day, 12 November 2002, available at BED ewe .jonesday.com/news/detail.asp?language=English&newsid=264. 7 Tbid.

The Proliferation of National Regulatory Authorities

207

48 See Recital 26 of Directive 97/33, supra note 28; See Recital 3 of Directive 96/92, supra note 33; See Recital 6 of Directive 98/30 of the European Parliament and of the Council of 22 June 1998 concerning Common Rules for the Internal Market in Natural Gas, O.J, L 245 of 4 September 1998, pp. 1-12: “Whereas the provisions of this Directive should not affect the full application of the Treaty, in particular the provisions concerning [...] the rules on competition’. See Recital 41 of See Directive 97/67, supra note 6: ‘Whereas this Directive does not affect the application of the rules of the Treaty, and in particular its rules on competition and the freedom to provide services’. 49 See Article 20(5) of the Framework Directive, supra note 6; see also Article 23(11) of Directive 2003/54, supra note 9:’Complaints referred to in paragraphs 5 and 6 shall be without prejudice to the exercise of rights of appeal under Community and national law’. °° The possibility to bring an action before national courts will not be analysed in this chapter. It is, however, a potential source of conflicts. See on this C. Lemaire, supra note 27, pp. 374-375. 5! See on this CFI, T-111/96, ITT Promedia vy. Commission, ECR [1998], II-2937. >? See Article 2(g) of Framework Directive, supra note 6. See Commission’s Decision setting up the European Regulators’ Group in the field of telecommunications, electricity and gas; also recall that the common mission to all NRAs’ Commission Decision of 29 July 2002 establishing the European Regulators Group for Electronic Communications Networks and Services, JOCE L 200 of 30 July 2002, pp. 38-40, §3. See also Commission Decision of 11 November 2003 establishing the European Regulators Group for Electricity and Gas Official Journal L 296 of 14 November

2003, pp. 34-35, §3. 53 This is, however, subject to several exceptions. First, in the UK, Ofcom and Ofgem have, for

instance, jurisdiction to enforce the national competition legislation. Second, in Greece, the NRA in charge of the telecommunications sector has the power to enforce competition legislation with respect to telecommunications markets. Third, in the postal sector, the EC framework recognises the possibility that in addition to the compliance with the obligations arising from the Directive, the NRA ‘may also be charged with ensuring compliance with competition rules in the postal sector’. See Article 22 of Directive 97/67, supra note 6. 34 See ECJ, 16 November 1977, SA G.B.-INNO-B.M. v. Association des détaillants en tabac (ATAB),

Case 13-77, ECR [1977]-2115, §31. 35 See Ibid., §§32-33. 5 See, for instance, ECJ, 11 April 1989, Ahmed Saeed Flugreisen and Silver Line Reisebtiro GmbH v. Zentrale zur Bekimpfung unlauteren Wettbewerbs e.V., Case 66/86, ECR [1989]-803, §§48 and

52. A significant analogy could be drawn here with the Framework’s Directive provisions on site sharing which leads to encourage the conclusion of agreements between undertakings, supra note

6. 57 See, ECJ 9 September 2003, Consorzio Industrie Fiammiferi (CIF) vy. Autorita Garante della Concorrenza e del Mercato, Case C-198/01, §§48-49. 58 Tt shall be noted, in that respect, that the new sets of Directives in the telecommunications but also

in the energy sector no longer contain the explicit reference to the fact that sector-specific regulation is without prejudice to the application of competition rules. Arguably, the EC legislator seems to have tried to mitigate the concern over the cumulative application of the two bodies of rules by omitting to explicitly reproduce this provision. The goal of the Directives is probably to promote a more integrated approach between the two bodies of rules, in that sector-specific regulation should increasingly internalise competition concerns so that the risk of conflict is limited. The various references, within the NRAs’ mandates, to the protection and promotion of competition, support this view. This shall not, however, imply that competition rules are not applicable to regulated sectors. 59 See Article 8(2)b of the Framework Directive, supra note 6. This Directive also contains several references to the duty to ‘safeguard competition’. Similar references were already made in the first

generation directives. See, for instance, Article 9(5) of Directive 97/33, supra note 28 that provides that one of the NRAs’ missions is the promotion of competition. 60 See Article 23 of Directive 2003/54, supra note 9 and Article 25 of Directive 2003/55 of the European Parliament and the Council of 26 June 2003 concerning common rules for the internal market in natural gas and repealing Directive 98/30, O.J. L 176 of 15 July 2003, pp. 57-78. 61 See P. Larouche, supra note 7, p. 300. A noticeable exception to this can nonetheless be found in the postal sector where NRAs may be charged with ensuring compliance with the competition rules. See Article 22 of Directive 97/67, supra note 6. This is, however, purely optional.

208

Regulation through Agencies in the

EU

62 Also, actions could be initiated before national courts. See J. Faull and A. Nikpay, supra note 13,

. 11.48. © See Michael Albers (2000), ‘Competition Law Issues Arising from the Liberalisation Process’, Journal of Network Industries, (3), 269, p. 284.

64 See Commission Notice on the Application of Competition Rules to Access Agreements in the Telecommunications Sector O.J. C 265 of 22 August 1998, §§13, 18 and 22. The Commission’s document does not, in any case, modify the state of the law. In line with the ECJ’s case law, the

Commission’s notice recalls the importance that NRAs do not approve any practice or agreement contrary to the competition rules. Also, it confirms that competition rules and sector-specific regulation apply cumulatively and that compliance with one does not ensure compliance with the other. These questions had already been evoked in the Guidelines on the application of the competition rules to the telecommunications sector in 1991. 65 Tbid., §§ 28 and 150. 66 Tbid., §§30 and 33. 67 Tbid., §31. 68 See European Commission, XXVIIth Report on the Application of Competition Rules in the European Union, 1997, p. 108.

69 See Commission Press Release IP/O2/792 of 31 May 2002, ‘Commission clears Synergen Venture between ESB and Statoil following strict commitments’. A similar approach has been followed in the Newscorp/Telepiu merger where a number of commitments were placed under the monitoring and enforcement of the Italian media authorities, i.e. the Ministry of Communications of the Republic

of Italy and the Autorita per le Garanzie nelle Comunicazioni. See infra note 77. 70 NRAs are better placed to decide over access pricing and access conditions because of their expert knowledge of the industry, the important volume of data they handle and the large amount of resources they can dedicate to these issues. See OECD, “Relationship between Regulators and Competition Authorities’, DAFFE/CLP(99)8, p. 8, available at http://www.oecd.org/dataoecd/ 35/37/1920556.pdf. Also, access pricing issues involve a number of policy making elements (investment incentives etc.) which go beyond the mission of a competition authority. For a sceptical view of competition authorities taking decisions in terms of pricing, see Richard Whish, Competition Law, 4th Edn, (2001) Butterworths, p. 480. See also P. Larouche, supra note 7, pp. 259 and 319. 71 See Article 20 of Framework Directive, supra note 6; Article 23(5) of Directive 2003/54, supra note 9; Article 23(5) of Directive 2003/55, supra note 60; Article 21(5) of Directive 2001/14, supra

note 6 on scheduling disputes before the infrastructure manager. See also A. Laget-Annamayer, supra note 1, p. 362, who underlines that a main advantage of the dispute settlement systems before NRAs is the short time limits within which a settlement must be reached. 7 This issue will become increasingly problematic in light of the modernisation of competition law. 73 In the telecommunications sector, in Spain, a number of procedures are organised through legislative measures, i.e. a Royal Decree Law 6/1996 as replaced by Act 12/1997 of 24 April on the liberalisation of telecommunications. In Finland and in the Netherlands, these issues are dealt with under an agreement/protocol between the competition authority and the NRA. 4 See, for instance, in the telecommunications sector, the requirements of the law in Italy, Spain or

France. See A. Laget-Annamayer, supra note 1, p. 435. See also, in the Energy sector, the requirements of the CRE, C. Lemaire, supra note 27 p. 377. Also, the NRA may consult the NCA when applying competition law concepts within the implementation of their regulatory missions. The regulator must obtain consent of the competition authority to take some decisions or, at least, is under a duty to consultation, in Germany, for the definition of the dominant position and the delimitation of markets. See A. Laget-Annamayer, supra note 1, p. 436. 75 See Decision 00-MC-19 of the Conseil de la Concurrence, 5 December 2000, relative a une demande de mesures conservatoires présentée par |’Autorité de Régulation des Télécommunications, available at http://www.conseil-concurrence.ft/. 76 e.g. in France, Italy, Spain. In France, the ART is regularly consulted by the NCA for advice within the context of disputes between competitors. See, for instance, Avis 03-64 of the ART of 14

January 2003 relating to anticompetitive practices by France Telecom, available at www.arttelecom.fr.

The Proliferation of National Regulatory Authorities

209

77 An example of this can be found in the field of merger control where a competition authority could entrust the monitoring of commitments to the NRA. In the recent Newscorp/Telepiu case, the Commission entrusted the monitoring of the commitments submitted by the parties for obtaining clearance to the Italian telecommunications regulator. See Case COMP/M.2876 —Newscorp/Telepiti of 2 April 2003 C (2003) 1082 final; Cristina Caffarra and Andrea Coscelli (2003), ‘Merger to Monopoly: Newscorp/Telepiv’, (11) European Competition Law Review, 265. 78 See C. Lemaire, supra note 27, p. 378. However, the author considers that there is no duty, for the NCA to stay proceedings and to refer the case to the NRA. Two proceedings can thus take place. See Laurence Idot (1998), “Régles de concurrence et régulations sectorielles’, in La régulation des services publics en Europe, TEPSA. ™ See A. Laget-Annamayer, supra note 1, p. 434. Even in the UK, conflicts have been reported. 80 See Bruno Lassere (1999), ‘Competition or Regulatory Authority? European or National Level?’, in Claus Dieter Ehlermann and Louisa Gossling (eds), European Competition Law Annual 1998: Regulating Communications Markets, Hart Publishing, p. 635. 81 See Article 3(4) and Article 3(5) of the Framework

Directive, supra note 6. However, the

cooperation mechanisms are not far-reaching. Interaction between NRAs and NCAs is simply envisioned under the form of exchange of information. It is required that the receiving authority should ensure the same level of confidentiality as the originating authority. 82 See Recital 13 of the Access Directive, supra note 8. 83 See, on this principle, M.L. Struys (Mars 2000), ‘Le droit communautaire et l’application des régles procédurales nationales’, JTD Européen, 67, p. 1. 84 See Commission Decision of 21 May 2003, COMP/C-1/37.451, 37.578, 37.579, Deutsche Telekoin AG, O.J. L 263 of 14 October 2003, p. 9. See also Robert Klotz and Jér6me Fehrenbach (2003), “Two Commission Decisions on Price Abuse in the Telecommunication Sector’, Competition Policy Newsletter, 3, 8.

85 This confirms that competing operators seek to use all procedural avenues. 86 This is called a margin squeeze because even if competitors are as efficient as the infrastructure owner, they have to bear the costs of access. 87 Let us recall that the RegTP is the German NRA in the telecommunications sector. Pursuant to German law, the pricing of access to the local loop must be geared towards costs and have to be approved by the government. 88 See Commission Decision, supra note 84, §54. 89 See Commission Decision, supra note 84, §57. This solution is in line with the ECJ case law on

the restrictions of competition induced by State interventions. In short, competition rules can be enforced against undertakings as long as they keep some discretion. However, it shall be discussed whether these principles should be transposed in the field of regulation. See XXII Report on Competition Policy, §193; see also Michel Waelbroeck and Aldo Frignani, Commentaire J. Megret, Le droit de la CE, Vol. 4 Concurrence, 2nd Edn (1997), p. 149. 90 See Commission Decision, supra note 84, §212. This led to a 10% reduction of the fine.

91 See infra. 92 See Regulatory Scorecard Report, Jones Day, supra note 46. 93 See Recital 13 of the Access Directive, supra note 8. Also, the operators that should be submitted to sector-specific obligations are incited to ask the NRA the imposition of detailed and rigid remedies to bear no responsibility of an infringement of competition law. This is not in harmony with the approach trying to favour proportionate and flexible regulation. 94 See A de Streel, R. Queck et P. Vernet, supra note 2, ‘evoking the risks of forum shopping and duplication of proceedings’, p. 260. It can be done so before competition authorities (Commission and NCAs) as well as national courts. 95 This is particularly true if the regulatory remedy took the form of guidelines, but is probably also true when NRAs approve access conditions if NRAs, in practice, try to limit the straightjacket effects of their decisions. 96 See Commission Decision, §2, supra note 84.

97 This is confirmed in the Commission Decision, §165: ‘Accordingly, each time DT submitted applications for the approval of adjustments to charges under the price cap system, the regulatory authority made only a rough assessment to check that the applicable index figures were respected and that the proposed charges did not manifestly breach the requirements of the Telecommunications

210

Regulation through Agencies in the

EU

Act. In the majority of all six tariff adjustment applications between 1998 and 2001 this was the case’. This statement shows that the intensity of the RegTP control over DT’s tariffs was extremely

limited and could accordingly be criticized. 98 See OECD, supra note 70, p. 10. 99 Thid. ve 100 This is especially true for issues of access regulation where the overlap between competition rules and sector-specific regulation is maximal.

101 See 540 U. S., Verizon Communications, Inc. v. Law Offices of Curtis Trinko, LLP (2004) No. 02-682, US Supreme Court of January 13, 2004. 102 That may, indeed, slow down the pace of regulatory reforms, reduce the efficiency of regulatory schemes, increase legal uncertainty and affect investments in the sector. 103 To the difference of the Deutsche Telekom case, the following illustrations relate to cases where the Commission refrained from intervening ex ante because sector-specific regulation provisions were available and could be applied. 104 See Commission Decision, supra note 23. The agreement could have been used as a blocking tactic against competitors to slow down the pace of rolling out their networks.

105 A similar approach was taken in BT/MCI I, where in case of a strategic alliance, the Commission concluded that no conditions or obligations were needed, in view of the national regulatory frameworks to which both parties were submitted. The Commission nonetheless reserved the application of competition rules, if regulatory action proved unsatisfactory. See Commission Decision of 27 July 1994, BT-MCI O.J. L 223 of 27 August 1994, pp. 36-55, §57. See on this P. Larouche,

supra note 7, p. 312. This suggests, to a certain extent, that the Deutsche Telekom case should be read on its facts and, arguably, the Commission will only follow this line of action in a narrow set of exceptional circumstances. 106 See Commission Press Release IP/02/1852 of 11 December 2002, ‘Prices decrease of up to 40% lead Commission to close telecom leased lines inquiry’. 107 See Commission Press Release IP/98/707 of 27 July 1998, ‘Commission concentrates on nine cases of mobile telephony prices’. 108 Tn addition, similarly to the findings of the Supreme Court in the US, the risk of false positives before national courts in the EC should not be underestimated.

109 See, for a similar point of view, D. Geradin and G. Sidak, supra note 4. 110 Pursuant to the principle of ‘procedural autonomy’, in the absence of Community rules on the subject at stake, it is for the domestic legal system of each MS to designate the courts having jurisdiction and to determine the procedural conditions governing actions at law intended to ensure the protection of the rights which citizens enjoy by virtue of the direct effect of community law. See ECJ, 16 December 1976, Rewe-Zentralfinanz eG et Rewe-Zentral AG v. Landwirtschafiskammer fiir das Saarland, 33/76, ECR 1976, p. 1989. See, on this principle, Michel L. Struys (2000), ‘Le droit communautaire et l’application des régles procédurales nationales’, JT Droit Européen, 67, 1.

‘ll See Article 4 of Framework Directive, supra note 6: ‘Member States shall ensure that effective mechanisms exist at national level under which any user or undertaking providing electronic networks and/or services who is affected by a decision of a national regulatory authority has the right to appeal against the decision to an appeal body that is independent of the parties involved’. Also, it is necessary that, where the controlling body is not a tribunal, a further possibility of appeal before a court of law be instituted. This is so in order to safeguard the possibility to ask for a preliminary

reference to the Court. '2 Tn the energy sector, see Article 23(11) of Directive 2003/54, supra note 9 and Article 25(11) of Directive 2003/55, supra note 60. Appeal procedures must only be set up for cases of refusal of authorisations, see Article 6(4) of Directive 2003/54, supra note 9 and Article 4(3) of Directive 2003/55, supra note 60. In the postal and in the rail sectors, EC legislation does not provide any requirement to set up judicial review mechanisms at the national level.

"3, See e.g. the Ahmeed Saeed case, where a court had raised the question of the compatibility of a sector-specific entity’s decision with competition rules. 14 The practice, by some MS, to centralise these matters to a specialised chamber within a court of law shall thus be welcomed. 115 Tn practice, the ECJ and the CFI have never had the opportunity of ruling that an MS infringed EC law because of unlawful action by an NRA. However, there is nothing to prevent an MS being

The Proliferation of National Regulatory Authorities

211

held liable for unlawful action by an NRA. Indeed, the ECJ’s case law makes a very broad interpretation of the categories of public entities that can, on behalf of an MS, be held liable for an infringement of EC law on the basis of Article 226. See ECJ, 22 June 1, 89, Fratelli Costanzo SpA v. Comune di Milano, 103/88, ECR [1989] 1839; see Denys Simon (2002), Le systéme juridique communautaire, 3rd Edn, PUF, pp. 617-619. 6 See José Luis Buendia Sierra, Exclusive Rights and State Monopolies under EC Law (1999), Oxford University Press, $7.23 and p. 263. However, the author does not make reference to the CNSD case where the Commission initiated Article 226 EC proceedings against Italy for an infringement of the competition rules. See ECJ, 18 June 1998, Commission v. Italy, C-35/96, ECR [1998] I-3851. Also, it should be noted that most of the cases of combined infringements ofArticle 3(g), 10 and of the competition rules of the Treaty concern Article 81 and not Article 82 EC. "7 See J. Faull and A. Nikpay, supra note 13, §11.60. Also, on questions that have implications with regard to investments, the Commission might be reluctant to initiate Article 226 EC proceedings against MS. 8 See P. Larouche, supra note 7, p. 294. This does not, however, take into account the fact that infringement proceedings are generally not brought to an end and it is less than sure that procedures against undertakings be shorter than infringement proceedings. Only one case out of five leads to the introduction of an action before the ECJ and the rest is solved after the issuance of a letter of formal notice or of a reasoned opinion. See Denys Simon, Le systéme juridique communautaire,

3rd Edn, PUF, p. 658. f '19 See Commission Decision 95/489 of 4 October 1995 concerning the conditions imposed on the second operator of GSM radiotelephony services in Italy, O.J. L 280 of 23 November 1995, pp. 4957 and Commission Decision 97/181 of 18 December 1996 concerning the conditions imposed on the second operator of GSM radiotelephony services in Spain, O.J. L 076 of 18 March 1997, pp. 19-

29. 120 Of course, this procedural route will in the future become less attractive. The number of public undertakings or of undertakings enjoying exclusive and special rights is substantially decreasing in most network industries. 21

See ECJ, 30 June 1988, Commission v. Hellenic Republic, Case 226/87, [1988] ECR-3611,

§§11-12. 122 See ECJ, 19 November

1991, Andrea Francovich and Danila Bonifaci and others v. Italian

Republic, Case C 6-9/90, [1991] ECR I-5357. See also for a recent illustration ECJ, 30 September 2003, Gerhard Kébler v. Republik Osterreich, Case C-224/01, not yet published.

123 See CFI, 30 January 2002, Max.Mobil v. Commission, Case T-54/99, [2002] ECR II-313. 124 Tn several network industries’ sectors, the Commission has opted for a model of ‘managed

decentralisation’ whereby NRAs are entrusted with important regulatory powers with, however, a certain degree of supervision by the Commission. In the telecommunications sector, for instance, the new Framework Directive on electronic communications strengthens the powers of the NRAs, which will be ultimately responsible for defining markets and identifying the operator significant market power to which the heavier regulatory obligations contained in the specific directives will apply. In exchange for the increased discretionary powers granted to NRAs, the directive provides for a cooperation mechanism whereby whenever a NRA intends to take a specific measure ina field where it is granted a certain margin of appreciation (definition of the market, analysis of the market, etc.), and whenever this measure is liable to affect trade between MS, it has to inform the Commission

and the other NRAs. The Commission as well as the other NRAs may then make comments to the relevant NRA concerned within a certain period of time. The NRA has to take the utmost account of these comments when adopting its measure. In some cases, the Commission can even force the NRA to withdraw its draft measure. In addition, cooperation between authorities has been further institutionalised with the setting up of formal network. A Commission’s decision of July 2002 established a European Regulators Group (ERG) for Electronic Communications Networks and Services with a view to ensure the consistent application of the regulatory framework, see supra note 52. In the electricity and gas sectors, a similar ERG has been set up, see supra note 52. The networking approach is, however, more limited than for telecommunications. The EC framework

in the field of energy does not set up far-reaching procedures for ensuring cooperation between NRAs.

247.

Regulation through Agencies in the

EU

125 Tn Deutsche Telekom, for instance, there was clearly a possibility to comply with the two sets of rules. 126 Recently, the Finnish Telecom Authority (hereafter, ‘Ficora’) notified several measures (i.e. market definition, designation of Significant Market Power operators and imposition of remedies) to the Commission on the basis of Article 7 of the Framework Directive. The Commission made comments in early 2004 announcing that the remedies proposals were not in line with EC sectorspecific legislation. Given that the Framework Directive does not entrust the Commission with the power to invalidate measures proposed by NRAs, the Commission announced that it would not hesitate to commence 226 proceedings for bringing Ficora in line with EC sector-specific regulation requirements. It is not clear whether the Commission intends to develop such practice. The solution might instead have been motivated by the specific features of the case and, in particular, that the problems were of a legislative nature (Ficora bound by Finnish rules of transposition that define remedies in contrariety with the EC framework). See on this European Telecoms Newsletter, Amold & Porter and http://forum.europa.eu.int/Public/irc/infso/ecctf/home. 127 See Communication from the Commission, ‘Unbundled Access to the Local Loop: Enabling the Competitive Provision of a Full Range of Electronic Communication Services, including Broadband, Multimedia and High Speed Internet’, JOCE C 272 of 23 September 2000, pp. 55-66,

86. 128 See, for instance, Santiago Martinez Lage and Helmut Brokelmann, ‘The Respective Roles of

Sector Specific Regulation and Competition Law and the Institutional Implications’, in C.D. Ehlermann and L. Gosling, supra note 80. The authors evoke the danger of a ‘sector-specific body of competition law’, p. 662. 129 See, on this, Damien Geradin and Michel Kerf (2003), Controlling Market Power in Telecommunications — Antitrust v. Sector Specific Regulation, Oxford University Press, pp. 214-216 and pp. 350-352. See also, OECD, Relationship between Regulators and Competition Authorities, supra note 70, p. 30. 130 See Giuseppe Tesauro, ‘Relations between Regulatory and Competition Authorities in the Communications Industry: the Italian Experience’ in C.D. Ehlermann and L. Gosling, supra note 80, p. 701.

PART

FOUR

Agencies in a Comparative and International Perspective

|

: te

nn

oF Fywees is

AS Si

Nise! i) Ve, Veit

ne ws

s =

7 were

:

=

a

eee Piero

i

1

es iy Aeon : 7

Pe

Sale aware re ti Ibe

a

ihe)

4 ner

Con)

:

A

ee

vo

ay

;

*

weyers

“a,

ay

wieSge oa

as

a

arian

=) ae

=

P >

Laker

ee

a

e 6,

:

cane es "Ke

ee SS ae

,

-



Gi

yriesar®

es

Ce

OD

tg

©

Ye:

This expansion,

which

will be discussed below,

includes

the

recognition of ‘association standing’, which allows well-staffed and experienced groups to defend the interest of their members against the mistakes made by regulatory bodies.

228

Regulation through Agencies in the

EU

In sum, what matters to the regulated industries and citizens potentially affected by regulatory decisions is not so much to know that agencies are organized under a coherent and uniform model, but the ability to benefit from clearly established procedural rights, and have regulatory decisions reviewed when they go against their interests. Improving the EU model From the preceding analysis, one could draw the conclusion that European scholars and policy-makers should not waste any time on attempting to come up with a uniform organizational model for agencies, but should instead focus on what matters to individuals, i.e. procedural rights and judicial review. Yet, in the paragraphs that follow, I propose my own classification of EAs. Although this may appear in contradiction with the conclusion drawn from the prior section, this effort is not vain in the current European context. As we have seen, the European Commission is busy trying to define categories of agencies, but its proposed categories fail to convince and drive the debate in wrong directions. I will thus first address this problem of categorization with the objective of steering debate in the right direction and then move to the issue of procedural rights. In order to clarify and add analytical rigour to the debate, I propose my own classification of EAs, which no longer relies on a functional approach, but on a legal criterion, i.e. the intensity of the prerogatives entrusted to the agencies for carrying out their missions. Pursuant to this criterion, three categories of EAs can be identified. The first category refers to the ‘executive’ agencies, which would comprise all European agencies that are responsible for (i) purely managerial tasks (e.g. the EAR, the ETF and the CdT), or (ii) observatory roles (the EEA, the EMCDDA, and the EUMC), or (111) missions of cooperation (e.g. the Cedefop, the Eurofound or the EUOSHA). The common feature between these agencies is that they would have no decision-making power, all relevant decisions being taken by the Commission. The second category would refer to the ‘decision-making’ agencies, which would comprise all agencies that have the power to enact legal instruments binding on third parties (e.g. the OHIM, the CVPO, the EASA) or of the agencies that without enjoying a formal decision-making power, in fact enjoy a considerable influence over the adoption of final decisions by the Commission (e.g. the EMEA and possibly the EFSA and the EMSA). This category is being relied upon by the Commission in its Communication on the Operating Framework for the European Regulatory Agencies as a subcategory of the category entitled ‘regulatory’ agencies. I believe, however, that the interpretation of the concept of regulatory agencies given by the Commission is misleading as this concept is generally used among experts

The Development of European Regulatory Agencies

229

and in most jurisdictions to designate agencies, which have discretionary powers in translating political choices into action. My third category would thus refer to true ‘regulatory’ agencies, which would comprise all European agencies that enjoy the types of powers enjoyed by the NRAs, including a discretionary power to translate broad legislation guidelines into concrete regulatory instruments. Of course, such a category could not be created under the current restrictive interpretation laid down in Meroni. As indicated above, the creation of such a category of agencies would require either an overruling of Meroni by the ECJ or a modification of the Treaty. Although I am very much in favour of this evolution, several years may pass before it takes place. In the meantime, the Commission should refrain from calling regulatory agencies bodies that do not share the characteristics that are generally associated with such bodies. Once a classification has been proposed, the next relevant question is to determine the legal consequences, which will be attached to it. In adopting Regulation No. 58/2003, the Council has already addressed most of the relevant issues regarding the ‘executive’ agencies. An important question is whether comparable regulations should be adopted for the other categories proposed, i.e. the decision-making agencies and the regulatory agencies. In fact, at this stage, there is no need for a regulation dealing with my category called ‘regulatory’ agencies since the creation of such agencies would require an overtruling of the Meroni doctrine by the ECJ or a Treaty amendment. The question we thus face is whether there is a need to adopt a second regulation dealing with the ‘decision-making’ agencies. The main advantage of responding positively to this question is that adopting a second regulation would be easy. Indeed, the Commission is itself proposing to adopt an instrument that would address the issues raised by regulatory agencies, a concept I prefer to reserve to the yet-to-be-created bodies enjoying discretionary powers, but which by and large cover my proposed category of decision-making agencies. Another approach would be to expand Regulation No. 58/2003 to my proposed category of decision-making authorities, which would then deal with both executive and decision-making agencies. As these categories of agencies differ on a number of key issues, this regulation would thus have to be characterized by a high degree of flexibility. The regulation should provide a series of core obligations that would be applicable to both categories of agencies. As is the case in the US, EU legislation establishing EAs could apply additional requirements to them depending on the particularities of a subject matter. For the sake of clarity, I believe that this latter approach is

preferable as all procedural and other issues regarding EAs would be addressed in a single piece of legislation, which I would propose to refer to as the European Agencies Procedural Regulation (the “EAPR’).

230

Regulation through Agencies in the

EU

Principles of Good Governance

Over the past five years, concerns over good governance, understood in a broad sense, have considerably risen. A number of international organizations have, for instance, sought to promote the development of good governance principles throughout the world. Similar preoccupations have also increased in the EU. In 2002, the Commission’s White Paper on European Governance has identified a number of core principles of good governance, which include the requirements of independence, accountability, transparency and

participation.*° In this section, I seek to determine whether principles of good governance are generally complied with by the EAs. I also propose measures that could be adopted to ensure better compliance of the EAs with these principles. I address good governance requirements in general, although in some places I indicate that variations might be needed to cater for differences between the different categories of agencies. The EU Model

Independence Independence of the agencies is often seen as the most central principle of good governance.*’ It is generally understood as the absence of pressures from political and industry interests. As far as independence from political interests is concerned, it is important to outline a major difference between the EU and the US, which relates to the peripheral role of political parties in the EU model of governance. As pointed out by Martin Shapiro, the core of the idea of agency independence in the US was not so much to shelter agencies from the Congress and the administration, which, as will be seen below, are always able to influence these bodies, but to avoid political control

by one party.** In contrast, EU politics is not party politics. It is instead Member States politics (i.e., conflicts between Member States, but also tensions between the EU level and the national level), as well as institutional politics (i.e., power struggle between EU institutions). When I speak about independence in the EU, I thus essentially mean independence vis-a-vis industry interests, as well as independence vis-a-vis the Commission and the Member States. Independence is not an abstract principle. Its implementation requires the adoption of a series of measures, which will shelter the agency against undue pressures. A first set of measures relate to the appointment of the management of the agencies. With some limited exceptions, the Commission generally does not enjoy the power to appoint executive directors. It is only entrusted with the power to propose candidates, the appointment of the

The Development of European Regulatory Agencies

2a)

director being generally a prerogative of the management board of the agencies. Moreover, the power over appointments has recently been reduced with: (i) the insertion of professional criteria in the regulations setting up new agencies; and with (ii) the implementation of public calls for candidates in some cases. This should provide guarantees that the appointment of candidates is not biased (or at least perceived as such) in favour of the Commission. Second, it is subject to the question of whether the agencies are sufficiently independent vis-a-vis the MS. I believe it is not the case. Indeed, most of the members of the EAs’ management board are MS representatives.*’ It therefore cannot be precluded that considerations linked with national interests may sometimes be taken into account within the decision-making process. This situation has been explicitly recognized by the Communication on the Operating Framework for the European Regulatory Agencies in which the Commission considers that the national representation principle insufficiently takes into consideration the Community interest. ; Third, the constituent regulations do not generally provide for rules designed to prevent the so-called ‘revolving doors’ practices. Members of the agencies are merely required to comply with a confidentiality obligation over professional secrets, although more demanding requirements have started to be imposed in the most recent constituent regulations. Finally, since agencies are established pursuant to legislative procedures and are not so to speak, “Treaty-framed’, they remain exposed to a possible amendment of their status. The EU legislative institutions can therefore, at any time, limit the attributions of the agencies and could even decide to dismantle them. This could well be used in order to exercise influence over the agencies. Lack of accountability The existing measures designed to ensure that EAs are accountable appear to be insufficient. This observation may be difficult to reconcile with the previous observation that they should be more independent. Indeed, it is sometimes considered that accountability and independence are conflicting concepts.*” However, independence and accountability can and must be combined. The control exercised by public representative institutions may indeed result in greater legitimacy for the agencies. This may in turn enhance their independence and ability to resist to undue political pressures. ¢ Judicial Accountability In the EC legal order, the principal mechanism for ensuring judicial accountability of EAs takes the form of a review of the legality of their acts. The situation is here again extremely disparate. While some constituent regulations explicitly provide that the acts taken by the agencies are

PY

Regulation through Agencies in the

EU

challengeable before the ECJ on the basis of Article 230 of the EC Treaty (annulment proceedings), other constituent regulations entrust the Commission or a specific agency chamber with the review of the legality of the EAs’ decisions. In other cases, the constituent regulations do not even provide for the possibility of judicial review. In the latter hypothesis, the classic rules governing the annulment action shall be applied (Article 230 of the EC Treaty). Reliance on this Treaty provision raises several problems. First, the acts of the EAs are not mentioned within the scope of the acts reviewable pursuant to Article 230 of the EC Treaty.*! This problem, however, does not appear insurmountable given the ECJ’s broad interpretation of the acts reviewable under Article 230.” In addition, the rules governing locus standi are not satisfactory. According to Article 230 of the EC Treaty, individuals can launch proceedings at the CFI in two sets of circumstances. First, they can challenge the validity of a decision that is directly addressed to them. Second, they can launch proceedings against a decision addressed to another person (e.g. a competitor) or a regulation provided that they are ‘individually’ and ‘directly’ affected by the measure in question. The interpretation of the notion of ‘individual’ concern by the ECJ is, however, particularly restrictive. For instance, in order to allow individuals to challenge a regulation, the ECJ requires that this regulation affects them “by reason of certain attributes, which are particular to them or by reason of circumstances in which they are differentiated from all other persons and by virtue of these factors distinguished individually’.* In practice, this means that only regulations, which specifically identify a limited number of individuals will be challengeable at the CFI. In addition, the ECJ seems equally restrictive when it comes to allowing associations to challenge measures that would affect the interests of their members. In Federolio, the Court stated that associations may only have standing under Article 230 when

they represent the interests of individuals that would be entitled to bring proceedings in their own rights.” As the standing test for individuals is extremely restrictive, the circumstances in which association standing is recognized will also be very limited. ¢ Financial Accountability As far as financial accountability is concerned, EAs are submitted to several controls. A first mechanism is the voting of the EP on the resources of the agencies. As many agencies depend ultimately on Community funds, the budgetary power enjoyed by the EP on expenditures is rather important. The EP could well use such power to push agencies towards more transparency and to increase parliamentary scrutiny of their activities. However, this has not happened in practice and the analysis of the evolution of budgetary procedures does not support this argument. Second, a control is exercised by the Court of

The Development of European Regulatory Agencies

233

Auditors by virtue of Article 248 of the EC Treaty. The Court’s main duty is to determine whether the annual accounts are ‘reliable’ and the transactions ‘legal and regular’.** However, this control has a limited preventive value since the reports are issued by the Court of auditors more than one year after the operations have been made. The financial control of the EAs is thus limited in that it only takes place on an ex-post annual basis. Major improvements could, however, be made with the adoption of a budgetary mechanism similar to the executive oversight function carried out by the Office of Management and Budget (the ‘OMB’) in the US. This body, which is directly connected to the US President, prepares inter alia the budget of the various administrative bodies, controls their implementation and reviews their expenditures and purchasing policies on a permanent basis. This gives considerable incentives for diligent management. * Political Accountability The importance of accountability before representative institutions is widely acknowledged. However, few constituent regulations provide for obligations of regular reporting before the EP.*° The absence of such a mechanism has been heavily criticized in the context of the Treaty-framed European Central Bank. This criticism should be applied to EAs as well. The submission of entities to the scrutiny of parliamentary committees may prove useful as it may oblige the agencies to provide reasoned explanations of their initiatives. This may result in an overall improvement of the decision-making process. Another aspect of accountability is linked to the public. Seen from that standpoint, the accountability of EAs is also insufficient. In most instances, the constituent regulations only require the publication and presentation of an annual report of activities. In contrast with agencies in the US and in other countries, EAs are under no obligation to publish explanatory documents, issue regular reports or organize public consultations prior to decisionmaking. Mechanisms designed to ensure public accountability are thus very

limited. ¢ ‘Performance’ Accountability Finally, the absence of a regular assessment mechanism of the regulatory activity of EAs is regrettable. In fact, the impact of the regulatory activity in terms of indirect socio-economics costs and benefits should be examined regularly. This would first help in determining whether the entity has acted efficiently or has failed to implement efficient solutions. In this case, improvements of the regulatory output would be required. As will be seen below, review procedures of that kind have been implemented in the US, where the Office of Information and Regulatory Affairs (the ‘OIRA’) is in

234

Regulation through Agencies in.the

EU

charge of the assessment of regulatory programmes and of the ‘Regulatory Impact Analysis’ prepared by the federal agencies.’ This solution has also been implemented in some MS such as France, where ‘Parliamentary Offices of Assessment of Legislation and of Public Policies’ are carrying out regular reviews of the policies implemented by the IAA. Interestingly, some of the new constituent regulations provide for an ‘evaluation’ instrument pursuant to which the functioning, as well as the regulatory outcome of the agency have to be assessed after a varying period of time. This procedure should be generalized to all agencies. Lack of transparency Article 255 of the EC Treaty, which provides for a general right of access to documents of the Commission, the Council and the EP, is not clear as to whether

EAs have to communicate the documents produced in the regulatory process. However, Regulation 1049/2001 regarding public access to the European Parliament, Council and Commission documents explicitly provides for its application to the EAs.** In addition, the constituent regulations setting up the most recent agencies provide for the application of Regulation No. 1049/2001 to their own activities pursuant to arrangements decided by the management board. This new regime is thus to be welcomed. Transparency also implies a certain degree of clarity and simplicity of the regulatory processes. In this respect, the degree of transparency of certain EAs is far from ideal since much of the regulatory work is carried out by opaque committees. This can be illustrated by the functioning of the EMEA. Its processes involve four different kinds of actors. First, the EMEA is composed of two committees (the CMPP and the CVMP), which are in turn assisted by working parties that provide advice on specific issues and in some instances, specialized ad-hoc groups can be set up to address particular matters. Second, the agency is expected to cooperate with several transnational committees. Third, some issues have to be dealt with by the EMEA in cooperation with the national authorities. Fourth, the decision-making powers belong to the Commission. Hence, the combination of these levels of intervention, the composition of the committees, as well as their precise functioning is obscure. Lack ofparticipation Stakeholders, civil society and professional bodies are insufficiently involved in the decision procedures of the agencies. No mandatory requirements are set up in this respect and the extent of public participation solely depends on the agencies’ own commitment. This lack of participation may prove counter-

productive.” Indeed, the participation of stakeholders in the adoption of regulatory decisions involving technical matters presents several advantages. First, in highly specialized fields, it tends to help address the problem of

The Development of European Regulatory Agencies

235

asymmetry of information (between regulatory institutions on the one hand, and economic operators on the other hand), and it helps to ensure that the practical consequences of decisions are well evaluated. Second, the implementation of regulatory decisions is likely to be better accepted by its addressees if the body that has taken the decision is composed of well-recognized and informed professionals. Better participation can be achieved through a variety of measures. For instance, the insertion within the constituent regulation of professional criteria for the appointment of the Members of the Management board and of representation rules (i.e. rules ensuring some representation of each of the relevant category of stakeholders subject, however, to the limits placed in the next paragraph) could be a positive measure. In addition, participation may be improved by the organization of conferences, workshops, regular meetings, preliminary consultations (under the form of Green Papers, call for comments etc.), as well as the posting of draft documents on the Internet with the possibility to comment. The US Model

This is probably with respect to the issue of good governance that the EU could gain the most by looking at the US experience. Indeed, in the US, a wide range of mechanisms have been developed over time to address public concerns over independence, accountability, transparency and participation. Independence Among all the concerns listed above, independence is perhaps the issue, which is the most challenging when it comes to engaging in an EU-US comparison. Important contextual differences are indeed present. First, we have seen that

‘independence’ from politics had a different meaning in the US and the EU. While, in the US, agency independence refers to the absence of control from a single political party, in the EU, agency independence means a certain degree of independence from the legislative authorities and the Commission. Yet, it would be naive to think that the US is free of the kind of ‘institutional’ politics, which can be observed in the EU. In a piece entitled ‘Presidential

Administration’, Elena Kagan shows, for instance, the growing influence played by Presidents over agencies since the Reagan administration.*’ As will be seen below, the efforts made by the Commission (i.e. the EU executive) to retain a high degree of control on the agencies present some common features with US Presidential efforts to control the administration. In this regard, something that might trouble European observers concerns the high degree of politicization of some US regulatory agencies. The current FCC is a case in point. Because its decisions may have extremely important

236

Regulation through Agencies in the

EU

economic significance and touch social issues, such as universal service, the FCC has always been politicized.*' Yet, over the last couple of years, politicization has reached new heights. The bitter division between Chairman Powell and Commissioner Martin (both republicans) has been largely reported in the media. In fact, agencies may become deeply political bodies when they are run by politically ambitious individuals who may use their position in an agency as a stepping stone for a future political career. Another well-documented issue, which may strike Europeans, relates to the so-called revolving door practices, whereby high-level government officials, including Commissioners of regulatory agencies, accept positions in the industries they were called to regulate once they have left government service. By contrast, there is little tradition in the Commission to hire, for instance, an accomplished antitrust lawyer to head one of its divisions as senior appointments are generally made from the ranks. The establishment of EAs may, however, reverse this tendency as such agencies are able to recruit their executive director, as well as their staff from the private sector. As these agencies are relatively young and most of them have little powers, there are few documented stories of officials defending positions favourable to potential future employers. However, as the number of agencies grows, several of which deal with matters of economic significance and their power increase, rules preventing, or at least reducing the risks of, revolving door practices will have to be developed. The US regime, where rules have been designed to prevent conflict of interest during government service, as well as after government service, could offer some guidance to the EU, although it is subject to question whether these rules are sufficiently stringent.°> When drafting such rules a subtle balance must, however, be found between firmness and flexibility. Excessively restrictive rules whereby, for instance, government officials would be banned for a long period of time (e.g. five years) to influence their former agency after leaving government service, may deprive agencies from attracting the brightest individuals for which government service will often be only a first step in their career. Accountability US law imposes a high degree of accountability on regulatory agencies and this is probably one of the main strengths of the regime. After all, the best way to prevent agencies from abusing their powers is to make them accountable. In the paragraphs that follow, I explore various ways in which agencies are held accountable in the US, including controls imposed by the Judiciary, Congress and the President. As we have seen above, during the 1960s and 1970s, courts relaxed standing

tules, thereby enlarging the class of interests entitled to obtain judicial review

The Development of European Regulatory Agencies

237

of agency actions. This enlargement was interpreted by some scholars as a judicial response to the growing scope of regulatory intervention (and thus interests affected by regulation) combined with the growing distrust in the fairness of administrative processes.* As we have seen in the preceding section, another form of accountability relates to the control exercised by the political authorities. Congress is able to exert an important degree of control on federal agencies. It has several ways of exercising its oversight function. First, Congress decides the scope of the level of discretion it is willing to delegate to agencies. In the earlier age of the regulatory state, Congress controlled administrative action by adopting clear and precise legislative standards. However, as the administrative state became larger, Congress routinely resorted to broad delegations of power, giving a large amount of discretionary powers to agency officials. Nevertheless, Congress possesses several other ways to control agency discretion. For instance, Congress has the ‘power of the purse’ through the appropriation process on which it can rely to express its displeasure with agency actions. It can indeed place limits on agency action by forbidding expenditures for certain purposes. The appropriation process thus gives congress a powerful tool to influence the activities of regulatory agencies. Although the European Parliament has the power to vote on the resources of EAs, it is hard to say it has used this tool to control their actions. For instance, the questionnaire sent by the European Parliament’s Committee on Budgetary Control (the Committee, which looks at the way money is being spent) to the EAs in December 2003 regarding the discharge of the 2002 budget only seeks to verify whether budgetary procedures have been respected. The discharge of the budgets of the EAs is thus not used by the Parliament to enquire into agency’s policies and programmes, as well as to influence the direction of their work. Congressional oversight is another tool through which Congress can monitor the performance of an agency.*° Many constituent statutes provide for periodic oversight hearings by Congress. Moreover, a congressional committee can request a hearing at any time to investigate a particular policy of an agency. Once again, the situation is different in the EU where only few constituent regulations provide for regular reporting to the European Parliament. This is to be regretted since serious questioning by a well-informed parliamentary committee can be one of the most effective tools to assess the quality of the work performed by an agency, as well as the ability to reach the objectives set in its constituent regulation. The influence of the executive and administrative processes has expanded significantly over the last 20 years. This is to a large extent a response to the growth of the bureaucracy that has occurred during that period, which requires greater policy coordination to avoid policy conflicts, as well as excessive overlap, between agencies. The supervisory role played by the President is essentially

238

Regulation through Agencies in the

EU

exercised through the use of executive orders, which are primarily implemented through the Office of Management and Budget (the “OMB’), an entity within the Executive Office of the President (the ‘EOP’). Several executive orders present a particular interest in the context of this chapter. First, President Reagan adopted Executive Order 12,291, which requires agencies to submit to the OMB’s Office of Information and Regulatory Affairs (the ‘OIRA’) for pre-publication any proposed major rule, accompanied by a regulatory impact assessment (the ‘RIA’) of the measure in question. Four years later, President Reagan issued Executive Order 12,498, which requires that each agency submits for OMB review an annual regulatory plan listing proposed actions for the year, thereby giving the OMB an opportunity to influence agency action very early on in the regulatory processes. The two executive orders raised a great deal of controversy at the time as they were perceived as a tool to realize the Reagan administration’s agenda. In 1993, President Clinton replaced Executive Orders 12,291 and 12,498 by Executive Order 12,866 which retained the main features of the two Executive Orders with, however, some

variations, which moderated the most controversial

aspects of the Reagan Orders. Ten years after the adoption of Executive 12,866 RIAs, and the cost-benefit

analysis they imply, remain a subject of controversy among scholars. Although cost-benefit analysis has many partisans, it has also many opponents questioning such analysis on methodological and ethical grounds. Whether one likes cost-benefit analysis or not, the costs imposed by regulation on the US economy is such that one may question whether RIAs are effectively performed by the agencies. For instance, a research whereby the AEI Brookings Joint Center on Regulatory Studies reviewed 106 RIAs between 1981 and mid-1996 showed that less than half of final regulations passed a neutral benefit-cost test.*° According to that study, the net benefit (i.e., the dollar value of benefits minus costs) of the agencies’ activity during that period could have been increased by approximately $300 billion in present value terms if they had rejected such regulations.*’ These findings suggest that agencies generally fail to comply with Executive Order 12,866. Transparency

As we have seen in the prior section, lack of transparency is one of the main criticisms, which can be made against the decision-making processes of EAs. True, the EU recently adopted a Regulation regarding public access to European Parliament, Council and Commission documents, which also applies to the documents produced by the EAs.** Although the content of these statutes is not identical, one could thus say that the EU has its ‘Freedom of Information Act’.

But transparency means more than access to documents. It also means that regulatory processes should be as simple and clear as possible, which is an

The Development of European Regulatory Agencies

239

aspect that is particularly deficient in the EU. First, regulatory processes tend to vary from one EA to the other and, as pointed out above, there is not in the EU a set of general procedural requirements, such as those that can be found in the APA or in other horizontal legislation providing for such requirements. One great value of the APA is that it provides individuals with some basic procedural requirements, which can then be enforced by federal courts in case of non-compliance. Second, the regulatory processes of some of the EAs are of an extreme complexity (and opacity) as they involve several categories of actors, such as scientific committees, working groups, etc. Things are made even worse in areas where the Council authorizes the Commission to adopt regulatory measures on the basis of the so-called ‘comitology’. Although US administrative processes are not models of simplicity, agencies seem to be more autonomous in their decision-making processes than their EU counterparts, although we have seen above that Congress and the President will impose some checks on their policies. Moreover, the advantage of these processes is that they seem to involve much more transparent debates than those taking place in EU committees. Participation The APA provides for rules designed to ensure that potentially affected interests have a right to intervene in regulatory processes. However, in the 1970s, federal courts imposed additional requirements on agencies with a view to ensure greater participation. As with the extension of standing described above, the expansion of participatory rights could be seen as a response to the distrust vis-

a-vis administrative agencies, which were seen as ‘captured’ by vested interests. Advocates of extended participatory rights believed that if agencies had become captured by powerful interest groups, administrative processes would become fair again by allowing participation of traditionally underrepresented interests.” Compared with the EU system, the US administrative regime seems remarkably open to the participation of affected interests in regulatory processes. Improving the EU model There are important contextual differences between the EU and US regimes and thus there are some limits to what the EU could draw from the US experience. However, the US model contains several interesting features from which the EU could draw ideas for reform. One is the importance of having a broad legislation, such as the APA, which sets baseline regulatory requirements applicable to all agencies, except where otherwise provided. This legislation not only offers procedural guarantees to individuals affected by regulatory decisions, but they also introduce clarity into regulatory processes. The APA has been complemented by a range of other federal statutes such as, for instance, the Regulatory Flexibility Act, the

240

Regulation through Agencies in the

EU

Paperwork Reduction Act, the Freedom of Information Act, the Government

in the Sunshine Act, which apply across regulatory matters with a view to promote good governance principles in regulatory processes. By contrast, the EU has opted for a piecemeal approach where the procedural guarantees offered to individuals will essentially depend on what is provided for in each constituent regulation. Some improvements have been observed in the constituent regulations of the last EAs created, which require these agencies to comply with a range of requirements designed to further principles, such as accountability, transparency or public participation. For a variety of reasons, this piecemeal approach is, however, insufficient. First, the insertion of these new principles only affects the newly created agencies. Second, the good governance requirements included by new generation constituent regulations remain too limited. For instance, the provisions of the recent constituent regulations generally fail to provide sufficient requirements regarding the assessment of regulatory outputs. The better approach is the adoption of a European Agency Regulation (EAPR), which would contain baseline procedural requirements designed to promote compliance with good governance principles. This EAPR could also contain provisions regarding the legal basis on which agencies could be created, their structure, appointment mechanisms, etc. This regulation should, however, be flexible in order to accommodate different regulatory structures. In any event, as in the US regime, this EAPR could be completed by additional legislation to deal with specific problems of regulation. Another important line of reform for the EU regulatory processes relates to the reliance on RIAs. As we have seen, RIAs are a common feature of agency practice in the US, although some doubt can be expressed as to the quality of such assessments. In June 2002, the Commission adopted a Communication on

Impact Assessment, which indicates its ambition to develop an integrated assessment process, which would ‘integrate, reinforce, streamline, and replace’

all the existing separated impact assessment mechanisms.” It seems, however, that the Commission would limit the application of its integrated assessment process to major initiatives only, understood as those presented by the Commission in its annual policy strategy or its work programme. Agency work is not mentioned in the list of actions, which should be submitted to such assessments, although we have seen that some of the constituent regulation of some of the most recent agencies mentioned it. Once again, it would be preferable to adopt rules applying to all agencies. A third important area for possible reform relates to government ethics. Revolving door practices have become a concern in the EU, but here again EU institutions seem to follow a piecemeal fashion. An integrated approach to this issue, such as, for instance, through a horizontal conduct of conduct applicable to all EA officials, would be preferable.

The Development of European Regulatory Agencies

241

A fourth area where the EU could look at the US for possible reform concerns judicial review. All the constituent regulations of EAs should provide for judicial review mechanisms. This issue could also be dealt with through a horizontal regulation. A more flexible interpretation of standing requirements would also be desirable.

CONCLUSION The creation of a growing number of agencies at both national and European levels is one of the most significant developments in the administrative structure of the EU and its Member States. These agencies have generally played a useful role as they have allowed the Commission to decentralize a number of scientific, technical or observatory functions to specialized bodies. Yet, I believe that several steps could be taken to enhance the methods of

functioning of these agencies with a view to improve their performance, as well as to ensure a greater degree of coherence in the way they are organized, interact with each other, and comply with good governance principles. First, the Meroni doctrine should be revisited so as to allow the EU institutions to create fully-fledged regulatory bodies. The US experience teaches that a flexible interpretation of the delegation doctrine is not only compatible with the institutional balance between the branches of government, but can even strengthen such balance by allowing these branches to focus on their core business. The current system whereby the Council delegates implementing powers to the Commission, which relies in turn on opaque committees for guidance is ineffective and strengthen the perception that the EU is an obscure institution. Second, the Commission

should revisit its classification of EAs, as well

as the procedural regimes applicable to them. The classification proposed by the Commission in its Communication on the Operating Framework for the European Regulatory Agencies fails to convince and risks introducing additional confusion on an issue that is already plagued by conceptual inconsistencies. I thus suggest that the Commission should opt for my proposed classification pursuant to which EAs would be divided into three categories: (1) the ‘executive’ agencies, which would comprise all European agencies that are responsible for (1) purely managerial tasks, or (ii) observatory roles, or (iii) missions of cooperation; (2) the ‘decision-making’ agencies, which would comprise all agencies that have the power to enact legal instruments binding on third parties or of the agencies that without enjoying a formal decision-making power, in fact enjoy a considerable influence over the adoption of final decisions by the Commission; and (3) the ‘regulatory’ agencies, which would comprise all European agencies that enjoy the types of powers

242

Regulation through Agencies in the

EU

granted to the NRAs, including a discretionary power to translate broad legislation guidelines into concrete regulatory instruments. As far as the procedural regime applicable to these agencies is concerned, the US experience shows the importance of having a general legislation, such as the APA, establishing basic procedural rights applicable to agency processes. The piecemeal approach followed by the EU is inadequate and I would thus favour adoption of a more integrated piece of legislation regulating agency processes. In practice, the Commission should expand Regulation No. 58/2003 to our proposed category of decision-making authorities, which would then deal with both executive and decision-making agencies. Ideally, this new expanded regulation, which I would refer to as the ‘EAPR’, would not only deal with the organizational and procedural matters currently dealt with by Regulation No. 58/2003, but also to good governance principles, the importance of which is increasingly significant. In sum, many useful insights can be gained by looking at the US experience with regulatory agencies. Of course, the limited scope of this study did not allow a full critical assessment of the US regime. Instead, I have focused on identifying procedures, which could be used as a source of inspiration for regulatory reform in the EU. Needless to say that the US will need important regulatory reforms in the future as well, especially considering the amount of over-regulation plaguing the US economy. Perhaps at that time, US scholars and policy-makers will start looking at EU experience as well.

NOTES ! See White Paper on European Governance, COM (2001) 428 final. ? The word constitutional is placed between quotations marks as the EU does not yet formally have a Constitution. 3 See, e.g., Peter L. Strauss (1984), ‘The Place of Agencies in Government: Separation of Powers and the Fourth Branch’, Columbia Law Review, 84, 573. 4 See Article 251 of the EC Treaty, 5 In 1988, a Court of First Instance (hereafter, the ‘CFI’) was attached to the ECJ. This Court knows of all direct actions against Community legal acts brought by natural or legal persons, albeit subject to the control of the ECJ. See Article 225 of the EC Treaty. ® See Xenophon A. Yataganas (2001), ‘Delegation of Regulatory Authority in the European Union ~The relevance of the American Model of Independent Agencies’, Jean Monnet Working Paper, 3/

O1, p. 13. 7 Details on existing agencies are available at http://europa.int/agencies/index_en.htm. 8 See Meroni v. High Authority (Case 9/56), E.C.R.,1957/1958, p. 133.

> Thid., §40. 10 Tbid., §44. 'l See, for instance, the extensive interpretation given by the Commission of Article 86 of the EC Treaty. See Jose Luis Buendia Sierra (2000), Exclusive Rights and State Monopolies under EC Law, Oxford University Press.

'2 Under Article 202 of the EC Treaty, it is for the Commission to implement legislation at EU level. In practice, each legislation specifies the scope of the implementing powers granted to the

The Development of European Regulatory Agencies

243

Commission and how the Commission is to use them. Often, legislative instruments also provide

that the Commission is to be assisted by a committee pursuant to the mechanisms provided for by the Comitology decision. '3 See A.L.A Schechter Poultry Corp. v. United States, 295 U.S. 495 (1935) and Panama Refining

Co. v. Ryan, 293 U.S. 388 (1935). '4 See, e.g., Mistretta v. United States, 488 U.S. 361, 372 (1989). 'S See Industrial Union Dept. AFL-CIO vy. American Petroleum Institute, 448 U.S. 607, 685

(Rehnquist, J concurring). '© See, e.g., Lawrence Lessig and Cass R. Sunstein (1994), ‘The President and the Administration’, Columbia Law Review, 94, 1, 98. '7 See Koen Lenaerts (1993), ‘Regulating the Regulatory Process: “Delegation of Powers” in the European Community’, European Law Review, 18, 24, 42. '8 See, for instance, Ellen Vos (2000), ‘Reforming the European Commission: What Role to Play for the European Agencies’, Common Market Law Review, 37, 1113, 1124. '9 A similar proposal had also been formulated in the feasibility study (known under the name of *Penelope’ project) prepared by some Commission’s officials under the responsibility of President Prodi. in Article 71, the Penelope project provided a legal basis for the setting up of EAs, but limited their role to ‘exercising tasks of programme management for the Union or for providing scientific expertise’. They could also be given responsibility for taking decisions for the application of laws. g 20 With respect to legal proceedings, the draft Constitution realizes a number of improvements: see Article I1I-270 (annulment proceedings), Article III-270 al.1. (reviewable acts), Article II]-272 (failure to act proceedings) and Article IIJ-273 (compliance with ECJ rulings) of the proposed Constitution. The acts of agencies that are intended to produce legal effects vis-a-vis third parties are thus challengeable on the basis of annulment proceedings. 21 See Communication from the Commission, ‘European Governance: Better Law Making’, COM

(2002) 275 final, p. 5. 22 See Paul Craig (2003), ‘The Community Political Order’, Indiana Journal of Global Legal Studies, 1, 131. The exceptions are the EEA (Article 175 of the Council Regulation No. 1210/90 of 7 May 1990, O.J. L 120 of 11 May 1990 as amended by Council Regulation No. 933/1999 of 29 April 1999, O.J. L 117 of S May 1999, as well as the more recent agencies, i.e. the EFSA (Articles 37, 95, 133 and 152(4)b of Regulation No. 178/2002, O.J. L 31 of 1 February 2002, the EASA (Article 80(2) of European Parliament and Council Regulation No.

1592/2002, O.J. L 240 of

7 September 2002, the EMSA (Article 80 (2) of European Parliament and Council Regulation No. 1406/2002, O.J. L 208 of 5August 2002 ), as well as the two proposed agencies, i.e. the NISA (Articles 95 and 156) and the ERA (Article 71 (1)).

23 See Yatanagas, supra note 7. 24 See Council Regulation No. 58/2003 of 19 December 2002, O.J. L 11 of 16 January 2003, pp. 1-

8. 2S This is, for instance, the case of the EASA or of the proposed NISA, which could also belong to the implementation category as they are in charge of monitoring the implementation of the EU legislation in the field of air transport and information security. Similarly, the EMSA could be classified in the cooperation category in that it functions on the basis of collaboration between MS experts and the EFSA could be classified in the regulatory model in that it shows analogies with the functioning of the EMSA. 26 See White Paper on European Governance, supra note 2, p. 24. 27 See Communication from the Commission, ‘European Governance: Better Law Making’, supra note 21, p. 5.

28 See Communication from the Commission ‘an Operating Framework for European Regulatory Agencies’, COM (2002) 718 final. 29 Thid., p. 3. 30 Tbid., p. 4. 31 See Alfred C. Aman and William T. Mayton (1993), Administrative Law, West Publishing, 603. 32 Tbid., p. 605. 33 This Act is codified at 5 U.S.C.A., §§551-559, 701-706, 1305, 3105, 3344, 5372, 7521.

244

Regulation through Agencies in the

EU

34 The basic principle of judicial review of agency decision can be found in the APA, 5 U.S.C.A., §701-706. u See Richard B. Stewart (1975), ‘The Reformation of American Administrative Law’, Harvard Law Review, 88, 1669.

36 See Commission’s White Paper on Governance, supra note 2. 37 Yet, it is not sure that all kinds of EAs need to be independent. For instance, this is probably not the case of the purely executive agencies. In addition, the optimal degree of independence shall certainly be different from one category of EA to another. 38 See Martin Shapiro (1997), ‘Independent Agencies: US and EU’, Journal of European Public Policy, 4, 276.

39 Dehousse observes that there is an evolution on this point. Recent proposals suggest that the administrative boards of future agencies might not include representatives of the MS. See Renaud Dehousse (2002), ‘Misfits: EU Law and the Transformation of European Governance’, Jean Monnet Working Paper 2/02. 40 See Michelle Everson (1995), ‘Independent Agencies: Hierarchy Beaters?’, European Law Journal, 2 (180), 183, speaking of a ‘dilemma for any lawyer concerned with the institutional design of independent agencies’. 41 See Paul Craig (2003), ‘The Constitutionalization of Community Administration’, Jean Monnet Working Paper 3/03. 42 See Les Verts v. Parliament (Case 294/83), E.C.R., 1986, p.1339 and Maurissen v. Court of

Auditors (Case 193-4/87), E.C.R., 1989, p. 1045. 43 See Greenpeace v. Commission (Case C-321/95), E.C.R., 1995, Il, p. 2205; Plaumann v. Commission (Case 25/62), E.C.R., 1963, p. 95. 44 See Federolio v. Commission (Case T-122/6), E.C.R., 1997, Ul, p. 1559; Greenpeace v.

Commission, C-321/95 P, E.C.R., 1995, Il, p. 2205. 45 See Article 248 (1) al 2 of the EC Treaty. 46 This is the case of the Regulation establishing the EASA (see Article 29 of European Parliament and Council Regulation No. 1592/2002, O.J. L 240 of 7 September 2002). However, it must not be forgotten that the classic mechanisms of parliamentary questions remain available. In such a hypothesis, controls would only take place on a non-regular basis. Hence, the Commission has proposed that agencies be submitted to regular parliamentary hearings. See Communication from the Commission ‘an Operating Framework for European Regulatory Agencies’, supra note 28, pals? 47 Tn the US, the OIRA reviews the agencies output on the basis of a cost-benefit analysis. For a good description of the OIRA, see Alfred C. Aman Jr. (1993), Administrative Law and Process, Mathew Bender, Casebook series, p. 634.

48 See European Parliament and Council Regulation No. 1049/2001 of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, O.J. L 145 of 31 May

2001 p. 43. 49 The degree of participation varies with the agencies. Some agencies include representatives from the public in their decision-making structures (EUOSHA), while others rely on voluntarily mechanisms such as conferences and meetings (EEA). Finally some agencies do not really involve stakeholders in their activities. 50 See Elena Kagan (2000), ‘Presidential Administration’, Harvard Law Review, 114, 1245. 5! For instance, former FCC Chairman Reed Hundt was criticized for working very closely with the White House during its tenure. See Randolph J. May (2002), ‘An Agency for Bill and Al’, Legal Times, 2 June 2000.

52 This conflict reached its peak in February 2003 where Commissioner Martin sided with two democrats to block the deregulatory plan supported by Chairman Powell. See ‘FCC Loosens Broadband Rules’, 20 February 2003, CNET News.com, available at http://news.com.com/21001033-985313.html?tag=fd_lede2_hed. °3 For a good summary of such rules, see United States of Government Ethics, Understanding the Revolving Door, May 2000. %4 See, for instance, the theory of capture developed by George Stigler. George Stigler (1971), ‘The Theory of Economic Regulation’, Bell Journal of Economics, 2, 3-21. 5° See Aman and Mayton, supra note 31, p. 607.

The Development of European Regulatory Agencies

245

56 See Robert W. Hahn and Robert E. Litan (March 2003), ‘Recommendations for Improving Regulatory Accountability and Transparency’, Testimony before the House Government reform Committee Subcommittee on Energy Policy, Natural Resources and Regulatory Affairs, p. 5.

57 Tid. 58 European Parliament and Council Regulation No.1049/2001 of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, supra note 48. °° For a discussion of that vision, see Stewart, supra note 35, p. 1760. 60 See Communication from the Commission on Impact Assessment, COM (2002) 276 final, p. 5.

11.

Regulation and Globalization: Interactions between International Standard-setting Agencies and the European Union Jan Wouters and Sten Verhoeven

INTRODUCTION Regulation and globalization are often seen as antonyms: regulatory measures, especially at the national level, often impede globalization, particularly in the field of international trade. However, rather than being an obstacle to international trade, the creation of international standards may be conducive of globalization and economic integration at regional and global levels. As many other contributions to this book have already shown, at the level of the European Union (EU),' one sees a steadily increasing number of agencies emerging that contribute to standard-setting at the European level. However, this phenomenon should not make us forget that, at the global level too, standard-setting by a variety of international bodies — further on referred to as ‘international standardsetting agencies’ — is on the rise. While the rules adopted by international standard-setting agencies are only exceptionally binding in and of themselves and are rarely visible in daily life, their impact on international, EU and national law should by no means be underestimated. This chapter will examine, by means of some examples,’ the impact of these international standard-setting agencies on the EU as well as the EU’s position vis-a-vis, and influence on, such international standard-setting agencies. In addition, it will make the case for the need of such international standard-setting agencies by weighing the merits of international standard-setting vis-a-vis mutual recognition agreements.

246

Regulation and Globalization

247

THE RELATIONSHIP BETWEEN THE EU AND INTERNATIONAL STANDARD-SETTING AGENCIES More than may be readily seen or willingly admitted, international standardsetting agencies have an impact on EU law-making. We will seek to demonstrate this by using four examples in a variety of fields: we will explore the impact on EU law of the standards of (i): the Codex Alimentarius Commission, the International Office of Epizootics and the Secretariat of the

International Plant Protection Convention; (ii) the Basel Committee on Banking Supervision; (iii) the Financial Action Task Force (FATF); and (iv) the United Nations Economic Commission for Europe (ECE). On the other hand, international standard-setting agencies often provide a forum to allow the EU to persuade other countries and international organizations to apply certain standards. The Impact of International Standard-setting Agencies on the EU The Codex Alimentarius Commission, the International Office of Epizootics and the Secretariat of the International Plant Protection

Convention: the WTO route The Codex Alimentarius Commission was established in Rome in 1961, as the

principal executive body of the Food Standards Programme jointly developed by the United Nations Food and Agriculture Organization (FAO) and the World Health Organization (WHO).* The purposes of this programme are to protect consumers’ health and to ensure fair trade practices in the food trade, as well as to promote coordination of all food standards work undertaken by international governmental and nongovernmental organizations (NGOs).° To achieve these goals, the Codex Alimentarius Commission develops food standards and guidelines, along with codes of good practice. The Codex Alimentarius Commission became fully operational in 1963 and today has 170 Member

Countries,

including, since the end of 2003, the

European Community (EC).° Standards are adopted according to a complex procedure,’ but this has not hindered the Commission in adopting a multitude of food-related standards in numerous fields, dealing inter alia with hygiene, labelling and pesticide residues, which it publishes in the Codex Alimentarius.* Much of the work is progressed by various committees, manned by representatives of Member Countries.’ The full Codex Alimentarius Commission meets annually. The adopted standards are not binding on the Member Countries; they are simply recommendations and, in that sense, soft law.'° The International Office of Epizootics is the oldest of the three bodies. It was set up as early as 1924 by the International Agreement for the creation of an Office International des Epizooties in Paris.'' This intergovernmental

248

Regulation through Agencies in the

EU

organization differs from the other two examples in this section in that it does not operate under the auspices of the FAO." Its objectives consist largely of informing governments of the existence and evolution of animal diseases in the world, recommending measures to combat animal diseases, coordinating different studies on surveillance and control of animal diseases on the international level’? and promoting the harmonization of regulations dealing with trade in animals and animal-related products.'* The organization currently has 167 Member Countries. The EC is not a Member but in 2004 an agreement was concluded between the organization and the European Commission establishing official relations between them.'° The International Committee of the International Office of Epizootics, consisting of delegates from all Member Countries, convenes at least once a year to approve new standards. The organization also has five regional commissions that encourage cooperation on animal health issues in their respective regions. Like the Codex Alimentarius Commission, this veterinary body has a long tradition of establishing international standards. These can be found in its Code, which lists standards for international trade, and its Manual, which sets out standard diagnostic

procedures and vaccine standards. In addition, the International Office of Epizootics has established a separate Code and Manual on aquatic animals.’® The International Plant Protection Convention, to which currently 132 States are a party, was drafted and signed in Rome in 1929,’ with the aims of preventing the introduction and spread of pests and diseases of plants and plant products and of promoting measures for their control. However, its entry into force only came in 1952 — after the FAO Conference had approved the Convention for submission to Governments for acceptance in 1951 — and even then it did not provide for an international standard-setting body. Instead, the Convention obliges each contracting Government to create national plant protection organizations'* and lays down conditions for the issuing of phytosanitary certificates by the contracting Governments." The latter have the right to limit or prohibit the imports of plants to prevent parasites gaining access to their territory. This right is not absolute, though, since restrictions can only be based on phytosanitary considerations and must be published and communicated to the FAO.” In case of a dispute it is for the contracting Government or Governments concerned to ask the Director-General of the FAO to appoint a committee of experts. The experts will draft a report with recommendations, which, although not binding, should be the starting point for a mutually acceptable solution.”! The regime of the International Plant Protection Convention proved not very effective. As a result, the FAO started to reform the Convention. A Secretariat, located in Rome, was created in 1989 by the FAO Conference, but it did not function until 1993. The purpose of the Secretariat is to coordinate international efforts concerning plant quarantine issues, to compile information

Regulation and Globalization

249

concerning plant pest outbreaks, and to provide technical assistance on phytosanitary issues.** The Secretariat also drafts recommended international standards with the aid of nine regional bodies* — which are sometimes more active than the Secretariat in this respect** — and it coordinates the implementation of the International Plant Protection Convention by these nine regional bodies. Despite these reforms, the regime of the International Plant Protection Convention has proved not to be very effective. A revised text of the Convention has been adopted by the FAO Conference in 1997 in order to, notably, bring the Convention in line with the Agreement on Sanitary and Phytosanitary Measures (SPS), one of the multilateral agreements of the World Trade Organization

(WTO), and to allow Member Organizations of the FAO (such as the EC) to becoming a Party to the Convention. The new text has not yet entered into force, though.” Although the standards

of the Codex Alimentarius

Commission,

the

International Office of Epizootics and the Secretariat of the International Plant Protection Convention are not binding, their practical significance is strongly enhanced through the SPS. SPS recognizes the basic right of every WTO Member to protect public health within its territory.*° As a result, Members can take measures affecting international trade in goods which they deem to be a danger to the health of human beings and animals or to the preservation of plants.?’ These measures should be non-discriminatory, necessary for the protection of public health and based on scientific evidence.** Since this may open the door to unnecessary trade restrictions, SPS provides for a mandatory mutual recognition clause” of each other’s measures as long as these measures are compatible with the SPS, and an obligation to harmonize sanitary and phytosanitary measures on the basis of international standards harmonization is the correct term; the article is aimed at members of the WTO individually, although in practice they follow collectively the international standards set out by the three agencies; they can adopt stricter measures if based on scientific evidence, but in reality this is mostly impossible to prove.*° These international standards are defined in Annex A of SPS as the standards, guidelines and recommendations of the Codex Alimentarius Commission, the International

Office of Epizootics and the Secretariat of the International Plant Protection Convention in cooperation with regional organizations operating within the framework of the International Plant Protection Convention, and ‘for matters

not covered by the above organizations, appropriate standards, guidelines and recommendations promulgated by other relevant international organizations open for membership to all Members, as identified by the [Committee on Sanitary and Phytosanitary Measures]’.*' Since the latter Committee has not yet identified other relevant international organizations, only the standards of the three named international bodies are incorporated in the regime of SPS. Consequently, this international agreement obliges WTO Members to use

250

Regulation through Agencies in the

EU

these relevant standards as guidelines for enacting trade-restricting measures, unless there is scientific justification for adopting higher standards.** Although, as stated above, the standards of the Codex Alimentarius Commission, the

International Office of Epizootics and the Secretariat of the International Plant Protection Convention are not legally binding in themselves, in practice, Article 3 SPS has the effect of converting these standards into obligations for WTO Members, since it is often too difficult or even impossible to scientifically prove that higher standards are required to protect public health.* As a result, the standards of the Codex Alimentarius

Commission,

the International

Office of Epizootics and the Secretariat of the International Plant Protection Convention play an important role in the WTO’s dispute settlement mechanism. This has led to decisions holding certain measures taken by WTO Members to be incompatible with WTO law. The EU had its share in the famous Beef Hormones cases.*4 In that case, the EU had banned the use of growth-inducing hormones in beef production, resulting in an import prohibition in 1989. Canada and the United States opposed this policy and contended that the hormones used were safe and posed no threat to human health. As a result, the decision of the EU was scientifically flawed and rather a means of protecting its own beef production. Since consultations did not result in mutually satisfactory solutions for the parties involved, WTO dispute settlement panels were formed. The two panels released their reports in 1997,*° concluding that the EU measures deviated from the international standards of the Codex Alimentarius and thus were not consistent with Article 3.1 SPS. However,

the EU argued that a

higher standard was desired based on risk assessment as is required by Article 5 SPS. The Panels disagreed, and held that the EU had failed to demonstrate that its measures

were indeed based on risk assessment. The EU appealed, but

although the Appellate Body rejected a number of the Panels’ arguments, it largely confirmed the Panels’ decision by holding that the EU policy was not based on adequate risk assessment.*° Apart from the SPS, the standards of the Codex Alimentarius Commission,

the International Office of Epizootics and the Secretariat of the International Plant Protection Convention are also relevant in relation to the WTO’s Agreement on Technical Barriers to Trade (TBT). Like the SPS, the TBT is

part of the multilateral agreements on trade in goods negotiated during the Uruguay Round. It regulates the use by WTO Members of technical regulations and standards in order to diminish the resulting obstacles to international trade. Under the TBT, a technical regulation is defined as a document which lays down product characteristics or their related processes and production methods, including the applicable administrative provisions, compliance with which

is mandatory,*’

whereas a technical standard is a

document approved by a recognized body, that provides, for common and

Regulation and Globalization

Pay |

repeated use, rules, guidelines or characteristics for products or related processes and production methods, compliance with which is not mandatory.** Sanitary and phytosanitary measures in the form of a technical regulation or technical standards fall under the scope of the TBT. As a result, the Agreement is not applicable to measures covered by the SPS and vice versa.*” In any event, the TBT prohibits technical regulations and standards which are more restrictive than necessary to realize a legitimate objective, such as the protection of human and animal health and the preservation of plants.*° Furthermore, the TBT provides that if technical regulations are necessary, WTO Members have to base these technical barriers on relevant existing or nearly existing international standards,"' in particular those of the Codex Alimentarius Commission, the International Office of Epizootics, and the Secretariat of the

International Plant Protection Convention. An example of the application of these standards under the TBT can be found in the case concerning the EC Trade Description of Sardines.” This case dealt with the sale of certain types of sardines. An EU Regulation® required that a certain type of sardines should be used in order to be sold as ‘canned sardines’. However, this species was mainly found in the North-East Atlantic Ocean, in the Mediterranean Sea and in the Black Sea. Peru was essentially dependent on another type of sardines and could not sell its canned sardines in the EU. Since both types of sardines were allowed under the Codex Alimentarius, Peru filed a complaint with the WTO. On 29 May 2002 the Panel decided that the EU Regulation concerned constituted a technical regulation, inconsistent with the TBT. The Appellate Body confirmed the Panel’s decision holding that the Codex Alimenarius Commission was indeed a regulatory body within the scope of the TBT and that the disputed standard — Codex Stan. 94 — was relevant to decide the case. Whereas the examples above demonstrate how the standards of the Codex Alimentarius Commission, the International Office of Epizootics and the Secretariat of the International Plant Protection Convention become legally binding for the EU through the operation of WTO agreements, these standards are also frequently referred to in EU law. These technical norms appear not

only in regulations“ and directives,’ where they often serve as minimum standards, but also in agreements between the EU and third countries. For example, the Framework Agreement for Trade and Cooperation between the European Community and its Member States and the Republic of Korea expressly refers in its Article 6 to the Codex Alimentarius.*° Moreover, the European Court of Justice takes these standards into account in determining whether a national measure restricting the free movement of goods comes under the exception of Article 30 (ex Article 36) of the EC Treaty.

Doe

Regulation through Agencies in the

EU

The Basel Committee on Banking Supervision

The Basel Committee on Banking Supervision was established by the central bank governors of the Group of Ten (Belgium, Canada, France, Germany, Italy, Japan, Luxembourg, the Netherlands, Spain, Sweden, Switzerland, the United

Kingdom and the United States) at the end of 1974 in the face of the Herstatt Bank collapse*’ and it meets regularly four times a year. The Committee is supported by 30 technical working groups and task forces. The Committee does not have a formal status and its conclusions do not — and were never intended to — have legal force. Rather, it formulates broad supervisory standards and guidelines and recommends statements of best practice in the area of banking regulation** in the expectation that individual authorities will take steps to implement them through detailed arrangements — statutory or otherwise — which are best suited to their own national systems.” In this way, the Committee encourages convergence towards common approaches and common standards without attempting detailed harmonization of Member Countries’ supervisory techniques. The Committee reports to the central bank Governors of the Group of Ten countries and seeks the Governors’ approval for its major initiatives.*° In addition, however, as representatives from supervisory authorities in the financial sector which are not central banks participate to the Committee’s works, the standards which it develops carry the commitment of many national authorities outside the central banking community.*' These standards cover a very wide range of financial issues. One of the most important objectives of the Committee’s work has been to close gaps in international supervisory coverage in pursuit of two basic principles: that no foreign banking establishment should escape supervision; and that supervision should be adequate. To achieve this, the Committee has issued a large series of documents since 1975. In 1988, the Committee introduced a capital measurement system commonly referred to as the Basle Capital Accord.*? Since then, this framework has been progressively introduced, not only in Member Countries but also in virtually all other countries with active international banks. In June 1999, the Committee issued a proposal for a New Capital Adequacy Framework to replace the 1988 Accord, which was after years of consultations finally approved in April 2003, with the aim to introduce it at the end of 2006." The new capital framework consists of three pillars: minimum capital requirements, which seek to refine the standardized rules set forth in the 1988 Accord; supervisory review of an institution’s internal assessment process and capital adequacy, and effective use of disclosure to strengthen market discipline as a complement to supervisory efforts.** Over the past few years, the Committee has moved more assertively to promote sound supervisory standards worldwide. In close collaboration with many non-G-10 supervisory

Regulation and Globalization

253

authorities, the Committee developed a set of Core Principles for Effective Banking Supervision in 1997, which provides a comprehensive blueprint for an effective supervisory system.*° To facilitate implementation and assessment, the Committee developed the Core Principles Methodology in October 1999.*”

The Basel Committee has been indisputably influential.** While it lacks the authority to actually make law, its pronouncements are nearly always faithfully incorporated into the national regulatory regimes of the G-10 States and many other countries around the world.’ The EU, the United States and

many non-G-10 countries have adopted the basic principles of the Basel Accords, although slight deviations exist. For example, the EU and most of the non-G-10 countries have adopted a Minimum Capital Adequacy Ratio similar to that of the Basel Committee® and have utilized the Basel recommendations to set regulatory limits on loans and liquidity ratios. In 2002, the EU widened its capital adequacy rules for banks, insurance companies and investment wo that are part of a financial conglomerate.°! The Financial Action Task Force

The Financial Action Task Force (FATF) was established by the G-7 Summit that was held in Paris in 1989 in response to mounting concern over money laundering. The G-7 Heads of State and Government and the President of the European Commission recognized this threat and the dangers to the banking system and financial institutions and convened the task force from the G-7 Member States, the European Commission and eight other countries.” The FATF is an intergovernmental body that operates in the context of the Organization for Economic Co-operation and Development (OECD) and whose purpose is to develop and promote policies, at both national and international levels, in order to combat money laundering and terrorist financing.® It is therefore a policy-making body, which works to generate the necessary political will to bring about national legislative and regulatory reforms in these areas. The FATF monitors Member Countries’ progress in implementing necessary measures, reviews money laundering and terrorist financing techniques and countermeasures, and promotes the adoption and implementation of appropriate measures globally. In performing these activities, the FATF collaborates with other international bodies involved in combating money laundering and the financing of terrorism. The task force does not have a tightly defined constitution or an unlimited life span and thus reviews its mission every five years.® It will only continue to exist and to perform its functions after that ifthe Member Countries’ governments agree that this is necessary. Its previous mandate ended in 2004 but the representatives of the 53 Member Countries have extended the mandate in May 2004 for a further eight years.

254

Regulation through Agencies in the

EU

The task force was given the mandate of examining money laundering techniques and trends, reviewing action which had already been taken at a national or international level, and setting out the measures that still need to be taken to combat money laundering. In April 1990, less than one year after its creation, the FATF issued a report containing a set of Forty Recommendations,” which provide a comprehensive plan of action needed to combat money laundering. The Forty Recommendations set out framework for anti-money laundering efforts and are designed for universal application. They provide a complete set of countermeasures against money laundering covering the criminal justice system and law enforcement, the financial system and its regulation, and international cooperation. The Forty Recommendations have been recognized, endorsed or adopted by many international bodies. The recommendations are neither complex nor difficult, nor do they compromise the freedom to engage in legitimate transactions or threaten economic development. They set out the principles for action and allow countries a measure of flexibility in implementing these principles according to their particular circumstances and constitutional frameworks. While the Forty Recommendations do not constitute a binding international convention, many countries in the world have made a political commitment to combat money laundering by implementing them.” Initially developed in 1990, the recommendations were revised for the first time in 1996 to take into account changes in money laundering trends and to anticipate potential future threats. More recently, the FATF has completed a thorough review and update of the Forty Recommendations (2003). The FATF has also issued various interpretative notes, which are designed to clarify the application of specific recommendations and to provide additional guidance. After the 11 September 2001 attacks in the United States, governments worldwide called for an immediate and coordinated effort to detect and prevent the misuse of the international financial system by terrorists. The EU Finance and Economics Ministers and the G-7 Finance Ministers suggested, for example, that such an initiative be pursued within the framework of measures already taken by the international community to combat money laundering.” At an extraordinary plenary meeting on the financing of terrorism held in Washington, DC, in October 2001, the FATF thus expanded its mission beyond money laundering to focus its energy and expertise on a worldwide effort to combat terrorist financing. The FATF issued new international standards for combating terrorist financing — the Eight Special Recommendations’! — and called on all countries to adopt and implement them. Implementing these special recommendations should deny terrorists and their supporters access to the international financial system. FATF Member Countries are strongly committed to the discipline of multilateral monitoring and peer review. The self-assessment exercise and the

Regulation and Globalization

255

mutual evaluation procedure are the primary instruments by which the FATF monitors progress made by Member Governments in implementing the FATF Recommendations. In the self-assessment exercise, every Member Country provides information on the status of its implementation of the Forty Recommendations and Eight Special Recommendations by responding each year to a standard questionnaire. This information is then compiled and analysed, and provides the basis for assessing the extent to which the recommendations have been implemented by both individual countries and the group as a whole. The second element for monitoring the implementation of the Forty Recommendations is the mutual evaluation process. Each Member Country is examined in turn by the FATF on the basis of an on-site visit conducted by a team of three or four selected experts in the legal, financial and law enforcement fields from other Member Governments. The purpose of the Visit is to draw up a report assessing the extent to which the evaluated country has moved forward in implementing an effective system to counter money laundering and to highlight areas in which further progress may still be required. The mutual evaluation process is enhanced by the FATF’s policy for dealing with Member Countries not complying with the Forty Recommendations. The measures specified in this policy represent a graduated approach aimed at reinforcing peer pressure on Member Countries to take action to tighten their anti-money laundering systems. The policy starts by requiring the country to deliver a progress report at plenary meetings. Further steps include a letter from the FATF President or to the non-complying Member Country by a high-level delegation. The FATF can also apply Recommendation 21, which results in issuing a statement calling on financial institutions to give special attention to business relations and transactions with persons, companies and financial institutions domiciled in the non-complying country. Then, as a final measure, the FATF membership of the country in question can be suspended. Summaries of each mutual evaluation are contained in the FATF Annual Report for the year in which the evaluation took place. The EU joined the fight against money laundering in 1991 with the adoption of the Directive on the Prevention of the Use of the Financial System for the Purpose of Money Laundering,’* which implements the Forty Recommendations of the FATF.” The stated purpose of the 1991 Directive is to defend the integrity, safety and soundness of the EU’s financial system.”* It harmonizes anti-money laundering legislation in EU Member States and requires the Member States to penalize infringements of the measures taken.” In so doing, the Directive overlaps to a considerable extent with other international acts,”° including the Basel Statement of Principle” and the Council of Europe Convention of 1990 on Laundering, Search, Seizure and Confiscation of Proceeds from Crime.”

256

Regulation through Agencies in the

EU

According to the Directive, Member States’ financial and credit institutions are obliged to identify customers,” keep records of transactions,” and report suspicious transactions without breaching customer disclosure restrictions.*! Moreover, the Directive requires financial and credit institutions to cooperate with law enforcement agencies and authorities,” and to establish internal control and communication procedures for monitoring suspicious transactions.** The Directive also forbids a financial institution from further participating in a transaction if it suspects that money laundering is taking place. The Directive adopts a suspicion-based reporting system. However, certain loopholes continued to exist in the 1991 Directive. These have been the subject of much debate, and many of them have been rectified in the recently adopted 2001 Directive.™ In the latter Directive, the Parliament and Council noted that there has been a trend in recent years towards a much broader definition of money laundering, based on a more extensive list of predicate offences, as reflected in the Forty Recommendations of the FATF.* Therefore, the list of predicate offences listed in the amended Directive is extended to include proceeds from participation in activities connected with organized crime, fraud as defined in the Convention on the protection of the European Communities’ financial interests, and corruption.** Apart from the issue of money laundering, the EU required in Accession Partnerships that candidate Member States had to ensure compliance with the recommendations of the FATF, as part of the priorities and intermediate objectives in the field of the free movement of capital, in order to demonstrate their abilities to assume the legal obligations attached to membership of the EU.*’ The United Nations Economic Commission for Europe (ECE)

The ECE was set up in 1947 by ECOSOC, as one of the five regional commissions of the United Nations. Its primary goal is to encourage greater economic cooperation among its Member States and to that effect it focuses on economic analysis, environment and human settlements, statistics, sustainable

energy, trade, industry and enterprise development, timber and transport. It analyses policies, develops conventions, regulations and standards and provides technical assistance. The ECE has 55 Member States mainly from Europe, but the Russian Federation,

Canada

and the United

States,8* and around 70

professional organizations and non-governmental organizations also take part in its activities.*” The ECE is comprised of an Ad Hoc Group of Experts and seven principal subsidiary bodies, namely the Committee on Environmental Policy, the Inland Transport Committee, the Committee for Trade, Industry and Enterprise Development, the Timber Committee, the Committee on Human Settlement,

the Committee on Sustainable Energy and the Conference of European

Regulation and Globalization

Zot

Statisticians.” These bodies contribute to the development of conventions, regulations and standards, which is an important means of achieving the goals of the ECE. It has drawn up more than 30 conventions and protocols and over 250 regulations and standards. These instruments eliminate obstacles and simplify procedures for trade throughout the region and with the rest of the world. A number of them also aim at improving the environment. In this way, the ECE provides consumer guarantees of safety and quality, helps to protect the environment, and facilitates trade. It also strengthens the integration of Member States at the regional and international level.”! In 2003, for example, in the field of technical cooperation, the ECE carried out more than 30 projects in environment, energy, transport, trade, statistics, entrepreneurship and property markets. The ECE also offered advisory services to a large number of transition and emerging market economies on issues such as trade facilitation, investment promotion, energy, environment, transport, youth and women’s entrepreneurship, and statistics. The main workload of the ECE in the field of technical cooperation dealt with the exchange and transfer of expertise, mainly in the form of workshops, seminars or forums, but also in the form of technical missions or study tours. Other capacity-building activities were associated with the implementation of projects and the build-up of institutional capacities in countries in transition to help them implement the ECE conventions.” The EU incorporates most standards of the ECE in its legislation. For example, the Economic Commission for Europe Working Party on Facilitation of International Trade Procedures developed a tool to assist in the electronic transfer of cash globally, the Electronic Data Interchange for Administration, Commerce and Transport (EDIFACT), which has been accepted for integration into the EU.” Furthermore, the EU has frequently integrated the standards of the ECE.” Moreover, it has acceded to the Agreement Concerning the Adoption of Uniform Technical Prescriptions for Wheeled Vehicles, Equipment and Parts which can be fitted and/or be used on Wheeled Vehicles and the Conditions for Reciprocal Recognition of Approvals Granted on the Basis of these Prescriptions,*> and has incorporated several regulations with regard to vehicles.*° The EU’s Influence on International Standard-setting Agencies While the various international agencies unquestionably have an influence on the EU, the EU itself often tries to use these international agencies as a forum to strengthen international technical standards or to impose its point of view. In the first place, some EU agencies have been explicitly given the task of promoting technical standards on the international level within their field of competence. The newly established European Aviation Safety Agency, for

258

Regulation through Agencies in the

EU

example, will seek to strengthen worldwide support for the EU’s views on civil aviation safety standards.” Moreover, this agency must assist the EU and its Member States in their relations with third countries in accordance with the relevant rules of Community law;”* it will provide its technical assistance in contacts and negotiations with the aeronautical authorities of third countries and international organizations competent in civil aviation safety and environmental protection. The agency may also cooperate with the aeronautical authorities of third countries and the international organizations expert in matters covered in the framework of working arrangements concluded with those bodies, in accordance with the relevant provisions of the EC Treaty.” In addition, the European Agency for the Evaluation of Medicinal Products 100 is a good example of how European agencies assist the EU and its Member States on the international level. Indeed, one of its tasks is to provide technical and scientific support for steps to improve cooperation between the EU, its Member States, international organizations and third countries on scientific and technical issues relating to the evaluation of medicinal products.’®' This involves close cooperation of the agency with international partners, thus reinforcing the EU contribution to global harmonization on this issue. Another way the EU tries to impose its view on certain issues is by using international agencies and cooperating with them. For example, the EU sparred with the United States in the continuing controversy over genetically modified organisms. This has encompassed such issues as safety, animal welfare, long-term effects, lack of adequate standards, social effects and threats to biodiversity and the environment. All of these issues have serious

international trade implications since countries can use them (and have done so) to ban products.'° The United States, on the one hand, is committed to

advancing the interests of the biotechnology industry; it has applied pressure on foreign governments and continues to treat genetically modified food labelling as an unjustified barrier to trade.’ The EU, on the other hand, has mounted stiff resistance in the international standard-setting committee, the

Codex Alimentarius Commission Committee on Food Labelling. As a result, this committee has been unable to form a consensus on the labelling of genetically modified food.'“ This has an impact on other countries. For example, many Latin American countries support biotechnology, but in practice they follow the requirements of the EU on this issue.'°> Although the Codex Alimentarius Commission recently adopted standards to assess the risk of genetically modified organisms,'°° it has not completed its work on the specific issue of labelling products containing these organisms.'”’ At first sight the EU regulations seem generally to be consistent with the adopted standards and the SPS.'* In any event, the dispute between the EU and the United States, Canada and Argentina continues to exist and in the end, it is up to the WTO to rule on the entire controversy.

Regulation and Globalization

259

The EU is also cooperating with other international agencies. For example, the EU participates in all the meetings of the ECE. Cooperation is regarded as profitable to both institutions: the ECE benefits from the work of the EU and, at the same time, brings to the EU its long experience of East European countries and its wider geographical coverage. Regular contacts on specific programmes or projects take place between the different divisions of the ECE and the corresponding directorates-general of the European Commission.'” The EU sometimes applies for full membership of an international agency. It is already a Member of the WTO and FAO and has recently become a Member of the Codex Alimentarius Commission.'!° It is also set to become a party to the International Plant Protection Convention,'!! alongside individual EU Member States, when the amended version of the Convention of 1997 enters into force. This would allow the EU to exercise its own competence and play a more significant role in the preparation, negotiation and adoption of standards. Accession should also help to reinforce coherence between these standards and other relevant international obligations of the EU. An important aspect will be the division of competences. With regard to the Codex Alimentarius Commission, the EU will be exclusively competent for areas which it has harmonized fully or to a large extent by legislation and for issues dealing with international trade.''? Member States have competence on organizational and procedural matters of the Codex Alimentarius Commission, while both the EU and its Member States have competence in issues which are only harmonized to a limited extent.''’ As a result, the European Commission will vote and speak on issues within the exclusive competence of the EU, while Member States may speak in support. A common position is taken in areas where both Member

States and the EU are competent.''* The accession of the EU to international organizations is an applauded evolution. It will allow the EU to act more coherently and to take firmer positions on the international level. Furthermore, it will strengthen the image of the EU as a global actor of significance.

IS THERE A NEED FOR INTERNATIONAL STANDARDSETTING AGENCIES IN A GLOBALIZED WORLD? The previous section demonstrated the importance and impact of international standard-setting agencies. This section will examine

the need, place, role,

functions and desired image of these agencies in a globalized world. Emphasis will be placed on agencies operating in the economic sphere, since globalization is most highly developed in that area. It will be argued that international agencies have a function, namely to liberalize international trade, and that they have

certain advantages over their main alternative — mutual recognition agreements. Their main problem, however, lies in the issue of their legitimacy.

260

Regulation through Agencies in the

EU

The Need for International Standard-setting Agencies for the Liberalization of Trade

Trade is becoming increasingly globalized. Barriers to trade, however, continue to exist. Some barriers try to shield national markets and protect domestic industries, but others have more noble goals — e.g. to protect consumers from dangerous goods and substances, to promote public health, to safeguard intellectual property or to impede unfair competition. A central goal and challenge is to reconcile the objectives of protective social regulation, on the one hand, and free competition and trade, facilitated through open trade policies, on the other.''® International standard-setting agencies play an important role in this matter and are even more advantageous than their main alternative: mutual recognition agreements. Mutual recognition agreements Mutual recognition can be defined as reciprocal agreement among jurisdictions to accept the others’ regulatory standards that govern the creation

and conduct of companies and businesses.''® The effect of mutual recognition is to allow a company to establish and offer its services in host jurisdictions without the necessity of having to comply with the individual regulatory controls of each jurisdiction. Rather, the company is subject to the regulatory control of its home State. As a result, mutual recognition agreements between States have the effect of liberalizing trade by removing certain barriers. The EU is quite familiar with this kind of means of removing trade barriers. Internally, mutual recognition of each Member State’s standards is a key ingredient of the free movement of goods, enshrined in Articles 28-31 EC Treaty. The European Court of Justice was quick to emphasize this principle in its famous Cassis de Dijon Case.''’? Consequently, Member States have to recognize each other’s standards as equivalent and can only narrowly limit trade. Furthermore, the EU has opted for mutual recognition agreements with third countries and has recently concluded mutual recognition agreements with the

United States,''* Japan'!? and Canada.'° One of the major advantages of mutual recognition is that it replaces centralized decision-making with decentralized decision-making, thus indirectly giving expression to the principle of subsidiarity.'*! Moreover, it promotes competition among different regulatory jurisdictions, which may lead to more efficient regulation:'” centralized decision-making tends to enact one set of standards applicable to a variety of situations, leading to inefficiency. The resulting competition also leads to the development of a dynamic regulatory environment that encourages the creation of business, without the restraints of unnecessary regulation’? and hence maximizes the possibilities for the creation of State wealth. It also contributes to innovation, which is a byproduct of

Regulation and Globalization

261

competition. Furthermore, mutual recognition agreements allow domestic producers to continue producing goods under their domestic standards, without the costs of changing production procedures. Mutual recognition agreements not only liberalize international trade; they also encourage these effects, thanks to their decentralized character. However, mutual recognition also has certain disadvantages. Since mutual

recognition agreements by their decentralized nature encourage competition between regulatory jurisdictions, this often leads to a general lowering of standards.'* The potentially negative effect of simple mutual recognition is that it encourages a gradual erosion of prudential standards, with States slowly responding to initiatives adopted in other States and so on, creating a spiral of decline.'** Furthermore, some standards may seem undesirable in the short term but have advantageous effects in the long term. It is questionable whether market participants and regulators are truly able to take a long-term view. With investors demanding faster returns from their investments and globalization requiring faster decision-making, responsible regulatory positions from a longterm point of view tend to be the first victims of this race to the bottom. Moreover, this race to the bottom risks endangering consumers,'*® and sometimes the economic health of States. Another disadvantage of mutual recognition agreements is that an exporter cannot always know whether the technical regulations in his home State are in fact equivalent to those in another country. '?’ Rule-making by international standard-setting agencies International standard-setting agencies generally design international standards

to establish a level playing field in a certain context.'** Unlike mutual recognition agreements, these standards frequently harmonize the divergent regulations of different States, thus enhancing and liberalizing international trade. Industries and businesses no longer need different production lines, since it is sufficient to respect a single standard. Although such standards are often only ‘soft law’, and their acceptance is therefore voluntary, some have evolved into de facto hard law by incorporation into a treaty'”’, such as, for example, the standards of the Codex Alimentarius Commission (supra). Moreover, despite their voluntary nature, they are often respected and incorporated into domestic

legislation on the strength of the fact that they have been devised by an international agency staffed by experts in the field (for example: the Capital Adequacy Standards of the Basel Committee). Furthermore, standards of international agencies bring to a halt the abovementioned race to the bottom in domestic regulation.'*° A standard lower than the international standard would only result in the exclusion of products from the foreign (and domestic) market. By adopting standards, international agencies also try to create greater transparency in, for example, corporate and

262

Regulation through Agencies in the

EU

financial institutions; the elimination of conflicts of interest and corruption;

honest reporting and financial statements, and even-handed enforcement of legal rules and standards.'! In general, the rule-making of international agencies is aimed at the creation of a legal framework by which behaviour can be measured. This has many advantages. First, it enables investors to make better cross-country comparisons and more informed decisions. Second, corporate decision-makers and regulators are provided with the right incentives to adopt norms on certain issues and to promote transparency, better ethics and discipline in the conduct of their affairs.'*? Finally, the existence of harmonized international standards facilitates regulatory control and law enforcement. The absence of or the lack of clarity in applicable standards is likely to give regulators and law enforcement officials too much latitude to exercise considerable discretion in the exercise of their function.'*? In conclusion, rule-making by international agencies has many advantages. However, an important drawback is the ‘soft law’ nature of such rules, which means that the standards are adopted on a voluntary basis. Normally this is not a problem, because of the legitimacy conferred by the expert status of the members of most international agencies. Unfortunately, or luckily, the legitimacy of these institutions is open to question. The Problems with International Standard-setting Agencies

Legitimacy International standard-setting agencies are not always highly regarded and are sometimes even considered as a threat to national sovereignty. Indeed, their

legitimacy is often questioned. This is certainly the case where specialized agencies with an informal structure and a limited membership are concerned (e.g. the Basel Committee). A major legitimacy problem concerns which States constitute the membership of some international standard-setting agencies. In general, international standard-setting agencies with a universal membership of all States are often quite well legitimized.'** However, not all of them have such a broad membership. Much of the work on the development of international standards has been undertaken and accomplished under the aegis of agencies consisting primarily of a small group of Western industrial countries.'*° The Basel Committee, for example, is now only composed of the representatives of 13 central banks that regulate the world’s largest banking markets, none of them coming from a developing country. Rule-making in international agencies is often the task not of Member States but of experts, given that the standards they develop require a high level of technical expertise. Even here questions of legitimacy may arise. The Codex Alimentarius Commission has been criticized for the role of food industry giants in the adoption of standards. These industries regularly attend

Regulation and Globalization

263

Codex Alimentarius meetings as part of official delegations and often have a central role in developing Members’ positions taken at such meetings, including providing scientific data and scientists.'*° Since standards adopted by the Codex Alimentarius Commission have de facto binding effects on WTO Members through the SPS and the TBT (supra), the influence of companies on the standard-setting process is at least debatable. Fortunately, developing countries are given assistance to attend the meetings of the Codex Alimentarius Commission by the FAO/WHO

Codex Trust Fund, so that their voice, too, is

being heard. Since March 2004, this fund has supported the participation of nine countries in Codex technical committee meetings and more than 30 countries have been funded to attend the 2004 session of the Codex Commission.!°” The argument that legitimacy does not constitute a problem since international standard-setting agencies only enact soft law, which countries voluntarily adopt, is partially false. First, as has been seen above, some rules of soft law have evolved into hard law. Second, even if the standards of-the

international agency are only soft law, they still can have a binding effect. The standards of the Basel Committee have been incorporated into the laws of the Members of the G-10 and other States.'°* Finally, legitimacy is sometimes questioned because of a lack of transparency. The Basel Committee in particular has lacked transparency. It is by no means a public forum: it operates secretly and has sought throughout its existence to maintain an unpublicized existence and a low profile.'”? This is illustrated by the Basel Committee’s 1975 founding agreement, which was not made public until over five years after the central bankers adopted it.'*° Moreover, the Basel Committee works informally and operates by consensus. It was not constituted by treaty, has no legal personality, no headquarters and no stationery.'*'! But more institutionalized agencies also have a transparency problem: the Codex Alimentarius Commission holds meetings that are closed to the general public, and standards are not made public until well into the process.|” Disordered conglomerate of international standard-setting agencies Apart from legitimacy, a second general problem of international standardsetting agencies is the lack of a structured and organized system. This situation grew historically from case to case and from specific need to specific need, mostly a special set of different economic, political and social factors. The International Telegraph Union, the predecessor of the International Telecommunications Union, for example, was established as far back as 1865

to develop rules to amend the first International Telegraph Convention. Because international telegraph networks had hitherto been regulated on a bilateral basis, the aim of this convention and the Union was to regulate the fast telegraph networks, to standardize equipment to facilitate international

264

Regulation through Agencies in the

EU

interconnection, to adopt uniform operating instructions which would apply to all countries, and to lay down common international tariff and accounting tules.'# This still applies today to the modern International Telecommunications Union, and the fundamental objectives of the organization have remained basically unchanged throughout the development of telecommunications technology.'* Furthermore, the various predecessors of the World Health Organization were established inter alia because of the fear that the disease epidemic would be spread by international trade and international travel.'* While the creation of an agency because of a combination of certain factors has some advantages, it also leads to overlaps between competences, a disorganized spectrum of international entities, and confusion, which is certainly not beneficial to international trade. A role can and should be played by the United Nations. Unfortunately, the United Nations’ institutional structure has proven to be unable to coordinate the creation and operation of the various international regulatory and specialized bodies in its own abdomen.'* Besides the disordered situation at the global level, there is moreover no uniform standard acceptance procedure. Such procedures, of course, relate to the specific field an international standard-setting agency is operating in. In a minority of cases, the enacted standards are binding on the members of the agency, unless the members indicate that they do not want to be bound within a certain time frame. An example of this procedure of standard setting is the International Civil Aviation Organization (ICAO), and the ICAO Council. The Council, composed of 33 States of primary importance in air transport, or which make the largest contribution to the provision of facilities for air navigation, or whose designation will ensure that all major areas of the world are represented,'*’ adopts recommendations and standards by a two-third’s vote and submits them to the Member States.'** These regulations then become new annexes or amended annexes to the ICAO treaty, unless a majority of the Member States indicate their disapproval within three months.'*? This scenario is quite unlikely to materialize, since the Council discusses new standards with Member States in the preparatory stage.'°° After four months the standards become effective unless States have opted out by notifying the differences between their existing rules and the newly created standards. '*! Consequently, a small group of Member States is able to legislate new standards, which become effective for every Member State unless a State opts out. In most cases, however, standards are only recommendatory.

While the

process for the adoption of standards in most agencies is quite similar, the status of the adopted standards is not. Recommended standards have the advantage of more rapid adoption since they do not bind the Member States of a specific agency unless they indicate their intention to accept them. In the case of the ICAO, however, direct informal contact is constantly being made to ensure

Regulation and Globalization

265

that a standard will be adopted, hence the longer duration of the period before adoption. On the other hand, it becomes (psychologically) more problematic to opt out and to disagree with the decision of the Council of the ICAO, given its composition. In conclusion, the disorganized conglomerate of international standard-setting agencies and standards that are only recommendations hinder a transparent regime of standards to guide expectations and decision-making in international trade for companies, consumer organizations and other players. A Global International Standard-setting Agency: a Possible Solution? An appropriate solution to these problems of legitimacy and the chaotic situation described above is necessary. In the opinion of the authors, it would be worth while to consider the idea of setting up a global international standardsetting agency under the auspices of the UN, which could be named the Internationai Standard-setting Agency (ISSA). The main objective of this agency would be the regulation of trade in goods, which is perhaps the easiest sector of international trade in which to develop standards, and services,

which is probably more politically sensitive. However, with the growing importance of trade in services and the phenomenon of delocalization of services, it is imperative to include trade in services as well, despite the political sensitivity. ‘Services’ is best defined as it is in GATS, as any service in any sector except services supplied in the exercise of governmental authority, i.e. supplied neither on a commercial basis nor in competition with one or more service suppliers.'** As a result, the scope of GATS and the ISSA would be identical, which would promote conformity between the two regimes and lessen political susceptibility. The ‘global’ character of ISSA indicates not only that the membership should be universal, with all States participating, but also that all relevant players should be present, including relevant non-State actors. One could think of the proposed agency as combining elements that already exist in other agencies. The structure of the ISSA could reflect the general composition of most specialized agencies of the United Nations, namely a plenary representative body, a restrictive representative and executive body (often called ‘Council’), and a Secretariat. The plenary organ, it is proposed, should consist of three groups: States, industry representatives and NGOs, with the possibility that regional trade organizations like the EU can become Members. Although this is not commonly found throughout UN specialized agencies, a similar structure exists in the International Labour Organization (ILO), where the International Labour Conference is composed of representatives from Member States, employers and workers’ organizations.'** This universal membership, and the involvement of other relevant players such as companies, NGOs and consumer

266

Regulation through Agencies in the

EU

organizations, will exclude the main criticisms levelled at the legitimacy of international agencies, namely that they are dominated by industrial Western States and industry giants.'** The main problems lie in the selection of these non-State actors. One possibility is that representatives of industry organizations and NGOs are selected on a national level, like the nomination procedure in the ILO.'5 This has the advantage that specific local concerns can be taken into account on the international level and that adopted standards reflect these concerns, hence enhancing their legitimacy. On the other hand, some problems are global and risk being ignored. Moreover, national NGOs and industry representatives do not always have the expertise and funds to tackle and discuss global issues. Preferably a mixture of national, regional and international non-State actors should be achieved. The representative plenary organ should preferably meet annually, which should be sufficient to enable it to respond to rapid developments in the global trade area and to allow standards to be drafted. It would also decide on the budget, general policies, elect the executive body and adopt the proposed standards'** by a single majority in each group and with an overall majority in the plenary body. The Council would be a more restrictive, but representative, body, entrusted

with the tasks of considering future policy issues (to be voted by the plenary body); facilitating implementation of the constitution of the agency and of the standards; coordinating the work of the agency and exercising financial control over the subsidiary bodies and the Secretariat.'*’ Specialized committees would be responsible for drafting the actual standards. Ideally, there should be a distinction between regional committees and thematic committees (e.g. transport, food products, trade in goods, trade in services, etc), which should be staffed

by experts in the field. It seems desirable that the experts should not be appointed by the plenary body, but should be independent. The final organ would be the Secretariat, which would take care of the day-to-day functioning of the agency, the organization of conferences, studies and administration and would act as the legal representative of the agency.'*$ The standards would only be recommendatory. This has the advantage that they would be more easily accepted and adopted by the Member States. It does not mean, however, that they would have no effect. Some States would undoubtedly incorporate them spontaneously into their national legislation because of self-interest. Other States that are important trading partners would follow. Another possibility is the incorporation technique, used in the TBT and SPS (supra).'* This technique could be used in other treaties, such as treaties adopted under the ECE or UNCITRAL.'® Slowly and spontaneously greater harmonization would occur, which would eventually benefit international trade. The last remaining problem is the relation between the proposed ISSA and other international agencies within or outside the UN. The creation of ISSA

Regulation and Globalization

267

would not mean the absorption of all other existing agencies. Some agencies would continue to exist, since their main objective is not the regulation of international trade, like the FAO or the FATF.'®! Other agencies could be integrated completely into the ISSA: the Codex Alimentarius Commission, for example, is mainly occupied with health-related aspects of trade in food, and could therefore probably become a sub-organ of the ISSA, dealing with standard-setting in that area of international trade. Other bodies could be partially integrated: e.g. the ECE and other regional commissions in the UN could transfer their standardization programmes relating to trade to the ISSA, but would continue to be active in drafting treaties, and concerning energy and environmental issues. The aim of the creation of the ISSA would therefore not be to create a super agency, but to create one single agency that adopts standards in the field of international trade in goods and services.

CONCLUSION This chapter has sought to clarify the relationship between globalization and regulation. International standard-setting agencies attempt at harmonizing rules and regulations in various domains of international cooperation. Particularly in the field of international trade and finance, they smooth international commercial transactions by harmonizing national technical regulations and by making financial markets more transparent. As a result, they have an impact on the daily life of citizens, even if they are or that impact is not directly visible. It would, however, be wrong to consider international regulatory agencies as transcendent organs imposing standards upon countries. First, they mostly only enact soft law, which countries voluntarily adopt, although their standards do sometimes become hard law and thus binding. Second, States and international organizations also have an impact on the functioning and rule-making of international standard-setting agencies. However, such agencies do sometimes have a problem of legitimacy, especially where their membership is limited, and their competences are dispersed. Inclusion of more countries, particularly developing countries, and of all stakeholders in a given issue, together with transparency, could remove most of this concern, although a delicate balancing act will remain between ensuring flexibility, high quality and expertise in rulemaking on the one hand, and broadening the participation and accessibility of the rule-making process on the other hand. We have argued that a possible solution could lie in the creation of a new ‘International Standard Setting Agency’, which would regulate international trade and operate under UN auspices. In addition to States, industry organizations and NGOs also should participate to enhance its legitimacy. This could perhaps ensure that the role of the UN in the field of international trade would increase.

268

Regulation through Agencies in the

EU

Universal State membership of this agency and the inclusion of all relevant players should lead to a counterbalance to the WTO and encourage a dynamic relationship between the UN and the WTO. Non-members to the WTO and NGOs would thereby have a forum and an impact on global trade. This would probably lead to a softening of criticisms of globalization and to a better acceptance of the way in which international trade functions. The EU, which is an important player in the WTO and the UN, could be a bridge in this matter and help to coordinate initiatives of new proposed United Nations agency and the WTO.

NOTES 1 For reasons of simplicity and clarity, this contribution uses the term “EU’ also in cases where it would be legally more correct to use ‘European Community’ or ‘EC’, such as in matters relating to the World Trade Organization, of which the EC (and not the EU) is a Member.

2 Over the last few years the EU established various new agencies: the European Food Safety Authority, established by Regulation (EC) No. 178/2002 O.J., 2002, L 031/1; the European Maritime Security Agency, established by Regulation (EC) No. 1406/2002 O.J., 2002, L 208/1; the European Aviation Safety Agency, established by Regulation (EC) No. 1592/2002 O.J., 2002, L 240/1; the European Network and Information Security Agency, established by Regulation (EC) No. 460/ 2004 O.J., 2004, L 77/1; and the European Agency for the Management of Operational Cooperation at the External Borders of the Member States of the European Union O.J., 2004, L 349/1. 3 Jt should be stressed that we do not aim at being exhaustive nor even representative of the interactions between international and EU standard-setting processes and agencies. For instance, within the UN family there exists a great variety of organs and specialized agencies (from the International Labour Organization to the International Postal Union to the World Bank) whose activities, through standard-setting and/or their policies, have regulatory implications. 4 M. Iynedjian (2002), L Accord de l’Organisation Mondiale du Commerce sur |Application des Mesures Sanitaires et Phytosanitaires, Paris, L.G.D.J., p. 60. 5 See http://www.codexalimentarius.net. 6 See Council Decision 2003/822/EC of 17 November 2003 on the accession of the European Community to the Codex Alimentarius Commission O.J., 2003, L 309/14. Membership of the EC was possible under Article 2 of the Statutes of the Codex Alimentarius Commission, as membership of the Commission is open to all Member Nations and Associate Members of FAO and WHO and the EC is a full Member of FAO since 1991. 7 The procedural rules for adopting food-related standards can be found at ftp://ftp.fao.org/codex/ PM/Manual13e.pdf. 8 The Codex can be consulted in part at http://www.codexalimentarius.net/standard_list.asp. ® T.P. Stewart and D.S. Johanson (1998), ‘The SPS Agreement of the World Trade Organization and International Organizations: the Roles of the Codex Alimentarius Commission, the International Plant Protection Convention, and the International Office of Epizootics’, Syracuse Journal of International Law and Commerce, 27, 41.

10 M. Iynedjian (2002), L’Accord de l'Organisation Mondiale du Commerce sur |’Application des Mesures Sanitaires et Phytosanitaires, Paris, L.G.D.J., p. 61.

| The text of the treaty can be found at: http://www.oie.int/eng/OIE/textfond/en_arrangement_

1924.htn. 12 T.P. Stewart and D.S. Johanson, supra note 9, 49. '3 Article 4 of the Organic Statutes of the Office International de Epizooties, Annex to International

Agreement of an Office International des Epizooties in Paris at: http://www.oie.int/eng/OIE/textfond/ en_statuts_organiques.htm for the creation.

Regulation and Globalization

269

'4 Although not mentioned in the Organic Statute of the International Office of Epizootics, this was subsequently developed in its functioning. '5 See the exchange of letters between the Commission of the European Communities and the Office international des Epizooties of 23 February 2004 O.J., 2004, C 215/3, also available at http://www.oie.int/eng/oie/accords/en_accord_ce.htm. 16 'T.P. Stewart and D.S. Johanson, supra note 9, 49.

'7 The text of the Convention, which was amended for the first time by the FAO Conference in 1979, is available at: http://www. ippc.int/servlet/BinaryDownloaderServlet/13767_1952_ English.pdf? filename=/publications/13767.The_International_Plant_Protection Convention.2001-3-

26.pdf&refID=13767. '8 Article IV International Plant Protection Convention. '9 Article V International Plant Protection Convention. 0 Article VI International Plant Protection Convention. 21 Article IX International Plant Protection Convention. 22 T.P. Stewart and D.S. Johanson, supra note 9, 47.

23M. Iynedjian (2002), L Accord de l'Organisation Mondiale du Commerce sur l’Application des Mesures Sanitaires et Phytosanitaires, Paris, L.G.D.J., p. 64.

24 T.P. Stewart and D.S. Johanson, supra note 9, 47. *5 The EC approved its accession to the revised International Plant Protection hee Decision 2004/S97/EC of the Council of 19 July 2004 O.J., 2004, L 267/39. 26 See first recital Preamble SPS. 27 Article 2 SPS. 28 Article 2 SPS. 29 Article 4 SPS. 30 Article 3 SPS.

dd by

31 Annex A, 3(d) SPS. eon Ne 32owe Article 3.3 SPS.

33Go See J.E. Alvarez and S. Charnovitz (2003), ‘Triangulating the World Trade Organization’, American Journal of International Law, 50; S. Piciotto, ‘Rights, Responsibilities and Regulation of International Business’, Columbia Journal of Transnational Law, 146.

34 WTO Panel Report, 18 August 1997, EC Measures Concerning Meat and Meat Products (Hormones), WT/DS48/R/CAN; WTO Panel Report, 18 August 1997, EC Measures Concerning Meat and Meat Products (Hormones), WT/DS48/R/USA; WTO Appellate Body Report, 16 January 1998, EC Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB/R

& WT/

DS48/AB/R. 35 WTO Panel Report, 18 August 1997, EC Measures Concerning Meat and Meat Products (Hormones), WT/DS48/R/CAN; WTO Panel Report, 18 August 1997, EC Measures Concerning Meat and Meat Products (Hormones), WT/DS48/R/USA.

36 WTO Appellate Body Report, 16 January 1998, EC Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB/R & WT/DS48/AB/R. 37 Article 1, Annex 1, TBT.

38 Article 2, Annex 1, TBT. 39 Article 1.5, TBT and Article 1.4, SPS.

40 Article 2.2, TBT. 41 Article 2.4, TBT. 42 WTO Panel Report, 29 May 2002, EC Trade Description of Sardines, WT/DS231/R; WTO Appellate Body Report, 26 September 2002, EC Trade Description of Sardines, WT/DS231/AB/R; for a thorough analysis of the case and its implications see: R. Mufioz (2003), “La Communauté entre les Mains des Normes Internationales: les Consequences de la Décision “Sardines” au Sein de VO.M.C.’, Revue du Droit de l’Union Européenne, 457-484. 43. Regulation (EC) No. 2136/89, O.J. L 212/22 of July 1989. 44 For example: Regulation (EC) No. 1181/2003, O.J., 2003, L 165/3; Regulation (EC) No. 1989/

2003, O.J., 2003, L 295/157; Regulation (EC) No. 2205/2003, O.J. 2003, L 330/20. 45 For example: Directive 91/67/EEC, O.3., 1991, L 46/1; Directive 91/494/EEC, O.J. 1991,

L 268/35; Directive 2001/110/EC, O.J., 2002, L 10/47; Directive 2002/82/EC, O.J., 2003, L292/1; Directive 2003/60/EC, O.J., 2003, L 155/15; Directive 2003/95/EC, O.J., 2003, L 283/71;

270

Regulation through Agencies in the

EU

Directive 2004/46/EC, O.J., 2004, L 114/15; Directive 2004/105/EC, O.J., 2004, L 319/9. 46 Framework Agreement for Trade and Cooperation between the European Community and its Member States and the Republic of Korea, OJ 30 March 2001, L 90.

47 A.M. Slaughter (2003), ‘Global Government Networks, Global Information Agencies, Disaggregated Democracies’, Michigan Journal of International Law, 1046; D. Zaring (1998), ‘International Law by Other Means: the Twilight Existence of International Financial Regulatory Organizations’, Texas International Law Journal, 287. 48 J. Laberi (2003), ‘The Financial Stability Forum: a Step in the Right Direction ... not Far Enough’, University of Pennsylvania Journal of International Economic Law, 557. 49 D. Zaring (1998), ‘International Law by Other Means: the Twilight Existence of International Financial Regulatory Organizations’, Texas International Law Journal, 289. 50 At http://www.bis.org/bcbs/aboutbebs. htm. 51 At http://www.bis.org/bcbs/aboutbcbs. htm. 52 At http://www.bis.org/bcbs/aboutbebs. htm. 53 A.M. Slaughter (2003), ‘Global Government Networks, Global Information Agencies, Disaggregated Democracies’, Michigan Journal of International Law, 1046. 54 At http://www.bis.org/bcbs/aboutbebs. htm. 55 At http://www.bis.org/bcbs/aboutbebs. htm. 56 At http://www.bis.org/bcbs/aboutbcbs.htm. 57 At http://www.bis.org/publ/bcbs61 .htm. 58 HP. Tarbert (2001), ‘Rethinking Capital Adequacy: The “Basle” Accord and the New Framework’, Business Lawyer, 780. 5° Tbid., 781. 60 Directive 89/299/EEC, O.J., 1989, L 124/5; Directive 89/647/EEC, O.J., 1989, L 386/14; Regulation (EC) no. 2086/2004, O.J., 2004, L 363/1.

61 Directive 2002/87/EC of the European Parliament and of the Council of 16 December 2002 on the supplementary supervision of credit institutions, insurance undertakings and investment firms in a financial conglomerate and amending Council Directives 73/239/EEC, 79/267/EEC, 92/49/ EEC, 92/96/EEC, 93/6/EEC and 93/22/EEC, and Directives 98/78/EC and 2000/12/EC European Parliament and of the Council O.J., 2003, L 35/1.

of the

62 At http://www.fatf-gafi.org/AboutFATF_en.htm#What%20is; G-7 Task Force to Make Recommendations on Curbing Money Laundering, Paris Economic Summit, 16 July 1989, 1.L.M. 1989, 1299. 63 A.L. Rause (2003), ‘USA Patriot Act: Anti-Money Laundering and Terrorist Financing Legislation

in the US and European Union since September 11th’, University of Miami International and Comparative Law Review, 185-186.

64 At http://www. fatf-gafi.org/AboutFATF_en.htm#What%20is. 65 At http://www. fatf-gafi.org/AboutFATF_en.htm#What%20is. 66 At http://www. fatf-gafi.org/pdf/PR-20040514 en.pdf. 67 At http://www. fatf-gafi.org/AboutFATF_en.htm#History. 68 At http://www.fatf-gafi.org/AboutFATF_en.htm#Forty. 69 A. Walters (2003), ‘The Financial Action Task Force on Money Laundering: the World strikes back on Terrorist Financing’, Law and Business Review of the Americas, 169. 70 A.L. Rause (2003), ‘USA Patriot Act: Anti-Money Laundering and Terrorist Financing Legislation in the US and European Union since September 11th’, University of Miami International and Comparative Law Review, 186.

1 At http://www 1.oecd.org/fatf/pdf/SRecTF_en.pdf. ® Directive 91/308/EEC, O.J., 1991, L 166/77; T. Doyle (2002), ‘Cleaning Up Anti-Money

Laundering Strategies: Current FATF Tactics Needlessly Violate International Law’, Houston Journal of International Law, 292-293. 3 L.L.C. Lee (1996), ‘Combating Illicit Narcotic Traffic in Taiwan: the Proposed Money Laundering Control Act’, Tulane Journal of International and Comparative Law, 227. 74 Preamble, Directive 91/308/EEC, O.J., 1991, L 166/77.

75 Article 14 of the Directive. 76 L.L.C. Lee (1996), ‘Combating Illicit Narcotic Traffic in Taiwan: the Proposed Money Laundering Control Act’, Tulane Journal of International and Comparative Law, 227.

Regulation and Globalization

8

7 Basel Committee on Banking and Supervisory Practices, Statement of Principles: Prevention of Criminal Use of the Banking System for the Purpose of Money-Laundering, 1988 at: http:// www.bis.org/publ/bebse137.pdf. 78 Council of Europe Convention of 1990 on Laundering, Search, Seizure and Confiscation of Proceeds from Crime, ETS no. 141. 1 Article 3, Directive 91/308/EEC, O.J., 1991, L 166/77. 80 Article 4, Directive 91/308/EEC, O.J., 1991, L 166/77. 81 Article 9, Directive 91/308/EEC, O.J., 1991, L 166/77. 82 Articles 6-7, Directive 91/308/EEC, O.J., 1991, L 166/77. a > Article 11.1, Directive 91/308/EEC, O.J., 1991, L 166/77. 84 Directive 2001/97/EC, O.J., 2001, L 344/76. 85 K.A. Lacey and B.C. George (2003), ‘Crackdown on Money Laundering: a Comparative Analysis of the Feasibility and Effectiveness of Domestic and Multilateral Policies Reforms’, Northwestern Journal of International Law and Business, 317-318.

86 Article 7 of Directive 2001/97/EC, supra note 84. 87 For example: Council Decision of 28 January 2002 on the principles, priorities, intermediate objectives and conditions contained in the Accession Partnership with Cyprus, O.J., 2002, L 44/12; Council Decision of 28 January 2002 on the principles, priorities, intermediate objectives and conditions contained in the Accession Partnership with the Czech Republic, O.J., 2002, L Aa20. 88 At http://www.unece.org/oes/member_countries/member_countries.htm. 89 At http://www.unece.org/oes/eceintro.htm. 90 At http://www.unece.org/oes/about/structure.htm. 91 At http://www.unece.org/oes/about/mandate.htm. 92 At http://www.unece.org/oes/nutshell/technical_cooperation.htm. 93 C. Rowe (1998), ‘Technological Advances in Banking: A Move to A Global Economy’, JLSA Journal of International and Comparative Law, 1311. °4 For example: Regulation (EC) no. 1466/2003, O.J. L 210 of 20 August 2003; Regulation (EC) No. 86/2004, O.J. L 013 of 20 January 2004; Regulation (EC) no. 214/2004, O.J. L 036 of 7February 2004; Regulation (EC) No. 1673/2004, O.J. L 300 of 25 September 2004.

°5 Council Decision of 27 November 1997 with a view to accession by the European Community to the Agreement of the United Nations Economic Commission for Europe concerning the adoption of uniform technical prescriptions for wheeled vehicles, equipment and parts which can be fitted to and/or be used on wheeled vehicles and the conditions for reciprocal recognition of approvals granted on the basis of these prescriptions, O.J. L 346 of 17 December 1997. 96 Council Decision of 26 June 2001 on the accession of the European Community to Regulation 109 of the United Nations Economic Commission for Europe concerning the approval for the production of retreaded pneumatic tyres for commercial vehicles and their trailers, O.J. L 183 of 6 July 2001; Council Decision of 26 June 2001 on the accession of the European Community to United Nations Economic Commission for Europe Regulation no. 104 on the approval of retroreflecting markings for heavy and long vehicles and their trailers, O.J. L 183 of 6 July 2001; Directive

2001/43/EC, O.J. L 211 of 4August 2001. 97 At http://www.easa.eu.int/. 98 Article 18., Regulation (EC) no. 2592/2002, O.J. L 24 of 7September 2002.

99 Article 18.2, Regulation (EC) no. 2592/2002, O.J. L 24 of 7 September 2002. 100 Established by Council Regulation (EEC) no. 2309/93, O.J. L 214 of 24 August 1993. 91 Article 51 (f), Council Regulation (EEC) no. 2309/93, O.J. L 214 of24 August 1993. 102 LD. Thue-Vasquez (2000), ‘Genetic Engineering and Food Labelling: a Continuing Controversy’, San Joaquin Agricultural Law Review, 110. The EU even established a moratorium to the import of genetically modified organisms in 1998, which led to the United States, Canada and Argentina taking steps under the dispute settlement procedure of the WTO in 2003. The same year the EU lifted the moratorium but introduced new regulations dealing with the labelling and traceability of genetically modified organisms, laying down strict requirements, Regulation (EC) no. 1829/2003, O.J. L 268 of 18 October 2003, and Regulation (EC) O.J. L 268 of 18 October 2003; see also B. Schwartz (2004), ‘WTO and GMO’s: Analyzing the European Community’s Recent Regulations Covering the Labeling of Genetically Modified Organism’, Mich.J.1.L., 7713-811.

103 Tbid., 110.

272

Regulation through Agencies in the

EU

104 Tbid., 111. 105: Thid., 111. 106 Codex Alimentarius Commission, ‘Principles for the Risk Analysis of Foods Derived from Modern Biotechnology’, at www.codexalimentarius.net/ download/standards/10007/CXG_044e.pdf.

107 Codex Alimentarius Commission, ‘Report on the Thirty-Second Session of the Codex Committee on Food Labelling’, 79-94, at www.codexalimentarius.net/ download/report/619/al04_22e.pdf. 108 Fora more in-depth study on this issue see A.A. Ostrovsky (2004), ‘The New Codex Alimentarius Commission Standards for Foods Created with Modern Biotechnology: Implications for the EC GMO Framework’s Compliance with the SPS Agreement’, Mich. J..L., 813-843. 109 At http://www.unece.org/oes/about/cooperation.htm.

110 Council Decision of 17 November 2003 on the accession of the European Community to the Codex Alimentarius Commission, O.J. L 309 of 26 November 2003.

11 Supra note 25. 112 Annex II.1 Council Decision of 17 November 2003 on the accession of the European Community to the Codex Alimentarius Commission. 113, Annex II.2-3 Council Decision of 17 November 2003 on the accession of the European Community to the Codex Alimentarius Commission. 114 Annex III.4 Council Decision of 17 November 2003 on the accession of the European Community to the Codex Alimentarius Commission. 115 G, Shaffer (2000), ‘Reconciling Trade and Regulatory Goals: the Prospects and Limits of New Approaches to Transatlantic Governance through Mutual Recognition Agreements and Safe Harbor Agreements’, Columbia Journal of European Law, 29. 116 PB. Griffin (2001), ‘The Delaware Effect: Keeping the Tiger in Its Cage. The European Experience of Mutual Recognition in Financial Services’, Columbia Journal of European Law, seh 17 Case 102/78 Cassis de Dijon [1979] ECR 649. 118 Agreement on mutual recognition between the European Community and the United States of America, O.J. L 31 of 4February 1999. 119 Agreement on mutual recognition between the European Community and Japan, O.J. L 284 of 29 October 2001. 120 Agreement on mutual recognition between the European Community and Canada, O.J. L 280 of 16 October 1998. 121 P.B. Griffin (2001), ‘The Delaware Effect: Keeping the Tiger in Its Cage. The European Experience of Mutual Recognition in Financial Services’, Columbia Journal of European Law, 338. 122 Tbid., 338; D. Charney (1991), ‘Competition Among Jurisdictions in Formulating Corporate Law Rules: an American Perspective on the “Race to the Bottom” in the European Communities’, Harvard International Law Journal, 440.

123 PB. Griffin (2001), ‘The Delaware Effect: Keeping the Tiger in Its Cage. The European Experience of Mutual Recognition in Financial Services’, Columbia Journal of European Law, 340. 124 Tbid., 338; FH. Kung (2002), ‘The Rationalization of Regulatory Internationalization’, Law and Policy in International Business, 462. 125 P.B. Griffin (2001), ‘The Delaware Effect: Keeping the Tiger in Its Cage. The European Experience of Mutual Recognition in Financial Services’, Columbia Journal of European Law, 341. 126 Tbid., 341. 127 §. Farr (1996), Harmonisation of Technical Standards in the EC, Chichester, Wiley, 12. 128 H.V. Morais (2002), ‘The Quest for International Standards: Global Governance vs Sovereignty’, University of Kansas Law Review, 790.

129 Tbid., 780-781. 130 HV. Morais (2002), ‘The Quest for International Standards: Global Governance vs Sovereignty’, University of Kansas Law Review, 790.

131 Thid., 791. 132° Tbid., 791. MEST bid’ 794:

Regulation and Globalization

23

'34 HV. Morais (2002), ‘The Quest for International Standards: Global Governance vs. Sovereignty’, University ofKansas Law Review, 806-807. 135 Tbid., 806. 136 L.M. Wallach (2002), ‘Accountable Governance in the Era of Globalization: the WTO, NAFTA,

and International Harmonization of Standards’, University of Kansas Law Review, 836. 137. At http://www.who.int/mediacentre/notes/2004/np1 S/en. 138 HP. Tarbert (2001), ‘Rethinking Capital Adequacy: the Basle Accord and the New Framework’, Business Lawyer, 780-781. 139 D. Zaring (1998), ‘International Law by Other Means: the Twilight Existence of International Financial Regulatory Organizations’, Texas International Law Journal, 288. 140 Tbid., 288. '41_ A.-M. Slaughter (2003), ‘Global Government Networks, Global Information Agencies, and Disaggregated Democracy’, Michigan Journal of International Law, 1052. 142 L.M. Wallach (2002), ‘Accountable Governance in the Era of Globalization: the WTO, NAFTA,

and 43 144) 145

International Harmonization of Standards’, University of Kansas Law Review, 837. At http://www. itu.int/aboutitu/overview/history.html. Thid. C.H. Alexandrowicz (1973), The Law-Making Functions of the Specialised Agencies of the

United Nations,

Cremorne, Angus and Robertson, 47.

7

146 §.A. Voitovitch (1995), International Economic Organizations in the International Legal Process, Dordrecht, Martinus Nijhoff Publishers, 31.

'47 Article 50 of the Convention on International Civil Aviation at: http://www.icao.int/cgi/ goto_m.pl?/icao/en/anb/mais/index.html. 148 Tbid., Article 54.

149 Thid., Article 38. 150 C.H. Alexandrowicz (1973), The Law-Making Functions of the Specialised Agencies of the United Nations, Cremorne, Angus and Robertson, 46.

1Sl_ At http://www.icao.int/cgi/goto_m.p1?/icao/en/anb/mais/index.html. 152 Article I (3) GATS.

153 Article 3.1 ILO Constitution. 134 See section II.B.1. 155 Article 3.5 ILO Constitution. 156 Comparable to Article 8 ITU Constitution, Article 49 Convention on International Civil Aviation. !57 Comparable to Article 10, ITU Constitution, Article 54, Convention on International Civil Aviation. 158 Comparable to Article 1, ITU Constitution. 159 Article 3.4 SPS; Article 2.2 juncto Article 2 Annex 1, TBT.

160 The United Nations Commission on International Trade Law (UNCITRAL) was established by General Assembly Resolution 2205 (XXI) in 1966 with the general mandate to further the progressive harmonization and unification of the law of international trade. As a result, the General Assembly

recognized that disparities in national laws governing international trade created obstacles to the flow of trade, and it regarded the Commission as the vehicle by which the United Nations could play a more active role in reducing or removing these obstacles. For further information see: http://www.uncitral.org. 161 The FAO has the objective of enhancing nutrition, food and agriculture (Article 1, FAO Constitution); the purpose of the FATF is to combat money laundering (see Section II.B.3).

Index

access regulation 182, 184-6, 187,

190-91, 192 Deutsche Telekom case 192-5, 200

accountability 12-15, 17, 18-20, 33, 80, 125-9, 156-8 concerns about 121 EU model 231-4 European Central Bank (ECB) 15-16 meaning of 88-9 and transmission belt model 149 and transparency 129 US model 236-8 see also evaluation and monitoring accountability and delegation 88-90, 113 Commission as principal 94-6 conceptual and terminological issues 90-94 contract design 100-102, 106 Council of Ministers as principal 96—9

Ahmed Saeed case 188, 199

appeals against regulatory decisions 34-5, 198-9 appointments 14, 102-5, 106, 230-31, 235 and politicians 52, 53, 56, 127-8

revolving door practices 236 asymmetric information 8, 9, 26, 27,

234-5

Basel Committee on Banking Supervision 252-3, 262, 263 Beef Hormones case 250 best practices, adoption of 35-6, 41 Bovens, M. 88-9

Bronner case 185 cartels 26—7 Cassis de Dijon case 260 Codex Alimentarius Commission 247,

institutional checks and balances 107,

249-51, 258, 259, 262-3 Commission

109-12, 147-50 monitoring and reporting requirements 105-9 public accountability, conceptualizing

accountability 17-18, 125, 127, 128, 129

agencies, views on 71—2, 121-2 appointments 102-3, 104, 106, 127, 230-31 and competition law 164, 165-6, 171-4, 176 conflicts with NRAs, avoiding 190-91 and contract design 100 and convergence of rules 41 and coordination of regulatory policy 171-4, 176

screening and selection mechanisms 102-5, 106 administration, historical failure of 151-3 Administrative Procedure Act (US) 227 adverse selection 26 agency losses 48, 49, 99 agency, meaning of 10, 50, 122-3

agents, types of 8-10, 50 Agreement on Sanitary and Phytosanitary Measures (SPS) 249-51 Agreement on Technical Barriers to Trade (TBT) 250-51

delegation 18, 89, 90, 92-6, 219-20,

2222. Deutsche Telekom case 192-5, 200 evaluation and monitoring 35—6, 69-71,

108

274

ZI

Index framework for regulatory agencies 81-2, 110, 122-3, 225-6, 228-9, 231, 241-2 governance 32-3, 141-2, 150, 155, 161-2, 230 independence of 123 and independence of agencies 72-3, 79, 123-5 as metaregulator 70-71 as multi-purpose authority 17—18, 70 nature of 77 non-enforcement of competition rules, where sector-specific remedy applied 196-8 NRA decisions, action against 199-201 and open method of coordination (OMC) 74 and policy making 82 as principal 94-6 reform of 120 as regulator 69-70, 77-8 and regulatory impact assessments 240 role of 8, 9, 35—6, 72, 76—7, 110, 125, 128, 217 social dialogue 75-6 Committee of Independent Experts 120 Community Fisheries Control Agency 95, 96 ‘Community method’ (policy making and implementation) 70—72 Community Plant Varieties Office 72,

224, 228 competition authorities see national competition authorities (NCAs); national regulatory authorities (NRAs) and competition authorities, overlaps and inconsistencies competition law 69-70, 164-6 and coordination 168—9, 170—76, 177

decentralized approach 167-8 complexity of systems 18 control 52, 62, 125-9

financial 111-12, 124—5, 126, 232-3, 237 principal-agent theory 11, 48, 49 US model 226-7, 237-8

convergence and divergence (regulatory), model of 3641 cooperation 9 NCAs and NRAs 191-2

coordination of regulatory policy 164-5, 176-7 horizontal and vertical coordination 167-9, 172-4 methods of 170-76 need for 166-70 proliferation of authorities, factors

leading to 165—6 transnational regulatory networks 74 transversal coordination 169—70, 174-6 coregulation 75—6 cost-benefit analysis 5, 238 Council 92, 217 appointments 102, 104—5, 106 contract design 100, 101-2, 106

and delegation 219-20 and Europol 101—2, 104-5, 108

as principal 96—9 transparency of 17 Court of Auditors 107, 111, 232-3 Court of First Instance (CFI) 17, 149,

159-60, 200, 232 courts, national 164-5, 166

credibility of commitment 9, 10 Curtin, D. 130 Curtis v. Trinko (US Supreme Court) 195-6, 197

Dahl, R. 19 decisions challenging 34—5, 198-201, 231-2 ‘deliberative’ decision making 157-8, 160-61 explanation of 14, 16 overruling 52, 59 review of 15 delegation 5—6, 7—10, 18 chains of 90—99 consequences of 10—17 criticisms of 121 and decision making process 160-61 definition of 91—2 and democracy 19-20 EU and US models 219-22, 241 Meroni doctrine 69 non-delegation 147—9, 150, 153, 159,

220 see also accountability and delegation ‘deliberative’ decision making 157-8, 160-61

276

Index

demerit goods 27 democratic deficit 89 democratic principles/theory 13-14, 19-20 Deutsche Telekom case 192-5, 200 Dewey, J. 19 discretion 93-4 IRAs 58-9 member states 31-3, 41 dismissal of regulators 52, 53, 56, 127 EC Trade Description of Sardines case 251 Eisner, M.A. 68 enforcement 34, 69-71, 73, 78, 165—6, 170 private 76, 166 essential facilities doctrine 182, 186

EU Fraud Office (OLAF) 73, 112 Eurojust 98 European agencies (EAs), lessons from US experience 215-16, 241-2 accountability 231-4, 236-8

categorization of EAs 223-9, 241-2 Commission framework for regulatory agencies 225-6 common features of EAs 223 delegation of powers 219-22 formation of EAs 218-19 governance 230-41 homogeneity vs. diversity 222-9 improving EU model 221-2, 228-9,

239-41 independence 230-31, 235-6

overview of models 216-17 stakeholder participation 234-5, 239 transparency 234, 238-9 European Agency for Reconstruction 224, 228 European Agency for Safety and Health at Work 224, 228

European Agency for the Evaluation of Medicinal Products (EMEA) 72, 106-7, 112, 113, 130-31, 224, 228, 234, 258 European Agency for the Management of Operational Cooperation at the External Borders 98—9, 103, 104, 105-7 European Aviation Safety Agency 103, 105, 106—7, 113, 224, 225, 228, 257-8

European Central Bank (ECB) 6, 14, 15-16, 73, 221 European Centre for the Development of Vocational Training 224, 228 European Chemicals Agency (ECA) 224 European Council 90, 102 European Court of Justice (ECJ) 16-17, 107, 109-11 appealing/challenging NRA decisions 198-9, 200, 231-2 and decision making process 160 and delegation 18, 219 and harmonization of laws 168 institutional balance 159-60, 219 and international standards 251 judicial review 232 negative integration and mutual recognition 75 NRA’s duty to observe competition tules 188-9 European Defence Agency (EDA) 105, 106-7, 112 European Environment Agency 102, 103-4, 106-7, 111, 146, 224, 228 European Food Safety Authority (EFSA) 72, 103, 106-7, 124, 130, 131-2, 143-7, 155-6, 157, 158, 224, 228 European Foundation for the Improvement of Living and Working Conditions 224, 228

European Maritime Safety Agency 224, 228 European Monitoring Centre for Drugs and Drug Addiction 109, 224, 228 European Monitoring Centre on Racism and Xenophobia (EUMC) 103, 224, 228 European Ombudsman 74, 112, 129 European Parliament 91, 106, 217

and accountability 126-8 and agency budgets 111-12, 232, 237 and appointments 102, 103-4, 127 and contract design 100, 101-2

and Europol 101—2, 104-5, 108, 112 reporting requirements 105, 108, 233 European Railway Agency 224-5 European Regulators Group (ERG) 35-6, 173 European Securities Regulators Committee (RESCO) 74

204

Index European Training Foundation 224, 228 Europol 98, 100, 101-2, 104-5, 106-7, 108, 110-11, 112 Eurostat 72-3 evaluation and monitoring 34—6, 80, 106, 108-9 and accountability 33 Commission 35-6, 69-71, 108 Financial Action Task Force (FATF)

254-5 reporting requirements 105, 107, 108, 233, 237 Everson, M. 18, 79

executive deliberation 158 expectations of IRAs 52 externalities 4-5, 25, 37, 78

federalism 217 fiduciary principles 12-14 Financial Action Task Force (FATF) 253-6 financial control 111-12, 124-5, 126,

232-3, 237 Fisher, E. 89 Food and Veterinary Office (FVO) 73 France average tenure of IRA members 58 background of IRA members 56 government policy and IRAs 59, 60 overview of IRAs 50-51 party politics 54 resignations of IRA members 54 resources of IRAs 57 free riding 26 functionalist approach, assumptions of 5-6 GB-Inno-BM/ATAB case 188 General Secretariat of the Council 96-8 genetically modified organisms 258 Geradin, D. 78 Germany average tenure of IRA members 58 background of IRA members 60 overturning IRA decisions 59 overview of IRAs 50-51 party politics 54 resignations of IRA members 54 resources of IRAs 57 globalization 12

see also standard-setting agencies, international governance 29, 31-3, 95, 141-3, 161-2

core principles of 230 EU and US models 230-41 limited 157 political administration 153-61 and public interest 143-54 see also regulatory governance, regimes approach Hamilton, A. 18-19 harmonization 168, 171

regulatory convergence and divergence, model of 36-41 see also standard-setting agencies, international HFC Bank pic/British Gas Trading Ltd’ case 190

ideological context of regulation 4—5 income/wealth distribution 27 independence 11, 29-30, 31, 33, 79-80, 123-5 and accountability 12-15, 156, 231 from elected politicians 48—9, 58-9

EU model 230-31 expectations of 52 within institutional balance 155-6 and public interest 143-4, 153 US model 235-6 independent regulatory agencies (IRAs) advantages of 121 average tenure of senior members 58 European, concept of 122-3 expectations of 52 expertise 56-8 further creation of 122 membership of 52-6, 59-60, 124 normative case for 76-81 overview of, selected countries 50-51 resources of 57 role of 120-21 types of 122-3 information 61, 146 access to documents 130-31, 238 asymmetric 8, 9, 26, 27, 234-5 incomplete 26 institutional models, EU regulation 69-76

278

Index

International Civil Aviation Organization (ICAO) 264-5 International Labour Organization (ILO) 265, 266 International Office of Epizootics 247-8, 249-51 International Plant Protection Convention 248-51, 259 international standard-setting agencies see standard-setting agencies,

international International Telegraph Union 263-4 investment 30 Italian GSM fees case 199 Italian Matches case 188-9 Italy average tenure of IRA members 58 background of IRA members 60 government policy and IRAs 59, 61 overview of IRAs 50-51 party politics 53, 54, 56 resignations of IRA members 54 resources of [RAs 57 ITTPromedia/Belgacom decision 183 judicial accountability 23 1—2 judicial review 33, 227-8, 231-2, 236-7,

241

Meroni doctrine 69, 95-6, 147, 148, 159,

219, 221, 222, 229, 241 Mill, J.S. 19 money laundering 253, 254-6 monitoring see evaluation and monitoring monopoly 26—7, 28 moral hazard 26, 27 Moravesik, A. 7, 89

mutual recognition 75, 78, 260-61 national competition authorities (NCAs) 164-5, 166, 186-7, 201 advantages of 167—8 coordination of 168-71, 172-6 see also national regulatory authorities (NRAs) and competition authorities, overlaps and inconsistencies national regulatory authorities (NRAs) 164-5, 166 advantages of 167-8 coordination of 168-71, 173-6 national regulatory authorities (NRAs) and competition authorities, overlaps and inconsistencies 180-81, 201-2

challenging NRA decisions 198-201 Commission, conflict avoidance 190-91 concurrent jurisdiction, mechanisms

learning-by-doing 167 legitimacy 11-12, 13, 141-2, 150, 158, 161-2 international standard-setting agencies 262-3 procedural, limits of 17—20 liberalized markets see state, role in liberalized markets

Majone, G. 10, 12-13, 18, 77, 79 mandates, specification of 14 Marathon case 184 market failure 25-8 markets

functioning of 25 incomplete 26 see also state, role in liberalized markets Max.Mobil case 200 McNamara, K. 15

merit goods 27

for limiting risks of 187—92 cumulative jurisdiction 192—201 Deutsche Telekom case 192-5, 200

duplication of proceedings, mechanisms to avoid 191-2 increasing overlap between 182-7 non-enforcement of competition rules, where sector-specific remedy applied 195-8 NRA’s duty to observe competition tules 188-9 potential for inconsistencies and jurisdictional confusion 185—7 sector-specific regulation and competition rules 182-3 specific mechanisms for conflict avoidance 189 national variations in approach to regulation and agencies 6-7, 35-41 nationalization 24 negative integration 75

219

Index Network and Information Security Agency 224 network sharing 183, 184, 186, 187 Deutsche Telekom case 192-5, 200 new public law, evolution of 153-4 Nicolaides, P. 30 nomination of IRA members 52, 53, 56 O2/T-Mobile case 196, 197 Office for Harmonization in the Internal Market (OHIM) 72, 110, 224, 228

Office of Management and Budget (OMB, USA) 238 open method of coordination (OMC) 74

party politics 53-6, 127-8, 230, 235-6 Peterson, J. 17 Petit, N. 78 policy frameworks, EU, gaps in 31-3 political administration 153-61 politicians (elected), and independent regulatory agencies (IRAs) 47-8,

61-2, 79-80, 126-8, 230 expertise 56-8 informal relationships and wider networks 59-61 institutional frameworks and initial expectations 50-52 membership of IRAs 52-6 principal-agent models 48-9 resources and use of discretion 52-61 USA 230, 235-6, 237-8 politics of regulation 3-4 delegation, consequences of 10-17 functionalist approach, assumptions or 5-6 general trends and national variations 6-7 ideological context 4-5 procedural legitimacy, limits of 17—20 supranational agencies 7—10 Pollack, M. 8 Presidency of the Council 96—7 prices 28-9 principal-agent approach/theory 5—6, 8-10 agency losses 48, 49, 99 Commission as principal 94-6 contract design 100-102, 106 control mechanisms 11, 48, 49

Council of Ministers as principal 96-9 criticisms of 49 definitions 92 elected politicians and IRAs 48-9 evaluation and monitoring 34, 105—9 institutional checks and balances 107, 109-12 screening and selection mechanisms 102-5, 106 privatization 4, 24 proportionality 160 public goods 25-6 public interest, and governance 143-54, 156-7, 160-61 regulatory capture 37, 239 regulatory governance, regimes approach 67-9 future 81—2 institutional models 69-76 normative case for agencies 76-81 regulatory impact assessments 238, 240 rent-seeking 28 reporting requirements 105, 107, 108,

2335237) resignations of IRA members 53, 54, 55,

56 revolving door practices 236, 240 rules common 30 and independence 30 regulatory convergence and divergence, model of 36-41 substantive 28—9, 31 Sapir report 67, 77-8 Scharpf, F. 11-12 Scott, C. 129

self-regulation 75—6 separation of powers principle 148-9, 152220 Shackleton, M. 17

SNEPLD decision 183 social dialogue 75—6 social policy objectives 25 Spanish GSM fees case 199 standard-setting agencies, international 246, 267-8 Basel Committee on Banking Supervision 252-3, 262, 263

280

Index

Codex Alimentarius Commission 247, 249-51, 258, 259, 262-3 and EU 247-59 Financial Action Task Force (FATF)

253-6 global agency, proposed creation of 265-7 International Civil Aviation Organization (ICAO) 264-5 International Office of Epizootics 247-8, 249-51 International Plant Protection Convention 248-51, 259

lack of organized system 263-5 legitimacy 262-3 mutual recognition agreements 260-61 need for 259-67 problems with 262-5 rule-making by 261-2 United Nations Economic Commission for Europe (ECE) 256-7, 259 state, role in liberalized markets 23-5,

42 EU policy frameworks, gaps in 31-3 market behaviour and regulatory institutions 28-31 monitoring and measuring regulatory performance 34-6 reasons for regulation 25—8 regulatory convergence and divergence, model of 36-41 Stewart, R. 227 Stone-Sweet, A. 10, 13, 91-2, 94

subsidiarity 69, 78-9, 260 substantive rules 28-9, 31 Synergen case 190 Tallberg, J. 11 terrorist financing 253, 254

Thatcher, M. 6-7, 10, 13, 91-2, 94 trade liberalization, and international standard-setting agencies 260-61

Translation Centre for the Bodies of the EU 224, 228 transmission and non-delegation 148-9 rejection of 143-5, 151-3 transnational regulatory networks 74 transparency 16-17, 120, 129-32, 234

EU model 238-9 and international standard-setting 261-2, 263 US model 239

UK average tenure of IRA members 58 background of IRA members 56 government policy and IRAs 59, 61 overturning IRA decisions 59 overview of IRAs 50-S1 party politics 53, 54 resignations of IRA members 54 resources of IRAs 57 United Nations, and international

standard-setting 264 United Nations Economic Commission for Europe (ECE) 256-7, 259 US regulatory model 68, 80, 215, 239-40,

242 accountability 233-4, 236-8 categories of agencies 226-7 delegation 220 financial control 233 independence 230, 235-6 judicial review 227 overview of 216-17 procedural rules 227 transparency 239 USA genetically modified organisms 258 sector-specific regulation 195-6 state, views on 4 Weber, Max 152

Leaming Resources entre

; ay

|

:

ey

ea \

VM. ‘a

13

Asesqry

naw) aq

jo Ayisuaniuy

SODIMIOS Fae

Sk < At

The past decade has witnessed a proliferation of regulatory agencies at both the national and the EU level. This coherent and clearly structured book is the first of its kind to analyse in equal measure, and interdependently, both national

regulatory authorities and European agencies. It brings together a select group of highly esteemed contributors — authorities in their fields - to provide a

systematic and over-arching view of regulation in the EU. Unlike many of the previous attempts to shed light on this increasingly opaque and complex co-existence of regulatory systems, this book takes a genuinely multi-disciplinary approach with integrated perspectives from law, politics and economics. Exploring firstly the rationales for the existence of agencies, the book then goes on to examine how agencies are designed in the EU before considering the legal and political challenges they raise, and finally comparing them with international agencies and agencies in an enlarged Europe and the wider world.

Academic researchers in the fields of law, economics and politics will find Regulation through Agencies in the EU of great interest as will EU law practitioners, policymakers and regulators in Europe.

Damien Geradin is Professor of Law at the University of Liége and the College of Europe, Belgium, Rodolphe Mufioz is Research Fellow and Nicolas Petit is Research Fellow at the University of Liége, Belgium.

EDWARD

ELGAR

PUBLISHING

ISBN

1-84542-267-8

Glensanda House, Montpellier Parade Cheltenham, Glos, GL50 1UA, UK Tel: +44 (0) 1242 226934 Fax: +44 (0) 1242 262111 Email: [email protected]

136 West Street, Suite 202, Northampton, MA 01060, USA Tel: +1 413 584 5551 Fax: +1 413 584 9933 Email: [email protected] www e-elgar com

| |

: |

9

78184

| |