Legislation in Europe: A Comprehensive Guide for Scholars and Practitioners 9781509908752, 9781509908783, 9781509908776

Table of contents :
Preface
TABLE OF CONTENTS
List of Contributors
1
Introduction
1.1. What This Book Intends
1.2. Legislation and Legisprudence
1.3. Process of Legislation
1.4. Policy Making, Targets and Instruments of Legislation, Evaluation
1.5. Techniques
1.6. Some Trends of Legislation and Legisprudence
Further Reading
2
Legislation and Regulation
2.1. Legislation and Regulation2
2.2. Normative and Positive Theories of Regulation
2.3. Functions of Legislation as Yardsticks for Notions of "Good" Legislation
2.4. The Regulatory State and Its Consequences
2.5. Regulation: From a Solution to a Problem
2.6. Conclusion
Further Reading
3
Legislative Process
3.1. Introduction
3.2. Regulation and Organisation
3.3. Legal Measures, Other Bodies and Participants Taking Part in Legislative Processes
3.4. Legislative Process
3.5. Summary of the Legislative Process, Trends and Challenges
Further Reading
4
Management of Legislation
4.1. Introduction
4.2. Aspects of Regulatory Management
4.3. Regulatory Management in Multi-tiered Legal Systems
4.4. Conclusion
Further Reading
5
Participation
5.1. Definition and Delimitations
5.2. Functions of Participation
5.3. Limits
5.4. Forms of Participation63
5.5. Regulation on Consultation
Further Reading
6
Goals and Measures of Legislation: Evaluation
6.1. Introduction
6.2. What a Legislative Goal is Not
6.3. Structuring the Legislative Goal
6.4. The Function(s) of the Legislation and Its Goals
6.5. Locating the Legislative Goals
6.6. Conclusion
Further Reading
7
Legislative Drafting Techniques/Formal Legistics
7.1. Structure and Systematics as Elements of Quality of Law in Context
7.2. The Standardised Elements of Legal Texts: Between Logic and Historical Legacy
7.3. Cross-references
7.4. Amendments
7.5. Transposition of European Guidelines
7.6. Innovation: A Glimpse Into the Future
Further Reading
8
Legislative Language and Style
8.1. Characteristics of Normative Language
8.2. Legal Definitions
8.3. Techniques in Normative Language
8.4. Style of Legislation
8.5. Common Law and Civil Law Approaches to Legislation
Further Reading
9
Legislative Drafting
9.1. Introduction
9.2. The Drafting Process and Its Stages
9.3. Organisational Models and Their Actors
9.4. Models of Multilingual Legislative Drafting
9.5. Summary
Further Reading
10
Publication
10.1. Publication as a Basic Requirement of Legislation
10.2. Publication Methods
10.3. Legal Effect; Vacatio Legis
10.4. Publication Defects and Corrections
10.5. Consolidation and Codification
10.6. Conclusions
Further Reading
11
Regulatory Evaluation Ex Ante and Ex Post: Best Practice, Guidance and Methods
11.1. Introduction and Definitions
11.2. An Overview
Box 1
11.3. Problem Definition
11.4. Setting the Criteria
11.5. Measuring the Baseline
11.6. The Formulation of Options
11.7. Data Gathering
11.8. Organising Consultation
11.9. The Analysis of Options
Box 2
11.10. Reporting and Communication
11.11. Quality Assurance
11.12. Effectiveness
Further Reading
12
Maintenance of Rules
12.1. Which Maintenance for Which Rules?
12.2. Maintenance of Rules: Purposes and Object
12.3. Historical Roots of Maintenance of Rules
12.4. How Maintenance Is Changing
12.5. The Modern Legislator Walks on Two Legs: Law Making and Law Maintenance
Further Reading
13
EU Legislation
13.1. Introduction
13.2. Types of Legal Acts
13.3. Legislative Procedure
13.4. Languages
13.5. Accessibility
13.6. Drafting Process and Drafting Rules
13.7. Structure and Style
13.8. Role of the CJEU and Interpretation of EU Law
13.9. Quality of EU Legislation
13.10. Condensing and Updating the Statute Book
13.11. Final Words
Further Reading
14
Legislative Training
14.1. Specialised Legislative Training for Legislative Production
14.2. Legislative Training Programmes at European Level
14.3. Summary
Further Reading
15
Emerging Trends in Legislation in Europe
15.1. The EU"s Regulatory Framework for Legislative Quality1
15.2. The Post-Lisbon Smart Regulation Initiative
15.3. Legislative Quality in the Future: Blue-sky Initiatives or a Realistic Way Forward?
15.4. Conclusions
Index

Citation preview

LEGISLATION IN EUROPE This book provides a practical introduction to legislation. Written by a team of experts, made up of both practitioners and scholars, it will allow national institutions to devise their own drafting manuals and to draft laws. The examination focuses on general principles and best practice within the context of the different governmental systems in Europe. Questions explored include subsidiarity, legitimacy, efficacy, effectiveness, efficiency, proportionality, monitoring and regulatory impact assessment. Taking a practical approach, which starts from evidence-based rationality, it represents essential reading for all practitioners in the field of legislative drafting.

Completed with the generous support of the Volkswagen Foundation.

Legislation in Europe A Comprehensive Guide for Scholars and Practitioners

Ulrich Karpen and Helen Xanthaki in cooperation with Luzius Mader and Wim Voermans assisted by Ronan Cormacain

OXFORD AND PORTLAND, OREGON 2017

Hart Publishing An imprint of Bloomsbury Publishing Plc Hart Publishing Ltd Kemp House Chawley Park Cumnor Hill Oxford OX2 9PH UK

Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

www.hartpub.co.uk www.bloomsbury.com Published in North America (US and Canada) by Hart Publishing c/o International Specialized Book Services 920 NE 58th Avenue, Suite 300 Portland, OR 97213-3786 USA www.isbs.com HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2017 © The Editors and Contributors 2017 The Editors and contributors have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as Authors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the authors, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3) except where otherwise stated. All Eur-lex material used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2016. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: HB: 978-1-50990-875-2 ePDF: 978-1-50990-877-6 ePub: 978-1-50990-876-9 Library of Congress Cataloging-in-Publication Data Names: Karpen, Ulrich, author.  |  Xanthaki, Helen, author. Title: Legislation in Europe : a comprehensive guide for scholars and practitioners / Ulrich Karpen and Helen Xanthaki ; in cooperation with Luzius Mader and Wim Voermans ; assisted by Ronan Cormacain. Description: Oxford [UK] ; Portland, Oregon : Hart Publishing, 2017.  |  Includes bibliographical references and index. Identifiers: LCCN 2016046245 (print)  |  LCCN 2016046777 (ebook)  |  ISBN 9781509908752 (hardback : alk. paper)  |  ISBN 9781509908769 (Epub) Subjects: LCSH: Legislation—Europe. | Law—Europe—Philosophy. | Bill drafting—Europe. |  Representative government and representation—Europe. Classification: LCC KJC5349 .K39 2017 (print)  |  LCC KJC5349 (ebook)  |  DDC 328.4/077—dc23 LC record available at https://lccn.loc.gov/2016046245 Typeset by Compuscript Ltd, Shannon To find out more about our authors and books visit www.hartpublishing.co.uk. Here you will find extracts, author information, details of forthcoming events and the option to sign up for our newsletters.

PREFACE

Literature in the area of legislative studies suffers from the crucial lack of a leading ­overarching work, a point of international reference. This is the gap addressed by this book. Current materials on legislative drafting are highly jurisdiction specific. Our book approaches legislative studies through the lens of legislation in Europe, both at the Member State and at the EU level. The book concentrates on general principles and best practices, wherever they occur. Overarching issues and approaches further enhance the detailed analysis of specific topics. Aspects of regulation are discussed on a comparative and interdisciplinary basis. The book therefore brings together a collection of essays, written by multijurisdictional and multidisciplinary author teams representing a wide spectrum of European countries and roles within law making. The book is addressed to both academics and practitioners, namely policy officers, legal officers, drafters, parliamentary officers and departmental officers. It can serve as a guidebook for national and European drafters and legislators. It can also serve as a guidebook for national and European drafters and legislators. This volume could not have come about without the active meaningful support of the International Association of Legislation, whose membership includes the majority of academics and practitioners in the field of legislation in Europe. It is from this pool of experts that authors were selected to write on the current philosophy and practice of law making in Europe and the EU.

vi 

TABLE OF CONTENTS

Preface����������������������������������������������������������������������������������������������������������������������������������������v List of Contributors������������������������������������������������������������������������������������������������������������������xv

1. Introduction�����������������������������������������������������������������������������������������������������������������������1 Ulrich Karpen 1.1. What This Book Intends������������������������������������������������������������������������������������������1 1.2. Legislation and Legisprudence��������������������������������������������������������������������������������2 1.2.1. Laws and Other Regulations�����������������������������������������������������������������������2 1.2.2. Legisprudence���������������������������������������������������������������������������������������������3 1.2.3. Good Governance, Better Regulation, Better Legislation�������������������������3 1.2.4. Comparative Legisprudence as a Tool of ‘Better Legislation’�������������������4 1.3. Process of Legislation�����������������������������������������������������������������������������������������������5 1.3.1. Competences, Organisation and Procedure of Legislation�����������������������5 1.3.2. Legislation in Multi-level Systems��������������������������������������������������������������6 1.3.3. Organisation of Legislation and Separation of Powers�����������������������������7 1.3.4. Legislation in the Regulatory Cycle������������������������������������������������������������9 1.4. Policy Making, Targets and Instruments of Legislation, Evaluation�������������������10 1.4.1. Policy Making��������������������������������������������������������������������������������������������10 1.4.2. Targets of Legislation��������������������������������������������������������������������������������11 1.4.3. Instruments of Legislation�����������������������������������������������������������������������11 1.4.4. Evaluation��������������������������������������������������������������������������������������������������11 1.5. Techniques��������������������������������������������������������������������������������������������������������������13 1.5.1. Law Drafting����������������������������������������������������������������������������������������������13 1.5.2. Structure, Language, References���������������������������������������������������������������13 1.5.3. Special Categories of Laws������������������������������������������������������������������������14 1.5.4. Guidelines and Manuals for Law Making������������������������������������������������15 1.6. Some Trends of Legislation and Legisprudence���������������������������������������������������15 Further Reading���������������������������������������������������������������������������������������������������������������16 2. Legislation and Regulation����������������������������������������������������������������������������������������������17 Wim JM Voermans 2.1. Legislation and Regulation������������������������������������������������������������������������������������17 2.1.1. The Difference Between Legislation and Regulation������������������������������19 2.2. Normative and Positive Theories of Regulation��������������������������������������������������20 2.3. Functions of Legislation as Yardsticks for Notions of ‘Good’ Legislation�����������23 2.3.1. Legislative Functions���������������������������������������������������������������������������������23 2.3.2. Requirements and Demands on Legislation in Constitutional Democracies����������������������������������������������������������������������������������������������24

viii 

Table of Contents 2.4. The Regulatory State and Its Consequences���������������������������������������������������������26 2.5. Regulation: From a Solution to a Problem�����������������������������������������������������������27 2.6. Conclusion�������������������������������������������������������������������������������������������������������������29 2.6.1. Assessing the Effect of Better Law Making and Better Regulation Policies�������������������������������������������������������������������������30 2.6.2. The True Test of Values Involved in Better Regulation and Better Law Making�������������������������������������������������������������������������������������������������������31 Further Reading���������������������������������������������������������������������������������������������������������������32

3. Legislative Process������������������������������������������������������������������������������������������������������������33 Tímea Drinóczi 3.1. Introduction�����������������������������������������������������������������������������������������������������������33 3.2. Regulation and Organisation��������������������������������������������������������������������������������35 3.2.1. Basic Rules of the Legislative Process�������������������������������������������������������35 3.2.2. Constitutional Context and the Status of Parliament and Government in the Legislative Process���������������������������������������������36 3.2.2.1. Parliament���������������������������������������������������������������������������������36 3.2.2.2. Government������������������������������������������������������������������������������37 3.2.2.3. On the Loss of the Legislative Character���������������������������������38 3.3. Legal Measures, Other Bodies and Participants Taking Part in Legislative Processes����������������������������������������������������������������������39 3.3.1. Legal Measures������������������������������������������������������������������������������������������39 3.3.2. Other Actors and the Legislative Process�������������������������������������������������41 3.4. Legislative Process��������������������������������������������������������������������������������������������������44 3.4.1. Preparatory Stage or Pre-parliamentary Stage����������������������������������������44 3.4.2. Legislation in Legislative Bodies���������������������������������������������������������������46 3.4.3. Post-parliamentary Stage and Constitutional Review����������������������������47 3.5. Summary of the Legislative Process, Trends and Challenges�������������������������������������������������������������������������������������������48 Further Reading���������������������������������������������������������������������������������������������������������������51 4. Management of Legislation���������������������������������������������������������������������������������������������53 Patricia Popelier 4.1. Introduction�����������������������������������������������������������������������������������������������������������53 4.2. Aspects of Regulatory Management���������������������������������������������������������������������56 4.2.1. Culture�������������������������������������������������������������������������������������������������������56 4.2.2. Structures��������������������������������������������������������������������������������������������������59 4.2.3. Processes����������������������������������������������������������������������������������������������������64 4.2.4. Instruments�����������������������������������������������������������������������������������������������65 4.2.5. Competences���������������������������������������������������������������������������������������������66 4.3. Regulatory Management in Multi-tiered Legal Systems��������������������������������������68 4.3.1. Regulatory Management in EU Member States��������������������������������������68 4.3.2. Regulatory Management in Multi-tiered EU Member States�����������������69 4.4. Conclusion�������������������������������������������������������������������������������������������������������������71 Further Reading���������������������������������������������������������������������������������������������������������������72

Table of Contents

 ix

5. Participation���������������������������������������������������������������������������������������������������������������������73 Felix Uhlmann and Christoph Konrath 5.1. Definition and Delimitations��������������������������������������������������������������������������������73 5.2. Functions of Participation�������������������������������������������������������������������������������������76 5.2.1. Information and Transparency����������������������������������������������������������������76 5.2.2. Legitimacy�������������������������������������������������������������������������������������������������77 5.2.3. Political Feasibility������������������������������������������������������������������������������������78 5.2.4. Fact Finding�����������������������������������������������������������������������������������������������79 5.2.5. Implementation and Practicability����������������������������������������������������������80 5.2.6. Legislative Quality�������������������������������������������������������������������������������������81 5.3. Limits����������������������������������������������������������������������������������������������������������������������81 5.4. Forms of Participation�������������������������������������������������������������������������������������������82 5.4.1. Consultation on the Timeline of the Legislative Process������������������������83 5.4.2. Authorship and Responsibilities��������������������������������������������������������������85 5.4.3. Objects of Consultation����������������������������������������������������������������������������85 5.4.4. Participants������������������������������������������������������������������������������������������������87 5.4.5. Format�������������������������������������������������������������������������������������������������������88 5.4.6. Deadlines���������������������������������������������������������������������������������������������������90 5.4.7. Assessment of Feedback and Use of Results��������������������������������������������91 5.4.8. Consultation and Parliamentary Procedures������������������������������������������92 5.5. Regulation on Consultation����������������������������������������������������������������������������������93 5.5.1. National Constitutions and Laws�������������������������������������������������������������93 5.5.2. National Ordinances, Manuals and Practice�������������������������������������������94 Further Reading���������������������������������������������������������������������������������������������������������������95 6. Goals and Measures of Legislation: Evaluation��������������������������������������������������������������97 Mauro Zamboni 6.1. Introduction�����������������������������������������������������������������������������������������������������������97 6.2. What a Legislative Goal is Not������������������������������������������������������������������������������98 6.3. Structuring the Legislative Goal����������������������������������������������������������������������������99 6.4. The Function(s) of the Legislation and Its Goals�����������������������������������������������102 6.5. Locating the Legislative Goals�����������������������������������������������������������������������������105 6.6. Conclusion�����������������������������������������������������������������������������������������������������������107 Further Reading�������������������������������������������������������������������������������������������������������������107 7. Legislative Drafting Techniques/Formal Legistics��������������������������������������������������������109 Jean-Pierre Duprat and Helen Xanthaki 7.1. Structure and Systematics as Elements of Quality of Law in Context��������������109 7.1.1. The Hold on Formal Legistics and Legal Certainty������������������������������110 7.1.1.1. A Main Interest in Economic and Social Effects of Regulation���������������������������������������������������������������110 7.1.1.2. Accessibility to Law and Democratic Prerequisite�����������������110 7.1.1.3. Formal Legistics and the ‘Legislative Cycle’���������������������������111 7.1.2. Historical Point of View: The Various Dimensions Afforded to Formal Legistics������������������������������������������������������������������111

x 

Table of Contents 7.1.3. Systematisation and Inner Structure of Statutes�����������������������������������113 7.1.3.1. Enhancement of Codification and Repealing Statutes to Simplify Law���������������������������������������������������������113 7.1.3.2. Formal Systematisation of Statutes����������������������������������������114 7.2. The Standardised Elements of Legal Texts: Between Logic and Historical Legacy�����������������������������������������������������������������116 7.2.1. The Title of Statutes��������������������������������������������������������������������������������116 7.2.1.1. The Practice of a Short Title: Information and Citation����������������������������������������������������������������������������116 7.2.1.2. The Short Title as the Dominant Continental Practice���������������������������������������������������������������117 7.2.2. The Decreased Usage of the Preamble and Its Modern Substitutes���������������������������������������������������������������������������������118 7.2.3. The Search for Clear Presentation of Provision: Numbering and Heading of Divisions and Subdivisions���������������������119 7.2.3.1. Continuous Numbering of Sections and Differentiation from Other Divisions of the Statute������������119 7.2.3.2. Headings, Understandability for Users and Consistency of Drafting����������������������������������������������������������120 7.2.3.3. Transitional Provisions and Legal Certainty�������������������������121 7.3. Cross-references���������������������������������������������������������������������������������������������������121 7.4. Amendments��������������������������������������������������������������������������������������������������������122 7.5. Transposition of European Guidelines���������������������������������������������������������������123 7.6. Innovation: A Glimpse Into the Future��������������������������������������������������������������124 Further Reading�������������������������������������������������������������������������������������������������������������127

8. Legislative Language and Style��������������������������������������������������������������������������������������129 Mario Hernández Ramos and Volker Heydt 8.1. Characteristics of Normative Language��������������������������������������������������������������129 8.1.1. Clarity������������������������������������������������������������������������������������������������������130 8.1.2. Conciseness���������������������������������������������������������������������������������������������132 8.2. Legal Definitions��������������������������������������������������������������������������������������������������133 8.2.1. Functions�������������������������������������������������������������������������������������������������134 8.2.2. Rules��������������������������������������������������������������������������������������������������������134 8.3. Techniques in Normative Language��������������������������������������������������������������������135 8.3.1. Legal Lexicon�������������������������������������������������������������������������������������������136 8.3.2. Non-legal Lexicon�����������������������������������������������������������������������������������136 8.4. Style of Legislation�����������������������������������������������������������������������������������������������137 8.4.1. Design of the Legal Order and Access to the Law���������������������������������137 8.4.2. Structure and Style of the Individual Piece of Legislation��������������������138 8.4.3. Linguistic Aspects������������������������������������������������������������������������������������139 8.5. Common Law and Civil Law Approaches to Legislation�����������������������������������140 Further Reading�������������������������������������������������������������������������������������������������������������142 9. Legislative Drafting��������������������������������������������������������������������������������������������������������145 Stefan Höfler, Markus Nussbaumer and Helen Xanthaki 9.1. Introduction���������������������������������������������������������������������������������������������������������145

Table of Contents

 xi

9.2.

The Drafting Process and Its Stages����������������������������������������������������������������146 9.2.1. Planning���������������������������������������������������������������������������������������������147 9.2.2. Composing�����������������������������������������������������������������������������������������149 9.2.3. Revising����������������������������������������������������������������������������������������������150 9.2.4. Editing������������������������������������������������������������������������������������������������152 9.2.5. Conclusions����������������������������������������������������������������������������������������152 9.3. Organisational Models and Their Actors��������������������������������������������������������153 9.3.1. Decentralisation���������������������������������������������������������������������������������153 9.3.2. Centralisation�������������������������������������������������������������������������������������155 9.3.3. Examples���������������������������������������������������������������������������������������������156 9.3.3.1. United Kingdom����������������������������������������������������������������156 9.3.3.2. Switzerland�������������������������������������������������������������������������157 9.3.3.3. Conclusions������������������������������������������������������������������������158 9.4. Models of Multilingual Legislative Drafting���������������������������������������������������159 9.4.1. Co-drafting�����������������������������������������������������������������������������������������159 9.4.2. Co-revision�����������������������������������������������������������������������������������������160 9.4.3. Co-editing������������������������������������������������������������������������������������������161 9.4.4. Conclusions����������������������������������������������������������������������������������������161 9.5. Summary����������������������������������������������������������������������������������������������������������161 Further Reading������������������������������������������������������������������������������������������������������������163 10. Publication�������������������������������������������������������������������������������������������������������������������165 Karl Irresberger and Anna Jasiak 10.1. Publication as a Basic Requirement of Legislation�����������������������������������������165 10.2. Publication Methods����������������������������������������������������������������������������������������167 10.2.1. Paper Publication�������������������������������������������������������������������������������167 10.2.2. Electronic Publication�����������������������������������������������������������������������170 10.2.3. Publication Requirements in Case of Legislation by Reference�������174 10.2.4. Deposit for Public Inspection Instead of Official Publication���������177 10.2.5. Extraordinary Publication�����������������������������������������������������������������177 10.3. Legal Effect; Vacatio Legis��������������������������������������������������������������������������������178 10.4. Publication Defects and Corrections��������������������������������������������������������������179 10.5. Consolidation and Codification����������������������������������������������������������������������180 10.5.1. Codification and Consolidation at the National Level��������������������180 10.5.2. Consolidation and Codification in the EU���������������������������������������183 10.6. Conclusions������������������������������������������������������������������������������������������������������184 Further Reading������������������������������������������������������������������������������������������������������������184 11. Regulatory Evaluation Ex Ante and Ex Post: Best Practice, Guidance and Methods�����������������������������������������������������������������������������������������������187 Stephan Naundorf and Claudio M. Radaelli 11.1. Introduction and Definitions��������������������������������������������������������������������������187 11.2. An Overview�����������������������������������������������������������������������������������������������������190 11.3. Problem Definition������������������������������������������������������������������������������������������194 11.3.1. Common Myths About Problems Are Hard to Dispel���������������������195 11.4. Setting the Criteria�������������������������������������������������������������������������������������������195

xii 

Table of Contents 11.5. Measuring the Baseline����������������������������������������������������������������������������������196 11.6. The Formulation of Options�������������������������������������������������������������������������197 11.7. Data Gathering����������������������������������������������������������������������������������������������201 11.8. Organising Consultation�������������������������������������������������������������������������������202 11.9. The Analysis of Options��������������������������������������������������������������������������������204 11.10. Reporting and Communication�������������������������������������������������������������������208 11.11. Quality Assurance������������������������������������������������������������������������������������������209 11.12. Effectiveness���������������������������������������������������������������������������������������������������211 Further Reading������������������������������������������������������������������������������������������������������������211

12. Maintenance of Rules��������������������������������������������������������������������������������������������������215 Maria De Benedetto 12.1 Which Maintenance for Which Rules?���������������������������������������������������������215 12.1.1. What Is Meant by the Reference to Rules?�������������������������������������216 12.2. Maintenance of Rules: Purposes and Object������������������������������������������������216 12.3. Historical Roots of Maintenance of Rules����������������������������������������������������220 12.4. How Maintenance Is Changing���������������������������������������������������������������������221 12.4.1. Reasons for Change������������������������������������������������������������������������221 12.4.2. Maintenance and Quality of Rules������������������������������������������������222 12.4.3. Kinds of Maintenance (Compilation, Consolidation, Revision, Reform)���������������������������������������������������������������������������224 12.5. The Modern Legislator Walks on Two Legs: Law Making and Law Maintenance������������������������������������������������������������������������������������226 Further Reading������������������������������������������������������������������������������������������������������������227 13. EU Legislation��������������������������������������������������������������������������������������������������������������229 William Robinson 13.1. Introduction���������������������������������������������������������������������������������������������������229 13.1.1. Origins���������������������������������������������������������������������������������������������229 13.1.2. Present Position������������������������������������������������������������������������������230 13.1.3. A New Legal Order�������������������������������������������������������������������������231 13.1.4. General Principles���������������������������������������������������������������������������231 13.2. Types of Legal Acts����������������������������������������������������������������������������������������233 13.2.1. General��������������������������������������������������������������������������������������������233 13.2.2. Regulations��������������������������������������������������������������������������������������233 13.2.3. Directives�����������������������������������������������������������������������������������������234 13.2.4. Decisions�����������������������������������������������������������������������������������������235 13.2.5. Recommendations��������������������������������������������������������������������������235 13.2.6 Opinions�����������������������������������������������������������������������������������������236 13.2.7. Interinstitutional Agreements��������������������������������������������������������236 13.2.8. Legislative and Non-legislative Procedure�������������������������������������236 13.2.9. Delegated and Implementing Acts�������������������������������������������������237 13.2.9.1. Delegated Acts���������������������������������������������������������������237 13.2.9.2. Implementing Acts��������������������������������������������������������237 13.3. Legislative Procedure�������������������������������������������������������������������������������������238 13.3.1. General��������������������������������������������������������������������������������������������238

Table of Contents

 xiii

13.3.2. Commission Proposal������������������������������������������������������������������239 13.3.3. European Parliament��������������������������������������������������������������������240 13.3.4. Council������������������������������������������������������������������������������������������240 13.3.5. Trilogues����������������������������������������������������������������������������������������241 13.3.6. Role of the Commission in the Codecision Procedure��������������241 13.4. Languages�������������������������������������������������������������������������������������������������������241 13.5. Accessibility����������������������������������������������������������������������������������������������������242 13.6. Drafting Process and Drafting Rules������������������������������������������������������������243 13.6.1. Drafting Process����������������������������������������������������������������������������243 13.6.2. Drafting Rules�������������������������������������������������������������������������������244 13.6.3. Rules on Legal-linguistic Revision�����������������������������������������������245 13.7. Structure and Style����������������������������������������������������������������������������������������246 13.7.1. Structure���������������������������������������������������������������������������������������246 13.7.2. Drafting Style��������������������������������������������������������������������������������247 13.8. Role of the CJEU and Interpretation of EU Law�����������������������������������������248 13.8.1. General������������������������������������������������������������������������������������������248 13.8.2. Actions for Annulment�����������������������������������������������������������������249 13.8.3. References for Preliminary Rulings���������������������������������������������249 13.8.4. Infringement Proceedings������������������������������������������������������������249 13.8.5. Interpretation of EU Law�������������������������������������������������������������249 13.9. Quality of EU Legislation������������������������������������������������������������������������������250 13.9.1. Calls for Improvement and Response of the EU Institutions�����������������������������������������������������������������������250 13.9.2. Governance Initiative�������������������������������������������������������������������251 13.9.3. Agreement on Better Law Making 2003��������������������������������������251 13.9.4. Smart Regulation��������������������������������������������������������������������������252 13.9.5. REFIT��������������������������������������������������������������������������������������������252 13.9.6. Better Regulation Package 2015���������������������������������������������������253 13.9.7. Agreement on Better Law Making 2016��������������������������������������253 13.10. Condensing and Updating the Statute Book������������������������������������������������253 13.10.1. Amendment����������������������������������������������������������������������������������253 13.10.2. Corrigenda������������������������������������������������������������������������������������253 13.10.3. Consolidation�������������������������������������������������������������������������������254 13.10.4. Codification����������������������������������������������������������������������������������254 13.10.5. Recasting���������������������������������������������������������������������������������������254 13.10.6. Repeal��������������������������������������������������������������������������������������������255 13.11. Final Words����������������������������������������������������������������������������������������������������255 Further Reading������������������������������������������������������������������������������������������������������������255 14. Legislative Training������������������������������������������������������������������������������������������������������257 Marta Tavares de Almeida and Chris Moll 14.1. Specialised Legislative Training for Legislative Production�������������������������257 14.1.1. The Importance of Specialised Legislative Training for the Quality of Legislation�������������������������������������������������������257 14.1.2. Development of Legislative Training Programmes at European Level�������������������������������������������������������������������������258

xiv 

Table of Contents 14.1.3. Academic and Professional Legislative Training������������������������������259 14.1.4. Training Activities by International and National Associations���������������������������������������������������������������������������������������259 14.1.5. The Two Stages of Law Preparation�������������������������������������������������260 14.1.6. Law Preparation in Common Law and Civil Law Juridictions��������������������������������������������������������������������������������260 14.1.7. EU Legislation: Law Making and Training��������������������������������������262 14.1.8. Regulatory Management Training����������������������������������������������������263 14.2. Legislative Training Programmes at European Level�������������������������������������264 14.2.1. Origin and Aims��������������������������������������������������������������������������������264 14.2.2. The Organisation of Training in Member States�����������������������������266 14.2.3. Training Activities Programme Design��������������������������������������������269 14.2.4. Training Activities Target Audiences������������������������������������������������271 14.3. Summary����������������������������������������������������������������������������������������������������������272 Further Reading������������������������������������������������������������������������������������������������������������273

15. Emerging Trends in Legislation in Europe�����������������������������������������������������������������275 Helen Xanthaki 15.1. The EU’s Regulatory Framework for Legislative Quality�������������������������������275 15.2. The Post-Lisbon Smart Regulation Initiative�������������������������������������������������278 15.3. Legislative Quality in the Future: Blue-sky Initiatives or a Realistic Way Forward?������������������������������������������������������������284 15.3.1. The Layered Approach to Structure�������������������������������������������������289 15.3.2. Legislative Image: Presentation, Layout, Pictures����������������������������291 15.3.3. The Statute Book or the Body of Legislation as a Whole: Quantity and Beyond������������������������������������������������������������������������292 15.3.4. The Theoretical Umbrella: Phronetic Legislative Drafting�������������294 15.4. Conclusions������������������������������������������������������������������������������������������������������296

Index��������������������������������������������������������������������������������������������������������������������������������������297

LIST OF CONTRIBUTORS

Cormacain Ronan, LLM (QUB), LLM (UoLondon), BL, Consultant Legislative Counsel, Co-Director of the Sir William Dale Legislative Drafting Clinic, Institute of Advanced Legal Studies, University of London, UK; author of ‘Retroactivity, Retrospectivity, and ­Legislative Competence in Northern Iceland: Determining the Validity of Janus-Faced L ­ egislation’ [2015] Statute Law Review; and ‘An Empirical Study of the Usefulness of Legislative ­Drafting Manuals’ [2013] Theory and Practice of Legislation 205. De Almeida Tavares Marta, Jurist; editor of the periodical Legislação (1991–2010); ViceChairwoman of the International Association of Legislation, IAL (1991–2012); member of the Advisory Board of IAL (since 2013), Portugal. De Benedetto Maria, Professor of Administrative Law at Roma Tre University, Italy; legal advisor for the Italian Anti-trust Authority (1996), the Presidency of the Italian Council of Ministers—Department for Civil Service (2000), the Italian School of Public ­Administration (2002–04); member of institutional committees on institutional reforms, such as the Italian Unit for Simplification and Quality of Regulation (2006–08); editor in chief of Studi Parlamentari e di Politica Costituzionale; co-author of N Rangone, M Martelli and M de Benedetto, La qualita delle regole (Rome, Mulino, 2011); author of ‘Maintenance of Rules’, CREI Working Paper 2/2014. Drinóczi Tímea, PhD, dr Habil; Associate Professor, Department of Constitutional Law, Faculty of Law, University of Pécs, Hungary; member of the Advisory Board of the International Association for Legislation; author of ‘Concept of Quality of Legislation— Revisited: Matter of Perspective and a General Overview’ [2015] Statute Law Review 211; and ‘Linguistic Approach in Legisprudence—Terminology, Translation Studies and Databases’ (2015) 3 Theory and Practice of Legislation 113 (with Barnabas Novak). Duprat Jean-Pierre, PhD in law, Aggregation in Public Law, Emeritus Professor, University of Bordeaux (CERDARE), France; Deputy Scientific Director (Law and Political Science), Ministry of National Education and Research (1998–2002), Expert for Parliamentary Office of Scientific and Technological Assessment; Member of the Société de législation comparée, of the International Association of Legislation, Associate member of the International Academy of Comparative Law; author of ‘Le parlement évaluateur’ [1998] Revue internationale de droit comparé No 68; co-author in R Drago (ed), La confection de la loi (PUF, 2005); La répartition des compétences normatives et la qualité de la loi (Annuaire international de justice constitutionnelle, Economica, 2007); ‘The Judicial Review of Ex Ante Impact Assessment in France’ (2012) 6(3) Legisprudence; ‘Reform of Legislative Regulation in Economics’ in L Mader and S Kapyshev (eds), Regulatory Reforms—Implementation and Compliance (Baden-Baden, Nomos, 2014). Hernandez Ramos Mario, Dr, Assistant Professor of Constitutional Law at ­Salamanca University, Spain; Research Fellow, Institute of Global Governance, University of ­Salamanca; lecturer in constitutional law since 2007.

xvi 

List of Contributors

Höfler Stefan, Dr, Research Fellow, Centre for Legislative Studies, Institute of Law, ­ niversity of Zurich, Switzerland; Legal Editor, Internal Drafting Committee of the Swiss U Federal Administration, Swiss Federal Chancellery, Bern, Switzerland (2013–16); author of ‘Die Redaktion von Verweisen unter dem Aspekt der Verstandlichkeit’ [2015] LeGes 325; and ‘Between Conciseness and Transparency: Presuppositions in Legislative Texts’ [2014] International Journal of Semiotics of Law 627. Irresberger Karl, Dr iur, Head of Legistics, Legal Information etc Division, Federal Chancellery, Vienna, Austria; author of ‘Die Publikation von Rechtsvorshcriften in Osterreich’ [2013] LeGes 141. Jasiak Anna, PhD, Council of State of the Netherlands, seconded expert to the Venice Commission of the Council of Europe in Strasbourg (2014); author of Constitutional ­Constraints on Ad Hoc Legislation (Antwerp, Intersentia, 2010). Karpen Ulrich Karpen, Dr, Professor of Constitutional and Administrative Law, University of Hamburg, Germany; member of the Hamburg State Parliament ­ (1991–2001); former chairman of the International Association of Legislation and of the German ­Association of Legislation, member of the editorial board of Zeitschrift für Gesetzgebung (ZG); author of Legistics—Freshly Evaluated, 2nd enlarged edn (Baden-Baden, Nomos, 2008); and ‘Comparative Law: Perspectives of Legislation’ [2012] Legisprudence 149. Konrath Christoph, Dr iur, MSc (LSE), Legal, Legislative and Research Service of the ­Austrian Parliamentary Administration, Vienna (Austria); lecturer of constitutional law and political science, University of Vienna. Mader Luzius, Dr iur, LLM (London), Deputy Director of the Federal Office of Justice, Berne (Switzerland), Professor at the Swiss Graduate School of Public Administration (Law Faculty of the University of Lausanne), former chairman of the International Association of Legislation and of the Swiss Society of Legislation, member of the editorial board of LeGes. Moll Chris, LLM, Consultant Regulatory Governance, Anti-corruption and Reform at Lexchange Ltd, the Hague, the Netherlands; co-editor (with Luzius Mader) of Proceedings of the 7th Congress of the European Association of Legislation (EAL) (Baden-Baden, Nomos, 2009). Naundorf Stephan, Dipl sc pol Univ, Dip-Kaufmann University, Better Regulation Unit, Federal Chancellery, Germany (Potsdam); author of ‘Law simplification en Republique ­federale d’Allemagne’ [2016] Revue Francaise d’Administration Publique 157. Nussbaumer Markus, Dr Phil, Head of the Internal Drafting Committee of the Swiss Federal Administration, Swiss Federal Chancellery, Bern, Switzerland; member of the C ­ ommittee of the Swiss Association of Legislation; co-editor of the journal LeGes –Gestzgebung und Evaluation; author of ‘Rhetorisch-stilistische Eigenschaften der Sprache des Rechtwesens’ in U Fix, A Gardt and J Kanpe (eds), Rhetorik und Stilistik/ Rhetoric and ­Stylistics, HSK 31.2 (Berlin, de Gruyter, 2009) 2132–50; ‘Der Verstandlickeit eine Anwaltin! Die Redaktionskommission der scheizerischen Bundesverwaltung und ihre Arbeit an der Gesetzsprache’ in KM Eichhoff-Cyrus and G Antos (eds), Verstadnlsichkeit als ­Burgerrecht? Die Rechts- und Verwaltungssprache in der offentlichen Diskussion (Mannheim, Duden, 2008) 301–23. Popelier Patricia, Dr, Full Professor of Constitutional Law, University of Antwerp; ViceDean for Research; Vice-Chair of the International Association of Legislation; Chair of the Interuniversity Center of Legislation Flanders (ICW); member of the editorial board of

List of Contributors

 xvii

Theory and Practice of Legislation; guest editor of the Legisprudence special issue on ‘The Role of Courts as Regulatory Watchdogs’. Radaelli Claudio, Dr, Professor of Political Science; Director, Centre for European ­Governance, University of Exeter, UK; Jean Monnet Chair in Political Economy; co-editor (with Claire Dunlop) of the Handbook of Regulatory Impact Assessment (Cheltenham, Elgar, 2016). Robinson William, LLB, Associate Research Fellow, Institute of Advanced Legal ­Studies, University of London; formerly a Legal Reviser at the Court of Justice of the EU; and ­Coordinator in the Quality of Legislation Team of the European Commission Legal S­ ervice; editor of the special issue of The Theory and Practice of Legislation on European Union ­legislation (2014); author of ‘Time for Coherent Rules on EU Regulation’ [2015] Theory and Practice of Legislation 257. Uhlmann Felix, Dr, LLM (Harvard), Professor of Constitutional and Administrative Law and Legislation, University of Zurich, Switzerland; co-author (with Georg Müller) of ­Elemente einer Rechtssetzungslehre, 3rd edn (Zurich, Schulthess Verlag, 2013). Voermans Wim, Dr, Professor of Constitutional and Administrative Law at Leiden ­University, the Netherlands; Scientific Director of the Institute of Public Law at Leiden Law School; Co-Director of the Profile Area Political Legitimacy of Leiden University; President of the International Association for Legislation. Volker Heydt, Rechtsanwalt (Hamburg bar), retired official of the European Commission, Tervuren (Belgium); author of ‘Zum Verkundungswesen im demokratischen Rechtsstaat’ in Demokratie und Verwaltung, 25 Jahre Hochschule fur Verwaltungswissenschaften Speyer (Berlin, 1972) 463–82. Xanthaki Helen, Dr, Professor of Law, University College London; Director, International PG Laws Programme, University of London; author of Drafting Legislation: Art and ­Technology of Rules for Regulation (Oxford, Hart Publishing, 2014); and Thornton’s ­Legislative Drafting, 5th edn (London, Bloomsbury, 2013); member of the editorial board of Theory and Practice of Legislation. Zamboni Mauro, Dr, Professor of Law, University of Stockholm, Senior Associate Research Fellow at the Institute of Advanced Legal Studies, University of London (2015–), general editor for Theory and Practice of Legislation (2014–); author of ‘Legislating Politics’ (2009) 2(3) Legisprudence: An International Journal for the Study of Legislation.

xviii 

1 Introduction ULRICH KARPEN

This book on legislation and legisprudence offers a survey in the structure and process of law making, policy, targets, instruments, and methods and techniques of modern legislation in Europe (both in the Member States of the European Union and in the Union itself). Chapters aim to give an overview of the current state of legisprudence in the research and teaching of legislative drafting and best practice application. Legisprudence in theory and practice contributes to ‘better regulation’. Its main purpose is to monitor and improve the quality of laws and to reduce their quantity. Transparency of legislation in an ever more complex world is a prerequisite of ‘better governance’. The chapters of this book take a comparative and generic perspective in the main aspects of legislation and legisprudence. As an introduction, Chapter 1 focuses on linking the chapters that follow. Chapter 2 highlights the basic question ‘What is legislation and the law?’ (analysis of the law). The organisation and procedure of legislative bodies and legislation, namely the management of legislation, are analysed in Chapters 2–5. Chapter 6 discusses what the legislator intends and how this is to be accomplished, that is, goals and measures (methodology of legislation). The techniques of drafting laws, their structure and language are discussed in Chapters 7–10. Monitoring of legislative quality and necessary amendments are depicted in Chapters 11 and 12. Finally, prominent themes of legislative studies are addressed in Chapters 13–15: training of drafters, procedures of European legislation, and emerging trends in the quantity and quality of legislation.

1.1.  What This Book Intends The authors of this book on legislation present a survey of subjects, procedures, targets, instruments and methods of modern legislation in the Member States of the European Union. They give an overview of the current state of legisprudence as a research topic, concerning the teaching and lecturing of law production and the application of standards in practice. The aim of legisprudence in theory and practice is to contribute to ‘better regulation’. Its main purpose is to monitor and improve the quality of laws and reduce their quantity. Transparency of legislation in an increasingly complex world is a prerequisite of ‘better governance’. Legislation today is still predominantly a national matter. Nevertheless, supranational regulations, such as EU law, and global ones, like international economic and ­environmental law, are implemented. To render this multi-level system of regulations more transparent,

2 

Ulrich Karpen

comparative law efforts are required, and steps towards unified legislation have already been taken. This book focuses on the principles and standards of legislation in constitutional ruleof-law democracies. The authors do not intend to present a ‘cookbook for better legislation’. Procedures and methods of drafting as well as monitoring are still too divergent in the Member States of the EU. Since these standards, as presented here, are the result of ‘good practice’, this book claims to be applicable to national legislators. Practically speaking, every country has its own ‘guidelines for (good) legislation’ and its own manuals. To study them carefully was a necessity for all chapters. Such a comparative perspective might widen the horizon for readers of topics and problems of legislation. A well-known aphorism of the psychiatrist Scott Peck suggests ‘that we should share our similarities and celebrate our differences’.1 If we understand why and how, in applying common standards, others legislate differently than how we do, we are encouraged to compare and learn, and to improve our methods or retain our own legislative style.2

1.2.  Legislation and Legisprudence 1.2.1.  Laws and Other Regulations Laws are general, abstract norms, as opposed to decisions in a particular case. The law is the primary and central instrument of government in a democratic rule-of-law state. The law regulates the organisation and procedures of state institutions, protects individual freedoms and serves as the single most important instrument to distribute and allot social expenditures. The term ‘legislation’ does not just cover formal laws, as enacted by ­parliament (ie statutes); legisprudence is also interested in norm, which rank below statutory law in the hierarchy of regulations, like secondary (delegated) legislation, by-laws and administrative regulations.3 Requirements of competence and procedure, goals and drafting techniques apply to them as well. The subject of legisprudence is law making in a broader sense, ie regulation. However, this book focuses on primary and secondary legislation. All constitutions regulate on the phases of enactment of statutory law: initiative, approval by parliament and/or referenda, implementation, amendment, etc. The admissibility and use of regulations on a hierarchical level below statute law differ from state to state, and are often not very transparent (secondary, subordinate, delegated legislation, governmental or administrative rule making, rules, ordonnances, decrets, decrets-lois, etc). Article 290(1) and (2) of the Treaty on the Functioning of the European Union (TFEU) requires that the legislative act which delegates (non-statutory) legislative power of general application to the executive must explicitly define the objects, content, scope and duration of the delegation: ‘The essential elements of an area shall be reserved for the legislative act.’4 In all countries, 1 

See W Voermans, ‘Styles of Legislation and Their Effects’ (2010) 32 Statute Law Review 38. See U Karpen, ‘Comparative Law: Perspectives of Legislation’ (2012) 6 Legisprudence 149. 3  See L Mader, ‘Die Rechtsetzung’ in A Ladner et al (eds), Handbuch der öffentlichen Verwaltung in der Schweiz (Zurich, Verlag Neue Zürcher Zeitung, 2013) 249. 4  Art 290 (1)(2) TFEU. 2 

Introduction

 3

secondary legislation by far outnumbers statutory law. This is a problem in a parliamentary system. On the one hand, it sheds light on the (often necessary at first sight) increase in public tasks which can only be managed properly by the executive; on the other hand, it contributes to the creeping loss of parliamentary power.

1.2.2. Legisprudence Legisprudence (legistics, legistique) looks into the whole regulation cycle, from impulse to amendment. It does so through the analysis of norms (b), through research and practice of organisation and procedure (c), and later by describing methods for setting policies and adequate goals, as well as by the choice of effective and efficient means of regulation (d), (e) with the intent to regulate on matters in a precise, clear, understandable form and by applying technical instruments for better regulation, like references, settlements or codifications, in order to keep the body of law small and transparent. Legisprudence is an interdisciplinary theoretical and practical science. Juridical sciences, economics, social and political sciences, philologies, information theory and cybernetics cooperate to better understand and improve legislation, to facilitate the implementation of constitutional values and laws, and ‘social engineering’. Legisprudence is first and foremost a theoretical science. It is descriptive, applying the hermeneutical methods of humanities. It writes and interprets legal texts, and uses empirical methods of social sciences. Legisprudence is also, however, a practical science. It is prescriptive and normative. The practical approach does not primarily accumulate knowledge, but, rather, aims to direct actions. It strives for political rationality. Politics needs to gain and keep democratic legitimation in order to realise policies by legislation. Finally, legisprudence deals with handicraft or even arts, and—at its best— requires intuition, talent and gift. ‘We hold these truths to be self-evident, that all men are created equal, that they are endowed by their creator with certain unalienable rights, that among them are life, liberty and the pursuit of happiness.’ These well-known words of the ­American Declaration of Independence are an outstanding piece of legislation, and GK Chesterton did not hesitate to evaluate it as ‘great literature’.5

1.2.3.  Good Governance, Better Regulation, Better Legislation The practical side of legislation, namely the policy elements of planning and decision making in the legislative process, has been neglected for a long time in European legisprudence. It was left to political sciences. In the meantime this has changed. It is obvious that legitimation of law does not come from rationality of its matter, but from democratic sources. Parliament and government as legislators need the majority support of the people. At best, rationality and majority coincide as a solid basis for implementing the law. If this is not the case, then the majority vote in politics takes precedence. This is one of the main reasons why legisprudence in recent times is more interested in the broader approaches of ‘good government’ and ‘better regulation’—better policy. Within the focus of ‘better legislation’, it cooperates in reducing the quantity and improving the quality of legislation. 5 

GK Chesterton, What I Saw in America (New York, Dodd, Mead & Co, 1922) 7.

4 

Ulrich Karpen

The term ‘better regulation’ is more oriented towards political management. The evaluation of political bodies, such as states, measures the steering capacity of governments and institutions of society on their way to democracy and a social market economy. Legislation is just one of a number of regulatory instruments, like (financial) support, subsidies, contracts and various forms of coordination and cooperation.6 As ‘better law making’, legisprudence strives at reducing the number of statutes and improving the transparency of legal systems, as well as increasing their efficacy and efficiency. It does so by more participation with and self-regulation by stakeholders, accelerated procedures of legislation and increased numbers of Regulatory Impact Assessments (RIAs).7 For example, in 2002, the EU adopted an interinstitutional agreement on better legislation.8 Much has been written on the quantity of modern legislation. Reasons may be the increase in the number of public tasks in welfare and intervention states, technology and the juridification of many areas of life. Undoubtedly, excessive legislation is a criterion of quality deficits, for laws which one cannot know, which one does not understand cannot be implemented effectively. Turning to the quality of laws, there is no such thing as an ‘ideal law’, ie a law which, according to targets and efficacy, transparency and good wording, is perfect, which is drafted and adopted in a well-organised, speedy procedure and meets all requirements of content and form. The drafter can, however, write a ‘good’ or ‘relatively’ good piece of legislation which regulates properly a new field of legislation or—with amendments—deletes recognised or recognisable deficits.9 Constitution and law provide quality standards for procedure, goals and contents, as well as for the textual form of legislation. There are quality standards which may be traced from substantive and material as well as procedural principles of due process.10

1.2.4.  Comparative Legisprudence as a Tool of ‘Better Legislation’ Comparative law is an essential element of legisprudence, in particular in the search for methods for better regulation. Comparative legisprudence analyses the functions of (statutory) law in legal systems as instruments of regulation, organisation and stabilisation of societies and distribution of social services. This is particularly true for comparative constitutional law. This book is not the proper place to do detailed research in the thicket of national constitutions. It is, however, necessary and possible to describe the principles of the democratic rule-of-law state which determine the procedure and contents of legislation and laws. These standards are vested in the organic and dogmatic parts of the constitution: the separation of powers and the (partly) federal decentralisation on the one hand, and the human rights, social and ecological state requirements on the other. Styles of legislation and methods of interpretation and implementation may differ significantly: civil and case law, codes or precedents, deductive or inductive methods of understanding the body of law. The main elements of national constitutions—as well as the treaties of the EU—are, 6 

See Chapter 14. See Chapter 10. 8  Interinstitutional Agreement (2003/C321/01). 9  See Chapters 5, 6 and 7. 10  See L Tribe, American Constitutional Law, 2nd edn (New York, Foundation Press, 1988) 679, 793, 947, 1333 and 1672. 7 

Introduction

 5

however, comparable, if not identical. The chapters of this book start from them. Moreover, it is remarkable that, notwithstanding different styles of legislation, legal solutions to problems are, if not identical, at least very similar.11 The growing convergence with European law supports this convergence, as do the requirements of the subject matter. The latter is particularly true for tax, social security, building, environmental and food legislation. On the national, supranational and international levels, one would find well-known ‘clubs of experts’ on any given subject, who, of course, intensively influence legislation. The transition from harmonisation of law in contents and style to unification is inevitable and desirable. Constitutional provisions allow for the streaming in of supra- and international law into national law, and even transfer parts of legislative sovereignty onto the transnational level. One important goal of the emergence of ‘open states’ is to foster peaceful cooperation. The EU itself is a communauté de droit, with legal principles, statutory law, rules, standards and best practices—together, ‘general principles common to the law of the Member States’.12 Probably the most important factors of this process are the court rulings of the European Court of Justice and the European Court of Human Rights. Other important factors are legal literature and teaching. Similarly, international standards, rules and best practices emerge. Issues of health protection, trafficking in human beings and drugs, and new techniques (information and communications technology (ICT), biotechnology) require political and legal cooperation. It is difficult to manage this new situation with case law, so judge-made laws are falling increasingly into disfavour in the body of statutory legislation. Universal rules are being developed in contract law, social security law and labour law. Legisprudence has to follow, because its basis is independent of particularities of legal and legislative cultures. Legisprudence, as inspired by comparative law, is at stake. It has to develop concepts of qualified legislation and insights into knowledge and capabilities which improve quality of drafting law everywhere.13 To do so, comparative legisprudence has to deal with the problems of (i) ‘who’ legislates, (ii) ‘on what’ and (iii) ‘how’ laws are produced.

1.3.  Process of Legislation 1.3.1.  Competences, Organisation and Procedure of Legislation Whether in a multi-layer system or in the frame of a national state constitution, the first question must be ‘Who is the legislator?’ Three principles of organisation and procedure have to be taken into account. First, there is the vertical separation of powers. In some states, legislative competences are left to the centre and subnational units, such as the states in the USA, the Länder in Germany, the cantons in Switzerland, the regions in Spain and other autonomous bodies. Similarly, on the vertical axis there are three layers of legislation: national, supranational and international. Secondly, on each level there is a horizontal 11  K Zweigert and H Kötz, Einführung in die Rechtsvergleichung auf dem Gebiete des Privatrechts, 3rd edn (Tübingen, Mohr Siebeck, 1996) 38. 12  Art 340(3) TFEU. 13  H Xanthaki, ‘Duncan Berry, a Visionary of Training in Legislative Drafting’ (February 2011) Special Issue 9 The Loophole, Journal of the Commonwealth Association of Legislative Counsel 18.

6 

Ulrich Karpen

separation of powers. On the national level, all three branches of government are actors in legislation: legislative, executive and even the constitutional judiciary. On the supranational (European) and international levels, the separation of powers has not been fully developed yet. In the EU, democratic progress has been accomplished by increased participation of the European Parliament in legislation. The ‘legislators’ are the European Council, the European Commission and the European Parliament; the executive are the European Commission and the judiciary (eg the European Court of Justice). Instead of ‘separation of powers’, one may—as Härtel does14—speak of an ‘institutional balance’ of the organs. The same lack of fully developed horizontal separation of powers is even more true on the international level,15 and its role as the central function of government is characterised by cooperation between all its branches and units. Cooperation is obviously needed in the horizontal direction, such as in the form of ‘cooperative federalism’. It is, therefore, legitimate to speak of a network of decision-making processes in legislation. Three problems of that dense network are currently being intensively discussed in comparative legisprudence: the shift of competences upwards in multi-layer systems; the loss of participation of parliaments due to the dominance of the executive; and the lack of transparency and deficits of democratic participation (people, stakeholders) in the legislative process.

1.3.2.  Legislation in Multi-level Systems Public tasks are accomplished primarily on the national level. According to national constitutions, competences are split and partly attributed to central as well as deconcentrated or decentralised lower levels. In unitarian, although deconcentrated, states, like the ­Netherlands, Sweden and Hungary, all legislation derives from the highest level, in contrast to administration. Federal systems are pluriform. Austrian federalism is weak, German federalism much stronger. In Switzerland, the subsidiarity principle is taken very seriously; in the USA, the state exists in very different and vivid subnational constitutional spaces. Hence the distribution of legislative competences varies quite widely. A supremacy of federal law (like in Germany) does not exist in every federation. The same is true for a general assumption for decentralised units, as in the EU: ‘Competences not conferred upon the Union in these Treaties remain with the Member States.’16 There are adversary trends in distribution of legislative competences: unitarisation and decentralisation. In all vertically structured systems, socio-cultural unity in a shrinking world is a predominant factor. On the other hand, democratic participation can be managed more easily in regional units, which citizens can oversee. Cooperative federalism, implied powers of the centre, interstate compacts and unified legislation of units (‘diversity in unity’) characterises the first trend. The second trend—decentralisation— may be observed in the emergence of the regional state (Italy), the union of autonomies (Spain) and the devolution of tasks (UK).17 Beyond all these structures in centralised as 14 

J Härtel, Handbuch Europäische Rechtsetzung (Berlin, Springer, 2006) 12. Möllers, Gewaltengliederung, Legitimation und Dogmatik im Nationalen und Internationalen Vergleich (Tübingen, Mohr, 2005) 253. 16  Art 4(1) EU Treaty. 17  See Scotland Act 1998 and Government of Wales Act 1998, as well as the reorganisation of the autonomy of Northern Ireland in 1988. 15 C

Introduction

 7

well as decentralised states is territorial autonomy of municipalities, counties or other local entities. The European Union as the supreme legislator above those of Member States is a sort of ‘state in being’, whether a supranational federation or however else one describes it. EU legislation is increasingly acquiring the status of constitutional law, mainly by way of the decisions of the two European courts. Legislative competences and other acts are laid down in Articles 288ff TFEU.18 European Parliament and Council are entitled to authorise the Commission to legislate, but only under strict preconditions of delegation. It is also true, however, that the centre of gravity in legislation has shifted more and more towards the executive. This applies to the European Commission as well as to national governments. The transfer of EU law to Member States generally requires a parliamentary law of the given state. In the UK and Portugal, a piece of delegated legislation might be sufficient. The European Court of Justice holds that EU law takes precedence over the law of the Member States. However, this opinion is not shared by the majority of the Member States’ constitutional courts. Nevertheless, there is a European constitutional and administrative space already. In addition, the ever-increasing internationalisation of legislation is taking place by virtue of coordination, consultation and cooperation. There are gateways for opening national and supranational law to international legal orders, like the United Nations, the International Labour Organization (ILO) and the World Trade Organization (WTO). These organisations, in general, do not produce binding law, as statutes, but rather ‘soft law’. The ILO, for example, releases conventions and recommendations. They are norms of a general character which set standards and international customary law. Principles of due process are undoubtedly applicable for the functions of these organisations. Very often they have general conferences, which are parliament-like, and further conferences of ministers (executive) and decision-making bodies of a court-like character (like the Panels and Appellate Bodies of the WTO) to solve legal disputes. Here one can observe the evolution of constitutional state structures on the international level, which allows for a comparison with the historical development of the EU as a supranational entity.

1.3.3.  Organisation of Legislation and Separation of Powers Legislation is primarily a responsibility of parliament. The initiative for law making rests in most countries with the government. Parliament adopts statutory laws (and international treaties) and approves the budget. The principle of statutory reservation (réserve de loi) requires parliamentary legislation for any administrative encroachment on human rights. Some countries have organic laws, which rank in between constitutions and statutory law. Sub-legal regulation in most countries requires authorisation by statutory law. Initiatives may originate from the floor of the house. Every parliament has structures for division of work (eg committees, caucuses, parliamentary services). Parliaments in many countries consist of a single chamber; in others, there is a second one. Some countries (like Denmark, Sweden and Iceland) have abolished the bicameral system for reasons of transparent democratic legislation and better functional efficiency. Second chambers are

18 

See Chapter 12 below.

8 

Ulrich Karpen

often federal/regional representations; otherwise they represent professional, social or economic groups of society, or are elements of a corporate state (as is the Irish Senate). There are symmetric and asymmetric bicameral systems, depending on whether both chambers participate equally in the legislative process or whether the second has only suspensive veto rights. Bicameral systems may have advantages or may be dispensable. Regional representations are certainly an instrument for avoiding the majorisation of small versus larger units. They often exercise intra-organ control of legislation. They can broaden the rationality of deliberations and the quality of decisions. Other voices say that two chambers cause a timeconsuming duplication of debate and can be misused to delay important projects without improving their quality. In countries with direct-democratic elements, the people participate in adopting constitutional and statutory laws, and to some extent may even initiate legislation: for example, Switzerland has the strongest form of referenda. Direct democracy is an addition to parliamentary action. However, without a parliamentary basis and support, a referendum democracy cannot function. To initiate a facultative referendum, in addition to obligatory ones, is a common ‘threat’ used in political debates. Nevertheless, referenda are increasingly appreciated as an innovative procedure for fostering the participation of citizens in the decision-making process. Government, within its remit of setting guidelines for policy, initiates most legislative drafts. It requires laws (and a budget) for running the country. Some constitutions have a catalogue of government functions, others do not, but undoubtedly planning and coordination of state functions includes directing legislation. Delegated legislation by government by far outnumbers statutory law. It deals with details of law, as enacted. Laws are mostly drafted in ministries and approved by cabinet. In the UK, the Office of Parliamentary Counsel in the Prime Minister’s office realises the government’s programme in drafts.19 In Sweden, the main work on the texts is done by Royal Commissions. For a long time, Switzerland discussed the establishment of a Central Draft Service, but finally abolished the idea due to the egoism of resort ministries.20 Courts and other state organs may have a ‘negative legislative competence’ in the sense that they are entitled to declare a law void or to refuse the application of a given law in a single case. This is a ‘juridification of legislation’ (juridification progressive des règles de methode législative).21 Juridicial review of laws is a tool for protecting the constitution, a kerbstone of precedence of the constitution as a basic law. As such, it is one of the most important elements of modern constitutionalism. In the case of concentrated review of laws, it is exclusively the highest (constitutional) court that is entitled to declare laws void as conflicting with the constitution. If supervision of laws is diffuse, any court in any given case may refuse the application of a piece of law. There are models of preventive abstract review, like in Sweden or in Germany, or case-linked concrete review. Since Marbury v Madison,22 the Supreme Court of the United States claims authority of concrete review of law. The same is true for the European Court of Justice and the European Court of Human Rights. In Switzerland, review of statutory legislation is explicitly excluded, in full respect of the instruments of law review by the people (referenda). 19 J Bates, ‘United Kingdom’ in U Karpen (ed), Legislation in European Countries (Baden-Baden, Nomos, 1996) 542. 20  G Müller, Elemente einer Rechtssetzungslehre, 2nd ed (Zurich, Schulthess, 2006) 227. 21  CA Morand (ed), Légistique Formelle et Materielle, Formal and Material Legistics (Aix-en-Provence, Presses Universitaires d’Aix-Marseilles, 1999), 39. 22  5 US 137 (1803).

Introduction

 9

1.3.4.  Legislation in the Regulatory Cycle Laws are planned, drafted, adopted, implemented and (if necessary) amended in a multistep procedure, in which all constitutional organs participate.23 The main steps are: impulse, analysis of the social problem, policy-setting, definition of targets and instruments of regulation, drafting, initiating, deliberation and adoption of the draft in parliament, implementation, monitoring of implementation, amendment (if necessary) in a new procedure. The regulatory cycle includes, in short, four phases: (i) drafting and initiating; (ii) deliberation and adoption; (iii) implementation and enforcement control; and (iv) amendment. The basic structures of the legislative process are part of the constitution. Details are regulated in rules of procedure of parliament and government. Governments and (constitutional) courts have established directives for good legislation: to check the facts in detail, to make a sound prognosis, to balance the interests at stake, to carefully monitor the impact of regulation, to induce amendments (if required). This catalogue is applicable in every modern legislation, including European legislation. Impulses for legislation in most states originate primarily in government, in parliament or in parliamentary committees, or may be given by citizens, regional representatives, governmental commissions, enquêtes, the media, etc. Drafting takes place in line ministries—in the UK in the Office of Parliamentary Counsel, in France often in the Sécrétariat Général du Gouvernement. The initiative for the next step—the parliamentary procedure—is taken by the government or parliament itself. In the EU, the Commission has the monopoly for initiatives in the Council and Parliament. Deliberations and adoption of the drafts are the responsibility of the Parliament. In general, between the first, second and third readings in the plenary, detailed work takes place in committees. There may also be hearings. According to the constitution of the country, the draft may be transferred to the second chamber (if there is one). In some states, if the decisions of both chambers diverge, a joint committee is set up to reach a compromise. A referendum may be mandatory or optional. The draft is then signed, in general by the head of state, promulgated and published. Effective implementation is the predominant quality criterion of every law. Execution may require secondary legislation, administrative directives, administrative acts or plans. Transformation and implementation of EU law is, in general, an institutional and procedural matter for the Member States, which distribute competences to levels, organs and institutions according to their own constitutions. As a ‘watchdog of the treaties’, the Commission is responsible for ensuring the Union-wide effective implementation of European regulations. It is primarily the Parliament itself which monitors the effective implementation of laws. It keeps its ‘legal production’ under control by obliging governments to regularly report on the effects of the law, by ‘sunset legislation’ or by regular consolidation of law. In some countries, including the Netherlands and Germany, an independent body undertakes an evaluation, a Regulatory Impact Assessment,24 which analyses, for example, bureaucratic burdens and environmental costs. Finally, there is the supervision of laws by their very application: by courts, administrative bodies and lawyers, as well as by law schools which deal with them in theory and practice. The same is true for offices of ombudsmen.

23  24 

See Chapters 3–11. See Chapter 10.

10 

Ulrich Karpen

In view of the rule of law and legislation, there are two main challenges for the legislative process in addition to the obvious deficiencies of representative government: transparency and participation. Modern ICT facilitates the distribution of facts and knowledge; the possibilities are, however, only partly exhausted. Procedures for preparing and adopting laws are often accelerated. The language and form of the final product may be poor and indigestible; in other words, intransparent. Examples are national security acts after 9/11 or financial packages ‘to save the euro’ (the European Stability Mechanism) or ‘new energy’ legislation after Fukushima. Better participation should help to avoid that sort of legislation and bring in the expertise and opinions of legal academics, legal practitioners and ‘intermediate agents’, like interest groups (eg autonomous bodies of local self-government, the media and other stakeholders).25

1.4.  Policy Making, Targets and Instruments of Legislation, Evaluation 1.4.1.  Policy Making According to the questions of legisprudence (‘who’, ‘what’, ‘how’), the focus is now on ‘why’ legislate. This section deals with the purposes of a legal project, the programme (policy), the intent (objectives, legislative targets, goals) of a law and the instruments used to realise it. The first step in preparing regulation must be a political directive of the government (policy). The boundaries of the project should be determined. The drafter requires the main lines to be drawn in order to understand what is the intended legal situation. It is vital that the legal project fits into the governmental programme. A political directive is a reaction to a problem—impulse, a thorough analysis of a deficit, wherever it originates. To put it simply: ‘What is the matter? Why is governmental action needed? Where is the problem?’26 Diagnosis and corresponding policy-setting should follow a ‘policy checklist’: What is the proper reaction to the problem? Is the best solution to enact a law? Who are the stakeholders? Which sociological, economic or cultural factors have to be taken into account? On the basis of this analysis, the intended legal situation must be framed: What are its main factors? Which category of law: public, civil, penal? Which effects are to be avoided? Policy setting has to take into account challenges, chances and limits of political action in a wider sense: governmental guidelines, consent and compromises, elections, the budget, etc. In some countries, special units in the governmental centre are in charge of the policy development, which, as an essential first step, should be clearly distinguished from the subsequent ones, namely the choice of targets and instruments of a draft. Normative directives of the constitution may have to be obeyed: the social state and proportionality principles, necessities of economy, ecology, education and research (as far as they are matters of government), support of disadvantaged groups of society, etc. In particular, a feeling 25  C Bergeal, Rédiger un Texte Normatif, 6th edn (Paris, Berger-Levrault, 2008) 206; the Dutch Woestinwet (‘Law of the Desert’) of 1995. 26 Gesetzgebungsleitfaden, Leitfaden für die Ausarbeitung von Erlassen des Bundes, 3rd edn (Bern, Bundesamt für Justiz, 2007) 113.

Introduction

 11

for the ‘manageables’ is required. As K von Beyme puts it: ‘Policy-making is not a topic [only]for lawyers.’27

1.4.2.  Targets of Legislation Within the established policy frame, the possibilities of regulation have to be considered. This means (if a law is indispensable) setting the targets for drafting. It is necessary to choose an option for tackling the problem: cause- or system-oriented, punctual or comprehensive, preventive or repressive? The best level of regulation requires consideration: statutory law or sub-legal regulation? A general regulation or a very detailed one? Decisions have to be taken on the concept of the law, the contents of provisions, the addressees, whether the law should be applicable directly or only after the release of sub-legal acts, conclusion of contracts, etc.28 As to the content of regulation: command or inhibition (with sanctions)? Authorisation or suggestion? The ‘goal’ of the law is often mentioned in the preamble, mostly in the arguments or even in the form of a directive norm in the text itself.

1.4.3.  Instruments of Legislation The legislator often prefers causal steering of the economy, society and the actions of administrative agencies. In the face of scepticism about ‘planning and steering’, the government may choose to use directives. In this case, the law is binding as to the results to be achieved, but leaves room as to the choice of measures, forms and methods the authorities can use to implement the law.29 Other instruments of legislation may be of a benefiting, stimulating, prohibiting or even punishing nature (‘carrots and sticks’). Governments may cooperate with stakeholders (co-regulation) or leave regulation to them completely (self-regulation). Once a choice has been made, agreements are drawn up and goals and limits are set or licences and certificates edited. There are forms of symbolic legislation which strive to have an educational effect: hints and indications, information, warnings, ‘shop-window laws’ (eg concerning the protection of the environment, waste morals). Such provisions may also be found in the constitution. This is a transition field to soft law, which contains non-regulatory measures, like standards or principles distinguishing quality or origin marks, and tokens or codes of conduct which originate from nongovernmental actions.

1.4.4. Evaluation Legisprudence is interested in policy making, the decision-making process for intent and instruments in a procedural sense, but also in developing value criteria for effective and efficient legislation; in sum, ‘good legislation’ (substance, merits, substantial due ­process).30 27 

Der Gesetzgeber (Opladen,Westdeutscher Verlag, 1997) 73ff. U Karpen, ‘Instructions for Law Drafting’ (2008) 10 European Journal of Law Reform 163, 213–14. 29  Eg, art 288(3) TFEU. 30  See Chapter 14. 28 

12 

Ulrich Karpen

This requires the consideration of standards and an evaluation of quality.31 The constitution of a rule-of-law democracy includes material and procedural principles for legislation.32 The contents of laws must be oriented towards public welfare. Legisprudence should apply constitutional values, like social, ecological and cultural state targets. Subsidiarity and proportionality are to be obeyed at all steps when approaching legal goals, formulating contents or choosing instruments for implementation. The policy of legislation follows the political programmes and necessities of the day. Goals and instruments are evaluated against three criteria: conformity with the constitution and law; effectiveness; and trust.33 Every piece of law must first obey the organic constitution and international law, as well as the dogmatic constitution (structural principles and human rights). The main constitutional values of national constitutions, namely democracy and rule of law, are in conformity with European standards.34 Secondly, important criteria of a qualified law are efficacy, effectiveness and efficiency (‘the three Es’). The law should meet the legislator’s intent, should be applied and obeyed by as many addressees as possible, and should stand the rationality test, as it has a positive cost–result (input/output) relation. The three Es are facets of proportionality, which is an element of rule of law. The effectiveness of law requires stability, limited quantity and transparency, reduction of bureaucracy and deregulation. Thirdly, effective regulation is built on trust. Trust is an important resource of stability which combats frequent legislation to amend, adjust and repair law as enacted. Trust contrasts a ‘motorised legislator’. Undoubtedly, the reduction of ‘legalisation’ in all European states is a mandate of the time. Combating the deluge of laws is an aim of ­debureaucratisation.35 Instruments of New Public Management, devolution, decentralisation, co-regulation—all sorts of ­participation—contribute to that end. To improve the quality (ie the rationality) of legislation, new methods of RIA have been introduced. They assess both the intended and unintended impacts of the law. These are financial, bureaucratic and other effects, originating in the economic or social sectors, or in the citizens’ sphere. RIA of a draft law is generally done prior to its adoption (prospective, ex ante), concurrently with the process of parliamentary deliberations in committees and the plenary, or after implementation (retrospective, ex post). The latter form of RIA is required for monitoring and possibly amending purposes. The former is done by those institutions which initiate the draft or by independent authorities. Concurrent RIA is the responsibility of all parliamentary organs, including scientific parliamentary staff. Ex post RIA is a matter for everybody who has to implement (administration, businesses), obey (citizens, businesses) or otherwise handle the law (courts, teaching and research). Deficits of legislation as well as high costs are reported by everybody who applies and the media. There is no uniform method of RIA. For decades, prognoses, simulations, expert interviews, experimental legislation, report duties of government and sunset legislation were in use. Following the Dutch model, the Standard Cost Model was later employed in most countries. It focuses on administrative burdens. This narrow perspective on information neglected checking and evaluating costs of other legal obligations, although these outweigh 31 

P Popelier, De Wet: Juridischbekeken (Brugge, die Keure, 2004) 23ff. European Commission, Better Regulation: Simply Explained (Brussels, Office of Official Publications, 2006). 33  See Chapter 10. 34  See arts 2 and 6 TFEU. 35 OECD, Cutting Red Tape. Why is Administrative Simplification so Complicated? Looking Beyond 2010 (Paris, OECD Publishing, 2010)15. 32 

Introduction

 13

the costs of the executive. These costs are better caught by the Net Administrative Cost Model. Within this model, RIA of several aspects of implementing the law is united in a single procedure. This leads to greater transparency of the economic, ecological and social effects of new legislation.

1.5. Techniques 1.5.1.  Law Drafting Finally, legisprudence deals with issues of ‘how’ to draft norms. In a formal sense, the law should be as simple as possible and formulated in a clear, plain language with a coherent structure. The questions of formal quality involve the style, structure, language, use of references, general clauses, etc. The drafting technique, however, not only has a formal character. It refers to the content of the draft, shapes it and is influenced by it. Form and content are in a dialectic tension. Method, procedure and technique of norm drafting are linked. Systematics, form of regulation, wording, publication and other technicalities are dependent on content. Choice of language depends on addressees, procedure on form and content. An optional connection of these elements strengthens the effectiveness of the norm and renders a law ‘good’. I am not being entirely frivolous when I remind You that Moses was an excellent draftsman. When he said, ‘Thou shall not steal’, everybody knew what he meant. But we have the Theft Act 1978 with 36 sections, which have given rise to many problems of interpretation, which have had … considered many times by the Court of Appeal.36

Clarity, precision, avoidance of contradictions, readability and comprehensibility are constitutional, ie rule-of-law, directives. Applying these rules for linguistically and technically good norms increases trust in legislation as a precondition for acceptance and implementation of the law. However, in reality, legislative standards of good legislation are often neglected. This is a standing complaint in all countries, and applies to European legislation as well. Such legislation is not transparent, nor is it ‘close to the citizens’. Instead, it is often superfluous and irritating. The bewildering amount of detail is often a result of the lack of consent of Member States. Some only agree to a regulation when the peculiarities of their national legal system is taken into account by special regulations or reservations of that particular piece of law. Some improvements in transparency are expected by electronic publication, but this seems to be insufficient.

1.5.2.  Structure, Language, References Good disposition and structure of a draft and transparent systematics are the prerequisites for comprehension. Systematics foster the unity of the legal order. Old laws, by virtue of 36 Lord Renton, The Evolution of Modern State Law and Its Future, Inaugural Statute Law Society Lecture ­(London, University College, 1995) 8.

14 

Ulrich Karpen

many amendments, may sufficiently regulate on the matter, but readability may degenerate into an intransparent and incomprehensible patchwork. All countries want to strengthen legislation by the consolidation and codification of laws. However, urgent daily political problems do not favour good legislation; rather, the result is often hectic, superfluous and too detailed. The language of norms is an inexhaustible topic. ‘Whatever is well conceived is clearly said. And the words to say it flow with ease.’37 The language must be oriented towards the function and addressees of the law. Traffic, criminal and contract law must be understandable to everybody. Provisions on limits of air pollution, noise, obnoxious materials and protection against ionising radiation are made for experts. Texts of this kind are supplemented by matrixes, graphs, formulas and plans. Some countries, such as Canada and Switzerland, must legislate in more than one language. The EU is a multilingual supranational entity of states; and the battle on gender-neutral language goes on: ‘chairman–chairwoman–chair’, ‘he–she–one’. All draftsmen use drafting techniques. Among them are legal definitions, references and repetitions. Definitions are ambivalent. They could be lighthouses in the ocean of norms, but provide extremely fine details as well. The latter is true for EU legislation, since 28 separate legal systems must be brought together. References are an instrument of integration of legal order and the insertion of new regulations into the existing body of law. Other drafting techniques are formula compromises, legal assumptions and fictions. As far as transparent publication of laws is concerned, sometimes ‘the diligence of an archive keeper’ is required to trace them, and there are pieces of law which can only be found ‘with subtle expert knowledge, extraordinary methodological expertise and some pleasure to solve brainteasers’.38

1.5.3.  Special Categories of Laws Due to the torrent of laws and their hectic sequence, amendments are of particular interest. To avoid intransparency, they have to be drafted with special care, and should be detached, over time, by consolidation of the whole law. The drafting literature covers details on ‘legislative packages’, time-limited and experimental laws, symbolic and ‘shop-window’ legislation, authorisation legislation (for delegated legislation), budget laws, ‘coat laws’, and laws which transform international and European law into national law. Laws on measures in individual cases (ad hoc laws) contradict many constitutional principles, which are based on the rule of law:39 general application of law, equality, trust, proportionality, separation of powers, fair procedure, access to courts and others. Measurement laws and exceptional provisions need to be avoided. In view of the intransparency of huge numbers of laws, special legislation is used in most countries to abolish (outdated) laws or republish them in a purified form, or to consolidate large areas of law. These techniques to keep the body of law up to date, smooth and transparent have been made easier by applying IaC-technology.

37 

Nicolais Boileau-Déspereaux, The Art of Poetry (1674) Canto I, I 153, German edn (Stuttgart, Reclam, 1967). Court Decision, E Wiederin, in A von Bogdandy et al (eds), Handbuch Ius Publicum Europaeum vol 1 (Heidelberg, CF Mueller, 2007) 432, no 174. 39  A Jasiak, Constitutional Constraints on Ad Hoc Legislation (Nijmegen, Wolf Legal Publishers, 2010) 314. 38  Austrian

Introduction

 15

1.5.4.  Guidelines and Manuals for Law Making To coordinate legislation and to assist drafters, most governments have implemented legislative guidelines.40 Many of them are based on the same principles, and include similar examples of good or bad legislation. These guidelines are conventions, or rules of best practice, which could be addressed as ‘loi des lois’. Codified law in common law countries is often uses based on judge-made interpretation. The judge adheres to precedents and principles as derived from litigation. Only in reasoning from case to case are they developed. Broad goal-oriented or historical interpretation is rarely attempted. The legislator follows this type of application: ‘le rédacteur doît parler la langue de l’interprêt’. He drafts statutes that are fine-meshed and very detailed, in order to ensure strict interpretation along the lines of the wording. A legitimate reference to common law is almost impossible. By this ‘formal method’ of drafting—in fact, cases are ‘wrapped up’ in statutes—the legislator intends to bind the judge as closely as possible to the law. By additional interpretation guidelines,41 how a law should be understood is prescribed exactly. This long-term teamwork between the legislator and the interpreter (the judge) continues until the law changes. Some countries have gone a step further by adopting a law on legislation.42 Unanimity in the principles of drafting—whether practiced throughout or not—has developed so far that it might well be possible to mandate a group of experts to gather them together under the umbrella of a model of better legislation. But even if applied, the legislator will not always succeed in drafting first-class bills. There remain differences between practice and rules, even when the latter are taken from the former.

1.6.  Some Trends of Legislation and Legisprudence From a comparative law perspective, four trends of legislation in modern states may be noted. First, legislation and its problems as analysed and described by legisprudence are similar but not the same in constitutional states. Astonishment over differences is the engine of comparative law. Progress of harmonisation and unification of legal procedures, content and form of laws will proceed. Europeanisation and globalisation of legislation provide a common basis by opening up constitutions to the world. European and national constitutional states can be understood as a ‘complimentary constitution’ of a ‘constitutional compound’, not yet a confederation or even a federation. European countries and others constitute a constitutional family, in which constitutions enrich each other. Secondly, legislation is a matter for parliament, which is the centre of power in a democratic state. However, its role may change. Parliament is under permanent political pressure to guarantee the stability and flexibility of laws at the same time. Effectiveness of norms 40  Eg, JP Duprat, ‘The Recent Evolution in French Legislative Practice: A Managerial Conception of Statutes’ (2009) 50 Legislação: Cadernos de Ciência de Legislação 261; W Robinson, ‘Drafting of EU Acts: A View from the European Commission’ in C Stefanou and H Xanthaki (eds), Drafting Legislation, A Modern Approach (Bodmin, Ashgate, 2008) 177. 41  Loi concernant l’interprétation des lois et des règlements (Canadian Ministry of Justice, 1985); cf Morand (n 21) 153, note 9. 42  Like Bulgaria, Law on Normative Acts, State Gaz 27/1979, as amended SDtate Gaz 65/1995.

16 

Ulrich Karpen

is the primary goal—quantity of law production versus deregulation, the lack of quality, insufficient transparency, increasing participation, RIA, governmentalisation on the one hand, and regional and federal as well as autonomous legislative competences (devolution) on the other. Thirdly, the juridification of legislation is proceeding. The judge is part of the legislative cycle. He ultimately measures procedures, targets, instruments and forms of legislation against the constitution. National courts, the European Court of Justice and the European Court of Human Rights are gaining increasing directive influence on national as well as European legislation: ‘Gulliver enchaîné’.43 Finally, legisprudence should be aware of its limitations. Legislation should be as good, precise, effective, efficient and rational as possible, but it will never be mathematics. As John Dickinson said on 13 August 1787 in the Constitutional Assembly of the United States of America in Philadelphia: ‘The life of the Law has not been logic: it has been experience.’44

Further Reading Benedetto, M De, Martelli, M and Rangone, N, La qualità delle regole (Bologna, Il Mulino, 2011). Bergeal, C, Rédiger un Texte Normatif: Loi, Décret, Arrêt, Circulaire, 6th edn (Paris, BergerLevrault, 2008). Bundesministerium der Justiz (ed), Handbuch der Rechtsförmlichkeit, Empfehlungen zur Gestaltung von Gesetzen und Rechtsverordnungen, 3rd edn (Cologne, Bundesanzeiger Verlag, 2008). Böhret, C and Konzendorf, G, Handbuch Gesetzesfolgenabschätzung (GFA): Gesetze, Verordnungen, Verwaltungsvorschriften (Baden-Baden, Nomos, 2001). Dale, Sir William, Legislative Drafting: A New Approach (London, Butterworths, 1977). Bundesamt für Justiz (ed), Gesetzgebungsleitfaden, Leitfaden für die Ausarbeitung von ­Erlassen des Bundes, 3rd edn (Bern, Bundesamt für Justiz, 2007). Karpen, U, ‘Comparative Law: Perspectives of Legislation’ (2012) 6 Legisprudence 149. Karpen, U, Gesetzgebungslehre—neu evaluiert/Legistics—Freshly Evaluated, 2nd edn (BadenBaden, Nomos, 2008). Mader, L, ‘Législation et légisprudence’ in J-L Chappelet (ed), Contributions à l’action publique (Bern, Haupt, 2006). Mader, L and Tavares de Almeida, M (eds), Quality of Legislation: Principles and Instruments (Baden-Baden, Nomos, 2011). Müller, G, Elemente einer Rechtssetzungslehre, 2nd edn (Zurich, Schulthess, 2006). Noll, P, Gesetzgebungslehre (Reinbek, Rowohlt, 1973). Schneider, H, Gesetzgebung, 3rd edn (Heidelberg, Müller, CF, 2002). Schuppert, G, Governance und Rechtsetzung (Baden-Baden, Nomos, 2011).

43 

Morand (n 21) 43. Century of Law Making for a New Nation: US Congressional Docments and Debates, 1774–1875, Farrand’s Records, vol 2 (1991), 278. 44 

2 Legislation and Regulation WIM JM VOERMANS1

‘Therefore, every rational being must so act as if he were through his maxim always a legislating member in the universal kingdom of ends.’ Immanuel Kant, Grounding for the Metaphysics of ­Morals (1785) In most civil law countries, legislation is the single most important medium through which law is expressed. In order to understand the concept of legislation, and the different meanings and uses it has nowadays, one needs to understand the functions legislation performs in modern states. This understanding is key to targeted and effective use of legislation in practice. To be able to use recent academic insights to full effect in legislative practice, an understanding of the difference between the concept of legislation and regulation is important. By ‘legislation’ we mean the authoritative and constitutionally controlled form in which law is cast and the procedure leading up to its enactment (the decision). By ‘regulation’ we mean a public intervention in a market or in society. Regulation is an instrument of governance (not governance itself). Legislation and regulation coincide in a lot of instances. A lot of regulation is cast in the form of legislation. By this regulation acquires the power of law, legitimacy and other effects. But not all regulation needs to be cast in the form of legislation (think of so-called policy rules), and not all legislation is regulation (eg legislation establishing an advisory body for the government). Legislation and regulation have become ever more important as means of governance and vehicles of rights, entitlements and legitimacy—we live our lives in regulatory states. The quality of legislation and regulation is therefore of the utmost importance. The chapter argues that it is important to take due notice of the difference between legislation and regulation, and the different functions they perform, if one wants to put policies and instruments in place to improve the overall quality of legislation or regulation, as the case may be.

2.1.  Legislation and Regulation2 In most civil law countries, legislation is the single most important medium through which law is expressed. In order to understand the concept of legislation, and the different meanings and uses it has nowadays, one needs to understand the functions legislation performs 1  Prof Dr Wim Voermans is professor of Constitutional and Administrative Law at Leiden University. He is Director of the Institute of Public Law at Leiden Law School, one of the directors of the Profile Area Political Legitimacy of Leiden University and the acting President of the International Association for Legislation. 2  Parts of this chapter have been lifted from W Voermans, ‘Concern about the Quality of EU Legislation: What Kind of Problem, By What Kind of Standards?’ (2009) 2(1) Erasmus Law Review 59.

18 

Wim JM Voermans

in modern states. Legislation comes with different meanings in different contexts at different times. In antiquity, for instance, the notion of legislation had a total different connotation than it has in modern history,3 and even in modern history the meaning of the concept varies widely. Where legislation in post-revolutionary continental Western Europe of the nineteenth century was predominantly perceived as an benign instrument of codification through which hitherto scattered and inaccessible common law could be systematised and made accessible for everyone, nowadays we predominantly use it as an instrument for modification—ie engineering society by setting out new law.4 Law expressed by legislation, enacted by the people’s representatives in parliaments, is nowadays often used as an instrument to effect societal, economic and institutional change; it gives governmental interventions both the punch of the law and the legitimacy of its democratic underpinning. We live, according to Seeley, in a legislation state,5 which is not the same as a Rechtsstaat (a state governed by the rule of law), but rather a form of state devoted to the business of making continual improvements in the life of the community by means of explicit legal innovations, ie by parliamentary legislation.6 This ‘interventionist’ element of law wrapped in the form of legislation (which as a concept refers to procedure—the nominalisation of the verb is no coincidence) makes today’s professionals and academics inclined to equate the concept of (modern-day) legislation with that of regulation. Regulation, like legislation, bears with it the notion of general measures or rules, stemming from government, aimed at institutional or behavioural change or control. This can, however, lead to confusion. The notions of regulation and legislation have different roots and also different meanings, whatever their superficial overlaps. The concept of ‘legislation’ is predominantly a legal concept drawing on the scholarship of law, whereas ‘regulation’ is a concept stemming from and rooted in economic scholarship as well as (more recently) in the social sciences.7 Hood and Scott, for instance, point out that: the term ‘regulation’ is conventionally used to refer to government intervention in markets. The term usually denotes a form of intervention that consists in setting and enforcing rules of behaviour for organizations and individuals. It thus contrasts with other forms of state intervention such as public ownership, taxes and subsidies or physical alteration of the environment.8

The focus of and underlying notion of regulation therefore is on government intervention in markets, ie on acts private actors cannot perform with private capital, on interventions beyond regular market mechanisms (eg of supply and demand) and (most of the time) below market prices—in short, what interventions do and can do in markets and societies living in them. Legislation, on the other hand, focuses primarily not on markets but on the human psyche, especially morale and social relations: the ‘oughts’ of our existence. 3 

See J Harries, Law and Empire in Late Antiquity (Cambridge, Cambridge University Press, 1999) esp ch 1, 1–35. T Koopmans, ‘De rol van de wetgever’ in HCF Schoordijk et al (eds), Honderd jaar rechtsleven: de ­Nederlandse Juristen Vereniging (Tjeenk Willink 1970). 5  JR Seeley, Introduction to Political Science: Two Series of Lectures (London, Macmillan & Co, 1896) 146. 6  See J Waldron, The Dignity of Legislation (Cambridge, Cambridge University Press, 1999) 7. 7  On the different meanings and backgrounds of the concepts of legislation and regulation, see B Morgan and K Yeung, An Introduction on Law and Regulation (Cambridge, Cambridge University Press, 2007). 8  C Hood and C Scott, ‘Regulating Government in a ‘Managerial’ Age; Towards a Cross-National Perspective’ (2000) 1 CARR Discussion Papers 1, available at www.lse.ac.uk/accounting/CARR/pdf/DPs/Disspaper1.pdf (accessed on 21 September 2015). 4 See

Legislation and Regulation

 19

2.1.1.  The Difference Between Legislation and Regulation The concept of two perspectives of one phenomenon under two slightly different names— legislation and regulation—gives rise to possible confusion. I will attempt here to come up with a more or less clear-cut definition. By ‘legislation’, we mean the authoritative and constitutionally controlled form in which law is cast and the procedure leading up to the enactment of it (the decision). By ‘regulation’, we mean a public intervention in a market or in society.9 Regulation is an instrument of governance (not governance itself). Legislation and regulation coincide in a lot of instances. A lot of regulation is cast in the form of legislation. By this, regulation acquires the power of law, legitimacy and other effects. But not all regulation needs to be cast in the form of legislation (think of so-called policy rules), and not all legislation is regulation (eg legislation establishing an advisory body for the government). The importance of legislation in our lives cannot be underestimated. Our modern societies based on (free or otherwise) market transactions10 are to a large degree dependent on clear and acknowledged agreements that are set out in advance in regulatory acts cast in the form of legislation to generate the trust that is necessary for our kind of economic societies. Regulation is a prerequisite for economics and economic growth and, because economic growth has become so important, it is also important for our political society. What is special about the history of the last century is that we don’t just recognise the standards written down in legal regulations as a necessary condition for trustworthy transactions within markets, but we increasingly use these regulatory acts to intervene in these markets, and to steer the economic society in this way. Regulation no longer acts merely as a prerequisite, but as an independent instrument for government and administration.11 This is occurring to the extent that many Western industrialised societies have developed into what Majone refers to as a ‘regulatory state’, because of the importance the interventions by regulatory measures play in economic, political and social life.12 In the past decade, however, the realisation has grown that regulatory acts are not just a blessing for economic growth, but that an incorrect or excessive use of regulation can cause restrictions in economic growth: (outdated) regulation can unnecessarily restrict (new) economic activities; regulation can disrupt spontaneous processes; a pile up of regulation can have undesired knock-on effects (particularly when they are not or insufficiently aligned to each other); and regulation can cause such a tremendous administrative burden that economic growth is slowed down. What exact effects regulation has on economic

9  See, eg, R Baldwin, M Cave and M Lodge (eds), The Oxford Handbook of Regulation (Oxford, Oxford University Press 2010) 6ff; Morgan and Yeung (n 7). 10  Contrary to self-supporting societies, where life and existence depend on being able to provide the essentials of life by hunting or gathering, or by forms of agriculture. 11  This is actually quite remarkable because rules and regulations are a kind of (authoritative) ‘mould’ into which we pour agreements. However, rules and regulations are often identified with the agreements, or ‘interventions’, themselves. This identification goes so far that in modern social scientific literature ‘regulation’ is usually seen as any public intervention in markets and societies. See, eg, Baldwin et al (n 9) 6ff; Morgan and Yeung (n 7). 12  G Majone, Regulating Europe (London, Routledge, 1996).

20 

Wim JM Voermans

activities and economic behaviour is a field of investigation in which the tradition of academic research is still quite young. Veljanovski states: The economics of regulation … is not a settled body of facts but an approach. It uses the economist’s toolkit to develop political economy theories of regulation, and assist regulators with the technical details of framing effective regulation. It is however in its formative phases. There remain many puzzles and paradoxes to be explained such as the tremendous growth in regulation in parallel with greater role by markets and economics …13

Even though we don’t know exactly what the exact effects of (different types of) regulation are, there are a number of theories on the economics of regulation. These basically fall into two categories: normative and positive theories.

2.2.  Normative and Positive Theories of Regulation Veljanovski summarises the difference between the normative and positive approaches as ­follows.14 Normative theories of regulation, he holds, principally build on the concepts of (i) economic efficiency and (ii) market failure. This is against the backdrop of the mainstream belief—that efficient allocation of goods, resources and services can best be reached in a market-based economy—an idea which, in turn, is based on Adam Smith’s maxim that when an individual pursues his self-interest, he indirectly (and unintentionally most of the time) promotes the good of society, because self-interested competition in the free market tends to benefit society as a whole by keeping prices low, while still building in an incentive for a wide variety of goods and services.15 This is Smith’s famous ‘invisible hand’. In this vein of thought, economic efficiency refers to the situation of perfect balance, in which nothing can be improved without something else being hurt. Two sorts of efficiency are feasible. First, there is allocative efficiency if the price of a product that the market is supplying is equal to the marginal value consumers place on it, and equals marginal cost. Secondly, there is productive efficiency when units of goods are being supplied at the lowest possible average total cost. The exact distinction does not need to bother us here, but the underlying notion is important –that a free (ie perfectly competitive) market is in itself capable of reaching a situation where the greatest good for all can be attained (a Pareto efficient outcome)16 without the need of any external interference (such as government intervention in the form of regulation). A Pareto efficient outcome theoretically occurs where the welfare of one individual cannot be improved without reducing the welfare of others. Pareto efficiency refers to the situation where all parties benefit, or nobody is harmed, by a reallocation of resources, goods, assets or a change in the law.17 13  C Veljanovski, ‘Economic Approaches to Regulation’ in R Baldwin, M Cave and M Lodge (eds), The Oxford Handbook of Regulation (Oxford, Oxford University Press, 2010) 34–35. 14  See Veljanovski (ibid) 19ff. 15  A Smith, The Glasgow Edition of the Works and Correspondence of Adam Smith, vol 2a, RH Campbell and AS Skinner (eds) (Oxford, Clarendon Press, 1976) 145; 158 and—of course—also to be found in A Smith, The Wealth of Nations (1776). 16  Another form of efficiency is called the Kaldor–Hicks efficiency, in which it is not the perfect equilibrium of no-one-being-harmed that is achieved, but the next best situation, in which those that gain can in principle compensate those that have been ‘harmed’ and still be better off. 17  The no-one-harmed principle is, of course, very hard to achieve, but that does not have to rule out efficiency attainment even if some individuals are harmed; if they can be compensated, by reallocation of resources, for

Legislation and Regulation

 21

In reality, of course, free market economies do not automatically produce these benign outcomes. Market ­economies generate many inequalities and harms to parties and individuals by ‘inefficient’ allocations of goods, resources and services. According to most economic theories, this is not an inherent flaw of the concept of free, competitive markets, but due to market distortions and/or market failures. This brings us to the second concept of normative approaches to regulation (ie theories that try to explain why governments can be entitled to or ought to intervene in markets). Markets do not automatically balance out and yield Pareto (or other) efficiency effects— they are prone to failures. According to normative regulatory theory, there are four main reasons why free (and thus competitive) markets fail to produce efficient outcomes: 1. unequal accumulation of market power (in the form of monopolies, oligarchies or cartels) distorting fair and open competition; 2. externalities (costs or benefits that fall beyond the scope of the normal market costs and benefits of production processes and supply and demand relations. Externalities are ‘social’ costs, like pollution, global warming, anti-social behaviour and crime, that are not or cannot be borne by private actors—‘you use and profit, society foots the bill’); 3. public goods (like fresh air, knowledge, public infrastructure, national security, education, common language(s), widespread and high public literacy levels, potable water, flood control systems, lighthouses and street lighting. A public good is one for which consumption of the good by one individual does not detract from that of any other individual, ie there is non-rivalrous consumption and thereby a disruption of competiveness—‘the bill has already been paid, you and everyone else uses and profits’)18; and 4. imperfect information, which can result in inefficient market outcomes and choices. If, for instance, you do not know what kinds of summer are to be expected in a region in which you want to start a farm, nor the volume and sort of costumers you might expect to buy your products, it is very hard to make investment decisions that will need to yield a profit in a competitive market. A market can also fail because of asymmetric information—for instance, situations in which the buyer or seller has better information. Governments have means to level the playing field for markets when it comes to the supply of information. If governments intervene in markets, for example, to correct or prevent market failure, this is most of the time not a zero sum game. Interventions by way of regulation are costly, and affect the competiveness in markets as well. A well-known example of side-effects of government intervention is the development of quasi-property rights. For instance, when the government curbs the taxi market by requiring a licence to operate a taxi business, taxi licences may in themselves become an asset that can be traded. This happened in the EU with the different quota systems (eg dairy and manure quotas). Getting the corrective intervention (regulation) right in an open and competitive market without unbalancing

their harm or loss, efficiency might still be improved. This approach is what is called Kaldor–Hicks efficiency. A ­reallocation is a Kaldor–Hicks improvement if those that are made better off could hypothetically compensate those that are made worse off and lead to a Pareto-improving outcome. BCJ van Velthoven and PW van Wijck, Recht en efficiëntie [Law and Efficiency], 4th edn (Dordrecht, Kluwer, 2007). 18 

Veljanovski (n 13) 20.

22 

Wim JM Voermans

the market and/or frustrating Pareto-efficient outcomes is difficult and most of the time will involve use of the law. According to Coase, this drift to the law follows from the fact that: Since, by and large, people choose to perform those actions which they think will promote their interests, the way to alter their behaviour in the economic sphere is to make it in their interest to do so. The only means available to government for doing this (apart from exhortation, which is commonly ineffective) is the law or its administration.19

But it is not only market mathematics and law that need to be considered in view of regulation: principles, perspectives and politics all play a role as well. As Veljanovski puts it: Markets and regulation generate winners and losers. It follows that individuals as market participants and as citizens will often be equally if not more concerned by whether laws increase their wealth than by their impact on others or society in general. Moreover, there will often be views about the fairness and acceptability of market and regulatory outcomes. Thus it would be surprising if the economics of regulation, especially in its guise of a normative theory, was able to ignore these ‘non-economic’ factors.20

These non-economic factors and their resulting (re)distributive decisions and policies fit economic theory and models poorly; they mess up the calculus. (Efficient) market outcomes are partially determined by the ex ante distribution of income and wealth in society. This means that there is an inextricable link between wealth distribution and economic efficiency and market outcomes—and, by implication, regulation. In this respect, it is remarkable, according to Veljanovski,21 that most economists view the use of regulation to redistribute income as a blunt and inefficient instrument. In their view, it distorts prices and incentives, and hence leads to substantial efficiency losses and unintended effects which often harm those whom it is designed to benefit. Positive theories of regulation try to have an open eye for distributive principles, ex ante distributions, and the roles politics and interest groups play in decisions on wealth transfers. Positive economic theories seek to explain regulation as it is. As Stigler has put it: ‘The central tasks of the theory of economic regulation are to explain who will receive the benefits of burdens of regulation, what form regulation will take, and the effects of regulation upon the allocation of resources.’22 This handbook is not a textbook on economic theory, nor is it one on regulation proper (for those purposes, the Oxford Handbook of Regulation, edited by Robert Baldwin, Martin Cave and Martin Lodge (Oxford, Oxford University Press, 2010), is much better suited). Instead, what this book tries to do is to present legislation as the vehicle of law and legal principles, and the effects law expressed via legislation has on markets and societies. To understand this requires a basic understanding of the way governments use interventions (regulation) to correct or guide markets, and how they employ the law for that. That tells a lot about the use and meaning of modern-day legislation, but it does not tell the whole story.

19 

RH Coase, The Firm, the Market and the Law (Chicago, Chicago University Press, 1988) 27–28. Veljanovski (n 13) 23. 21 ibid. 22  GJ Stigler, ‘The Theory of Economic Regulation’ (1971) 2(1) The Bell Journal of Economics and Management Science 3, 3. 20 

Legislation and Regulation

 23

2.3.  Functions of Legislation as Yardsticks for Notions of ‘Good’ Legislation If we were to summarise the former paragraph on the function and uses of regulation, we can, with Anthony Ogus, define ‘regulation’ as ‘the necessary exercise of collective power through government in order to cure “market failures” to protect the public from such evils as monopoly behaviour, “destructive” competition, the abuse of private economic power, or the effects of externalities’.23 And for a really proper understanding, we must add to it: ‘with the law as government’s weapon of choice’. Regulation therefore predominantly functions as a market intervention aiming for a correction. Legislation, on the other hand, serves other, broader functions. This is important to keep in mind if we want to make assessments or appraisals of whether or not regulation on the one hand or legislation on the other has succeeded or met its purpose, or whether it meets the relevant requirements (in other words, is of good quality). The differences between the concepts of legislation and regulation are important if we want to distinguish their effects or requirements, and the demands set on these phenomena. If we truly want to understand the nature of legislation, there is more than mere market effects to consider. Legislation as the single most important container of law in constitutional democracies performs different important functions.

2.3.1.  Legislative Functions In constitutional democracies governed by the rule of law (such as many European continental systems), legislation provides both the basis and the framework for government action (constitutional function).24 Legislation also works as a safeguard against government action by enshrining rights and obligations that can be invoked in legal procedures while at the same time providing legal certainty (legal or legal certainty function).25 At the same time, legislation, and the law expressed by it, serves as an instrument to further government policies (instrumental function),26 fixes trade-offs between (opposing) interests 23 

A Ogus, Regulation: Legal, Form and Economic Theory (Oxford, Hart Publishing, 2004). important procedural element of the rule of law is that it requires a mandate in law for governmental action and the obligation to act in accordance with the law. 25  See P Eijlander and W Voermans, Wetgevingsleer [On Legislative Drafting] (The Hague, Boom Juridische Uitgevers, 2000) 18–19. 26  The instrumental functions of legislation in modern societies are manifold. Summers distinguishes five common techniques to give effect to government policies using the law as an instrument. To further policy or political ends, law is commonly used as: (i) a grievance remedial instrument (recognition of claims to enforceable remedies for grievances, actual or threatened); (ii) a penal instrument (prohibition, prosecution and punishment of bad conduct); (iii) an administrative-regulatory instrument (regulation of generally wholesome activities, business or otherwise); (iv) an instrument for ordering governmental/authoritative conferral of public benefits (education, welfare, economic infrastructure, etc); and (v) an instrument for facilitating and effectuating private arrangements (facilitation and protection of private voluntary arrangements, economic and otherwise). EU legislation— and especially EC legislative instruments—basically perform most of these instrumental functions, albeit that the administrative-regulatory function (eg in agriculture, health and environment legislation) and facilitating and effectuating private arrangement (eg consumer protection and telecommunication services) functions are somewhat predominant. See RS Summers, ‘The Technique Element in Law’ (1971) 59 California Law Review 733, 736. 24  An

24 

Wim JM Voermans

in political arenas (political function), provides the basis for popular participation—and so the legitimisation—in the framing of legislation (democratic function), offers the basic framework for the operation of a bureaucracy (bureaucratic function),27 and communicates and reaffirms the public morals, values and public goods (symbolic function) which structure the legislative debate and provide the authoritative aura for the overall legitimacy of legislation.28

2.3.2. Requirements and Demands on Legislation in Constitutional Democracies In order to be able to perform these functions, legislation in a constitutional state must meet a number of basic requirements. Since we are focusing on legislation as an expression of law and as a legal act, these demands and requirements will mostly originate from legal concepts and theory. Subject to the rule of law, as any other institution or agent in a constitutional state, the activity of legislating is subject to the law itself. This means that, in order to legislate, a constitutional power to legislate is a prerequisite, and legislative processes as well as legislative discretion are fixed and set out by law itself (preparation and enactment according to the due procedure, not acting contrary to higher ranking laws, and some form of accommodation of existing law). Aside from this principle of legality, the rule of law also imposes a duty on the legislator to consider in some way, or respect, the implementation and enforcement of legislation to be enacted (principle of effectiveness). The last requirement the rule of law sets upon a legislative act results from the principle of legal certainty, and is what we can call the principle of intelligibility, the principle that legislative acts need to some extent to be readable and intelligible to its addressees. Most constitutional states are not only governed by the rule of law but are in effect democratic states too. The latter feature results in separate, additional requirements for legislatures connected to the democratic and political function of legislation. Legislative authorities are subject to the duty to give reasons, the duty to consult or involve interested parties, be it directly or indirectly, and the duty to inform (transparency and accessibility) during the course of legislative processes resulting in primary legislation.29 The functions are inspired by an article by H Kelsen, ‘The Law as a Specific Social Technique; the Essence of Legal Technique’ (1941) The University of Chicago Law Review 75, 75. 27  This in fact is more or less a subsection of the ‘constitutional function’. The constitutional function is commonly believed to be made up of the constituent, institutional function (establishing institutions), the attributive function (attributing power to institutions) and the regulatory function (outlining and limiting the scope of the use of power). See, eg, CAJM Kortmann, Constitutioneel recht [Constitutional Law], revised (7th) edn by PPT Bovend’Eert, JWL Broeksteeg, SNJ Kortmann and BPM Vermeulen (Dordrecht, Kluwer, 2012) 21. 28  The communicative function of law (and legislation, as its expression) is, as observed by the Dutch scholar Wibren van der Burg, a complex one. Law may create a normative framework and a vocabulary to structure normative discussions, as well as institutions and procedures that promote further discussion. The expressive function of law is at stake when it expresses which fundamental standards and which values are regarded as important. See W van der Burg, ‘The Expressive and Communicative Functions of Law, Especially with Regard to Moral Issues’ (2001) 20(1) Journal of Law and Philosophy 31. See also BZ Tamanaha, Law as a Means to an End; Threat to the Rule of Law (Cambridge, Cambridge University Press, 2006). 29  ‘Primary’ is used here as opposed to ‘subordinate’ or ‘delegated’ legislation, in the form of statutory instruments, government decrees or ministerial regulations that are based on higher ranking regulations and merely detail the norms of the higher ranking laws.

Legislation and Regulation

 25

In Europe and elsewhere, these requirements resulting from the rule-of-law base of constitutional democracies have been made the cornerstones of legislative quality policies.30 The EU’s ‘Better Law Making’ policy also provides a good illustration of this approach.31 Although to some it will be quite obvious that the act of legislation expresses constitutional and democratic values, and is governed, in turn, by (constitutional) law (the legalconstitutional perspective), others take a different view. The Organization for Economic Co-operation and Development (OECD), for instance, approaches the idea of what they call regulatory quality from a more economic angle. The overall OECD perception of regulatory activity—often taking the form of legislation—is largely instrumental. The OECD and a large part of the regulatory governance community32 understand legislation primarily as a regulatory instrument, as a means to attain public good and to provide prerequisites for stable institutions, fair market conditions, citizen’s satisfaction, and economic growth and welfare.33 Taken from this perspective, legislation performs well if it maximises the net benefits of regulatory measures and citizen’s wealth. Legislative quality, according this view, is not primarily the extent to which legislation complies with constitutional principles or conveys symbolic notions, but rather the way legislation rates in terms of enhancing economic performance, or the dynamics of trade-offs of interests. Over the last decade, much effort has been invested in defining (a wide range of) regulatory quality indicators, in order to make regulatory quality measurable. Performance on such indicators gives us an idea of the ability of a government to formulate and implement sound policies and regulations that permit and promote private sector development.34 The different functions of legislation translate into different views on legislation. When one adopts an instrumental or political view of legislation, the conceptual lens to problems of and the proper standards for legislation will be different than from a more constitutional or symbolic perspective. It is therefore worthwhile to distinguish between the concept of ‘legislative quality’ and that of ‘regulatory quality’. From a constitutional point of view (and the symbolic function that is closely related to it), the only right measure for the quality 30  For instance, in the Netherlands the notion of legislative quality has traditionally been closely related to the demands of the rule of law and the demands of the so-called administrative quality of legislation ever since the 1990s. See, amongst others, the legislative policy document ‘Outlooks for Legislation’, Zicht op wetgeving [Dutch Parliamentary Papers] (Kamerstukken II 1990/91, 22 008, Nos 1–2). Quality of legislation in the Netherlands is perceived as the degree to which a regulation complies with the requirements (so-called ‘quality pairs’) of: (i) legality and lawfulness; (ii) implementation and enforcement; (iii) effectiveness and efficiency; (ii) subsidiarity and proportionality; (ii) harmonisation and coordination; and (ii) simplicity, readability and accessibility. These requirements are elaborated in policies and dedicated instruments, like reviews, manuals and (voluminous) drafting directives. 31  See European Commission, ‘Better Law Making’ COM (2002) 275 final, followed by European Commission, ‘Action Plan for Simplifying and Improving the Regulatory Environment’ COM (2002) 278 final and the Interinstitutional Agreement on Better Law Making of 31 December 2003, OJ C321/01. 32 Regulatory governance is topical in political science and the literature is still developing rapidly. The ­European Policy Group for political research has an active standing group on regulatory governance. See http:// regulation.upf.edu/ (accessed on 28 August 2015). It is quite strange that scholars in constitutional law are not catching on: a state’s regulatory capacity is the core business of constitutional law. 33  See, eg, OECD, Regulatory Policies in OECD Countries; from Interventionism to Regulatory Governance (Paris, OECD Publishing, 2002). 34  See D Kaufmann, A Kraay and M Mastruzzi, ‘Governance Matters VI: Aggregate and Individual Governance Indicators 1996–2006’, World Bank Policy Research Working Paper No 4280 (2007), available at http://papers.ssrn. com/sol3/papers.cfm?abstract_id=999979 (accessed on 21 September 2015). See also Centre for European Studies Bradford University, ‘Final Report on Indicators of Regulatory Quality’ (report for DG Enterprise, European Commission, 2004).

26 

Wim JM Voermans

of legislation is its ability to express law.35 The quality of legislation is the extent to which the criteria36 emanating from constitutional principles are met. Regulatory quality, on the other hand, is the extent to which legislation as a means to express public policies is successful in implementing policies to permit and promote private sector development, fair market conditions, stable institutions, citizens’ satisfaction, etc. These different notions, however, are not mutually exclusive; in fact, they coincide in some respects. One might, for instance, argue that the regulatory quality of legislation is a part of an overall notion of legislative quality, since it deals with effectiveness and efficiency of legislation. This would not, however, do justice to the very different perspectives on the function of legislation in the two different notions.

2.4.  The Regulatory State and Its Consequences There is more to regulation and legislation than functions and quality policies, if we consider their uses and requirements. Regulation is changing the face of our states and governments. If we boil it down to the essentials, then actually what states and governments do is manage resources.37 For this purpose, modern states perform public interventions in the economy. These public interventions are of three basic types: income redistribution (eg taxing and spending); macroeconomic stabilisation (influencing employment rates, monetary policies, central bank interventions); and, as discussed in sections II and III, market regulation (attempts to correct market failures like monopolies, public goods, externalities and information deficits). Majone has demonstrated that in the last four decades in Europe and elsewhere the positive (Keynesian) welfare state (predominantly focusing on taxing and spending) has given way to the regulatory state, in which governments more or less indirectly intervene in markets by regulating them in an attempt to make them more efficient and—consequently—more profitable to everyone.38 This shift towards the regulatory state has had different, and quite dramatic, consequences, especially for the executive branch of government (or administration), which has witnessed a rise in specialised agencies and commissions39 in favour of the formerly centralised bureaucracy. The rise of the regulatory state has affected the balance of power between the three branches of government as well: there has been a significant reallocation of powers.40

35  See, eg, Waldron (n 6) esp ch 2. See also CM Radaelli and F de Francesco, Regulatory Quality in Europe: Concepts, Measures and Policy Processes (Manchester, Manchester University Press, 2007). 36  See D Kaufmann, A Kraay and M Mastruzzi, ‘Governance Matters VI: Aggregate and Individual Governance Indicators 1996–2006’, World Bank Policy Research Working Paper No 4280 (July 2007). See also Centre for European Studies Bradford University (n 34). 37  And therefore politics, as Lasswell famously put it, is about ‘who gets what, when, and how’: H Lasswell, Politics: Who Gets What, When, How (New York, McGraw-Hill Book Company, 1936). According to Fukuyama, the phenomenon of states originates from the need to pool resources in some way or other. F Fukuyama, The Origins of Political Order (London, Profile Books, 2011). 38  G Majone, ‘From the Positive to the Regulatory State: Causes and Consequences of Changes in the Mode of Governance’ (1997) 17(2) Journal of Public Policy 139, notably 141–42. See also Majone (n 12). 39  Most of the time working more or less autonomously of the central government. 40  Majone (n 12) 155.

Legislation and Regulation

 27

First of all, more and more of the actual ‘legislating’—traditionally with parliamentary participation—has passed to the administration. This against the backdrop of an increasing symbiosis of the legislative and executive branches of government. So much so that some say we have arrived at a situation of duas politica, or duo politicum, instead of the traditional trias politica—ie the original idea of three branches of governmental power, notably legislative, executive and judicial.41 Secondly, law making is becoming less and less the reserve of the parliamentary–­ governmental complex, with the locus of regulatory activity moving away from the traditional command-and-control regulation, centred on the state, to self-regulatory processes by private actors in multiple locations. This notion of decentralized regulation, put forward by Black,42 is a process that results in the ‘decoupling of regulation from government’ and ‘the removal of government and administration from the conceptual centre of society’.43 Especially in the field of regulation,44 which has become so important, traditional state-based law making is ebbing away horizontally (to non-elected officials, in a process of bureaucratisation), vertically (to international organisations, the EU, etc) and laterally (to private actors, eg in processes of self-regulation), as Corkin observes.45 Thirdly, the rise of the regulatory state affects the judiciary as well. Seidman and Gilmour have argued that the growth of the regulatory state has converted the judiciary from a relative neutral referee to an active player in the administrative game.46 Malleson corroborates this analysis by pointing out that recently judges and judiciaries have extended their judicial review into areas hitherto considered the reserve of politics, and that (both as a cause and result) political life itself has become more judicialised.47 This transformation results in what Malleson calls a new judiciary. This new judiciary is activist, with new responsibilities in the field of law making and even policy making.

2.5.  Regulation: From a Solution to a Problem It is hard to imagine today, but there was once a time when societies were proud of their many regulations. These regulations, whether written down or not, were viewed as a sign of civilisation—one of the reasons to record them for posterity, as was done in Roman law in the Justinian Code. Good and acknowledged regulation was also a prerequisite for

41 

T Koopmans, Courts and Political Institutions (Cambridge, Cambridge University Press, 2003) 247. Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-­ Regulatory” World’ (2001) 54(1) Current Legal Problems 103. 43  As cited by J Corkin, ‘Constitutionalism in 3D: Mapping and Legitimating Our Law Making Underworld’ (2013) 19(5) European Law Journal 636, 640. 44  Generally ‘regulation’ is quite simply defined as any public intervention in markets and societies. See, eg, Baldwin et al (n 9) 6ff; Morgan and Yeung (n 7). Anthony Ogus defines regulation in somewhat more detail as ‘the necessary exercise of collective power through government in order to cure “market failures” to protect the public from such evils as monopoly behaviour, “destructive” competition, the abuse of private economic power, or the effects of externalities’: A Ogus, Regulation: Legal, Form and Economic Theory (Oxford, Hart Publishing, 2004). 45  Corkin (n 43) 638ff. 46  H Seidman and R Gilmour, Politics, Position and Power, 4th edn (Oxford, Oxford University Press, 1986). 47  K Malleson, The New Judiciary: The Effects of Expansionism and Activism (Aldershot, Ashgate, 1999). 42 J

28 

Wim JM Voermans

a flourishing trade sector. According to authors such as Leon Trakman,48 the Renaissance and the industrial revolution were made possible because of the booming development of the lex mercatoria (trade law or private law) from the early Middle Ages. Trade in the large commercial centres (the towns) could not rely on a patchwork of differing local legal rules, but needed acknowledged general rules with legal certainty for trustworthy transactions. Italian towns were proud of their large volume of written regulation: they used them as a kind of advertisement. Before our modern societies based on (free or otherwise) market transactions,49 we were dependent on clear and acknowledged agreements that were set out in advance in regulatory acts to generate the trust that was necessary for our kind of economic societies to function. Regulation is a prerequisite for economics and economic growth, and because economic growth has become so important, it is also important for our political society. What is special about the history of the past century, is that we don’t just recognise the standards written down in legal regulations as a necessary condition for trustworthy transactions within markets, but we increasingly use these regulatory acts to intervene in these markets, and thereby steer the economic society. Regulation no longer acts merely as a prerequisite, but as an independent instrument for government and administration.50 Many Western industrialised societies have developed into regulatory states because of the important role that interventions by regulatory measures play in economic, political and social life.51 Even though we don’t know exactly what effects different types of regulation have on the economy, government policies throughout the world have been trying for the past 40 years to stop overregulation. At the base of these policies often lie basic, but no less valid, theoretical concepts, such as the notion that by excessive regulatory intervention the market freedom of economic actors is reduced to the extent that innovation and competition may be compromised. In economies that are generally free and have an international focus, such as those that developed throughout the world in the 1970s, these are two factors that are essential for growth opportunities in the new global economy. It is therefore not surprising that the roots of most governmental deregulation projects go back to this period. We now live in a world in which countries and trade blocs compete in economic activities, but actually also compete when it comes to regulation. Regulatory competition is seen not just in attempts to create a favourable business or investment climate in a country, but also (in its wake) in the extent to which countries accept each other’s solutions and policies on regulation. It is noticeable that, since 1997, many of the countries that are affiliated with the OECD have actively adopted each other’s ‘best practices’ in the field of regulation policy. Almost all countries affiliated to the OECD today have a form of impact assessment for proposed regulatory acts. Practices to safeguard this impact assessment

48  LE Trakman, The Law Merchant: The Evolution of Commercial Law (Littleton, Fred B Rothman & Co, 1983) 11ff. 49  Contrary to self-supporting societies, where life and existence depend on being able to provide the essentials of life by hunting or gathering, or forms of agriculture. 50  This is actually quite remarkable because rules and regulations are a kind of (authoritative) ‘mould’ into which we pour agreements. But rules and regulations are often identified with the agreements (or ‘interventions’) themselves. This identification goes so far that in modern social scientific literature ‘regulation’ is usually seen as any public intervention in markets and societies. See, eg, Baldwin et al (n 9) 6ff; Morgan and Yeung (n 7). 51  Majone (n 12).

 29

Legislation and Regulation

(for example, making it compulsory), and methods to refine it and to (institutionally) incorporate it better in decision-making arrangements, are increasingly adopted from other affiliates.52 35

Number of jurisdictions

30 25 20 15 10

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

1978

1976

0

1974

5

Note: this represents the trend in the number of countries with a formal requirement for regulatory impact analysis (beyond a simple budget or facal impact).

Figure 1: The trend in the number of countries where a Regulatory Impact Analysis is required during the preparation of legislation/regulation53

2.6. Conclusion The last few decades have witnessed a remarkable increase in awareness of the problems and consequences of defective legislation, as well as a sense of urgency for ‘Better Law Making’ (or, similarly, ‘Better or Smart Regulation’). A host of regulatory and legislative policies consisting of criteria, tests, institutions and methods have been put in place in order to improve the quality of regulation and legislation. But did the medicine cure or prevent the disease, and is it properly dosed? This is a particularly hard question to answer, not only because these better law making or better regulation policies are still fairly new and have not yet been fully evaluated. It is also difficult because any appraisal of these policies depends, as we have seen, on a perception of the basic functions attributed to legislative instruments and the standards

52  See G Bounds, ‘Recent Trends in RIA Implementation’, Seminar on Regulatory Impact Assessment (RIA) in the Czech Republic and in other OECD Countries, Prague, 1 June 2012, available at www.oecd.org/gov/regulatorypolicy/50526827.pdf (accessed on 15 January 2013). The OECD is very active in this area and supports countries in their regulation policy for example with practical directives. See, eg, ‘Building an Institutional Framework for Regulatory Impact Analysis (RIA): Guidance for Policy Makers’ (OECD 2008), available at www.oecd.org/gov/ regulatory-policy/40984990.pdf (accessed on 21 September 2015). 53  Source: www.oecd.org/gov/regulatorypolicy/ria.htm (accessed on 15 January 2013).

30 

Wim JM Voermans

derived from them. Debate, academic or otherwise, on the standards for ‘good’ legislation and its theoretical underpinnings is still quite scarce. Different visions regarding the quality of legislation and the proper standards seem to have amassed in the present quality policies.54 For instance, the EU’s Better Law Making initiative (2002–2006), in its standards and policies, seems to reflect and emphasise the constitutional, democratic, bureaucratic and instrumental functions of EU legislation, whereas the Better Regulation initiative (2006–2014)55 seems to favour the instrumental and political functions.56 In the current Better Regulation strategy, however, strands of both perceptions, and the standards resulting from it, are still present. This leads to the paradoxical situation in which the Better Regulation strategy aims to improve the interventionist performance of legislation, whilst a great deal of the legislative standards reflect the constitutional, democratic and bureaucratic functions of legislation. Although, admittedly, there is some relation between standards and aim, this can lead to indetermination or even confusion about the objectives of the quality policies. This makes it very hard to assess the success of these policies, and to learn from experiences.

2.6.1. Assessing the Effect of Better Law Making and Better Regulation Policies The key to achieving better laws or regulation is hard to find. Baldwin has argued that there are no easy routes to regulatory improvement, and I think he is right.57 The success or failure of better law making or regulation policies is notoriously hard to review and assess. First of all, there is the problem of the yardstick. What are the right standards for legislation? What is the right scale? From a better regulation perspective, one may want to take a set of economic scales and try to calculate whether legislation strikes the right balance when it comes to minimising costs (eg not creating any more burdens than is strictly necessary) and maximising benefits (increasing productivity, employment, etc). The benefits, however, may prove hard to calculate. Especially if we consider that European legislation also creates trust, security, legal protection and all kinds of other, more or less imponderable, benefits for the internal market, the ‘pricing’ of pros and cons may prove to be extremely difficult. In the update of the project ‘Governance Matters’, the World Bank presents an impressive set of data on the scoring rates of 215 countries and regions on six dimensions of governance between 1996 and 2013.58 One of these dimensions was ‘regulatory quality’, which

54  De Francesco and Radaelli label this ‘the proliferation of objectives and goals of the regulatory reform’ after the Lisbon Agenda (including competitiveness, the completion of the single markets and the participatory-­ transparent government). F de Francesco and CM Radaelli, ‘Indicators of Regulatory Quality’ in C Kirkpatrick and D Parker (eds), Regulatory Impact Assessment; Towards Better Regulation? (Cheltenham, Edward Elgar Publishing, 2007) 50. 55  European Commission, ‘A Strategic Review of Better Regulation in the European Union’ COM (2006) 689, culminating, after different new initiatives, in the REFIT approach of 2015. See European Commission, ‘Better Regulation for Better Results—An EU Agenda’ COM (2015) 215 final. 56  See AJ Harcourt and CM Radaelli, ‘Limits to EU Technocratic Legislation’ (1999) 35 European Journal of Political Research 107, 109. 57  R Baldwin, ‘Is Better Regulation Smarter Regulation?’ [2005] Public Law 485. 58  The indicators are produced by D Kaufmann, Natural Resource Governance Institute (NRGI) and Brookings Institution; A Kraay, World Bank Development Research Group; and M Mastruzzi, World Bank Institute. http:// info.worldbank.org/governance/wgi/index.aspx#home.

Legislation and Regulation

 31

was measured by more than 25 indicators for quality. Based on these indicators, different groups and organisations were asked questions about the way legislation impedes, permits and promotes economic life in a country or region (Does the exchange rate policy hinder the competitiveness of firms? Is it easy to start a business? Etc). These data were compared with statistics on economic (and other types of) performance, giving an overall score on regulatory quality over a course of ten years. This allows a review of a country’s performance over the years or a comparison to other countries. But does it really tell us something about regulatory quality and can it be used to judge the effectiveness of better regulation policies?59 If we examine the data closely, we can see that performance indicators and outcomes do not calculate the effect of legislation to economic life as such, but merely estimate the risks of some regulatory impediments and the effect of laws directly pertaining to the economy, like business, tax and financial law. The quality of traffic regulations or of human rights in a country, for instance, is not considered separately in this project. Traffic regulations and human rights do, however, form an inextricable part of a country’s legal system, but it is difficult to assess what their particular contribution is to the quality of other legislation, or to the overall legal framework. This may yet prove imponderable, which can blur the total assessment of the regulatory quality of other, more measurable, domains of law. Regulatory quality assessment may be prone to a measurability myopia with a bias to the quality and effects of legislation that does not directly pertain to or affect economic life. Even if we were to take a somewhat more general, utilitarian approach, it would remain difficult to find a quantitative net result of better law making or better regulation efforts in terms of impact or success of the measures on the overall quality of the legislation. The legislative standards themselves seemingly resist an objective calculation of success. What is the right measure for proportionality, what are the degrees of legality and drafting quality of a directive, and what are the right benchmarks?

2.6.2. The True Test of Values Involved in Better Regulation and Better Law Making It may even be undesirable to assess legislation and policies to improve them in this way. Tamanaha argues that a one-sided instrumental use and perception of law, and one-sided debates on law as a mere means to an end, can threaten the very idea of legality itself.60 Mere instrumentalism may in the end undermine important social and symbolic functions pertaining to legislation. Regulatory quality and legislative quality are, as we have seen, different concepts, expressing and highlighting different functions of legislation. Where regulatory quality is about the extent to which legislation, as an instrument of public policy, permits and promotes ‘private sector development’, legislative quality is about the ability of legislation to express law. Improving legislative quality may require a different approach than the improvement of regulatory quality. Sometimes these objectives may even run counter to or prove incompatible with each other. Does that mean there is no way of assessing the success of (EU or other) better law making policies? On the contrary, we believe there is. First, one can take a more procedural 59  60 

See also Radaelli and de Francesco (n 35) 48–49. Tamanaha (n 28).

32 

Wim JM Voermans

point of view to the assessment: have the relevant procedures been complied with, have the programme targets been met, is the message hitting home and are the actors satisfied with the result? One may also take the level of institutional embeddedness as an indicator of the success of the methods and institutions held in the policies. The ultimate test for better law making or better regulation will, of course, be the political appraisal of the outcome of the policies. Better regulation and better law making policies are most of the time essentially political programmes resting on political perceptions as to the overriding values of legislation and regulation in a market economy. This does not mean that the political assessment of the success is purely subjective or unstructured. It only means that, ultimately, the effect and success of better law making can only be weighed politically. On that note, it is important that better law making and better regulation policies are monitored well and are evaluated regularly. And on the occasion of such evaluation, it is important to pay due attention to the differences between regulatory quality and legislative quality.

Further Reading Baldwin, R, Cave, M and Lodge, M (eds), The Oxford Handbook of Regulation (Oxford, Oxford University Press, 2010). Baldwin, R, ‘Is Better Regulation Smarter Regulation?’ [2005] Public Law 485. Black, J, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-regulatory” World’ (2001) 54(1) Current Legal Problems 103. Harcourt, AJ and Radaelli, CM, ‘The Limits to EU Technocratic Legislation’ (1999) 23 ­European Journal of Political Research 109. Kelsen, H, ‘The Law as a Specific Social Technique; the Essence of Legal Technique’ (1941) 9 University of Chicago Law Review 75. Majone, G, Regulating Europe (London, Routledge, 1996). Morgan, B and Yeung, K, An Introduction on Law and Regulation (Cambridge, Cambridge University Press, 2007). Ogus, A, Regulation: Legal, Form and Economic Theory (Oxford, Hart Publishing, 2004). Radaelli, CM and de Francesco, F, Regulatory Quality in Europe: Concepts, Measures and Policy Processes (Manchester, Manchester University Press, 2007). Tamanaha, BZ, Law as Means to an End; Threat to the Rule of Law (Cambridge, Cambridge University Press, 2006). Voermans, W, ‘Concern About the Quality of EU Legislation: What Kind of Problem, by What Kind of Standards?’ (2009) 2(1) Erasmus Law Review 59. Waldron, J, The Dignity of Legislation (Cambridge, Cambridge University Press, 1999).

3 Legislative Process TÍMEA DRINÓCZI

This chapter gives a rough overview of the main characteristics of legislative processes, mainly concentrating on the regulatory framework, legislative organs and stages of legislative processes. Some trends and challenges (such as political commitment toward legislative quality, the claim about the loss of the legislative character of parliaments, the tendency to use more information and communications technology, the emerging need for evidence-based legislation and regulatory management) are also addressed. It is not easy, or even possible, to strictly divide the processes and the organs taking part in the processes, though to elaborate on the topic of legislative processes and address the issues mentioned it seems necessary, howsoever artificial it may seem.

3.1. Introduction Comparative legisprudence deals with the issue of ‘who’ legislates, and ‘on what’ and ‘how’ laws are produced. This subject matter encompasses the entire legislative process, from the very beginning (initial development of the broad policy ideas and conceptualisation of a draft) to the very end (implementation and monitoring), including all interim procedural steps. These steps are policy making and preparation, either by governmental agencies or parliament, drafting,1 consultation,2 impact assessments, readings, hearings in parliament or in its committees, adoption, promulgation and publication,3 implementation, monitoring, ex post impact assessments4 and maintenance5—and, if necessary, due to policy or other changes, as well as the need for amendment, the entire process starts all over again. This is mainly how laws are made; but this process cannot be dealt with without answering the ‘who’ question. Those who are taking part in the legislative process determine and monitor the main policy goals and their way of achievement (political decision ­makers), draft the law (drafters, being either staff members or external drafters, employed by the executive or legislative branch), organise and participate in consultation (responsible staff members and various stakeholders), assess the impact of draft laws and already ­implemented laws, do analytical research (staff of the executive and legislative bodies or 1 

See Chapters 7–9 of this book. See Chapter 5 of this book. 3  See Chapter 10 of this book. 4  See Chapter 11 of this book. 5  See Chapter 12 of this book. 2 

34 

Tímea Drinóczi

independent agencies, institutions, experts), adopt laws and by-laws or secondary legislation (parliament, government, heads of state, if available, local governments, independent legislative bodies), implement and adjudicate the law (public administration and courts), and determine the constitutionality of the law (supreme courts, or constitutional courts if established). Obviously, such actors as the executive and the parliament may be further classified as states may have a unicameral or bicameral parliament. Different committee structures and compositions may be applied, different tasks and competences may be allocated to line ministries, the internal working method in ministries may differ from state to state and work within the parliament may be organised differently. The establishment of external institutions or involvement of independent agencies or experts, even academia, depends on the state’s own decision on how to organise or even manage its legislative/lawmaking process, best fitting its constitutional framework, governmental system, political and social structure, and culture. It may also be the case, however, that some procedural steps are missing from a state’s legislative process; consequently, it is not a general rule that all the above-mentioned actors actually take part in legislation or that all the procedural steps are really implemented. Nevertheless, in a constitutional democracy, the legislative process should comply with the basic principles of the rule of law and democracy, which are transparency, participation, knowledge-based decision making, the possibility to choose (having alternatives), accountability, constitutionality and legality, direct representation and presence of the legitimacy chain, the division of powers, and respect for the parliamentary minority’s rights and for human rights. Compliance with these principles has the potential to enhance legislative quality, provided that there is an (undeniably necessary) political commitment towards better regulation and legislative quality at the political level, and the staff dealing with ­legislative issues are well trained, and have sufficient resources and capacity. Active citizenship and active stakeholder engagement are also needed, as they are the basis of involvement and participation, and, to a certain extent, of monitoring the (legislative) activity of those having the political power. Legislation and the legislative process may be seen from a more procedural aspect and the focus can be put either on the peculiarities of the pre-parliamentary (preparatory phase), parliamentary and post-parliamentary processes or on the procedures applied during these processes, such as the consultation and involvement of experts, which may occur in both the preparatory and parliamentary processes, and impact assessment, which is conducted not exclusively in one of the legislative phases. A more structural approach can be applied when one is interested in the organisational structure of a particular state related to legislation: constitutional organs vested with regulatory power, independent regulators (their constitutional status and powers), and bodies responsible for impact assessment and measuring administrative burdens, including advisory, consultant and coordinating organs, etc. Against this background, and taking the other chapters of this book into account, only a rough overview of the main characteristics of legislative processes is given in this ­chapter, which concentrates mainly on the regulatory framework, legislative organs and stages of legislative processes. Some of the trends and challenges (such as political commitment toward legislative quality, the claim about the loss of the legislative character of parliaments, the tendency to use more information and communications technology (ICT), the emerging need for evidence-based legislation and regulatory management) that are presented in a more systematic manner in section 3.5 are also addressed. It is not easy, or even possible,

Legislative Process

 35

to strictly divide the processes and organs taking part in the processes, though to elaborate the topic of legislative processes and address the issues mentioned it seems necessary, howsoever artificial it may seem.

3.2.  Regulation and Organisation Legislation itself is a cyclic action of the state, but each legislative action (such as policy making, drafting, negotiating, consulting) is also cyclic to a certain extent as they require a continuous decision-making process fitted to the particular legislative action. Thus, components of ordinary decision making are applied throughout the legislative process, in each phase and by each body involved. These elements are: identifying the problem; analysing the start-up situation, as well as the reasons leading to that situation; determining the legislative aim; analysing the possibilities of reaching the aim; evaluating alternatives based on the aim; decision making; and monitoring the result. As can be seen, the greater part of the law-making process is a political as well as ordinary decision-making process and, in principle, cannot be regulated by law.6

3.2.1.  Basic Rules of the Legislative Process The regulation of the legislative process extends to the most important issues about ­legislation, such as (legislative) principles, competences, bodies, processes, reservation, legislative initiative, types of statutory laws (organic, ordinary), hierarchy, validity of the legal norm, referendum, and distribution of legislative competences. The sources of law and their contents may differ from state to state, but it is certain that general principles, such as representative democracy, participation, the rule of law, legal hierarchy and legislative competences, law-making competences of the executive and other organs, the main rules on promulgation and publication, and constitutional review of legal norms are stipulated at the constitutional level. It is also a new phenomenon that in the newer constitutions (newer revisions) legislation-specific principles and procedures are prescribed: for example, the Swiss constitution (2000) stipulates the efficiency principle in connection with legislation. Another legislation-related principle is the prohibition of detrimental retroactivity (other than the prohibition of retroactive criminal legislation). In some constitutions this prohibition is explicitly laid down (for instance, in the Croatian basic law), and in other ones it is implied by the rule of law principle (Italy, France and Poland). Unfortunately, it may also happen that, due to sheer, one-shot political intention, a constitution contains a rule on the possibility of retroactive taxation (eg in Hungary in 2010–12).7 The newer and more 6 I Bogdanovskaia, ‘The Legislative Bodies in the Law Making Process’, available at www.nato.int/acad/ fellow/97-99/bogdanovskaia.pdf (accessed on 8 March 2014). See also the five stages in law making in GC Thornton, Legislative drafting (London, Butterworths, 1987). 7  For more on this, see T Drinóczi, ‘Constitutional Dialogue Theories—Extension of the Concept and Examples from Hungary’ (2013) 1 Zeitschrift für öffentliches Recht 87. For common law, see R Cormacain, ‘Retroactivity, Retrospectivity, and Legislative Competence in Northern Ireland: Determining the Validity of Janus-Faced Legislation’ [2015] Statute Law Review 134, 10.1093/slr/hmu054.

36 

Tímea Drinóczi

detailed the constitution, the more likely it is that it contains some references to particular legislative principles, such as efficiency or prohibition of detrimental retroactive legislation. The only question here is enforceability, which should be based on the demand for evidence-based legislation, developed mainly by the judicial power. The main procedural phases of legislation,8 more detailed rules on coordination, planning, management, policy making, publication and official journals, as well as access to laws, participation, consultation, the need for impact assessment and related responsibilities, are mainly regulated in acts on legislation, and in the UK by conventions and customs. Parliaments have their own rules of procedure, while in the executive bodies more flexible9 rules of procedure may be adopted. More detailed technical issues relating to all of the above-mentioned activities are regulated at sub-legislative level, and their legal source is either a legal norm or a guideline. This depends on the choice of the state, taking into account the legal culture and level of compliance.

3.2.2. Constitutional Context and the Status of Parliament and Government in the Legislative Process 3.2.2.1. Parliament In a parliamentary democracy, it is the constitution that prescribes the law making competences of state organs, reserving the legislative power to the parliament as a representative body and giving a detailed description of the rules and the basis on which the executive can make a legal norm. The structure of the parliament (unicameral or bicameral)10 and the main competences of chambers, as well as the distribution of legislative tasks to territorial structures of the state, be it a federal subject, region, autonomous region or local government, are also stipulated in the constitution in a detailed manner, depending on the structure of the state. In a federal (eg Germany, Belgium) or regional state (eg Italy), or a state with devolution (eg UK), legislative competences are classified into exclusive competences, reserved to the state parliament, and concurring legislation and subject matters, in which sub-state legislative bodies are free to make law and/or concerning which the constitution refers to the principle of subsidiarity in legislative matters. The state with devolution differs slightly as its different territorial units have different powers, and this difference in status creates a different balance in relation to the central government. In a unitary state, whether centralised or decentralised, a similar vertical division of legislative power can be found, due to the recognition of the right of self-governance, even though these territorial units cannot be compared to those having a more complex territorial structure. Due to historical, ethnic, cultural etc reasons, special territorial structures having an autonomous status (under different names, such as province, region, community and country, reflecting differences in legal status) exist in some other states, such as Finland, Spain and Denmark. This status is often accompanied by the structure having its own legislative power stipulated by the constitution. Whatever the territorial structure of a state, sub-state legislative power has to be in harmony with the laws of the central legislative body and the constitution. 8  9 

10 

This topic is discussed below in section 3.4. Bogdanovskaia (n 6). See Chapter 1 of this book.

Legislative Process

 37

Parliaments and governments take part in the legislative processes of the EU and international organisations. This involvement is not constitutionally regulated in a detailed manner but certain other legal measures also regulate its procedural aspects, especially parliament’s involvement in the decision-making process and oversight of the government.11 As for the regulatory power of international organisations, compliance with rules, processes, principles developed by nation states and best practices studied by legisprudence shall mutatis mutandis be observed at international level as well. The reason is the emerging phenomenon of global governance12 and transnational governance,13 which justifies the recognition that states are not able to perform some functions traditionally belonging to the state due to global problems and challenges. That is why many of the state’s functions have necessarily been transferred beyond the state’s power, so states need to continuously cooperate within international organisations, and comply with rules entailed by their own involvement in order to solve global problems and/or regulate matters that are beyond the state level but which affect states as well as individuals living within the borders of the states.14 Moreover, there is a phenomenon of growing participation of non-state actors (non-governmental organisations (NGOs), transnational cooperation and individuals) in international governance. Therefore, law making and law enforcement at this level call for applying existing notions and solutions of constitutional law as well as legisprudence in terms of principles, best practices, and legislative solutions and methodologies. Simultaneously, in this respect parliaments need to strengthen their oversight function over actors taking part in cooperation organised beyond the state level. Due to the very functions of the parliament, such as its representative, legislative and educational functions, and the role of ensuring publicity and legitimacy, it is a fundamental requirement that processes and documentation be transparent and the latter be easily accessible. This characteristic of the parliamentary process and working method should trigger better understanding of legislative processes, more willingness to participate and more trust in the democratic institutions. Transparency is best guaranteed by freedom-of-information laws and using properly developed and fully serviced information technologies.15

3.2.2.2. Government In parliamentary democracies, the phenomenon that the government has the capacity, power and possibility to provide input to legislation is acknowledged by constitutions. The constitution confers the right to initiate legislation and imposes the obligation to implement laws on the government. At the same time, it makes the government politically responsible, even for its policy-making activities, to the parliament. A government may produce legal norms by iure proprio (within its constitutional competence) and iure delegato (based on 11 

See Chapter 1 of this book. It is a wide category aiming at ‘creating political order in the absence of a state with a legitimate monopoly over the use of force and the capacity to authoritatively enforce the law and other rules’, or governance without government. See T Risse, ‘Transnational Governance and Legitimacy’, available at userpage.fu-berlin.de/~atasp/ texte/tn_governance_benz.pdf 2 (accessed on 8 March 2014). 13  It refers to those governance arrangements that are beyond the nation-state and in which private actors are systematically involved. ibid 3. 14  See S Hainsworth, ‘Sovereignty, Economic Integration, and the World Trade Organization’ (1995) 33 Osgoode Hall Law Journal 590. 15  For integrity studies see Transparency International at www.transparency.org/enis/national_results. 12 

38 

Tímea Drinóczi

delegation and authorisation by statutory law) in the following manner: in the form of (i) secondary legislation16 (based on iure proprio), (ii) a legal norm having the force of a Act of Parliament (based on and within the framework of the constitution) or (iii) in the form of the implementation of an Act of Parliament (based on iure delegato). In this latter case, the Act sets out the broad principles, and leaves it to secondary legislation to set out some of the technical details. In a constitutional democracy, legality and constitutionality means that legally binding acts cannot be produced disregarding related constitutional and legislative rules. The decision of what issues should belong to the law-making power of the executive is to be made during the constitution-making process. The constitution needs to define both the scope and limits of the competence of executive organs to make secondary legislation. This is usually realised by constitutional guarantees ensuring the dominance of statutory law/Acts of Parliament by stipulating legal hierarchy or reservations, enumerating subject matters reserved exclusively for the legislative power or guaranteeing the primacy of legislative norms. In some countries, eg Italy and Georgia, the constitution may also empower the government or the head of state (eg in Estonia)17 to issue a legal norm that has the force of law. This may occur only in cases determined by the constitution and in compliance with procedural rules prescribed therein. Such a legal norm may be issued only in exceptional and urgent cases. The validity of this norm depends on the subsequent approval of the parliament within a period of time determined by the constitution. The lack of approval means that the legal norm is null and void. There is a possibility of misusing this power and, if the executive lacks self-restraint, it may issue a large number of emergency ordinances. In the Romanian constitutional practice, the exercise of this power in a well-extended manner can be considered ‘very easily as an abuse of (almost) every government’.18 In other states, due mainly to historical experiences (Hungary,19 the Czech Republic20 and Bulgaria,21 for instance), the government is constitutionally not allowed to adopt legal norms having the force of law at all.

3.2.2.3.  On the Loss of the Legislative Character The exercise of its decree-issuing competence by the executive does not lead to the loss of the legislative power of the parliament as the parliament still possesses the legislative power relating to the most important political issues and can focus on the politically most important matters, because the executive takes over regulatory burdens. Besides, reservations make it possible for the regulatory power of the executive to remain bound by statutes, 16  Also referred to as subordinate legislation or executive legislation. The name of the legal act may differ according to jurisdiction. 17  J Pöld and B Aaviksoo and R Laffranque, ‘The Governmental System of Estonia’ in N Chronowski, T Drinóczi and T Takács (eds), Governmental Systems of Central and Eastern European States (Warsaw, Wolters Kluwer Polska—OFICYNA, 2011) 267. According to the authors, no such decrees had been issued as of 2010. 18  E Veress, ‘The Governmental System of Romania’ in Chronowski et al (ibid) 565. 19  In the socialist era, the collective head of state (Presidential Council of the Peoples’ Republic of Hungary) had the power to issue law decrees which usurped the functions of the parliament and that of statutory law. The reason was not only the composition of this organ (prominent party members and party secretary), but also the fact that the parliament was not acting as the main legislative body at all. 20  J Filip, P Molek and L Vyhnánek, ‘Governance in the Czech Republic’ in Chronowski et al (n 17) 194. 21  There have often been cases of abuse by the tzar, the Presidium of the National Assembly and the State C ­ ouncil of the right to issue decrees with the force of a parliamentary act. E Tanchev and M Belov, ‘The ­Governmental System of the Republic of Bulgaria’ in Chronowski et al (n 17) 93.

Legislative Process

 39

as certain subject matters can only be regulated by parliament and in a statute. Decrees may regulate the same subject only to the extent to which they have explicitly been by the piece of primary legislation. Substantial regulation is thus not transferred to the regulatory power of the executive. It is also required that the legislature give proper authorisation or delegation of power, and that the executive (government) should make proper implementation and not misuse its constitutional entitlement to adopt legal norms having the force of law. Nevertheless, in a well-functioning constitutional democracy, there are monitoring and reviewing mechanisms available to check if the legislative or law-making power has been used properly. These mechanisms (courts, the constitutional court and the ombudsman) can provide a remedy where the principle of leaving intact competences (more simply: the separation of powers) has been infringed. With this constitutional framework (and each constitutional democracy has something similar), the trend described as ‘the executive is more “legislative” in character than the legislative power’ and that there is a loss of parliamentary power22 does not seem to be adequate. In (most) parliamentary systems, due to their special characteristic that the parliamentary majority is in a very close relationship with the government in terms of trust and responsibility, the government has always been the one to make policy decisions and then channel them into the parliamentary process.23

3.3.  Legal Measures, Other Bodies and Participants Taking Part in Legislative Processes 3.3.1.  Legal Measures The policy maker has to decide on the normative formulation of the rules of drafting, consultation and impact assessment. This basic choice depends on the constitutional provisions and relevant decisions of the constitutional court on sources of law and legal certainty, and other basic values determining the governmental system, such as the division of powers and the attitude of the legislature towards better/smart regulation policy. The non-normative rules on these issues may be used and are an advantage in a state having an explicit commitment to better regulation and a high level of awareness and understanding24 of the necessity of legislative quality. Without doubt, there are advantages to this regulatory method. One of them is flexibility, as deviation from rules, if necessary and justified, can be made without any complication. Another advantage is that the legal system is not overloaded with norms that are deemed to be generally applied by all means. A disadvantage, however, is that the 22 Especially because a survey conducted in 2001 showed an increasing role of the parliaments. See A Kasemets, ‘Impact Assessment of Legislation for Parliament and Civil Society: A Comparative Study’, available at www.siseministeerium.ee/public/Kasemets_Aare.pdf 51 (accessed on 5 March 2014). 23  The essence of the parliamentary governmental system is that it is the government that plays an active and proactive role in governance, which is dependent upon the will and trust of the parliamentary majority and is responsible to this majority. The parliamentary majority may rely on the proposition that the government, when making policy choices and drafting, takes the actual situation into account. Further, the ministerial and ­governmental staff are capable of coordinating a comprehensive and complex legislative process, and in this respect, can take over some of the burden from the parliament. 24  Legal culture can be enumerated here as well.

40 

Tímea Drinóczi

legal consequence of non-compliance is hardly tangible, and this is not in conformity with the principle of legal certainty. The advantage of regulation in a legally binding norm is the binding force itself, which forces the personnel to comply with the rules, and this may result in a greater personal commitment. It may lead to changing attitudes; an implicit commitment can change to a more explicit one. In Croatia, there is a separate Act on impact assessment, in Hungary it is mentioned in the Act on legislation and is elaborated in a decree of the minister of justice. However, in these countries, the very fact that there is a legal obligation to do Regulatory Impact Analysis (RIA) has not altered the attitude towards it—the number of RIAs conducted and their quality have not increased at all. As for the normative regulation of drafting rules, there can be two choices: detailed regulation for the entire drafting process and the formulation of a draft could be laid down in a legal norm such as an Act or decree; or only the main rules could be included in the Act or decree, and specifications could be laid down in a decree or a manual or guideline, respectively.25 The advantage of the latter is that it could lead to a more detailed and comprehensive measure that assists the work of legislative jurists. The advantage of the former (and disadvantage of the latter) is that, being an Act, it has a ‘more normative binding force’. If the drafting rules are regulated in an Act, a lower-level norm (governmental or ministerial decree) cannot be in conflict with that Act. If, however, these rules are in a decree, the rule of hierarchy does not apply. Should the rules on drafting be infringed, it is up to the constitutional court to decide if the infringement reaches the level of unconstitutionality. Unconstitutionality is more likely to be stated when the rules infringed are located in an Act. However, this does not mean that each infringement is declared unconstitutional; it depends on the tests developed by the constitutional judiciary in connection with legislation. The detailed regulation concerning drafting legislation can usually be found in the so-called ‘legislative guides’, which are used in several Western countries (eg Austria, Germany) and the EU. In Hungary, however, there is a decree of the minister of justice on drafting legislation. Consultation and impact assessment need to be completed either at the preparatory or the parliamentary phase; and different phases may necessitate different procedural rules and approaches. These appear in the rules of procedure or other internal regulations of governments and parliaments. What is important, though, is that the same methodology should be used. It seems that the normative formulation of the rules of specific legislation-related activities should have two or, depending on the internal regulatory background of the state concerned, three levels: statutory law, secondary legislation (decree), and non-binding manuals or guidelines. The use of RIA is not legally regulated in many old democracies, while in many of the so-called new democracies its use is legally regulated to bring about an institutional change through the power of law. The distinguishing factor between Central and Eastern European (CEE) and other European countries is constituted not by the borderline between legal/ normative and informal regulation of requirements on RIA, but by the actual implementation of the rules and the differences in the institutional framework and methodology. CEE countries tend to lack both proper implementation and methodology. As for consultation, survey shows that there is a legal obligation to seek the opinion of different state organs, but there is usually no legal obligation to ask for an opinion from NGOs. It seems to be a trend that public authorities are building up stronger cooperation with civil society based on 25 

See T Drinóczi, ‘Quality Drafting—The Case of Hungary’ (2010) 2 Legisprudence 157.

Legislative Process

 41

legally non-binding multilateral agreements, which may raise the question of enforceability. Enforceability depends on the decision of the political decision maker on involvement/ non-involvement, as well as on the approach of the (constitutional) court to evidencebased legislation—whether they demand it or not. The comparatively young principles and procedures of involvement of NGOs in the EU institutions and some European countries can nevertheless be observed.26

3.3.2.  Other Actors and the Legislative Process It is not only the parliament and the government that take part in legislative processes, but other constitutional bodies, agencies, other advisory bodies, civil society and individual citizens, interested parties and stakeholders, different experts within and outside the governmental structure, academia and independent bodies as well. Shaping the legislative structure and mechanism, as well as the employment and involvement of the abovementioned actors, depends on the actual constitutional arrangement of the state and the political will towards legislative or regulatory quality. Further, some examples of bodies and other actors that are involved in the legislative and decision-making process according to some fields of legislation (RIA, consultation, drafting, legislative initiatives and referenda) are provided. In most countries, there is no central unit for conducting RIA, and usually each ministry carries out its assessment in the range of the ministry’s competence. The Netherlands, Poland, Russia and Slovenia are the only exceptions, as they have such a type of central unit.27 For draft legislation originating from parliament, the following ways to analyse impact are used: impact assessment conducted (i) primarily by the author of the proposed legislation, (ii) by the government at the request of the parliament, (iii) by parliamentary bodies (standing committees, research services, etc) or (iv) by research institutes, universities, etc at the request of parliament. The parliament can order different studies if there are special funds allocated for that purpose in the parliamentary budget (it depends on how financial autonomy is achieved). It seems to be advisable to develop knowledge and skill both in the government (ministries) and in the parliamentary services in order to implement RIA. States opting for this solution need to reconsider the procedural rules they apply in the preparatory, governmental and parliamentary phases. Thus, as far as parliaments are concerned, consideration should be given to: (i) allocating additional resources to parliamentarians (Members of Parliament; MPs), either by adding more staff or by giving them funds to remunerate external experts consulted on an ad hoc basis; or (ii) indicating methods by which MPs can draw upon the resources of ministries, for instance by asking a minister to draft a bill.28 There are different rules and practices on the involvement of governmental or other state bodies and NGOs in the legislative processes.29 In some states there is no legal obligation

26 

Kasemets (n 22) 54, 61, 62, 90, 98. For more detail, see Chapter 11 of this book. 28  Kasemets (n 22) 72, 73, 93, 94. 29  ibid 62–65. 27 

42 

Tímea Drinóczi

at all to involve any governmental body or agency or civil society organisation (Austria, Slovenia, UK), though in practice they tend to be involved on a regular basis. Austrian ministries send their draft to such bodies and the Nationalrat (second chamber) may carry out additional consultation. In Slovenia, consultation is performed in specific areas (budget, research policy, etc), and governments often seek the opinion of social partners (professional associations and unions). In the UK, consultation is required under a guide that has a non-legally binding character. There are states where there is a legal requirement to involve state bodies, but it is not compulsory to involve NGOs (eg Romania, the Netherlands, Lithuania); in other states, involvement of both state bodies and NGOs is an obligation (eg Czech Republic, Poland, Sweden). Even in these countries, there are differences as to which body is to be consulted obligatorily. In states where a Council of State is established as a preparatory and advisory body of the government or head of state, its involvement in the preparatory phase is required, even if its opinion is not binding (eg Belgium, Bulgaria, France).30 In Sweden, the majority of bills are submitted for a legal-technical check to the Council of Legislation, which is made up of judges. In some states, the governmental bodies affected by the draft have to be consulted (in the Czech Republic: the National Bank, Audit Office, telecommunications regulatory body); in others, this involvement is more restricted (for example, in Hungary, local government or other (state) organ(isation)s may give their opinion on a specific law if they are expressly authorised by an Act to do so, and even in this case they are entitled to do so only with regard to draft legislation relating to their legal status or scope of duties). In some countries, separate advisory councils dealing with legislative and other issues are established, such as the Romanian Legislative Council and Economic and Social Council,31 or the Legal and Economic Council established by the government of Macedonia. The actual involvement of NGOs, besides the legal regulation, depends mainly on the level of organisation of society. For instance, in the Netherlands, where there is no legal obligation to involve civil society, it is highly unusual and regarded as unacceptable for NGOs not to be involved in the policy-making process from the very beginning. On the other hand, eg in Hungary, where civil society is comparatively not as well organised, if NGOs are involved at all, this certainly does not happen at the very beginning of the policy making, and the actual added value of their involvement is very low, due to the policy already having been decided or the law drafted. Another example is that of Poland, where the following bodies have the right to be consulted regarding bills under a binding procedure: countrywide trade unions, federations and confederations of employers, chambers of commerce and the Union of Polish Craftsmen. In the majority of states, laws provide for direct citizen participation as well.32 Drafting is mainly made within a special unit of the ministries, and within the internal structure of parliaments. Usually it is the parliamentary staff who help MPs to draft their

30  It would, however, be strange if the binding character of opinion of a council of state had been acknowledged, as it would mean a decision-making or veto power, depending on the perspective. 31  These bodies are constitutionally established advisory bodies; the Legislative Council is the advisory body of the Romanian parliament, whereas the Economic and Social Council is an advisory body of both the government and the parliament. 32  See Chapter 5 of this book.

Legislative Process

 43

initiatives, though in some cases external drafters can be involved, depending on the rules of procedure of the parliament. In certain cases, drafting is outsourced to externals even at the preparatory phase (eg in Hungary). In this case, active and continuous cooperation with the line ministry (or parliament) and other actors is necessary, which may not be achieved very easily. In addition, the skill of the drafters and the quality of the draft may be justifiably criticised. That is why the outsourcing of drafting does not seem to be a good practice. Decisions on the organisation and operation of the drafting activity depend on the common law or civil law nature of the legal system. The most powerful way for a citizen to influence the legislative process and make an impact on the content of the law is the right to legislative initiative and participation in a referendum. The main difference between these two options is that, whilst, with a legislative (or constitutional) initiative, citizens can determine the text and the very content of the draft, this is not possible at all in the case of referenda: during a referendum, the citizens’ influence is restricted to a simple ‘yes’ or ‘no’ answer to a specific question. The legal effects differ as well. By submitting a legislative (or constitutional) initiative, citizens only give an impulse to the legislative power; their draft may be drastically changed during the processes. In the case of a referendum, however, voters determine for a certain time the legislative obligation (to legislate or not legislate on a certain subject matter) of the parliament,33 or they adopt, reject or abrogate a certain law, depending on the national regulation.34 The power of legislative initiative belongs to a group of citizens whose exact number is determined by the constitution of the respective state, like in Lithuania, Italy, Romania, Serbia and Slovenia.35 This power is usually given to citizens having the right to vote. It is also a common requirement that a properly formulated draft has to be submitted. This institution is not recognised in all European states, eg in Hungary. With the exception of Switzerland, an even more rarely used institution is where a group of voters are entitled to initiate a constitutional amendment. In Lithuania, such a power is given to 300,000 voters; however, this kind of process has never succeeded.36 In Switzerland, however, amending the constitution and adopting laws with the active involvement of the voters is standard practice. By contrast, in Hungary, the constitution cannot be changed at all via a referendum, so even the initiation of a constitutional referendum is inadmissible. As for the referendum, according to constitutional and statutory rules, citizens having an active right to vote may take part in constitutional and legislative referenda. In the case of a constitutional referendum, a referendum may be required by the constitution for all kinds of amendments (eg Switzerland) or, most commonly, for amendments of only the most important constitutional rules, values or principles (eg the Baltic states, Serbia).

33  This kind of direct exercise of popular sovereignty is called plebiscite, as the decision of the people is not preceded by any final decision on the draft by the representative body. The expression of referendum is suggested to be used only for cases involving the joint exercise of direct and indirect popular sovereignty: parliament decides on the law, ie adopts it, and the people vote for or against it during the referendum. 34  The same applies at the local level as well. 35  Just to give some examples of the required number of voters: a legislative initiative may be submitted by 50,000 Lithuanian citizens (though until 2010 legislation had never been adopted using this popular initiative; see VA Vaičaitis, ‘The Lithuanian Governmental System’ in Chronowski et al (n 17) 452], at least 100,000 Romanian voters, 30,000 Serbian voters, 5,000 Slovenian voters or 50,000 Italian voters. 36  The data are as of 2010. See Vaičaitis (ibid) 443.

44 

Tímea Drinóczi

In France, for instance, a practice to change the constitution via referendum has been developed, but a referendum is also required when the two chambers of the French ­Parliament adopt the same text on a constitutional amendment. In the case of legislative referenda, their application and the detailed rules relating to them are rather nation-specific, thus the main rules covering them (types, entitled actors, excluded subject matter, validity of the vote) can be found in the national constitutions.

3.4.  Legislative Process The law-making process consists of several stages.37 As a rule, after policy making, a draft is prepared and scrutinized, then a bill is submitted. The power to initiate legislation is usually given to MPs, groups of MPs, the government, groups of citizens, autonomous provinces, if there are any (Spain), and in certain European states (eg Denmark, Hungary and Latvia) also to the head of state,38 and in the UK the Law Commission for England and Wales. It may also happen that the parliament receives from the president of the republic (eg Latvia) only an initiative for legislation but not the detailed text of a bill; in this case, the parliament decides whether to instruct the particular ministry or its own committee to draft the text of the bill.39 Once the bill by been adopted by the parliament, the law is published. A constitutional review of the adopted law may be requested.

3.4.1.  Preparatory Stage or Pre-parliamentary Stage In order to organise and plan legislative activity, the government is required to make a legislative programme, and it sends this plan to the parliament, preferably before the beginning of an ordinary session. From a purely procedural point of view (in civil law states), the first stage of law making starts with the preparation of the first version of the draft, in which an idea on law is realised. A governmental body may take an official decision concerning the elaboration of a project, give the task to its internal structures (committees, departments) to write a bill or prepare a preliminary analysis of the public interests, the need for legislation, and the correspondence of a prospective Act to the current legislation and to the constitution. The project is discussed by other governmental agencies and bodies during the intergovernmental cooperation process, and experts, associations and interest groups may also be involved. The working committee (or, in the case of draft reform laws, codification committees or inter-ministerial committees) analyses the results of the discussion and occasionally changes the text. The next stage consists of the scrutiny of the project by a governmental body. In a well-managed preparatory phase, the matter in question, determined by the instructions of the policy maker (government), is evaluated and analysed by a unit or committee, or a group of experts. The instructions and the report and proposal are published and 37  38  39 

This summary is mainly based on Bogdanovskaia (n 6). However, in Hungary, for instance, this power is not exercised in practice at all. D Iljanova, ‘The Governmental System of the Republic of Latvia’ in Chronowski et al (n 17) 404.

Legislative Process

 45

circulated among the relevant consultation bodies and governmental actors. After consultation, the draft is prepared by the relevant ministry and discussed within the government, before it is finally agreed with the government whether the draft law can be submitted to the parliament. The lack of proper preparation of a draft law may manifest in a number of ways. For example, there may be the lack of preparatory documents: written terms of reference made by the government on the matter in hand are replaced by oral instructions. Reports and proposals of assigned committees, whether independent, constituting an organisational unit of the ministry or a group of experts, may also be missing. Reasons may be that there is no such unit assigned or, for instance, the drafting is outsourced and the documentation is not monitored or requested. It may also happen that the drafting is conducted by organisational units of the ministry, but these units do not make proper documentation of their activities because there is no regulation requiring it or they simply do not comply. RIA reports and documentation of consultation may also be lacking, even though there is a legal obligation to have them because there are no legal consequences and political will to enforce compliance. In the parliamentary system, most of the legislative proposals come from the government. Surveys show40 that all or the vast majority of bills (above 90 per cent) are initiated by the government in the UK, Finland, Macedonia, Romania and Slovenia. This practice strongly suggests that the rules on the preparation of drafts mentioned above, such as consultation and impact assessment, are most probably complied with, and the policy is made with the involvement of different actors. There is, however, an adverse trend in Hungary, where the previously measured 80 per cent of laws initiated by the government in 2006–10 decreased drastically to approximately 50 per cent in 2010–14. Instead, more and more bills were initiated by MPs. This, in itself, would not be an issue, but considering that the majority of these MPs belonged to the governing majority, it gives rise to the suspicion that the intention behind this change in practice was to avoid the necessary consultation and other processes.41 Nowadays, information technologies are used for interdepartmental cooperation, circulation and discussion of drafts among ministers (eg Georgia, Czech Republic), government meetings (eg Estonia) and decisions (eg Finland)—making the entire process more manageable. There are also tools to facilitate the drafting process. Such dedicated drafting aids are used in the Netherlands, Belgium and Italy, and they help to observe the drafting rules.42 However, even though there is high e-government capacity, including high utilisation levels of connected services and e-participation in, for example, from the CEE region, Hungary43 (in contrast with the Czech Republic and Estonia), there is no record of using more advanced and interactive ICT tools in legislative processes. It is probable that the development of ICT tools and the facilitation of legislative processes are promoted more in states where there is a greater commitment towards improving the quality of legislation. 40 

Kasemets (n 22) 53. See also G Bárándy, Centralizált Magyarország—megtépázott jogvédelem [Centralised Hungary—Manhandled Legal Protection] (Budapest, Scolar Kiadó, 2014). 42  For more, see W Voermans, W Fokkema and R VanWijk, ‘Free the Legislative Process of its Paper Chains: IT-Inspired Redesign of The Legislative Procedure Cycle’ (2012) 1 The Loophole 62. 43  United Nations E-Government Survey 2010, ‘Leveraging e-Government at a Time of Financial and Economic Crisis’ (New York, United Nations, 2010) 72, 86, 88, available at unpan1.un.org/intradoc/groups/public/ documents/un/unpan038851.pdf (accessed on 10 October 2011). 41 

46 

Tímea Drinóczi

3.4.2.  Legislation in Legislative Bodies Policy making also takes place in parliaments as, depending on the national rules, MPs and certain groups of MPs (committees, factions) may initiate a draft law or amendments to already submitted drafts. After the submission of the draft to the parliament, the parliament discusses the bill in different readings and debates, either in committees or at a plenary session, and, in the case of bicameral states, in the other chamber as well. The required majority and procedural rules for passing the bill are stipulated mainly by the constitution and the rules of procedure for the parliament; non-compliance with these rules leads to invalidity. Normally, bills are discussed first at the level of general principles, then at the detailed level. When the legislative initiative is submitted to the parliament, it goes through an electronic registration process and is published on the parliament’s website. The leading committee dealing with the bill is assigned by the speaker of the parliament. The parliamentary legislative process shows great variety in how bills are dealt with: some may be discussed in different readings or, if there is only one reading, the bill may be discussed in different debates; readings may have different goals and different numbers of readings may be required; alternatively, there may be no readings, but instead, general and detailed debates may be applied; and amendments may or may not be presented during each hearing, etc. Whatever the organisational and procedural rules of the parliamentary process may be, the purpose of the regulatory framework is to ensure that a bill is elaborated as much as ­possible. Committee hearings are usually followed by plenary meetings and discussion. Depending of the national parliamentary rules, in a unicameral parliament, bills are ­discussed in two or more readings or in two kinds of debates. Initially, the plenum may discuss only the main principles and necessity of the law (first reading), or it may discuss the entire draft. This latter is called a general debate and is based on the committee report. In each case, the initiator of the draft presents the draft law and explains its necessity. If the presenter is not a representative of the government, the ­government—along with the committee and others—may also present its opinion. It may happen that in the general debate there is no reference to any amendments discussed in the previous committee hearing. In this case, the plenum may adopt the draft law and the parliamentary stage of the legislative process ends here. However, in the majority of cases the process does not end like this, as there are proposed amendments to be discussed in the detailed debate of the plenum. The voting at the end of the general debate is about the acceptance of the principles, the necessity of the draft law and its suitability for a detailed debate. In the detailed debate, the discussion is about amendments (here, some kinds of amendments need to be supported by the proponent and the assigned committee) and amendments to the amendments; then a vote is taken on the amendments. There is then a final discussion of the draft and a final voting. Amendments substantially affecting the draft law may be allowed to be submitted or these kinds of amendments may be restricted. Usually in this phase only amendments of a technical nature are allowed in order to maintain and ensure coherence within the draft and in the legal system. There is a clear disadvantage of a rule allowing substantial changes (it is allowed in Hungary), because these newly proposed amendments cannot be properly considered and debated.

Legislative Process

 47

In other states (eg Bulgaria), the discussion of laws is carried out through different readings. As opposed to the above-mentioned process, the first reading in the committees and in the plenum concerns the principal need for and the general ideas and logic of the draft. The second reading in both parliamentary forums is related to the legal and technical details of the draft law. There is also a final reading, where the plenum discusses and votes on the proposed Act chapter by chapter, section by section and/or paragraph by paragraph. The approval of the draft in the second or final reading concludes the parliamentary phase of the legislative process. In bicameral parliaments, draft laws are discussed in up to a maximum of three readings. Depending on the national legislation, drafts may be submitted only to the lower chamber or to both chambers, and the first reading may be either of a technical character—where no discussion takes place, like in the UK—or may have a more substantial role. The best example for presenting the functions of the three readings is the German parliament. Here, draft laws have to be submitted to the lower chamber. In the first reading, the draft is evaluated and a decision is made as to whether the draft can be forwarded to the committee hearings. Amending proposals cannot be submitted in this phase. After the report has been made by the committee, a second reading is held. In this reading, amendments can be submitted. Amendments can also be submitted in the third reading, but only concerning those articles that were affected by amendments in the second reading. The third reading also serves as a final and global debate on the draft law. Final voting may be preceded by a committee hearing if the committee itself has made amendments in the second reading. Each reading ends with a vote; the third reading is closed by a final vote. In Germany, this process is carried out in the lower house of the parliament (Bundestag), which is followed by a debate in the other chamber representing the territorial units (Bundesrat). Whether there is a further process depends on the law at issue, as it may have a subject matter on which the Bundesrat has a suspensive veto or on which its approval is necessary. If it is necessary for a compromise to be reached, a joint committee is convened. The upper house in Austria functions in a similar manner. The process is similar in other parliaments. Some have only slight differences in the function of the readings: for example, in Austria, the second reading in the Nationalrat (lower chamber) is used to conduct the general and the detailed debates together, and in the third reading only technical amendments can be submitted. Other parliaments differ more widely: for example, draft laws may be submitted to both the lower and, in certain cases, the upper house of the parliament (eg Belgium, France), or the speakers of the chambers may decide which of them should start discussing the draft (eg Switzerland). In these states, chambers discuss texts and can submit amendments that are sent to the other chamber for further discussion. Sending texts back and forth is characteristic of these processes, and distinguishes such processes from that applied in Germany and Austria. Conciliation bodies may also assist in the decision-making process in these parliaments if needed. It should further be noted that the detailed rules of the national parliaments may reveal more significant differences in the legislative work of the upper houses.

3.4.3.  Post-parliamentary Stage and Constitutional Review When the final vote has been taken, each draft law is considered to have been adopted in any of the legislative processes mentioned above. When the Act/statute has been adopted,

48 

Tímea Drinóczi

it is generally signed by the speaker of the parliament. In some states, such as Germany, it is countersigned by the competent minister. In keeping with the determined deadlines, the adopted law is sent to the president of the republic for promulgation and publication, or to the monarch for royal assent (eg UK, Norway; a countersignature may also be needed). The promulgation is used to certify that the text sent for publication is identical to that of the adopted law (authentic). This ensures that the will of the representative body would not be unduly influenced. Publication is made in the official journal (or gazette) of the state. In this last, post-parliamentary, stage, the head of state may have the power to send the law back to the parliament for further consideration, which means a legislative impulse, which is either accepted by the parliament or refused. Another power that may be assigned to the president of the republic is to send the law for preliminary constitutional review. When there is a decision to put the law to referendum or the law is on a constitutional amendment that requires a referendum, the head of state must not promulgate and publish the law before knowing the result of the referendum. An initiative for preliminary constitutional review may also emerge at the end of the parliamentary process. In Hungary, for instance, it may also be initiated before the final voting and voted on after the final voting. ICT tools are used more and more to publish legislation.44 Electronic publishing cannot, however, be used exclusively—paper-based access still needs to be ensured, especially due to the digital divide and the right to access to laws. The volume of the piece of legislation that has to be printed cannot be a reason for not making paper-based official journals (as was the reason in Georgia). Were only electronic publishing to be available, then free access to copies (through printing) would also have to be provided. Electronic access has to be provided for all, free of charge, through a comprehensive system, including for secondary and other legislation. E-publication should be given an authentic status, even though not all states have authentic online copies yet; many have disclaimers on their website that they are not authentic. That is why the question of a mechanism remedying editorial (typing, numbering, etc) mistakes needs to be addressed not only in connection with paper-based publishing, but with electronic publishing as well.

3.5.  Summary of the Legislative Process, Trends and Challenges Against this background, researchers have detected some trends and addressed some issues in the field of legisprudence. When we talk about trends and the challenges a law maker (in a broad sense) needs to face, we should remember that what seems to be a new trend and challenge in Western Europe might (or might not) have45 reached the Eastern part of Europe.

44  For more, see Voermans et al (n 42); R Cormacain, ‘Have the Renton Committee’s Recommendations on Electronic Access to Legislation Been Fulfilled?’ (2013) 3 European Journal of Current Legal Issues, available at http://webjcli.org/article/view/264/361. 45  The case is, however, that almost all political and legal ‘innovations’ were introduced to the Eastern part with a considerable delay, mainly due to historical reasons.

Legislative Process

 49

Likewise, we should not forget about the necessity of political will. It has become a trend for governments and supra- and international organisations to pay much greater ­attention to better regulation, smart regulation and legislative quality, using IT tools in as many phases of legislation as possible, etc. Nevertheless, where the political commitment is missing, it is hardly possible to enact laws creating a proper framework for good legislation, and thus neither the process itself nor the content of the law will be of good quality. It is also difficult to implement any better regulation policy where the entire legal system is about inverse processes: centralisation against decentralisation, prioritising state efficiency against involvement and participation. The same may apply to transitional countries where the (legal) culture of democracy, rule of law and advocacy of human rights still do not seem to be mature. In both cases, policy makers need to find a balance between their policy goals and the requirements of better or smart regulation and legislative quality, and researchers of legisprudence and organisations advocating legislative quality need to offer as much assistance as possible in a graded manner that takes the actual situation of the state into account. Another issue concerning researchers is that there is a shift in the division of work between the parliament and the government in a constitutional democracy, resulting in the phenomenon that the executive is more ‘legislative’ in character than the legislative power. There are also complaints that secondary legislation outweighs statutory law by far, and this is a problem in a parliamentary system that contributes to the creeping loss of parliamentary power. These are quite challenging observations, and highlight the issues we need to address when dealing with the legislative process. Due to the challenges of the twenty-first century, there is an inevitable move towards more active law-making activity of the executive of states having a parliamentary governmental system. This move does not seem to be as clear in semi-presidential states, as the constitution itself renders the legislative competence of the parliament narrower and the legislative competence of the executive power broader, and this may not be an issue in a presidential governmental system, due to the strict constitutional definition of the system of checks and balances between the executive, legislative and judicial branches of the government.46 Each state action and interaction needs to be analysed within the respective constitutional framework, as all conduct, even that of the state, or the daily operation of any (state) organ needs to remain within the constitutionally prescribed context with its special country-specific checks and balances, monitoring and reviewing mechanisms. It may happen that a government starts to claim more (‘legislative’) power or even usurp parliament’s power, but in this case, unless the constitution is changed, the government is acting unconstitutionally. Maintaining this practice may lead the country out of the community of constitutional democracies or, if the constitution is changed, the model is then different, and requires different approaches and legislative rules and practice. Evidence-based or rational legislation47 is also a trend that is intended to ensure that policy making, the preparation of drafts and legislative decisions take place properly and 46  This chapter does not deal with presidential governmental systems, as it is unusual in European constitutional democracies. 47  See more in P Popelier, A Mazmanyan and W Vandenbruwane (eds), The Role of Constitutional Courts in Multilevel Governance (Cambridge, Intersentia, 2013).

50 

Tímea Drinóczi

are discussed based on alternatives and well-founded arguments, scientific evidence, and data stemming from, for instance, impact assessments and consultations. Evidence-based legislation is also a more advanced stage of ‘better regulation’, and as such requires not only political will and commitment to provide a proper regulatory framework and ensure proper implementation, but also awareness on the side of adjudication bodies, such as ordinary courts or the constitutional court. Without their clear demand that legislation needs to be based on evidence, state legislative actors may not comply with this requirement due to other political goals requiring faster legislation. It is also difficult to speak about any impact assessment on human rights in states where a proper cost–benefit RIA and/or an impact assessment on the budget have not been adequately conducted. Using ICT tools in legislative processes is also a trend that accompanies the development of e-democracy and e-government. It facilitates involvement, participation, decision making, communication and even drafting. The more commitment that can be seen at the political level, the more useful ICT tools may be in different legislative processes, especially when these processes are well managed. The challenge here is to create processes in a way that does not entail any duplication of activities and uses all the possibilities new technologies may offer, while taking into account the actual state of development of the respective state. For instance, it does not make much sense to introduce exclusively e-government initiatives (at the legislative level) when the digital divide is considerably huge in a state, even if using e-tools is a trend. Legislation (regulation) today is seen as a process to be managed, so the preparation and implementation of legislation can be seen as a project to which the concepts and techniques of project management can be applied. Regulatory management consists of culture (shared norms and values, and implicit assumptions that are held by a group in an organisation), structures (allocation of competences, tasks and responsibilities), processes (succession of regulatory activities), instruments (practical tools for regulatory policy and processes) and competences (the development and management of the knowledge, expertise and skills of the different actors in the regulatory process).48 Regulatory actions can only be managed when you are well aware of what is to be managed, why and how. Thus, a certain level of good legislation or legislative quality is needed for dealing with legislation and a project to be managed. It obviously presupposes an already existing practice of the organisation and implementation of drafting tasks, impact assessment, consultation, procedural rules, planning and coordination. In the parliamentary legislative process, which is to a certain extent a political (decision-making) process, there is always some room for improvement,49 especially because of the development of ICT, the need for addressing legislative quality issues (for example, Romania and Slovakia should enhance the transparency of their legislative processes; Latvia and Poland should enhance the transparency of committee hearings; committee bills need to be published in Germany; simplification should be introduced in Greece; more involvement is required in Latvia, France and Romania; and reduction of accelerated procedures is highly recommended in Greece and Romania), the emerging new mainstreams (for example, the gender budgeting approach and RIA are missing in Romania; ethical issues need to be addressed in the Netherlands and Slovenia; and in most European states 48  49 

See Chapter 4 of this book. See n 15.

Legislative Process

 51

lobbying activity has to be (more properly) regulated) and the always-present possibilities of encumbering the proper functioning of the legislature. As a very general observation, it may be stated that it seems that most of the basic legislative issues need to be addressed by most of the CEE states. In other states, the issue is rather the fine tuning of existing best practices. The daily operation of the parliament, which mainly depends on the related regulatory framework (which thus must also be addressed), may be organised in a way which could potentially impede, for instance, genuine discussion and involvement not only of civil society and stakeholders, but of MPs as well. These techniques, which should be avoided, are the following: extensive use of accelerated processes; not complying with parliamentary deadlines; distributing background materials and proposals at the very last minute; creating a regulatory background that does not offer sufficient time for genuine consideration and discussion; not inviting interested parties (no time for proper consideration and discussion within the particular interest groups and even parliamentary groups); scheduling plenary discussions at night; not publishing relevant information on meetings, agenda, invited persons, composition of committees, involved experts and preparatory notes; not having a properly operating intranet (limited transparency); not having units for analytical, impact assessment and drafting work; not providing staff or budget for hiring staff for these activities (lack of expertise or of an evidence-based approach); and not allowing clause-by-clause debate and voting on modifications, but providing only for a debate and vote on the whole package of modification. These practices may lead to the loss of legislative power and limitation of representative democracy, and may justify the claim that there is a trend towards the loss of parliament’s legislative power.

Further Reading Bogdanovskaia, I, ‘The Legislative Bodies in the Law Making Process’, available at www. nato.int/acad/fellow/97-99/bogdanovskaia.pdf. Chronowski, N, Drinóczi, T and Takács, T (eds), Governmental Systems of Central and Eastern European States (Warsawa, Wolters Kluwer Polska—OFICYNA, 2011). Cormacain, R, ‘Have the Renton Committee’s Recommendations on Electronic Access to Legislation Been Fulfilled?’ (2013) 3 European Journal of Current Legal Issues. Cormacain, R, ‘Retroactivity, Retrospectivity, and Legislative Competence in Northern Ireland: Determining the Validity of Janus-Faced Legislation’ [2015] Statute Law Review, doi: 10.1093/slr/hmu054. Drinóczi, T, ‘Concept of Quality in Legislation—Revisited: Matter of Perspective and a General Overview’ [2015] Statute Law Review, doi: 10.1093/slr/hmv008. Flückiger, Alexandre, ‘Effectiveness: A New Constitutional Principle’ (2009) No 50 Legislação. Karpen, U (ed), Evaluation of Legislation. Proceedings of the Fourth Congress of the European Association of Legislation (EAL) in Warsaw (Poland), June 15th–16th, 2000 (Baden-Baden, Nomos Verlag, 2002). Karpen, U, Gesetzgebungslehre—neu evaluirt. Legislatics—Freshly Evaluated (Baden-Baden, Nomos Verlag, 2006).

52 

Tímea Drinóczi

Kasemets, A, ‘Impact Assessment of Legislation for Parliament and Civil Society: A ­Comparative Study’, available at www.siseministeerium.ee/public/Kasemets_Aare.pdf. Kortmann, CAJM, Fleuren, JWA and Voermans, W, Constitutional Law of 10 EU Member States: The 2004 Enlargement (Dordrecht, Kluwer, 2006). Almeida, MT and Mader, L (eds), Proceedings of the 9th IAL Congress Quality of Legislation—Principles and Instruments. Lisbon, 24–25 June 2010 (Baden-Baden, Nomos Verlag, 2011). Müller, G and Uhlmann, F, Elemente einer Rechtssetzungslehre, 3rd edn (Zürich, Sculthess, 2013). Prakke, L and Kortmann, K, Constitutional Law of 15 Member States (Dordrecht, Kluwer, 2004). Risse, T, ‘Transnational Governance and Legitimacy’, available at http://userpage.fu-berlin. de/~atasp/texte/tn_governance_benz. Thornton, GC, Legislative Drafting (London, Butterworths, 1987).

4 Management of Legislation PATRICIA POPELIER

This chapter analyses the different elements that constitute regulatory management aimed at informed decision making. Failure or success in managing the decision-making process so as to produce high-quality regulation depends upon regulatory culture, structures, processes, instruments and competences. Overall, political support is a crucial point, as this determines regulatory culture, the budget for the development and maintenance of regulatory instruments, the effective implementation of regulatory processes and room for the building of competences. Particular attention is paid to regulatory management in multi-tiered legal systems, and EU Member States in particular. EU integration complicates regulatory management in all Member States, but the implications are even stronger in multi-tiered Member States, which face specific co-ordination issues.

4.1. Introduction Constitutions, when describing the legislative process, usually confine themselves to the parliamentary procedure. In practice, and in particular in parliamentary systems, debate and votes in parliament are usually based on decisions that were negotiated and concluded in earlier stages. Hence, in order to guard the quality of legislative outcomes, guarantees should be built into the entire legislative process. The Organization for Economic Co-­ operation and Development (OECD) encourages its member states to adopt an effective regulatory policy for reasons of economic competitiveness. However, there are also legal concerns. National constitutional courts, as well as the European Court of Justice and the European Court of Human Rights, tend to adopt a procedural-type review of legislative acts, in particular if they wish to avoid the delicate balancing of interests for which (national) parliaments are considered to be better placed.1 The legislator should therefore keep in mind that preparatory documents or the lack thereof can be used to either challenge or justify the proportionality of its laws before court.

1  See, eg, K Lenaerts, ‘The European Court of Justice and Process-oriented Review’, European Legal Studies Research Paper No 1/2012 (College of Europe, 2012); P Popelier and C Van De Heyning, ‘Procedural Rationality: Giving Teeth to the Proportionality Principle’ (2013) European Constitutional Law Review 230; P Popelier, ‘The Court as Regulatory Watchdog: The Procedural Approach in the Case Law of the ECtHR’ in P Popelier, A Mazmanyan and W Vandenbruwaene, The Role of Constitutional Courts in Multilevel Governance (Cambridge, Intersentia, 2013) 249–67.

54 

Patricia Popelier

The regulatory model en vogue points to an evidence-based type of law making. Underlying this is the rationalist assumption that political decisions can be made on the basis of objective information.2 In reality, time limits and political rationality and culture outweigh the concern for informed decision making.3 Careful preparation of laws is time-consuming, whereas politicians are under constant pressure4 and are therefore more interested in visible short-term results; and the political elite consists of a disproportionate number of lawyers who often lack methodological training.5 Also, the very concept of ‘objective information’ is challenged, as the framing of questions or the choice of methodology can lead to different outcomes,6 and even scientific experts are hindered by ‘bounded rationality’.7 Moreover, decisions need public support in order to be effective; therefore the most ‘rational’ decision is not necessarily the best one.8 All this should be kept in mind when designing a regulatory policy. For example, guidelines should be provided to ensure that experts are transparent about their methodology and assumptions. Formalisation of procedures is required where policy culture entails a risk of regulatory capture by privileged interest groups. In the literature, it is even suggested that ex ante evaluations be made mandatory only in the case of information deficits (when information can be derived from standardised sources or non-regulatees with no special interest), not in the case of information asymmetries (when information is in the possession of interested parties).9 Data gathering should be institutionalised so as to enable speedy fact-finding. Mixed commissions, consisting of lawyers, economists, social scientists and public administration representatives, may find useful compromises between scientific and political rationality.10 This is the case in Sweden, for example, where ‘Committees of Inquiry’ for the preparation of specific regulatory projects are often composed of experts, public administration representatives and even lobbyists.11 Finally, expectations should be kept low: representative democracy will always allow for a certain margin of discretion for policy makers to let political arguments prevail over arguments of scientific rationality or economical effectiveness. Hence, while acknowledging that law makers do not make perfect decisions,12 regulatory policy and the management of legislation should aim to adopt responsive and justifiable decisions.

2  C Knill and J Tossun, Public Policy. A New Introduction (London, Palgrave Macmillan, 2012) 195. See also E Bohne, ‘The Politics of the Ex Ante Evaluation of Legislation’ in J Verschuuren (ed), The Impact of Legislation (Leiden, Martinus Nijhoff, 2009) 63–64. 3  See Bohne (ibid) 65. 4  Knill and Tossun (n 2) 195. 5  K Van Aeken, ‘Pushing Evaluation Forward. Institutionalization as a Means to Foster Methodological Growth of Legislative Ex Ante Evaluation’ in Verschuuren (n 2) 122–23. 6  S Breyer, Breaking the Vicious Circle. Towards Effective Risk Regulation (Cambridge, MA, Harvard University Press, 1993) 42–47; Knill and Tossun (n 2) 195–96. 7  JJ Rachlinski and CR Farina, ‘Cognitive Psychology and Optimal Government Design’ [2002] Cornell Law Review 558. 8  ibid 592. 9  Bohne (n 2) 77–78. 10  See the recommendation in R Van Gestel and J Vranken, ‘Assessing the Accuracy of Ex Ante Evaluation through Feedback Research: A Case Study’ in Verschuuren (n 2) 225. 11 PA Svensson, ‘The Swedish Way to Create Lawful, Consistent and Uniform Legislation—a Summary Presentation’ in L Mader and C Moll (eds), The Learning Legislator (Baden-Baden, Nomos, 2009) 137–38; S Veit, ‘National Experiences with General Forms of Ex Ante Evaluation of Legislation: the Cases of Germany and S­ weden’ in Verschuuren (n 2) 184–86. 12  H Schäffer, ‘Towards a More Rational and Responsible Law Making Process’ in U Karpen (ed), Evaluation of Legislation (Baden-Baden, Nomos, 2002) 151.

Management of Legislation

 55

Quality guarantees are analysed in administrative sciences and organisational theories. In the last decades, New Public Management (NPM) has introduced organisational models in public management, as well as managerial methods taken from the private sector.13 But public organisations differ from private organisations in a variety of ways: they must consider a broad set of democratic and constitutional norms and values; they are accountable to a large group of citizens; and they must adhere to principles of transparancy, impartiality and predictability.14 For example, democratic values and accountability imply a parliamentary chain of governance, which restricts the possibilities to delegate legislative powers to the executive or regulatory agencies. In the present governance models, described as networks of state and non-state actors, these values are completed with notions of collaboration, subsidiarity, flexibility, non-coerciveness, co-ordination and adaptability.15 This brings a shift from traditional command-and-control rules to government-supported self-regulation.16 These values are often visible in the quality standards formulated by national governments and European institutions. For example, in the EU, transparency, subsidiarity and accountability are among the standards enumerated in the Mandelkern Report on Better Regulation and the 2001 White Paper on European Governance.17 Transparency and accountably are equally part of the UK and Irish principles of good regulation.18 The OECD criteria stress that regulation should produce benefits for society, taking into account economic, social and environmental effects.19 In Finland, quality standards even include the requirements that legislation must promote innovation, access to the market and productivity, and create competitive advantages for Finnish business, as well as promote good working conditions and well-being at work.20 It is important to formulate such standards: if regulatory management aims to adopt high-quality rules, it needs to build on some common shared ideas about which quality standards should be used. These should determine the management of the legislative process. For example, the impact assessment instrument may aim at better-informed decision making, as well as legitimate decision making through participation, transparency and justification.21 Essentially, better regulatory programmes, even where they are established primarily for reasons of economic competitiveness, imply management of the regulatory process.22 This encompasses culture, structures, processes, instruments and competences. In what follows,

13  On the relation between NPM and Better Regulation programs, see CM Radaelli and A Meuwese, ‘Better Regulation in Europe: Between Public Management and Regulatory Reform’ (2009) 87 Public Administration 646–48; P Popelier and V Verlinden, ‘The Context of the Rise of Ex Ante Evaluation’ in Verschuuren (n 2) 20–22. 14  T Christensen, P Laegreid, PG Roness and KA Rovik, Organization Theory and the Pubglic Sector (London, Routledge, 2007) 4. 15  O Lobel, ‘The Renew Deal: The Fall of Regulation and the Rise of Governance in Contemporary Legal Thought’ (2004) 89 Minnesota Law Review 371; O Lobel, ‘New Governance as Regulatory Governance’ in D LeviFaur, Oxford Handbook of Governance (Oxford, Oxford University Press, 2012) 66. 16  Lobel, ‘New Governance as Regulatory Governance’ (ibid) 65. 17  Mandelkern Group on Better Regulation, Final Report (13 November 2001) 9–10; Commission, White Paper on European Governance, COM (2001) 428 final, 10–11. 18  As laid down in Arts 2 and 21 of the Legislative and Regulatory Reform Act 2006 (UK); Roinn an Taoisigh, Better Regulation (Dublin, 2004). 19 OECD, Recommendation of the Council on Regulatory Policy and Governance (Paris, OECD, 2012) 4. 20 OECD, Better Regulation in Europe: Finland (Paris, OECD, 2010) 43. 21  See the rationales laid down in the EU Impact Assessment Guide, as analysed by P Larouche, ‘Ex Ante Evaluation of Legislation Torn among its Rationales’ in Verschuuren (n 2) 42. 22  Radaelli and Meuwese (n 13) 646.

56 

Patricia Popelier

I will go through these aspects, which are borrowed from organisational studies (section II). I will then focus on regulatory management in multi-tiered legal systems (section III). Most of the examples in what follows derive from regulatory management systems developed at the executive level. The development of regulatory policies at this level derives naturally from the predominant role of governments in the making of laws. However, the role of parliament should not be underestimated. The development of regulatory institutions and skills at the parliamentary level is important for securing the quality of parliamentary initiatives, but above all for checking the quality of bills and pointing to the government’s responsibility in this respect. In Switzerland, the constitution even explicitly assigns the task of the evaluation of laws to parliament. Article 170 of the Swiss Constitution states that ‘The Federal Assembly shall ensure that federal measures are evaluated with regard to their effectiveness’. In France, the National Assembly’s Commission for Laws issued the Warsmann report,23 which revived simplification and ex ante evaluation policy.24 Therefore, aspects of regulatory policy at the parliamentary level will also be mentioned.

4.2.  Aspects of Regulatory Management Regulatory management is the implementation of policies and the monitoring of processes and staff to ensure that an organisation can achieve the objective of producing high-quality regulation. Failure or success depends upon regulatory culture, structures, processes, instruments and competences. Political support is crucial for the success of regulatory management in all these aspects. It is political leadership that determines the regulatory culture; the budget for the development and maintenance of regulatory instruments; the implementation of regulatory processes; and the building of competence through training and help desks, etc.

4.2.1. Culture ‘Culture’ refers to shared norms, values and implicit assumptions that are held by a group in an organisation and determine how this group perceives its environment and how it acts.25 Applied to regulatory policy, it requires that legislative quality standards are valued and used to guide action and behaviour in the public body. It does not suffice to couch regulatory quality standards in official documents and formal procedures; they should also be internalised as commonly shared, informal values. Leadership is of pivotal importance in this respect: leaders play an important role in both the protection and the changing of informal norms and values.26 Therefore, the commitment of ministers and leading officials

23  La simplification du droit au service de la croissance et de l’emploi (Paris, July 2011), albeit at the request of the President, M Sarkozy. 24 OECD, Better Regulation in Europe: France (Paris, OECD, 2010) 47, 60. 25 EH Schein, ‘The Missing Concept in Organization Studies’ (1996) 41 Administrative Science Quarterly 229, 236. 26  Christensen et al (n 14) 47.

Management of Legislation

 57

to regulatory quality standards and procedures is determinative of the failure or success of regulatory policies. For example, it was stated that ‘The content of, and formal political commitment to regulatory reform policies in the UK demonstrate a high level of consistency’ with the OECD recommendations on the quality of government regulations.27 In Germany, it is observed that political support for better regulation remains limited, resulting in scarce resources and weak organisational anchoring; instead, the political focus is on the reduction of administrative burdens caused by regulations.28 As a result, regulatory impact analysis in Germany often consists of a mere ‘box-ticking’ exercise, except when it comes to the description of administrative costs for enterprises.29 The OECD country reports on Better Regulation reveal the importance of regulatory culture and leadership commitment. For example, in Belgium, it was observed that ‘the involvement of politicians in rule drafting makes the implementation of impact assessment particularly difficult’. Impact assessments are drafted ex post to justify decisions that are already reached, instead of guiding the decision-making process. The conclusion was that ‘strengthening impact assessments will require strong high-level commitment and further culture change’.30 In Finland, the political culture based on consensus and corporatism was found to frustrate transparency and the taking into account of interests of all ­stakeholders.31 The need for a change in administrative culture is experienced in particular in East European countries, along with the transformation to a market democracy.32 In contrast, the Danish political culture, based on values of openness, consensus and widespread participation, was found to be fundamental for the development of legislative quality improvement programmes.33 In Sweden as well, long-standing consultation practices are rooted in an administrative culture of openness.34 Although nation-specific variations in policy style should not be overestimated,35 policy style does seem to play a role in the establishment of political and administrative culture, which impacts on the regulatory practice. In the UK, Better Regulation programmes prospered in a culture of market openness, low government interference and (usually) a ruling party with a clear majority in Parliament.36 On the other hand, consociational practices, characteristic of corporatist consensus democracies and resulting in partitocracy and opaque and elitist decision making,37 do not provide the most fertile soil for transparent, plural and evidence-based regulatory policy. The OECD reports on Belgium and Finland mentioned above are illustrative thereof. In particular, the high level of

27 OECD, Regulatory

Reform in the UK (Paris, OECD, 2002) 15. Veit (n 11) 182–84. ibid, 189–96. 30 OECD, Better Regulation in Europe: Belgium (Paris, OECD, 2010) 51. 31 OECD, Regulatory Reform in Finland (Paris, OECD, 2003) 6–7, 47. In 2010, the Finnish political culture was still described as leading to informality and corporist attitudes, and, although the financial crisis had triggered a more strategic approach to better regulation and effective consultation, key stakeholders were still favoured, OECD, Better Regulation in Europe: Finland (n 20) 56. 32 OECD, Regulatory Reform in the Czech Republic (Paris, OECD, 2001) 37. 33 OECD, Regulatory Reform in Denmark (Paris, OECD, 2000) 5, 8. 34 OECD, Better Regulation in Europe: Sweden (Paris, OECD, 2010) 49, 59. 35  See Knill and Tossun (n 2) 34. 36 OECD, Better Regulation in Europe: United Kingdom (Paris, OECD, 2010) 60. 37  For Belgium, see BG Peters, ‘Consociationalism, Corruption and Chocolate: Belgian Exceptionalism’ [2006] West European Politics 1079. 28  29 

58 

Patricia Popelier

partitocracy in ­Belgium seems to hinder the institutionalisation of an effective evaluation policy.38 The complex negotiation process in order to find compromises across ideologically diverse coalition partners, across language borders and with social partners fears regulation actors, such as evaluators, who might function as additional veto players, make the negotiation positions of the consociational partners transparent and strengthen the position of the administrative services.39 In Germany as well, the political culture, with a highly politicised and closed consensus-building process, is suspected of hindering serious impact assessment.40 On the other hand, consensus democracy can fuel regulatory programmes if it is organised on values of openness and participation, as the Danish example shows. Legal culture also plays a role. For example, the UK constitutional culture, non-codified and based on conventions, seems to result in a less formal legal system that is open to selfregulation,41 whereas self-regulation is less customary in formalised legal systems such as those of Belgium and France. It is suggested in the literature that common law systems take a more instrumental perspective on rule making, and are therefore more open to better regulation programmes which imply regulatory review.42 In Germany, the OECD noted that the legal culture resulted in an extremely formal-legalistic concept of regulatory quality.43 In 2010, improvements were observed that resulted from a determined policy to appoint non-legal civil servants.44 In Finland, a legalistic tradition combined with a large number of high officials with a legal background is said to explain the equation of policy and law making, and the adoption of specific and detailed command and control laws.45 Consultation practices in various legal orders reveal more—in particular, the effects of culture. In Flanders, characterised by partitocracy, corporatism and elitist decision making, informal consultations are often conducted by Cabinet staff rather than administrative staff, and lack any transparency as to access, participants and content. While they appear to have more impact than formal consultations with standing advisory councils, they are also highly sensitive to regulatory capture, allowing for specific interests to manipulate the content of laws to their advantage.46 Formal consultations are considered constitutive of unpleasant delay, as they are obligatory but occur at the end of the preparatory process, when decisions have already been taken after negotiation with political and social partners. Hence, transparent consultations with expert and representative bodies are conducted pro forma.47 If the consultation requirements were lifted, the administrative staff would expect that the advisory councils would not be consulted, for lack of a spontaneous ­deliberative culture.48 In contrast, in the Netherlands, due to the widespread consultation culture,

38  F Varone, S Jacob and De Winter, L, ‘Polity, Politics and Policy Evaluation in Belgium’ (2005) 11 Evaluation Review 261, 261–64. 39  ibid 262. 40 OECD, Bessere Rechtsetzung in Europa: Deutschland 2010 (Paris, OECD, 2010) 85. 41 OECD, Regulatory Reform in the UK (n 27) 31. 42  R Hoppe, ‘Ex Ante Evaluation of Legislation: Between Puzzling and Powering’ in Verschuuren (n 2) 84. 43 OECD, Regulatory Reform in Germany (Paris, OECD, 2004) 8, 52. 44 OECD, Bessere Rechtsetzung in Europa: Deutschland 2010 (Paris, OECD, 2010) 58. 45 OECD, Better Regulation in Europe: Finland (n 20) 56. 46  P Popelier, P Van Humbeeck, A Meuwese and K Van Aeken, ‘Transparant consulteren in Vlaanderen: de spanning tussen rationeel wetgevingsmodel en besluitvormingspraktijk’ [2012] Tijdschrift voor Wetgeving 9. 47  P Popelier, K Van Aeken, P Van Humbeeck and A Meuwese, Kwaliteitsvolle regelgeving door middel van consultatie van belanghebbenden (Brussels, Flemish Government, 2010) 95–96. 48  ibid 109. Results from a focus group of administrative staff and members of standing advisory councils.

Management of Legislation

 59

a radical cutting in formal consultation requirements did not fundamentally change consultation practices.49 In Denmark as well, culture ensures that processes of consultation, while not formally specified in legislation, ‘are widely observed and there is a high level of public confidence that they are followed’.50 Likewise, in the UK, there is no general legal requirement for consultation, but public consultation is ‘highly used’ and ‘based on a long tradition of pragmatic and flexible approaches to effective consultation’.51 In Ireland as well, a strong tradition of public consultation ensures the embedding of consultations on draft regulation in the administrative culture.52 Where culture, then, becomes an obstacle for transparent, participatory and evidencebased law making and is difficult to change, for example because of the political context, resort must be taken to institutionalisation:53 to systematisation through formal organisations, rules and procedures,54 and transparent measurement of ministries’ performances.55 For example, when, in France, strong scepticism towards the use of regulatory impact assessments grew among government departments and senior officials,56 an Institutional Act inserted the obligation to submit executive bills to impact studies,57 under the close ex ante scrutiny of the Council of State.58 Empirical study has indeed observed a close correlation between institutionalisation and culture: those countries that intensively institutionalised the evaluation process experienced an increase in the degree of maturity of evaluation praxis.59 Further, the creation of new regulatory bodies may advance regulatory culture, as they establish a new administrative culture in specialised areas. For example, in Portugal, it was observed that the energy regulator provided a best practice regarding open consultation.60

4.2.2. Structures ‘Structure’ refers to the allocation of competences, tasks and responsibilities for the implementation of regulatory policy. In particular, the quality of regulation and the observance of regulatory processes should be clearly identified as a responsibility of ministers and leading officials.61 For example,

49  P Popelier, R Van Gestel, K Van Aeken, V Verlinden and P Van Humbeeck, Consultaties in de w ­ etgevingspraktijk (Brussels, Politeia, 2008) 101, 105. 50 OECD, Regulatory Reform in Denmark (n 33) 9. Meanwhile, standing procedures including public consultations are laid down in ministerial guidelines, OECD, Better Regulation in Europe. Denmark (Paris, OECD, 2010) 60. 51 OECD, Regulatory Reform in the UK (n 27) 25. 52 OECD, Better Regulation in Europe: Ireland (Paris, OECD, 2010) 81. 53  Van Aeken (n 5) 127. 54  Varone et al (n 38) 255. 55 OECD, Better Regulation in Europe: Netherlands (Paris, OECD, 2010) 50. 56 OECD, Regulatory Reform in France (Paris, OECD, 2004) 10. 57 Art 8 of the Institutional Act of 15 April 2009 (Loi organique n° 2009–403 relative à l’application des articles 34–1, 39 et 44 de la Constitution), specifying the content of the impact assessments. 58  And to a much lesser degree the Constitutional Court. See J-P Duprat, ‘The Judicial Review of Ex Ante Impact Assessment in France: An Attempt to Fuse the Principles of Legal Certainty and Institutional Balance’ [2012] Legisprudence 384, 384–86, 388. 59  Varone et al (n 38) 260. 60 OECD, Better Regulation in Europe: Portugal (Paris, OECD, 2010) 58. 61  See, for example, the recommendation to give regulatory quality responsibility to a ‘political champion’ who will provide a voice in Cabinet meetings, follow up on Better Regulation issues and support the administrative structures in fulfilling their mandates, OECD, Better Regulation in Europe: Greece (Paris, OECD, 2012) 17, 47.

60 

Patricia Popelier

in the Netherlands, the Minister of Justice is responsible for the overall legislative quality policy and assesses the quality of proposed regulatory measures.62 However, general responsibility is best placed at an overarching entity, such as a federal chancellery or Prime Minister’s Cabinet, with a co-ordination function, able to impose policy across administrative departments.63 In legal systems with strong ministerial autonomy, like the Nordic ones, an alternative is found in the construction of a co-operative cross-ministerial network, where ministers with a key role in the Better Regulation policy meet.64 Responsibility for particular tasks can be spread amongst administrative staff, including help desks, consultation staff and drafters, and autonomous bodies, such as advisory councils and impact assessment audit bodies. The existence of formal bodies for regulatory tasks is an indicator of the level of institutionalisation of regulatory policy.65 This, in turn, ensures professionalisation, facilitates co-ordination in multi-tiered systems and fosters a regulatory culture.66 Legislative policy implies choosing the type of structure for each regulatory task. A fundamental question is whether these tasks should be entrusted to internal staff or external bodies. In the Netherlands, former plans to turn the Ad Hoc Independent Advisory Board on Administrative Burdens into a permanent supervisory and strategic advisory body on all aspects of regulatory pressure were recently changed by the new government, which felt that these aspects should remain an integral part of the departmental policy processes.67 Another question is whether or not to centralise regulatory tasks. For example, in the UK, a Cabinet subcommittee, the ‘Reducing Regulation subCommittee’, has to give clearance whenever a proposed measure regulates or deregulates business, or concerns the regulation of business.68 It is part of the UK Better Regulation Executive, which is considered ‘one of the best examples of an effective central regulatory unit across the OECD’.69 In Germany, the Better Regulation Unit is situated within the Federal Chancellor’s Office.70 Centralisation may serve the political role of avoiding deviating or duplicating departmental policies and influencing the exercise of departmental policies.71 Meanwhile, it ensures the consistent implementation of regulatory policy. Therefore, it is recommended that, for example, the task of assessing the quality of impact assessments is situated at the centre of government.72 In practice, indeed, this is where regulatory oversight bodies are usually located.73 The Flemish example shows that decentralisation leaves the success of structural innovations such as departmental ‘cells for legislative quality’ dependent on the support of the departmental leading officials. In what 62  Guideline 254 of the circular Aanwijzingen voor de Regelgeving. See SE Zijstra (ed), Wetgeven (Deventer, Kluwer, 2012) 113–14. 63  See the recommendations in OECD, Better Regulation in Europe: Austria (Paris, OECD, 2010) 56–57. 64  See, eg, in Finland: OECD, Better Regulation in Europe: Finland (n 20) 63. 65  Varone et al (n 38) 255. 66  ibid 255. 67  ‘Goed geregeld, een verantwoorde vermindering van regeldruk 2012–2017’ (25 April 2013) 12, available at www.rijksoverheid.nl/documenten-en-publicaties/kamerstukken/2013/04/24/kamerbrief-goed-geregeld-­eenverantwoorde-vermindering-van-regeldruk-2012-2017.html. 68  Department for Business Innovation and Skills, ‘Better Regulation Framework Manual’ (London, 2013) 8–9. 69 OECD, Better Regulation in Europe: United Kingdom (n 36) 56. 70  Veit (n 11) 182–83. 71  Hoppe (n 42) 96–97. 72 OECD, Better Regulation in Europe: Austria (n 63) 57. 73 JB Wiener and A Alemanno, ‘Comparing Regulatory Oversight Bodies across the Atlantic’ in S Rose-­ Ackerman and P Lindseth, Comparative Administrative Law (Edward Elgar 2010).

Management of Legislation

 61

follows, I give but a few examples of ‘structures’ with specific tasks in the implementation of regulatory policy. Ex ante scrutiny of compliance with higher rules is often conferred to a specialised body. This can be an independent body or even, within the parliamentary procedure, a standing parliamentary commission. In Belgium, France, Luxembourg and the Netherlands, an independent ‘Council of State’ composed of legal experts provides non-binding opinions on the legal and drafting quality of draft government bills and regulatory acts and—on a voluntary basis—proposals of Members of Parliament. In Sweden, a special Council on Legislation, consisting of acting and retired judges of the Supreme Court and the Supreme Administrative Court, provides non-binding opinions on the constitutionality of draft legislation.74 A mixed body exists in Bulgaria, where the Legislative Council that examines draft legislation consists of university professors, judges, attorneys at law and high government officials, and is chaired by a member of the executive.75 In Finland, on the other hand, a parliamentary commission, the Constitutional Law Committee, delivers authoritative opinions on the conformity of bills with the constitution and international human rights treaties.76 These opinions constitute a political review, based upon opinions by constitutional experts and binding on Parliament unless the Committee’s ruling is overruled by a Statute of Exception adopted with a qualified majority similar to an amendment of the Constitution.77 In the UK, ex ante scrutiny is also conferred to parliamentary committees assisted by legal experts. The House of Lords’ Select Committee on the Constitution examines bills for constitutional implications;78 the Joint Committee on Human Rights has the specific task to examine the human rights implications of draft legislation.79 The Committee on the Constitution’s growing influence is attributed partly to the appointment of permanent legal advisors to the Committee.80 The drafting of legislation is, in several legal systems, the responsibility of a specialised team of drafters. In particular, the Netherlands and the UK are worth mentioning, as in both of these states drafters receive substantial training. In the Netherlands, a specialised ‘academy’ was even established to provide training for drafters. In contrast, in, for example, Belgium and France, ‘drafting’ is not identified as a specific function, and the drafting of legislation is considered part of the many tasks of the administrative staff.81 This makes it 74  T Bull, ‘Judges without a Court—Judicial Preview in Sweden’ in T Campbell, KD Ewing and A Tomkins (eds), The Legal Protection of Human Rights. Sceptical Essays (Oxford, Oxford University Press, 2011) 398–407; G Edelstam, ‘National Legal Tradition—Sweden’ in S Galera (ed), Judicial Review. A Comparative Analysis inside the European Legal System (Strasbourg, Council of Europe Publishing, 2010) 139; Svensson (n 11) 141. 75  F Novák, ‘The Role of Legal Experts in the Legislative Process in the Czech Republic’ in L Mader and U Karpen (eds), The Participation of Civil Society in the Legislative Process (Baden-Baden, Nomos, 2005) 178–79. 76  Arts 42(2) and 74 Finnish Constitution. 77  K Tuori, ‘Judicial Constitutional Review as a Last Resort’ in Campbell et al (n 73) 380. 78  Based on a Code of Constitutional Standards developed by the Committee, see J Simson Caird, R Hazell and D Oliver, The Constitutional Standards of the House of Lords Select Committee on the Constitution, 2nd edn (London, UCL, 2015), available at www.ucl.ac.uk/constitution-unit/publications/tabs/unit-publications/164 (last accessed on 16 September 2015). 79  See D Feldman, ‘The Impact of Human Rights on the UK Legislative Process’ (2004) 25 Statute Law Review 91. 80  J Simson Caird, ‘Parliamentary Constitutional Review: Ten Years of the House of Lords Select Committee on the Constitution’ [2012] Public Law 10. 81  For France: OECD, Better Regulation in Europe: France (n 24) 66. For Belgium: P Popelier and V Verlinden, ‘Framework for a Country Comparative Perspective: Belgium’ in Mader and Moll (n 11) 122. See the survey on drafting in the Flemish Community, V Verlinden, ‘Wie stelt de wet in Vlaanderen? Een reflectie naar aanleiding van een enquête over de regelgevende capaciteit bij de Vlaamse administratie’ [2005] Tijdschrift voor Wetgeving 378.

62 

Patricia Popelier

difficult to build expertise and to give specialised training. On rare occasions, the drafting of legislation is contracted out. This is more often the case in smaller jurisdictions, with limited personnel, resources and expertise.82 Another example concerns ex post evaluation policy. Varone, Jacob and De Winter developed an institutionalisation index for measuring the level of institutionalisation of policy evaluation, assigning a score to the existence of evaluation bodies within the Executive, within Parliament and within the supreme audit institution; and the existence of an epistemic community, including a national society of evaluators, scientific review on evaluation and quality standards.83 They combined this with an index measuring the maturity of an evaluation culture. Australia, Canada, France, the Netherlands, Norway, Sweden, Switzerland, the UK and the US were indicated as countries with a high level of institutionalisation and maturity. Hence, these countries may serve as models for other legal systems that wish to implement better evaluation practices. When institutionalising evaluations, it is important to carefully consider the type of body to be established, and in particular the distance created between legislator and evaluator, as this is bound to impact on the utilisation of the evaluation outcomes. Distance increases if (i) the evaluation is triggered, (ii) launched and followed through, (iii) financed and (iv) carried out by a unit external to the body or department responsible for the measure under evaluation; if (v) an order to conduct the evaluation is allocated through an invitation to tender; and if (vi) the evaluation report is published.84 Balthasar distinguishes between instrumental, process-related, conceptual and symbolic utilisation. His empirical study conducted in Switzerland gave evidence that the first two types benefit from a close link between legislator and evaluator, while the latter two types increase with the distance between them. Instrumental utilisation concerns the implementation of recommendations. This is enhanced when those responsible for the measures have to pay for the costs of the evaluation. Process-related utilisation implies the development of new insights in the area of the evaluated measure and the building of new networks, which outlast the evaluation as such. Evaluations prepared and conducted within the organisation have a better chance of producing process-related utilisation effects. Conceptual utilisation takes place if the legislator changes his opinion, attitude or ideas regarding the law as the consequence of an evaluation. Here, the quality of the evaluation determines utilisation, which is expected to benefit from external evaluators as they have better methodological knowledge and experience. Finally, symbolic utilisation refers to the legitimacy-providing effect of the evaluation, for example when it has strengthened support for the measure.85 Here, indications are not clear, but it is considered that symbolic utilisation increases with the distance between legislator and evaluator.86

82  See in this respect T St J Bates, ‘Experts in the Legislative Process in the British Isles’ in Mader and Karpen (n 75) 124–25; see also OECD, Better Regulation in Europe: Luxembourg (Paris, OECD, 2010) 49. 83  Varone et al (n 38) 257. 84  A Balthasar, ‘Institutional Design and Utilization of Evaluation: A Contribution to a Theory of Evaluation Influence Based on Swiss Experience’ [2009] Evaluation Review 232. 85  But compare with the more contemptuous term of ‘legislative entrepreneurship’ in Larouche (n 21) 58, regarding ex ante evaluations: these can be used as a business plan, with the purpose to convince potential partners, rather than an objective analysis. 86  Balthasar (n 84) 235–56.

Management of Legislation

 63

In the literature, the crucial role of evaluation institutions at the parliamentary level is underlined, in order to activate a disinterested government.87 For example, in Switzerland, the parliament, aided by the Parliamentary Administrative Audit Unit, may ask the executive to conduct evaluations, examine the quality of evaluations conducted by the executive or conduct its own evaluations.88 In Ireland, parliament regularly asks questions about impact assessments and parliamentary researchers use impact assessments as the basis for briefing on bills, which has raised the profile of impact assessments.89 However, if the position of parliament is particularly weak—for example, if the stability of a dyadic, multinational polity such as Belgium requires parliament to sustain the stability of the coalition—the existence of a parliamentary evaluation body does not suffice. In Belgium, a law creating a standing interparliamentary committee for the evaluation of laws was adopted in 2007,90 but the committee only started functioning in November 2011, and year reports do not reveal a significant level of activity.91 ‘Distance’ is also a factor to be considered when institutionalising ex ante evaluation. It is recommended that impact assessments are carried out by the administration, so as to encourage deliberation on the various options.92 In addition, however, several legal systems opt for a more independent form of check on the quality of the assessments, seeking a balance between ‘distance’ in order to guarantee technocratic neutrality and ‘closeness’ to those in power in order to have authority over ministries.93 For example, in the UK, an advisory ‘Regulatory Policy Committee’ of independent experts assesses the quality of evidence and analysis in impact assessments. In the EU, an Impact Assessment Board, composed of high-level officials independent of the policy-making departments under the authority of the Commission President, issues opinions on the Commission’s impact assessments.94 Another type of control can be found in France, where, as mentioned above, the Council of State is explicitly involved in the control of impact assessments. The Council’s check on whether a proposal is accompanied by an impact assessment is not of a mere formal nature, as the Council does not hesitate to criticise failings or incompleteness of the impact assessment.95 In the Netherlands as well, the Council of State plays a role: having been given the explicit task of assessing the policy-analytical aspects of legislative drafts, it substitutes for an impact assessment board.96 Here as well, utilisation of the evaluation is linked with the structure of the ex ante evaluation process. For example, regulatory quality bodies that perform systematic

87 

Varone et al (n 38) 264–66. L Mader, ‘The Drafting Procedure for Laws and Ordinances in Switzerland’ in Mader and Moll (n 11) 187. 89 OECD, Better Regulation in Europe: Ireland (n 52) 76. 90  Law of 25 April 2007 concerning the establishment of a Parliamentary Committee regarding the legislative evaluation, Official Gazette 11 May 2007. 91  Year reports for 2011, 2012 and 2013–April 2014 can be found at the Commission’s website, http://www. comitesuivilegislatif.be/indexF.html (last accessed on 16 September 2015). 92 OECD, Better Regulation in Europe: France (n 24) 92: this requires investments in training so as to develop assessment skills. 93  Wiener and Alemanno (n 73). 94  See A Meuwese and L Senden, ‘European Impact Assessment and the Choice of Alternative Regulatory Instruments’ in Verschuuren (n 2) 147–48; Wiener and Alemanno (n 73). 95  Duprat (n 58) 385. 96  Van Gestel and Vranken (n 10) 202. 88 

64 

Patricia Popelier

e­ valuations of the impact assessments foster instrumental utilisation.97 Political-strategic use—for which some evidence is found in the UK and the Netherlands—implies a topdown approach, with the purpose of enhancing political control by the central executive over the regulatory departments.98

4.2.3. Processes The term ‘processes’ refers to the succession of regulatory activities. Proceduralisation of regulatory activities such as impact assessments and consultations makes these activities a natural step in the law-making procedure. This enhances both the use and the impact of regulatory instruments. For example, empirical study in Switzerland has demonstrated that the general utilisation of evaluations increases if the evaluation is part of the routine.99 Process management implies the co-ordination of the various activities, aimed at the implementation and improvement of the regulatory process. The public administration literature visualises this in the so-called PDCA cycle, consisting of four phases: planning (Plan), implementation (Do), evaluation (Check) and adjusting (Act).100 Several parts of this cycle are discussed in other chapters in this book, such as impact assessments and consultation as parts of the planning phase, as well as ex post evaluation. Processing implies that they form part of a default procedure that is written out and followed as a daily practice. In the Netherlands, an overview of this procedure, with references to responsible departments and contact details, is laid down in a ‘scenario book’ (Draaiboek voor de regelgeving). Process management allows for the careful planning of separate regulatory projects, inserted in a regulatory agenda. An example can be found at the EU level. This starts with a yearly working programme,101 presenting proposals for law making, White Papers, simplification initiatives, etc, and continues with updated reports presenting a calendar for proposed measures to be introduced to other bodies as well as overviews of measures that have been taken, or withdrawn initiatives—for example, if the impact assessments make clear that the proposal is not ready for adoption.102 For individual regulatory initiatives, a roadmap is constructed, with a description of the planned initiative and the planned impact assessment. In turn, a blueprint for impact assessment procedures, including consultation activities, is laid down in guidelines.103 Process management has the additional advantage of furthering legitimacy by enhancing the role of parliament in the legislative process. For example, the European Parliament discusses the Commission’s working programmes, which allows it to insert its own priorities.104 In Denmark, an annual law programme that

97  CM Radaelli, ‘Measuring Policy Learning: Regulatory Impact Assessment in Europe’ (2009) 16 Journal of European Public Policy 1148. 98  ibid 1152, 1157–59. 99  Balthasar (n 84) 250. 100  For the application of Deming’s PDCA circle on the law making process, see Popelier and Verlinden (n 13) 25–26. 101  On the basis of Art 2 of the Rules of Procedures of the Commission, C[2000] 3614. 102  For a more detailed overview, see A Meuwese, ‘De EU aanpak van de wetgevingsagenda: werkprogramma’s & roadmaps’ in P Popelier and J Van Nieuwenhove (eds), Regelgeving & Tijd (Bruges, Die Keure 2009) 87–98. 103 Commission, Impact Assessment Guidelines, SEC(2009) 92. 104  ibid 91.

Management of Legislation

 65

includes a detailed list of bills to be initiated in parliament in the year to come is discussed in the parliament and aims at the early and close engagement of parliament.105 Process management is concerned with the sequence of regulatory activities, and hence with timing. For example, regulatory review usually has little impact if it takes place at a late stage of the legislative process. This is the case for the policy-analytic review by the Dutch Council of State,106 the consultation reports by formal advisory bodies in Belgium107 and public consultations in Finland.108 If processes are not sustained by culture, they can be phrased as a formal requirement in legal texts. In France, the constitution provides that parliament can refuse discussion of a bill if the impact assessment is of inferior quality. If government and the assembly’s ‘Conference’ disagree, the Constitutional Court can settle the dispute. In practice, however, parliament misses controlling opportunities because the ‘Conference’ is usually of the same majority as the government.109

4.2.4. Instruments ‘Instruments’ are practical tools for shaping regulatory policy and processes.110 Such tools include drafting guidelines, roadmaps, impact assessment forms and guidelines for impact assessments,111 guidelines for consultations, training modules, help desks and even computer programs.112 With respect to the latter, Austria has played a pioneering role in the development of the E-LAW system for the management of the drafting of laws.113 Guidelines are probably the first instruments that come to mind. They are usually laid down in internal documents and ministerial circulars. In some cases, basic principles or a legal basis are inserted in legislative acts. For example, principles of good regulation are inserted in the UK Legislative and Regulatory Reform Act 2006. In France, the Institutional Act provides that government bills are submitted to impact studies and enumerates the information to be included in the impact study report.114 Article 39 of the French Constitution provides that government bills are not to be tabled if they do not comply with the Institutional Act. Prescriptive rules, however, should remain limited and not go into much 105 OECD, Better

Regulation in Europe. Denmark (n 50) 70, 81. Van Gestel and Vranken (n 10) 223. Popelier, ‘Consultation on Draft Regulations: Best Practices and Political Objections’ in L Mader et al, Quality of Legislation: Principles and Instruments (Baden-Baden, Nomos, 2011) 136–47; OECD, Better Regulation in Europe: Belgium (n 30) 97. 108 OECD, Better Regulation in Europe: Finland (n 20) 77, 85. 109  Duprat (n 58) 387–88. 110  For an overview, see G Müller and F Uhlmann, Elemente einer Rechtssetzungslehre (Zurich, Schulthess, 2013) 58–71. See also Radaelli and Meuwese (n 13) 640. 111 We use the term for different types of assessments of potential impact and effects of proposals. In Switzerland the term ‘impact assessment’ did not find root to replace the ‘Gesetzesfolgenabschätzung’, Müller and Uhlmann (ibid) 62, but there are no fundamental differences between both instruments, L Mader, ‘Stand und Entwicklungstendenzen der Gesetzesfolgenabschätzung in der Schweiz’ in S Hensel, K Bizer, M Führ and J Lange (eds), Gesetzesfolgenabschätzung in der Anwendung (Baden-Baden, Nomos, 2010) 249. 112  For the drafting of legislation as well as document management, see R De Busser, ‘Automated Legislative Drafting: an Overview’ in M-F Moens, Digital Legislation (Bruges, die Keure 2002) 111–32; C Magnusson Sjöberg, ‘The Melting Pot Paradox of Structured Documents: New Tools for Digital Legislation’ in ibid 133–47. 113 OECD, Better Regulation in Europe: Austria (n 63) 90, 98–99. 114  Art 8 Institutional Act No 2009-403 of 15 April 2009. 106 

107  P

66 

Patricia Popelier

detail. It is, for example, feared that mandatory drafting manuals would inhibit creative and innovating rule making.115 Guidelines should not remain the only or most important tools, though a survey has indicated that drafting guidelines are particularly useful in assisting new drafters.116 This finding can be safely extended to all drafters in jurisdictions, such as Belgium and France, without a specific staff of legal drafters, as drafters in these jurisdictions are less experienced. However, training and platforms for the exchange of information are much more important than drafting manuals for the development of drafting skills.117 In most legal systems, such instruments are elaborated and issued by the rule-making authorities, with or without the help of external bodies.118 This, however, is not always the case. For example, while drafting manuals are usually prepared by a centralised drafting unit or the department responsible for the quality of legislation,119 in Belgium, the federal authorities do not create their own drafting guidelines. Instead, they rely on the guidelines drawn up by the Council of State addressed at the federal and subnational legislators. For this work, the Council relies on its advisory practices, which include a check on the drafting quality of bills and proposals.120 It is nevertheless recommended that legislators draw up their own guidelines, for several reasons. First, in order to secure their implementation into practice, guidelines should link up with the specific practice of the separate legislators. They should be conceived as an important part of the regulatory policy, and passed on through training and daily practice. This will sooner be the case if the guidelines are drawn up internally, instead of received from an external body. Also, by drawing up its own guidelines, the authority is made conscious of its responsibility regarding the drafting quality of its own legal norms. Finally, guidelines inserted in ministerial circulars have greater authority than documents originating from external bodies. For example, in the Netherlands, guidelines (Aanwijzingen voor de regelgeving) are determined by the Minister-President, which makes them in principle binding upon the drafters.121

4.2.5. Competences ‘Competences’ refers to the knowledge, expertise and skills of the different actors in the regulatory process, which are essential for them to fulfil their tasks. Competence management implies the development and deployment of competences. This includes, as has been insisted upon in the literature, knowledge management.122

115  R Cormacain, ‘An Empirical Study of the Usefulness of Legislative Drafting Manuals’ (2013) 1 Theory and Practice of Legislation 210–11. 116  ibid 217, 218. 117  ibid 218. 118  For example, in France drafting guidelines are prepared by the Cabinet Office in collaboration with the Council of State, J-P Duprat, ‘The French “Learning” Legislator: The Improvements of an Indirect Process’ in Mader and Moll (n 11) 161. 119  Cormacain (n 215) 205. 120  Although nowadays most advice is given under a time restraint, which reduces the scope of the advice. As a result, commenting on drafting quality no longer belongs to the kernel of the Council’s activities. 121  P Eijlander and W Voermans, Wetgevingsleer (Deventer, WEJ Tjeenk Willink, 1999) 374. 122  W Voermans, ‘A Learning Legislator? Dutch Attempts to Prevent Brain-Drain in the Legislative Process’ in Mader and Moll (n 11) 192.

Management of Legislation

 67

To this end, a clear organisation chart is a minimum requirement. If drafting legislation, consulting, conducting impact assessments, etc are not clearly defined as tasks for specific types of staff,123 it will be difficult to identify target groups for training and for forums for the exchange of experiences and good practices. Additionally, if, as is the case in Belgium, laws are often drafted within the minister’s Cabinet staff of personal advisors,124 it is difficult to build expertise. Also, the identification of specific regulatory tasks will demonstrate the need for an interdisciplinary staff. Nowadays, civil servants engaged in law making are often lawyers who lack the relevant skills for performing consultations and evaluations.125 A comparative study has revealed that in Germany, where positions in ministries are mainly occupied by jurists, most emphasis is put on legal questions, and impact assessments on factual grounds remain a merely formal exercise, whereas the interdisciplinary composition of the Swedish administration results in more substantial impact assessments.126 In multilingual systems, the Canadian method, with professional drafters from each language group who co-draft the official versions,127 stands out as a best practice. Actors in the regulatory process are not limited to internal staff. For example, in order to deal with the complexity of problems in current societies, consultative documents may be issued to seek expert advice along with public opinion.128 In continental countries, expert advisors are mostly institutionalised in formal advisory bodies.129 In neo-corporatist systems, specific interest groups—in particular, social partners—occupy a key position in this respect. Governments are keen to collaborate with interest groups, as these are able to supply evidence-based information and at the same time ensure acceptance by their adherents.130 Moreover, specific tasks such as ex ante evaluation and impact assessments, as well ex post evaluations, are often outsourced to external specialists. While the legislator cannot ‘manage’ external actors, regulatory management does include the establishment of procedures and the training of staff to deal with external actors, in order to avoid regulatory capture. It also includes the development of a policy regarding outsourcing, dependent upon the availability of budget and learning potential within the administration.131 In particular, it implies a careful consideration of the desired utilisation, keeping in mind the types of utilisation mentioned above and the extent to which ‘distance’ is a determinant factor.

123 

As is the case, for example, in Belgium: see above, para. 3.2.2. Regulation in Europe: Belgium (n 30) 64, 111. 125  Van Aeken (n 5) 126, referring to the situation in Belgium. A similar observation was made regarding Spain, see V Vinas, ‘The European Union’s Drive towards Public Policy Evaluation’ (2009) 15 Evaluation Review 461. For Austria, see the observation in OECD, Better Regulation in Europe: Austria (n 63) 23, 92. In France, however, the opposite trend is noticed: bills are drafted by officials with a technical background, but lacking necessary legal skills, Duprat (n 118) 160. 126  Veit (n 11) 189–95. 127  RC Bergeron, ‘The Bilingual and Bijural Learning Legislator in Canada’ in Mader and Moll (n 11) 88–89. 128  Bates (n 82) 123. 129  Knill and Tosun (n 2). 130  ibid 127. 131  See Van Aeken (n 5) 125. 124 OECD, Better

68 

Patricia Popelier

4.3.  Regulatory Management in Multi-tiered Legal Systems The Securitel judgment,132 although concerning the lack of notification of a Belgian statute, made the Dutch government conscious of the implications of EU integration on regulatory quality and procedures. As a result, a comprehensive legislative quality programme was introduced, including purely domestic regulatory issues.133 In other countries as well, the EU has triggered domestic regulatory management initiatives.134 However, EU integration also complicates regulatory management. In this section, I will very briefly provide an overview of the regulatory implications of EU membership (section IV.A). These implications are even stronger in multi-tiered Member States, which face specific co-ordination issues within the domestic as well as the EU legal system (section IV.B).

4.3.1.  Regulatory Management in EU Member States EU Member States are embedded in the larger EU regulatory system in many ways: national laws and regulations have to comply with EU law; they often have to fulfil procedural requirements, such as notification to the European Commission; and they have to execute EU regulations or transpose EU directives. In this respect, Member States are co-actors in the legislative cycle which share, with the EU institutions, responsibility for the quality of European law.135 At the EU level, there are several ‘structures’ and procedures for the involvement of national legislators in the EU legislative process or the correct transposition of EU directives, such as the composition of the EU Council of Ministers, the involvement of national parliaments via the subsidiarity procedure or the EU Pilot, an online platform to give assistance if Member States face transposition problems. At the national level, it is up to the Member States to incorporate EU obligations in national regulatory policy. Essential aspects in this regard are: the conduct of impact assessments on draft EU legislation; structures and procedures for examining the compliance of drafts with EU law; correlation tables; the domestic subsidiarity procedure; and participation in international networks which provide information and best practices. The European Commission’s reports on monitoring the application of EU law give insight into those legal systems best equipped to comply with EU law.136 In 2012, the best performances were situated in Latvia, Lithuania, Estonia, Malta and Denmark. By contrast, the worst results were recorded in Italy, Belgium, Spain, Poland and Greece. The OECD accordingly praises Denmark for having ‘an effective, well-managed and highly-institutionalised internal co-ordination system for EU affairs’.137 This includes co-operation between 132 

CJEU, Case C-194/94 CIA Security International SA [1996] ECR I-2201. JPM van Lochem, ‘The Dutch Academy for Legislation’ in Mader and Moll (n 11) 63–64. 134  See, eg, in Finland: OECD, Better Regulation in Europe: Finland (n 20) 117. 135  L Senden, ‘The Member States and the Quality of European Legislation: Not Consumers, but Actors’ in Mader and Moll (n 11) 103–05. 136  See Commission, 30th Annual Report on Monitoring the Application of EU Law (Brussels, European Commission, 2013). 137 OECD, Better Regulation in Europe. Denmark (n 50) 115. 133 

Management of Legislation

 69

ministries, parliamentary control138 and early involvement of stakeholders so as to identify ­problems at an early stage of EU law making, and to give Danish negotiators in Brussels a clear mandate.139 Transposition of EU directives is facilitated by the ‘centralized, co-­ operative and simple structure’ of the Danish government.140 It is therefore not surprising that the five best performing countries are small, unitary legal systems.141 However, transposition is also facilitated by the early involvement of parliament and stakeholders, ensuring early detection of difficulties and allowing for transposition by executive orders, considering that political scrutiny by parliament has already occurred at an earlier stage.142 By contrast, a weak presence at the early EU decision-making phase, lesser efforts in building alliances with other Member States and underdeveloped impact assessments on draft EU legislation are considered to account for Finland’s poorer record.143

4.3.2.  Regulatory Management in Multi-tiered EU Member States Apart from ‘old model’ federal states such as Austria, Germany, and Switzerland, there is a growing tendency towards devolutionary processes, for example in Belgium, Italy, Spain and the UK. In the governance literature, decision-making capacity of political systems is expected to increase with the delegation of decision-making powers to decentralised levels and non-state actors.144 This is especially the case in multinational states, where national or ethnic groups may act as veto-players at the central level. It is, for example, the rationale behind the gradual transfer of legislative powers to the subnational levels in Belgium.145 Devolution, however, also brings about specific regulatory problems. The OECD notices a link between devolution of competences and regulatory inflation.146 Moreover, in Belgium, dual federalism is considered an important barrier to policy evaluation, as the transfer of competences and multinational conflict management rest on large package deals and so-called ‘wafer iron politics’ incompatible with assessments on objective grounds.147 This is different in co-operative federalism. In the literature, the hypothesis was formulated that where federal programmes are implemented by or with subnational authorities, evaluations are favoured so as to ensure that subnational programmes meet national standards.148 This is particularly the case in systems of executive federalism, where federal

138 For a more critical stance on the Danish parliamentary mandate, see M Moller Sousa, ‘Learning in Denmark? The Case of Danish Parliamentary Control over European Union Policy’ (2008) 31 Scandinavian ­Political Studies 31, 434–447. 139 OECD, Better Regulation in Europe. Denmark (n 50) 117. 140 OECD, Better Regulation in Europe. Denmark (n 50) 118. 141  Although problems arise in very small countries with small governments, where EU obligations impose a heavy burden in relation to management capacities. See OECD, Better Regulation in Europe: Luxembourg (n 82) 42. Still, the infraction rate in Luxembourg is low. 142 OECD, Better Regulation in Europe. Denmark (n 50) 118. 143 OECD, Better Regulation in Europe: Finland (n 20) 118. 144  Knill and Tossun (n 2) 215–16. 145  P Popelier and K Lemmens, The Constitution of Belgium. A Contextual Analysis (Oxford, Hart Publishing, 2015) 76–77. 146 OECD, Better Regulation in Europe. Spain (Paris, OECD, 2010) 36. 147  Varone et al (n 38) 266–67. 148 HU Derlien and RC Rist, ‘Conclusion. Policy Evaluation in International Comparison’ in J-E Furubo, RC Rist and R Dansahl (eds), International Atlas of Evaluation (New Brunswick, Transaction Publishers, 2002) 445.

70 

Patricia Popelier

laws are, as a rule, executed by subnational state authorities. Executive federalism necessitates the involvement of subnational authorities in the enactment of the law, so as to ensure implementation and allow subnational authorities to adjust their regulatory agenda, as well as inter-federal co-operation for the evaluation of regulatory projects. In practice, the first is often effectuated, in contrast to the latter. In Austria, executive federalism results in detailed federal laws, so as to ensure uniform implementation, and in uncertainty as to the implementation and effectiveness of legislation.149 Next, a correlation can be noticed between the level of regionalism, on the one hand, and non-compliance with EU directives, on the other.150 With the exception of the UK, in all federal or regionalised EU Member States (Austria, Belgium, Germany, Italy and Spain) the numbers of declared infringements in 2010–14 are slightly to significantly above the average.151 The division of competences requires more co-ordination mechanisms for the effective implementation of EU law. Those countries that exhibit a high degree of regionalisation but an average number of infringement cases dispose of a better functioning monitoring system or better intergovernmental co-operation. This is the case in Austria,152 Germany and the UK, in contrast to Belgium, Italy and Spain.153 The putting into place of such monitoring or co-operation systems can even be phrased as a legal obligation: the European Court of Justice requires that supervisory and inspection procedures are established in order to ensure implementation and enforcement of EU law.154 For example, in Austria, the federal Chancellory takes up a co-ordinating function, collecting information from federal ministries and Länder on directives to be transposed as well as timetables, and stimulating underperforming ministries or states.155 Instruments provided by the EU to involve national authorities in the EU decision making procedure, deserve specific attention in the case of multi-tiered Member States. It was mentioned above (section IV.A) that early involvement in the decision-making process is pivotal for the effective and timely implementation of EU rules. However, EU directives often deal with matters that, under domestic rule, are allocated to regional authorities. Hence, if regional authorities are responsible for the implementation of EU directives, they should be involved early in the EU decision-making process. Article 16(2) TFEU allows for Member States to delegate a regional minister to the Council of Ministers. Moreover, Article 5(3) TFEU permits an extended delegation, enabling a mixed delegation with central as well as regional representatives. It is for the Member States to use these options and to choose the formula. The Member States have indeed made use of these options to varying degrees,156 but co-ordination mechanisms

149 OECD, Better

Regulation in Europe: Austria (n 63) 144. W Vandenbruwaene, P Popelier and C Jansens, ‘Article 260 TFEU Sanctions in Multi-tiered Member States’, based on a study for the Belgian government, on file with the authors. 151  Based on the Annual Report of the Court of Justice 2014, 106. 152 OECD, Better Regulation in Europe: Austria (n 63) 135. 153  In 2010–14, the CJEU declared an average of 10 infringements per Member State. In Austria, Germany and the UK the number was, respectively, 14, 13 and 9. In contrast, in Belgium, Italy and Spain, it was, respectively, 26, 34 and 32. 154  Joined Cases C-227–230/85, Commission v Belgium [1988] ECR 8, para 13. 155 OECD, Better Regulation in Europe: Austria (n 63) 139, 140–41. 156  For a comparative overview, see N Skoutaris, ‘The Role of Sub-state Entities in the EU Decision-Making Processes: A Comparative Constitutional Law Approach’ in E Cloots, G De Baere and S Sottiaux (eds), Federalism in the European Union (Oxford, Hart Publishing, 2012) 216–22. 150 

Management of Legislation

 71

are always present. For example, in the UK, devolved entities are intensely consulted, but the final decision and responsibility remains with the UK minister.157 In Germany, the Bundesrat plays an important role as mediator between Länder and federal government. Specific interests of separate Länder, however, do not prevail.158 In Belgium, the separate authorities have a stronger position. In exclusive regional matters, when a regional minister represents Belgium in the European Council on the basis of a rotation system, as well as in mixed matters, where Belgium is represented by a mixed delegation, regional and federal ministers meet beforehand in order to agree upon the Belgian stance.159 Every actor has a veto right, although a gentlemen’s agreement inhibits the use of a veto by an actor who is not competent in the concrete case.160 The EU subsidiarity mechanism explicitly provides for the involvement of regional ­parliaments in the subsidiarity review by national parliaments.161 Again, it is for the Member States to decide whether and how to involve regional entities.162 The obvious ­solution is for the second chamber to act as a representative of the regional authorities. Remarkably, despite the transformation of the Belgian Senate into a chamber of the substates in 2014, the Senate has no competences in European or international matters. Instead, the regional parliaments are considered ‘national assemblies’.163 In the Belgian procedure, federal and regional opinions are positioned next to each other, on an equal footing, and without institutional dialogue in order to take a better reasoned stance.164 This is different in Austria, which provides for an intense co-operative procedure.165

4.4. Conclusion Any legal order that pursues the production of high-quality laws needs a comprehensive management system that addresses structures, processes and competences, but also culture. The latter implies that political actors and leading officials value legislative quality standards and support the implementation of a regulatory policy. Much, however, depends on the legal and political culture. Transparent and evidence-based law making fits poorly with 157  ibid 219; N Skoutaris, ‘Comparing the Subnational Constitutional Space of the European Sub-state Entities in the Area of Foreign Affairs’ [2012] Perspectives of Federalism 260. See also S Bulmer, M Burch, P Hogwood and A Scott, ‘UK Devolution and the European Union: A Tale of Cooperative Asymmetry?’ [2006] Publius 86, describing participation as ‘dependent’ and ‘conditional’. 158  A Puttler, ‘Die Deutschen Länder in der Europäischen Union’ in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts (Heidelberg, CF Muller, 2008) 1089. For a detailed arrangement, see Art 23 of the German Constitution. 159  Cooperation agreement of 8 March 1994 between the Federal State, the Communities and the Regions, regarding the representation of the Kingdom of Belgium in the Council of Ministers of the European Union. 160  P Bursens, ‘Het Europese beleid in de Belgische federatie. Standpuntbepaling en vertegenwoordiging van Belgische belangen’ [2005] Res Publica 67. 161  Art 6 of Protocol No 2 on the application of the principles of subsidiarity and proportionality. 162  For an overview of diverse arrangements in the various Member States, see K Boronska-Hryniewiecka, ‘Regions and Subsidiarity after the Treaty of Lisbon: Overcoming the “Regional Blindness”?’ in M Cartabia, N Lupo and A Simoncini (eds), Democracy and Subsidiarity in the EU (Bologna, il Mulino, 2013) 358–61. 163  Declaration No 51, [2008] OJ C-1115/355. 164 P Popelier and W Vandenbruwaene, ‘The Subsidiary Mechanism as a Tool for Inter-level Dialogue in Belgium: On “Regional Blindness” and Cooperative Flaws’ [2011] European Constitutional Law Review 223. 165  For an overview of the procedure, see ibid 224–25.

72 

Patricia Popelier

consociational policies, but consociational policy making may prove vital for the functioning of a political system, for example in multinational societies. Overall, political reason may hinder the process of informed, rational law making. Standardisation and institutionalisation, then, may be of service to guarantee the best possible legislative output given the political context. There is no single best model for the construction of a regulatory management system. This chapter has merely pointed out some choices that can be made and has drawn attention to the risks and benefits. In multi-tiered systems, and especially in the context of European integration, the implementation of a regulatory management becomes even more complex. Co-operation and co-ordination mechanisms are revealed to be of crucial importance in these cases.

Further Reading Balthasar, A, ‘Institutional Design and Utilization of Evaluation: A Contribution to a Theory of Evaluation Influence Based on Swiss Experience’ (2009) 33 Evaluation Review 1. Christensen, T, Laegreid, P, Roness, PG and Rovik, KA, Organization Theory and the Pulic Sector (London, Routledge, 2007). Mader, L and Moll, C (eds) The Learning Legislator (Baden-Baden, Nomos, 2009). OECD, Recommendation of the Council on Regulatory Policy and Governance (Paris, OECD, 2012). OECD, Better Regulation in Europe. Country Reports (Paris, OECD, 2010). Radaelli, CM and Meuwese, A, ‘Better Regulation in Europe: Between Public Management and Regulatory Reform’ (2009) 87 Public Administration 639. Varone, F, Jacob, S and De Winter L, ‘Polity, Politics and Policy Evaluation in Belgium’ (2005) 11 Evaluation Review 253. Verschuuren, J (ed), The Impact of Legislation (Leiden, Martinus Nijhoff, 2009). Wiener, JB and Alemanno, A, ‘Comparing Regulatory Oversight Bodies across the Atlantic’ in S Rose-Ackerman and P Lindseth (eds), Comparative Administrative Law (Cheltenham, Edward Elgar Publishing, 2011) 309–35.

5 Participation FELIX UHLMANN AND CHRISTOPH KONRATH

Participation offers citizens the possibility to comment on a proposal for a law or on a draft law. It provides for information and transparency of the legislative process. At the same time, it offers the possibility of a kind of dialogue between the government, the parliament and citizens. In that sense, it can be seen as an essential element of a ‘due process of lawmaking’. Further opportunities offered by participation are tests of political feasibility of a draft law, enhanced fact finding and assessment of practicability. However, it is important to realise that participation is not a panacea for defective drafts. It is not a scientific method, and participants may be biased and may not fairly represent civil society. In practice, participation of the civil society in the legislative process will most often be in the form of consultation. Throughout Europe, a wide variety of forms of consultation can be found, many of which have been subject of recent criticism and reform with the aim of creating more formalised, transparent and inclusive practices.

5.1.  Definition and Delimitations At its broadest, participation (or consultation) may be understood as the involvement of the civil society in the legislative process.1 The process of participation is often referred as ‘consultation’. Participation may be more precisely defined by the following elements:2 —— The object of participation is a draft law. The draft may be accompanied by explanatory notes, reports, a questionnaire, etc, but its core is a legal text (which does not exclude that similar mechanisms as explained in this text may apply with regard to other forms of government action, such as policies and white books). The legal text 1  Art 2, para 1 CPA (Swiss Federal Act on the Consultation Procedure of 18 March 2005, Consultation Procedure Act, CPA; SR 172.061): ‘The consultation procedure has the aim of allowing the cantons, political parties and interested groups to participate in the shaping of opinion and the decision-making of the Confederation’; T Sägesser, Vernehmlassungsgesetz (Bern, Stämpflis Handkommentar, 2006), Art 2 N 4ff, with further references; for an extensive view of the term ‘consultation’ see P Nanz and C Leggewie, Die Konsultative, Mehr Demokratie durch Bürgerbeteiligung (Berlin, Klaus Wagenbach Verlag, 2016) 1ff; on the reasons of involving civil society in policy making see OECD, ‘Consultation and Participation’, Open Governement: Fostering Dialogue with Civil Society (Paris, OECD, 2004) 17; U Karpen, ‘Die Begutachtung von Rechtsetzungsvorhaben in der Bundesrepublik Deutschland’ in L Mader and U Karpen (eds), The Participation of Civil Society in the Legislative Process (BadenBaden, Nomos, 2005) 238. 2  For the ‘best practices’ in several countries compare P Popelier, ‘Consultation on Draft Regulation—Best Practices and Political Objections’ in L Mader and M Tavares de Almeida (eds), Quality of Legislations, Principles and Instruments (Baden-Baden, Nomos, 2011) 139ff.

74 

Felix Uhlmann and Christoph Konrath

may be at the level of a constitution, of primary legislation (laws), of secondary legislation (ordinances) or even of an international treaty. It does not matter at what time in the legislative process the participation takes place, so long as it takes place before the legislation becomes law. Thus, it may be before the government or in parliament. In any case, it must be a draft law, that is, a law that has not yet been passed by the legislator, which means that public opinion may still influence the outcome.3 —— Participation has no binding effect on government.4 Participation is understood as ‘advice’ to the sponsors of the draft, whether they should pursue, abort or amend the project. This means that the law may be passed against the fierce resistance of the respondents. The participation process may require that the feedback be considered, but it need not be followed by the government.5 In this respect, participation clearly differs from instruments of direct democracy, such as referenda. It is also the government that does (most of) the drafting, not the private individuals (crowdsourcing of legislation). —— The typical participants are private individuals. The ‘sounding board’ of the draft law is civil society.6 However, the group of respondents may also be extended to include other governmental bodies or state entities.7 —— Participation is a formalised process.8 Participation is more than random petitions from citizens commenting on legal projects. It has some kind of opening, an invitation to respond to a draft law, and typically has some ending, for example a deadline for comments.9 The formalisation may be low; it may be that the draft laws are published only on a governmental website open to remarks.10 However, the participation should not 3 

See, eg, T Sägesser, ‘The Consultation Procedure in Switzerland’ in Mader and Karpen (n 1) 238. See, eg, Art 8 CPA (n 1). the ‘general principles and minimum standards for consultation’ from the Commission, ‘Towards a Reinforced Culture of Consultation and Dialogue—Proposal for General Principles and Minimum Standards for Consultation of Interested Parties by the Commission’ (Communication) COM (2002) 704 (11 December 2002); Commission, ‘Review of the Commission Consultation Policy’ (Staff Working Document) SWD (2012) 422) (December 2012), available at http://ec.europa.eu/smart-regulation/better_regulation/key_docs_en.htm#_ consultation (accessed on 22 November 2015). 6  See Art 4, para 1 CPA (n 1); see also S Salzborn, Demokratie (Baden-Baden, Nomos, 2012) 86–87. 7  See, for instance, the differentiation between ‘consultation’ involving mainly bodies established by the state and ‘concertation’ open to civil society and interest groups in France: S Rose-Ackerman and T Perroud, ‘Policymaking and Public Law in France: Public Participation, Agency Independence, and Impact Assessment’ (2013) 19 Columbia Journal of European Law 223, 254ff. 8  See, eg, L Mader, ‘Die Rolle der Zivilgesellschaft und die Bedeutung der Konsultationsverfahren für die nationale Rechtsetzung in Europa’ in H Schäffer and J Iliopoulos-Strangas (eds), State modernization in Europe (Athens, Ant N Sakkoulas, 2007) 56; see also Art 5ff CPA (n 1). 9  T Divjak and G Forbici, Public Participation in Decision-Making Process, International Analysis of the Legal Framework with a Collection of Good Practices (Ljubljana, Center for Information Service, Co-operation and Development of NGOs, 2014) 9. 10  See the ‘minimum standards for consultation’ from the European Commission (n 5); regarding electronic/ online rule making and public participation in the US see BS Minor, ‘Evaluation E-Rulemaking: Public Participation and Political Institutions’ (2006) 55 Duke Law Journal 893; C Coglianese, ‘Enhancing Public Access to Online Rulemaking Information’ (2012) 2 Michigan Journal of Environmental & Administrative Law 7; C Coglianese, ‘The Internet and Citizen Participation in Rulemaking’ (2005) 1 Journal of Law and Policy for the Information Society 33; CR Farina, M Newhart, C Cardie and D Cosley, ‘Rulemaking 2.0’ (2011) 65 University of Miami Law Review 1001; PL Strauss, ‘Legal Frameworks and Institutional Contexts for Public Consultation Regarding Administrative Action: The United States’, Columbia Public Law & Legal Theory Working Papers (New York, Columbia Law School, 2009); for the EU see C Cobo, ‘Networks for Citizen Consultation and Citizen Sourcing of Expertise: Exploring Innovations in the Public Sector’ [2012] Contemporary Social Science: Journal of the Academy of Social Sciences 4; N Rodean, ‘Participatory Democracy: Mechanism of Better Regulation in Europe’ (2011) No 2 Acta Universitatis Danubis 31, 42. 4 

5  See

Participation

 75

be conducted arbitrarily and randomly, but should follow certain rules. These rules are often, though not necessarily, written law; they should be rooted at least in administrative practice (see section 5.5.4 below).11 It may be that the duty to consult is mirrored by a right to participate, allowing citizens to enforce consultation before a court. It may even be that the law will be invalidated by the courts in case of a violation of the participation procedure (‘due process’). Such a right may substantially enhance the effects of participation, but is not an indispensable prerequisite.12 —— There is no standard format for participation.13 The typical format will be an opportunity for written observations within a certain time limit, but participation may also take place orally (eg conferences) or via the internet.14 Usually, participation is open to everyone, but some formats require restrictions (confidential materials, hearings, etc). Participation is typically unsolicited and unpaid by the government. One may add further elements that delimit participation towards similar forms of interaction between state and private actors: —— Participation is not an opinion poll.15 Opinion polls are conducted to scientific standards in order to determine the attitude of society (or groups thereof) towards a certain question. Participation does not follow this pattern. It is usually open to everybody and may not reflect a valid cross-section of society. Of course, participation may deliver pertinent information on the political feasibility of a project. —— Participation is not an expert opinion.16 There may be forms of participation that come close to expert opinions, such as public hearings of parliament, when mainly experts 11 As an example one could mention the ‘Grenelle’ organized by the French government since the 1960s, responding to the calls for more direct public participation, but which ‘has no formal legal definition and is used rather indiscriminately’: Rose-Ackerman and Perroud (n 7) 258; for the ‘Grenelle’ see also Divjak and Forbici (n 9) 64–65; for public participation in the EU see generally O Mross, Bürgerbeteiligung am Rechtsetzungsprozess in der Europäischen Union (Berlin, Duncker & Humbolt, 2010) 175; see also A Classen, Interessenvertretung in der Europäischen Union (Wiesbaden, Springer, 2014) 31; for the statutory implementation in Central and Eastern Europe, eg in Hungary, see T Drinóczi and M Kocsis, ‘Public Consultation—Theory and Hungarian Practice’ in L Mader and S Kabyshev (eds), Regulatory reforms, Implementation and compliance (Baden-Baden, Nomos, 2012) 303, 314ff. 12  On the right to participation see generally J Temperman, ‘Public Participation in Times of Privatisation: A Human Rights Analysis’ (2011) 4(2) Erasmus Law Review 43; see also K Czapansky and R Manjoo, ‘The Right of Public Participation in the Law Making Process and the Role of the Legislature in the Promotion of this Right’ (2009) 19(1) Duke Journal of Comparative & International Law 1. 13  See generally L Blomgren Amsler, ‘The Next Generation of Administrative Law: Building the Legal Infrastructure for Collaborative Governance’ (2010) 2010 Wisconsin Law Review 297, 316–17; see also DC Brabham, ‘The Effectiveness of Crowdsourcing Public Participation in a Planning Context’ (2010) 2ff for the appreciation of crowdsourcing as an alternative method of public participation. 14  C Finney, ‘Extending public consultation via the Internet: the experience of the UK Advisory Committee on Genetic Testing electronic consultation’ (1999) 26 Science and Public Policy 363 quoted by Popelier (n 2) 142 fn 28; Blomgren Amsler (ibid) 331ff; OECD, ‘Citizens as Partners’, OECD Handbook on Information, Consultation and Public Participation in Policy-Making (Paris, OECD, 2001) 67ff, available at www.oecd-ilibrary.org/governance/citizens-as-partners_9789264195578-en (accessed on 25 November 2015); OECD, Open Government: Fostering Dialogue with Civil Society (Paris, OECD, 2004) 17, available at www.oecd-ilibrary.org/governance/opengovernment_9789264019959-en (accessed on 25 November 2015). 15  Compare, eg, the critical assessment of participants’ attitudes and opinions as criteria for evaluating public participation by C Coglianese, ‘Is Satisfaction Success? Evaluating Public Participation in Regulatory Policymaking’ in R O’Leary and L Bingham (eds), The Promise and Performance of Environmental Conflict Resolution ­(Washington, Resources for the Future, 2003) 4. 16  For a delimitation of ‘experts’ and ‘civil society’ see TStJN Bates, ‘Experts in the Legislative Process in the ­British Isles’ in Mader and Karpen (n 1) 121–22; see also J-P Duprat, ‘La marginalisation de la pratique r­ éférendaire

76 

Felix Uhlmann and Christoph Konrath

are invited. However, participation should be open to everyone or, if that is not practically possible, it should fairly represent the public.17 Private actors will typically voice political comments on the project, but the draft is usually open to any comments. Technical and scientific advice may fall under the general heading of participation. —— Participation is not lobbying. Although lobbying activities may take place during participation, lobbying is much broader, encompassing any form of influence on governmental action. It may well be that the formalised process of participation is less attractive for lobbyists, as their input is often made public. One may even speculate whether lobbyists prefer secretive channels of influence. Still, this does not exclude that the participation process may be used to display the ‘official’ position of a certain interest group.18 Participation is transparent, lobbying generally less so.

5.2.  Functions of Participation 5.2.1.  Information and Transparency Participation offers citizens the possibility to comment on a draft law. It is often the first time that a legal text leaves the narrow circle of civil servants dealing with the subject. It may be the first time that citizens get to know of a government project. Thus, first of all, participation may serve the function of informing the public.19 Once the law has been passed, it is possible to retrospectively trace the changes the law has been subject to. This improves the transparency20 of the legislative process, for one may compare versions before and after the public consultation or the hearing. If the contributions of civil society and possibly also the government’s assessment thereof are available, these materials may prove to be helpful in the process of statutory interpretation. They may explain the reasons for amendments of the law that may otherwise remain foggy if the

en France et les solutions alternatives de la participation de la Société civile au processus législatif ’ in Mader and Karpen (n 1) 145–46, 148–49 and 162ff with further references. 17  eg, the environmental ‘Grenelle’ in France ‘did not involve open-ended public hearings, but there were no explicit quotas on who could participate’: Rose-Ackerman and Perroud (n 7) 258–59; for the Swiss Consultation Procedure see Sägesser (n 3) 238. 18  P Van Nuffel, ‘Le rôle de la société civile et les procédures de consultation à caractère supranational’ in Schäffer and Iliopoulos-Strangas (n 8) 44. 19  See, eg, the references in the Bundesblatt (BBl = Swiss Federal Gazette): BBl 1998 3791 and BBl 2004 546; see also Mader (n 8) 65; Popelier (n 2) 137; Divjak and Forbici (n 9) 6; but see J Rossi, ‘Participation Run Amok: The Costs of Mass Participation for Deliberative Agency Decisionmaking’ (1997), 92 Northwestern University Law Review, 173, 185ff. 20  See especially C Coglianese, H Kilmartin and E Mendelsohn, ‘Transparency and Public Participation in the Rulemaking Process: Recommendations for the New Administration’ (2009) 77(4) George Washington Law Review 924; W Funk, ‘Public Participation and Transparency in Administrative Law—Three Examples as an Object Lesson’ (2009) 61 Administrative Law Review 171; see also Blomgren Amsler (n 13) 334ff; E Sutherland, The European Union Approach to Better Regulation: Impact Assessment (Communications Regulators Association of Southern Africa, 2010) 4ff; RK Craig and JB Ruhl, ‘Designing Administrative Law for Adaptive Management’ (2014) 67(1) Vanderbilt Law Review 29; OECD, ‘Background Document on Public Consultation’, 1ff, available at https://www. oecd.org/mena/governance/36785341.pdf.

Participation

 77

political logic behind the changes is not disclosed.21 Participation enhances the transparency of the legislative process and helps (slightly) to prevent corruption through interest groups.22

5.2.2. Legitimacy Participation is clearly more than government information and the production of legal materials. It offers the possibility of a kind of dialogue,23 as citizens may comment on the draft law. Even if their propositions are not adopted by the legislator, the possibility for citizens to voice their concerns may be seen as a valuable element.24 It enhances the legitimacy of the law.25 This is especially valuable in systems in which there is no formal role for citizens in the legislative process via referenda or other mechanisms of direct democracy. Hence, the mere act of participation serves a goal.26 It may be structurally compared with the right to be heard in administrative procedures. The citizen should not be the object of administrative action, or treated as an object by the law. If one transposes the term ‘due process’ into the field of legislation and speaks of the ‘due process of lawmaking’,27 the right to be heard on a draft law should be a crucial element. However, one should not overlook the fact that legislative action is often structurally different from administrative action. Many laws need concretion through administrative and court action, and their direct effects on citizens may not yet be assessable. It is more the general idea that citizens should be heard before state action takes place that links due process in individual cases to citizen participation in the legislative process. It may also be said that participation is valuable even if it does not help to find an adequate legislative answer. The draft law may trigger discussions in the civil society that per se

21 

Blomgren Amsler (n 13) 298ff; regarding the Swiss Consultation Procedure see Sägesser (n 3) 239. an assessment of the role of civil society in the consultation procedure at the European level see Van Nuffel (n 18) 44; eg, compare the critical appreciation of the numerous established bodies and recognized interest groups involved in the consultation and participation procedures in France by Rose-Ackerman and Perroud (n 7) 261–62, 307. 23  cf Popelier (n 2) 137; see also OECD, ‘Citizens as Partners’ (n 14) 15ff; for more details, see T Drinóczi, ‘Constitutional Dialogue Theories—Extension of the Concept and Examples from Hungary’ (2013) 68(1) Zeitschrift für öffentliches Recht 87. 24  LN Mulligan and G Stazewski, ‘The Supreme Court’s Regulation of Civil Procedure: Lessons from Administrative Law’ (2012) 59 UCLA Law Review 1188, 1244; W Voermans, ‘Concern about the Quality of EU Legislation: What Kind of Problem, by What Kind of Standards?’ (2008) 2 Erasmus Law Review 60, 86–87. 25 Van Nuffel (n 18) 44–45; regarding the Swiss consultation procedure see Mader (n 8) 63; M Kägi, ‘Die Rolle von Argumenten im Vernehmlassungsverfahren’ (2011) 2 LeGes 161, 170–71; NA Mendelson, ‘Rulemaking, Democracy, and Torrents of E-Mail’ (2011) 79 George Washington Law Review 101; JE Gersen and AJ O’Connell, ‘Deadlines in Administrative Law’ (2008) 156 University of Pennsylvania Law Review 923, 945, 972; OECD, ‘Citizens as Partners’ (n 14) 18. 26  See, eg, Craig and Ruhl (n 20) 28ff. 27  See especially HA Linde, ‘Due Process of Law Making’ (1976) 55 Nebraska Law Review 176. For the USA see I Bar-Simon-Tov, ‘The Puzzling Resistance to Judicial Review of the Legislative Process’ (2011) 91 Boston University Law Review 1915; V Goldfield, ‘Legislative Due Process and Simple Interest Group Politics: Ensuring Minimal Deliberation through Judicial Review of Congressional Process’ (2004) 79 New York University Law Review 367; PP Frickey and StS Smith, ‘Judicial Review and the Legislative Process: Some Empirical and Normative Aspects of Due Process of Law Making’, Boalt Working Papers in Public Law (University of California, Berkeley, 2001) 1ff; J Waldron, The Dignity of Legislation (Cambridge, Cambridge University Press, 1999) 1ff; G Müller and F ­Uhlmann, Elemente einer Rechtssetzungslehre, 3rd edn (Zurich, Schulthess Verlag, 2013) 42. 22  For

78 

Felix Uhlmann and Christoph Konrath

are part of the solution.28 Prostitution and the sex trade, assisted suicide and AIDS are all tough challenges for the legislator, and it may well be that restrained governmental action is advisable. Public awareness through discussion on draft laws may help. Obviously, the goal of better legitimacy may only be served if the process of participation satisfies some basic requirements. If citizens believe that participation is a mere fig leaf for governmental action, or even worse, if they feel manipulated to serve the (hidden) agenda of the ruling party, the process will be worthless, if not detrimental. By the same token, participants who have repeatedly observed that their contributions were not even considered by the legislator will abstain from participation in the near future. If not taken seriously, participation becomes an idle ritual that only prolongs the legislative process. There are more subtle dangers to better legitimacy by participation. It may be that citizens are called too often to comment on a draft law and that they get frustrated by the sheer number of consultations.29 It may also be that the draft laws are so complicated that only specialists may reasonably participate. In this respect, it may be advisable for draft laws to be accompanied by reports—but not propaganda—of the government, or for the government to invite specialists who are able to answer complex questions in hearings and,30 ideally, disseminate these findings to the public. In sum, the function of better legitimacy of laws through participation may be achieved only if the channels of communication between government and citizens are working properly.31 Even under ideal circumstances, however, one must critically assess whether participation truly enhances the legitimacy of a law.32 It may well be that the asymmetries in civil society are echoed in the results of the participation process. One may assume that general public interests are voiced less than those of specialised—typically well-funded—interest groups.33 This explains why the results of public consultation are not binding on government and, indeed, why they should not be binding. The feedback from the civil society must be critically assessed and evaluated by the government, not blindly followed.34 Participation should not be used to blur responsibilities and accountability.

5.2.3.  Political Feasibility The procedure of participation is often the first ‘baptism of fire’ for a legal text that has hitherto only been circulated within the administrative or legislative branch.35 Its sponsors should be able to assess what the chances are of having the law passed after the consultation process. As already described, feedback from the civil society is not an opinion poll or a blueprint of the outcome in a popular referendum. The more concerned—or better funded—will 28 

Rossi (n 19) 187. Regarding the Swiss consultation procedure see Mader (n 8) 66. 30  OECD, ‘Citizens as Partners’ (n 14) 59. 31  Sutherland (n 20) 5–6. 32  Coglianese (n 15) 2ff. 33  See, eg, Duprat (n 16) 173. 34  See especially Mader (n 8) 69; cf G Rowe and LJ Frewer, ‘Public Participation Methods: A Framework for Evaluation’ (2000) 25 Science, Technology & Human Values 11. 35  See, eg, Y Papadopoulos, ‘La consultation: Un outil de gouvernabilité?—Fonctions et dysfonctions de la phase préparlementaire’ (1997) 2 LeGes 41. 29 

Participation

 79

raise their voices louder;36 the possible political outcome must be assessed thoroughly. One may steer to a more balanced response through a special format (eg in hearings) or through regulation, but every open process is subject to possible hijacking by lobbyists.37 Participation should help to localise political enemies and allies with regard to a specific piece of legislation. It may help the legislation’s sponsors to fine-tune the draft if they can identify the political stumbling blocks, hence ideally turning opponents into supporters. The sponsors of the law may also intensify their efforts to convince the public of the need for the proposed legislation.38 Finally, if the participation takes place at an early stage of the legislative process, it can prevent legislation from being stillborn and hence can save governmental resources.39 The test of political feasibility40 may come into conflict with the role of parliament, especially if the parliamentary phase is followed by a popular referendum. If participation takes place at an early stage on drafts prepared by the government—which is often the case—a political discussion may emerge before the draft law has reached parliament. The parliament as addressee of draft legislation from the government does not participate in public consultation (or only through the political parties represented in parliament).41 Thus, it could well be that the political positions are carved in stone before the parliamentary phase. Irreconcilable positions may be detrimental to an open debate in parliament. They may reduce the marge de manœuvre of parliament. Of course, a public debate on a draft law may also serve parliamentarians, as they can better assess what their voters think of the draft law.

5.2.4.  Fact Finding The process of fact finding42 should take place before the draft law has been put on public display. (In fact, it should take place even before the law has been written at all.)43 Nevertheless, participation may give government new factual insights. Ideally, the process of participation helps the government to access the aggregated knowledge of the civil society. This is a powerful resource.44 It should also be noted that participation of the public is typically unpaid, which means that the government may collect valuable data ‘for free’.45 There are caveats. Participants in the consultation process need not serve the public interest; they may reasonably pursue their own private interests.46 This implies that private data

36 

Rose-Ackerman and Perroud (n 7) 262. See T Drinóczi, ‘Concept of Quality in Legislation—Revisited: Matter of Perspective and a General Overview’ (2015) 36 Statute Law Review 211, available at https://slr.oxfordjournals.org/content/36/3/211.full (accessed on 24 March 2016). 38  See, eg, Van Nuffel (n 18) 44. 39  See, eg, Rossi (n 19) 187; Rose-Ackerman and Perroud (n 7) 307; see also OECD, ‘Focus on Citizens’, OECD Studies on Public Engagement for Better Policy and Services (Paris, OECD, 2000) 17. For a critical reflection on the Swiss consultation procedure in this respect see Mader (n 8) 62, 64–65. 40  See Art 2, para 2 CPA (n 1). 41  For the possible interference with the parliament see also Mader (n 8) 67. 42  See Art 2, para 2 CPA (n 1); Mader (n 8) 63–64. 43  cf Müller and Uhlmann (n 27) 78ff. 44  See, eg, Divjak and Forbici (n 9) 4; see also OECD, ‘Evaluating Public Participation in Policy Making’ (Paris, OECD, 2005) 115ff; OECD, ‘Citizens as Partners’ (n 14) 17ff. 45  cf Blomgren Amsler (n 13) 350. 46  Duprat (n 16) 146; Coglianese (n 15) 1ff. 37 

80 

Felix Uhlmann and Christoph Konrath

must be critically analysed. Still, the data may be helpful. This holds especially true if there are data offered from both sides of the political spectrum. The government may reasonably assume that the truth lies somewhere in between, and that the facts are correctly presented if both sides agree on certain premises.

5.2.5.  Implementation and Practicability One area of knowledge from the civil society that seems particularly worthy is the assessment of practicability.47 Civil servants preparing legislation may be distanced from the implementation of the law, and it does not seem unfair to say that problems of implementation are often underestimated if one does not work with the law on a daily basis. There is also the danger that questions of implementation are regarded as ‘boring technicalities’ by the lawmaker, which does a disservice to these often important issues. Many laws fail not because of their legal quality, but because of poor implementation or poor legal provisions for implementation, or both. The law by itself may not redress all of the problems of implementation—this would be another perilous assumption that could lead the government to make futile amendments to the law instead of improvements to the implementation. Successful implementation depends on many factors, such as administrative resources and expertise, and acceptance of the law by the public.48 It should not be doubted that the law itself may pose problems of implementation, such as unrealisable procedures, unreasonable expectations and burdensome data collection. Hence, it is advisable that the challenges of implementation be known while drafting a new law. Civil servants dealing with the law or its predecessors should be able to participate in the process, or at least be consulted. Even if civil servants familiar with implementation assist the process of law drafting, their participation does not render the participation of civil society superfluous, as the assessment of civil servants may not match that of private actors. What the administration considers a smooth process may be exceptionally burdensome to private parties. Thus, feedback from the civil society can provide new insights on implementation.49 Private actors will typically ask what the practical consequences for them are. They will ‘test’ the law on the basis of the presumed daily use of it, and will undoubtedly react if they find the consequences unpractical or unduly burdensome.50 Ideally, participation offers an ex ante evaluation51 on the basis of countless individual applications of the law.52 Again, there are caveats. As already pointed out, participants in the consultation process may not serve the public interest.53 They may exaggerate the burdens of implementation

47 

See Art 2, para 2 CPA (n 1); Mader (n 8) 64. See, eg, Craig and Ruhl (n 20) 29; Rossi (n 19) 185ff; Divjak and Forbici (n 9) 5. 49  Mulligan and Stazewski (n 24) 1244. 50  OECD, ‘Evaluating Public Participation’ (n 44) 120. 51  Regarding Regulatory Impact Assessment see Müller and Uhlmann (n 27) 62–63; see also Sustainable Governance Indicators (SGI), ‘Evidence-based Instruments’, available at www.sgi-network.org/2014/Governance/ Executive_Capacity/Evidence-based_Instruments/Quality_of_RIA_Process (accessed on 25 November 2015). 52  See, eg, OECD, ‘Background Document on Public Consultation’ (n 20) 2ff. 53  Coglianese (n 15) 2ff. 48 

Participation

 81

for the sake of political advantage—a problem that is similar to a well-known critique to the standard cost model.54 They may even try to raise fake problems of implementation in order to prevent legislation unsympathetic to them from being passed. Even so, the government should certainly take objections to the practicability of the law seriously.

5.2.6.  Legislative Quality55 Questions of legislative technique are not usually the focus of participation. The greater public and political parties tend to see them as mere ‘technicalities’, and matters rather for specialists within the government. There are, however, exceptions. Other administrative bodies (typically in a federal system) or—less often—courts may also be participants in public consultations. Some laws concern private actors that are highly competent in legal questions (eg legislation on solicitors). These respondents will typically look at the legislative details and will add valuable contributions to the legal text. In practice, though, such contributions are relatively rare. From the general public, one may expect remarks on the legislative quality concerning vagueness and gross errors: if the law is too vague, the private actors will wonder how it will apply to them. It seems fair to assume that the public is sceptical of norms that give substantial leeway to the government and administrative authorities. A layperson’s approach to the law is often one of literal interpretation. A norm may be unambiguous for a lawyer, who sees it embedded within the legal order, whereas a layperson expects all of his or her questions to be answered by the law itself. If the attitude of the civil society is one of distrust towards the government,56 it will typically insist on dense norms that tie up the government’s hands. Participation may also eliminate gross errors in the legal text. If a substantial number of respondents have misunderstood the draft, it may be safely assumed that the wording is unclear and needs improvement.57 However, if the legal text is accompanied by a report and if the civil society generally believes that the report accurately summarises the content of the draft law, it is doubtful whether the legal text will be read at all. Obviously, in this scenario, even gross errors in the text will go unnoticed.

5.3. Limits As described above, participation may serve numerous functions. If properly executed, it enhances the transparency and legitimacy of the legislative process. It provides for helpful 54  Müller and Uhlmann (n 27) 52–53; M Schorn, ‘Bürokratiekostenabbau: Die Illusion des Unpolitischen’ in Hensel et al (eds), Gesetzesfolgenabschätzung in der Anwendung: Perspektiven und Enwicklungstendenzen (BadenBaden, Nomos, 2010), 99ff. 55  For the Europen Union see, eg C Grüner, Quantität und Qualität der europäischen Rechtsetzung (Frankfurt, Lang, 2011) 359ff; Mandelkern Group, Better Regulation: Final Report, SEC(2010) 114 (Brussels, European Commission, 2010). 56  OECD, ‘Evaluating Public Participation’ (n 44) 85ff. 57  OECD, ‘Focus on Citizens’ (n 39) 23.

82 

Felix Uhlmann and Christoph Konrath

insights on political feasibility, facts, the practicability of the law and, in some cases, even legislative quality. However, it is also important to see that participation is not a panacea for defective drafts.58 Participation is not a scientific method. Participants may be biased and may not fairly represent civil society.59 This does not mean that the results of the process are useless, but it does underline the necessity for the results to be assessed open-mindedly but critically by the sponsors of the law.60 In this respect, participation is a crafting tool for successful legislation rather than a detector of legislative ‘truth’. It should also not be concealed that participation comes with a price. It consumes two of the most valuable resources of the legislative process, that is, time and governmental manpower.61 Civil society must have a fair timeline to react to a draft.62 Its feedback must be analysed carefully by the sponsors of the law, a process that also takes time. The deployment of these resources is crucial. Participation rightly raises expectations that the feedback from civil society will be taken seriously by the legislator. If not, civil society will get frustrated and participation will become a waste of time to the legislative process. This suggests that participation should be aptly dosed and envisaged only if the necessary resources can be properly deployed.

5.4.  Forms of Participation63 In practice, the participation of civil society in the legislative process will most often be in the form of consultation. As such, it can be an autonomous process or complement other, more traditional, forms of consultation, like the consultation mechanisms in federal states64 or the consultation of certain state organs or public bodies.65 58 

Popelier (n 2) 143ff. cf A Flückiger, ‘Consulter pour mieux légiferer: Utilité des procédures de consultation pré-parlementaires’ (2007) 3 LeGes 185. See the criticism regarding the Swiss system of H Muralt Müller, ‘Vernehmlassungsverfahren— helvetisches Ritual oder modernes partizipatorisches Instrument der Konkordanzdemokratie’ (1997) 2 LeGes 17. For further references see OECD, ‘Evaluating Public Participation in Policy Making’ (n 44) 57–58; S Kravchenko, ‘The Myth of Public Participation in a World of Poverty’ (2009) 32 Tulane Environmental Journal 34; CR Farina, M Newhart and J Heidt, ‘Rulemaking vs Democracy: Judging and Nudging Public Participation that Counts’ (2012) 2 Michigan Journal of Environmental and Administrative Law 124. 60  See especially Müller and Uhlmann (n 27) 105; see also Funk (n 20) 179–80. 61  See, eg, Mader (n 8) 68; Popelier (n 2)136; see also OECD, ‘Background Document on Public Consultation’ (n 20) 1; OECD, ‘Citizens as Partners’ (n 14) 45. 62  cf the ‘general principles and minimum standards for consultation’ from the European Commission (n 5) and Forbici (n 9) 9. 63  A comparison of consultation practices is faced with a number of obstacles, especially as many practices are neither codified by law nor explicated in guidelines or conventions. The country cases and best practices we refer to in this section were collected in 2013/14 in a survey among the correspondents of the European Centre of Parliamentary Documentation and Research. We are grateful for the information provided by the parliaments of Belgium, Bosnia and Herzegovina, Croatia, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Norway, Poland, Portugal, Serbia, Slovakia, Slovenia, Spain and Sweden. The information on Austria and Switzerland is based on our own research. 64  cf the Austrian Consultation Mechanism (Federal Law Gazette I 35/1999), an intra-state treaty between the Austrian Federal State, the Laender and the Municipalities providing for an early warning mechanism in fiscal affairs. The Constitutional Court acts as an arbitrator in case of conflict. 65  See, eg, France (n 7); see also Art 4, paras 1 and 2 CPA (n 1). 59 

Participation

 83

Today, participation is a common practice in Europe, and there are few countries where public consultations on draft laws are an exception (eg Spain). However, there are various understandings of the concepts of participation and consultation, and a plurality of regulations and practices.66 First, we may differentiate between traditional forms of consultation and new developments. While the roots of the former can be traced back to the nineteenth century,67 it was mostly developed and established in the interwar period. It is a practice rooted in corporatism and consociationalism, in which certain groups, for example employee and employer representatives or language groups, have a privileged position in the (­pre-)legislative process. They retain this position, at least formally, even if the consultation has become a public affair.68 In such a context, the line between consultation and participation (in the sense of active involvement and negotiation) can often be blurred, and consultation can be one of the more formal elements in a system characterised by informal arrangements.69 In contrast, the development of new practices can be associated with new regulatory models, such as Regulatory Impact Assessment (RIA) on the one hand and democratic change on the other hand. Consultation, whether of stakeholders, experts or the wider public, has become a standard regulation tool since the 1980s.70 The efforts to establish and strengthen democracy and civil society in post-communist countries led to the development of a number of models of public consultation from the late 1990s onwards. Unlike traditional practices, they are characterised by quite detailed legal procedures (see section 5.5 below). Almost at the same time, a growing scepticism was expressed by citizens about their involvement in consultation exercises,71 which resulted in a number of innovative reforms in established democracies, too (see sections 5.4.1 and 5.4.5 below). From this rough sketch we can already discern other features that characterise the practices and regulation of consultation. Among them are the underlying concepts and goals of consultation, the participants in the process and the impact of the consultation. In what follows, we try to present the diverse elements of consultation procedures and practices in a systematic way.

5.4.1.  Consultation on the Timeline of the Legislative Process Best practices and comparative studies emphasise the importance of relatively open and transparent multi-layered consultation processes starting at an early phase of the

66  Divjak and Forbici (n 9) 14ff give a comparative overview for several countries. For a ‘bad’ example of nonparticipation see R Cormacain, ‘Prerogative Legislation as the Paradigm of Bad Law Making: The Chagos Islands’ (2013) 39 Commonwealth Law Bulletin 487. 67  cf G Schefbeck, ‘E-Partizipation und elektronisches Begutachtungsverfahren’ (Bildungsprotokolle: 5. Klagenfurter Legistikgespräche, 2008) 11ff. 68  This is the case, for example, in Austria, Belgium, Portugal, the Scandinavian countries and to some extent Switzerland. When consultation exercises are open to the public, as is standard today, such bodies will still receive formal invitations to take part, etc. 69  On participants see section 5.1 above. 70  Popelier (n 2) 136. 71  G Smith, Democratic Innovations (Cambridge, Cambridge University Press, 2009) 17; J Newman et al, ‘Public Participation and Collaborative Governance’ (2004) 33 Journal of Social Policy 211.

84 

Felix Uhlmann and Christoph Konrath

­decision-making process.72 The UK Consultation Principles73 are quite specific on this issue and state that consultation should take place ‘at an early stage to enable contestability and challenge’. The UK Principles also distinguish between informal dialogue with stakeholders prior to a formal consultation to obtain initial evidence and gain an understanding of the issues that will need to be raised in formal consultation and formal consultations as a means to seek out views or evidence to influence policy. The Principles subsequently foster a pragmatic approach when it comes to multi-stage consultations. This should be decided according to the subject matter, the burden of running several consultation exercises and any potential delay in implementing the policy.74 The general idea of consultation as exemplified by the UK Consultation Principles is shared in most European countries. However, there are huge differences when it comes to the actual practice of consultation, especially with regard to methods applied and publicity. We observe a general practice of policy makers (including the administration) to have multiple consultation rounds with selected partners at the informal and non-public level. At the same time, formal public consultations—on the basis of either laws or conventions—tend to occur only at the end of the preparatory process. In effect, consultation often takes place when the core decisions have already been taken and a draft regulation has been finalised. This combination of informal and formal procedures and its influence on the consultation schedule seems to be typical for traditional modes of consultation in Europe and is an important factor for public acceptance and support of consultations. In such models, public consultation is obligatory or conventional at one point in the pre-parliamentary process. In most countries it can be complemented at parliamentary level if a committee decides to consult selected experts either in written form or in a committee hearing, or if it decides to organise a public consultation on the draft bill. It should be noted, though, that parliamentary activities will mostly require the support of a majority and are therefore not used very frequently. In contrast, best practice recommendations are reflected in examples of new developments and countries with a specific consultation culture. The consultation praxis of the European Commission can serve as a model for these developments. Here, all stages of policy formulation and legislative drafting are accompanied by expert and/or open consultations, and are marked by milestones, for example Green Papers, White Papers and drafts.75 Other examples include Finland,76 Norway,77 Poland and the UK. All those models share the pragmatic approach of multi-stage consultations already outlined and use such procedures especially for legislative projects that are deemed important. 72 

Popelier (n 2) 140. HM Government, ‘Consultation Principles: Guidance’ (London, Cabinet Office, 2013) 7, available at https:// www.gov.uk/government/publications/consultation-principles-guidance (accessed on 25 November 2015). 74  A similar approach can be found in Finland: ‘Decision of Government: Consultation in Legislative Drafting: Guidelines’ 2 (2010), available at http://oikeusministerio.fi/material/attachments/om/toiminta/laitjalainvalmistelunkehittaminen/WvbIuN9RH/Consultation_in_legislative_drafting_guidelines.pdf (accessed on 25 November 2015). 75  cf on the development of such milestones in the context of the EU B Kohler-Koch and C Quittkat, De-Mystification of Participatory Democracy: EU-Governance and Civil Society (Oxford, Oxford University Press, 2013) 41ff. 76 ‘Decision of Government: Consultation in Legislative Drafting: Guidelines’ (n 74) 2; see also Divjak and Forbici (n 9) 21ff. 77  Royal Decree of 24 June 2005: Instructions for Official Studies and Reports; see also Divjak and Forbici (n 9) 23ff. 73 

Participation

 85

The UK Principles include another important rule regarding the date of consultation, holding that such exercises should not generally be launched during election periods.78 There may be exceptional circumstances, but in general departments should avoid taking action during election periods as it will cause competition with candidates for the attention of the public.

5.4.2.  Authorship and Responsibilities As we have already argued (see section 5.1), consultation on draft laws is primarily viewed as the responsibility of the executive branch of government. Thus, the government as a whole or competent ministers will be responsible for carrying out consultation procedures, including the preparation and sending out or publication of materials, the analysis of contributions, etc. An exemplary model of these responsibilities can be found in the Swiss Federal Act on Consultation Procedures.79 Generally, special bodies are not responsible for consultation procedures anywhere in Europe. However, the EU and Ireland feature two notable models. In the EU, only the Commission can make legislative proposals. This unique role is closely connected with the development of a specific consultation regime that provides the Commission with a comparatively powerful position in policy formulation through creating its own network of stakeholder relations.80 In contrast, the reform of the rather intransparent consultation practices of Ireland81 has made the Dáil (the lower house of the Irish parliament) the central actor in the pre-legislative consultation stage. The reason for this was to enhance the democratic process by allowing MPs to consider and debate bills at an earlier stage and ‘to allow for the extensive engagement of the public in law making’.82

5.4.3.  Objects of Consultation Legal regulations, manuals and practices on consultation throughout Europe focus on draft laws as objects of consultations. This is especially true of countries that adhere to traditional modes of consultation (see section 5.4 above). Objects of consultation are broader in the EU Member States, and such countries that have formal and open multi-stage consultation procedures or feature a specific consultation culture. There, Green Papers, White Papers or other kinds of policy proposals may be at the centre of consultation. However, it is often difficult to make a clear-cut distinction between a smaller or greater number of objects of consultation. A good example is Austria, which adheres to a rather traditional model that foresees a strong role for state organs and social partners in consultation procedures,83 but

78  UK Consultation Principles (n 73) 3; for a critical appreciation of the practice in Hungary see Drinóczi and Kocsis (n 11) 303ff. 79  Art 5 and 6 CPA (n 1). 80  cf B Kohler-Koch, ‘Civil Society Participation’ in Kohler-Koch and Quittkat (n 75) 173ff. 81 OECD, Better Regulation in Europe: Ireland (Paris, OECD, 2010) 83ff. 82  Oireachtas Library & Research Service, ‘Pre-legislative Scrutiny (PLS) by Parliament’, Spotlight 2014/8, 4. 83 H Schäffer, ‘Allgemeine Begutachtung und Konsultationsmechanismus in Österreich’ in L Mader and U Karpen (eds), The Participation of Civil Society in the Legislative Process (Baden-Baden, Nomos, 2005) 244ff.

86 

Felix Uhlmann and Christoph Konrath

uses new approaches in selected policy fields like environmental or social policy at a comparatively early stage of policy formulation.84 Another development can be discerned in the context of RIA: there may be separate consultations for the purpose of conducting an RIA or specifically on an RIA report, as is practised in Croatia.85 In other cases, an RIA report will be sent out along with a draft law. There is a close connection between the materials sent out for consultation and the format of the consultation (see section 5.4.5 below). Their combination seems to be crucial for the acceptance and impact of the whole process. If the objects of consultation are written and presented in a way that presupposes expert knowledge, it will be difficult for nongovernmental organisations (NGOs) and citizens with limited resources to successfully take part in the consultation. If we leave the cases of policy proposals, Green Papers, etc aside, we find that most countries in Europe have sent out draft laws and explanatory materials for consultation. In many countries, it is common to include a brief overview of the draft and its central goals.86 Explanatory materials range from explanatory notes on the draft, which can be found throughout Europe, to (extensive) reports on RIA and on the impact on the state budget, the environment, employment and equality issues.87 In some countries, it is necessary to explain which alternative modes of regulation were considered and why they were rejected.88 Further on in the process, preparatory studies, comparative tables and other background documents might be sent out along with the draft. If consultation is an open process, participants can comment on all materials sent out and are not restricted to the normative elements. There is an underlying assumption that any piece that will be submitted to the following stages of the legislative process might be changed and adapted. It should not be overlooked that such a practice may lead to ­crucial questions on the relationship between the draft law and the explanatory materials (see section 5.4.5 below). We refer to the question if and to what extent laws need to be ­justified,89 and how to deal with incongruences between draft law and materials. Another set of problems arise where legal drafts and RIA or budgetary reports are based on a different regulatory logic (ie a strictly legal and hierarchical mode of the former, as in the traditional Austro-Germanic conception of administrative law, and a performance and output-based mode of the latter). Here, it can be difficult to combine the concepts and formulations of the draft law with the goals and measures set out in the materials. In many countries, consultation will also be common with regard to delegated legislation. This is in line with efforts for better regulation that do not necessarily distinguish 84  Bundeskanzleramt Österreich, Standards der Öffentlichkeitsbeteiligung 2009, available at www.partizipation. at/fileadmin/media_data/Downloads/Standards_OeB/standards_der_oeffentlichkeitsbeteiligung_2008_druck. pdf (accessed on 25 November 2015). These standards were developed in a participatory process, see K Arbter and R Trattnigg, ‘Standards zur Öffentlichkeitsbeteiligung. Auf dem Weg zu effizienter und effektiver Partizipation’ in H Bauer et al (eds), Public Governance—Öffentliche Aufgaben gemeinsam erfüllen und effektiv steuern (Wien-Graz, Neuer Wissenschaftlicher Verlag, 2005) 295ff. 85  Act on Regulatory Impact Assessment (Official Gazette-Narodne novine No 90/11). 86  This is the case in Austria, Finland, Germany, Switzerland and the UK. 87  Such reports are common, for example, in Austria, Croatia, Poland and Slovakia. 88  In Austria, this is regulated in a decree of the Federal Chancellor, which is binding for all Federal Ministries. See § 5 Verordnung des Bundeskanzlers über Grundsätze der wirkungsorientierten Folgenabschätzung bei Regelungsvorhaben und sonstigen Vorhaben (WFA-Grundsatz-Verordnung—WFA-GV), BGBl II 489/2012. 89  cf on the debate in Germany C Waldhoff, ‘Gesetzesmaterialien aus verfassungsrechtlicher Perspektive’ in H Fleischer (ed), Mysterium ‘Gesetzesmaterialien’ (Tübingen, Mohr Siebeck, 2013) 78ff.

Participation

 87

between different forms and hierarchies of legislation.90 Examples for such practices can be found in Austria and in the UK, where drafts for delegated legislation are sent out in a similar form and with the same kind of explanatory materials as draft laws.91

5.4.4. Participants Best practices stress the importance of open and transparent consultation processes where the threshold for participation is low and the range of consulted persons and institutions is transparent.92 Generally, we can distinguish between three models of participation in consultation exercises: —— The consultation process is restricted to a fixed group of institutions. This is typical of traditional modes of consultations which aim to integrate federal or administrative units, advisory boards,93 or privileged public or private actors, for example trade unions, in the pre-legislative process. Here, consultation will most often be legally prescribed and failure to consult may lead to legal or political consequences (see section 5.5 below). —— The participants of the consultation process are defined according to the scope and purpose of the draft law in question. —— Consultation is an open procedure. In practice, many countries combine all three elements. This reflects the historical development of consultation practices, with some institutions or stakeholders having a longstanding right to be heard, and other consultation practices expanding and becoming more flexible in due course without anyone seeing the need to enshrine this in legal acts. Also, it seems quite common that certain institutions or persons are explicitly invited to take part in consultation procedures, while at the same time draft laws are published on the internet and anyone is free to comment on the draft. The Organization for Economic Co-operation and Development (OECD) has pointed out in a number of country reports that it views the selection of participants as crucial for the (perceived) success of consultation procedures.94 If consultation is restricted to a certain group, it becomes a closed procedure with a strong leaning towards negotiations. Here, the government can expect to confer with a well-informed group of consulted groups that will often share the underlying policy assumptions of the draft. If, however, consultation can involve other groups or is open to the public, the situation can change and formal consultation procedures can become a means to foster a broad public debate and citizen engagement. This can also be in line with research on institutional learning that profits

90 

Though in most cases it will be handled with a focus on expert partners and less publicity. See the examples on the dedicated websites: Austria: www.parlament/PAKT/MESN/ for draft laws and www.ris. bka.gv.at/Begut/ for draft laws and draft decrees; UK: https://www.gov.uk/government/publications?publication_ filter_option=consultations (accessed on 25 November 2015). 92  Popelier (n 2) 140. 93  This is a typical element of consultations in Belgium, see OECD, Better Regulation in Europe: Belgium (Paris, OECD, 2010) 95ff. 94  cf OECD, Better Regulation in Europe: Ireland (n 81) 91ff; OECD, Better Regulation in Europe: Belgium (ibid) 99ff. 91 

88 

Felix Uhlmann and Christoph Konrath

from open procedures and wide, even controversial, input.95 The key element is how participants are identified and involved. Here, laws and regulations throughout Europe remain vague, and seem to leave a lot to the discretion of the organiser of a consultation. The organiser may thus attain a rather powerful position if the invitation of participants is of importance for the assessment of replies.96 Transparency is therefore a crucial element of the invitation process, and guidelines such as those of Finland and the UK emphasise that organisers should act in a careful manner and be able to account for any invitation.97

5.4.5. Format This brings us to crucial questions regarding the practice of consultation, namely the format in which consultation procedures are conducted and the methods used thereby. Best practices emphasise the development of consultation policies and consultation techniques that help to get the best out of the consultation process for the organiser, the consulted and the general public, including parliament or other state organs that will discuss the draft law at later stages.98 Thus, successful consultations seem to depend on pragmatic and flexible approaches that have to be developed and evaluated in situ. What can be provided are a kind of methodical toolkit and the exchange of examples and experiences, as is done in consultation manuals or by the documentation of consultation procedures in databases.99 While many European countries have a basic set of laws or conventions organising and regulating consultation procedures, most remain silent on how consultations should be conducted. Even where there are more detailed regulations enacted in the wake of democratic reforms in former communist states, there is not much guidance on methods of consultation. In most cases, a draft law is sent out or published on a website and consulted institutions or the general public can make written statements. Apart from basic electronic tools, no further methods or instruments are used. In such a context, we can see that it will mainly be administrative bodies, corporatist institutions or NGOs with significant resources that actively take part and make substantial contributions in consultation procedures. The question remains whether the line between participation in negotiations and consultation, or lobbying and consultation, may not be blurred when a significant impact depends primarily on the resources available. It is interesting to look at the Austrian case in this context. Here, draft laws and materials will be sent out in a more technical than political style and consultation periods will usually be short. Thus, most replies will contain general remarks and only a few (mainly administrative or corporatist bodies) will be expected to answer in more detail. If a legislative project is deemed important, a large number of institutions or individuals will tend to share their replies by copying and pasting them. Further on in the process, a significant number of institutions will abstain from comments as they simply do not have the time or resources to participate.

95 

cf Smith (n 71) 173ff. cf for the research on the European Commission’s role in this context Kohler-Koch (n 80) 173ff. 97  See UK Consultation Principles (n 73) 2. 98  Popelier (n 2) 140. 99  See www.participedia.net (accessed on 25 November 2015) for such a collection. 96 

Participation

 89

The format and practice of consultation will more often than not be matters of specific traditions and culture. However, there are some examples where deliberate steps have been taken to enhance consultation procedures. A driving force seems to be growing disagreement with and abstention from consultations, criticism from international organisations or the ambition to implement better regulation initiatives. The development of consultation practices in the UK is often cited in literature on democratic innovations and deliberative democracy.100 Now-abandoned consultation practices are quoted as examples of superfluous exercises with little to no impact. From 2000, the government committed itself to effective consultation ‘targeted at, and easily accessible to, those with a clear interest in the policy in question’.101 A Code of Practice on Consultation was developed in three versions until 2008. In 2012, this was replaced by Consultation Principles that hold more or less the same seven consultation criteria and a set of standardised methods, ranging from basic information on a draft law, a mixture of open and closed questions on the draft, and group-specific approaches. The potentials of this approach become evident in highly contested matters that receive wide public attention. A well-documented example is the consultation on the Human Fertilisation and Embryology (Mitochondrial Donations) Regulations 2015. It comprised three scientific reviews between 2011 and 2014 and a public consultation carried out for the government by the Human Embryology and Fertilisation Authority, complemented by a number of parliamentary hearings.102 The public consultation included deliberative public workshops in several cities, in which participants were recruited to reflect a broad spectrum of the population. The aim was to explore public attitudes in depth. A public representative survey was then carried out, to be followed by an open consultation questionnaire that allowed responses from self-selected members of the public and stakeholder organisations on selected questions. Later on, there were open consultation meetings and patient focus groups.103 A second round of consultations asking for written statements to selected questions was then led by the government itself and lasted for 12 weeks.104 Overall, this process was deemed important for the following political, parliamentary and public debates. Another example is Ireland. The established consultation practices received a lot of criticism, especially from the OECD, with regard to their intransparency and closed character. This led to a complete restructuring of the process, which now lies in the hands of parliament, as has already been mentioned (section 5.4.2 above). Now, consultation is a parliamentary process focused on broad debate and combines parliamentary committee work with written statements, expert hearings and the possibility of conferences, seminars, etc.105

100 

cf Smith (n 71) 17–18. Government, ‘Code of Practice on Consultation’ (London, Department for Business, Enterprise and Regulatory Reform, 2008) 5. 102  cf S Barber and P Border, ‘Mitochondrial Donation’ (House of Commons Library Note SN/SC/6833, 2015) for a detailed account of the whole process. 103  cf Human Fertilisation and Embryology Authority, ‘Mitochondria Replacement Consultation, Advice to Government’ (London, Human Fertilisation and Embryology Authority, 2013). 104  See Department of Health, ‘Mitochondrial Donation. Government Response to the Consultation on Draft Regulations to Permit the Use of New Treatment Techniques to Prevent the Transmission of a Serious M ­ itochondrial Disease from Mother to Child’ (London, Department of Health, 2014). 105  A detailed account of the practice until 2014 is provided in Oireachtas Library & Research Service (n 81); see also Divjak and Forbici (n 9) 17ff. 101  HM

90 

Felix Uhlmann and Christoph Konrath

A distinct ‘better regulation’ approach has been used in Finland since 2011. Here, the explicit aim was to improve the clarity of legislation and thus promote the welfare of citizens and the competitiveness of businesses.106 This involved reforming the law drafting process in general, with special emphasis on consultation. The Finnish Guidelines on Consultation in Legislative Drafting focus on the planning of consultation as an important part of planning the whole legislative process.107 This involves careful planning of the timing of consultation and the combination of different consultation methods, such as setting up preparatory bodies consisting of several members, written statements, discussion meetings, negotiations, unofficial communication, surveys and online discussions, the compilation of summaries, and the possibility to review and comment on other statements. Transparency and online documentation of the process is self-evident. A comparatively new approach is crowdsourced knowledge generation and evaluation, which is used for consultation, policy formulation and even legislative drafting. Since 2013, Finland has carried out three experiments in ‘crowd-sourced law making’.108 The goal was to include as many citizens as possible in an open process fostered by social media and cloud-based IT tools. Expert knowledge was readily available for participants, but the aim was also that people should become experts in the field through active participation.

5.4.6. Deadlines As we have already seen, consultation depends to a large degree on the resources available, and has to take into account that it is mostly institutions that take part in consultation procedures. Therefore, best practices stress the importance of reasonable terms for consultation which are aligned with the needs of the consulting and consulted actors.109 It is important to have regard to both parties as a common objection to consultation is that it is time-consuming. However, empirical research110 shows that consultation procedures take up only a limited proportion of the entire decision-making process. Therefore, it is recommended that part of the preparatory processes be reserved for public consultation and that superfluous consultation procedures held only after political decisions have been reached be avoided. The latter is perceived as a deceleration of the process. Providing ample time is also important for respondents, as they might have to undertake research and coordinate internal decision-making processes. In countries with a strong consultation culture, such as Denmark, it is common to give reasonable time for comments without specifying particular deadlines.111 In Switzerland, the usual duration of the consultation period is three months.112 Other countries foresee 106  Government Programme of Prime Minister Jyrki Katainen’s Cabinet 2011, p 39; cf for a critical evaluation of the Finnish approach with special regard to the role of public consultations A Pakarinen, Taupautsutkimuksia lainvalmistelun kehittämisestä ja sääntelyn toimivuudesta (Helsinki, Publications of the University of Eastern ­Finland, 2012). 107  cf the overview at http://lainvalmistelu.finlex.fi/en/. 108  See http://thefinnishexperiment.com; cf T Aitamurto et al, Crowdsourced Off-Road Traffic Law Experiment in Finland. Report about Idea Crowdsourcing and Evaluation (Helsinki, Committee for the Future, 2014). 109  Popelier (n 2) 140. 110  Popelier (n 2) 144 provides examples from Belgium. 111  cf s 5.5 Guideline No 9801 of 3 June 2005 from the Ministry of Justice about good law quality. 112  Art 7, para 2 CPA (n 1).

Participation

 91

minimum periods that range from 30 days to 12 weeks. Recently, reformed or established consultation schemes, such as those in Finland or the UK, have not only stressed longer consultation periods, but have foreseen their extension if they overlap with holiday seasons so that voluntary organisations will have ample time to participate.113 For the EU, the timeframe is between eight and 12 weeks.114 Shorter deadlines of one week to 20 days will sometimes be possible, as in Poland, if justifications are given.115 If urgent decisions have to be taken, consultation procedures can sometimes be cancelled, or consultation can be organised by means of a conference of stakeholders, as practised in Switzerland.116 An obvious problem is the lack of any (self-)binding rules on deadlines, as this may lead to arbitrariness in setting the consultation agenda. Again, Austria serves as a good case. Originally, it was recommended that consultation periods be of at least eight weeks. ­However, this has subsequently been ignored by the government. In consequence, the practice of two- to four-week consultation periods has led to a significant decrease in non-state and non-corporatist participants in consultation procedures, which, in turn, has narrowed the results one gets out of the process.117

5.4.7.  Assessment of Feedback and Use of Results Best practices emphasise the importance of giving feedback to respondents about the impact of their contributions on decisions.118 This does not mean that the government has to take every respondent into account, or that it must use consultation in a ‘scientific’ way. Consultation is not necessarily politically neutral and its results may serve various functions (see section 5.2 above). The UK Consultation Principles provide for a detailed outline on how responses should be analysed in a qualitative way, how to give ample consideration to those who responded and how the response has been collated. The Principles ask the government to provide a summary of the views expressed, and to set out the decisions it has made and how it has taken into account the responses from the consultation. The report shall be published before or alongside any further action by the government.119 These reports are compiled in the weeks following the closing of the consultation and then published on the government’s website. Usually, they comprise relevant extracts of the statements from respondents. The full responses are, however, no longer available once the report has been published. There are similar procedures in other states, for example Croatia, Denmark, Iceland, Norway, Poland and Sweden, where the government or the government unit responsible

113 

Consultation Principles (n 73) 2. See the ‘minimum standards for consultation’ from the European Commission (n 5). 115  Resolution No 190 of the Council of Ministers of 29 October 2013 on the Rules of Functioning of the ­Council of Ministers, in Monitor Polski 2013, position 979. 116  Art 7, para 3 CPA (n 1). 117  A detailed study is lacking, but the development of participation in consultation exercises can be traced at the permanent documentation of all such exercises at www.parlament/PAKT/MESN/ (accessed on 25 November 2015). 118  Popelier (n 2) 140 and passim; see also Art 6, para 1 CPA (n 1). 119  Consultation Principles (n 73) 2–3. 114 

92 

Felix Uhlmann and Christoph Konrath

for drafting shall provide a report on the consultation. In Croatia and Poland, there is a duty to comment on all proposals and objections submitted in the consultation exercise, and to give reasons why a comment has been rejected.120 Such comments may include clarifications of obvious misinterpretations by the participants. The reports and comments shall be submitted to parliament along with the draft bill and materials. In other countries, the assessment of consultations is not formally regulated and it remains at the discretion of the government to publish its considerations or reply to ­participants in the consultation procedure. This can, as pointed out in section 5.2.2 above, be a source of discontent, as it will often remain unclear if and how the government has used responses from the consultation exercise. However, it is standard practice for contributions to remain publicly available and can be referred to in public debates and parliamentary procedures. It should also be noted that making contributions available can have special significance for legal debates and court proceedings. In Austria, for example, it is a common practice for complainants before the Constitutional Court and the Administrative Court to refer to arguments presented in the consultation procedure, and in some cases even the courts have used them for historical interpretation.121

5.4.8.  Consultation and Parliamentary Procedures In general, consultation exercises are regarded as a valuable resource for parliamentary ­proceedings. They provide parliamentarians with practical insights and knowledge, and with first-hand opinions of stakeholders and the general public. This can supplement (or compensate for) impartial research notes provided by parliamentary research services. Usually, parliamentary rules of procedure enable parliamentary committees to organise their own consultations by means of written statements, expert hearings or conferences. There is, however, no general duty to do so. Consultation in parliamentary proceedings occurs on a case-to-case basis and is more narrow than public consultations in the prelegislative phase. Usually, parliaments may select only a small number of institutions and persons for consultation, and will turn to specific stakeholders or widely acknowledged experts. Contributions from pre-parliamentary consultations may nevertheless serve as a valuable source for whom to consult and hear in parliament. Some parliaments have special parliamentary consultation procedures. We have already mentioned Ireland, where parliament is now responsible for pre-legislative scrutiny, which will often include public consultations. In Croatia, the parliamentary rules of procedure not only encourage the invitation of public officials, scholars and stakeholders, but also foresee their active involvement. In principle, parliamentary bodies may appoint up to six ‘external members of working bodies’. Apart from voting, they have the same rights as Members of Parliament.122 In Poland, the parliamentary rules of p ­ rocedure of

120  cf Art 11, para 3 of the Croatian Right of Access to Information Act (Official Gazette-Narodne novine No 25/13); Resolution No 190 of the Council of Ministers of 29 October 2013 (n 115). 121  cf Collection of the Decisions of the Administrative Court VwSlg 17953 A/2010, a reference case in asylum matters that emphasises the influence of the consultation procedure on the interpretation of certain regulations. 122  Art 57 of the Standing Orders of the Croatian Parliament.

Participation

 93

the Sejm distinguish between draft bills presented by state organs and by the people and those presented by members of the Sejm or Sejm-committees. As a rule, the f­ormer can only be presented after a public consultation, but there are no legal consequences if such consultation is not undertaken. The latter have to undergo public consultation organised by the Sejm before a first reading can be held if such obligation is provided by a specific law (depending on the subject of the bill) or if a resolution to hold a public hearing has been adopted by a committee considering a particular bill.123

5.5.  Regulation on Consultation As already mentioned, the regulation and practice of consultation are not necessarily coherent. More often than not, legal regulation may only cover (very) general aspects or prescribe the consultation only with selected state organs or stakeholders. Throughout Europe there is a complex interplay between legal regulations, official guidelines and established practices.

5.5.1.  National Constitutions and Laws In Sweden, Chapter 7, Article 2 of the Instrument of Government stipulates at the constitutional level a duty of the government to obtain the necessary information for the preparing of government business. This includes that organisations and individuals shall be given an opportunity to express their opinions. The Portuguese constitution does not foresee such a general duty, but stipulates the right of workers commissions, trade unions, certain national associations of local authorities and the Data Protection Authority to be heard in relation to laws and regulations that fall within their competences. Failure to do so may render a legal text unconstitutional.124 These two models—a general duty to consult or specified rights of certain bodies to be consulted—can be found throughout Europe. General duties are regulated in acts on the organisation and duties of the public administration,125 rules of procedure of executive governments,126 rules or guidelines on procedural and organisational aspects of the 123  Arts 34.3 and 70a.1 and 2 of the Standing Orders of the Polish Sejm; Art 79a1 and 80.1b Standing Orders of the Senate. 124  See Arts 54 n 5, 56 n 2, 60 n 3, 65 n 5, 80, 227 n 1, 229 n 2 and 249 of the Portuguese Constitution. 125  cf Bosnian Law on Administration (Official Gazette of Bosnia and Herzegovina Nos 32/02 and 102/09); the Instructions for Official Studies and Reports (ch 5) in Norway (n 77); Art 77 of the Serbian State Administration Act (Official Gazette No 79/2005 in the version of 99/2014); Art 41 of the Rules of Procedure of the Serbian Government (Official Gazette No 61/2006 in the version of 76/2014). 126  cf Rules of Procedure of the Council of Ministers (Official Gazette of Bosnia and Herzegovina No 22/03); Uniform Rules for Legislative Drafting in the Institutions of Bosnia and Herzegovina (Official Gazette of Bosnia and Herzegovina Nos 11/05, 58/14 and 60/14); Rules on Consultation in Legislative Drafting (Official Gazette of Bosnia and Herzegovina Nos 81/06 and 80/14); Resolution No 190 of the Council of Ministers of 29 October 2013 (n 115). Rules of Procedure of the Government of Slovenia (Poslovnik Vlade Republike Slovenije, Uradni list RS št 43/01 in the version 10/14).

94 

Felix Uhlmann and Christoph Konrath

­legislative process,127 acts regulating access to information128 or acts regulating public consultation in particular.129 Specified rights of certain bodies or entities will be enshrined either in acts that establish and privilege such bodies or in acts regulating particular affairs in which such bodies shall have a say.130 If such acts hold more than a duty to consult or a statement on the importance of consultation, there will be regulations on responsibilities, timelines and participants, and sometimes on the format and assessment of consultations. Usually, such acts will also consider emergency situations or other events in which a consultation may be abandoned completely or carried out in a special format, for example a conference of stakeholders as in the Swiss example.131 Failure to conduct proper consultations or to consult at all may interrupt or delay the legislative process in a number of countries. In Denmark, a bill cannot be tabled for the first reading by convention unless the parliament has received the responsible ministry’s summary of the consultation. In Croatia, similar practices are regulated by law, but depend on the insistence of all involved state organs.132 In general, however, the regulations and practices on consultation do not foresee stronger sanctions or consequences. In contrast, the obligatory consultation of federal entities (as in Austria)133 or certain public bodies (as in Portugal)134 may lead to legal consequences as the conferral of financial burdens or even cause the unconstitutionality of an act.

5.5.2.  National Ordinances, Manuals and Practice In many cases, such acts will only address certain aspects of the consultation practice and the government will be free to broaden the scope of participants, provide technical and organisational support, etc. New practices, as exemplified here by reference to the UK and Finland, tend to be based on guidelines that allow for an approach that is both more detailed and flexible. In that way, it is possible to express the importance of consultation procedures, the commitment of the government, and methodological issues such as ­consultation techniques or assessment of contributions in plain and widely accessible ­language. It may also allow one to view consultation procedures as a democratic, viz a legal/ bureaucratic, exercise, as often expressed in legal acts. On the other hand, such informal practices rely heavily on the willingness of the executive government and the 127  cf the Guideline No 9801 of 3 June 2005 of the Danish Ministry of Justice on law quality, available at https://www.retsinformation.dk/Forms/R0710.aspx?id=2272 (accessed on 25 November 2015); the Ministry of Justice Finland: Legislative Drafting Process Guide, available at http://lainvalmistelu.finlex.fi/en/ (accessed on 25 ­November 2015); Legislative Rules of the Government of the Slovak Republic first adopted on 25 May 2010 by Government Resolution 352. 128  Right of Access to Information Act (Official Gazette-Narodne novine No 25/13) (n 120). 129  cf the Hungarian Law on Public Consultation, 2010, évi CXXXI törvény, or the Swiss Federal Act on the Consultation Procedure (n 1). 130  See the examples in Schefbeck (n 67). 131  See n 116. 132  See Art 23, para 2 of the Act on Regulatory Impact Assessment (n 82); Art 174 (para 4) of Standing Orders of the Parliament of Croatia (Official Gazette-Narodne novine No 81/13). 133  See n 64. 134  See n 124.

Participation

 95

administration to follow and develop them. They are recommendations that do not involve any formal consequences if they are not followed.135

Further Reading Divjak, T and Forbici, G, ‘Public Participation in Decision-Making Process, International Analysis of the Legal Framework with a Collection of Good Practices’ (Ljubljana, Center for Information Service, Co-operation and Development of NGOs, 2014) 4. European Commission, ‘Towards a Reinforced Culture of Consultation and Dialogue— Proposal for General Principles and Minimum Standards for Consultation of Interested Parties by the Commission’, Consultation Document COM [2002] 704 (11 December 2002). European Commission, ‘Review of the Commission Consultation Policy’, Staff Working Document SWD [2012] 422 (December 2012), available at http://ec.europa.eu/ smart-regulation/better_regulation/key_docs_en.htm#_consultation. Flückiger, A, ‘Consulter pour mieux légiferer: Utilité des procédures de consultation préparlementaires’ (2007) 3 LeGes 185. Mader, L, ‘Die Rolle der Zivilgesellschaft und die Bedeutung der Konsultationsverfahren für die nationale Rechtsetzung in Europa’ in H Schäffer and J Iliopoulos-Strangas (eds), State Modernization in Europe (Athens, Ant N Sakkoulas, 2007) 56. Mader, L and Karpen, U (eds), The Participation of Civil Society in the Legislative Process (Baden-Baden, Nomos, 2005). Mross, O, Bürgerbeteiligung am Rechtsetzungsprozess in der Europäischen Union (Berlin, Duncker & Humbolt, 2010). Müller, G and Uhlmann, F, Elemente einer Rechtssetzungslehre, 3rd edn (Zurich, 2013) 101ff. Van Nuffel P, ‘Le rôle de la société civile et les procédures de consultation à caractère ­supranational’ in H Schäffer and J Iliopoulos-Strangas (eds), State Modernization in Europe (Athens, Ant N Sakkoulas, 2007) 44. OECD, OECD Handbook on Information, Consultation and Public Participation in PolicyMaking (Paris, OECD, 2001), available at http://www.oecd-ilibrary.org/governance/ citizens-as-partners_9789264195578-en. OECD, ‘Consultation and Participation’ in Open Government: Fostering Dialogue with Civil Society (Paris, OECD, 2004), available at http://www.oecd-ilibrary.org/governance/ open-government_9789264019959-en. Popelier, P, ‘Consultation on Draft Regulation—Best Practices and Political Objections’ in L Mader and M Tavares de Almeida (eds), Quality of Legislations, Principles and Instruments (Baden-Baden, Nomos, 2011) 139ff. UK Government, ‘Consultation Principles: Guidance’ (2013), available at www.gov.uk/ government/publications/consultation-principles-guidance. 135  For a debate on a complaint about the actual practice in Finland see J Pekannen, ‘Saivatko virkamiehet luvan pyyhkiä takapuolensa oikeusministeriön ohjeilla?’ (2014), available at http://jpekkanen.puheenvuoro.uusisuomi. fi/177657-saivatko-virkamiehet-luvan-pyyhkia-takapuolensa-oikeusministerion-ohjeilla.

96 

6 Goals and Measures of Legislation: Evaluation MAURO ZAMBONI

This chapter assesses a process for evaluating the connection between legislation and its goals (or intended effects) in a certain community. In particular, the purpose of the chapter is to clarify both the terms of and modalities through which the connection between ideals (goals) and evaluation of the final product (measures) of a piece of legislation is created. Three main aspects of legislative goals can be relevant to the evaluative process: the structure of the legislative goals, their function and their location. The structure of the legislative goals comprises three ideal-typical categorisations: where the legislative goal is intended to be realised (positioning), when the legislative goal is intended to be realised (perspective) and how the legislative goal can be traced (visibility). The functions of the goals of legislation can be divided based on a chronological criterion (and can be subdivide between preliminary expectations and final expectations) or by assessing where the intended impact is located in time (the micro-, meso- and macro-functions of the legislation). The third and final perspective in evaluating the goals of legislation involves their location, ie considering the list of documents that can be relevant (as either an exhaustive or an open list) and examining this material using either traditional or non-traditional analytical instruments.

6.1. Introduction Legislation is not an object with a value in and of itself; it is not something that can be evaluated regardless of the premises or consequences, as can be the case, for instance, for the welfare of a society. The essence of legislation lies primarily in its function as a tool which various actors (primarily political actors, but also public agencies, judges and so on) use to implement certain ideas, or ideal visions, in a certain community of people. This designation of legislation as a tool means that it cannot be evaluated as being ‘good’ as such; it must be evaluated in respect of the results the legislative process is intended to achieve (eg the original intentions of the legislators) and those that it actually has achieved (eg whether or not the results correspond to the original intentions). In other words, the essential component of evaluating the quality of legislation is its relational nature: putting the ideas of a legislator, for instance, in relation to its concrete results (eg in terms of concrete changes in the society). In this respect, the overall aim of this chapter is not so much to discuss the criteria used to assess what is traditionally defined as the ‘internal’ quality of legislation—namely, all

98 

Mauro Zamboni

the technical or drafting qualities of the legislative acts per se (eg clarity, consistency, their ­efficient and timely production, or their consistency with basic legal principles). Instead, the focus here is on evaluation of the ‘external’ connection between legislation and its goals (or intended effects) in a certain community. Moreover, the goal of this chapter is not primarily to improve this connection between goals and measures of legislation. The task is much humbler, but to some extent more fundamental—namely, to clarify both the terms of and modalities through which the connection between ideals (goals) and evaluation of the final product (measures) of a piece of legislation is created. This analytical approach to the issue can be considered fundamental because, by clarifying the terms of the discussion (eg what a legislative goal is) and the way these terms relate to each other, it is possible to establish the operative platform and methodological tools needed to investigate and evaluate legislation in different countries within a common conceptual framework. For this reason, this chapter focuses on three main aspects of legislative goals which can be relevant for evaluation: the structure of the legislative goals, their function and their location.1 At the same time, while distinguishing these aspects of the goals, each of the three sections of the chapter will also clarify how such features affect (and sometimes determine) the kind of evaluative criteria one should use to measure the final product of the legislative processes.

6.2.  What a Legislative Goal is Not Before beginning a proper investigation of the various aspects of legislative goals, it is ­necessary to define the fundamental difference between ‘goals’ and ‘effects’ of a piece of ­legislation. In both public debate and scientific discussion, these two terms are often erroneously used as synonyms to indicate the impact that legislative measures have, or could have, upon a certain community and/or legal system. However, this practice is incorrect because the two terms refer to two different phenomena associated with the legislative process. This distinction between legislative goals and legislative effects is based on a distinction made in the socio-legal literature (and the sociological literature in general) between ‘function as purposes (or goals)’ and ‘functions as effects’. To summarise this roughly, a distinction has been drawn between functions of law as the effects or actual consequences of the law (or a specific branch of law) on a community (‘what law does’) and the functions of law as the purposes or goals the law is intended to have in a community (‘what law is thought to do’).2 For example, the function-effect of labour law can be the maintenance of existing

1  These three features are ideal-typical points of observation of the legislative process; in reality, they tend to overlap and interrelate. As to potential interrelation and overlapping, one should consider (as is presented below) such aspects as the natural correlation between the visibility of legislative goals (section II) and the paths to be taken to find these very goals (section IV). However, for all the ideal-typical constructs, the elucidation of these features can be helpful when dissecting and tackling the extremely complicated issue of identifying and measuring goals of legislation. 2  For an example of rgw terminological confusion between ‘functions as purposes’ and ‘functions as effects,’ see, eg, H Hansmann and U Mattei, ‘The Functions of Trust Law: A Comparative Legal And Economic Analysis’ (1998) 73 New York University Law Review 472.

Goals and Measures of Legislation

 99

economic forces in the labour market (eg employer associations and trade unions), whereas the function-purposes are the promotion and defence of all individual employee rights. By shifting this distinction into the world of legislative studies, one notices an interesting feature: while legislative goals somehow indicate the modifications that legislators aspire to produce, the effects of legislation in general denote the modifications that legislative acts bring about or are capable of bringing about in a community or a legal system—regardless of (and sometimes in contrast to) the original aspirations of the legislative bodies. In other words, while the goal of legislation points only to the intended effects of the law making (‘what law is intended to do’), effects can (and often do) also include the unintended consequences (‘what law does’, which often takes place instead of or regardless of the legislator’s original goals).3 For example, in social-democratic Sweden, the goals of legislating on the matter of equality in labour relations have always involved the promotion and defence of all individual employees’ right of not being discriminated against because of gender or race. However, the effects—in particular, through the ‘narrow’ application by the labour courts—have been the preservation, as much as possible, of the right of the individual employer to determine the structure of his or her working organisation, while leaving the equality goal to be reached through mass media (eg through boycott) or political pressures (eg by denying to private companies the possibility of entering into contracts with public agencies).4

6.3.  Structuring the Legislative Goal Once the distinction between legislative goals and legislative effects has been clarified, the first essential task is to provide a definition of a ‘goal’ of legislation, in the sense of the intended effects of the legislative law-making process. There are three potential idealtypical categorisations that can form the structural framework of such a definition: where the legislative goal is intended to be realised (positioning), when the legislative goal is intended to be realised (perspective) and how the legislative goal is traceable (visibility). The first categorisation deals with the positioning of the goals of legislation either outside or within the legal system, a distinction that to a certain extent tends to overlap with the distinction between goals and instruments of legislation. In this respect, a necessary analytical step would be the distinction between the goal of legislation as intended outputs and the legislative goal as intended outcomes.5 In this context, outputs are the intended impacts of the legislative measures in the legal arena where the process itself has taken place (eg changes of the legal system concerning hiring procedures). The outcomes of the legislation mark the intended effects that such impacts have on the surrounding environments (eg changes expected in the concrete behaviours of employers). It is important to distinguish 3 

See R Cotterrell, The Sociology of Law: An Introduction, 2nd edn (London, Butterworths, 1992) 72–73. ‘gap’, at least from the labour union perspective, between purposes and effects in labour law has been particularly pointed out by critical legal theory. See, eg, KE Klare, ‘Critical Theory and Labor Relations Law’ in D Kairys (ed), The Politics of Law: A Progressive Critique, 2nd edn (New York, Basic Books, 1990) 64–69, 80–81. 5  This separation of outputs from outcomes is actually an adaptation of the results reached through a long series of studies developed in political science. See, eg, FG Castles, Comparative Public Policy: Patterns of Post-war Transformation (Cheltenham, Edward Elgar Publishing, 1998) 248–92. 4  This

100 

Mauro Zamboni

between goal as intended outputs and goal as intended outcomes in order to understand that, sometimes, a single intended legislative output is created in order to produce multiple and diverse legislative outcomes. For example, the change of a single regulation within the financial market (a single goal-output) is produced with the intention of stabilising the internal financial market, increasing the trust that the public places on the financial system as a whole, and gaining legitimacy in the international arena (several goal-outcomes). On the contrary, legislators sometimes produce several coordinated changes in the regulatory landscape, ie several legislative outputs, in order to generate a single outcome in the community. For example, legislative bodies can enact several legislative measures to affect the taxation law (by granting certain taxation benefits), administrative law (by limiting public procurement) and corporate law of a country (by having a legislative say in corporate governance)—all done to promote a more gender-equal structure in the board of directors. As to the second categorisation, this type of structuring of legislative goals springs from the perspective adopted while investigating and evaluating these goals. The second evaluation platform looks at the outputs and outcomes in relation to time. In this respect, one can distinguish between a diachronic dimension of legislative goals (when observing them from their placement in a specific period in a certain context) and a chronological dimension (when looking at the goals from a more in-time perspective). The diachronic dimension indicates that the evaluation of a certain legislative goal is focused on a specific, narrow period of time, both within and outside a certain legal field. Using a military metaphor, this can be defined as the tactical goal of the legislation; an example would be the goal of lawmakers to put the financial reporting of larger corporations under public scrutiny by having the companies’ annual reports controlled and approved by a state agency. All that is needed to evaluate the goal from a diachronic perspective, eg during an evaluation of the impact assessment, is to connect the intended outputs (or outcomes) behind a certain act (eg the goals expressed in the parliamentary discussion or in the preparatory works of this specific act) to the actual outputs (or outcomes) that the act has produced. The advantage of using the diachronic dimension in evaluating a certain legislative act is the relative facility in performing it, due to the relative facility in tracing both the intended goal and the actual effects. However, this kind of evaluation can be used only when evaluating the goals of ‘focused legislative reforms’—changes in the regulatory regime that tend to be limited in time (ie where their actual results can be evaluated within a very narrow period of time) or in scope (ie the reform does not significantly affect ­surrounding regulatory regimes). There is also a second possible perspective from which to evaluate the legislative measures, a point of observation that can be defined as ‘chronological’ because it tends to observe the measurement of legislative goals with respect to a timeline. From this chronological standpoint, the goals of legislative work in a certain area should encompass the broader intent of a group of legislative measures, over time and upon a certain area. In this respect, the chronological dimension of these legislative measures tends to overlap with the investigation of the legislative policy that presides over the regulatory field under observation. Using the previous metaphor, the chronological dimension then tends to force the observer to move to higher ground and look instead at the strategic goal of the legislation in a regulatory area. For example, this long-range vantage point can lead to investigation and evaluation of the goal behind multiple, coordinated legislative measures which result in increasing direct intervention of state agencies in matters of financial-market regulation.

Goals and Measures of Legislation

 101

It thus appears quite natural that investigating the overarching goal of a certain reform is advantageous in that it offers a more comprehensive image of what the legislative body is aiming to achieve; the process connects all the dots (ie specific legislative measures in specific legal areas) on the law-making map. However, defined as such, the chronological aspect of legislative goals is naturally associated with problems stemming from their being overarching and problems that arise over time. The overarching aspect means that the chronological structuring of legislative measures tends to assume that there is always a ‘grand goal’ behind each legislative measure. In this way, it tends to bypass legislative measures that are intended to have only a limited impact, ie all new regulatory provisions directed at offering ad hoc solutions. One example would be finding the goals of specific legislation designed to deal with specific natural disasters (eg legislative measures directed to postpone the collection of income taxes in an area afflicted by an earthquake) or unexpected financial crises (eg a legislative act directed to nationalise a bank in financial trouble). Secondly, if looking to their chronological structure, the evaluation of legislative goals can present a problem connected to their ‘over time’ character. This broader approach always entails the risk that, in evaluating a legislative reform, the analyst will reconstruct a unique goal ex post by artificially connecting various legislative measures, which in reality are expressions of different (and often conflicting) goals intended by the different legislators. For instance, evaluating the goal and measures behind legislative reform in higher education can be an impossible task. For several complex historical and cultural reasons (at least in Western countries), attempts at legislative reform of higher education often comprise an inconsistent puzzle of partial reforms (eg some reforms through parliamentary acts and others through governmental decrees), made by different actors (eg different parliamentary majorities during a period) with different (and conflicting) goals (eg increasing student participation v strengthening ties between education and markets). The third categorisation needed to structure a common investigative and evaluative platform for defining the goal of legislation involves evaluative tools around the visibility of legislative goals. Visibility refers to the capacity of an external observer to determine the position of the goal of a new or existing statutory regulation, in particular with respect to what are traditionally considered the legally relevant documents. Based on this idea of goals’ visibility, one can trace two major ideal types of goals. The first one the patent goal: these are the goals of legislation that have been formally endorsed by the legislative agencies and therefore are clearly traceable in the very legal (or quasi-legal) documents that are connected to the process leading to the enactment (or rejection) of certain legislative provisions. For instance, patent goals can be evaluated by looking at the legal texts, including the preambles, or by tracing them back to the preparatory works and all the official documents recording the parliamentary debate behind a certain legislative process.6 The second ideal- type of goal, based on goal visibility, is the hidden goal: this is the goal embraced by the legislative players, but not in a formal way; hidden goals are not present in directly relevant legal or quasi-legal documents. Instead, this type of legislative goal tends to be traceable only by exploring non-legal documents and material. For instance, the goals

6  It is worth mentioning that among the documents where the patent goals are traceable, one should consider both the legal documents connected to the legislation (eg travaux preparatoires, explanatory memoranda) and those instances where the legislation itself actually sets out the goals (eg the preamble to legislation).

102 

Mauro Zamboni

of a legislative regulation can be hidden in the text of debates in general assemblies (often in a minister’s oral presentation of a bill); they can also be reconstructed and/or discovered through qualitative interviews with law-making officials or by consulting the press releases published by the political parties in connection with the promulgation of a certain act. In this respect, it should be pointed out that coexistence and even tension are qualities that can characterise the relationships between goals with different visibility. Contradictions and even conflict between patent goals and hidden goals endorsed in the same legislative process can be quite common—a situation created by poor legislative drafting, for instance, or by extensive politicisation of the legislative issues at stake, or simply by contradictory political goals (where a compromise around a legislative act is reached, but the different legislative actors aim at conflicting goals based on the same text). For example, a political party’s press release can reveal the ‘true’ (or hidden) goal (limiting the immigration quotas for people coming from certain countries) of a legislative reform (increasing the fees for obtaining a visa) with a different patent goal (raising revenues to cover public agencies’ increasing costs due to the mounting immigration). At the same time, tension may exist among patent goals of the same promulgated act; legislative regulation of penal procedure may be carried out with the goal of having ‘swift and efficient’ trials, while also having the explicit task of offering comprehensive and in-depth legal protection for the accused.7 To sum up, the first step in evaluating legislative goals is to look at their structure. This can be determined by positioning the goals (as outputs in the legal system or outcomes in the areas surrounding the legal world); determining their time framework (as the product of a diachronic process or in the tracing of a longer chronological path of several changes in regulatory regimes); or defining the visibility of the legislative goals (whether they are explicitly stated in legal or quasi-legal sources or are somehow hidden in legally irrelevant documents).

6.4.  The Function(s) of the Legislation and Its Goals The second aspect through which it is possible to evaluate the policy and goals of legislative processes concerns the function of the legislation. In tackling this complex question, two fundamental (even philosophical) issues must be confronted by all legislative studies. First, one should question whether legislation, or even law in general, needs a goal.8 For many legal scholars, such as those applying all the various critical approaches, the law—and consequently its creation—has a function (eg the legitimisation of the elite’s power), but

7  It is worth noting that the internal conflict among different patent goals may be resolved by the ­‘practical application’ of the enacted legislative regulation, ie by having the judicial bodies and/or the public agencies (through various legal interpretative tools) favour one goal over the other. In other words, conflicting legislative (and constitutional) goals are the typical parameters for a balancing test. 8  For instance, as pointed out by Max Weber, and more recently by Niklas Luhmann, the very goal of legal ­systems is to somehow ‘stabilise’ the element of uncertainty that is typical of non-legally regulated relations among actors. See, eg, N Luhmann, Law as a Social System (Oxford, Oxford University Press, 1994) 148, 152–53; M Weber, Economy and Society: An Outline of Interpretative Sociology, ed G Roth and C Wittich (Berkeley, University of California Press, 1978) 34.

Goals and Measures of Legislation

 103

lacks a goal (eg it is simply a show for the masses). In other words, for many scholars (and for some of the people working within the legislative process, such as the representatives of certain political parties), the presence of a function of law making does not necessarily require the law making to have certain goals. With regard to the second philosophical question, even if one argues that the law and its legislative making must have a goal, this standpoint leaves the door open to another (and ideologically deeper) issue: what are the reasons, or, more accurately, the arguments, that support the idea that each piece of legislation must have a goal to pursue? The answers to such a fundamental question tend to fall across a broad spectrum. At one end of the range are those claiming that legislative processes are ‘naturally’ goaloriented. This is the ‘structural’ argument. For these persons, every law-making cycle is a process requiring a direction, whether explicit (eg in the preparatory works) or implicit (eg in the hidden agenda of a political party). At the other end, others maintain that goals belong to the legislative process because this process is part of a broader environment. This is the ‘institutional’ argument. For instance, the legislation is necessarily part of a broader context of beliefs (in particular, in the political arena) and the primary objective of these other environments (eg the political party) is the very implementation of these ideas upon a certain community through legislative measures. Irrespective of the position one takes in these philosophical questions, if one shifts one’s attention to the investigation and evaluation of the function of legislation, it is possible to distinguish at least two major ideal typologies according to which one can consider the functions of legislative work in relation to its goals. In other words, one can evaluate the goal of a legislative process according to two basic functional parameters: time and intended impact. The first typology focuses upon the time of consideration when evaluating the function of legislation. On the one hand, there are the evaluations of the function of law in relation to its preliminary expectations: these are assessments of the role expected to be played by new legislative regulation (in a certain legal system and/or in a certain community) as soon as it becomes valid law, ie as soon as it acquires the formal validity of statutory provision. In particular, such preliminary expectations of the function of legislation can be increasing (or establishing) the addressees’ political legitimacy, or the legal legitimacy of lawmaking agencies and/or of parties sitting in national assemblies. For example, the ­preliminary expectation of a legislative measure allowing state intervention in the banking system could be that the legislation will guarantee the solidity of the national financial market towards the stock market’s operators (regardless of whether or not the state will actually intervene in the near future). A preliminary expectation may also be seen in the general effects of (good) legislation, such as legal certainty or legal predictability, as well as adequate statutory interpretation. For example, the immediate expectation of a legislative provision clarifying certain terms of a previous act could be that the provision sends a clear signal to the public and other legal actors that the political arena intends to clearly limit the discretionary power of judicial bodies (regardless of whether judges will then adhere to this or instead persist, for instance, in their activist approach). On the other hand, each legislative law-making process is imbued with final expectations, ie the function that the legislation will have when it becomes law. In other words, the law will concretely affect the behaviour of the addressees. In this respect, the observer’s ­attention will then be drawn to evaluating the kind of function the legislative act will have

104 

Mauro Zamboni

when it has (or is going to have) a concrete impact upon the behaviour of the ­community and/or the legal actors. The issues of statutory interpretation or those of Regulatory Impact Assessment can be considered as the traditional instruments for investigating and ­evaluating the final expectations of a legislative measure. Using the previous examples, the final expectation of a legislative measure allowing state intervention in the banking system could be that the measure will legislate an area previously left to other kinds of regulation, such as self-regulation or transnational law. As to the other example, the final expectation of a legislative provision clarifying a previous act can be that the provision will fulfil the ‘true intentions’ of the legislator, that is, achieving the results intended by the drafters of the act ‘under attack’ by judicial activism. The second typology that can be used to investigate and evaluate the goals of legislation is based on the areas of intended impact of the legislative product, namely the extension of the field that the legislation is intended to affect. In this respect, and looking at it from a legal actor’s perspective, one can observe how the goal of legislation can be intended to have a micro-function, a meso-function and a macro-function. The micro-functional goals of legislation refer to the goals of certain legislative measures that are intended to influence a specific legal area; in other words, in the evaluation process of the legislative goals, attention is focused upon how, and the extent to which the legislative measures are intended, to change a certain part of a regulatory regime. For example, looking at the regulation of corporate law, a micro-functional goal of a country’s new statutory provision cold be the introduction of a requirement on corporations: in order to be registered as incorporated and to have headquarters there, these companies must ensure that at least 30% of the directors on the board are women. Meso-functional goals are legislative aims intended to affect an entire legal area; these goals tend to identify the changes that are directed at modifying an entire regulatory regime (or at least a substantial part of it). Examples include legislative measures which affect the entire structure of corporate law by allowing public agencies to have a say in all areas (eg the internal configuration of corporate governance) which traditionally have been considered by the legal world as being the exclusive domain of the private parties concerned. Finally, the macro-functional aspects of legislative goals point out the intended effects that lawmakers plan for society at large; in this respect, measuring this type of goal implies the evaluation of not only the outputs of the legislative process (ie how the process is intended to affect the legal landscape), but also its outcomes (ie how the legal changes are intended to affect the world surrounding the legal field). This can be the case, for example, of a new corporate law: imposing a quota on female membership in the board of directors and thereby changing the corporate governance regulatory regime also aims in general at encouraging gender equality on the higher levels of the corporate world. Before concluding this section, it is necessary to observe that not all legislative measures present goals that are intended to cover all three (micro-, meso- and macro-) dimensions of the legislative function. Not all legislative processes are set into motion with the intention of changing a certain legal area, the entire regulatory regime and the world surrounding the law. By evaluating legislative measures, it is sometimes possible to observe how goals in legislation are constructed with the intention of changing only one functional dimension. For instance, legislation on technical issues of civil procedure often have an intended impact only at the micro-level, for example to make the procedure smoother for the parties

Goals and Measures of Legislation

 105

in the process or in terms of offering a better opportunity for one of the parties to expose his or her claims. It is also common for legislative processes to aim at changing only two functional dimensions. For example, so-called framework legislation is often intended by its very nature to have only meso- and macro-functions, while giving other actors (mostly public agencies) the task of setting the micro-goals via other regulations than statutory provisions.

6.5.  Locating the Legislative Goals The third and final aspect of evaluating the goals of legislation has to do with determining their location, ie where these goals can be found. Although it might seem trivial, there are some fundamental normative questions that must be answered before investigating and evaluating the goals of a piece of legislation: which documents or material should the legal actor or legal scholar consult to find the goal of the legislation? In other words, a preliminary step in the evaluative enterprise involves creating a list of document types that can be considered fundamental for grasping the goals of a legislative regulation. For example, can the preparatory works be considered as a reliable source, or should one also consider the records of the parliamentary debates? In this respect, it is very difficult to establish a definitive list of documents, because this depends very much on the legal system under consideration and the legal culture dominating it. For instance, certain legal systems (like the Nordic ones) give the preparatory works equal status with the legislative acts, whereas for others (eg Italy) the preparatory works play a more limited (if not insignificant) role when it comes to the enterprise of identifying the goals of the legislation. However, despite these differences, it is possible to at least set some operational parameters, ie some criteria for organising the very process of where to look for the legislative goals. The first operational parameter has to do with the very nature of the list of documents and material that could be relevant in the search for legislative goals. One must consider whether it is possible to present an exhaustive list of documents and materials that the legal actors should use. For example, an exhaustive list would be the one offered by a legislative act which, in its preamble, explicitly and specifically indicates the goals or documents in which the legislative goals of this very text can be found. Another example is the use of decisions by supreme courts in which the judges specifically ‘reveal’ the legislative goals of a certain act. The alternative to this operational parameter is to consider a group of documents and materials as an open list of sources in which the goals of legislation are traceable. The data indicated in the list would then tend to have the value of a guideline rather than a normative (or quasi-binding) status towards addressees and public agencies. The open list of documents would suggest possible sources where the goals of the legislation could be found and the starting point for goal evaluation. For example, when lacking a clear normative indication (from either the legislators or the court’s judges), it is common to rely on legal scholarship, which very frequently includes the goals of a certain piece of legislation in its analysis. Similarly, decisions taken by lower courts can be used to sketch an open list of legislative goals.

106 

Mauro Zamboni

Somehow overlapping this typology of how to find legislative goals is the second possible classification of documents and materials used to define these very objectives. In this case, the criterion for building up a platform from which to start evaluation of legislative goals is the perspective (or, rather, the analytical tools) applied by the evaluator for the task. In particular, more traditional analytical instruments may be used by legal actors or legal scholars, ie instruments included in the history of Western legal thinking as primary tools for discovering the will of legislative bodies. Such is the case, for instance, in the investigation and evaluation of legislative goals based on the ‘discovery’ of the purposes in the ‘goal norm’ (in German, Zweckartikel) in the preamble of the statute or, in certain legal traditions, in the preparatory works. In addition, it should not be forgotten that legal texts themselves present a primary source for identifying goals. A penal law may not explicitly state the goal of protecting human life when stipulating sentences for murder; nevertheless, this goal can easily be brought to the surface by a legal investigation of the various legislative acts prohibiting criminal behaviours. This type of traditional analytical inquiry will tend to use all the investigatory apparatus of paradigms and methodological principles that are typical of disciplines legitimised as being ‘legal’ by their nature. For example, the traditional legal black-letter approach (in German, Rechtsdogmatik), legal theory and legal philosophy can be counted among the primary perspectives used to discover the goal of legislation in traditional, legally relevant documents. However, in evaluating legislative goals, a broader perspective is also available to both legal scholars and legal practitioners: namely, a perspective that allows legal actors to make use of non-traditional analytical instruments. These include investigation and evaluation of legislative goals by analysing the various political actors’ debate in the media preceding the promulgation of a certain act. Directly connected to this broader choice of material and documents is the possibility of using methodological tools and paradigms provided by non-legal disciplines, such as political science and sociology. For instance, a possible nontraditional instrument for discovering and/or establishing the goal of a statutory measure can be qualitative interviews with the members of the drafting committee. It is clear that opting for this second, non-traditional, route will raise the question of the extent to which the findings can have a normative value—and choosing traditional legal instruments will present a larger bone of contention in this regard. For instance, many might point out how the best (and only) source of information about the intention of a ­legislative body is what this body actually enacted; therefore, many may question whether the results of an investigation based on qualitative interviews can be used as direct supporting argumentation by judges when interpreting statutory provisions that are consistent with the goals ‘discovered’ through these interviews. This questioning of ‘unorthodox’ (at least from a strictly legal perspective) instruments used to discover legislative goals is more than legitimate (in particular, from a legal perspective). However, the non-traditional analytical instruments can be useful, if not vital, in cases of opacity in traditional legal documents (owing, for instance, to attempts at finding a compromise in the bill that satisfies highly polarised parties in the drafting committee) or in cases of ambiguity in the legislative product (resulting, for instance, from a radical change of the text promulgated from the one originally drafted).

Goals and Measures of Legislation

 107

6.6. Conclusion To conclude, it is worth remembering that the purpose of this chapter has not been to clarify how one should evaluate legislative goals. The task has been much humbler, and has consisted of offering to the reader some analytical clarifications on the starting point for an evaluative enterprise. In other words, the main purpose has been not so much to answer the question ‘How does an evaluation of the legislative goals work?’ as to clarify the issues to be considered before tackling such an enquiry. In order to do so, several ideal-typical clarifications have been offered, ie typologies sketched merely as suggestions for solving the problem, regardless of the paths chosen, and central preliminary steps to be taken by the analyst before delving into the investigation and evaluation of the goals in legislation. In particular, typologies have been offered to answer the questions ‘What is a legislative goal?’, ‘What is a function of the legislative measures?’ (and how is the function related to their goals?) and finally ‘Where are the legislative goals located?’ (and how can they be discovered and/or reconstructed?). These three fundamental aspects to be considered before evaluating legislative goals are not exhaustive. In addition to these main features, further studies of goals in legislation can and must consider issues which are unique to a certain piece of legislation and its goals— such as the differences and similarities between the goals when formulated in constitutional documents (and their judicial interpretation in the highest courts) and those which are instead traceable in the secondary legislation. Another example of a specific legislative issue is the indication of, and in some cases the reasoning and argumentation supporting, the existence of inherent goals of every legislation, such as transparency, equality, legal certainty or the enhancement of fundamental rights. However, regardless of the specific path taken by each enterprise aimed at the evaluation of legislative goals, it is still necessary find a common analytical platform upon (and from) which each evaluator must start—namely, a common definition of a legislative goal, the functions it has in the legislative process and where one can find it. This chapter does not directly answer these questions, but it is hoped that it indicates where the answers can be found.

Further Reading Editorial, ‘Regulatory Creep’ (2005) 26(1) Statute Law Review iii. Gunningham, N, ‘Two Cheers for Prescription? Lessons for the Red Tape Reduction Agenda’ (2015) 38(3) University of New South Wales Law Journal 936. Maley, Y, ‘The Language of Legislation’ (1987) 16(1) Language in Society 25. Mousmouti, M and Crispi, G, ‘“Good” Legislation as a Means of Ensuring Voice, Accountability, and the Delivery of Results in Urban Development’ (2015) 6 The World Bank Legal Review 257. Mousmouti, M, ‘Operationalising Quality of Legislation through the Effectiveness Test’ (2012) 6(2) Legisprudence 191.

108 

Mauro Zamboni

Mousmouti, M, ‘Effectiveness as an Aspect of Quality of EU Legislation: Is it Feasible?’ (2014) 2(3) The Theory and Practice of Legislation 309. Reid, E, ‘Montgomery v Lanarkshire Health Board and the Rights of the Reasonable Patient’ (2015) 19(3) Edinburgh Law Review 360. Westerman, P, ‘Arguing about Goals: The Diminishing Scope of Legal Reasoning’ (2010) 24 Argumentation 211. Westerman, P, ‘Governing by Goals: Governance as a Legal Style’ (2007) 1(1) Legisprudence 51. Westerman, P, ‘Legal or Non-legal Reasoning: The Problems of Arguing about Goals’ in ET Feteris, H Kloosterhuis and HJ Plug (eds), Argumentation and the Application of Legal Rules (Amsterdam, Rozenberg Publishers, 2009) 85–103. Westerman, P, ‘Breaking the Circle: Goal-Legislation and the Need for Empirical Research’ (2013) 1(3) The Theory and Practice of Legislation 395. Xanthaki, H, ‘European Union Legislative Quality after the Lisbon Treaty: The Challenges of Smart Regulation’ (2013) 35(1) Statute Law Review 66.

7 Legislative Drafting Techniques/ Formal Legistics JEAN-PIERRE DUPRAT AND HELEN XANTHAKI

This chapter attempts to address one of the most intricate and traditionally jurisdiction-specific topics of legislation: drafting techniques. The term is used interchangeably with formal ­legistics. They are used, and are believed to be, synonymous. Within the common goal of legislative approachability as an expression of the rule of law agenda, the supranational picture of legislative drafting techniques in Europe is first set out, starting from the discovery of unexpected commonality of the theoretical framework behind the choice of technique within Europe. Next, an attempt is made to identify the historical roots of any national specificities. A selection of these specificities are identified and national approaches compared. A glimpse into the common E ­ uropean future in legislative drafting is offered by referring to the main relevant advances offered both at the EU level (such as the Smart Regulation Agenda) and at the national level (such as Good Law). France and the UK are used as representative case studies of the civil and common law drafting traditions in Europe, along with references to other jurisdictions.

7.1.  Structure and Systematics as Elements of Quality of Law in Context It is an indication of the jurisdictional specificity of drafting techniques that even the term used across Europe differs. This chapter uses legislative drafting and formal legistics (a transliteration from the French legistique formelle) interchangeably in order to denote the jurisdictional abstraction of its normative approach. Thus, formal legistics, known in English as legislative drafting techniques, determine the extent of accessibility of legislation as communication, thus making it clearer for users.1 These techniques can resolve difficulties arising from the expression of technical (legal) concepts. Despite the increased interest in material legistics, and the accessibility and intelligibility of legislation, the formal aspect appears prominent in the simplification-of-law movement calling for standardisation of the structure of statutes and codes, affording the possibility of coherence in updating the norms in force. Material legistics refers to the whole life cycle of legislation, focusing mainly

1  CA Morand, ‘Eléments de légistique formelle et matérielle’ in CA Morand (ed), Legistique formelle et materielle (Formal and Material Legistic) (Aix-en Provence, Presses universitaires d’Aix-Marseille, 1999) 19.

110 

Jean-Pierre Duprat and Helen Xanthaki

on the substantive issues of policy, politics and constitutionality. Legisprudence covers both formal and materials legistics in that it encompasses all aspects of legislation as a process and as a product.

7.1.1.  The Hold on Formal Legistics and Legal Certainty 7.1.1.1.  A Main Interest in Economic and Social Effects of Regulation The main points of official reports, issued, for example, by the OECD or the EU, reflect an increased interest in material legistics, and particularly Regulatory Impact Assessment (RIA), reflected by institutional reforms for the reduction of the volume of regulations.2 Not only governments, but parliaments too, have created specific bodies vested with this competence to review bills against pre-legislative scrutiny of their regulatory necessity. Traditionally it was governments that were first involved in the scrutiny of ministerial departments, as in Germany with the National Regulatory Control Council at the Chancellery.3 But several parliaments tried to protect their autonomy with committees in charge of RIA, for example at the National Assembly and the Senate in France,4 after the experience of pre-legislative scrutiny in Great Britain. An interesting question is the linkage between RIA and the formal aspects of legistics in the implementation of legislative regulation policy, as described in a guideline issued by the Italian Prime Minister on 10 September 2008.5

7.1.1.2.  Accessibility to Law and Democratic Prerequisite Formal legistics aims to make legislation more accessible to the public, while at the same time facilitating an interpretation by administrations and courts that does not betray the intent of the legislation. The involvement of technical bodies, including the Department of Justice, or the review by the Conseil d’État in the name of expert knowledge, can only reinforce, but not replace, the legitimacy of democratically elected chambers. If courts ground their interpretation on ‘presumptions’, they need to read the provisions as a whole, including purely formal aspects, such as titles, headings and even marginal notes, at least under the UK tradition.6 This problem of legitimacy explains the debate arising from the French constitutional reform of 2008 related to the legislative process (section 39 of the

2 

For the influence of OECD see F de Francesco, Transnational Policy Innovation (Colchester ECPR Press, 2013) 6. Act on the Establishment of a National Regulatory Control Council, 14 August 2006, amended by s 1 of the Act of 16 March 2011. The standard cost model had been introduced in April 2006. See the report by the Better Regulation Unit at the Federal Chancellery, ‘A Foundation for Better Law’ (April 2012). On the NRCC as an independent watching: OECD, Better Regulation in Europe: Germany (Paris, OECD, 2010) 54. 4  For reforms in France on the government side see OECD, Mieux légiférer en Europe: France (Paris, OECD, 2010) 56, 65. The Comité d’évaluation et de contrôle des politiques publiques was established at the National Assembly by a ‘resolution’ of 27 May 2009, with new sections 146-2 and 146-3 of the Standing Orders; at the Senate, the Commission senatoriale pour le contrôle de l’application des lois results of an ‘arrêté’ of the Board, 16 Novembrer 2011. 5  Camera dei deputati, XVI Legislatura, Documentazione e richerche, No 27, 11 November 2008. 6  I McLeod, Principles of Legislative and Regulatory Drafting (Oxford, Hart Publishing, 2009) 24. 3 

Legislative Drafting Techniques

 111

Constitution and Organic (Institutional) Statute of 15 April 2009) and opening the way for consultation of the Conseil d’État on members’ bills.7 But the main service offered by the Conseil d’État lies in its comprehensive review of legislative texts in the pre-opinion stage: this may extend to a new draft of the government bill or some of its provisions, albeit on a purely advisory basis for the government. Beside this aspect, the reform of 2008–09 implies a second opinion about the quality of the assessment. Of course, when dealing with a parliamentary bill, the review is less detailed, respecting the democratic will of the parliament. Nevertheless, to refer to an independent body outside the circle generating the bill seems a better way to achieve the aim of legal certainty, especially8 as to the formal aspect of legistics, than inner committees, at least for prior assessment: after all, in the common law tradition, external verification of the draft is supported by Thornton’s fifth stage of legislative drafting.

7.1.1.3.  Formal Legistics and the ‘Legislative Cycle’ Within the framework of better regulation programmes, international and European institutions developed the concept of the ‘legislative cycle’ in relation to the economic and social effects of a regulation, but with a linkage between formal and material legistics. This holistic methodological approach to legistics strengthens the implementation of provisions.9 At the same time, however, a political choice has to be made between several available technical options, such as isolated statute, codification and consolidation, or coordination.10

7.1.2. Historical Point of View: The Various Dimensions Afforded to Formal Legistics Formal legistics is defined by reference to the quality of statutes, and specifically their effectiveness and simplicity as a means of facilitating effective implementation and accessibility. Several aspects are noticeable in the works of authors since the eighteenth century with regard to this. First, only general principles determining formal legistics were settled, with a greater priority being placed on style and vocabulary than on technique. On the eve of the French Revolution, the drafting process of royal ordinances was still ruled by the principle of secrecy, even if partly disclosed through the writings of some chancellors. This was a different approach to the rules determining the drafting of judgments in search of homogeneity.

7  Organic Statute No 2009-03, 15 April 2009, OJ 6528; J-P Duprat, ‘The Judicial Review of Ex Ante Impact Assessment in France’ (2012) 6 Legisprudence 379; J-P Duprat, ‘Le Conseil d’Etat et la qualité de la legislation: vers la consécration de nouveaux aspects de la légitimité’ in AFHIP XXIII, Justice et Etat (Aix-en Provence, Presses universitaires d’Aix-Marseille, 2014). 8  As to the two options in balance, technical and independent body versus a more comprehensive one working ‘alongside government’, see the example of the Law Commission: S Wilson, ‘Reforming the Law (Commission): A Crisis of Identity’ (2012) 1 Public Law 20. 9  See contra D Greenberg, Craies on Legislation (London, Sweet & Maxwell, 2012) 387. 10  For a distinction between the several types see C Lambotte, Technique legislative et codification (Brussels, Story-Scientia, 1988) 188ff; Lord Simon of Gaisdale and JVD Webb, ‘Consolidation and Statue Law Revision’ (1975) 4 Public Law 285; V Lasserre-Kiesow, La technique legislative—Etude sur les codes civils français et allemand (Paris, LGDJ, 2002) 57.

112 

Jean-Pierre Duprat and Helen Xanthaki

This last approach generated an interest in philosophical background and sociological requirements.11 Nevertheless, in practice, the lawyers close to the king, mainly the chancellors, shaped the structure of legal texts, defining fixed parts as a tradition. Later, codification, as in the Napoleonic Code, was considered to be an instrument with which to modernise the most important fields in law and to reach a new balance in the civil rights of citizens, together with a rationalised presentation of the enactments to make them more intelligible. But this effort was due to special commissions of lawyers rather than representative bodies and, contrary to Bentham’s goals,12 the movement involved only some areas. Linked to the effects of a surge in political liberalism, a second affirmation paid more attention to the legislative process in chambers in order to enhance the quality of legislation and legal certainty.13 But this approach underlined the importance of parliamentary practice and of provisions to achieve such an aim, and formal legistics was rather subordinate. From the nineteenth century until the middle of the twentieth, clerks of chambers or heads of specialised offices (including the British Parliamentary Counsel) issued treatises or books on ‘parliamentary law’.14 The strong interest in formal legistics as a technique was due to consciousness about a new decline in quality of law after 1945 in European countries, clearly expressed by the title of the book Le déclin du droit.15 L Mader explains clearly the efforts made by academics to define more precisely the contents of formal legistics, for example with P Noll (1973).16 More recently, the criticisms of legislative inflation and poor quality of drafting have led to a stricter listing of the prerequisites for the formal aspects of statutes, including their internal structure and their scrutiny by several institutions.17 But, under the influence of the OECD and the EU in developing legislative scrutiny, the formal aspect is determined by the material one, as the statute is seen merely as an instrument of public policy. The principal policy aim of legislation is of an economic and financial nature: to reduce the societal 11  C de Secondat, Baron de Montesquieu, The Spirit of Laws: A Compendium of the First English Edition, ed DW Carrithers (Berkeley, University of California Press, 1977), Book 29, ch 16. See the introduction by V Zapatero to J Bentham, Nomografia (Madrid, Centro de estudios politicos y constitucionales, 2004). 12  J Bentham, The Works of Jeremy Bentham, vol. 1 (Principles of Morals and Legislation, Fragment on Government, Civil Code, Penal Law) [1843] ed J Bowring (Edinburgh, William Tait, 1843), Pannomial Fragments, ch 1; Nomography or the Art of Indicting Laws. 13  M James, C Blamires and C Pease-Watkin (eds), The Collected Works of Jeremy Bentham. Political Tactics (Oxford, Clarendon Press, 1999).; JP Duprat, ‘Bentham et la diffusion du droit parlementaire’, AFHIP XXI, L’influence politique et juridique de l’Angleterre en Europe (Aix-en-Provence, PUAM, 2012) 297. 14  P Valette and B Saint-Marsy, Traité de la confection des lois (Paris, Joubert, 1839); E Pierre, Traité de Droit politique, électoral et parlementaire (1893; Paris, Loysel, 1989); C Ilbert, Legislative Methods and Forms (Oxford, Clarendon Press, 1901) ch XI (devoted to formal aspect)s; also see H Thring, Practical Legislation; or, The Composition and Language of Acts of Parliament (London, HMSO, 1878). 15  G Ripert, Le déclin du droit: études sur la législation contemporaine (Paris, Librairie générale de droit et de jurisprudence, 1949). 16 See P Noll, Gesetzgebungslehre (Reinbek, Rowohlt, 1973). For the legislative technique see L Mader, L’évaluation législative (Lausanne, Payot, 1985) 15. 17  A Fluckiger and C Guy-Ecabert (eds), Guider les parlements et les gouvernements pour mieux légiférer (Geneva, Schultess, 2008) 16. In his famous report, Lord Renton made clear the criticism about illogical arrangements of statutes: The Preparation of Legislation (Cmnd 6053, 1975) 30–31. After the Conseil d’Etat Annual Report 2006, a working group looked at actions to follow and analysed the causes of the phenomenon with an insistence on the numerous amendments. This resulted in the reform of legislative procedure of 2008–2009 (available on the website of the Conseil). See www.oecd.org/gov/regulartory-policy; www.ec.europa.eu; ‘Proposal for an Interinstitutional Agreement on Better Regulation’ COM (2015) 216 final.

Legislative Drafting Techniques

 113

cost of legislation.18 Such a concept cannot serve as a basis for the significance of formal legistics in formal legistics without consideration of the democratic requirements and the position of each state institution. Finally, the internal structure of statutes has to set out the law in force as its main communication with the citizens.

7.1.3.  Systematisation and Inner Structure of Statutes With the use of supra-jurisdictional solutions to simplify laws, a preliminary consideration of instruments such as codification, consolidation and repeal statutes is relevant to the pursuit of proper structure. But the common tool for good structure is the use of a set inner structure for statutes with a common intention, as a means of making them approachable and user friendly.

7.1.3.1.  Enhancement of Codification and Repealing Statutes to Simplify Law The first wave of modern codification took place in the nineteenth century, especially in France and Germany, with their respective Civil Codes.19 In spite of the theoretical reflections of Cambacerès20 and Portalis,21 who were more interested in practice, a true discussion of technique was due to the adoption of the German Civil Code of 1896 and the Swiss project derived from E Huber’s works.22 F Geny published comments on Huber’s contribution to the modernisation of codification, but more as a matter of vocabulary (definitions) and style than of technique.23 The new wave of codification that emerged in France in 1989 gave rise to about 60 codes that are currently in force, but at the same time they had new limits.24 The rather narrow specialisation of many codes shows a change from the original solution of a systematic technique that applied to a wide range of topics for the same code.25 Nevertheless, the authors followed the thinking that a logical organisation of legal elements pertaining to the same area preserves the ideal of codification.26

18  A reference checklist for regulatory decision making was defined by the OECD in 1995, see de Francisco (n 2) 7. 19  The Civil Code of the French (Paris, Imprimerie de la République, An XII-1804) was the culmination of a series of proposals prepared since 1793, three by Cambaceres (in 1793, 1794 and 1796), and the last one by a Committee with Portalis (An IX). See JL Halperin, Regards de l’historien, Le Code Civil, 1804–2004 (Paris, Dalloz-Litec, 2004) 50. 20 L Chatel de Brancion, Cambacérès: maître d’œuvre de Napoléon (Paris, Perrin, 2001) 642; J-P Delbert, ­Cambacérès: unificateur de la franc-maçonnerie sous le Premier Empire, préface de Jean Tulard (Lille, Athos, 2005) 120. 21 JEM Portalis, ‘Discours préliminaire sur le Code Civil’, Traités et discours juridiques et politiques (Aixen-Provence, Presses Universitaires d’Aix-Marseille, 1988) 26. 22  Schutzmantelmadonna (bronze sculpture), designed in 1960, execution posthumously in 2000: LernetHolenia Park, Vienna Dornbach. 23  F Geny, ‘La technique législative dans la codification civile moderne’, Le Code Civil—Livre du Centenaire (reprint, Paris, Dalloz, 2004) 1025. For the debate about the ‘scientific’ dimension of codification see LasserreKiesow (n 10) 53. 24  M Van de Kerchove and F Ost, Le système juridique entre ordre et désordre (Paris, PUF, 1988) 111. 25  For the evolution and difficulties met in French codification see C Bergeal, Rédiger un texte normatif— Manuel de légistique (Paris, Berger-Levrault, 2012) 31. 26  A Zaradny, ‘Codification et Etat de droit’, Thesis Paris II, 30 March 2011, 326.

114 

Jean-Pierre Duprat and Helen Xanthaki

Numerous codes in France are divided into two parts, one legislative and the other regulatory, but the continuous numbering of sections in each of them and the parallelism of it safeguards the idea of a system. Each division includes: Parts, Books, Titles, Chapters and lastly sections, although a ‘three level division’ is also possible.27 The entire process is under the control of the Commission supérieure de codification, which is expected to issue a new guide soon, modernising a text that was issued in 1996 as a circular. In some aspects, the method of ‘codification à droit constant’ integrates the elements necessary to consolidate legislation. As the policy of simplification tends to imply a reduction in the provisions in force and an assessment of their utility and effectiveness, the UK has introduced the possibility of repealing or consolidating an act. Consolidation in the UK carries an inherent debate on the extent of possible amendment allowed to former provisions.28 A recent example is provided by the most extensive instrument of this kind, the Statute Law (Repeals) Act, which repeals 817 acts in their entirety and parts of a further 50 acts. The relevant report of the Law Commission of England and Wales and the Scottish Law Commission describes and explains the structure of the bill, and then of the act.29 The formal provisions of an act are mentioned, albeit very briefly: section 1 refers to repeals (as mentioned in the Schedules), section 2 defines the extent, and section 3 sets the short title and the commencement. Thus, the most extensive elements pertain to the Schedules: one for repeals and revocations, the other for savings. A very long explanatory note explains the scope of every item repealed.30 Special procedures are dedicated to accelerate the process of codification, consolidation and repeal or coordination, with a strong involvement of executive or specialised bodies. In France, many codes have been adopted through ordinances, under authorisation by Parliament, and ratification is necessary by a legislative provision. In Belgium, a specific statute with a general scope provides the executive with this jurisdiction.31 However, according to one expert, G Braibant, codification is just the first step on the way to a better legislation.32

7.1.3.2.  Formal Systematisation of Statutes The Swiss Guide of Legislation sets out an interesting presentation of the topic, with the distinction between internal and external structure. Only the external structure is directly

27  Precisions in the 21st Report of the Commission supérieure de codification, 2010 (Paris, Journaux Officiels, 2011) 10; D Labetoulle, ‘Quel avenir pour les codes à la française?’ ibid, Schedule 18, 58. For the formal criteria from the Belgian point of view see C Lambotte, Technique legislative et codification (Brussels, Story-Scientia, 1988) 186. 28  P Matthews, ‘Consolidation, Coroners and Democracy’ (1989) 86(21) Law Society’s Gazette 28. 29  The Law Commission and the Scottish Law Commission (1989) Statute Law repeal, nineteenth report, Cm 8330—SG/2012/39. 30  Oral evidence of J Saunders, HL, Consolidation Bills Joint Committee, 21 November 2012. 31  For criticisms of a too much intensive French practice see P Delvolvé, ‘L’été des ordonnances’, Revue française de Droit administratif (Paris, Dalloz, 2005) 909. For a comparative point of view see JP Duprat, ‘Les ordonnances de l’article 38 et la législation secondaire en Grande-Bretagne’, Mélanges JF Lachaume. Le droit administratif: permanences et convergences (Paris, Dalloz, 2007) 467. About the precisions to be fixed by Parliament see National Assembly, second sitting, 1 October 2013, Debates OJ 9160. For Belgium see M Uyttendale, Précis de Droit constitutionnel belge (Brussels, Bruylant, 2001), No 20, 288. 32  G Braibant, ‘Utilité et difficultés de la codification’ (1996) 24 Droits 72. Braibant underlines the linkage between the decree of 12 September 1989 (Creating of the Upper Commission) and the criticisms of the Annual Report 1991 of the Conseil d’Etat.

Legislative Drafting Techniques

 115

related to formal legistics. Several principles are involved with the division of the text: ­intelligibility and accessibility on one hand, and efficacy on the other.33 But efficacy and effectiveness relate directly to the issue under regulation. This can explain the unusual Greenberg format: in the UK, the traditional structure of statutes was ignored in the Health and Social Care Act 2012, which places substantive provisions immediately after the long title and the enacting clause (‘Be it enacted …’).34 The main idea was to facilitate immediate access to the provisions of interest in response to specific questions from the public, with preliminary provisions drawn down at the end of the act, with just a short title, commencement and extent. However, the ambition to facilitate users must be balanced with the equally significant necessity to adapt to the context of the issue under regulation. For example, definitions may still be placed in preliminary provisions if they determine the meaning of substantive provisions. Even a purpose clause that guides statutory interpretation may assist users in making sense of special enactments. It is therefore difficult to deprive every statute of the traditional provisional clauses opening the acts. But the 2015 ‘Drafting Guide from the Office of Parliamentary Counsel’,35 whilst reiterating the tolerance and, indeed, encouragement of individuality in the drafters’ battle with structure, confirms the UK trend of depriving the start of the text from any preliminary provisions that do not relate to the main regulatory message: users—actually mainly non-lawyers—are at the forefront of their consideration, and therefore definitions and purpose seem to have now been transferred to Schedules, perhaps with a hanging clause at the start of the text referring the user to them. As for substantive provisions, guidelines usually maintain a logical approach from general to special ones,36 from substantive to administrative ones, and from primary to supplementary, according to Thring’s classification and doctrine. When principally aimed at modifying existing provisions, the enactments of a statute have to follow the order of the legislation that they amend. Otherwise, they are introduced in the order that best serves the logical sequence of issues arising from the underlying policy.37 Amongst the several national guidelines, the Belgian one is particularly precise about formal aspects, and has useful tables. In Germany, section 2 of Chapter 6 of the Joint Rules of Procedure of the Federal Ministries, dealing with the structure of federal government bills, mentions the two guidelines used: one issued by the Ministry of Interior (‘Guidelines for Drafting Legal Provisions and Administrative Regulations’), the other by the Ministry of Justice (‘Manual of Legal Drafting’). In addition, the Ministry of Finance is involved in budgetary impacts with its ‘General Instructions’.38 As exemplified by the Polish Sejm,

33 OFJ, Guide

de législation, Guide pour l’élaboration de la législation fédérale (Bern, OFJ, 2007) 351. Xanthaki, Thornton’s Legislative Drafting (London, Bloomsbury, 2013) 224. About the idea that there is ‘no right structure’ see G Bowman, ‘The Art of Legislative Drafting’ (2006) Amicus Curiae; for the formula see Greenberg (n 7) 387. 35  Office of Parliamentary Counsel, ‘Drafting Guide’ (August 2015), available at gov.uk/government/uploads/ system/uploads/attachment_data/file/454628/guidancebook_August_2015.pdf. 36  JC Bécane, M Couderc and JL Hérin, La loi (Paris, Dalloz, 2010) 184. With the experience of the French authors as clerks of the chambers, they reflect more a general conception shared by Continental countries. Sometimes, other schemes of a chronological or functional nature are better. 37  Conseil d’Etat, SGG, Guide pour l’élaboration des textes législatifs et règlementaires (Paris, Documentation française, 2007) 180. 38  On the recent aspects of the standards see Federal Government Report 2011 to Bundestag, ‘A Foundation for Better Law’ (Berlin, Federal Chancellery, Better Regulation Unit, April 2012) 11. 34  H

116 

Jean-Pierre Duprat and Helen Xanthaki

the standing orders of a chamber require special documents joined to the Bill for I­ nformation of Members (eg section 34). Drafting practices are improving (see, eg, the Good Law initiative in the UK) and the importance of easy access to legislation is widely recognised. However, this does not negate the importance of interpretation,39 especially where there exists a special obligation for judges to further a particular interpretation, even where a legal text is silent on a point (see, eg, section 4 of the French Civil Code). It must also be noted, however, that in a number of jurisdictions not all provisions of a statute have the same legal force, particularly the title.

7.2.  The Standardised Elements of Legal Texts: Between Logic and Historical Legacy 7.2.1.  The Title of Statutes For any legal text, the title is an element of its identity and provides the public with the first necessary information about the subject matter and the nature of the text, and what type of rule it is (statute, decree, simple regulation, etc).40 Thus, in the UK, the short title is viewed as a tool indicating the main issue under regulation whilst distinguishing the act from similar existing acts: emphasis is placed on the indexing of the statute book and facilitation for the lay user, which leads to the traditional use of generic titles, with parenthetical matters of detail and relevance to existing acts, and the year of passing. But the title is not vested with the same effect in all jurisdictions; in Belgium and France, for example, it does not have a normative effect.41 These national discrepancies are justified on the basis of legal culture, but also on the basis of the parliamentary practice: in jurisdictions where titles have no normative effect, the title is not debated in Parliament and it is also non-amendable.

7.2.1.1.  The Practice of a Short Title: Information and Citation The short title of legislation serves a dual purpose: it specifies its unique identity as a legislative text within the legal system, whilst at the same time it distinguishes it from other similar legislative texts. In the UK, the short title was introduced at the beginning of the nineteenth century as being easier for citation in the context of the movement of legal reform. For example, in the past, one would refer to ‘An Act passed in the Sixth year of the reign of His Majesty King George the Fourth, entitled An Act to render valid marriages solemnised in certain churches and public chapels in which Banns have not usually been published 6 39  The subject of interpretation is particularly well developed in W Dale, Legislative Drafting: A New Approach (London, Butterworths, 1977) 292. 40  Belgian Conseil d’Etat, Principes de technique législative (Brussels, Belgian Conseil d’Etat, 2008) 35–36. 41  In Belgium and France, the title is deprived of any normative effect. For France, it is the result of an old solution of the Cour de Cassation (1811) confirmed by a decision (Ch Civ) 20 April 1920 (Dalloz Rec Périodique, 1923—I—87), as the title is not debated. In a litigation context, the Conseil d’Etat decides on the same argument, but when delivering an opinion to the government in the normative process it reviews the full text and can refer to the adequacy of provisions to the title. Contrary to the expectations of authors, the Conseil d‘Etat kept on with the traditional solution for the title of a decree that is ‘deprived’ of normative value (CE-7 October 2015, No 386436).

Legislative Drafting Techniques

 117

G 4, c. 92’. Now, under the Short Title Act 1896, the Act is simply the Marriages ­Confirmation Act 1825. The practice had become usual practice by the middle of the nineteenth century, before being formalised by the Short Title Act 1896.42 Traditionally, short titles were placed at the beginning of the text but recently in the UK, as opposed to many other common law countries,43 they have been placed in final provisions. Short titles carry the name of the act [by which it is to be known] plus the date of passing: for example, ‘This Act may be cited as the Health and Social Care Act 2012’. The short title may carry parenthetical information on the specific aspect of the statute at hand, for example the fictional Evidence (Documents) Act 2025; an indication of whether this is an Amending Act and which act it amends, for example the fictional Evidence (Document) Amendment Act 2025; and an indication of its geographic extent, for example the fictional Evidence (Document) (Amendment) ­(Scotland) Act 2025. The intent is clearly easy identification, citation and distinction from existing legislation. The drafting of short titles in the UK seems to be at odds with the ‘European Joint Practical Guide’, which states that such a short title ‘is less useful in Community law where acts are identified by a combination of numbers and letters’ (p 26).44 One has to wonder, however, if a random combination of numbers and letters offers lay users an equal insight into the topic of the legislation: Act 135 of 2015 means little to a lay user, whereas the Evidence (Document) Act communicates a clear message. What is important to note here is that currently (and opposite to what Lord Moulton said in 1913) short titles are ‘part of the act’ and fit for interpretation.45 On the Continent, Belgian and Swiss guides allow a concise short title for practical reasons when the normal title is too long.46 France first knew the practice of a long title chosen by the Department of Justice at the beginning of the Revolution, but a political conflict of importance was generated by an inappropriate long title about enactments dealing with the civic oath taken by ecclesiastics (27 November 1790) so that another statute (5–19 January 1791) made it an obligation to exclusively use a short title, with a concise indication of the subject of the text.47

7.2.1.2.  The Short Title as the Dominant Continental Practice In spite of limited exceptions due to pragmatic considerations related to the features of a statute, most countries use a short title. In France, the title has to describe in an objective and concise manner the subject of the text. In addition to its nature and date, a particular number appears just after the word ‘loi’, comprising the year and the number of the statute as delivered by the Secrétariat Général du gouvernement (Cabinet Office).48 The date is the 42  With this act a short title was given to ‘all the public general Acts passed since the date of the union with S­ cotland, which had not been previously given short titles’: C Ilbert, Legislative Methods and Forms (Oxford, Clarendon Press, 1901) 272. 43  Xanthaki (n 34) 244. 44  ‘Joint Practical Guide of the European Parliament, the Council and the Commission for persons involved in the drafting of European Union Legislation’ (Publication Office, December 2014) 26, available at http://eur-lex. europa.eu/content/techleg/KB0213228ENN.pdf. 45  McLeod (n 6) 23. 46  OFJ (n 33) No 8241, 357, n 32. 47  Décret relatif au titre des lois: ‘le titre qui sera mis en tête de chaque loi en indiquera simplement l’objet’, Duvergier, Collection des lois, t.1, Paris 1824, 142; for the debates at the National Assembly see Archives parlementaires, t.22 (Paris, 1855) 14, 21; P Valette and B Saint-Marsy, Traité de la confection des lois (Paris, 1839) ch IX, 216. 48  Bergeal (n 25) 286; Bécane et al (n 36) 177. This numbering depends on the order of issue.

118 

Jean-Pierre Duprat and Helen Xanthaki

same one as the decree for promulgation issued by the President of the Republic. Although the ‘Guide in Legistics’49 recommends avoiding any subjective expression in the title, the practice reveals that several statutes do not respect that point.50 But the title does not have any legal effect and is not referred to by judges for interpretation. The same applies in Belgium. In contrast, as an element of the text itself, in Spain the title has three functions: identification, interpretation and citation.51 For interpretation, the French courts use the ‘travaux préparatoires’, a compound of exposition of motives of bills, reports and debates in chambers.

7.2.2.  The Decreased Usage of the Preamble and Its Modern Substitutes The preamble is defined by the ‘Practical Guide of Community Institutions’52 in a rather simple and vague manner: ‘“Preamble” means everything between the title and the enacting terms of the act’. In the frame of a representative regime, it was used to deliver a rather subjective explanation of the context and aim of the text.53 As the practice of using preambles was increasing, F Bacon criticised it as a means ‘to please the people’.54 (During the French Revolution, it had been common practice.) The UK has abandoned the practice of using preambles in acts, but retains them in instruments with a ceremonial character.55 In statutory instruments, very brief preambles confirm the enabling power and clause of the original act that allows the intra vires delegation.56 Spain is in the original position of reproducing the exposition of motives of bills in the statutes. In this country, two possible headings are used for statutes: ‘Préambulo’ or ‘Exposicion de motivos’. Normally, the second one is attached to the bill for information when presented to the chambers, as noted by the Constitutional court, in addition to other documents. The French practice is sometimes described as being similar, but an important difference lies in the total absence of such an element in statutes; the ‘Exposé des motifs’ is attached to bills, but is referred to for interpretation when necessary to determine the government’s intent, under the general terms of ‘travaux préparatoires’.57 The recent organic (or institutional) statute of 15 April 49 www.legifrance.gouv.fr/Droit-francais/Guide-de-legistique.

50  As an example of the title: Loi No 2014-58 du 27 Janvier 2014 de modernisation de l’action publique territoriale et d’affirmation des métropoles. Two words (modernisation and affirmation) do not fulfil the condition of objective writing. 51  ‘Resolution’ of 28 July 2005 to publish the Agreement of the Cabinet of 22 July for approbation of the Directrices de tecnica normativa (BOE, No 180, 29 July 2005, 26880) related to the preparation of bills; the same effects result for statutes. 52  ‘Joint practical guide of the European Parliament, the Council and the Commission for persons involved in the drafting of legislation within the Community institutions’, KA-45-02-094-EN-C, 2003. 53  A Viandier, Recherche de légistique comparée (Berlin, Springer Verlag, 1988) 83. 54  Dignitate et augmentatis scientarum, Book 8, ch 3, part 2, with a French translation of an extract: Archives de philosophie du droit 20 (Paris, Sirey, 1985), 357–58. 55  ‘Whereas the Commons of the United Kingdom in Parliament assembled have resolved to authorize the use of resources and the issue of sums out of the Consolidated Fund towards making good which they have granted to her Majesty in this session of Parliament’. See Consolidated Fund Act 2012. 56  Greenberg (n 7) 113. The Belgian Code states that statutes, decrees and ordinances are deprived of a preamble (s 18). The same is true for statutes in Germany, Italy and Sweden, see Viandier (n 53) 83; R Pagano, Introduzione alla legistica (Milan, Giuffre, 2004) 113–14. 57  For a reference to them in a decision of the Conseil d’Etat (30 December 2013, No 347047): ‘il résulte de ces dispositions, éclairées par les travaux préparatoires de la loi’, see B Génevois, ‘Le Conseil d’Etat et l’interprétation de la loi’ (2002) 5 Revue française de Droit administratif 880.

Legislative Drafting Techniques

 119

2009 (section 7) confirms the obligation to deliver a ‘statement of their motives’ ahead of the bill itself. Thus, a preamble exists at this preliminary stage, but not thereafter. For statutes, generally the modern ‘preamble’ is a much reduced one and presents some elements of an enacting clause, with mention of the authorities involved in the adoption of the statute. The Swiss Guide of Legislation introduces an additional reference to constitutional or legal grounds.58 Since 2012, French decrees have been prefaced with a notice clarifying the scope of the text to the public.59 The function of a formal preamble now pertains to different types of documents joined to bills: exposition of motives, explanatory notes, memoranda or documents of RIA. The notes and memoranda are parts of the bill itself. In France, the ‘Exposé des motifs’ and ‘Etude d’impact’ are gathered in the same document with the bill for introduction to the first chamber, and can subsequently be consulted via an electronic application. In the UK, explanatory notes are an authoritative commentary of the statute but do not always refer to each of the provisions separately: they are drafted by the client team, they clarify the law from their perspective, but they do not form part of the text and carry a relevant disclaimer. In Poland, section 34 of the standing orders of the Sejm requires an explanatory statement to be produced with statements about preliminary consultation.

7.2.3. The Search for Clear Presentation of Provision: Numbering and Heading of Divisions and Subdivisions Some types of statutes require only a very simple presentation, when they are just delivering an authorisation, for example to ratify a treaty or an agreement. The same is true when a provision is intended to deliver an approbation of an institution, with an annex presenting extracts of the full text. When a statute is of greater extent, the choice of divisions and subdivisions is important to make the provisions easily seen by the public and to simplify the interpretation of enactments. Besides the use of numbers or letters, the use of typography can also help distinguish divisions and subdivision. Usually the titles given to the several divisions are designed to improve the accessibility to the law. In spite of special requirements for codification, the main statements are linked to ordinary statutes.

7.2.3.1. Continuous Numbering of Sections and Differentiation from Other Divisions of the Statute A continuous numbering system is used for all sections (or articles as named on the Continent), from the first one to the last, with Arabic numbers. However, to implement the European treaty of 2 March 2012, related to public finance balance, the French Finance Acts present at the beginning a general table describing the evolution of this balance for all public administrations (state, social security, local governments) under a section with the title ‘Article liminaire’ (introductory section), before the normal numbering commences.60

58 

OFJ (n 33) No 8242, 358, n 32. Conseil d’Etat, SGG (n 37) 166–67, updated on the legifrance website, www.legifrance.gouv.fr/. 60  Finance Act for 2014, 29 December 2013, OJ 21829. It is the direct implementation of a provision of the organic statute of 17 December 2012 (s 7). 59 

120 

Jean-Pierre Duprat and Helen Xanthaki

The use of higher levels of division depends on the length of the statute: if it is quite short, sections are sufficient, but longer ones must identify clearly the several matters covered. For a very long statute, such as the above-mentioned Health and Social Care Act (319 sections) in Great Britain, there are 12 parts. Further subdivisions are possible, for example into Chapters, and also cross-headings. The length of the statute and the different types of subject matter will determine what sorts of division are used. The goal is clear organisation, allowing the users to find their way more easily around the statute. The divisions are numbered in a continuous manner, but differences of typography can afford assistance in clarifying the scheme. In several continental countries (eg France), several levels of division are used: Titles, Chapters and ‘sections’, before ‘articles’ (known as ‘sections’ in the common law tradition). Each of them can contain several or just one subdivision. Titles and Chapters are numbered with Roman figures, whereas ‘sections’ and ‘articles’ use Arabic ones.61 In addition, the headings of ‘sections’ and Chapters are in bold type. The most debated topic is how to label subdivisions, This is often by a mixture of numbers and letters, sometimes using brackets. Several guidelines warn against including too many different ideas in the same section, recommending one concept per clause in a subdivision. More precisely, the Swiss Guide of Legislation gives a limit of no more than three ‘alinéas’ per section. In France, Roman numerals are used, and ‘alinéas’ can be grouped under them (I, II, …) in the same section. Since 2000, statutes and decrees have used the same subdivisions, but as C Bergeal has noted, the use of Roman numerals at this level is an indication of poor drafting for too long a section.62 Due to the parallelism between the two main parts of codes in France, legislative and regulatory, perfect correspondence for the higher divisions is required (Parts, Books, Titles and Chapters), though there is more freedom for sections, so long as they are numbered continuously within the Chapter.63

7.2.3.2.  Headings, Understandability for Users and Consistency of Drafting Headings are vested with a multifunctional dimension: self-guidance for drafters, help for users and sometimes for statutory interpretation. Some countries use headings for each division of a statute, with Roman capitals used for the higher levels (Parts generally) and italics used for others: for example, for the group of sections (known in the UK as crossheadings) under chapters in the UK. However, the value of headings in statutory interpretation is under debate. They can play a part only when ambiguities need more elements to resolve the difficulty.64 In Spain, the predominant practice is to use headings at each level for ordinary statutes.65 The Swiss Guide of Legislation underlines two requirements for headings: to be precise and to be concise.66 In Germany, headings are used generally for upper divisions and scarcely for sections, but when they are so used, each section must have 61 

As an example, Statute No 2014-58, 27 January 2014, OJ 1562. Bergeal (n 25) 299. 63  Conseil d’Etat, SGG (n 37) 73–74. In the section number, the first part refers to the upper divisions of the Code (section L.1232-1) and the last to the section itself. 64  About historical dimension in common law systems see N Horn, ‘Legislative Sections Headings’ (2011) 32(3) Statute Law Review 189. 65  Directrices de tecnica normativa (28/7/2005) No 28. 66  OFJ (n 33) No 8246, 359. 62 

Legislative Drafting Techniques

 121

a heading. In the UK, headings are part of the text and are thus fit for use as tools for statutory interpretation. For some types of legislation, the scheme of the act is determined by legislative provisions of superior normative value, as in France for Finance Acts under the enactments of organic statutes.67 So if Part I of the act is devoted to resources and Part II to appropriations, with an important section at the end of Part I dealing with the balance between the two, which Part allows for debate on appropriations? Again, each Part is divided into titles. Several types of headings are used to allow a logical presentation of sections, expressed in letters or in numbers. Headings usually involve a group of sections and can be of a different typography, depending on their position, to make the identification of provisions easier. Schedules or annexes are important, and in the UK are of the same legal value as provisions within the main body of the text. Sometimes they consist of a report (mainly for statutes of a programming nature), but in France they are not vested with any legal effect, in contrast to Finance Acts, when the annexes (‘Etats annexes’) determine the ceilings for appropriations.

7.2.3.3.  Transitional Provisions and Legal Certainty Constitutes a main element of the rule of law (or of the rechstaat) and requires that the legitimate interests of the public, particularly tax payers or contracting parties, are allotted special protection against normative changes. Transitional provisions are a solution to this problem.68 Often statutes gather all such provisions under a single heading, though not always, as seen in Thornton’s Legislative Drafting.69 Changes consist of new provisions, repeals and savings. Various positions are adopted by drafters as a means of dealing with transitional provisions. The prevalent aim is to maintain compliance with the principle of legal certainty; this is best facilitated when transitional provisions are collected together with or close to commencement provisions.70 Another possibility is to vest a minister with a competence in secondary legislation or to issue a special decree about it. To conclude, there does not seem to be a dominant practice on transitional provisions, though various solutions are offered to deal with the specific problem at hand.

7.3. Cross-references Drafting by reference induces some positive effects, particularly for codification as recommended by the French commission supérieure de codification. The advantages of this method are greater precision, concision and coherence of each particular section of a code. 67  Mainly the organic statute (loi organique) of 1 August 2001 and, for the introductory section, the organic statute of 17 December 2012. 68  JP Duprat, ‘Reform of Legislation in Economics Condition of a Better Balance between Necessary Adaptations to Change and Interest of Operator for Legal Certainty’ in L Mader and S Kabyshev (eds), Regulatory Reforms (Baden-Baden, Nomos, 2014) 228. 69  Xanthaki (n 34) 474. 70  Bergeal (n 25) No 205, 250. The Belgian guide (Principes de technique législative) recommends placing the transitory provisions just after the repealing ones (No 145, 90). In Spain, the transitory provisions come just before the final provisions.

122 

Jean-Pierre Duprat and Helen Xanthaki

Nevertheless, for the public it is often more difficult to read and understand the true meaning of the provision as modified. For this purpose, the Swiss Guide of Legislation underlines a number of prerequisites for the use of cross-references: they must simplify the statute and not affect the intelligibility of the provisions. The UK’s 2014 Guide also supports the measured use of cross-references as a means of balancing the need to avoid referring to instruments that may, in due course, be amended or repealed before an amendment or repeal of the cross-referring text, along with the need to reduce the length of the statute as means of achieving user friendliness, approachability and intelligibility. In general, cross-references are not recommended. References to European provisions are usual in statutes implementing such rules and are under national enactments defining conditions for interpretation.71

7.4. Amendments The definition of amending provisions is clearly introduced by the Guide in Legistics issued by the Belgian Conseil d’Etat.72 Amending provisions are those which intervene in existing acts by: supplementing their provisions; replacing their provisions in total or in part; inserting new provisions; abrogating prospectively their provisions in whole or in part; or retroactively amending their provisions as if they had never existed or as if they had never produced the effect in question. This definition of amending provisions is shared by all jurisdictions. What differs is the drafting techniques used for the purposes of the amendment. One can distinguish between two clusters of jurisdictions: those who opt for textual amendments and those who opt for a repeal and re-enactment of the whole section or subsection under modification. Typically, the UK does not choose between the two: each case is decided on its merits, with the aim of balancing the parliamentary practicality offered by the surgical precision of the textual amendments with the distinct advantage of intelligibility of the repeal and re-enactment scenario. The clear advantage of textual amendment is that control over the debate and drafting of the amending text remains with the drafter, who need only touch the words required, thus avoiding having to make any changes that could respark the debate of irrelevant policy issues. The advantage of repeal and re-enactment lies with the distinct clarity offered to the user, who is not lost amidst instructions to replace former words with new ones, but is offered the luxury of reading the whole section or subsection as it is now meant to read. Most civil law jurisdictions seem inclined to go for the second method: this is the case in France, Luxembourg, Italy, Germany and Switzerland. However, they all seem to revert to textual amendments if the practicalities of the case so require, thus leading to the conclusion that the prevalence of repeal and re-enactment does not necessarily indicate exclusivity of this practice.73 71 

See Greenberg (n 7) 827, for the Interpretation Act 1978. See http://www.raadvst-consetat.be/?lang=fr&page=technique_legislative. 73  For example, at http://www.legilux.public.lu/leg/textescoordonnes/guides/traite_legistique_formelle/traite_ legistique_formelle.pdf the Luxemburg Guidelines refer to both techniques in parallel: 72 

art. 1er. l’article 5 du projet de loi … est amendé comme suit: ‘art. 5. …’ art. 2. a l’article 6 du même projet, les mots ‘…’ sont remplacés par ceux de ‘…’.

Legislative Drafting Techniques

 123

Another discrepancy between civil and common law traditions in Europe concerns implied repeals or amendments, namely the practice where the introduction of a new provision consequently amends or repeals an existing one. For the UK, Ireland, Malta and Cyprus, the concept of an implied repeal is anathema to drafters, and should be avoided at all costs. As a result, consequential amendments and repeals are gathered together in an, in principle, exhaustive list of affected texts. This is not a practice shared in civil law jurisdictions, where statutory interpretation includes an element of tidying up the competing provisions of the statute book.

7.5.  Transposition of European Guidelines Transposition of EU legislation is another issue of relevance here, in the sense that transposition measures are widely accepted to be special provisions requiring separate treatment. What sort of treatment, however, is a matter of continuing debate. In the past, there seemed to be a huge clash between Member States who chose the copy-and-paste technique of transposition, thus simply enacting the EU measure within their national legal order without any or much intervention, and those Member States who opted for the effective creative approach to drafting known as elaboration.74 Until 2012, the UK led the cause for the second group, calling upon the effectiveness of the transposing measures as a cause demanding and deserving creative reception into the legal system of the country in a manner appropriate and relevant to the legal tradition, existing laws and administrative structures of the country. These were values juxtaposed with the practicability of copying and pasting as a means of superficial transposition. However, the ideological debate between elaboration versus copying and pasting for the purposes of transposition has been surpassed by the practicability of delegated versus primary legislation. For example, Ireland took the policy decision to transpose all EU legislation by means of delegated legislation: the EU Accession Act was viewed as an adequate enabling clause that could be used to ease the lengthy delays in the passing of transposition legislation that were overwhelming the Irish Assembly and had led to a very low implementation score for the country. Recently, the UK also gave in to practicability: most EU legislation is now copied and pasted there. It is now accepted that, under the same structure of delegation from Parliament to the government as used by the Irish, EU legislation can be, and in its vast majority is, transposed by means of delegated legislation. This position is shared by Denmark, Germany, Estonia, Spain (albeit to a lesser extent), France, Italy, ­Ireland, Poland, Portugal and Romania. The advantage of this position is an end to the dangers of gold-plating, partial or incomplete transposition, and consequent mismatches in the legal regimes of Member States. The disadvantage is an end to effective in-depth analysis of EU legislation, juxtaposition with national existing law, and a smooth reception by the legal system in a format familiar to the system and its users. 74  B Steunenberg and W Voermans ‘The Transposition of EC Directives: A Comparative Study of Instruments, Techniques and Processes in Six EU Member States’ (Leiden, Leiden University/the Hague, WODC of the Ministry of Justice, 2006).

124 

Jean-Pierre Duprat and Helen Xanthaki

At this point, it is worth noting that the position in Luxembourg seems to be moving to transposition without the necessity of national regulatory or legislative intervention: the supremacy of EU law dictates that EU instruments are part of the national law of each Member State. On that basis, what is the purpose of repeating verbatim the provisions of EU instruments? This radical view, which is not, however, used in practice in the majority of transposition tasks, is not shared by many Member States. One wonders whether this is due to a consideration of the users, whose access to EU legal sources and consequent awareness of EU legislation is limited, or is a result of their preoccupation with the dreaded transposition tables. In France, for example, the Conseil d’Etat wonders whether one can transpose solely by reference. Transposition by reference means that the national transposing measure makes reference to the EU instrument without reproducing its content in any further provision in national law. The French Conseil d’Etat is adamant that this is inexcusable, as it invokes ineffectiveness of transposition. However, it is an acceptable means of transposition when reference to national agencies in the rest of the Member States would make transposition by reference a somewhat ridiculous exercise. Transposition by reference is possible in Cyprus (where it is very rarely used), the Czech Republic, Germany, Latvia, the Netherlands (where it is actually used in practice), Slovenia (mainly used for the transposition of technical annexes) and Spain (exceptionally used with Royal Decree-Law 1/1/1997 of 31 January, in relation to the technical specifications of Directive 95/47/CE). The transposition of EU legislation is indeed a special task. But the treatment reserved for it in the vast majority of Member States is one of technical and superficial placement into the volume of national legislation. Although the trend for quick transposition is served by copying and pasting by means of delegated legislation, it is very doubtful that the cause of effective legislation for the achievement of EU policies can be served equally well.

7.6.  Innovation: A Glimpse Into the Future Formal legistics are not currently at the forefront of the regulation debate. Within the EU the opportunity to address and shed some light on them was missed both in the recent EU Smart Regulation Agenda and in the EU 2020 Agenda. Smart does refer to simplification of EU law, but delimits it to the reduction of administrative burdens. It also calls for an ex ante evaluation of law effectiveness and efficiency, but delimits the concept to fitness checks and general policy evaluations. Smart further calls for an improvement in the implementation record, though the tools for its achievement are restricted to postlegislative scrutiny, SOLVIT and the EU Pilot on clarification and assistance with the application of EU legislation. Smart calls for ‘the best possible’ legislation to be chosen, but the tools proposed are only impact assessments, clearer and accessible legislation, simple language, codification, recasting and e-access. Thus, as in many EU policies, the regulatory framework for EU legislative quality is a puzzle of fragmented principles introduced rather cryptically in texts of political compromise. Consequently, what is still missing from this lengthy grapple with EU legislative quality is a comprehensive set of concrete standards by which EU legislative texts are classified as good or bad laws. Effectiveness is the

Legislative Drafting Techniques

 125

prevalent theory.75 A good law is one that is capable of leading to efficacy of regulation; but how effectiveness is to be achieved is left vague and unanswered. It is therefore important for legislative techniques or formal legistics to come back into the limelight of the agenda at both the EU and national levels. Perhaps some encouragement can be drawn from the current UK government’s innovations in legislative drafting. What has changed? Legislation is now a discipline in its own right. Its prevalent theory, phronetic legislative drafting, links quality of legislation with effective legislation that produces the types, extent and level of regulation required by the government. Against this background there is incredible innovation in the field of legislative drafting, especially in the UK: the Good Law project; TNA initiatives; and the AHRC Project on Big Data in Legislation. We now look at each of these in some detail. The Good Law project76 is a collaboration between the Office of Parliamentary Counsel and the National Archives. Its main objectives are to gain further insight into how different user groups approach and use legislation and their expectations; to understand how different drafting techniques may help or hinder users to answer a comprehension question quickly and accurately; and to investigate how additional information presented on legislation.gov.uk helps users understand what they are reading. Perhaps the most important finding of the project for the purposes of legislative studies and legislative drafting reform is the final identification of who is the audience of legislation in the UK.77 Lawyers are, of course, some of the users of legislation, but they only constitute a small percentage of the 2 million visitors per month who access the 400,000,000 page impressions of www.legislation.gov.uk annually. In fact, users of legislation are represented by three personas (based on evidence and as developed by the Good Law project). ‘Mark Green’ is a non-lawyer who needs to use legislation for work: non-law-trained professionals constitute about 60% of users of legislation. gov.uk. ‘Heather Cole’ is a member of the public seeking to enforce her rights or those of a relative or friend; she uses legislation to answer questions related to those rights, to reinforce her arguments before the authorities, or to understand what the law quoted to her by others really says. And ‘Jane Booker’ is a law-trained persona representing legal practitioners, legal researchers, law librarians, etc. Usability studies were conducted under the auspices of the project. These bring to light impressive empirical data. It is interesting to note that most users expected that legislation will be hard to read—even by experienced barristers. Few participants answered all comprehension questions correctly. Participants did not know what a Schedule is or where to find one. Participants did not understand the date of entry into force: many assumed that all laws on legislation.gov.uk are in force. Several participants did not know what SI stood for: the abbreviation and the numbers were very confusing; some participants also did not know what secondary legislation is. All participants asked why they had to hover over the text to get the date an act was passed, rather than having the date written within the text.

75  H Xanthaki, ‘Quality of Legislation: An Achievable Universal Concept or a Utopian Pursuit?’ in L Mader and M Tavares de Almeida (eds), Quality of Legislation (Baden-Baden, Nomos, 2011) 75–85. 76 https://www.gov.uk/good-law. 77 https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/326937/Loophole_-_20142__2014-05-09_-What_works_best_for_the_reader.pdf.

126 

Jean-Pierre Duprat and Helen Xanthaki

The conclusions of the Good Law project so far are that users prefer different styles and interact with text in different ways: it is unlikely that further testing will provide conclusive results about which style is more effective. Having extra information and links to key sections of a piece of legislation makes navigating and understanding it easier. Less experienced users would benefit from prominent information providing explanations of how legislation is put together, such as what a section is, what happens to legislation after it is enacted (eg how/why provisions come into force) and what secondary legislation is. Moreover, the UK government feels that tackling the complexity of legislation does not stop with the language, structure and architecture of the statute book. A lot has to do with the publication of legislation. The National Archives has departed on a journey of innovation that is leading to the use of IT concepts in the layout of published legislation. Most is reflected in the current version of the publication of an act, which is yet to go live. Its main characteristic lies in the linkages between the text and the surrounding information, and in the many clickable options that allow the reader to understand the text by means of information from explanatory notes, extent, entry into force, geography, and more. Finally, the Big Data in Legislation initiative of the Arts and Humanities Research Council supports the Legislation Data Research Infrastructure project (LDRI). Led by the National Archives, with participation from the Office of Parliamentary Counsel, the Institute of Advanced Legal Studies of the University of London and the University of Cambridge, the project aims to make big data research possible with legislation by bringing together the data, online tools for end users tailored to their needs and open source tools for researchers to download, adapt and use. The LDRI is about conducting research across the Statute Book. The project is still in progress, but the team expects the following datasets to be made available: sets of pre-packaged data analyses accompanied by the methodology used to p ­ roduce those analyses (‘annual census of the Statute Book’); downloadable data sets with tried and tested tools and examples of data analysis methods: researchers can use or adapt tried and tested resources and approaches to their own research questions; and downloadable data sets: the entire Statute Book, and other open data created as part of the service (case law or anonymised usage data for legislation.gov.uk): researchers will be given access to a lot of data, so they can do their own analyses from scratch. Thus, there is a distinct focus on drafting techniques in the UK, albeit one that manifests as an open-minded innovative quest to offer answers by means of methodologically sound empirical analyses on a number of fundamental legistics issues. First, who uses legislation: understanding the audience can lead to an educated approach to how communication between law and user is to take place. And this can lead to further innovation, such as the layered approach to legislation that demands the division of legislative texts into three parts, each speaking to each of the three user groups identified by Good Law.78 Secondly, how can legislation be presented to aid accessibility: after all, accessibility is not solely a matter of language; it is a matter of context—indeed, a context which must be offered to all users of legislation, trained or not. Thirdly, how can further empirical work be invoked with reference not to single legislative instruments but with reference to the whole of the Statute Book. 78 H Xanthaki, Drafting Legislation: Art and Technology of Rules for Regulation (Oxford, Hart Publishing, 2014); H Xanthaki (ed), Enhancing Legislative Drafting in the Commonwealth: A Wealth of Innovation (London, ­Routledge, 2014).

Legislative Drafting Techniques

 127

Ultimately, one can only admit to the misery of defeat when it comes to legislation: nobody has managed to do it perfectly so far. The future lies with innovation that shakes the foundations of legistics. And this can only be a source of excitement.

Further Reading Asprey, M, Plain Language for Lawyers, 4th edn (Sydney, Federation Press, 2010). Becane, M, Couderc, M and Hérin, JL, La loi (Paris, Dalloz, 2010). Bergeal, C, Rédiger un texte normatif, Manuel de légistique (Paris, Berger-Levrault, 2012). Butt, P and Castle, R, Modern Legal Drafting—A Guide to Using Clearer Language, 2nd edn (Cambridge, Cambridge University Press, 2006). Dickerson, FR, Fundamentals of Legal Drafting (Aspen Publishers, 1965). Drago, R (ed), La confection de la loi (Paris, Presses universitaires de France, 2005). Greenberg, D, Craies on Legislation, 9th edn (London, Sweet & Maxwell, 2008). McLeod, I, Principles of Legislative and Regulatory Drafting (Oxford, Hart Publishing, 2009). Office fédéral de la Justice, Guide de légistique (Bern, OFJ, 2007). Pagano, R, Introduzzionne alla legistica (Milan, Giuffre, 2004). Secrétariat général du gouvernement, Conseil d’Etat, Guide pour l’élaboration des textes législatifs et réglementaires (Paris, Documentation française, 2007). Xanthaki, H, Thornton’s Legislative Drafting, 5th edn (London, Bloomsbury Professional, 2013). Xanthaki, H, Drafting Legislation: Art and Technology of Rules for Regulation (Oxford, Hart Publishing, 2014). Xanthaki, H (ed), Enhancing Legislative Drafting in the Commonwealth: A Wealth of Innovation (London, Routledge, 2014).

128 

8 Legislative Language and Style MARIO HERNÁNDEZ RAMOS AND VOLKER HEYDT

Access to the law is an important feature in democratic societies. In this sense, the language used to express the law is a key factor because it is the tool utilised to express the will of the legislator. These are the reasons why plain language should be used as far as possible. Clarity and precision are the predominant principles to be respected when rules are drafted, and definitions are a useful tool discharging a text of the repetition of detailed and lengthy concepts or situations. The respective requirements of legal documentation must be taken into account during the drafting process. In addition, specific problems arise from the coexistence of national and EU legislation and from multilingual legislation.

8.1.  Characteristics of Normative Language When drafting a rule, one should keep in mind that the text is a vehicle for communication. Language is code, whether speech, sign, or writing. That code has a purpose: to transmit a message. The efficacy of that transmission depends on the will and the ability of the transmitter to code the message in such a way as to exploit the ability of the recipient to understand it.1

Any imbalance between the intended meaning of the law and its expression causes uncertainty and thus affects legal certainty. The importance of normative language is paramount because its errors become entrenched and extend by linguistic osmosis to the language of jurisprudence, doctrine and forensic practice. For this reason, it is important for legal rules to be expressed according to specific characteristics.2 According to the various concepts of law, normative language has passed through different stages. Initially it was persuasive, attempting to persuade or dissuade as to behaviour; then it became a language of conviction, as in enlightened despotism legislators liked to

1  D Roebuck, ‘Plain, Clear and Something More? Criteria for Communication in Legal Language’ (2014) 16 European Journal of Law Reform 633, 650. 2  The Spanish Constitutional Court has repeatedly highlighted the relationship between legislative technique and legal certainty, but only if this principle is affected the Spanish Constitutional Court would judge the challenged law before him as a matter of legislative technique, eg STC 150/1990, 4 October 1990, § 8. This view contrasts with the classic position of American courts that have appreciated since long time ago the unconstitutionality of legal provisions for vagueness or uncertainty in the language used. See, eg RW Aigler, ‘Legislation in Vague or General Terms’ (1923) 8 Michigan Law Review 831.

130 

Mario Hernández Ramos and Volker Heydt

express the ratio legis necessary to achieve the obedience of their subjects thanks to conviction itself; subsequently it superseded this didactic style when the objective of the rule was to instruct, inform and train, and laws included examples and comments. Today, normative language avoids a persuasive style (lex iubeat, non suadeat), a didactic style of conviction (lex iubeat, non doceat) or a descriptive style, being above all imperative. Legislation must simply state what the law is, not why it is. However, even though the language and style of drafting may not be standardised, it ought to be reasonably consistent in style. This will depend on the text type (constitution, statutory law, sub-legal provisions such as decrees or administrative regulation, etc), the subject, the recipients, and the social or professional environment at which the laws are aimed.3 Every standardisation or unification in the wording of laws would separate the law from the reality which it has to adapt to, which would perhaps make it less useful. It is difficult to determine what a unified legal language would look like, or how desirable it would be. Despite these reservations, normative language must satisfy a series of basic essential conditions. The main requirements of normative language are clarity, brevity, ­precision, coherence, consistency, simplicity and certainty. All doctrine and institutions specialising in legisprudence agree that normative language must be at least clear, simple and precise. Moreover, clarity and precision in normative language are a necessary proviso for legal certainty.4

8.1.1. Clarity Laws must be drawn up clearly, that is, they must be as simple as possible and easy to understand. They must also be drafted precisely, which means that a good drafting minimises doubt in the reader who has to observe or apply them. At this point, determination of who is going to be the primary or secondary user or recipient (audience) of the rule is essential in order to establish the rule’s style, precision and details. Moreover, laws must be drafted in a coherent manner, that is, the same terms must be used to express the same concepts, and the law must hold no contradictions. In short, laws must be understandable. Semantic clarity requires that standard or ordinary language should be used in the rules whenever possible. The rules should be drafted in such a way as to make them understandable to all citizens. In recent times, common law doctrine has also referred to modern or plain English, and has advocated as a general rule the use of plain language as a reaction against the incomprehensibility and complexity of traditional normative language. So as to ensure clarity and maintain technical correctness, normative language must approach as far as possible that plain language, from which it is not really different but rather a modulation produced by its use in the processes of the creation and application of the law. For this reason, the meaning of the words used in normative texts must in principle retain that of the plain language from which they come, as there is no worse trap for the unwary reader,

3 

U Karpen, ‘Comparative Law: Perspectives of Legislation’ (2012) 6 Legisprudence 149, 182–84. These characteristics are always highlighted by the Spanish Constitutional Court as essential for legal norms. See, eg, STC 96/2002, § 5; STC 235/2000, § 8; STC 104/2000, § 7. 4 

Legislative Language and Style

 131

or even the careful drafter, than the use of a term in a sense that is significantly different from that normally attributed by the target audience.5 Despite these affirmations, it is not possible to renounce either the conceptual heritage or the semantic precision deposited in normative language as it provides semantic univocity, lexical economy and conceptual precision. Precision on the one hand and clarity and simplicity on the other are characteristics that are difficult to combine in the same text as they are related to each other in a dialectical manner. For example, the use of references and that of definitions are techniques that contribute to the precision of the text but at the same time involve a loss of clarity and simplicity.6 Simplicity and clarity may lead to ambiguity and vagueness, which means it is recommended to avoid as far as possible indeterminate legal concepts and polysemous words. Nevertheless, at times ambiguity and vagueness are deliberately sought in the rule by covering as many cases as possible to which it can be applied, or owing to the inherent characteristics of language, so as to facilitate political agreement. The use of synonyms may affect the precision of the rule and should therefore be avoided. Uniform language is precise language and therefore uniformity is an essential component for precision in the communication of the rule. In this sense, creative style that varies the terminology to draw the reader’s attention is not appropriate for normative language. It is also important that the same terms always express the same concepts, unless in a specific context the desired meaning is perfectly clear and no other term is available; this principle is important especially in subsequent modifications of the law. The use of technical terms should be reduced as far as possible, and, when necessary, recourse should be made to the use of technical terms that have their own meaning, adding definitions to clarify them and using them throughout the law with the same meaning. In the same way, one should avoid the use of neologisms, foreign terms and unusual linguistic constructions. In these contradictory and dialectical relations that the drafter has to face, a general answer cannot be given to this type of problem, as this will depend to a large extent on the type of rule in question and on the part played by the specific measure in question, but above all it will depend on the effectiveness of the rule. If the recipients are trained legal professionals, precision can predominate; if the recipients are laypersons, then clarity and the lack of ambiguity must prevail. As to the type of rule, coercive rules require detailed and precise drafting. This is not so necessary, however, when the legislator merely aims to set down general principles and leave their development case by case to the courts. As for the type of measure, definitions, for example, require more precision than clarity. However, accuracy and clarity are the two most obvious goals. In recent years, and in the interests of the clarity of normative texts, more and more importance is being given to ‘gender-neutral language’. Therefore, new legislation is being drafted in more and more ‘gender-neutral’ terms, although this comes with the risk of loss

5 

For Canada see also Will-Kare Paving (2000) 1 SCR 915. definitions may affect legal certainty and cause the unconstitutionality of a law, such as Legislative Decree 1/2005, 10 June 2005, ruled by STC 234/2012, 13 December 2012. For unclear references leading to confusion, the Spanish Constitutional Court ruled that Canary Parliament Law 14/1987, 29 December 1987 was unconstitutional because of the complex and unclear remissions used by that law: STC 46/1990, 15 March 1990, § 4 in fine. 6  Deficient

132 

Mario Hernández Ramos and Volker Heydt

of simplicity and elegance.7 Gender-neutral language aims to avoid gender specificity in the pronoun used. Consequently, using the masculine pronoun should be avoided. However, clarity, precision and unambiguity predominate if there is a choice to be made. Ultimately, these three aspects serve effectiveness, which, in turn, along with efficiency, promotes efficacy in the legislative process. For its part, normative clarity requires the texts to formulate unequivocally their normative value, that is, their nature and scope; it requires the content of each rule to have a complete and logical homogeneous argumentative structure and a typified structure following a logical order; and it requires appropriate publicising of the rules.

8.1.2. Conciseness Laws should be drafted concisely, without superfluous elements. The principle of economy in language, which plays an important role, imposes brevity and simplicity on the wording. The key phrase is ‘use only necessary words’. The capacity of comprehension is proportionally opposed to the length of the sentences. Moreover, the longer the initial sentence, the harder it will be to absorb incorporations without coherence or intelligibility being lost. Short sentences encourage the logical organisation of thought, arranging the ideas and separating them suitably by means of punctuation marks. The rule of brevity must extend to the law as a whole: the shorter the law, the more efficient it is likely to be. The elimination of unnecessary words enhances the readability and understanding of what is written. In addition to this, it is important to stress that the courts will endeavour to give effect to every word of legislation. Therefore the legislative body must choose each word with great care and should not include a word unless it intends the word to have some effect. This being the case, the directive to eliminate unnecessary words rises from a mere principle of good writing to an imperative in legislative drafting. Each word in legislation adds to or detracts from the meaning of other words used. Each word invites those who must interpret the legislation, particularly courts, to engage in statutory construction to find legislative intent when no obvious intent exists. Laws should therefore be formulated briefly and simply. Superfluous wording, unnecessary repetition and the use of examples should be avoided except when they are strictly necessary to clarify concepts or specify general clauses. Sentences must be restricted to expressing a single idea, while articles must bring together a series of ideas that have a logical connection between themselves ‘cada artículo, un tema; cada párrafo, un enunciado; cada enunciado, una idea’ (each article a theme; each paragraph a principle; each principle an idea).8

7  A negative example is s 23 of the German Road Circulation Regulation (StVO), which has been adapted to gender-neutral language by replacing the male noun ‘Fahrzeugführer’ (vehicle conductor) by the gender-neutral expressions ‘Fahrzeugführende’ and ‘wer ein Fahrzeug führt’ (those conducting a vehicle) in order to avoid the addition of the female noun ‘Fahrzeugführerin’. However, these new expressions are artificial; nobody would ever use them in plain language. 8  Directive No 26 of the Directrices de técnica normativa de 28 de julio de 2005 del Ministerio de Presidencia, Gobierno de España.

Legislative Language and Style

 133

8.2.  Legal Definitions Texts that are too long resist accessibility and comprehensibility. That is why drafters and framers of legislation must avoid repetition in legislative texts. Techniques used for this are definitions and references. Definitions in normative texts are used for brevity, clarity and precision. They make it possible to avoid the repetition of lengthy expressions, and eliminate ambiguity where the defined term is intended to have a narrower or broader meaning than the one provided by dictionaries or the one customarily assigned to it. The details of an idea can be put into a definition so that the defined term will express them each time it is used. This may also simplify the drafting of the provisions that use the term, rather than loading them down with details.9 Definitions are directly linked to the principles of the rule of law, of the legal certainty and the separation of powers, because a definition is the coercive determination of the exclusive manner by which a term must be understood and used; it would be not fair to wait until a difference arises and makes a random victim bear the burden of interpretation. Scholasticism, developing the ideas of Aristotle, already distinguished between actual definitions that described or attempted to capture the essence or nature of a thing (quid rei) and nominal definitions that clarified the meaning of a linguistic sign (quid nominis). Within the category of nominal definitions, one can make a distinction between lexical definitions (quid verbis), which attempt to ascertain the meaning of a specific linguistic expression and which are explained in dictionaries concentrating on the linguistic uses of a community of speakers, and stipulative definitions (quid conventionis), which establish a certain meaning for a specific expression. By means of these stipulations, the decision is made to use a specific term in a certain sense that may go beyond the definitions given in dictionaries and linguistic uses. This is the type of definition most frequently found in legislation, which explicitly allocates a meaning to a word, phrase or symbol, and stipulates that throughout a specific law, in specific sections of it, this symbol must be interpreted according to this meaning. Stipulative definitions are not subject to former uses or meanings previously given, thus setting the lexical definitions against each other. The use of definitions is only useful insofar as it can be used for the fundamental aims of determining the meaning of the legislation and communicating it to its recipients. Drafters should avoid defining terms which do not depart from their ordinary meaning. They should only define terms if their ordinary meaning is unclear or overly broad, or if some technical meaning is intended. For this reason, lexical and stipulative definitions are generally related, as it is logical for a legislator to define a technical term starting from its common meaning and subsequently specifying, restricting or widening it in the sense he or she seeks and intends, avoiding forced and artificial definitions. Definitions may clear up doubts as to the technical meaning of the concept defined, but they cannot be considered a panacea, as in themselves they do not transform a text in specialised language into one in plain language. In any case, one must be aware that clarity is not a quality intrinsic to the definition but, rather, the result of a plain and logical way of arranging and expressing the thought defined. The use of definitions in the law must be

9 

See definitions in a specific section (eg UK, EU) or in the legislative text itself (eg Germany, Austria).

134 

Mario Hernández Ramos and Volker Heydt

moderate in the case of specifically legal terminology. Excessive reliance on the legal definition of the specialised terms used will cause more problems than it may solve. Balance has to be marked by the rule of speciality. A rule can and should define stipulatively the basic legal concept or concepts, which are the specific and central objective of the regulation. Under no circumstances, however, should one venture to give general definitions of the legal terminology that is instrumental for a given regulation and which has its specific enclave in other laws. The more specialised the subject matter of the regulation, the more numerous definitions are acceptable. However, what is unequivocally defined in the dictionary should not be defined in the regulation.

8.2.1. Functions In legislation, definitions play two main roles. The more important one is to avoid ambiguity, and the other one is to avoid tedious repetition. Before drafting a rule, the drafter ought to have a clear idea of the sense and purpose of the definition: is it used to avoid ambiguity or should it serve as abbreviation? In the first place, ambiguity is avoided by a process of delineation, the drawing of boundaries around the stipulated meaning of the term. Definitions that aim to avoid ambiguity may be of three types: delimiting, extending or narrowing.10 A delimiting definition determines completely the limits of the significance to be attached to the term defined. The purpose is not to alter conventional significances, but to provide a desirable degree of definiteness that may be of value in the case of ambiguous words having a number of different meanings. An extending definition stipulates for the defined term a meaning that in some respect goes beyond the meaning conveyed by the term in common usage. The second function of definition in legislation, referred to as labelling definition, is to abbreviate a group of words by stipulating that one word or phrase is to signify the longer group of words. To conclude, definitions are a useful way of making the law simpler, more readable, more consistent and clearer.

8.2.2. Rules Currently, the two basic rules of definitions are eliminability and non-creativity. The former refers to the fact that all that can be established with the defined expression must also be able to be established without it, if it is replaced by the definiens. The second implies that a new sign is introduced (or the pre-existing vocabulary is increased), but it may not be used to introduce new concepts (which are not already deducible from the original act). Both rules are thus characterised by interchangeability between the definiens and the definiendum.

10 

H Xanthaki, Thornton’s Legislative Drafting, 5th edn (London, Bloomsbury Professional, 2013) 165.

Legislative Language and Style

 135

To ensure that a definition is in keeping with these two basic rules, another series of traditional rules or principles can be mentioned: —— Define only when necessary. It is necessary when a legal meaning needs to be established or when abbreviation is required. —— Make clear what is being defined. —— Make the definiendum clear when it is used in the sense defined. —— Do not define the same expression several times in different ways. —— When defining, one must delimit the scope of application of the definition as well as possible. —— Definitions should be as self-contained and complete as possible, but if a reference is necessary, its objective must be clearly specified. —— The definitions should be located systematically at the beginning of the law or the part of the law they affect. —— Rules establishing definitions may not contain precepts of another nature. Definitions have been very common in English law. The practice of continuous recourse to definitions has penetrated EU legislation11 and also continental regulations, like in Spain, but mostly with laws related to technical objects.12

8.3.  Techniques in Normative Language Bentham criticised the use of normative language for being obscure and for causing, whether intentionally or not, the distancing, the lack of knowledge and the errors of citizens as to the content of laws.13 He saw a direct link between the firmness in the meaning of words and the degree of political security and good government. For this reason, one of the most controversial matters in legislative technique is the use of technical or specialised terms and the extent to which they should be used, as the more technical the language used the more precise their meaning for specialists on the subject but the less understandable they are for most of the population. Complex legislation not only complicates comprehension and interpretation, but also irritates its audience and may otherwise tend to undermine its authority. Moreover, it creates an additional burden by an increased need for administrative explanation. The more vulgar or colloquial the terms used, the easier it will be for the

11  Guideline 14 of the Joint Practical Guide (of the European Parliament, the Council and the Commission) for persons involved in the drafting of European Union legislation: ‘Where the terms used in the act are not unambiguous, they should be defined together in a single article at the beginning of the act. The definitions shall not contain autonomous normative provisions.’ See, eg, Art 2 of Directive 2014/49/EU of the European Parliament and of the Council on deposit guarantee schemes: ‘Definitions: 1. For the purposes of this Directive the following definitions apply: …’ 12  In Spain, for instance, see Law 22/2015, 20 July 2015, about audit, Art 3; Law 5/2015, 26 June 2015, about terminally ill people’s rights and dignity guarantees, Art 3; Royal Legislative Decree 1/2015, 24 July 2015, approving the consolidated text of the law won guarantees and rational use of medicines and medical devices, Art 2; Organic Law 2/2012, 3 March 2010, about sexual and reproductive health and abortion, Art 2; Catalan Law 14/2015, 21 July 2015, about tax on empty houses, Art 5. 13  J Bentham, Nomografía o el arte de redactar leyes (Madrid, Centro de Estudios Políticos y Constitucionales, 2004) 89–91.

136 

Mario Hernández Ramos and Volker Heydt

layman to understand the legislator’s message. The objective is therefore to find a balance between these two drafting styles. The conciliation between them could be found by using quotidian terms in the title and more legally precise words in the rule.

8.3.1.  Legal Lexicon The starting point is that the draft must be written in legal language. It is the technical, scientific language of law. It must be precise. The core of the specificity of legal language can be found in a storehouse of words rather than in specific characteristics or in a certain style. In this way, normative language fulfils the function of condensing into a series of words the basic concepts on which experience and methodological knowledge of law are based. The reason for its use has already been mentioned throughout this chapter: semantic univocity, lexical economy and, above all, conceptual precision and clarity. The inevitable technical servitude of normative language is an obstacle to its clarity to the citizen. Nevertheless, the drafter has tools to help overcome this difficulty as far as possible and attempts to avoid the unnecessary saturation of the law with legal technicalities.14

8.3.2.  Non-legal Lexicon Legal rules also use non-legal technical terminology. This is the case when law deals with specialised technical sectors of social reality which have their own language and vocabulary. In this sense, normative language absorbs the most varied technical nomenclature. The use of this non-legal technical vocabulary represents a serious problem for the clarity of language. However, this difficulty is partly overcome because the more specialised the content of a rule, the more restricted and specialised the circle of immediate recipients. In any case, the text of a law must also show communicative qualities and must be understandable by anyone, which means that legal terminology tends to contain words taken from standard language that have acquired a particular legal meaning that is completely different from that of ordinary use. The fact is that, in contrast to other forms of technical language, the proportion of words used exclusively in law is very low; most of them are terms from plain language and even from other technical languages that law modulates semantically with its own meanings. In short, when drafting laws, plain language should be preferred when its use is suitable; however, in the interests of clarity, precision and accuracy, technical language should be used whenever necessary. Because of this, the language of the drafter tends to be plain language plus a series of technical terms redefined stipulatively based on their common meanings.

14  Art 1296 Spanish Civil Code: ‘Words which may have different meanings shall be understood in the meaning which is most in accordance with the nature and subject matter of the agreement.’

Legislative Language and Style

 137

8.4.  Style of Legislation Although the aim of legislation is the same throughout the world—to provide rules for regulating society and government—there are considerably different styles with regard to —— the overall design of the legal order; —— its organisation into individual pieces of legislation; and —— the fine-tuned structure of each individual piece of legislation as well as of the rules contained therein.

8.4.1.  Design of the Legal Order and Access to the Law The overall design of the body of law is influenced and partly determined by constitutional rules, in particular on the allocation of functions and competences between different actors (parliament, government, head of state) and different levels (central state, regions, EU), and between types of regulation (parliamentary or subordinate legislation). Moreover, under the rule of law, there are not only substantive conditions, such as respect for fundamental rights and for international law, but also formal requirements for the validity of an act, such as its publication, words of enactment, preambles or, in the case of restriction of a fundamental rights, the mention in the act of the affected right. The constitutional characteristics have to be taken into account for the breakdown of the law into different sets of rules; they may take the form of comprehensive codes or piecemeal legislation. One might dream of conceiving a single instrument of law for the whole of the legal order, but this is not possible due to the frequency and volume of new or amending regulations, which are dictated by political requirements and considerations. However, sectoral codifications, prominently realised in the French Civil Code of 1803/04, and the specific codes which were still being adopted under Napoleon are common features of many legal orders. Instead of attempting to have a single code, a systematic classification of the ever increasing number of rules has been developed in many states, because the chronological order, based on publication in the official gazette, is neither sufficient nor satisfactory. A systematic order is even more necessary today as modern legislation consists of a great number of individual acts without explicit relations between them. Moreover, most of them are designed to modify existing rules and do not therefore have a substantive standing on their own. Various official or private tools have been produced in order to ease the access to the law, such as alphabetical indices, systematic directories and collections of consolidated legislation. Modern information technology has considerably improved the possibilities of a structured access to the law. A very useful method is the immediate incorporation of modifying acts ex officio into the existing rules so that the accessed rules are always up to date in an authentic version (eg in Switzerland, Slovenia).

138 

Mario Hernández Ramos and Volker Heydt

8.4.2.  Structure and Style of the Individual Piece of Legislation Where a national15 classification plan exists for the legislation, the drafter should examine and determine from the outset where in that system the place of the new rules to be drafted will be. Transparency is enhanced when a reference to that place in the classification scheme is kept as an integral part of that piece of legislation or is at least officially communicated together with the publication. The structure of an act is mainly a matter of common sense and depends on the subject matter and the addressee, the dominating requirements being transparency and clarity. The existing varieties and specific (normally national) features and patterns result from autonomous decisions of the respective legislators, sometimes drawing on examples from of other states. Astonishingly, the (meta-)rules on the structure of legal acts are not normally laid down in an Act of Parliament (although they apply to these acts in the first place), but rather are contained in administrative instructions or just follow customary practice. However, sometimes isolated instructions are found in legislative texts.16 The principle of clarity leads to more specific requirements with regard to the structure of acts and its subdivisions. Rules must be drafted in a way which allows them to be quoted easily and immutably. This implies that future modifications should not lead to a renumbering of original articles17 or paragraphs; instead, intermediate identifiers (eg Article 4a after Article 4) should be used. At the same time, attention must be paid to the aptitude of the structure for legal documentation, essential for the access to the law. A modifying act should also take into account the need to produce an officially or privately consolidated text. It is not acceptable that, in EU legislation, recitals, although an integral part of the act, are omitted from the consolidated version because the amending act does not provide for a corresponding modification of the recitals. As regards the substantive contents of the law, the above-mentioned principles should determine the language used. For some matters, traditional structures have become sector specific: penal law is often codified in a catalogue of rules each of which contains the sanction for an incriminated behaviour, which is described in detail; however, the implied prohibition is not expressed directly. By contrast, in public law, obligations and respective sanctions are mainly found in different chapters. A considerable difference in style can be evident between the legal rules and the explanatory notes by which the author of draft legislation clarifies their meaning and envisaged effects; they rather resemble a commentary. A similar difference in style can exist between legislation and court judgments. Judgments have in their style hardly anything in common with legislative rules, although they are designed to be a concretisation of them. This is particularly remarkable because judgments of the highest courts are respected and applied as if they were emanating from the legislator.

15  Different from libraries, which can use the Universal Decimal Classification for books, there is no recognised international classification system for legislation. 16  Some specific drafting instructions are made, for example, in Art 1(4)(f) of EU Council Regulation No 539/2001 as amended by Regulation No 1289/2013; cf judgment of the CJEU of 16 July 2015 in the Case C-88/14 European Commission v EP and Council. 17  The repeated renumbering of the EU Treaties, by the Treaty of Amsterdam (1998) and again by the Treaty of Lisbon (2007), has led not only to considerable confusion in communications of lawyers, but also in databases.

Legislative Language and Style

 139

A general feature of all legislation is the style of the title of an act: should it be a (full or abbreviated) title, a number in a classification system or a publication number, or a combination of them? If the title of an amending act includes the title of the amended act, it should be made clear, even by layout or typography, which part is a quote from the original act. A useful tool enhancing transparency is the addition of a table of contents, in particular in the case of long and/or complex acts. If such tables are conceived as an integral part of the act (and not just added by administrative services), they require a legislative amendment if they become incorrect due to changes in the substantive rules. When legislation affects and therefore modifies many other existing acts, such as an annual act on the budget, it should be mandatory to include a table of contents in the act, listing all modified acts irrespective of the importance of changes for them. Specific challenges result from the need to incorporate European law into the national legal orders of the 28 EU Member States. The 24 linguistic versions of EU law are not only equally valid and binding, but are also drafted in a single style, which thus cannot take on board the different styles used in the Member States’ national legislations. This leads to a cohabitation of styles in the same territory between national and directly applicable EU regulations. An appealing way to overcome this problem would be a harmonisation of styles following an exchange of best practices between national draftsmen, which could usefully be organised at EU level. When legislation is made for the implementation of EU directives or international treaties, which normally do not cover the full range of rules of a given subject, the occasion is often used for the modification of purely domestic rules in the same area. Unfortunately, this makes it more complicated to distinguish whether an individual rule may be autonomously modified or whether it is bound to exist under EU or international law. Further issues are the inclusion of references to the different steps of the legislative procedure and to the justification for the legislation. At EU level, the legislative procedure, as enshrined in the TFEU, requires that the motivation for the legislation be included in the act (this is done by recitals); however, the normative legal value of the latter is not clear. The addition of references to the different stages of the legislative process preceding the final adoption and publication, as, for example, in French legislation, is useful for research, but has only a documentary purpose without legal value.

8.4.3.  Linguistic Aspects A need for multilingual legislation in specific areas has existed since the first international treaties were concluded between states with different languages. The French language has lost its exclusive role in diplomacy, but still influences the specific style of international treaties. Even if negotiations are conducted in a single language—agreed upon by the contracting parties—and this language is the one that should be relevant for settling divergent interpretations, the texts have to be drawn up in the official languages of all parties. By far the most complex situation is found in respect of EU legislation (see Chapter 13). And as the terminology in the European rules is not necessarily identical to that used in national laws, there is a risk of legal confusion or uncertainty. As an example, consider the use of the word ‘codification’. This expression is normally used for a systematic reorganisation of the

140 

Mario Hernández Ramos and Volker Heydt

rules in a given field, whereas the EU legislation uses it for amended and consolidated rules of any kind and size which are made subject to a fresh legislative procedure; and in the UK, ‘codification’ means taking the rules determined by judges in case law and turning them into legislation. In states where several languages are officially in use (mostly federated states such as Switzerland, Canada and Belgium), there is an obvious need for legally equivalent corresponding linguistic versions of the applicable legislation. Under the rule of law, it is inconceivable to confront citizens with rules written in a language which they do not understand. Therefore the whole body of legislation needs to be available and accessible in all official languages. An optimal way to prevent problems resulting from legislation of a text written in one language being transcribed into another is simultaneous drafting, as is done in Canada, where two native speakers sit at the same table and discuss the given legal contents during the drafting process. At EU level, the joint check of translations by legal revisers at round table sessions, which took place until the 2004 enlargement, has been abandoned because it became unworkable after the increase in the number of relevant languages. An upcoming challenge could result from the integration of linguistic minorities, which are likely to become more widespread due to globally increased mobility, in particular immigration. This problem is sometimes addressed at administrative level for specific purposes (eg by multilingual forms for social security purposes or driving licences), but not yet at the level of legislation.

8.5.  Common Law and Civil Law Approaches to Legislation The legal-cultural identity of legislation is an important style feature, determined by the constitutional context, the nature of the legal system in which the legislation is found and the origins of that system.18 In the Western world and in the jurisdictions under its influence, most legal systems derive from either the civil law or common law tradition.19 The most distinctive character of civil law has traditionally been that rules are predominantly codified and as a consequence the code prevails over individual judicial decisions. By ­contrast, common law was largely developed by judicial decisions. Although this is a simplification of a very complex legal situation, it is useful to illustrate the difference in focus that influences legislation and the style in which it is drafted.

18 

W Voermans, ‘Styles of Legislation and their Effects’ (2011) 32 Statute Law Review 38, 51. See generally GC Thornton, Legislative Drafting (London, Butterworths, 1987); W Dale, Legislative Drafting: A New Approach (London, Butterworths, 1977); JA Clarence Smith, ‘Legislative Drafting: English and Continental (1980) 1 Statute Law Review 14; R Dickerson, The Fundamentals of Legal Drafting, 2nd edn (Boston, Little Brown & Co, 1986); D Melinkoff, Legal Writing: Sense and Nousense, (New York, Charles Scribner’s Sons, 1982); W Dale, ‘The European Legislative Scene’ (1992) 13 Statute Law Review 79; F Bennion, Bennion on Statute Law, 3rd edn (Longman, 1990); P Noll, Gesetzgebungslehre (Reiberck, Ro-Ro, 1973); J. Rödig (ed), Studien zur einer Theorie der Gesetzgebung (Berlin, Springer, 1976); M Gotti, ‘Linguistic Insights into Legislative Drafting’ (2014) 2 The Theory and Practice of Legislation 123; H Xanthaki, ‘Editorial. Burying the Hatchet between Common and Civil Law ­Drafting Styles in Europe’ (2012) 6 Legisprudence 133. 19 

Legislative Language and Style

 141

In continental Europe, which comprises predominantly civil law jurisdictions, the traditional approach to drafting laws has been that of expressing the law in general principles, with it being up to the courts and the executive to specify the details for the application of the normative precepts to individual cases, taking into account the general intention of parliament as expressed in the preambles, discussions and other documents. Continental law tends to be expressed in shorter sentences, with less use of paragraphing but with more understandable principles, despite the fact that the relationship between several principles may be more complex. This approach leads on the one hand to simpler, clearer, and more general legislation, with the objective being for the rules to withstand the passage of time. On the other hand it is a legislation that lacks the certainty that would be afforded by a more detailed application of the principles. In contrast, in the common law tradition, legislative drafting has always been intended to be precise, unambiguous, all-inclusive and perhaps even exhaustive, with as detailed specification of legal scope as possible, even at the cost of accessibility to its audiences, both specialists and laymen. The approach adopted by common law jurisdictions consists of expressing in the laws themselves the precise manner in which they should be applied to certain circumstances without it being necessary to await the decisions of the courts in specific cases. The aim of the drafter is to draw up a legislative text containing all the necessary details for the text itself to regulate the actual situations for which it has been drafted. As common law is based on the principle of precedence, this system regards certainty of expression as the most valued quality in legal drafting. This approach leads to a more complex legislation that is less clear to the reader. This is because the civil-code draftsman is eager to be widely understood by the ordinary readership, whereas the common law draftsman seems to be more worried about not being misunderstood by the specialist community. This style puts its faith in precision and pursues the certainty of expression to the detriment of clarity and comprehensibility. English laws seem to be long-winded and lacking in transparency, with a wealth of detail and few principles. Indeed, the continental style is more often based on principles, comprehensive approaches and conceptual legal thinking.20 In more recent times, however, legislative drafting in the common law tradition has started adopting a number of ‘easification’ and ‘plain language’ devices in an attempt to better manage legal messages, leading to more accessible legislation. The great advantage of legislation based on the general principles of civil law is that it is easier to read and its objectives are easy to understand. However, the disadvantage is its higher degree of uncertainty, as it is the executive and the courts that will a posteriori provide content and sense to such general formulae. In these cases, the legislator is renouncing control of the actual facts and passing them to the judiciary, that is, is abandoning (implicitly delegates) his own competence in favour of judges. This has been severely criticised from the point of view of common law, as it is hard to predict how the courts will do part of the work which the legislator has hitherto done. The reason for choosing one system of drafting or another lies in a whole series of considerations that are deeply rooted, in particular in history, culture and the institutional system. This background is beyond the control of the drafter, who is merely responsible for warning the corresponding politician of the effects in each case of using general clauses,

20 

Dale (ibid) 331–33.

142 

Mario Hernández Ramos and Volker Heydt

principles or excessively vague terms. Moreover, the choice of approach depends both on whether the legislation is in the field of private or public law and on civic culture in general and the culture of judges in particular. Despite all these reflections, the integration of the UK and Ireland (common law jurisdictions) in the EU is giving rise to reciprocal influences that are allowing a greater rapprochement in the manner of drafting the EU’s rules. This is called the ‘unification of the law’. The transition from equalisation of law in style and contents to unification is proceeding step by step, is inevitable and desirable. This transition is supported by two facts: the opening of states and their constitutions to supra- and international organisations as well as its effects: the emergence of European and International Unitary Law.21

Important factors in this process are the judgments of the Court of Justice of the EU and the European Court of Human Rights. The difference in legislation and drafting style between common and civil law is becoming ever weaker and is leading to the convergence of the two systems from conceptual approaches to drafting conventions. The rules for drafting quality legislation are the same in the two legal systems. The ‘drafting style’ is actually less defined by legal tradition and increasingly more by common standards of good quality of legislation, namely clarity, simplicity, precision, accuracy and plain language.

Further Reading Assemblée nationale [France] (ed), Guide pour la rédaction des textes législatifs (May 2009). Barnes, J, ‘The Continuing Debate About “Plain Language” Legislation: A Law Reform Conundrum’ (2006) 27 Statute Law Review 83. Beigner, B (ed), La codification (Paris, Dalloz, 1996). Bennion, F, Bennion on Statute Law, 6th edn (New York, LexisNexis, 2013). Bentham, J, ‘Nomography or the Art of Inditing Laws’ in J Bowring (ed), The Works of Jeremy Bentham (Edinburgh, William Tait, 1843). Bhatia, VK, ‘Linguistic and Socio-pragmatic Considerations in Legislative Drafting’ (2014) 2 Theory and Practice of Legislation 14, 169. Biel, L, ‘Translation of Multilingual EU Legislation as a Sub-genre of Legal Translation’ in D Kierzkowska (ed), Court Interpreting and Legal Translation in the Enlarged Europe (Warsaw, Translegis, 2006) 144–63. Bundesministerium der Justiz und für Verbraucherschutz [Germany], Handbuch der Rechtsförmlichkeit, 3rd edn (2011). Bundesministerium des Innern [Germany] (ed), Handbuch zur Vorbereitung von Rechts- und Verwaltungsvorschriften. Empfehlungen zur Vorbereitung von Rechts- und Verwaltungsvorschriften nach § 42 Abs 3 und § 69 Abs 2 der Gemeinsamen Geschäftsordnung der Bundesministerien, 2nd edn (Bonn, Bundesanzeiger, 2012). Butt, P, Modern Legal Drafting. A Guide to Using Clearer Language, 3rd edn (Cambridge, Cambridge University Press, 2013). 21 

Karpen (n 3) 162–63.

Legislative Language and Style

 143

Conseil d’Etat [Belgium], Principes de technique législative. Guide de rédaction des textes législatifs et réglementaires, 2008. Cormacain, R, ‘Accessing Legislation: 40 Years Post Renton’ (2013) 9 European Journal of Current Legal Issues. Dale, W, Legislative Drafting: A New Approach (London, Butterworths, 1977). Dale, W, ‘The European Legislative Scene’ (1992) 13 Statute Law Review 79. Dickerson, R, The Fundamentals of Legal Drafting, 2nd edn (Boston, Little Brown & Co, 1986). García-Escudero Márquez, P, Manual de técnica legislativa (Madrid, Civitas, 2011). García Mexía, P and Rodríguez Arana, J (eds), La técnica normativa: una necesidad para la realización de la justicia (Madrid, CEU, 2010). Göbel, L, ‘Gesetzessprache heute’ in Von der Stimme aus den Wolken zum Bundesgesetzblatt. Beilage zum Bundesanzeiger Nr 145 vom 6 August 1983, Beilage 37/83 (Bonn, Bundesanzeiger, 1983) 59–74. Gotti, M, ‘Linguistic Insights into Legislative Drafting’ (2014) 2 Theory and Practice of Legislation 123. Greenberg, D, ‘The Techniques of Gender-Neutral Drafting’ in S Constantin and H Xanthaki (eds), Drafting Legislation. A Modern Approach (Aldershot, Ashgate-Dartmouth, 2008) 63–76. Holzinger, G, ‘Die Technik der Rechtssetzung’ in H Schäffer, Theorie der Rechtssetzung (Vienna, Manz, 1988) 275–302. Karpen, U, ‘Comparative Law: Perspectives of Legislation’ (2012) 6 Legisprudence 149. Karpen, U, ‘Instructions for Law Drafting’ (2008) 2 European Journal of Law Reform 163. Müller, G and Uhlmann, F, Elemente einer Rechtssetzungslehre (Zurich, Schulthess, 2013). Prieto de Pedro, J, Lenguas, lenguaje y Derecho (Madrid, Cuadernos Civitas, 1991). Publications Office [European Union], Interinstitutional Style Guide (Brussels, Europa, 2011). Roebuck, D, ‘Plain, Clear and Something More? Criteria for Communication in Legal Language’ (2014) 16 European Journal of Law Reform 633. Roznai, Y, ‘A Bird is Known by its Feathers. On the Importance and Complexities of Definitions in Legislation’ (2014) 2 Theory and Practice of Legislation 145. Special Issue: Law in Translation (2014) 20 The Translator 1. Sullivan, R, Construction of Statutes, 5th edn (New York, LexisNexis, 2008). Tanner, E, ‘Clear, Simple and Precise Legislative Drafting: Australian Guidelines Explicated Using an EC Directive’ (2004) 25 Statute Law Review 223. Vidal Marín, T, ‘Técnica legislativa, inserción de la norma en el ordenamiento jurídico y Tribunal Constitucional’ (2013) 31 Teoría y Realidad Constitucional 323. Voermans, W, ‘Styles of Legislation and their Effects’ (2011) 32 Statute Law Review 38. Xanthaki, H, ‘Editorial. Burying the Hatchet between Common and Civil Law Drafting Styles in Europe’ (2012) 6 Legisprudence 133. Xanthaki, H, Thornton’s Legislative Drafting, 5th edn (Haywards Heath, Bloomsbury Professional, 2013). Xanthaki, H, Drafting Legislation. Art and Technology of Rules for Regulations (Oxford, Hart Publishing, 2014). Zapatero, V, El Arte de Legislar (Pamplona, Thomson Aranzadi, 2009).

144 

9 Legislative Drafting STEFAN HÖFLER, MARKUS NUSSBAUMER AND HELEN XANTHAKI

Legislative drafting is the process whereby the conceptualisation of a new legislation is transformed into an actual legislative text. This process can be broken down into several, potentially repeated stages (planning, composing, revising, editing) and involves a range of actors with different skills (policy makers and domain experts, drafting specialists and language experts, translators). We argue that the way in which the drafting process is organised can have a substantial impact on the quality of the product. For this reason, we first discuss the requirements that each stage of the drafting process must meet in order to facilitate the production of high-quality legislative texts. We then introduce the major models by which the drafting process has been organised—in civil law and common law countries as well as in multilingual jurisdictions—and we identify their respective strengths and weaknesses.

9.1. Introduction Legislative drafting is the process whereby the conceptualisation of some new legislation is transformed into an actual legislative text.1 The aim of legislative drafting is to produce a text that meets the formal and linguistic properties typically exhibited by ‘good’ legislation: clarity, precision, conciseness and coherence, absence of redundancies and inconsistencies, as well as conformity with the formal and linguistic conventions that have emerged for this particular class of texts.2 At the same time, the texts have to accurately implement the intended policies. The drafting process falls into several, potentially repeated stages (planning, composing, revising, editing) and may involve a range of actors, each of them

1  Note that we use the term ‘legislative drafting’ in its narrow sense here, ie, to denote the actual process of writing a legislative text. In its broad sense, as the term is used, eg, in H Xanthaki, Thornton’s Legislative Drafting, 5th edn (London, Bloomsbury, 2013) and in H Xanthanki, Drafting Legislation: Art and Technology of Rules for Regulation (Oxford, Hart Publishing, 2014), ‘legislative drafting’ encompasses the five stages of understanding the drafting instructions, analysing them, designing the new legislation, composing and developing the text (ie ‘legislative drafting’ in the narrow sense in which the term is used in this chapter), and submitting the product to scrutiny and testing. 2  See chapters 7 and 8.

146 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

bringing their own specific expertise to the process (eg policy makers and domain experts, drafting specialists and language experts, translators). The way in which the drafting process is organised can have a substantial impact on the quality of the product: whether legislative drafting results in good legislative texts also—and crucially—depends on the conditions under which it takes place. This chapter elaborates on this insight. To this end, we first look closely at the individual stages of the drafting process and discuss the requirements each stage must meet if it is to facilitate the production of high-quality legislative texts (section 9.2). Next, we present the two major models according to which the process of legislative drafting has been organised in civil law and common law countries, and discuss the ways in which the different actors contribute their specific expertise in each of these models (section 9.3). We then complement this discussion with an analysis of the peculiarities of legislative drafting in multilingual legal systems (section 9.4).

9.2.  The Drafting Process and Its Stages First and foremost, the process of legislative drafting is one of producing a written text. As such, it is not fundamentally different from the production of other types of texts. Independent of how its specifics are organised, the drafting process naturally comprises four major stages: planning, composing, revising and editing.3 These stages are not necessarily completely distinct: there may be overlaps between two consecutive steps and each stage may be repeated if required. Occasionally, it is even necessary, at some point in the process, to go back to an earlier stage and start afresh. In a similar fashion, each individual stage usually consists of repeated acts of contemplating, specifying, generating, annotating, editing and re-interpreting notes, plans and text (cf figure 9.1). In sum, producing a text is hardly ever a straightforward, linear process but rather a recursive one, that is often required to undergo several iterations before it reaches completion. Legislative drafting is no different in this regard.4 We take this observation as the starting point of our discussion. Bearing in mind that drafting is an inherently non-linear process, we will now briefly explain the characteristics of each of the aforementioned stages (planning, composing, revising, editing) and discuss some of the specific requirements each stage must meet in order to facilitate the production of high-quality legislative texts.

3  See, eg, S Molitor-Lübert, ‘Die Produktion schriftlicher Texte’ in G Hartmut and O Ludwig (eds), Schrift und Schriftlichkeit, HSK 10.2 (Berlin, Walter de Gruyter, 1996) 1005–27; M Sharples, How We Write: Writing as ­Creative Design (London, Routledge, 1999); A Wrobel, ‘Phasen und Verfahren der Produktion schriftlicher Texte’ in K Brinker, G Antos, W Heinemann and SF Sager (eds), Text- und Gesprächslinguistik, HSK 16.1 (Berlin, Walter de Gruyter, 2000) 458–72. 4  cf K Pattchett, Law Drafting and Regulatory Management in Central and Eastern Europe, Sigma Papers, 18 (Paris, OECD, 1997), available at www.oecd.org/officialdocuments; A Seidman and N Abeyesekere, Legislative Drafting for Democratic Change: A Manual for Drafters (The Hague, Kluwer, 2000); Xanthaki, Thornton’s ­Legislative Drafting (n 1).

Legislative Drafting

 147

Figure 9.1: The recursiveness of the first three stages of the writing process as represented by Sharples (n 3) 72

9.2.1. Planning Before the process of legislative drafting can be initiated, the normative contents of the intended legislation must have been properly conceptualised. Drafting problems can frequently be traced back to a lack of such a conceptualisation. It is thus necessary that the policy makers’ intentions have been fully understood, that the current legal situation has been properly analysed and, in particular, that the intended normative contents have been sufficiently developed before the process of drafting a text is started. If, at the onset of the drafting process or during its course, it becomes evident that the conceptualisation of the intended legislation exhibits major deficiencies, it may be advisable to put the process on hold and return to the conceptual phase.5 Once the intended legislation has been sufficiently conceptualised, the process of legislative drafting then begins with a stage in which the future legislative text is planned and outlined. This stage comprises, among other things, the following tasks: (i) determining the appropriate location of the new legislation within the existing corpus of law; (ii) outlining the overall structure of the future text; (iii) defining the key terms around which the text evolves; and (iv) searching for precedents. Some of these questions—especially determining the appropriate location of the new legislation within the existing corpus of law—should already have been addressed in the conceptualisation phase; however, the results may need to be refined at this pre-writing stage.

5 

See Chapter 6.

148 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

The task of determining the appropriate location for the new legislation involves, in the first instance, deciding on the level in the hierarchy of legal norms at which the new contents are to be placed: can they go in an ordinance, must they be enacted by statute, or do they even warrant a new constitutional provision? Depending on the chosen level, the norms will have to exhibit a greater or lesser degree of specificity. Where they are to be cast at the level of statutes and ordinances, one will further have to determine whether a new text needs to be composed or an existing one must be amended, or, respectively, whether a new text needs to be composed and existing texts need to be amended or repealed along with it. If some new legislation involves more than one text, one will have to decide which content needs to be put into which text. In sum, legislating always amounts to working not just with one isolated text but with a whole system of interconnected texts: any new text will necessarily have an impact on existing texts if the coherence and systematicity of the law is to be maintained. Once it has been determined where in the existing corpus of law the new legislation is to be placed, one can proceed to outlining the overall structure of the text (or, respectively, defining where the norms to be created are to be inserted into the structure of an existing text). In general, this task implies identifying the main parts into which the new text shall be organised and deciding on the linear order in which the contents are to be presented.6 The structure of the text must make it as easy as possible for readers to access and comprehend the content. To achieve this aim, two perspectives on the text must be borne in mind: on the one hand, the structure of the text is meant to convey the logical structure of its content, for example, by reflecting the chronological sequences, causal relationships and organisational hierarchies inherent to its subject matter; on the other hand, the structure of the text must be geared towards its primary addressees, that is, to those who are affected by the text and those who will have to work with it in the future. It is therefore important to consider, at this early stage of the process, who the primary addressees of the text are and what perspective they will have on the subject matter. As a further planning task, one will have to identify and define the key terms around which the text evolves. While most terminological questions will have to be addressed during the actual composition of the text and its revision, it is helpful to assemble at least the most important terms beforehand: being aware of the most central terminological questions greatly facilitates the writing process that is to follow. Some of these terms one will have to derive from existing legislation in order to maintain coherence within the law as a whole, while others will have to be coined anew, potentially having to express a crucial difference to an existing legal concept. To prevent inconsistency, the textual and terminological context of the location at which the new legislation is to be included in the corpus of law will have to be studied carefully. Nowadays, specialised terminology databases can often provide additional support.7 The aforementioned tasks can be facilitated by resorting to precedents—that is, by identifying existing legislative texts that can inform the legislation at hand. In many cases, there will be texts that had to solve problems similar to those arising from the current project; 6 

cf Chapters 7 and 8. See, eg, European Commission, Directorate-General for Translation (ed), Document Quality Control in Public Administrations and International Organisations, Studies on Translation and Multilingualism (Luxembourg, ­Publications Office of the European Union, 2013) 168–74. 7 

Legislative Drafting

 149

these texts can—and should—be used as a source of inspiration. Legislative texts do not aim for originality but often reuse phrasings and textual patterns that have already been established. Following textual conventions has the double advantage that, on the one hand, the content of the text becomes more easily recognisable to the addressees and, on the other hand, the drafters’ task becomes easier as they will not have to reinvent the wheel. However, using other texts as templates does not mean that one should simply copy and paste bits and pieces of text: doing so is likely to introduce ambiguities, inconsistencies and loopholes into the law. Any new text will still have to be properly conceptualised and thought through. Precedents can be used for inspiration, but in order to guarantee a coherent text, they must be adapted to the current project where necessary.8 Ultimately, the effectiveness of the legislation depends greatly on the process of creative transplantation to its new legislative, legal, social, economic and political environment.

9.2.2. Composing Once the text has been planned and outlined, a first draft can be composed. Composing a legislative text is special in so far as it must take place within a fine-grained system of structural units: legislative texts do not just fall into chapters and sections, but are strictly organised down to the level of individual articles, paragraphs, subparagraphs and sentences. Any content must be fitted into this framework of structural units. Nowadays, drafters often resort to the support provided by specialised authoring tools to achieve this task.9 The task of composing a first draft is best approached by proceeding from the general to the specific.10 Badly drafted legislative texts are often the result of the authors getting tangled up in the details and failing to clearly state the underlying principles. To avoid this pitfall, one should begin by formulating the title of the text and the clauses describing the purpose, subject matter and scope of the legislation. Only then should one proceed to writing down the actual normative content. However, there, too, it helps to clearly phrase the general rules before adding details, provisos, exemptions and procedural provisions. The same technique can even be applied when formulating the individual sections or articles.11 Working one’s way through the text in a more or less linear fashion, along with proceeding from the general to the specific, also helps keep track of what has already been said. It thus serves as a technique to avoid repetitions that would introduce redundancy12 and to prevent

8  See also G Müller and F Uhlmann, Elemente einer Rechtssetzungslehre, 3rd edn (Zurich, Schulthess, 2013) 97–99; H Xanthaki, ‘Legal Transplants in Legislation: Defusing the Trap’ [2008] International and Comparative Law Quarterly 659. 9  See, eg, C Biagioli, E Francesconi, P Spinosa and M Taddei, ‘Legislative Drafting Support Tool Based on XML Standards’ in Proceedings of the International Conference on E-Government (Academic Conferences Limited, 2005) 390–97; E Francesconi, ‘Structuring Legal Semantics in Legislation’ [2011] Jusletter IT 22; Bundesministerium für Justiz und Verbraucherschutz (ed), ‘eNorm: Software Designed to Aid Drafting of Legislation’ (2014), available at www.enorm.bund.de > Download. 10  See, eg, H Thring, Practical Legislation: The Composition and Language of Acts of Parliament and Business Documents (London, John Muray, 1902) 38. 11 See also S Höfler, ‘“Ein Artikel—eine Norm.” Redaktionelle Überlegungen zur Diskursstruktur von ­Gesetzesartikeln’ [2012] LeGes 311. 12  cf Chapter 8.

150 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

the creation of norms that are merely presupposed rather than stated explicitly.13 It also makes it easier to circumvent the need for complicated internal cross-referencing that can jeopardise the text’s readability. When formulating individual provisions, one first has to determine the type of norm the respective provision shall express—for example, whether it shall place an obligation, grant a right, confer some responsibility, define a term, name a goal, give rise to a claim or impose some sanction.14 There are conventional ways to phrase each type of norm; in some cases, there may even be prescribed wordings. Existing legislative texts can provide drafters with ready-made building blocks that can be adapted to the present case. However, before one can decide on the norm type of a provision, two preliminary issues need to be addressed. First, one has to analyse how many (and which) normative statements the respective provision consists of. Legislative texts are frequently difficult to comprehend and the individual norms they contain difficult to discern, because too much information has been conflated into single sentences. Disentangling the individual normative statements that a provision consists of can help prevent this problem.15 Secondly, one has to become clear about who the primary addressees of each normative statement are. Depending on the addressee, one and the same norm can appear, for instance, either as a general obligation or as an assignment to be carried out by the authorities. Legislative texts are often hard to understand because they have not been written from the perspective of the primary addressees. In particular, the view of the authorities is often erroneously chosen over the view of those who are actually affected by the respective provisions. It has occasionally been pointed out that it is advisable for the task of composing a first draft to be carried out by a single person: patchwork texts rarely make good legislation as the authors of the individual parts lack an understanding of the whole and are thus prone to introducing clauses that are incoherent or inconsistent with the remainder of the text.16 In practice, however, this requirement often cannot be met, especially where large, multifaceted legislative texts have to be composed. If it is unavoidable that multiple authors provide their input, it is thus essential that one person is in charge of bringing the pieces together and turning them into a homogeneous text.

9.2.3. Revising Once a first draft has been composed, it needs to undergo reviewing, both internally and externally, which generally leads to a revision of the text. Three aspects of the draft need to be assessed: (i) its accuracy with regard to the intended content; (ii) its compliance with the rules of substantial and formal legistics;17 and (iii) its linguistic quality, particularly its clarity and readability.18 Revising is thus an inherently interdisciplinary process that 13 See S Höfler, ‘Between Conciseness and Transparency: Presuppositions in Legislative Texts’ [2014] ­International Journal for the Semiotics of Law 627. 14  See also T Fleiner-Gerster, Wie soll man Gesetze schreiben? Leitfaden für die Redaktion normativer Texte (Bern, Haupt, 1985) 34–40. 15 See also S Höfler, ‘“Ein Satz—eine Aussage.” Multipropositionale Rechtssätze an der Sprache erkennen’ [2011] LeGes 259. 16  See, eg, Müller and Uhlmann (n 8) 301. 17  See Chapters 6 and 7. 18  See Chapter 8.

Legislative Drafting

 151

requires in-depth knowledge of the subject matter at hand, the rules of substantial and formal l­egistics, and the techniques of plain-language writing. Each of these aspects must be given a voice in the process of revising—ideally by specialists from each respective field. Revising offers an opportunity to gain a view from the outside: failure to produce clear texts can often be traced back to the authors having been too heavily involved in matters of content and policy, and thus failing to adopt the perspective of the target audience. Everyday experience has proved time and again that it is often difficult for authors to judge whether their own text is clear enough for others to understand. It is thus crucial that, during revision, the draft is (also) reviewed by unbiased persons who have not been involved in composing the text, developing its contents or shaping the policy it is meant to implement. One aspect of revision is to anticipate how the text will be perceived by its future readers. Experience has also shown that for reviewing and revising to be effective, it must be without limitations or taboos: reviewers must be at liberty to comment on all aspects of the text. Revising must not be confused with simple editing: it may be important that a text is grammatically and orthographically correct and stylistically pleasing; however, according to linguistic research, the factor that influences a text’s understandability the most is the logical structuring of its contents and the order in which these contents are presented.19 It follows that, at the stage of revision, it must still be possible to rearrange the overall structure of the text. Moreover, reviewers must be at liberty to comment not only on the form of the text, but also on its content: the form and the content of a text are never completely distinct, but rather mutually affect each other in a dialectic manner. It is thus inevitable that the drafting process may also have an impact on the content. Language specialists in particular, by asking the authors to clarify certain points so that they can be phrased in a fashion that is more easily understandable, frequently unearth loopholes, ambiguities and inconsistencies in a text and in the conceptualisation on which it is based. All these observations entail that revising—as opposed to editing—must begin early in the process: the input provided by reviewers can only be effective if the formal and informal decision-making processes are still in their early stages—that is, if the political process has not yet been completed and no arrangements have been made on the basis of the text and its content.20 To guarantee a timely consultation of specialists, it is advisable to institutionalise the procedures of reviewing and revision. It is not a good idea to leave it to the authors to decide whether their text would benefit from such input. An institutionalisation of the process of reviewing and revision must furthermore take into account that texts are rarely composed at a single cast and that revisions frequently give rise to follow-up problems that also need to be addressed. It is thus necessary that there is room and time for several iterations that allow for a continued improvement of the text and for potential revisions of revisions. Any single iteration must then be organised in a way that reflects the inherently co-­ operative and discursive nature of the process of reviewing and revising, during which ­various different perspectives of the text must be brought together. Therefore, sufficient 19  See, eg, N Groeben, Leserpsychologie: Textverständnis—Textverständlichkeit (Münster, Aschendorff, 1982); W Kintsch, ‘The Role of Rhetorical Structure in Text Comprehension’ [1982] 74 Journal of Educational Psychology 828; I Langer, F Schulz von Thun and R Tausch, Sich verständlich ausdrücken, 9th edn (Munich, Reinhardt, 2011). 20  See European Commission (n 7) 211.

152 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

space and time must be provided for the authors and the reviewers to be able to talk about the text, identify problems, and together search for alternatives and improvements.21 At the same time, one must also bear in mind that too many cooks may spoil the broth. For this reason, clear decision-making and conflict-resolution procedures must equally be put in place. The stage of revision must be complemented with a validation of the new legislation— that is, with the text being submitted to scrutiny and testing to ensure that potential future cases actually lead to the desired outcome.

9.2.4. Editing The last phase of the drafting process is concerned with editing the text that has been developed by then. As we have already pointed out, editing serves a different function than revision. As it occurs at the very end of the drafting process, editing cannot induce any major changes to the text anymore; enhancing clarity can thus not be its primary goal. Editing must rather content itself with ensuring that the text is grammatically and orthographically correct, and that it adheres to the rules of formal legistics. The latter includes, for instance, the task of examining all cross-references that occur in the text and making sure that they point to the right place. In short, editing provides the formal and linguistic finishing necessary for a draft to become publishable.

9.2.5. Conclusions Legislative drafting—that is, planning, composing, revising and editing a legislative text— is an inherently co-operative and interdisciplinary process that requires various kinds of expertise: an understanding of the subject matter at hand, in-depth knowledge of the rules of substantial and formal legistics, and experience in plain-language writing. These kinds of expertise will rarely be united in a single person or group of people. And even if they were, the process of drafting a text would still have to be a process that provides sufficient time and space for the consultation of third parties who can review the text from the unbiased perspective of its primary addressees, for discussions of the draft in its various stages and, if necessary, for potentially extensive revisions. The main challenge in institutionalising the process of legislative drafting is thus to create an environment where, on the one hand, a recursive, discussion-oriented unfolding of the individual stages of text production is possible and all necessary perspectives are included and, on the other hand, clear decision-making procedures are in place so that the enterprise will eventually reach its goal. In what follows, we shall outline how this challenge has been approached across the civil law–common law divide.

21  See also S Thieme, ‘Recht verständlich? Recht verstehen? Möglichkeiten und Grenzen einer sprachlichen Optimierung von Gesetzen’ in KM Eichhoff-Cyrus and G Antos (eds), Verständlichkeit als Bürgerrecht? Die Rechts- und Verwaltungssprache in der öffentlichen Diskussion (Mannheim, Duden, 2008) 230–43, 240.

Legislative Drafting

 153

9.3.  Organisational Models and Their Actors The main organisational choices to be made when institutionalising the process of legislative drafting refer to how, and at which stage, the individual actors and their expertise are brought in and what role they are given. In particular, one needs to decide on (i) who composes the text, (ii) who reviews and revises it, at what point in the process, from which perspective and to what effect, and (iii) who is in charge of the text throughout the process and how conflicting views on the text’s development are to be resolved. (In multilingual legal systems, one will additionally have to consider at which point in the drafting process the text is to be translated into the remaining official languages. We will deal with this aspect separately, in section 9.4 below.) In light of the above questions, two major types of organisational models can be identified: models where drafting is a general, decentralised function and models where drafting is a specific, centralised function. In what follows, we will briefly describe the main characteristics of each type of model and discuss their respective strengths and weaknesses.

9.3.1. Decentralisation Where legislative drafting is a general, decentralised function, it is primarily carried out by people with expert knowledge of the subject matter concerned. Normally, the drafts are planned and composed by civil servants in the respective department of government; occasionally, if there is not sufficient expertise available within the administration, the task of composing a draft may be outsourced to external experts, such as scholars or advocates who have specialised in the respective domain of law. In either case, the drafts are composed by individuals for whom legislative drafting is not their main occupation; rather, drafting is just one task among others, and generally not one that they perform on a daily basis. Systems in which drafting is a general, decentralised function can often be found in civil law countries (eg in France, Germany, Sweden, Switzerland) and in the drafting of secondary legislation in European common law countries (eg in Cyprus, Ireland, Malta, the UK); it is also the model that has been adopted by the EU.22 The advantage of a model where drafting is a general, decentralised function is that it facilitates the exchange of information between those in charge of drafting the law and those applying it. Domain knowledge and practical experience can easily be accessed at all stages of the drafting process. As domain specialists are in charge of drafting, special emphasis is given to the applicability of the new legislation and its compatibility with current practice. The main disadvantage is that domain specialists are usually less familiar with the techniques of formal legistics and the principles of plain-language writing than professional drafters. They are also more prone to getting tangled up in matters of policy; as a result, general legal constraints may be neglected and the authors may fail to adopt the perspective of the target audience and thus infringe on the clarity of the law. One way to tackle this problem is to provide sufficient and regular training for those involved in legislative drafting. Such training must include learning how to develop a first 22 

See, eg, European Commission (n 7).

154 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

draft from the conceptualisation of some legislation, gaining a better understanding of the rules of formal legistics and acquiring techniques of plain-language writing that help improve the clarity of a text. However, such training can only be effective if the knowledge acquired is regularly put to use. It also does not offer sufficient guarantees that the legislative texts produced by the respective persons will meet the required qualitative standards. For this reason, most administrations provide guidelines detailing the rules of formal legistics and explaining linguistic pitfalls that must be avoided.23 These guidelines are meant to be available to the persons in charge of drafting independent of whether they have undergone any vocational training. However, such guidelines are easily disregarded by civil servants drafting a new text as they are preoccupied with the content and the policy issues surrounding it.24 In this model, it is therefore especially important that the texts are reviewed and revised by drafting specialists. In contrast to the initial composition of the text, this function is usually performed by a special, centralised unit of government. As we have already pointed out,25 revision requires legal as well as linguistic expertise; it is a fundamentally interdisciplinary task. In some systems (eg in the EU)26, these two types of expertise are united in a special category of professionals, sometimes called ‘lawyer-linguists’ (in French: ‘jurilinguistes’) or ‘legal revisors’ (in French: ‘juristes-réviseurs’). These professionals are lawyers who have undergone specialised training in formal and substantial legistics, as well as in the principles of plain-language writing. In other systems (eg in Germany,27 Sweden,28 Switzerland29), the two functions are carried out by separate individuals, lawyers and linguistics, who work as a team and revise the drafts together. This latter solution has the advantage that either aspect of good legislation—compliance with the rules of formal and substantial legistics or adherence to the principles of plain language—is given its own voice in the process of revision; experience shows that otherwise linguistic issues are often not given due consideration in light of potential legal problems arising from a draft.30 The advantages of having lawyers and linguists work in pairs are stronger when their positions in the revision process are equal: the linguists in particular must be given the same procedural possibilities to draw the authors’ attention to issues relating to the clarity of the text that the lawyers have to raise problems concerning formal or substantial legistics. In systems where drafts are composed by domain experts in the department of government that is in charge of the respective policy area, that department usually remains in control of the text throughout the whole of the drafting process. As a consequence, the effectiveness of any revision depends on the drafting specialists who review the text being

23 

See also Müller and Uhlmann (n 8) 65–66. also R Cormacain, ‘An Empirical Study of the Usefulness of Legislative Drafting Manuals’ [2013] The Theory and Practice of Legislation 205. 25  See section 9.2.3. 26  See, eg, W Robinson, ‘L’amélioration de la qualité de la législation européenne: La rôle des juristes-­réviseurs de la Commission’ [2007] LeGes 439; S Šarčević and C Robertson, ‘The Work of Lawyer-Linguists in the EU ­Institutions’ in A Borja Albi and F Prieto Ramos (eds), Legal Translation in Context: Professional Issues and P ­ rospects (Oxford, Lang, 2013) 181–202. 27  See Thieme (n 21). 28  See European Commission (n 7) 210–11. 29  See M Nussbaumer, ‘Der Verständlichkeit eine Anwältin! Die Redaktionskommission der schweizerischen Bundesverwaltung und ihre Arbeit an der Gesetzessprache’ in Eichhoff-Cyrus and Antos (n 21) 301–23. 30 ibid. 24  See

Legislative Drafting

 155

given sufficient time and space to make their case to the department and to argue for the changes they propose. It is also important that they are consulted as early as possible (and then continuously throughout the process): since the first draft is composed by individuals who are not equipped with specialised drafting skills, far-reaching alterations to the structure of the text may be necessary that can only be initiated if the text is still in its early stages of development.31 Common law countries generally discourage what is considered ‘lay drafting’.32 The view usually held is that a model where drafting is a general, decentralised function carried out by individuals not specifically trained for the task is a waste of resources as professional drafters come into the equation late. As a result, they either end up merely editing the text, which, as we have pointed out, is inadequate, or they discard the text and start afresh. However, experience from countries that have implemented such systems shows that these risks can be minimised if, as we have pointed out above, professional drafters are included early in the process.

9.3.2. Centralisation An alternative approach is offered by models where legislative drafting is a specific, centralised function. Here, policy makers in the individual departments of the administration, or in parliament, directly commission drafting specialists with planning and composing the text for the intended legislation. These drafting specialists usually belong to a central government agency that focuses on legislative drafting (sometimes referred to as ‘drafting service’ or ‘legislative counsel’) and drafting is consequently their main occupation. They have usually received specialised training for this task, either during their education or on the job. Systems in which drafting is a specific, centralised function are typically found in common law countries (eg in Australia, Canada,33 New Zealand, the UK, the US). This second model has the advantage that it guarantees that legislative texts are drafted by professionals who have the skills and experience that enable them to draw up high-quality legislative texts. Professional drafters will be more familiar with the rules of formal and substantial legistics, and with the techniques of plain-language writing, than civil servants working in other domains, and they will be less likely to get tangled up in policy issues. In this model, it is the authors of the text themselves who provide a first view from the outside: once they have been commissioned with drafting some particular legislation, they compose the respective text relatively independently of the policy makers. Their role is rather to convert a developed policy into a legislative text that is both clear and legally sound, that accurately reflects the intended policy and that is sufficient to carry it out.34 The main challenge for systems in which drafting is a specific, centralised function is to organise the communication between policy makers (organised as a so-called ‘Bill team’)

31 

See also section 9.2.3. See W Dale, ‘Canadian Draftsmanship, and the French Connection’ (1984) 10 Commonwealth Law Bulletin 1865, 1866. 33  See RC Bergeron, ‘La centralisation des fonctions rédactionnelles dans la filière législative fédérale au Canada’ [2003] LeGes 85. 34  See, eg, TA Dorsey, Drafter’s Deskbook: A Practical Guide (Alexandria, TheCapitol.Net, 2006) 8. 32 

156 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

and drafters. It is essential that the former provide the latter with detailed and sound instructions; drafters should not be expected to develop incomplete policy and to work out the details of how the policy is to be implemented. Installing a domain specialist who is familiar with the aims of the intended legislation, and who is senior enough to be able to take certain policy decisions, as an instructive officer can further facilitate communication between drafters and policy makers throughout the process.35 Furthermore, it is important that these drafts are thoroughly reviewed by policy makers and domain experts in order to make sure that the texts composed by the drafters are accurate with regard to the subject matter concerned and to initiate changes where this is not the case. In this model, the focus of revision is thus on the draft’s accuracy with regard to the intended content, whereas in the previous one, it was on its compliance with the rules of substantial and formal legistics and its adherence to the principles of plain-language writing. However, it would be wrong to assume that, as these texts are composed by professional drafters, they would not require any revision from the perspective of legistics and plain language anymore. On the contrary, it has been noted that the issue of plain-language writing, in particular, is prone to being jeopardised in this model: As most drafters join [the drafting service] early in their career and remain involved in its work for their whole working experience, they develop a specific style that some commentators consider removed from the day-to-day application and use of statutes in legal practice.36

It thus seems advisable, in this model too, that all drafts are also reviewed and revised by trained linguists who specifically focus on the clarity of the texts and their compliance with the principles of plain-language writing.

9.3.3. Examples The aforementioned general models of legislative drafting can, naturally, be implemented in manifold ways. We will briefly illustrate this variety of possible implementations by providing two specific examples: we will describe how the process of drafting primary legislation is organised in the UK (Westminster Bills) and in Switzerland (Federal Acts). The former is an example of a monolingual common law system with centralised drafting, the latter one of a multilingual civil law system with decentralised drafting.

9.3.3.1.  United Kingdom In the UK, the drafting of a new Westminster Bill is preceded by the consultation of interested parties both within the government (ie departments with an interest in the subject matter) and outwith (eg professional bodies, voluntary organisations, pressure groups).37 As there is no draft yet at this stage of the process, such consultation is based on a legislative proposal. Following the process of consultation, the government department in charge will put together a team of policy makers and legal advisors to co-ordinate the preparation of the bill and its passage through Parliament (the Bill team). This team 35  36  37 

See Xanthaki, Thornton’s Legislative Drafting (n 1) 142–44. European Commission (n 7) 43. cf Chapter 5.

Legislative Drafting

 157

will prepare instructions for the professional drafters at the Office of the Parliamentary Counsel (OPC), who are then commissioned with planning and composing the bill. The drafting instructions describe the current legal situation as well as the goals to be achieved with the new bill. The OPC is responsible for drafting bills that are clearly worded, accessible and effectively implement the intended policies.38 Usually, at least two drafters are assigned to a bill. These drafters will first analyse the instructions and try to clarify any questions they have with the policy makers in charge. They will then plan and compose a draft, and submit it for revision within the drafting team. The draft will also be sent to the sponsoring department for review. Policy officials and legal advisors in the Bill team will examine the draft and comment on it, and the drafters at the OPC will revise the text accordingly. This procedure will be repeated until the policy makers are satisfied with the resulting text; the final decision on the text is with them. Some provisions may make it necessary that the department or the drafters obtain consent from other government departments (eg the Ministry of Justice) or the devolved government in Wales, Scotland or Northern Ireland. As a last preparatory step, the Clerk of Legislation in the House of Commons or the Public Bill Office in the House of Lords will check whether the text adheres to certain specific rules of formal and substantial legistics. Once approved, the bill will be submitted to parliamentary procedure.39

9.3.3.2. Switzerland In Switzerland, a Federal Act is usually drafted by policy makers and legal advisors in the government office responsible for the policy area concerned.40 The drafts are then reviewed by other government offices with an interest in the subject matter, as well as by three centralised institutions tasked with guaranteeing the legal and linguistic quality of the texts: the Federal Office of Justice examines whether the texts comply with the constitution and any other superordinate law; the Federal Chancellery assesses the texts’ adherence to the rules of formal legistics; and the Internal Drafting Committee (IDC) of the Federal Administration makes sure that the texts are written in clear and understandable language. The IDC is an interdepartmental and interdisciplinary body, made up of language specialists from the Federal Chancellery and legal experts from the Federal Office of Justice (notably those who also examine the text’s legal quality), thus combining linguistic and legal knowledge. While the drafts may be planned and composed in just one language or in multiple languages, at the stage of revision, the German and French versions of the texts must both be available, and will be examined and revised in parallel.41 For these purposes, the IDC 38  cf Office of the Parliamentary Counsel, Drafting Guidance (London, 2014), available at https://www.gov.uk/ government/publications/drafting-bills-for-parliament. 39  See, eg, R Rogers and R Walters, How Parliament Works, 6th edn (London, Longman, 2006) 190–196; Cabinet Office, ‘Legislative Process: Taking a Bill Through Parliament’, available at www.gov.uk/legislative-processtaking-a-bill-through-parliament. 40  See Bundesamt für Justiz, Gesetzgebungsleitfaden: Leitfaden für die Ausarbeitung von Erlassen des Bundes, 3rd edn (Bern, 2007), available online at www.bj.admin.ch > Staat & Bürger > Legistik > Legistische Hauptinstrumente. 41  Switzerland has three official languages at its federal level: German, French and Italian. All Federal Acts are revised in both German and in French but, for practical reasons (there are considerably fewer speakers of Italian than there are speakers of German and French), are only translated into Italian once a definitive text has been established.

158 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

works in teams of four, consisting of a German-speaking and a French-speaking language specialist and a German-speaking and a French-speaking legal expert. These teams review the draft and propose to the office in charge the changes that they deem fit. In most cases, the texts are then revised in a collaborative effort, with the domain experts and the drafting specialists sitting at the same table. After this first round of revision, the drafts are sent out to the public (including the political parties and other stakeholders) for consultation.42 After the consultation process, the office in charge adapts the drafts where necessary and submits them to the IDC and the other aforementioned government units for a second round of revision.43 Once revision has been completed, the drafts are proofread and edited by translators and legal experts at the Federal Chancellery, who are charged with ensuring that all language versions are orthographically and grammatically correct and equivalent in form and meaning. At this point, the texts are also translated into Italian. Only then are the texts moved up to cabinet level for preliminary decision-making and submitted to parliament. At the end of the parliamentary procedures, the texts are checked by the Drafting Committee of the Federal Assembly, a standing parliamentary committee made up of select members of both chambers of parliament. The committee examines, for a last time, the formal and linguistic quality of the texts, and checks that all language versions are equivalent, before they are put before the chambers for the final vote.44

9.3.4. Conclusions One can argue that what distinguishes the two aforementioned types of organisational models at an abstract level is the fact that, to some extent, they assign inverse roles to domain experts and drafting specialists: one model has domain experts compose the texts and drafting specialists review them, the other has drafting specialists compose the texts and domain experts review them. In the former, legislative drafting is typically decentralised; in the latter, it is centralised. In reality, one will, of course, often find mixed models, where some stages of the process have been centralised but not all, as is the case with the centralisation of revision discussed in the context of the example of Switzerland. However, the individual stages that drafts need to go through—planning, composing, revising, editing—are the same regardless of the specifics of the model one chooses to implement. The example of Switzerland has also shown that in multilingual legal systems the situation is more complex, as another task needs to be added to the picture (translation) and an additional group of actors and their specific expertise must be brought into the drafting process (translators). In the following section, we discuss the different ways in which this can be done.

42 

cf Chapter 5. See S Höfler, Die verwaltungsinterne Verständlichkeitskontrolle im Rechtsetzungsverfahren des Bundes (thesis, University of Bern, 2015), available at http://dx.doi.org/10.5167/uzh-110371; Nussbaumer (n 29). 44  See S Steiner, ‘5. Abschnitt: Redaktionskommission (Art. 56–59)’ in M Graf, C Theler, M von Wyss (eds), Parlamentsrecht und Parlamentspraxis der Schweizerischen Bundesversammlung: Kommentar zum Parlamentsgesetz (ParlG) (Basel, Helbing Lichtenhahn, 2014) 465–89. 43 

Legislative Drafting

 159

9.4.  Models of Multilingual Legislative Drafting In multilingual legal systems, that is, in systems with more than one official language, the process of legislative drafting has to meet an additional goal: it needs to produce highquality legislative texts in several languages, and all language versions must be equivalent in both form and content.45 This additional requirement is owing to the fact that in multilingual legal systems a legislative text is considered equally authentic in each of the official languages—independent of whether a particular language version started off as the original or is merely a translation. At the end of the legislation process, all language versions have the same legal force; in theory, the specifics of the drafting process by which they have come about are not relevant anymore.46 At first glance, one might primarily notice that multilingual drafting means extra work, and that it bears the risk of resulting in different language versions of a legislative text that, while being equally binding, diverge either in form or in meaning. These observations may be true, but they do not mark the end of the story. Rather, the requirement to produce legislative texts in multiple language versions also offers an opportunity to incorporate an effective additional type of quality control into the drafting process: when legislative texts are translated into other languages, and when the different language versions of these texts are examined for formal and semantic equivalence, they are automatically subjected to particularly thorough linguistic scrutiny. As a consequence, translators frequently unearth ambiguities and inconsistencies that would otherwise have gone unnoticed. Furthermore, approaching legal norms from a multilingual perspective heightens awareness of the fact that there are always several different ways in which some piece of content can be phrased. In the ideal case, the different language versions enter a healthy competition, inspiring drafters to come up with wordings that are as clear as possible. As we shall see, such positive side effects are best exploited if the translation process is initiated early in the course of drafting some new legislation, when the ‘translation’ can still influence the ‘original’, and not just vice versa. The different models of legislative drafting in multilingual systems are therefore best distinguished by the stage at which multiple language versions of a text first come into play: whether this happens at the stages of planning and composition (co-drafting), during revision (co-revision) or only for the final editing of the text at the end of the process (co-editing). In what follows, we will briefly summarise the characteristics of each of these prototypical models.47

9.4.1. Co-drafting In a ‘co-drafting’ model, multiple language versions of a text are planned and composed in parallel. Even the conceptualisation of the new legislation may be developed in multiple 45  See, eg, R Bratschi and M Nussbaumer, ‘Mehrsprachige Gesetzgebung’ in E Felder and F Vogel, Handbuch Sprache und Recht, HSW (Berlin, De Gruyter, in press) on what it means for the different language versions of a legislative text to be ‘equivalent’. 46  Note, however, that multilingual international treaties sometimes contain a clause indicating that, in case of doubt, one specific language version takes precedence over the others, usually the one used during negotiations. 47  See Bratschi and Nussbaumer (n 45) for a more detailed discussion of the individual models.

160 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

languages simultaneously. Later, at the planning stage, the most important terminological issues can already be addressed from a multilingual perspective, thus pre-empting potential future translation problems. During composition, the individual norms are then also phrased more or less in parallel. This practice helps drafters avoid, right from the start, wordings that are hard to translate, and it continuously highlights potential discrepancies between the individual language versions. Co-drafting implies that the multilinguality of the law is fully represented in the drafting process: as several language versions are developed at the same time, none of them can be identified as the ‘original’ or as the ‘translation’. Rather, there is a continued crossfertilisation between the versions throughout the whole drafting process. The best documented case of a country that employs co-drafting is that of Canada.48 Co-drafting is to be distinguished from mixed-language composition. The latter denotes an approach whereby different parts of a text are initially written in different languages— for example, when multiple authors with different mother tongues contribute to the text. As opposed to what happens in co-drafting, in mixed-language composition, the individual norms are only ever cast in a single language, even though that language is not always the same throughout the text. The different languages may still influence each other to a certain degree as they occur in one and the same text. Fully parallel texts, however, will only emerge at a later stage, namely when the individual parts of the text are translated into the respective other languages and the initial mixed-language draft is converted into multiple monolingual versions of the text.

9.4.2. Co-revision In a ‘co-revision’ model, the texts are composed in just one language but, at the stage of revision, multiple language versions are made available and the different language versions are revised together. In this model, revising therefore includes not only the task of examining whether the texts exhibit the required formal, legal and linguistic properties, but also that of assessing whether the different language versions are equivalent in form and content. In addition, revisers usually aim to strip the texts of any wordings that identify them as translations. Since the different language versions are revised in parallel, there can be crossfertilisation not just from the original to the translations, but in all directions. As a result, all language versions must be considered ‘originals’ at the end of the process; none of them can be characterised as a mere translation anymore. Through co-revision, language versions that started off as mere translations thus become fully integrated in the drafting process. The process of co-revision anticipates, at a linguistic level, the fact that, from a legal point of view, all language versions will be considered equally authentic once the respective provisions will have come into force. An example of a model of co-revision that has been adopted is the drafting of primary legislation in Switzerland.49

48  See, eg, JC Gémar, ‘Translating vs Co-drafting Law in Multilingual Countries: Beyond the Canadian Odyssey’ in A Borja Albi and F Prieto Ramos, Legal Translation in Context: Professional Issues and Prospects (Oxford, Lang, 2013) 155–78. 49  See, eg, F Bertagnollo and C Laurent, ‘Unkraut vergeht nicht: La corédaction dans l’administration f­ édérale suisse’ in JC Gémar, Jean-Claude and N Kasirer (eds), Jurilinguistique: entre langues et droits—Jurilinguistics: Between

Legislative Drafting

 161

9.4.3. Co-editing In a ‘co-editing’ model, the text is only translated towards the end of the drafting process, when its content and form have more or less been agreed on. In this model, examining the formal and semantic equivalence of the original text and its translation(s) is conceived as part of the text’s editing. Co-editing cannot initiate a creative process anymore, but is usually limited to adjusting the translation to the original; only very minor changes to the original text are still possible at this final stage of the drafting process. In some cases, the text may even only be translated post hoc, that is, after it has been enacted. In reality, one will often find mixed models that combine elements of co-drafting or corevision with elements of co-editing or post-hoc translation. If three (eg in Switzerland) or more (eg in the EU) official languages are in play, co-drafting or co-revision procedures involving all of the languages will be infeasible. In such cases, the texts will be planned, composed and revised either monolingually (eg in the EU)50 or in only the two most important languages (eg primary legislation in Switzerland)51, while being translated into the other official languages at a later stage in the process.

9.4.4. Conclusions The positive side effects of multilingual drafting—the detection of ambiguities and inconsistencies that would otherwise have gone unnoticed and the cross-fertilisation between the languages—can be exploited better when the texts are translated earlier in the process. The earlier in the process the texts are translated, the more likely it is that the language communities that were not included in the preceding stages of drafting can still make an impact on the final product. The interests of linguistic minorities in particular are thus best represented in co-drafting and co-revision models. However, since multilingual drafting at any stage requires additional personnel with special linguistic training, co-drafting and co-revision are most likely to be encountered in countries where the processes of planning and composing (in the case of co-drafting) or revising (in the case of co-revision) have been centralised (eg in Canada and in Switzerland, respectively).

9.5. Summary Legislative drafting is first and foremost a process of writing a text—a very particular kind of text, but a text nonetheless—and as such it necessarily comprises the four stages that characterise any writing process: planning, composing, revising and editing. Just like any Law and Language (Brussels, Bruyland, 2005) 119–26; A Lötscher, ‘Multilingual Law Drafting in S­ witzerland’ in G Grewendorf and M Rathert (eds), Formal Linguistics and Law (Berlin, De Gruyter, 2009) 371–400. 50 See, eg, M Guggeis and W Robinson, ‘“Co-revision”: Legal-Linguistic Revision in the European Union “­ Co-decision” Process’ in CJW Baaij (ed), The Role of Legal Translation in Legal Harmonization (Alphen aan den Rijn, Kluwer, 2012) 51–81; Šarčević and Robertson (n 26). 51  See above, section 9.3.3.

162 

Stefan Höfler, Markus Nussbaumer and Helen Xanthaki

writing process, legislative drafting will also rarely be a straightforward, linear process. It will almost certainly exhibit some degree of recursivity, potentially requiring several rounds of planning, composing and, above all, revising. What distinguishes legislative drafting from other writing processes, however, is that it draws from a very specific set of skills: (i) expert knowledge of the subject matter at hand, (ii) a comprehensive understanding of the rules of substantial and formal legistics, and (iii) experience at plain-language writing and, as the case may be, legal translation. As these different kinds of skills are hardly ever united in one person, legislative drafting is an intrinsically collaborative, discussion-oriented enterprise that requires ample space and time for the actors involved to talk about the text-in-themaking, identify problems and together search for appropriate solutions. Legislative drafting is also characterised by its being embedded in the highly institutionalised and relatively constrained procedures of legislation, which, to a certain degree, runs counter to its somewhat unpredictable nature. The way in which the process of legislative drafting is organised within this broader setting is therefore crucial to the quality of the product: whether legislative drafting results in good legislative texts also depends on the conditions under which it takes place. The main challenge is to create an environment where, on the one hand, a recursive, discussion-oriented unfolding of the individual stages of text production is possible, with all relevant perspectives being included, and where, on the other hand, clear decision-making procedures are in place so that the enterprise will eventually achieve its goal. There is no single best way of institutionalising such a process. While, at an abstract level, two major types of models—models where drafting is a specific, centralised function and models where drafting is a general, decentralised function—can be identified in common law and civil law countries, in reality, manifold variations of these models have been implemented across different legal systems. The models mainly differ in the stages of the drafting process at which they involve the relevant actors (policy makers and domain experts, drafting specialists and language experts, translators), that is, whether the individual actors are brought in for the planning and composition of the text or only for its revision and editing. While the roles of policy makers and legal experts may be more evident in this context, the contribution that language specialists and translators can make to the quality of a legislative text—and also to its legal quality—must not be underestimated. Ideally, their role extends beyond that of mere editors. Their close reading of the text and their working on rephrasing provisions to make them more easily understandable and to facilitate their translation into other languages frequently unearths ambiguities and inconsistencies both in the text itself and in the conceptualisation on which it is based. Linguistic revision, whether monolingual or multilingual, can thus provide effective additional quality control, above all and crucially if it is initiated at an early stage of the drafting process and if it enters a constructive dialogue with the legal and policy-related revision of the text.

Legislative Drafting

 163

Further Reading Bergeron RC, ‘La centralisation des fonctions rédactionnelles dans la filière législative ­fédérale au Canada’ [2003] LeGes 85. Borja Albi, A and Prieto Ramos, F (eds), Legal Translation in Context: Professional Issues and Prospects (Oxford, Lang, 2013). Bratschi, R and Nussbaumer, M, ‘Mehrsprachige Gesetzgebung’ in E Felder and F Vogel (eds), Handbuch Sprache und Recht, HSW (Berlin, De Gruyter, in press). European Commission, Directorate-General for Translation (ed), Document Quality ­Control in Public Administrations and International Organisations, Studies on Translation and Multilingualism (Luxembourg, Publications Office of the EU, 2013). Guggeis, M and Robinson, W, ‘“Co-revision”: Legal-Linguistic Revision in the EU ­“Co-decision” Process’ in CJW Baaij (ed), The Role of Legal Translation in Legal Harmonization (Alphen aan den Rijn, Kluwer, 2012) 51–81. Lötscher, A, ‘Multilingual Law Drafting in Switzerland’ in G Grewendorf and M Rathert (eds), Formal Linguistics and Law (Berlin, De Gruyter, 2009) 371–400. Müller, G and Uhlmann, F, Elemente einer Rechtssetzungslehre, 3rd edn (Zurich, Schulthess, 2013). Nussbaumer, M, ‘Der Verständlichkeit eine Anwältin! Die Redaktionskommission der schweizerischen Bundesverwaltung und ihre Arbeit an der Gesetzessprache’ in KM Eichhoff-Cyrus and G Antos (eds), Verständlichkeit als Bürgerrecht? Die Rechts- und Verwaltungssprache in der öffentlichen Diskussion (Mannheim, Duden, 2008) 301–23. Pattchett, K, Law Drafting and Regulatory Management in Central and Eastern Europe, OECD Sigma Papers, 18 (Paris, 1997), available at www.oecd.org/officialdocuments. Robinson, W, ‘L’amélioration de la qualité de la législation européenne: La rôle des juristesréviseurs de la Commission’ [2007] LeGes 439. Seidman, A and Abeyesekere, N, Legislative Drafting for Democratic Change: A Manual for Drafters (The Hague, Kluwer, 2000). Thieme, S, ‘Recht verständlich? Recht verstehen? Möglichkeiten und Grenzen einer sprachlichen Optimierung von Gesetzen’ in KM Eichhoff-Cyrus and G Antos (eds), Verständlichkeit als Bürgerrecht? Die Rechts- und Verwaltungssprache in der öffentlichen Diskussion (Mannheim, Duden, 2008) 230–43. Thring, H, Practical Legislation: The Composition and Language of Acts of Parliament and Business Documents (London, John Murray, 1902). Uhlmann, F and Höfler, S (eds), Professional Legislative Drafters: Status, Roles, Education (Zurich, Dike, 2016). Xanthaki, H, Thornton’s Legislative Drafting, 5th edn (London, Bloomsbury, 2013). Xanthaki H, Drafting Legislation: Art and Technology of Rules for Regulation (Oxford, Hart Publishing, 2014).

164 

10 Publication KARL IRRESBERGER AND ANNA JASIAK

Publication is one of the essential requirements of legislation, allowing the people to discover what their rights and obligations are and to be able to control public power. Traditionally, official publication concerns printed copies of laws and regulations grouped into official journals or legal gazettes. However, electronic publication is on the rise. It is a simple, fast and inexpensive publication method. Its attractiveness has made some jurisdictions decide to abolish the official paper publication, electronic publication through the internet becoming the standard. Although electronic publication is a serious step towards improving the accessibility of law, it does not offer all solutions in this respect. Additional efforts to make the law more accessible have been taken up by the governments, consolidation practice being one of them.

10.1.  Publication as a Basic Requirement of Legislation That a law may be obeyed, it is necessary that it should be known: that it may be known, it is ­necessary that it be promulgated. (Jeremy Bentham)1

Law, and positive law (as the result of legislation) in particular, can be described as a set of rules meant (by a legislator) to steer the behaviour of individuals. It, therefore, can fulfil its task only to the extent individuals are aware of it. Consequently, it should be brought to the attention of individuals when it comes into existence and it should be accessible by individuals throughout the whole period of its existence. Secondly, awareness of both existing and new legislation is in the interest not only of the state but also of individuals, so that they can avoid harmful consequences of ignorance, and take benefit from knowledge of law. Thirdly, in a democratic society, those governing are chosen, and need to be controlled, by those governed. Publication serves public control of the legislature, which is a major branch of government. At least three aspects demand the publication of legislation: efficacy, legal security and democratic control. Accordingly, modern legal systems take up the stance that publication is essential for the creation of law. More specifically, legislation can acquire

1  J Bentham, ‘Essay on Promulgation of the Laws and Promulgation of the Reasons Thereof ’ in J Bowring (ed), The Works of Jeremy Bentham, Now First Collected (Edinburgh, Tait, 1838) vol I, 157.

166 

Karl Irresberger and Anna Jasiak

legal effect only through publication,2 and it is an obligation of public institutions to assure publication. If (or given that) legislation can acquire legal effect only through publication, publication is the final step of a legislative procedure. It is not just information about a change of a legal situation, but it is the prescribed, and therefore the only, way for that legal situation to come about. Information about what has been legislated may reach addressees in one way or another, but only official publication provides legislation with legal effect. Conversely, what has not reached a certain minimum publicity through publication may not be regarded as law at all. What is more, the question of whether such minimum publicity has been given to legislation will play a role in the assessment of the legality of the infringement of individuals’ human rights such as those under the European Convention of Human Rights. According to the European Court of Human Rights’ settled case law, interference by national authorities with the rights enshrined in the Convention must have a basis in domestic law.3 The Court thereby requires that such legal basis must be accessible for the persons concerned.4 This means that ‘the citizen must be able to have an indication that is adequate in the circumstances of the legal rules applicable to a given case’. This requirement will be met if the legislative acts, other official regulations and (judicial) decisions that are the basis of the limitation of the rights and freedoms from the Convention are officially published.5 Sufficient perceptibility by those concerned of the law in force is so important that it has been a feature of the constitutional system of all states governed by the rule of law.6

2  In many countries this is a constitutional requirement. For a few random examples, see Art 88 of the Dutch Constitution: ‘The publication and entry into force of Acts of Parliament shall be regulated by Act of Parliament. They shall not enter into force before they have been published’; and Art 72(4) of the Constitution of Malta: ‘(4) When a law has been assented to by the President it shall without delay be published in the Gazette and shall not come into operation until it has been so published, but Parliament may postpone the coming into operation of any such law and may make laws with retrospective effect.’ 3  However, the concept of law also embraces common law. Therefore, as Arai correctly observes, the test of accessibility does not require that every law be codified, Y Arai, ‘The System of Restrictions’ in P van Dijk et al (eds), Theory and Practice of the European Convention on Human Rights (Antwerp, Intersentia, 2006) 336–37. 4  The other quality requirement of ‘law’ is that it is foreseeable. For both requirements see European Court of Human Rights (ECtHR) Sunday Times v the United Kingdom, judgment of 26 April 1979, No 6538/74, paras 43–53. On the concept of quality of law in the case law of the ECtHR see A Jasiak, ‘Reasonable Legislation: A Matter of the Rule of Law and Human Rights’ in M van Roosmalen et al (eds), Fundamental Rights and Principles, Liber Amicorum Pieter van Dijk (Cambridge, Intersentia, 2013) 481–96; for procedural quality in particular see P Popelier, ‘The Court as Regulatory Watchdog. The Procedural Approach in the Case Law of the European Court of Human Rights’ in P Popelier et al (eds), The Role of Constitutional Courts in Multilevel Governance (Cambridge, Intersentia, 2013) 249–67. 5  See, eg, Piroğlu and Karakaya v Turkey, judgment of 18 March 2008, No 36370/02 and 37581/02, para 53. See also recently the case of Kasymakhunov and Saybatalov v Russia, judgment of 14 March 2013, Nos 26261/05 and 26377/06, para 92 (‘Only a publication emanating from an official source can give an adequate and reliable indication of the legal rules applicable in a given case’) and Korbely v Hungary, judgment of 19 September 2008, No 9174/02, paras 74–75 (when a conviction is based solely on an international law, that law must be accessible, which means published, at the domestic level). 6  Common law jurisdictions, however, traditionally deny a legal requirement of publication (of parliamentary acts). In the UK, legal doctrine adopted the view that because the population is represented by the members of parliament the knowledge of the representatives can be counted as that of those represented.

Publication

 167

10.2.  Publication Methods There are various ways of bringing (statutory) law to the attention of those to whom it is addressed. Early famous examples of publication of legislation are Hammurabi’s Code and the Roman Law of the Twelve Tables, but printed paper later became the standard. Methods and rules for publication have been grouped by the terms ‘substantial publication’ and ‘formal publication’,7 the former describing an effort to effectively bring the legislation to the knowledge of persons concerned and the latter referring to an action creating the minimum publicity (limited as it may be) that is legally sufficient for a legislative act to enter into force. For obvious reasons, bringing the content of the original document containing the ­legislative act to the knowledge of (a large number of) addressees implies the making of copies thereof, those (paper) copies playing the key role in the process of publication. In early times, but still in the eighteenth century, these were meant to be read out—given the illiteracy of the general population—in public places, sometimes in churches (where virtually the entire population could be addressed), or were designed to be posted in public places, so that pedestrians (those who were able to read) could read them. Publication was not considered to be completed before the public reading or displaying, respectively. Consequently, it was not the distribution of papers, but the posting or oral proclamation that was publication proper (though not always in the strict legal sense, ie as a legal requirement). Later on, the notion of publication shifted to the mere release of printed copies, in particular those of the issues of official gazettes designed as publication media. Oral announcement has been reduced to an emergency form of publication, while its range has been extended through the means of broadcasting. In the course of their evolution, legal systems have moved towards ‘formal publication’, using a variety of means. Nowadays electronic publication is on the rise. An increasing number of jurisdictions have chosen the internet as a forum for the publication of legislative acts, thereby making publication less ‘formal’ and more ‘substantial’.

10.2.1.  Paper Publication One standard publication method is that paper copies of legal acts are printed and spread throughout the country, more specifically being distributed to public offices and subscribers, including libraries and retailers, so that ultimately everybody has a chance to take note of the content. In the course of the nineteenth century, it became a general standard that publication in a specific periodical was required to give binding effect to a legislative act.

7  ‘Materielle Publikation’ and ‘formelle Publikation’ in German language sources; terms introduced by J Lukas, Über die Gesetzes-Publikation in Österreich und dem Deutschen Reiche (Graz, Leuschner & Lubensky, 1903), 7ff.

168 

Karl Irresberger and Anna Jasiak

As Wiederin8 pointed out, a proper organ of publication of legislation has to fulfil the following requirements: —— Its use must be obligatory, so that legally binding effect cannot be achieved otherwise. —— It must be in a monopoly position, so that keeping track of it is sufficient to not overlook any relevant provision. —— It must contain all legislative acts and provide for each of them to be found as quickly and conveniently as possible. —— It must be issued to everybody at the same time. —— Addressees must be in a position to rely on the authenticity of the organ in question. Choosing a specific periodical for the publication of legislation implies making further choices about its content. There are different options: such periodicals can be limited to the publication of legislative acts only; they can also contain official announcements of any kind; or they can carry legislative acts along with other information relevant for a certain public. It may even be a newspaper (of a somewhat official character) that carries both pieces of journalism (news of the day, reports and comments) and official announcements, including new legislation. Given the two main goals of publication, namely bringing new legislation to the attention of those concerned and forming a collection of ongoing legislation which can be easily looked up even decades later, there seems to be a tendency to produce a collection from the start rather than combining the publication of legislation with other information. In historical terms, the ‘bulletin des lois’ created in 1791 in revolutionary France is the common ancestor of a large family of official gazettes or law gazettes that have become the predominant type of medium for the publication of legislation in continental Europe and beyond. It also paved the way for what has become the legal standard in Europe and beyond: that publication in a specialised periodical is the ordinary way of giving legal effect to new legislative acts. Today, two main types of periodicals designed for the publication of legislation can be distinguished: those which publish legislative acts among other official announcements and those which are devoted to the publication of legislative acts (more or less) exclusively. It has been suggested9 that the former kind be grouped as ‘official journals’ and the latter as ‘legal gazettes’, which leave the non-legislative official content to ‘official gazettes’. A 2009 survey covering the jurisdictions of the European Union (EU) and its Member States, the European Economic Area (EEA) and the member states of the European Free Trade Association (EFTA) counts 17 single official journals, 13 separate legal gazettes and three official ‘collections of legislation’ (UK, Ireland, Sweden). Given the increasing number and size of legislative acts, legal gazettes may be divided into several series, allowing for different subscription options. For example, the German Federal Law Gazette (Bundesgesetzblatt) has two series (for domestic and for interna8  E Wiederin, ‘Kundmachung im Internet’ (2004) 2 Forum Parlament 43, 43–44, available at www.parlament. gv.at/ZUSD/PDF/forumParl2-2004.pdf; E Wiederin, ‘Die Kundmachung von Rechtsvorschriften im Internet’ in M Gruber (ed), Die rechtliche Dimension des Internet (Vienna, Manz, 2001) 25, 32–40. 9  For the following, see chiefly A Berger (ed), Access to Legislation in Europe (Luxembourg, Publications Office of the European Union, 2009) 11ff; a similar inquiry, also covering the EU, the EFTA and states of South-Eastern Europe, was conducted by the Publication Office in 2014/15, the results of which have not been published but were available for the purposes of the present work.

Publication

 169

tional law), and the Austrian and Icelandic law gazettes have three (for parliamentary laws, secondary legislation and international law). Official journals, which publish many official announcements beyond legislative acts, have an even greater need to be divided into various series than legal gazettes. A well-known example is the Official Journal of the European Union, which is divided (chiefly) into the series L (legislation) and C (communications). As another example, the Italian Gazzetta Ufficiale della Repubblica Italiana falls into two parts:10 Part I comprises the General Series and five Special Series devoted to (1) the Constitutional Court, (2) the European Union, (3) the Regions, (4) Contests and Examinations and (5) Public Contracts, whereas Part II contains other notices. It can be assumed that, by joining other current announcements and information, new legislation, when released, will attract the attention of a wider public than a gazette devoted exclusively to new legislation, though later it will be more difficult to retrieve it. A legal gazette, on the other hand, will appeal chiefly to lawyers and officials, and will thus be less likely to reach a broader public. In the long run, both a legal gazette and a ‘legislation series’ of an official journal, by virtue of their nature, form a chronological collection of legislation (formerly or still in force) which allows for easy access to the sources of legislation. However, it is characteristic of legal gazettes (but not necessarily of legislation series of an official journal; see the Official Journal of the European Union) that the privilege to invest acts with the force of law is reserved to them. In the case of paper publication, publication is not a single act but a process the core of which is bringing a sufficient number of copies into the reach of the intended readers. This process implies transportation over some distance—usually involving the public postal s­ ervice—so that the process of publication, inasmuch as the possibility to take note of what is published is considered as essential (hardly from a legal but from a principal point of view), is completed at different places and at different times (which cannot always be predicted). Given the importance of publication as the final step of producing legislation, it is essential to know what in the legal sense is to be considered as the completion of the publication process. Further importance is given to this question by the legislation itself, which, either in general rules or in the respective act, refers to the publication as the starting point of a delay (calculated by days, weeks, months) after the expiry of which a published act takes legal effect (vacatio legis). To meet such need, the date of the publication is indicated on each issue of any official gazette. Of course, the date to be imprinted has to be determined before the start of printing, and what needs to happen on this day must be clear to the print and publication offices. For example, at the time when only the printed version of the Official Journal of the European Union was considered authentic, one copy of each language had to be displayed in the entrance hall of the Publications Office (the display was registered and the delivery sheet of the display copies was timestamped) with a view to proving that the printed date corresponded with the availability of the Official Journal on its publication date.11 10  Decreto del Presidente della Repubblica 28 dicembre 1985, No 1092 (Approvazione del testo unico delle disposizioni sulla promulgazione delle leggi, sulla emanazione dei decreti del Presidente della Repubblica e sulle pubblicazioni ufficiali della Repubblica italiana), GU No 123 del 29-5-1986—Suppl Ordinario. 11  The arrival (to be registered) from the printing house was instituted as the event relevant for the legal effect of publication (for the French ‘Bulletin des lois’) by the Ordonnance du Roi du 27 novembre 1816 concernant la promulgation des lois et des ordonnances.

170 

Karl Irresberger and Anna Jasiak

What is considered as the publication date depends on what is considered as the legally decisive step in the course of the production and dissemination process; this varies from country to country and may change over time. In Germany, it is the time when the issue of the legal gazette is ‘released’12 that is relevant. ‘Release’ used to be understood to mean delivery to the central post office from which the copies would be dispatched.13 At present, in the light of criticisms voiced by legal science and case law developed by the Federal Constitutional Court, the day on which the issue will probably be delivered to the majority of subscribers, which is normally the day after the dispatch, is considered to be the relevant one, so the publication date imprinted on the issues (which is vested with a presumption of correctness)14 of the Federal Law Gazette has been moved by one day.15 Obviously it is contrary to the underlying principle of publication to give immediate legal effect to an event which actually marks only the beginning of a process the end of which—the chance for individuals to gather the information—is crucial. This was recognised both by the German court decisions referred to above and by the legislators, who provided for some delay between ‘formal’ publication and entry into force. There are a number of exceptions to the principle that legislation has to be published in a specific periodical, which are mostly due to practical considerations. One source of law concerned is international treaties (whether or not they are given immediate legal effect in national law), and especially language versions thereof which do not use an official language of the respective state; these may be relegated to some other—less far-reaching but cost-effective—means of publication or be exempted from any publication requirement.

10.2.2.  Electronic Publication Electronic publication—more specifically, online publication via the internet—can be seen as a simple, fast and inexpensive way of informing the people about new law, and is thus growing throughout Europe. Initiated by Norway in 2001, the process passed a milestone when the electronic version of the Official Journal of the European Union was made legally binding in mid-2013.16 While the shift from printing to electronic publishing was a revolutionary move, the traditional features of publication remained relatively unchanged. The electronic publication media tend to be exact copies of their paper predecessors, edited as ‘gazettes’ and ‘journals’. This is almost inevitable where printed editions of the same gazettes and journals

12 

In German: ‘ausgegeben’, which may be rendered as ‘given out’ as well. This was ruled, for instance, in a 1922 court case in which an act punishable under the newly issued legislation was committed two hours after the edition had been dispatched at the central post office of the capital. In Austria before the introduction of authentic electronic publication of the Federal Law Gazette, the legal effect of publication was linked to the release and dispatch (‘ausgegeben und versendet’) of the relevant edition of the Federal Law Gazette; consequently it was deemed essential that the copies of the Federal Law Gazette be dispatched at a post office in the capital on the day that the copies themselves indicated was the publication date. 14  BVerfGE 16, 6. 15  BVerfGE 81, 10 (83). cf H Schneider, Gesetzgebung, 3rd edn (Heidelberg, Müller, 2002) 289–90 (No 485ff); H Bauer in H Dreier, Grundgesetz, Kommentar, 3rd edn, Vol II (Tübingen, Mohr Siebeck, 2009), Art 82, No 18. 16  Council Regulation (EU) No 216/2013 of 7 March 2013 on the electronic publication of the Official Journal of the European Union, [2013] OJ L69/1. According to the 2014/15 survey mentioned above (n 9), 22 out of 40 countries already had an official gazette legally binding in its online format. 13 

Publication

 171

continue to be published; what has shifted—though not always entirely—is the legal effect that accompanied the paper publication, which is now attributed to the electronic version. Paper and online editions of the same official gazette do not always contain the same documents. For example, in France, several groups of ordinances which are of lesser interest for the general public were excluded from the paper version (ordinances concerning the internal organisation, members of the public service, etc) when both versions were made authentic.17 In some jurisdictions, both the paper and online editions of the same official gazette have legal value (eg Greece, Italy, Serbia, Slovakia); for a number of jurisdictions, this proved to be just a transitory solution on the way to exclusive authenticity of the electronic version (Estonia, 2002–06; Slovenia, 2006–10; Poland, 2006–11; Switzerland, 2004–15;18 France, 2004–15). Where both editions have legal value, the question arises which of the versions shall prevail in case of differences between them. Slovenia, for instance, gave precedence to the electronic version, Switzerland to the paper version. An increasing number of jurisdictions give legal effect to the electronic version exclusively (eg EU, Austria,19 Denmark, Finland, France,20 Hungary, Switzerland), save (in the case of the EU) for the exceptional disruption of the electronic publication, which event would confer legal effect to the printed edition of the official journal. Some countries, such as Estonia, France, the Netherlands, Poland, Portugal and Slovenia, have decided to abolish the official paper publication of the official gazette—sometimes after the equally valid paper version had suffered a substantial decline in circulation numbers.21 Instead, they have chosen electronic publication as the only manner of official publication of legislation. A number of the states that have opted for authentic online publication have ensured that full accessibility of legislation is available for those who have no access to the internet. For example, in Estonia and Spain, certain institutions, such as local governments or public libraries, are required to offer free access to the online legislation. Another option would be to designate locations where individuals can obtain paper copies of the requested official publication. This is the case, for example, in the Netherlands, where such copies can be collected at one spot in the Hague after payment of the production costs.22 In Belgium, the legislature has agreed to establish a hotline that people can call free of charge when they need help in finding some legal texts, after the Constitutional Court (Grondwettelijk Hof) quashed the legal provisions which provided for electronic publication as the sole manner

17  Décret No 2004–617 du 29 juin 2004 relatif aux modalités et effets de la publication sous forme électronique de certains actes administratifs au Journal officiel de la République française, JORF No 150 du 30 juin 2004, 11849, Arts 1–2. 18  Bundesgesetz über die Sammlungen des Bundesrechts und das Bundesblatt (Publikationsgesetz, PublG), Änderung vom 26 September 2014; Verordnung über die Sammlungen des Bundesrechts und das Bundesblatt (Publikationsverordnung, PublV) vom 7 Oktober 2015. 19  Bundesgesetzblattgesetz (Federal Law Gazette Act), BGBl I Nr 100/2003. 20  Publication in the Journal officiel de la République française is governed by Arts L221-9ff of the Ordonnance No 2015-1341 du 23 octobre 2015 relative aux dispositions législatives du code des relations entre le public et l’administration, JORF No 0248 du 25 octobre 2015. 21  In France, for example, the number of subscribers declined from 33,500 to less than 2,300 between 2004 and 2015, see www.lcp.fr/actualites/politique/176372-la-commission-des-lois-vote-la-suppression-de-la-versionpapier-du-journal-officiel (4 November 2015). 22  See Art 10 of the Official Publication Act (Bekendmakingswet) and Art 4 of the Accompanying ministerial regulation (Bekendmakingsregeling).

172 

Karl Irresberger and Anna Jasiak

of official publication of the official gazette (Belgisch Staatsblad) in view of equality and non-discrimination principles.23 They can also order a copy via the hotline and, as in the Netherlands, pay just the actual production costs, a solution which has been accepted by the Court. In the literature, the question has been raised whether this measure and its further, rather cumbersome, implementation indeed fully addresses the needs of those persons in Belgium who still do not have access to the internet. It has been suggested that enabling persons to consult the electronic version of the official gazette free of charge, for example in public libraries, would better serve their interests in getting to know the rules that are applicable in their case in a fast and easy way.24 While paper publication relies on printing and dissemination, and after that on continuous availability of a large number of paper copies which will survive a century without the need for considerable precautions, it is different with electronic publication. Whether addressees have sufficient access to such a system is only the first concern, which can be sidelined when the extension of internet access and its widespread use for everyday purposes is compared with the need to buy paper copies or resort to libraries and other institutions holding such copies. Online publication relies on a complex system which requires the continuous existence and functioning of certain facilities on the part of both the supplier (the legislator) and the consumer of the information (the addressees) and of a certain infrastructure. One more important difference between paper and online publication is that, while the core of paper publication is the dissemination of a number of hardcopies throughout the country, online publication means that the state (or the responsible public body) holds files representing legislative acts which can be accessed at (almost) any time via the internet. When shifting to electronic publication and basing legal systems on complex, inherently unstable and vulnerable structures, legislators are well advised to take (legal) precautions to secure documents holding the authentic electronic versions of legislative acts. Among the security attributes relevant to electronic documents used for publication are:25 —— integrity, that is, freedom from unauthorised (whether deliberate or accidental) modification of data; —— authenticity, meaning, with regard to records, that a record can be proven to be what it purports to be—to have been created or sent by the person purported and at the time purported; —— reliability, designating records whose contents can be trusted as a full and accurate representation of the transactions, activities or facts to which they attest and can be depended upon in the course of subsequent transactions or activities; and —— availability, meaning timely access to information.

23  This decision has been criticised for overestimating the negative effects of the replacement of the official publication on paper by the electronic official publication. 24  See J van Nieuwenhove, ‘Elektronische bekendmaking en beschikbaarstelling van geconsolideerde wetgeving in elektronische vorm in België’ (2008) 23 RegelMaat 209. 25 A Hietanen ‘Electronic Publishing of Legislation—Towards Authenticity’, Final Report of the Working Group, European Forum of Official Gazettes, 4th Meeting, Helsinki–Tallinn, 14–15 June 2007, 4ff, referring, inter alia, to the ISO standard on record management (ISO 15489).

Publication

 173

Legal precautions taken by jurisdictions that have opted for authentic electronic publication include the following: —— Documents shall be made available in a non-obsolete format (eg in the EU), or must have a format that is upwardly compatible (Austria). —— Documents must have been produced in a reliable process (Austria). —— Documents shall bear an electronic signature (Austria), in particular an ‘advanced electronic signature’ (EU, Hungary, Spain), and a timestamp (Hungary). —— Subsequent modification or deletion is prohibited (Austria, Hungary). —— Back-up copies shall be held (Austria, Portugal), or (certified) paper copies must be archived (Estonia) with different institutions (Austria, Belgium, Hungary, Poland, Portugal, Spain). Sometimes such precautions are instituted by legislatures in more or less general terms, with emphasis on the aspects of authenticity and integrity.26 Naturally, when instituting authentic online publication, legislators should consider possible disruptions of the publication system, be it for reasons on the part of the publishing institutions or because of a dysfunction of the internet. The most obvious solution is a temporary return to authentic paper publication (EU,27 Spain). Where online publication provides the legally binding version of legislation, the time when the document is made accessible via the internet is legally relevant; in particular, it marks the entry into force or the beginning of the delay fixed by legislation. Therefore, publication laws require the day of the publication to be published along with the legislative act in question, which in some jurisdictions (Portugal) must also be entered into a separate register. As the nature of online publication is not (active) transmission by the publisher of copies but just granting access to files held by the publisher, it is important that access be granted more or less ‘at any time’ (Austria), for an unlimited period of time (eg EU). This is a requirement stipulated regularly by publication laws. It is not, however, clear from the various national legislations what might be considered the minimum duration of accessibility that fulfils the requirement of publication and what degree of failure to give (permanent) access to documents endowed with integrity and authenticity might damage the legal validity of documents once properly published. It is also worth mentioning that almost all jurisdictions that have opted for electronic (online) publication give access free of charge— which is usually explicitly stated in the publication laws—and free of possible administrative barriers (eg registration). The internet has a number of impacts on the publication of legislation. Not only has the legislation of most jurisdictions been made accessible online, and not only have an increasing number of jurisdictions opted for authentic electronic publication on the internet, but also features that have been characteristic of gazettes are likely to be dropped in the case 26  cf France: ‘dans des conditions de nature à garantir leur authenticité’, ‘sous forme électronique de manière permanente et gratuite’; Switzerland: ‘Le Conseil fédéral arrête les mesures garantissant l’authenticité, l’intégrité et la conservation des textes publiés sur la plate-forme ainsi que le bon fonctionnement de celle-ci, en tenant compte de l’état de la technique’; Spain: Real Decreto 181/2008, esp Art 12. 27  See Council Regulation (EU) No 216/2013 which provides, inter alia, that an electronic version of that exclusively authentic paper edition shall be published later for information purposes only and shall contain a notice to that effect.

174 

Karl Irresberger and Anna Jasiak

of electronic publication, such as gathering several new legislative acts into a single issue (which in the digital world makes no sense). Issues of gazettes are less likely to be bound into books that would be provided with annual indexes and arranged on shelves, and without books there is no need for annual continuous pagination. Without annual continuous pagination, citation styles based thereon are obsolete. The true destination of digital versions of legislative acts is to be part of a database, so the publication of legislative acts will be adapted to the needs and benefits of databases. One advantage of electronic publication is that costs are low, so whereas the size of a document, the frequency of changes or a very technical content may have deterred paper publication, these are not serious obstacles to electronic publication. Authorities making secondary legislation do not necessarily just edit gazettes, but regularly operate websites offering various other sorts of information. In case an entirely centralised publication system is lacking, they may succeed in making their homepage a place of authentic electronic publication of their own acts. However, such self-made electronic publication is likely not to fulfil the same quality standards as publication in official gazettes in terms of guarantees of authenticity, long-term accessibility or even detectability (not to mention legal and drafting quality, on which the mere requirement of publication in the general official gazette may have favourable effects). It seems recommendable for jurisdictions not to disregard such demands but to specify legal criteria for all cases of permissible electronic publication.

10.2.3.  Publication Requirements in Case of Legislation by Reference It is not unusual for legislation (parliamentary legislation, but also subordinate legislation) to make references to other sources, whereby it just names them instead of providing full citations. This old legislative technique is attractive, because it is an effective means to provide for harmonisation between various acts, concepts and standards, and therefore consistency within the body of legislation. At the same time, it is attractive because it allows a reduction in the volume of legislative texts.28 References can be made to various documents (cross-references in the same statute, references to other statutes and/or regulations, international treaties, international agreements, codes, handbooks, international and national technical standards) and can have various functions and effects.29 Despite the attractiveness of this technique, there are also disadvantages, which means that it should be used with prudence.30 Most importantly, legislation by reference raises questions from the rule-of-law perspective: the principle of legal certainty, and with it the accessibility and foreseeability of laws. Particular attention should be paid to the constitutional requirements for the publication of legislation. Which

28  This is attractive especially nowadays, given the latest trend in the field of legislative policy, which is ‘fewer regulations!’. 29  See F Scott Boyd, ‘Looking Glass Law: Legislation by Reference in the States’ (2008) 68 Louisiana Law Review 1204. His study of legislation by reference is based on the American doctrine, but the general notions apply also to the European legislation. 30  At the EU level see also the Interinstitutional Agreement of 22 December 1998 on common guidelines for the quality of drafting of Community legislation (1999/C 73/01), Art 16: ‘References to other acts should be kept to a minimum’.

Publication

 175

standards concerning the publication of legislation should apply to documents to which references are made?31 In many cases of incorporation by reference, the referenced material will concern another piece of legislation which has already been published in conformity with constitutional or legislative publication requirements. The accessibility and publication of referenced documents in such cases is not problematic. The same can be said for references made to international treaties or agreements, since national constitutions or legislation on publication normally provide rules for approval, ratification and publication of such material before it can be legally binding. As far as references to legally binding resolutions, decisions etc of organs or institutions which are constructed by international treaties are concerned (secondary law), the treaties themselves may indicate the manner in which these documents should be made public. The question of accessibility and legal effect through publication becomes more important, however, when references are made to documents which do not originate from any public regulatory authority, meaning that they are not officially published and therefore are not legally binding. This is the case, for example, with references to technical standards (normalisation), which have been developed by either international or national private bodies and which are often subject to copyrights, meaning that they cannot be obtained without payment. At the European level, the 1998 Institutional Agreement on common guidelines for the quality of Community legislation recommends that a reference made in the enacting terms of a binding act to a non-binding act shall not have the effect of making the latter binding. Should the drafters wish to render binding the whole or part of the content of the non-binding act, its terms should as far as possible be set forth as part of the binding act.

If references to such standards are just informative and are not meant to bind individuals, the question of publication and accessibility of such standards is less important. However, even if legislation does not make any compulsory reference to these standards but presents them as one of the ways for individuals and legal entities to comply with the relevant legal provisions (‘acceptable means of compliance’), there must be a real alternative for the public to comply with the law. In many cases, such an alternative can be illusory, if the standards are still the point of reference: one can only know whether the alternative way leads to comparable results to the standards to which the legislation refers if one has the knowledge of these standards. And they cost money. In Germany, the Bundesgerichtshof, as confirmed by the Bundesverfassungsgericht, decided that even though the standards were officially not binding, they formed a basis for the assessment by the civil servant deciding whether the building licence in a specific case should be granted. Therefore the standards became the own standards of the authorities who referred to them.32 In consequence, they cannot be subject to copyright. In reaction to these judgments, the federal legislature amended the copyright law by adopting a provision stating that a reference to private norms does not mean that the copyright no longer exists. However, the authors of these private norms

31  Disadvantages accompanying this legislative technique are clearly listed by JM Keyes, ‘Incorporation by Reference in Legislation’ in (2004) 25 Statute Law Review 194. 32  See 1 BvR 1143/90, §§ 12–13.

176 

Karl Irresberger and Anna Jasiak

must, under reasonable conditions, allow publishers to publish them.33 In Austria, the state and the two major standardisation institutes agreed that the state will publish the standards to which obligatory references in legislation are made in the official state journal, following the payment of an annual contribution.34 In the US, a more radical approach has been taken. If the standards, even if it is by reference, have been adopted as law, they must be freely accessible, which means that no copyrights apply. This was decided in Veeck v SBCCI (Southern Building Code Congress International), where the court gave consideration to the fact that such (building) standards were developed by the SBCCI ‘with the sole motive and purpose of creating codes that will become obligatory in law’.35 In other cases, a single reference to a copyrighted text does not mean that the text becomes public and therefore loses its copyright protection: If a statute refers to the Red Book or to specific school books, the law requires citizens to consult or use a copyrighted work in the process of fulfilling their obligations. The copyrighted works do not ‘become law’ merely because a statute refers to them.36

In the Netherlands, in a similar case, the Dutch Supreme Court came to an opposite decision. In Knooble v State and NNI (Netherlands Standardisation Institute),37 the Court decided that a reference to normalisation norms does not transform them into public law norms. Therefore, they do not need to be published like laws and do not lose their copyright protection. The above examples show that there is not one universal answer to the question of publication of technical standards being referred to and to that of copyrights.38 Nevertheless, the crucial question seems to be whether such standards have become a part of the law through the reference. Ideally, from the perspective of accessibility of such standards, they are freely accessible if they are a compulsory part of the law39 and are meant to

33  § 5(3) of the Urheberrechtsgesetz. See also LM Zentner, ‘Die Ausnahme von Urheberrechtsschutz für amtliche Werke’ (2009) 1 ZGE/IPJ 94. 34  K Irresberger, ‘Die Publikation von Rechtsvorschriften in Österreich’ [2013] LeGes 144. Meanwhile new legislation (Normengesetz 2016, BGBl I 153/2015, §§ 9 and 15) allows for certain standards of ‘purely domestic origin’ to be made obligatory and published freely while granting an annual contribution to that specific standardisation institute; this followed a decision by the Austrian Constitutional Court (G 104/2013 of 10 December 2014) ruling that a technical standard which has been made obligatory by legislation is not subject to copyright. 35  Veeck v SBCCI 293 F.3d 791 (5th Cir 2002), cert denied, US 123 SCt 2636, § 45 with references. 36 ibid. 37  For a description of the case see K Stuurman, ‘Public Access to Standards: Some Fundamental Issues and Recent Developments’ in L Mommers et al, Het binnenste buiten, Liber Amicorum Aernout HJ Schmidt (Leiden, eLaw@Leiden, 2010) 408, available at openaccess.leidenuniv.nl/handle/1887/15206. For the analysis of the entire case up to the proceedings before the Dutch Supreme Court see RAJ van Gestel, ‘Hoge Raad is duur. Over het verwijzen naar normalisatienormen in wetgeving’ (2012) 27 RegelMaat 249. See also RAJ van Gestel, ‘Primacy of the European Legislature? Delegated Rule-Making and the Decline of the “Transmission Belt” Theory’ (2014) 2 Theory and Practice of Legislation 51. 38  See also a country report from 2000: H Schepel and J Falke, ‘Legal Aspects of Standardization in the Member States of the EC and EFTA’ in Comparative Report (Luxembourg, Office for Official Publications of the European Communities, 2000) vol 1. 39  Uhlmann makes, in this respect, a refined distinction between two questions: (i) Did the standards became a part of the general state law through the reference? and (ii) Are the standards only used in practice by the administrative authorities in order to decide on the compliance with the law (‘blosse Konkretisierung staatlichen Rechts’)? He admits, however, that this distinction will often not be easy to make: F Uhlmann, ‘Die Normen können bei … bezogen werden. Gedanken zur Publikation und Verbindlichkeit privater Normen’ [2013] LeGes 89, available at www.zora.uzh.ch/87484.

Publication

 177

bind individuals, whether de jure or de facto. Such free accessibility can be achieved by means of, for example, state publication in the official journal (which, however, does not contribute to the reduction of the volume of legislation), on the website of a standardisation institute or at another official deposit point. This last possibility would particularly apply to such standards because of their technical character—a possibility which, as will be discussed in the next section, some countries explicitly provide for.

10.2.4.  Deposit for Public Inspection Instead of Official Publication As a general rule, official public law regulations which are legally binding upon individuals or legal entities are integrally published in the official publication source (mostly in the official gazette). However, instead of full publication, documents may exceptionally be placed on deposit for public inspection. This method may apply to an entire legislative act or to an external source which is explicitly awarded a legal binding effect by way of the reference made to it in a legislative act. Obviously this publication method opens only an arduous way to the legal content. It can therefore be deemed a sufficient way of publication only in limited circumstances, and only if the fact of its deposition for public inspection is made public by ordinary means of publication, such as reference in a legislative act or separate announcement published in the official journal. This particular publication technique may be chosen in cases where other forms of publication meet with particular difficulties or the object in question is of very limited interest. The first such situation concerns documents such as maps and model forms (eg in the Netherlands). Legislation of ‘very limited interest’ may include exotic language versions of international treaties (eg in Austria). Further cases which may justify publication by reference to the title and the location where the text can be found are (eg in Switzerland) the fact that the documents (i) concern only a small group of people, (ii) are technical by nature and only intended for specialists, (iii) must be published in a different format or (iv) are published in an official organ of publication that is available in the state concerned. The Swiss Publication Act thereby requires that all texts published in the official gazette by reference are published in a different organ of publication or as a separate edition, or are provided by the office responsible on request.40

10.2.5.  Extraordinary Publication In extraordinary circumstances, ordinary publication may be impossible or not speedy enough to enable an urgent reaction to such circumstances. To cope with possible difficulties concerning regular publication, states are well advised to establish rules that address such cases. As it is difficult to predict the scale of such extraordinary circumstances, the corresponding national rules are likely to remain rather vague when describing alternative forms of publication. Some jurisdictions, however, provide for detailed regulation, such as 40  Comparable exceptions to the principle of integral publication can also be found in, for example, Belgium. See B de Sutter, ‘Regelgeving door verwijzing en de bekendmakingsplicht van normen. Een analyse aan de hand van de adviespraktijk van de afdeling wetgeving van de Raad van State’ [2012] Tijdschrift voor wetgeving 249.

178 

Karl Irresberger and Anna Jasiak

Germany for the ‘state of defence’: in such cases, if it is impossible to insert a new piece of legislation in the Federal Law Gazette or such insertion would come too late, publication can be made by broadcasting, or by insertion in daily newspapers or on the bulletin boards of the local authorities.41 Publication rules may stipulate that the publication must then be repeated in the regular way,37 or that the regular publication medium must include a notice pointing to the extraordinary publication. In case the regular publication is not feasible due to the non-functioning of the publication medium, an alternative medium will be indicated, as is the case, for example, in Europe: when it is impossible to publish the electronic version of the Official Journal of the European Union ‘due to an unforeseen and exceptional disruption of the information system of the Publications Office’, only the printed edition of the Official Journal ‘shall be authentic and shall produce legal effects’.42 Conversely, jurisdictions still adhering to paper publication may exceptionally allow electronic publication, for such reasons as urgency or in selected areas of legislation.43

10.3.  Legal Effect; Vacatio Legis It was indicated above that in (continental) European jurisdictions legislation can acquire legal effect only as a result of official publication. However, a few nuances are worth mentioning. Examples are the French legal doctrine according to which a law can be applied—provided that citizens are not thereby concerned (in the legal sense)—but is not enforceable (in French: ‘opposable’) against citizens before publication44 and that of Swiss legal doctrine45 suggesting that a legislative act can create rights of citizens vis-à-vis the state even without publication (because failure by the state to fulfil the requirement of publication should not be to the detriment of individuals); similarly, the Court of Justice of the European Union holds that a regulation which has not been properly published is not enforceable against individuals.46 It was mentioned above that jurisdictions set up general rules that determine the date of entry into force of a legislative act in relation to the time of publication. The general rule might even be that acts come into force on the date of publication (Finland). Standard delays between publication and entry into force range from less than a day (from midnight following the publication, eg in Austria, France) to one month (Norway) following publication. Other standard delays are five (Portugal), 10 (Estonia), 14 (Germany, Poland), 15 (Italy, Slovakia) and 20 (EU) days. In Switzerland, publication has to occur at least five

41 

Gesetz über vereinfachte Verkündungen und Bekanntgaben v 18 Juli 1975 (BGBl I 1919). After restoration of that system and making the corresponding electronic version available, the exclusive legal value remains with the printed edition, see Art 3 of Regulation (EU) No 216/2013 and the list of ‘print editions with legal effect’ at eur-lex.europa.eu/oj/all/auth-direct-access.html. The said exception, undesirably, appears to have been invoked even in the case of disruptions of less than a day. 43  Germany: eg § 66 Straßenverkehrsgesetz (as of 2009). 44  cf Waline in JB Herzog, G Vlachos and M Waline (eds), La promulgation, la signature et la publication des textes législatifs en droit comparé (Paris, Éditions de l’Épargne, 1961) 3ff. 45  cf M Roth, Die Veröffentlichung von Rechtsnormen in der Schweiz (Zurich, Dike, 2011) 266. 46 C-161/06 Skoma-Lux sro v Celní ředitelství Olomouc [2007] ECR I-10841; C-345/06 Heinrich [2009] ECR I-1659. 42 

Publication

 179

days before the act enters into force.47 The Netherlands has chosen for two fixed calendar days for legislative acts and delegated legislation to enter into force, namely 1 January and 1 July. The reason for this is to ensure that those to whom the law is addressed are not confronted with too frequent changes in law and have sufficient time to prepare for the new legal regime. It also allows for a reduction in the administrative burden. However, the period between the publication of the law and the moment of entering into force may not be less than two months.48 Exceptions to both rules are possible in cases of emergency legislation, legislation correcting legislative deficiencies, when European or international law needs to be implemented or when it is justified in order to prevent undesirable private or public (dis)advantages.49 Should an individual legislative act not contain any information about the moment of its entry into force, the rule in the Netherlands is that it will enter into force on the first day of the second calendar month after the date of publication.50 This wide range of national vacatio legis standards indicates that the general rule seems less important as long as the law gives certainty about the moment of entry into force, thus permitting the addressees to prepare themselves for the new regime.

10.4.  Publication Defects and Corrections It is not unusual that after the publication of a legislative act mistakes are discovered by readers. Some of them may have occurred during the process of development and adoption of the act, some of them in the course of publication. In the first case, there is little space, if any, for corrections to be made by anyone other than the legislator. In the second case, it is desirable that deviations from the original document that was to be published be eliminated. It is the publisher’s main task to render the content of the act faithfully, that is, without any omissions or degradations. Improvements, if permissible, must be confined to aspects of formal uniformity of texts in terms of orthography, citation and abbreviation rules, and printing styles, to the extent that this is understood to be part of the tasks of an editor and publisher of legal texts. Any interference that alters the meaning must be considered illicit. Sometimes mistakes subject to correction are described as ‘obvious’, meaning that it is clear from the context or by applying the relevant interpretation rules how the provision should actually read. Such mistakes do not alter the substance of the act, nor do the corrections of the mistakes. Corrections of a ‘minor’ nature which are in the interest of comprehensibility may be effected, according to jurisdictions such as that of Germany, in a special procedure which does not involve the legislators themselves but rather representatives, such as, in the case of parliamentary legislation, the presidents of both houses of parliament or, in the case of secondary legislation, senior officials of the respective ministry.51 Such corrections can be made both before and after publication. 47 

Art 7 of the Publication Law (Publikationsgesetz), SR 170.512. If the new law is also relevant for local governments, the period should not be shorter than three months. 49  Directive 174 of the Dutch Drafting Directives (Aanwijzingen voor de regelgeving), available at wetten.overheid.nl/BWBR0005730/Hoofdstuk4/415/Aanwijzing174. 50  Art 7 of the Publication Law (Bekendmakingswet). 51  cf Schneider (n 15) 295ff (Nos 495ff). 48 

180 

Karl Irresberger and Anna Jasiak

Sometimes the published version of a legislative act deviates from the adopted act in such a way that it fails to provide the original meaning, eg by omitting entire sentences or paragraphs. It may be obvious to the legislator and the publisher what should have been published, but not to the reader of the legal gazette. In such cases, the mistake is not a minor one and it is highly desirable that a corrigendum be published with a view to rendering the real text of the act; such correction may be the responsibility of the same person or organisation that was responsible for the publication or for the preparation of the respective act itself; this may be a rather informal action taken by the publication office or a formal action taken by the responsible minister. To take full effect, the correction should come into force at the same time as the act undergoing correction, but will frequently be retroactive. Such retroactivity, which may be evoked even years after the defective publication, may conflict with fundamental principles like legal certainty.52 In this context, there seems to be some wisdom in Portuguese publication law, which limits the period for corrections of publications to 60 days;53 it is also noteworthy that the Austrian constitutional court denied the admissibility of corrections (made by the administration to the original text adopted by parliament) changing the meaning, thereby necessitating a new enactment by parliament to remedy the publisher’s mistake.54

10.5.  Consolidation and Codification Nowadays, legislative bodies produce enormous amounts of legislative texts, and many countries, and even more so the EU, are struggling with it. In most cases, these texts do not concern new coherent pieces of legislation designed from scratch, but rather amend already existing legislation. This often makes it very difficult for the reader to understand from the amending act what the new legal situation is. To improve the accessibility of legislation in force and to guarantee its coherency, but sometimes also to manage its volume, states apply various techniques. Most common are those of codification and consolidation. These methods are applied by the EU as well, even though the EU adopts its own autonomous concept of these two techniques.

10.5.1.  Codification and Consolidation at the National Level Codification refers traditionally to laying down in a code all the rules relating to a specific subject or field of law. These rules stem not only from statues, but may also originate from customary and case law. Codification in these cases goes beyond just bringing the 52  On the use of and limits of curative legislation see A Jasiak, Constitutional Constraints on Ad Hoc Legislation (Antwerp, Intersentia, 2011) 265ff. 53  Lei No 42/2007. DR 163 Série I de 2007-08-24, Anexo: Republicação da Lei No 74/98, de 11 de Novembro, sobre a publicação, a identificação e o formulário dos diplomas. 54  VfSlg 16.852/2002. The reasoning, in short, is that such correction altering the substance of what has been published would amount to powers of the government to legislate with unlimited retroactive effect (or to discretionary powers to avoid unconstitutional retroactive effects), powers which have not been granted by the constitution to the government.

Publication

 181

rules together; harmonisation of these rules, defining certain concepts, etc, may be necessary with a view to creating an integrated and coherent body of rules in a specific field, such as criminal law, civil law or administrative law, based on common principles of law. Most European jurisdictions apply the concept of codification primarily in this sense, as appears from Voermans et al’s comparative survey.55 Voermans et al call this type of codification a ‘classic of big codification’. Other types of codification that the authors discern are ‘pure codification’, ‘corrective codification’ and ‘codification with recast’. Pure codification, in their definition,56 is meant to reduce the volume of legislation or to improve the accessibility to it. It brings together different statutes on the same subject in one act without amending the existing law. Codification programmes are known, for example, in the UK (where they are called consolidations) and France. Specific bodies, such as the Law Commissions57 in the UK and the Commission supérieure de codification in France, may be entrusted with the task of proposing codifications to parliament, as well as with facilitating and coordinating the works.58 Some jurisdictions, such as Spain, delegate the authority to produce and enact codifications to the government, and codifications might be subject to a simplified, expedited procedure.59 ‘Corrective codification’ allows minor amendments such as correcting errors and ambiguities,60 while ‘codification with recast’ concerns a situation in which, on the occasion of the codification, the original or preexisting acts are amended. From Voermans et al’s comparative survey, it appears that in various European jurisdictions recodification where substantial parts of an original act have been amended is seen as good practice.61 Codifications repeal the original, codified legislative acts and officially replace them. These are new pieces of legislation which have legal force and are therefore legally binding upon individuals. Unlike codifications, consolidations usually do not qualify as new legislation as such; they normally do not substitute the original acts. Consolidations are about integrating all changes into the original, basic act up to a certain date. Such an integration is not part of the legislature’s task—which normally ends after issuing an amendment to a legal act— but, rather, concerns a technical operation that does not affect the content of the basic act. This means that the authority which performs consolidations does not make any other changes to the original act other than those which the legislature has agreed upon. As an exception to this rule, consolidations in the form of official republications going beyond pure integration of amendments may be authorised by the legislature (eg in Austria,62 the

55  WJM Voermans, CP Moll, NA Florijn and P JPM van Lochem, ‘Codification and Consolidation in Europe as Means to Untie Red Tape’ (2008) 29 Statute Law Review 65. 56  For a more elaborated description see ibid 79. 57  There are separate Law Commissions for England and Wales, Northern Ireland and Scotland. cf S Wilson ‘Reforming the Law (Commission): A Crisis of Identity?’ [2013] Public Law 20. 58  See the French Décret No 89–647 du 12 septembre 1989 relatif à la composition et au fonctionnement de la Commission supérieure de codification. 59  This is the case both in the UK and in France. For a more detailed description of this parliamentary procedure see the UK Guide to Making Legislation (July 2015) 278ff, the French Guide de légistique (March 2014) and the Circulaire du 30 mai 1996 relative à la codification des textes législatifs et réglementaires. 60  In the UK, for example, this type of codification is regulated by the Consolidation of Enactments (Procedure) Act 1949. 61  Voermans (n 55) 80. 62  Art 49a Bundes-Verfassungsgesetz (general authorisation by constitution).

182 

Karl Irresberger and Anna Jasiak

Netherlands63 and Malta).64 They may include, among other things, renumbering, restructuring, corrections of orthographic mistakes or rectification of obsolete provisions. Such authorised republications are published in the official gazettes and are legally binding (see below). Preparation of consolidated texts might be put in the hands of a variety of actors. In many cases, the responsible ministers (Austria, Hungary, Romania) or high public authorities are entrusted with this task (eg the Chancellery of the Parliament in Lithuania, the Speaker of the Parliament (Sejm) in Poland or—when requested by the Speaker of the Parliament or by the government—the Council of State in Belgium). The question of whether a consolidation is due is usually decided on a case-by-case basis by the official publisher of the state gazettes, official journals, etc, and it will be performed when the original act has been modified substantially, seriously diminishing its accessibility.65 In some countries, the publication of consolidations is mandatory and prescribed. In Poland, the Speaker of the Sejm announces the consolidated text at least once a year if the law concerned has been amended.66 In Portugal, laws have to be republished after at least three amendments.67 In the digital age, the online presentation of the bulk of legislation in force in a consolidated version has become an unquestioned standard for today’s governments. However, most countries do not treat the consolidated texts in the same or a similar manner as they treat the original acts. The latter still count as authentic documents; the consolidated texts are not meant to replace them and therefore have little, if any, legal value; nevertheless, online accessibility of non-authentic consolidated texts weakens the need and demand for authentic ones.68 It appears that so far only a few jurisdictions have moved towards comprehensive consolidation with legally binding effect and with no particular institutional constraints for issuing such legally binding consolidated acts. For example: —— In Estonia, consolidated texts are published (in electronic version) after each modification. They have a legally binding nature, but a court may refuse to apply a consolidated text which does not correspond to the original text.69 —— Maltese law provides for a revised printed or electronic edition, vested with legal effect of the statute laws of Malta, being published at least every ten years,70 but goes further by giving legal effect to consolidated online versions of laws that have encountered amendments after the release of the latest printed edition. Article 11A. section 4 of the Statute Law Revision Act 1980 states that ‘Unless proof is brought to the contrary

63 

Such reprints are based on the Dutch Drafting Directives (Directive 246). In Malta it is the Law Commission which is authorised to make such revisions under the Statute Law Revision Act. 65  Voermans (n 55) 76. 66  Ustawa z dnia 20 lipca 2000 r. o ogłaszaniu aktów normatywnych i niektórych innych aktów prawnych (The Act on publication of normative acts and some other legal acts), Art 16, as of 1 January 2016. 67  See n 53. 68  In Austria, for example, republications (which are issued by the Federal Chancellor together with the competent minister) have become rare since the creation (in the 1990s) of the federal legal information system, which provides non-authentic consolidated versions of legislation in a comprehensive way: cf Irresberger (n 34) 149. 69  M Seppius, ‘Consolidation’, Interim Report of the Working Group. European Forum of Official Gazettes, 5th meeting, Madrid, 25–26 September 2008, 6ff, available at circabc.europa.eu/webdav/CircaBC/OPOCE/ojf/Information/prod/data/pdf/madrid2008-consolidation-interimreport.pdf. Riigi Teataja seadus (Riigi Teataja Act [State Gazette Act]) § 3 and § 9, paras (3) and (4). 70  A tradition retained from the times of the British Empire. 64 

Publication

 183

the text of any law published on an internet site in accordance with this article shall be deemed to be a true representation of the law incorporating all amendments up to the date indicated on the internet site’. —— In Poland, consolidated texts of parliamentary laws are published in the electronic Law Gazette71 and have an official character. Although they do not substitute the original acts, they are presumed to be authentic. The case of Slovenia deserves particular attention. There, an ‘official consolidated text’ of a parliamentary law is prepared if the National Assembly so decides. Such text is ‘confirmed’ by the National Assembly without debate. The National Assembly may decide that all official consolidated texts be confirmed by single voting.72 An ‘official consolidated text’ is published in the Official Gazette and is made available in electronic form on the website of the National Assembly (consolidation is therefore a part of a promulgation process). Consolidated texts are viewed as representing but not repealing the contributing acts (basic act and amending acts). Subsequent legislation will refer to and proceed from the official consolidated text, which is legally binding and presumed to be authentic. However, in case of errors that might have occurred during the consolidation process, it has been considered that the validity of contributing acts prevails over the validity of the consolidated version, taking into account that the contributing acts were adopted by the parliament (via debates and following first, second and third readings) while the consolidation version was only ‘confirmed’ without any debate.73 The Slovenian example represents a practice which offers a high level of accessibility to the national legislation in force for the public. Given that the consolidation procedure is a part of the official promulgation process and consolidated texts are authorised by the legislature, this practice gives the public the confidence that the consolidated text is reliable.

10.5.2.  Consolidation and Codification in the EU While most of the national jurisdictions use the term ‘codification’ for laying down in a code all the rules relating to a specific subject or field of law, as stated above, at the European level the term ‘codification’ refers to a situation where a new legislative act is created by bringing together a basic legislative act and its amendments (also called formal or official consolidation). Being more than just consolidation, such codification provides an authoritative text by also deleting obsolete and overlapping provisions, by harmonising terms and definitions, and by correcting errors without substantive change. The texts being codified are repealed and replaced by the new act, which needs to pass through the full legislative process. However, the interinstitutional agreement on an ‘Accelerated working method for 71  Ustawa z dnia 20 lipca 2000 r. o ogłaszaniu aktów normatywnych i niektórych innych aktów prawnych (n 66), Art 9. 72  Art 153 of the Rules of Procedure of the National Assembly, available at https://www.dz-rs.si/wps/portal/ en/Home/ODrzavnemZboru/PristojnostiInFunkcije/RulesoftheProcedureText/!ut/p/z1/04_Sj9CPykssy0xPLMnMz0vMAfIjo8zinfyCTD293Q0N3MOczAw8QwJcXE0tLIwsgk31wwkpiAJKG-AAjgb6XvpRmUm5euXJuXoGeiYWlpbGFhaWhkBoZmBoZK4f7qQfWZWZpmrgG6hfkBtR5emoqAgAjJ15Tw!!/dz/d5/ L2dBISEvZ0FBIS9nQSEh/#_ftn1. 73  E-mail information given to the author by the Office of Legislation of the Government of the Republic of Slovenia on 12 June 2014.

184 

Karl Irresberger and Anna Jasiak

official codification of legislative texts’ provides for a simplified procedure.74 This method was used when the project75 of codifying the whole ‘acquis communautaire’ was performed from 2001 to 2009, and several codifications are still produced each year. Two types of codification can be discerned: vertical and horizontal. Vertical codification means that the original act and its amendments are incorporated into a single new act, while in the case of horizontal codification the incorporation concerns two or more original acts covering related subjects and the amendments to them.76 Unlike codification, consolidation at the European level does not concern enactment of any new legally binding act; rather, it is just an integration of subsequent amendments and correction of errors into the basic text. Consolidations are prepared by the Publication Office of the EU and they do not have any legal value, but are used as a documentation tool (they are therefore also called informal codifications).77

10.6. Conclusions By making law accessible to the public, the publication of laws is one of the essential requirements of proper legislation. Governments have therefore taken up the challenge to provide individuals and other legal subjects with adequate and timely information (free of charge, wherever possible) on the legal rules in force. ICT and the internet played and still do play an important role in this process, and their full potential is still being explored. However, these technologies are unable to offer a solution to all the problems concerning the accessibility of law. Real access to law remains restricted, for example, because of the great volume of legislation that states enact nowadays, or in cases of legislation by reference. Accordingly, complementary efforts of governments to really bring the law to the citizens (non-experts and non-lawyers) are still needed. The above descriptions of practices in some countries, like a decision to give legal value to consolidated legislative texts, show that a little bit of creativity and daring can bring about serious improvements in this respect.

Further Reading Berger, A, Access to Legislation in Europe. Guide to the Legal Gazettes and Other Official Information Sources in the European Union and the European Free Trade Association (Luxembourg, Publications Office, 2009).

74 

Interinstitutional Agreement of 20 December 1994 (96/C 102/02). Communication from the Commission to the European Parliament and the Council—Codification of the Acquis communautaire, COM (2001) 0645 final. 76  See the website of the EC: ec.europa.eu/dgs/legal_service/codifica_en.htm. 77  Voermans (n 55) 68–9. Discussions between the European Commission and the Publication Office at the level of the Management Committee about the wish to provide legal value to consolidated texts are ongoing, as appears from the 2014 Annual Management Report (Luxemburg, Publication Office of the European Union, 2014) 28. 75 

Publication

 185

Herzog, JB, Vlachos, G and Waline, M (eds), La promulgation, la signature et la publication des textes législatifs en droit comparé (Paris, Éditions de l’Épargne, 1961) (including country reports on selected European [BE, CH, CZ, DE, GR, FR, IT, PL, SE, UK, YU] and non-European countries). Hietanen, A, ‘Electronic Publishing of Legislation—Towards Authenticity’, Final Report of the Working Group, European Forum of Official Gazettes, 4th Meeting, Helsinki– Tallinn, 14–15 June 2007. Seppius, M, ‘Consolidation’, Interim Report of the Working Group. European Forum of Official Gazettes, 5th meeting, Madrid, 25–26 September 2008, available at circabc. europa.eu/webdav/CircaBC/OPOCE/ojf/Information/prod/data/pdf/madrid2008consolidation-interimreport.pdf. AT: Holzinger, G, ‘Die Kundmachung von Rechtsvorschriften in Österreich’ in H Schäffer (ed), Theorie der Rechtssetzung (Vienna, Manz, 1988) 303–41. AT: Irresberger, K, ‘Die Publikation von Rechtsvorschriften in Österreich’ [2013] LeGes 141. CH: Roth, M, Die Veröffentlichung von Rechtsnormen in der Schweiz (Zurich, Dike, 2011). CH: ‘Themenheft “Publikationsrecht”’ [2013] LeGes. DE: Schneider, H, Gesetzgebung, 3rd edn (Heidelberg, Müller, 2002). DE: Guckelberger A, Der Übergang zur (ausschließlich) elektronischen Gesetzesverkündung. Dargestellt am Amtsblatt des Saarlandes unter Einbeziehung der Bundes- und Gemeindeebene sowie anderer E-Government-Prozesse (Baden-Baden, Nomos, 2009). DE: Severin U, Das Bundesgesetzblatt. Eine staatsrechtliche Untersuchung (Bonn, Deutscher Bundesverlag, 1962). UK: Cormacain R, ‘Have the Renton Committee’s Recommendations on Electronic Access to Legislation been Fulfilled?’ (2013) 19(3) Web JCLI available at webjcli.org/article/ view/264/361%22%20//l%20%22_ftn30.

186 

11 Regulatory Evaluation Ex Ante and Ex Post: Best Practice, Guidance and Methods STEPHAN NAUNDORF AND CLAUDIO M. RADAELLI

This chapter is dedicated to concepts, methods and guidance for regulatory policy evaluation. We address evaluation across the policy cycle: rules should be evaluated before they enter into force (typically with Regulatory Impact Assessment, RIA), but the core principles and methods of RIA can also be used ex post to carry out tests on existing regulations. Regulatory policy evaluation draws on a wide range of methods, hence tests and analyses informed by RIA are only a component of this task. However, by focusing on RIA we can present a coherent discussion of analytical methods and discuss best practice across OECD countries. We first review the available evidence on regulatory policy evaluation, focusing on the literature on RIA. Then we introduce and discuss concepts, methods and examples. S­ pecifically, we discuss the main components of RIA, covering problem definition, the identification of criteria for analysis, the measurement of the baseline, the formulation of alternative options, data gathering, consultation, the economic analysis of feasible options, monitoring and quality assurance.

11.1.  Introduction and Definitions In this chapter, we introduce our readers to best practice and evidence-based policy ­instruments for evaluating regulation ex ante and ex post. In some countries, the methods we describe apply only to ‘regulation’ (secondary legislation), whilst in others they are used to scrutinise primary legislation (see Chapter 2 by Wim Voermans on legislation and regulation). At the outset, we explain what we mean by the concept of high-quality regulation: for us, high-quality regulation as ‘output’ of the legislative and regulatory processes is responsive, efficient, easy to understand, accessible and socially legitimate. However, we also care about the process that produces the regulations. From the perspective of ‘process’, high-quality regulation is based on an open and transparent process of decision making under the rule of law. When we say ‘open’, we are referring to diffuse consultation (ex ante and ex post) and the obligation of regulators to notify and inform those who are targeted by proposed regulation. This double understanding of high-quality regulation is in line with international standards and the literature, although in the real world there are trade-offs among all these qualities of regulation.

188 

Stephan Naundorf and Claudio M. Radaelli

The importance of ensuring the quality of regulations systematically is broadly connected to issues of accountability and legitimacy that affect both independent regulators and public administration in general. Essentially, this is what is at stake in the drive towards better regulation (see Chapter 15 by Luzius Mader). Micklethwait and Wooldrige1 talk about the need to reinvent the state as they see Western democracies and emerging states in a vicious circle: ‘Having overloaded the state with their demands, voters are furious that it works so badly.’2 They call for another revolution: a revolution that will have to improve state management and harness technology, particularly information technology, while at the same time will ‘think hard about the proper role of state in a fast changing society’.3 Fukuyama4 argues that successful nations are grounded in three pillars: a strong, competent state; rule of law; and mechanisms of accountability. The same author5 also reminds us of the well-known correlation between the quality of government and good economic and social outcomes. Evaluation provides the decision makers with the most relevant information about whether, to what extent and at what costs policies reach their objectives.6 This is why regulatory evaluation matters. However, evaluation not only needs to meet professional standards, as they are usually defined by the scientific community and professional associations of all kinds; it also needs to be effective in itself. Evaluation, understood as the ‘systematic application of social research methods to the assessment of social intervention programs’,7 is a means to an end in political contexts. This end is regulatory quality (see Chapter 15) and management of regulation across the policy cycle (see Chapter 5 by Patricia Popelier). Following an extensive self-evaluation of the tools that OECD member states utilise to develop the quality of their regulation,8 the OECD has argued that the implementation of regulatory policies should build on a clear methodology, on instruments for oversight and quality control, on stakeholder engagement and transparency, and finally on the systematic adoption of the respective practices.9 In the following, we use the term ‘regulation’ to cover both primary and secondary ­legislation, hence we also cover regulations that are usually issued by independent agencies. The main difference is that proposals for primary legislation are subject to parliamentary scrutiny and decision making, while secondary legislation is usually to be adopted by the authorised and issuing agency. In many jurisdictions, secondary legislation is also more likely to be revised ex post by judicial review. In order to deliver high-quality regulation, government intervention should rely on rules that are properly evaluated both before they enter into force (typically ex ante via

1 J Micklethwait and A Wooldridge, The Fourth Revolution: The Global Race to Reinvent the State (Harmondsworth, Penguin Books, 2015). 2  ibid 11. 3  Micklethwait and Wooldridge (n 1) 21. 4  F Fukuyama, The Origins of Political Order: From Prehuman Times to the French Revolution (London, Profile Books, 2011). 5 F Fukuyama, Political Order and Political Decay: From the Industrial Revolution to the Globalisation of ­Democracy (London, Profile Books, 2014). 6  OECD, ‘OECD Framework for Regulatory Policy Evaluation’ (Paris, OECD Publishing, 2014). 7 PH Rossi, MW Lipsey and HE Freeman, Evaluation: A Systematic Approach, 5th edn (New York, Sage ­Publications, 1993) 54. 8  See data at http://qdd.oecd.org/subject.aspx?Subject=f39ac642-3247-4140-acd2-e65dac5b8aaf. 9  OECD, ‘Regulatory Policy Outlook 2015’ (Paris, OECD Publishing, 2015).

Regulatory Evaluation

 189

­ egulatory Impact Assessment, RIA) and after they have operated for a certain period of R time (ex post evaluation).10 Although RIA is mostly used ex ante, its logic and main components can be deployed ex post too, for example to carry out cost–efficiency tests of existing regulations. The core principles and methods of RIA do not change whether we apply it ex ante or ex post. Hence, in the remainder of the chapter the reader should bear in mind that RIA principles and methods are easily transferrable to ex post evaluations. Indeed, the notion of regulatory management, or ‘managing regulation through the life cycle’, captures the two dimensions, ex ante and ex post. The main difference between ex ante and ex post evaluation is that it is generally easier to collect data for ex post evaluations. Many methods described below perform both ex ante and/or ex post. RIA is a systematic, comparative appraisal of how proposed primary and/or secondary legislation might affect stakeholders, society, economic sectors and the environment. In its ex post version, RIA is an appraisal of how existing rules have affected stakeholders, society, the economy and the environment. In ex post usages of RIA, the main problem is to establish the counterfactual: we compare data on how regulations have affected X or Y with the value that outcomes would have taken without such regulations. Since we cannot observe the counterfactual, this comparison of the state of play with the counterfactual is the most difficult issue to deal with in establishing causality. Going back to our definition of RIA, a few words of clarification are in order. ‘Systematic’ means coherent—not episodic or random. ‘Comparative’ means that more than one option is appraised, including the option of not altering the status quo (baseline). Some RIA procedures draw attention to the unintended effects and their measurement. The complexity and analytic breadth of RIA and ex post evaluations (carried out via RIA or otherwise) vary, depending on the issues at stake and the resources available—the degree of complexity should be proportional to the salience and expected effects of the regulation. Complex RIAs normally require quantification of compliance costs and at least some estimation of benefits. Simple RIAs can be limited to the analysis of main administrative costs. In any case, the thrust of RIA is to improve on the peripheral vision of decision makers by asking them to define what the problem is and to appraise the status quo correctly. This is because even the best economic analysis of options may be flawed if the problem is not defined correctly. Indeed, sometimes regulators identify only certain categories of costs (rather than performing comprehensive analyses), such as the measurement of administrative burdens or, in more ambitious versions, compliance costs.11 RIA can also be used to appraise the effects of proposed regulations on public administration (eg other departments, schools, hospitals, prisons, universities) and sub-national governments. The comparative evidence supports the notion that RIA is a flexible instrument. Previous research12 shows that governments have adopted RIA for different aims: —— Identification and selection of policy options, for example, to explore new policy proposals. Options can be developed and examined without too much prior knowledge.

10  For the different methods used, see C Coglianese, ‘Measuring Regulatory Performance Evaluating the Impact of Regulation and Regulatory Policy’, OECD Expert Paper No 1 (Paris, OECD Publishing, 2012). 11  See OECD, ‘Regulatory Policy Outlook 2015’ (n 9). 12  K Jacob et al, ‘Evaluating Impact Assessment, Final Activity Report’ (2006), available at http://cordis.europa. eu/docs/publications/1243/124376871–6_en.pdf (accessed on 16 February 2016).

190 

Stephan Naundorf and Claudio M. Radaelli

—— Policy fine-tuning. In this case, the government has already taken major decisions, but needs to adjust the current policy instruments. In this scenario, the government may have produced reviews of existing policy or scenario documents—ex ante and ex post evaluations are connected. —— Involvement of stakeholders. RIA is the major tool for collecting evidence and engaging stakeholders. In this regard, it operates as a structure of incentives and procedures to increase the participatory dimension either in policy formulation or in the ex post evaluation of existing rules or whole policy regimes. —— Explaining policies. Finally, RIA can also explain and justify choices. Together with appropriate indicators,13 the results from different RIAs (eg the total net benefits delivered in a period of x years) is a tool to communicate the content of policy and to report on regulatory reform. RIA is usually used to estimate the expected effects of proposed regulations. It could be deployed consecutively during the process of developing regulations from the problem definition or when first conceptual ideas are developed until its final approval by the authorised bodies. In this case, RIA should be understood primarily as a process. This view reflects in particular that consultation, evidence gathering, economic analysis and the proposed regulation undergo subsequent changes during the phases of their respective developments. In the very beginning of the process, the core ideas will be described in a more general way. In this situation, the inception RIA can hardly cover more than some of the basic facts about the addressed entities and the expected effects. During the process of elaborating the proposed regulation, more and more detailed data will be included. Process and documentation reporting depend on each other. The process and the documentation are equally important, since processes that are not documented are not transparent and cannot be challenged by stakeholders. In ex post evaluation or monitoring exercises, the ex ante RIA provides the baseline document for comparisons with the results achieved. Throughout this chapter, therefore, we will examine RIA in both ways as a process and as a document that enables the decision makers involved to come to informed decisions and the stakeholders involved to understand, challenge and manage (i) the reason(s) why politicians, the government or agencies intervene, (ii) how the new regulations are most likely to affect a broad range of stakeholders and the environment, and (iii) the estimated economic impact of alternative options and of the actual measures.

11.2.  An Overview RIA is a flexible tool fostering evidence-based learning14 and decision processes. Ex post legislative evaluation is a relatively new field, at least in Europe, where there have been 13  See CM Radaelli and F de Francesco, Regulatory Quality in Europe: Concepts, Measures and Policy Processes (Manchester, Manchester University Press, 2011); CM Radaelli and O Fritsch, ‘Measuring Regulatory Performance: Evaluating Regulatory Management Tools and Programmes’, OECD Expert Paper No 2 (Paris, OECD Publishing, 2012). 14  CM Radaelli, ‘Measuring Policy Learning: Regulatory Impact Assessment in Europe’ (2009) 16 Journal of European Public Policy 1145.

Regulatory Evaluation

 191

attempts to carry out post-implementation reviews of regulations and, at the EU level, ­fitness tests of regulatory frameworks. Although RIA has been adopted throughout the OECD and in several developing ­countries, its implementation varies markedly. Often even in a single country there are ­variations across sectors. We do not provide a literature review here because there are already at least two good sources that the reader can refer to. On ex post legislative evaluation, the European Journal of Risk and Regulation dedicated a symposium issue to the topic, with a comprehensive introductory article by Stijn Smismans.15 Turning to RIA, Dunlop and Radaelli16 review the state of play with this policy instrument—their handbook provides systematic information on cross-country implementation and practices across OECD and developing countries, as well as chapters on specific methods that we cannot cover here, such as gender impact assessment and social impact assessment. For our purposes, it is sufficient to illustrate the main characteristics of the instrument rather than its cross-country variation: —— RIA is a process for assessing the likely effects of a proposed policy change against the benchmark of the status quo (or the hypothesis of ‘no change’); when the no-change option is not appraised, the RIA is weakened; —— it is eminently comparative, considering one risk against others, or a policy option against other options; —— it involves an assessment of the most direct effects, as well as the consequences of side effects, for example effects that occur beyond the specific sector being regulated; —— it fosters structured consultation with stakeholders (citizens, business sector, administration), thus providing a possible framework for ‘regulatory conversations’17 and for the reality check of data and underlying concepts of regulations; —— it should be carried out and documented with the aim of producing knowledge (eg data and empirical evidence) that can be used in policy decisions, thus making RIA different from a classic background study, a literature review or an academic report; and —— it should consider both design and delivery implementation, since most of the effectiveness of rules depends on how they are delivered. The OECD Council produced the first-ever international benchmark on high-quality decision making with the 1995 Declaration.18 This checklist is still widely used in the OECD and developing countries for training purposes and for the design of guidelines on RIA. It is essentially an overview of the main steps of RIA. It has been updated within the Recommendation of the Council of the OECD on Regulatory Policy and Governance, which was adopted on 22 March 2012. Recommendation number 4 focuses on RIA (see Box 1).

15  S Smismans, ‘Symposium on Policy Evaluation in the EU Policy Evaluation in the EU: The Challenges of Linking Ex Ante and Ex Post Appraisal’ (2015) 6 European Journal of Risk and Regulation 6. 16  CA Dunlop and CM Radaelli (eds), Handbook of Regulatory Impact Assessment (Cheltenham, Edward Elgar Publishing, 2016). 17  J Black, ‘Regulatory Conversations’ (2002) 29 Journal of Law and Society 163. 18  Recommendation of the Council on improving the quality of government regulation, 9 March 1995, C(95)21 Final (Paris, OECD, 1995).

192 

Stephan Naundorf and Claudio M. Radaelli

Box 1 Recommendation No. 4 of the Council of the OECD on Regulatory Policy and Governance; Adopted 22 March 2012 4. Integrate Regulatory Impact Assessment (RIA) into the early stages of the policy process for the formulation of new regulatory proposals. Clearly identify policy goals, and evaluate if regulation is necessary and how it can be most effective and efficient in achieving those goals. Consider means other than regulation and identify the tradeoffs of the different approaches analysed to identify the best approach. 4.1 Adopt ex ante impact assessment practices that are proportional to the significance of the regulation, and include benefit cost analyses that consider the welfare impacts of regulation taking into account economic, social and environmental impacts including the distributional effects over time, identifying who is likely to benefit and who is likely to bear costs. 4.2 Ex ante assessment policies should require the identification of a specific policy need, and the objective of the regulation such as the correction of a market failure, or the need to protect citizen’s rights that justifies the use of regulation. 4.3 Ex ante assessment policies should include a consideration of alternative ways of addressing the public policy objectives, including regulatory and non regulatory alternatives to identify and select the most appropriate instrument, or mix of instruments to achieve policy goals. The no action option or baseline scenario should always be considered. Ex ante assessment should in most cases identify approaches likely to deliver the greatest net benefit to society, including complementary approaches such as through a combination of regulation, education and voluntary standards. 4.4 When regulatory proposals would have significant impacts, ex ante assessment of costs, benefits and risks should be quantitative whenever possible. Regulatory costs include direct costs (administrative, financial and capital costs) as well as indirect costs (opportunity costs) whether borne by businesses, citizens or government. Ex ante assessments should, where relevant, provide qualitative descriptions of those impacts that are difficult or impossible to quantify, such as equity, fairness, and distributional effects.

Regulatory Evaluation

 193

4.5 Regulatory Impact Analysis should as far as possible be made publicly available along with regulatory proposals. The analysis should be prepared in a suitable form and within ­adequate time to gain input from stakeholders and assist political decision making. Good practice would involve using the Regulatory Impact Analysis as part of the consultation process. 4.6 Ex ante assessment policies should indicate that regulation should seek to enhance, not deter, competition and consumer welfare, and that to the extent that regulations dictated by public interest benefits may affect the competitive process, authorities should explore ways to limit adverse effects and carefully evaluate them against the claimed benefits of the ­regulation. This includes exploring whether the objectives of the regulation cannot be achieved by other less restrictive means. 4.7 When carrying out an assessment, officials should: —— Assess economic, social and environmental impacts (where possible in quantitative and monetised terms), taking into account possible long term and spatial effects; —— Evaluate if the adoption of common international instruments will efficiently address the identified policy issues and foster coherence at a global level with minimal disruption to national and international markets; —— Evaluate the impact on small to medium sized enterprises and demonstrate how administrative and compliance costs are minimised. 4.8 RIA should be supported with clear policies, training programmes, guidance and quality control mechanisms for data collection and use. It should be integrated early in the processes for the development of policy and supported within agencies and at the centre of government. Source: OECD, ‘Recommendation of the Council of the OECD on Regulatory Policy and Governance, Which Was Adopted 22 March 2012’ (Paris, OECD Publishing, 2012) 10–11.

194 

Stephan Naundorf and Claudio M. Radaelli

However, early examples of quantitative ex post evaluation indicate that evaluating ­policies quantitatively is clearly no modern invention. Following the self-evaluation of the implementation of certain regulatory policy tools by the OECD member states, the OECD proposes a composite indicator for regulatory management policies which contains three instruments:19 RIA; stakeholder engagement; and ex post analysis. The implementation of each of these instruments is anchored to four pillars: methodology; oversight and quality control; stakeholder engagement and t­ransparency; and systematic adoption of the respective practices. (see Fig. 1).

RIA methodology

RIA oversight and quality control RIA transparency

RIA systematic adoption

Stakeholder engagement methodology

Ex post evaluation methodology

Stakeholder engagement oversight and quality control Stakeholder Stakeholder engagement engagement systematic transparency adoption

Ex post evaluation oversight and quality Ex post Ex post control evaluation evaluation systematic transparency adoption

Figure 1: Overview of regulatory indicators: main categories. Source: C Arndt et al, ‘2015 Indicators of Regulatory Policy and Governance’, OECD Expert Paper No 2 (Paris, OECD Publishing, 2015), available at http://dx.doi.org/10.1787/5jrnwqm3zp43-en. Permission granted by the OECD in February 2016

11.3.  Problem Definition Problem definition is arguably the most important determinant of regulatory quality. Define the problem wrongly and, no matter what analytic methods are used, the regulations will be flawed. But what does problem definition mean? For some classic, routine problems, regulators are often in a position to sit down and define the problem without too much political interference, as shown by the studies on public policy making in the UK.20 For more highprofile issues, the problems will land on the desk of the regulators already ‘defined’ by political consensus (like a coalition treaty in countries like the Netherlands), parliamentary hearings, general political attention and initiative, the media or civil society organisations. However, these ‘definitions’, especially the ones communicated by professional media like newspapers or social media after some high-visibility episodes, are often not the best point of departure for an RIA. They are limited, they often call for ‘immediate intervention’ and they address the current wave of emotions rather than undertaking a thorough analysis of the problem.21 At 19 

OECD, ‘Regulatory Policy Outlook 2015’ (n 9). Page, ‘The Civil Servant as Legislator: Law Making in British Administration’ (2003) 81 Public ­Administration 651; EC Page and B Jenkins, Policy Bureaucracy: Government with a Cast of Thousands (Oxford, Oxford University Press, 2005). 21  The classic case is the reaction to ‘dangerous dogs’ attacking children or those workers delivering the post: C Hood, H Rothstein and R Baldwin, The Government of Risk: Understanding Risk Regulation Regimes (Oxford, Oxford University Press, 2001). 20 EC

Regulatory Evaluation

 195

the outset, the process of impact assessment takes us back to the fundamental questions of what is the problem, what is wrong and what needs to be fixed? This brings us to the conceptual framework of problem definition. However, the very act of ‘defining’ a problem is not straightforward, as shown by a number of popular myths.

11.3.1.  Common Myths About Problems Are Hard to Dispel 1. The regulator defines the problem. This is true in some cases, but, as we have just seen, there are often many social actors that define the problem in different ways—­ consultation often exposes these different social definitions. 2. The state is the only regulator. Actually, regulation is also produced by professional associations like the British Medical Association, firms, lobbying groups, non-governmental organisations and self-administrating bodies. 3. Information about the problem is inside the department. In most cases, information and data on the problem is at least equally rich (if not better) outside the department. Lobbying groups and other state levels are examples of obvious sources of information, although such information may be biased. However, it will usually be hard work for the regulators to collect and quality check the empirical evidence. 4. All problems have basically the same causal thread. Some problems have simple causeand-effect patterns, others are complicated and some can be complex. There are also problems that pose questions that science and public policy cannot solve at the moment. Analytical devices like ‘problem-tree’ analysis support regulators and stakeholders in the analysis of causal relationships. The problem definition will also provide the main reference for the ex post evaluation. However, the ex post evaluation might come to the conclusion that the regulation did not solve the problem described, though it might have had other beneficial effects.

11.4.  Setting the Criteria The general criteria for the design of RIA should be set prior to the beginning of the appraisal. Usually some features of regulatory policy are already set, legally and/or politically: the legal framework, general political decisions, organisational and structural conditions, and specific political goals for the regulation intended are all provided before the appraisal work begins. The possibility to learn from RIA is reduced substantially if one looks at the policy options first, makes a choice and then retrofits the decision-making criterion to justify that choice. It is also inconsistent practice in terms of transparency and accountability. Thus, criteria for the design of a piece of regulation should be clear, transparent and set in advance. Some governments have broad criteria,22 within which there is a range of discretion. This is because some criteria are appropriate for major, complex RIAs, whilst others satisfy less complicated appraisals. Again, this is a question of proportionate analysis, since some 22  See OECD ‘Regulatory Policy Outlook 2015’ (Paris, OECD Publishing, 2015), results of a self-evaluating survey of member states regarding their implemented tools for regulatory policy. Data publicly available at

196 

Stephan Naundorf and Claudio M. Radaelli

criteria are very demanding in terms of the analysis to be performed. Some criteria are easy to be standardised in simple models (eg administrative burden, compliance costs), other criteria, such as job effects, rely on the chosen model for calculation and the underlying (macro-economic) assumptions, and yet other criteria, such as effects on sustainability, are very difficult to evaluate ex ante at all because of their multivariate character. Especially in the cases with a multivariate character, it is helpful to recall one of the propositions of evaluation: ‘Evaluation is an art, and every evaluation represents, or should represent, an idiosyncratic effort to meet the needs of program sponsors and stakeholders.’23 Among the criteria used in different countries are the following: —— Threshold criteria—such as ‘an independent statistical evaluation of data provided in the proposal is to be done, if the compliance costs exceed xxx’. —— Cost-effectiveness criteria—to be used when there is a given and measurable benefit to be achieved. In this case, the aim is to choose the option that delivers on the given objective in the most efficient way. —— Cost–benefit principles. A classic cost–benefit principle is that the flow of discounted benefits should outweigh the flow of costs. This is typical for economic feasibility analysis. Variations of this principle include the consideration of benefits and costs for a particular category of stakeholders, such as businesses. In any case, it is important to ensure that the quantification of benefits and all assumptions of the used models are transparent, clear and consistent. Otherwise, there is a high risk of gaming.

11.5.  Measuring the Baseline RIA could be done in an ideal setting as a comparative exercise, depending on the target audience, the available options for the design of the proposed regulation and—last but not least—the resources available (time and capacity). A fundamental comparison is between the hypothesis of intervention and the null hypothesis of not altering the status quo (the do-nothing option). In several policy domains, it is not entirely clear what is wrong with the status quo. There are always pressures to intervene and regulate, both from civil society and from elected politicians who want to increase their popularity by championing a new bill in parliament or drive home a new regulation. But further analysis on what motivates political decision makers specifically to call for change (and how much, in terms of undesirable effects) is needed intellectually before one can proceed with the recommendation to issue a new regulation. In democratic systems, the elected majority will have the right of decision within the given constitutional and legal framework. However, evaluators and policy officers must accept that the logic of political decision making might lead to ­different results

http://qdd.oecd.org/subject.aspx?Subject=f39ac642–3247–4140-acd2-e65dac5b8aaf. For example, the Joint Rules of Procedures of the Federal Ministries in Germany require that the explanatory note of legislative draft reflects on up to 37 different aspects (see Chapter 6; in German: http://www.verwaltungsvorschriften-im-internet.de/ bsvwvbund_21072009_O11313012.htm) These range from effects on public households to effects on gender equality, sustainability, benefits and compliance costs. 23 PH Rossi, MW Lipsey and HE Freeman, Evaluation: A Systematic Approach, 5th edn (New York, Sage ­Publications, 1993) 30.

Regulatory Evaluation

 197

compared to the logic of evaluations.24 Policy officers support these processes by providing expertise and capacity,25 but they are not usually entitled to control or veto the elected politicians. The control function lies with the judiciary. Problem definition is a way in which RIA generates information on the status quo. It is good practice to capture the baseline in its overall qualitative–quantitative dimensions. The identification of a baseline requires in addition some form of counterfactual reasoning: what would happen if nothing is done? This requires the policy team to plot the baseline against the future and to design a scenario. There are consequences of the status quo that may not be a concern in the short term, but should be avoided in the long term: for example, demographic projections on the robustness of pension systems or long-term environmental damage created by current regulations. To analyse a baseline in a forward-looking scenario, one has to make conjectures about the evolution of the market, whether courts will affect the regulatory regime under discussion (for example, some rules may be found in breach of World Trade Organization standards) and whether other rules in different policy sectors will most likely change, thus having an effect on the sector being examined. The work done in analysing a baseline provides a double dividend. On the one hand, it assists the policy formulation team in providing information on what leads the public, the media or stakeholders specifically to call for change. On the other, by identifying some sources and types of information, it provides usable knowledge and analytic capacity that will later be used to appraise alternative options. Baseline scenarios are not always disclosed explicitly. Nevertheless, there is always an underlying rationale which motivates decision makers to engage in regulatory action.

11.6.  The Formulation of Options Having established that RIA considers more than one option (regulatory and non-­ regulatory), and that the empirical description of a baseline is indeed a way to appraise the do-nothing option, how are options identified and appraised? It is useful to start by considering the overall policy agenda of the department or agency. In several countries, departments formulate either a broad policy agenda or a specific regulatory policy agenda for discussion within the cabinet and in parliament. Agendas set the overall policy aims, as well as provide information on the political reasons and values behind the choice to intervene in a given policy domain, the legal constraints and the limits for proposed regulation arising out of existing laws. The empirical description of a baseline generates evidence on how to model marginal changes to the status quo given the current laws and policies, as well as the proposals formulated in the departmental or agency-level agendas. A caveat is that when a policy is entirely novel or untried, such as in the case of licensing entirely new chemicals, it is difficult to find objective measures of success or failure.26 24 See in greater detail TD Cook, ‘Lessons Learned in Evaluation over the Past 25’ in E Chelimsky and WR Shadish (eds), Evaluation for the 21st Century: A Handbook (New York, Sage Publications, 1997). 25  See F Fukuyama, ‘What is Governance?’ (2013) 26 Governance 347. 26  This point is explored by JM Mendeloff, The Dilemma of Toxic Substance Regulation: How Over-Regulation Causes Under-Regulation at OSHA (Cambridge MA, MIT Press, 1988).

198 

Stephan Naundorf and Claudio M. Radaelli

With the identification of alternative options, we can move one step forward by assessing the pros and cons of the different choices against the baseline. Problem trees used in the previous steps can be used to establish the cause–effect relationships of negative and undesirable features of the baseline. Given these various elements, that is, the legal-political context, problem identification and overall objectives, the various options identified are evaluated based on a causal logic of intervention, in the form of hypotheses such as ‘if we use option A, we will have these effects on stakeholders 1, 2, 3, income distribution, gender and the environment’. It is exactly this set of expected results of the intervention that have to be appraised transparently and by use of proportionate analysis. (Note that further analysis, although often advisable, has to be justified in relation to the magnitude of the potential effects.) Specific objectives underlying the options are often designed drawing on the SMART template, that is, objectives have to be: —— Specific: precise and concrete enough not to be open to drastically different interpretations. —— Measurable: the future state of the world is presented in measurable terms, and verifiable in terms of results achieved by the intervention. —— Accepted: if objectives and targets are to influence behaviour, they must be accepted, understood and interpreted. —— Realistic: showing how the logic of intervention is realistically leading to results, given the resources available, the enforcement strategy chosen, and the assumptions of compliance. —— Time-dependent: it is useful to set a time limit or the target(s) will remain vague and difficult to monitor. This process moves the RIA from the general strategy underlying the logic of intervention to the identification of relevant options, and finally to the in-depth analysis of options that are not only ‘relevant’ in the sense that they can achieve specific objectives, but also feasible. This step therefore leads to the identification of feasible options. The literature27 has catalogued a number of options. Technically, this catalogue is known as the list of regulatory instruments, or simply policy instruments. Recent experiments show how regulatory policy objectives can be reached by using social interaction rather than regulation. There is a whole agenda concerned with the adoption of cognitive and behavioural economics within impact assessment methods in the EU.28 Options (regulatory and non-regulatory) can be situated on a continuum from more intrusive to more respectful of markets and individual choice: 1. Command and control regulation. This is the domain of prescriptive regulation. Although prima facie attractive because clear, standardised and malleable in relation to given public health and environmental policy objectives, this type of regulation, especially uniform standards and generalised prohibitions, has been criticised. Following Gunningham and Grabosky,29 excessive command and control regulation leads to increasing administrative complexity and the proliferation of laws and regulation. One 27 N Gunningham and P Grabosky, Smart Regulation: Designing Environmental Policy (Oxford, Oxford ­University Press, 1998). 28  See A Alemanno, A-L Sibony, Nudge and the Law (Oxford, Hart Publishing, 2015). 29  Gunningham and Grabosky (n 27) 46.

Regulatory Evaluation

2.

3.

4.

5.

6.

 199

smart way to use command and control regulation is within the technique of regulatory pyramids (see below). Self-regulation. Typically, self-regulation involves an industry association setting rules, codes of conduct and standards relating to the behaviour of firms in the industry. Selfregulation techniques are prominent in environmental policy, but have been tried in other domains too. In the aftermath of the recent economic recession, there has been a discussion as to whether federations of banks should issue codes of conduct to minimise tax avoidance, maximise transparency and ensure banks have a clear understanding of the behaviours that tax authorities expect from them. Deregulation and simplification. It is often forgotten that one regulatory policy option is deregulation, for example in areas where utilities or media are heavily regulated by domestic laws, possibly in conflict with the paradigm of open markets and free trade set by the World Trade Organization. In the UK, the Advocate General ruled in 2015 that legislation for minimum unit pricing for alcohol in Scotland breaches free trade and is disproportionate to the public health protection claims. In the EU, the European Court of Justice has been a very active player in forcing governments to deregulate their media and telecoms markets. Voluntary programmes that, instead of entailing social control by an industry association, are carried out unilaterally by individual businesses. Typically these programmes are voluntary contracts between equal partners, often the government and the firm. This approach has been used in the Australian Landcare programme and the US Environmental Protection Agency’s Green Lights initiative on energy-efficient lighting. Negotiated regulation30 (reg-neg) brings voluntarism forward to set rules through which standards, levels of compliance and (sometimes) legal definitions of best available technology are negotiated between agencies and firms. In the USA, the Administrative Procedure Act was amended in 1990 to allow for reg-neg to become a means of setting regulatory standards. Two well-known (US) projects in the reg-neg family are the Common Sense Initiative and Project XL (dealing with the negotiated flexible implementation of regulatory standards). (Government/agency-sponsored) Education and training campaigns. Since the aim of regulatory policy is to change behaviour, education campaigns have strong appeal as an option. Especially if they exploit the self-interest of the target audience, educational instruments deliver improved management practices.31 Regulators can also ‘nudge’ the target audience by educating and releasing information on the consequences of individual behaviour. Apparently small alterations of the information set available to individuals may have large consequences for regulatory results.32 In Europe, the UK has used nudge-type approaches to regulation with the Behavioral Insights Team.33 Germany is also applying similar methods with the ‘wirksam regieren’ team at the ­Federal Chancellery, which supports ministries with behavioural insights. There is also a lively debate on the uses of nudge in EU public policy34 and the difference between

30  C Coglianese, ‘Assessing Consensus: The Promise and Performance of Negotiated Rulemaking’ (1997) 46 Duke Law Journal 1255. 31  Gunningham and Grabosky (n 27) 61. 32  CR Sunstein and RH Thaler, Nudge: Improving Decisions about Health, Wealth, and Happiness (New Haven, Yale University Press, 2008). 33  P John, ‘Policy Entrepreneurship in UK Central Government: The Behavioural Insights Team and the Use of Randomized Controlled Trials’ (2013) 29 Public Policy and Administration 257. 34  Alemanno et al (n 28).

200 

Stephan Naundorf and Claudio M. Radaelli

nudging and persuasion—the latter providing stronger foundations for permanent behavioural change.35 7. Information, such as corporate reporting, award schemes and public registry of toxic substances. A major study on the US Toxic Release Inventory Program by James ­Hamilton36 shows that the publication of data affects the stock value of firms, influences newspapers stories and voting behaviour, and most importantly affects the behaviour of firms. 8. Economic and market-based instruments, such as property rights, market creation (where no market previously existed), liability and performance bonds (which ‘require regulatees to post a security deposit which is redeemable upon satisfactory completion of a required task’).37 An area where property rights have been used effectively is the protection of wildlife with commercial value. In this domain, property rights create an incentive to maintain the species and their habitats.38 Market creation is theoretically strong, but practically there are difficulties such as transboundary effects, monitoring and enforcing permits, and income distribution inequality. 9. Free markets. This argument has been put forward with reference to environmental policy by Terry Anderson and Donald Leal.39 It goes further than the creation of market-based instruments by the government. It entails the substitution of free markets and commercialisation for any type of government-led and/or legislative solution. The position can be extended from environmental policy to any type of regulatory policy. It should be considered as one of the feasible options only under conditions of relatively perfect market information (this is the extreme and unusual situation in which ­transaction costs are low, and there are no externalities and difficulties in pricing goods). We know from the literature on smart regulation40 that effective regulatory choice in ­complex areas is rarely based on a single option, such as voluntarism or command and control regulation. More often than not, when problems are relatively complex, regulators resort to a mix of options. They may use a modicum of classic regulation in combination with substantial degrees of self-regulation, for example. One way to produce smart RIAs is to assemble options within a mix. Obviously, a policy mix is more or less effective depending on the internal coherence of its ingredients. The individual options within a policy mix can interact synergistically or create contradictory effects. This is yet another opportunity to reconsider critically the logic of intervention in the light of a policy mix. At the stage of consultation, it should be clear to the stakeholders if the government is consulting on individual options or on policy mixes. In major RIAs carried out in the USA, Canada and the EU, policy mixes are often considered and chosen.

35  F Mols et al, ‘Why a Nudge Is Not Enough: A Social Identity Critique of Governance by Stealth’ (2014) 54 European Journal of Political Research 81. 36  JT Hamilton, Regulation through Revelation: The Origin, Politics, and Impacts of the Toxics Release Inventory Program (Cambridge, Cambridge University Press, 2005). 37  Gunningham and Grabosky (n 27) 79. 38  Gunningham and Grabosky (n 27) 70–71. 39  TL Anderson and DR Leal, Free Market Environmentalism (New York, Westview Press, 1991). 40  R Baldwin, ‘Is Better Regulation Smarter Regulation?’ [2012] Autumn Public Law.

Regulatory Evaluation

 201

A final point on the identification of options, partly connected to the notion of policy mix, is to outline options within a logic of intervention shaped by a responsive, dynamic approach.41 The interaction starts with the regulator offering to the regulatee trust and soft options, such as codes of conduct. If the regulatee complies with the code, the regulator rewards this behaviour and does not escalate up towards harder regulatory options. If, in a given industry, the number of firms that do not comply with the soft option is greater than expected, the threat of a command and control regulation is used, up to the point of withdrawing licences and criminal enforcement of law. This strategy is often called a regulatory pyramid, since the regulator starts from soft and market-conforming options at the base of the pyramid, and moves up only if compliance does not corroborate the assumptions about trust and virtuosity made initially. In practice, this type of evidence is not always disclosed explicitly. However, everybody involved in the preparation of regulations considers different options in a more or less systematic way. Transparency and the involvement of stakeholders will help to enhance the quality of the regulation itself and provide the basis for the political accountability which is core for democratic systems following Fukuyama.42

11.7.  Data Gathering Data collection is a delicate step in regulatory evaluation (ex ante and ex post), especially in countries with limited statistical information. It is therefore essential to plan how to collect the data and how to use them. At the same time, the establishment of proper RIA structures and processes is a capacity-enhancing strategy. RIA, in fact, creates the framework within which regulators can increase their capacity to collect and understand data. Some countries utilise their statistical offices for these purposes. Open data and/or open government initiatives also lead to increased accessibility of data. In all situations, consultation with stakeholders helps to raise the quality of data, as long as there are good mechanisms to avoid being captured by the regulated firms. Those who are likely to be regulated have an incentive to magnify the costs and distort the time profile of the costs and benefits— showing that costs hit fast and hard, whilst benefits belong to a distant and very uncertain future. Bureaucracies must develop their own internal capacity to appraise the quality of data provided by the regulated firms. The major steps in data gathering are the identification of sources, the selection and validation of data, and the link between data gathering and the consultation stage. Transparency and accountability are fundamental in this stage of RIA. Policy officers have to report on where the data come from, and how they have been validated and interpreted. In some major RIAs, both data and analysis will be subjected to external peer review. If some data are not available, they can be obtained by commissioning studies, organising focus groups,

41  I Ayres and J Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (Oxford, Oxford University Press, 1992). 42  F Fukuyama, The Origins of Political Order: From Prehuman Times to the French Revolution (Macmillan, 2012).

202 

Stephan Naundorf and Claudio M. Radaelli

running surveys—all these approaches have to be considered in light of the principle of proportionate analysis, since they can be very expensive. A cost-effective way to handle data collection is to ask very specific questions on costs in consultation, thus shifting the cost of producing data to those who are being consulted. At the outset, it is useful to start with an inventory of data sources, and keep the record ‘alive’ by constantly updating it, RIA after RIA, and by sharing it with other colleagues and departments. The type of data needed vary in relation to the options that are being considered. If options include regulation by information, we must have some data on how people use or misuse information, such as labelling certain medicines. If we are to consider tradable permits, the first and most essential data gathering requirement is to establish if there is a market for this type of permit, and what its extension or volume is. If the RIA considers self-regulation, we need data on the self-regulating capacity of the key actors, whether organisations of stakeholders have already established some capacity for self-regulation and what the evidence says on this. If we regulate by using prices, we need to know how people respond in term of elasticity of demand and other basic economic indicators. Data related to human behaviour (in markets as well as in relation to risk) have to be handled sensibly. Political psychology and experimental economics have demonstrated that people use heuristics (or ‘rule of thumb’ or educated guesses) to shape their responses to regulation and risk.43 These heuristic methods differ from classic rational behaviour in different ways. When we use data to model the response to a proposed regulation, especially in sectors where uncertainty and risk are important features, we should be aware of these important deviations from classic rational behaviour.

11.8.  Organising Consultation Consultation is a fundamental component of regulatory evaluation and participatory governance. We have already seen several references to consultation in Chapters 4 and 5 of this handbook, hence we shall be brief here. Instead of seeing consultation as one of the steps of RIA and ex post legislative evaluation, it should be considered an ongoing instrument of quality management. Its aim is to provide information, not to bargain or negotiate the substance of the policy. Consultation obliges the regulator to provide information on the likely effects of new regulations, and to ask stakeholders and citizens to provide opinions on how and whether the new regulations will work. Of course, this is not the same as negotiating regulatory policy, or voting for or against a certain clause—or the whole regulation. In the ex post analysis of legislation, consultation is often used to gauge the effectiveness of policy, together with other empirical measures. This way, consultation may reveal the depth of unforeseen effects of existing regulations, thus widening the peripheral view of decision makers. In ex ante analysis, consultation can be activated during problem

43 A Schneider and H Ingram, ‘Systematically Pinching Ideas: A Comparative Approach to Policy Design’ (1988) 8 Journal of Public Policy 61; Sunstein and Thaler (n 32).

Regulatory Evaluation

 203

­ efinition, during the measurement of the baseline or in the process of identification of d feasible options, but it can also be used to check on the logic of intervention, the coherence of a policy mix, the assumptions of enforcement and compliance underlying alternative options, and so on. Consultation can take a special form in the case of experts’ advice. Several countries have adopted rules on the transparency of expertise used in RIA. One important principle of consultation of experts is pluralism and discursive representation, meaning that both mainstream and dissenting voices will be heard in the process. Consultation, however, is both a principle of transparency and a formal requirement. The latter features in administrative procedures acts around the world or in special codes of conduct for consultation. In consequence, when we talk about consultation we refer to the general idea of consulting on all aspects of the RIA and on different stages, including the informal notion of consultation but also the formal process that takes place after the feasible options have been identified. Formal consultation typically lasts between eight and twelve weeks. It is strengthened considerably by publication of the full draft RIA for ‘notice and comment’—a classic and powerful notion of administrative law. In some countries, regulators do not follow the strict notion of notice and comment, but still publish White Papers and consultation papers that provide essential information on feasible options to the stakeholders. In any case, consultation, to be real and effective, has to involve more than one option or one policy mix to allow for a comparative consideration of alternative courses of action. Baseline and status quo should be included and discussed in consultation documents that go out for notice and comment. Consultation has the following roles and plays the following functions: —— It reduces asymmetric information between regulator and regulates. —— It enriches the empirical basis for decision making. A good regulation should always take into account different points of view. —— It strengthens the information basis of the regulatory process, including experts’ advice. —— It increases opportunities for citizens’ participation and democratises the input provided by experts, increasing transparency and accountability. —— Overall, it raises the transparency of the decision-making process. —— It reduces the risk of unintended consequences. —— It improves implementation by spotting at an early stage what can go wrong with implementation. A consultation plan identifies the following elements: —— —— —— —— ——

the objective of the consultation; the elements of the RIA for which the consultation is necessary; the target groups; the appropriate consultation tool(s); and the appropriate time-line for consultation.

The consultation plan should be prepared as soon as possible. However, it should not be interpreted as a rigid tool. Indeed, it is often modified according to the results of the early conversations with the stakeholders and the needs that emerge during the process of appraising options.

204 

Stephan Naundorf and Claudio M. Radaelli

11.9.  The Analysis of Options In this section we deal with the concept of (mainly economic) analysis of the feasible options. The concept of analysis should not be restricted to theoretical conjectures about the impact of different options. To be effective, analysis has to include specific information of compliance, enforcement and the responses of those who will most likely be affected by the policy. To avoid considerations about implementation is a source of major bias in RIA. To anticipate how market players and citizens will respond to policy is, however, particularly difficult, and for this reason most real-world RIAs augment the economic analysis of options with more general conjectures about behaviour—conjectures that are often difficult to encapsulate within the template of economic analysis. The types of methods which are used typically in different policy sectors, on different levels of a political-administrative system and in different jurisdictions vary widely.44 In general, there is some evidence that the intensity, frequency and direct effectiveness of evaluation is higher when sectors are rather non-political and/or rather technical. In these environments, there is a predominant need to get the best value for money, or at least to collect good arguments to defend the chosen solution. Techniques like cost–benefit analysis have been used in specific domains of environmental policy and foodstuff regulation, where there are independent regulators in charge of a clearly demarcated policy domain, with relatively stable audiences of stakeholders. Even where they work well, all techniques are contingent on the quality of data: issues of reliability, construct validity and (for r­ andomised trails) external validity loom large. Best practice also suggests that sensitivity analysis should be performed to check whether the results are robust.45 The major elaborated techniques are: —— —— —— —— —— ——

cost–benefit analysis (CBA); multi-criteria; cost-effectiveness analysis; compliance cost assessment (CCA); risk–risk analysis; and modelling highly improbable high impact risks with workshop scenarios (including experts and stakeholders).

They can be used together in iteration in the same RIA process. As shown by the case study in Box 2, ‘options’ are not stable, fixed entities in the process. Analysis may show that a given option leads to the fine-grained consideration of sub-options A and B, as shown by the high hedges example from the UK below.

44  T Widmer, W Beywl and C Fabian, Evaluation (Wiesbaden, Verlag für Sozialwissenschaften, 2009) gives unique comparative insights in their systematic handbook on evaluation. They offer a systematic comparison of evaluation experiences in Austria, Switzerland and Germany, not only regarding the different institutional settings in each of the countries, but also in each of the following policy sectors: agriculture, labor market, education, energy and environment, international cooperation for development, science and technology, health, institutional politics, special development, and social care. 45  A Saltelli, K Chan and EM Scott, Sensitivity Analysis (New York, Wiley-Blackwell, 2008).

Regulatory Evaluation

 205

Box 2 Analysis of Options: High Hedges An estimate in the UK revealed that in 1999 there were around 17,000 unresolved neighbour disputes over problems caused by overgrown plant/shrub hedges. The most frequent issues were light obstruction and loss of visual amenity. A consultation paper found that the procedures in place were not effective in settling these disputes. As reported by the RIA carried out by the office of the Prime Minister, common law rights entitled people only to cut overhanging branches; they did not allow them to reduce the height of a neighbouring hedge. Existing regulatory controls, such as those relating to statutory nuisance, were unlikely to be appropriate. And people were reluctant to pursue action through the civil courts, not only because of the time and cost but also because the outcome was uncertain.

Alternative options were identified and put forward for consultation in a consultation document on ‘high edges: possible solutions’. The consultation exercise revealed a consistent body of opinion in favour of giving local authorities the right to settle these disputes. Accordingly, provisions were included in the 2003 Anti-Social Behavior Act to give local authority the power to intervene in these disputes with orders to take action against high hedges and fine owners for not taking action. The 2003 provisions also included rights of appeal of the citizens. This case is interesting for three reasons. First, an RIA was carried out on whether to implement the provisions of the 2003 or not, thus giving additional time for reflecting on the regulatory choice and collecting additional evidence on costs and benefits. Second, the RIA both reconsidered options and re-calibrated the option previously chosen, that is, to give power to local authorities. In fact, during the RIA [it] became evident that, apart for the null option of not implementing the 2003 provisions on high hedges, the option of intervention of local authorities could be split in[to] two alternatives. One was to follow classic regulatory procedures in dealing with neighbourhood issues. This would involve the steps of publicity of complaints, comments and representation between parties, and ultimately protracted exchanges and relatively wide consultation on the dispute. Another option, however, was to create a streamlined procedure based on written statements and limited exchanges, allowing for consultation only if necessary. Third, both the non-implementation hypothesis and the two alternatives were subjected to an analysis of economic, social and environmental benefits, trying to integrate the different dimensions of this problem, from the cost of making a complaint to the advantages of community-based mediation, into a cost–benefit framework. The economic costs to local authorities seemed to vary depending on the use of classic administrative procedures or the alternative of the streamlined process. Differences were also factored in, concerning the lower impact in rural areas than in cities. On monitoring, the RIA proposed to review the legislation in five years. Consultation included government departments and agencies, and the RIA showed how the issues raised during consultation were handled. In the end, the lighter option 2B was recommended for choice (see table 20 of the original RIA below).

206 

Stephan Naundorf and Claudio M. Radaelli

Table 20:  Summary of costs and benefits Option 1. Do nothing.

Annual Cost £0.09m to £0.66m17

£1.14m to £5.83m 2A. Implement part 8 of the Antisocial Behaviour Act 2003 following established procedures.

2B. As option 2 but with more streamlined procedures.

£0.74m to £4.35m

Benefits None identified. Transparent process to resolve outstanding hedge disputes. Role of local authority, plus appeals procedure, provide certainly of fair and impartial resolution. improved amently for complainants. Better neighbour relations. Better managed hedges. Fewer new hedge disputes. Benifits as for Option 2A but with a lighter regulatory touch.

Source: Office of the Deputy Prime Minister, RIA High Hedges, Implementing part 8 of the Anti-Social Behavior Act, 2003.

CBA as a procedure is based on eight steps: 1. The analysis of the types of stakeholders, and sectors, which are going to be affected by individual options or a policy mix. 2. The quantification of those who are affected—how many firms, how many sectors, how many small firms and so on. 3. The categorisation of costs and benefits for each category of actors. 4. The quantification of costs and benefits. 5. The monetisation of costs and benefits. 6. The identification of a cash flow of costs and benefits across time. 7. The selection of a time-frame relevant to policy, including the usage of a proper ­discount rate, often mandated by government or agency policy. 8. The calculation of CBA indicators that provide information on the net present value and the internal rate of return of different options. Specific handbooks are available on different tools of economic analysis, such as risk–risk analysis and CBA.46 An attempt to combine CBA thinking with ‘progressive’ beliefs about health and the environment has been made by Revesz and Livermore.47 As mentioned, this is the full procedure, but according to the principle of proportionate analysis it may be desirable to stop after the first four or five steps. Clearly, if major data are lacking, it does not make sense to proceed with the calculation of indicators. Another ­variation of CBA is to use probability (rather than deterministic assessments), sensitivity analysis and confidence intervals, and make the level of uncertainty transparent. This is good practice in terms of transparency, although when confidence intervals are p ­ articularly large and overlapping, the choice between one option and the others may become indeterminate. 46  KW Viscusi, JE Harrington and JM Vernon, Economics of Regulation and Antitrust, 4th edn (Cambridge, MA, MIT Press, 2005); EA Posner and MD Adler, Cost–Benefit Analysis: Economic, Philosophical, and Legal ­Perspectives (Chicago, University of Chicago Press Journals, 2001). 47 RL Revesz and MA Livermore, Retaking Rationality: How Cost–Benefit Analysis Can Better Protect the ­Environment and Our Health (Oxford, Oxford University Press, 2008).

Regulatory Evaluation

 207

This discussion of the usage of CBA and the presentation of its results are in line with the discussion of cold and warm CBA.48 Warm CBA is a usage of the technique that is open to the argument that not all costs and benefits are measurable and that the choice of discount rates, and the notion of using discount rates across generations is contested.49 Additionally, warm usages of CBA involve using CBA to show the benefits of regulatory intervention. In essence, the idea is to draw on CBA to inform choice knowing that, for example, some benefits are relevant and must be described although they may not be measurable.50 For example, CBA has been used to protect forests and human life, and some academics have argued passionately for CBA as a major tool for the protection of human health and the environment.51 Monetisation is problematic when some effects can only be described but not measured. In the US, the rule is ‘Where appropriate and permitted by law, each agency may consider (and discuss qualitatively) values that are difficult or impossible to quantify, including equity, human dignity, fairness, and distributive impacts.’52 In a sense, multi-criteria analysis is an extension of CBA, since it appraises options according to different criteria, one being efficiency (the CBA criterion) and the others being, depending on the problem, income distribution, gender balance, CO2 emissions, etc. Multi-criteria analysis is based on a performance or pay-off matrix in which the performance of options A, B, etc are measured against different criteria. One problem is using this technique to finalise the choice of one option. It is often the case that the option that dominated according to one criterion does not dominate the other options when other ­criteria are considered. Some multi-criteria analyses apply weighting to the different criteria (for example, after consultation with the head of agency or elected politicians, efficiency may be valued at 80% and income distribution at 20%). The questions of who should apply the weighting and how are difficult, however. Cost-effectiveness analysis looks at different options assuming that they will deliver the same benefit (which is not measured). It compares the costs of these different options. Examples are the costs of health and safety regulation per number of lives, and the cost of market regulation per new job created. CCA is often limited to the analysis of administrative burdens (standard cost model) today, but its rationale is to capture the overall cost of complying with a regulation. ­Practically, it is used with regard to a single category of actor: the firm or the citizen. This technique shows the direct costs arising from the introduction of new rules. Simple and easy to grasp, it can be used to set targets of cost reduction for the government; however, it remains conceptually weak when compared to CBA or multi-criteria in that it only considers costs and not benefits. The OECD has issued guidance about how to implement CCA.53 How to make a choice among all these methods? As a rule of thumb, CCA can be used to assess the costs of simple rules when there is no need to invest a lot of time and resources in

48 

JB Wiener, ‘Better Regulation in Europe’ (2006) 59 Current Legal Problems 447. Harrington, L Heinzerling and RD Morgenstern, Reforming Regulatory Impact Analysis (Washington, Resources for the Future Press, 2009). 50 For an elaboration of this point, see JB Wiener, ‘Better Regulation in Europe’ (2006) 59 Current Legal ­Problems 447. 51  The best example is Revesz and Livermore (n 47). 52 The White House Office of the Press Secretary, ‘Executive Order 13563—Improving Regulation and Regulatory Review’ (18 January 2011), available at www.whitehouse.gov/the-press-office/2011/01/18/­improvingregulation-and-regulatory-review-executive-order (accessed on 31 May 2015). 53  See OECD, ‘OECD Regulatory Compliance Cost Assessment Guidance’ (Paris, OECD Publishing, 2014). 49  W

208 

Stephan Naundorf and Claudio M. Radaelli

appraising the impact of regulations. CBA can be used for major decisions. Multi-criteria is useful when there are contrasting preferences and goals in the community and the decision makers need more information on the major trade-offs at work in the process of policy formulation. At the same time, it should always be borne in mind that methods and instruments should never be utilised to substitute accountable political decision making, nor should these instruments be utilised to simply justify a solution which has been already defined. For example, certain decisions are based on values, and it is not acceptable to reappraise them in terms of CBA. Think of slavery: its rejection is a political decision that no CBA should ever reconsider. Some human rights receive protection—be it universally, in jurisdictions like the EU or in a single country. This is the highest constitutional protection, and this means ‘protection’ from CBA reasoning and arguments. In conclusion, all these instruments should do nothing more than support the legitimate political process of decision making.

11.10.  Reporting and Communication RIA is a process as well as a document. As a document, it provides information on the rationale behind the policy, the empirical evidence, and how policy choice should be understood and challenged. The presentation of RIA is intimately linked to its usages. It has been noted that ‘aggressive’ presentations of RIA increase the likelihood that elected politicians will not use them. Anne Meuwese reports that the European Parliament has de facto rejected the contents of RIA when the European Commission has presented its findings as the ‘best’, ‘almost incontrovertible’ evidence supporting regulatory choice.54 Sensitivity analysis and more generally high levels of transparency on uncertainty (underlying data and the applicability of methods) are key to a good acceptance and understanding of the RIA. In brief, clear communication maximises the probability that RIA will actually inform governmental decision making, an agency’s policy choices and parliamentary deliberations. Clear communication needs a narrative that takes the reader, with an appropriate degree of effort, through the various stages of the process, from problem definition to the examination and analysis of feasible options. It is also fundamental to spell out as clearly as possible how the concerns raised by the stakeholders during consultation were addressed, why certain data provided by certain associations were or were not considered, and which opinions (of sectors, experts or individual firms) were disregarded in the final choice and why. In general, the issue of communicating RIA involves both internal and external accountability. Internal accountability entails the relationship between the agency or department that carried out the RIA and the decision makers/RIA oversight bodies, often represented by specific quality control units like the Office for Management and Budget in the USA or the Better Regulation Executive in the UK. External accountability is a concern of those readers of RIA who belong to parliaments/legislative branches, civil society organisations and the business community. Given these complex layers of accountability, the p ­ ublication 54 

ACM Meuwese, Impact Assessment in EU Law Making (Dordrecht, Kluwer Law International, BV, 2008).

Regulatory Evaluation

 209

of full preliminary analysis for notice and comment is a major step in the direction of clear and effective communication. Countries that have the requirement to publish RIA for notice and comment before regulations reach the statute book are in a better position. In some countries there are also legal obligations to document the usage of expertise, consultants and special expert bodies. Another example of good practice comes from the presentation of the main findings and the narrative of the RIA process in a succinct document, followed by more specific documents on consultation, economic analysis and other aspects, in e-dockets hierarchically organised for easy access by different types of audiences of RIA. Single RIA portals, like the ones used in the UK and by the European Commission, greatly facilitate both communication and the overall accessibility of individual RIAs, ex post evaluations and supporting documentation.55

11.11.  Quality Assurance Quality assurance systems consist of different aspects, from central scrutiny units that support and check on the quality of regulatory evaluation to external bodies like independent ‘watchdogs’ and national audit offices.56 Historically, courts have been quite important in the development of US policy on risk and cost–benefit analysis, and recently have shown how they could become more visible in setting quality standards for RIA.57 Several European countries have established regulatory oversight bodies, such as Actal in the Netherlands, the Normenkontrollrat in Germany, the Regulatory Policy Committee in the UK and the Impact Assessment Board at the European Commission. In 2015, the Commission revamped the Board, transforming it into the Regulatory Scrutiny Board (RSB). Whilst the Impact Assessment Board of the past was staffed solely by Commission officers, the RSB includes three independent experts. Interestingly for the purposes of this chapter, both the RSB and the Regulatory Policy Committee in the UK consider both RIA and ex post legislative evaluation. Their regulatory scrutiny functions thus cover the whole policy cycle—at least, that is their mandate. In addition to regulatory oversight, quality assurance is facilitated by systems of regulatory indicators that capture the quality of RIAs as well as the quality aspects of overall regulatory reform policies.58 The OECD has published a Regulatory Compliance Cost Assessment Guidance,59 which provides a hands-on guidance for calculating compliance costs. In addition, the OECD Framework for Regulatory Policy Evaluation provides conceptual orientation on ­evaluation across the policy cycle.

55  Commission, ‘Communication from the Commission to the European Parliament, the Council, the ­European Economic and Social Committee and the Committee of the Regions: Better Regulation for Better Results—An EU Agenda’, COM (2015) 215 final. 56  JB Wiener and A Alemanno, ‘Comparing Regulatory Oversight Bodies across the Atlantic: The Office of Information and Regulatory Affairs in the US and the Impact Assessment Board in the EU’ in S Rose-Ackerman and PL Lindseth (eds), Comparative Administrative Law (Cheltenham, Edward Elgar Publishing, 2010). 57  Alemanno et al (n 28). 58  Radaelli and Fritsch (n 13). 59  OECD, ‘OECD Regulatory Compliance Cost Assessment Guidance’ (n 53).

hard Implementation

Design I

II

III

IV

V

Input

Process

Output

Intermediate outcome

Regulatory quality

What resources are committed?

Are requirements for good regulatory practices in place?

Have the good practices been implemented?

Have the good practices helped to get quality regulation?

◾ Eg % of RIAs that comply with formal requirements ◾ Eg quality of postimplementation reviews

◾ Eg % of those involved in the regulatory process that think RIA has improved the quality of regulation as opposed to being a tick-the box-exercise; comparison of costs/benefits of initial regulatory proposal to those of actual regulation that was passed

◾ Eg budget, staffing, expertise

Set strategic objectives for regulatory policy in specific sectors (eg health, environment, education)

Strategic outcome

◾ Eg requirements for objective setting, consultation, evidence-based analysis, simplification, risk assessments (eg indicators of regulatory management type), aligning regulatory changes internationally

◾ Regulatory system

Have strategic objectives for regulatory policy in general been achieved?

Feedback loop

◾ Economic: Net benefits (possibly in comparison with counterfactual and alternatives) ◾ Efficient and streamlined: Compliance costs and burdens (possibly integrated into net benefits; Reduced enforcement costs) ◾ Perception of regulatory quality in general ◾ Improved compliance rate ◾ Transparent and easy to access

Regulatory output Have strategic objectives for regulatory policy in a specific sector been achieved? ◾ Effectiveness – market failure/problem/risk was solved or mitigated ◾ Protection and benefits of the public, responsive and accountable Fact-based and perception-based sector-specific indicators: ◾ Eg Environment (eg SOx and NOx emissions, concentration of air pollutants in urban areas) ◾ Eg Safety ◾ Eg Health

◾ Regulatory impacts ◾ Other factors that influence outcome

Figure 2: OECD Framework for Regulatory Policy Evaluation. Source: OECD, ‘OECD Framework for Regulatory Policy Evaluation’ (Paris, OECD Publishing, 2014), available at http://dx.doi.org/10.1787/9789264214453-en. Permission granted by the OECD in February 2016

Stephan Naundorf and Claudio M. Radaelli

Set strategic objectives for regulatory policy in general

210 

easy

Regulatory Evaluation

 211

11.12. Effectiveness Evaluators are always concerned about whether their results will lead to change of policies—if recommended.60 Evidence-based policy design is the conceptual framework for this type of question. In the real life of ministries, agencies and administrations around the world, the effectiveness of evaluation depends partly on the quality of the evaluation in itself. This chapter suggests approaches and methods in this direction. This is part of capacity building in public administrations and regulatory agencies.61 Then there is the issue of demand. This might be the demand of engaged voters for some political action, the demand of politicians for some convincing solutions, the demand of top administrators for self-­implementing solutions or the demand of regulators to defend their preferred solution. The issue is whether the evaluation results, both ex ante and ex post, meet the demands of decision makers. For this reason, there is a need for evaluators to understand clearly what is expected from them. This includes in particular (besides all methodological capacities) a clear notion of the relationship with elected politicians, and the capacity to communicate facts and figures in clear language.

Further Reading Alemanno, A, Sibony, A-L and Sunstein, CR, Nudge and the Law (Oxford, Hart Publishing, 2015). Anderson, TL and Leal, DR, Free Market Environmentalism (Boulder, Westview Press, 1991). Arndt C et al, ‘2015 Indicators of Regulatory Policy and Governance’, OECD Expert Paper No 2 (Paris, OECD Publishing, 2015). Ayres, I and Braithwaite, J, Responsive Regulation: Transcending the Deregulation Debate (Oxford, Oxford University Press, 1992). Baldwin, R, ‘Is Better Regulation Smarter Regulation?’ [2012] Autumn Public Law 485. Black, J, ‘Regulatory Conversations’ (2002) 29 Journal of Law and Society 163. Coglianese, C, ‘Assessing Consensus: The Promise and Performance of Negotiated ­Rulemaking’ (1997) 46 Duke Law Journal 1255. Coglianese, C, ‘Measuring Regulatory Performance Evaluating the Impact of Regulation and Regulatory Policy’ OECD Expert Paper No 1 (Paris, OECD Publishing, 2012). Cook, TD, ‘Lessons Learned in Evaluation over the Past 25’ in E Chelimsky and WR Shadish (eds), Evaluation for the 21st Century: A Handbook (New York, Sage Publications, 1997). Davies, HTO, Nutley, SM and Smith, PC, ‘Editorial: What Works? The Role of Evidence in Public Sector Policy and Practice’ (1999) 19 Public Money and Management 3.

60  See in greater detail Cook (n 24); I Sanderson, ‘Evaluation, Policy Learning and Evidence-Based Policy ­Making’ (2002) 80 Public Administration 1. 61  Fukuyama (n 25).

212 

Stephan Naundorf and Claudio M. Radaelli

Dunlop CA and Radaelli CM (eds), Handbook of Regulatory Impact Assessment (Cheltenham, Edward Elgar Publishing, 2016). European Commission,‘Communication from the Commission to the European ­Parliament, the Council, the European Economic and Social Committee and the ­Committee of the Regions: Better Regulation for Better Results—An EU Agenda’ COM(2015) 215 final. Fukuyama, F, The Origins of Political Order: From Prehuman Times to the French Revolution (London, Profile Books, 2011). Fukuyama, F, ‘What Is Governance?’ (2013) 26 Governance 347. Fukuyama, F, Political Order and Political Decay: From the Industrial Revolution to the ­Globalisation of Democracy (London, Profile Books, 2014). Gunningham, N and Grabosky, P, Smart Regulation: Designing Environmental Policy (Oxford, Oxford University Press, 1998). Hamilton, JT, Regulation through Revelation: The Origin, Politics, and Impacts of the Toxics Release Inventory Program (Cambridge, Cambridge University Press, 2005). Harrington, W, Heinzerling, L and Morgenstern, RD, Reforming Regulatory Impact Analysis (Washington, Resources for the Future Press, 2009). Heilmeyer, M, Schurig, G and Wimmer, C, Äpfel fürs Volk (Potsdam, Vacat, 2002). Hood, C, Rothstein, H and Baldwin, R, The Government of Risk: Understanding Risk ­Regulation Regimes (Oxford, Oxford University Press, 2001). Jacob, K, ‘Evaluating Impact Assessment, Final Activity Report’ (2006), available at http:// cordis.europa.eu/docs/publications/1243/124376871–6_en.pdf (accessed on 16 May 2016). John, P, ‘Policy Entrepreneurship in UK Central Government: The Behavioural Insights Team and the Use of Randomized Controlled Trials’ (2013) 29 Public Policy and ­Administration 257. Meuwese, ACM, Impact Assessment in EU Law Making (Dordrecht, Kluwer Law International, BV, 2008). Micklethwait, J and Wooldridge, A, The Fourth Revolution: The Global Race to Reinvent the State (Harmondsworth, Penguin Books, 2015). Mols, F et al, ‘Why a Nudge Is Not Enough: A Social Identity Critique of Governance by Stealth’ (2014) 54 European Journal of Political Research 81. OECD, ‘Recommendation of the Council of the OECD on Regulatory Policy and ­Governance, Which Was Adopted 22 March 2012’ (Paris, OECD Publishing, 2012). OECD, ‘OECD Framework for Regulatory Policy Evaluation’ (Paris, OECD Publishing, 2014). OECD, ‘OECD Publishing’ (Paris, OECD Publishing, 2014). OECD, ‘OECD Regulatory Compliance Cost Assessment Guidance’ (Paris, OECD ­Publishing, 2014). OECD, ‘Regulatory Policy Outlook 2015’ (Paris, OECD Publishing, 2015). Page, EC, ‘The Civil Servant as Legislator: Law Making in British Administration’ (2003) 81 Public Administration 651. Page, EC and Jenkins, B, Policy Bureaucracy: Government with a Cast of Thousands (Oxford, Oxford University Press, 2005). Posner, EA and Adler, MD, Cost–Benefit Analysis: Economic, Philosophical, and Legal ­Perspectives (Chicago, University of Chicago Press, 2001).

Regulatory Evaluation

 213

Radaelli, CM, ‘Measuring Policy Learning: Regulatory Impact Assessment in Europe’ (2009) 16 Journal of European Public Policy 1145. Radaelli, CM and de Francesco, F, Regulatory Quality in Europe: Concepts, Measures and Policy Processes (Manchester, Manchester University Press, 2011). Radaelli, CM and Fritsch, O, ‘Measuring Regulatory Performance: Evaluating ­Regulatory Management Tools and Programmes’, OECD Expert Paper No 2 (Paris, OECD Publishing, 2012). Revesz, RL and Livermore, MA, Retaking Rationality: How Cost–Benefit Analysis Can Better Protect the Environment and Our Health (Oxford, Oxford University Press, 2008). Rossi, PH, Lipsey, MW and Freeman, HE, Evaluation: A Systematic Approach, 5th edn (New York, Sage Publications, 1993). Saltelli, A, Chan, K and Scott, EM, Sensitivity Analysis (New York, Wiley-Blackwell, 2008). Sanderson, I, ‘Evaluation, Policy Learning and Evidence-Based Policy Making’ (2002) 80 Public Administration 1. Schneider, A and Ingram, H, ‘Systematically Pinching Ideas: A Comparative Approach to Policy Design’ (1988) 8 Journal of Public Policy 61. Smismans, S, ‘Symposium on Policy Evaluation in the EU Policy Evaluation in the EU: The Challenges of Linking Ex Ante and Ex Post Appraisal’ (2015) 6 European Journal of Risk and Regulation 6. Sunstein, CR and Thaler, RH, Nudge: Improving Decisions about Health, Wealth, and ­Happiness (New Haven, Yale University Press, 2008). Viscusi, KW, Harrington, JE and Vernon, JM, Economics of Regulation and Antitrust, 4th edn (Massachuesetts, MIT Press, 2005). Widmer, T, Beywl, W and Fabian, C, Evaluation (Wiesbaden, Verlag für ­Sozialwissenschaften, 2009). Wiener, JB, ‘Better Regulation in Europe’ (2006) 59 Current Legal Problems 447. Wiener, JB and Alemanno, A, ‘Comparing Regulatory Oversight Bodies across the Atlantic: The Office of Information and Regulatory Affairs in the US and the Impact Assessment Board in the EU’ in S Rose-Ackerman and PL Lindseth (eds), Comparative Administrative Law (Cheltenham, Edward Elgar Publishing, 2010).

214 

12 Maintenance of Rules MARIA DE BENEDETTO

The maintenance of rules is an increasingly relevant function of every kind of regulator in contemporary legal systems. Historically, it has always been present in order to respond to various needs, such as making rules accessible, or correcting and reforming them. ­Alongside maintenance in the strict sense (the specific interventions of compilation, consolidation and revision), there is an increasing need for maintenance in order to ensure the quality of rules: it is necessary to evaluate the effects of rules, mainly because of the fallibility of regulation. Maintenance of rules has therefore become an institutional function which completes (and no longer serves) legislation.

12.1.  Which Maintenance for Which Rules? This chapter argues that new attention has to be paid to the maintenance of rules, where ‘maintenance’ refers to both the narrow traditional processes of reviewing, revising and updating rules (always present and also termed ‘maintenance of the statute book’) and the wide emerging processes of regulatory evaluation, implemented in order to ensure the effective results of rules. Legal systems can be likened to public gardens. They work best when they are properly kept and maintained, and this allows them to be easily used. Without intervention and maintenance, they become unkempt and overgrown, making it difficult for the population to derive any benefit from them. In the same way as in a city garden service, it is indispensable for a government to establish maintenance task forces of ‘gardeners’, experts in the field of drafting as well as in regulatory analysis, in charge of planned activities to support law making and the maintenance of rules. Sometimes, such ‘gardeners’ will intervene with daily activities (narrow maintenance), sometimes they will implement stronger interventions (wide maintenance).

216 

Maria De Benedetto

12.1.1.  What Is Meant by the Reference to Rules? First, the term applies to individual rules, ie legal standards,1 which could be parts of different kinds of legislation.2 An individual rule may also express a regulatory content which imposes obligations (ie a command), which, in turn, affect the activities and organisation of its regulation/legislation targets.3 This regulatory content strongly links the rule to its consequences,4 and also requires a regulatory evaluation of its impact.5 The problem of maintenance has been considered from a more general point of view within the framework of regulatory reform, which demands that governments implement specific activities in order to maintain the stock of rules,6 while regulatory evaluation has become the way by which specifically unintended and unexpected consequences7 of rules are identified. Indeed, if legislation is a tool for regulation, it should be viewed not only as the sum of binding legal provisions designed to achieve political consensus on the good formal quality of a normative text, but also as a collection of regulatory rules which aim to achieve its objectives, solve problems8 and avoid (as far as possible) regulatory failures.9 In this sense, legislation should be structured by ex ante and ex post evaluation, strongly oriented towards being effective,10 and requiring not simply narrow maintenance activities but also the already mentioned wide maintenance: when legislation has a regulatory content it would be better to adopt a regulatory logic.

12.2.  Maintenance of Rules: Purposes and Object Functions of maintenance of rules have been present in legal systems for centuries with the purpose of responding to a diversity of needs. Traditionally, maintenance has been oriented 1  See R Dworkin, Taking Rights Seriously (Cambridge, MA, Harvard University Press, 1978) 24 onwards: ‘rules are applicable in an all-or-nothing fashion’. 2  See C Coglianese, ‘Evaluating the Impact of Regulation and Regulatory Policy’, OECD Expert Paper No 1 (Paris, OECD Publishing, 2012) 8. 3  See M De Benedetto, M Martelli and N Rangone, La qualità delle regole (Bologna, Il Mulino, 2011) 12–13. 4  See C Coglianese and R Kagan (eds), Regulation and Regulatory Processes (Aldershot, Ashgate, 2007) xi; on this point see J Black, ‘Enrolling Actors in Regulatory Systems: Examples from UK Financial Service Regulation’ [2003] Public Law 69. 5  On this point see Coglianese (n 2) 9. 6 Eighth Workshop of the APEC-OECD Co-operative Initiative on Regulatory Reform, Gyeongju, Korea, 9 September 2005, 29. See also, OECD, ‘From Intervention to Regulatory Governance’ Report. 7  Commission, ‘Strengthening the Foundations of Smart Regulation—Improving Evaluation’, COM (2013) 686 final, 3. On this point see C Radaelli and O Fritsch, ‘Evaluating Regulatory Management Tools and Programs’, OECD Expert Paper No 2 (Paris, OECD Publishing, 2012). 8  See SG Breyer and RB Stewart, Administrative Law and Regulatory Policy (New York, Little Brown & Co, 1992) 5–6. 9  See R Baldwin, M Cave and M Lodge, Understanding Regulation. Theory, Strategies and Practice (Oxford, Oxford University Press, 2012) 68. 10  A Ogus, Regulation: Legal Form and Economic Theory (Oxford, Hart Publishing, 2004) 90. See also J-B Auby and T Perroud (eds), Regulatory Impact Assessment/La evaluación de impacto regulatorio (Global Law Press/INAP, 2013).

Maintenance of Rules

 217

towards increasing public accessibility of rules and allowing their correction and reform, as will be discussed later (section III). However, even in a revolutionary historical period, legal systems are never completely new and, in contemporary law, maintenance of the stock of rules is increasingly relevant. Maintenance of rules is indeed a cross-system question, even if comparison of law ­making in different legal systems is not a simple task because ‘the terms used for these activities vary from country to country and should often be understood within a particular legal or administrative context’.11 Maintenance of rules is, however, becoming more important because it operates nowadays in a context characterised by an exponential growth of legislation:12 legislative inflation represents a real menace for the so-called securité juridique,13 which consists of clarity, accessibility and predictability of rules. Maintenance of rules today aims to achieve a further objective by contributing to make law ‘capable of leading to efficacy of regulation’:14 in other words, today, maintenance looks at and focuses on the effects of rules because legislation is an indispensable means of carrying out public policies which require implementation and evaluation.15 In this regard, over the last 20 years, there has been a pressure at international as well as European level to increase the quality of regulation.16 The OECD has stated that ‘better regulation means to adopt regulations that meet concrete quality standards, avoids ­unnecessary regulatory burdens and effectively meets clear objectives’.17 The EU first adopted the concept of ‘smart regulation’18 (which means a regulation ‘about the whole policy cycle—from the design of a piece of legislation to implementation, enforcement, evaluation and revision’) and later the idea of ‘fit for purpose’ regulation19 (because of ‘the current economic situation demands the EU legislation be even more effective and efficient in achieving its public policy objectives … REFIT will identify burdens, inconsistencies, gaps and ineffective measures’).

11  E Donelan, ‘European Approaches to Improving Access to and Managing the Stock of Legislation’ [2009] Statute Law Review 167. See also D Tallon, ‘Codification and Consolidation of the Law at the Present Time’ [1979] Israel Law Review 1; G Drewry, ‘Law Making Systems—How to Compare’ [2008] Statute Law Review 100. See finally the question analysed in W Voermans, ‘Styles of Legislation and Their Effects’ [2011] Statute Law Review 38. 12  On this point see S Eng, ‘Legislative Inflation and the Quality of Law’ in L Wintgens (ed), Legisprudence (Oxford, Hart Publishing, 2002) 65 onwards. 13  Conseil d’Etat, Rapport public 2006—Sécurité juridique et complexité du droit (Paris, La Documentation ­française, 2006). 14  H Xanthaki, ‘Quality of Legislation: An Achievable Universal Concept or a Utopian Pursuit?’ in L Mader and M Tavares de Almeida (eds), Quality of Legislation (Baden-Baden, Nomos, 2011) 81 15  On this point see M Howlett and M Ramesh, Studying Public Policy: Policy Cycle and Policy Subsystems (Oxford, Oxford University Press, 1995). 16  On this topic, an analysis was first developed in De Benedetto et al (n 3) and later in M De Benedetto, ‘“Good Regulation”: Organizational and Procedural Tools’ (2013) 2 Italian Journal of Public Law 235. 17 OECD, Overcoming Barriers to Administrative Simplification Strategies: Guidance for Policy Makers (Paris, OECD Publishing, 2009) 44; OECD, ‘Recommendation on Improving the Quality of Government Regulation’ (Paris, OECD, 9 March 1995); OECD, ‘The OECD Report on Regulatory Reform Synthesis’ (Paris, OECD, 1997) 8. 18  Commission, ‘Smart Regulation in the European Union’ (Communication) COM (2010) 0543 final, 3. See also R Baldwin, ‘Is Better Regulation Smarter Regulation?’ [2005] Public Law 485. 19 See Commission, ‘EU Regulatory Fitness’ (Communication) COM (2012) 746 final, 3; Commission, ‘Strengthening the Foundations of Smart Regulation—Improving Evaluation’ (Communication), COM (2013) 686 final, 3.

218 

Maria De Benedetto

The question could be synthetically described in terms of a ‘good’ regulatory regime,20 strictly connected to enforcement (and compliance) of rules.21 In other words, the issue is that regulation ‘seeks to change behaviour in order to produce desired outcomes’.22 This led to the consideration of the quality of legislation as a problem of making ­legislation clear and accessible, but also of making it as ‘easy to comply with as possible’.23 We could say that there is a ‘symbiotic relationship between the formulation of regulatory rules and their application’.24 The primary object of maintenance activities, as already mentioned, is the individual rule. It is clear, however, that ‘regulation can refer either to individual rules or collections of rules’,25 but the rule constitutes the basic element of regulation and so is the object of enforcement,26 of evaluation27 and ultimately of maintenance. In this light, the legislative process (which traditionally focuses on the creation of statutes or other kinds of legislation) should adopt the larger and more comprehensive logic of regulation: ‘a good law is simply a law that is capable of achieving the regulatory reform that it was released to effectuate or support’.28 It is useful, at this point, to use some examples. First, a constitutional reform in France stated in 2008 that the French Parliament is obliged to carry out public policy evaluation sessions, dedicating one week per month to this kind of activity.29 For this purpose, the Comité d’évaluation et de contrôle des ­politiques publiques30 was established inside the Assemblée Nationale: it works on the basis of an annual programme, and can be supported by external experts and by the Cour des Comptes, giving its advice on the études d’impact presented by the government.31 In the period between July 2010 and February 2012, 11 reports were adopted and published.32 20 

Baldwin et al (n 9) 38. OECD, ‘Reducing the Risk of Policy Failure: Challenges for Regulatory Compliance’ (Paris, OECD, 2000). See also W Voermans, ‘Motive-Based Enforcement’, Working Paper (Leiden University, 23 March 2013). 22  C Coglianese, ‘Measuring Regulatory Performance: Evaluating the Impact of Regulation and Regulatory Policy’, OECD Expert Paper No 1 (Paris, OECD Publishing, 2012) 9. 23  Commission, ‘EU Regulatory Fitness’ (n 19) 9. See also OECD, Better Regulation in Europe: Italy 2012—Revised Edition (Paris, OECD Publishing, 2013) 101 (compliance, enforcement and appeals). 24 Ogus (n 10) 90. See also K Hawkins and JM Thomas, Enforcing Regulation (Boston, Kluwer-Nijhoff ­Publishing, 1984) 173. 25  Coglianese (n 22) 8. 26  See Baldwin et al (n 9) 230: ‘Not all kinds of rule can be enforced with the same degree of success.’ 27  Coglianese (n 22) 9. 28  Xanthaki (n 14) 81. See also S Weatherill, ‘The Challenge of Better Regulation’ in S Weatherill (ed), Better Regulation (Oxford, Hart Publishing, 2007) 19. 29  The loi constitutionelle du 23 Juillet 2008 established in Art 24 that ‘Le Parlement vote les lois, contrôle l’action du Gouvernement et évalue les politiques publiques’. On the specific point of Impact Assessment see J Maia, ‘Outline of the French Practice of Regulatory Impact Assessment (RIA) System Preliminary to the Legislative Process, One Year After the Enactment of the New 2008 Constitutional Framework’ in Auby and Perroud (n 10) 167. 30  The Règlement de l’Assemblée National was reformed in 2009 (Art 146-2/146-7). Also, the Loi Accoyer (Loi No 2011-140 du 3 février 2011) was adopted ‘tendent à renforcer le moyens du Parlement en matèrie de contrôle de l’action du Gouvernement et d’évaluation des politiques publiques’. 31  The recommendations of the CEC are transmitted to the First Minister or to the Minister responsible. On this point see P Avril, ‘Le contrôle. Exemple du Comité d’évaluation et de contrôle de politiques publiques’, available at in www.juspoliticum.com. 32  On this point see ‘Rapport d’information déposé en application de l’article 146–3 du Règlement sur le bilan d’activité du Comité d’évaluation et de contrôle des politiques publiques de 2009 à 2012’, where the presented reports are indicated. 21 

Maintenance of Rules

 219

The presented reports show interesting evidence33 about specific issues and seem to express a serious commitment to public policy implementation, with an attempt to fix and describe policy objectives, indicating specific recommendations: for example, in the case of the Rapport sur les autorités administratives indépendantes, after information had been ­collected,34 27 recommendations on specific aspects of rationalisation, independence and control were indicated. In other words, the French Parliament, traditionally occupied with law making, is now in charge (at least, more so than in the past) of the task of public policy evaluation and ­evaluation of legislation, performing a sort of law maintenance activity. A second example regards the already mentioned EU Regulatory Fitness (REFIT) programme, which established that, starting from 2014, ‘the Commission will not examine proposals in areas of existing legislation until the regulatory mapping and appropriate subsequent evaluation work has been conducted’.35 Furthermore, in order to guarantee EU regulatory quality, the REFIT programme conclusively provides that ‘EU legislation and the national rules that implement it, must be managed in a manner that ensure it continues to efficiently achieve its public policy objectives’.36 Even in this example, such management of legislation evokes a maintenance activity which should be performed by EU institutions and by Member State governments. Many other examples could be made from specific national experiences in which ex ante as well as ex post evaluation of legislation have been provided in the EU,37 the UK or the US,38 Australia, Canada or the Netherlands,39 Eastern Europe40 or elsewhere. In this regard, many comparative reports have been prepared by national41 and international42 ­organisations, as well as by academics.43 In conclusion, maintenance, which was in the past an additional and incidental activity, is becoming an essential part of ‘good’ legislative process (as a consequence of ‘good’

33 

Reports are often accompanied by an annexe. Rapport d’information fait au nom du Comité d’évaluation et de contrôle des politiques publiques sur les autorités administrative indépendantes, 28 octobre 2010, No 2925, tome I, 24. 35  Commission, ‘EU Regulatory Fitness’ (n 19) 4. 36  Commission, ‘EU Regulatory Fitness’ (n 19) 11. 37  On this point see C Radaelli and F De Francesco, Regulatory Quality in Europe, Concepts, Measures, and Policy Processes (Manchester, Manchester University Press, 2007). See also House of Lords, European Union Committee, 4th Report of Session 2009–10, ‘Impact Assessment in the EU: Room for Improvement’ (9 March 2010). 38  See C Radaelli, O Fritsch, L Schrefler and A Renda, ‘Comparing the Content of Regulatory Impact ­Assessments in the UK and the EU’ (2013) 6(33) Public Money and Management 445. See, also A Renda, Impact Assessment in the EU. The State of the Art and the Art of the State (Brussels, Centre for European Policy Studies, 2006). 39  An interesting comparison is available in Australian Government, ‘Identifying and Evaluating Regulation Reforms’, Productivity Commission Research Report (December 2011) Appendix K. 40  See K Staronova, ‘Regulatory Impact Assessment: Formal Institutionalization and Practice’ (2010) 30(1) Journal of Public Policy 117. 41  See Australian Government (n 39). 42  See S Jacobzone, C-W Choi and C Miguet, ‘Indicators of Regulatory Management Systems’, OECD Project on Quality Indicators in Government Working Paper 4 (Paris, OECD Publishing, 2007). 43  See C Radaelli, ‘The Diffusion of Regulatory Impact Analysis in OECD Countries: Best Practices or LessonDrawing?’ [2004] European Journal of Political Research 725. See, recently, E Donelan, ‘Progress and Challenges in Selected OECD and EU Countries in Developing and Using Regulatory Impact Assessement (RIA)’ in Auby and Perroud (n 10) 125. 34 

220 

Maria De Benedetto

­regulation),44 achieving formal relevance and requiring a specific time in order to be performed.

12.3.  Historical Roots of Maintenance of Rules Over the centuries, the reasons that have prompted maintenance activities have very often been responses to the need to ensure concrete accessibility to rules, so that citizens could both avail themselves of them (at different stages—law as enacted or as amended—and at ‘a particular point in time in the past’)45 and understand them. The greatest problem in the Roman legal system was precisely the excessive number of statutes, from which was born the need to bring together legislation.46 The idea of the irreducibility of law to codification had already been conceived at this time, and has come down to us through the centuries. In fact, it has been affirmed that codification is a simulation of a definitive statement and of a completeness which in reality it does not hold, and that codification makes for rigid legal systems.47 In the Middle Ages, in Italian communes, statutarii (or correctores or emendatores) were charged with the task of updating, reviewing, correcting and making coherent legislation.48 Their ‘delicate activity’—necessary in order to oppose the enormous growth of statutes— was sometimes undertaken in a sort of conclave, without any opportunity to communicate with the external world until the work had been finished. In other cases, statutarii were compelled to engage in real, though rudimentary, consultation processes. For a long time, maintenance of rules has been considered as a less noble activity than law making, and as being in the service of law making. This was the consequence of a ‘lawcentred’ approach, or legolatria49—an inheritance of the Enlightenment tradition.50 The modern idea of code was affirmed by the Napoleonic codifications, which aimed at completeness, systematisation and durability of law.51 As a consequence, it reduced the legitimacy and the concrete feasibility of maintenance, in favour of statutory interpretation.

44  On this point see L Mader, ‘From the Struggle for Law to the Nurture of law Making—Recent Efforts by the Swiss Confederation to Improve the Quality of Legislation’ (2009) 50 Legislaçao 296. 45  R Cormacain, ‘Accessing Legislation: 40 Years Post-Renton’ (2013) 19(3) European Journal of Current Legal Issues 11. 46  Tacitus (Annales, 3, 25) talked about a ‘moltitudo infinita ac varietas legum’. On the topic of leges collections in Roman law see F Schulz, Principles of Roman Law (Oxford, Clarendon Press, 1936) 9. See also Donelan (n 11) 167; E Steiner, ‘Codification in England: The Need to Move from an Ideological to a Functional Approach— A Bridge Too Far?’ [2004] Statute Law Review 218. 47  See Schulz (ibid) 13. 48  On this issue, see F Calasso, Medioevo del diritto—I. Le fonti (Milan, Giuffrè, 1954) 424. See also L Armstrong and J Kirshner, The Politics of Law in Late Medieval and Renaissance Italy (Toronto, University of Toronto Press, 2011) 79. 49  The expression ‘legolatria’ has been used by P Grossi, L’Europa del diritto (Rome, Laterza, 2007) 134; see also P Grossi, A History of European Law (Oxford, Blackwell Publishing, 2010). 50  Conseil d’Etat (n 13) 259, reference to ‘la croyance solon la quelle la loi vien à bout des difficultés, qu’elle est une solution, une garantie ou, en quelque sorte, une thérapeutique’. 51  On this issue, see Steiner (n 46) 209. See also Donelan (n 11) 167.

Maintenance of Rules

 221

In contrast, in the legal systems of English-speaking countries, an esprit de non ­codification,52 oriented towards not codifying law, was affirmed and law revision tools were introduced early on, in order to respond to the need for the maintenance of rules. In civil law legal systems, codification aims, in a certain way, to reduce or even eliminate the need for maintenance. Nonetheless, alongside the activity of law making, the activity of maintenance of rules has become progressively more relevant, working in parallel with the process called ‘decodification’,53 in particular since the 1980s.54 Nowadays, the code has gained renewed vitality, albeit with fewer ambitions of completeness, specifically in order to achieve narrow maintenance objectives and the formal quality of rules: ‘today the aim of the codes is to make the law more accessible and coherent’.55 Sectorial codes have frequently been adopted in public law issues, in order to maintain law in specific and limited fields of regulation, which are characterised by copious regulation.

12.4.  How Maintenance Is Changing The question of maintenance has regained centre stage due mostly to pressure from supranational and international organisations, in particular the OECD, which has proposed a number of tools to achieve wider maintenance, including impact assessment, consultations and ex post evaluation.56 For this reason, the concept of maintenance has also been reconceived: it is no longer a remedy for a pathology (such as difficulty in accessing rules or the need for correcting and revising them), but has become a real institutional function, an essential part of the legislative process. The objectives of maintenance are now more defined and more structured. They not only respond to the generic requirement of accessibility, correction (or bringing up to date) and simplification of rules, but also aim to ensure the continuous adequacy of rules through evidence-based decision making and evaluation. Maintenance has become functionally autonomous in the sense that it is no longer in the service of law making but is an essential part of the legislative process and expects its own rules, specific experts and dedicated offices, ie law revision commissions (see section IV.C).

12.4.1.  Reasons for Change As we have seen, the enforcement of a rule is necessary in order to allow the rule to achieve its objectives. This is determined both by the formal quality of the rule (the rule is formulated in a coherent, simple and clear way) and by its substantive quality (the rule is 52 

G Samuel, ‘L’esprit de non-codification: le common law face au code Napoléon’ [2005] Droits 123. concept of the ‘decodification era’ was developed by N Irti, L’età della decodificazione (Milan, Giuffrè, 1979). On this point see ML Murillo, ‘The Evolution of Codification in the Civil Law Legal Systems: Towards Decodification and Recodification’ (2001) 11 Journal of Transnational Law and Policy 1. 54  R Sacco, ‘Codificare: modo superato di legiferare?’ (1983) I Rivista di diritto civile 117. 55  Donelan (n 11) 147. 56  See C Radaelli, ‘How Context Matters: Regulatory Quality in the European Union’, paper presented for the PSA Conference, Lincoln, 5–8 April 2008, 4. 53  The

222 

Maria De Benedetto

f­ ormulated through ex ante analysis and ex post evaluation, according to the principles of good regulation).57 However, the more rapid obsolescence of rules than in the past contributes to the increased relevance of maintenance. In fact, frequent abrogation or updating of ­anachronistic rules could be necessary because it is impossible to preserve any static equilibrium for long.58 Furthermore, rules might contain defects produced during the law-making process; this could impose their regularisation, integration, correction or even reform. Further, there is an insidious ‘creative compliance’, whereby the scope of a rule is circumvented and its spirit breached to achieve desired results without breaking the formal terms of the rule.59 Finally, the speed of technological innovation has increased, with the parodoxical effect of legislation always playing ‘catch up’, but without success. Obsolescence of rules, defects, creative compliance and technical innovations require adequate responses. Looking in depth at the problem, one of the most pressing needs for maintenance is ­epistemological in nature: the limits of reasoning and of knowledge and the question of unintentional consequences of human behaviour affect even legislation.60 It follows that the consequences of behaviour depend (in part) on what other individuals might choose.61 The perspective that should be adopted by regulators must be fallibilist62 (or antiperfectionist).63 When we adopt a regulation, we have to ensure its effectiveness64 and monitor the life cycle of its individual rules. Legislation should be followed just as we would follow the course of a sickness after a medicine has been administered.

12.4.2.  Maintenance and Quality of Rules Quality of regulation, as mentioned above, is not a new problem, either in the formal quality of rules (related to their coherence, clarity and understandability) or in their substantial quality (related to the effects of the rules). Formal quality is the objective of the modern legislative technique, which aims to ensure the coherence, clarity and understandability of rules: drafting in English-speaking 57 ‘The OECD Report on Regulatory Reform Synthesis’ (n 17) established principles of good regulation. In 2006, the UK Legislative and Regulatory Reform Act, Art 21, established principles of good regulation, based on the Better Regulation Commission’s Principles. These principles inform a Code of Practice issued by the minister: Art 21(2) states that ‘regulatory activities should be carried out in a way which is transparent, accountable, proportionate and consistent and also that regulatory activities should be targeted only at cases in which action is needed’. 58 The issue is analysed in G O’Driscoll and MJ Rizzo, The Economics of Time and Ignorance (London, Routledge, 1996). 59  On creative compliance see R Baldwin, Rules and Government (Oxford, Clarendon Press, 1995) 185; Baldwin et al (n 9) 232. 60  J Rawls, A Theory of Justice [Italian translation] (Cambridge, MA, The Belknap Press of Harvard University Press, 1971) 299. The issue has also been developed by F Von Hayek, The Counter-revolution of Science. Studies on the Abuse of Reason (London, The Free Press, 1952). 61  O’Driscoll and Rizzo (n 58) in particular ‘Introduction’. 62  K Popper, The Open Society and Its Enemies, vol II (London, Routledge & Kegan Paul, 1945) addendum: ‘Every discovery of a mistake constitutes a real advance in our knowledge.’ 63  Antiperfectionism is a theory developed by A Rosmini, Filosofia della politica (1837, Milan, Rusconi, 1985), 137. See also R Dahrendorf, ‘Economic Opportunity, Civil Society, and Political Liberty’ (1996) 27(2) Development and Change 229. 64  On this point see F Jullien, Traité de l’efficacité (Paris, Édition Grosset & Fasquelle, 1996).

Maintenance of Rules

 223

countries,65 legistique formelle in France,66 tecnica legislativa (which has been defined as ‘craftsmanship’)67 in Italy and in Spain. Formal quality has become the object of specific provisions all over the world. For example, many provisions have been set out by legislative assemblies, by governments and by the different Conseils d’Etat, at subnational level (by federal and regional systems), at metanational level (as in the case of the EU) and at the international level (as in the case of OECD and other international organisations).68 Alongside drafting, the science of legislation, starting from the work of Filangieri,69 aims to redesign the way in which ‘good’ legislation is achieved, from the point of view of its content. The science of legislation has gained particular theoretical importance in Germanspeaking countries, with the Gesetzgebungslehre.70 ‘Good’ legislation is also an object of legisprudence, a normative theory of rational legislation, which takes into account both the form of legislation and its content.71 It is widely agreed that quality of rules is nowadays a ‘comprehensive’ question, which implies both formal quality and substantial quality aspects. Law takes charge in the field of law making and law maintenance72 because it regulates the genesis and the effects of regulation. Many legal provisions establish, in some way, that rules must be efficient,73 and that their application is to be measured and evaluated: efficiency becomes subject to quality rules in the regulatory process because ‘the process of law making is, as a rule, regulated by law’.74 There is, in other words, an ‘evaluate first’ principle.75 This situation means that experts are of increasing relevance76 in the law making and law maintaining processes, and there is increasing pressure on political decision makers to justify rules.77 65  See H Xanthaki, Thornton’s Legislative Drafting, 5th edn (London, Bloomsbury, 2013). See also W Dale, Legislative Drafting: A New Approach, a Comparative Study of Methods in France, Germany, Sweden and the United Kingdom (London, Butterworths, 1977). 66  J Chevallier, ‘L’évaluation législative: un enjeu politique’ in A Delcamp, JL Bergel and A Dupas, Contrôle parlementaire et évaluation (Paris, La Documentation française, 1995) 15. See also (for Belgium) Conseil d’Etat, Legistique formelle. Recommandations et formules (November 2001). 67  See (1985) V Il Foro Italiano special edition (entitled ‘La tecnica legislativa: un artigianato da valorizzare’) 20. 68  Broad references on this point are given in De Benedetto (n 16) 240. 69 G Filangeri, La scienza della legislazione [1780-1785] – Benjamin Constant, Comento sulla scienza della ­legislazione, a cura di Vittorio Frosini; revisione critica dei testi di Francesco Riccobono, vols I and II (Rome, IPZS, 1984). 70  The most important contribution on this topic is by P Nöll, Gesetzgebungslehre (Reinbek, Rowohlt, 1973). 71  ‘It concerns both the content of legislation and its form’, L Mader, ‘Evaluating the Effects: A Contribution to the Quality of Legislation’ [2001] Statute Law Review 119. See also L Wintgens, ‘Freedom and Legisprudence— A More Substantial View: A Reply to Professor Perju’ [2009] Boston University Law Review 1795. 72  S Cassese, ‘Introduzione allo studio della normazione’ [1992] Rivista Trimestrale di diritto Pubblico 309, affirmed: ‘la progettazione della legge è estranea all’essenza del diritto’. 73  On this issue see F Denozza, Norme efficienti. L’analisi economica delle regole giuridiche (Milan, Giuffrè, 2002). 74 J Wroblewski, ‘The Rational Law Maker. General Theory and Socialist Experience’ in A Giuliani and N Picardi (eds), L’educazione giuridica, V—Modelli di legislatore e scienza della legislazione, tomo III—la discussione contemporanea (Perugia, Università di Perugia, 1987) 60. 75  Commission, ‘EU Regulatory Fitness’ (n 19) 7. See also Coglianese (n 22). 76  See U Karpen, ‘Law Drafting and the legislative Process: Outline of a Training Course for Law Drafters’ in L Mader and C Moll (eds), The Learning Legislator: Proceedings of the 7th Congress of the Association of Legislation (Eal) (The Hague, Nomos, 2006). 77  See JO Freedman, Crisis and Legitimacy: The Administrative Process and American Government (London, Cambridge University Press, 1978); M Shapiro, ‘The Giving Reason Requirement’ [1992] University of Chicago Legal Forum 179.

224 

Maria De Benedetto

12.4.3. Kinds of Maintenance (Compilation, Consolidation, Revision, Reform) In the different systems of law making there are various uses of terms like codification, consolidation, revision and reform, and there is a real risk that the same word could refer to different activities. This difficulty can be compounded by (often frequent) contamination of different degrees of maintenance inside a single maintenance intervention. It is possible, for example, that a compilation activity would be the occasion for consolidation; in a reform, there would also be rules which were subject to revision alone.78 Difficult as it may be to recognise the kinds of maintenance applied to rules, it is nonetheless possible to distinguish those interventions that have quantitative effects on the stock of legislation from the interventions which are limited to ordering regulation. The first kind regards activities that remove legislation, carried out in some Englishspeaking countries with the Statute Law (Repeal) Act (a normative act with a typical content), which is echoed to a degree by the Italian taglia-leggi (a sort of legislative guillotine).79 Regarding the second kind of maintenance, it is possible to proceed towards a graduation based on the objectives pursued. If the objective pursued is the accessibility of rules, there will be interventions of compilation or consolidation. If it deals with their correction or updating, there will be a revision. If it is the innovation (or simplification) of a regulation, there will be a real reform (or codification). The objective of accessibility to regulation80 could require compilation,81 which is the first type of maintenance: in compilation, normative texts are gathered into a single collection, which retains their separate identities. The Roman leges collection was inspired by the logic of compilation, as we have seen, in the same way as the United States Code, which collates federal legislation. In contrast, through consolidation, scattered rules are gathered together into a single, up-to-date, normative text. Consolidation constitutes a very important category, because it stands in contrast with modern codification, from theoretical and historical perspectives, and because the two are very often confused.82

78  A compilation activity could be the opportunity for a revision, as a reform could consist of a consolidation of existing norms. For example, the US Code refers to activities which are defined as ‘consolidation and revision’ and which consist of compilation and revision, sometimes consolidation; the French ‘codification’ is often consolidation and revision, sometimes a real reform; in the UK, ‘revision’ means abrogation of norms, as described by Donelan (n 11) 167. In Australia, the Legislation Revision and Publication Act (2002) promoted ‘an ongoing program for the revision and publication of legislation’, with the aim ‘to consolidate public general legislation and make up-to-date copies of the public general legislation’. In Italy, there is a distinction between ‘riordino’ and ‘riassetto’ of norms: the first has a conservative character; the second has the objective of a substantial revision of legislation. 79  On this issue, see N Lupo and B Mattarella, ‘Gli sviluppi del “taglia-leggi”: per ora molta immagine e poca sostanza’ [2008] No 161–62 Studi parlamentari e di politica costituzionale 7. 80 Donelan (n 11) 149. The French Conseil d’Etat, in the Guide pour l’élaboration des textes législatifs et ­réglementaires (2007), affirmed: ‘Essentiellement fondée sur une consolidation et une meilleure organisation des normes existantes, la codification tend à faciliter la connaissance et la communication des règles de droit.’ 81 Steiner (n 46) 219: ‘Compilations, designed to bring together into one statute existing laws, either in ­chronological order or by subject, but without altering their form.’ 82  On this issue the historical contribution of Mario Viora is very important: Consolidazioni e codificazioni. Contributo alla storia della codificazione (Turin, Giappichelli, 1967).

Maintenance of Rules

 225

In France, for example, codification—which operates à droit constant—is a consolidation.83 Even in Italy, codification is today substantially a consolidation: here the ‘single text’ (testo unico) has been affirmed. Furthermore, consolidation is also used in the UK system and in the legal systems of other English-speaking countries. If maintenance responds to the need for correcting or updating rules, then it is necessary to resort to revision84 in order to eliminate ‘defects or anachronisms’, such as in the many commissions which operate in the US (eg the State of New York Law Revision Commission).85 In the UK, revision is the technique which imposes the abrogation of useless (sometimes obsolete) rules, the fusion of similar rules and the elimination of anomalies. This process is the specific degree of maintenance which better corresponds to the logic of quantitative interventions on the stock of legislation (the Statute Law Revision Act or the Italian taglia-leggi). Finally, if there is a real need for ‘much more substantial changes’86 in regulation—in other words, if there is a need to codify (and simplify) rules—it is possible to implement a reform (intervention which would take the form of codification, also known as riassetto in Italy), which implies a wide maintenance activity based on regulatory evaluation. The relationship between the technical phase of the decision and the political choice in the maintenance of rules has a different relevance in the first three kinds of maintenance (compilation, consolidation and revision) than it does in the fourth (reform). In the first three kinds, formal quality dominates, as well as technical evaluation and law-making expertise. In the fourth, policy evaluation, guided by the regulatory options, is necessary in order to achieve substantial quality: a regulatory analyst (for the technical decision) and the regulator itself (for the political decision) are necessary here.87

83  The Commission supérieure de codification, established by Décret No 89-647 (1989) was in charge of the task to schedule the codification activities and to adopt and transmit the draft codes to the government. On this point see G Braibant and A Zaradny, ‘L’action de la Commission superiéure de codification’ [2004] Actualite Juridique: Droit Administratif 1856, who have written that ‘la codification actuelle apparaît comme un besoin d’une société démocratique au bord de l’asphyxie normative’. See also OECD, ‘La capacité du gouvernment a produire des réglementations de grande qualité’ (Paris, OECD, 2004), 6. 84  Steiner (n 46) 219 used the word ‘restatement’ in this regard: ‘this type of codification does not necessitate reconsideration of the relevant law with a view to reform, although it may include minor improvements such as the repeal of obsolete texts or the elimination of inconsistencies’. The concept of ‘revision’ was described by Alfred Clapp in the foreword to Title 2A of New Jersey Statutes, as reported in the Annual Report of the New Jersey Law Revision Commission (1989), 16: ‘The task of making a revision is primarily to rewrite where language can be improved upon materially without ripping up the fabric of the law satisfactorily settled, to boil down, to clarify, to eliminate the obsolete, to reconcile the inconsistent, to correct clear errors.’ 85  The State of New York Law Revision Commission was established by the Legislative Law, Art 4-A, implemented with Chapter 597 of the Laws in 1934. Law Revision Commissions have such important tasks as studying and proposing new legislation, revising current legislation and drafting legislative processes. In California there is the California Law Revision Commission (1953), in Michigan the Law Revision Commission (1965), in New Jersey the Law Revision Commission (1985) and in Oregon the Law Commission (1997). 86  A Samuels, ‘Consolidation: A Plea’ [2005] Statute Law Review 59. A codification reform ‘means a complete reconsideration of the law in a particular field with a view to its reform’, Steiner (n 46) 220. 87  De Benedetto et al (n 3) 111.

226 

Maria De Benedetto

12.5.  The Modern Legislator Walks on Two Legs: Law Making and Law Maintenance Having reached the end of this chapter, it is now clear that the modern legislator should walk on two legs—law making and law maintenance—in order to ensure formal quality of legislation as well as the continuous adequacy of rules. This comprehensive viewpoint is the direct consequence of a public policy approach to legislation: if legislation is a tool for public policies and regulation, and if public policies and regulation are processes, then it should be obligatory to produce good-quality legislation as well as to evaluate legislation even after its adoption. In this sense, maintenance of rules is not only a way to ensure accessibility, clarity and coherence of legislation (where rules must have formal quality as a result of the application of drafting techniques), but also the way to ensure the continuous adequacy of rules in achieving regulatory objectives expressing and demanding a logic of continuation.88 ­Legislation should be built to be maintained, thanks to evidence-based decision making89 and to economic analysis of regulation (as in the case for cost–benefit analysis), which could bring greater rationality to public decisions.90 This is also possible through effective consultations, which may anticipate the criticisms or failures of regulation, strengthening the legitimacy of regulation and reducing ­litigation.91 In effect, consultation could be currently perceived as a weak form of public participation, incapable of counterbalancing special interest groups which strongly influence legislation. Nonetheless, they are crucial in a maintenance perspective because they give relevant information about enforcement and compliance. There is, in other words, a real need to improve the positive outcomes of consultation. It is also important to reduce the negative effects of lobbying, ‘labelling’ legislation to make the interests in any given legislation clear and traceable (like ingredients for food): it should also be clear, for each rule, ‘who pays and who benefits’, because it make it simpler to evaluate and maintain legislation. This new, more complex and modern perspective for maintenance of rules goes against the opinion that studying and planning rules is better (and easier) than verifying and implementing them.92 Alongside a number of examples of narrow maintenance in history, there are even more cases of wide maintenance that have been undertaken in order to achieve the continuous adequacy of rules: the already mentioned French constitutional reform (2008), the EU Regulatory Fitness programme (2010), but also the widespread recourse to ex ante and ex post analysis of regulation all over the world.

88 

F Jullien, Traité de l’efficacité (Paris, Grosset & Fasquelle, 1996). See Baldwin et al (n 9) esp 25. On this point see C Coglianese, ‘Empirical Analysis and Administrative Law’ [2002] University of Illinois Law Review 1111. 90  On this point see RW Hahn and PC Tetlock, ‘Has Economic Analysis Improved Regulatory Decisions?’ (2008) 22 Journal of Economic Perspectives 67. 91 Commission, ‘Towards a Reinforced Culture of Consultation and Dialogue’ (Communication) COM (2002) 704. 92  See B Mattarella, La trappola delle leggi. Molte, oscure, complicate (Bologna, Il Mulino, 2011) 8. 89 

Maintenance of Rules

 227

Moreover, it will be important for governments (and any other regulator) to adopt a legislation maintenance strategy that includes both the stock of rules and the regulatory flow. In any case, it is becoming increasingly clear that legislation founded only on the leg of law making seriously risks limping and, so doing, risks failing in its role. In the already mentioned metaphor of legislation as a public garden, the legislator sometimes operates like a city garden service which is responsible for an originally well-designed and planted garden that has later been completely abandoned.

Further Reading Auby, JB and Perroud, T (eds), Regulatory Impact Assessment/La evaluación de impacto regulatorio (Seville, Global Law Press/INAP, 2013). Baldwin, R, Cave, M and Lodge, M, Understanding Regulation. Theory, Strategies and practice (Oxford, Oxford University Press, 2012). Chevallier, J, ‘L’évaluation législative: un enjeu politique’ in A Delcamp, JL Bergel and A Dupas (eds), Contrôle parlementaire et évaluation (Paris, La Documentation française, 1995). Coglianese, C, ‘Measuring Regulatory Performance: Evaluating the Impact of Regulation and Regulatory Policy’, OECD Expert Paper No 1 (Paris, OECD Publishing, 2012). Coglianese, C and Kagan, RA (eds), Regulation and Regulatory Processes (Aldershot, Ashgate, 2007). De Benedetto, M, Martelli, M and Rangone, N, La qualità delle regole (Bologna, Il Mulino, 2011). Karpen, U, ‘Comparative Law: Perspectives of Legislation’ (2012) 6 Legisprudence 149. Mader L, ‘Evaluating the Effects: A Contribution to the Quality of Legislation’ [2001] Statute Law Review 119. Mader, L and Tavares de Almeida, M (eds), Quality of Legislation. Principles and Instruments (Baden-Baden, Nomos, 2011). Ogus, A, Regulation. Legal Form and Economic Theory (Oxford, Hart Publishing, 2004). Radaelli, CM, ‘Diffusion without Convergence: How Political Context Shapes the Adoption of Regulatory Impact Assessment’ (2005) 12 Journal of European Public Policy 924. Radaelli, CM and Fritsch, O, ‘Evaluating Regulatory Management Tools and Programs’, OECD Expert Paper No 2 (Paris, OECD Publishing, 2012). Samuels, A, ‘Consolidation: A Plea’ [2005] Statute Law Review 56. Voermans, W, ‘Styles of Legislation and Their Effects’ [2011] Statute Law Review 38. Voermans, W, ‘Motive-Based Enforcement’, Working Paper (Leiden University, 23 March 2013). Weatherhill, S (ed), Better Regulation (Oxford, Hart Publishing, 2007). Xantaki, H, ‘Quality of Legislation: An Achievable Universal Concept or an Utopian ­Pursuit?’ in L Mader and M Tavares de Almeida (eds), Quality of Legislation. Principles and Instruments (Baden-Baden, Nomos, 2011), 75.

228 

13 EU Legislation WILLIAM ROBINSON*

The EU has, since the 1950s, developed a new legal order which includes a body of legislation applying equally to all 28 Member States. That legislation is largely autonomous, although it generally relies on the rules and authorities of the Member States for its implementation, application and enforcement. EU legislation is adopted by the institutions established by the founding treaties in accordance with a procedure laid down by those treaties. The institutions have developed their own drafting style and established their own drafting rules to ensure consistency. Particular features of EU legislation are that it is authentic in 24 official languages and must be made accessible to users in all 28 Member States, including over 500 million EU citizens.The ultimate authority for interpreting EU legislation is the Court of Justice of the EU, which has developed principles of interpretation taking account of the particular features of EU legislation, which is in some respects comparable to public international law.

13.1. Introduction 13.1.1. Origins The law of the European Union (EU) can trace its origins back to 1951, when the European Coal and Steel Community (ECSC) was established by the Paris Treaty that was concluded by just six countries.1 That Treaty set the institutional framework that is still at the core of the EU today. It consisted of: —— the Assembly, composed of ‘representatives of the peoples’ of the Member States; —— the Council, composed of one minister from each Member State; —— the High Authority, the independent body representing the interests of the ECSC and having the power to adopt acts which applied directly to Member States or to coal and steel undertakings; and —— the Court of Justice, composed of independent judges.

*  Associate Research Fellow, Institute of Advanced Legal Studies, University of London, formerly a coordinator in the Quality of Legislation Team of the European Commission Legal Service. I wish to thank Manuela Guggeis for her invaluable assistance. 1  All EU Treaties and legislation can be found on the EUR-Lex website, http://eur-lex.europa.eu/en/index.htm (accessed on 20 May 2015).

230 

William Robinson

By the two Rome Treaties of 1957, two further European Communities were established to stand alongside the ECSC: the European Economic Community (EEC) and the European Atomic Energy Community (EAEC or Euratom). Those founding Treaties were amended and added to over the years as new countries joined, and the scope of the Treaties was extended to new sectors. The entry into force of the Maastricht Treaty in 1993 saw the creation of the EU and the renaming of the EEC as the European Community (EC).

13.1.2.  Present Position The Treaty of Lisbon, which came into force in 2009, was in the form of an amending Treaty. It made textual amendments to the Treaty on European Union (TEU) and to the Treaty establishing the European Community, which is renamed the Treaty on the Functioning of the European Union (TFEU).2 Following the accession of Croatia in 2013, the EU now has 28 Member States.3 The institutional framework of the EU now comprises, in so far as is relevant to the adoption of legislation, the European Parliament (EP), the Council, the European Council and the European Commission (the Commission). The Court of Justice has now become the Court of Justice of the European Union (CJEU).4 The European Parliament is composed of 751 members directly elected by EU citizens. It holds its plenary sessions in Strasbourg or Brussels and its committee meetings in Brussels. It exercises the EU’s legislative and budgetary functions jointly with the Council. The Council is composed of one minister from each Member State. It meets in different configurations depending on the subject under discussion. A Committee of Permanent Representatives of the Governments of the Member States (COREPER) prepares the work of the Council. The presidency of the Council is held by each Member State on a six-month rotating basis. The European Council is composed of the heads of state or government of the Member States, together with its permanent President and the President of the European Commission. It does not exercise legislative functions, but defines the general political directions and priorities of the EU and provides the ‘necessary impetus for its development’.5 The Commission consists of one Member from each Member State. The Members are appointed for a term of five years and are to be wholly independent. The Commission’s work is led by its President, who is proposed by the European Council and elected by the European Parliament. The Commission acts as a collegiate body in ensuring that the EU Treaties and EU law are applied, in executing the budget, and in exercising executive and management functions.

2 

The ECSC Treaty expired in July 2003. The EAEC Treaty continues its separate existence. In June 2016 a majority of voters in the United Kingdom voted to leave the EU. It is not possible at the time of finalising this text in September 2016 to foresee precisely the consequences of that vote. 4  See Art 13ff TEU. Under that article the other EU institutions are the European Central Bank and the Court of Auditors, while the European Economic and Social Committee and the Committee of the Regions are advisory bodies under Art 300 TFEU. Under Art 132 TFEU the European Central Bank may adopt legal acts in the field of monetary policy. 5  Art 15(1) TEU. 3 

EU Legislation

 231

The CJEU includes the Court of Justice and the General Court comprising judges from each Member State. The Court of Justice is also assisted by Advocates-General, who give advisory opinions before the final judgment. The judges and Advocates-General are appointed for a term of six years and are to be wholly independent. The CJEU ensures that ‘in the interpretation and application of the Treaties the law is observed’.6 It rules on actions brought by Member States, the EU institutions or natural or legal persons and gives preliminary rulings at the request of national courts on the interpretation of EU law or the validity of acts of the EU institutions (see section 8 below).

13.1.3.  A New Legal Order The first president of the Commission, Walter Hallstein, wrote that the European ­Community, the precursor of the EU, ‘is a legal phenomenon in three respects: it is a creation of the law, it is a source of law and it is a legal order’.7 It was created by the basic ­treaties, which have now been superseded by the TEU and the TFEU. But those treaties do little more than set out a basic institutional framework and basic principles, together with goals and mechanisms, and much of the actual construction is done by the legal acts adopted by the EU institutions. It was the CJEU that held that the Treaties had established: a new legal order for the benefit of which the States have limited their sovereign rights, in ever wider fields, and the subjects of which comprise not only Member States but also their nationals … The essential characteristics of the Community legal order which has thus been established are in particular its primacy over the law of the Member States and the direct effect of a whole series of provisions which are applicable to their nationals and to the Member States themselves.8

The EU is a legal order in the sense that the system of rules it has created is not confined to regulating the operation of the EU’s internal market but also guarantees the lawfulness of the actions of the institutions and legal protection of those subject to EU rules.

13.1.4.  General Principles Article 5 TEU sets out basic principles of the EU relevant to its legislative activities: the principles of conferral, subsidiarity and proportionality. Under the principle of conferral, the EU is to ‘act only within the limits of the competences conferred upon it by the Member States in the Treaties to attain the objectives set out therein’. That is a long-standing formulation that was already in the EC Treaty, but the Lisbon Treaty added a further provision: ‘Competences not conferred upon the Union in the Treaties remain with the Member States.’ 6 

Art 19(1) TEU. W Hallstein, Die Europaeische Gemeinschaft (Düsseldorf, Econ Verlag, 1973) ch 2 (I). 8  Opinion 1/91 (Draft Treaty on a European Economic Area) [1991] ECR I-1061. The first reference to a new legal order was in Case 26/62 van Gend en Loos [1963] ECR 1. The case law on primacy dates back to Case 6/64 Costa v ENEL [1964] ECR 585 and that on direct effect to Case 26/62 van Gend en Loos [1963] ECR 1. 7 

232 

William Robinson

Accordingly, the EU institutions cannot adopt any legal act unless there is a legal basis for it in the Treaties, and that legal basis must be specified in the citations in the preamble to the act.9 The choice of legal basis must be objectively justifiable.10 Under the principle of subsidiarity, which was first introduced by the Maastricht Treaty in 1993, the EU is to act only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States, either at central level or at regional and local level, but can rather, by reason of the scale or effects of the proposed action, be better achieved at Union level.

Under the principle of proportionality, the ‘content and form of Union action shall not exceed what is necessary to achieve the objectives of the Treaties’. That principle was formulated by the CJEU as long ago as 195611 and was already in the EC Treaty. The principles of conferral and proportionality apply to all Union action, while the principle of subsidiarity applies only in areas that do not fall within the exclusive competence of the Union.12 The EU institutions must apply the principles of subsidiarity and proportionality as laid down in Protocol No 2. Under Article 5 of that Protocol, draft legislative acts must be justified with regard to those principles and must contain a detailed statement making appraisal of compliance with those principles possible. A key role in monitoring compliance with the principle of subsidiarity before any legislative act is adopted is given to the national parliaments by Protocol No 1. The CJEU takes account of compliance with the principles of subsidiarity and proportionality in actions for review of the legality of EU legal acts under Article 263 TFEU (see section VIII.B below). The Commission publishes annual reports on the subsidiarity and proportionality principles and Better Law making on its website.13 Article 4(3) TEU refers to the principle of sincere cooperation under which: The Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks which flow from the Treaties. The Member States shall take any appropriate measure, general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting from the acts of the institutions of the Union. The Member States shall facilitate the achievement of the Union’s tasks and refrain from any measures which could jeopardise the attainment of the Union’s objectives.

That principle was already set out in the EC Treaty. It forms the basis for the principle of indirect administration under which the EU relies to a large extent on the administration and courts of the Member States to apply, implement and enforce Union law.14

9  Under the Council’s Rules of Procedure, Decision 2009/937/EU [2009] OJ L325/35; see Annex VI, point A.1(c). 10  Case C-300/89 Commission v Council (titanium dioxide) [1991] ECR 2867, para 10. 11  Case 8/55 Federation Charbonniere de Belgique (judgment of 29.11.1956) [1954] ECR 292, 299. 12  The categories and areas of Union competence are defined in Arts 2–6 TFEU. 13  http://ec.europa.eu/smart-regulation/better_regulation/reports_en.htm (accessed on 20 May 2015). 14  See also section II.I below; JC Piris, The Lisbon Treaty, A Legal and Political Analysis (Cambridge, Cambridge University Press, 2010) 97ff.

EU Legislation

 233

Other principles relevant to EU legislation were established by the CJEU, including the principle of the primacy of EU law15 and the principle of direct effect.16

13.2.  Types of Legal Acts 13.2.1. General The EEC Treaty provided for five legal acts, whose essential features remain unchanged today. Article 189 lists three types of binding acts, regulations, directives and decisions, and two types of non-binding acts, recommendations and opinions. In 1993, the Maastricht Treaty added two new sectors of competence, the Common Foreign and Security Policy (CFSP), and Justice and Home Affairs (JHA), for which nine new legal acts were created,17 bringing the total number of legal acts to 15. The differences between the acts were often minor and not self-evident. The Lisbon Treaty rationalised the types of act by cutting them back to the original five: regulations, directives, decisions, recommendations and opinions.

13.2.2. Regulations Under Article 288 TFEU: ‘A regulation shall have general application. It shall be binding in its entirety and directly applicable in all Member States.’18 On the meaning of ‘general application’ the CJEU has stated: ‘a regulation, being essentially of a legislative nature, is applicable not to a limited number of persons, defined or identifiable, but to categories of persons viewed abstractly and in their entirety’.19 The CJEU has further held: because of its nature and its purpose within the system of sources of Community law [a regulation] has direct effect and is, as such, capable of creating individual rights which national courts must protect. … these rights arise when the conditions set out in the regulation are complied with and it is not possible at a national level to render the exercise of them subject to implementing provisions other than those which might be required by the regulation itself. …

15  See the landmark case Case 6/64 Costa v ENEL [1964] ECR 585; see also Case 11/70 Internationale Handelsgesellschaft v [1970] ECR 1125. 16  See section II.B below. 17  CFSP and JHA Decisions, with binding force; JHA Framework Decisions, similar to EC Directives; EC and JHA conventions; CFSP common strategies; CFSP and JHA common positions, with binding force; and CFSP joint actions, with binding force. 18  Regulations adopted in relation to monetary union, Schengen or areas of enhanced cooperation may be applicable only in the Member States which are part of those areas. 19 Joined Cases 16/62 and 17/62 Confédération nationale des producteurs de fruits et légumes [1962] ECR 471, 478.

234 

William Robinson

So as to apply with equal force with regard to nationals of all the Member States, Community regulations become part of the legal system applicable within the national territory, which must permit the direct effect provided for in Article 189 to operate in such a way that reliance thereon by individuals may not be frustrated by domestic provisions or practices.20

By the same token, regulations can impose obligations directly on individuals.21 As part of the principle of direct applicability of regulations, Member States are barred from purporting to convert them into national law.22 A regulation may, however, expressly call for implementing measures by Member States.23 Since a regulation is binding in its entirety, a Member State does not have the power to exempt its operators from obligations laid down in the regulation.24 Nor may a Member State refuse to apply a regulation on the ground that it had opposed the adoption of the regulation in the course of the pre-adoption negotiations or on the grounds of practical difficulties in putting the regulation into effect.25 Under Article 297(2) TFEU, regulations must be published in the Official Journal and enter into force on the date specified in them or on the default date of the twentieth day following their publication. In certain cases they may have retroactive effect.

13.2.3. Directives Under Article 288 TFEU: ‘A directive shall be binding, as to the result to be achieved, upon each Member State to which it is addressed, but shall leave to the national authorities the choice of form and methods.’ The Member States must transpose a directive into their national systems by virtue of the principle of sincere cooperation enshrined in Article 4(3) TEU and by virtue of specific provisions incorporated in each directive (transposition article). While the Member States cannot simply rely on administrative practices to fulfil their transposition obligations,26 the existence of general principles of constitutional or administrative law may suffice.27 Broadly speaking, Member States can transpose directives in one of two ways: by reproducing the provisions of the directive more or less word for word in their national measures (copy out); or by reworking the provisions of the directive to adapt them to the national legal system (elaboration). Increasingly, the trend is to opt for copy out in order in particular to avoid two dangers: —— underimplementation of the directive, with a risk of infringement proceedings; and —— gold plating—that is, adding additional obligations not in the directive, which would place operators in the Member State concerned at a competitive disadvantage to those in other Member States.28 20 

Case 93/71 Orsolina Leonesio [1972] ECR 287, paras 5, 6, 22 and 23. Case 34/73 Fratelli Variola [1973] ECR 981, para 10. 22  Case 34/73 Fratelli Variola [1973] ECR 981, paras 10 and 11. 23  Case 230/78 Eridania-Zuccherifici nazionali [1979] ECR 2749, para 34. 24  Case 18/72 Granaria Graaninkoopmaatschappij [1972] ECR 1163, paras 15–18. 25  Case 39/72 Commission v Italy [1973] ECR 101, in particular para 22. 26  Case 102/79 Commission v Belgium [1980] ECR 1473, para 11. 27  Case C-475/08 Commission v Belgium [2009] ECR I-11503, para 41. 28  See, eg, p 10 of the UK Transposition Guidance, available at https://www.gov.uk/government/publications/ implementing-eu-directives-into-uk-law (accessed on 20 May 2015). 21 

EU Legislation

 235

One major problem with the copy out approach is that directives are drafted in a less detailed manner than regulations precisely so that Member States are left with sufficient latitude when transposing them, with the result that copy out will leave the national provisions lacking the requisite precision.29 Unlike regulations, directives are not directly applicable, but the CJEU has developed the doctrine of ‘direct effect’ to hold that directives can confer rights on individuals if they are clear, precise and unconditional.30 The principle of directives conferring rights directly on individuals in their dealings with state authorities (vertical direct effect) was confirmed in the van Duyn case31 and in numerous subsequent cases, but directives cannot be relied on by individuals as against other individuals (horizontal direct effect). A Member State which has failed to transpose a directive cannot rely on the principle of direct effect to escape liability for breach of its obligations.32 Under Article 297(2) TFEU, directives which are addressed to all Member States must be published in the Official Journal and enter into force on the date specified in them or on the default date of the twentieth day following their publication. By their nature, they generally cannot have retroactive effect.

13.2.4. Decisions Under Article 288 TFEU: ‘A decision shall be binding in its entirety. A decision which specifies those to whom it is addressed shall be binding only on them.’ Decisions specifying addressees may be addressed to one or more Member States or to one or more natural or legal persons. The CJEU has held that where a decision is addressed to Member States it may still have direct effect.33 Decisions do not generally include specific transposition provisions, although they may require national implementing measures. Under Article 297(2) TFEU, decisions which do not specify addresses must be published in the Official Journal and enter into force on the date specified in them or on the default date of the twentieth day following their publication.

13.2.5. Recommendations Under Article 288 TFEU, recommendations have no binding force. They are generally addressed to all the Member States. They are classic instruments of soft law whose use has been encouraged as part of efforts to reduce recourse to legislation at EU level.34 They may also be part of efforts to coordinate policies of the Member States in an area within the scope of the Treaties which may later lead to EU legislation. 29 

See Joint Practical Guide, point 2.2.2. The first expression of the doctrine was in relation to an article of the Treaty in Case 26/62 Algemene Transport- en Expeditie Onderneming van Gend en Loos [1963] ECR 1. It was extended to directives (and decisions) on the basis of the principle of the effectiveness (effet utile) of EU measures in Case 9/70 Grad [1970] ECR 825. 31  Case 41/74 van Duyn [1974] ECR 1337, see paras 12–14. 32  Case C-475/08 Commission v Belgium [2009] ECR I-11503, para 44. 33  Case C-156/91 Hansa Fleisch Ernst Mundt [1992] ECR I-5567, paras 12–15. 34  See Commission, ‘European Governance—A White Paper’ (Communication) COM (2001) 428, point III.3.2. 30 

236 

William Robinson

The CJEU has accepted that a national court may make a reference for a preliminary ruling regarding the interpretation of a recommendation.35

13.2.6. Opinions Under Article 288 TFEU, opinions have no binding force. They are generally adopted by an institution to set out its views on a situation or process. They may form part of preparations leading to the adoption of legislation and are mentioned in the list of procedural acts that must, under Article 298 TFEU, be referred to in all legal acts.

13.2.7.  Interinstitutional Agreements The European Parliament, the Council and the Commission have in the past concluded interinstitutional agreements covering various aspects of the legislative process and EU legislation. Article 295 TFEU, a new provision inserted by the Lisbon Treaty, provides that those institutions may make arrangements for their cooperation by means of interinstitutional agreements ‘which may be of a binding nature’. Such agreements will be binding only on the institutions.

13.2.8.  Legislative and Non-legislative Procedure Binding legal acts of the Union may, since the entry into force of the Lisbon Treaty, be divided into legislative and non-legislative acts. Legislative acts can be adopted only by the European Parliament or the Council, either jointly or separately, but not all the legal acts that they adopt are legislative acts. The decisive criterion is the adoption procedure. Under Article 289(3) TFEU: ‘Legal acts adopted by legislative procedure shall constitute legislative acts.’ Two types of legislative procedure are referred to in Article 289. Under paragraph 1, the ordinary legislative procedure consists in ‘the joint adoption by the European Parliament and the Council of a regulation, directive or decision on a proposal from the Commission’, as set out in detail in Article 294 TFEU. Under paragraph 2, the special legislative procedure consists in the adoption in ‘the specific cases provided for by the Treaties’ of ‘a regulation, directive or decision by the European Parliament with the participation of the Council, or by the latter with the participation of the European Parliament’. Thus, the distinction between legislative and non-legislative acts is now explicitly set out in the Treaties: if the article in the Treaty conferring competence on the Union provides that an act has to be adopted by the ordinary legislative procedure or a special legislative procedure, then the act is legislative. If the Treaty article does not mention a legislative procedure, then the act is non-legislative. One of the consequences of this distinction is the increased transparency of the legislative procedure: when the Council deliberates and votes on a draft legislative act, it must meet in public (Article 16(8) TEU) and all the related documents are public. 35 

Case C-322/88 Grimaldi [1989] ECR 4407.

EU Legislation

 237

13.2.9.  Delegated and Implementing Acts The implementation of EU law is in principle left to the Member States, under the supervision of the Commission and the CJEU.36 This structure was chosen first of all in order to avoid the need for a large central European administration and secondly to comply with the principle of subsidiarity, by leaving the Member States to decide on the most appropriate measures to ensure the fulfilment of obligations resulting from the Treaties or acts of the institutions. There are exceptions to the principle of ‘indirect administration’:37 in some cases the Treaty confers implementing powers directly on the Commission (eg Articles 105 and 106 TFEU in the field of competition law) and in others such powers are conferred on the Commission by legislative acts. Before the Lisbon Treaty, the Council could decide to confer on the Commission the power to implement the rules it had laid down or, in specific cases, to exercise implementing powers itself (Article 202 ECT). A system of supervision of the Commission’s activities by committees composed of representatives of the Member States was put in place (‘Comitology’), under which the issue could be referred to the Council in case of deadlock.38 The Lisbon Treaty has reaffirmed the principle of indirect administration but also the implementing role of the Commission by incorporating specific provisions on ‘delegated acts’ and ‘implementing acts’.

13.2.9.1.  Delegated Acts Under Article 290 TFEU: ‘A legislative act may delegate to the Commission the power to adopt non-legislative acts of general application to supplement or amend certain non essential elements of the legislative act.’ Key features, therefore, are that the power to adopt delegated acts can be conferred only by legislative acts, it can be conferred only on the Commission, and it can relate only to supplementing or amending non-essential elements.39 The legislative act must explicitly lay down the conditions to which the delegation is subject, which may include the right for the legislative authority to revoke the delegation or to block the entry into force of a delegated act by raising an objection.

13.2.9.2.  Implementing Acts Article 291 TFEU first provides that Member States are to adopt the necessary national implementing measures and then continues: ‘Where uniform conditions for implementing legally binding Union acts are needed, those acts shall confer implementing powers on the 36  See Declaration No 43 annexed to the Lisbon Treaty relating to the Protocol on the application of the principles of subsidiarity and proportionality. 37  See further Piris (n 14) 97ff. 38  The Comitology decision, Council Decision 1999/468/EC [1999] OJ L184/23, was amended by Council Decision 2006/512/EC [2006] OJ L200/11 to introduce a new procedure which gave the European Parliament a right of scrutiny. 39  See Case C-355/10 European Parliament v Council [2012] ECR I-0000 (ECLI:EU:C:2012:516), in which the Court of Justice states in particular that provisions entailing political choices cannot be delegated and that ascertaining which elements are essential ‘must be based on objective factors amenable to judicial review’ (paras 64–68).

238 

William Robinson

Commission or, in duly justified specific cases and in the cases provided for in Articles 24 and 26 of the Treaty on European Union [on the Common Foreign and Security Policy], on the Council.’ The power to adopt implementing acts can therefore be conferred by non-legislative acts as well as legislative acts and, while it will generally be conferred on the Commission, it may in specified cases be conferred on the Council. In both cases, the implementing measure has to respect the limits and the principles established in the act conferring the power. The exercise by the Commission of implementing powers conferred upon it under this procedure is governed by rules laid down by Regulation (EU) No 182/2011.40 The Commission acts in close consultation with a large number of committees composed of representatives from the Member States. The system is similar to the old Comitology system save that the European Parliament is now fully involved alongside the Council both in the adoption of the regulation on the supervision of the Commission’s exercise of implementing powers and in the scrutiny of draft implementing acts. The distinction between delegated acts and implementing acts is not clearly drawn, although some light was shed by the judgment of the CJEU in the Biocides case.41 In March 2016, the European Parliament, the Council and the Commission adopted a Common Understanding on Delegated Acts setting out the criteria for their use and detailed procedural arrangements.42

13.3.  Legislative Procedure 13.3.1. General As part of the ‘Community method’ established by the first European Treaties, basic legislation was adopted by the Council on the basis of a proposal from the Commission. The Assembly had only a consultative role. With successive amendments to the Treaties, the role of the European Parliament has been reinforced. Since the entry into force of the Lisbon Treaty, almost all basic legislation is adopted jointly by the Parliament and Council on the basis of a proposal from the Commission (known as the ‘codecision procedure’). The codecision procedure, which was first introduced by the Maastricht Treaty in 1992, is now the ‘ordinary legislative procedure’ under Article 294 TFEU. There are just a few fields where codecision does not apply, such as taxation, the common foreign and security policy, and the conclusion of international agreements.43

40 

[2011] OJ L55/11. Case C-427/12 Commission v European Parliament and Council [2014] ECR I-0000 (EU:C:2014:170), paras 33–40. 42  See Annex 1 to the 2016 Interinstitutional Agreement on Better Law_Making referred to in section 9.7 below. The new Common Understanding replaces the former text of 2011 (Council document 8640/11 of 4 April 2011). 43  See Art 113 TFEU, Art 24 TEU and Art 218 TFEU, respectively. 41 

EU Legislation

 239

The basic principle of the ordinary legislative procedure is that the same text is adopted by the two institutions.44

13.3.2.  Commission Proposal In formulating its proposal, the Commission seeks to ‘promote the general interest of the Union’ and enjoys full independence from the other institutions and the Member States.45 It is responsible for all the preparatory work. At an early stage, the Commission department responsible will consult the Member States and other interested parties about the policy options and may seek advice on technical issues from groups of experts. It may publish a Green Paper setting out the issues and policy options under consideration and inviting comments or a White Paper to outline tentative proposals and to invite comments. In 2003, the European Parliament, the Council and the Commission agreed on increased use of public consultations and impact assessments to improve the preparation of legislation.46 The Commission now systematically carries out public consultations on legislative initiatives and has set up a website giving access to all ongoing consultations.47 For all legislative initiatives, the Commission publishes a roadmap describing the problem that the initiative aims to address and possible policy options, and giving an overview of the planned stages in the development of the initiative.48 The Commission department responsible carries out an impact assessment to help structure and develop policies following Commission Guidelines. The quality of its impact assessment is checked by a semiindependent Regulatory Scrutiny Board.49 The Commission adopts its proposal in all the official languages.50 It is published as a ‘COM document’ on EUR-Lex and in the Commission’s own register of documents.51 The proposal must generally be submitted for consultation to the European Economic and Social Committee,52 and in appropriate cases to the Committee of the Regions.53 It is also forwarded to national parliaments, which may deliver a reasoned opinion on whether the draft act complies with the principle of subsidiarity.54

44 

See Art 294(3) and (4) TFEU. See Art 17 TEU. 46  Interinstitutional Agreement on Better Law-Making [2003] OJ C321/1, points 25 and 26. The 2003 Agreement has now been replaced by the 2016 Agreement referred to in section 9.7 below. 47  http://ec.europa.eu/yourvoice/consultations/index_en.htm (accessed on 20 May 2015). 48  A list of all roadmaps is available on the website of the Commission Secretariat-General: http://ec.europa.eu/ smart-regulation/impact/planned_ia/planned_ia_en.htm (accessed on 20 May 2015). 49  The Commission maintains a website containing the text of the Impact Assessment Guidelines (SEC (2009) 92) and a list of all impact assessments, together with the reports of the Regulatory Scrutiny Board, available at http://ec.europa.eu/smart-regulation/impact/index_en.htm (accessed on 20 May 2015). The European Parliament, the Council and the Commission have adopted common rules on impact assessments in section III of the 2016 Agreement referred to in section 9.7 below. 50  See section 4 below. 51  http://ec.europa.eu/transparency/regdoc/recherche.cfm?CL=en (accessed on 20 May 2015). 52  See Art 304 TFEU. 53  See Art 307 TFEU. 54  In accordance with Arts 2 and 3 of Protocol No 1, which refers in turn to Protocol No 2. 45 

240 

William Robinson

The text of the Commission’s proposal forms the basis for all discussions in the other institutions. Article 293(1) TFEU states: ‘Where, pursuant to the Treaties, the Council acts on a proposal from the Commission, it may amend that proposal only by acting unanimously, except in the cases referred to in’ a small number of Treaty provisions.

13.3.3.  European Parliament Within the European Parliament, the proposal is assigned to the relevant committee55 and a rapporteur is chosen. The rapporteur is given a mandate by the committee (or in some cases a mandate by the plenary)56 and represents the European Parliament in the negotiations with the Council and Commission. When those negotiations are concluded, the rapporteur will present to the committee a draft report comprising a draft legislative resolution and, where appropriate, amendments to the draft act. After debating that draft, the committee submits to the plenary for voting its final report, again comprising a draft legislative resolution and any amendments to the draft act.57 In most cases, the Parliament proposes textual changes (known as ‘amendments’) to the Commission’s proposal. In the procedure at first reading, amendments may be tabled by any committee member at the committee stage, while in the plenary amendments are admissible only if they are tabled by the committee responsible, by a political group or by 40 MEPs.58 The European Parliament has now established within its Directorate-General for Parliamentary Research Services a Directorate for Impact Assessment and European Added Value to provide ex ante and ex post impact assessment services to parliamentary committees, together with assessment of the added value of future or current EU policies.

13.3.4. Council Within the Council, the proposal is examined by one of the specialised working parties and committees known as the ‘Council preparatory bodies’, composed of national experts from all the Member States and chaired by a representative of the country holding the six-monthly presidency of the Union.59 Administrative support is provided by the General Secretariat of the Council. The chair seeks to facilitate political compromises in the working party. After all technical aspects have been discussed by the national experts, the chair will ask COREPER or the Council itself (in the configuration of the ministers for the subject concerned) to define the political position in a ‘general approach’. The general approach constitutes the mandate for the negotiations with the European Parliament and the Commission.

55  There are currently 22 standing committees and two special committees: see www.europarl.europa.eu/committees/en/full-list.html (accessed on 20 May 2015). 56  EP Rules of Procedure, Rule 73. 57  Full information on the parliamentary procedure can be found in the European Parliament Legislative Observatory: www.europarl.europa.eu/oeil/index.jsp?language=en (accessed on 20 May 2015). 58  EP Rules of Procedure, Rules 208 and 169, respectively. 59 There are currently over 150 such bodies: www.consilium.europa.eu/council/committees-and-workingparties?lang=en (accessed on 20 May 2015).

EU Legislation

 241

To accommodate the different interests of all Member States, numerous textual changes to the Commission’s proposal are generally suggested.60

13.3.5. Trilogues The negotiations between the three institutions are conducted in the ‘informal trilogues’ whose use was formalised in 2007.61 The trilogues serve to bring the positions of the three institutions closer together and to enable the representatives of each institution to keep it informed of the direction of the negotiations. The trilogues have proved so effective that over 80% of legislative acts are now adopted at first reading.62 However, if the European Parliament and the Council do not agree on the Commission proposal at first reading, it passes to a second reading,63 after which, if there is still no agreement, a conciliation procedure is launched.64 While the first reading has no time limitations, the second reading and the conciliation stage are subject to strict time limits.65 If the European Parliament or the Council fails to approve the text within those time limits, the proposed act is deemed not to have been approved.

13.3.6.  Role of the Commission in the Codecision Procedure If the Commission takes the view that, as a result of the amendments made by the European Parliament and the Council, its proposal has been ‘denatured’, that is, changed to such an extent that it no longer reflects the original intention, it may withdraw its proposal at any time before the adoption of the Council position,66 after which the act may no longer be adopted. That power, while rarely exercised, gives the Commission a stronger position during the negotiations. If it becomes evident that substantial changes to a proposal are needed to take account of the other institutions’ concerns, the Commission may agree to submit an amended proposal.

13.4. Languages Multilingualism is one of the fundamental principles of the EU. Article 3(3) of the TEU provides that the Union is to ‘respect its rich cultural and linguistic diversity’.67 More 60 

Council Rules of Procedure [2009] OJ L325/35, Annex V, point 15. See the Joint Declaration on practical arrangements for the codecision procedure [2008] OJ C102E/111. 62  For the details of the procedure, see Art 294(3)–(6) TFEU. For the statistics, see the European Parliament’s Conciliation and Codecision site: www.europarl.europa.eu/ code/about/statistics_en.htm (accessed on 27 September 2015). 63  See Art 294(7)–(9) TFEU. 64  See Art 294(10)–(12) TFEU. 65  For the second reading, see Art 294(7) and (8); for the third reading, see Art 294(12) and (13). Extensions are possible under Art 294(14). 66  See Art 293(2) TFEU. Some limits on the power of withdrawal have been laid down by the CJEU in Case C-409/13 Council v Commission [2015] ECR I-0000 (ECLI:EU:C:2015:217). 67  The Charter of Fundamental Rights reiterates that obligation in Art 22, and in Art 21 prohibits any discrimination on grounds of, amongst other things, language. 61 

242 

William Robinson

s­pecifically, Article 342 of the TFEU provides: ‘The rules governing the languages of the institutions of the Union shall … be determined by the Council, acting unanimously by means of regulations.’ The importance attached to language matters and their sensitivity are attested by the fact that the rules are adopted by the Council acting unanimously and without the involvement of the Parliament or the Commission. Council Regulation No 1 of 1958 lays down the list of the ‘official languages’ and provides that: ‘Regulations and other documents of general application shall be drafted in the official languages’ and that the Official Journal of the European Union is to be published in the official languages.68 Each time new Member States have joined the EU, Regulation No 1 has been amended to add their languages to the list, which now includes 24 languages.69 The Treaties are expressly declared to be authentic in all the official languages.70 The CJEU has long recognised the importance of EU legislation being interpreted and applied uniformly in all the Member States and hence of taking account of the different language versions.71 In the CILFIT case,72 the CJEU stated ‘it must be borne in mind that Community legislation is drafted in several languages and that the different language versions are all equally authentic’.

13.5. Accessibility The TEU, as amended by the Lisbon Treaty, makes various references to the need for openness and transparency, and to the rule of law.73 A general right of access to EU documents is given by Article 15 TFEU, with detailed rules being laid down in a 2001 Regulation.74 Under Article 297 TFEU, all EU legislative acts, all regulations and all directives which are addressed to all the Member States must be published in the Official Journal of the European Union. The Publications Office of the European Union (Publications Office) is responsible for publishing all EU documents. The electronic version of the Official Journal is now the authentic version.75 The Publications Office makes available without charge in all the official languages a system of websites and databases on all aspects of EU law through the EUR-Lex portal.76 It includes: the electronic version of the Official Journal; collections of treaties and legislation, including summaries and consolidated texts of EU legislation; search engines; and databases on legislative procedures. 68 

[1958] OJ 17/385; see Arts 1, 4 and 5. Ireland joined in 1973, the Irish language was not given the status of an official language. That was eventually done by Council Regulation (EC) No 920/2005 [2005] OJ L156/3. 70  See Art 55 TEU and Art 358 TFEU. 71  For some of the earliest expressions of this principle see Case 19/67 van der Vecht [1967] ECR 345; Case 29/69 Stauder [1969] ECR 419, para 3. 72  Case 283/81 CILFIT [1982] ECR 3415, para 18. 73  On openness see the preamble and Arts 1, 10 and 11; on the rule of law see the preamble and Art 2, and the Charter of Fundamental Rights. 74  Regulation (EC) No 1049/2001 of the European Parliament and of the Council [2001] OJ L145/43. 75  See Council Regulation (EU) No 216/2013 [2013] OJ L69/1. 76  http://eur-lex.europa.eu/homepage.html (accessed on 20 May 2015). 69  When

EU Legislation

 243

The EU institutions also maintain large databases of information about the legislative process and its preparatory stages.77 The European Parliament goes to great lengths to offer access to its sessions.78 When the Council deliberates and votes on draft legislation, its sessions are open to the public.79 Most of the departments of the Commission maintain public websites offering explanatory material about legislation in their respective sectors.80 The CJEU ensures that the rules on publication are strictly complied with.81

13.6.  Drafting Process and Drafting Rules 13.6.1.  Drafting Process The first draft is written, usually in English, by technical experts in the Commission Directorate-General (DG) responsible for the subject matter, and discussed and agreed within that DG. It is then submitted to all the other Commission DGs concerned, a process known as the Inter-Service Consultation (ISC). The ISC serves to ensure that the Commission acts in a coordinated manner and it is supervised by the Commission Secretariat-General, which makes sure that all measures proposed correspond to the Commission’s political priorities. The draft is also examined by two separate departments of the Commission Legal Service. In one, a lawyer specialising in the subject matter checks the lawfulness of the proposal and whether it is consistent with other EU legislation. In the other, a lawyer specialising in legislative drafting checks that the draft is clear and complies with all the EU drafting rules. The draft may also be edited by an Editing Service to ensure that it is grammatically correct. At the end of the procedure within the Commission, the draft is translated into the other languages so that it can be adopted in all the official languages by the 28 Commissioners. The text thus adopted together with an explanatory memorandum constitutes the Commission proposal which is submitted to the European Parliament and the Council. From this point, all changes to the Commission’s proposal require the agreement of the representatives of the European Parliament, the Council and the Commission. For practical reasons, the negotiations between the three institutions focus on a single language version,

77  See the websites of the EU institutions accessible from their common server Europa: http://europa.eu/abouteu/institutions-bodies/index_en.htm (accessed on 20 May 2015); European Parliament Legislative Observatory: www.europarl.europa.eu/oeil/index.jsp?language=en (accessed on 20 May 2015). Access to Council documents: Public Register: www.consilium.europa.eu/register/en/content/int/?typ=ADV (accessed on 20 May 2015). The EU institutions have undertaken in s VI of the 2016 Agreement referred to in section 9.7 below to establish a joint database on legislative files. 78  See www.europarl.europa.eu/eng-internet-publisher/eplive/public/default.do?language=en (accessed on 20 May 2015). 79  See the Council’s Rules of Procedure, Decision 2009/937/EU [2009] OJ L325/35, Art 7. 80  http://ec.europa.eu/about/ds_en.htm (accessed on 20 May 2015). 81  See, in particular, Case C-370/96 Covita [1998] ECR I-7711, paras 26 and 27 (non-publication of one language version of a regulation); Case C-345/06 Gottfried Heinrich [2009] ECR I-1659, para 63 (provisions not published on security grounds); Case C-161/06 Skoma-Lux [2007] ECR I-10841, paras 33, 34, 38, 48 and 49 (failure to publish in the prescribed form).

244 

William Robinson

now almost invariably the English version. Any changes to the proposal must be in the form of suggested amendments to the actual text.82 All the negotiations focus on which of the amendments suggested by the European Parliament or the Council can be accepted for inclusion in the Commission proposal. Once agreement in principle has been reached on those amendments, the text in the original language incorporating all the amendments is translated into the other languages so that the legislative act can be formally adopted by the European Parliament and the Council in all the official languages. The work of translating the final text is shared by the translation services of the European Parliament and of the Council. Both the European Parliament and the Council have teams of lawyer-linguists who are responsible for ensuring that all legislation complies with the drafting rules and all the language versions produce the same legal effect. While they were formerly involved only at the final stage, when agreement had been reached in principle, they are now increasingly offering drafting guidance at earlier stages.83

13.6.2.  Drafting Rules The Treaties themselves lay down only minimal requirements for the form and presentation of EU legal acts,84 leaving it to the institutions to establish more detailed rules. The Rules of Procedure of the European Parliament85 and those of the Council86 set out the basic format for EU legislative acts. In 1997, the Intergovernmental Conference adopted Declaration 39 annexed to the Amsterdam Treaty, which emphasises the importance of good quality legislation and calls on the EU institutions to draw up by ‘common accord guidelines for improving the quality of the drafting of Community legislation’.87 The European Parliament, the Council and the Commission duly adopted an Agreement, with 22 drafting guidelines drawn up in collaboration with the Member States.88 While little in them was actually new, they were of significance in reflecting a consensus on the basic requirements for drafting EU legislation, ironing out divergent practices and recording expressly some unwritten conventions. The lawyer-linguists of the three institutions then drew up a Joint Practical Guide for persons involved in the drafting of legislation. The Guide developed the content and explained the implications of the guidelines in the 1998 Agreement. The Guide has been translated into all the official languages and has been widely disseminated both within the

82 

See Annex V, point 15 to the Council’s Rules of Procedure, Decision 2009/937/EU [2009] OJ L325/35. M Guggeis and W Robinson, ‘“Corevision”: Legal-Linguistic Revision in the European Union “Codecision” Process’ in CJW Baaij (ed), The Role of Legal Translation in Legal Harmonisation (Alphen aan den Rijn, Kluwer Law International, 2012). 84  See Arts 288, 296 and 297 TFEU. 85  See Rule 77 on ‘Requirements for the drafting of legislative acts’, available at www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//TEXT+RULES-EP+20140701+RULE-077+DOC+XML+V0//EN&language=EN&navi gationBar=YES (accessed on 20 May 2015). 86  [2009] OJ L325/35, Annex VI. 87  [1997] OJ C340/139. 88  [1999] OJ C73/1. 83  See

EU Legislation

 245

institutions and beyond. It was updated in 2013 to take account of the Lisbon Treaty and other developments.89 Since 1963, the Council’s Legal Service has produced its own drafting manual, the Manual of Precedents. It was produced in all the official languages but was not publicly available. It set out all types of acts in schematic form, showing the usual method of presentation and the most common formal provisions. It also contained general information on the structure of acts and the way they should be drafted. It was updated regularly, with more material on legal drafting problems being included over the years. It was eventually split into two parts, one on legal acts and the other on international agreements. The Manual on Legislative Drafting of the Commission Legal Service was first drawn up in 1985 and the current edition, the third, was produced in 1997. It is aimed mainly at lawyers and staff closely involved in the drafting process, and is written in technical language with a quite opaque structure.90 To take account of the special characteristics of its decentralised drafting system, the Commission in the 1990s developed a word-processing program, called LegisWrite, to standardise the basic structure and presentation of legal acts. In 2014, a Drafter’s Assistance Package was incorporated into LegisWrite. It offers standard formulations and guidance, as well as links to the Commission Legal Revisers’ internal website, giving drafters more assistance. To ensure that texts published by the different EU institutions, bodies and agencies are presented in a uniform manner to make them more accessible to users, the Publications Office has issued an Interinstitutional Style Guide in all the official languages.91 While the Guide focuses mainly on rules concerning the presentation of texts, it does include sections touching on legislative matters. The lawyer-linguists of the European Parliament and of the Council worked with the legal revisers of the Commission to produce, in 2016, a Joint Handbook for the Presentation and Drafting of Acts Subject to the Ordinary Legislative Procedure. It sets out to harmonise the drafting practices of the three institutions, give guidance and suggest standard formulas to be used.

13.6.3.  Rules on Legal-linguistic Revision In the 1998 Agreement on drafting quality, the EU institutions responded to the call in Declaration 39 for practical measures to ensure that the drafting guidelines are properly applied by undertaking to ‘organise their respective internal procedures in such a way that their legal services, including their legal/linguistic experts, may, each for their own institution, make drafting suggestions in good time’. In the 2003 Agreement on better law making, the institutions undertook to improve their scrutiny of the wording of texts adopted, if necessary by allowing extra time before an act was finally adopted.92 89 

http://eur-lex.europa.eu/content/techleg/KB0213228ENN.pdf (accessed on 20 May 2015). The English text is available at http://ec.europa.eu/smart-regulation/better_regulation/key_docs_en.htm#_ acces_eu_law (accessed on 20 May 2015). 91  http://publications.europa.eu/code/en/en-000500.htm (accessed on 20 May 2015). 92  [2003] OJ C321/1, points 25 and 31. That provision is, however, not reproduced in the 2016 Agreement (referred to in section 9.7 below), which repealed the 2003 Agreement. 90 

246 

William Robinson

In 2007, the institutions adopted a Joint Declaration on practical arrangements for the codecision procedure, which emphasised the importance of a consistent and consensual approach to revision.93 The Rules of Procedure of both the European Parliament94 and the Council95 now include provisions on the importance of the scrutiny of the drafting of their legislation by their lawyer-linguists. The emphasis is now on improving drafting quality as early as possible, since the more advanced the stage in the legislative process, the less scope there is for improving a draft. And once a text has been adopted, there is little scope for making any changes affecting its content, except to correct obvious mistakes and formal errors.96

13.7.  Structure and Style 13.7.1. Structure Under Guideline 7 in the 1998 Agreement, all EU legal acts should have the same basic structure: title, preamble, enacting terms and, where necessary, an annex. The title of an EU act comprises: the indication of the type of act, the name of the institution or institutions concerned, a serial number,97 the date of adoption and an indication of the subject matter, as in Regulation (EU) No 211/2011 of the European Parliament and of the Council of 16 February 2011 on the citizens’ initiative.98 The full title is the primary means of identifying the act in subsequent references, although it may be abbreviated. Acts may also be given a short title (Guideline 8 and JPG 8.7 and 8.8), though this is not often done in practice. Preambles consist of two main parts: the citations and the recitals. Under Guideline 9, the citations set out the legal basis in the Treaties for the adoption of the act,99 and any procedural requirements under that legal basis.100 Under Guideline 10, the recitals set out the reasons on which the act is based, as required by Article 296 TFEU. The enacting terms are divided into articles, which may be subdivided into numbered or unnumbered paragraphs and subparagraphs. In longer acts, the articles may be grouped in chapters, which may consist of different sections. In extremely complex texts, the chapters may be grouped in titles, which themselves may be grouped in parts (JPG 15.4). A standard order of the articles has been laid down in Guideline 15.

93 

[2007] OJ C145/5. See Rule 193(2). See also Rules 75 and 78, available at www.europarl.europa.eu/sides/getDoc.do?pubRef=-// EP//TEXT+RULES-EP+20140701+RULE-193+DOC+XML+V0//EN&language=EN&navigationBar=YES (accessed on 20 May 2015). 95  [2009] OJ L325/35; see Art 22. 96  See Case 131/86 United Kingdom v Council [1988] ECR 905, in particular paras 35 and 37. 97  Since 1 January 2015 all EU acts are numbered according to a single simplified numbering system: http://eur-lex.europa.eu/content/tools/elaw/OA0614022END.pdf (accessed on 20 May 2015). 98  [2011] OJ L65/1. 99  Council Rules of Procedure [2009] OJ L325/35, Annex VI, point A.1(c). 100  See Art 296 TFEU. 94 

EU Legislation

 247

An annex may be attached to an EU act to set out material which is technical or especially lengthy (Guideline 22). An annex is an integral part of the act, but is separate from the enacting terms. In practice, annexes are often used to present material which is destined to be amended frequently. The structure of EU rules is a hierarchical one, with the most fundamental rules being laid down in the Treaties (TEU and TFEU) and then the basic rules being laid down in legislative acts adopted jointly by the European Parliament and the Council under the ‘ordinary legislative procedure’ under Article 289 TFEU. Those basic acts may then be fleshed out by delegated acts adopted by the Commission under Article 290 TFEU or by implementing acts under Article 291 TFEU. The structure is not always wholly rational. EU rules have often been adopted piecemeal, as and when needs were perceived and as and when progress with a legislative initiative was possible. The rules in one sector may also be fragmented because of the need to respect the confines of different legal bases, which may entail different procedures and even different legislative authorities. Sometimes it is only after the adoption of an act that the need for exceptions and special provisions becomes apparent, and they may be adopted by ad hoc acts. Efforts are now being made to tackle the problem by means of recasts or reviewing and rationalising all EU law in one sector as part of the REFIT programme (see section 9.5 below).

13.7.2.  Drafting Style Early European acts were drafted in very broad terms, leaving much latitude in interpretation and application to the executive and the courts. In 1974, an English judge famously said of EC legislation: ‘The enactments give only an outline plan. The details are to be filled in by the judges.’101 Over the years, there has been a shift from the ‘fuzzy’ or general-principles style familiar to many countries of continental Europe towards the ‘fussy’ or more detailed drafting style familiar to common lawyers. EU acts have become longer and more complex, not just because of the entry of Member States with different drafting traditions, but also because EU rules are entering ever more areas which have to be regulated in detail and because of the existence of more international law that must be integrated in the EU system. Another reason is that it is necessary to embed any new act in the acquis, the substantial body of existing EU legislation, so a new act has to make clear how it relates to acts already in existence in the same area. Finally, more ‘special provisions’ and ‘exemptions’ have to be inserted to accommodate the needs and wishes of more Member States. This is inevitable in a Union which is committed to ‘respect its rich cultural … diversity’102 and in a legislative procedure in which any amendment to the Commission’s proposal requires unanimity amongst all the Member States.103 The enacting terms have become more precise and detailed. Many more definitions are now included (Guideline 13) and specific references have replaced vague ones 101  102  103 

Lord Denning in Bulmer Ltd v Bollinger SA [1974] 4 Ch 401, 411. Art 3(3) TEU. Art 293(1) TFEU.

248 

William Robinson

(­Guideline 16). While early legislative acts often left the Commission very wide discretion in implementing them, the TFEU itself now stipulates that ‘the objectives, content, scope and duration of the delegation of power shall be explicitly defined in the legislative acts’ (Article 290). The drafting rules encourage the spelling out of things that once might have been left open, such as transitional provisions and repeals (see, respectively, JPG 15.1.5 and Guidelines 20 and 21). In particular, the number of recitals has increased noticeably. In the early days, when legislation was adopted by the Council alone and there was no publicly accessible record of the debates leading up to adoption, the recitals had to explain why EU acts had been adopted. Now that the EU legislative procedure involves the European Parliament, and is more open and transparent, that need is less strongly felt, but other factors have come into play. To forestall any challenge of an EU act under Article 263 TFEU on the grounds that the statement of reasons for an act is not sufficient to satisfy the requirements of Article 296 TFEU, drafters may include more material in the recitals as a precaution. Standard formulas are inserted to confirm that the act complies with basic EU principles such as proportionality and subsidiarity under Article 5 TEU (JPG 10.15) and the Charter of Fundamental Rights as referred to in Article 6 TEU. Recitals have also come to serve other functions. They have become bargaining chips in the negotiating process and a means of communication between the legislative a­ uthority and the user of the legislation. Recitals are included to mention all the other EU acts referred to in the act, and to mention certain consultations and financial arrangements.104 To counterbalance the trend for EU acts to become more complex and lengthy, various steps have been taken to make them more reader-friendly, in particular by means of instructions in the Joint Practical Guide (JPG). A standard order for the articles has been formalised (Guideline 15). Drafters are recommended to keep sentences and articles short (Guideline 4 and JPG 4.4 to 4.5.2 and 5.2.2), and to number items in a list (Guideline 15) and all recitals (Guideline 11 and JPG 11.1 to 11.4). Each article is now generally given a heading and most subdivisions of articles are numbered. Drafters are told to use simple terms and everyday language (Guideline 1 and JPG 1.4), and to avoid jargon, vogue words, non-standard Latin expressions and expressions linked to one language or one national legal system (JPG 5.2.4 to 5.3.2).

13.8.  Role of the CJEU and Interpretation of EU Law 13.8.1. General Most cases brought before the CJEU are heard at first instance by the General Court, with a possible subsequent appeal to the Court of Justice.However, some types of case considered to be particularly important are heard directly by the Court of Justice itself. 104 

See, respectively, JPG 16.10.3; 10.17; and 10.18 and 10.19.

EU Legislation

 249

The main heads of jurisdiction relevant to EU legal acts are actions for annulment, references for preliminary rulings and infringement proceedings brought by the Commission against Member States.105

13.8.2.  Actions for Annulment Under Article 263 and 264 TFEU, an action may be brought before the CJEU for a review of the legality of an EU legal act, and if the action is well founded the act will be declared void. The Court of Justice has exclusive jurisdiction over actions brought by a Member State against the European Parliament and the Council or against either one of them (apart from Council acts in respect of state aid, dumping and implementing powers) and over actions brought by one EU institution against another. The General Court has jurisdiction, at first instance, in all other actions of this type and particularly in actions brought by individuals.

13.8.3.  References for Preliminary Rulings Under Article 267 TFEU, the Court of Justice has jurisdiction to give preliminary rulings in response to questions referred to it by national courts on the interpretation of the Treaties or on the validity or interpretation of EU legal acts. It thus cooperates with the courts of the Member States, which are the ordinary courts in matters of EU law, to ensure the effective and uniform application of EU law and to prevent divergent interpretations. Its preliminary rulings are binding on the national court which referred the question to it and on other national courts before which the same problem is raised.

13.8.4.  Infringement Proceedings Under Article 258 TFEU, if the Commission considers that a Member State is failing to fulfil an obligation under the Treaties, it may initiate proceeding against it. The pre-litigation stage first entails a letter of formal notice and then a reasoned opinion. Most disputes are resolved at this stage; if not, the Commission may bring proceedings before the CJEU for judgment against the Member State. In the very rare cases where the Member State fails to comply with that judgment, the Commission may again bring the case before the CJEU, which then has jurisdiction to impose financial penalties on the Member State.

13.8.5.  Interpretation of EU Law The EU legislative authority has laid down no rules on the interpretation of EU legislation (apart from one short regulation on time limits).106 All the principles of interpretation have been developed by the CJEU on the basis of principles derived from the law of the Member

105  106 

For other heads of jurisdiction see Art 259ff TFEU. Regulation (EEC, Euratom) No 1182/71 of the Council [1971] OJ L124/1.

250 

William Robinson

States and international law, in particular the Vienna Convention on the Law of Treaties of 1986. The approach of the CJEU to interpretation of EU law differs from that of courts in Member States to interpreting their national law. Particularly striking is the difference from the literalist approach of courts in common law countries. The CJEU itself referred in the CILFIT case to ‘the characteristic features of Community law and the particular difficulties to which its interpretation gives rise’ and went on: Community legislation is drafted in several languages and … the different language versions are all equally authentic. An interpretation of a provision of Community law thus involves a comparison of the different language versions. … even where the different language versions are entirely in accord with one another, … Community law uses terminology which is peculiar to it. … legal concepts do not necessarily have the same meaning in Community law and in the law of the various Member States. Finally, every provision of Community law must be placed in its context and interpreted in the light of the provisions of Community law as a whole, regard being had to the objectives thereof and to its state of evolution at the date on which the provision in question is to be applied.107

The CJEU accordingly takes a broad approach to interpreting a text, relying on core meanings of terms. It may compare the different language versions and attempt to find a ‘common interpretation which best reflects the sense in all the languages’.108 It has established the principle that all the language versions must be treated equally109 and ruled that: ‘the Community legal order does not, in principle, aim to define concepts on the basis of one or more national legal systems unless there is express provision to that effect’.110 The CJEU looks at a provision in its context and in the light of the aims of the act.111 It has regard to the reasons for the provision as stated in the preamble.112 It may look at the travaux préparatoires, such as the Commission proposal and the explanatory memorandum. It also takes account of general principles of EU law, such as legal certainty113 and fundamental rights.114

13.9.  Quality of EU Legislation 13.9.1.  Calls for Improvement and Response of the EU Institutions In 1992, the French Conseil d’État drew up a report which looked at the growing influence of Community legislation on French law and expressed concern at the volume of C ­ ommunity 107 

CILFIT (n 72) paras 17–20. Case 80/76 North Kerry Milk Products [1977] ECR 425. 109  Case C-152/01 Kyocera Electronics Europe [2003] ECR I-13821, para 32. 110  Case C-103/01 Commission v Germany [2003] I-5369, para 33. 111  See, eg, Case C-136/91 Findling Walzlager [1993] ECR I-1793, para 11. 112  See, eg, Case C-355/95 P TWD [1997] ECR I-2549, para 21. 113  See, eg, Joined Cases 42 and 49/59 SNUPAT [1961] ECR 53; Case T-171/00 Spruyt [2001] ECR FP I-A-187, II-855, paras 70, 71 and 72. 114 Joined Cases C-465/00, C-138/01 and C-139/01 Rechnungshof v Österreichischer Rundfunk and Others [2003] ECR I-4989, para 68. 108 

EU Legislation

 251

rules and how difficult they were to understand.115 In the same year, the European Council adopted the Birmingham declaration, with the pithy demand: ‘We want Community legislation to be clearer and simpler.’ In response, the Council adopted a Resolution setting out what are known as the ten commandments of legislative drafting,116 but the results were limited. In 1995, the Koopmans committee of senior Dutch civil servants recommended the introduction of guidelines on legislative quality and the establishment of an independent vetting committee.117 The Netherlands and the European institutions then organised a conference on the quality of legislation, the proceedings of which were published.118 In 1997, the Intergovernmental Conference adopted Declaration 39 on the quality of the drafting of Community legislation, which was annexed to the Amsterdam Treaty.119 That led the EU institutions to adopt the 1998 Agreement on drafting guidelines referred to in section 6.2 above.

13.9.2.  Governance Initiative In 2000, the European Council in Lisbon ‘set itself a new strategic goal for the next decade: to become the most competitive and dynamic knowledge-based economy in the world’.120 In 2001, the Mandelkern Group established by the EU Member States to examine regulatory quality in Europe called for an action plan of over 30 measures.121 The Commission launched its governance initiative in 2001, stating that the EU ‘must pay constant attention to improving the quality, effectiveness and simplicity of regulatory acts’.122 It later called for a new strategy and a new culture of simplification of regulation,123 and the following year adopted a package of measures to promote better law making.124

13.9.3.  Agreement on Better Law Making 2003 Responding to an invitation from the European Council in Seville in 2002, the EU institutions adopted another Agreement in 2003125 affirming their commitment to improving all aspects of law making: better preparation of legislation, greater transparency, drafting quality, improved accessibility of EU legislation, keeping the regulatory burden as light as possible and improved follow-up to legislation adopted. That Agreement became a cornerstone of the Commission’s Better Regulation programme.126 115 

French Conseil d’État, ‘Rapport public 1992, Le droit communautaire’, Etudes et documents No 44. [1993] OJ C166/1. Koopmans, T, De kwaliteit van EG-regelgeving—Aandachtspunkten en voorstellen. 118  AE Kellermann, Improving the quality of legislation in Europe (Leiden, Martinus Nijhoff Publishers, 1998). 119  [1997] OJ C340/139, see section 6.2 above. 120  Point 5 of the Council Conclusions, available at www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/en/ec/00100-r1.en0.htm (accessed on 20 May 2015). 121 http://ec.europa.eu/smart-regulation/better_regulation/documents/mandelkern_report.pdf (accessed on 20 May 2015). 122  COM (2001) 428 (n 34), point 3.2. 123  See COM (2001) 130, 3; COM (2001) 726, 2. 124  COM (2002) 275, 276, 277 and 278. 125  [2003] OJ C321/1. It was repealed by the 2016 Agreement referred to in section 9.7 below. 126  The Commission publishes annual reports on Better Law making, available at http://ec.europa.eu/smartregulation/better_regulation/reports_en.htm (accessed on 20 May 2015). 116  117 

252 

William Robinson

The Member States continued to call for further steps to be taken to improve the quality of EU regulatory policy.127 In 2005, the Commission adopted a strategy for the simplification of the regulatory environment128 and later ran a rolling programme for simplifying and improving existing Union law by identifying areas where action should be taken with input from stakeholders.129

13.9.4.  Smart Regulation In 2010, the Commission launched the Smart Regulation project, which embraces the whole cycle of regulation, including implementation and post-adoption scrutiny.130 Smart Regulation forms part of the Europe 2020 strategy, which seeks a way out of the economic crisis by setting various goals to be achieved by initiatives to deliver smart, sustainable and inclusive growth, including: ‘considering the wider use of regulations rather than directives, launching ex post evaluation of existing legislation, pursuing market monitoring, reducing administrative burdens, removing tax obstacles, improving the business environment, particularly for SMEs, and supporting entrepreneurship’.131 Specific measures taken include a simplification agenda covering the current Multiannual Financial Framework, measures to improve the business environment for small businesses and a red-tape reduction programme for the period 2007–12.132

13.9.5. REFIT In 2012 the Commission launched its Regulatory Fitness and Performance programme, REFIT, designed make EU law ‘fit for purpose’ and to simplify and reduce regulatory costs.133 It includes actions to simplify legislation, reduce regulatory burdens, withdraw pending proposals, repeal laws no longer necessary, and evaluate the efficiency and effectiveness of EU legislation. Key tools are evaluations and fitness checks which make a holistic and critical appraisal of the regulatory framework in a given area.134 127 See the 2004 Joint Initiative on Regulatory Reform, available at http://collections.europarchive.org/ tna/20050217120000/http:/www.hm-treasury.gov.uk/media/95A/52/6presidencies.pdf (accessed on 20 May 2015). See also the Davidson Review on implementation of EU legislation in the United Kingdom (London, HMSO, 2006) point 5.2.6, available at http://webarchive.nationalarchives.gov.uk/+/http:/www.bis.gov.uk/policies/ better-regulation/reviewing-regulation/simplifying-eu-legislation-davidson-review (accessed on 20 May 2015). 128  COM (2005) 535. 129  COM (2006) 690, COM (2008) 33, COM (2009) 17. More recently, the emphasis has been on reducing burdens on business, see COM (2011) 803. See also http://ec.europa.eu/smart-regulation/better_regulation/key_ docs_en.htm#_simplification (accessed on 20 May 2015). 130  COM (2010) 543, point 2.4. See http://ec.europa.eu/smart-regulation/better_regulation/key_docs_en.htm (accessed on 20 May 2015). 131  COM (2010) 2020, point 2.1. 132  Commission, ‘Simplification Agenda for the MFF 2014–2020’ COM (2012) 42; Commission, ‘Minimising Regulatory Burden for SMEs’ COM (2011) 803. 133  See the Commission, ‘EU Regulatory Fitness’ (Communication) COM (2012) 746. It drew, amongst other sources, on OECD, ‘Recommendation on Regulatory Policy and Governance’ (Paris, OECD, 2012), available at www.oecd.org/gov/regulatorypolicy/49990817.pdf (accessed on 20 May 2015). 134 See the Commission’s summary document on fitness checks, available at http://ec.europa.eu/smartregulation/evaluation/docs/fitness_checks_2012_en.pdf (accessed on 20 May 2015).

EU Legislation

 253

13.9.6.  Better Regulation Package 2015 In May 2015, the new Commission (the Juncker Commission) adopted a package of measures ‘to deliver better rules for better results’, including a new agenda, better regulation guidelines and a better regulation toolbox, the establishment of a REFIT platform to make suggestions for reducing regulatory burdens, an independent Regulatory Scrutiny Board to advise the Commission and a proposal for an Interinstitutional Agreement on Better Regulation.135

13.9.7.  Agreement on Better Law Making 2016 On the basis of that proposal, in March 2016 the three institutions adopted a new Interinstitutional Agreement on Better Law Making. It replaces the 2003 Agreement, but addresses many of the same themes. In particular, it gives the European Parliament and the Council more influence over the EU’s legislative programming, stresses the importance of an evidence-based approach to legislation by means of reinforced impact assessments and ex post evaluation of legislation, establishes some ground rules for delegated and implementing acts, and outlines measures to improve transparency and to keep regulatory burdens in check.136

13.10.  Condensing and Updating the Statute Book 13.10.1. Amendment It is estimated that over one-third of Commission proposals each year concern amendments to existing legislative acts. Since it is clearly vital to keep EU legislation as accessible as possible, the institutions have agreed that amendments should ‘take the form of text to be inserted in the act to be amended’.137 This makes it possible to produce almost mechanically consolidated texts of the law in force at any given moment and also to codify and recast legislative acts that have been amended (see sections 10.3–5 below).

13.10.2. Corrigenda A particular problem with EU acts is the number of times they have to be corrected after adoption by means of correcting acts or of corrigenda published in the Official Journal for one or more language versions. A higher risk of errors is perhaps inevitable in a system with

135  COM (2015) 215, COM (2015) 2016 and other documents available at http://ec.europa.eu/smart-regulation/ better_regulation/key_docs_en.htm (accessed on 20 May 2015). 136  [2016] OJ L123/1. 137  Guideline 18. Recourse to substantive amendment is permissible in limited circumstances (JPG 18.14).

254 

William Robinson

24 official languages, but it is certainly liable to undermine the trust of citizens and traders in EU law.138

13.10.3. Consolidation Consolidation is a mechanical process whereby the provisions of a legal act (the articles and any annexes, but not the recitals) and all amendments and corrections to them are brought together in a single new text. That text is published on EUR-Lex for information only and the original act remains in force. Consolidation in all the official languages is carried out by the Publications Office under the supervision of the inter-institutional Working Group on Consolidation.139

13.10.4. Codification Codification is the adoption of a new act which brings together in a single text all the provisions of an existing act and all amendments and corrections already made to those provisions, without making any new amendment. The new act includes a complete and coherent statement of reasons. When the new act is adopted, the original act and all the amendments to it are repealed. In 1994, the EU institutions agreed on an accelerated working method for codification.140 In 2001, the Commission launched a project to codify the entire acquis,141 but the results were disappointing because of the technical problems involved.142

13.10.5. Recasting Recasting is the process whereby a new act is adopted which brings together in a single text all the provisions of an existing act and all amendments to those provisions. It differs from codification in that new amendments are also made. When the new act is adopted, the original act and all the amendments to it are repealed. In 2001, the institutions adopted an agreement on recasting.143 In recasting, the part of the text that corresponds to the existing provisions as already amended is treated as a codification and the legislative authority undertakes not to reopen discussion on those parts. The parts that are new are subject to the normal legislative procedure.

138  See M Bobek, ‘Corrigenda in the Official Journal of the European Union: Community Law as Quicksand’, European Law Review 2009, 950. 139  See COM (2001) 645, point 2.2. 140  [1996] OJ C102/2, point 5. 141  COM (2001) 645, point 1.3. 142  COM (2009) 17 final, point 5. 143  [2002] OJ C77/1.

EU Legislation

 255

13.10.6. Repeal Concern at the increasing volume of EU legislation led to a commitment from the EU institutions to simplify and repeal legislation wherever possible.144 The rule has been laid down that whenever a new act is adopted, any redundant provisions or acts should be repealed (Guideline 21). In addition, independently of the adoption of a new act, the acquis is being screened to identify acts that are no longer applied and to repeal them. Despite all the efforts made, the acquis continues to grow. In 2001, it was estimated at some 80,000 pages.145 Unofficial estimates of the current size of the acquis in connection with the translation of all binding EU law into Croatian put it at some 130,000 pages. The Commission is reviewing certain areas of EU law with a view to repealing unnecessary and irrelevant laws as part of its REFIT programme (see section 9.5 above).

13.11.  Final Words It is over 50 years since the CJEU first described the European Community as a new legal order. Since then, the European Union has continued to develop further the features of a legal order. It is in a symbiotic relationship with the legal orders of the 28 Member States. In many respects, it depends on them for the implementation and enforcement of EU legislation. It also draws inspiration from them for its development. But, in turn, it exercises a significant influence on all those legal orders. These elements are doubtless contributing to the development of a common European legal culture that goes beyond the actual texts themselves.

Further Reading Baaij, CJW (ed), The Role of Legal Translation in Legal Harmonisation (Alphen aan den Rijn, Kluwer Law International, 2012). Blanquet, M and Isaac, G, Droit général de l’Union européenne, 10th edn (Sirey, Sirey University, 2012). French Conseil d’État, ‘Rapport public 1992, Le droit communautaire’, Etudes et documents No 44. Jacqué, JP, Droit institutionnel de l’Union européenne, 7th edn (Paris, Dalloz, 2012). Kellerman, AE, Azzi, GC, Jacobs, SH and Deighton-Smith, R (eds), Improving the Quality of Legislation in Europe (The Hague, Kluwer Law International, 1998). 144  Agreement on Better Law Making [2003] OJ C321/1, point 35. The 2016 Agreement refers in s VIII more generally to simplification and the REFIT programme. 145  See COM (2001) 645, point 1.3.

256 

William Robinson

Koopmans, T, De kwaliteit van EG-regelgeving—Aandachtspunkten en voorstellen, 1995. Piris, JC, The Lisbon Treaty, A Legal and Political Analysis (Cambridge, Cambridge University Press, 2010). Stefanou, C and Xanthaki, H, Drafting Legislation—A Modern Approach (Aldershot, ­Ashgate, 2008). Steunenberg, B and Voermans, W, The Transposition of EC Directives (Leiden, Leiden ­University, 2006).

14 Legislative Training MARTA TAVARES DE ALMEIDA AND CHRIS MOLL

Regulatory policy is currently considered to be one of the three tools of state power, alongside budgetary and monetary policies. This explains the various measures adopted by different countries to improve the quality of regulatory policy. Amongst the different measures considered, legislative training is of major relevance. We analyse legislative training in two different sections: a survey on the importance of specialised legislative training for the quality of law, including descriptions of the different aspects of legislative training (cycle of regulation and law drafting), law making and training at EU level, and regulatory management training; and a presentation of legislative training programmes in five EU Member States: Germany, Poland, Switzerland, the Netherlands and UK. All of these are countries that are recognised by having developed a regulatory policy in order to achieve better regulation, but are also countries with different traditions, which naturally influence the organisation of training activities, allowing a comprehensive analysis of different training programmes.

14.1.  Specialised Legislative Training for Legislative Production 14.1.1. The Importance of Specialised Legislative Training for the Quality of Legislation In accordance with OECD studies, regulatory policy is one of the three tools of state power, alongside budgetary and monetary policies. It is for this reason that governments, parliaments and supranational organisations continuously work to improve the quality of their regulations. The quality of the legislative work depends on a number of aspects, namely: the institutional framework; the time allocated to the project development; the stages of legislative work; the entities involved in law making; training; the professionalism of the law production; a good relationship between the legislators (or politicians), who have the legislative power, and the legists (or legislative lawyers), who have the responsibility for proposing the norm’s content. In a recent text, it was underlined that Political decisions depend on the quality of the legislative work done by experts. Qualitatively good drafts are a prerequisite for appropriate political decisions to be made by the legislator … What is

258 

Marta Tavares de Almeida and Chris Moll

needed and what really contributes to the overall quality of the product is a good, constructive cooperation between the legists and the legislator.1

Another expert on legisprudence considers that a ‘Good quality of laws facilitates acceptance and implementation, poor quality creates problems of understanding, implementation, interpretation, and enforcement by executive authorities and the courts’.2

14.1.2. Development of Legislative Training Programmes at European Level Specialised training for legislative work is nowadays considered essential in various ­European countries, even though each country adopts a different model and has different priorities. Legislators (from parliaments and governments), as well as experts from administration, universities and the civil society, are interested in innovative methods for improving the quality of the legislative process. However, this approach is recent. The traditional methodology, the science of law, was focused on the interpretation and application of law, not on the production of law. In fact, the ‘genetic moments of law’3 were excluded from jurists’ education. In order to understand the current professional production of law, we must take into consideration developments in legisprudence. We can say that this leads to the professionalism of the law production. For a long time, training was only considered in relation to legislative drafting (formal legistics). This was the tradition in the UK. In fact, since 1869, with the creation of the Office of Parliamentary Counsel—a specialised service of legislative drafting for primary legislation—great expertise on legislative drafting has been developed. Expertise on legislative drafting is a feature of the Commonwealth countries, based on the English experience. The importance of specific training for the different stages of law making—namely Regulatory Impact Assessment (RIA)—is in fact fairly recent. It originated in the 1970s, in countries such as Germany, Switzerland and Austria, influenced by the publication, in Germany, of Gesetzgebungslehre,4 a fundamental work for law preparation. In 1998, the UK adopted a legislative policy from which has emerged the specific training for the different stages of law making. Later, in 1999, a seminal publication on the meaning of formal and material legistics5 was published. This publication also had a vital role in the development of specialised training for law making.

1  L Mader, ‘Congress Conclusions’ in L Mader and M Tavares de Almeida (eds), Quality of Legislation—­ Principles and Instruments—Proceedings of the Ninth Congress of International Association of Legislation (EAL) (Baden-Baden, Nomos, 2011) 324. 2  U Karpen, ‘Law Drafting and the Legislative Process: Outline of a Training Course for Law Drafters’ in L Mader and C Moll (eds), The Learning Legislator—Proceedings of the 7th Congress of the European Association of Legislation (EAL) (Baden-Baden, Nomos, 2009) 10. 3  JJ Gomes Canotilho, ‘Relatório sobre Programa, Conteúdos e Métodos de um Curso de Teoria da Legislação’ (1987) LXIII Boletim da Faculdade de Direito 5. 4  P Noll, Gesetzgebungslehre (Reinbek, Rowohlt, 1973). 5  C-A Morand, Eléments de Légistique Formelle et Matérielle—Formal and Material Legistic (Aix-en-Provence, Presse Universitaire D’Aix-Marseille, 1999).

Legislative Training

 259

There is now a clear distinction between ‘formal legistics’, which relates to legislative drafting, and ‘material legistics’, which relates to the norm’s content. It is recognised that specialised training is necessary for both.

14.1.3.  Academic and Professional Legislative Training Nowadays, we can make a significant distinction between the academic teaching offered by universities and the professional legislative training offered by governmental institutions (namely national schools of government) and specialised agencies. These two types of programmes have different approaches and objectives. In general terms, we note that at university level there is a more theoretical approach, and great emphasis is placed upon the international dimension of law making. Some universities offer legislative courses within their Master’s programmes, namely on public law. On the other hand, governmental institutions’ legislative programmes have, as a major objective, the training of experts for law making. Therefore, practical skills, namely assessment of legislation and drafting techniques, are of the essence. We must underline that, in the UK, drafting training is mainly ‘on the job’. There are varying degrees of structure, but in essence it is ‘learning by doing’, with junior drafters shadowing senior drafters and learning as they watch them do the job. The first training programmes on legistics were promoted by public institutions responsible for the general training of civil servants. The introduction of law-making programmes in the curricula of law schools occurred, in most universities, at a later date.

14.1.4.  Training Activities by International and National Associations Another important role in legislative training activities is played by international and national associations. At international level, it is important to emphasise the activity of the European Association of Legislation,6 a non-profit association, founded in Germany in 1991 and later renamed the International Association of Legislation (IAL). The name was changed in 2008, due to the significant number of new members from non-European countries and to the various activities of the Association at international level. The main goal of IAL is to promote the quality of legislation—through consultation, dissemination of studies and training activities on legistics. IAL also organises a biannual congress, and publishes the respective proceedings.7 For the Commonwealth, there is the Commonwealth Association of Legislative Counsel, which has members in many Commonwealth and non-Commonwealth countries around the world. In essence, it is an association of legislative drafters around the world who share ideas, publish articles and hold conferences.8 In Europe, there are national associations of legislation, namely in Austria, Estonia, Germany, Greece, Italy, Poland, Russia, Scotland and Switzerland, which promote lectures, conferences and training programmes on formal and material legistics.9 6 

See the website at www.ial-online.org. So far, IAL has published through Nomos then Congresses proceedings. 8  See the website at www.opc.gov.au/CALC/ and its journal, The Loophole. 9  See the IAL website at www.ial-online.org. 7 

260 

Marta Tavares de Almeida and Chris Moll

These associations are of great interest. It is important to create such associations in other European countries, as they are an efficient way of organising legislative training programmes at national level, diffuse new training approaches, and research on formal and material legistics.

14.1.5.  The Two Stages of Law Preparation Following the above-mentioned distinction between formal and material legistics, we can easily understand that we must consider two main stages and two different legislative training programmes: —— The initial stage of law preparation: identification of the problem; definition of objectives; identification of the options to solve the problem; consultation procedure; impact assessment of the different options. The training programme should concentrate on the cycle of regulation and public policy evaluation (‘emerging training’, as designated by some authors). —— The second stage: law drafting (traditional training). During the initial stage, it is necessary to have a multidisciplinary team (jurists, economists, sociologists) in order to obtain an interdisciplinary approach to the different aspects of the problem—which originated the legislative initiative—and to identify and evaluate the different options for solving the problem. Jurists may analyse the juridical frame at national, European and international level to achieve the best juridical solution; sociologists should present the data available which would allow a better comprehension of the problem and contribute to the efficacy and effectiveness of the new legislation; and economists should present cost–benefits analyses of the different possible solutions, in order to contribute to the efficiency of the new legislation. During this stage, it will be necessary, most of the time, to consult experts in the field under discussion. Due to the complexity of our societies, we note the increasing role played by non-lawyers in law making. As recommended by experts in the matter, ‘the art of legislation can not ignore the multiple knowledge and competences offered by the current sciences’.10 And specific training programmes, mainly on RIA, are essential. During the second stage—law drafting—jurists have a central and vital role. Therefore, it is necessary to design a precise training programme addressed especially to them. Jurists should have an in-depth knowledge of the ‘legislative history’ of the subject under discussion; the courts’ jurisdiction on the subject; and the direct and indirect relations of the subject with any other legislation. Furthermore, jurists have the onerous task of drafting laws in a clear and precise way, according to the rules of formal legistics.

14.1.6.  Law Preparation in Common Law and Civil Law Juridictions The training programmes are influenced by the way the legislative activity is organised. 10 JJ Gomes Canotilho, ‘Sobre o Clube dos Nomófilos/Nomógrafos e a difícil Arte de Legislar no Estado ­Contemporâneo’ (2009) No 50 Legislação: Cadernos de Ciência de Legislação 110.

Legislative Training

 261

In various European countries, most bills presented to the parliament are prepared at governmental level, due to the governmental initiative or the governmental participation in parliamentary bills. Therefore, legislative training programmes have a vital importance at governmental level. However, parliaments should also be involved in legislative training programmes, to achieve legislative quality. Very interesting references to the contribution of modern parliaments to legislative quality are made by Helen Xanthaki in a recent publication.11 Xanthaki also publishes the important conclusion of a report on ‘Standards in the Quality of Legislation’, prepared by the House of Commons Political and Constitutional Reform Committee, which considers: Parliament must have a stronger role as a partner with the Government in setting and monitoring standards of legislation. This will require a change of attitude by parliamentarians in asserting their role and ‘caring more’ about legislative standards, and in using existing processes and documents, such as Impact Assessments, more effectively. It may also require the creation of new mechanisms to assist them in the performance of their legislative duties. A change in attitude by Government is also required in its work with Parliament.

Xanthaki concludes that for Parliaments ‘the effort of training in legislative quality is great’.12 In fact, an institutional framework at political level, involving both parliament and the government, to allow the development of good practices on law making, including training, is considered to be essential. In most European countries, law preparation (policy process and drafting) at the governmental level is the responsibility of a single entity, a government department. In the UK, however, there is a separation between policy process and drafting (material and formal law preparation) of primary legislation. This means that the policy process of primary legislation is the responsibility of different government departments, whereas the bills are drafted centrally by the Office of Parliamentary Counsel. Drafting statues is just one part of the process of legislation, whereby an idea or concept concerning the social framework of society becomes government policy, is transformed to legislative shape by means of the drafting process, and eventually passes through the legislative machinery to reach the statute book as law.13

Drafters can have different roles in small and large jurisdictions, as recently analysed: Clearly there is a difference in the way drafting offices operate in large and small jurisdictions. Large jurisdictions usually have dedicated units or individuals who specialise in drafting. It is, therefore, possible for drafters to ‘insulate’ themselves from the different stages of the policy process. In contrast, in small jurisdictions the drafters’ position is more sui generis – depending on the idiosyncrasies of the jurisdiction and, of course, of the volume of work.14

In common law and civil law jurisdictions, there are training programmes and general principles addressed to legists and drafters responsible for law making. It is important to make the distinction between legists and drafters: the broader term of ‘legists’ can be employed to describe those who are involved in the different stages of the legislative policy process 11  H Xanthaki, Drafting Legislation—Art and Technology of Rules for Regulation (Oxford, Hart Publishing, 2014) 351. 12  ibid 352. 13  GC Thornton, Legislative Drafting, 2nd edn (London, Butterworths, 1979) 102. 14 C Stefanou, ‘Drafters, Drafting and the Policy Process’ in C Stefanou and H Xanthaki (eds), Drafting ­Legislation a Modern Approach (Aldershot, Ashgate, 2008) 321–33.

262 

Marta Tavares de Almeida and Chris Moll

(material and formal legistics), while ‘drafters’ describes those who are in charge only of the drafting procedure (formal legistics), according to the UK tradition. There is a general consensus that ‘good laws can be taught and learned’,15 but it is also recognised that training is not sufficient to produce good legislation. In fact, training is of utmost importance in achieving a good quality of legislation, but experience, mainly concerning the drafting process (formal legistics), is likewise necessary, as recognised by different experts from common law and civil law jurisdictions. In the UK, experts on legislative education and training consider that drafting entails both elements of art and elements of science, the drafter’s task entails both identification of all relevant circumstances and rules; and promotion of the most appropriate rule. And so the skills required are both an understanding of all relevant rules and wisdom through experience in the application of the most appropriate rule. These are the main skills that training in drafting must deliver. And they form the core of the reasoning behind the argument that training in drafting must be both academic and practical, both formal and via mentoring.16

The importance of practical training is also underlined by experts from civil law jurisdictions, considering that ‘the learning process still is and probably will always be “learning by doing” and “training on the job”’.17

14.1.7.  EU Legislation: Law Making and Training When we discuss legislative training in different European countries, the preparation of EU legislation, and consequently legislative training at EU level,18 cannot be ignored. In fact, the influence of European legislation on the legislation of Member States is well known. Therefore, the formal and material qualities of EU legislation are of great importance in achieving legislative quality at the national level. As mentioned in the Mandelkern Report, ‘The lack of simplicity, clarity and accessibility of European provisions … constitutes significant problems’.19 As is well known, at EU level there are three institutions involved in law making: the Commission—the only institution with legislative initiative, with some rare exceptions- the Parliament and the Council. Therefore, the first draft of a legal act is produced by the Commission ‘in the technical department by experts in the technical subject matter, not drafting experts … drafting expertise in the Commission is largely concentrated in the Legal Revisers Group, which almost always revises a draft produced by others’.20 The Legal Revisers Group is now known as the Quality of Legislation Team. It is part of the Legal Service of the Commission. The legal revisers (lawyers-linguists) of the Parliament and the Council have a later intervention in the legislative process, with all the problems of late revision. As mentioned, within the Commission, drafters generally are not jurists, but technical experts. Therefore, generally, they do not have any specific training in legislative draft15 

Karpen (n 2) 9. Xanthaki (n 11) 357. L Mader, ‘Legistic Training and Education in Switzerland’ in The Learning Legislator (n 2) 47. 18  We thank William Robinson (Institute of Advanced Legal Studies, University of London) for the information concerning training at EU level. 19  See http://ec.europe.eu/governance/better- regulation/documents/mandelkern-report.pdf. 20  W Robinson, ‘Making EU Legislation more Accessible’ in Quality of Legislation (n 1) 268. 16  17 

Legislative Training

 263

ing. They will learn from their more experienced colleagues and are often told to follow ­precedents in the same field as the one in which they are working. Very few training courses are given by legal revisers. Usually, legal revisers, upon starting such work, learn on the job while working in collaboration with the more experienced revisers. We underline two training activities addressed to legal revisers: —— A joint programme established by both the Parliament and the Council. Although the Commission is not associated with this programme, some of its experts join the training sessions. The programme includes aspects of material legistics (impact assessment and consultation) and formal legistics (legislative drafting). However, nowadays the main part of the programme is focused on legislative drafting. —— The organisation of ‘seminars on quality of legislation’.21 These seminars allow Member States’ experts to present comments on EU legislation and the EU experts to have a better understanding of the difficulties in interpreting and implementing EU legislation. According to William Robinson, investment in improving the very first drafts would pay off at every subsequent stage of the legislative process. If the first draft is good, the legal revisers can polish it to an even higher standard but if the first draft is poor, they struggle to identify the intended meaning and often have to work in the dark, as there is rarely the dialogue between the originating department and the revisers which should be the ideal. To make that possible, the Commission should act to establish centres of drafting expertise in every DG which is involved in the legislative process.22

The creation of specialist drafting units was one of the implementing measures agreed on by the European Parliament, the Council and the Commission.23

14.1.8.  Regulatory Management Training Another important dimension which deserves attention is that of ‘regulatory management training’. Claudio Radaelli addresses interesting questions on this subject:24 what are the approaches to training and what is the link between training and better regulation leadership? When referring to the approaches to training, Radaelli underlines the trend for ‘teaching in a top-down style’, which assumes that, in different countries and situations, the same model applies to consultation, risk assessment and cost–benefit analysis. Radaelli underlines the risks of this strategy because, in this situation, ‘the administrative context,

21  Details of the seminar programme are available on the website of the Commission Legal Service: http:// ec.europe.eu/dgs/legal service/legal reviser en.htm #3. 22  Robinson (n 20) 268, 269. 23  Interinstitutional Agreement of 22 December 1998 on common guidelines for the quality of drafting of Community legislation [1999] OJ C73/1. 24  C Radaelli, ‘Regulatory Management: Training Whom, How and For What Purposes?’ (2009) No 50 Legislação: Cadernos de Ciência de Legislação 221.

264 

Marta Tavares de Almeida and Chris Moll

the relationship between the core executive and the regulators, the hearing –consultation tradition’ are not normally considered. On the other hand, the trend to teach according with the peculiarities of each country is what Radaelli names the ‘bottom-up approach’. This approach is considered more realistic because it is not possible to address, in the same way, better regulation tools in countries which have a very different political, economic and social context.25 Radaelli recognises that training programmes have to adapt to the individual country conditions, taking into consideration, by intensive interaction, the knowledge and possible suggestions made by the individual trainees for achieving better regulation in their countries. But a strategy is also necessary that allows the introduction of a new approach to law making, considering tools that were not used, which can contribute to achieving better regulation. The idea should be to demonstrate to the participants ‘that better regulation training is about practical knowledge, not abstract and generic knowledge’.26 Concerning the role of leadership, the expressions ‘training the leaders’ and ‘training the trainees’ are nowadays discussed in better regulation circles. What does they mean? It is considered that ‘senior management’ should be involved in workshops to allow them to understand at first hand and support better regulation policies—namely RIA, law enforcement, implementation and risk assessment. Regarding the idea of ‘training the trainees’, it is recognised that the trainees do not have associations that can discuss regulatory management, including methodologies for different aspects of law making in different countries and situations, and it is considered that such associations would be very helpful in the discussion of new training approaches.

14.2.  Legislative Training Programmes at European Level Following the introductory survey, indicating the relevance of specialised training for the quality of legislation, in this section we analyse how the in abstracto striving for legislation of enhanced quality has been transferred to and implemented in training practices. The country cases of choice have been limited, for brevity, to five countries: Germany, Poland, Switzerland, the Netherlands and the UK. The selected practices (in development) are those that have shown distinctive features in terms of their origin and aims, governance structure and knowledge base (academic or practitioner-oriented), programme design and target audience.

14.2.1.  Origin and Aims Which problems or shortcomings spark the need for training? And what does a training programme aim at? 25  ibid: ‘The rationale for this approach is provided by recent research that has empirically demonstrated the existence of different types of RIA across Europe. This body of research argues that we should not encapsulate the substance of different types of assessment into one formal definition of RIA … and therefore we should teach the subject accordingly’ (226). 26  Radaelli (n 24) 227.

Legislative Training

 265

In a comprehensive outline, Ulrich Karpen diagnoses the process of legislative drafting as suffering from a number of deficiencies. Many parliaments, unicameral and bicameral alike, are unable to organise a strong secretariat tasked with the examination of a bill, or to render support for the law-making process. Units of the responsible ministries apply ad hoc methods to draw up the technical, material details of draft laws. In addition, it is clear that there is a lack of coordination between ministries and agencies in law making. There is a generally observed lack of co-operation between governmental actors and organised interest groups. This lack of public participatory input leads to a loss of information, but moreover it creates problems in implementation. The inability to organise a well-functioning legislative unit, or cell, in the government is rooted in a lack of resources and technical manpower. Not least, overburdened MPs lack sufficient assistance in law-making labour and generally are not very knowledgeable in the ways of writing laws. Addressing these deficiencies involves interventions of an organisational nature, as well as the teaching and learning of methods of good drafting.27 Addressing the organisational and procedural matters involved is one aim (which will not be dealt with in this context). The other aim relates to facilitating the creation of sound legal drafts through expert input and appropriate consultative tools and methods. This concerns the offering of learning opportunities for those who will have to prepare workable, technically acceptable laws. The professionals involved will have to become familiar with the various stages and forms of the legislative process. The purpose of a training programme originates from the deficiencies in the legislative drafting process. Another interesting approach comes from Luzius Mader, who considers that it is necessary to adopt a realistic approach and to avoid an overestimation of the contribution that can be expected from teaching or training, be it academic or professional. Academic teaching may be instrumental in preparing for the learning-by-doing experience; professional experience can support it. However, neither will be able to substantially replace practical experience. On balance, these imposed limits align with the aims for training in legislative drafting in the context of Switzerland.28 In the Netherlands, a decision of the European Court of Justice helps one to understand the developments for training legal drafters. Without this decision, the organised training of legislative drafters in the Netherlands would probably still be non-existent. In the Securitel case,29 the European Court decided that technical requirements, laid down in national law without taking into account certain procedural requirements stemming from European Directive 83/189, are not binding and must not be applied. Details of the particular European procedural requirement at stake are as follows: a Member of the European Union must notify the Commission of every plan concerning the issuing or amendment of technical regulations. Effectively, this is how the EC prevents Member States from imposing new national barriers that could—potentially—endanger the proper working of the European International Market. With its ruling, the Court in fact decided that national statutes, which are made in accordance with national rules covering legislative procedure but which overlook European procedural rules, are not really statutes at all because they do not have any effect. 27  28  29 

Karpen (n 2) 9. Mader (n 17) 47. Court of Justice of the European Union, Case C-194/94 Securitel Sprl [1996] ECR I-2201.

266 

Marta Tavares de Almeida and Chris Moll

The case had a serious impact on Dutch legislation, as the Dutch legislator apparently failed to issue these required notifications systematically. Legal experts in The Hague ­realised that more than 100 laws that had been applied for years were suddenly judged to be illegal because procedural demands had not been met. In a major effort, all Dutch statutes were investigated in order to find out which ones were affected, and this was followed by a repair operation. As a result of all this, a committee was established to review the lawmaking departments. This Visitation Committee for Legislation was tasked with making recommendations on how to improve the quality of legislation. The Committee concluded that the quality of legislation also concerned the quality of the legislators. It recommended that the legislative departments of the ministries should strive for more professionalism. It also criticised the fact that there were too few training facilities for legislative drafters, and no structured system of courses and programmes was offered by universities or other institutes. The ministries and the Council of State endorsed the conclusions and recommendations of the Committee, resulting in the founding of the Dutch Academy for Legislation. On balance, the Securitel case brought to light the fact that the legislative drafters in the ministries are critical actors, because their malfunctioning may have fatal consequences. Furthermore, it was acknowledged that the rules and procedures relating to legislation and law making had become so vast and complex that it had become necessary to train legal drafters in how to apply them when engaging in law-drafting labour. Finally, even if a Dutch legal drafter applies the Dutch rules and procedures correctly, failure may be around the corner if the European law and its consequences are overlooked. Though the Court of Justice of the European Union is probably unaware of this, its ruling triggered the foundation of the Dutch Academy for Legislation, and articulated the aims for an organised programme for the training of legal drafters.30 In the UK, when we speak of legislative training, we must think about the policy process and legislative drafting.31 The approach to law drafting is heavily influenced by general common law traditions. This is evident in the treatment of drafting as a specific activity, distinct from law preparation. In the UK, the development of centralised drafting of primary legislation—through the Office of Parliamentary Counsel—has contributed to the growth of common standards and practices. This distinction has important consequences in the way that training programmes are considered and designed.

14.2.2.  The Organisation of Training in Member States How is the development of training programmes organised? This question refers to the governance of the knowledge base. Evidently, the input of qualified experts is a critical factor for the development of successful training. Where to find skilful, knowledgeable people to train legal drafters? Are they to be found in academia, amongst experienced practitioners, or perhaps in both circles? In Germany, in the proposed comprehensive outline for developing a training programme, reference is made to involving a wide circle of stakeholders, such as MPs, officials employed at parliamentary secretariats, governmental agencies, members of the judiciary 30  31 

P van Lochem, ‘The Dutch Academy for Legislation’ in The Learning Legislator (n 2) 63–64. See above.

Legislative Training

 267

and representatives of non-governmental organisations (NGOs). Members of these entities would ideally be represented in a joint Programme Management Committee, and decide about the content and format of training. From an analysis workshop, competent manpower in the drafting field will be created for preparing the courses, teachers and coordinators appointed and the curriculum designed. Furthermore, the Programme Management Committee will establish ex ante indicators for a sound assessment. Finally, the Committee will also monitor the development of a training programme and conduct an open review of it. Training must rely heavily on practical exercises. It is the practitioner who shares and exchanges individual drafting experiences with his or her peers and connects law and policy. In Poland,32 as in the other countries considered, there is a distinction between training programmes offered at governmental and academic level. Both programmes have as their main objective the professional preparation of legists and drafters. The Government Legislative Centre, a governmental agency, organises every year a course on law making. The Faculty of Law and Administration of Warsaw University has organised annually, since 1973, a course on legislation practice. We believe that this course is one of the first at academic level. In Switzerland, priority is given to practitioner experience as input for a training programme. However, this preference for practice over theory as a knowledge base does not exclude the involvement of academia in training for legal drafters. The University of Geneva has been providing courses on legistics for about 40 years, and more recently the University of Zurich has incorporated it as an optional course in their pubic law curriculum. The Universities of Bern and Lucerne offer legislative courses in their Master’s programme. Finally, legisprudence is part of a Master of Public Administration, offered jointly by co-operating French and Italian speaking universities, including the University of Bern. A Master’s programme with a focus on legistics or legislation is deemed to be too narrow in concept. Furthermore, professional training is provided by the Swiss Association of Legislation (SAL), the Swiss Federal Office and the University of Zurich. All entities organise seminars. Finally, the Swiss Graduate School of Public Administration (IDHEAP) has combined academic teaching and professional training, and has incorporated this mix in some of its (Executive) Master’s programmes. The University of Geneva developed the CEFOLEG diploma programme—a distance-learning programme.33 The SAL established a governing committee for legislative education and training (Ausbildungsrat), tasking it with the analysis of those employed in public administration, with regard to education and legislation. The committee takes part in the development of concepts for seminars and courses, and sets ‘terms of reference’ for their form and content. Finally, it decides where the seminars are to be organised, and issues –for seminars and courses—merit awards, amongst which is the SAL label. The SAL agrees contractually with institutes, organisations and individuals on the offering of courses and seminars. With the introduction of a new concept for the seminars in Murten and Montreux, developed under the auspices of the SAL, it looks for the involvement of the Universities of Fribourg, Geneva, Berne, Zurich and Lucerne as organising partners.34 With this new concept, the SAL aims at better addressing the stakeholder needs for training offerings. Yet another education and 32  We thank Professor Wieslaw Staskiewicz, Faculty of Law and Administration, Warsaw University, for the information concerning Poland. 33  Mader (n 17) 48–49. 34  Mader (n 17) 50–52.

268 

Marta Tavares de Almeida and Chris Moll

training concern for it may arise at federal level, as the Swiss Federal Office has been considering the long-term outsourcing of the organisation of its seminars. This would be in line with the aim of the Association to unite all entities and individuals that are willing and capable of contributing to the improvement of legistics training in Switzerland.35 In the Netherlands, the fact that the Academy of Legislation is tasked with the recruitment, selection and training of trainee legislative lawyers is an indication that training is not motivated by academic interest, but by the interest of educating legal professionals to make them capable of doing their job. The stakeholders (ministries, Council of State, Parliament) keep closely in touch with the Academy in a governance structure that consisting of three bodies: the Management Board, with the involvement of a representative of the Council of State, the Ministry of Justice and the Dean of the Academy, meet annually to reflect on developments that are undertaken. The Council of Participants, in which the directors of legislation of the stakeholder organisations are represented, acts as an advisory board, and submits advice on desired policy and action. Finally, the Curriculum Committee advises on the content and didactic concepts of the Academy’s training programmes. The Academy’s knowledge base is dual in nature. In the development of its training programme, it incorporates predominantly practitioner experience by committing seasoned experts of stakeholder organisations, together with academic expertise, provided by nearly all (law faculties of) Dutch universities.36 Legislative Studies is part of the academic curriculum of several Dutch universities. The Faculties of Law of the Universities of Maastricht, Tilburg, Leiden and Groningen offer courses on Legistics. At the Universities of Maastricht and Leiden, Legislative Studies is taught within the framework of the Constitutional and Administrative Law Masters. At Tilburg University, Legislation and Regulation is an (optional) part of the Administrative Law and Governance Law Masters, as well as the LLM Law programme. Finally, Philosophy of Legislation and Regulation is being taught at the University of Groningen. In the UK, we must consider two sides of the legislative policy process. We first analyse drafting activity, due to its specificities and importance. The Office of Parliamentary Counsel centralises the drafting of primary legislation for the government.37 The Office of Parliamentary Counsel is composed of jurists who are highly competent at law preparation, with a deep knowledge of law and legislation and much experience of legislative drafting. This knowledge and practical experience are transmitted to the new members of the Office of Parliamentary Counsel, who are recruited from good jurists with some practice in the private sector and who, when admitted, initially work with one of the experienced permanent jurists. Generally, it is considered that this system allows uniformity in law drafting and a legislative style based on a pool of experience. There are, however, some critics of the system, who point out that nowadays there are not many jurists interested in this activity, because it is a very specialised and demanding activity. It is also emphasised that the system depends on the quality of the instructions coming from the different governmental departments. In 1975, the Renton Committee38 suggested that a preliminary training programme should be of the utmost importance, as it would reduce the apprenticeship period and 35 

Mader (n 17) 52. van Lochem (n 30) 66–67. 37  There are three further offices which draft primary legislation (for Scotland, Wales and Northern Ireland, respectively) and which follow a similar model. 38  Lord Renton, The Preparation of Legislation, 1975, para 8.15. 36 

Legislative Training

 269

l­iberate the senior members from their training obligations. This proposal was refused by the Office of Parliamentary Counsel, as mentioned by Professor Keith Patchett.39 In the UK, there is great expertise on legislative drafting and various training programmes on that subject. The Civil Service College offers short training programmes, namely on drafting secondary legislation, addressed to jurists from different departments and also from the Commonwealth countries. The Institute of Advanced Legal Studies (IALS), University of London, has offered, since its foundation, in 1947, advanced research and studies in law, namely law and legislative studies. Special reference must be made to the Legislative Drafting course, directed from 1964 till 1999 by Sir William Dale, a well-known expert on the subject. Since that date, the course has had a new director and orientation. The new programme, organised with intensive daily lectures and tutorial seminars, analyses the legal and political framework of legislative drafting. Besides the sessions on theory of drafting, intensive drafting exercises are the main tools of the tuition. Nowadays, the IALS offers an LLM in Advanced Legislative Studies in the traditional taught format and as a distance-learning programme. IALS also runs a Legislative Drafting Clinic, which offers a chance for students to practice their drafting skills on real-life problems for real clients.40 The University of Strathclyde offers a Legislation course, and a Postgraduate degree is offered by its Centre for Parliamentary and Legislative Studies. An interesting programme is now offered by the University of Exeter: Regulation and Reform: Analysis and Policy, an optional postgraduate module of the Master of Public Administration, in which arises the question of whether better regulation has also produced efficient and democratically legitimate regulation.41

14.2.3.  Training Activities Programme Design What are the ingredients of a training programme, and what form does it take? From the German perspective, the following components are valued as indispensable. The first concerns better understanding of the legal systems and legislative techniques. This covers a range of topics: an overview of the legal framework is evidently part of this, as is the relationship between policy and law. The various stages for analytical purposes are relevant in this respect as well. Improving the quality of legislation comes second, and is relatively straightforward. Drafters can be taught drafting rules and conventions, and can be given practical exercises in order to improve the technical quality of draft laws. The quality of the substance of the drafts is another issue to be addressed, with the help of (stakeholder) experts and others, embedded in the legislative process.42 A series of interrelated seminars and workshops has been suggested as a desirable format for training legal drafters. The seminars would be incorporated directly into the training of legislative drafters. During the seminars and workshops, the participants would act, in

39 

K Patchett, ‘Legislação e Redação Legislativa no Reino Unido’ (1991) No 2 Legislação 41. See http://ials.sas.ac.uk/postgrad/LDclinic/LDclinic.htm. 41  See http://www.exeter.ac.uk/postgraduate/taught/politics/public-administration/#Programme-structure. 42  Karpen (n 2) 11. 40 

270 

Marta Tavares de Almeida and Chris Moll

a simulation, as legal drafters attending public hearings, round-tables and parliamentary debates. There are research and teaching centres at the Universities of Berlin, Speyer and Augsburg. In Poland, the annual programme organised by the Government Legislative Centre is a priority for the preparation of civil servants from the government, Parliament and the Presidential Cabinet. This programme is also open to law graduate students. The course involves an average of 40 participants. The Legislative postgraduate programme of the Faculty of Law and Administration, offered each academic year, lasts for three semesters, ending with a dissertation. It is open to graduate students from different subject areas, so it is not necessary to have a law degree. This is a popular programme, being attended by an average of 80 students. In Switzerland, under the auspices of the SAL (founded in 1982), seminars are organised in Murten and Montreux. The seminars in Murten, organised by the University of Fribourg, basically consisted of three parts: an introductory course, which was succeeded by more specific parts, covering legal, linguistic and methodological issues, such as European legislation, the structure of legal acts and legislative evaluation. The seminars in Montreux were managed by the University of Geneva, and divided into two parts: material aspects of legislation (problems, goals, instruments) on the one hand, and legal drafting and linguistic aspects on the other. In both sets of seminars (Murten and Montreux), scholars and practitioners were involved as teachers. In addition, the offerings were similar in their combinations of lectures and exercises, and had a shared target audience. However, Murten and Montreux differed in number of modules (four versus two), and Murten mainly dealt with legal and linguistic issues, while Montreux covered more material legistics. In 2004, the SLA introduced a new concept developed by its special committee, amending both the content and the format of the Murten and Montreux seminars. Regarding the content, a methodological approach has been developed, as has the formal legistics. Additional seminars are optional. Furthermore, each seminar is now the result of contractual agreements between interested administrative entities, organising institutes and university faculties. Finally, the number of participants is limited, to ensure maximal interaction and exchange. On yet another level, the Swiss Federal Office develops a seminar to address the needs of legists in the federal administration. The seminar consists of two parts: the first part concerns legal theory; the second part focuses on specific aspects of the legislative procedure, on special law drafting relating to formal aspects of logistics and on an introduction to the Manual of Federal Law Making, developed by the Swiss Federal Office.43 CEFOLEG completes the Swiss menu of training programmes for legal drafters. Unlike the seminars, this is an online (distance-learning) French language diploma course, developed by the University of Geneva for legal professionals. The internet feature of the course, an innovative initiative, made it the first of its kind (in 1998) in Switzerland.44

43  Federal Office of Justice, Guide de Legislation [Guide for the Elaboration of Federal Legislation], 3rd edn (Bern, Federal Office of Justice, 2007), available at www.bj.admin.ch/dam/data/bj/staat/legistik/hauptinstrumente/­gleitf-f.pdf. 44  Mader (n 17) 50–52.

Legislative Training

 271

In the Netherlands, though the growing relevance of theory is acknowledged, there is a firm belief that practice is the best way for the legal drafter to learn his or her profession. Basically, the lecturer at the Academy for Legislation and the tutor at the working place jointly have the task of establishing the conditions for the trainee to become a professional legal drafter. The accredited Master of Legislation takes a period of two years, in which the trainees spend 30 per cent of their time in class, while spending the remaining 70 per cent in their job at the ministry, Council of State or Parliament. A substantial part of the training deals with theoretical and practical aspects of legislation, in particular the theory of legislation, legislative procedure, the technique of legislation and the philosophical approach to legislation. A next part of the programme deals extensively with administrative law, constitutional law and European law. Another part consists of modules on political and social issues. The aim of these modules is to make trainees aware of the political process, the political culture and the way the law works in society. The training course consists of lectures, practical exercises, (international) working visits and internships. In addition, the trainees have to fulfil an internship at the House of Representatives of the Dutch Parliament. Many of the lecturers are professionals in the field of legislation, mostly employed in the central government. Other input is provided by university experts, mostly working as (associate) professors at Dutch universities. The curriculum is implemented in co-operation with the programme managers and a small coordinating staff.45 The Academy also develops courses for other professionals as well as for trainees, and codevelops special and tailor-made courses for entities that are part of the central government of the Netherlands. To a large extent, the content of the courses resembles the modules that are part of the master programme.46 In the UK, legislative drafting training is an old tradition, supported by different training programmes offered by public and private institutions and by a large number of publications about legal writing and drafting legislation. The Institute of Advanced Legal Studies plays an important role—sessions on theory of drafting, small classes and intensive drafting exercises are the main tools of tuition. The Statute Law Society, a charitable body, is also developing a series of lectures and conferences about the legislative process, addressed to the legal professions and the public. A more theoretical approach to Legislative Studies is offered at academic level.

14.2.4.  Training Activities Target Audiences A highly relevant question for those involved in the development of a training programme is: who are the addressees of the offering? In other words: what is the target audience of a particular training programme? In Germany, in the proposed model for a training course for legal drafters, the addressees coincide with a wide circle of stakeholders, such as MPs, officials employed at parliamentary secretariats, governmental agencies, members of the judiciary and representatives of NGOs. Alternatively, the target audience can be sub-categorised into key government actors and key non-governmental actors.47 45  46  47 

van Lochem (n 30) 66. van Lochem (n 30) 66. Karpen (n 2) 9, 16. The latter page lists the stakeholders groups, divided in sub-categories.

272 

Marta Tavares de Almeida and Chris Moll

In Poland, the target audience is composed of civil servants in the programme organised by the government, and by graduate students in the postgraduate course offered by the Faculty of Law and Administration, Warsaw University. In Switzerland, the target audience of the seminars typically consists of professionals employed in federal, cantonal and municipal administration. However, there is growing acknowledgement that the training should be addressed not only to jurists, but also to others professionals.48 In the Netherlands, the admission and selection criteria for the Master of Legislation programme clearly states who the target audience of the Dutch Academy for Legislation is. Aspiring trainees must hold a Dutch law degree, obtained at a Dutch university, and must also meet some additional demands. With regard to other training opportunities, the participants have to be employees of an entity of the Dutch central government (departments, councils, agencies), though not necessarily in the capacity of a legal expert. In the UK, the legislative drafting programmes are mainly addressed to jurists of the different government departments responsible for drafting secondary legislation, and also jurists from Commonwealth countries. At academic level, postgraduate programmes on legislative issues are mainly addressed to students of law and social sciences.

14.3. Summary The origin of the training programme for legal drafters stems from either the threat of ‘legislative disaster’, as happened in the Netherlands, or from a general effort for better legislation amongst the professionals, as notable in Germany, Poland, Switzerland and the UK. A shared aim of the training programmes is to equip legislative experts better for the job. Training programmes are organised with substantial involvement of the professional community. In Germany, a wide circle of stakeholder groups is thought to be involved in defining the ‘terms of reference’ of a training programme in the Project Management Committee. This governing role for practice-oriented training in Switzerland has been adopted by the SAL, which is gradually seeking a more comprehensive coordinating role for the development of training at all levels of the Swiss government. Another feature common to the training programmes in the various country cases is the predominant orientation on professional experience prior to training: —— In the German model, training relies heavily on practical exercises. —— In Poland, the programme organised by the Governmental Agency and addressed especially to civil servants is based on legislation practice. —— In Switzerland, learning by doing and training on the job are indispensable parts of learning the legislative process. Additionally, the Swiss adopt the approach that an overestimation of the contribution of training, be it academic or professional, must be avoided. 48 

Mader (n 17) 51 and 53.

Legislative Training

 273

—— In the Netherlands, the rationale for training legal professionals comes mainly from the practical necessity of creating a well-skilled and educated workforce. Therefore, professionals play a very important role in the curriculum. —— In the UK, training on the job by senior professionals with much experience is an important characteristic of training programmes. The other significant aspect of legislative studies programmes is the practical training approach based on the presentation of case studies, exercises and tutorial sessions. However, in the above-mentioned countries, academia is also involved. This is more pronounced in Switzerland, Poland and the Netherlands, where universities or members of their academic staff are substantially involved in the development of training. The training offered ranges from workshops, seminars and courses to fully developed modular Master’s programmes. The content is often of a practical, technical legistics nature, and to a lesser extent supported by an overall theoretical framework, though its perceived relevance is growing. Finally, a note on the target audience of governmental training programmes: this should comprise of drafters and legists grouped under the common nomination of ‘legislative professionals’. However, it is acknowledged that ‘the participant’ should not be defined too narrowly. In a professional sense, more categories than simply (partially) legally trained professionals could also be interested in taking up some of the training programmes on offer. The target audience should come from all levels of government, as favoured in Germany and Switzerland. However, in the Netherlands and the UK, the circle of participants for training is composed mainly of those working in central government. In Poland, the participants come from the government and the Presidential Cabinet.

Further Reading Almeida, Marta Tavares de, ‘Légiste- une nouvelle profession?’ in Liber Amicorum Paul ­Delnoy, Larcier, 2005’, 1123–1132. Bergeron, Robert ‘The training of Drafters’ in Ulrich Karpen/Paul Delnoy (eds), ­Contributions to the Methodology of the Creation of Written Law (Baden-Baden, Nomos, 1996) 101–105. Delnoy, Paul, ‘The role of Legistics and Legists in the Determination of the Norm Content’, in Ulrich Karpen/Paul Delnoy (eds), Contributions to the Methodology of the Creation of Written Law (Baden-Baden, Nomos, 1996) 25–33. Gwizdz, Andrzej, ‘Organisation of the Post-Graduate Course in Legislation at the U ­ niversity of Warsaw’, in Ulrich Karpen/Paul Delnoy (eds), Contributions to the Methodology of the Creation of Written Law (Baden-Baden, Nomos, 1996) 106–110. Patchet, Keith, ‘Legislação e Redação Legislativa no Reino Unido’, Legislação, Cadernos de Ciência de Legislação, nº2, Out-Dez. 1991. Robinson, William, ‘Drafting of EU Acts: A View From the European Commission’, in ­Constantin Stefanou/Helen Xanthaki (eds), Drafting Legislation –A Modern Approach (Ashgate, 2008) 177–197.

274 

Marta Tavares de Almeida and Chris Moll

Stefanou, Constantin/ Xanthaki, Helen (eds), Drafting Legislation – A Modern Approach (Ashgate, 2008). Mader, Luzius/ Moll, Chris (eds), The Learning Legislator-Proceedings of the 7th Congress of the European Association of Legislation (EAL), Baden-Baden, Nomos, 2009. Xanthaki, Helen, Drafting Legislation-Art and Technology of Rules for Regulation (Hart ­Publishing, 2014). Zapatero, Virgílio, El Arte de Legislar (Editorial Aranzadi, 2009).

15 Emerging Trends in Legislation in Europe HELEN XANTHAKI

This chapter attempts to use the wisdom offered by the authors in the previous chapters in order to define legislative quality in the EU—and to contemplate future initiatives in the field. ‘­Blue-sky’ choices or future reality, only time will tell. The book ends with a look at the future, in hope of more work and more success in legislation for academics and practitioners.

15.1.  The EU’s Regulatory Framework for Legislative Quality1 The EU’s long engagement with legislative quality is characterised by a strong political driving force behind the regulatory agenda by a number of Member States.2 Although this could have led to a common law prevalence in drafting conventions, the emerging legislative style is a unique EU one. It admirably balances common and civil law structures, thus accommodating both families of law. At the same time, it respects the legal intricacies of EU law, such as the prevalence of consistent EU case law as a source of binding legal norms. The framework within which this dynamic process takes place is via a long list of documents, mainly of a non-binding nature, that set rules for the process and format of legislative drafting applicable to all EU institutions. The first set of concrete quality-promoting rules is expressed in the 1992 Sutherland Report. Lord Sutherland sets five criteria that each new legislative measure must meet as a means of achieving legislative quality. These are: the need for regulatory action; the choice of the most effective course of action, be it legislative or alternative; proportionality of the measure to the desired regulatory results; consistency of the proposed measure with existing measures; and wide consultation at the preparatory stage. But how could one put the Sutherland Report into effect, with specific reference to drafting conventions and rules? Two months after Lord Sutherland’s report, in December 1992, 1  H Xanthaki, ‘The Problem of Quality in EU Legislation: What on Earth is Really Wrong?’ (2001) 38 Common Market Law Review 651. 2  For example, the UK initiatives below, the Dutch contribution with the Koopmans Report, the Scheveningen Conference and the Amsterdam IGC.

276 

Helen Xanthaki

the European Council demanded clearer and simpler legislation.3 In June 1993, the Council clarified these concepts, and introduced concrete measures for their attainment by use of a Resolution on legislative quality: the Resolution touched upon wording, structure and consistency with the content of existing legislation.4 In an attempt to make legislation more accessible, the Resolution called for clear, simple, concise and unambiguous wording. Practices encouraged by the Resolution were: the use of the same term throughout the act, the use of the accepted structure of chapters—sections—articles—paragraphs, compliance with the role of the preamble as a means of justification of the enacting provisions in simple terms, clear determination of the rights and obligations deriving from the act, clear reference to the act’s date of entry into force, and consistency of the provisions of various acts. Practices discouraged by the Resolution were: the use of unnecessary abbreviations, Community jargon, long sentences, imprecise references to other texts, too many crossreferences, political statements without legislative character, pointless repetitions of existing provisions, and inconsistencies with existing legislation. With specific reference to acts amending earlier legislative texts, the Resolution discouraged the inclusion of autonomous provisions, which could not be directly incorporated into the existing act. The Resolution is very similar to a legislative manual in the national context, but it is much shorter, and includes an imbalanced range of concrete conventions and general principles for drafters. Although the Resolution cannot be attributed with originality or innovation in the prescribed drafting rules and conventions, its great value lies in its unprecedented (for the EU) focus on legislative quality, accessibility of legislation and the concept of set, clear criteria for the evaluation of EU legislative texts.5 But a Resolution carries little binding force, and therefore limited legal value; and in any case this Resolution was of a general prescriptive character. Therefore, the onus was for the various actors of the legislative process to introduce concrete measures that put the Sutherland Report and the Resolution into effect in specific areas of law making. Thus, in December 1994, the Council, the Commission and the Parliament set out to detail the guidelines of the Resolution in a concrete, albeit still not binding, Interinstitutional Agreement on the official codification of legislative texts. The Resolution and Agreement are, of course, directly linked: the Resolution declares that ‘the general objective of making Community legislation more accessible should be pursued not only by making systematic use of EU consolidation but also by implementing the following guidelines’.6 In June 1995, the Molitor Group applied the Sutherland Report criteria to legislative and administrative simplification.7 This was supplemented in 2003 by the notion of prioritisation for the purposes of simplification.8 3 

See Edinburgh European Council, ‘Conclusions of the Presidency’, Bulletin EC 12/92, 7. See Council Resolution of 8 June 1993 on the quality of drafting of Community legislation [1993] OJ C166/1, 1. 5  These rules are also included in the Council Decision 93/662/EC of 6 December 1993 adopting the Council’s Rules of Procedure [1993] OJ L304/1. 6  See Interinstitutional Agreement of 20 December 1994, ‘Accelerated working method for official codification of legislative texts’ [1995] OJ C293/2, [1996] OJ C102/2. 7  See Commission, ‘Report of the Group of Independent Experts on Legislative and Administrative Simplification: summary and proposals’ COM (95) 288 final; see also the Commission’s comments on the report SEC (95) 2121 final; the European Parliament Resolution of 4 July 1996 [1996] OJ C211/23. 8 See Commission, ‘Communication from the Commission to the Council, the European Parliament, the ­European Economic and Social Committee and the Committee of the Regions—updating and simplifying the Community acquis (SEC (2003) 165)’ COM (2003) 0071 final. 4 

Emerging Trends in Legislation in Europe

 277

Concurrently, the Commission took it upon itself to provide a more principled, though still not binding, stance on law making. In 1995 and 1996, the Better Law-Making Reports9 identified the regulatory aims of the EU’s legislative policy: legislative quality, consistency, openness in the drafting process, carefully planned and co-ordinated legislative process, and thorough monitoring and evaluation of the legislation. In parallel to a focus on legislative quality at the EU level, the Commission identified the crucial role of legislative quality in the national implementing measures as a factor of quality of EU regulation. This is evident in the Commission’s 1996 SLIM (Simpler Legislation for the Internal Market) initiative, which, with the strong encouragement of Internal Market Ministers, targets simplification of both EU and national implementing legislation.10 The theme was confirmed in June 1997, when the Commission’s approach to simplification was expressly extended to national measures,11 and also in the Commission’s 1997 Better Law-Making Report, which called upon the drafting authorities of the Member States ‘to work towards the effective implementation of Community law and play an active part in the process of improving the quality of legislation’.12 Again, in the Informal Meeting of Internal Ministers of 13–14 February 1998, the ministers unanimously stressed their strong support for legislative simplification of national and Community rules. The rationale for this policy was expressed clearly in the 1998 Better Law-Making Report 1998: ‘A Shared Responsibility’,13 which pronounced Member States as ‘the main producers of legislation and hence the most direct cause of the burden [on firms]’. In a series of documents, the Commission extended its emphasis on the quality of legislative measures by the Member States to the need for monitoring and evaluation of transposition of EU instruments both pre- and post-accession.14 But what exactly are the elements of legislative quality that the Commission so eloquently extended both to the EU and to its Member States? The 1998 Commission Staff Working Paper entitled ‘Making Single Market Rules More Effective, Quality in Implementation and Enforcement’15 clarifies the necessity for, purpose and content of quality in legislation: clear and simple legislation helps businesses and citizens to comply with the law without excessive burdens; it facilitates its enforcement; and it addresses complaints about excessive red tape which often leads to cases for damages, such as Francovich.16 All this can be achieved by use of legislation that is easy to transpose and apply and that takes into account the views of interested parties expressed in consultation under the 1998 Regulatory Policy Guidelines of

9  See Commission, ‘Better Law-Making 1995’ CSE (95) 580, Bulletin, 11-1995, point 1.9.2; see also ­Commission, ‘Better Law-Making 1996’ CSE (96) 7. 10  See Commission, ‘Simpler legislation for the Internal market (SLIM): a pilot project’ (Communication) COM (96) 204 final; see also Resolution of the EP of 10 April 1997 on the communication from the Commission to the Council and the EP ‘Simpler legislation for the internal market (SLIM): a pilot project’, COM (96)204 fin (A4-0108/97) [1997] OJ C132/213. 11  See COM (97) 618 final, 2. 12  See Bulletin EU 11-1997, point 1.1.1. 13  See COM (1998) 715 final. 14  See Commission, ‘25th annual Report from the Commission on monitoring the application of community law (2007)’ COM (2008) 777; Commission, ‘A Europe of results—applying Community law COM (2007) 502 (5 September 2007); Commission, ‘24th annual Report from the Commission on monitoring the application of community law (2006)’ COM (2007) 398 and Annex SEC (2007) 976; Commission, ‘23rd annual Report 2005 on monitoring the application of Community law’ COM (2006) 416; Commission, ‘Better monitoring the application of Community law’ (Communication) COM (2002) 725. 15  See SEC (1998) 903, 3–5. 16  Case C-6/90 Andrea Francovich and Danila Bonifaci and others v Italian Republic ECLI:EU:C:1991:428.

278 

Helen Xanthaki

the Commission and the 2002 General principles and minimum standards for consultation of interested parties by the Commission. Accountability, effectiveness and proportionality were put forward as the main elements of better law making in the 2002 Communication on European Governance: ‘Better Lawmaking’.17 Moreover, the Commission Communication to the European Council ‘Legislate Less to Act Better: the Facts’ emphasised the need to concentrate on policy priorities with strict application of the subsidiarity and proportionality principles (legislate less), the need for improved consultation procedures and the need for clearer, simpler and more accessible legislation (act better). These aims are achievable through the reduction of legislative proposals; the use of alternatives to legislation as a regulatory tool; the quality of legislative drafting through the introduction of drafting guidelines for clear, coherent and unambiguous legislation; the simplification of legislation through SLIM; the appropriate use of formal consolidation [or codification, in today’s terminology], recasting and informal consolidation; easier access to information; proper transposition; shared responsibility amongst institutions; and rationalising of national legislation. The culmination of rules for legislative quality came with the 1998 Interinstitutional Agreement on common guidelines for the quality of drafting of Community legislation:18 Community acts must be clear, simple, precise, concise and with homogeneous content. Drafting must be appropriate to the type of act concerned and to the audience to whom it is addressed. Terminology must be internally and externally consistent. The standard structure of title—preamble—enacting terms—annexes (if necessary) applies. The title offers a full indication of the subject matter. Citations set out the legal basis of the act. Recitals set out concise reasons for the chief provisions of the enacting terms without paraphrasing or reproducing them. Only clauses of a normative nature can be included in legislation. Internal and external references must be kept to a minimum. Repeals must be introduced expressly. Dates of transposition or enforcement must be introduced clearly as day/ month/year. This was supplemented by the 2003 Interinstitutional Agreement on Better Law-­Making. However, the 2003 Agreement focuses only on: improving interinstitutional co-ordination and transparency; providing a framework for alternative regulatory instruments; increasing the use of impact assessments; and the simplification of EU law.19 The 1998 Agreement remains non-binding and somewhat fragmented in its approach to legislative quality. It seems that on its own, and even in combination with the many relevant EU texts referred to here, the Agreement and therefore the EU have failed to demonstrate what the elements of legislative quality must be.

15.2.  The Post-Lisbon Smart Regulation Initiative The flurry of action for the achievement of legislative quality in the EU ended some time in 2003. Numerous policy documents continued to refer to Better Regulation, which 17 

See Commission, ‘European Governance: Better lawmaking’ (Communication) COM (2002) 0275 final. See [1999] OJ C73/1. 19  See Parliament, Council and Commission Interinstitutional Agreement on better law making [2003] OJ C321/01. 18 

Emerging Trends in Legislation in Europe

 279

c­ ontinued to be at the forefront of the EU’s governance debate, but a closer look at them shows beyond doubt the transposition from legislation as an autonomous product to legislation as a regulatory tool. This is, of course, exceptionally insightful, as legislation is indeed a tool for regulation. Unfortunately, however, in the EU this conceptual move led to a notable ignorance of legislation altogether, at least from a drafting perspective. One could source the conceptual framework behind this movement as far back as 2001 and the Mandelkern Group Report on Better Regulation. It is notable that since 2001 the Better Regulation Reports are pursuant to and therefore limited to proportionality and subsidiarity, whereas from 2007 even the title of the Better Regulation reports has been usurped by subsidiarity and proportionality, thus excluding scope for legislative quality conventions and innovative assessments. The 2008 Second strategic review of Better Regulation in the EU focuses solely on administrative burdens, legislative scrutiny, reducing the number of legislative instruments and emphasising the shared responsibility of the EU and Member States. Similarly, the 2009 Third strategic review of Better Regulation in the EU is another prime example of this move from legislation to regulatory agendas: although clear, precise and accessible legislation lies at the heart of regulatory reform, none of the recommendations address it: pre- and post-legislative scrutiny is the only focus. The most recent innovation in the field comes with the Smart Regulation Agenda.20 The October 2010 Commission Communication on Smart Regulation constitutes the formal passing from the old Better Regulation Agenda to the new Smart Regulation Agenda.21 The Commission identified three key messages in the Smart Regulation Agenda. First, Smart Regulation is about the whole policy cycle, and thus touches upon the design of a piece of legislation and its implementation, enforcement, evaluation and revision. Secondly, Smart Regulation remains a shared responsibility between the EU institutions and the Member States. Thirdly, the views of users of regulation have a key role to play in Smart Regulation, as consultation is an element of democracy. In other words, the novelty of Smart Regulation refers to three main themes: smart regulation throughout the policy cycle; shared responsibility; and stakeholder engagement.22 There is no doubt that Smart Regulation is revolutionary. The reaction of the experts to the Commission’s agenda has been positive, although the need for further action has already been identified. This includes the need to carry out impact assessments for every new regulatory proposal; to improve the informative value of roadmaps; to make the Commission’s Impact Assessment Board more independent; to conduct systematic ex post evaluations from the end users’ perspective; to strengthen the role of the High Level Group; and to consult the public.23 Smart Regulation presents obvious positive points: it follows Stefanou’s identification of the drafting process as a part of the legislative process, which is

20  See H Xanthaki, ‘The Regulatory Reform Agenda and Modern Innovations in Drafting Style’ in L Mader (ed), Regulatory Reform (Baden-Baden, Nomos, 2013) 128. 21  See Commission, ‘Smart Regulation in the European Union’ (Communication) COM (2010) 543, available at http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:52010DC0543:EN:NOT. 22  See H McColm, ‘Smart Regulation: The European Commission’s Updated Strategy’ [2011] European Journal of Risk Regulation 9, 9. 23  See ‘Common Position Paper of the Five European Independent Advisory Boards for Cutting Red Tape and Better Regulation’ (Adviescollege toetsing regeldruk (ACTAL), Nationaler Normenkontrollrat (NKR), Regelrådet, Regulatory Impact Assessment Board (RIAB) and Regulatory Policy Committee (RPC), 2012), available at www. regelradet.se/wp-content/uploads/2012/09/2012-Reaction-on-consultation-Smart-Regulation.pdf.

280 

Helen Xanthaki

a part of the policy process;24 it confirms that EU regulation is a shared responsibility of the institutions and Member States; and it affirms the need for in-depth consultation. Focus is placed on the simplification of EU law via the reduction of administrative burdens past the expected 25 per cent cuts in red tape by 2012; the evaluation of law effectiveness and efficiency ex ante via fitness checks on key areas (environment, etc) and via strategic general policy evaluations; selection of the ‘best possible’ legislation through Impact Assessment, improvement of implementation record, via post-legislative scrutiny, SOLVIT (the service of national administrations offering solutions to citizens’ problems with their EU rights)25 and EU Pilot; and achieving clearer and accessible legislation via simple language, codification, recasting and e-access. These are worthy aims—and the Commission can show considerable success. For example, the EU’s Impact Assessment system has been praised as being first class.26 The question is how these noble aims are going to be achieved. Simplification of EU law is indeed a wonderful goal. A depressing [or it is impressive?] 74 per cent of Europeans believe that the EU generates too much red tape.27 The Commission has responded to this via impact assessments and stakeholder consultation, and via regular fitness checks undertaken within the Regulatory Fitness and Performance Programme (REFIT) of December 2012. These initiatives have contributed to a reduction of red tape by well above the 25 per cent target set out in the Administrative Burden Reduction programme:28 the precise figures correspond to a decrease of 25 per cent of burden in 13 priority areas, equivalent to savings of €30.8 billion, with a further €5 billion still pending adoption by the co-legislator. But simplification cannot be taken to mean simply a streamlining of legislation and a reduction of administrative burdens. In fact, the Smart Regulation agenda fails to address a number of crucial aspects of simplification without which reduction of administrative burdens cannot be achieved. What about simplicity of the chosen policies? If a policy choice is complex in itself, then the reduction of red tape will not suffice to make it accessible to the citizens. What about simplicity in the selected regulatory means? Is the notoriously user-unfriendly legislation not a most complex regulatory tool? So reducing administrative burdens via legislation carries inherent complexity, which may well endanger the end result. What about the chosen drafting style as a means of simplifying the expression of a simply policy when a simpler regulatory tool is considered ineffective? A complex legislative style would diminish the actual effect of any simplification effort irrespective of what the percentage of administrative burdens in the simple policy chosen may be: if the users cannot understand the language of the law, how can they benefit from the opportunities created by a policy of reduced red tape? What about straightforward enforcement methods? And could one rely on a reduction of administrative burdens within the EU text if the national 24  See C Stefanou, ‘Drafters, Drafting and the Policy Process’ in C Stefanou and H Xanthaki (eds), Drafting Legislation: A Modern Approach (Aldershot, Ashgate, 2008) 321. 25 http://ec.europa.eu/solvit/what-is-solvit/index_en.htm. 26  See OECD, ‘Sustainability in Impact Assessments—A Review of Impact Assessment Systems in Selected OECD Countries and the European Commission’ (2011); European Parliament, ‘Comparative Study on the Purpose, Scope and Procedures of Impact Assessments Carried Out in the Member States of the EU’ (2011); CEPS/ University of Exeter, ‘Regulatory Quality in the European Commission and the UK: Old Questions and New Findings’ (2012). 27 See Standard Eurobarometer 79 (Spring 2013), Question QA16.4, available at http://ec.europa.eu/ public_opinion/archives/eb/eb79/eb79_anx_en.pdf. 28  See SWD (2013) 401 final.

Emerging Trends in Legislation in Europe

 281

implementing measures are complex and seize on any opportunity for discretion in the EU text in order to add further burdens? Finally, could EU law be considered simple when the methods of pre- and post-legislative scrutiny utilised are inherently complex and therefore inaccessible for the users? Similarly, the evaluation of effectiveness and efficiency of law ex ante is a fantastic initiative in theory. It is defined as a judgment of interventions according to their results and impacts, and the needs they aim to satisfy.29 But can it be achieved simply via fitness checks and general policy evaluations? In a legislative environment where the definition of effectiveness has not been provided conclusively, the goal that the regulatory team is supposed to be trying to achieve becomes a moving target, a vague and ambiguous goalpost, which carries a different meaning for the different actors in the regulatory process, and much more so between the EU and national levels. But even where the concept of effectiveness appears clear, which are the specific criteria of effectiveness by way of successful regulation that must be used in reference to a specific piece of legislation? The application of the generic elements of the semantic field of the concept of effectiveness in the specific context of a piece of legislation as applied in the specific legal system that serves a specific society in a specific time is not an easy task, and should not really be ignored or left to chance. There is a dire need to ensure that the criteria for effectiveness of any piece of legislation are agreed between policy makers, law experts and legislative drafters, and that they are clearly expressed in the legislation itself, perhaps via their inclusion in a purpose clause or an objectives article. These can then be carried through to post-legislative scrutiny and utilised to confirm effectiveness, thus allowing the text to continue its legislative life. But how will ineffectiveness be addressed? What if the effectiveness criteria are not met by the piece of legislation at the pre-set time of post-legislative monitoring: will the legislation die an automatic death via perhaps a sunset clause, will it continue to plague the statute book as it stands until an enlightened decision maker decides to address the problem or will it lead to an automatic exercise of fine-tuning via perhaps an amending piece of legislation? Smart Regulation fails to identify the way in which evaluation will take place, if at all, at the Member State level. Here, other considerations should also come into play. Will national scrutiny be compulsory or could the Member States be offered discretion on the basis of national sovereignty in the legislation-making process? Moreover, Smart Regulation fails to address the extent of any national scrutiny process, whether it must relate strictly to the national implementing measures or whether it can refer to the original EU text. This is crucial, especially in reference to national legislation, which departs from the policy and law of the EU by means of either a direct or indirect breach, or even legitimately by means of an acceptable exercise of discretion, as would be the case with the implementation of a directive. And what if the national scrutiny exercise identifies a flaw in the EU policy? Can that be reported back to the EU and will this result in any action at the EU level? The Smart Regulation initiative also fails to define efficiency. It is unclear whether that refers to a mathematical exercise involving financial cost or whether social and other impact must be calculated towards the reduction of burdens, or indeed how these can be calculated. The improvement of the implementation record is a third worthy point of reference for Smart Regulation, but once again one has to distinguish between the aim and the proposed 29 See G Luchetta, ‘Impact Assessment and the Policy Cycle in the EU’ [2012] European Journal of Risk ­Regulation 561, 572–73.

282 

Helen Xanthaki

methods for its achievement. Can implementation be improved solely via post-legislative scrutiny, the availability of SOLVIT and the EU Pilot on clarification and assistance with the application of EU legislation? What about clear guidance on the definition of complete transposition for new, older and aspiring Member States? What about clear guidance on the definition of quality in legislation for the purposes of EU drafting but also EU transposition? What about the establishment of national drafting offices with trained specialist drafters vetting (if not drafting) implementation measures on the basis of the UK model for drafting national primary legislation (the Office of Parliamentary Counsel, OPC)? What about extending the scrutiny of implementation beyond substantive transposition to technical quality of transposition? Finally, who would disagree with the election of the ‘best possible’ legislation? But is this really achievable simply via impact assessments, clearer and accessible legislation, simple language, codification, recasting and e-access? What about opening the debate for a holistic approach to effectiveness in the sense of the use of legislation as a tool for regulation? What about setting a hierarchy of goals for the drafter? What about training drafters to achieve these goals? What about considering a central drafting office with trained drafters within EU institutions, including the Commission, the Council and the Parliament? What about training national drafters to contribute to the effort?30 Thus, even post-Lisbon EU legislative quality continues to face challenges which have remained unaddressed, for the most part. The analysis of the relevant EU initiatives, namely the Better Regulation agenda, identified a pause somewhere around 2003. Despite the flurry of activity in the field of legislative quality until then, all initiatives from the EU after that date refer to legislative quality as an aim but have transferred the focus of attention to holistic regulatory quality, leaving legislative quality aside.31 The move to regulatory quality as the main focus of the EU’s better governance campaign is by no means a negative development in the field. Indeed, legislative quality is an intrinsic part of regulatory quality: without regulatory quality one cannot perceive the notion of legislative quality. As legislative drafting can only aspire to effectiveness, namely to the production of a legislative text that, with the co-operation and synergy of all other actors in the policy process, can achieve the regulatory aims, one cannot possibly expect it to produce quality legislation when the regulatory aims are erroneous or illegal, when the choice of legislation as a regulatory tool is inappropriate or, indeed, when the implementation of the legislation has not been thought through. At the same time, one cannot possibly perceive regulatory quality without, in cases where legislation is the appropriate regulatory tool, legislative quality. To me, this is the great error of the EU’s current strategy. By turning the focus of attention from legislation to regulation, the EU seems to have forgotten about legislation altogether, somehow trying to simply wish away the continuing problems of legislative quality.32 The same conclusion is reached by the analysis of Smart Regulation, the 2010 EU initiative constituting the successor of Better Regulation. This conclusion is confirmed by an analysis of the new Interinstitutional agreement on Better Law-Making 30  For a full presentation of my critique to Smart Regulation see H Xanthaki, ‘Implementation of EU Legislation’, oral evidence, public hearing of the Legal Affairs Committee of the European Parliament ‘Better Regulation’ Hearing, Brussels, 21 June 2011. 31  See M Mousmouti, ‘The Effectiveness Test’ [2012] Legisprudence 191, 196. 32  See C-H Montin, ‘Smart Regulation in the European Union’ (29 November 2012), available at http://montin. com/documents/smartregulation.pdf.

Emerging Trends in Legislation in Europe

 283

between the European Parliament, the Council of the European Union and the European Commission on Better Law-Making made on 13 April 2016. Once again, the institutions reaffirm their common goal for effectiveness of legislation, but they fail to address their drafting altogether: emphasis is on procedural issues (such as planning and monitoring), and on pre- and post-legislative scrutiny. The actual drafting is ignored, and therefore the criteria by which legislation as a product is to be assessed remains vague and therefore inapplicable in practice.33 What is even more disappointing is the 2020 Agenda for Europe, where not only Better Regulation but also Smart Regulation are ignored. The question is why the EU felt that they could move from Better Regulation to Smart Regulation and then to a strategy for growth and competitiveness. From the perspective of legislative studies at least, neither of the two moves make sense. Why would the EU decide to leave Better Regulation and Smart Regulation aside when neither of these agendas has borne the desired fruits? Surely, one could not seriously support the argument that EU legislation and national implementing legislation, or indeed the EU regulatory environment, has reached perfection? One is tempted to attribute this move to an underlying policy of competitiveness for businesses rather than a policy for better legislation in Europe. Perhaps the legislative and then regulatory quality agendas are viewed by the EU solely as means of pursuing the policy aim of growth and competitiveness. It is precisely this aim which is repeated, and expressed even more clearly, in the 2020 Agenda for Europe. In view of this, where is Europe going with reference to legislative quality and quantity? What the EU has clumsily missed here is a unique opportunity to finally balance its focus of attention between businesses and citizens. While the first enjoy the fruits of Better Regulation, Smart Regulation and the 2020 Agenda for Europe, citizens are still facing the same issues of confusion stemming from the multitude of bad EU laws (not tackled by SLIM as they did not relate to small and medium-sized enterprises) and bad regulation (not tackled by Smart Regulation since administrative burdens were not applicable). This imbalance may well be intentional, but it remains unjustifiable within the focus on citizen and citizenship expressed clearly in the Treaty of Lisbon. It also cannot be accommodated within the emphasis on social equity for EU citizens so eloquently professed in the Treaty of Lisbon, which declares the passage from the Internal Market to a forum of citizenship, international justice and peace. It appears, therefore, that the new challenge for Europe in the field of EU legislative studies is to apply Better Regulation, Smart Regulation and the 2020 Agenda for Europe to citizens, along with businesses, thus showing that the Treaty of Lisbon is not a list of good political intentions but an accurate reflection of the new EU for its citizens and peoples. The challenge is to go back to Better Regulation and Smart Regulation, and assess their success from the point of view of the citizens using the Treaty’s citizenship concept as a focus. Transferring the focus from businesses to citizens would tint the picture of the effectiveness of these three regulatory initiatives with much darker colours, because the amount of work that remains in order to make EU legislation and EU regulation palatable is daunting. But absolutely necessary.

33  See also Commission, ‘Better regulation for better results—An EU agenda’ (Communication) COM (2015) 215 final.

284 

Helen Xanthaki

Does this mean that there is only gloom and doom ahead for Europe and its legislation? Far from it. The EU as a regional organisation is lagging behind, whilst a number of ground-breaking initiatives are pursued within the Member States. At the same time, the academic discipline of legislative studies is enriching Europe with innovative concepts for better legislative quality. In what follows, I will refer to a selection of these concepts. Bluesky ideas or a realistic way forward? Only time will tell.

15.3.  Legislative Quality in the Future: Blue-sky Initiatives or a Realistic Way Forward? The first innovation in the area of drafting within the Member States is an unprecedented agreement over what constitutes legislative quality. From a legislative studies perspective, good legislation is legislation that manages to achieve the desired regulatory results.34 Since governments use legislation as a tool of successful governing,35 namely as a tool for putting into effect policies that produce the desired regulatory results,36 the qualitative measure of successful legislation coincides with the prevalent measure of policy success, which is the extent of production of the desired results.37 This is assuming that the government’s choice is indeed to put a policy to effect rather than only on paper.38 Within this context, regulation is the process of putting government policies into effect to the degree and extent intended by government.39 Legislation, as one of the many regulatory tools available to government,40 is the means by which the production of the desired regulatory results is pursued. Also, in application of Stefanou’s scheme on the policy, legislative and drafting processes,41 legislative quality is a partial but crucial contribution to regulatory quality.42 This promotes the

34  See H Xanthaki, Drafting Legislation: Art and Technology of Rules for Regulation (Oxford, Hart Publishing, 2014) ch 1. 35  See OECD, ‘Recommendation of the Council on Improving the Quality of Government Regulation’ (9 March 1995) C(95)21/Final. 36  The executive branch of government is no longer expected to confine itself to the mere making of proposals: it has to see them through. See JC Peacock, Notes on Legislative Drafting (Washington, REC Foundation, 1961) 3. 37  See N Staem, ‘Governance, Democracy and Evaluation’ (2006) 12(7) Evaluation 7, 7. 38  And the choice is the governments, not the drafters: see P Delnoy, ‘Le rôle des légistes dans la détermination du contenu des norms’, 2013 Report for the International Cooperation Group, Department of Justice, Canada, 3, available at www.justice.gc.ca/fra/apd-abt/gci-icg/publications.html. 39  See National Audit Office, Department for Business, Innovations and Skills, ‘Delivering regulatory reform’, 10 February 2011, para 1. 40  Tools for regulation vary from flexible forms of traditional regulation (such as performance-based and incentive approaches), to co-regulation and self-regulation schemes, incentive and market-based instruments (such as tax breaks and tradable permits) and information approaches. See Better Regulation Task Force, ‘Routes to Better Regulation: A Guide to Alternatives to Classic Regulation’ (December 2005); see also JC Miller, ‘The FTC and Voluntary Standards: Maximizing the Net Benefits of Self-Regulation’ (1985) 4 Cato Journal 897; OECD Report, ‘Alternatives to traditional regulation’, para 0.3; OECD, Regulatory Policies in OECD Countries: From ­Interventionism to Regulatory Governance (Paris, OECD, 2002). 41  See C Stefanou, ‘Legislative Drafting as a Form of Communication’ in L Mader and M Tavares de Almeida (eds), Quality of Legislation Principles and Instruments (Baden-Baden, Nomos, 2011) 308; see also C Stefanou, ‘Drafters, Drafting and the Policy Process’ in Stefanou and Xanthaki (n 24) 321. 42  In fact, there is an emergence of a public interest in good quality of rules: see M De Benedetto, M Martelli and N Rangone, La Qualità delle Regole (Bologna, SE il Mulino, 2011) 23.

Emerging Trends in Legislation in Europe

 285

current synergetic approach to legislation eloquently expressed by Richard Heaton, former First Parliamentary Counsel and Permanent Secretary of the Cabinet Office: I believe that we need to establish a sense of shared accountability, within and beyond government, for the quality of what (perhaps misleadingly) we call our statute book, and to promote a shared professional pride in it. In doing so, I hope we can create confidence among users that legislation is for them.43

This approach feeds into the diagram of elements of regulatory and legislative quality in Figure 1.44

Efficacy Effectiveness Cost Efficiency Clarity Precision Unambiguity Simplicity/plain language Gender-neutral language

Figure 1:  Elements of regulatory and legislative quality.

Efficacy as synonymous with regulatory quality is the extent to which regulators achieve their goal.45 Regulatory efficacy is achieved via legislative effectiveness.46 The OPC repeats its aspiration to effectiveness as a contribution to or in balance with accuracy, but does not define the term.47 Effectiveness is the ultimate measure of quality in legislation.48 If one subjects effectiveness of legislation to the wider semantic field of efficacy of regulation as its element, effectiveness manages to hold true even with reference to diverse legislative phenomena, such as symbol legislation, or even the role of law as a ritual. If the purpose of legislation is to serve as a symbol, then effectiveness becomes the measure of achieved inspiration of the users by the symbol legislation. If the legislation is to be used as a ritual, effectiveness takes the robe of persuasion of the users who bow down to its appropriate rituality. Effectiveness requires a legislative text that can (i) foresee the main projected ­outcomes 43  See R Heaton, ‘Foreword’ in Cabinet Office, Office of Parliamentary Counsel, ‘When Laws Become Too ­Complex’ (16 April 2013). 44  See H Xanthaki, ‘On Transferability of Legislative Solutions: The Functionality Test’ in Stefanou and X ­ anthaki (n 24) 1. 45  ibid 126. 46  See C Timmermans, ‘How Can One Improve the Quality of Community Legislation?’ (1997) 34 Common Market Law Review 1229, 1236–37. 47  See Office of Parliamentary Counsel, ‘Working with OPC’ (6 December 2011); OPC, ‘Drafting Guidance’ (16 December 2011). 48  See H Xanthaki, ‘On Transferability of Legal Solutions’ in Stefanou and Xanthaki (n 24) 6.

286 

Helen Xanthaki

and use them in the drafting and formulation process; (ii) state clearly its objectives and purpose; (iii) provide for necessary and appropriate means and enforcement measures; and (iv) assess and evaluate real-life effectiveness in a consistent and timely manner.49 Leaving cost efficiency out of the equation, since it is an economic–political rather than purely legal choice,50 effectiveness is promoted by clarity, precision and unambiguity. In turn, clarity, precision and unambiguity are promoted by plain language and gender-neutral language. Plain language has been promoted as the main tool for achieving clarity and, in turn, effectiveness of legislation, but its meaning has been transformed. Plain language is defined by Peter Butt as clear and effective for its audience.51 In its traditional definition, plain language is a general and inevitably vague pursuit for techniques that can produce a text that may be understood by the users in the first reading. This, in turn, enhances the clarity of the text, an attribute that makes it possible for users to adhere with the legislation if they so wish. It consequently promotes implementation, which is necessary for effectiveness. This is the crucial link between plain language and good legislation. If, however, plain language is all about facilitating implementation, does it really matter if successful communication of the legislative message takes place in the first reading? Moreover, plain language is not solely about language. Words, syntax and punctuation are very important elements—but so are the structure of the legislative text, its layout on paper and screen, and the architecture of the whole statute book as a means of facilitating awareness of the interconnections between texts. Thus, plain language begins to take effect during the analysis of the policy and the initial translation into legislation, with the selection and prioritisation of the information that readers need to receive. It continues with choices related to structure during the selection and design of the legislative solution, with simplification of the policy, simplification of the legal concepts involved in putting the policy to effect, and initial plain language choices of legislative expression (for example, a decision for direct textual amendments combined by a Keeling schedule, or a repeal and reenactment when possible). Plain language enters very much into the agenda during composition of the legislative text and remains there during the text verification, where additional confirmation of appropriate layout and visually appeal come into play. Thus, plain language extends from policy to law to drafting. Recent trends in Europe have developed plain language further. Plain language is a tool promoting uninhibited communication between the text and its users—or, to personify the communication, between the drafter and the user. The drafter is a trained lawyer with drafting training and experience. The user of the legislative text can be anything from a senior judge to an illiterate citizen of below average intellectual capacity: the inequality in the understanding of both common terms (whatever these may be) and legal terms renders communication via a single text a hopeless task. What can facilitate communication is the identification of the possible precise users of the specific legislative text: identifying who the users of the text will be allows the text to ‘speak’ to them in a language that they can generally understand. Until now, identifying the users was a hypothetical and rather 49 

This is Mousmouti’s effectiveness test: Mousmouti (n 31) 202. See R Posner, ‘Cost Benefit Analysis: Definition, Justification, and Comments on Conference Papers’ (2000) 29 The Journal of Legal Studies 1153. 51  See P Butt and R Castle, Modern Legal Drafting (New York, Cambridge University Press, 2006). 50 

Emerging Trends in Legislation in Europe

 287

academic exercise. Recent empirical data offered by a revolutionary survey of UK’s The National Archives (TNA) in co-operation with the OPC, however, have provided much needed answers.52 Speaking to the users is a noble pursuit, but one that presupposes an understanding of who uses the legislation and what level of legal awareness those users have. Ultimately, identifying the people whose choice to act or not makes government policy a success or a failure53 is crucial in establishing effective communication with them. But is there one audience of legislation? Can a drafter rely on the common notion of the ‘lay person’, the ‘average man on the street’,54 the ‘user’? The theoretical debate over this point has now been answered by the Good Law Initiative survey: at least three categories of people constitute the audience of legislation. These are: sophisticated non-lawyers using the law in the process of their professional activities; lay persons reading the legislation to make it work for them;55 and lawyers and judges. In more detail, the three categories of users of legislation are: 1. Non-lawyers who need to use legislation for work, such as law enforcers, human resources professionals, or local council officials; the ‘Mark Green’ of the survey represents about 60 per cent of users of legislation; 2. Lay persons who seek answers to questions relating to their personal or familial situation; ‘Heather Cole’ represents about 20 per cent of users of legislation; and 3. Lawyers, judges and senior law librarians; the ‘Jane Booker’ persona represents about 20 per cent of users of legislation.56 The significance of the survey for plain language and good legislation cannot be understated. The survey provides, for the first time in UK legislative practice, empirical evidence from a huge sample of the 2,000,000 visitors to www.legislation.gov.uk per month. The survey, whose data relate to users of electronic versions of the free government database of legislation only, destroys the myth that legislation is for legal professionals alone. In fact, legal professionals are very much in the minority of users, although their precise percentage may well be skewed by their tendency to use subscription databases rather than the government database, which is not annotated and often not updated. Whatever the exact percentages of each category are, there is significant empirical evidence to suggest that in the UK, and most probably in the whole of Europe, for which the UK is a representative case study, legislation speaks to three distinct groups of users, whose legal awareness varies from none to some to much. But is the legal awareness of the users the only parameter for plain language as a means of effective legislative communication?

52 

See www.gov.uk/good-law. See D Berry, ‘Audience Analysis in the Legislative Drafting Process’ [2000] The Loophole, available at www. opc.gov.au/calc/docs/calc-june/audience.htm. 54 See D Murphy, ‘Plain English—Principles and Practice’, Conference on Legislative Drafting, Canberra, ­Australia, 15 July 1992. 55  See JJE Gracia, A Theory of Textuality: The Logic and Epistemology (Albany, State University of New York Press, 1995) 159–63 and 164–65; see also GL Pi and V Schmolka, ‘A Report on Results of Usability Testing Research on Plain Language Draft Sections of the Employment Insurance Act: A Report to Department of Justice Canada and Human Resources Development Canada’ (unpublished, August 2000); V Schmolka, ‘Consumer Fireworks Regulations: Usability Testing, TR1995-2e (Department of Justice Canada, unpublished, 1995). 56  See A Bertlin, ‘What Works Best for the Reader? A Study on Drafting and Presenting Legislation’ [2014] The Loophole 27, available at www.gov.uk/government/uploads/system/uploads/attachment_data/file/326937/ Loophole_-_2014-2__2014-05-09_-What_works_best_for_the_reader.pdf. 53 

288 

Helen Xanthaki

Pitching the legislative text at the right level requires an additional consideration. ­Having determined what the rough profiles of the audience are, the next parameter for plain communication is the topic of the legislative text. Legislative texts are not all aimed at the same readers. Their primary audience varies. For example, the main users of rules of evidence are probably judges and lawyers.57 Thus, the language and terminology used can be sophisticated: paraphrasing the terms ‘intent’ or ‘mens rea’ with a plain language equivalent such as ‘meaning to’ would lead the primarily legal audience to the legitimate assumption that the legislation means something other than ‘intent’ and would not easily carry the interpretative case law of ‘intent’ on to ‘meaning to’. So rules of evidence can be drafted in specialist language, albeit with a caveat: a primarily legally sophisticated audience cannot give the drafter carte blanche for the use of legalese, since non-lawyers may need, or must have, access to the legislation too. As audiences become more specialised and more educated in technical areas, they expect texts that are targeted to their particular needs.58 Moreover, since accessibility of legislation is directly linked to Bingham’s rule of law,59 passing inaccessible legislation under the feeble excuse that its primary audience possesses legal sophistication is not easily acceptable. There are thus arguments for the continued use of legal terminology or for the provision of a definition of the new plain language equivalent referring to the legal term hitherto used. But how ‘plain’ must legislation be? Even within the ‘Heather Cole’ persona there is plenty of diversity. There is a given commonality in the lack of legal training, but the sophistication, general and legal, of Heather Coles can range from a fiercely intelligent and generally sophisticated user to a rather naïve, perhaps illiterate or even intellectually challenged individual. Which of those Heather Coles is the legislation speaking to? It certainly is not the commonly described ‘average man on the street’. To start with, there are also women on our streets, and they are users of legislation too. And then, why are the above or below averages amongst us excluded from legislative communication?60 Since effectiveness is the goal of legislative texts, should legislation not speak to each and every user who falls within the circle of subjects of the policy solution expressed by this specific legislative text? This includes a wide range of people, above average, average and below. This is a rather revolutionary innovation. Identifying the users of legislation has led to not one but two earthquakes in legislative studies: yes, the law does not speak to lawyers alone; but neither does it speak to the traditional plain-language ‘average man’. The significance of this innovation cannot be sidelined. Identifying the users has provided irrefutable empirical evidence on who uses legislation, and for what purpose. If applied in practice, this new knowledge will change the way in which legislation is drafted in Europe and beyond. First, legislative language can no longer be gauged at legal and regulatory professionals. Although great advances have already taken place, legislation now tends to be pitched to ‘Mark Green’: further simplification for the benefit of ‘Heather Cole’ needs to take place with immediate effect.

57 

See BA Garner, ‘Guidelines for Drafting and Editing Court Rules’ [1997] Federal Rules Decisions 169, 187. KA Schriver, ‘Plain Language through Protocol-Aided Revision’ in ER Steinberg (ed), Plain Language: Principles and Practice (Detroit, Wayne State University Press, 1991) 148, 152. 59  See Lord Simon of Glaisdale, ‘The Renton Report—Ten Years On’ [1985] Statute Law Review 133. 60  See J Kimble, ‘Answering the Critics of Plain Language’ (1994–95) 5 The Scribes Journal of Legal Writing 51, 59. 58  See

Emerging Trends in Legislation in Europe

 289

Dealing with language is not enough, especially when the modern holistic concept of plain language is taken into account. Scientists outside the discipline of law can and must contribute to the introduction of novel mechanisms for the production of plain and effective legislation. Having established the concept of effectiveness as synonymous with good legislation and the new holistic mandate of plain language in legislation, and armed with the new empirical data offered by the TNA and the OPC, we can now discuss further possibilities. I have identified three blue-sky mechanisms for better law. They respond to widely accepted weaknesses in European legislation stemming from the newly identified need for legislation to speak to three diverse user groups with a single text. The layered structure promotes a three-tier structure for legislative texts, each addressed to one of the three user groups; the typography-inspired presentation and layout responds to the need to bring to light the main regulatory messages in legislation; and the interactive electronic statute book highlights the interconnectivity between legislative texts within the statute book as a whole.

15.3.1.  The Layered Approach to Structure Currently legislative texts are structured by the application of Lord Thring’s Five Rules of Drafting,61 which offer precedence to provisions declaring the law versus provisions relating to the administration of the law; to simpler versus the more complex proposition; and to principal versus subordinate provisions. Exceptional and temporary provisions, and provisions relating to the repeal of acts, procedure and matters of detail, should be set apart. The application of Lord Thring’s Rules has led to a traditional legislative structure of preliminary, miscellaneous and final provisions. There is much scope for blue-sky innovation by use of the layered approach.62 The rationale behind the modern approach lies with the logical sequence of provisions within the text, which reflects logic, and philosophical and linguistic approaches to language and thought. This basis has now been overcome by the crucial evidence on the three user groups for legislation. Heather Cole, Mark Greene and Jane Booker are diverse users who require diverse pitches of the legislative text. Speaking to all three of them at the same time is a rather complex, for some impossible, task. Introducing three versions of the same legislative text is a possibility, but it is a recipe for disaster on a diverse range of grounds, moral, ethical, constitutional and practical: rule of law, issues of interpretation between versions, identifying which version corresponds to which user, using that version as opposed to the one selected by the user, who identifies each user with their corresponding persona, and ethical and moral consequences of the application of a diverse version for each user. The parallel existence of three different texts could also be counter-productive: users currently choose to use the complex but official legislative text over any of the many interpretation aids offered by government. If the plethora of attractive user-friendly manuals and policy documents are shunned in favour of legislative texts, what

61  See Lord Thring, Practical Legislation, The Composition and Language of Acts of Parliament and Business Documents (London, 1902), 38; see also VCRAC Crabbe, Legislative Drafting (Oxford, Cavendish Publishing, 1998) 148–50. 62  The term and, to a certain extent, the concept are attributed to John Witing, Tax Director at the Tax Simplification Office. I am very grateful to John for his inspiration and the generosity with which he has shared it with me.

290 

Helen Xanthaki

makes it probable that users will go to the simple Heather Cole text as opposed to the legal Jane Booker one, which reflects users’ perception of legislation? Thus, remaining with a single text is the only real option. But this is exactly what has imprisoned legislative drafters in the struggle for simplicity within legislative texts. It is now possible to see that each user group has its own requirements for legislative information that are distinct from those of the other user groups. Identifying the needs for legislative information for each user group at a provision, rather than text, level would allow drafters to imitate oral communication, and to pitch the legislative text to specific abilities and requirements. Drafters of legislative texts can now begin to think about what regulatory or legal message is relevant to each group, and structure the text accordingly. The layered approach promotes the division of legislation into three parts, corresponding to each of the three profiles of legislative users. Part 1 can speak to lay persons: the content is limited to the main regulatory messages, thus conveying the essence of law reform attempted by the legislation, focusing gravely on the information that lay persons need in order to become aware of a new regulation, to comply with new obligations or to enjoy new rights. Part 2 can speak to non-legally trained professionals who use the legislation in the course of their employment. Here one can see scope for further detail in the regulatory messages introduced, and for language that is balanced (technical, yet approachable to the professionals in question). Part 3 of the legislation can then deal with issues of legislative interpretation, issues of procedure and issues of application, in a language that is complex but not quite legalese, so there is nothing to prevent all groups from reading all parts. The layered approach is revolutionary, as it shifts the criterion for legislative structure from the content and nature of provisions to the profile of the users. It switches to a usercentred structure, thus promoting a link between policy and its effecting legislative text, but also enhancing and personalising the channel of communication between drafters and users. It further applies and reflects the modern doctrine of contextualism in language and philosophy. However, it cannot be viewed as a complete departure from tradition, as it continues to apply Lord Thring’s Five Rules. By requiring that Part 1 includes the primary regulatory message, it promotes Lord Thring’s Rules that give precedence to the simpler proposition. And by structuring legislation into three parts, the layered approach complies with the other of Lord Thring’s Rules that require: division of provisions declaring the law (in Part 1 or 2) with provisions administrating the law (in Part 2 or 3 accordingly); that principal provisions be separated from subordinate (in Parts 1 and 2); that exceptional, temporary and provisions relating to the repeal of Acts be separated from the other enactments, and placed by themselves under separate headings (in Part 3); and that procedure and matters of detail be set apart by themselves (either in Part 3 of the layered approach, or in a Schedule). The layered approach seems to be one of the most promising initiatives in the field of legislation. However, three points need to be clarified. First, the layered approach may, though will not necessarily, lead to a partial, fragmented or incomplete legislative communication to Heather Cole. There is no doubt that an erroneous application of the approach could result to that. But the placement of the main messages in Part 1 per se must be seen as an added bonus to lay users compared with the current state of affairs: in the layered approach the now frequently elusive main regulatory message will be easily identified, will be brought forward in a pronounced place at the beginning of the legislative text and will be expressed in a language that is accessible to lay users. Compared to the current state of affairs, where

Emerging Trends in Legislation in Europe

 291

the main message is communicated somewhere within the legislative text and is expressed in the layered approach’s Part 2 or 3 language, this is certainly an improvement. And of course, there is nothing preventing Heather Cole from reading the rest of the text: in fact, an inviting Part 1 can only encourage Heather Cole to keep reading, whilst offering her a clear context within which her understanding of complex and detailed messages can only be enhanced. Secondly, although Part 1, carrying the main regulatory message, is distinctly different from Parts 2 and 3, it may be unclear what really distinguishes between Part 2 data and Part 3 data: both Mark Green and Jane Booker are able to handle complexity and technicality of legislative data. However, they do not both require the same data, as demonstrated by their motives when using www.legislation.gov.uk: Mark Green is interested in answers that allow him to perform his professional but non-legal duties, whereas Jane Booker seeks legal information. As a result, what Mark Green needs is a clear understanding of substantive and procedural requirements imposed by the legislation, whereas Jane Booker seeks deeper statutory interpretation, often coupled with a holistic view of the statute book. As a result, Part 2 of the layered approach involves answers to questions such as who must do what by when, and what happens if they don’t. Part 3 will delve deeper into intricate distinctions and possible exceptions that relate to statutory interpretation and interconnections between legislative texts within the statute book. There are two caveats here: (i) Mark Green must still read the text as a whole; and (ii) Part 3 cannot be viewed as a mere shell of definitions, repeals and consequential amendments, as this would deprive the readers from at least part of the benefits of the layered approach. Thirdly, it would be inappropriate to consider that the simplification serviced by the layered approach would result in an abolition of the need for explanatory materials for legislation. In fact, as the layered approach results in an inherent fragmentation of data, it renders the use of explanatory materials and notes reinstating the fluidity of information and the cross-fertilisation between parts a crucial requirement. Ultimately, the proof of the layered approach is in its application. User testing can prove whether it works, which user group it works for, and how it can be amended or fine-tuned to serve users better.

15.3.2.  Legislative Image: Presentation, Layout, Pictures Looking now at the image of the legislative text, namely at the picture that the user receives when looking at the text, it is necessary to distinguish between paper and electronic. It is noteworthy that in New Zealand legislation is only published electronically: paper publication ceased in 2015. In Europe, no government has yet declared an intent to abolish paper publication, but there is certainly a move to a more user-friendly electronic version of legislation. Plain language has always advocated the need to rethink the layout of legislative texts.63 The single font, the lack of adequate contrast between paper and text, and the unique ­format are elements of the current legislative image that prevent the user from identifying

63  See Office of Scottish Parliamentary Counsel, ‘Plain language and legislation’ (February 2006), available at www.gov.scot/resource/doc/93488/0022476.pdf.

292 

Helen Xanthaki

the important aspects of the regulatory message, thus reducing the readability of legislative texts. Legislative texts attempt to convey a ‘legislative story’ to the user, thus allowing them to identify and then understand the underlying policy, the legislative choices made and the rationale behind the text. This offers them the ability to read and interpret the text in context, thus making accessibility easier and more secure. However, there is plenty of scope for further progress. Layout is now at the forefront of practitioners’ agenda. And quite rightly so—it has been overlooked and there is great scope for change. However, layout alone cannot respond to a complex text, a complex regulatory message or, indeed, a complex policy. It can contribute to simplification, but with the aid of additional visual tools. One of those tools that have been ignored by even the most visionary of legislative academics and practitioners is the use of image in legislation. Images have been used in legislation that introduces national flags, traffic signs or planning regulations. However, the relationship between picture and legislation has not been explored fully. The visual arts could play a significant role here: there is nothing more direct, relevant to a wide range of users and time resistant than Cain swinging his club above the prostrate Abel in Titian’s painting Cain and Abel in Santa Maria della Salute in Venice. The visual representations of themes relating to wrongdoing are so emotionally charged and the characters shown in such magnification that, combined with beauty and other aesthetic values, the picture has a tremendous impact on the viewer. Perhaps the inclusion of images in legislation can enhance the quality of communication. An example could be drawn from criminal provisions. The picture accompanying the legislation in the form of a Schedule may show: —— what behaviour is to be condemned (show the action and specify whether the person knows that it is bad, suspects that it is bad or is ignorant of the badness of the behaviour); —— that this is an offence (for example, show a stop sign or show societal disapproval); and —— that it carries a sanction (for example, show the penalty and its adverse effect). The use of typographical and visual aids in legislation can enhance readability64 immensely. Such aids can address textual limitations and can take the user further by banishing the barriers of written textual communication. User testing is the only way to assess if and how useful they are. But academic research—indeed, inter-disciplinary academic research— is the only forum for analysis at a theoretical level first, then in application to actual legislation.

15.3.3. The Statute Book or the Body of Legislation as a Whole: Quantity and Beyond Reforming the structure and layout of individual legislative texts may bear little fruit without changes in the statute book as a whole. Addressing the issue of legislative volume that 64  See G Jones, P Rice, J Sherwood and J Whiting, ‘Developing a Tax Complexity Index for the UK’, available at www.gov.uk/government/uploads/system/uploads/attachment_data/file/285944/OTS_Developing_a_Tax_ Complexity_Index_for_the_UK.pdf.

Emerging Trends in Legislation in Europe

 293

enhances complexity65 has been at the forefront of the agendas of the last two governments as the epicentre of regulatory quality. The volume of legislation came under review in the UK in 2003. The Better Regulation Task Force’s ‘Principles of Good Regulation’66 linked better regulation with less legislation, and offered a number of regulatory alternatives: do nothing; advertising campaigns and education; using the market; financial incentives; selfregulation and voluntary codes of practice; and prescriptive regulation. In ‘The Coalition: Our Programme for Government’,67 the coalition government undertook to: cut red tape68 by introducing a ‘one-in, one-out’ rule whereby no new regulation can be brought in without another regulation being cut by a greater amount;69 to impose sunset clauses on regulations; and to give the public the opportunity to challenge the worst regulations. Such was the importance attributed to legislative volume that the Prime Minister, in his letter of 6 April 2011 to all Cabinet Ministers, declared: ‘I want us to be the first Government in modern history to leave office having reduced the overall burden of regulation, rather than increasing it.’ In order to achieve this aim, the UK government went one step further and introduced a one-in, two-out approach. It undertook to use regulation for the achievement of its policy objectives only where: non-regulatory approaches cannot lead to satisfactory outcomes; cost–benefit analysis demonstrates a clear margin of superiority of regulation to alternative, self-regulatory or non-regulatory approaches; or the regulation and the enforcement framework can be implemented in a fashion which is demonstrably proportionate, accountable, consistent, transparent and targeted.70 The number of acts passed in 2012 was only 20, with a total number of pages of 1,886;71 this was a new low, after the peak of the late 1990s and early 2000s. But, whilst the number of acts has decreased since the 1980s, the mean average number of pages per act has increased significantly, from 37 and 47 pages during the 1980s and 1990s, respectively, to 85 in the past decade; if one compares these numbers with the 1950s, when the average was 16, the trend for fewer but longer acts becomes evident.72 One could contribute this increase to plain-language drafting and to the increasing amounts of white space and bigger margins, leading to 20 per cent fewer words on a page.73 However, there is a crucial contributing factor: over the last 30–40 years, the number of Statutory Instruments has steadily increased.74 Thus, the volume of legislation, 65  See Office of Parliamentary Counsel, ‘When Laws Become Too Complex: A review into the causes of complex legislation’ (London, Cabinet Office, 2013) 6–7, available at www.gov.uk/government/uploads/system/uploads/ attachment_data/file/187015/GoodLaw_report_8April_AP.pdf. 66  See http://webarchive.nationalarchives.gov.uk/20100407162704/http:/archive.cabinetoffice.gov.uk/brc/upload/ assets/www.brc.gov.uk/principlesleaflet.pdf. 67  See ‘The Coalition: our programme for government’, available at www.gov.uk/government/uploads/system/ uploads/attachment_data/file/78977/coalition_programme_for_government.pdf. 68  For further information on the Red Tape Challenge, see www.redtapechallenge.cabinetoffice.gov.uk/home/ index. 69  See www.bis.gov.uk/assets/biscore/better-regulation/docs/o/11–671-one-in-one-out-methodology. 70  See Department for Business, Innovation and Skills, ‘Better Regulation Framework Manual’ (July 2013) 4, available at www.gov.uk/government/uploads/system/uploads/attachment_data/file/211981/bis-13-1038-betterregulation-framework-manual-guidance-for-officials.pdf. 71  See HoL Library Note 2013/008, Volume of Legislation, 4. 72  See HoL Library Note, Volume of Legislation, LLN 2011/028, September 2011. 73  See R Heaton, House of Commons Political and Constitutional Reform Committee, ‘Ensuring standards in the quality of legislation’, First Report of Session 2013–14, HC 85 incorporating HC 74-i to vii, Session 2012–13, 20 May 2013, Question 64. 74  See R Cracknell and R Clements ‘Acts and Statutory Instruments: the volume of UK legislation 1950 to 2012’ HoC Standard Note SN/SG/2911 (15 November 2012) 2.

294 

Helen Xanthaki

including primary and delegated, seems to be holding ground in practice.75 Nonetheless, a recent OECD Review pronounces the regulatory reforms in the UK as impressive.76 Points of excellence include the effective balance between policy breadth, and the stock and flow of regulation, and the extensive application of EU’s Better Regulation initiatives in the UK.77 But of course, innovations to the statute book do not end with legislative volume. There are also blue-sky proposals, which in this case may be put into effect much quicker than one might expect.

15.3.4.  The Theoretical Umbrella: Phronetic Legislative Drafting Legislative studies and legislative practice are rapidly progressing to the age of maturity via a number of innovations. These are not only found in Europe; in fact, legislative innovation is happening all over the world. This rampage of fresh and innovative thinking is not haphazard: it reflects, and is evidence of, academic innovation in legislative studies theory. Until recently, legislative drafting was viewed as a mere skill, normally and mostly served by government lawyers. But things have changed. Legislation has become the focus of regulation, replacing the common law. There are a number of possible causes for this phenomenon: the Europeanisation of law offered common law systems the opportunity to appreciate more the feared statutory law; legal globalisation led to an emphasis on international statutory law (treaties, etc) that required national implementation via national statutory law; and finally the realisation that regulation was passed for the purposes of achieving measurable results led to the inevitable (and not always fortunate) use of statutory law as a method of regulation. Whatever the reason, it invited a detailed study of statutory law from its conceptualisation to its implementation, and it paved the way for a new theory for legislative drafting.78 The traditional view, mostly within the common law world, is that drafting is a pure form of art79 or a quasi-craft:80 if drafting is an art or a craft, then creativity and innovation lies at the core of the task; rules and conventions bear relative value. In the civil law world, drafting is viewed as science81 or technique:82 it carries formal rules and conventions whose inherent nomoteleia manages to produce predictable results. But, if drafting is viewed as a sub-discipline of law, then there is a third option: law is not part of the arts, nor is it part of the sciences83 in the positivist sense.84 In science, rules apply with universality and 75  And not just in the UK: see R Pagano Introduzione alla legistica—L’arte di preparare le leggi (Milan, Giuffre, 1999) 6. 76  See www.oecd.org/dataoecd/61/60/44912018.pdf. 77 For a listing of such policies and their implementation in the UK, see www.bis.gov.uk/policies/bre/ improving-eu-regulation/guiding-principles-eu-legislation. 78  See H Xanthaki, ‘Duncan Berry: A True Visionary of Training in Legislative Drafting’ [2011] The Loophole 18. 79  See BG Scharffs, ‘Law as Craft’ (2001) 45 Vanderbilt Law Review 2339. 80  See C Nutting, ‘Legislative Drafting: A Review’ (1955) 41 American Bar Association Journal 76. 81  In opposition see Editorial Review (1903) 22 Canadian Law Times 437. 82  In opposition see J-C Piris, ‘The Legal Orders of the European Union and of the Member States: Peculiarities and Influences in Drafting’ [2006] European Journal of Law Reform 1. 83  For an analysis of the counter-argument on law as a science, see M Speziale, ‘Langdell’s Concept of Law as Science: The Beginning of Anti-Formalism in American Legal Theory’ 5 [1980] Vanderbilt Law Review 1. 84  See RR Formoy, ‘Special Drafting’ 21 [1938] Bell Yard: Journal of the Law Society’s School of Law 3; but against see C Langdell, ‘Harvard Celebration Speeches’ (1887) 3 Law Quarterly Review 123.

Emerging Trends in Legislation in Europe

 295

infallibility: gravity will always make an object fall down. Law is different: ‘All law is universal but about some things it is not possible to make a universal statement which will be correct … the error is not in the law nor in the legislator but in the nature of the thing.’85 But rejecting the view that drafting is a science does not necessarily confirm that drafting is an art. Art tends to lack any sense of rules. In the pursuit of aesthetic pleasure, art uses whatever tools are available. Art is anarchic. Drafting is not. Its rules may not be rigid, but they are present. There may be exceptions to all rules of drafting, but this does not mean that there are no rules. And these rules carry with them a degree of relevant predictability, since the latter is one of the six elements of theory.86 For Aristotle,87 all human intellectuality can be classified as88 science as episteme; art as techne; or phronesis89 as the praxis of subjective decision making on factual circumstances or the practical wisdom of the subjective classification of factual circumstances to principals and wisdom as episteme.90 Law and drafting seem to be classical examples of phronesis, as they are liberal disciplines with loose but prevalent rules and conventions whose correct application comes through knowledge and experience. Drafting as phronesis is ‘akin to practical wisdom that comes from an intimate familiarity with contingencies and uncertainties of various forms of social practice embedded in complex social settings’.91 The art of drafting lies with the subjective use and application of its science, with the conscious subjective Aristotelian application and implementation of its universal theoretical principles to the concrete circumstances of the problem.92 Phronesis supports the selection of solutions made on the basis of informed yet subjective application of principles on set circumstances.93 Phronesis is ‘practical wisdom that responds to nuance and a sense of the concrete, outstripping abstract or general theories of what is right. In this way, practical wisdom relies on a kind of immediate insight, rather than more formal inferential processes.’94 Thus, drafting legislation simply involves the choice of the appropriate rule or convention that delivers the desired results within the unique circumstances of the specific problem at any given time. Under this functional prism, successful drafting is the production of a good law, namely an effective law that contributes to regulatory efficacy.95 There is nothing technical with qualitative functionality here: what counts is the ability of the law to achieve the reforms requested by the policy officers. In view of the myriad parameters that are unique in each dossier, there are no precise elements of quality at this level.

85 

See Aristotle, EN, 5.10.1137b13–24. B Flyvbjerg, Making Social Science Matter: Why Social Inquiry Fails and How It Can Succeed Again (­Cambridge, Cambridge University Press, 2001) 39. 87  See Aristotle, EN, bk VI, chs 5–11 (D Ross trans, 1980). 88  See M Griffiths and G Macleod, ‘Personal Narratives and Policy: Never the Twain?’ (2008) 42 Journal of ­Philosophy of Education 121, 126. 89  See Aristotle (n 85). 90  See S-U von Kirchmann, Die Werlosigkeit der Jursprudenz als Wissenschaft (Berlin, Verlage von Julius Springer, 1848). 91  See B Caterino and SF Schram, ‘Introduction’ in SF Schram and B Caterino, Making Political Science Matter: Debating Knowledge, Research, and Method (New York, New York University Press, 2006) 8. 92  See W Eskridge Jr, ‘Gadamer/Statutory Interpretation’ (1990) 90 Columbia Law Review 635. 93  See E Engle, ‘Aristotle, Law and Justice: the tragic hero’ (2008) 35 Northern Kentucky Law Review 4. 94  See C Rideout, ‘Storytelling, Narrative Rationality, and Legal Persuasion’ (2008) 14 Legal Writing: Journal of the Legal Writing Institute 75. 95  See H Xanthaki, ‘Drafting Manuals and Quality in Legislation: Positive Contribution towards Certainty in the Law or Impediment to the Necessity for Dynamism of Rules?’ (2010) 4 Legisprudence 111. 86  See

296 

Helen Xanthaki

This qualitative definition of quality in legislation respects and embraces the subjectivity and flexibility of phronetic legislative drafting.96 Phronetic legislative drafting does not ignore the elements of art and science identified within the discipline; it focuses on the subjectivity of prioritisation in the selection of the most appropriate virtue to be applied by the drafter in cases of a clash between equal virtues. But subjectivity is not anarchic: it is qualified by means of recognising effectiveness as the sole overriding criterion for that choice. In phronetic legislative drafting, one must be able to identify basic principles which, as a rule, can render a law good. The pyramid at the beginning of this paper presents such principles: when applied, at least in the majority of cases, they lead to good law. Yet the ultimate criterion of good law is its effectiveness, at least under the prism of phronetic legislative theory, a theory that has innovated legislative study and legislative practice in Europe and beyond.

15.4. Conclusions Perhaps the biggest innovation in legislation and legislative studies is the realisation that the partnership between legislative professionals and legislative academics provides a dynamic combination of appropriate research methodology and internally available governmentheld empirical legislative data: when the two gel, they can produce academically valid and practically useable know-how whose empirical impact can change our whole perception of legislation and the statute book. Challenging as it is, the new research agenda offers academics the comfort of a sound theoretical framework within which any co-operation can flourish: phronetic legislative drafting views the study of legislation as a new sub-discipline of legal science, thus allowing it to benefit from the wealth of theoretical and empirical analyses in substantive fields of law that can serve as persuasive case studies for the further development of both the substantive law and the legislative fields of study. Blue skies await ahead, and Europe has a leading role to play both via its academics and via its practitioners. Perhaps this book constitutes the first step to the right direction.

96  See H Xanthaki, ‘Quality of Legislation: An Achievable Universal Concept or a Utopian Pursuit?’ in Mader and Almeida (n 41) 75–85.

INDEX

2020 Agenda for Europe  283 accessibility to law  110, 242–3 Agreement on Better Law Making 2003  251 Agreement on Better Law Making 2016  253 AHRC Project on Big Data in Legislation  126 amendments  14, 122–3, 253 analysis of the options  204, 206–8 example of  205–6 Anderson, T.  200 annexes  121 Arndt, C.  194 Bacon, F.  118 Baldwin, R.  30 Balthasar, A.  62 Bentham, J.  112, 135, 165 Bergeal, C.  120 ‘better legislation’  3, 4 better law making policies  29, 30, 31 assessing  31–2 EU Better Law Making initiative  30 ‘better regulation’  4 better regulation policies  29, 30, 31 assessing  31–2 EU Better Regulation initiative  30 Biocides  238 Black, J.  27 Braibant, G.  114 Butt, P.  286 Cambacerès, J.-J.-R. de  113 categories of laws  14 centralisation  60, 155–6 Chesterton, G. K.  3 CILFIT  242, 250 CJEU see  Court of Justice of the European Union Coase, R.  22 codification  113–14, 180, 181, 254 ‘codification with recast’  181 ‘corrective codification’  181 EU level  183–4 horizontal codification  184 national level  181 ‘pure codification’  181 vertical codification  184 Commonwealth Association of Legislative Counsel  259 comparative legisprudence  4–5

compilation  maintenance  224 composing drafting process  149–50 conceptualisation drafting process  147 conferral, principle of  231, 232 consolidation  114, 180, 224–5, 254 EU level  184 national level  181–3 constitutional amendment, right to initiate  43 ‘constitutional family’  15 consultation see participation Corkin, J.  27 corrigenda  253–4 Court of Justice of the European Union (CJEU) actions for annulment  249 infringement proceedings  249 interpretation of EU Law  249–50 references for preliminary rulings  249 role of  248 cross-references  121–2 ‘crowd-sourced law making’  90 Dale, W.  269 De Winter, L.  62 decentralisation  6, 60, 153–5 decisions EU legislation  235 delegated acts  237, 238 delimiting definitions  134 deposit for public inspection, documents placed on  177 Dickinson, J.  16 direct effect EU law  233 directives EU legislation  234–5 drafting process  145–6, 152, 161–2 multilingual legal systems  159, 161 co-drafting model  159–60 co-editing model  161 co-revision model  160 mixed-language composition  160 organisational models  153, 158 centralisation  155–6 decentralisation  153–5

298 

Index

examples  156–8 Switzerland  157–8 United Kingdom  156–7 stages of  146–7 composing  149–50 conceptualisation  147 editing  152 planning  147–9 reviewing and revising  150–52 see also EU legislation drafting techniques  13, 14, 109–10 accessibility to law  110 amendments  14, 122–3 annexes  121 cross-references  121–2 distinguishing divisions and subdivisions  119–21 headings  120–21 history of  111–13 innovation  125, 126–7 AHRC Project on Big Data in Legislation  126 Good Law project  125–6 TNA initiatives  126 legal certainty  111 ‘legislative cycle’  111 numbering  119–20 preambles  118–19 schedules  121 structure and systematics  13–14, 113 codification  113–14 consolidation  114 internal and external structure  114–15 preliminary provisions  115 repealing  114 substantive provisions  115 title of legislation  116 short title  116–18 transitional provisions  121 transposition of EU legislation  123–4 see also EU legislation Dunlop, C.  191 economics of regulation  19–20 normative theories of regulation  20–22 positive theories of regulation  22 editing  drafting process  152 electronic publication  170–74 EU legislation  229 accessibility  242–3 basic principles  conferral  231, 232 direct effect  233 primacy of EU law  233 proportionality  232 sincere cooperation  232 subsidiarity  232 condensing and updating amendments  253 codification  254 consolidation  254 corrigenda  253–4

recasting  254 repealing  255 Court of Justice of the European Union actions for annulment  249 infringement proceedings  249 interpretation of EU Law  249–50 references for preliminary rulings  249 role of  248 drafting drafting process  243–4 drafting rules  244–5 drafting style  247–8 scrutiny and revision  245–6 structure  246–7 implementing powers  237, 238 institutional framework of the EU  230–31 interinstitutional agreements  236 languages  241–2 legal order, EU as  231, 255 legislative procedure codecision procedure  238 Commission proposal  239–40 Commission withdrawal or amendment of proposal  241 Council of the European Union, role of  240–41 European Parliament, role of  240 ‘informal trilogues’  241 ordinary legislative procedure  238, 239 time limits  241 origins of  229–30 publication  242, 243 quality of EU legislation Agreement on Better Law Making 2003  251 Agreement on Better Law Making 2016  253 Better Regulation Package 2015  253 calls for improvement  250–51 governance initiative  251 REFIT  252 Smart Regulation  252 transposition of EU legislation  123–4 types of legal acts  233 decisions  235 delegated acts  237, 238 directives  234–5 implementing acts  237–8 legislative acts  236 non-legislative acts  236 opinions  236 recommendations  235–6 regulations  233–4 evaluation of legislation  11–12 conformity with the constitution and law  12 effectiveness  12 RIA  12–13 trust  12 see also legislative goals; legislative quality; regulatory policy evaluation evidence-based legislation  49–50, 54 extending definitions  134 extraordinary publication  177–8

Index Filangieri, G.  223 formal legistics see drafting techniques ‘formal publication’  167 France public policy evaluation  218–19 Francovich  277 Fukuyama, F.  188, 201 Geny, F.  113 Gilmour, R.  27 Good Law project  125–6 government functions of  8 Grabosky, P.  198 Gunningham, N.  198 Hallstein, W.  231 Hamilton, J.  200 Härtel, J.  6 headings  120–21 Heaton, R.  285 hidden goals  101–2 Hood, C.  18 horizontal separation of powers  5–6 Huber, E.  113 IAL  259 implementing acts  237–8 instruments of legislation  11 interinstitutional agreements  236, 276, 278 International Association of Legislation (IAL)  259 Jacob, S.  62 juridification of legislation  8, 16 Kant, I.  17 Karpen, U.  265 Knooble v State and NNI  176 language  14 clarity and precision  130–32 conciseness  132 consistency, need for 130 gender-neutral language  131–2 importance of  129 multilingual legislation  139–40 plain language  130, 136, 141, 142 purpose of  129 requirements of  130 technical or specialised terms, use of  135–6 legal lexicon  136 non-legal lexicon  136 see also style of legislation law role of  2 Leal, D.  200 legal certainty  111 legal definitions  14, 133–4 delimiting definitions  134 extending definitions  134 functions of  134

 299

lexical definitions  133 rules of  134–5 stipulative definitions  133 legal gazettes  168, 169, 170 legislation comparing legislation and regulation  19 concept of  17, 18 functions of  18, 23, 24 importance of  19 requirements and demands on  24–6 rule of law  24–5 ‘legislation state’  18 legislative acts  236 ‘legislative cycle’  111 legislative goals  97, 107 distinction between legislative goals and legislative effects  98–9 function of the legislation  102–3 intended impact  104–5 time  103–4 hidden goals  101–2 location  105–6 macro-functional goals  104 meso-functional goals  104 micro-functional goals  104 patent goals  101, 102 structure of  99, 102 perspective  100–101 positioning  99–100 visibility  101–2 legislative guidelines  15 legislative initiative  43 legislative process  33–4 actors involved in the process  41–4 ‘creeping loss’ of parliamentary power  49 evidence-based legislation  49–50 general principles  35–6 governments legislative powers  37–8, 39 ICT, use of  50 necessity of political will  49 Parliaments  involvement in legislative processes of EU and international organisations  37 legislative competences 36, 38–9 need for improvements to parliamentary process  50–51 transparency, need for  37 political and ordinary decision-making process  35 project management  50 RIA  40, 41 rules of drafting, consultation and impact assessment  36, 39–41 stages of  44 constitutional review  48 parliamentary stage  46–7 post-parliamentary, stage  47–8 preparatory stage  44–5 publication  48 sub-state legislative powers  36 see also EU legislation

300  legislative quality  4, 25–6, 31, 32 2020 Agenda for Europe  283 agreement over what constitutes legislative quality  284 citizens of the EU, and  283 drafting as art / science / phronesis  294–6 elements of regulatory and legislative quality  285–9 clarity, precision, and unambiguity  286 effectiveness  285–6 efficacy  285 language and terminology used  288 plain language  286, 287, 288, 289 users of the text  286–7, 288 EU’s regulatory framework  275–8 Better Law-Making Reports  277 Interinstitutional Agreements  276, 278 Resolution on legislative quality  276 SLIM Initiative  277, 278 Sutherland Report  275 layered approach to structure  289–91 partnership between legislative professionals and legislative academics  296 presentation and layout of legislative texts  291–2 regulatory quality and legislative quality  282 Smart Regulation  279–82 evaluation of law effectiveness and efficiency ex ante  281 improvement of implementation record  281–2 main themes of  279 selection of the ‘best possible’ legislation  282 simplification of EU law  280–81 volume of legislation  292–4 see also EU legislation legisprudence aim of  1 limitations  16 practical science, as  3 role of  3 theoretical science, as  3 lexical definitions  133 Livermore, M.  206 macro-functional goals  104 Mader, L.  112, 265 maintenance of rules  215 consultation  226 history of  220–21 increased relevance of maintenance  221, 222 individual rules  216 kinds of maintenance  224 compilation  224 consolidation  224–5 ordering regulation  224 quantitative effects on the stock of legislation  224 reform  225 revision  225 maintenance strategy, need to adopt  227

Index narrow maintenance  215, 216 purpose and object of  216–20, 221, 226 quality of rules  222 formal quality  222–3 substantial quality  223 wide maintenance  215, 216, 226 see also regulatory policy evaluation Majone, G.  19, 26 Malleson, K.  27 meso-functional goals  104 Meuwese, A.  208 Micklethwait, J.  188 micro-functional goals  104 multi-level systems  6–7 multilingual legal systems  159, 161  co-drafting model  159–60 co-editing model  161 co-revision model  160 mixed-language composition  160 narrow maintenance  215, 216 Noll, P.  112 non-legislative acts  236 numbering  119–20 official journals  168, 169 Ogus, A.  23 opinions EU legislation  236 paper publication  167–70 parliament responsibility for legislation  7–8, 15–16 see also legislative process participation  10, 73 election periods, during  85 features of  73–6 forms of  82–3 duration of the consultation period  90–91 feedback and use of results  91–2 format and methods  88–90 multi-stage consultations  83–4 objects of consultation  85–7 parliamentary procedures, and  92–3 range of participants  87–8 responsibility for carrying out consultation procedure  85 functions of fact finding  79–80 implementation and practicability  80–81 informing the public  76 legislative quality  81 legitimacy, enhancing  77–8 political feasibility  78–9 transparency, enhancing  76–7 limitations of  82 regulation and practice  93 guidelines  94–5 national constitutions and laws  93–4 Patchett, K.  269 patent goals  101, 102 PDCA cycle  64

Index Peck, S.  2 planning drafting process  147–9 policy setting  10–11 Portalis, J.  113 preambles  118–19 preliminary provisions  115 primacy of EU law  233 proportionality principle  232 publication codification  180, 181 EU level  183–4 national level  181 consolidation  180 EU level  184 national level  181–3 defects and corrections  179–80 EU legislation  242, 243 legal effect  178–9 legal gazettes  168, 169, 170 legislation by reference  174–7 methods  167 documents placed on deposit for public inspection  177 electronic publication  170–74 extraordinary publication  177–8 ‘formal publication’  167 paper publication  167–70 ‘substantial publication’  167 official journals  168, 169 publication date  169–70, 173, 178, 179 requirement for  165–6, 184 Radaelli, C.  191, 263, 264 recasting  254 recommendations EU legislation  235–6 referenda  8, 43, 44 REFIT  219, 252 reform  maintenance of rules  225 regulation comparing legislation and regulation  19 concept of  17, 18 economics of regulation  19–20 normative theories of regulation  20–22 positive theories of regulation  22 functions of  23, 28 importance of  19, 28 problem of excessive regulation  28, 110 regulations type of EU legislation  233–4 regulatory competition  28–9 regulatory cycle elements of  9 Regulatory Fitness (REFIT) and performance programme  219, 252 regulatory impact assessment (RIA)  12–13, 28–9, 40, 41 analysis of the options  204–8

 301

baseline analysis  196–7 characteristics of  191 consultation  202–3 data collection  201–2 ex ante and ex post evaluation  188–9 formulation of options  197–201 OECD Recommendation  191–3 problem definition  194–5 quality assurance  209 reporting and communication  208–9 setting criteria for the design of  195–6 use of  189–90 see also regulatory policy evaluation regulatory management  53, 56, 71–2 competence management  66–7 culture  56–9 consultation practices  58–9 institutionalisation  59 leadership commitment  56, 57 legal culture  58 designing a regulatory policy  54 EU Member States  68–9 multi-tiered Member States  69–71 evidence-based legislation  54 instruments  65–6 parliament, role of  56 political support, need for  56 process management  64–5 quality standards  55 structures  59–64 centralisation / decentralisation  60 institutionalisation of evaluation  62–4 specialised bodies  61–2 regulatory policy evaluation  188 effectiveness of  211 ex ante and ex post evaluation  188–9, 190–91, 194 high-quality regulation concept of  187 OECD Framework for Regulatory Policy Evaluation  209–10 guidance  191, 194 regulatory oversight bodies  209 see also regulatory impact assessment regulatory pyramid  201 regulatory quality  26, 31, 32 ‘regulatory state’  19, 26 consequences of  26–7 repealing  114, 255 Revesz, R.  206 reviewing and revising  drafting process  150–52 maintenance  225 RIA see regulatory impact assessment Robinson, W.  263 schedules  121 Scott, C.  18 secondary legislation extent of  2–3 Securitel  68, 265, 266

302  Seeley, J.  18 Seidman, H.  27 separation of powers  horizontal separation of powers  5–6 vertical separation of powers  5 sincere cooperation, principle of  232 SLIM (Simpler Legislation for the Internal Market) Initiative  277, 278 Slovenia consolidated texts  183 Smart Regulation  252, 279–82 evaluation of law effectiveness and efficiency ex ante  281 improvement of implementation record  281–2 main themes of  279 selection of the ‘best possible’ legislation  282 simplification of EU law  280–81 Smismans, S.  191 Smith, A.  20 ‘soft law’  7, 11 Stefanou, C.  279–80, 284 Stigler, G.  22 stipulative definitions  133 style of legislation  137 legal-cultural identity of legislation  140–42 overall design of the legal order  137 structure and style of the individual piece of legislation  138–9 ‘unification of the law’  142 see also language subsidiarity principle  232 ‘substantial publication’  167 substantive provisions  115 Sutherland Report  275 Switzerland drafting process  157–8 Tamanaha, B.  31 targets of legislation  11 Thring, H.  115, 289, 290 title of legislation  116 short title purpose of  116–17 use of  117–18 TNA initiatives  126 training academic teaching  259

Index country analysis of training practices  264 design of training programmes  269–71, 272, 273 organisation of training programmes  266–9, 272, 273 origin and aims  264–6, 272 target audience of training  271–2, 273 governmental level  261 international associations, role of  259 ‘learning by doing’  262 legislative training at EU level  262–3 legislative training programmes development of  258, 259 national associations of legislation, role of  259–60 parliamentary level  261 professional legislative training  259 regulatory management training  263–4 specialised legislative training importance of  257–8 stages of law preparation  260, 261–2 Trakman, L.  28 transitional provisions  121 transparency  10, 37, 76–7 transposition of EU legislation  123–4 trends in legislation  15–16 unitarisation  6 United Kingdom drafting process  156–7 van Duyn  235 Varone, F.  62 Veeck v SBCCI  176 Veljanovski, C.  20, 22 vertical separation of powers  5 Voermans, W.  181 volume of legislation  4, 292–4 von Beyme, K.  11 wide maintenance  215, 216, 226 Wiederin, E.  168 Wooldridge, A.  188 World Bank regulatory quality, assessment of  30-31 Xanthaki, H.  261